Retour à la page d'accueil
FORSE (Fonds Rabelais et ses Sources En ligne)

A Rabelais database and his sources
Works and Images, main Editions and Manuscripts



Janvier 2008

Abstract

A primary "Rabelais" database, a pioneering project, was published online in 1995, but at the present time, neither researchers nor the public can find on the Web, or even in a printed  publication, the whole corpus of texts, written, or presumably written, by Rabelais. The FORSE project has  been developed by several contributors who have an ample experience both of electronic publishing and of  problems associated with editing of such a complex author. This research team (from Tours, Paris, Limoges, Lille and Poitiers, but also from other countries) wishes to extend this expertise to a broad field consisting of the context in which the works of Rabelais were produced. Thanks to funding from local governments, which brings considerable co-financing, the Centre for Renaissance Studies in Tours possesses a suitable insfrastructure and will assume the task of gathering around Rabelais an important body of material, the kernel of which is the complete works themselves in their major versions. From this kernel an access to two secondary corpora will be possible, one involving antecedents —the most important sources— the other decendants, with the early imitators of Rabelais. The members of the team gathered around this main corpus, most of whom have been involved since 2003 in the "Bibliothèques Virtuelles Humanistes" (CESR), will consider FORSE as the main determinant, during three years, for selecting works to be digitized. The context is particularly encouraging, since the CESR has been accepted in December 2006 as an "associate member" of the Bibliothèque nationale de  France: this membership should facilitate access to rare documents concerning Rabelais and their integration into the corpus.
Three workpackages will be pursued in parallel. The first one involves assembling a corpus of documents and various databases in progress, and completing the collection with documents not available for the moment in digitized format (manuscripts or printed texts): the point of departure will be Rabelais’ works, together with his readings and the circles he may have haunted. As many transcriptions as possible will be offered, so that access to full texts will be provided. The corpus itself is divided into three parts:
1) more than 30 editions of Rabelais' works, published between 1532 and 1564, the main items being in textual form, accompanied by manuscripts, autograph manuscripts (or so-called), and by a hitherto unpublished iconography ;
2) a hundred sourcebooks, some of them being very well-known (such as Erasmus' Adagia), or relatively unknown as the Lectiones antiquae of Caelius Rhodiginus: to them we join unpublished manuscripts originating in humanist circles, some of which will merit a full text transcription ;
3) between 30 and 50 books either showing the obvious influence of Rabelais (translations or adaptations), or mentioning Rabelais to criticize or to praise him. The purpose of making these documents available is to provide primary sources, for interpretation as much as for biographical information. In order to avoid producing a gigantic encyclopedia, the terminus ad quem will be marked by the publication of the most peculiar imitator of Rabelais, François Béroalde de Verville, in his last work, the Moyen de parvenir (1610-1616), a moment that coincides with the end of the era of Henri IV. The oldest translations (in German, Dutch, and English) will also be part of the corpus.
This group of nearly 200 documents, comprising 400 000 images and 15 000 transcribed pages (30 Mo), will supply material for the second workpackage, that is the task of processing, which will be carried out using specialized tools, all of which are in operation in 2008 : if image acquisition is not longer a major problem, thanks to the establishment of technical standards (slightly compressed jpeg files), text acquisition will be notably speeded up, and less expensive, using the software that have been developed with the computing teams that are our partners in the Navidomass project. For automatic page layout, AGORA helps by dividing graphic elements and text blocks; for automatic old-character recognition (RETRO), manual intervention will be limited to verification and encoding. Thus, the project will offer innovative features: the images will comprise not only text in image form, but all the graphic elements extracted from the digitized documents, properly indexed and accessible by means of a double standard search engine, connected to the Iconclass thesaurus. Also unprecedented is the work of manuscript identification and authentification, that will be undertaken using specific tools for handwriting analysis, and by comparison with a database of humanistic "hands", developed by the Institute for Research and Text History (IRHT). It will be thus possible to reconsider the problem of authenticity for some documents, with the aid of this new technological tool. Finally, the third workpackage will involve a simultaneous visualisation of images and texts, rendered possible by the organisation of metadata and XML/TEI encoding: during the academic year 2007-2008, in collaboration with other research teams dealing with old and middle French texts, the team has been working on a style sheet, submitted to this community and compatible with the different existing OAI protocols, and with the standards of the Europeana portal. It is important that these image-and-text resources constitute a true corpus with its own distinctiveness: it will also be the base for a wider repository about Rabelais and his era, which will be an efficient means of diffusing the results of the research.

Marie-Luce Demonet
marie-luce.demonet@univ-tours.fr