Save

A Newly Discovered Fragment of the Babylonian Talmud, Tractate Soṭah, in the Vatican Library

In: The Vatican Library Review
Author:
Judith Olszowy-Schlanger University of Oxford Corpus Christi College Oxford UK
CNRS-IRHT Paris France
École Pratique des Hautes Études-PSL Paris France

Search for other papers by Judith Olszowy-Schlanger in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0000-0001-6070-9386
Open Access

Abstract

In the process of systematic search for reused Hebrew fragments in the bindings of printed books in the collections of the Vatican Library, as a part of the collaborative project “Books within Books,” two previously unknown fragments of a manuscript of the Babylonian Talmud (Tractate Soṭah) have been discovered. This paper is a preliminary study of these fragments and focuses on their material features.

Alongside a rich collection of codices and scrolls, the Hebrew language collections of the Biblioteca Apostolica Vaticana contain several dozens of fragments that were recycled as binding supports for other books. During several conservation and rebinding campaigns in the BAV, many such fragments were removed from the bindings they served and then gathered together into modern miscellaneous volumes and assigned a shelfmark among the Hebrew manuscript collections. These volumes can be accessed online in the Vatican Library’s virtual reading room.1 Their description is included in the most recent catalogue of the Hebrew manuscripts in the BAV,2 and a more detailed analysis can be found in the database of the “Books within Books: Hebrew Fragments in European Libraries” project, for which the BAV is one of the founding partners.3 In addition to these long-identified items, the BAV clearly holds further Hebrew fragments within the bindings of early printed books. Together with the Books within Books project, the BAV has been undertaking for the past six years a systematic search and inventory of previously undetected binding fragments.

This brief note concerns two fragments from the same medieval copy of the Babylonian Talmud, recently identified by Delio Proverbio, the Vatican Library’s scriptor Orientalis. Generally speaking, each and every page of recovered medieval Talmud manuscripts represents a precious testimony to this founding Jewish text. Indeed, only quite a small percentage of medieval Talmud manuscripts have been preserved, largely because of the persecution and systematic destruction inflicted on this book, which Christian authorities considered blasphemous. As for these Talmud manuscripts which somehow escaped the auto da fé, they often suffered from natural wear and tear due to their intensive and regular use. Besides providing new evidence on the medieval Talmudic text, the analysis of the material features of these fragments can provide some insights into the history of the volume which they served to bind.

The call number “R.G.Oriente.II.82” corresponds to a composite volume: the present binding combines two different early printed works. The first unit is the biblical book of Job with a commentary by the Saragossa-born scholar Meir ben Isaac Arama (c. 1460–1550). The commentary was written in 1506 in Salonica, after Meir Arama fled from Naples following the French invasion in 1495. It is also in Salonica that the commentary was first printed in 1517, under the evocative title of Meir Iyyov, “Illuminating Job,” a pun linking the commentator’s interpretation of this complex biblical book to his first name, Meir. Although the Vatican scriptor who described the volume on the blank recto of the Job commentary (f. 1r) specified that the place of printing is not mentioned, the copy in the BAV is, in fact, the editio princeps.4 The second unit in the volume is the Sefer ha-Rokeah, a famous compendium of Ashkenazi legal lore and customs by Elazar ben Judah of Worms (c. 1176–1238). It was first printed in Fano in 1505 at the press of Gershon Soncino, and then reprinted twice in the sixteenth century, in Venice in 1549 and in Cremona in 1557.5 The copy of this work is also the editio princeps, as confirmed by the printer’s colophon on f. 226v: “The Rokeah, composed by the Rabbi our teacher Elazar son of the Rabbi Judah of blessed memory, was completed in the town of Fano on the evening of Passover of the year 265” (corresponding to Wednesday, 29 March 1505).6 This edition is among the earliest Hebrew books, if not the earliest, printed with a title page.7 Although the title page in question (f. 1v) is actually a modest blank space with the mention of the title and the name of the author, it also indicates the name of the book’s proofreader, Meir son of Judah from Pesaro.

