TextmanuscriptTextmanuscripts - Les Enluminures

les Enluminures

Grammatical Miscellany, including BENJAMIN, Mevo ha-dikduk [The Introduction to Grammar], copied by Samuel; MEIR BEN SOLOMON BEN DAVID, Supplement to Petah devarai [The Beginning of My Words]; excerpts from SOLOMON BEN ABRAHAM IBN PARHON, Mahberet he-arukh [The Organized Notebook]; JOSEPH KIMHI, Sefer zikkaron [Book of Remembrance]; excerpt from PROFIAT DURAN, Sefer ma‘aseh efod [The Work of Profiat Duran]; and other grammatical, exegetical, and calendrical treatises

In Hebrew, manuscript on paper
Italy and Northern France or Germany, c. 1400-1450, c. 1450-1500; with additions c. 1500-1550

TM 983


i + 49 folios on paper, modern pagination in pencil in Arabic numerals in upper-outer corner on ff. 21v-46 (numbered 1-50), modern foliation in pencil in Arabic numerals in upper-left corner of recto throughout (cited), (collation i4 ii12 iii4 (through f. 20); i12, ii4, iii8, iv4 [+1]), layout varies: (I.),  ff. 1-20v, on paper, watermarks similar to Briquet 2777, “boeuf,” Turin, 1435, horizontal catchwords lower margin verso throughout (except ff. 7v, 14v, 15v, 17v), mostly ruled in blind with margins intermittently ruled in ink (justification 164 x 131 mm), written in two different compact Italian square and semi-cursive (incipits) and cursive (text body) scripts in a single-column text of 38-39 (ff. 1-16v) or 41 (ff. 17-20v) lines, enlarged incipits, new paragraphs sometimes indicated via indentation and/or triangular dot arrangements over the incipits, periodic vocalization, justification via dilation of final letters and use of anticipatory letters and space fillers, marginalia in hand of primary scribe throughout, sometimes partially cropped, episodic strikethroughs, poem on f. 1, manicules on ff. 2, 7, 9, diagram showing Hebrew grammatical paradigms laid out like the map of a city on f. 7v, marginal paragraph numeration on f. 10rv, decorative devices on ff. 11v-15; and (II), ff. 21-49v, watermarks similar to Briquet 3055, “cercle,” Innsbruck, 1452, no catchwords, ff. 47v-48 ruled in blind, otherwise unruled (justification varies), written in compact Ashkenazic (ff. 21v-46) and Italian (ff. 21, 46v-48, 49v) semi-cursive (incipits) and cursive (text body) scripts in a single-column text of 32-39 lines (ff. 21v-46 only; otherwise inconsistent), enlarged incipits, new paragraphs sometimes indicated via indentation, frequent vocalization, justification via dilation of final letters and use of anticipatory letters and space fillers, marginalia in later hand through f. 29 (see especially f. 24), sometimes partially cropped, episodic strikethroughs, pen trials on ff. 21, 48v, poems on ff. 21v, 46, marginal paragraph numeration on f. 47v, censors’ signatures on ff. 44v (Domenico Carretto, 1618) and 49v (Domenico J[e]rosolomitano; Aless[and]ro Scipione, 1597), slight scattered staining and/or smudging (see especially ff. 8-10v, 18-19v, 21v-22), minor dampstaining, several leaves reinforced along gutter, small holes in outer margin of f. 1, in center and outer edge of f. 21, and in gutter at head of f. 22, cuts in outer edge of f. 17, a folio following f. 20 was glued to f. 21 and subsequently partially torn so that only traces of the original page now remain, tape repair to f. 21 at head, ff. 48-49, once glued together, have been separated, although not entirely, resulting in damage to both folios, f. 49 almost completely loose now, tape repair to f. 49 at foot. Bound in dark blue library buckram, Montefiore name lettered in gilt along spine, Halberstam (157) and Montefiore (410) shelf marks taped to spine, spine slightly split along joints, light damage to headcap and tailcap, corners rounded, red speckled paper edges, modern paper pastedowns and flyleaves. Dimensions 197 x 147 mm.

