Printer Friendly

Bridging the Gap: Two Early Texts of Islamic Legal Theory.


The purpose of this paper is twofold: to introduce two surviving, very early legal-theoretical texts, written by Ibn Surayj (d. 306/918) and al-Khaffaf (fl. first half of the fourth/tenth century) respectively, and to demonstrate that the existence and content of these texts can critically inform current debates on the early history of Islamic legal theory (usul al-fiqh). Specifically, I seek to make two contributions--first, to show that these early texts were not self-standing works but rather formed part of larger works on positive law, thus calling into question the almost exclusive focus on legal theory as a stand-alone genre in previous historiography; and second, to shed new light on the continuities and discontinuities in this genre in its earliest stages since the two texts fall squarely within what has been seen as a gap in the literature between the Risala of al-Shafi[??]i (d. 204/820) and the so-called classical literature on legal theory that arose in the late fourth/tenth century. (1) To this end I first introduce each text, its surviving manuscripts, and its author. I then discuss the topics each work treats and describe how the contents relate to al-Shafi[??]i's Risala, to each other, and to later legal theory in order to highlight similarities, differences, and change over time. Finally, I provide a complete edition of both texts.

One of the most vexing questions in the historiography of Islamic legal theory is the development of the discipline between its first surviving articulation by al-Shafi[??]i at the beginning of the third/ninth century and the next cluster of surviving legal-theoretical works written in the second half of the fourth/tenth century. (2) The long hiatus separating al-Shafi[??]i's Risala from its successors and the obvious differences in structure and content that distinguish the later works from the Risala have prompted several speculative explanations. In the 1980s George Makdisi argued that al-Shafi[??]i's legal theory had been conceived as an alternative to theology and that what we see in the second stage of legal-theoretical writing in the late fourth/tenth century is the theologians' Trojan horse-like appropriation of legal theory through the injection of their own concerns and methods into the discipline. (3) In a 1993 article Wael Hallaq used the observation that al-Shafi[??]i's work differs in important respects from later works to deny the existence of a genealogical relationship between the Risala and the classical genre of legal theory. (4) He argued instead that the Risala is primarily a theoretical work whose aim is to establish the authority of hadith, while the discipline of legal theory concerns itself with theorizing the relationship between reason and revelation. His search in Ibn al-Nadim's Fihrist for early works with the word usul in the title having been unsuccessful, Hallaq concluded that there is no evidence of legal-theoretical literature before the generation of Ibn Surayj, a full century later. Accordingly, Hallaq proposed an alternative genealogy of the discipline by locating its origins in the work of Ibn Surayj and his students.

In the last decade, careful studies of various legal-theoretical issues by, for example, Hans-Thomas Tillschneider, (5) David Vishanoff, (6) and Ahmet Temel (7) have demonstrated that the thesis of a radical discontinuity between the Risala and the later literature is not tenable and that beyond the undeniable differences a clear sense of continuity can be detected. In order to illuminate the intervening period, Devin Stewart has successfully reconstructed several fragments of otherwise lost legal-theoretical works from the third/ninth century, including writings by Dawud al-Zahiri (d. 270/883f.), his son Abu Bakr Muhammad al-Zahiri (d. ca. 297/909), and Muhammad b. Jarir al-Tabari (d. 310/923). (8) In addition, I have demonstrated elsewhere that extensive legal-theoretical discussions can be found in surviving third-/ninth-century texts of positive law (fiqh), exegesis (tafsir), and hadith. (9) The following analysis of two treatises on legal theory, written in the late third/ninth or early fourth/tenth century as either introductions or conclusions to short works of positive law, continues this line of inquiry. The first treatise is contained in Ibn Surayj's al-Wada[??]i[??] li-mansus al-shara[??]i[??], the second in al-Khaffaf's al-Aqsam wa-l-khisal.


Abu l-[??]Abbas Ibn Surayj was an important Shafi[??]i jurist in Iraq. He served as a judge in Shiraz and trained the most significant Shafi[??]i jurists of the next generation, who would subsequently make Iraq and Khurasan the centers of Shafi[??]i thought. In recent scholarship Ibn Surayj has emerged as an indispensable figure in the historiography of Islamic law: he has been lauded as the true father of legal schools in general and of the Shafi[??]i school in particular, (10) and as the crucial node through which legal-theoretical thought passed to his students, who were the originators of legal theory proper. (11) However, the claims of Ibn Surayj's significance are based on biographical works written long after his death; there seems to have been no attempt to corroborate them through an analysis of his actual writings on law or legal theory. (12) The present article attempts to address this lacuna.

Ibn Surayj's Kitab al-Wada[??]i[??] li-mansus al-shara[??]i[??] comprises 128 folios in the most complete extant manuscript and contains both positive law and legal theory. It served as one of the main sources on Ibn Surayj's ideas on the latter subject for Badr al-Din al-Zarkashi (d. 794/1392), the Mamluk-era historian of Islamic and particularly Shafi[??]i usul al-fiqh. (13) At least two manuscripts of the work survive: a complete version in Istanbul's Aya Sofya collection and a significant fragment in the collection of [??]Abd al-Hayy al-Kattani (d. 1962), now housed in the Moroccan National Library in Rabat. (14) It was edited as an MA thesis by Salih b. [??]Abd Allah b. Ibrahim al-Duwaysh at the Imam Muhammad b. Sa[??]ud University in Saudi Arabia, but the edition was never published. (15) I have used both this edition and the Istanbul manuscript. The legal theory sections are also quoted almost in full, albeit piecemeal, in Husayn al-Juburi's al-Imam Abu l-[??]Abbas ibn Surayj wa-ara[??]uhu al-usuliyya. (16)