The present binding of smooth cream-white tawed leather on cardboard covers is relatively modern and was crafted at the BAV. Its spine contains the gold coat of arms of the Pope Pius VI (in office between 1779 and 1799), stamped above the title ‘Meir Aram. Comment. In Iob’ and the number ‘47’ (thus indicating that only one text is included in the volume). At the foot of the spine, there are the insignia of the Library’s twenty-fourth Librarian, Franciscus Saverio de Zelada, who held this position between 1779 and 1801. Therefore, the book in its present form was bound in the last quarter of the eighteenth century.

It is unclear whether the two printed Hebrew books had already been joined together before the present binding. One indication is that, on the last folio, after the colophon of the Rokeaḥ, the famous censor Domenico Jerosolimitano left his signature in 1597. Since no such censorship inscription appears in the Job commentary, it is possible that Domenico Jerosolimitano’s censorship mark concerned both books, already united then.

We cannot confirm whether or not the two books were transmitted and possibly bound together between 1517 (the date of the printing of the commentary on Job) and Jerosolimitano’s intervention. What is certain is that at least one of the printed books in the volume was bound when still in Jewish ownership.

The preliminary study of the two Hebrew parchment fragments of a Talmudic manuscript recycled as the front and back flyleaves of the present volume reveals a part of the story.8 It indicates that these two leaves were formerly pasted down on the inside of a previous book binding. This is shown by the traces of vegetal glue which had been spread evenly on the outer edges of what is now respectively the recto of the front flyleaf and the verso of the back flyleaf. The state of the text also provides some indications: rubbed and damaged on these pages. The damaged and glue-covered side of the folio was the one facing the earlier book cover boards. Pending a further material analysis, the first visual observation allows us to identify the traces of a dark brown material on the glue, suggesting that the book was probably once covered in dark brown leather. More importantly, there is a strong likelihood that the first brown binding onto which the Hebrew manuscript fragments were pasted covered the present printed volume or at least its first textual unit. This is indicated by a few Hebrew notes on the well-preserved side of the front flyleaf, the one which faced the book and which was free from glue in the previous binding. One of the notes, written by an early modern Sephardi Jewish hand, reads: ‮איוב ומאיר איוב‬‎, “Job and Meir Job”—which corresponds to the contents of the first textual unit of the present volume (Fig. 4).9 Thus, the sheets of a Hebrew manuscript were used to strengthen the binding of a Jewish printed book which was a part of a library of a Jewish reader.

The two parchment leaves come from the same original codex. The lower outer corner of the front leaf was damaged and restored during the most recent conservation. The leaves are relatively large. They measure 29 × 20 cm., and come from an even larger book, whose outer edges had been trimmed to fit the size of the newly bound book block. The text is written in long lines, in one block per page. The written space measures 20 × 16.5 cm. While for most recycled fragments no attention was paid to the text of this cheap, reused binding material (with the folios cut across the text, bound upside down, etc.), in this case, there is no damage to the text area. All four margins around the text are partly preserved.

Figure 1: BAV, R.G.Oriente.II.82, front flyleaf, hair side, detail (UV photograph)

Figure 1

BAV, R.G.Oriente.II.82, front flyleaf, hair side, detail (UV photograph)

Citation: The Vatican Library Review 3, 2 (2024) ; 10.1163/27728641-00302006

© 2024 Biblioteca Apostolica Vaticana
Figure 2: BAV, R.G.Oriente.II.82, front flyleaf, flesh side, detail

Figure 2

BAV, R.G.Oriente.II.82, front flyleaf, flesh side, detail

Citation: The Vatican Library Review 3, 2 (2024) ; 10.1163/27728641-00302006

© 2024 Biblioteca Apostolica Vaticana

The text is written on good quality parchment, which fared well despite its recycling. The parchment is cream-white on the flesh side and yellow on the hair side. In both fragments, it was the darker, hair side that was glued on the previous binding, while the ‘white’ flesh side remained visible and legible: this can be explained by aesthetics—the wish to expose a brighter and smoother side to the reader’s eye when the book is opened. In both fragments, the yellow hair side contains traces of animal follicles. This grain forms a long patch of very dark tiny spots, placed horizontally in the middle of the page, crossing it from left to right on both fragments. These are groupings of the follicles of a dark stripe along the animal’s (probably goat’s) spine (Fig. 1).

