Fragments of a New Manuscript of Sefer Ha-Eshkol

A few months ago, I saw on the Ktiv website a description of two fragments of Tractate Berakhot of the Babylonian Talmud in the holdings of the Cambridge University library (Add. 474,2). My attention was drawn to this description because it contained no additional details, and it was stated that the Institute of Microfilmed Hebrew Manuscripts in the National Library of Israel in Jerusalem did not possess a microfilm of these fragments. I checked in the Sussmann catalogue, which includes descriptions of all the manuscripts of the Babylonian Talmud, and I noted that these fragments were not recorded there[2]. This only further piqued my curiosity, and a search in the catalogue by Professor Stefan Reif revealed that these fragments are presented as an unknown commentary to tractate Berakhot that mentions R. Hai Gaon[3]. The Reif catalogue states that the fragments are on parchment, and written in Spanish script dated to the fourteenth century. Although these fragments are not found in the Cairo Genizah collection, I originally thought that they might have been removed from the Genizah before Solomon Schechter transferred the bulk of its contents to the Cambridge University Library in 1897. After a closer examination of the fragments this thought was found less likely.

I contacted Dr. Ben Outhwaite, the head of the Genizah Research Unit in Cambridge, and Dr. Ezra Chwat of the Institute of Microfilmed Hebrew Manuscripts in the National Library of Israel, asking for their aid in obtaining photographs of the fragments, in order to examine them. They both kindly agreed to assist me in this, and I received good photographs of the two fragments. I lost no time in beginning to study them. It soon became evident to me that these fragments do not contain a commentary of the Babylonian Talmud, but are rather fragments of an important halakhic work, Sefer Ha-Eshkol, known from other sources. This work was composed by R. Abraham ben Isaac, the head of the Narbonne rabbinical court, in the middle of the twelfth century. The fragments were cut on at least three sides – in the upper, lower and inner sides with only the outer margin remaining. Although the margin on the inner side is cut off, the text is preserved in its entirety in most of the lines of p. 1, while a few letters are cut from each line on p. 2. A number of lines were cut off at the top and/or bottom parts of the page. The cuts are almost in a straight line. Horizontal folds are noticeable on both pages. The fold in p. 1 is in its lower part, and in p. 2, in its upper part. Professor Judith Olszowy-Schlanger examined the photographs of the fragments, and she is of the opinion that the fragments contain traces of glue. It is quite probable that in the past these fragments had been glued together or to a support, and were part of a book binding; they might have been removed from the binding of MS Add. 474,1, which contains responsa by Geonim and by R. Isaiah di Trani. In other instances, the Reif catalogue describes the connection between two manuscripts listed under a single shelf-mark, for instance – whether they were bound together, or whether one was removed from the binding of the other[4]. Here, however, the catalogue is silent on this question. It notes that the Add. 474,1 and 2 were purchased from Hirsch Lipschütz, a Polish bookseller, in 1869.

There are two printed editions of Sefer Ha-Eshkol, which differ greatly from one another. The first edition was published by R. Benjamin Zevi Auerbach in 1868-69. Already in 1869, an anonymous writer in the He-Halutz journal claimed that this edition contained many additions and falsifications. A stormy controversy surrounding the Auerbach edition erupted in 1909, after Shalom Albeck raised the question anew[5]. A few decades later (1935-1938), Shalom and Hanokh Albeck published the second edition of the book. At present, scholars generally believe that the Albeck’s edition renders faihtfully the original text, while the Auerbach edition contains false readings, but there is no consensus concerning the time of these distortions. Contrary to the prevailing view that Auerbach’s edition’s falsification dates to the nineteenth century, Israel Ta-Shma gave it an earlier dating, placing it in the fourteenth century[6]. The fragments discovered now at Cambridge are of identical content to the Albeck edition. The discovery of fragments of an additional copy identical to the text published by Albeck provides some additional support for the argument that the distortions of the Auerbach’s edition are not of medieval origin. The following is a description of the fragments’ contents, in comparison with the pages in the Albeck edition:

fol. 1a: Ha-Eshkol, 1, Hilkhot Seudah (The Laws of Meals), p. 75: “וקאמרי רבנן” – p. 76: “אשכחינהו דקאמרי ברוך”. Reproduced by kind permission of the Syndics of Cambridge University Library
fol. 1b: Ha-Eshkol, 1, Hilkhot Seudah, p. 77: “לו ולברכו” (thus at the beginning of the first truncated line) – p. 78: “לא לכתוב ולא לקרות”. Reproduced by kind permission of the Syndics of Cambridge University Library
fol. 2a: Ha-Eshkol, 1, Likkutei Hilkhot Tefilah (Collected Laws of Prayer), p. 91: “משיחזיק בדרך” – p. 92: “דאיכא שהות בין”. Reproduced by kind permission of the Syndics of Cambridge University Library

fol. 2b: Ha-Eshkol, 1, Likkutei Hilkhot Tefilah, p. 93: “דרב אמר שמע” – p. 94: “תכסיסא אינון”. Reproduced by kind permission of the Syndics of Cambridge University Library

