6
Al-Uṣūl Al-Arba'umi'a

Etan Kohlberg

When Imāmī scholars describe the beginnings of their literary tradition, they point to two early genres out of which it grew. The first are works by the Imams themselves, such as the celebrated Nahj al-balāgha and al-Ṣaḥīfa al-sajjādiyya, attributed respectively to ʿAlī b. Abī Ṭālib and ʿAlī Zayn al-ʿĀbidīn, and various writings ascribed to Jaʿfar al-Ṣādiq and ʿAlī al-Riḍā.¹ The problem of the authenticity of these works is an issue that has still to be examined. The second genre consists of collections of Imāmī ḥadīth aptly known as uṣūl (sing. aṣl, 'source'), and it is their nature and significance which it is proposed to examine in what follows.²

I

An aṣl is a collection of a particular kind: in contrast to other collections of ḥadīth, it consists exclusively of utterances of an Imam which are committed to writing for the first time. In some cases the author of an aṣl reports traditions which he himself heard directly from the Imam,³ in others he relies on the authority of a ḥadīth scholar who transmits what he heard the Imam say.4 All uṣūl authors lived during the period of the Imams. Most are said to have been disciples of Jaʿfar al-Ṣādiq, though the compilation of uṣūl continued for several generations after his death.

Several factors combine to complicate any discussion of the uṣūl. There is first the question of terminology. The term normally used for a collection of ḥadīths is kitāb (pl. kutub). Now the word kitāb may refer either to a work based on earlier written material, or it may be used interchangeably with aṣl. Hence many uṣūl are sometimes referred to as kutub, and it is often impossible to tell whether a work called a kitāb is also an aṣl. (In fact, some authors show a marked preference for the term kitāb, even when the work to which they refer is known from other sources to be an aṣl.) Only where a particular ḥadīth scholar is credited (in the same source) with both a kitāb and an aṣl can we be reasonably certain that the kitāb in question is not an aṣl.

In the second place, there is some confusion in the sources as to the number of uṣūl. The one most commonly given is 400, hence the collective title al-uṣūl al-arbaʿumi'a. The earliest surviving reference to this number may well be that of al-Shaykh al-Mufid (d. 413/1022), who is quoted as declaring: 'From the time of ʿAlī to that of ai-Ḥasan al-ʿAskarī the Imāmis composed four hundred books called uṣūl'.⁵ These books were said to have been composed by 400 disciples of the Imams.⁶ According to al-Muḥaqqiq al-Ḥillī (d. 676/1277), however, the number 400 refers only to the uṣūl compiled by 400 disciples of Jaʿfar al-Ṣādiq; the total number was larger.7 For his part, Ibn Shahrāshūb (d. 588/1192) speaks of more than 700 uṣūl written by the Imams' disciples.⁸ Further complication is caused by the fact that Aḥmad b. Muḥammad Ibn ʿUqda (d. 333/944-5), in his work on the disciples of Jaʿfar al-Ṣādiq (Kitāb al-rijāl wa huwa kitāb man rawā ʿan Jaʿfar b. Muḥammad),⁹ is said to have listed 4,000 names;¹⁰ and although the original work is lost, it was used by Abū al-Ṭūsī (d. 460/1067) in his Kitāb al-rijāl,¹¹ and so also in subsequent Rijāl works which depend on al-Ṭūsī. Now it could be, and indeed was, argued that not all disciples of Jaʿfar al-Ṣādiq composed uṣūl;¹² but it is only natural that some scholars attributed an aṣl to each disciple mentioned by Ibn ʿUqda, thus raising the number of uṣūl to 4,000.13

Closely linked to the issue of the number of uṣūl is the identity of their compilers. As will be shown below, considerably less than 400 names can be culled from the sources available at present. Lists devoted entirely to authors of uṣūl were in fact composed, but have not come down to us in their original form. According to al-Ṭūsī, these lists were partial, and did not aim at comprehensiveness.¹⁴ The only exception was Aḥmad b. al-Ḥusayn Ibn al-Ghaḍā'irī (fl. beginning of 5th/11th century),¹⁵ who composed two lists: one of authors of uṣūl, the other of authors of other works (termed by al-Ṭūsī taṣānīf or muṣannafāt).¹⁶ Yet these lists fell into oblivion, since no one copied them during Ibn al-Ghaḍā'irī's lifetime, and they were destroyed (for some unexplained reason) by members¹⁷ of his family after his death.¹⁸ It appears in fact that none of the early lists of uṣūl authors was directly available to al-Ṭūsī: when he set out in his Fihrist to provide the names of all authors of uṣūl and muḍannafṣt known to him, he relied for information about the uṣūl solely on oral sources.¹⁹ Al-Ṭūsī acknowledges that he cannot guarantee to have covered all names, for 'it is well-nigh impossible to count the writings (taṣānīf) and uṣūl of our colleagues, since these colleagues are scattered in all corners of the earth.'²⁰

By their very nature, the uṣūl were not as a rule compiled according to subject matter; in this they were no different from most other early Imāmī works. The next phase in the history of early Imāmī literature was thus naturally the arrangement of all such works by subject matter (or according to some other principle of organization). This was achieved either by rearranging an existing text, or by collecting several early works into a larger compilation. Works of the first type are often called mubawwab, and those of the second, jāmiʿ (pl. jawāmiʿ).21 Since there are very few instances of a work surviving in both of these forms, it is usually impossible to tell whether a particular mubawwab or jāmiʿ is based directly on an aṣl (or several uṣūl) or on other intervening texts. In the majority of cases, all we have to go by are the titles of the works. From these we learn that mubawwab works often dealt with legal matters. We thus find a Kitāb mubawwab fī l-ḥalāl wa l-ḥarām,²² a Kitāb fī l-sharā'iʿ mubawwab,²³ and a Kitāb al-qaḍāyā mubawwab.²⁴ Occasionally such works are devoted to a specific subject, as in the case of a Kitāb mubawwab fī l-farā'iḍ.²⁵ Some authors are credited both with a mubawwab and a ghayr mubawwab work;²⁶ although these might conceivably be two independent texts, the former is in all likelihood a rearrangement of the latter. Indeed, it is at times explicitly stated that the same work was transmitted in both forms;²⁷ on other occasions the name of the redactor is also supplied. The Kitāb al-mashyakha of al-Ḥasan b. Maḥbūb al-Sarrād (d. 224/838-9), for example, had at least two redactors: one, Ahmad b. al-Ḥusayn b. ʿAbd al-Malik, arranged it by alphabetical order of the transmitters (ʿalā asmā' al-rijāl) (the musnad principle),28 and another, al-Kulīnī's teacher Dāwūd b. Kūra (or Kūza) al-Qummī, arranged it by subject matter (ʿalā maʿānī l-fiqh) (the muṣannaf principle).²⁹ Dāwūd b. Kūra is also credited with turning the Kitāb al-nawādir of Aḥmad b. Muḥammad b. ʿĪsā from a ghayr mubawwab into a mubawwab work.³⁰

Works of the jāmiʿ type seem to have been more widespread than the mubawwabs. Like the mubawwabs they often dealt with legal matters, either generally, as in the Kitāb al-jāmiʿ al-kabīr fī l-fiqh of Ibrāhim b. Muḥammad b. Saʿīd al-Thaqafī (d. 283/896)³¹ or the Kitāb al-jāmiʿ fī l-fiqh of Muḥammad b. Aḥmad b. ʿAbd Allāh b. Quḍāʿa,³² or with specific matters: Ibn Bābawayhi (d. 381/991), for one, wrote a Kitāb jāmiʿ nawādir al-ḥajj, Kitāb jāmiʿ li-ziyārāt al-Riḍā, etc.³³ There were some uṣūl authors who also compiled jawāmiʿ; they include ʿAlī b. Asbāṭ³⁴ and ʿAlī b. Abī Ḥamza al-Baṭā'inī, who is credited with a Kitāb jāmiʿ fī abwāb al-fiqh.³⁵

Despite the increasing predominance of the mubawwabs and jawāmiʿ, certain Imāmī scholars persisted in transmitting the uṣūl in their original form. Perhaps the most important of these in the latter half of the 3rd/9th century was Abū l-Qāsim Ḥumayd b. Ziyād (d. 310/922-3), a Kūfan who moved to Ninawā (near Karbalā),³⁶ whom al-Ṭūsī credits with transmitting most of the uṣūl (rawā l-uṣūl aktharahā).³⁷ Ḥumayd, who was active during the crucial period of the Small Occultation, appears not to have transmitted directly from uṣūl authors. The names of his most important authorities appear in al-Ṭūsī's Rijāl in the chapter devoted to those who did not transmit directly from an Imam, where they are introduced with the formula rawā ʿanhu Ḥumayd uṣūlan kathīra / kutuban kathīra min al-uṣūl. They include (in alphabetical order) Aḥmad b. Muḥammad b. Maslama,38 Aḥmad b. Muḥammad b. Zayd (d. 262/875-6),³⁹ ʿAlī b. Buzurg,⁴⁰ Ibrāhīm b. Sulaymān b. Hayyān al-Tamīmī,⁴¹ Muḥammad b. ʿAbbās (or ʿAyyāsh) b. ʿĪsā,⁴² Muḥammad b. Aḥmad b. Rajā (d. Dhū l-Ḥijja 266/July-Aug. 880),⁴³ Muḥammad b. Ḥasan (or Ḥusayn) b. Ḥāzim (d. 261/874-5),⁴⁴ Muḥammad b. Khālid al-Ṭayālisī (d. Jumādā II 259/April 873),⁴⁵ al-Qāsim b. Ismāʿīl al-Qurashī,⁴⁶ and ʿUbayd Allāh (or ʿAbd Allāh) b. Aḥmad b, Nahīk,⁴⁷ Other authorities mentioned by al-Ṭūsī include Aḥmad b. al-Ḥusayn b. Muflis,⁴⁸ Aḥmad b. Mītham b. Abī Nuʿaym,⁴⁹ ʿAlī b. Ibrāhim al-Khayyāṭ,⁵⁰ and Yūnus b. ʿAlī al-ʿAṭṭār.⁵¹ Some of these scholars were themselves authors of nawādir or kutub which are likely to have included material from the uṣūl.

The most prominent transmitters of uṣūl in the generation after Ḥumayd include the already mentioned Ibn ʿUqda,⁵² as well as Muḥammad b. Hammām al-Iskāfī (d. 332/943-4), Muḥammad b. Ḥasan Ibn al-Walīd al-Qummī (d. 343/954-5), whose Kitāb al-jāmiʿ⁵³ is one of the sources of Ibn Bābawayhi's Man lā yaḥḍuruhu l-faqīh,⁵⁴ Jaʿfar b. Muḥammad Ibn Qūlawayhi (d. 368/978-9),⁵⁵ who also composed a Fihrist of the kutub and uṣūl which he had transmitted,⁵⁶ and the lesser known Abū ʿAbd Allāh al-Ḥusayn b. Aḥmad b. Idrīs al-Qummī and Muḥammad b. Nuʿaym al-Samarqandī.57 Some of these were the authorities of Hārūn b. Mūsā al-Tallʿukbarī (d. 385/995-6), who is a pivotal figure in that he transmitted all the uṣūl available at his time.⁵⁸ It is of course impossible to know precisely how many uṣūl were at his disposal; but judging by some indirect evidence, there may not have been very many. The evidence concerns al-Tallʿukbarī's contemporary, the leading Imāmī scholar Aḥmad b. Muḥammad Abū Ghālib al-Zurārī (d. Jumādā I 368/Dec. 978), who addressed an epistle (risāla) to his grandson Abū Ṭāhir Muḥammad b. ʿUbayd Allāh. The epistle includes a list of early Imāmi works (kutub qadīma) in the possession of the family, the Āl Aʿyan, which the grandson is urged to safeguard. The list was drawn up in Dhū l-Qaʿda 356/Oct.-Nov. 967 and brought up to date in Rajab 367/Feb.-March 978.⁵⁹ Of some one hundred works in it, only 13 can be said with some certainty to be uṣūl. It is perhaps symbolic that one of the latest items included in the list is al-Kulīnī's Kāfī, the first of the 'Four Books' which constitute the basis for all subsequent Imāmī legal texts. The need for such large, comprehensive works, conveniently arranged by subject matter, became pressing after the disappearance of the Twelfth Imam; once these were available, the original uṣūl became dispensable from a practical point of view. It may be recalled at this point that Ibn al-Ghaḍā'irī compiled separate lists of authors of uṣūl and of other works, and that al-Ṭūsī in his Fihrist abandoned this method and instead combined the names of authors of uṣūl and muṣannafāt. Al-Ṭūsī's stated reason for doing so was his wish to avoid repetition;⁶⁰ an additional factor may well have been the declining importance of the uṣūl as independent sources. A further blow came from an entirely different quarter, when many (presumably rare) copies of uṣūl manuscripts were destroyed in the fire which swept the Sābūr library in al-Karkh in 451/1059.⁶¹

During the Seljuq period, little is heard of the uṣūl; Muḥammad b. Aḥmad Ibn Idrīs (d. 598/1202), who in the closing chapter of his Sarā'ir quotes from various early works, includes at least one aṣl among them.62 It took a devoted bibliophile such as ʿAlī b. Mūsā Ibn Ṭāwūs (d. Dhū l-Qaʿda 664/August 1266) to salvage some of the surviving uṣūl in the last years of the ʿAbbāsid caliphate and incorporate them in his writings. On occasion Ibn Ṭāwūs quotes from an aṣl without identifying it further, as when he declares, 'I have found in one of the ancient uṣūl of our colleagues' (wajadtu fī aṣl ʿatīq min uṣūl aṣḥābinā),⁶³ or 'I have found in an aṣl.^,64 Mostly, however, he identifies the aṣl by name, as when he quotes from the aṣl of Rabīʿ b. Muḥammad al-Aṣamm,⁶⁵ or from that of ʿUbayd Allāh b. ʿAlī al-Ḥalabī.⁶⁶ At times he refers to a work known to be an aṣl by the broader term kitāb,⁶⁷ In contrast, he may use the term aṣl in a rather loose fashion to refer to a work which is based on an aṣl but which is not, strictly speaking, an aṣl itself, as when he mentions an aṣl of Ḥasan b. Maḥbūb.68 Some of the uṣūl in Ibn Ṭāwūs's possession contained supererogatory prayers (adʿiya) which he incorporated in works such as Falāḥ al-sā'il and Muhaj al-daʿwāt.⁶⁹