The section on legal theory is situated at the end of the work, but Ibn Surayj contextualizes its incorporation in the introduction, where he states:
When I examined how many paths of transmission the people of knowledge
have gathered, and how much they have differed concerning reports, and
how extensive the [legal discussions] are that experts have engaged in
despite the paucity of authentic prophetic traditions and [historic]
precedents, I found this to be due to their refraining from clarifying
their method (ibanat al-usul) and from indicating what [evidence?] they
exclude. Instead, in their drive to cover all possible events they fail
to provide unified guidance. I therefore decided to compose a work that
brings together methods (usul al-din) and individual rules (furu[??]),
so that it is easier to understand and memorize for the reader, and God
is the giver of success in what he loves and is pleased by. You [...]
asked me about knowledge of traditions (sunan) and rulings contained in
the Quran and transmitted of the sunna of God's Messenger, peace be
with him. I will make my explanation to you in the clearest possible
manner, through an articulation that facilitates understanding and
plants its knowledge in everyone, specialist or layman. (17)

Although the bulk of the work is dedicated to fiqh, the role of a methodology (here called usul al-din) is significant. Ibn Surayj stresses that scholars have accumulated a wealth of prophetic reports and gone to great lengths in investigating the minutiae of the law, but he argues that in both of these exercises they have neglected to devote sufficient attention to justifying and structuring their work with reference to a unified methodology. In other words, the shortcomings of both the hadith scholars (ahl al-hadith) and the jurists (ahl al-fiqh) can, in his opinion, be remedied through an explicit engagement with methodology.

The methodology that Ibn Surayj presents in his final section is remarkably similar to that found in al-Shafi[??]i's Risala. Most noticeably, the range of topics Ibn Surayj covers does not extend beyond the issues introduced by al-Shafi[??]i. The work addresses abrogation (naskh), prophetic tradition (sunna), the distinction between the general and the particular ([??]amm and khass), indeterminate vs. determinate revelatory statements (mujmal and mufassar), the status of single-transmitter reports (akhbar ahad, sg. khabar wahid), consensus (ijma[??]), and analogy (qiyas), and it concludes with an exposition on the obligation to seek knowledge. The high degree of similarity between the contents of the ###Wada[??]i[??] and those of the roughly contemporary legal-theoretical writings of Dawud and Muhammad al-Zahiri, as reconstructed by Stewart, indicates that these topics constituted the realm of legal-theoretical discussions until Ibn Surayj's generation.

Parallels with al-Shafi[??]i's work are also evident in the treatise's substantive positions. Ibn Surayj's discussion of abrogation, which opens the section on legal theory, sets out three forms of abrogation: a text (1) superseding another text's legal implication, but leaving the latter text itself in place; (2) superseding (and erasing) the legal implication as well as the text; and (3) superseding (and erasing) the text without canceling its legal implication. Al-Shafi[??]i does not discuss the second type in his extant writings--unsurprisingly, since it does not represent a challenge to interpretation and al-Shafi[??]i showed no inclination to mention issues simply for the sake of completeness. He acknowledges but does not endorse or theorize the third type, (18) and his discussion focuses primarily on the first type. (19) Ibn Surayj's position that only the Quran and not the Sunna can abrogate the Quran is identical to al-Shafi[??]i's, (20) and he also uses the same proof text, (21) despite the fact that this position had already been criticized even by sympathetic scholars such as Muhammad b. Nasr al-Marwazi (d. 294/906), whom later Shafi[??]is counted among their own. (22)

On the authority of single-transmitter hadith, Ibn Surayj again uses a proof text already employed by al-Shafi[??]i in the Risala, namely, the instructions given by Muhammad to Mu[??]adh b. Jabal (d. 18/639) upon sending the latter to Yemen. (23)

In the section on qiyas, Ibn Surayj also closely parallels al-Shafi[??]i. Ibn Surayj employs the root [??]-q-l ("to reason") only once, in the same context in which al-Shafi[??]i uses it: in justifying analogy with reference to obligations imposed by God that can be discharged only through reasoning (bi-taqdir al-[??]uqul) or, as al-Shafi[??]i put it, "by means of the reason implanted in them" (bi-l-[??]uqul allati rakkaba fihim). (24) In both scholars' discussions the term [??]illa ("cause"), which was to dominate discussions of qiyas in classical legal theory, is missing. In its stead they use the term ma[??]na ("reason"). (25) Besides this ma[??]na-based analogy, al-Shafi[??]i also uses analogy based on shabah ("resemblance"), in which a novel case is analogized to the precedent that most closely resembles it: al-shay[??] yushbih al-shay[??]. (26) Ibn Surayj's discussion parallels al-Shafi[??]i's shabah-based analogy, which Ibn Surayj describes as tamthil al-shay[??] bi-l-shay[??] wa-tashbih al-shay[??] bi-l-shay[??]. (21) Al-Shafi[??]i uses the roots m-th-l and sh-b-h to describe the raison d'etre of qiyas and the verbal noun tashbih to denote the process of analogy. (28) The term tamthil is not found in the Risala, but it appears already in Abu Ya[??]qub al-Buwayti's (d. 231/846) abridgment of the Risala (tamthil al-shay[??] bi-l-shay[??]), (29) which may have had a direct or indirect influence on Ibn Surayj. The Isma[??]ili jurist al-Qadi Nu[??]man (d. 363/974) seems to have been familiar with Ibn Surayj's text, since he offers in Ikhtilaf usul al-madhahib a refutation of the latter's specific arguments for qiyas. (30)

