A sociolinguistic study of grammatical variation in Martinique French

Abstract

This thesis is the first quantitative sociolinguistic study of grammatical variation in the French département et région d’outre-mer (DROM) of Martinique. Although Canadian varieties of French have been extensively studied within a Labovian framework, there is currently a dearth of variationist research focusing on varieties genetically related to European French. My doctoral work aims to address this gap in the research literature by presenting a sociolinguistic description of selected aspects of the variable grammar in a previously under-researched variety of French. The data for the present study were extracted from a corpus of spoken Martinique French which comprises approximately 16 hours of semi-directed sociolinguistic interviews conducted between December 2010 and February 2011. The analysis is based on a judgment sample of 32 native islanders from the Saint-Pierre arrondissement in the Northwest of Martinique, who were stratified by age, sex and educational level. Due to high levels of French/créole martiniquais bilingualism, informants’ frequency of use of French in interpersonal communication was measured using a modified version of Mougeon and Beniak’s (1991) languagerestriction index. My thesis specifically focuses on three morphosyntatic variables, namely the alternation between doubled and non-doubled subject NPs, the use/non-use of the morpheme ne in verbal negation and the variable expression of future temporal reference. I investigate the overall distribution of variant forms and determine whether the constraint systems for other varieties of French also hold in a Caribbean context. I also test and compare a range of statistical methods currently used in variationist research with the aim of providing a more comprehensive picture of the variable grammar in this regional DROM variety. Fixed-effects logistic regression models demonstrate that this Caribbean variety exhibits patterns of variation that distinguish it from other French speech communities. Mixed models further reorder and refine the respective constraint hierarchies, demonstrating the importance of considering random effects—such as individual speaker or lexical verb—when analysing sociolinguistic data. The combination of these statistical tools thus allows me to assess the extent to which such random effects constrain variation in Martinique French grammar. By contrasting variable usage in Martinique with that reported for communities in mainland France and francophone Canada, my doctoral thesis provides a localised as well as a global perspective on French morphosyntactic variability. This study therefore contributes to our understanding of the linguistic and social factors that unite and divide the French-speaking world. As such, it adds a French perspective to the extant literature on global linguistic trends (Meyerhoff & Niedzielski 2003; Buchstaller & D’Arcy 2009).