In the generations following Ibn Ṭāwūs, uṣūl works continued to be copied,⁷⁰ but the practice of quoting directly from them all but ceased. The first to revive it was Muḥammad Bāqir al-Majlisī (d. 1110/1699 or 1111/1700) who, in his indefatigable search for early Imāmī works to be included in his Biḥār al-anwār, discovered - and extensively quoted from - various uṣūl. These may be divided into two groups. There are, first, anonymous uṣūl which are not mentioned in the introduction to the Biḥār, but which are used in the work itself.⁷¹ The second, much more significant, group consists of 13 uṣūl of identifiable authorship, which al-Majlisī unearthed in a majmūʿa copied in 374/985 by Manṣūr b. al-Ḥasan al-Ābī; he in turn had based himself on a manuscript written by al-Tallʿukbarī.⁷² These 13 uṣūl (plus a few others) are in all likelihood the only ones to have survived, and it is thanks to al-Majlisī that this genre regained its place in the mainstream of Imāmī literature. As will be shown below, manuscripts of the surviving uṣūl began to be copied with increasing frequency. In addition, other authors followed al-Majlisī's lead and incorporated material from the uṣūl into their writings. An interesting case in this connection is that of Muḥammad b, al-Ḥasan al-Ḥurr al-ʿĀmilī (d. 21 Ramaḍān 1104/26 May 1693), al-Majlisī's contemporary and author of the Tafṣīl wasā'il al-shīʿa ilā aḥkām al-sharīʿa, which rivals the Biḥār in importance. This work, which was almost twenty years in the making,73 was completed in mid-Rajab 1082/Nov. 1671.⁷⁴ Al-Ḥurr al-ʿAmilī's stated aim was to include in it all Imāmī texts which were available to him either directly or through quotations in later sources (bi l-wāsiṭa);⁷⁵ yet none of the 13 uṣūl used by al-Majlisī is quoted in the Wasā'il, either directly or indirectly.⁷⁶ However, in a later work, the Ithbāt al-hudāt bi l-nuṣūṣ wa l-muʿjizāt (completed in 1096/1685), al-ʿĀmilī cites from seven of these uṣūl (and would surely also have cited from the other six, had they included material relevant to his theme: the divine right of the twelve Imams to rule, and the miracles performed by them). Now another source used by al-ʿĀmilī in the Ithbāt, but not in the Wasā'il, is al-Majlisī's Biḥār; and so it would be tempting to conclude that the uṣūl traditions in the Ithbāt were taken from the Biḥār. This, however, is not the case, for in the Ithbāt al-ʿĀmilī includes the uṣūl in his list of works available to him directly.⁷⁷ He does not reveal how he suddenly gained access to these uṣūl; yet one can infer that al-Majlisī had something to do with it. The most important joumey in al-ʿĀmilī's life was his move from his native Jabal ʿĀmil to Mashhad, which he reached in 1073/1662-3. On his way he stopped in Iṣfahān, where he befriended al-Majlisī; in fact, the two gave each other ijāzas to transmit traditions which each of them had collected.⁷⁸ It is conceivable that al-Majlisī lent al-ʿĀmilī (or even gave him as a present) the majmūʿa of uṣūl at his disposal, or a copy thereof. Another, more likely, possibility is that al-ʿĀmilī learnt of the existence of these uṣūl from the Biḥār, only gaining access to a manuscript of the uṣūl after he had completed the Wasā'il.

The omission of the uṣūl from the Wasā'il was rectified by al-Nūrī al-Ṭabarsī (d. 1320/1902) in his Mustadrak al-wasā'il.79 Direct quotations from the uṣūl are to be found throughout this work, as also in other works of his, such as the Dār al-salām fīmā yataʿallaqu bi l-ru'yā wa l-manām.⁸⁰

II

The prolonged absence of the uṣūl from general use did not signal their disappearance from Imāmī consciousness. In fact, the precise significance to be attached to them proved one of the bones of contention between the Akhbārī and Uṣūlī schools. The main argument concerned the reliability to be accorded individual traditions appearing in a particular aṣl, especially when such traditions were isolated (khabar wāḥid) or were transmitted on the authority of persons not wholly reliable. For the Akhbārīs, such traditions did not present any problem. Muḥammad Amīn al-Astarābādī (d. 1033/1624), considered as the founder of the later Akhbārī school, maintains that all traditions in the uṣūl derive with certainty from an Imam (i.e. they are qaṭʿī al-ṣudūr or al-wurūd), and are therefore to be accepted without reservation as the basis for legal rulings.⁸1 In fact, al-Astarābādī insists that the uṣūl are the single most important source for Imāmī law and doctrine.⁸² They were composed by order of the Imams, who told their disciples to learn them by heart and propagate them among all Imāmīs, in preparation for the period of the Greater Occultation.⁸³ The uṣūl compiled during the lifetime of the early Imams were continually presented (ʿarḍ) to the later Imams, who verified their contents;⁸⁴ in addition, the uṣūl circulated widely and were readily available, so that no extraneous material could possibly have infiltrated them.85 Al-Astarābādī declares that the material in the Four Books derives in its entirety from the uṣūl.⁸⁶

The attitude of the Uṣūlīs (or Mujtahids) is, in contrast, more skeptical. Already al-Ṭūsī (whose writings reveal an interesting amalgam of Uṣūlī and Akhbārī tendencies)⁸⁷ acknowledges in the introduction to his Fihrist that many uṣūl authors belonged to deviant religious persuasions (al-madhāhib al-fāsida), yet he insists that the uṣūl which they compiled are nevertheless reliable (muʿtamada).⁸⁸ The deviations most often mentioned in the Rijāl works are waqf ('stopping' with a particular Imam and refusing to recognize his successor) and ghuluww (extremism). Al-Ṭūsī maintains that traditions transmitted by trustworthy wāqifīs and ghālīs may be accepted even though the deviant religious beliefs of the transmitters are to be condemned.⁸⁹ As for isolated traditions found in the uṣūl, al-Ṭūsī's position is that they are acceptable as a basis for legal rulings provided their transmitters are trustworthy (thiqa).⁹⁰

A more purely Uṣūlī view is presented by the Shahīd al-Thānī Zayn al-Dīn al-ʿĀmilī (d. 966/1558) and Bahā' al-Dīn al-ʿĀmilī (d. 1030/1621). They agree with the Akhbārīs that during the lifetime of the Imams the reliability of a particular tradition or aṣl could be checked by referring it to an Imam. The situation changed radically, however, with the occultation of the Twelfth Imam: some uṣūl disappeared (indarasat) because people, fearing the Sunnī rulers, refrained from exhibiting (iẓhār) or copying them; at the same time, unreliable uṣūl began circulating, and it became increasingly difficult to distinguish them from the reliable ones.91 The problem was compounded by the incorporation of the extant uṣūl into the Four Books. It was this state of affairs which prompted Imāmī scholars, beginning with the ʿAllāma al-Ḥillī (d. 726/1325), to devise a system based on ʿilm al-rijāl for testing the reliability of Imāmī traditions.⁹² In other words: the Uṣūlīs do not automatically accept a tradition appearing in an aṣl as reflecting the view of an Imam; in fact they subject it to the same rules as any other tradition. The uṣūl are not to be endorsed en bloc: the standing of each is determined by the reliability of its compiler.⁹³ Furthermore, not all knowledge is stored in the uṣūl; considerable scope is left for ijtihād,⁹⁴ The outstanding representative of Uṣūlī Shīʿism in the 18th century, Muḥammad Bāqir al-Bihbihānī (d. 1205/1790-1), is particularly critical of the Akhbārī contention that all uṣūl are qaṭʿī al-wurūd.⁹⁵ To prove the fallacy of this contention, he quotes Ibn Bābawayhi's claim (on which more below) that the uṣūl of Zayd al-Zarrād and Zayd al-Narsī are forged.⁹⁶ (Other uṣūl authors whose reliability is doubted are ʿAlī b. Abī Ḥamza al-Baṭā'inī,⁹⁷ Ḥasan b. Ṣāliḥ⁹⁸ and Wahb b. Wahb al-Qurashī.⁹⁹) Also, the author of an aṣl may commit a mistake while compiling his aṣl; for however high his standing may be said to be, it cannot be claimed that he enjoys ʿiṣma.¹⁰⁰ Al-Bihbihānī quotes from Ibn Bābawayhi's introduction to the Man lā yaḥ- ḍuruhu l-faqīh, in which the author states that all the material in his work derives from 'al-uṣūl wa l-muṣannafāt'. For al-Bihbihānī, this constitutes proof that this work (and, by implication, the other three of the Four Books) includes material not found in the uṣūl.101 Even the appearance of the author of an aṣl in an isnād does not necessarily mean that the tradition in question was included in his aṣl.¹⁰² In a similar vein, Shaykh Murtaḍā al-Anṣārī (d. 1281/1864) stresses that the compilers of the jawāmiʿ were aware of the infiltration of unreliable traditions into the uṣūl, and that the transition to jawāmiʿ was accordingly accompanied by a process of sifting and editing.¹⁰³ A somewhat different Uṣūlī method of tackling the same question consists in defining the term aṣl in a more restrictive way, applying it only to those collections of traditions which, in addition to not drawing on earlier works, are also reliable (majmaʿ al-aḥādīth al-ghayr muntazaʿ min ghayrihi maʿa kawnihi muʿtamadan).¹⁰⁴

III

Our knowledge of the names of uṣūl authors derives in the main from the classical Rijāl works, although the information available there can at times be supplemented by notices in other sources. It is impossible in the present state of our knowledge to recover the names of all these authors. The following list is merely an attempt to pull together the information scattered in the most important sources. As noted above, an aṣl may occasionally be concealed under the word kitāb; for this reason, whenever a particular author is credited with an aṣl in one source and with a kitāb in another, both sources appear on the list. Two asterisks denote the explicit use of the term aṣl; one asterisk denotes the use of kitāb (or kutub). Sources not marked by an asterisk are those which provide additional biographical information on the author. When it is known from which Imam (or Imams) the author transmitted, the Imam is assigned a number (in square brackets) according to his position in the line of Twelve Imams.

