On examine les étapes par lesquelles en français le sujet non exprimé (S0) et le sujet postposé au verbe (VS) ont régressé, sans disparaître, jusqu’à l’état moderne, entre le 9e s. et le premier tiers du 17e s. Dès le milieu du 16e s., les taux d'expression et d'antéposition du sujet au verbe sont comparables à ceux du français moderne, la syntaxe du sujet conservant par ailleurs des constructions archaïques. Sur un corpus de dix-sept extraits de textes diversifiés (11e–20e s.), on analyse dans quel ordre et selon quelle chronologie les facteurs corrélés à S0 ou VS depuis l'ancien français ont disparu, à quel moment les traits régressifs S0 et VS sont devenus des traits distinctifs caractérisant des constructions spécifiques, à quel moment et dans quelles constructions se situent les derniers développements de ce changement, et à partir de quelle période se trouvent des textes offrant – et n'offrant que – la syntaxe du sujet moderne.

The present study investigates the functions and forms of conversational self-reformulation in spoken French. (Self-)Reformulations in general are a typical feature of unplanned and spontaneous conceptional orality (as opposed to conceptional distance; Koch and Oesterreicher, 1985). They exhibit retrospective modification of a reference expression, which is semantically equivalent in paraphrases and semantically different in corrections. The latter are therefore communicatively more problematic with regard to discourse intervention and turn-taking. As for the linguistic marking of self-reformulation, paraphrases are preferably introduced by lexically polyfunctional markers and prosodic deaccentuation, while corrections are marked by lexically monofunctional and prosodically overaccented structures. Since the accessibility to context-dependent forms is specifically related to conceptional orality, a more important linguistic marking of self-reformulation is hypothesized to occur in conceptional orality when compared to conceptional distance. The results of an empirical study contrasting two conceptionally different corpora suggest a generalization of paraphrastic markers in conceptional orality. This tendency is attributed to speaker-strategic routinization in which corrections are re-marked as paraphrases in order to avoid conversational intervention and, as a consequence, turn-taking. When taken over by other speakers, this routine may cause variation and, eventually, linguistic change.

The papers in this Special Issue present some of the results of the Multicultural London English/Multicultural Paris French project, supported by the UK Economic and Social Research Council (ESRC) from October 2010 to December 2014 and by the French Agence Nationale de la Recherche (ANR) from 2010–2012. The project compared language variation and change in multilingual areas of London and Paris, focusing on the language of young people of recent immigrant origin as well as that of young people whose families had lived in London or Paris for many generations. Similar projects in other European cities have documented the emergence of new ways of speaking and rapid language change in the dominant ‘host’ language, which are attributed to the direct and indirect effects of language contact; see, for example, Wiese 2009 on young people's language in Berlin, Quist 2008 on youth language in Copenhagen, and Svendsen and Røyneland 2008 on Norwegian). In London, young children from diverse linguistic backgrounds tend to acquire English in their peer groups at nursery school rather than from their parents, many of whom do not speak English or are in the early stages of learning English. Since their peers speak a wide range of different languages, the only language the young children have in common is English; and since many of their friends are also acquiring English, there is no clear target model, a high tolerance of linguistic variation, and plenty of scope for linguistic innovation. By the time they reach adolescence, young people's English has stabilized, and many innovations have become part of a new London dialect, now known as Multicultural London English (Cheshire et al., 2013). New urban dialects and language practices such as these have been termed ‘multiethnolects’: they contain a variable repertoire of innovative phonetic, grammatical, and discourse-pragmatic features. In multiethnic peer groups, where local children from many different linguistic backgrounds grow up together, the innovative features are used by speakers of all ethnicities, including those of local descent such as, in London, young monolingual English speakers from Cockney families. Nevertheless they tend to be more frequent in the speech of bilingual young people of recent immigrant origin, and by young speakers with highly multiethnic friendship groups (see further Quist 2008 for an account of the use of features associated with a multiethnolect in conjunction with nonlinguistic ‘markers’ of style, such as tastes in music and preferred ways of dressing). Our project aimed to determine whether a similar outcome had occurred in multicultural areas of Paris.