Finally, Ibn Surayj also takes some very unusual positions that diverge both from al-Shafi[??]i's Risala and from the later legal-theoretical literature. These include his stance that prophetic actions should be treated as imperatives (awamir, sg. amr) that confer obligations (31) and his statement that the definition of ijma[??] is simply the "true position" (qawl al-haqq) and that therefore the opinion of a single person can constitute ijma[??]. (32)


The Chester Beatty Library in Dublin houses a short manuscript on law titled al-Aqsam wa-l-khisal that the catalogue ascribes to Ibn Surayj (33)--probably because Hajji Khalifa (d. 1058/1657) attributes a work with the title al-Khisal to him. (34) However, it appears that the author of the manuscript is in fact a fourth-/tenth-century Shafi[??]i called al-Khaffaf. We have little biographical information about this scholar. Abu Ishaq al-Shirazi (d. 476/1083) mentions al-Khaffaf among the Shafi[??]is in his Tabaqat al-fuqaha[??], but notes merely that he is the author of al-Khisal, giving no death date or further biographical details. (35) (Hajji Khalifa provides al-Khaffaf's death date as 261/874, but this is clearly a mistake: he has confused al-Khaffaf with the Hanafi jurist al-Khassaf.) (36) However, since al-Shirazi lists his entries according to generation, and al-Khaffaf's entry appears immediately after that of Ibn Haddad, who died in 345/956f., together with a cluster of entries for students of Ibn Surayj, it is likely that al-Khaffaf was active in the first half of the fourth/tenth century. Later biographers seem to have been unable to find out more about al-Khaffaf than what could be gleaned from al-Shirazi's entry and the contents of the Khisal--al-Zarkashi, Taj al-Din al-Subki (d. 771/1370), and Ibn al-Mulaqqin (d. 804/1402) cite al-Aqsam wa-l-khisal and attribute it to al-Khaffaf. (37) Ibn Qadi Shuhba (d. 851/1448) also attributes the text to al-Khaffaf, mentions its pattern of introducing each section with "the clarification of such-and-such [a topic]," and notes that the introduction of the text deals with legal theory. (38)

The Chester Beatty manuscript is, as far as I can see, a unicum. Unfortunately it is severely water-damaged, and the transcript consequently contains lacunae, but the name "al-Khaffaf" is faintly visible on the title page. Internal evidence from the text, discussed below, further supports the attribution to al-Khaffaf.

Al-Aqsam wa-l-khisal is primarily a book of fiqh, but its 3,000-word introduction (henceforth Muqaddima) contains a concise but complete exposition on legal theory. The author begins with the following explanation:
I have received your request included in your letter, namely, that you,
may God perpetuate your honor, wish that I write a book in the mold of
the people of Iraq according to the school of al-Shafi[??]i, may God
lighten his face, and according to al-Muzani's chapter ordering, and
that I preface it with a section on legal theory (usul). I have heeded
your request, trusting in God's support, illustrious and mighty is he.
I wrote the book and called it the "book of parts and properties"
(kitab al-aqsam wa-l-khisal), and I have explained its meanings
according to my knowledge, aiming at conciseness in my efforts to
clarify positions so as to facilitate comprehension and enhance its
usefulness to the one consulting it. I hope that it will be a remedy
for the one plagued by doubts and an increase for the one who is
certain of what is correct. (para. 2)

The most plausible explanation for al-Khaffaf's reference to writing the work "in the mold of the Iraqis" is that he is referring to Iraqi Shafi[??]is, who, at this time, would have been represented by Ibn Surayj's students such as Ibn al-Qass (d. 335/946) and al-Sayrafi (d. 330/941f.). The only surviving comparable work is Ibn al-Qass's al-Talkhis, an abridgment of fiqh with a short introductory paragraph related to legal theory. (39) The overlap of al-Aqsam wa-l-khisal with this section suggests that al-Khaffaf was familiar with Ibn al-Qass's text.

The last lines of the above-quoted passage are unusually melodramatic for a work dedicated to law: jurists generally do not describe themselves and their kind as "plagued by doubts." However, in a curious instance of intertextuality, exactly the same sentence appears in a work on the argument from design composed by the Christian physician Jibril b. Nuh al-Anbari (fl. 240/850) for the Abbasid court. (40)

The style of the work is fiercely systematic, breaking down every term and process into a list of constituent properties (khisal). This drive to systematize is also evident in Ibn Surayj's text in, for example, his discussion of abrogation and of types of imperative, and it contrasts sharply with the style of al-Shafi[??]i's Risala, which is much less concerned with this kind of systematic comprehensiveness. The range of topics covered in al-Khaffaf's text is considerably wider than that discussed by Ibn Surayj, and it approaches the topical range of later works of legal theory. The overall structure of the treatise rests on the distinction between knowledge derived by reason, on the one hand, and knowledge derived through revelation, on the other. Reason imposes universal ethical imperatives (such as "Be grateful to your benefactor") and prohibitions (such as "Do not lie"), but most things are permissible when judged by reason alone, and it is these issues that are regulated through revelation as it can be gleaned by applying the tools of legal theory. Al-Khaffaf thus locates legal theory within an overall rational structure of ethical obligation. Furthermore, he appeals to rational knowledge throughout in the form of recurring non-contradiction clauses: a report has to be acted upon if, among other criteria, it does not violate reason (ghayr kharij min al-ma[??]qul), (41) and an imperative imparts obligation if, among other criteria, it is "good in itself and not deemed evil" (hasanan fi nafsihi ghayra mustaqbah). (42) This insistence that laws may not contradict reason emphasizes that both divine revelation and the applicability of the tools of legal theory are limited to an area delineated by a firm and epistemologically superior ethical framework of reason. By contrast, as noted earlier, the root [??]-q-l both in the Risala and in Ibn Surayj's text relates not to reason as an independent and universal set of prepositions but to the mental process of analogizing.

