Diatopic Variation and the Study of Regional French

For many French speakers, the most salient form of linguistic variation is probably the way in which so-called ‘regional accents’ differ within France. Geolinguistic variation is certainly well known to students of French linguistics and forms a cornerstone of most introductory courses and coursebooks.¹ It is all the more striking, therefore, that no fully developed theoretical analysis of the precise nature and origins of regional French has yet been elaborated. This survey article reports on the directions that have been taken in past and current research in the field. It should be stressed at the outset that it is intended as an overview rather than as a comprehensive account.

The linguistic history of France in the early modern period is shot through with the ideology of standardization, or the formation of definite ideas about what is ‘correct’ in language use and the belief that all people should use language in the same ‘correct’ way. Thus, thanks to institutions as diverse as the salons and the Académie française, and to the grammarians and remarqueurs of the seventeenth century, France emerged into the modern period with a strong normative linguistic tradition, and an apparent idée fixe that any form of French other than le bon usage was unworthy of serious study. As has been commented upon elsewhere, such an ideology seems to have inhibited the elaboration of Labovian-style co-variationist studies of French within the Hexagon, a hallmark that has served to distinguish the study of linguistics in France from its counterpart in the Anglo-Saxon world. As Gadet noted in 1996, ‘il n’existe à peu près pas à ce jour de sociolinguistique variationniste française’.²

Despite this (perhaps ideologically rooted) reluctance to engage in studies of Labovian diastratic variation, the study of diatopic variation is far more established within the French context. Acknowledgement of regional variation has, after all, inevitably gone hand in hand with (though has been stigmatized by) the ideology of a ‘langue unifiée et homogène’. For the Abbé Grégoire, for instance, in his Rapport of 1794, ‘la nécéssité [. . .] d’universaliser l’usage de la langue française’ is coupled with that of ‘anéantir les patois’. Given the preoccupation with standardization that permeated France in the early modern period, it is hardly surprising that, as Pooley points out, many early works on French diatopic variation were normative, not to say pedagogical, in intent.³ Titles of such studies as Molard’s Le Mauvais Langage corrigé, ou recueil par ordre alphabétique, d’expressions et de phrases vicieuses usitées en France, et notamment à Lyon (1810) and Desgrouais’s Les Gasconismes corrigés, ouvrage utile à toutes les personnes qui veulent parler et écrire correctement, et principalement aux jeunes gens dont l’éducation n’est point encore formée (1812) reveal an all too clear aim of helping readers correct their ‘gasconismes’ and their ‘phrases vicieuses’ as they learnt Standard French.

The figure most often associated with the early days of the study of diatopic variation in France is Jules Gilliéron, co-founder of the Revue des patois gallo-romains (1887). Gilliéron’s most important legacy, of course, was the Atlas linguistique de la France, which he co-published with Edmond Edmont between 1902 and 1910 and which was largely responsible for shaping the study of regional variation in France during much of the twentieth century.⁴ The Atlas linguistique can boast a dual legacy within the Hexagon. First, it spawned a large number of regional linguistic atlases by professional linguists, recording for posterity many dialect forms now threatened by extinction.⁵ Second, it provoked a more general interest in lexical regionalisms among the broader French public, which has led to the production of a number of non-specialist dictionaries⁶ and other reference works.⁷

The concept of linguistic regionalism has therefore been a familiar one in France for over a century. It is notable, however, that during a large part of the twentieth century it tended to be studied within a largely descriptive rather than an analytical framework.

It is...