1.	Abān (al-Sindī) b. Muḥammad al-Bazzāz	N*. R, F*. ISh*, Ibn
	al-Bajalī [10]	Ṭāwūs (Kitāb al-iqbāl, Kashf al-maḥajja, p. 156)**
2.	Abān b. Taghlib (d. 141/758-9) [4, 5, 6]	K, IN**,105 N*, R,F**, ISh**
3.	Abān b. ʿUthmān al-Bajalī [6, 7]	K, N*, R, F**, ISh*
4.	ʿAbbād Abū Saʿīd al-ʿUṣfurī (d. 250/864)	N*. F*. Biḥar**
5.	ʿAbd Allāh b. Ḥammad	N*, R*. F*, ISh*, Ibn
	al-Anṣarī [6, 7]	Ṭāwūs (Kitāb al-iqbāl, Kashf al-maḥajja, pp. 18f)**
6.	ʿAbd Allāh b. al-Haytham	N**
7.	ʿAbd Allāh b. Sulaymān al-Ṣayrafi [4, 5, 6]	N**, R
8.	ʿAbd Allāh b. Yaḥyā al-Kāhilī [6, 7]	K, N*, R, F*, ISh*, Biḥār**
9.	ʿAbd al-Malik b. Ḥakīm aI-Khathʿamī [6, 7]	N*, F*, ISh*, Biḥar**
10.	Abū Muḥammad al-Khazzāz	N*. F**, ISh**
11.	Ādam b. al-Ḥusayn al-Nakhkhās (or al-Naḥḥās) [6]	N**, R
12.	Ādam b. al-Mutawakkil [6]	N**, F*, ISh*
13.	Aḥmad b. al-Ḥusayn b. ʿUmar [6, 7]	N*, Ibn Ṭawūs (Kitāb al-luhūf, Tehran, 1348 Sh, p. 63)**
14.	Ahmad b. Yūsuf b. Yaʿqūb al-Juʿfi [8]	N** (s.v. Jamil b. Darrāj), R, F* (lahu riwāyāt), ISh*
15.	ʿAlā' b. Razin al-Qallā' [6]	N*, R, F*, ISh*, Dhaīʿa**
16.	ʿAli b. Abi Ḥamza al-Baṭa'inī [6, 7]	K, G, N*. R*, F**, ISh**
17.	ʿAli b. Asbāṭ (d. ʿa. 250/864) [8, 9]	K, N* (nawādir), R, F**, ISh**, Biḥar**
18.	ʿAli b. Ri'āb al-Ṭaḥḥān [6, 7]	K, IN*, N*, R, F**, ISh** (lahu aṣl kabīr)
19.	Asbāṭ b. Sālim al-Kūfī Bayyāʿ al-Zuṭṭī [6, 7]	N*, R, F**,106 ISh**
20.	ʿ#X0100;ṣim b. Ḥumayd [6]	K, N*, R, F*, ISh*, Biḥar**
21.	Ayyūb b. al-Ḥurr [6, 7]	N**, R, F*
22.	Bakr b. Muḥammad al-Azdī [6, 7, 8]	K, N*, R*, F**, ISh**
23.	Bashshār b. Yasār al-Ḍubaʿi al-ʿIjli [6, 7]	N*, R,F**, ISh**
24.	Bishr (or Bashīr) b. Maslama al-Kūfī [6]	N*, R, F**, (s.v. Bashshār b. Yasār), ISh**
25.	Dāwūd b. Kathīr al-Raqqī [6, 7, 8]	K, G, N*, R, F**,107 ISh**
26.	Dāwūd b. Zirbī [6, 7]	K, N* R, F**, ISh**
27.	Dharīḥ b. Yazīd al-Muḥāribī [6, 7]	K, G, N*, R, F**, ISh**
28.	Ḥabīb b. al-Muʿallil al-Khathʿamī [6, 7, 8]	K, N*, R, F**, ISh**
29.	Ḥafṣ b. al-Bakhtarī [6, 7]	N*, R, F**, ISh**
30.	Ḥafṣ b. Sālim [6]	N*, R, F**
31.	Ḥafṣ b. Sūqa al-ʿAmrī [6, 7]	N*, R, F**,ISh**
32.	al-Ḥakam b. Ayman al-Ḥannāṭ [6, 7]	N*, R, F**
33.	al-Ḥakam al-Aʿmā b. Miskīn [6]	N*, R, F**
34.	al-Ḥārith b. Muḥammad b. ʿAlī al-Aḥwal Mu'min al-Ṭāq [6]	N*, R, F**, ISh**
35.	al-Ḥasan b. Ayyūb (d. 179/795-6) [7]	N** (lahu kitāb aṣl), R, F*.ISh*
36.	al-Ḥasan b. Mūsā b. Sālim [6]	N*, R, F**, ISh**
37.	al-Ḥasan b. Ribāṭ (or al-Ribāṭī) [5,6]	K, N*, R, F**, ISh**
38.	al-Ḥasan b. Ṣāliḥ b. Ḥayy (d. 168/784-5) (EI2, s.v.) [5, 6, 7]	K, IN*, N*, R, F** ISh**
39.	al-Ḥasan b. al-Sarī [5, 6]	N*, R, F*, ISh*, Dharīʿa**
40.	al-Ḥasan b. Ziyād al-ʿAṭṭār [6]	K, N*, R, F**, ISh**
41.	Hishām b. al-Ḥakam (d. 179/795-6) (EI2, s.v.) [6, 7]	K, IN*, N*, R, F** ISh*
42.	Hishām b. Sālim al-Jawāliqi [6, 7]	K, IN*, N*, R, F**, ISh**
43.	Ḥumayd b. al-Muthannā Abū 1-Maghrā' [6, 7]	N*, R, F**, ISh**, Ibn Ṭāwūs (al-Malāḥim wa l-fitan, Najaf, 1365/1946, p. 174)**
44.	Ḥurayz (or Ḥarīz, or Ḥafṣ) b. ʿAbd Allāh al-Sijistānī [6]	K, IN*, N*, R, F**, ISh** (lahu kutub... wa tuʿaddu kulluhā fi l-uṣūl), Ibn Idrīs (Sarā'ir)**
45.	al-Ḥusayn b. Abī l-ʿAlā' [5, 6]	F**, ISh** (lahu kitāb yuʿaddu fi l-uṣūl)
46.	al-Ḥusayn b. Abī Ghundur [7]	N*, F**, ISh**
47.	al-Ḥusayn b. Sayf al-Kindi [6]	N*, R, F*, ISh*, Ibn Ṭāwūs (Falāḥ al-sā'il, p. 95)**
48.	al-Ḥusayn b. ʿUthmān b. Sharīk (or Shurayk) [6, 7]	N*, F*, ISh*, Biḥār**
49.	Ibrāhīm b. ʿAbd al-Ḥamīd [6 , 7]	K, N*, R*, F**, ISh**
50.	Ibrāhīm b. Abī 1-Bilād [6, 7, 8]	K, N*, R*, F**, ISh**
51.	Ibrāhīm b. Mihzam al-Asadī [6, 7]	N*, R, F**, ISh**
52.	Ibrāhim b. Nuʿaym Abū l-Ṣabāḥ al-Kināni al-ʿAbdī [5, 6, 7]	K, N*, R**, F*, ISh*
53.	Ibrāhīm b. Ṣaliḥ [8, 9]	K, N*, R, F*, ISh**108
54.	Ibrahim b. ʿUmar (or ʿUmayr) al-Yamānī al-Ṣanʿānī [5, 6]	K, IN*, G*, N*, R** (lahu uṣūI), F**, ISh**
55.	Ibrahim b. ʿUthmān (or ʿIsā) Abū Ayyūb al-Khazzāz [6, 7]	K, N*, R, F**, ISh**
56.	Ibrahim b. Yaḥyā109	F**, ISh**
57.	Isḥāq b. ʿAmmār b. Mūsā al-Sābāṭī [6]	F**, ISh**
58.	Isḥāq b. Jarīr al-Bajali al-Kūfi [6, 7]	N*, R, F**, ISh**
59.	Ismāʿīl b. Abān al-Ḥannāṭ [6]	N*, R, F**,110 ISh**
60.	Ismāʿī'l b. ʿAmmār al-Kūfi [6]	R, ISh**
61.	Ismāʿīl b. Bukayr (or Bakr)	N*, F**, ISh** (both s.v. Ismāʿīl b. Dīnār)
62.	Ismāʿīl b. Dīnār	N*, F**, ISh**
63.	Ismāʿīl b. Jābir al-Juʿfi [5, 6, 7]	K, N*, R** (lahu uṣūl), F*, ISh**
64.	Ismāʿīl b. Mihrān [8]	K, IN*, G, N*, R, F**, ISh**
65.	Ismāʿīl b. Muḥammad	F**, ISh**
66.	Ismāʿīl b. ʿUthmān b. Abān	R, F**, ISh**
67.	Jābir b. Yazīd al-Juʿfi (d. 128/745-6) (EI2, Suppl., s.v. Djabir al-Djuʿfi) [5, 6]	K, N*, R, F**, ISh*
68.	Jaʿar b. Muḥammad b. Shurayḥ al-Ḥaḍramī	F*, ISh*, Biḥār**
69.	Jamīl b. Darrāj Abī l-Ṣubayḥ (or Abi ʿAlī) [6, 7]	K, N*, R, F**, ISh**
70.	Jamīl b. Ṣaliḥ al-Asadī [6, 7]	N*, R, F**, ISh**
71.	Khālid b. [Abī] Ismāʿīl [6]	N*, R, F**, ISh**
72.	Khālid b. Ṣubayḥ [6]	N*, F**, ISh**
73.	Khalḷd al-Sindī (or al-Suddī) al-Bazzāz [6]	N*, F*, ISh*, Bihar**
74.	Marwak b. ʿUbayd b. Sālim [9]	K, N** (qāla aṣḥābunā lqummiyyūn: nawādiruhu aṣl), R, F*, ISh*
75.	Masʿada b. Ziyād [5, 6]	N*, R, F* ISh*, Ibn Ṭāwūs (Risāla fi muḥāsabat alnafs)**
76.	Muhammad b. Jaʿar al-Bazzāz (or al-Razzāz) al-Qurashī	Dhariʿa**
77.	Muḥammad b. al-Muthannā b. al-Qāsim (or Muḥammad b. al-Qāsim b. al-Muthannā) al-Ḥaḍramī [6]	N*, R, F*, ISh*, Biḥār**
78.	Muḥammad b. Qays al-Asadi [5, 6]	N, R, F*, ISh*, al-Shahid al-Thānī (Sharḥ al-dirāya)**
79.	Muḥammad b. Qays al-Bajali (d. 151/768) [5, 6]	N*, R*, F**, ISh*, al-Shahid al-Thāni (Sharḥ al-dirāya)**
80.	Muthannā b. Al-Walīd al-Ḥannāṭ [6]	K, N*, F*, ISh*, Biḥār**
81.	Rabīʿ b. Muḥammad al-Muslī al-Aṣamm [6]	N*, R, F**, ISh**, Ibn Ṭāwūs (Falāḥ al-sā'il, p. 202)**
82.	Ribʿī b. ʿAbd Allāh b. al-Jārūd al-ʿAbdī al-Baṣrī [6, 7]	K, N*, R, F**, ISh**
83.	Rifāʿa b. Mūsā al-Naḥḥās (or al-Nakhkhās) al-Asadī [6, 7]	N*, R, F*, ISh**
84.	Saʿd b. Abī Khalaf al-Zām al-Zuhrī [6, 7]	K, N*, R, F**,ISh**
85.	Saʿdān b. Muslim al-ʿĀmirī [6, 7]	N*, R, F**, ISh**
86.	Saʿīd b. ʿAbd al-Raḥmān al-Aʿraj al-Sammān [6]	K, N*, R*, F**, ISh**
87.	Saʿīd b. Ghazwān al-Asadī [6]	K, N*, R, F**, ISh**
88.	Saʿīd b. Maslama [6]	N*, R, F**, ISh**
89.	Saʿīd b. Yasār al-Ḍubaʿī [6, 7]	K, N*, R, F**, ISh**
90.	Salām b. Abi ʿAmra (or ʿAmr) al-Khurāsāni [5, 6]	K, N*, F*, ISh*, Biḥār**
91.	Ṣāliḥ b. Razin [6]	N*, F**, ISh**
92.	Shihb b. ʿAbd Rabbihi al-Asadi [5, 6]	K, N*, R, F**, ISh**
93.	Shuʿayb b. Aʿyan al-Ḥaddād [6]	K, N*, R, F**, ISh**
94.	Shuʿayb b. Yaʿqūb al-ʿAqarqūqī [6, 7]	K, N*, R, F**, ISh**
95.	Sufyān b. Ṣāliḥ [6]	N*, R, F**, ISh**
96.	Sulaym b. Qays al-Hilali [1, 2, 3, 4, 5]	al-Nuʿmāni (Ghayba, Tehran 1318/1900, p. 47)**, K*, IN*, G*, N*, R, F*, ISh*
97.	Thābit b. Dīnār Abū Ḥamza al-Thumālī (d. 150/767) [4, 5, 6, 7]	K, N*, R**, (s.v. Yūnus b. ʿAli), F*, ISh*
98.	ʿUbayd Allāh b. ʿAli al-Ḥalabī [6]	N*, R, F*, Ibn Ṭāwūs, Risālat al-muwāsaʿa fi lqaḍā' (cit. Bihar. XVIII/ii, 677)**
99.	Udaym b. al-Ḥurr al-Juʿfi [6]	K, N**, R
100.	Wahb b. ʿAbd Rabbihi [5, 6]	K, N*, R, F**,111 ISh**
101.	Wahb b. Wahb Abū 1-Bakhtarī al-Qurashi [6]	K, G, N*, R, F*, ISh*, al-Māmaqāni, Tanqīḥ, I, 179**
102.	Yaʿqūb b. Shuʿayb b. Mītham [6, 7]	Al-Mufid, Risāla fi l-raddʿalā l-Ṣadūq,112 cit. al-Bihbihāni, al-Ijtihād wa l-akhbār, p.76**, N*, R*, F*, ISh*
103.	Zakariyyā b. Muḥammad al-Mu'min [6, 7, 8]	IN*, N*, R** (s.v. Aḥmad b. al-Ḥusayn), F*, ISh*
104.	Zakkār b. Yaḥyā al-Wāsiṭī [6]	IN*; R*, F**, ISh**
105.	Zayd al-Narsī [6, 7]	G*, N*, R, F**, ISh**, Bihār**
106.	Zayd al-Zarrād [6]	G*, N*, R, F**, ISh** (s.v. Zayd al-Narsī), Biḥār**
107.	Ziyād b. Marwān al-Qandī [6, 7]	K, N*, R** (s.v. Aḥmad b. Muḥammad b. Salama), F*, ISh*
108.	Ziyād b. al-Mundhir Abū l-Jārūd [5, 6]	K, IN, G, N*, R, F**, ISh**112a
109.	Zurʿa b. Muḥammad al-Ḥaḍramī [6, 7]	K, N*, R, F**, ISh**

All the names on this list appear in the Dharīʿa,113 with the exception of nos. 47, 98, 102, 103.^113a In contrast, several works presented in the Dharīʿa as uṣūl have not been included here. These works are of two kinds. First, there are those whose status was already unclear to the early Imāmī scholars. Thus al-Ṭūsī says of Aḥmad [b. Muḥammad] b. Ḥusayn (or Ḥasan) b. Saʿīd al-Qurashī: 'He is the author of a Kitāb al-nawādir, which some of our colleagues have counted among the uṣūl.'¹¹⁴ Secondly, there are works which al-Ṭihrānī would seem to regard as uṣūl through misinterpretation of his sources. For example: a number of works of Bundār b. Muḥammad b. ʿAbd Allāh are said by him to be uṣūl.¹¹⁵ This claim is based on a sentence in Ibn al-Nadīm's Fihrist¹¹⁶ which refers to 'some books of his [= Bundār] in the manner of the uṣūl (ʿalā nasaq al-uṣūl).' This expression can hardly mean that Bundār composed several uṣūl. It seems rather to carry the same meaning as ʿalā tartīb al-uṣūl, an expression used by al-Ṭūsī¹¹⁷ and explained by al-Ṭihrāni¹¹⁸ (correctly, I believe) as referring to a work which, like the uṣūl, is not arranged by subject matter.¹¹⁹ This interpretation seems to be corroborated by the fact that, according to Ibn al-Nadīm, Bundār also composed works on specific subjects (Kitāb al-ṭahāra, Kitāb al-ṣalāt, etc.).¹³⁰ Another example concerns Ibrāhim b. Muslim b. Hilāl al-Ḍarīr al-Kūfī, whom al-Ṭihrāanī mentions among the uṣūl authors¹²¹ on the basis of a passage in al-Najāshī's Rijāl122 according to which 'our masters have mentioned him among the aṣḥāb al-uṣūl.' It is doubtful, however, whether this expression is to be taken as meaning that Ibrāhīm composed an aṣl. The word ṣāḥib is ambiguous: in addition to 'composer', it can also mean 'owner' (or 'transmitter');¹²³ and the fact that Ḥumayd b. Ziyād transmitted directly from Ibrāhīm b. Muslim¹²⁴ makes it unlikely that we are dealing with the author of an aṣl.

Several further issues raised by our list deserve a brief mention. First, the names on it prove how right al-Ṭūsī was in referring to the presence of adherents of al-madhāhib al-fāsida among uṣūl authors. The list includes six wāqifīs (nos. 16, 49, 64, 103, 107, 109), three faṭḥīs (nos. 17, 57, 70), two Zaydīs (nos. 38, 108), two suspected ghulāt (nos. 25, 64), one nāwūsī (no. 3), and at least one ʿāmmī (no. 101; on no. 4 see below). Secondly, some authors (nos. 44, 54, 63) are credited with more than one aṣl each, a fact which puts paid to the neat formula '400 uṣūl by 400 authors.' Thirdly, it is noteworthy that some works described in the sources as uṣūl were devoted to a single topic. This is particularly obvious in the case of Ḥurayz b. ʿAbd Allāh al-Sijistānī (no. 44), who is said to have compiled four uṣūl, of which three are entitled fī l-ṣalāt, fī l-zakāt and fī l-ṣiyām, respectively.¹²⁵ There is no reason, indeed, why an aṢl should not be devoted to a single subject;¹²⁶ after all, an Imam could have spent a complete session answering questions on a particular issue, and his answers could then have been committed to writing by one of the disciples present. In al-Sijistānī's case, the fact that al-Ṭūsī (followed by Ibn Shahrāshūb) uses the cautious formulation 'his books are considered as uṣūl' makes it tempting to speculate that he was not entirely certain that these books really were uṣūl. It could then be further suggested that al-Sijistānī originally noted down in one aṣl all the traditions which he had heard, and that he (or one of his transmitters) subsequently divided this aṣl into several works according to subject matter. If this suggestion were correct, then these secondary works would hardly deserve the term aṣl, since they would not represent the earliest published version of the traditions included in them. The trouble with such speculation is that al-Ṭūsī employs practically identical language in referring to al-Ḥusayn b. Abī l-ʿAlā' (no. 45), whose aṣl does not seem to present any problem. Further examples of works on particular subjects which are referred to as uṣūl are the Kitāb al-malāḥim and the Kitāb al-dalāla (or al-dalā'il), both ascribed to Aḥmad b. Mītham b. Abī Nuʿaym.127 The ascription of uṣūl to this scholar is, however, problematic since, as already mentioned, he is one of the authorities of Ḥumayd b. Ziyād, and as such unlikely to have himself compiled uṣūl. Indeed, in his Fihrist al-Ṭūsī refers to these and other works of Aḥmad b. Mītham as muṣannafāt.¹²⁸

IV

As noted above, the revival of interest in the uṣūl followed al-Majlisī's discovery of a compilation in the recension of al-Tallʿukbarī. This final section consists of a brief survey of uṣūl manuscripts found in public or private libraries today, or used by Imāmī scholars from the time of al-Majlisī onwards.

I. Tehran University MS no. 962¹²⁹

This is the only compilation of uṣūl which I have seen, so I will treat it at some length. It was copied in Shawwāl 1192/Oct.-Nov. 1778 by Aḥmad b. Ḥusayn b. ʿAbd al-Jabbār al-Baḥrānī al-Khaṭṭī from a copy made in Karbalā' in 1015/1606-7 by Naṣr Allāh al-Ḥusaynī, who had based himself on a manuscript completed in Mosul in 25 Dhū l-Ḥijja 374/18 May 985 by Manṣūr b. al-Ḥasan al-Ābī from a copy of Muḥammad b. al-Ḥasan al-Qummī based on a copy made by al-Tallʿukbarī.¹³⁰ The compilation contains the following uṣūl:

1. Zayd al-Zarrād (fols. 1b-9a).¹³¹ Chain of transmission: al-Tallʿukbarī - Muḥammad b. Hammām b. Suhayl al-Kātib (d. 332/943-4) - Ḥumayd b. Ziyād (d. 310/922-3) - Abū l-ʿAbbās ʿUbayd (or ʿAbd) Allāh b. Aḥmad b. Nahīk - Muḥammad b. Abī ʿUmayr (d. 217/832-3) - Zayd. The work contains 34 traditions, in all but one of which Zayd transmits directly from Jaʿfar al-Ṣādiq. The one exception is tradition no. 10,132 which Zayd transmits from al-Bāqir through Jābir b. Yazīd al-Juʿfī.

