Sunteți pe pagina 1din 12

Two Unrecognized Karrm Texts Author(s): Aron Zysow Source: Journal of the American Oriental Society, Vol.

108, No. 4 (Oct. - Dec., 1988), pp. 577587 Published by: American Oriental Society Stable URL: Accessed: 04/05/2009 08:55
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact

American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Oriental Society.



In a recent monograph Josef van Ess has drawn attention to several works of apparent KarramT origin, two of which Massignon had already examined. The Karrami provenance of two additional texts, both in print, has gone unrecognized. One is the introduction to the Qur'an commentary K. al-MabdnT li-nazm al-macnl (begun in 425/1033), a major source of orientalist research on the Qur'an since the nineteenth century. The author of al-Mabdn is unidentified but was associated with the same circles as the famous KarramT theologian Muhammad b. al-Haysam (d. 419/1019). The other text of apparent KarramTorigin is al-Nutaffi'l-fatdwd, a legal treatise attributed to the HanafT Qad. al-qudat Abu'l-Hasan 'All b. al-Husayn al-SughdT (d. 461/1068). The Abu 'Abdallah whose legal opinions are reported throughout this work is none other than Muhammad b. Karram (d. 255/869). Al-Nutaf is thus a unique source for the fiqh of the Karramiyya.


RECOGNIZE the social and political significance of the Karramiyya, the study of the place of the Karramiyya in Islamic intellectual history until recently remained where Massignon left it.' However, in a recent monograph Josef van Ess, with characteristic industry, has assembled an impressive body of new information on this sect.2 Following Massignon's lead, van Ess has re-examined two works of apparent KarramT origin and brought to light at least two others.3 This paper is

See particularly L. Massignon, Essai sur les origines du lexique technique de la mystique musulmane, 2d ed. (Paris, 1954), 260-68. In recent years, the social and political role of the Karramiyya has been examined by C. E. Bosworth, a summary of whose work can be found in EI2, s.v. Karramiyya, and R. W. Bulliet, The Patricians of Nishapur: A Study in Medieval Islamic Social History (Cambridge, Mass., 1972), General Index, s.v. Karraml. 2 Ungeniitzte Texte zur Karrdmiya:eine Materialsammlung, Sitzungsberichte d. Heidelberger Akad. d. Wiss., Phil.-hist. Kl., Jg. 1980, Abh. 6 ("Texte"). 3 CUmar al-Samarqandl, K. Rawnaq al-qulub, Bibliotheque Nationale Ar. 4929 and 6674 (Texte, 30-41); the acephalous Qur'an commentary, British Museum Or. 8049 (Texte, 4155); the collection of stories about the prophets of Abu'lHasan al-Haysam b. Muhammad, Princeton, Yahuda 439 (Texte, 68-73); and the Qur'an commentary of Abu Bakr 'AtTq b. Muhammad al-Surabadi (d. 454/1101), of which two facsimile editions have been published in Tehran (Texte, 73-74). The Karrami provenance of Abu Mutic Makhul al577

in the nature of an addendum to van Ess' study and confirms the correctness of his observation that "the KarramTya have left behind more traces than we have hitherto believed." Unfortunately, van Ess is equally correct in stating that "everything remains to be done before we can read these traces with understanding."4 Of the two works that I propose to ascribe to the Karramiyya, the first has been a mainstay of orientalist research on the Qur'an since the nineteenth century. This is Berlin MS 910 (Wetzstein 103), K. al-MabanT li-nazm al-macdnT, a Qur'an commentary of 273 folios, extending to Sura 15; the loss of the first folio has so far prevented identification of the author. The introduction to this work, covering eighty-five folios, was a major source for both editions of Noldeke's Geschichte des Qordns and has continued to be used in Qur'anic research.5 It was published by Arthur

NasafT (d. 318/930), K. al-Radd 'ala al-bida', Oxford, Bodleiana Or. 421, which van Ess also examined (Texte, 55-60), appears highly doubtful. The text has since been published by Marie Bernand in Annales islamologiques 16 (1980): 39-126. 4 Texte, 81. 5 T. Noldeke, Geschichte des Qorins (G6ttingen, 1860), xxx-xxxi and passim; T. N6ldeke, Geschichte des Qorans, 2d ed., yols. 1 and 2, ed. F. Schwally, vol. 3, ed. G. Bergstrasser and O. Pretzl (Leipzig, 1909-38), 2:184, 3: Index, s.v. MabanE.Al-Mabdan is frequently cited in J. Burton, The Collection of the Quradn(Cambridge, 1977).


Journal of the American Oriental Society 108.4 (1988)

identity of the author of al-Mabadn remains although numerous clues exist. The work, the tells us, was begun in ShaCban of 425/1033.9 time that he wrote al-MabdnT, the author had already established his mastery of the Qur'anic sciences in a series of writings, of which he names no fewer than four,'1 and must have been of middle age, The elusive, author By the

Jeffery in Cairo in 1954.6 My discussion of al-Mabani will be confined to the introduction, the unpublished portion of the work having remained inaccessible to me. Despite the fact that Jeffery's edition contains a full index of proper names, the Karrami provenance of al-Mabanl appears to have gone undetected. The strongest evidence for a Karrami attribution comes on page 209 of the printed text, where the author briefly quotes al-Imam al-Hadi Abtu 'Abdallah Muhammad b. Karram (d. 255/869), the founder of the Karramiyya.7 Of other recognizable Karramis, the name that appears with greatest frequency is that of the famous theologian Muhammad b. al-Haysam, who is quoted a number of times.8 With the introduction to Ibn 'Atiyya's tafsir as Two Muqaddimas to the Qur'anic Sciences. A second edition, with corrections, was published by cAbdallah Isma1TI alSawl (Cairo, 1392/1972). This edition preserves the pagination of the first. The fullest discussion in print of al-MabdnT is that of V. Cornell, "'Ilm al-Qur'an in al-Andalus: The Tafsir Muharrar in the Works of Three Authors," Jusur 2 (1986): 65-72. Influenced perhaps by Jeffery's reference to the "local Andalusian style of Arabic" used in al-Mabain (English preface to Two Muqaddimas), Cornell (p. 66) writes that "an examination of the literary style used in the text reveals a relative sophistication of vocabulary tempered by the influence of Andalusian or Maghribi dialectical elements." It is true that the unique Berlin manuscript is written in a Maghribi script, but Ahlwardt, Verzeichnis der arabischen Handschriften der koniglichen Bibliothek zu Berlin (Berlin, 1887-99, reprinted Hildesheim, 1980-81), 1:361, estimated that this copy dated from about 700/1300, or nearly two hundred years after the work was written. Apart from the script of the manuscript, everything else about the text points to an origin in the Eastern regions of Islam. 7 In this quotation Ibn Karram enumerates the contents of the Qur'an by genre and warns against al-tafsir bi'l-ra'y. The author of al-Mabanmadmits that he is not entirely sure of Ibn Karram's intention, although he is familiar with the full context of the citation (Mabdni, 209-10). 8 Mabadn,47-48 (Ubayy's mushaf), 170-71 (qira'at), 188ff. (tafsir), 194 (tafsTr),207-8 (traditions on seven ahruf), 21718 (seven lughat), 240-41 (traditions on asbac al-Qur'in). In some cases it is not clear where the quotation from Ibn alHaysam ends. On Ibn al-Haysam, see Texte, 60-67. To the list of Ibn al-Haysam's writings compiled by van Ess (Texte, 67), add a work on faith (Tman) as consisting of the utterance of the tongue (qawl al-lisdn) alone, seen by Ibn Taymiyya (d.