As both fragments contain this trace of the dark hair stripe, it is immediately apparent that both fragments come from the same original bifolio cut in two—the hypothesis will be able to be verified when studying the text. The manuscript is written in dark grey ink, probably of the iron-gall type. The text is guided by a grid of ruled lines made with a hard colourless point (traditionally a piece of reed or sharpened wood). The ruling pattern consists of one vertical line on each side of the block of text crossed by horizontal lines, which protrude in the inner margin. There is the same number of ruled and written lines (25 on each folio, on both recto and verso—naturally as the furrows on one parchment side automatically left relief lines on the reverse). These codicological features of parchment and ruling techniques are consistent with the Sephardi tradition. This is corroborated by other graphic aspects, which include the page layout as well as the script type, which, as we will see, is Sephardi square.

Even before reading the text, the first glance at the page and text layout suffices to define this work as a part of the Talmud. When the Talmud is concerned, Sephardi book makers continued an ancient Eastern and Maghrebi tradition of displaying the text in one large block on the page. The long lines are interrupted by blank spaces of the width of about six signs to separate paragraphs. The paragraphs end here with a punctuation mark consisting of two very short, slanted strokes parallel to each other, placed at the headline after the last letter of the paragraph (Fig. 2).

When the Gemara begins the discussion on a mishna of the Mishna chapter, it is marked by a blank space filled with the abbreviation ‮מתני׳‬‎ (for ‮מתניתא‬‎, “Mishna”), in conformity with the Sephardi design for Talmud manuscripts.10 This is attested on the hair side of the front leaf (Fig. 1).

A new baraita (ancient anonymous opinion not included in the Mishna) is introduced by the abbreviation ‮תנ׳‬‎ with a stroke on the left, followed by two sof pasuq-like dots (on front flyleaf, hair side, line 20).

The back flyleaf, on the flesh side, contains a larger space which is used to write the explicit formula of a Talmudic chapter ‮סליק פירקא‬‎, a usual text division marker in Sephardi Talmud manuscripts (Fig. 3).

The reading of the text gives further details on the structure and tradition to which these fragments belong. They contain together a continuous text of the Mishnah and Gemara of the tractate Soṭah, Mishna chapter 5. The preserved Mishna part corresponds to Soṭah 5: 1 (end of the mishna) to 5; the corresponding Gemara is BT Soṭah 27b to 30b. The Mishnah chapter is written uninterrupted, starting on what is now the flesh side of the back flyleaf, and the text on the hair sides follows directly: this means that the two flyleaves belong to one and the same bifolio and that they were the inner bifolio of a quire (see the order of the text in Table 1).

Table 1

The order of the text

Back flyleaf, Flesh side (Fig. 3)

Mishna Soṭah 5:1 (end)–5

BT Soṭah 27b (end)–28a

Incipit: ‮אחד לבעל ואחד לבועל‬‎

Explicit: ‮אסורה לבעל שנ׳ אם תטמאה‬‎

Back flyleaf, Hair side

BT Soṭah 28a–29a

Incipit: ‮[ונט]מאה ועדען תבעי‬‎

Explicit: ‮ארבעה קראי כתיבי‬‎

Front flyleaf, Hair Side (Fig. 1)

BT Soṭah 29a–29b

Incipit: ‮וחד לבועל ו[חד] לכהנה [ו]חד לתרומה‬‎

Explicit: ‮בתרומה ואינו עוש[ה]‬‎

Front flyleaf, Flesh Side (Fig. 2, 4)

BT Soṭah 29b–30b

Incipit: ‮רביעי בקדש‬‎

Explicit: ‮טומאה לחולין שבארץ ישראל‬‎

The text of this manuscript of the Babylonian Talmud is subdivided into chapters. Each Mishna chapter is copied in full. The corresponding Gemara follows. The incipits of each individual legal teaching (mishna) in the Mishna chapter appear within the Gemara, to introduce the discussions, and is preceded by the abbreviation ‮מתני׳‬‎, as mentioned above.