2. Abū Saʿīd ʿAbbād al-ʿUṣfurī (fols. 9b-12a).¹³³ Chain of transmission: al-Tallʿukbarī - Ibn Hammām - Muḥammad b. Aḥmad b. Khāqān al-Nahdī al-Qalānisī - Abū Sumayna Muḥammad b. ʿAlī b. Ibrāhīm al-Ṣayrafī - ʿAbbād. Its 19 traditions comprise a variety of isnāds in which ʿAbbād transmits from various Imams (al-Ṣādiq being the latest) through one or more intermediaries. In addition, there are three traditions (nos. 13 to 15) on the authority of persons who are not Imams, and three Prophetic traditions (nos. 17 to 19) with isnāds which give the impression of being Sunnī. Only the third of these has a distinctly Shiʿī matn, in which mention is made of the Prophet's order to have Muʿāwiya and al-Ḥakam b. al-ʿĀṣ killed, and of ʿUthmān's crime in allowing the latter back into Medina in contravention of Muhammad's wishes.134

3. ʿĀṣim b. Ḥumayd (fols. 12b-25a). There are two chains of transmission: (a) Abū l-Ḥasan Muḥammad b. al-Ḥasan b. al-Ḥusayn b. Ayyūb al-Qummī - al-Tallʿukbarī - Ibn Hammām - Ḥumayd b. Ziyād (in 309/921-2) - Ibn Nahīk - Musāwir and Maslama (or Salama) - ʿĀṣim; (b) Ibn Ayyūb al-Qummī - al-Tallʿukbarī - Abū l-Qāsim Jaʿfar b. Muḥammad b. Ibrāhīm al-Mūsawī (in Egypt, in 341/952-3) - Ibn Nahīk - Musāwir and Maslama - ʿĀṣim. Although ʿĀṣim was reportedly a disciple of al-Ṣādiq, he transmits from him directlly only the first of the 97 traditions. Elsewhere, he transmits from him through various authorities, the most frequent being Muḥammad b. Muslim and Abū Baṣīr. Abū Baṣīr is also by far the most prominent authority for traditions from al-Bāqir, appearing in 26 such traditions. All the traditions in this aṣl are cited from the fifth or sixth Imam, with the exception of three Prophetic traditions, three traditions from ʿAlī, and one from Zayn al-ʿĀbidīn.

4. Zayd al-Narsī (fols. 25b-35a). Chain of transmission: al-Tallʿukbarī - Ibn ʿUqda (d. 333/944-5) - Abū ʿAbd Allāh Jaʿfar b. ^cAbd Allāh al-ʿAlawī al-Muḥammadī - Ibn Abī ʿUmayr - Zayd. Of its 49 traditions, Zayd transmits 18 directly from al-Ṣādiq and 13 directly from Mūsā al-Kāẓim. The rest are transmitted from al-Ṣādiq through one intermediary, with the exception of nos. 27 (two intermediaries), 3 (two intermediaries, from al-Bāqir) and 49 (Abū Baṣīr from al-Bāqir). This work, and the aṣl of Zayd al-Zarrād, posed a problem for Imāmī scholars ever since the previously mentioned Ibn al-Walīd al-Qummī (d. 343/954-5)135 refused to transmit them, arguing that they (together with the Kitāb Khālid b. ʿAbd Allāh b. Sudayr) were forgeries made by Muḥammad b, Mūsā b. ʿĪsā al-Sammān al-Hamdānī, who was rejected by the Qummī scholars as an extremist (daʿʿafahū l-qummiyyūn bi l-ghuluww).¹³⁶ Ibn al-Walīd's stand gained added weight since it was supported in the Fihrist of his famous disciple Ibn Bābawayhi.¹³⁷ In contrast, Ibn al-Ghaḍā'irī, ¹³⁸ known as a stern critic, argues for the authenticity of the two uṣūl, pointing out that he saw manuscripts of these works which (like the one described here) were transmitted on the authority of Ibn Abī ʿUmayr, a disciple of the Imams al-Riḍā and al-Jawād who was acclaimed for his trustworthiness and reliability.¹³⁹ Ibn al-Ghaḍā'irī's judgment seems to have won the day with many Imāmī scholars, from al-Majlisī,¹⁴⁰ the ʿAllāma Baḥr al-ʿUlūm al-Ṭabāṭabā'ī (d. 1212/1797-8)¹⁴¹ and Ṣadr al-Dīn Muḥammad b. Ṣāliḥ al-ʿĀmilī (d. 1264/1848)142 to al-Nūrī al-Ṭabarsī¹⁴³ and Muḥsin al-Amīn (d. 1952).¹⁴⁴ Al-Ṭabāṭabā'ī is particularly critical of the Qummī scholars for 'hastening to criticize for no apparent reason.'¹⁴⁵ Other voices, however, can also be heard. As might be expected, they include Uṣūlī scholars such as al-Bihbihānī, for whom the existence of a forged aṣl is useful ammunition in the battle with the Akhbārīs.¹⁴⁶ Another, perhaps less biased, voice is that of the contemporary Imāmī scholar Muḥammad Taqī al-Tustarī (born 1321/1903-4), who declares that he read the two uṣūl and detected in the Aṣl Zayd al-Narsī various traditions which, from a strict Imāmī viewpoint, are objectionable. Most of these are traditions of an anthropomorphic tendency; the rest are traditions containing legal opinions not found in other Imāmī texts. Al-Tustarī acknowledges that he could find nothing wrong with the contents of the Aṣl Zayd al-Zarrād,¹⁴⁷ He insists, however, that Ibn al-Walīd was a renowned authority whose views cannot simply be discarded; he therefore suggests that the original version of the aṣl also contained objectionable material, but that this material has been deleted from the version which has come down to us.¹⁴⁸

5. Jaʿfar b. Muḥammad b. Shurayḥ al-Ḥaḍramī (fols. 35b-48b). Chain of transmission: al-Tallʿukbarī - Ibn Hammām - Ḥumayd b. Ziyād - Abū Jaʿfar Aḥmad b. Zayd b. Jaʿfar al-Azdī al-Bazzāz (known as Bazīʿ) - Muḥammad b. al-Muthannā al-Ḥaḍdramī - Jaʿfar. With 123 traditions, this is the longest of the uṣūl in this compilation. Almost all the traditions go back to al-Bāqir or al-Ṣādiq, normally through two intermediaries; the exceptions are a bloc of traditions (nos. 87-104) transmitted from al-Ṣādiq through ʿAbd Allāh b. Ṭalḥa al-Nahdī. In the first half of the work, practically all traditions carry the isnād: Ḥumayd b. Shuʿayb al-Subayʿī from Jābir b. Yazīd al-Juʿfī; it is used for traditions from both al-Bāqir (nos. 1-18, 29-38) and al-Ṣādiq (nos. 19-28, 41-86).

6. Muḥammad b. al-Muthannā al-Ḥaḍramī (fols. 49a-57a). Chain of transmission: as in the previous aṣl, as far as Muḥammad b. al-Muthannā. There is in fact a striking connection between these two uṣūl for, as Ibn al-Muthannā himself declares (at the end of tradition no. 54), Ibn Shurayḥ was his authority for all but two traditions in this aṣl (nos. 53, 54). Ibn Shurayḥ in turn transmits the bulk of the traditions from al-Ṣādiq via Dharīḥ al-Muḥāribī (nos. 1-22, 33-46, 48), and all the traditions from al-Bāqir via Dharīḥ and ʿUmar b. Ḥanẓala (nos. 23-34). For the last two traditions see the Addenda.

7. ʿAbd al-Malik b. Ḥakīm (fols. 57b-59b). Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — ʿAlī b. al-Ḥasan b. ʿAlī b. Faḍḍāl al-Taymulī — Jaʿfar b. Muḥammad b. Ḥakim — his uncle ʿAbd al-Malik. This is the first of a group of short uṣūl in the latter part of the majmūʿa. It contains six traditions, all quoting an Imam (ʿAlī, al-Bāqir or al-Ṣādiq) through one or two intermediaries.

8. Muthannā b. al-Walīd al-Ḥannāṭ (fols. 59b-61b). Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — Ibn Faḍḍāl — ʿAbbās b. ʿĀmir al-Qaṣabānī (or al-Qaṣabī) — Muthannā. Only one of the 23 traditions (no. 5) is quoted directly from al-Ṣādiq; the rest are quoted (almost all from al-Ṣādiq) via one intermediary, most commonly Abū Baṣīr (nos. 4, 7, 11-14, 17-19, 23).

9. Khallād al-Sindī (or al-Suddī) (fols. 61b-62a). Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — Yaḥyā b. Zakariyyā b. Shaybān — Ibn Abī ʿUmayr - Khallād. Of its 8 traditions, the first two are quoted directly from al-Ṣādiq.

10. Al-Ḥusayn b. ʿUthmān (fols. 62b-65a) Chain of transmission: as in the Aṣl Zayd al-Narsī (but ending of course with al-Ḥusayn). Three of its 44 traditions (nos. 3, 35, 37) are quoted directly from Mūsā al-Kāẓim, the rest from the fifth, sixth or seventh Imam through one intermediary, except for two traditions having two intermediaries. This aṣl is characterized by a particularly large proportion (over a third) of traditions quoted on the authority of anonymous transmitters (nos. 5-7, 10, 16, 21-23, 25-28, 30, 32, 33, 36, 44).

11. ʿAbd Allāh b. Yaḥyā al-Kāhilī (fols. 65a-66b). Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — Muḥammad b. Aḥmad b. al-Ḥasan b. al-Ḥakam al-Qaṭawānī (or al-Qaṭirānī) — Aḥmad b. Muḥammad b. Abī Naṣr al-Bizanṭī — ʿAbd Allāh. It contains 13 traditions, the first of them transmitted directly from al-Ṣādiq, three others (nos. 3, 4, 10) transmitted directly from al-Kāẓim.

12. Salām b. Abī 'Amra (fols. 66b-68a). Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — al-Qāsim b. Muḥammad b. Ḥasan (or Ḥusayn) b. Ḥāzim — ʿAbd Allāh b. Jabala al-Kattānī (or Kinānī) (d. 219/834) — Salām. None of its 10 traditions is transmitted directly from an Imam.

13. An excerpt from the Nawādir ʿAlī b. Asbāṭ (fols. 68a-73b). It seems fairly safe to assume that this work is identical with the aṣl ascribed to the same author. Some residual doubts must remain, however, owing to the fact that an identity between ʿAlī b. Asbāṭ's aṣl and his nawādir (the former mentioned by al-Ṭūsī, the latter by al-Najāshī) is claimed only by recent authors.149 In fact, the Persian title at the head of the majmūʿa bespeaks some caution, referring as it does to 'Twelve of the four hundred uṣūl, together with an excerpt from the Nawādir ʿAlī b. Asbāṭ.' Chain of transmission: al-Tallʿukbarī — Ibn ʿUqda — Ibn Faḍḍāl — ʿAlī. There are 30 traditions, most of them quoted from the Imams through two intermediaries.

The compilation ends with a story (khabar fī l-malāḥim) (fols. 73b-74b) in which various calamities foretold by al-Ṣādiq are shown actually to have occurred. This khabar is clearly unconnected to the preceding Nawādir,¹⁵⁰ except that, like the rest of the majmūʿa, it is transmitted on the authority of al-Tallʿukbarī. He heard it in Muḥarram 328/Oct.-Nov. 939 from Abū l-Qāsim ʿAlī b. al-Ḥasan b. al-Qāsim al-Yashkurī al-Khazzāz al-Kūfī, known as Ibn al-Ṭabbāl,¹⁵¹ who had it from Muḥammad b. Maʿrūf al-Khazzāz al-Hilālī, a centenarian disciple of al-Ṣādiq.¹⁵²

Several points should be noted concerning the uṣūl in this majmūʿa. First, many appear to be incomplete. This is stated explicitly in the case of ʿAlī b. Asbāṭ, suggested by al-Tustarī for Zayd al-Zarrād, and claimed by al-Ṭihrānī for seven further uṣūl.¹⁵³ Secondly, of the 13 authors, three (ʿAbbād al-ʿUṣfurī, ʿAlī b. Asbāṭ and Jaʿfar b. Muḥammad b. Shurayḥ) are relatively late. The other ten are disciples of Jaʿfar al-Ṣādiq (and sometimes also of al-Bāqir or al-Kāẓim); yet what is particularly noticeable is the scarcity of direct transmission from an Imam (the uṣūl of the two Zayds excepted). Now this fact, though somewhat surprising, can still be accommodated within the terms of reference of the genre. Of greater interest is the fact that some of the intermediaries between author and Imam are themselves known to have compiled uṣūl. Cases in point are Jābir b. Yazīd al-Juʿfī, who figures prominently in the aṣl of Jaʿfar b. Muḥammad b. Shurayḥ, and Dharīh al-Muḥāribī, who appears in more than half the traditions of the aṣl of Muḥammad b. al-Muthannā. In both cases, the traditions in question are grouped together; and it is only natural to suspect that they belonged originally to the uṣūl of al-Juʿfī and al-Muḥāribī, whence they were taken into the later texts. If this is in fact what happened (and there is no way to establish it with certainty), then the later works, or at least those parts so derived, could not properly be called uṣūl; the traditions contained in them would have been taken from earlier written works. Thirdly, it is perhaps not superfluous to note that only three of the thirteen works are called uṣūl by authors preceding al-Majlisī; it thus remains an open question whether the remaining ten are genuine uṣūl which had escaped the notice of the early biographers (or which these biographers, knowing that they were uṣūl, none the less preferred to call kutub), or whether their designation as uṣūl is a relatively late development, calculated to emphasize their antiquity. What remains clear, however, is that all 13 texts are truly representative of Imāmī literature as a whole: they contain a great deal of legal material, as well as traditions on the imamate, faḍā'il of the Imams, prayers, anti-Sunnī polemics and eschatology.