in Majmu'at 728/1328) (al-Furqdnbayn al-haqqwa'l-bdtil,

al-rasdail al-kubrd [Cairo, 1385/1966], 1:43) and a compilation on the praises (mandqib) of Ibn Karram, which Ibn alSalah (d. 643/1243) saw in Nishapur and mentioned in his Fawa'id al-rihla (Zayn al-din al-'Iraqi, al-TaqyTdwa'l-diidh sharh muqaddimat Ibn al-Salah, ed. 'Abd al-Rahman M. 'Uthman [Medina, 1389/1969], 139 [al-'Iraqi's comment]). This work of Ibn al-Salah's is likely to have been the source from which later authors quote Ibn al-Salah on the proper vocalization of the name Karram (al-DhahabT, MTzdn al-i'tidal, ed. CAllM. al-Bijawi [Cairo, 1382/1963], 4:22; Taj al-Din al-Subki, Tabaqdtal-shfi'iyya al-kubra, ed. CAbdalFattah M. al-Hulw and Mahmuid M. al-Tanahl [Cairo, 1383/1963], 2:305 (from majamTrof Ibn al-Salah). On this philological problem, see the exhaustive discussion in Texte, 8-11. It may be noted that Ibn al-Haysam's interest in the Qur anic sciences is revealed not only by the quotations from his writings that appear in al-MabanT and in the work of his grandson al-Haysam b. Muhammad, identified by van Ess (Texte, 68-73), but also by his activity in the field of qird9at. As reported by Ibn al-Jazari, Ghdyat al-nihaya fi tabaqat al-qurra', ed. G. Bergstrasser(Cairo, 1351-52/193233), 2:274, Ibn al-Haysam's learning in this area, as transmitted by Abu Muhammad Hamid b. Ahmad (Ibn al-Jazari, 1:202), was relied upon by Abu 'Abdallah Ahmad b. Abi 'Umar al-Khurasani (Ibn al-Jazari, 1:93) in his K. al-.dadh on the ten qirda't. Since, according to Ibn al-Jazari, alKhurasani died after 500 A.H., he is probably not to be identified with the Karrami expert on qir'adt Ahmad b. AbT 'Umar al-Misri known as al-Zahid al-Andarani (d. 470/ 1077) (Ibrahim b. Muhammad al-Sarifini, al-Muntakhab min k. al-siydq li-ta'rikh Ntsdbur, in The Histories of Nishapur, ed. R. N. Frye [Cambridge, Mass., 1965], f. 33a). In addition to Ibn al-Haysam, al-Mabant (173-74) quotes al-Shaykh Abui 'Amr 'Uthman b. Baqiyya al-Mazini, who is probably the Karrami Abui 'Amr al-Mazili or al-MazulT (Texte, 77-79). 9 MabnT, 6. 'o K. al-Ghurar fi asami al-suwar (MabdnT, 64); al-lbdna wa'l-i'rdb (Mabdan, 116); al-ZTnafi su'aldt al-qur'dn (MabdnT, 116) (an unfinished work); al-Durar ft tarftF al-suwar (Mabdnt, 172).

ZYsow: Two Unrecognized Karrdmi Texts since he reports having heard traditions from Ibn alHaysam, who died in 409/1019.11 Whatever theological tendency al-MabanTmay represent is kept well in the background.'2 The style throughout is that of conservative Islamic learning, rooted in the prophetic traditions. The work contains numerous isndds, replete with names that are now unidentifiable. Notable is the author's interest in furnishing an isndd of his own for the traditions that he cites from one of his chief sources, the K. Fihi mdfih of Abui Sahl Muhammad b. Muhammad b. 'All alTalaqani an-Anmari.'3 Our present ignorance of the Karramiyya is such that we cannot assign our author with definiteness to any particular branch of the sect. We do not know, for example, what significance, if any, is to be attached to the author's use of the epithet al-Imam al-Hadi for Ibn Karram.14 It is clear enough, however, that the author of al-Mabanl stood in close relation to the circles from which Ibn al-Haysam emerged, and one of the author's teachers, Abu Ja'far Muhammad b. Ahmad b. Ja'far is known to have transmitted traditions to Ibn al-Haysam.15 " MabdnT, date of death,see 207, 240. On Ibn al-Haysam's Texte,61-62. 12 Thereis a brief to the claim made reference disparaging by a Shliite imamin the mountainous regionsof Isfahanto be in possessionof a fullertext of the Qur'an(MabanT, 40). On the evolution of ImamT doctrine on this issue, see E. Kohlberg,"Some Notes on the Imamite Attitude to the Qur'an,"in IslamicPhilosophyand the ClassicalTradition: ... to RichardWalzer, ed. S. M. Stern,A. EssaysPresented
Hourani and V. Brown (Oxford, 1972), 209-24. 13 Mabdan, 16. 14 See n. 20 below. 5 Al-'Iraqi, al-TaqyTdwa'l-Lda.h, 139. This scholar may be the Muhammad b. Ja'far who was Ibn al-Haysam's teacher (Texte, 28-29) and is perhaps to be identified with the


Hanafijurist and asceticAbu Ja'far Muhammad b. Ahmad

b. Ja'far, who died in 413/1022 (Fasih Ahmad b. Jalal alDin KhwafT, Mujmal-efasLhi, ed. Mahmud Farrokh [Mashhad, 1339-41 solar/ 1960-62], 2:126). Another of the author's teachers, Abu'l-Qasim 'Abdallah b. Mahmashadh, also appears to have been older than Ibn al-Haysam, for he reports traditions directly from Abui Sahl al-Anmari (Mabain, 8), the author of K. Fthi mafih, whereas Ibn al-Haysam reports from Abu Sahl through Abu'l-Nadr Muhammad b. 'AlI al-Talaqani (Mabani, 240), as does another of the author's teachers, Abui 'Abdallah Muhammad b. 'All (MabanT, 8). On the Mahmashadh family, see C. E. Bosworth, "The Rise

The second of the works to which I wish to draw attention is al-Nutaffi'l-fatawd, edited about a decade ago by Salah al-Din al-Nahl, and attributed by him to the Hanafi QadT Abu'l-Hasan 'All b. al-Husayn alSughdi (d. 461/ 1068).16This is a work on Islamic law and covers the standard topics in two volumes of over 800 printed pages. It is characterized by a highly articulated presentation of the rules of law, with little attention to legal argument. There is, however, constant reference to the points of disagreement among the leading jurists such as Abui Hanifa, Abui Yiisuf, Muhammad al-Shaybani, al-Shafici and Malik. If every book has its secret, as the saying goes, then the secret of al-Nutaf is the identity of the jurist Abui 'Abdallah, whose opinions are as fully reported as those of the famous jurists named in the preceding paragraph, a fact which no attentive reader could fail to note. In fact, the manuscript tradition of al-Nutaf, as presented by Professor al-Nahi's edition, provides no less than three suggested solutions to this problem. But of these only one was recognized as such by the editor and discussed by him in the footnote that he devotes to the identity of Abii 'Abdallah.17 According to the editor, a marginal note in the Izmir manuscript of al-Nutafidentifies Abu 'Abdallah as Abui 'Abdallah al-Bukharf (d. 546/1151). This identification is rejected by the editor on the ground that al-SughdT (d. 461/1068), whom he regards as the author of al-Nutaf, died well before this Abui 'Abdallah. The editor himself suggests, only to reject, the identification of Abui cAbdallah with Imam alHaramayn al-Juwayni, whose death date, 478/1085, is also too late. He finally settles upon the identification of Abtu Abdallah as one of al-Sughdi's teachers. Although the editor fails to note it, the identification of Abiu 'Abdallah as al-Juwayni already appears in the text of al-Nutaf, in the form of what is clearly a copyist's gloss that has worked its way into the body of the work, producing the otherwise unknown Abu 'Abdallah al-Juwaynl.'8 Since Imam al-Haramayn's kunya was Abu'l-Ma'al, it is not clear how the editor came to consider the famous al-Juwayni as a possible
of the Karamiyyah in Khurasan," Muslim World (50) 1960, reprinted in Bosworth, The Medieval History of Iran, Afghanistan and Central Asia (London, 1977), 7-12. Texte, 33-35, furnishes an annotated family tree. 16 Baghdad,1975-76. 17 Nutaf, 1:8, n. 1.
18 Nutaf, 1:79. The interpolated words are on the second line from the bottom: al-JuwaynTmin ashab al-Shafi'T.