Feature Inheritance (Richards 2007) entails that uninterpretable features originate on the phase head C or v* and are then transferred to the associated Agree head, T or V. In the present article, it is argued that the French que–qui alternation is the locus of a Case contrast, implying that nominative Case originates on the complementiser and only becomes associated with T as a consequence of feature transfer. Que–qui thus provides new, Case-based empirical support for the theory of Feature Inheritance. The article also suggests that the que–qui alternation has an important implication for Chomsky's recent application of dynamic antisymmetry, reinterpreted in terms of labelling, to the issue of subject extraction failure. Specifically, the alternation appears to indicate that Case-matching is required, in addition to phi-feature agreement, in order for extraction to be blocked by labelling.

This article presents the results of a corpus study of prosodic rhythm in the urban vernaculars of 24 female and male adolescents featured in the MPF corpus (Gardner-Chloros et al., 2014). Using canonical rhythm metrics, among them the normalized Pairwise Variability Index (nPVI), we show that there is no clear effect of gender and only a small effect of cultural background on the variability of adjacent vocalic and consonantal duration intervals, typically correlated with more or less syllable-timed rhythm. However, female and male teens with multicultural background who clearly dominated their conversational exchanges with their peers and also used multiple phonetic features attributed to adolescent urban-vernaculars in French tended to show more variability in interval durations, pointing to more stress-timed rhythm. We discuss these findings in comparison with other urban contact varieties in Europe and North America. We speculate that rather than the leveling of phonological contrasts, as in London English, or societal pressures to conform to monolingual norms, as in French spoken in minority contexts in Ontario, Canada, rhythm-type differences in the present corpus are tied to speakers’ allophonic repertoires and best thought of as elements of interactional styles.

This article, based on a French oral corpus, discusses the standing and the interpretation of some vocalic false starts. It recaps the characteristics of false starts and presents the spoken data on which this study is grounded. It details the false starts in the corpus that begin with a vowel in order to draw some conclusions about their form, focusing on the differences between what would be expected and a few discrepancies. The analysis starts from the auditors' perception and proposes an interpretation of the effects of morphophonological constraints.

Nous décrivons dans cet article l’évolution des formes adverbiales ailleurs, d'ailleurs et par ailleurs de l'ancien français au français moderne. A l'aide d'une étude sur un large corpus diachronique, portant sur des occurrences tirées des bases de données BFM, DMF et Frantext, et combinant une approche qualitative et une approche quantitative, nous montrons l’évolution sémantique de l'adverbe ailleurs du concret à l'abstrait, et l’émergence successive de deux locutions adverbiales, d'ailleurs en français classique et par ailleurs en français moderne. Plus précisément, nous montrons que d'ailleurs puis par ailleurs, une fois lexicalisés, se grammaticalisent en marqueurs de discours, signalant un changement de topique de discours, avec des emplois proches mais distincts.

Les Canadiens francophones sont naturellement exposés à un environnement bilingue constant. L'influence de la langue officielle dominante – l'anglais – sur leur langue maternelle ne peut être ignorée. Vu la politique linguistique proactive actuellement en place dans la province de Québec, on peut s'attendre à ce que la presse écrite francophone soit très attentive à la protection de la langue qu'elle utilise. Cependant, plusieurs études (Théoret, 1991; Martel et al., 2001; Chaput, 2009) ont montré que la presse n’échappait pas au phénomène d'anglicisation tant les traces laissées par les contacts entre les deux cultures étaient visibles à l’écrit dans l'utilisation de termes anglais ou anglicisés. Se basant sur deux visions s'opposant sur le sujet, cet article se propose d’étudier si le contexte culturel et linguistique des lectorats a une influence sur le degré d'anglicisation de la presse écrite canadienne francophone. La première hypothèse admet que plus une communauté est exposée à l'anglais de façon répétée, plus la fréquence d'utilisation d'emprunts à l'anglais dans les journaux est élevée ; la deuxième se base sur le principe de survie de la langue qui veut que lorsqu'il y a insécurité linguistique, la protection du lexique est beaucoup plus importante. Basée sur un corpus composé d'une année entière de publications de trois quotidiens destinés à trois communautés francophones québécoises et outaouaise, cette étude analyse la fréquence d'utilisation de plus de 500 « anglicismes », plus précisément des emprunts lexicaux intégraux et hybrides. Les résultats obtenus semblent confirmer la deuxième hypothèse, vu que la région la plus bilingue des trois communautés affiche les chiffres les plus bas.