II. Tehran University Law Faculty MS no. 182j, fols. 1b-72a.¹⁵⁴

This manuscript was copied in 1308/1890-1 by Shaykh Naṣr Allāh Qazwīnī, and is also based on the manuscript completed in 374/985 by Manṣūr b. al-Ḥasan al-Ābī from a copy of Muḥammad b. al-Ḥasan al-Qummī, based in turn on al-Tallʿukbarī's copy.¹⁵⁵ The uṣūl in this compilation (together with the khabar fī l-malāḥim) appear in the same order as in the previously mentioned manuscript, except that four additional items are included: the Kitāb Durust, which appears at the beginning (fols. 1b-7b); the Kitāb Muḥammad b. Jaʿfar al-Qurashi (fols. 38a-43a), placed between the Kitāb Muḥammad b. Muthannā al-Ḥaḍramī and the Kitāb ʿAbd al-Malik b. Ḥakīm; and, at the end of the majmūʿa, a marthiya on Ḥusayn, and the Kitāb al-diyāt, transmitted156 by Ẓarīf b. Nāṣiḥ.¹⁵⁷ The marthiya, of course, is not an aṣl; nor can any of the three remaining items be said with certainty to be one. Durust b. Abī Manṣūr al-Wāsiṭī, a disciple of the sixth and seventh Imams, is credited with a kitāb by both al-Najāshī¹⁵⁸ and al-Ṭūsī,¹⁵⁹ but nowhere is an aṣl of his mentioned. Thus the only reason to suppose that the Kitāb Durust is also an aṣl is its inclusion in this majmūʿa. The impression gained from the incipit (āghāz), as quoted in the Catalogue, is that the original beginning of the work is missing.¹⁶⁰ The author of the second item appears to be Muḥammad b. Jaʿfar al-Bazzāz (variants: al-Razzāz, al-Zarrār), a great-uncle of Abū Ghālib al-Zurārī.¹⁶¹ In the classical sources he is not credited with a kitāb, let alone an aṣl; indeed, al-Ṭihrānī is apparently the first to mention an aṣl of his.¹⁶² Contrary to his normal practice, al-Ṭihrānī does not state on which sources he bases himself in this case, and merely maintains that this is one of the uṣūl of which only excerpts have come down to us (min al-uṣūl al-mukhtaṣara l-mawjūda). The conclusion must therefore be that al-Ṭihrānī drew his information from a manuscript collection of uṣūl, perhaps the one under consideration.¹⁶³ In fact, Muḥammad b. Jaʿfar's work is not included in any other manuscript of uṣūl that I know of, nor do I know of any Imāmī author other than al-Ṭihrānī who includes it in his survey of uṣūl (extant or otherwise). It thus remains a moot point whether it was dropped (by accident or design) from al-Tallʿukbarī's version at some stage in the transmission, or whether by contrast it was introduced into the majmūʿa at some unknown date. The Kitāb al-diyāt, finally, is often ascribed to ʿAlī b. Abī Ṭālib, but it is not clear whether ʿAlī is also to be taken to have written it down himself, or whether this was done by one of his disciples (or by a disciple of a later Imam). Only in the latter case would we be justified in referring to it as an aṣl.164

III. A majmū˓a in the possession of Muḥammad ˓Alī al-Ya˓qūbī in Najaf¹⁶⁵

This is a fairly recent manuscript, copied in Shawwāl 1336/July-Aug. 1918 by Muḥammad Ḥusayn b. al-Kāẓim al-Mūsawī al-Qazwīnī. Judging from the information in RIMA, it contains all the items appearing in Tehran University MS no. 962, though in a somewhat different order. It also includes excerpts from the aṣl of al-ʿAlā' b. Razīn¹⁶⁶ and the Kitāb al-zuhd of al-Ḥusayn b. Saʿīd al-Ahwāzī (d. after 300/913).¹⁶⁷

In addition to these manuscripts, compilations of uṣūl were seen and utilized by the following late Imāmī authors:

1. Al-Majlisī in his Biḥār.¹⁶⁸ The uṣūl mentioned by him are the same as those in Tehran University MS no. 962.

2. Al-Ḥurr al-ʿĀmilī in his Ithbāt al-hudāt.¹⁶⁹ Traditions from the uṣūl of the following are cited: ʿAbbād al-ʿUṣfurī,¹⁷⁰ Zayd al-Narsī,¹⁷¹ Jaʿfar b. Muḥammad b, Shurayḥ,¹⁷² Muḥammad b. al-Muthannā,¹⁷³ Salām b. Abī ʿAmra,174 ʿAbd al-Malik b. Ḥakīm,¹⁷⁵ the Nawādir of ʿAlī b. Asbāt.¹⁷⁶

3. The Indian scholar Iʿjāz Ḥusayn al-Kantūrī (d. 17 Shawwāl 1286/20 Jan. 1870). He clearly had access to the same uṣūl as al-Majlisī, since he quotes the incipits of all 13 uṣūl in his list of Imāmī books.¹⁷⁷ Al-Kantūrī also includes the malāḥim story, presenting Ibn al-Ṭabbāl as its author.¹⁷⁸ A peculiarity of al-Kantūrī's text is the conflation of Zayd al-Zarrād and Zayd al-Narsī: their uṣūl are presented as two works by the same author, one Zayd al-Zarrād al-Nabarsī (sic).¹⁷⁹ Since we are not told which manuscript was used, it remains unclear whether al-Kantūrī misread his source, or whether he was merely reproducing an error in the manuscript.

4. Al-Nūrī al-Ṭabarsī in his Mustadrak al-wasā'il (and other works),¹⁸⁰ The uṣūl used in the Mustadrak are likewise those of the Biḥār, but include in addition the final section of the Kitāb Durust, as well as the Mukhtaṣar kitāb al-ʿAlā' b. Razīn, and the Kitāb al-diyāt.¹⁸¹

Each aṣl mentioned by these four authors is introduced by its chain of transmission; all chains are identical to those of Tehran University MS no. 962.

5. Muḥammad Taqī al-Tustarī mentions in his biographical dictionary Qāmūs al-rijāl¹⁸² that he had seen a majmūʿa of 14 uṣūl in the library of the Muḥaddith Niʿmat Allāh al-Jazā'irī (d. 23 Shawwāl 1112/2 April 1701).¹⁸³ The information which al-Tustarī provides on this library appears to be restricted to the observation that 'what remains of it is to be found in our town.'184 The town in question must be Tustar, where the Muḥaddith al-Jazā'irī occupied the position of shaykh al-islām towards the end of his life.¹⁸⁵ I have been unable to discover further details about the library, or the specific manuscript referred to by al-Tustarī.

6. According to the Dharīʿa,¹⁸⁶ a compilation of 16 uṣūl was published by Ḥasan Muṣṭafawī in Tehran (1371/1951-2, 174 pp.). No details are given of these uṣūl, except that one of them is the excerpt from the Nawādir ʿAlī b. Asbāṭ already referred to; nor is any information provided about the manuscript(s) on which this publication is based. My attempts to discover a copy of this book have so far proved unsuccessful; but it stands to reason that most (if not all) uṣūl in it will be those found in one or more of the manuscripts mentioned above.¹⁸⁷ If Muṣṭafawī's is a critical edition, then efforts should be made to make it more accessible; if it is not, then such an edition remains a desideratum.¹⁸⁸

Appendix: Al-Khumayanī on the Uṣūl

Arguably the most radical approach to the uṣūl is that of Rūḥ Allāh al-Khumaynī. I came across his analysis of the subject after this paper had gone to the press, during a visit (in October 1984) to the Shīʿa library of the Oriental Seminar at Cologne University. Al-Khumaynī's analysis is tucked away in an unlikely place: the second volume of his Kitāb al-ṭahāra (Qumm, 1376/1956-7, pp. 184ff.) Discussing the legal status of various beverages, he refers to a virtual ijmāʿ among Imāmī jurists concerning juice produced from raisins: it is not an intoxicant and may therefore be consumed. The only dissenting opinion appears in a tradition included in the Aṣl Zayd al-Narsī (fol. 35a in the Tehran University MS 962; cit. al-Nūrī al-Ṭabarsī, Mustadrak, III, 135), according to which such juice is to be treated as an intoxicant and so may not be consumed until two-thirds have evaporated. Al-Khumaynī's objection to this tradition leads him to a general appraisal of the uṣūl. He points out that none of the early masters is on record as having defined an aṣl. All we know from al-Mufīd and al-Ṭūsī is that an aṣl is a particular kind of kitāb; nowhere do they state that it consists of traditions written down for the first time, or that traditions included in an aṣl are ipso facto reliable (Kitāb al-ṭahāra, II, 197f). Indeed, some renowned disciples of the Imams (such as Abū Baṣīr Layth al-Murādī) are not credited with an aṣl, while others of dubious reliability are known to have compiled uṣūl. The fact that some uṣūl are ascribed to persons who did not meet an Imam also undermines the traditional interpretation of the term; al-Khumaynī implicitly rejects the view that such persons only transmitted uṣūl and did not actually compile them (ibid., pp. 201f). Al-Khumaynī's own hypothesis is that an aṣl is a collection of traditions heard from an Imam, directly or through an intermediary, often — but not always — written down for the first time (ibid., pp. 198, 203). Such a collection might consist of several kutub which relate to the whole as do branches (furūʿ) to their root (aṣl) (ibid., pp. 197f, 200). Regarding contents, al-Khumaynī originally assumed that the term aṣl indicated a preoccupation with doctrinal and theological issues (uṣūl al-dīn) (ibid., pp. 202f); he later changed his mind and opted for the view that while the uṣūl merely preserved Imāmī traditions, other works (called muṣannafāt) elaborated upon the legal and doctrinal elements embedded in these traditions (ibid., pp. 201, 204f). The message, at any rate, is clear: the uṣūl enjoy no privileged status in Imāmī literature. Al-Khumaynī shares the suspicion of some scholars (referred to in this article) that the uṣūl manuscripts available to al-Majlisī and later authorities do not reflect the original text. Specifically, he points out that nothing is known of the copyist Manṣūr b. al-Ḥasan al-ābī (ibid., pp. 206-8).

Not surprisingly, al-Khumaynī reserves his harshest criticism for the aṣl of Zayd al-Narsī. In his view, the inclusion of Ibn Abī ʿUmayr in its isnād says nothing about the reliability of the matn; all it means is that Ibn Abī ʿUmayr can be trusted actually to have transmitted this aṣl from al-Narsī (ibid., pp. 186f). Defenders of al-Narsī use the fact that two traditions from his aṣl were incorporated by al-Kulīnī in his Kāfī to plump for the reliability of this aṣl; al-Khumaynī turns this argument on its head by maintaining that the inclusion in the Kāfī of two traditions only must mean that al-Kulīnī did not entertain a high regard for al-Narsī's aṣl (ibid., pp. 205f).

Addenda

I. I have now been able to consult Muṣṭafawī's edition of the uṣūl, which was kindly put at my disposal by Professor Wilferd Madelung. As I had supposed, the 16 texts in it correspond to those used by al-Nūrī al-Ṭabarsī. They consist of the 13 uṣūl of Tehran University MS no. 962, followed by the Kitāb al-diyāt (pp. 134-148), the Mukhtaṣar aṣl ʿAlā' b. Razīn (pp. 150-157), and the excerpt from the Kitāb Durust b. Abī Manṣūr (pp. 158-169). The edition is based on a single manuscript, on which no information is provided. The following points may be worth noting:

(a) All chains of the first 13 uṣūl are identical to those of MS 962. The printed edition contains some minor additions and differs slightly in the arrangement of the material: in the Aṣl Zayd al-Zarrād, tradition no. 25 (according to my numbering of MS 962) follows no. 26; in the Aṣl ʿĀṣim b. Ḥumayd, there is an additional tradition after no. 8; there are two additional traditions in the Aṣl Zayd al-Narsī (following nos. 6 and 37); in the Aṣl Jaʿfar b. Muḥammad b. Shurayḥ, a comparison of no. 12 with the printed edition reveals that there is an omission in the MS, and that no. 12 consists in fact of two separate traditions. Also, in the printed edition, nos. 49-56 follow no. 62.

(b) Re Aṣl Muḥammad b. al-Muthannā al-Ḥadramī: in MS 962, no. 52 is followed by the cryptic statement: 'This is the last tradition of Muḥammad b. al-Muthannā al-Ḥadramī from (ʿan) the tradition of Muḥammad b. Jaʿfar al-Qurashī'. The same sentence occurs in the printed edition (p. 93), but with a crucial difference: for ʿan it has wa yatlūhu. In other words: what follows is supposedly the Ḥadīth Muḥammad b. Jaʿfar al-Qurashī. In fact, however, only no. 55 qualifies for this title, as is confirmed by the statement at its end. It is thus probable that the sentence in question was misplaced, appearing originally after no. 54. It is this Ḥadīth which in all likelihood is referred to as an aṣl in the Dharīʿa. The last tradition (no. 56) does not seem to belong with any of the preceding material.

(c) As in MS 962, the Khabar fī l-malāḥim is attached to the Nawādir ʿAlī b. Asbāṭ.

(d) At the end of the book (p. 170) there appears a comment by Muḥammad b. al-Ḥasan al-Ḥurr al-ʿĀmilī, in which he states that most of the traditions in the 14 (!) books are cited in the Kāfī and other reliable works, while the rest have their contents confirmed by these works. Al-ʿĀmilī says that he came across two objectionable traditions only, and that these can be explained away (e.g. as instances of taqiyya). In this comment, al-ʿĀmilī does not reveal which two traditions he has in mind. Their identity can, however, be established by reference to the marginal notes of MS 962, many of which were written by al-ʿĀmilī and initialled M-D-Ḥ (= Muḥammad al-Ḥurr). Two of these notes deal with problematic traditions: nos. 15 and 30 in the Aṣl Zayd al-Narsī. No. 15 maintains, against the classical Imāmī doctrine, that there may be periods without a ḥujja; al-ʿĀmilī in his note interprets this as taqiyya, or as a reference to the period of the ghayba. No. 30 describes how God, riding a camel, descends to earth on the day of ʿArafa; and al-ʿĀmilī suggests that this should be interpreted metaphorically. Al-ʿĀmilī was not alone worrying about these two traditions: they are among several to be later criticized by Muḥammad Taqī al-Tustarī.

My thanks are also due to Professor Madelung for lending me his copy of al-Jalālī's Dirāsa (see above, n. 2), which provides a useful survey of much of the available information on the uṣūl.

II. The uṣūl are used as a weapon in a strongly anti-Shīʿī (and anti-Khumaynī) tract by Dr. Aḥmad al-Afghānī. In a work entitled Sarāb fī Īrān: kalimāt sarīʿa ḥawla l-Khumaynī wa dīn al-shīʿa (n. pl., 1402/1982, p. 36) the author recalls that the Four Books of the Shīʿa developed from 400 works written by 400 Shīʿīs. He then maintains that this is an exact parallel to the situation obtaining in the Christian world, where 400 gospels (anājīl) which existed in Europe in the days before Martin Luther gradually underwent changes until they became four gospels. It is not clear whether this is al-Afghānī's invention or whether he heard it from someone else; the anti-Shīʿī message is, however, clear enough.

List of Abbreviations

• Biḥār — Muḥammad Bāqir al-Majlisī, Biḥār al-anwār, lith. ed., [Iran], 1305/1887-8/1315-1897-8.
• Dharīʿa — Āghā Buzurg al-Ṭihrānī, al-Dharīʿa ilā taṣānīf al-shīʿa, Najaf, 1936-8, Tehran, 1941-78.
• F — Abū Jaʿfar al-Ṭūsī, Fihrist, ed. A. Sprenger a. o., Calcutta, 1853-5; F², ed. Muḥ. Ṣādiq Āl Baḥr al-ʿUlūm, Najaf, 1380/1961 (usually cited only where the text deviates from Sprenger's edition).
• Fawā'id — Muḥ. Amīn al-Astarābādī, al-Fawā'id al-madaniyya, [Iran], 1321/1903.
• G — Ibn al-Ghaḍā'irī, Rijāl, as cited in Q.
• IN — Ibn al-Nadīm, Fihrist, ed. R. Tajaddud, Tehran, 1391/1971.
• ISh — Ibn Shahrāshūb, Maʿālim al-ʿulamā', Najaf, 1380/1961.
• K — al-Kishshī, Rijāl, ed. A. al-Ḥusaynī, Najaf, n.d.
• N — al-Najāshi, Rijāl, Bombay, 1317/1899-1900.
• Q — al-Quhpā'ī, Majmaʿ al-rijāl, Isfahan, 1384/1964-5 - 1387/1967-8.
• R — al-Ṭūsī, Rijāl, Najaf, 1381/1961.