Journal of the American Oriental Society 108.4 (1988) Our knowledge of Karrami fiqh apart from alNutaf is sparse to say the least, but sufficient to corroborate that the legal doctrine attributed by alNutaf to Abiu 'Abdallah is Karrami. The sources to which we have to turn are not the standard works on legal differences (ikhtilaf), but writings of other genres. In the first place stand the references to Karrami
fiqh found in Ahsan al-taqasTm (completed about

candidate for the solution of the Abii CAbdallahmystery, unless this solution was suggested to him by this reference in the text of al-Nutaf itself. But if so, it is somewhat surprising that he was not disturbed by the anachronistic (on his own view) appearance of Imam al-Haramayn in the text he was editing. Another copyist's gloss interpolated into the text of al-Nutaf, which the editor has again failed to note, is the one that points in the right direction. This gloss In fact, Abil identifies Abii 'Abdallah as al-KarramT.19 CAbdallahis none other than Muhammad b. Karram, the founder of the Karramiyya, and al-Nutaf is more than simply a work of Hanafi fiqh, as the editor thought, but a unique source for the fiqh of the Karramiyya.20
19 Nutaf, 1:372. In K. al-Siydq li-ta'rTkh Nisdbur of CAbdal-Ghafir alFarisi (d. 529/1135) in The Histories of Nishapur, ed. R. N. Frye (Cambridge, Mass., 1965), the Karramiyya are routinely referred to as ashab AbTCAbdallah,a usage already noted by W. Madelung, "The Spread of Maturidism and the Turks," Actas do IV congresso de estudios arabes e isliamicos,

1968 (Leiden, 1971), 121, n. 32a, and by Coimbra-Lisboa

C. E. Bosworth in an addendum to the 1977 reprint of his "Rise of the Karamiyyah in Khurasan," 13. The evidence presently available suggests that this usage became prevalent in the course of the fifth/eleventh century. Reports of Ibn Karram's teachings by his contemporaries consistently refer to him as Ibn Karram, to judge by the al-islam 241-350, examples collected in al-Dhahabi, Ta'rTkh Leiden Or. 2363, f. 74a. In British Museum Or. 8049, which van Ess dates to about 400 A.H., the reference is consistently to Abu CAbdallah(Texte, 55, n. 236). So also al-Haysam b. Muhammad (first half of fifth/eleventh century) quotes a legal work of Abu 'Abdallah (Texte, 72, n. 325). In K. Rawnaq al-qulub (ca. second half of fifth/eleventh century) the usage alternates between al-Imam al-Zahid Abu CAbdallah and al-Zahid Muhammad b. Karram (Texte, 31). The evolution in usage is evident in the writings of nonKarramis. In CAbdal-Qahir al-Baghdadi (d. 429/1037), alFarq bayn al-firaq, ed. Muhammad Muhyi al-Din CAbd al-Hamid (Cairo, n.d.), 215-25, the reference is exclusively to Ibn Karram. The Zaydl al-Hakim al-Jushami (d. 494/1101) in his Sharh CUyunal-masd'il (passages translated in Texte, 20-29) usually refers to Abu CAbdallah, sometimes to Ibn Karram. The Imami Sayyid Murtada b. DaCi al-Razi (first half of sixth/twelfth century), K. Tabsirat al-'awamm ft maCrifatmaqalat al-anam, ed. CAbbas'Iqbal (Tehran, 1313), 64-74, almost always uses the form Abii 'Abdallah-e Karram. Al-Shahrastani (d. 548/1153), al-Milal wa'l-nihal, ed. 'Abd al-'Aziz M. al-Wakil (Cairo, 1387/1968), 1:108,112, refers to

375/985) of the geographer al-Maqdisi, whose account bears no signs of the hostility to the Karramiyya that we find elsewhere.21 Al-Maqdisi informs us about several specific Karrami legal teachings. Like ashab al-hadrth, the Karramiyya admit the wiping of the headcovering ('imtma) as part of ritual ablution (wu.du').22 On four points, al-Maqdisi tells us, the Karramiyya differ from all others: in their leniency (musamaha) with respect to the formulation of an intention (niyya) for acts of worship,23 in allowing mandatory prayers to be performed on horseback,24 in recognizing the validity of the fast of someone who eats unwittingly after daybreak,25 and in recognizing the validity of morning prayer which is still in progress when the sun rises.26 A fifth point is found in one of the manuscripts: the Karramiyya allow the jum'a prayer to be conducted with fewer than forty persons and outside of a settled town (misr jamiC).27For each of these points of law we can find a teaching attributed to Abu cAbdallah in al-Nutaf which is either directly on point or which at least furnishes the basis for a plausible explanation of al-Maqdisi's formulation.28
Abui cAbdallah. In his K. Nihayat al-aqddm, ed. A. Guillaume (Oxford, 1934), 105 (Arabic), we find the ibn-less Persian form Abui cAbdallah al-Karram (unless the proper reading is al-Karrami). 21 Al-MaqdisT says of the Karramiyya that they are ahl zuhd wa-tacabbud wa-marjicuhum ild AbT Hanmfawa-kull man rajaca ila AbT Hanifa ... fa-laysa bi-mubtadiC(Ahsan al-taqasfm fi ma'rifat al-aqalfm, ed. M. J. de Goeje, 2d ed. [Leiden, 1906, reprinted Baghdad, n.d.], 365). 22 Maqdisi, 40. Cf. Nutaf, 1:19. 23 MaqdisT, 40. Cf. Nutaf, 1:56-57, and see discussion in text below. 24 Maqdisi, 40. Cf. Nutaf, 1:83 (salat al-musayafa). 25 Maqdisi, 40. Cf. Nutaf, 1:157 (point 14). 26 MaqdisT,40. Cf. Nutaf, 1:80. 27 MaqdisT, 40. Cf. Nutaf, 1:90-93. This point is the only one that is troublesome. According to Abu 'Abdallah the jumca prayer may only be conducted in a town (qarya) that contains at least forty men (Nutaf, 1:92), but the prayer itself may be performed by two men (Nutaf, 1:93). 28 See references in nn. 22-27 preceding.

ZYSOW: Two Unrecognized KarramTTexts Other accounts of Karrami fiqh give us further details, but at the same time raise problems of a new dimension, since all come from sources with an undisguised hostility toward the Karramiyya. The brief account provided by the Ashcari Shafi'l heresiographer 'Abd al-Qahir al-BaghdadT (d. 429/1037-8) of what he terms "the unprecedented imbecilities of Ibn Karram in law" is characteristic of these sources.29 Al-Baghdadi cites four points of law. Ibn Karram, he tells us, held the view that the prayer of a traveler could be validly performed with two takbTrswithout the other elements that ordinarily constitute prayer. Ibn Karram further held that prayer could be validly performed by someone in a state of personal uncleanness, since he required cleansing (tahdra) only on account of ritual impurity (hadath). Ibn Karram also taught that washing and praying over the dead were only recommended, not obligatury like shrouding and burial. Finally, he taught that prayer and fasting, insofar as obligatory, required no intention (niyya), since he held that the intention of accepting Islam in the beginning (niyyat al-isldm fi'l-ibtida') was sufficient to satisfy the requirement of an intention for all obligatory rituals of Islam.30 Abu'l-Muzaffar al-Isfara'inT (d. 471/1078), who like al-Baghdadi was an Ash'ari Shafi'i and whose account differs but little from that of his predecessor, expands upon the role of intention in Karrami law. The "intention in the beginning," he tells us, is the intention expressed by all the future descendants of Adam when they acknowledged God as their Lord, as reported in the Qur'an (7:172). This primordial intention expressed by Adam's first progeny (al-niyya alsdbiqa fi'l-dharr al-awwal) obviates the requirement of any further intention with respect to obligatory rituals, but not with respect to supererogatory rituals, since these the first progeny did not accept. "Would that they knew," al-Isfara'inT adds, "on what basis they say this and on what basis they say that all obligations, in all their details, were set before them and accepted. If they base this on what is in the Qur'an, there is nothing more in the Qur'an than the offering to them of the formula of belief (kalimat alTmfn)."3' This account of Ibn Karram's teaching on