This article will review the parameters of a grammatical variable within the putative variety ‘Multicultural Paris French’, i.e. its distribution and use within a group of young banlieue speakers. The structure in question stands out as it has rarely been found in previous corpora in France: indirect questions following verbs like savoir, where the question word is post-verb (je sais pas il a dit quoi). We discuss which groups use the new forms in Paris, referring briefly to some comparable changes in London. This structure appears to be an instance of ‘change from below’ (Labov, 2007), which seems to have emerged in the speech of young people of immigrant background. It might also, on the other hand, be a long-standing vernacular variant, which has re-emerged, with specific identity-related significance, in this particular group of speakers. Its exceptional character in the Paris context highlights a lack of evidence for the emergence of a more wide-ranging, distinct multiethnolect, as found in London and other European capitals.

We analyse genre in the speech of young people with multi-ethnic friendship groups in Paris in order to address the as yet unresolved question of whether new quotatives equivalent in meaning to English BE LIKE result from simultaneous independent parallel developments in languages other than English or from a process of calquing. We conclude that French quotative genre results from independent internal developments, but that it enters the French quotative system in the same way that BE LIKE entered the English system, driven by the meanings of ‘similarity’ or ‘approximation’ that are shared by the lexical item genre in a range of syntactic categories. We propose that in order for a new similarity quotative to emerge, a lexical item with a meaning of ‘similarity’ or ‘approximation’ must become syntactically multifunctional, and that the use of that lexical item must reach a critical frequency threshold. In the case of genre we suggest that the necessary increase in frequency results from the development of the lexical item into a discourse marker. We also analyse another new French quotative, ETRE LA, a sequence that we find is used to highlight activity of many kinds (including, but not confined to, spoken behaviour). The trajectory followed by each of the new quotative expressions conforms to De Smet's proposals about how linguistic innovations spread through the grammar.

Cet article vise à examiner la mise en scène de l'accent marseillais dans le film le plus emblématique de Marseille, Marius (1931) et son remake de 2013 par une une analyse du comportement du schwa. La réalisation fréquente des e graphiques fait en effet partie des particularités phonético-phonologiques de l'accent prototypique de Marseille. Pourtant, aujourd'hui, le nombre de réalisations des schwas semble diminuer continuellement. Notre étude a pour but de découvrir à quel degré l'accent mis en scène diffère de l'accent des locuteurs marseillais actuels et si les deux films, qui ont été tournés avec 80 ans d’écart, soulignent le changement linguistique en cours dans le Midi. Les deux films ont été transcrits et codés selon le protocole du programme phonologie du Français contemporain (PFC). Par la suite, les données ont été comparées à celles des enquêtes PFC. Les résultats montrent que l'accent mis en scène se distingue nettement de l'accent authentique d'aujourd'hui. Le schwa ne se comporte cependant pas différemment dans les deux films.

This article examines early attestations of verlan and related backward slangs in French in the nineteenth century. Its main contribution is the edition and analysis of the only known text, a letter, written with features of verlan before the twentieth century. This largely predates other attested forms of verlan. The principles underlying this early form of verlan are shown to be different from contemporary verlan, as is much other early evidence, though all forms have the syllable as their basic unit. The letter is evidence that backward slangs can originate in the education system as much as in the underworld of thieves.

This study examines contact-induced change in Jèrriais, the severely endangered Norman variety currently spoken by some 1% of the population of Jersey, one of the British Channel Islands. Today, English dominates all linguistic domains of island life, and all speakers of Jèrriais are bilingual. The analysis uses original data to test empirically whether Myers-Scotton's (2002) five theoretical assumptions about the structural path of language attrition (broadly defined as language loss at the level of the individual) also have relevance for the process of language obsolescence (broadly defined as language loss at the level of the community). It explores i) whether Jèrriais is undergoing contact influenced language change owing to its abstract grammatical structure being split and recombined with English, a hypothesis related to Myers-Scotton's Abstract Level model; and ii) whether different morpheme types of Jèrriais are related to the production process in different ways and are, accordingly, more or less susceptible to change during the process of language obsolescence, a hypothesis related to Myers-Scotton's 4-M model. In addition to its contribution to linguistic theory, this study increases existing knowledge about Jèrriais and makes data from this language available for systematic comparison with other languages.

