Cet article examine l’évolution de la négation phrastique en ancien français dans une optique sociohistorique. En ancien français, la négation simple (ne + verbe) se voit de plus en plus concurrencée par la variante novatrice renforcée (ne + verbe + pas/mie/point). Partant de l'hypothèse que la répartition des variantes conservatrice et novatrice varie en partie selon le registre, leurs fréquences sont analysées dans l'oral représenté et dans le récit, deux registres distincts au sein d'un même texte. Dans certains textes, la distribution des variantes s'avère conditionnée par le registre, l'oral représenté contenant davantage de négation renforcée que le récit. En outre, les données laissent entrevoir un effet du sexe, l'oral représenté des hommes étant plus novateur quant à la négation que celui des femmes. Ces résultats indiqueraient que le choix de forme de négation possédait une valeur sociostylistique à l’époque en question.

This article investigates variable future-time expression among native speakers of Hexagonal French who participated in informal conversations. The quantitative analysis is the first to examine the inflectional future, periphrastic future, and present indicative as separate forms within a single statistical model of French oral production. Results indicate that temporal distance and presence/absence of a temporal expression predict use of these verb forms. The second phase of the analysis focused on the use of the present indicative in future-time contexts. The examination of each instance of the present indicative shows that an immediate lexical temporal indicator is not necessary for this form to convey futurity and that future-time reference is often established at the discourse level and occasionally through apparent shared knowledge between the interlocutors. This investigation suggests the value of including the present indicative in the analysis of future-time reference in Hexagonal French in order to fully capture variation.

This paper examines the extent to which second language (L2) speakers of French acquire the semantic and pragmatic (or interpretive) properties associated with the c'est-cleft, specifically the exhaustive inference. This phenomenon is relevant to theories of language acquisition because it is situated at the interface of syntax and pragmatics. The results from a forced-choice task challenge the empirical adequacy of the interface hypothesis (Sorace, 2011, 2012; Sorace & Filiaci, 2006), which claims that external interfaces between a linguistic module and a cognitive module remain problematic even at the highest levels of L2 acquisition. The results from 40 L2 learners at three proficiency levels reveal development from nontargetlike to nativelike behavior. In particular, the high-proficiency group interprets the c'est-cleft, as well as canonical subject–verb–object sentences and sentences with exclusives (i.e., seul(ement) “only”), in a statistically identical way to the French native speaker control group.

This study examined how native and near-native speakers of Hexagonal French make reference to future events in a corpus of informal conversations. A concept-oriented analysis reveals that no fewer than 13 different finite verb forms appeared in future-time contexts. A qualitative analysis of the use of the present in future-time contexts in the two portions of the corpus points to similarities in the native-speaker and near-native-speaker use. This analysis contributes to the understanding of future-time expression in Hexagonal French and to discussions concerning near-nativeness in second language acquisition.

The position of object and adverbial clitics remains problematic in Old Occitan syntax (Wanner 2010). This paper analyzes clitic position specifically in affirmative main declaratives with overt preverbal subjects, in which clitics are either preverbal or postverbal with no apparent semantic distinction. Thus, the phrases En Constantiss’enanet and En Constantis anet s’en are semantically equivalent, each meaning ‘Sir Constantine left’, whether the clitics s’en ‘himself.from-there’ appear before or after the verb anet ‘went’. Previous analyses have concluded that this variation is random (Mériz 1978) or due to regional or dialectal variation (Hinzelin 2007). Neither approach satisfactorily addresses the underlying grammar or the principles underlying the distribution of the variants. The present analysis draws on claims about the left periphery in medieval Romance (Benincà 2006) and reports empirical data from the troubadour biographies (vidas and razos) and the vida of Saint Douceline. Results from 470 subject–verb declaratives establish that the subject in subject–verb–clitic sequences is left-dislocated, albeit covertly so. This sequence is one of several instantiations of subject left dislocation in Old Occitan and usually signals topic shift.

This paper analyses aspects of the loss of verb-second (V2) in Old French in a historical sociolinguistic perspective. Data come from sequences in which a main declarative is preceded immediately by a tensed subordinate clause (Vance, Donaldson, and Steiner, 2010). Following Romaine (1982), represented speech and narrative are considered to represent distinct registers within a single text. An analysis of intra-textual variation between narrative and represented speech reveals stylistic variation, in that represented speech evinces higher rates of surface subject-verb-(object) orders in main clauses than narrative passages within the same text. The data also reveal gender variation, as the represented speech of women tends to be more linguistically conservative than that of men.

The present study is concerned with the upper limits of SLA—specifically, mastery of the syntax-discourse interface in successful endstate learners of second-language (L2) French (near-native speakers). Left dislocation (LD) is a syntactic means of structuring spoken French discourse by marking topic. Its use requires speakers to coordinate syntactic and pragmatic or discursive knowledge, an interface at which L2 learners have been shown to encounter difficulties (e.g., Sorace, 1993; Sorace & Filiaci, 2006). The data come from (a) an 8.5-hr corpus that consists of recordings of 10 dyadic conversations between near-native and native speakers of French and (b) two contextualized paper and audio tasks that tested intuitions and preferences regarding LD. Analyses of the near-native speakers’ production of LDs, the syntactic properties of their LDs, and their use of LDs to promote different types of discourse referents to topic status suggest that their mastery of this aspect of discourse organization converges on that of native speakers.

This study investigates the manner in which syntax, prosody, and context interact when second- and fourth-semester college-level English-French learners process relative clause (RC) attachment to either the first noun phrase (NP1) or the second noun phrase (NP2) in complex nominal expressions such as le secrétaire du psychologue qui se promène (au centre ville) “the secretary of the psychologist who takes a walk (downtown).” Learners' interpretations were affected by the length of the RC, specifically its phonological weight. Effects of intonation contour were found only in a subset of learners. In a response time (RT) experiment that manipulated contexts, fourth-semester learners showed a final bias for NP1 attachment in interpretation but an initial RT bias for NP2 attachment. Second-semester learners also produced a NP2 attachment bias in RTs, but no asymmetry in interpretation was found. We argue that the processing of RC attachment by English-French learners requires a task-specific algorithm that implicates autonomous syntactic and prosodic computations and specific interactions among them.