The critical study of the textual variants represented by these fragments of the BT Soṭah remains beyond the scope of this paper. Suffice to say that there are many differences from the modern Talmud editions, as represented by the Vilna Talmud edition.11 These variants include unintentional omissions of one or more words, different order of the passages, evident scribal mistakes, some of which have been corrected in the margins by a scrupulous reader. For example, on front flyleaf, flesh side, in TB 29b, an entire sentence (‮ממאי דר׳ יוסי כרבנן סבירא ליה דא׳מ׳ לטמא שנים ולפסל אחד אי כאבא שאול סבירא ליה‬‎) was forgotten by the scribe and added in the margin, with an editorial symbol in the shape of a circulus. There are as well genuine differences including the choice of different words (such as Aramaic ‮שרי‬‎ instead of ‮מותר‬‎, “permitted”, front flyleaf, flesh side, BT Soṭah 29b), slightly different expressions (such as ‮עפר בין עיניך‬‎ in the fragment, on back flyleaf, flesh side, Mishnah, Soṭah 5:2, versus ‮עפר בעיניך‬‎, as printed in the Vilna Talmud).

Even if some variants could be meaningful for the history of the transmission of the Talmudic text, most are fruit of the scribe’s inattention. It is noteworthy that the scribe who was so careless about the exact copy of the text displayed real calligraphic skills. His writing is of Sephardi type of calligraphic quality. The average letters are of medium size, ca. 3.5 × 3 mm., with larger words marking textual subdivisions: the letters of the introduction of a new mishna (‮מתני׳‬‎) measure ca. 8 × 7 mm. (tav), and the explicit of the Mishna chapter is even larger, with the letters reaching ca. 9 × 7–7.5 mm. The difference of the size allows the reader to navigate the manuscript more easily. Such emboldened letters are commonly found in Sephardi Talmud manuscripts from their earliest attested examples in the twelfth century, such as MS Hamburg, Cod. Hebr. 19, containing the tractates Bava Qamma, Bava Meṣiʿa and Bava Batra of the Babylonian Talmud, written in Gerona in 1184.

The characters are regular, vertically elongated, with a high degree of parallelism between vertical downstrokes, ascenders and descenders. The upper bars and the bases of the letters are parallel to the headline and baseline respectively. The characters have few additional elements: serifs of the upper bars of beth, daleth, he, kaph, samekh, resh and tav are short approach strokes rather than independent additional lines. The typically Sephardi features include the shape of the letter aleph, with its left-hand downstroke starting from some distance from the headline, descending to the baseline with a rounded, thickened ‘knee’ and ending with a foot turned to the left and descending slightly below the baseline with a tapering; gimel whose left-hand stroke is attached to the lower extremity of the stem and is almost parallel to the baseline; shin, with a base and whose one allograph contains the middle arm whose roof turns to the right. Although the dating of the Sephardi square calligraphic script is notoriously difficult due to its regularity and following of the scribal models, some feature, such as a ‘broken’ aspect of the apex of the mem and its body almost completely closed at the baseline, suggest the turn of the fourteenth century as a possible time of writing.

Figure 3: BAV, R.G.Oriente.II.82, back flyleaf, flesh side

Figure 3

BAV, R.G.Oriente.II.82, back flyleaf, flesh side

Citation: The Vatican Library Review 3, 2 (2024) ; 10.1163/27728641-00302006

© 2024 Biblioteca Apostolica Vaticana
Figure 4: BAV, R.G.Oriente.II.82, front flyleaf, flesh side

Figure 4

BAV, R.G.Oriente.II.82, front flyleaf, flesh side

Citation: The Vatican Library Review 3, 2 (2024) ; 10.1163/27728641-00302006

© 2024 Biblioteca Apostolica Vaticana
1

digi.vatlib.it.