The Muqaddima covers the following topics:

Rational categories of legal knowledge: obligatory, permissible, and forbidden (wajib, mumtana[??], mujawwaz)

Rationally permissible issues that are judged on the basis of revelation, the sources for such judgment being the Quran, the Sunna, consensus, and analogy on the previous three

Sources of Quranic authority: inimitability (i[??]jaz) and concurrency (tawatur)

Reports (akhbar): from prophets, proven by miracles; concurrent reports; single-transmitter reports (akhbar ahad); the difference between witness statements (shahada) and reports (akhbar); disconnected reports (marasil, sg. mursal)

Different typologies of language, including haqiqa/majaz

Consensus (ijma[??])

Analogy (qiyas)

Definitions of knowledge, ignorance, truth, falsehood, fiqh, jurist (faqih), and student of law (mutafaqqih)

Imperative (amr) and its nature and legal ramifications

General ([??]amm)

Particular (khass)

Bases of Islam

Sources of differences of opinion in legal theory

Forms of clarification from the Prophet

Undifferentiated statements (mujmal)


Particularization (takhsis)

Negation (nahy)

Religious laws of previous communities (shara[??]i[??] man kana qablana)

Status before revelation (ashya[??] qabl maji[??] al-shar[??])

Causes ([??]ilal): rational and legal

Abrogation (naskh)

Disputation theory (jadal): etiquette, types of questions, and signs of defeat

Types of acceptable juristic preference (istihsan)

Blind following (taqlid)

As this list shows, al-Khaffaf's text contains elements, not addressed by al-Shafi[??]i or Ibn Surayj, that would subsequently become standard in legal-theoretical literature. At the same time, however, clear continuities with al-Shafi[??]i's Risala and Ibn Surayj's ideas are visible.

The only work that al-Khaffaf refers to explicitly in his legal theory discussion is al-Shafi[??]i's old (Iraqi) version of the Risala, on the topic of taqlid of the Companions. He quotes the first part of al-Shafi[??]i's statement ("It is not permitted to follow blindly anyone after the Prophet") and omits the second part ("except one of the Companions"). (43) However, the omission most likely happened in the process of transmission, since al-Khaffaf introduces this discussion by expressing an intention to mention cases of permissible taqlid. Also, something is clearly missing from this section, as the number of cases currently does not add up: al-Khaffaf promises to enumerate ten cases but delivers only eight or nine, depending on how one counts them. (44)

As noted earlier, several sections of the text display close parallels with the legal-theoretical introduction of Ibn al-Qass's al-Talkhis, which deals with blind following, juristic preference, and disconnected reports (taqlid, istihsan, and marasil). These parallels strongly suggest direct influence, but it is worth noting that the quotations that Ibn al-Qass gives from his teacher Ibn Surayj are missing in al-Khaffaf's text. (45) This omission suggests that, unlike Ibn al-Qass, al-Khaffaf was not, pace Temel, (46) a student of Ibn Surayj. The fact that the biographical works that generally provide such rich information on Ibn Surayj's students are largely silent on al-Khaffaf also supports this conclusion.

The topic of ijtihad receives similar treatment in the Muqaddima and the Risala: both works equate it with qiyas. In his discussion of bayan (the conveyance of the divine message through revelation), al-Khaffaf echoes al-Shafi[??]i's insistence on the importance of familiarity with the Arabic language and its idiomatic usage (lisan al-[??]arab) and then mentions different classifications of speech (aqsam al-kalam). This topic would become standard in later legal-theoretical literature. Similar classifications had already been proposed by linguists--students of Sibawayh (d. 180/796)--in the late second or early third century. (47) As one such possible classification, al-Khaffaf mentions the distinction between haqiqa and majaz (literal and figurative use of language). This is the earliest currently known mention of this juxtaposition in a text on legal theory. Importantly, although the distinction is still less prominent here than it would become in later texts, (48) it makes an appearance in al-Khaffaf's discussion on imperatives, where he defines the imperative as an utterance coined (lafza mawdu[??]a) to express an obligation--a clear reference to the idea of the primordial act of coinage (wad[??]) that enables the identification of literal versus figurative usage.

The topic of the imperative (amr) contains a clear innovation in the writing of Ibn Surayj, who argues that the actions of the Prophet constitute imperatives and that the default case for an imperative is that it confers an obligation. By contrast, al-Khaffaf considers prophetic precedent to mark an action as merely encouraged (mandub) until other evidence indicates that the action is in fact obligatory. The difference of opinion appears to originate in the above-mentioned theory of literal and figurative speech, with its concept of primordial coinage. According to this theory, to which al-Khaffaf subscribes, the imperative is a verbal form whose default meaning is to confer obligation. This does not apply to actions, because the connection between words and their meanings established during wad[??] has no parallel in the case of actions. Interestingly, later writers on legal theory reported that Ibn Surayj had also made use of the literal/figurative dichotomy by claiming that at least one word in the Quran was used in a non-literal way (majaz), contra the son of Dawud al-Zahiri, Muhammad al-Zahiri, who had said that while figurative usage characterized normal language, it was not found in the Quran. (49) It is difficult to judge from this secondhand report in what context Ibn Surayj asserted figurative usage, but his stance on the imperative indicates that even had he endorsed the idea of figurative language, he had not systematically integrated it into his legal theory.