Hebrew grammatical miscellanies bringing multiple linguistic works together in one codex were highly popular premodern study tools, often introducing grammar to Jews in Christian lands and co-existing with Latin grammars. The present exemplar boasts copies of two works, Mevo ha-dikduk and the supplement to Petah devarai, each of which has only been preserved in one other recorded manuscript, as well as a relatively early witness to Rabbi Joseph Kimhi’s seminal Sefer zikkaron.


1. The two treatises (ff. 1-20 and 21-48) that make up the present codex were copied separately and likely combined in the late fifteenth century. Ff. 21v-46 were copied by a single fifteenth-century Ashkenazic scribe, while ff. 1-21, 46v-48, 49v are the work of three distinct Italian scribes of the fifteenth and sixteenth centuries. One of them, Samuel, responsible for ff. 1-9, 10-11, 11v-16, indicated his participation by dotting the letters of his name at the beginning of successive lines on ff. 5, 15. Another, later fifteenth-century scribe copied text on ff. 9rv, 11, 16-21, 46v-48 (after the two treatises had been bound together), seemingly in Northern Italy, since f. 18v states: “I found several [grammatical] roots in the city of Legnano/Legnago.” Finally, a sixteenth-century copyist added some grammatical paradigms on f. 49v.

2. The second treatise (ff. 21-48) came into the possession of Eliezer bar Isaac (f. 21), and possibly also of Samuel Cohen (f. 48v), before being bound together with the first treatise.

3. According to a note on f. i, citing a letter by Moritz Steinschneider dated July 21, 1871, Samuel Schönblum (1833-1900), the most prominent nineteenth-century Judaica book dealer in the Austrian Empire, acquired the manuscript at some stage and tried to sell it to A. Asher & Co. in Berlin, but the latter declined to buy it. Nevertheless, the manuscript was included “by mistake” as no. 9 in the A. Asher & Co. inventory prepared by Steinschneider and published in 1868 (see the note by Steinschneider included at the end of Halberstam, 1890).

4. Solomon Joachim Halberstam (1832-1900), a wealthy Polish Jewish scholar and bibliophile who had acquired hundreds of valuable manuscripts from the libraries of Leopold Zunz (1794-1886) and Samuel David Luzzatto (1800-1865), bought this manuscript from Schönblum, according to the note in his hand on f. i, and included it when cataloging his own personal collection. The codex was in Halberstam’s possession as early as 1871, as attested by the aforementioned letter sent to him by Steinschneider (see also Steinschneider, 1871). The front flyleaf contains Halberstam’s Hebrew initials in pen below the manuscript’s shelf mark: “No 157.”  The shelf mark also appears on the spine, on the pastedown of the upper board, and on f. 1.

5. The Judith Lady Montefiore College in Ramsgate, England, purchased 412 manuscripts from Halberstam’s collection, including ours. The transaction was carried out by Rabbi Moses Gaster (1856-1939), principal of the College between 1891 and 1896. The manuscript contains the library stamp of the institution, known in Hebrew as Yeshivat Ohel Mosheh vi-Yehudit, on ff. 1 and 49, as well as the library’s shelf mark (MS 410) on both its spine and the pastedown of the upper board.

6. Between 1898 and 2001, most of the Montefiore manuscripts, including ours, were placed on permanent loan at Jews’ College in London. In 2001, they were returned to the Montefiore Endowment Committee.

7. In 2004, part of the Montefiore Collection, including our manuscript, was sold at auction by Sotheby’s in New York (lot 297).

8. Ariel Toaff (b. 1942), professor emeritus of Italian Jewish history at Bar-Ilan University in Israel, acquired the manuscript and affixed his bookplate to the pastedown of the upper board.