intention attaches it to the well-known Karrami doctrine that all men are believers by virtue of the original allegiance to the lordship of God that they undertook in their pre-existence.32 Of the four points of Ibn Karram's fiqh touched upon by al-Baghdadi, two are in direct contradiction to the information supplied by al-Nutaf. The traveler's prayer for Abu cAbdallah consisted of two complete rak'as.33 As for Ibn Karram's indifference to tangible impurity, al-Nutaf expressly states that according to Abu CAbdallah prayer cannot be validly performed when tangible impurity greater in extent than a dirham adheres to the clothing or the body.34 On the other hand, al-BaghdadT's summary of Ibn Karram's teaching on the obligations due to the dead appears to be correct.35There is also some basis for al-BaghdadT's report on Ibn Karram's doctrine of intention. Thus al-BaghdadT's account exhibits no more than a fiftypercent correspondence to the information in al-Nutaf, and even less if we look closely at the doctrine of intention presented in al-Nutaf.36 Al-Nutaf distinguishes between a "prior intention" (niyya qadTma) and a "supervening, or subsequent, intention" (niyya hdditha, niyya jadida). The former Al-BaghdadT, al-Farq, 223; al-Isfara'InT, al-Tabs[r, 69; Abu'l-Yusr al-BazdawT, Usil al-dTn, ed. Hans Peter Linss 211-12. (Cairo,1383/1963),
33 Nutaf, 1:75. 3Nutaf, 1:34. 3Nutaf, 1:32 (ghusl al-sunna). 3The account of KarramTlaw in Ibn DaCT, Tabsirat al32

29 Al-BaghdadT, al-Farq, 223: abdaa fi'l-fiqh hamaqat lam yusbaq ilayha. Abu'l-Muzaffar al-Isfara?nT, al-TabsTrfi'ldTn, ed. M. Z. al-Kawthari (Cairo, 1359/1940), 69, speaks of khurafat, hamdqat and jahdlat. Ibn DgCi, Tabsirat al-

'awamm,coverssome dozen or so points, of whichonly two or three are corroborated by, or can be so construedas to accordwith, al-Nutaf.For example,it is correctly statedthat Ibn Karram did not regard the qucid and two tashahhuds as essentialelements (arkdn)of saldt(p. 67;cf. Nutaf, 1:47). A numberof the teachingsattributedto Ibn Karramin Tabsiratal-'awammare highly shocking:homosexualrelations with non-Muslims are an act of worship(cibadat) (68to Qur'an9:120and a verseby a KarramT 69, with reference poet; cf. Nutaf 1:269[homosexualacts subjectto hadd of zina]); intercoursewith emission but without penetration does not make ghusl obligatory(68; cf. Nutaf, 1:29);unnaturalintercourse with a menstruating womanis permitted (69; cf. Nutaf, 1:137).Needlessto say, al-Nutafprovidesno basisupon whichto attribute suchteachingsto Ibn Karram. It may be noted that opinionsfavoringthe permissibility of anal intercourse wereattributed to both Malik(Ibn Juzayy,
Qawiinrn al-ahkdm al-sharT'iyya, ed. T. Sa'd and M. al-

Cawamm, 67, refersto thefa.daih of the Karramiyya.


Al-BaghdadT,al-Farq, 223-24. Al-lsfarainl, al-Tabsir, 69.

HawwarT 22 [slander]) and al-ShaficT [Cairo,1395/1975], (alAmir al-SanCanT (Subul al-saldm sharh buliigh al-mardm [Cairo,1379/1965],3:138).


Journal of the American Oriental Society 108.4 (1988)

is defined as the "intention (irdda) to perform future obligations as laid down by God at their appropriate times." The latter is the "intention to perform obligations that one intends to perform at the time (fi'lwaqt)." A prior intention obviates the need for a subsequent intention, although a subsequent intention merits a reward of its own.37 The doctrine of a prior intention appears to have been designed to provide a solution to certain technical problems of ritual law, particularly prayer.38There is no trace in al-Nutaf of any theological substructure for the legal doctrine.39 In fact, the language of al-Nutaf makes it clear that not everyone will have a prior intention, which could not be the case if the prior intention were common to all descendants of Adam.40 The correspondence between the legal teachings ascribed to Abui cAbdallah in al-Nutaf and the reports on Karramifiqh in al-Maqdisi and the heresiographers is sufficient to corroborate the identification of the Abiu Abdallah of al-Nutafwith Ibn Karram. Another question is the relative reliability of al-Nutaf and the heresiographers where they fail to agree. It is, of course, possible to prefer the heresiographers. The doctrine of al-Nutaf, it might be argued, represents a reformulation of Karram1law designed to rid it of its more objectionable features, along the lines of the reformulation of Karrami theology which Ibn al-Haysam is supposed to have accomplished.4' According to this hypothesis, al-Nutaf would be inlaw to Ibn accurately ascribing reformulated Karrdami Karram, while the heresiographers preserve his original teachings. There is, however, evidence that runs counter to such a theory. In the first place, al-Nutaf carefully distinguishes between the teaching of Ibn Karram and that derived from it.42 In the second Nutaf, 1:56-57.A thirdintentionis labelled"discriminawhatis obligatory sinceit "discriminates tory"(mumayyiza), (sunna) and what is (farTda)from what is recommended from what is supererogatory recommended (fa.d'il) (Nutaf,
1:57). 38 Nutaf, 1:58-60 (saldt), 1:155 (sawm); 1:207 (talbiya). 39 Nor is there any such trace either in Maqdisi, 40, or in Ibn Da'!, Tabsirat al-cawamm, 67. 40 idhadkdnat lahu niyya qadima (Nutaf, 1:58, 11. 12-13); wa-in lam yakun lahu niyya (1:155, 1. 12). 41 Al-Shahrastani, al-Milal, 1:112 (wa-qad ijtahada Ibn alft kull mas'ala); Haysam fi irmdm maqalat AbT CAbdalldh al-Shahrastani, Nihayat al-aqdam, Arabic pp. 105, 122 (tazwtrat Ibn al-Haysam . . . wa-innama huwa islah madhhab la yaqbal al-islah wa-ibrdm mu'taqad la yaqbal al-ibrdm wa'l-ihkam). 42 See n. 59 below.

