Character and Authenticity of the Muslim Tradition on the Life of MuammadAuthor(s): Gregor SchoelerSource: Arabica, T. 49, Fasc. 3 (Jul., 2002), pp. 360-366Published by: BRILLStable URL: http://www.jstor.org/stable/4057655 .Accessed: 21/06/2014 10:06

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Arabica.

http://www.jstor.org

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

METHODES ET DEBATS

CHARACTER AND AUTHENTICITY OF THE MUSLIM TRADITION ON THE LIFE OF MUHAMMAD

BY

GREGOR SCHOELER*

A comprehensive discussion of my book Charakter und Authentie der muslimischen Uberlieferung fiber das Leben Mohammeds, by Herbert Berg, appeared in Journal of the Amenrcan Oriental Society 119 (2), 1999, pp. 315- 317. This discussion seems at first sight to be thorough and profes- sional, and it is to be commended that the criticism raised by the reviewer is presented in a tone which is moderate and not polemical.

This positive initial impression dwindles, however, when Berg's remarks are put under the magnifying glass. It is then seen that his discussion exhibits not only errors of fact as well as errors and imprecision in the portrayal of my methodology, my argumentation and my results, but also that he has failed to explicitly point out the primary instrument of my investigation and the most important arguments for the "mid- dle position" I maintain on the question of the authenticity of early Islamic tradition, let alone discussed them in detail.

As regards the primary instrument of my investigation, it consists in the compilation of complete corpora of traditions; in the present case, in particular all existant versions of two traditions traced back to 'Urwa b. al-Zubayr (the first revelatory experience of Muhammad and the narrative of the slander of 'A'isa [scandal story]). For assessing the zsndds, suffice it to say here that their correctness-including the cor- rectness of the common link that they display-is not just simply accepted, but rather tested in the investigation against the texts (mutuin). The analy-

* A Response to Herbert Berg's review of my book Charakter und Authentie der mus- limischen Uberlieferung uber das Leben Mohammeds (Berlin, New York: Walter de Gruyter 1996).

?D Koninklijke Brill NV, Leiden, 2002 Arabica, tome XLVIII,3 Also available online - www.brill.nl

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

METHODES ET DEBATS 361

sis of the texts is done analogously to the investigation of manuscripts whose interdependence is to be ascertained: interdependencies are deter- mined from the structure (abbreviations, additions, gaps) and wording of these texts. If the text analyses show that the isndds correctly indi- cate the interdependencies and that all of the present versions indeed lead back to a common archetype-the common link of the isndds- then-and only then-can it be regarded as proved that the tradition was disseminated by the transmitter who turned out to be the com- mon link. The content (generally, however, not the exact wording) of the tradition spread by the common link transmitter can then be approxi- mately determined by means of the diverse, often considerably differing, further transmissions of this tradition.

In a fundamental treatment of my theses, which is what Berg wanted to accomplish, the instrument of the corpus, its compilation, examina- tion and evaluation needed absolutely to have been clearly pointed out and discussed. This instrument, or rather the correct handling of it, is a protection against arbitrary assessments of individual traditions; as it enables the comparison of texts, it very often allows the distinguishing of good transmissions of a tradition from bad ones, it identifies adulter- ations of texts and forgeries, it shows the mutual core of a tradition which developed differently through different transmitters, etc.

After the presentation of my most important scientific instrument, my main argument will be expounded and clearly emphasized in the following: The total corpus of traditions traced back to 'Urwa b. al- Zubayr (died ca. 712 A.D.) must in great part be seen as being "gen- uine", i.e. the overwhelming majority of these reports do indeed have their origins with 'Urwa. (WVhether the contents of 'Urwa's traditions are historical in each case is another matter). A test of genuineness is always possible when an individual tradition attributed to cUrwa has been further transmitted by his two primary transmitters, his son Hisam and his master pupil al-Zuhri. A substantial portion of the 'Urwa tra- ditions has in this way undergone a "two-tracked" transmission (some- times a third transmission is even found). A comparison of texts of the same tradition in the two (or three) transmissions leads more often than not to the result that 1. both texts have indeed been transmitted inde- pendently of one another (they always have an "own face") and 2. they indeed can be traced back to a common source, namely 'Urwa. (The corpus does however also allow for the recognition of erroneous attri- butions, fraudulent traditions and forgeries!) But since 'Urwa as the son of one of the first disciples and close relatives of the Prophet and