This description shows that many pages are missing between these two fragments, which leads us to assume that they come from an outer bifolio of a quire. As was noted, some of the lines were cut off in their entirety, and the number of lines on each page cannot be determined. Of p. 1a, 20 lines are preserved; of p. 1b, 21 lines; of p. 2a, 20 lines; and of p. 2b, 19 lines. A regular line contains about 13-15 words. A comparison with the Albeck edition shows that about 116 words, that is, 8-9 lines, are missing between side a and side b of fol. 2. There is, however, a much greater lacuna between sides a and b of fol. 1. What is the meaning of this? The version in the Albeck edition is based on MS Jerusalem, National Library, MS Heb. 4o1159 (formerly MS Carmoli), and Albeck comments on page 76 note c that another identified manuscript of the work, MS Paris, Alliance Israelite Universelle library, MS 91, lacks a lengthy section. Its place is taken in the manuscript by the wording: “The meaning of this halakhah is explained in the Halakhot of the Gaon, of blessed memory.” Albeck understood that the copyist abridged the section since it inclueded a copy from Hilkhot ha-Rif (the halakhot of R. Isaac Alfasi) with no addition or change. If we omit the lacuna in MS Paris, and add the words quoted above, then MS Cambridge lacks 125 words between sides a and b in fol. 1, a finding which is quite similar to the parallel findings in fol. 2. In light of this reconstruction, we may assume that each page comprised 28-30 lines.

The significant abridgement, present in both MS Paris and the fragments from Cambridge, could well lead to the conclusion that these two copies belong to the same textual family. However, a comparison of the fragments with the other textual annotations in the Albeck edition teaches that, in almost every instance, the version of the fragments at Cambridge is identical to that in MS Jerusalem (see especially p. 77, n. 5). The Cambridge fragments might therefore appear to represent a more original version of the textual branch represented by  MS Paris. It seems to me, however, that it would be more correct to argue that the fragments at Cambridge do not belong to this branch at all, and that Albeck erred in thinking that the copyist of MS Paris abridged the section from Hilkhot ha-Rif, while the copyist of MS Jerusalem took pains to complete it. This proposal is surprising, but Albeck demonstrated in the introduction to his edition that Sefer Ha-Eshkol is based on Sefer Ha-Ittim by R. Judah Barceloni, who in turn based his work on Hilkhot ha-Rif. Albeck additionally showed in his introduction (pp. 54-55) that at times the author of Ha-Eshkol, R. Abraham ben Isaac, abridged the passages that R. Judah Barceloni had copied from Hilkhot ha-Rif, since he found nothing new in them. The version of the fragments discovered at Cambridge strengthens the possibility that, in this case as well, it was R. Abraham ben Isaac himself who abridged what had been copied from Hilkhot ha-Rif, rather than one of the copyists of his work, as Albeck thought. A clear answer of this question might be obtained after a complete comparison of the textual witnesses of Sefer Ha-Eshkol has been carried out.


[1] I am grateful to Prof. Simcha Emanuel, for his support in writing this article.

[2] Y. Sussman (in collaboration with Y. Rosenthal and A. Shweka), Thesaurus of Talmudic Manuscripts (Jerusalem: Yad Ben-Zvi, 2012) (Hebrew).

[3] S. C. Reif, Hebrew Manuscripts at Cambridge University Library: A Description and Introduction (Cambridge: Cambridge University Press, 1997), no. 185, p. 129.

[4] E.g., in the description of MS Add. 638,1 (no. 325, p. 198), and in that of MS Add. 497,3 (no. 327, p. 199).

[5] The controversy is presented in the first part of the article by S. Z. Havlin, “More on the R. Auerbach Edition of Ha-Eshkol,” Yeshurun 13 (5703), pp. 701-10 (Hebrew).

[6] I. M. Ta-Shma, Rabbi Zerachiah ha-Levi the Author of Ha-Maor and His Circle (Jerusalem: Mossad Harav Kook, 1993), pp. 40-41 (Hebrew).


You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Search OpenEdition Search

You will be redirected to OpenEdition Search