place, a theory of this sort fails to come to terms with the agreement between al-Maqdisi's account and alNutaf. It is surely simpler to posit carelessness or even maliciousness on the part of the heresiographers than inaccuracy on the part of the author of al-Nutaf. Another possible course in upholding the accuracy of the heresiographers is to hypothesize a KarramT esotericism which would not readily expose eccentric Karrami legal teachings to the scrutiny of the uninitiated. The possibility of Karrami esotericism is in fact raised by a reference in the Persian heresiographical Tabsirat al-'awdmm of the Imami Ibn Da'i (first

half of the 6/12th century) to a Kitdb al-Sirr among the writings of Ibn Karram. "They call it secret," we are told, "because they do not show it to any but the inner circle (khawdss) of his followers." Somewhat ironically, given the alleged laxity of the Karramiyya with respect to purity, Ibn Karram is supposed to have written in his own hand on the flyleaf of this book, "Let none but the purified touch it (Qurdan 56:79)."43 Without dismissing the existence of KarramTesotericism out of hand, here again it is simpler to trace the problem that confronts us to the heresiographers than to the Karramiyya. The discussion of Karrami law in the heresiographical sources is confined almost exclusively to ritual law. This is hardly fortuitous. By virtue of its relative imperviousness to rational analysis, this area of the law is a touchstone of fidelity to the Islamic revelation and the Muslim community. Remaining within the accepted parameters of ritual law was thus of considerable symbolic importance, and within ritual law, the laws of purity must have been especially sensitive barometers of social cohesion. These considerations emerge with particular clarity in the account of KarramT legal practice furnished by the Shafilc QadTAbu Ja'far al-Zawzani and preserved by Ibn DaC'.44 Al-Zawzani portrays the Karramiyya
43 Ibn DaCT,Tabsirat al-'awdmm, 65. Ibn Da'Ts informant here, as for much of his account, is QadT Abu Ja'far alZawzani, for the identification of whom see the following note. The problem of Karrami esotericism is raised not only by Ibn Karram's K. al-Sirr but also by al-Hakim al-Jushami's reference to the secret teachings of the Karramiyya, which they termed ahkdm (Texte, 25). The teachings al-Jushami mentions are, however, unlike those quoted from K. al-Sirr, exclusively theological. 44 Al-Zawzani is surely to be identified with QadT Abu Ja'far Muhammad b. Ishaq al-Zawzani al-Bahhathi (d. 463/ 1071) a famous lampoonist, not the otherwise obscure Muhammad b. Ishaq b. cUzayr, whom van Ess tentatively suggests (Texte, 77), but whose kunya, Abu Bakr, is wrong.

ZYsow: Two Unrecognized

Karrdmi Texts


as at one and the same time both lax and overscrupulous with respect to the laws of purity.45Their perception of purity is thus seriously out of focus by Islamic standards. Ibn Karram, as already observed, applied to his Kitab al-Sirr the standard of purity appropriate to the Qur'an. Even more telling is Ibn Karram's alleged teaching on the impurity of wine: Should one drop of wine fall into the CaspianSea and a sparrowdrink a drop of water from that sea and fly off and then reachthe ocean sevenyearslater, and a bit of excrementfrom that bird fall into the ocean, the water of the ocean and the flesh of every creaturein the ocean would be prohibited.So that if someone should eat a bit of the flesh of a fish from the ocean, he wouldbe liableto the haddpenalty[for consumingwine]. And shouldhe die, it would not be lawfulto prayover him. But he shouldbe cast onto a so that birdsmaydevourhim.46 Magianfire-temple For biographical references, see 'Umar Rida Kahhala, Mu'jam al-mu'allifin (Damascus, 1379/1960),9:41. Yaqut
wrote of al-Zawzani, wa-ma taraka ahad min al-kubarda wa'l-a'imma wa'l-fuqahd' wa-sd'ir al-asndf min al-nds illd hajdhu wa-waqa'a fih (Mu'jam al-udaba', ed. A. F. Rifac'

[Cairo, 1936-38], 18:18).For one of his victims,Muhammad

b. 'Abd al-Rahman al-KanjarudhT (d. 453/1061), see alSarTfTnT, al-Muntakhab, f. 10a, where the text (1. 9) refers to al-QadT al-Zawzani. It may be noted that Ibn Abi'l-Wafa' al-QurashT, al-Jawdhir al-mudiyya ft tabaqdt al-hanafiyya (Hyderabad, 1332), 2:31, lists al-Zawzani as a HanafT. 45 Ibn DadT, Tabsirat al-'awamm, 67 (prayer in unclean garment), 68 (impurity has no effect on small amount of water unless color, smell or taste is changed) (confirmed by Nutaf, 1:9), 69 (personal uncleanness in prayer). Cf. the description of the KarramT Ishaq b. Ahmad al-Muh.ammadabadT (d. 478/1085) as muhtdt fi'l-tahdra wa-nazdfat althiydb in al-SarffinT,al-Muntakhab, f. 47a. 46 Ibn Da'T, Tabsirat al-'awdmm, 68. The ultimate source for this quotation is Ibn Karram's son, who related it to alBazzazT,a KarramTleader contemporary with Abu 'Amr alMazini (Ibn Dac', Tabsirat al-'awdmm, 65-66). The striking name of Ibn Karram's son, CAbdal-Jasim, appears to have gone unnoticed. Ibn Karram's forceful personality is still discernible from the meager sources presently available, and it is perhaps not surprising that he should have found this provocative way of asserting his belief that God is a jism (al-BaghdadT, al-Farq, 216). Ibn Karram's fondness for theological neologisms was already a topic of medieval comment (El2, s.v. Karramiyya, 4:668b). On the question of the original context of the quotations from Ibn Karram in Ibn Da'T, see Texte, 16-19. Al-Nutaf may in fact prove of assistance in corroborating the authen-

We therefore need not go so far as to suppose a sustained KarramTesotericism to explain the absence from al-Nutaf of the more provocative details of Karramilaw supplied by the heresiographicalaccounts. These accounts are clearly the last place where we would look for a balanced discussion of Karraml teachings, and we can fully agree with Bosworth when he says that "theological prejudice in these writers made them exaggerate the differences between the Karamiyyah and the other legal and theological schools."47 The identification and reliability of al-Nutaf as a law having been dealt with, the repository of KarramT tasks remain of eliciting what al-Nutaf can tell us about Karrami law and of situating the work itself in the Karrami legal tradition. Al-Nutaf leaves no doubt that Ibn Karram's legal doctrine was closely affiliated with Hanafism, but this is hardly a surprise. In defending the Karramiyya against the charge of being innovators, al-Maqdisi already pointed to, among other things, their affiliation to Abu Hanifa.48 The problem lies in identifying more precisely the relation between Karramism and Hanafism, and with respect to this problem, al-Nutaf, far from providing an obvious answer, only suggests more questions. Al-MaqdisT tells us that the Karramiyya possessed both a legal system and a theology (fiqh wa-kaldm) of their own.49 But other sources point to a more complex situation. According to Ibn Da'T most of
ticity of some of these quotations. For example, in his K. al-Sirr Ibn Karram is alleged to have posed the question of why washing underwear is not prescribed for farting "since the wind is no more free of dust (bdd-Taz ghobdr khali nabovad) than a fart of moisture" (Ibn Da'T, Tabsirat al'awamm, 67, translated in full in Texte, 16). This notion is

akin to Ibn Karram's remarkable conceptually teachingthat as a last resort tayammummay be performedby waving one's hands in the air "sincethe air is not free of dirt"(lianna al-hawa' la yakhliu min al-turdb) (Nutaf, 1:39, which

identifiesthis teachingas the view of Wahbb. Sayyar).The closest teachingto that of Ibn Karram that I have been able to findis that of the Shafi'i Abu Hamidal-Isfara'inT (d. 406/
1016), who would permit tayammum in a blowing wind (alrThal-hdbba) (Ibn Hajar al-'AsqalanT, Fath al-bdrTbi-sharh al-Bukhd rT[Cairo, 1378/1959], 1:450). 47 Bosworth, "Rise of the Karamiyyah," 7, n. 5. 48 See n. 21 above. Ibn Karram is reported to have trans-

mitteda traditionforetelling Abu HanTfa's role as a renewer

of the Prophet's sunna (al-Dhahabi, Mizdn al-i'tidal, 1:1067. See Texte, 49).

37. MaqdisT,


Journal of the American Oriental Society 108.4 (1988) quickly summarized. References in al-Nutaf to the legal opinions of Ahmad b. Harb (d. 234/848-9), already known as Ibn Karram's master in asceticism, suggest that Ibn Harb may have played a role in Ibn Karram's formation as a jurist as well.54 Apart from Ibn Karram, who is mentioned throughout al-Nutaf, there are frequent references to the legal views of a
54 On Ibn Harb's relation to Ibn Karram, see Massignon,

the Karramiyya had their own legal system, while some of the Karramiyya in Ghur and Sind followed Abi HanTfa.50 The combination of Hanafi law and Karrami theology also appears in the line of the poet Abu'l-Fath al-Busti (d. 400/1010): "the only true legal system (fiqh) is Abu Hanifa's, just as the only true religious system (din) is that of Muhammad b. Karam."51The ShVite 'Abd al-Jalil al-RazT(ca. 540/ 1145) distinguishes among HanafTs according to their theological affiliation: KarramT,Mu'tazili, Najjari or orthodox HanafT.52 An interesting postscript to these sources comes from Mas'id b. Shayba al-Sindi, a staunch Hanafi of the 7th/13th century. According to Ibn Shayba, Ibn Karram combined the theology of the anthropomorphists (usuil al-mushabbiha) with most of the law (akthar furc') of Abiu Hanifa. In his own day, Ibn Shayba tells us, "the Karramiyya are a small group (shirdhima) living in the mountains of Ghur and the suburbs (sawad) of Ghazna, whom the HanafTs shun and whom on occasion some of them anathematize."53 Presumably Ibn Shayba's familiarity with Karrami law was gained in those regions where, as we learn from Ibn DaCi, the Karramiyya adhered to a more or less standard variety of HanafTlaw, which Ibn Shayba simply attributed to Ibn Karram. The positive information that al-Nutaf furnishes with respect to the background of Ibn Karram's legal doctrine and to the evolution of Karrami law can be

50 Ibn DaFT,Tabsirat al-'awdmm, 76, 90. In his account of and the troublesof 595/1199involvingFakhral-Din al-RazT Ibn al-Qudwa(on which see C. E. Bosworth, the KarramT AsiaticJournal Central "TheEarlyIslamicHistoryof Ghur," al-Ward Ibn no. 2 VI, (d. 749/1349)speaks [1961]:131-32),

of the presence of al-karramiyya min al-hanafiyya wa'lshdficiyya (Tatimmat al-mukhtasar fi akhbar al-bashar

[Cairo, 1285],2:114).But this is most likelydue to carelessness. Ibn al-Athir(d. 630/1233) in recountingthese events
mentions al-fuqaha' min al-karrdmiyya wa'l-hanafiyya wa'lshdfi'iyya (al-Kamilfi'l-ta'rikh, ed. Tornberg [Leiden, 1853,

Beirut,1386/1966],12:151). reprinted
5' The translation is that of Bosworth, "Rise of the Karamiyya," 8, which gives the sources. Al-Subki, Tabaqat alshdfi'iyya, 2:305, which Bosworth does not list, has ra'y in place offiqh. 52 'Abd al-Jalil al-Razi, K. al-Naqd, ed. Jalal al-Din HusaynTUrmawi, n.p., n.d., 74. 53 ya9naf minhum al-hanafiyya wa-rubbama yalcanahum ba'duhum (Mas'ud b. Shayba b. al-Husayn al-Sindi, Muqaddimat k. al-taclim, ed. Muhammad 'Abd al-Rashid alNu'man [Hyderabad, 1384/1965], 205).

Essai, 259-60, 318. References to Ibn Harb's legal opinions appear in Nutaf, 1:6 (purity of water), 47 (prayer), and 501-2 (pre-emption). In the first reference Ibn Harb is reported to have refined a classic HanafT problematic. In the last reference, Ibn Harb and Ibn Karram are reported as in agreement with a reported teaching (reading wa-riwayatin at 502, 1. 1) of al-Shaybani (cf. 'Abdallah b. Mahmfid b. Mawdiid alMawsili, al-Ikhtiyar li-taCll al-mukhtdr, ed. Mahmud Abf DaqTqa [Cairo, 1951/1370, reprinted Beirut, n.d.], 2:44-45, where this teaching is attributed to Abi Yusuf). According to al-KhatiTbal-BaghdadT, Ta'rfkh Baghdad (Cairo, 1349/1931), 4:118, the Karramiyya "claimed" (tantahil) Ahmad b. Harb. Jacqueline Chabbi has suggested that this association with the Karramiyya may have led to Ibn Harb's being ignored in Sufi historiography until the 5th/llth Century ("Remarques sur le d6veloppement historique des mouvements asc6tiques et mystiques au Khorasan: iiie/ixe siecle-ive/xe siecle," Studia Islamica 46 [1977]: 30, 48, n. 3). It should be noted, however, that Ahmad b. Harb appears at the head of the list of fuqahd' and sdlihuin in the Persian translation of al-Hakim al-Samarqandi's (d. 342/953) al-Sawdd al-aCzam (Tarjome-ye al-sawad ala4zam, ed. 'Abd al-Hayy Habibi [Tehran, 1348], 146) but is missing entirely in the Arabic original (al-Sawdd al-a'zam, n.p., n.d., 31). Since this translation is dated by its editor to about 370 A.H., this pushes back the chronology for Ahmad b. Harb's re-emergence put forward by Chabbi. The Persian text is not otherwise favorably disposed to the Karramiyya (Tarjome-ye al-sawad al-aCzam, 115 [Karrami teaching on the Qur'an is kufr]). It is possible that Ibn Harb had a role to play in Ibn alHaysam's attempt to trace Karriam doctrine back to cAl by way of Sufyan al-Thawri (d. 161/778) (Ibn Abi'l-HadTd, Sharh Nahj al-balagha, ed. Muhammad Abu'l-Fadl Ibrahim [Cairo, 1379/1960], 6:371). See the report, transmitted by way of Ahmad b. Harb and Sufyan al-Thawri, of 'Al's being taught a supplication at the Kacba by al-Khadir, in al-Khatib al-BaghdadT, Ta'rrkh Baghddd, 4:118-19. It is interesting to note that Jum'a b. 'Abdallah al-Balkhi, who is supposed to have exposed Ibn Harb's Murji'ism (Ibn Hajar al-'Asqalani, Lisan al-mizan [Hyderabad, 1329], 1:149; Massignon, Essai, 259) appears in an isnad in MabnT, 17 (Jum'a b. 'Abdallah Abu Bakr al-Sulami al-BalkhT).

ZYsow: Two Unrecognized KarramTTexts certain Muhammad b. Sahib, sometimes referred to simply as al-Shaykh.55 In the form Muhammad b. alSahib this name also appears in one of the isnads of the Qur'an commentary British Museum Or. 8049, studied by van Ess.56 On the basis of van Ess' dating of this work to about 400, the place of Ibn Sahib in the isndd in which he appears allows us to assign him to about the first half of the 4th/ 10th century.57 Ibn Sahib appears to have been quite independent in developing his own legal opinions and may have stood to Ibn Karram in the same relation that Abu Yusuf and al-Shaybani bore to Abu Hanifa.58 There is, moreover, language throughout al-Nutaf that indicates that Ibn Karram's legal teachings underwent a typical elaboration at the hands of his followers.59 Unfortunately, al-Nutaf does not ordinarily furnish the arguments for the rules of law that it states.60Any
55 Sometimes we findal-ShaykhMuhammad b. Sahib(e.g., Nutaf, 1:322). Muhammadb. Salih (Nutaf, 1:602, 612) is


reconstruction of the missing argumentation must depend upon general progress in the study of Islamic law. But the study of Islamic law, to which al-Nutaf has a notable contribution to make, has not advanced to the point of making such a reconstruction feasible. Still less are we in a position to attempt an overall assessment of Karrami law as contrasted with some other Islamic legal system. As far as legal theory (usul al-fiqh) is concerned, al-Mabian has more to tell us than al-Nutaf, which has only a few suggestive remarks here and there.61 From al-MabanT it is clear that the Karrami tradition represented by Ibn al-Haysam, which we can provisionally regard as mainstream Karramism, embraced a standard variety of legal theory, one that recognized the validity of rules of law that were only probable (zann).62 But there is evidence that other types of 36 (spermis pure becauseGod would not create a prophet from what is impure),39 (tayammum in air), 75 (permission
to break fast of Ramadan [iftdr] is ikrdm from God of which