The central concern of al-Shafi[??]i's Risala is to explain how God's will is clarified through revelation (the process of bayan). In contrast, al-Khaffaf's Khisal takes on the question of whether it is possible that revelation is not clear at every point of the revelatory process; in other words, whether clarification may, on occasion, be postponed (ta[??]khir al-bayan). Al-Khaffaf argues that such postponement was possible during Muhammad's prophetic mission, but only before it became necessary to act on the relevant piece of revelation, at which point clarification must have occurred. The question is not thematized in Ibn Surayj's text, but Abu Ishaq al-Shirazi attributes the same opinion to him. (50) This issue demonstrates how a concern of al-Shafi[??]i's (namely, bayan) was taken from a narrow legal-theoretical domain focused on epistemology in an ethical direction: accepting al-Shafi[??]i's claim that divine clarification happens through complex interactions of text (between Quranic verses, between hadith, or between verses and hadith) raised the possibility that there might be moments within the process of revelation in which obligations were not clear, as in the case of a general command that would eventually be particularized through another text revealed at a later stage.

Significant change over time on the subject of analogy can also be observed. While al-Shafi[??]i and Ibn Surayj describe qiyas as a method both of assimilating similar cases (shabah-based qiyas) and of identifying a legal reason (ma[??]na-based qiyas), al-Khaffaf classifies the former as a less precise form of analogy (calling it ghalabat al-ishtibah) and then introduces a form of analogy that he presents as scientifically rigorous. This he describes as a process of isolating the "legal cause" ([??]illa mustakhraja), a term whose precision contrasts with the more nebulous term "reason" (ma[??]na). (51)

The new terminology had significant methodological ramifications that immediately become clear in al-Khaffaf's extensive treatment of the rules for dialectics (jadal), which constitutes almost a fifth of the total text. Aristotelian dialectics, that is, rules for debating and deciding who has won a debate, appeared in Islamic thought by the mid-third/ninth century. (52) By the early fourth century other Shafi[??]is of al-Khaffaf's generation, including al-Qaffal al-Shashi (d. 365/976), Ibn al-Qass, Ibn Abi Hurayra (d. 345/956), and al-Sayrafi (d. 330/941f.), had composed independent treatises on jadal in law as possibly the first jurists to adopt this originally theological methodology. (53)

As I have described elsewhere, juristic jadal was applied first and foremost in the process of analogy and especially in the verification of the legal cause. (54) By the nature of the jadal method, the verification process was formalistic. The primary way of arguing for a legal cause was to demonstrate its consistency (tard, i.e., whenever the cause is present the legal qualification is present) and convertibility ([??]aks, i.e., whenever the cause is absent the legal qualification is absent). As Aron Zysow has argued, the fascination of early Shafi[??]is with consistency and convertibility in the ascertainment of legal causes probably sprang from the application of the standards of rational causes to legal ones. (55) The terminological shift from underlying reason (ma[??]na) in the writings of al-Shafi[??]i and Ibn Surayj to legal cause ([??]illa) in the works of al-Khaffaf and legal theorists after him was intimately connected to the new emphasis on jadal: a cause, unlike a "reason," could best be verified through the formalistic means developed for the ascertainment of rational causes--that is, through jadal. In addition, the adoption of the technical term [??]illa and the concomitant methodology of jadal probably served to defend the validity of analogy and the robustness of its results against its critics, who considered analogy to represent a subjective human imposition on the divine law. (56)

Finally, al-Khaffaf's Muqaddima is distinguished from both Ibn Surayj's Wada[??]i[??] and al-Shafi[??]i's Risala by the near absence of Quranic citations and prophetic hadith. Whereas al-Shafi[??]i, al-Buwayti, and Ibn Surayj still consistently justified their methods by appeal to scriptural proof texts, al-Khaffaf does not seem to feel a need to anchor and verify his theory through constant reference to scripture. This may be a reflection of al-Khaffaf's rationalism (which leads him to present his legal-theoretical tools as universal and not tied to scripture), or it may indicate a maturation of legal theory as a field of inquiry in which basic topics can be treated as axiomatic and no longer in need of justification.


My broad comparison of these two texts, probably produced close to (if on opposite sides of) the hijri year 300 (912 C.E.), yields a number of interrelated results. First, it indicates that already by this time Muslim jurists accepted as given an established discourse of legal theory--at this stage called usul al-din, not yet usul al-fiqh. Second, for these jurists the discourse encompassed al-Shafi[??]i's old Risala, authored in the late second/eighth century. Third, the legal-theoretical discourse was not carried on exclusively in the form of dedicated monographs, but also in introductions or conclusions to works of positive law. And, fourth, a gradual development from al-Shafi[??]i to the classical fourth-/tenth-century texts on legal theory can be discerned. Ibn Surayj neither founded legal theory independently of al-Shafi[??]i's Risala nor adopted al-Shafi[??]i's ideas without modification. Similarly, al-Khaffaf and his contemporaries, including the generation of Ibn Surayj's students, neither copied Ibn Surayj's ideas nor inaugurated a new field from scratch. Rather, they drew consciously on a disciplinary tradition going back to al-Shafi[??]i and developed it by expanding its range of topics and modifying previous positions according to ideas newly integrated from other fields. (57) As part of this process, they introduced a keen attention to delineating the relationship between reason and revelation, (58) reflecting an influx of theological concerns, particularly ethics and natural philosophy, (59) as well as theories of language. (60)

The title of this article is not meant to suggest that these texts represent a bridge between two distinct works or types of works. Rather, the results of my study strongly suggest that the alleged gap between the Risala and the classical usul al-fiqh tradition is primarily a phenomenon of perception. With the recent discovery of several early legal-theoretical writings, we now have extant texts on the subject from every generation between al-Shafi[??]i and al-Jassas. Instead of a gap between the Risala and the classical usul al-fiqh tradition, the present evidence paints a picture of change and expansion within a continuous discourse.


