Author's Note: This article is dedicated to Wendell “Gumby” Martin.

¹ al-Shāfiʿī, Muḥammad b. Idrīs, al-Risāla, ed. Shākir, Aḥmad Muḥammad (Beirut: al-Maktaba alʿIlmiyya, n.d.), 88Google Scholar.

² According to al-Shafiʿi, there were three types of Prophetic action: (1) where the Prophet executed explicit injunctions from the Qurʾan, (2) where he clarified or specified more general injunctions from the Qurʾan, and (3) where he instituted practices independent of any Qurʾanic injunction. According to al-Shafiʿi, no one disagreed that the first two types were authoritative. Disagreement obtained only regarding the third. See al-Risala, 91–92. It was the authority of actions of the third type that al-Shafiʿi would go on to defend.

³ See, for example, Abū al-Ḥasan ʿAli b., ʿUmar (d. 909), Muqaddimafl Uṣūl al-Fiqh (Arabic MSS [170] 5786, catalogue no. 2, al-Azhar collection), fols. 6r, 9rGoogle Scholar; al-Baṣri, Abś al-ḥusayn (d. 1044), Kitāb al-Muʿtamad fi Uṣūl al-Fiqh, 2 vols., ed. al-Mays, Khalīl (Beirut: Dār al-Kutub al-ʿllmiyya, 1403 A.H.), 1:343–59Google Scholar; ʿAbd Allāh al-Juwaynī, ʿAbd al-Mālik b. (Imām al-Haramayn) (d. 1085), al-Burhān fī Uṣūl al-Fiqh, 2 vols., ed. al-Dīb, ʿAbd al-ʿAẓim (Cairo: Dār al-Anṣāar, 1400 A.H.), 1:487—502Google Scholar; al-Ghazālī, Abū Ḥāmid, al-Mustaṣfā min ʿllm al-Uṣūl, 2 vols., ed. al-Shakūr, Muṣibb al-Dīn ʿAbd (Bulaq: al-Maṭbaca al-Amiriyya, 1322 A.H.), 1:129–32Google Scholar; ibid., 2:212–26; al-Āmidī, Sayf al-Dīn (d. 1233), al-Iḥkāmfī Uṣūl al-Aḥkām, 4 vols. (Cairo, n.d.), 1:155–73Google Scholar

⁴ In the period immediately following al-Shafiʿī's death, there was, for example, the infamous Inquisition (miḥna), which lasted for fifteen years (833–48) and involved the attempt by four successive Abbasid caliphs—al-Maʾmun, al-Muʿtasim, al-Wathiq and al-Mutawakkil (first two years)—to impose as orthodoxy the doctrine of the createdness of the Qurʾan (khalq al-Qurʾan). See Makdisi, G., The Rise of Colleges (Edinburgh: Edinburgh University Press, 1981), 6–7Google Scholar. Earlier, there had been the failed attempt by the Abbasid secretary of state, Ibn al-Muqaffaʿ (d. 756) to have the caliph codify the law, which would have effectively given the latter law-making powers. See Schacht, J., An Introduction to Islamic Law (Oxford: Clarendon Press, 1964), 55–56Google Scholar. Also, there was the ideological struggle between Sunnis and Shiʿites over the concept of the imamate, the latter holding the imam to be infallible (maʿṣūm). This struggle is reflected in the early writings of such diverse figures as Ibn Qutaybah, al-Jahiz, al-Baqilani, al-Baghdadi, al-Juwayni, and al-Ghazali. See Lambton, A. K. S., State and Government in Medieval Islam (Oxford: Oxford University Press, 1981), 65–66, 58–64, 70–71, 73 ff., 77–82, 105–6Google Scholar, respectively. On al-Ghazali, see, for example, “al-Qisṭās al-Mustaqīm,” in al-Quṣūr al-ʿAwālī, ed. al-ʿAlla, Muḥammad Muṣṭafā Abū (Cairo: Maktabat al-Jundi, n.d.), 15 ff., 55 ffGoogle Scholar.

