Latin authority and constructional transparency at work: neologisms in the French medical vocabulary of the Middle Ages and their fate.

The formation of Modern French scientific vocabulary has followed a number of strict morphological patterns since approx. the 18th century, called neoclassical composition (a.o. Cottez 1980, Corbin & Paul 2000, Fradin 2003, Namer 2009: 315 ff., Villoing 2012). Up till then, linguistic processes were much less uniform, during the Middle Ages in particular, a period of intensive translation activity characterized by a need for an adequate vocabulary in specific target languages. This project aims at investigating why certain French neologisms that emerged in the field of medicine during the Middle Ages managed to survive, while others disappeared after some time. Our hypothesis is that morphology, in particular constructional transparency, contributed in a crucial manner to lexical preservation: more specifically, words showing a close formal relation with the Latin equivalent from which they were borrowed, could stand the test of time better than original French creations, i.e. derivations or compositions on the basis of genuinely French morphemes. As such, they would sow the seeds of the neoclassical composition technique still in use today. The first reason why we hypothesize that a Latinized morphology guaranteed the survival of scientific terms is related to the specific situation of science communication in the Middle Ages, when science was almost exclusively expressed in Arabic, Greek or Latin (Lusignan 1989). In France, scholars learned to read and write in Latin and mostly worked with Latin texts. The first French translations were made by these scholars, for who it was easy to borrow a word from Latin and to interpret Latin loanwords when reading scientific texts in French. A second reason is the specific characteristic of Latin having a transparent lexical morphology, i.e. lexical morphemes (both stems and affixes) are stable throughout their combinations (e.g. noc-ere to harm, noc-ibilis harmful, noc-ivus hurtful all have an easily identifiable root noc-), whereas the (medieval) French lexicon displays an opaque structure (e.g. eau water aquatique aquatic évier sink, all derived from Lat. aqu-(a) water, but the common root is not identifiable anymore, cf. Goyens 2013). Several factors play an important role in the survival of scientific neologisms (Goyens 2013 and Goyens Van Tricht in press): success of a text, prestige of its author, the presence of explicative glosses, constancy of use of the neologism, frequency of the new word throughout several texts, etc. However, what we contend here is that morphological criteria also play a crucial role in this process. Although previous studies usually mention the importance of Latinisms or gallicized Latin words (e.g. Ducos 1998), scientific medieval vocabulary has never been the subject of an in-depth morphological analysis. A full-scale investigation of the medieval medical vocabulary from a morphological point of view is the objective ofthis research project. Our study will be based on a corpus of medieval medical texts, both translations from Latin and texts directly written in French, a great deal of which are already accessible in an electronic version. This corpus will be new in its kind, and will be available for linguistic study in its textual form, but also in a lemmatized form via our collaboration with the ATILF (DMF 2012: http://www.atilf.fr/dmf) in Nancy, avoiding problems linked to spelling irregularities. This collaboration will also provide us with tools for the semi-automatic extraction of words and word families from the texts, and will eventually lead to the creation of a database of medical terms which will be analyzed on the basis both of the above mentioned external criteria (success of the text, etc.) and of internal criteria such as (type and token) frequency of the word and morphological criteria. Our morphological criteria are partly based on psycholinguistic studies regarding the productivity of morphological constructions in Modern French (see Dal ed. 2003), such as the recognizability of affixes, the creation of word families with clear morphological links, the size of the word family, formal and semantic transparency of the morphological construction (composition or derivation), and the size of the word itself. These criteria will be implemented in the framework of Construction Morphology (in collaboration with Kristel Van Goethem,F.R.S.-FNRS & UC Louvain; cf. Booij 2010, Amiot & Van Goethem 2012). A multivariate statistical analysis should then reveal which criteria play the most significant role in the survival of neologisms, and allow us to verify whether our hypothesis (a word form close to Latin has the best odds of being preserved) is correct. Our results will eventually be confronted with the morphological characteristics of neoclassical compounds, in order to see if the seeds of these characteristics were already present in the French Middle Ages, and if so, to what extent. It should be stressed that the planned research has not yet been carried out with respect to the creation of the medieval French scientific, viz. medical vocabulary. This study will shed new light on the development and the fate of scientific terminology, and is ground-breaking from the point of view of its design, methodology and theoretical framework. The electronic database will serve the research community for other purposes as well.

Keywords:Middle Ages

Disciplines:Linguistics, Theory and methodology of linguistics, Other languages and literary studies