9. The manuscript was again offered at auction by Kedem Auctions in Jerusalem in 2017 (lot 11).


f. i, notes in Halberstam’s hand in pen about the purchase of the manuscript, as well as Kedem cataloging notes in pencil about the name of the scribe. On the verso is a table of the manuscript’s contents.

ff. 1-9, [Mevo ha-dikduk, an introduction to Hebrew grammar in ten chapters, authored by a certain Benjamin (as indicated by the acrostic of the poem at the beginning and in the lines immediately following)], incipit, “be-ezratekha elyon be-yadekha rum hevyon ahabber gillayon le-havin talmidim; explicit: berikh rahamana de-sayye‘an amen ve-amen ba[rukh] no[ten] la-[ya‘ef] ko[ah] u-[le-ein] o[nim] o[tsmah] ya[rbeh]”;

The work, clearly written by a seasoned teacher, draws on the philological scholarship of Rabbis Abraham ben Meir Ibn Ezra (1089-1164) and David Kimhi (1160?-1235?), as well as on Petah devarai, composed by an author named David (see TM 982).

In 1885, Wilhelm Bacher, the great Hungarian Jewish historian of Hebrew linguistics, published a paper in which he used our manuscript to identify the author of this treatise as the grammarian and Bible commentator Rabbi Benjamin ben Judah Bozecco (Bozecchi) of Rome (1290-1335) (see also Bacher, 1892). At the time, Bozecco was known to have composed a Hakdamah (introduction) to Hebrew grammar that was often printed together with Rabbi Moses Kimhi’s (d. c. 1190) Mahalakh shevilei ha-da‘at (first edition in which they were printed together: Pesaro, 1508). By comparing the (published) Hakdamah and (newly-discovered manuscript of) Mevo ha-dikduk, Bacher concluded that the two works were both composed by Bozecco and that the former was simply an earlier version of what the author would later revise and expand into Mevo ha-dikduk. Because Mevo ha-dikduk makes use of Petah devarai, Bacher was able to date the latter to no later than the second third of the thirteenth century. (He also noted that Mevo ha-dikduk is likely the first Hebrew grammar to use the vernacular – in this case, Judeo-Italian – to teach conjugation paradigms [see f. 7v].)

During his comprehensive, close textual studies of Petah devarai, David Sámuel Lőwinger, another Hungarian Jewish scholar, took issue with Bacher’s attribution of Mevo ha-dikduk to Bozecco (1931, 1937-1938). He demonstrated that Bozecco, in his published Hakdamah, and the author of Mevo ha-dikduk had different goals and writing styles and that each work addresses topics that the other does not (they also disagree on a few points). Lőwinger therefore asserted that although the Benjamin who wrote Mevo ha-dikduk was influenced by Benjamin Bozecco’s Hakdamah and that the two Benjamins probably lived around the same time, they were not, in fact, one and the same person.

The confusion caused by Bacher’s attribution of Mevo ha-dikduk to Bozecco has led catalogers of antiquarian Hebrew books to mistakenly refer to manuscripts of Bozecco’s Hakdamah as “Mevo ha-dikduk,” when in fact the two works should be kept separate. In that light, only two copies of the real Mevo ha-dikduk, ours and MS Budapest, Országos Rabbiképző Intézet K 53 (Italian, sixteenth century), have survived. Lőwinger printed the text of our manuscript in 1931, using the Budapest exemplar for variant readings;

f. 9rv, A series of Hebrew verbal conjugation paradigms using the root p-k-d accompanied by Judeo-Italian translation, written in a later hand;

ff. 10-11, [Short, anonymous, and apparently unpublished treatise on Hebrew grammar], incipit, “yir’at ha-shem reshit da‘at shelomoh ha-melekh a[lav] ha-[shalom] ramaz be-zeh ha-pasuk el yesod ha-dibbur … ve-ha-shevi‘i hi [sic!] ha-makor be-hidrug bakhlam immo”;

f. 11, Note in a later hand suggesting that Kimhi’s aforementioned Mahalakh shevilei ha-da‘at should follow here;