⁵ Yaʿlā, Abū (Muḥammad b. al-Ḥusayn b. Khalaf b. Aḥmad b. al-Farrāʾ), al-Aḥkām al-Sulṭāniyya, ed. al-Fiqī, Muḥammad Ḥamid (Beirut: Dār al-Kutub al-ʿIlmiyya, 1404 A.H.), 19–34Google Scholar; ʿAlī b., Abū al-ḥḤasan Muḥammad b. Ḥabīb al-Māwardī, al-Aḥkām al-Sulṭāniyya, ed. ʿAbd al-Qādir, Muḥammad (Bulaq: Matbaʿat al-Waṭan, 1298 A.H.), 3–28Google Scholar.

⁶ Lambton, , State and Government, 70–77Google Scholar.

⁷ ibid., 77–82.

⁸ Abū al-Ḥasan, Ismāʿīl al-Ashʿarī, ʿAlī b., Maqālāt al-lslāmiyyīn wa-Ikhtilāf al-Muṣallīn, 2 vols., ed. al-Majīd, Muḥammad Muḥyī al-Dīn ʿAbd (Cairo: Maktabat al-Nahḍa al-Miṣriyya, 1389 A.H.), 1:48, 50Google Scholar and passim; ibid., 2:141–45, 148–54, 155, 157–58, and passim.

⁹ Muḥammad, AbūAḥmad b. Saʿīd b. Hazm, ʿAlī b., al-Faṣl fī al-Milal wa-al Ahwāʾ wa-al-Nihạl, 5 vols. (Cairo: Maktabat al-Salām al-ʿĀlamiyya, n.d.), 4:72–90Google Scholar.

¹⁰ The latter appears to be the basis of al-Juwayni's theory in his Ghiyāth al-Umam fī Iltiyāth al-Ẓulam, as discussed by Hallaq, Wael, “Caliphs, Jurists and the Saljūqs in the Political Thought of Juwayni,” The Muslim World 74, 1 (1984): 26–41CrossRefGoogle Scholar.

¹¹ Lambton, , State and Government, 63–64Google Scholar.

¹² There are two printed editions of this work. The first was done by Maḥmūd Arnus and appeared in 1938 (Cairo: Maṭbaʿat al-Anwār, 1358 A.H.). It was done on the basis of a single, poorly preserved manuscript from the National Library in Egypt. It contains a number of corruptions and is unreliable. The second was done by ʿAbd al-Fattāḥ Abū Ghudda and appeared in 1967 (Aleppo: Maktab al-Maṭbūʿāt al-Islāamiyya, 1387 A.H.). Abu Ghuddah's is a fine edition, done on the basis of four different manuscripts and including some helpful notes and cross-references. There is a fifth manuscript of the work in the Yahudah collection (no. 86, shelf no. 488) of the Princeton University library. I am thankful to Dr. Chris Taylor of Drew University for his kind assistance in obtaining a copy of this manuscript. All references in the present paper are to the Abu Ghuddah edition under the abbreviated title, Tamyīz. Also, in order to reduce the number of footnotes, I will use the symbol “T,” followed by page number in the body of the paper to refer to quotations from the Tamyḥz.

¹³ None of the medieval biographers give a date of birth. The year 1228 is given by the modern biographer, al-Baghdadi, Ismaʿil Basha, Hadīyat al-ʿĀrifīn Asmaʾ al-Mu-ʾallifin wa-Āthār al-Muṣannifīn, 2 vols. (Istanbul, 1951), 1:99Google Scholar, but this date is suspect. According to Brockelman, al-Qarafi composed a work, al-Istibṣār fīmā Tudrikuhu al-Abṣār, which was reportedly “prompted by eighty seven questions from the ruler of Sicily (Frederick II?) to al-Malik al-Kamil.” Brockelmann, C., Geschichte der arabischen Litteralur (Suppl.), 3 vols. (Leiden: E. J. Brill, 1937–1942), SI, 665Google Scholar. As al-Kamil died in 1238, a birth date of 1228 would mean that al-Qarafi composed this work before the age of ten.

¹⁴ See Farhun, Ibn, al-Dībāj al-Mudhahhab fī Maʿrifat Aʿyān ʿVlamaʾ al-Madhhab (Beirut: Dār al-Kutub al-ʿIlmiyya, n.d.), 66Google Scholar.

