The Oral and the Written in Early Islam

Gregor Schoeler, Uwe Vagelpohl, James E. Montgomery

• Éditeur: Routledge
• Année: 2006
• ISBN:
  9781134158799
• Édition: 1
• Sujet: Literature
• Sous-sujet: Middle Eastern Literary

Made up of a number of seminal articles that are translated for the first time in English, this prestigious book from Gregor Schoeler gives a reasoned, informed and comprehensive overflow of how the written and the spoken interacted, diverged and received cultural articulation among the Muslim societies of the first two centuries of the Hijra.

1
THE TRANSMISSION OF THE SCIENCES IN EARLY ISLAM
Oral or written?

Hitherto, controversy has surrounded the issue of whether the major compila-tory works of the Arabo-Islamic sciences composed between the second/eighth and fourth/tenth centuries, marked by their use of כisnād (chain of transmitters), depended on mainly written or oral sources. Examples of such compilations are the Kitāb āl-muwaṭṭaכ (The Book of the Well-Trodden [Path]) by Mālik Ibn Anas (d. 179/796), the Kitāb al-maġāzī (The Book of the Campaigns) by Ibn Isḥāq (d. 150/767), the Ṣaḥīḥ ( The Sound [ Compilation ]) of al-Buḫārī (d. 256/870) and Muslim (d. 261/875), aṭ-Ṭabarī’s (d. 310/923) Taכrīḫ (History) and Tafsīr (Qur’ān Commentary), and Abū ’l-Faraǧ al-Iṣfahānī’s (d. 356/967) Kitāb al-כaġ ānī (The Book of Songs).⁵⁸

In her Studies in Arabic Literary Papyri,⁵⁹ Nabia Abbott advocated an early and incremental written tradition, based on a plethora of evidence such as Umayyad papyri fragments. Fuat Sezgin proposed in his Geschichte des arabischen Schrift-tums⁶⁰ a method for the reconstruction of the (as he maintains, exclusively written) sources of these compilations.⁶¹ He further maintained that he had discovered a number of early source texts on which the late compilations were based.⁶² With the work of these two scholars, earlier claims about a largely oral transmission of the Arabo-Islamic sciences up to the time of the major compilations⁶³ seemed to have been laid to rest.

[202] (The numbers in brackets refer to the pagination of the original articles on which the translation is based.) In the meantime, however, several studies testing Sezgin’s method and claims have cast doubt on the exclusively written character of these sources. At best, the newly discovered, purported source texts proved to be later arrangements or different, but by no means earlier recensions of those source texts, that is, recensions which were not drawn on in the well-known later compilations (e.g. aṭ-Ṭabarī’s Taכrīḫ [History]). One example is the so-called Qur’ān commentary of Muǧāhid (d. 104/722), actually the Tafsīr Warqāכ ^сan Ibnābī Naǧīḥכan Muǧāhid (The Qur’ān Commentary of Warqā’ on the Authority of Ibn Abī Naǧīḥ on the Authority of Muǧāhid).⁶⁴ At worst, they turned out to be extracts from later compilations, for example, Abuī MiḪnaf’s (d. 157/774) presumed Kitāb al-ġārāt (The Book of Raids), which is in fact a part of Muḥammad ibn A ‘ṯam al-Kūfī’s (d. after 204/819) Kitāb al-futūḥ (The Book of Conquests) in which Ibn A ‘ṯam exclusively quotes traditions from Abū Miḫnaf.⁶⁵

Moreover, studies of works extant solely in divergent later versions have uncovered a high degree of discrepancy between those different versions. For this reason, literal, and sometimes even complete, quotations of (more or less codified) books, which, according to Sezgin, had already taken place at an early date in the transmission of scientific knowledge,⁶⁶ seem highly unlikely. As a result, Sezgin’s optimism in claiming to be able “to reconstruct many old source texts in their entirety from later compilations”⁶⁷ was unjustified. Al-Samuk’s study dealing with the different extant recensions of Ibn Isḥāq’s biography of the Prophet (Ibn Hišām’s [d. 218/834] Sīra [Biography], aṭ-Ṭabarī’s Ibn Isḥāq- “quotations” [203] etc.) has shown that, due to the innumerable variants found in the different textual traditions, a reconstruction of Ibn Isḥāq’s material would evince confusing inconsistencies.⁶⁸

Werkmeister’s study on the sources of the Kitāb al-כiqd al-farīd (Book of the Unique Necklace) established that sources demonstrably available to the author in manuscript form had little impact on the work. Alleged borrowings by Ibn ‘Abd Rabbihī (d. 328/940) from actual books which previously had been considered his models and sources (al-Ğāḥiẓ’s [d. 255/868–869] Kitāb al-bayān [The Book of Eloquence (and Exposition)], Ibn Qutaybah’s [d. 276/889] Kitāb כuyīun al-כaḫbār [The Book of the Wellsprings of Reports]) for the most part exhibit substantial differences from their supposed counterparts in the aforementioned texts. Only an indirect connection can plausibly be posited.⁶⁹ All this seems to point towards oral transmission. Advocates of written transmission can, however, argue against these two studies as follows: in the case of Ibn Isḥāq, credible authority has it that he put his history down in writing,⁷⁰ while for Ibn ‘Abd Rabbihī, some of his supposed oral sources are texts which had been put into a fixed written form by their authors.

