Uncovering Unknown Judeo-Romance Dialects in a Bodleian Miscellany (MS. Heb. g. 1)

This post is based on research undertaken during Dr Peter Nahon’s stay at the OCHJS as an Oxford School of Rare Jewish Languages Visiting Fellow in Hilary Term 2023. He teaches Judeo-Provençal at the OSRJL.


In southern France, local dialects of the Romance family which developed from Latin are heterogeneous and exhibit considerable territorial variation. As in other parts of the Diaspora where Jews have spoken and written varieties of local languages, it can be assumed that in southern France this dialectal variation was mirrored in the languages used by Jews of that region. That being said, until recently, only Judeo-Provençal—the dialect used in one single, narrow area of south-eastern France—had received extensive scholarly attention. A focus of my research in the past few years has been documenting and describing other insufficiently studied or unknown Southern French Romance varieties used by Jews—Judeo-Gascon, Judeo-Niçart and diverse secondary Jewish linguistic varieties of Southern French—thus yielding unprecedented insights into the linguistic diversity of the field.

Yet much remains to be done to expand our knowledge of Jewish languages in this linguistic region. Indeed, a wealth of material still remains in need of identification, description and linguistic analysis. It so happens that the Bodleian Library holds a manuscript miscellany that has tremendous potential for shedding light on whole swathes of Jewish linguistic variation. Oxford Bodleian Ms. Heb. g. 1 (cat. n. 2792 in Neubauer and Cowley’s Catalogue, 1906, vol. 2, p. 203-204) is miscellany of 197 folios, containing 21 different texts in several independent codicological units, of which a vast majority seem to have been written in the south and the east of France during the 15th century. The content of these texts is chiefly technical: medical and surgical topics, astronomic treatises, calendrical indications, charts of measure and weights, pharmacologic recipes and lists of therapeutic plants (materia medica). It also contains various materials attesting to deep cultural contacts between Jews and Christians: inter alia, a list of the books of the Old Testament ‘according to the count of the Christians’ (fol. 47), a list of the names used by Christians for the seven branches of learning (trivium and quadrivium) in Latin written in Hebrew letters (f. 61v) and a Christian calendar (ff. 181v-188v).

Fig. 1. Folio 187a. Part of the calendar, for the month of Tišri (תשרי) and September (שטומברא). The left column gives the name days of the Christian calendar of saints: Line 1—Saint Giles and Saint Ladre or Lazarus (ק׳ גילא וק׳ לאדרא), Line 4—Saint Martial de Limoges (ק׳ מרצייל דלימונגייש), Line 13—Saint Vernier or Werner (ק׳ וורנייר), Line 6—reads יריד בריאנשון ‘the fair of Briançon’ and Line 14—יריד רבייל ‘fair of Revel’, on the day of the Holy Cross (ק׳ קרוייץ).

The calendar, the only part of the miscellany that has been thoroughly described,[1] gives important clues about the dating and localisation of the manuscript: it is dated 1493 and refers to fairs or market days in Dauphiné (e.g., Briançon, Revel). Although the manuscript also contains some slightly later Italian material, it is very likely that the largest part of the miscellany was written and compiled in that area of south-eastern France by a single person—certainly a physician well-versed in the liberal arts—probably originating from northern France (as demonstrated by the palaeographic type of Hebrew handwriting used), who spent time in Dauphiné before immigrating to northern Italy.

Fig. 2. Folio 142b. A list of vernacular names of ‘hot’ seeds or species (זרעים חמים) used as remedies. In the first column, next to the Hebrew indication למחול קר ‘against the cold disease’, one reads קאפריש capres ‘capers’, שיקורייאיי chicorié ‘chicory’, טימי timi ‘thyme’, etc.; in the second column, שטאפיזאגרא stafizagre ‘stavesacre’, אפי api ‘celery, lovage’, אניזיי anize ‘anise’, אנייטי aneti ‘dill’, etc.