Perhaps the exemplar from which the scribe was copying did in fact place Kimhi’s work directly after the previous treatise. Below the note is a three-part discussion of Rabbi Solomon ben Isaac’s (1040-1105) comment to Ex. 6:9, s.v. ve-rabboteinu derashuhu, probably copied from the same Tosafist collection from which material on ff. 12v-16 below was taken.

ff. 11v-12, Basic Jewish calendrical rules and calculations;

ff. 12v-16, [Comments of various Northern French medieval scholars, known as Tosafists, on the weekly Torah portions Be-reshit through Va-yiggash (Gen. 1:1-47:27; a note by the scribe at the top of f. 16 recognizes that he skipped the portion Mi-kets, Gen. 41:1-44:17)], incipit, “be-reshit bara amar ra[bbi] yitshak lo hayah tsarikh le-hathil et ha-torah ella me-ha-hodesh ha-zeh lakhem … ve-ye[sh] lo[mar] le-paresh gam attah ke-divreikhem ken hu kemo she-amartem aleikhem she-tihyu la-avadim”;

Much of the material here matches that printed in Minhat yehudah (editio princeps. Livorno, 1783), a commentary on the Torah written at the beginning of the fourteenth century by Rabbi Judah ben Eleazar. The comments on each portion (except Be-reshit and Va-yiggash) are followed by explanations of difficult words in R. Solomon ben Isaac’s commentary on that portion. A critical edition of Minhat yehudah on Genesis was published in 2012 by Hazoniel Touitou based on the editio princeps and seven manuscripts.

ff. 16-18v, [in a later hand, Meir ben Solomon ben David’s supplement to his grandfather’s Petah devarai (see TM 982) in seven brief chapters], incipit, “ani me’ir be-ra[bbi] shelomoh be-ra[bbi] david ha-katan ba-alafi[m] ve-ha-tsa‘ir be-beit avi le-ba’er inyanim hutsrakhti … me-asotah me-asot otah tam ve-nishlam shevah le-bore olam”;

The author writes that matters his grandfather had taken for granted in his own day were no longer generally well known and therefore had to be explained, prompting him to compose this treatise.

In 1885, Bacher proposed, based on our manuscript, that Petah devarai was written by a man named David, given Meir’s patronymics (i.e., son of Solomon son of David), although Bacher acknowledged that it is not clear from the author’s language whether Petah devarai was written by his paternal or maternal grandfather. As with Mevo ha-dikduk above, this treatise appears to survive in only two copies, ours and another in a private collection (formerly Budapest, Országos Rabbiképző Intézet, MS K 59). Lőwinger published the text of the Budapest exemplar in 1931, using our manuscript for variant readings.

ff. 18v-20v, In the same later hand, excerpts from Rabbi Solomon ben Abraham Ibn Parhon’s (d. c. 1167) Mahberet he-arukh, a biblical lexicon completed in Salerno, Italy, in 1160, based in large part on the work of Rabbis Judah ben David Hayyuj (c. 945-c. 1000), Jonah Ibn Janah (Abu al-Walid Marwan; c. 990-c. 1050), and Ibn Ezra;

These excerpts were taken from Ibn Parhon’s entries on five roots: n-s-h, ts-‘-d, h-t-n, sh-l-h, and g-y-d. Mahberet he-arukh, which played an important role in transmitting Sephardic Hebrew linguistic scholarship to the (non-Arabic-speaking) Jews of Christian lands, was only printed once, in Pressburg in 1844 (photo-offset: Jerusalem, 1970), and would benefit from a critical edition based on the approximately twelve substantial manuscripts that have come down to us.