¹⁵ Ibid., 62.

¹⁶ Ibid., 65.

¹⁷ Ḥusn al-Muḥāḍara fī Tārīkh Miṣr wa-al-Qāhira, 2 vols., ed. Ibrāhḥm, Muḥammad Abū al-Faḍl (Cairo: Īsā al-Bābī al-Ḥalabī, 1387 A.H.), 1:316 ffGoogle Scholar.

¹⁸ In addition to the traditional religious sciences, al-Qarafi was also adept in mathematics, astronomy, and a number of other disciplines—even magic. In one of his works, al-Ajwiba al-Fākhira ʿan al-Asʾila al-Fājira, a refutation of certain Jewish but especially Christian charges, he flaunts his knowledge of Hebrew, quoting (in Arabic script) and translating the Old Testament. He is reported to have written in excess of twenty works, the most important of which are al-Furūq (on legal precepts), al-Dhakhira (his opus on Maliki law), Sharḥ Tanqīḥ al-Fuṣūl (an important commentary on Fakhr al-Din al-Razi's al-Maḥṣūl on uṣūl al-fiqh), and the Kitāb al-Iḥhkām fī Tamyīz al-Fatāwā ʿan al-Aḥkām wa Taṣarrufāt al-Qāḍī wa al-lmām.

¹⁹ Farhun, , Dībāj, 66Google Scholar.

²⁰ The Tamyīz is divided into forty questions. Marvelling at al-Qarafi's response to this particular question, the editor of the 1938 edition wrote in his introduction: “I speak indeed the truth when I say that this book is uncommon and unique in its [treatment of its] subject. Indeed, it contains a number of issues never before contemplated by any other scholar. And the reader will find confirmation of this statement when he reads his [al-Qarafi's] response to Qu. #25;….”

²¹ I use the terms “messengership” and “prophethood” to differentiate respectively between risāla and nubūwwa, the two types of prophethood cited by al-Qarafi. According to al-Qarafi, nubūwwa occurs when an individual receives from God revelation embodying a command or countermand specific to that person. Risala, on the other hand, occurs when the revelation received includes the obligation to communicate something to others. Thus, in the case of the Prophet Muhammad, nubūwwa occurred when he received the verses, “Read, in the name of thy Lord” (96:1). Then, when he received, “O you wrapped in sheets: Rise and warn…” (74:1), he became a rasūl (T.92–93). It is interesting that, according to al-Qarafi, in order for a person to become a nabī the revelation received must embody a command or countermand (e.g., “Read!”). On this understanding, Mary, the mother of Jesus to whom the angel Jibril revealed, “I am but a messenger [sent by] your Lord that He may give you a pure lad” (19:19), was, according to al-Qarafi, not a prophetess (nabīyya) (T.91). It seems, however, that al-Qarafi overlooked such verses as, “And shake towards yourself the trunk of the date palm; fresh dates shall fall before you in abundance” (19:25), and, “So cast him into the basket, then cast it into the sea” (20:39), in the case of the mother of Moses.

²² See Fish, Stanley, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Durham, N.C.: Duke University Press, 1989), 1–3CrossRefGoogle Scholar.

²³ For a brief summary of what legal theoreticians refer to as “descriptive rulings” (al-aḥkām al-waḍʿiyya), which include legal causes (sing, sabab), legal rules (sing, ḥukm), legal impediments (sing. māniʿ), and legal prerequisites (sing, sharṭ), see Jackson, S., “In Defense of Two Tiered Orthodoxy: A Study of Shihab al-Dīn al-Qarāfī's Kitāb al-Iḥkām fī Tamyīz al-Fatāwā ʿan al-Aḥkām wa-Taṣarrufāt al-Qāḍī wa-l-Imam” (Ph.D. diss., University of Pennsylvania, 1991), 141–44Google Scholar.

²⁴ This does not apply of course in the case of the Prophet, whose fatwa was, again, an extension of his function as God's infallible messenger.