The paragraph numbering of the following editions is mine. The numbers in parentheses refer to pages of the Duwaysh M.A. thesis and to the folios of the Aya Sofya 1502 manuscript of Ibn Surayj's text, and to folios of the Chester Beatty manuscript of al-Khaffaf's text. Where there are dots within parentheses indicating a lacuna, each dot stands for approximately one word to give a rough indication of the lacuna length.



I would like to thank Bilal Aybakan, Peri Bearman, and two anonymous reviewers, as well as workshop and conference participants in New York, Boston, and Istanbul, for their comments on earlier versions of the manuscript. I am grateful to Najah Nadi for proofreading the Arabic texts.

(1.) Thus, I do not intend to write a comprehensive history of tenth-century usul, while a comparison with the Hanafi usul tradition, however helpful it will eventually be, would require both another article and a reliance on preliminary data such as these.

(2.) Notably, al-Fusul fi l-usul of al-Jassas (d. 370/980), al-[??]Udda fi usul al-fiqh of Abu Ya[??]la (d. 458/1066), and al-Taqrib wa-l-irshad of al-Baqillani (d. 403/1013).

(3.) G. Makdisi, "The Juridical Theology of Shafi[??]i: Origins and Significance of Usul al-Fiqh," Studia Islamica 59 (1984): 5-47.

(4.) W. B. Hallaq, "Was al-Shafi[??]i the Master Architect of Islamic Jurisprudence?" International Journal of Middle East Studies 25 (1993): 587-605.

(5.) H.-T. Tillschneider, Die Entstehung der juristischen Hermeneutik (usul al-fiqh) im fruhen Islam (Wurzburg: Ergon, 2006).

(6.) D. R. Vishanoff, The Formation of Islamic Hermeneutics: How Sunni Legal Theorists Imagined a Revealed Law (New Haven: American Oriental Society, 2011).

(7.) A. Temel, "The Missing Link in the History of Islamic Legal Theory: The Development of Usul al-Fiqh between al-Shafi[??]i and al-Jassas during the 3rd/9th and Early 4th/10th Centuries" (Ph.D. diss., Univ. of California at Santa Barbara, 2014).

(8.) D. Stewart, "Muhammad b. Da[??]ud al-Zahiri's Manual of Jurisprudence, al-Wusul ila ma[??]rifat al-usul," in Studies in Islamic Legal Theory, ed. B. G. Weiss (Salt Lake City: Univ. of Utah Press, 2002), 99-158; D. J. Stewart, "Muhammad b. Jarir al-Tabari's al-Bayan 'an usul al-ahkam and the Genre of Usul al-Fiqh in Ninth-Century Baghdad," in Abbasid Studies: Occasional Papers of the School of Abbas id Studies, Cambridge, 6-10 July 2002, ed. J. E. Montgomery (Leuven: Peeters, 2004), 321-49.

(9.) A. El Shamsy, The Canonization of Islamic Law: A Social and Intellectual History (New York: Cambridge Univ. Press, 2013), chap. 8; see also A. El Shamsy and A. Zysow, "Al-Buwayti's Abridgment of al-Shafi[??]i's Risala: Edition and Translation," Islamic Law and Society 19.4 (2012): 327-55.

(10.) See, for example, C. Melchert, The Formation of the Sunni Schools of Law, 9th-I0th Centuries C.E. (Leiden: Brill, 1997), 68-86, and following him, K. S. Vikor, Between God and the Sultan: A History of Islamic Law (Oxford: Oxford Univ. Press, 2005), 100-101.

(11.) Hallaq, "Was al-Shafi[??]i the Master Architect," 596: "Although there is no evidence to indicate that Ibn Surayj wrote a complete work on usul al-fiqh, he seems to have assimilated all teachings on the subject from within and without the Shafi[??]ite school." Hallaq's claim has been repeated by, for example, Gregor Schwarb in "Capturing the Meanings of God's Speech: The Relevance of Usul al-Fiqh to an Understanding of Usul al-Tafsir in Jewish and Muslim Kalam," in A Word Fitly Spoken: Studies in Mediaeval Exegesis of the Hebrew Bible and the Qur[??]an, ed. M. M. Bar-Asher et al. (Jerusalem: Ben Zvi Institute, 2007), 111-56 at 116-17.

(12.) A short creed attributed to Ibn Surayj has been published as Juz[??] fihi ajwiba fi usul al-din, ed. W. al-[??]Ali (Beirut: Dar al-Basha[??]ir al-Islamiyya, 2006); the text is also included in full in Ibn Qayyim al-Jawziyya's Ijtima[??] al-juyush al-islamiyya fi ghazw al-mu[??]attila wa-l-jahmiyya (Amritsar: Matba[??]at al-Qur[??]an wa-l-Sunna, 1870, and countless reprints and new editions since). Temel ("Missing Link," 23-24) argues that the attribution is probably incorrect. In 1993 Hallaq ("Was al-Shah[??]i the Master Architect," 595) asserted that none of Ibn Surayj's works has survived. In 1997 Melchert (Formation, 90 n. 11) mentioned the existence of a manuscript attributed to Ibn Surayj (incorrectly, as it turns out--see below).

