|next article in this issue|
Document Details :
Title: La sémantique du français technique
Subtitle: Une étude empirique et quantitative
Author(s): BERTELS, Ann
Journal: ITL - International Journal of Applied Linguistics
Volume: 157 Date: 2009
This article discusses the methodology and results of a quantitative semantic analysis (result of a PhD dissertation) of about 5000 pivotal terms (keywords) in the domain of machining terminology in French. Building on a double quantitative approach and corpus data (viz. a lemmatised corpus of French technical texts of about 1,7 million tokens), the investigation attempts to find out whether, and to what extent, pivotal lexical items are polysemous. The KeyWords Method was used in order to identify the most typical words. Next, a quantitative semantic analysis of the keywords determined their degree of monosemy, which was implemented in terms of degree of overlap between co-occurrents of co-occurrents of keywords. Finally, the quantitative data were submitted to a simple regression analysis, in order to check the hypothesis that the most typical terms are not always the most monosemous terms. In this article, we present the statistical results and linguistic interpretations of regression analyses on the 4717 keywords, on several subsets (nouns, adjectives, verbs, adverbs) and on the keywords of the four subcorpora (electronic reviews, technical files, technical standards, handbooks).