1 GALS, I, 705; GAS, I, 526-31, 535f. See also Mir Dâmâd, al-Rawāshiḥ al-samāwiyya, Tehran, 1311/1893-4, p. 99.

² For a discussion of the uṣūl consult Dharīʿa, II, 125-67, XXIV, 147f, 315-8; A. Falaturi, "Die Zwōlfer-Schia aus der Sicht eines Schiiten: Probleme ihrer Untersuchung", Festschrift Werner Caskel, ed. E. Grāf, Leiden, 1968. pp. 62-95, at pp. 64f, and the literature cited there; Ḥusayn al-Jalālī, Dirāsa ḥawl al-uṣūl al-arbaʿimi'a, Tehran, 1394/1974, mentioned in Dharīʿa, XXIV, 148 (see the Addenda); B. Scarcia Amoretti, "L'introduzione al Qâmûs ar-riǧâl di Tustarî: per una guida alla lettura dei testi prosopografici imamiti", Cahiers d'onomastique arabe, Paris, 1979, pp. 37-49, at pp. 40f.

³ This is similar to the amālī. For an analysis of the connection between the two genres see Muḥ. Mahdī al-Kharsān's introduction to Ibn Bâbawayhi's Amālī, Najaf, 1389/1970, pp. 13f, 18ff, 28.

⁴ Contrast the usage by some Sunnī authors of the term aṣl in the sense of 'an original copy' (of a particular work), M.M. Aʿẓamī, Studies in Early Ḥadīth Literature, 2nd ed., Indianapolis, 1978, p. 29. When al-Majlisī refers to the evil, lie-infested uṣūl of the Sunnīs which are transmitted on the authority of a group of munāfiqūn (Biḥār, VIII, 252), he is applying to the Sunnī world a purely Imāmī concept. — The term nuskha as used in Imāmī literature (especially by al-Najāshī) is usually assumed to have the same meaning as aṣl (see Dharīʿa, XXIV, 147f), and this assumption may well be correct. However, it might be suggested that a nuskha differs from an aṣl in that it was copied by a disciple from a text written by the Imam himself. If true, this suggestion could explain why works referred to as nuskha by al-Najāshī are not called aṣl by other authors. For non-Shīʿī usages of nuskha see Aʿẓamī, loc. cit.

⁵ ISh, p. 3. See also al-Ḥurr al-ʿĀmilī, Kitāb al-īqāẓ min al-hajʿa, ed. H. al-Rasūlī al-Mahallātī, Qumm, 1381, p. 25.

⁶ Al-Shahīd al-Thānī, Sharḥ risālat dirāyat al-hadīth, Najaf, n.d., p. 17, cit. Fawā'id, p. 59, Yūsuf al-Baḥrānī, al-Durra al-najafiyya, Tehran, 1314/1896, p, 169, al-Narāqī, Manāhij al-aḥkām wa l-uṣūl, Tehran, 1269/1878, unpaginated, chapter entitled al-Ijtihād wa l-taqlīd; al-Bahā'ī, al-Risāla l-wajīza, Tehran, 1356, p. 8, cit. Fawā'id, al-Narāqī, loc. cit.; Mīr Dāmād, op. cit., p. 98 (al-mashhūr anna l-uṣūl arbaʿumi 'at muṣannaf li arbaʿimi'at muṣannif). According to al-Narāqī (op. cit., chapter entitled al-Taʿabbud bi l-khabar al-wāḥid al-mujarradʿan al-qarā'in), these 400 were all disciples of either al-Bāqir or al-Ṣādiq. See also Mīr Dāmād, loc. cit.; Dharīʿa, II, 129f; GAS, I, 525.

⁷ Al-Muḥaqqiq al-Ḥillī, al-Muʿtabar, Tehran, 1318/1901, p. 5, cit. Fawā'id, p. 60.

⁸ Ibn Shahrāshūb, Manāqib āl Abī Ṭālib, Najaf, 1376/1956-7, I, 218. One hundred of these uṣūl were supposedly based on answers provided by Jaʿfar al-Ṣādiq to his disciples (ibid., II, 43). The number 700 clearly conflicts with al-Mufid's view as quoted (with apparent approval) by Ibn Shahrāshūb (see above n. 5). (Or is the sabʿumi'a of the text a corruption of arbaʿumi'a?). Ibn Shahrāshūb emphasizes that compilation of the uṣūl continued until the (Smaller?) Occultation (ibid., I, 258).

⁹ N, p. 69; F, pp. 42f (read, as in F², p. 53, wa huwa kitāb man for wa kitāb man); Dharīʿa, X, 86f, no. 161.

¹⁰ Al-ʿAllāma Ibn al-Muṭahhar al-Ḥillī, Khulāṣat al-aqwāl, Tehran, 1310/1892-3, p. 98, cit. al-Nūrī al-Ṭabarsī, Mustadrak al-wasā'il, Tehran, 1382/1962-3 - 1384/1964-5, III, 770. Al-Mufid (al-Irshād, n. pl., 1320/1902-3, p. 249 [= p. 408 in the translation of I.K.A. Howard, London, 1981]), al-Faḍl b. Ḥasan ai-Ṭabarsī (Iʿlām al-warā, Najaf, 1390/1970, p. 284) and Ibn Shahrāshūb (op. cit., II, 43) all refer to 4,000 disciples of al-Ṣādiq (without however mentioning Ibn ʿUqda), while al-Muḥaqqiq al-Ḥillī in his Muʿtabar, p. 5 (cit. Fawā'id, p. 85) speaks of 'almost 4,000 men.' All these sources are quoted in al-Nūrī al-Ṭabarsī, loc. cit. See also Hāshim al-Baḥrānī, Ḥilyat al-abrār, Qumm, 1397/1977, II, 145f. Cf. Seyyed Hossein Nasr's introduction to A Shiʿite Anthology, trans. W.C. Chittick, State University of New York Press, Albany, 1981, p. 9: 'The number of students, both Shiʿite and Sunni, trained by the sixth Imam Jaʿfar al-Ṣādiq has been estimated at four thousand.' See also Ḥasan al-Ṣadr, al-Shiʿa wa funūn al-islām, Beirut, n.d., pp. 45f. Ibn ʿUqda is held in high esteem by Imāmī scholars, despite his Jārūdī Zaydī beliefs (N, p. 68; F, p. 42). According to some authorities, the number of these disciples exceeded 4,000: Abū l-ʿAbbas Aḥmad b. ʿAlī b. ʿAbbās b. Nūḥ (or Aḥmad b. Muḥammad b. Nūḥ) (fl. beginning of 5th/11th century) is credited with a work on al-Ṣadiq's disciples, in which he added considerably to Ibn ʿUqda's list (F, p. 48). The work, however, like all others by this author, was considered lost already in al-Ṭūsī's lifetime (ibid.).

¹¹ R, p. 2. The list of al-Ṣādiq's disciples (R, pp. 142-342), though impressive, falls 777 names short of 4,000 (a fact which al-Nūrī al-Ṭabarsī, op. cit., III, 772, does his best to explain away).

¹² Ibn Shahrāshūb (loc. cit.), for one, counts among the 4,000 non-Shīʿites such as Abū Ḥanīfa and Mālik.

¹³ Al-Tanukābunī (Qiṣaṣ al-ʿulamā', n. pl., 1304/1886-7, p. 156) speaks of 4,000 (or even 6,000) uṣūl.

¹⁴ F, p. 1.

¹⁵ Al-Khwānsārī, Rawḍāt al-jannāt, ed. A. Ismāʿīliyān, Qumm, 1390/1970 - 1392/1972, I, 47-59; Āghā Buzurg al-Ṭihrānī, al-Nābis fi l-qarn al-khāmis, Beirut, 1391/1971, p. 15; Muḥsin al-Amīn, Aʿyān al-shīʿa, VIII, Beirut, 1380/1960, 201-5.

¹⁶ F, pp. 1f. Some late scholars, however, had no hesitation in using the term muṣannaf when referring to an aṣl. See above, n. 6; Muṣṭafā al-Iʿtimādi al-Tabrīzī, Sharḥ al-rasā'il, I/iii, Qumm, ca. 1390/1970, 85.

¹⁷ Or a member (baʿḍ).

¹⁸ F, p. 2. See also Dharīʿa, X, 87f, no. 163. Al-Ṭūsī may have been mistaken in his belief that Ibn al-Ghaḍā'irī's work had been lost (Muḥsin al-Amīn, op. cit., VIII, 204).

¹⁹ This can be confirmed by checking the references to uṣūl authors scattered throughout the Fihrist. Al-Ṭūsī's main authorities on uṣūl are his teacher al-Shaykh al-Mufid, Abū l-Ḥusayn ʿAlī b. Aḥmad Ibn Abī Jīd al-Qummī (cf. Q, IV, 164; VII, 203), Ibn al-Ghaḍā'irī's father al-Ḥusayn b. ʿUbayd Allāh (d. mid-Ṣafar 411/June 1020), Aḥmad b. ʿAbdūn (d. 423/1031-2) (cf. Q, I, 124), and al-Ḥusayn b. Ibrāhīm al-Qazwīnī. Al-Ṭūsī frequently relies on ʿidda/jamāʿa min aṣḥābinā, understood as referring usually to the first four authorities mentioned (see Muḥ. Ṣādiq Āl Baḥr al-ʿUlūm's introduction to al-Ṭūsī's Kitāb al-rijāl, pp. 66f).

²⁰ F, pp. 3f, cit. Dharīʿa, II, 128.

²¹ This distinction, however, is not always clear-cut, and the two terms are on occasion used synonymously. Thus the mubawwab of Ghiyāth b. Ibrāhīm (see next note) is also known as al-Jāmiʿ (or al-Jāmiʿa). See W. Madelung, "The sources of Ismāʿīlī law", JNES, 35, 1976, pp. 35f; cf. also al-Bihbihānī, al-Ijtihād wa l-akhbār, Tehran, 1313/1895-6, p. 14.

²² Ibrāhīm b. Muḥammad b. Abī Yaḥyā (F, p. 16), Ghiyāth b. Ibrāhim al-Tamīmī (N, p. 215), Muḥammad b. ʿAlī b. Abī Shuʿba al-Ḥalabī (N, p. 228), Masʿada b. Ziyād (N, p. 295), Yaʿqūb b. Sālim (Q, VI, 274, quoting N; but he is missing from the Bombay edition. See already al-Māmaqānī, Tanqīḥ al-maqāl, Najaf, 1349/1930-1352/1933, biography no. 13276.) For the first two see Madelung, art. cit., pp. 34-6.

²³ Wuhayb b. Ḥafṣ al-Jurayrī (N, p. 304).

²⁴ Ismāʿīl b. Abī Khālid Muḥammad b. Muhājir al-Azdī (N, p. 18; F, p. 55).

²⁵ Rifāʿa b. Mūsā (N, p. 119).

²⁶ Saʿd b. Saʿd b. al-Aḥwaṣ al-Ashʿarī al-Qummī (N, p. 127). Ibn Bābawayhi is said to have composed two works called Kitāb al-ʿilal, one (also known as ʿIlal al-sharā'iʿ) mubawwab, and the other not (N, pp. 276, 278). Since only the former has come down to us, it is impossible to examine the relationship between the two.

²⁷ For example, the Kitāb fi l-ḥalāl wa l-ḥarām of ʿAlī b. Jaʿfar al-Ṣādiq (N, p. 176).

²⁸ N, p. 58; F, p. 26; al-Astarābādī, Manhaj al-maqāl, Tehran, 1306/1888-9, p. 34.

²⁹ N, p. 114; F, p. 133.

³⁰ Ibid. See also N, p. 60; R, p. 472; F, p. 47.

³¹ N, p. 13; F, p. 17.

³² N, p. 280.

³³ N, pp. 276-9.

³⁴ ISh, p. 63 (lahu aṣl wa riwāyāt wa jāmiʿ). His Jāmiʿ is quoted in the fragment of the Kitāb al-īḍāḥ of al-Qāḍī al-Nuʿmān. See Madelung, art. cit., p. 34.

³⁵ N, p. 175. Some uṣūl authors also wrote mubawwab works, e.g. the above-mentioned Masʿada b. Ziyād and Rifāʿa b. Mūsā.

³⁶ N. pp. 95 f.

³⁷ F, p. 118. He did not, however, compose an aṣl, pace ISh, p. 43.

³⁸ R, p. 440.

³⁹ Ibid.

⁴⁰ R, p. 480.

⁴¹ R, p. 440.

⁴² R, p. 499.

⁴³ Ibid.

⁴⁴ R, p. 500.

⁴⁵ R, p. 499.

⁴⁶ R, p. 490.

⁴⁷ R, p. 480

⁴⁸ R, p. 441.

⁴⁹ R, p. 440,

⁵⁰ R, p. 480. His death date is given as 207/822-3, and the information is reproduced in Q (IV, 151). But this is clearly a mistake, since Ḥumayd could not possibly have met a person who died so early.

⁵¹ R, p. 517.

⁵² R, p. 442.

⁵³ N, p. 271; F, p. 284.

⁵⁴ Ibn Bābawayhi, Man lā yaḥḍuruhu l-faqīh, ed. ʿAlī al-Ākhundī, Najaf, 1377/1957 - 1378/1959, I, 4f.

⁵⁵ GAS, I, 544.

⁵⁶ F, p. 77.

⁵⁷ R, p. 463.

⁵⁸ R, p. 516. The same is also reported of Ḥaydar, the son of Muḥammad b. Nuʿaym al-Samarqandī (R, p. 463).

⁵⁹ This work, known as Risālat Abī Ghālib al-Zurārī ilā ḥafīdihi, is preserved in full in Yūsuf al-Baḥrānī, al-Kashkūl, Najaf, 1381/1961, I, 180ff. It was also published independently by Muh. Ḥasan Āl Yāsīn in Nafā'is al-makhṭūṭāt, II, Baghdad, 1373/1954, 53-73 (see GAS, I, 544). But the text given there, though purporting to be complete, includes in fact only the first half, and so does not contain the list appearing in the latter half of the Risāla.

⁶⁰ F, p. 2.

⁶¹ This is the date proposed by Y. Eche (Les bibliothèques arabes, Damascus, 1967, pp. 116f), who rejects Yāqūt's dating of the event to 447/1055. See in general Yāqūt, Muʿjam al- buldān, Beirut, 1374/1955 - 1376/1957, I, 534, whence Dharīʿa, II, 129, 134, VIII, 173f; Āl Baḥr al-ʿUlūm's introduction to al-Ṭūsī's Kitāb al-rijāl, pp. 13-7, 52 and the sources cited there.