Such context of production has historically and linguistically significant ramifications. From the beginning of the Jewish presence in the kingdom of France and Burgundy, 15th-century Dauphiné was a crossroads for Jewish migrants en route to Savoy, Provence and, ultimately, Piedmont. It was there in Dauphiné that cultural and linguistic encounters occurred between its centuries-old Jewish community and exiles from neighbouring areas. However, material written by Jews in this area is extremely scarce: the few known items are all from earlier periods and comprise mostly formal codices, such as Bibles (e.g., Bodleian ms. Opp. 14, a 1340 Bible) or liturgical manuscripts (Parma ms. De Rossi 654), or very short, non-literary Hebrew documents (such as Grenoble ms. 4376, a betrothal contract from 1347). No other document of Jewish origin containing substantial matter written in the vernaculars of this dialectal zone—which extends across the linguistic border between Occitan and Franco-Provençal—has ever been described. Linguistic analysis shows that this unique manuscript furnishes a wealth of linguistic evidence in various dialects of south-eastern Gallo-Romance. While Hebrew is the main language used in the miscellany, the technical nature of its texts induces a frequent use of vernacular glosses, names or more extensive textual fragments. Reflecting the complex migration route and linguistic aptitudes of its writers, it includes words or phrases in Northern French, Franco-Provençal, Occitan, Latin and even Italian, all written in Hebrew characters.

Fig. 3. Folio 68a. A pharmacological recipe in Hebrew mixed with Romance: קח בצלים מדבריים דל שָוַויְיא ובשלם ‘take wild onions “del sâvaje” (Occitan “from the wilderness”) and cook them’. 

A thorough study of this material is needed. In the course of my ongoing OSRJL Visiting Fellowship at the Oxford Centre for Hebrew and Jewish Studies, I am currently assessing the exact linguistic contents of the miscellany in order to then extract and describe all the vernacular linguistic data in it. This data deserves to be transcribed, edited and analysed with the methodological tools of Romance linguistics and dialectology (linguistic atlases, Gallo-Romance lexicography) in order to locate it within the complex map of dialect variation in southern France. Its critical edition and analysis will broaden considerably our knowledge of the languages used by Romance-speaking Jews in the late Middle Ages, as well as the history of cultural, scientific and linguistic contacts to which this important manuscript is a unique witness.

[1] Stern, Sacha, 2016. ‘Christian Calendars in Medieval Hebrew Manuscripts’, Medieval Encounters 22(1):236–265.

Dr Peter Nahon 
OSRJL Visiting Fellow Hilary Term 2023, OCHJS; OSRJL Judeo-Provençal Teacher; Research Fellow in Romance
Dialectology and Linguistics,
University of Neuchâtel
Announcements Judeo-Greek Judeo-Provençal

Applications for Classes in Trinity Term 2023!

Applications for classes beginning in Trinity Term 2023—Judeo-Greek and Judeo-Provençal—are now open! To apply, please use this form. The deadline to submit applications is 23 March 2023 at 12:00 noon UK time.

Note: Applications for classes taking place during the 2023-24 academic year will be posted prior to each term.

Judeo-Arabic Judeo-Italian

The ‘Maḳre dardeḳe’ Manuscript with Judeo-Italian and Judeo-Arabic Glosses

On 8 August 1488, Joseph ben Jacob Ashkenazi Gunzenhauser printed in Naples an abridgement of David Kimḥi’s Sefer ha-shorashim entitled Maḳre dardeḳe (מקרי דרדקי). This work is attributed to Menaḥem ben Peretz Trabot and includes a number of Judeo-Italian and Judeo-Arabic glosses. It seems that it was intended for a readership of Italian Jews, as well as Arabophone Jews in Sicily and Sephardic Jews who had come to Italy from the Iberian Peninsula beginning in the late thirteenth century. What is remarkable about Gunzenhauser’s edition is that it is not only the earliest printing of Maḳre dardeḳe, but also the earliest dated witness of this text with Judeo-Italian and Judeo-Arabic glosses.

Maḳre dardeḳe, printed by Joseph ben Jacob Ashkenazi Gunzenhauser (Naples, 1488), München, Bayerische Staatsbibliothek, Res/2 A.hebr. 183.