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

362 METHODES ET DEBATS

as nephew and close confident of his wife 'A'i-sawas in close if only indirect contact with the historical time of Muhammad (distance: about 35-70 years), the reports he collected, at least the events of the Medinan period-and their main outlines at any rate-would be all in all correctly transmitted. (With regard to the above, cf. also my article in EF, s.v. 'Urwa b. al-Zubayr.) And as the 'Urwa traditions form the basic frame- work of the entire Sira, the latter cannot be of such extreme inau- thenticity as is assumed by the (hyper-)critical school of Wansbrough, Crone and Cook.

This is my main argument for the middle position I take on the question of genuineness. As an additional argument for the genuine- ness of the scandal story I have stated that the main outlines of the story go against the usual pattern ('A'is'a as "mother of the believers"), even that the entire story (like the story of the satanic verses, f.i.) must have been a matter of extreme awkwardness for the Prophet, some- thing that his disciples would hardly have invented. In keeping with discoveries in oral tradition research, however, characteristics and anec- dotes of this sort should be seen as being reliable, since they have resisted the tendency towards idealization.

So much for my primary instrument and main arguments. I shall now point out a few of Berg's incorrect judgements and direct mis- takes, some of them grave.

On p. 317a the reviewer accuses me of being inconsequent; he main- tains that I treat the iqra' and the scandal stories differently, in that I do not recognize the former-despite the isnad which is traced back to 'A'is'a as having originated with 'A'is'a, but rather trace it back to qass material, whereas for the latter-in agreement with the isndd-I accept 'A'is'a as the original reporter (informant). Here it turns out that Berg has difficulty interpreting the corpora, because my assessment is based on a significant difference between the corpus of "al-Zuhri 'an 'Urwa" traditions on the one hand and the corpus of "Hisam 'an 'Urwa" traditions on the other. Namely, the former exhibits the spe- cial feature that al-Zuhri traces most of the 'Urwa traditions back to 'A'is'a as the original informant (isndd: al-Zuhri < 'Urwa < 'A'is-a); on the other hand the latter distinguishes itself in that His'am gener- ally has the isndds of the respective traditions end with his father 'Urwa (isnad: His'am < 'Urwa). As now the isnad in the iqra' account is only in al-Zuhri's version traced back via 'Urwa to 'A'isa, but in Hisam's version stops at 'Urwa, while for the scandal story the isndd with both

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

METHODES ET DEBATS 363

al-Zuhri and Hisam (against the rule for him!) is traced back to the Prophet's wife, I have concluded that the scandal story with probability indeed goes back to 'A'isa, yet do not believe that the same can be assumed for the iqra' story. (By no means, however, have I excluded their origin with 'A'isa; s. Char. p. 87f!) Why this stance is said to be inconsequent is beyond my comprehension. But it is wrong indeed when Berg says, "In both cases, not all the isnads cite her (i.e. 'A'isa)"; cor- rect is rather that, in the case of the iqra' account, the isnads only of the al-Zuhri's version(s)-these, however, without exception!-quote 'A'isa, while in the case of the scandal story the isnads in the al-Zuhri's version(s) as well as the Hisam version(s) go back to 'A'isa.

Furthermore, Berg insists that, "Schoeler accepts the claims of the isndds unless they disagree with his conclusions, in which case, he decides that the isndds have been manipulated. Why only the isndds of the tra- ditions that disagree with his reconstructions should be dismissed as false is unclear" (p. 317a). This is a complete distortion and erroneous portrayal of my method, according to which the isnads, in order to be regarded as correct, need to be checked against the texts. In some cases the (above described) method also even shows that the isnad in ques- tion cannot be correct. This is the case, e.g., for the isnad cited by al- Waqidi in his version of the scandal story (s. Char. p. 137f). Berg con- cealed the fact that the spuriousness of al-Waqidi's source information can be irrefutably proved: Through an exact comparison of the texts I could indeed prove and verify (s. ibid., the table opposite p. 139 in par- ticular) three of al-Waqidi's actual sources, none of which is mentioned by him in the isnad (among them an obscure tradition from a certain al-Muwaqqari). Moreover, I could prove in the same way that Ibn Ishaq actually used the sources he gives in the isnad, whereby the cor- rectness of his isnad is proved (ibid. p. 127ff).