²⁵ In the Tamyīz, 30–31, al-Qarafi states: “The judge relies upon courtroom evidence (ḥijāj; sing. ḥujja), while the mufti relies upon scriptural evidence (adilla; sing, dalīl). And the mufti does not rely upon courtroom evidence but rather upon scriptural evidence. And scriptural evidence includes the Qurʾan, the Sunna, and the like, while courtroom evidence includes testimonial proof (bayyina), confession and the like.” For more on the function of the judge according to al-Qarafi, see Jackson, “Defense,” 206–8.

²⁶ This appears to be a justification for the office and authority of the sultan, even in cases where a caliph exists.

²⁷ It will be noticed that these examples involve statements (aqwāl; sing, qawl) by the Prophet to the exclusions of actions (afʿai, sing. fiʿl). Al-Qarafi makes no apparent distinction between these two (at least not in terms of the application of his method). This is brought out clearly in an example given below involving Abu Bakr's act of enslaving the captives of Banu Hanifa during the wars of apostacy. Al-Qarafi refers to this action as a fatwa.

²⁸ Al-Shafiʿi, for example, who is considered the founder of uṣūl al-fiqh begins his al-Risāla with “Kitab al-bayān,” which in essence deals with formal linguistic conventions. See al-Risāla, 21 ff. In the uṣūl works of the period after al-Shafiʿi, this formalism takes on a more general application, as may be seen for example in al-Baṣrī's, Abū al-Ḥusayn, Kitāb al-Muʿtamad fī Uṣūl al-Fiqh, 2 vols., ed. al-Mays, Khalīl (Beirut: Dār al-Kutub al-ʿIlmiyya, 1983), 1:50–75Google Scholar. Later, in the 11th and 12th centuries, al-Ghazali would observe that bayān (clarification) “may also be realized through actions, gestures, and symbols, since these too are indicators and clarifiers. But according to the custom of the mutakallimūn [which includes the bulk of legal theoreticians] it has come to be specific to what is indicated by words….” See al-Ghazali, Abū Ḥāmid, al-Mustaṣfā min ʿIlm al-Uṣūl, 2 vols. (Bulaq: Amīriyya Press, 1322 A.H.), 1:366Google Scholar.

²⁹ al-Baṣrī, Al-Muʿtamad, 1:98 ffGoogle Scholar.

³⁰ See Skinner, Q., “Meaning and Understanding in the History of Ideas,” History and Theory 8 (1969): 47CrossRefGoogle Scholar. Emphasis Skinner's.

³¹ On the distinction between meaning and understanding, see Skinner, , “Meaning and Understanding,” 1–53Google Scholar. Among the points Skinner makes is that in order to understand an utterance one must understand also the illocutionary force behind it, which entails understanding not only the individual words and the context in which they are uttered but also the intention of the utterer in making the utterance. According to Skinner,

[The] question about what a given agent may be doing in uttering his utterance is not a question about meaning at all, but about a force co-ordinate with the meaning of the utterance itself, and yet essential to grasp in order to understand it. And … even if we could decode what a given statement must mean from a study of its social context, it follows that this would still leave us without any grasp of its intended illocutionary force and so eventually without any real understanding of the given statement at all. The point is, in short, that an unavoidable lacuna remains: even if the study of the social context of texts could serve to explain them, this would not amount to the same as providing the means to understand them.

Ibid., 46 (emphasis Skinner's).

³² Interestingly, this became one of the methods employed by 20th-century modernists in their quest for Islamic reform. For example, in an effort to offset the unilateral right of husbands to divorce their wives, Jordan, under the Law of Family Rights of 1951, stipulated that no plea of divorce could be heard from a husband (who, for example, claimed that he had divorced a woman who was now seeking maintenance) unless the divorce had been properly registered before a qadi. This measure was in effect a move toward making divorce a judicial proceeding. See Coulson, N. J., A History of Islamic Law (Edinburgh: Edinburgh University Press, 1964), 173–74Google Scholar.

