L’article se penche sur les influences entre les langues sur une aire plurilingue, dans le but de questionner la notion de convergence. Celle-ci présente a priori l’intérêt de tenir compte du fait que plusieurs facteurs sont présents ensemble et peuvent créer une situation propice au changement du fait de leur réunion. Par ailleurs, les situations plurilingues et pluriculturelles de l’Afrique d’aujourd’hui ont l’avantage d’être observables est vécues en direct et non à reconstituer à partir de recherches dans le passé. Après un examen général de la convergence des facteurs sociolinguistiques pour la Côte d’Ivoire, deux cas de réanalyses assez différents sont étudiés plus en détail dans le but d’explorer une possible convergence des langues : la possibilité pour des syntagmes prépositionnels d’avoir des fonctions de syntagmes nominaux, avec le cas particulier de pour, et la fonctionnalisation d’une forme verbale comme conditionnel périphrastique en français.
The article studies the influences between languages in a plurilingual area, in order to question the concept of convergence. This has the immediate advantage of taking into account the fact that several factors can together create a situation conducive to change by the very fact of their meeting. One of the goals of our contribution is to underline the impact of plurilingualism on the cognitive processing of languages by speakers. Indeed, postcolonial African territories make it possible to observe in real time what speakers do with languages, which is more advantageous than reconstituting situations from past research. The behaviors of multilingual speakers go beyond anything an observer of the language can imagine ex post. By multiplying the resources of their repertoire, plurilingualism multiplies the influx of possibilities of saying and ways of saying and, consequently, the factors of convergence so that a construction is privileged among others. However, there is immense heterogeneity and the practices have little stability (Boutin 2017), to the point that it is impossible to predict what, in the proliferation of syntactic and lexical creations, will remain as a result of code mixtures.Numerous facts about the three languages most commonly in contact in Ivory Coast, French, Jula and Baule, show two levels of convergence: between several factors of sociolinguistic and communicational order, so that an innovation first appears in a single language (Kriegel 2018; Mufwene 2005), and then between languages, when a construction scheme carrying meaning in one or more of these languages has served as a model for reanalysis of shapes in one or more other languages. The notion of reanalysis that I use is close to that of re-functionalization and is easily conceived as a result of unguided acquisition processes. Reanalysis is the result of comprehension strategies, a reinterpretation of the lexicon and the role of morphemes or constructions (Cf. Kriegel 2003; Detges 2003) which leads to an innovation including a re-categorization.The first part examines the convergence of factors for the Ivory Coast while the following parts detail the study of two cases of rather different analysis, with the aim of exploring a possible convergence of languages. The more widely used languages, in this case Ivorian French, Jula and Baule have mutually changed due to several factors such as the wide knowledge of their main constructions, the way in which the speakers manage an approximate inter-understanding by formal tolerance and diversified reformulation, the popularity of Nouchi, a mixed slang, and the more or less consensual practices of alternating languages. It is by using their repertoire and their plurilingual competence that speakers have operated reanalyses or adaptations (Gumperz 1964, Gumperz & Wilson 1971) by convergence between the various functionings of the languages present in the speeches.The second part presents the possibility for prepositional phrases to have functions of noun phrases, such as dans la cour, bɔ kɔ́nɔn, àwlo nun (in the courtyard) in examples (1) to (3), in which speakers talk about plantains.IF (1) c’est ici qui est mûr, dans la cour n’est pas mûr pres here rel cop ripe prep def courtyard cop.neg ripeJula (2) yàn lé mɔ̀nnan, bɔ kɔ́nɔn man mɔ̀n here cop ripe house in cop.neg ripeBaule (3) wàfan lo-li, àwlo nun wa-lo-man here ripen-pfv courtyard interior aux-ripe-neg (Those here are ripe, those in the courtyard are not ripe) While in French, the adpositions, in this case prepositions (which often derive from relational nouns), form a completely distinct category from nouns, Baule, like other Akan languages, has no adpositions. Baule uses relational nouns likely to provide localization with their anteposed noun complement, when French, like Romance languages, uses locative prepositional phrases. Jula represents an intermediate case since it has formal postpositions which have no nominal property, and locative postpositions, of the same form and of the same construction as relational nouns, but no longer having the specifying properties of names. These three languages could detail the scale of grammaticalization of nouns in adpositions of Hopper & Traugott (2003: 110-111) in a convergent process, without it being yet possible to say whether the change will be accelerated, for Baule for example, due to contact. On the other hand, speakers of Ivorian French who know the properties of languages such as Jula or Baule, easily grant nominal properties to locative prepositional phrases, even if the prepositions do not have a nominal origin as dans (1), seeming thus to downgrade the process.The third part analyzes, beyond prepositional phrases which have functions of noun phrases, the particular case of pour, in French, of tá in Jula and of liε in Baule in examples like (4) to (6) in which the third person (3sg) is a trader. IF (4) je n’achète pas pour lui 1sg buy.ipfv.neg prep 3sgJula (5) n tέ à tá sàn 1sg ipfv.neg 3sg postp buyBaule (6) n to man i liέ 1sg.nom buy.ipfv neg 3sg part (I don't buy his merchandise)The comparison with a vehicular French from the colonial period, of which we have written sources, and with Manding languages that have not known the vehicularization and contact that Jula has known, highlights various reanalysis processes making the properties of these morphemes converge.The fourth part studies the functionalization of a verbal form into a periphrastic conditional in French, illustrated by examples (7) to (9) in West African French.WAF (7) Si on partait en huitième de finale […] on allait arriver (If we started in the round of 16 […] we would make it) (8) Si j’étais riche sincèrement j’allais aider ceux qui sont dans le besoin (If I was rich honestly I would help those in need) (9) Donc vraiment si j’avais l’argent j’allais beaucoup aider ces enfants-là et leur maman (So, if I really had the money I would help these children and their mom a lot). In this case, no structural convergence is visible, but an influence of the languages in contact is possible at a deeper semantic level than the mere multilingualism of the speakers: the extralinguistic factors mentioned above converge to favor such reanalyses. In conclusion, the notion of convergence does not always prove to be discriminating in the case of a multiplicity of factors likely to join at different levels to bring about linguistic change. It needs to be anchored in the much broader consideration of linguistic ecology, such as Mufwene (2008) and Ludwig, Pagel & Mühlhäusler (2018) approach it. Their models try to show how these multiple factors interact, and are arguably better frameworks for going further in the analysis.