Finally Berg criticizes my use of the traditional biographical litera- ture for portraying the teaching and learning methods of the Medinan scholars (Ch. I of my book), literature written much later than the events (he mentions Ibn Ha'ar's Tahdib, which is known to be from the 15th century). Berg leaves unmentioned thereby that numerous rel- evant reports are also accessible in very much earlier works ('Abd al- Razzaq's Musannaf; al-Fasawil's K. al-Ma'rjfa, Ibn Sa'd's Tabaqdt) (and also subsequently cited by me), so that, if need be, al-Dahabil (14th century) and Ibn Hagar, who certainly comfortably compile a large quantity of material, can be done without. A comparison of the infor- mation of these later biographers with the corresponding versions of

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

364 METHODES ET DEBATS

the earlier ones shows what type the variations are that arise in the texts. In the following I have compared the oft cited report traced back to Ma'mar b. Rasid > Sahih b. Kaysan on al-Zuhrl's methods of col- lecting traditions in 'Abd al-Razzaq (d. 211/827) and al-Fasawi (d. 277/890) with the corresponding reports of the later biographers al- Hati-b al-Baghdadi (d. 463/1071) and Ibn Hagar (d. 852/1449).

'Abd al-Razzaq, Musannaf 11/258

Can Salih ibn Kaysan, qala: igtama'tu ana wa-bn Sihab, wa-nahnu natlubu al-cilm.

fa-gtamacna 'ala an naktuba al-sunan, fa-katabna kull say' sami'nahu Can al-nabi.

tumma katabna aydan ma ga'a can ashabihi, fa-qultu: Ia, laysa bi-sunna wa-qala huwa: bal, huwa sunna. fa-kataba wa-lam aktub. fa-angaha wa-dayyactu.

al-Fasawl-, K. al-Ma'rfa 1/637

Can Salih ibn Kaysan qala: igtamactu ana wal-Zuhri, wa-nahnu natlubu al-cilm.

fa-qulna: naktubu al-sunan, fa-katabtu ma ga'a can al-nabi. tumma qala: naktubu ma ga'a can ashabihi, fa-innahu sunna. qala, qultu ana: laysa bi-sunna, fa-la naktubuhii. qala: fa-kataba wa-lam aktub. fa-an-gaha wa-dayyactu.

al-HatT-b al-Baghdadi: Taqyrd al-'ilm, 106-107:

ahbarana Salih ibn Kaysan qala: igtamactu ana wal-Zuhri, wa-nahnu natlubu al-'ilm.

fa-qulna: naktubu al-sunan, fa-katabna ma ga'a 'an al-nabi. tumma qala: naktubu ma ga'a can ashabihi, fa-innahui sunna. fa-qultu ana: laysa bi-sunna, fa-la naktubuhii. qala: fa-kataba wa-lam aktub. fa-an-gaha wa-dayyactu.

Ibn Hagar: Tahib 9/396-397

Can $alih ibn Kaysan: kuntu atlubu al-cilm ana wal-Zuhrl. fa-qala: tacala naktub al-sunan. qala: fa-katabna ma ga'a can al-nabi.

tumma qala: ta'ala naktub ma ga'a 'an as-sahaba. qala: fa-kataba wa-lam naktub. fa-angalha wa-dayyactu.

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

METHODES ET DEBATS 365

The comparison shows that although the report contains many vari- ants in the different versions separated by very great lengths of time, these variants, however, restrict themselves to the wording and in no way affect the substance of the report. At any rate, the distance between the event and the definitive fixation of the event in writing is often by no means as great as Berg would have it appear. What we read, e.g., in Ibn Sa'd's (784-845) Tabaqdt about Ibn Ishaq (ca. 704-767) is sep- arated by a distance of two generations.

Be that as it may, it is much more important that I have expressly remarked that I, too, am of the opinion that a single report need not always absolutely correspond with the historical truth; I have only pos- tulated that the overall picture yielded by these stories about the learning and teaching methods in the Islamic world from the time of 'Urwa must for the most part be correct (Char. p. 7). A similar view, by the way, is also held of late by the "sceptic" M. Cook when in his long article "The Opponents of the Writing of Tradition in Early Islam" (in Arabica 44 [1997], pp. 437-530)-in full agreement with the tradition-sees the being-in-force of a prohibition on writing 4adits for the first half of the 8th century as assured, although no contemporary reports whatsoever are available on this matter.