⁶² Ibn Idrīs, al-Sarā'ir, Tehran, 1390/1970, pp. 471-94. These works include the Kitāb Abān b. Taghlib (p. 475) and the Kitāb Jamīl b. Darrāj (p. 476), both of which may well be uṣūl (see below), and also the Kitāb Ḥurayz b. ʿAbd Allāh al-Sijistānī (pp. 479-80), which Ibn Idrīs explicitly identifies as an aṣl (p. 480). Since the traditions cited from this work deal with prayer, it is probably identical with al-Sijistānī's Kitāb al-ṣalāt. Ibn Idrīs was commended by later authors for preserving these excerpts from early Imāmī works (al-Khwānsārī, op. cit., VI, 274). In addition, he is known to have copied the aṣl of ʿAlā' b. Razīn (Dharīʿa, II, 164, no. 604; see below, n. 166).

⁶³ Ibn Ṭāwūs, Fatḥ al-abwāb bayna dhawī l-albāb wa rabb al-arbāb (cf. Dharī'a, XVI, 103, no. 127), as quoted in Biḥār, XVIII/ii, 932. See also id., Muhaj al-daʿwāt, [Tehran] 1323/1905, p. 313; id., Faraj al-mahmūm, Najaf, 1368/1948-9, p. 99. Ibn Ṭāwūs was not the first to employ such a formula; al-Ṭūsī (Kitāb al-ghayba, Najaf, 1385/1965, p. 187) quotes Aḥmad b. ʿAlī b. ʿAbbās b. Nūḥ (see above, n. 10) as declaring: "I have found in an ancient aṣl which was copied in Ahwāz in Muḥarram 317 [ = Feb. - March 929]". Cf. al-Mufīd's statement (in al-Ṭūsī, Amālī, Najaf, 1384/1964-5, II, p. 27): 'I have read in one of the uṣūl a tradition whose isnād escapes me now.'

⁶⁴ Ibn Ṭāwūs, Falāh al-sā'il, Najaf, 1385/1965, p. 90; id., Kashf al-maḥajja, Najaf, 1370/1950, p. 124. Cf. ibid., p. 18, cit. al-Ḥurr al-ʿĀmilī, Wasā'il al-shīʿa, Tehran, 1378/1958-9-1389/1969-70, VI/i, 458, Biḥār, I, 107 (in the lithographed edition, erroneously, 106).

⁶⁵ Ibn Ṭāwūs, Falāḥ al-sā'il, pp. 202, 204.

⁶⁶ Id., Risālat al-muwāsaʿa fi l-qaḍā' (a work also known by other names), cit. Fawā'id, pp. 30f (where this work is not identified), Biḥār, XVIII/ii, 677.

⁶⁷ E.g., Kitāb Ibrāhīm al-Khazzāz (Falāḥ al-sā'il, p.89), or the Kitāb of Zakariyyā al-Mu'min (ibid., p. 246).

⁶⁸ Ibn Ṭāwūs, Fatḥ al-abwāb, cit. Biḥār, XVIII/ii, 932; see also Fawā'id, pp. 31f. Mīr Dāmād (op. cit., p. 98) emphasizes that works of Ḥasan b. Maḥbūb are not uṣūl. Ibn Maḥbūb did, however, transmit various uṣūl, including that of Wahb b. ʿAbd Rabbihi (F, p. 349; cf. N, p. 303; both cit. Q, VI, 196f. F², p. 201, however, has lahu kitāb). A tradition quoted by Ibn Ṭāwūs (Falāḥ al-sā'il, p. 126) which has in its isnād Ibn Maḥbūb transmitting from Wahb may well derive from Wahb's aṣl.

⁶⁹ Dharīʿa, VIII, 172-9.

⁷⁰ E.g. by al-Shahīd al-Awwal (d. 786/1384) (al-Nūrī al-Ṭabarsī, op. cit., III, 298f).

⁷¹ Thus, after quoting a long story on Salmān al-Fārisī and ʿUmar b. al-Khaṭṭāb, al-Majlisī remarks; 'We have cited this story, even though it is strange and not mentioned in reputable books, because we found it in an ancient aṣl' (Biḥār, VIII, 224). In this particular case, al-Majlisī further describes the work in question as a book on manāqib on the authority of al-Mufaḍḍal (ibid., 222). The story appears, in fact, with some variations, in the Kitāb al-haft wa l-aẓilla ascribed to al-Mufaḍḍal b. ʿUmar ai-Juʿfī (pp. 86-93 in the ʿA. Tāmir and Ign. - A. Khalifé edition, Beirut, 1960; on this work see H. Halm, "Das 'Buch der Schatten'", Der Islam, 55, 1978, pp. 219ff). None of al-Mufaḍḍal's works is explicitly referred to as an aṣl in the sources. A kitāb of his is, however, mentioned by al-Ṭūsī (F, p. 337; see Halm, art. cit., p. 222). The conclusion must therefore be either that al-Majlisī knew (from sources unavailable to us) that al-Mufaḍḍal's kitāb (which may or may not be identical with the Kitāb al-haft wa l-aẓilla) was in fact an aṣl, or else that al-Majlisī, in this case at least, was using the term aṣl to refer generally to an ancient work from the time of the Imams. — Cf. further Biḥār, XVIII/ii, 722, 787.

⁷² Biḥār, I, 10, 16f. See below, section IV.

⁷³ Wasā'il, I/i, 1.

⁷⁴ Ibid., IX/iii, 392.

⁷⁵ The list of sources of the first kind is given ibid., I/i, 4-6, and again in IX/iii, 36-47; of the second kind, ibid., I/i, 7f, and again in IX/iii, 47-9.

⁷⁶ Several other uṣūl are, however, quoted indirectly. The lists of direct and indirect sources in al-Ḥurr al-ʿĀmilī's Kitāb al-īqāẓ min al-hajʿa (completed in 20 Rabīʿ 1079/28 Aug. 1668) contain no uṣūl (see pp. 27-9).

⁷⁷ Ithbāt al-hudāt, Qumm, 1378/1958-9 - 1379/1959-60, I, 60.

⁷⁸ Al-Khwānsārī, op. cit., II, 84, VII, 103, whence Muḥsin al-Amīn, op. cit., XLIV, Beirut, 1378/1959, 54.

⁷⁹ See below, section IV.

⁸⁰ Qumm, 1378/1958 - 1380/1960 (4 vols.). See also al-Nūrī al-Ṭabarsī's comment on the significance of the uṣūl in his Nafas at-raḥmān fī faḍā'il Salmān, Tehran, 1285/1868-9, p. 29.

⁸¹ Fawā'id, p. 30. See in general my articles "Aḳbārīya' (Encyclopaedia Iranica, I, 716-8), 'Astarābādī' (ibid., II, 845f).

⁸² Fawā'id, pp. 52f.

⁸³ Ibid., pp. 3, 30.

⁸⁴ Ibid., p. 57. In fact, some authorities distinguish two kinds of uṣūl: those dictated by an Imam to his disciple; and those which include traditions heard by the disciple on various occasions and then presented (ʿarḍ) to the Imam (or to one of his successors) for approval (Dharīʿa, XXIV, 147-51, 318). The aṣl of ʿUbayd Allāh al-Ḥalabī is of the second kind (Fawā'id, p, 31; Biḥār, XVIII/ii, 677; Yūsuf al-Baḥrānī, al-Ḥadā'iq al-nāḍira, I, Najaf, 1377/1957, 9). Other works, not of the aṣl genre, were likewise presented to the Imams for approval. For instance, the Kitāb yawm wa layla of Yūnus b. ʿAbd al-Raḥmān and the Kitāb of al-Faḍl b. Shādhān were both approved by al-Ḥasan al-ʿAskarī (N, p. 312, cit Biḥār, I, 110; Bahā' al-Dīn al-ʿĀmilī, Kitāb mashriq al-shamsayn, cit. Fawā'id, p. 54; Muḥsin al-Fayḍ, al-Wāfī, Tehran, 1375/1955-6, 1/i, 11; Yūsuf al-Baḥrāni, loc. cit.)

⁸⁵ Fawā'id, p. 53.

⁸⁶ Ibid., pp. 52f, 58f, 65.

⁸⁷ Cf. Āl Baḥr al-ʿUlūm's introduction to al-Ṭūsī's Kitāb al-rijāl, p. 77.

⁸⁸ F, p. 3, cit. Fawā'id, p. 52.

⁸⁹ Al-Ṭūsī, ʿUddat al-uṣūl, Tehran, 1314/1896-7, pp. 57, 61f. In a typical comment, al-Ṭūsī says of the aṣl of Isḥāq b. ʿAmmār ai-Sābāṭī, 'he was a faṭḥī but is nevertheless trustworthy (thiqa) and his aṣl is reliable (muʿtamad ʿalayhi)' (F, p. 54). — According to ʿAbd Allāh al-Māmaqānī (d. Shawwāl 1351/Feb. 1933), transmission from a wāqifī is permitted when it is known that he composed an aṣl before adopting a position of waqf (Tanqīḥ al-maqāl, I, 171).

⁹⁰ Al-Ṭūsī, ʿUddat al-uṣūl, p. 51, cit. Fawā'id, p. 70. Al-Ṭūsī's position is echoed by later authors: for Muḥsin al-Fayḍ (d. 1091/1680), a tradition is ṣaḥīḥ when found in several uṣūl, or in a single aṣl only, but one which is known to be the work of a trustworthy Imāmī (al-Wāfī, I/i, 11). Aḥmad b. Muḥammad al-Narāqī (d. 1245/1829-30) says that an isolated tradition is acceptable if taken from an aṣl known with certainty to have been compiled by a trustworthy disciple of an Imam (Manāhij al-aḥkām, chapter on al-Taʿabbud bi l-khabar al-wāḥid).

⁹¹ Persons who allegedly introduced false traditions into the uṣūl include Abū l-Khaṭṭāb, al-Mughīra b. Saʿīd al-Anṣārī (Farā'id al-uṣūl, Qumm, 1374/1954-5, p. 103) and Ibn Abī l-ʿAwjā' (ibid., p. 93).

⁹² Fawā'id, pp. 53-5, 57f, quoting from al-Shahīd al-Thānī and from Bahā' al-Dīn al-ʿĀmilī's Kitāb mashriq al-shamsayn. The possible existence of forged uṣūl is also emphasized by Mīr Dāmād (op. cit., pp. 107ff). Al-Astarābādī (Fawā'id, pp. 57f) rejects the claim that some uṣūl were lost or became indistinguishable from other works.

⁹³ Al-Anṣārī, op. cit., pp. 70f, 102f; Dharīʿa, II, 306.

⁹⁴ Al-Māmaqānī, op. cit., I, 181.

⁹⁵ Al-Bihbihānī, op. cit., pp. 67f.

⁹⁶ Ibid., pp. 77, 78.

⁹⁷ Ibid., p. 52; al-Māmaqānī, op. cit., I, 179. See in general Q, IV, 153-9.

⁹⁸ Al-Māmaqānī, loc. cit.

⁹⁹ Ibid. Wahb b. Wahb is not generally regarded as the author of an aṣl. See Q, VI, 197-9.

¹⁰⁰ Al-Bihbihānī, op. cit., pp. 56f, 59.

¹⁰¹ Ibid., p. 60 (with examples of non-uṣūl material in Ibn Bābawayhi's book). The passage referred to by al-Bihbihānī appears in Man lā yaḥḍuruhu l-faqīh, I, 5. Al-Ṭūsī, too, emphasizes that he has included in his Tahdhīb al-aḥkām material from both uṣūl and muṣannafāt. See the section on the mashyakha, X, Najaf, 1382/1962, 4f, 88.

¹⁰² Al-Bihbihānī, op. cit., p. 51.

¹⁰³ Al-Anṣārī, op. cit., p. 70.

¹⁰⁴ Al-Narāqī, Kitāb ʿawā'id al-ayyām, Tehran, 1321/1903, p. 210.

¹⁰⁵ The text reads: kitāb min al-uṣūl fī l-riwāya ʿalā madhāhib al-shīʿa (IN, p. 276). The term uṣūl is perhaps used here in the sense of 'principles' (of religion), in which case this work does not belong in the list.

¹⁰⁶ So according to F², p. 63. Sprenger's edition (p. 52) has lahu kitāb, and al-Ṭūsī's text as reproduced in Q (I, 183) has lahu kitāb aṣl.

¹⁰⁷ F², p. 133 has lahu kitāb.

¹⁰⁸ Only according to the version cited in the Dhariʿa (II, 136, no. 507). My edition of ISh (p. 6, no. 21) only credits him with a kitāb.

¹⁰⁹ According to Q (I, 81), he is identical with Ibrāhīm b. Abī l-Bilād; but al-Ṭihrānī (Dharīʿa, II, 137f, no. 514) adduces evidence that the names refer to two different persons.

¹¹⁰ The reading lahu aṣl is attested in one (or several, baʿḍ) manuscripts of the Fihrist (apud Dharīʿa, II, 141, no. 525), and is supported by ISh. However, F, pp. 54f, F², p. 37 (in both of which two persons of this name are mentioned) and Q's reproduction (I, 203) all have lahu kitāb.

¹¹¹ F², p. 201 has lahu kitāb (see above, n. 68).

¹¹² On which see M.J. McDermott, The Theology of al-Shaikh al-Mufid (d. 413/1022), Beirut, 1978, p. 31 (bottom), Al-Mufid mentions having seen this aṣl.

^112aLegal traditions from this aṣl were incorporated in the Amālī of the Zaydī Aḥmad b. ʿĪsā b. Zayd (d. 247/861). See Madelung, 'Abū'l-Jārūd Zīād b. al-Mon/ḍer', Encyclopaedia Iranica, I, 327.

¹¹³ II, 135-67.

^ll3aAddendum: The following additional names are referred to by Ibn Ṭāwūs as uṣūl authors (though some of these apparently transmitted uṣūl rather than compiling them): ʿAbd Allāh b. Qāsim al-Ḥaḍramī (Faraj al-mahmūm, p. 93), Ḥammād b. ʿUthmān Dhū l-Nāb (al-Malāḥim wa l-fitan, p. 139), Muʿāwiya b. Ḥakīm (Faraj al-mahmūm, p. 91), Muḥammad b. Abī ʿUmayr (ibid., p. 93; see below, n. 139), Yūnus b. Bukayr (Muhaj al-da ʿwāt, p. 253), and the author of the Kitāb al-tajammul (Faraj al-mahmūm, pp. 24, 99, 100, 124), presumably Aḥmad b. Muḥammad b. al-Ḥusayn al-Qummī (cf. Dharīʿa, III, 359, no. 1293). None of these authors appears in the Dharīʿa list.

¹¹⁴ F, p. 27, whence Q, I, 107, Dharīʿa, II, 138, no. 517. The version min jumlat al-uṣūl (F², Q) (or min al-uṣūl) (Dharīʿa) is to be preferred to min jillat al-uṣūl (F). Al-Qurashī did not transmit directly from any Imam (R, p. 453).

¹¹⁵ Dharīʿa, II, 144, no. 538.

¹¹⁶ IN, p. 279.

117 F, p. 48, s.v. Aḥmad b. Muḥammad b. Nūḥ.

¹¹⁸ Dharīʿa, II, 134.

¹¹⁹ B. Dodge's rendition of this phrase (The Fihrist of al-Nadīm, New York and London, 1970, p. 543) appears to be mistaken.

¹²⁰ IN, loc. cit.

¹²¹ Dharīʿa, II, 137, no. 511.

¹²² N, p. 18.

¹²³ Cf. R, p. 446: ṣāḥib kitāb al-imāma min taṣnīfʿAlī b. Muḥammad al-Jaʿfarī; R, p. 483: ṣāḥib kutub al-Faḍl b. Shādhān.

¹²⁴ N, Dharīʿa, loc. cit.

¹²⁵ F, p. 85; ISh, p. 44.

¹²⁶ But I have found no evidence to corroborate Falaturi's statement (art. cit., p. 64, n. 3) that this was often the case.

¹²⁷ R, p. 440.

¹²⁸ F, p. 49.

¹²⁹ Fihrist-e kitābkhāne-ye ihdā'ī-ye... Mishkāt, III/iii, Tehran, 1335 SH/1955, 1088-95.

¹³⁰ Cf. above, section I.

¹³¹ Not 9b, as printed in Mishkāt, III/iii, 1089, whence GAS, I, 532.

¹³² The numbering here and in what follows is mine.

¹³³ The author of this aṣl is better known as Abū Saʿīd ʿAbbād b. Yaʿqūb al-Asadī al-Rawājinī (d. 250/864). Various Imāmī authorities are at pains to emphasize that the two names refer to the same person. They include Ibn Dāwūd al-Ḥillī (fl. second half of the 7th/13th century) in his Kitāb al-rijāl (Tehran, 1383/1963, p. 194, no. 795), Muḥammad b. Muḥsin al-Fayḍ in his Naḍd al-īḍāḥ (printed together with F, pp. 176f), and al-Nūrī al-Ṭabarsī in his Mustadrak (III, 299: wa ammā kitāb Abī Saʿīd ʿAbbād al-ʿUsfurī wa huwa bi ʿaynihi ʿAbbād b. Yaʿqūb al-Rawājinī...); likewise Sezgin (GAS, I, 316f, 537). Al-Māmaqāni, in contrast, reserves judgment (op. cit., II, 124). His doubts stem from the fact that al-Ṭūsī regarded the names al-ʿUṣfurī and al-Rawājinī as referring to two distinct persons, and therefore devoted a separate entry to each name (F, p. 176, nos. 372, 374). What led (or rather: misled) al-Ṭūsī to adopt this position? The key to the answer is found in a succinct remark of al-Ḥusayn b. ʿUbayd Allāh al-Ghaḍā'irī (see above, n. 19), who quotes unnamed Imāmī scholars for the view that Abū Sumayna concealed ʿAbbād's identity (N, p. 208). The verb used to describe Abū Sumayna's action is dallasa (ibid.), referring in this case to the method known as tadlīs al-shuyūkh (for which see e.g. al-Khaṭīb al-Baghdādī, al-Kifāya fī ʿilm al-riwāya, Hyderabad, 1357/1938, pp. 365-71; Mīr Dāmād, op. cit., pp. 186, 191; EI², III, s.v. 'Ḥadīth' [J. Robson], at p. 26; in general G. H. A. Juynboll, Muslim Tradition, C. U. P., 1983, index, s.v. tadlīs). It was apparently applied by Abū Sumayna by omitting the name of ʿAbbād's father (Yaʿqūb) and substituting 'al-ʿUṣfurī' for 'al-Rawājinī', the nisba by which ʿAbbād was generally known. Since Abū Sumayna transmitted ʿAbbād's aṣl, this work is invariably referred to as the aṣl (or kitāb) of ʿAbbād al-ʿUṣfurī. Abū Sumayna (i.e. Muḥammad b. ʿAlī b. Ibrāhīm al-Ṣayrafī) is described in Imāmī sources as an inveterate liar (K, p. 457, G, V, 264, whence al-Māmaqānī, op. cit., III, 151); and al-Ṭūsī knows of some works of his which included tadlīs (F, p. 303). In speculating as to why Abū Sumayna might have wished to conceal ʿAbbād's identity it is worth noting that ʿAbbād did not properly belong within either the Imāmī or the Sunnī camp. In Sunnī circles (where he is known as al-Rawājinī, not al-ʿUṣfurī) he was condemned as an extremist Rāfiḍī, but was none the less regarded as a trustworthy transmitter, and appears in fact as one of al-Bukhārī's authorities (al-Bukhārī, al-Ta'rīkh al-kabīr, Hyderabad, 1360/1941 - 1364/1945, VI, 44; al-Khaṭīb al-Baghdādī, op. cit., pp. 131f; al-Dhahabī, Mīzān al-iʿtidāl, Cairo, 1325/1907, II, 16f; Ibn al-Athīr, al-Lubāb fī tahdhīb al-ansāb, Beirut, n.d., II, 39, s.v. al-Rawājinī; Ibn Ḥajar al-ʿAsqalānī, Tahdhīb al-tahdhīb, Hyderabad, 1325/1907 - 1327/1909, V, 109f; further references in al-Ṣafadī, al-Wāfī bi l-wafayāt, XVI, ed. Wadād al-Qāḍī, Wiesbaden, 1982, pp. 614f; cf. F. Rosenthal, Knowledge Triumphant, Leiden, 1970, p. 146). Al-Ṭūsī, on the other hand (F, p. 176), followed by Ibn Ṭāwūs (al-Yaqīn fī imrat amīr al-mu'minīn, Najaf, 1369/1950, p. 74), Ibn al- Muṭahhar al-Ḥillī (Khulāṣat al-aqwāl, p. 117) and Ibn Dāwūd (op. cit., p. 465, no. 248), refers to al-Rawājinī as ʿāmmī al-madhhab. Al-Māmaqānī (op. cit., II, 123f) interprets this term as meaning 'Sunnī', and goes on to argue that in using it al-Ṭūsī was merely trying to conceal al-Rawājinī's true beliefs, since he knew that the Sunnīs transmitted from him (and obviously wanted them to go on doing so). Al-Ṭūsī, then, was practising taqiyya. It is just as likely, however, that in referring to ʿAbbād as ʿāmmī al-madhhab al-Ṭūsī meant that he was a Shīʿī, but of a different persuasion than the Imāmīs. This may well be an accurate description of ʿAbbād's true position (see Madelung, 'The Sources of Ismāʿīlī Law', p. 37); judging by his views on barā'a (al-Dhahabī, op. cit., II, 16), he may have been a Jarūdī Zaydī; the possibility of a Zaydī persuasion is in fact mentioned by Ibn Ṭāwūs (op. cit., p. 175). Such credentials could not have endeared him to Abū Sumayna, who was notorious for his extreme Shīʿī views (Halm, art. cit., pp. 241f; my article 'Barā'a in Shīʿī Doctrine', JSAI, 7, 1986, p. 165); so while Abū Sumayna did not refrain from transmitting from ʿAbbād, he did his best to disguise ʿAbbād's identity.

¹³⁴ An excerpt from this tradition (with a different isnād) is included by al-Dhahabī (op. cit., II, 17, whence Sharaf al-Dīn al-Mūsawī, al-Murājaʿāt, Najaf, 1384/1964, p. 108) and al-ʿAsqalāni (op. cit., V, 110) in al-Rawājinī's biography. This is another indication that al-Rawājinī = al-ʿUṣfurī.

¹³⁵ See above, p. 134.

¹³⁶ F, pp. 122, 147f. On Muḥammad b. Mūsā see N, p. 239; Q, VI, 59.

¹³⁷ F, pp. 147f; Ibn Dāwūd al-Ḥillī, op. cit., pp. 164, 455; Mīr Dāmād, op. cit., p. 97. Ibn Bābawayhi's Fihrist was known to al-Ṭūsī (F, p. 304), but is not mentioned by al-Najāshī. See also Dharīʿa, XVI, 374, no. 1738.

¹³⁸ As quoted in Q, III, 84.

¹³⁹ K, pp. 492-4; N, pp. 228-30; F, pp. 265f. Ibn Abī ʿUmayr is said to have transmitted the books (kutub) of 100 disciples of Jaʿfar al-Ṣādiq (F, p. 266); many of these books were uṣūl. He did not, however, himself compose an aṣl (pace Ibn Ṭāwūs, Faraj al-mahmūm, p. 87, cit. Muḥsin ai-Fayḍ, op. cit., III/xiv, 131). Ibn Abī ʿUmayr spent several years in prison by order of Hārūn al-Rashīd (or al-Ma'mūn), during which time his books were destroyed. Upon his release, he went on transmitting by memory traditions which had been included in them. But since he could not always recall the exact isnāds of each tradition, he transmitted some of them with incomplete isnāds (K, p. 493; N, p. 229). It is a measure of Ibn Abī ʿUmayr's standing that these marāsil are nevertheless regarded by Imāmī ḥadīth specialists as sound and reliable (N, loc. cit.; al-Ṭūsī, ʿUddat al-uṣūl, p. 63; Mīr Dāmād, op. cit., pp. 67 [marāsīl Ibn AbīʿUmayr tuʿaddu fī ḥukm al-masānīd], 175f; Biḥār, XVIII/ii, 784).

¹⁴⁰ Biḥār, I, 17.

¹⁴¹ In his Rijāl, cit. al-Nūrī al-Ṭabarsī, op. cit.. III, 300, al-Māmaqānī, op. cit., I, 464f, Muḥsin al-Amīn, op. cit., XXXII, Damascus, 1368/1949, 404-8.

¹⁴² In his commentary (taʿlīq, ḥawāshī) on the Muntahā l-maqāl of Abū ʿAlī Muḥammad b. Ismāʿīl al-Ḥā'irī (d. Rabīʿ I / July-Aug. 1215/1800 or 1216/1801), cit. al-Mamaqānī, op. cit., I, 465, Muḥsin al-Amīn, op. cit., XXXII, 403f.

¹⁴³ Mustadrak, III, 297-9, 300-3.

¹⁴⁴ Aʿyān al-shīʿa, XXXII, 401-8.

¹⁴⁵ Al-Nūrī al-Ṭabarsī, op. cit., III, 301; al-Māmaqānī, op. cit., I, 464; Muḥsin al-Amīn, op. cit., XXXII, 405.

¹⁴⁶ See above, p. 141.

¹⁴⁷ This is also the verdict of al-Nūrī al-Ṭabarsī (op. cit., III, 299).

¹⁴⁸ Al-Tustarī, Qāmus al-rijāl, Qumm, 1387/1967-1391/1971, IV, 248-51.

¹⁴⁹ Such as al-Ṭihrānī (Dharīʿa, II, 164) and al-Tustarī (op. cit., VI, 423).

¹⁵⁰ Although the fact that it begins on the same line in which the Nawādir ends misled the compiler of the Mishkāt Catalogue (III/iii, 1088-9, whence GAS, I, 537) into treating it as part of ʿAlī b. Asbāṭ's work.

¹⁵¹ R, p. 481, whence Q, IV, 182; al-Māmaqānī, op. cit., II, 280.

¹⁵² N, p. 257 (ʿummira). He was supposedly 128 years old when telling Ibn al-Ṭabbāl his story (MS, fol. 73b).

¹⁵³ Nos. 2, 7-12 of the majmūʿa. Al-Ṭihrānī refers to each of them as a Mukhtaṣar, but does not state his reasons for doing so.

¹⁵⁴ M. T. Daneche-Pajouh, Catalogue... des manuscrits de la Bibliothèque de la Faculté de droit et des sciencespolitiques et économiques de l'Université de Téhéran, Tehran, 1380/1961, pp. 246-9.

¹⁵⁵ Ibid., pp. 247, 249.

¹⁵⁶ Not compiled, pace GAS, I, 537.

¹⁵⁷ R, p. 306.

¹⁵⁸ N, p. 117.

¹⁵⁹ F, p. 134.

¹⁶⁰ P. 247. In fact, al-Nūrī al-Ṭabarsī states explicitly that the beginning of the Kitāb Durust at his disposal is missing (op. cit., III, 296). He must thus have used a manuscript of the same family as the one dealt with here.

¹⁶¹ Risālat Abī Ghālib al-Zurārī (above, n. 59), p. 67; Yūsuf al-Baḥrānī, al-Kashkūl, I, 188; al-Māmaqānī, op. cit., III, 93.

¹⁶² Dharīʿa, II, 165, no. 609.

¹⁶³ Al-Ṭihrānī does not, however, mention the Kitāb Durust.

¹⁶⁴ See the discussion in Dharī'a, II, 159-62, no. 595. The Kitāb al-diyāt is preserved in its entirety in Ibn Bābawayhi's Man lā yaḥḍuruhu l-faqīh (IV, 54-66) and elsewhere (Dharīʿa, II, 160, whence GAS, I, 537).

¹⁶⁵ RIMA, 4, 1958, p. 214.

¹⁶⁶ Dharīʿa, II, 164, no. 604, mentions the existence of this mukhtaṣar in a copy based on a manuscript written by Ibn Idrīs al-Ḥillī.

¹⁶⁷ GAS, I, 539.

¹⁶⁸ See above, p. 137.

¹⁶⁹ See above, p. 138.

¹⁷⁰ Ithbāt al-hudāt, III, 142f.

¹⁷¹ Ibid., V, 493f, VII, 176f.

¹⁷² Ibid., III, 679f, VII, 177f.

¹⁷³ Ibid., III, 680f, V, 93.

¹⁷⁴ Ibid., III, 681.

¹⁷⁵ Ibid., V, 465f.

¹⁷⁶ Ibid., III, 681f, V, 212. The khabar fī l-malāḥim is also cited (ibid., V, 467f). Quotations from the uṣūl invariably appear at the end of a given chapter of the Ithbāt, but I do not know what (if any) significance should be attached to this.

¹⁷⁷ Kashf al-ḥujub wa l-astār, Calcutta, 1330/1911-2, p. 431, no. 2396, p. 433, no. 2406, pp. 433f, no. 2410, p. 444, nos. 2501, 2502, p. 445, no. 2506, pp. 447f, no. 2520, pp. 448f, no. 2522, p. 449, no. 2523, pp. 449f, no. 2524, pp. 458f, no. 2585, p. 459, no. 2588, p. 463, no. 2616.

¹⁷⁸ Ibid., p. 549, no. 3090, whence Dharīʿa, XXII, 188, no. 6630.

¹⁷⁹ Ibid., p. 444.

¹⁸⁰ See above, p. 139.

¹⁸¹ Al-Nūrī al-Ṭabarsī, op. cit., III, 291, 296-308.

¹⁸² On which see B. Scarcia Amoretti, art. cit. (above, n. 2).

¹⁸³ Qāmūs al-rijāl, II, 416, IV, 21, 249, 408, V, 183, VI, 176, 185, 423, VIII, 358.

¹⁸⁴ Ibid., IV, 249.

¹⁸⁵ Muḥsin ai-Amīn, op. cit., L, Beirut, 1381/1961, 24. ʿAbd Allāh Efendī (d. after 1130/1718) mentions in his Riyāḍ al-ʿulamā' (Qumm, 1401, V, p. 256) having seen and utilized al-Jazā'irī's rich book collection in Tustar. Cf. al-Jazā'irī's Manbaʿ al-ḥayāt, Princeton MS N.S. no. 960, fols. 77a-78a, 85a-b.

¹⁸⁶ XXIV, 335.

¹⁸⁷ The 16 purported uṣūl might well be identical to the works mentioned by al-Nūrī al-Ṭabarsī (above, n. 181); as shown above, two of these are probably not uṣūl.

¹⁸⁸ My thanks are due to Professors Michael A. Cook and Frank H. Stewart for their comments on this paper.