(13.) Badr al-Din al-Zarkashi, al-Bahr al-muhit, ed. [??]A. [??]A. al-[??]Ani et al., 6 vols. (Kuwait: Wizarat al-Awqaf wa-l-Shu[??]un al-Islamiyya, 1992), 1: 204; 2: 256, 312; 4: 110, 201, 516.

(14.) Istanbul: Suleymaniye, Aya Sofya 1502 (127 fols., copied 591/1195); Rabat: al-Maktaba al-Wataniyya li-l-Mamlaka al-Maghribiyya, [??]Abd al-Hayy al-Kattani collection, part of majmu[??]a 250.

(15.) S. b. [??]A. b. I. al-Duwaysh, "al-Wada[??]i[??] li-mansus al-shara[??]i[??]" (M.A. thesis, 705 pages, Imam Muhammad b. Sa[??]ud Univ., Saudi Arabia, 1989).

(16.) H. al-Juburi, al-Imam Abu l-[??]Abbas ibn Surayj wa-ara[??]uhu al-usuliyya (Mecca: Matba[??]at al-Safa, 1994).

(17.) Al-Duwaysh, "Wada[??]i[??]," 85; Aya Sofya 1502, fol. lb.

(18.) Al-Shafi[??]i, Kitab al-Umm, ed. R. F. [??]Abd al-Muttalib, 11 vols. (Mansura: Dar al-Wafa[??], 2001), 8: 615-16 (using the same hadith regarding nursing that Ibn Surayj also uses).

(19.) Al-Shafi[??]i, al-Risala, ed. A. M. Shakir (Cairo: al-Babi al-Halabi, 1940), e.g., paras. 375-76. The earliest exposition of the full tripartite classification that I have found is in Abu [??]Ubayd al-Qasim b. Sallam's (d. 224?/838f.) al-Nasikh wa-l-mansukh, ed. M. al-Mudayfar (Riyadh: Maktabat al-Rushd), 14-15. He promises three categories but mentions only two, hukm duna l-tilawa and hukm together with tilawa; however, the latter is followed by a hiatus in the manuscript (which John Burton overlooks in his introduction to Abu [??]Ubaid al-Qasim b. Sallam's "Kitab al-nasikh wa-l-mansukh" [Cambridge: E. J. W. Gibb Memorial Trust, 1987], 19), and it seems plausible that this would have contained a reference to tilawa duna l-hukm.

(20.) Al-Shafi[??]i, Risala, para. 314.

(21.) Al-Shafi[??]i, Risala, para. 321.

(22.) Muhammad b. Nasr al-Marwazi, al-Sunna, ed. [??]A. al-Busayri (Riyadh: Dar al-[??]Asima, 2001).

(23.) Al-Shafi[??]i, Risala, para. 1140.

(24.) Al-Shafi[??]i, Risala, para. 65.

(25.) Al-Shafi[??]i, Risala, para. 1334.

(26.) See J. E. Lowry, Early Islamic Legal Theory: The Risala of Muhammad ibn Idris al-Shafi[??]i (Leiden: Brill, 2007), 150; al-Shafi[??]i, al-Risala, para. 125.

(27.) Ibn Surayj, Wada[??]i[??], para. 5.

(28.) Al-Shafi[??]i, Risala, paras. 125 and 1334.

(29.) El Shamsy and Zysow, "Al-Buwayti's Abridgment," para. 14. The last part of al-Buwayti's Mukhtasar, including the section on the Risala, has been (poorly) edited by [??]Abd al-Nasir Shahawi (M.A. thesis, 583 pages, al-Azhar University, Damanhur branch, 2007); the whole Mukhtasar has been edited by Ayman al-Salama (M.A. thesis, 1,252 pages, University of Medina, 1431/2010) and edited and published by [??]Ali al-Qarah-Daghi (Jeddah: Dar al-Minhaj, 2015).

(30.) Al-Qadi Nu[??]man, Ikhtilaf usul al-madhahib (Beirut: Dar al-Andalus, 1984), 177. I am grateful to Devin Stewart for this reference.

(31.) See al-Shirazi, Sharh al-Luma[??], ed. [??]A. al-M. Turki, 2 vols. (Beirut: Dar al-Gharb al-Islami, 1988), 1: 546.

(32.) For a historical contextualization of this position, see Temel, "Missing Link," 198-99. Al-Zarkashi (al-Bahr al-muhit, 4: 516) framed the position as referring to a situation in which only one mujtahid existed.

(33.) Dublin: Chester Beatty Library, MS Arabic 5115 (43 fols., copied 660/1262).

(34.) Hajji Khalifa, Kashf al-zunun, 2 vols. (Baghdad: Maktabat al-Muthanna, 1972), 1: 705.

(35.) Al-Shirazi, Tabaqat al-fuqaha[??], ed. I. [??]Abbas (Beirut: Dar al-Ra[??]id al-[??]Arabi, 1970), 114.

(36.) Hajji Khalifa, Kashf al-zunun, 2: 1416.

