The majority of previous studies on nasal coarticulation in French find an inversely proportionate relationship between vowel opening and nasality, such that high vowels are the most nasalized, sometimes exceeding 50% nasality. However, it has been unclear whether this is a mechanical or controlled property of French, given the typically short duration of high vowels in natural speech, as well as the aerodynamic and acoustic factors rendering them more susceptible to spontaneous nasalization. This study uses nasometric data to quantify progressive and regressive nasalization in 20 Northern Metropolitan French speakers as a function of vowel height. Furthermore, the relationship between degree of nasal coupling and overall vowel duration serves as a proxy for distinguishing mechanical from controlled nasalization, in the spirit of Solé (1992, 2007). This study finds evidence that high vowel nasalization in French is mechanical in pre-nasal position, but controlled in post-nasal position. Meanwhile, nasalization of mid and low vowels is blocked in pre-nasal position but, at most, mechanical in post-nasal position. In consequence, French appears to block nasalization in otherwise lexically impossible positions (*ṼN), while passively allowing, though not actively requiring, nasalizing in positions where conflation is possible (both NṼ and NV being permitted in the lexicon).

Dont has been claimed to be an exception to the ‘subject island’ constraint (Tellier, 1991; Sportiche and Bellier, 1989; Heck, 2009) and to contrast with true relative pronouns such as de qui. We provide corpus data from a literary corpus (Frantext), which show that relativizing out of the subject is possible with dont and de qui in French relative clauses, and is even the most frequent use of both relative clauses. We show that it is not a recent innovation by comparing subcorpora from the beginning of the twentieth century and from the beginning of the twenty-first century. We also show, with an acceptability judgement task, that extraction out of the subject with de qui is well accepted. Why has this possibility been overlooked? We suggest that it may be because de qui relatives in general are less frequent than dont relatives (about 60 times less in our corpus). Turning to de qui interrogatives, we show that extraction out of the subject is not attested, and propose an explanation of the contrast with relative clauses. We conclude that in this respect, French does not seem to differ from other Romance languages.

Cette étude propose d’explorer et d’identifier des moments particuliers où le changement linguistique se produit, afin de confirmer ou de rejeter l’idée d’une période spécifique désignée par le terme « français préclassique », avec une rupture – ou frontière chronolectale – détectable autour de 1630 (cf. Ayres-Bennett et Caron, 2016). Afin de vérifier dans quelle mesure cette chronologie peut être confirmée, il est nécessaire de multiplier des analyses fines et pointues sur des traits linguistiques qui ont subi des changements à l’époque en question et d’interroger une gamme de textes qui reflètent la variation discursive et pragmatique, au lieu de consulter le canon des traditions textuelles actuellement disponibles sur des bases numérisées, qui sont essentiellement littéraires. C’est pourquoi nous avons consulté des sources de nature différente, qui pourraient attester des usages émergents, à savoir les corpus du Réseau Corpus Français Préclassique et Classique (RCFC). Seront présentés les résultats de deux études de cas (la recatégorisation des formes dedans/dessous/dessus/dehors et la montée des clitiques), abondamment discutés par les remarqueurs.

This article examines patterns of variation and change in the phonology of the regional French of Alsace, within an overarching framework of regional dialect levelling (Kerswill, 2003) in the French of France. Data are drawn from an original corpus gathered in Strasbourg and a small village in a rural area of the Bas-Rhin. We analyse two well-known regional features in spontaneous speech: (h), the variable realisation of initial [h], and (ʒ), the non-assimilatory devoicing of /ʒ/. We focus on the effect on the variation observed of the major extra-linguistic variables of age, gender and social class as well as urban or rural community. While the results for class and location follow expected patterns, whereby working-class and rural speakers show higher rates of traditional non-standard variants, the principal observation is the decline and, in the case of (ʒ), apparent loss of such features. We thus provide new evidence in support of supralocalization, not only in the urban context but also in the rural location. The results for gender are however less clear-cut: there is an interaction with age, class and location, and disruption of the usual pattern of female-led adoption of supralocal norms.

Based on two syntactically annotated corpora, and within the theoretical tradition of dependency grammar, the current study investigates the quantitative differences and similarities between written and spoken French. Our findings support the assumption that spoken and written French are two realizations of one language that do not differ in the syntactic categories, but in the frequency of these categories, and also in their organization in sentence. The subjects in spoken French are mostly pronouns, whereas in written French the subjects are mostly nouns and pronouns. Spoken and written French share many syntactic relations, but with different frequencies. For instance, dislocations are more diverse and frequent in spoken French. Spoken French and written French differ in the word order of vocative nominal phrases. Finally, written French is slightly more difficult to process than spoken French.