f. 21, Remnants of a torn page on which are written Hebrew verbal conjugation paradigms using the root p-k-d. (The page is glued to the front of a second treatise featuring an owner’s mark and pen trials);

ff. 21v-46, [Sefer zikkaron, an important Hebrew grammar by exegete, translator, and polemist Rabbi Joseph Kimhi (c. 1111-c. 1170)], incipit, “mivta leshon kodesh ashav nokhhi u-le-ilgei safah ahav ruhi; explicit: ve-az tavin yesod ivri ke-dato asher tsivvah e-loheinu le-mosheh siyyamti sefer ha-zikkaron te[hillah] le-[e-l] rishon ve-aharon”;

Born in Spain, Kimhi escaped to Narbonne around 1150 in the wake of the Almohad persecutions. There he taught Bible, Talmud, and Hebrew grammar and composed Sefer zikkaron (whose title is taken from Mal. 3:16) in an early attempt, following Ibn Ezra’s lead, to introduce the Jews of Christian lands to Sephardic Hebrew linguistics. The book, drawing on the work of Hayyuj and Ibn Janah, surpasses Ibn Ezra’s in the clarity of its presentation and in the uniformity of its treatment of the material. Under the influence of Latin grammar, it also revolutionized the understanding of the vowel system by claiming that Hebrew had ten (not seven, as had been previously claimed) vowels: five qualities (a, e, i, o, u), each with two lengths (long and short). Through the agency of his two famous sons, Rabbis Moses and David Kimhi, Joseph’s vocalic classification system would gain acceptance throughout the Jewish world; indeed, it is still taught to this day.

Sefer zikkaron has survived in about twenty other complete or partial manuscripts housed in various public European and Russian collections, the earliest of which, Oxford, Bodleian Library, MS Opp. 625, was copied in Ashkenazic lands in the thirteenth century. The book was published for the first time in 1888 (photo-offset: Jerusalem, 1968) by Wilhelm Bacher using our copy as the base text, with variant readings taken from manuscripts in Berlin, Munich, and Parma. At the end of the present exemplar is a three-line riddle-like poem about the vocalization of the imperative, also apparently composed by Kimhi, followed by a paragraph of explanation.

ff. 46v-47, [Ha-binyanim ve-ha-peratim me-ha-dikduk, a short, anonymous, and apparently unpublished summary of the most important rules of Hebrew grammar], incipit, “ha-shem nehelak li-[shenei] halakim etsem u-mikreh; explicit: ve-shalo[m] al yisra’el bi-kelal u-perat golat ari’el ve-yesha yekarev”;

ff. 47v-48, Brief excerpt from physician, philosopher, and polemist Profiat Duran’s (d. c. 1414) Sefer ma‘aseh efod (editio princeps: Vienna, 1865; photo-offset: Jerusalem, 1970), completed in 1403, which at the time constituted one of the most significant works of Hebrew grammar composed since the appearance of David Kimhi’s Mikhlol over a century and a half earlier. (Aharon Maman notes the need to reprint this book based on manuscripts not used by its original editors [2003]);

The text lists twenty-seven or twenty-eight types of Hebrew verbal roots, with examples, and follows them with some quick observations on the positioning of “weak” letters within a given root.

f. 48, Short, anonymous epitome of Hebrew grammar, using the root p-‘-l in its verbal paradigms; f. 48v, pen trials and a possible owner’s mark; f. 49, Montefiore library stamp;

f. 49v, another series of Hebrew verbal conjugation paradigms using the root p-k-d accompanied by Judeo-Italian translation, followed by a brief remark on the difference between transitive and intransitive verbs.

Hebrew linguistics, the handmaiden of biblical study, has been an important field of Jewish scholarship at least since the late ninth or early tenth century. In cataloging the philological manuscript holdings of the Jewish Theological Seminary in New York, Aharon Maman has shown that works on Hebrew grammar and lexicology were composed and/or copied throughout the generations (2003, 2006). Indeed, of the Seminary’s approximately ten thousand Hebrew manuscripts and twenty-seven thousand Genizah fragments, 406 and 306, respectively, concern linguistics. The 406 full-length codices comprise 158 distinct grammatical works, the authors of 141 of which are known by name.