³³ al-Sarkhasī, Shams al-Dīn, al-Mabsūṭ, 30 vols. (Beirut: Dār al-Maʿrifa, n.d.)Google Scholar. This work catalogues the views of all three “founders” of Hanafi fiqh: Abu Hanifa, Abu Yusuf, and Muhammad b. al-Hasan al-Shaybani. Its significance in later times is reflected in the statement of the 15th-century Hanafi jurist, ʿAlaʾ al-Din al-Tarabulusi (d. 1440): “Whoever memorizes al-Mabsūṭ and the doctrine of the ancient scholars becomes thereby a mujtahid.” See al-Tarabulusī, ʿAlāʾ al-Dīn, Muʿīn al-ḥukkām fīmā Yataraddadu bayna al-Khaṣmayn min al-Aḥkām (Cairo: Muṣṭafa al-Bābī al-Ḥalabī, 1393 A.H.), 26Google Scholar.

³⁴ al-Sarkhasi, , al-Mabsūṭ, 23:167Google Scholar.

³⁵ Ibid.

³⁶ al-Tānūkhī, Sahnūn b. Saʿīd, al-Mudawwana al-Kubrā, 4 vols. (Beirut: Dār al-Fikr, 1406 A.H.), 4:377Google Scholar.

³⁷ al-Shāfiʿī, Muḥammad ibn Idrīs, al-Umm, 7 vols. (Cairo: Kitāb al-Shaʿb, n.d.), 7:130Google Scholar.

³⁸ See Jackson, , “Defense,” 56Google Scholar.

³⁹ I am indebted to Dr. Adel Allouche for his valuable insights in suggesting that this event was probably of major concern for al-Qarafi and that it possibly informed his thought concerning the caliphate.

⁴⁰ See Jackson, , “Defense,” 32–39Google Scholar, where I indicate that the Shafiʿis were the most powerful madhhab in Egypt during this period, whereas the Malikis were fast losing ground to the Hanafis.

⁴¹ For more on al-Qarafi's “Pure Law” doctrine, see ibid., 139–55.

⁴² For more on judges and obiter dictum, see ibid., 193–95.

⁴³ Jamāl al-Dīn Abū al-Faraj ʿAbd al-Raḥmān ʿAlī b. al-Jawzī, Taʿẓīm al-Futyā, Chester Beatty Library, Arabic MSS. no. 3829.

⁴⁴ In other words, one who had received a license (ijāza) to issue legal opinions. This license was a personal authorization from the individual master jurisconsult to the deserving candidate. Prior to being awarded this license, a candidate would pass through a number of other steps, the final one being taṣaddur, that is, “going to the head of the class.” Sometimes a student would set himself up to teach or give legal opinions prematurely, and this bad practice was also referred to as taṣaddur. Ibn al-Jawzi thus warns his students in the Taʿṣim: “man tasaddara wa-huwa saghīr fātahu ʿilm kathīr” (Whoever sets himself up to teach while still a neophyte forfeits an abundance of knowledge), Taʿẓīm, fol. 9r. On the license to teach and give legal responsa (al-ijāza li-l-tadrīs wa-al-iftāʿ), see the excellent study by Makdisi, G., The Rise of Colleges, 148–52Google Scholar; see also ibid., 203, on taṣaddur.

⁴⁵ This is only implied in the case of the judge. However, that this was al-Qarafi's understanding is clear, as may be seen in statements such as the following: “When a Mālikī judge orders a man to pay a severence gift (muṭʿa) upon divorcing his wife, and other such recommended acts … such statements constitute no more than legal opinions (fatwas), not binding decisions (aḥkām; sing, ḥukm) that may quell the dispute” (T.55–56). And, “If a man says to a woman, ‘If I marry you, you are thrice divorced’, and then marries her and a judge rules that the marriage is valid, a [subsequent] judge who holds that such statements necessitate divorce would have to uphold this marriage, and he could not issue afatwa obliging divorce….” See al-Furūq, 4 vols. (Beirut: ʿĀlam al-Kitāb, n.d.), 2:103Google Scholar. The implication of this latter statement is that anything a subsequent judge would say that did not reinforce the validity of the marriage would be no more than a fatwa which should be ignored by the litigants.