clearly a typographical error. That Muhammad b. Sahib is

the samepersonas al-Shaykhis corroborated by the appearance of the latteras a variantfor the former(Nutaf, 2:622; and see editor's note at Nutaf, 1:164).The only difficulty is the presentedby the text against such an identification attributionto Muhammadb. Sahib and al-Shaykhof incompatibleopinionson a singlepoint of law at Nutaf,2:699, 1. 13. Possibly al-Shaykhhere is a misinterpretation by a shTn for al-Shafi'cthat is used in copyist of the abbreviation some of the manuscriptsof al-Nutaf. But there is some question whether the teaching at issue (consuming the property of another rather than eating mayta) is to be attributed to al-ShaficT. The opposingopinionappearsin alShafi'c, al-Umm (Beirut, 1400/1980), 2:277, but see Abu al-Muhadhdhab Ishaq al-ShTrazi, (Cairo, n.d.), 1:250 and the commentary of al-Nawawi, al-Majmic'(Cairo, n.d.),9:46. The otherwiseunknownAhmad b. Muhammad b. Hilal al-Haddadi may also have been a KarramT jurist (Nutaf, 1:90, where the full name appears, 171 [al-Haddad],385 [pairedwith Ibn Sahib againstal-fuqaha'and Abu 'Abdaln. 78 below. lah], 2:702).See further
56 Texte, 44. 57 Texte, 54.

one travelling for a sinfulpurposeis unworthy), 93 (khutba), 128(ten reasons('ilal) whyfuneralserviceis "prayer" (saldt) in the strictsense). Certain of Ibn Karram'slegal opinions appearto bear some relationto the circumstances of his life. For example, Ibn Karram's of someone leniencywithrespectto the prayers
kept in confinement (Nutaf, 1:84 [tayammum in air]) is

reminiscent of Ibn Karram's own long imprisonment and his to leave for thejumCa repeatedpracticeof preparing prayer.
When stopped by the guard, he would say, "O Lord, I have

tried my best. The obstacleis not of my making(wa'l-man'

min ghayrr)." (al-Subki, Tabaqdt al-shdfi'iyya, 2:302, quoting al-Hakim al-Nisabuiri). Similarly, Ibn Karram's con-

siderablepowersas a preacher are suggested by his opinion, in agreement with that of Abu HanTfa, that both delivering
and listening to the Friday khutba were obligatory (farTda) but could be discharged by a single word "since the point of the khutba is in the moral lesson ('iza) and there can be a powerful lesson ('iza baligha) in a single word" (Nutaf, 1:93). 61 For example, Nutaf, 1:40 (qiyds), 83 (taklif md l yutdq), 436 (istihsan), 490 ('awn 'ala al-ma'siya), 501 (Muhammad b. Sahib: ahwat). Ibn Karram's use of discordant traditions (which are generally not cited in al-Nutaf) may reveal patterns discernible in other Islamic legal systems (e.g., Nutaf, 1:8-9 [purity of water]), but this point cannot be entered into here. 62 Mabdanl p. 47, 1. 13 (bi-khabar al-wdhid dun al-jamc alladh yulzim al-yaqTn), 195, 1. 19 (bi-akhbdr tunqal ilayhim cala alsinat al-ruwat mimmd la yanqatic (read yuqta) cald mughayyabihT bi'l-yaqTn). The important passages on usul al-fiqh in MabanT,200-206, are replete with the terminology characteristic of al-tahsTnal- aqli (hakTm,hikma, tajwz fi'l-

He often appearsholdingan opinionopposedto that of Ibn Karram(e.g., Nutaf, 1:164, 216, 224, 385), but his

opinions are reported even when in agreement with those of Ibn Karram (e.g., Nutaf, 1:403, 447). 59 Qiyas qawl AbT 'Abdallah (Nutaf, 1:154, 361, 2:699 [= qawl Muhammad b. Sahib]); ashbah bi-qawl AbT'Abdallah (Nutaf, 1:156, 202); riwdyatdn(Nutaf, 1:197, 310); qawlan (Nutaf, 1:418).

For examples of argumentscited in support of the

opinions of Abui 'Abdallah, see Nutaf, 1:14 (israf in wudui),


Journal of the American Oriental Society 108.4 (1988) certain al-Ghaznawi, al-Timurtashi, and Sharaf alDln Qasim b. Husayn al-Damraji (al-DamraghT)
(d. 854/1459).68 The manuscripts examined by the

legal theory won support among the Karramiyya.63 Al-Maqdis? tells us that some of the Karramiyya accepted general infallibilism (taswTb) and held that every mujtahid was correct with respect to both theology and law. This teaching, as al-Maqdisi notes, was associated with the Murji'a.64 And Abu'l-Muzaffar al-Isfara'lni writes of a leading KarramT, who, wishing to elevate theology (kalam) above law, used to say that the learning of al-Shafic" and Abu Hanlfa could all be fitted under a woman's hems.65This attitude is characteristic of legal infallibilism, which regards every mujtahid as correct on points of law, since law is not a discipline in which certainty is generally attainable. Certainty pertains to theology, which is the learning worth acquiring. Legal infallibilism was widespread in Mu'tazill circles and came to dominate cIraqi Hanafism, so that it is not surprising to find that it made headway among Karramijurists as well.66 The recognition of al-Nutaf as a unique source for the fiqh of Ibn Karram calls for a re-examination of the identity and legal affiliation of its author. The editor of al-Nutaf, Professor al-Nahi, accepts the attribution of the work to the Hanafi Qadi al-qudat Abu'l-Hasan CAl b. al-Husayn al-Sughdi (d. 461/ 1068), a teacher of al-Sarakhsl (d. 490/1096).67 But this attribution is hardly free from doubt. Hajjl Khalifa, in fact, lists the names of no fewer than four authors to whom al-Nutaf was attributed, with alSughdl standing in the first place, followed by a
'uqil) and thus confirm the Karrami espousal of this ethical doctrine reported elsewhere (e.g., al-Shahrastani, al-Milal, 1:113). On this point see Texte, 17. 63 It is not clear whether the denial that there is a category of indifferent (muhmal, mubdh) human conduct attributed

editor, not all of which bear the name of an author, do actually exhibit the different attributions noted by Hajji Khalifa, except for that to al-Timurtashi.69 To complicate matters, al-Nutaf appears to have also madhhab circulated under the title Nutafal-hisdn Cald AbT HanTfa al-NuCmdn, a work for which Hajji Khallfa has a separate entry and which was attributed to Abii Bakr al-Wasitl, a teacher of al-Sarakhsl, and to Abii 'Abdallah al-Barqi, as well as to others not named.70The foregoing makes it amply clear that the identity of the author of al-Nutaf was no longer familiar to the Hanafl circles in which the work circulated. Internal evidence that could assist in the identification of the author of al-Nutaf is very sparse. There is no introduction or epilogue to speak of. Of the identifiable jurists named in the work, most are early.7 The latest jurist mentioned appears to be the HanafT Abu JaCfar al-Hinduwani (d. 362/973).72 Unfortunately our knowledge of the organization and technical terminology of Islamic legal texts hardly permits

Hajji Khalifa, Kashf al-zuniun, ed. 5. Yaltkaya and R.

2:1925. Bilge(Istanbul,1362/1943), 69 See editor'sdiscussion,Nutaf,2:867-72,wherethe form al-Damirji,instead of al-Damrajl,appears,erroneouslyit seems, throughout. The printed edition is based on the following manuscripts:1. Ukta'T, Fihrist-e kitdbkhane-ye astdne-yeqods-e radawT (Mashhad, 1329),vol. 5, 884fiqh
(attributed to al-Sughdl); 2. Da'ud al-Chelebi, K. Makhtutit al-Mawsil (Baghdad, 1346/1927), MS186 (Madrasat 'Abd alRahman al-Sa'igh, without attribution); 3. Yeni Cami 500

ed. to the Karramiyya in K. al-Iktisdb fi'l-rizq al-mustatdb,

'Izzat Amin al-CAttar(Cairo 1357/1938), 68, is meant as a serious statement of legal theory. On K. al-Iktisdb, which purports to be an abridgment by Muhammad b. Sama'a (d. 233/847-48) of a work of al-Shaybani (d. 189/-805), see Texte, 75-76. 64 Maqdisi, 38-39. 65 Al-Isfar'lnli, al-TabsTr,69. 66 See Chapter 5 of my "Economy of Certainty: An Introduction to the Typology of Islamic Legal Theory," Ph.D. diss., Harvard University, 1984. 67 See the editor's discussion, Nutaf, 2:872-73, 877-78. A biographical notice on al-Sughdl's teacher Abu Muhammad CAbdallahb. Ahmad al-Kufini, which the editor was unable to locate, may be found in al-Samcani, K. al-Ansab, ed. D. Margoliouth (London, 1912), f. 485b, who gives Muhammad in place of Ahmad.

4. Izmir (attributedto al-Qasimb. al-Husaynal-Damraji); to Abu 'Abdallahal-Qasimb. al-Husayn 36/312 (attributed al-Ghaznawi).

Kashf al-zunfin, 2:1925. A. J. Arberry, The Chester

Beatty Library: A Handlist of the Arabic Manuscripts (Dublin, 1955-64), lists al-Nutaf al-hisdn,attributedto alQasim b. al-Husayn al-Damraghi(2:96, MS3473) and alNutaf CalamadhhabAbTHanifa, attributedto Abu Bakr
al-Wasiti (3:88, MS3697). 71 The HanafT Abu Ja'far al-Tahawi (d. 321/933) is in one (Nutaf, 1:54). See place referredto as bacd al-muta7akhkhirTn 'Ala' al-DTn al-SamarqandT, Tuhfat al-fuqahda, ed. Muhammad ZakT 'Abd al-Barr (Damascus, 1377/1958), 1:182, and al-Mawsill, al-Ikhtiydr, 1:39-40, for the identification. Al-Tahawi is elsewhere referred to by name (Nutaf, 2:700, 782). 72 Nutaf, 1:191.

ZYsow: Two Unrecognized

Karrdml Texts


the confident invocation of such criteria for the purposes of dating.73 The legal affiliation of the author is also uncertain. Al-Nutaf does not, any more than al-Maban-, bear an obviously Karrami character on its face, and it is not surprising that the editor, along with earlier Hanafi scholars and copyists, should have regarded al-Nutaf as a work of Hanafifiqh, insofar as the text, given its comparativist character, represents any particular school at all. Nonetheless, there are several reasons for believing that the author was a Karrami jurist. In the first place, the author, who, as observed, rarely provides the reasoning behind the rules of law that he sets forth, seems to do so most often in the case of Ibn Karram's teachings.74 And on occasion he even offers arguments of his own to bolster Ibn Karram'sviews.75Secondly, Karrami doctrine is sometimes presented without attribution as the accepted teaching.76 But the strongest argument for identifying the author of al-Nutaf as a Karrami is his interest in Ibn Karram's fiqh. There is no HanafT work with which I am familiar that even remotely suggests a comparable concern with Karrami law, and there is evidence, that from Ibn Shayba, that orthodox Hanafis shied away from the Karramiyya. The attribution of al-Nutaf to al-SughdT, a Hanafi of impeccable credentials, is therefore highly doubtful. The dating of the work to al-Sughdi's time, ca. 1050, is, however, not unreasonable in the absence of other evidence. If one has to hazard a guess, it may be suggested that al-Nutaf was written for Karrami students of law, some of whom were Hanafis and some of whom followed Ibn Karram or his successors such as Muhammad b. Sahib. The author's constant concern to

present the views of Ibn Karram, even when Ibn Karram is in agreement with the leading Hanafis, makes it clear that no attempt is being made to absorb Karrami law into mainstream Hanafism.77The
Hanafis represent a distinct group, of which Ibn Karram is not a part.78 It was the author's wellorganized and lucid presentation of HanafTfiqh that led to the popularity of al-Nutaf among generations

of Hanafi jurists who no longer knew the identity of Abi CAbdallahor had any other reason to suspect the
work's Karramir origin.

77 It is possible that Ibn Karram's transmitted legal opinions did not cover all areas of law. This would account for the absence of Abu 'Abdallah's name from various parts of the text of al-Nutaf and would explain such a remark as that made in connection with the discussion of -ld':hadha kulluhu qawl Abi HanTfawa-ashdbihT (Nutaf, 1:370). 78 The Hanafis as a group are sometimes referred to as Abti Hanifa wa-ashabuhu (e.g., Nutaf, 1:54, 229, 385 [four times]), but more frequently simply as al-fuqahd' (e.g., Nutaf, 1:16, 28 [as distinct from Malik, al-Shafi'T and Abi cAbdal-

lah]). At Nutaf, 1:75, the opposition is between the two farlqs, that is, the fuqahai and the ShafiCis,with Abu
cAbdallah in the middle. The opposition of al-fuqaha' and Abu 'Abdallah is frequent (e.g., Nutaf, 1:9 [wa-hddha qawl al-fuqaha' jamlcan wa-fT qawl AbT CAbdallah], 58 [twice]). Similarly, the fuqaha' are opposed to Muhammad b. Sahib (Nutaf, 1:404, 427 [twice]) and to al-Haddadi (Nutaf, 1:385). At Nutaf, 2:702, al-Haddadi is opposed to Abui Hanifa wa-ashdbihT.

At several points the author of al-Nutaf refers to "our

jurists" (fuqaha'und), a term which may include both Hanafis and Karramis (Nutaf, 1:106, 361, 436 ['ulama'una], 2:702 [al-amr bi'l-maCrtf]). Nutaf, 1:503, is the strongest argument for this interpretation. There the author contrasts the opinion of some fuqahda that a minor has no right of pre-emption (shuf a) with that of "ourfuqahd'," who give a minor equal rights with an adult in this matter, and then goes on to specify the teachings of Abu Hanifa, Abu Yfsuf, al-Shaybani and Abu cAbdallah. On the other hand, the antecedent of "our fuqaha"' at Nutaf, 1:374, 1. 7, is Abu .Hanfa wa-ashabuhu (as opposed to al-Shafi'i), although opinion from that of the HanafTs on the question at issue, that yaauduna in Qur'an 58:3 refers to the resumption of conjugal relations, not the repetition of the zihdr formula (cf. al-Samarqandi, Tuhfat al-fuqahda, 2:320-21 [for alShafi'i, 'awd is failure to pronounce talq; for the Zahiris, it is the repetition of the zihdr formula]).

7One example of an apparently unusualusage is khiyar for what is commonlyreferred to as khiyaral-qabfil al-Caqd (Nutaf, 1:433,editor'snote). 74 For examples, see n. 60 above. 75 128(waNutaf, 1:41(wa-yuqawwT qawl Abi CAbdallah), awkadmin dhalikakullihl). 76 The most strikingexampleof this is the teachingon intention(Nutaf 1:56-57)discussed above.See also Nutaf, 1:32 (washingof the dead is sunna; it is wajib for the Hanafis Tuhfat [al-Samarqandi, al-fuqahad, 1:378])and Nutaf, 1:497 (shuf'a availableto khallt only accordingto al-Shaykh; the standardHanafiteaching,whichrecognizes severalclassesof claimants(tartib), is not presented untilp. 502).The passage on qibla(Nutaf, 1:70)is also noteworthy.

there is no indicationthat Abu 'Abdallahheld a different