Maḳre dardeḳe exists in manuscript form in six different manuscripts, dating from the end of the fifteenth century to the seventeenth century, and presenting notorious variants. These variants include the replacement of Judeo-Italian and Judeo-Arabic glosses by Yiddish or Judeo-Spanish glosses, or even the omission of glosses altogether. Three manuscripts contain Yiddish or Judeo-German glosses (Dortmund, Museum für Kunst und Kulturgeschichte, F 212, the oldest Maḳre dardeḳe witness; München, Bayerische Staatsbibliothek, Cod.r. 63; and Oxford, Bodleian Library MS Opp. 634); one contains Judeo-Spanish and Judeo-Arabic glosses (Oxford, Bodleian Library, MS Hunt. 218); another one has no glosses at all (Vatican, Biblioteca Apostolica Vaticana, ebr. 429); and the last one contains Judeo-Italian and Judeo-Arabic glosses (San Lorenzo de El Escorial, Real Biblioteca, MS G-IV-9). In other words, this last manuscript is the only one containing Judeo-Italian and Judeo-Arabic glosses, like the Naples edition, and therefore deserves special attention.

Maḳre dardeḳe, San Lorenzo de El Escorial, Real Biblioteca, MS G-IV-9, fol. 22v. © Patrimonio Nacional, Ministerio de la Presidencia.

The manuscript in El Escorial does not contain the entire text of Maḳre dardeḳe, but only an excerpt of it. In fact, the manuscript, with 224 folios, is a composite or non-homogeneous miscellany containing several exegetical works, some materials relating to Jacob ibn Ḥaviv’s ‘En Ya‘aḳov, the printed work Sha’are teshuvah by Yonah ben Abraham Girondi (probably a copy of the Venice edition, 1544) and other miscellaneous materials. The excerpt from Maḳre dardeḳe is copied on folios 22v to 24v and includes the beginning of the work (roots אב to אפז from the letter א). Previous mentions of this manuscript described it as containing Judeo-Italian glosses only, but folio 24v includes several Judeo-Arabic glosses as well.

Being a non-homogeneous miscellany, this manuscript is the result of previous existing materials having been gathered into a single binding. Yet this process was not a random selection of material done at the library at El Escorial. Rather, it had already been completed in sixteenth-century Italy by one of the manuscript’s owners, who selected the different constituent parts according to their own interests. We know this because when the manuscript arrived at El Escorial at the end of the sixteenth century, it was already in its present form. The book contains two different colophons, which describe the copying of parts of it in 1516 and 1518, and we therefore can surmise that the gathering of all materials was probably done later in the sixteenth century, sometime prior to the arrival of the manuscript at El Escorial in 1599. The different watermarks in the paper used for its many constituent parts also confirm this dating.

David Curbelo, a student under my supervision at the Universidad Complutense de Madrid, edited the brief text of Maḳre dardeḳe in this manuscript in his Master’s thesis and discovered that the text of the El Escorial manuscript is an abridgment of the text printed by Gunzenhauser. The El Escorial text retains the biblical roots, the Judeo-Italian glosses (and on folio 24v, the Judeo-Arabic glosses, too) and the biblical verses, although most of the exegetical texts and references from the Naples edition are absent. For instance, the first root in both texts (אב) is followed by the first Judeo-Italian gloss (מיור/מיורי, ‘maggiore’, translated as ‘elder’) and the biblical verse including the word אב with this meaning, וישימני לאב לפרעה (Gen 45:8). Then, the text in the El Escorial manuscript skips Rashi’s explanations that appear in the Naples edition, and provides the reading for the following gloss for אב, which is פטרי (‘patre’, translated as ‘father’), and so on. 

Many questions arise from this manuscript witness of Maḳre dardeḳe. Was the scribe in possession of a copy of the Naples edition that served as a model? Was the scribe interested in the glosses only, given the fact that most of the exegetical passages were omitted? Or was the scribe copying from other manuscript witnesses not known to us but that were circulating along with the Naples edition? And why were the Judeo-Arabic glosses copied only on the last page, and not from the beginning of the text? This being the only manuscript version of Maḳre dardeḳe with both Judeo-Italian and Judeo-Arabic glosses that was produced shortly after Gunzenhauser’s printing, the El Escorial manuscript definitely merits a thorough study. This study should include the glosses, their comparison with the ones included in the Naples edition and the nature of the relationship between both witnesses.

Professor Javier del Barco
Associate Professor of Hebrew Language and Literature,
Research Fellow at Institute of Religious Sciences (,
Universidad Complutense de Madrid
Search OpenEdition Search

You will be redirected to OpenEdition Search