(37.) E.g., al-Zarkashi, al-Bahr al-muhit, 2: 378; 3: 494; 4: 91, 420; 5: 136; 6: 95, 110; al-Subki, al-Ashbah wa-l-naza[??]ir, 2 vols. (Beirut: Dar al-Kutub al-[??]Ilmiyya, 1991), 2: 304; Ibn al-Mulaqqin, al-Badr al-munir, ed. M. Abu l-Ghayt et al., 9 vols. (Riyadh: Dar al-Hijra, 2004), 1: 662-63.

(38.) Ibn Qadi Shuhba, Tabaqat al-shafi[??]iyya, 4 vols. (Hyderabad: Da[??]irat al-Ma[??]arif al-[??]Uthmaniyya, 1978), 1: 95.

(39.) Ibn al-Qass, al-Talkhis, ed. [??]A. A. [??]Abd al-Mawjud and [??]A. M. Mu[??]awwad (Mecca: Maktabat Nazzar Mustafa al-Baz, 1999), "Bab al-taqlid wa-l-istihsan wa-l-marasil [??]ala madhhab al-Shafi[??]i," 73-75.

(40.) Manuscript (Istanbul: Aya Sofya 4836, fols. 160a-187b), at 161a. Compare lines 1-3 with al-Khaffaf, Muqaddima, para. 2.

(41.) Al-Khaffaf, Muqaddima, para. 4.

(42.) Al-Khaffaf, Muqaddima, para. 11.

(43.) See Ibn al-Qass, Talkhis, 74.

(44.) Muqaddima, para. 39; a longer text from the old Risala on the authority of the Companions is reproduced by al-Bayhaqi in his Manaqib al-Shafi[??]i, ed. S. Saqr, 2 vols. (Cairo: Dar al-Turath, 1970), 1: 442-43.

(45.) Compare Ibn al-Qass, Talkhis, 73-75, with Muqaddima, paras. 38 (istihsan, marasil), 39 (taqlid).

(46.) Temel, "Missing Link," 23.

(47.) A. Zysow, "If Wishes Were... : Notes on Wishing in Islamic Texts," in Classical Arabic Humanities in Their Own Terms: Festschrift for Wolfhart Heinrichs on His 65th Birthday, ed. B. Gruendler (Leiden: Brill, 2008), 521-67, at 522-23.

(48.) See R. Gleave, Islam and Literalism: Literal Meaning and Interpretation in Islamic Legal Theory (Edinburgh: Edinburgh Univ. Press, 2012).

(49.) See al-Shirazi, Sharh al-Luma[??], 170.

(50.) Al-Juburi, Ibn Surayj, 171.

(51.) The appearance of the term [??]illa in al-Khaffaf's text supports Aron Zysow's hypothesis (The Economy of Certainty: An Introduction to the Typology of Islamic Legal Theory [Atlanta, GA: Lockwood Press, 2013], 255-56) that the term entered legal-theoretical use in the fourth/tenth century.

(52.) On jadal, see L. Miller, "Islamic Disputation Theory: A Study of the Development of Dialectic in Islam from the Tenth through Fourteenth Centuries" (Ph.D. diss., Princeton Univ., 1984), and W. E. Young, The Dialectical Forge: Juridical Disputation and the Evolution of Islamic Law (Cham, Switzerland: Springer, 2016).

(53.) For the writings, see al-Zarkashi, al-Bahr al-muhit, 1: 149 (Ibn al-Qass), 5: 136 (Ibn Abi Hurayra), and 5: 364 (al-Sayrafi); and al-Shirazi, Tabaqat al-fuqaha[??], 112 (al-Qaffal al-Shashi).

(54.) A. El Shamsy, "The Wisdom of God's Law: Two Theories," in Islamic Law in Theory: Studies on Jurisprudence in Honor of Bernard Weiss, ed. A. K. Reinhart and R. Gleave (Leiden: Brill, 2014), 19-37, at 29-30.

(55.) Zysow, Economy of Certainty, 220.

(56.) These were both Zahiri and Shi[??]i thinkers; see Stewart, "Muhammad b. Da[??]ud al-Zahiri's Manual," 114, 117. In the Muqaddima (para. 10), al-Khaffaf explicitly mentions Zahiris as accepting a fortiori arguments (fahwa l-khitab), which defenders of analogical reasoning considered a kind of qiyas.

(57.) Abu l-Hasan al-Ash[??]ari (d. 324/935f.) also drew on al-Shafi[??]i's Risala and Ibn Surayj's reception of it; see Ibn Furak, Mujarrad maqalat al-Ash[??]ari, ed. D. Gimaret (Beirut: Dar al-Mashriq, 1987), 193. For examples, see pp. 192, 201.

(58.) As Hallaq already suggested in "Was al-Shafi[??]i the Master Architect," 598.

(59.) As has been argued by Makdisi ("Juridical Theology of Shafi[??]i," 35) and Kevin Reinhart (Before Revelation: The Boundaries of Muslim Moral Thought [Albany: State Univ. of New York Press, 1985], 15-16).

(60.) This was pointed out by Ibn Taymiyya in Majmu[??]at al-fatawa, ed. [??]A. al-Jazzar and A. al-Baz, 37 vols. (Mansura: Dar al-Wafa[??], 1998), 20: 220-73.
COPYRIGHT 2017 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Shamsy, Ahmed El
Publication:The Journal of the American Oriental Society
Article Type:Critical essay
Geographic Code:1USA
Date:Jul 1, 2017
Previous Article:(bn) agptr / (Binu) Agaptarri's House: A Functional Analysis of an Ugaritic "Archive" (PH Room 10).
Next Article:Maimonides and the Habitus Concept.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters