Contenu du sommaire : Les marqueurs discursifs

This article deals with a group of discourse markers that resemble sentences containing a finite verb, from a formal point of view, but that function as discourse markers which are invariable in form and which cannot govern other sentence elements. A distinction can be made between two different groups of such sentential DMs, characterized respectively by the first person singular (je pense, je crois, je trouve) and the second person singular or plural (tu sais/vous savez, tu vois/vous voyez). Imperative constructions can be added to form a further group of such items. The aim is to show that these expressions have undergone a grammaticalization from free verbs to discourse markers, that they have lost their ability to be governing verbs ( “verbes recteurs”), and that this can be seen in the following traits : morphological invariability, optionality, free position, no propositional contribution to the utterance, and subjective meaning. Their function becomes that of a phrasal adverb. Semantically, the functions of the governing verb and the discourse marker are not totally different, but the basic meaning of the governing verb is modified or weakened in the discourse marker function. Sentential discourse markers have a phatic function in oral interactions and provide structure at a syntactic and thematic as well as at a discourse level. The first group is not fully grammaticalized as a discourse marker in conventional (written) French, since it is unacceptable in the first position where the use of the complementizer que is obligatory, even though the structure without que can be found in Middle French. In the written mode it is thus not possible to distinguish between a governing verb and a parenthetical use of the verb. A third group might be added since it is possible to discuss the dependency relation between quotation verbs, traditionally analysed as governing verbs, and quotes.

This article aims to contribute to the study of discourse markers (DMs) which arise out of structures which are syntactically (pseudo-) subordinated ; such markers are also known as “parentheticals” or “propositional discourse markers”. A working corpus of examples was built up on the basis of a thorough examination of several corpora of spoken French (mainly European varieties). First of all, the morpho-syntactic variation of the expression (il) faut dire (que) is analyzed, specifically in terms of the presence or absence of the neutral subject pronoun il and the complementizer que, and of the positional variability of (il) faut dire (que) with regard to the complement clause that it is supposed to govern. Then, the question is raised whether (il) faut dire (que) actually functions as a DM or if it is more akin to a textual connective, comparable in this to the expression je dois dire que studied by Kronning (1988). This part of the analysis leads to the conclusion that, although (il) faut dire (que) may occur in contexts permitting a connective reading, in most of the attested oral examples it works indeed as a real DM with an interlocutive value involving explicativity and subjectivity.

Although several studies have described the use of discourse markers (DMs) from a syntagmatic point of view, the possible reduplication of some DMs has never been studied in detail (ex. in French : bon bon, tiens tiens, voyons voyons, là là, bien bien and in English, OK OK, all right all right...). The present article has two aims. First, it seeks to analyze the role of the pragmatic reduplication (PR) of the là marker in Quebec French. It then aims to discover whether the PR of DMs is similar to the PR of other grammatical classes, such as that of adjectives, where the reduplication is generally considered to be an intensification process. The working hypothesis is that, if “intensification” is defined as a semantic category equivalent to “very X” or “really X”, the notion of “intensification” is not adequate to describe the PR of DMs. Intensification is only possible with words on a scale and DMs are not scalar. This being the case, it is proposed that the PR of DM is essentially related to the speaker's commitment to his message and that the use of a reduplication device allows him to add an emphatic modal quality to the performance of an illocutionary act.

The English discourse markers – but, so, well, because – have been incorporated into a variety of Acadian French which has developed in southeastern New-Brunswick (Canada) in close contact with English. This paper attempts to explain what has motivated the borrowing of these particular markers into the current system, as opposed to the failure to borrow and and or, or even the markers of opposition and consequence, still and then, which are almost synonymous with but and so. There is a strong body of evidence showing that the markers so and but have largely evicted their homologues in standard French (mais and alors or ça fait que in regional French), while ben (bien) and parce que have maintained their place in the system. The argumentative value of the markers within an utterance and their intersubjective value in turn-taking are borne out by our data, although a few potential counter examples remain to be investigated.

This article takes a sociolinguistic approach to the study of the co-variation of a set of frequently occurring markers in three corpora of spoken French, dated 1968, 1988 and 2002. A Factor Analysis yields three Factors, labelled “Normal/Deferent”, “Modern/ Camaraderie” and “Tradition/Formality”. The most recent corpus, and in particular its youngest speakers, have higher rates of “Modern/Camaraderie” and lower rates of “Tradition/Formality”. The Factor “Modern/Camaraderie”, interestingly, combines diachronic developments of two sorts, however ; the spread of the sociolinguistically stigmatised quoi and the more intralinguistically-motivated spread of bon. The gradual pragmaticalisation of the latter allows it to be used in a greater number of contexts and thus increase in frequency. Frequency rates of these Factors correlate to a greater extent with the generation of the speaker than with educational background, sex or age per se. This suggests not only that modes of politeness are shifting but that speakers identify more closely with their own generation than with those of a similar educational background.

This article analyzes the diachronic rise of complex discourse markers, that is, markers consisting of more than one word. First, a taxonomy of four types of complex markers is proposed. Two of these types lend themselves to a specific diachronic analysis. Second, two complex French markers are analyzed in detail : bon ben and enfin bref. The diachronic rise of these two markers confirm the hypothesis that complex markers arise in a process of “over-use” of a given sequence of source words. These sequences are over-used in response to a functional need for these particular combinations of words in structuring discourse and interactions, regardless of whether the combinations of words are syntactic phrases or just a series of adjacent markers in discourse.

Based on evidence presented by a number of authors since the 1980s concerning various pragmatic uses of contemporary French toujours (c'est toujours ça ; tu peux toujours essayer ; toujours est-il que ; etc.), this paper draws up an inventory of the individual (grammatical and pragmatic) meanings of this adverb and retraces their internal history. The author has tracked down the appearance of the different usages of toujours in a meta-corpus including various dictionaries and data-bases as well as personal readings. The study leads to a historical reconstruction of the different pragmatic meanings of toujours over time, the aim being to distinguish between usages coined directly on the temporal adverb from those originating from already pragmatic ones.