The difference between ‘car’ and ‘parce que’ is often explained in the literature by the type of causal relation (objective or subjective) that each connective prototypically conveys. Recent corpus studies have demonstrated, however, that this distinction does not hold in speech, and is fluctuating in writing. In this article, we present new empirical data to assess the status of this pair of connectives. In Experiment 1, we test French-speakers’ intuitions about ‘car’ and ‘parce que’ in a completion task, and compare these results with those of a similar experiment in Dutch. In Experiment 2, we measure the processing of objective and subjective causal relations containing ‘car’ and ‘parce que’ in an online reading experiment. Experiments 1 and 2 lead us to conclude that ‘car’ has to a large extent lost its specific procedural meaning. In the literature, the difference between ‘car’ and ‘parce que’ is also linked to a difference of register, ‘car’ being perceived as a formal equivalent of ‘parce que’. We assess the strength of this distinction in Experiment 3, by means of a completion task involving sentences from different registers. Results confirm the effect of register as a distinguishing factor between ‘car’ and ‘parce que’.

We report on a survey of language attitudes carried out as part of a project comparing youth language in Paris and London.1 As in similar studies carried out in London (Cheshire et al., 2008), Berlin (Wiese, 2009) and elsewhere (Boyd et al., 2015), the focus was on features considered typical of ‘contemporary urban vernaculars’ (Rampton, 2015).

The respondents were pupils aged 15–18 in two secondary schools in a working-class northern suburb of Paris. The survey included (1) a written questionnaire containing examples of features potentially undergoing change in contemporary French; (2) an analysis of reactions to extracts from the project data: participants were asked to comment on the speakers and the features identified.

Quantitative analysis had shown that some of these features are more widespread than others and are used by certain categories of speaker more than others (Gardner-Chloros and Secova, this volume). This study provides a qualitative dimension, showing that different features have different degrees of perceptual salience and acceptability. It demonstrates that youth varieties do not involve characteristic features being used as a ‘package’, and that such changes interact in a complex manner with attitudinal factors. The study also provides material for reflection on the role of attitude studies within sociolinguistic surveys.

This study highlights the complexity of French grammatical gender as a lexical property at the interface of morpho-phonology and the lexicon. French native speakers (n = 168) completed a gender assignment task with written stimuli illustrating common versus uncommon nouns, vowel-initial versus consonant-initial nouns, compounds and grammatical homonyms; they also indicated the strategies they used to assign a gender to stimuli. The findings showed strong lexical and gender effects suggesting that grammatical gender must be acquired for individual lexical items as morpho-phonological cues alone are unreliable and vary greatly.

This paper reports on language practices in the city of Strasbourg, in a multi-ethnic working class neighbourhood. This provides a comparative setting to identify whether linguistic features are spreading between French cities. Data were collected from young speakers (16 to 21) using an ethnographic approach over a year. First, this paper will briefly review the literature on language variation research in France. Second, a comparison of vernacular features will be carried out, focusing on lexical innovations, indirect questions following the verb savoir (Gardner-Chloros and Secova, this issue), quotative systems (Cheshire and Secova, this issue) and discourse markers. Finally, the ethnographic data collected as part of this research will be used to consider how multi-ethnic working class neighbourhoods in France are connected with each other, and how language may be travelling between settings.

Dans cette contribution, nous nous intéressons à l'origine et au développement de la construction [adverbe de degré + nom]: C'est très théâtre, c'est-à-dire très faux. De nos jours, cette construction se présente comme un puissant outil de recatégorisation contextuelle pour exprimer un rapport de ressemblance basé sur une analogie avec un concept nominal. Sur la base d'une recherche de corpus outillée, nous montrons que la construction-hôte [ADVdegré + ADJ] a pu s'ouvrir à la catégorie nominale à partir d'un petit noyau dur de noms humains, en partie qualitatifs, tirant profit de certaines propriétés structurelles du français de l’époque. Par la suite, la construction s'est diversifiée vers d'autres classes sémantiques (noms inanimés, noms propres, etc.) et a progressivement attiré une gamme plus diversifiée de types lexicaux dans son orbite. Elle n'a cessé de gagner en productivité, tout en réduisant son fonds lexical d'origine à la portion congrue.

This article assesses the sociolinguistic impact and importance of the other articles in this special issue on Paris, considering three main themes that are evoked. First, the contribution of the articles here to the development of work on language variation and change on Hexagonal French within the variationist paradigm. Second, I address what I see as the important contribution made to our understanding of the ‘city’ as a sociolinguistic site. Finally, I focus on ethnicity as a social construct in recent variationist work in cities and consider what the articles here, and in comparison with cities elsewhere, add to our understanding of the impact of immigration on local manifestations of language variability. In each case, I attempt to show how these articles foreground or even problematize these three issues, and provide a prospectus for further research that can address unresolved questions.