⁴⁶ By “legal process” I refer to that process that results in a decision (ḥukm) which is backed by force. For more on al-Qarafi's doctrine on the limits of the legal process, see Jackson, , “Defense,” 179–89Google Scholar.

⁴⁷ See Enayet, Hamid, Modern Islamic Political Thought (Austin: University of Texas Press, 1982), 81CrossRefGoogle Scholar. Rida had put forth his views on the caliphate in al-Khilāfa aw al-Imāma al-ʿUẓmā, which appeared sometime between 1922 and 1923—that is, prior to the abolition of the Ottoman caliphate.

⁴⁸ See al-Māwardī, , al-Aḥkām al-Sulṭāniyya, 15–16Google Scholar; Yaʿlā, Abu, al-Aḥkām al-Sulṭāniyya, 27–28Google Scholar. The ten basic responsibilities were (I) guard the religion in its mutually agreed upon principles, (2) resolve disputes through application of the religious law, (3) maintain public order and security, (4) apply to offenders the punishments contained in the religious law, (5) guard the outlying districts, (6) prosecute the military campaign so that non-Muslims become either Muslims or protected religious minorities (dhimmis), (7) collect and distribute taxes and booty, (8) determine the amount of and make disbursements from the public treasury to those who have a right to such, (9) appoint needed and competent people to public office, and (10) personally supervise the affairs of state.

⁴⁹ Malik and al-Shafiʿi held the majority of the Prophet's actions to have been fatwas. This allows al-Qarafi to imply that most of the actions of a caliph may likewise be treated as fatwas, meaning, however, obiter dicta.

⁵⁰ It is interesting that while al-Qarafi's response to Question 25 of the Tamyīz has a direct bearing on the legal implications of Prophetic actions as a source of law, not a trace of this doctrine is found in the section on Prophetic actions in his work on uṣūl al-fiqh, Sharḥ Tanqīḥ al-Fuṣūl. This work was written after the Tamyīz, and it is thus clear that al-Qarafi had developed these ideas by the time he came to write on uṣūl al-fiqh. See Sharḥ Tanqīḥ al-Fuṣūl, ed. Saʿd, Ṭāhā ʿAbd al-Raʾūf (Cairo: Maktabat Kulliyyāt al-Azhariyya, 1393 A.H.), 441Google Scholar, where the Tamyīz is cited; see also ibid., 460, where al-Qarafi states that he completed this work on Monday, 10 Shaʿban, 677 A.H. I am not as yet prepared to say for sure what all of this means. What it suggests, however, at least to my mind, is that (1) the Tamyīz was primarily a political treatise and not a legal one (although for al-Qarafi law is necessarily political, and to write about law is thus to write about politics), and (2) that by al-Qarafi's time uṣūl al-fiqh had become too insular a discipline to be used as a vehicle for furthering political causes; in other words, al-Qarafi's doctrine on Prophetic actions was simply too blatant (and perhaps too radical) in its political intent and implications to be admitted into the sanctum of legal science proper. True, the views of al-Shafiʿi, Malik, and Abu Hanifa cited in al-Qarafi's discussion show that the eponyms themselves did not shy away from the political implications of their legal deductions. But theirs was the formative period during which the political contours of the law were still being set. Later jurisprudence seems to have sanctioned the emergent order as the “Islamic” order and refused, subsequently, to contemplate any other possibilities for political organization—even in cases where new political orders, such as the sultanate, had actually evolved. Exceptions to this, such as reflected in the work by al-Juwayni cited earlier are rare (see n. 10). Finally, one may note that al-Qarafi's doctrine on Prophetic actions does show up in a later work on uṣūl al-fiqh by Taqī al-Dīn (d. 1355) and al-Subkī, Tāj al-Dīn (d. 1370), al-Ibhāj fī Sharā al-Minhāj, 3 vols., ed. Ismāʿṣl, Shaʿban Muḥammad (Cairo, 1402 A.H.), vol. 3Google Scholar. However, by this time the legal implications and intent behind his doctrine are no longer appreciated, and al-Subki (Taj al-Din, who completed the work begun by his father) adduces it merely to entertain his reader with what he presents as his own intellectual prowess.