In the following I shall correct some of Berg's clear errors and impre- cise statements:

P. 315b: "and his (i.e. al-Zuhri's) notes were destroyed before his death." This refers not to al-Zuhri but to 'Urwa (s. Char., p. 29f). Ibid. "According to Schoeler, this book (i.e. Ibn Ishaq's copy of the Sra, al-I?itdb al-kabfr, which was presumably or in fact made for the court), is lost." This book is no longer in existence in fact. According to M. Lecker (in ZDMG 149 [1999], p. 411), however, quotations from Ibn Ishiaq's al-Sra al-kubrd (= al-Kitdb al-kabFr [?]) are found in Ibn Hagar's Isdba.

P. 316a: "'Urwa is likely source (i.e. of the iqra' story), but his son Hisham transmitted a complete version.. ." In fact only a single, very short (possibly greatly abbreviated by Ibn Sa'd) version of the iqra' nar- rative is extant from Hisam (Char. p. 79f).

P. 316b: "Crone has argued that al-Waqidi did not copy Ibn Ishaq. Rather, both men had a common source of qdss material." This- erroneous-theory is not from Patricia Crone (she only quotes and adopts it), but from J.M.E. Jones. Wellhausen had already remarked, and Horovitz then subsequently irrefutably proved by means of 10

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

366 METHODES ET DEBATS

examples in his dissertation written in Latin that al-Waqidi's main source is Ibn Ishaq, although he never quotes him (Char. p. 141).

Ibid. "Schoeler... .concludes that the report of al-Zuhri is historically reliable (because it is an 'inverted common link')." The stating of this circumstance was not meant to be an argument. My main argument for this 'Urwa tradition having its origin with great probability in 'A'ilsa, however, consists in it having been transmitted not only by al-Zuhri-, but also by Hisam, with both of them indicating 'A'isa as the original informant (s. above).

P. 317a: "A more convincing strategy would have been to look at apparently related traditions and analyze their contents without recourse to their isnads. If the contents displayed a consistent correlation with their isndds, then at least some of the information contained in isnads might by authentic." Berg, after having ignored that I have based my investigation on complete corpora, suggests on his part at the end of his review-the preparation of a corpus as a better alternative to my pro- cedure. This, however, should be compiled without taking the isndds into account, and solely based on the texts (see above for my method!). It is against the content of the texts, then, that the isndds should be checked.

Agreed! If Berg succeeds in completely putting together a complex of traditions based solely on the similar content of traditions, and if he then tests the isnads accurately against the texts, his investigation, pro- vided that he works in an orderly fashion, must lead to the same results as mine with respect to the relationships of dependence. That this did not happen in his examination of my investigation of the iqra' narra- tion and scandal story can come as no surprise, considering the impre- cision of Berg's method of work.

This content downloaded from 193.231.35.250 on Sat, 21 Jun 2014 10:06:22 AMAll use subject to JSTOR Terms and Conditions

Article Contentsp. [360]p. 361p. 362p. 363p. 364p. 365p. 366

Issue Table of ContentsArabica, T. 49, Fasc. 3 (Jul., 2002), pp. 267-401Front Matter"Du ventre de la royaut aux jardins de la papaut": entre islam et chrtient, le lexique et la mtaphore du Corps [pp. 267-298]L'onagre dans le Diwn d'al-Aal [pp. 299-324]Modern Gendered Illustrations of the Life of the Prophet of Allah: tienne Dinet and Sliman Ben Ibrahim (1918) [pp. 325-359]Mthodes et DbatsCharacter and Authenticity of the Muslim Tradition on the Life of Muammad [pp. 360-366]

Notes et DocumentsDeux verbes singuliers de l'arabe classique: Istaraa et Istaa [pp. 367-375]Some Formal Characteristics of the Musnad Type of ad Collection [pp. 376-382]Sad B. ubayr: Piety, Chess and Rebellion [pp. 383-386]

Bulletin CritiqueReview: untitled [pp. 387-393]Review: untitled [pp. 393-394]Review: untitled [pp. 395-397]Review: untitled [pp. 397-398]Review: untitled [pp. 399-401]

Back Matter