Glosses and Commentaries on Aristotelian Logical Texts: The Syriac, Arabic and Medieval Latin Traditions.

• Author: Lameer, Joep

From the very outset, the Organon of Aristotle had a strong impact on logical studies, both in the West and in the East. The joint effort of numerous scholars, ever since the publication of the second volume of the celebrated CAG series in 1883, has done much to further our knowledge of the classical commentary tradition.(1) In contrast, the study of Syriac, Arabic, and medieval Latin glosses and commentaries on Aristotle's logical works is still in a relatively early phase; indeed, many texts are yet to be edited, let alone studied. It is for this reason that the collection of articles under review represents a welcome and much needed step forward in charting this complex field. Below, I shall be mainly concerned with the contributions of Henri Hugonnard-Roche and Dimitri Gutas, which treat of the medieval Arabic tradition, while restricting my comment on the others to a few marginal remarks.

The first article is Sebastian Brock's "The Syriac Commentary Tradition" (pp. 3-18). In this paper, Brock gives a condensed yet valuable overview of Syriac scholarship in the field of Aristotelian logic from the sixth to the thirteenth century. This overview is divided into the following sections: a) the Syriac translations of the Organon, b) the Syriac commentary tradition (where "commentary" is used in the widest sense of the term, ranging from general introductions to discourses in verse), c) a note on two prominent commentators: Probus and George, Bishop of the Arabs, and d) a brief listing of some outstanding problems. Here, I would like to make just three remarks.

First, it might be of interest to note that it was probably for theological rather than logical reasons that Syriac scholars in late antiquity restricted their studies to the Categories, the de Interpretatione and Prior Analytics I.1-7 (see Brock, p. 3). This, at any rate, seems to be the implication of Abu Nasr al-Farabi's (d. 950) account on the appearance of philosophy as preserved in Ibn Abi Usaybi a's thirteenth-century bio-bibliographical dictionary, the Uyun al-anba fi tabaqat al-atibba, from which the following passage deserves quotation:

The bishops assembled and took counsel together on which parts of the teaching of philosophy should be kept and which should be abolished. They decided that the books on logic could be taught up to the assertoric figures, but no further, since they believed that everything beyond would harm Christianity, whereas the permitted material could be used for the promotion of their creed.(2)

At page 4, in the entry on the Prior Analytics, 1 or 2a, Brock, with Nagy, describes the work in question as being by an anonymous author. But see also Anton Baumstark's Geschichte der syrischen Literatur, in which he suggested attributing this treatise to Probus (sixth cent.?).(3)

In connection with the dating of Probus' works (Brock, p. 7), I would draw attention to H. Suermann's "Die Ubersetzungen des Probus und eine Theorie zur Geschichte der syrischen Ubersetzung griechischer Texte," Oriens Christianus 74 (1990): 103-14 (which might be added to the references given in note 30). In essence, Suermann draws the same conclusion as Brock, namely that Probus' linguistic usage would appear to belong to the sixth century rather than to the fifth.

Last, attention should be drawn to Brock's appendix (pp. 11-15), which contains a systematic and chronological series of references to Syriac translations of the Organon and the main commentary tradition. These references inform us about manuscripts as well as about available studies and editions. Together with the main text, this listing (which does not pretend to be exhaustive) will prove very useful as a tool in future research.

Next, there is the article by Henri Hugonnard-Roche (hereafter H.-R.), entitled "Remarques sur la tradition arabe de L'Organon d'apres le manuscrit Paris, Bibliotheque nationale, ar. 2346" (pp. 19-28). In this brief but interesting and thought-provoking contribution, H.-R. invites us to take another look at some of the interlinear and marginal glosses and notes to the Arabic translations of the Organon as preserved in this famous manuscript. According to H.-R., Richard Walzer, in his classic "New Light on the Arabic Translations of Aristotle,"(4) was wrong to stress the supposedly merely philological character of these glosses and notes because in many instances they give evidence of a different understanding of the logic of Aristotle, be it at the level of concepts or at the level of logical theory.

Challenging though it is in proposing a new perspective, I am not entirely convinced that the examples given by H.-R. on pp. 22-27 provide him with a water-tight case. True, his remarks on the alternative renderings of the expression ex hypotheseos syllogismous at Top. I.18 108b 12 (see pp. 22-23) and the various translations of the definition of the syllogism and its explanation at APr. I.1 24b 18-22 (see pp. 23-24) seem to be well taken. But in the majority of the cases, H.-R.'s analyses and comments suggest differences in understanding where in fact they do not exist. Let me explain.

At page 22, H.-R. suggests interpreting the alternative translations of dialektike (methodos) at Top. 1.2 101b 2 by sina at al-mantiq and (sina at) al-jadal (given as a marginal gloss) as two different ways of looking at the relation between dialectic and the art of logic. H.-R.'s line of reasoning is based on the otherwise correct observation that, at least from al-Farabi onward, jadal in Arabic logical terminology refers to dialectic proper as a specific part of the art of logic in general, which is invariably referred to as mantiq. Yet I think he is begging the question where he intimates that the translator of the Topics, Aba Uthman al-Dimashqi, did not distinguish the two because he used the compound sina at al-mantiq in translating dialektike (methodos), whereas the glossator must have regarded dialectic as a subspecies of the art of logic in general since he preferred the compound (sina at) al-jadal. After all, elsewhere in the Topics, e.g., at Top I.14 105b 23, Aristotle uses the adjective logikos in the sense of dialektikos. Therefore it is not at all inconceivable that it was with this particular acceptance of logikos in mind that al-Dimashqi felt free to render dialektikos sometimes as jadali(5) and at others as mantiqi, which would then be a translation ad sensum rather than ad litteram. Now the interesting thing is that, at 105b 23, al-Dimashqi translates logikos as mantiqi(6) without being corrected by the glossator, which can only mean that the latter, or rather the ancient translator(s) upon whose (Syriac) translation(s) he drew in making his comments,(7) did not understand mantiqi (only) as referring to "logical" in the general acceptance of that term. Given these findings, I am inclined to regard al-Dimashqi's use of mantiqi for dialektike as explainable in terms of a translation ad sensum, whereas the glossator's source appears to have preferred a translation that is both ad sensum and ad litteram (i.e., dialektikos = jadali and logikos = mantiqi). Accordingly, I conclude that the translations differ only from a philological point of view (ad litteram) and not from a logical one (ad sensum), as intimated by H.-R.

The second case (see pp. 24-25) concerns the translation of the phrase ek gar tes antistrophes perainetai to anagkaion at APr. I.14 33a 20 ("for that which is necessary [i.e., the conclusion] results from a conversion").(8) In the Arabic translation of the Prior Analytics by Theodorus, this phrase was rendered as follows: li-anna-hu in-nama tajibu l-natija an-hu min al-ruju,(9) whereas the glossator gives the following alternative translation taken from the Syriac: wa-dhalika anna l-daruri innama ya-kunu min al-ruju al-idtirari.(10)

As stated by H.-R., Walzer explains the difference between these two translations by assuming that Theodorus read perainetai ("results") in the Greek (giving rise to the remark [endorsed by H.-R. but not so by me, on which see below, n. 14]) that we should read tantiju ["results"] rather than tajibu ["is necessary"]), whereas there must have stood ginetai ('comes about') in the text that was at the basis of the Syriac version that is quoted in Arabic translation in the margin.(11) According to H.-R., this assumption is wrong: in his view, the translations are much more likely to differ as a result of a difference in understanding the word anagkaion. For where Theodorus took this (rightly) to refer to the necessity, i.e., the...