Many linguistic treatises were copied or bound together in grammatical miscellanies like ours. These were often joined by calendrical or kabbalistic tracts, demonstrating their integration into the wide range of Jewish studies. The present exemplar boasts copies of two works, Mevo ha-dikduk and the supplement to Petah devarai, each of which has only been preserved in one other recorded manuscript, as well as a relatively early witness to Kimhi’s seminal Sefer zikkaron.


Bacher, Wilhelm. “Un abrégé de grammaire hébraïque de Benjamin ben Juda, de Rome, et le Pétah Debaraï,” Revue des Études Juives 10.19-20 (January-June 1885), pp. 123-144.

Bacher, Wilhelm. Die Hebräische Sprachwissenschaft vom 10. bis zum 16. Jahrhundert, pp. 71-74, 91-92, Trier, 1892.

Cohen, Mordechai Z. “The Qimhi Family,” in Hebrew Bible / Old Testament: The History of Its Interpretation, ed. by Magne Sæbø, vol. 1, pt. 2, pp. 388-415, Göttingen, 2000.

Duran, Profiat. Sefer ma‘aseh efod, ed. by Jonathan Friedländer and Jakob Kohn, pp. 87-88, Vienna, 1865.

Halberstam, Solomon Joachim. Kohelet shelomoh, pp. 21 (MS 157), 136, 145, Vienna, 1890.

Hirschfeld, Hartwig. Descriptive Catalogue of the Hebrew Mss. of the Montefiore Library, pp. 122-123 (MS 410), London and New York, 1904.

Ibn Parhon, Solomon ben Abraham. Mahberet he-arukh, ed. by Salomo Gottlieb Stern, Pressburg, 1844.

Judah ben Eleazar. Minhat yehudah: peirush la-torah me-et rabbi yehudah ben el‘azar me-rabboteinu ba‘alei ha-tosafot, vol. 1, ed. by Hazoniel Touitou, Jerusalem, 2012.

Kedem Auctions. Auction 55, Part I: Rare and Important Items, pp. 25-27 (lot 11), Jerusalem, 2017.

Kimhi, Joseph. Sefer zikkaron, ed. by Wilhelm Bacher, Berlin, 1888.

Lőwinger, David Sámuel. Két középkori héber grammatikáról (a Petach Debarai és a Mebó Hadikduk), Budapest, 1931.

Lőwinger, David Sámuel. “Le-toledot ha-dikduk ha-ivri (petah devarai – mevo ha-dikduk),” Ha-soker 5 (1937-1938), pp. 52-74.

Maman, Aharon. “Millonim u-gelosarim me-osef beit ha-midrash le-rabbanim ba-america: he‘arot mavo,” Leshonenu 65 (2003), pp. 303-314.

Maman, Aharon. Otserot lashon: kitvei ha-yad ve-kit‘ei ha-genizah be-hokhmat ha-lashon me-osef beit ha-midrash le-rabbanim ba-america, New York and Jerusalem, 2006.

Sotheby’s. Important Hebrew Manuscripts from the Montefiore Endowment, p. 351 (lot 297), New York, 2004.

Steinschneider, Moritz. Verzeichniss hebräischer Handschriften und seltener Drucke aus dem antiquarischen Lager von A. Asher & Co. in Berlin, p. 2 (no. 9), Berlin, 1868.

Steinschneider, Moritz. “Miscellen,” Ha-mazkir = Hebræische Bibliographie: Blätter für neuere und ältere Literatur des Judenthums 11 (64) (July-August 1871), pp. 103-106, at p. 104.

Online Resources

Our MS (accessible from within the National Library of Israel)

TM 983