2

Richler & Beit-Arié & Pasternak 2008.

3

www.hebrewmanuscript.com.

4

Cf. Steinschneider 1852–1860, col. 7. On this paper flyleaf, under the names of the book’s Jewish owners, there are two library entries: 1. Liber Job cum commentariis Rabbi Meir Aben Aràma impressus ubi nescitur in fortuna fano, anno mundi 5266, hoc A. Domini 1505. 2. Liber Rabbi Eliezer filii Juda quem nominem Rokeach. Tractat de judicialibus, qua hodie servantur. Fani anno mundi 5265, id est 1504. Incipit pag. 113. Omnes pag. sunt 226. For the Riva di Trento and Venice’s edition, cf. Heller 2004, 539.

5

Steinschneider 1852–1860, cols 916–917; Heller 2004, 11.

6

נשלם הרקח אשר חיבר הרב רבינו אלעזר בן רבי יהודה זל׳ בעיר פאנו בערב פסח שנת רסה׳‬‎

7

Heller 2013, 452–453.

8

I refer to the ‘front’ and ‘back’ flyleaves considering the book in the Hebrew order, from right to left.

9

One may infer that the note referred to two different books: a copy of Job as separate from Arama’s commentary, but the fact that the text of Job is printed in full in the body of the commentary justifies this description of the first textual unit of the volume.

10

A similar organisation of the text, with a block page layout and the Mishna text introduced by the abbreviation ‮מתנ׳‬‎ in large calligraphic characters is attested, for example, in the ‘Hamburg Talmud’: Staats- und Universitätsbibliothek Hamburg, Carl von Ossietzky, Cod. Hebr. 19, Gerona, 1184.

11

The Vilna Talmud (published between 1880 and 1886 by the publishing house of the Widow Romm and Brother in Vilna) represents to scholars today a convenient version of the Talmudic ‘textus receptus’ but is far from a proper critical edition. Unfortunately, the remarkable critical edition of Raphael Nathan Nata Rabbinovicz (Rabbinovicz 1867–1897) does not include the Tractate Soṭah. A great help for the study of the variants is the Hachi Garsinan project on the Friedberg Genizah Project website. It offers a synoptic edition in parallel columns of the major Talmudic manuscripts. For the relevant passage of the Tractate Soṭah, see https://bavli.genizah.org/ResultPages/Difference.

Bibliography

  • Heller, Marvin J. (2004): The Sixteenth Century Hebrew Book. An Abridged Thesaurus. Vol. I. Leiden: E.J. Brill (Brill’s Series in Jewish Studies 33.1).

    • Search Google Scholar
    • Export Citation
  • Heller, Marvin J. (2013): Essays on the Making of the Early Hebrew Book. Leiden: E.J. Brill (Brill’s Series in Jewish Studies 68).

  • Rabbinovicz, Raphael Nathan Nata (1867–1897): ספר דקדוקי סופרים […] עם הגהות דברי סופרים‬‎, Variae lectiones in Mischnam et in Talmud Babylonicum, quum ex aliis antiquissimis et scriptis et impressis tum e codice Monacensi praestantissimo collectae, annotationibus instructae. München: ex officina Aulae Regiae E. Huber.

    • Search Google Scholar
    • Export Citation
  • Richler, Benjamin & Beit-Arié, Malachi & Pasternak, Nurit (2008): Hebrew Manuscripts in the Vatican Library: Catalogue, Compiled by the Staff of the Institute of Microfilmed Hebrew Manuscripts in the Jewish National and University Library. Città del Vaticano: Biblioteca Apostolica Vaticana (Studi e Testi 438).

    • Search Google Scholar
    • Export Citation
  • Steinschneider, Moritz (1852–1860): Catalogus librorum Hebraeorum in Bibliotheca Bodleiana. Berlin: Typis Ad. Friedlaender.

Content Metrics

All Time Past 365 days Past 30 Days
Abstract Views 0 0 0
Full Text Views 56 56 19
PDF Views & Downloads 73 73 9