Today’s uncertainty about the question of oral versus written transmission is fittingly illustrated by M. Fleischhammer’s statements on the sources of the Kitāb al-כaġānī (The Book of Songs), a subject which he studied intensively. He maintains on the one hand that “Nowadays, . . . there is widespread agreement that, in most cases, these כisnāds conceal written sources” while on the other, he states: “Often enough, we cannot disprove beyond doubt the existence of a genuinely oral tradition.”⁷¹

[204] In what follows, we will attempt to solve this problem by proposing a theory which can, we believe, reconcile what seems to be diametrically opposed points of view. It should be added that this theory emerged as a result of a careful consideration of the results of previous, established research rather than renewed source studies and that, in the course of our examination, we felt compelled to return to the view of A. Sprenger on a number of essential points. He was the first Orientalist to deal with this question.⁷²

The theory will be formulated in six points. For a better understanding of our argument, it will be helpful to illustrate some of the characteristics of the Islamic practice in the teaching of the sciences. Modern academic lecture courses, the “Vorlesung,” shall serve us as a model. The institution of academic lecture courses, practised in antiquity (some of Aristotle’s works were only transmitted through lectures), was familiar to Muslims, too, under the label samā^с, namely, “audition.”⁷³ This form of teaching, which involved the students listening to a teacher’s (šayḫ) or his representative’s recitation given on the basis of written notes or from memory, is generally regarded as the superior mode of transmission. Only qirāכah, “recitation”, later also known as ^сard. , “presentation”, was considered equal. Like samā^с, it took the form of a lecture, in which the student, in the presence of his teacher, either recited material on a subject from memory or read it out from his written notes. The teacher listened and made corrections. These “lectures” were held in maǧālis or muǧālasāt (sessions) and ḥalaqāt (circles), which in earlier times often took place in mosques, sometimes also in other places, for example, a scholar’s home.⁷⁴ Apart from these two methods of transmitting information, simple copying of notebooks (wiǧādah, [205] kitābah, etc.)⁷⁵ emerged early on. Inasmuch as the text in question was not “heard” from an authority, its transmission was regarded as inferior.⁷⁶

I

On the basis of extensive evidence collected by Abbott and Sezgin, it has become clear that, in the very beginning, writing was used sporadically, and that, over time, its use to record ḥadīṯ, legal rulings, historical information, poetry, and so on became more and more widespread.

We should note in particular that this also applies to ḥadīṯ . Interestingly, academic discussion about written tradition in the earliest period is less heated than that concerning the phase immediately prior to the composition of the major compilations. On the one hand, Goldziher explicitly asserts that initially, ḥadīṯ was not exclusively intended to be orally transmitted and provides evidence that it had also been put into writing sporadically at a very early stage.⁷⁷ On the other, Abbott⁷⁸ and Sezgin⁷⁹ admit that after this earliest period, there were occasionally religious misgivings against putting ḥadīṯ into writing. This very early stage, however, will not be dealt with in the following discussion.⁸⁰

The existence of ḥadīṯ collections is a much more controversial issue: should we, with Goldziher,⁸¹ date the beginning of the muṣannafāt (works systematically arranged into thematic chapters) to the time of al-Buḫārī (d. 256/870) and Muslim (d. 261/875) or place it with Sezgin⁸² a century earlier? Similarly, we could for example inquire after the existence of fiqh literature before Mālik Ibn Anas (d. 179/796) or historical books before Ibn Isḥāq (d. 150/767) or even, substantially later, aṭ-ṭabarī (d. 310/923), as well as after the existence of codified works of literary history preceding Abū ’l-Faraǧ (d. 356/967) and so on.⁸³

[206] Against the existence of written ḥadīṯ collections prior to al-Buh1ārī (and of other contemporary works in different fields of learning), scholars have since Goldziher quoted certain topoi frequently found in the sources such as mā raכaytu/afī yadi-hī kitāban qaṭṭu ( “I [one] never saw a book in his hand”) or lam yakun la-hū kitābכinna-mā kāna yaḥfaẓu ( “he did not have a book, but used to memorise it/keep it in his memory”).⁸⁴ These topoi, obviously highly laudatory, have been reported in relation to exponents of several areas of learning, for example, ḥadīt (Sa ‘īd ibn Abī ‘Arīubah, d. 156/773⁸⁵; Wakī ‘ ibn al-Ǧarrāḥ, d. 197/812),⁸⁶ fiqh (Sufyān aṯ-Ṯawrī, d. 161/778)⁸⁷ and philology (Ḫalaf al-Aḥmar, d. c.180/769⁸⁸; Ḥammād ar-Rāwiyah, d. c.156/773⁸⁹; and Ibn al-A ‘rābī, d. 231/846).⁹⁰ ḥ

These expressions should not, however, be viewed in isolation from their context: reports about the teaching and learning methods of the respective scholars. Mostly, they indicate that an authority lectured without ...