Après un bref historique de l’utilisation de la notion de convergence essentiellement en études romanes et créoles (1.), je présenterai l’approche proposée par Kriegel et al. (2019) et je limiterai la convergence par rapport à d’autres notions centrales en linguistique de contact et en études créoles (2.). Ensuite, la contribution présentera une étude de cas (3.) : la marque du parfait (f)in/’n dans les créoles mauricien et seychellois. Il s’agit certes de « matériaux de construction » français parce qu’on y reconnaît la périphrase verbale française ‘finir de’. Cependant, le degré de grammaticalisation de (f)in/’n dans les créoles mauricien et seychellois est très élevé étant donné que cette marque fait partie du paradigme des particules TMA essentielles. Il n’en est pas de même dans les créoles français de la zone américano-caraïbe où la périphrase verbale ‘finir de’ a laissé des traces sans pour autant avoir connu une grammaticalisation poussée. Contrairement à Kriegel et al. (2019) qui présente un cas de convergence (la marque du réfléchi) en mettant l’accent sur la continuité avec les dialectes français, la présente contribution présente un cas de convergence (la marque du parfait) en insistant sur le rôle des langues parlées par les populations serviles dans l’océan Indien. Je montrerai que ces langues ont joué un rôle essentiel de catalyseur pour la grammaticalisation de (f)in/’n. L’interaction entre les langues en présence est d’autant plus remarquable qu’il s’agit de langues appartenant à des familles très différentes : d’un côté, les langues bantoues de l’Est et de l’autre côté le malgache, langue austronésienne. Dans le domaine de l’expression du parfait (ou de la complétivité), les locuteurs malgaches, puis les locuteurs de langues bantoues de l’Est auraient perçu des similarités (Jarvis & Pavlenko, 2008) entre leur langue première et la langue cible de leur apprentissage, le français en train de se transformer en créole au cours du 18e siècle. Pour l’évolution de la marque (f)in/’n, je propose donc de parler de double convergence dans l’évolution du français vers le créole mauricien, convergence avec le malgache et avec les langues bantoues de l’Est parlées par les populations serviles.
This contribution aims at a better understanding of the emergence of creole languages by examining the role of convergence in the evolution of the perfect marker (f)in/’n in Mauritian and Seychelles Creoles. Section 1 gives a brief historical outline of the use of the notion of convergence in Romance and in Creole studies. After a short discussion of early work e. g. by Gumperz & Wilson (1971), it introduces the first precise definition of the notion given by Bollée (1982) in the context of Creole studies: according to her, convergence characterizes the phenomenon (…) that similar structures of two languages in contact meet and reinforce one another, or that a tendency observable in one language becomes grammaticalised through the influence of a structure existing in the other. (Bollée, 1982: 392, my translation) Building on Bollée’s findings, I will then present the approach proposed by Kriegel et al. (2019) which develops the notions of similarity and grammaticalization. Language learners might perceive similarities between linguistic codes where linguists might not see them at first sight. In other words, convergence is a function of speakers who rely on perceiving similarities. A perceived similarity is a conscious or unconscious judgment that a form, structure, meaning, function, or pattern that an L2 user has encountered in the input of the recipient language is similar to a corresponding feature of the source language. (Jarvis & Pavlenko, 2008: 179) As to the second important point in Bollée’s definition, Kriegel et al. (2019) propose an extended notion of grammaticalization, traditionally used in a monolingual frame (e.g. Lehmann, 1995) and consider it in the plurilingual ecologies creole languages use to evolve in (e.g. Bruyn, 2009; Heine & Kuteva, 2005; Kriegel, 2003). In section 2, I will distinguish convergence from other central notions used in contact linguistics and in Creole studies: first, convergence processes are often described in terms of borrowing, replication, or transfer. Instead of using this terminology, Kriegel et al. (2019) work with an elaborated version of Johanson’s (2002, 2005, 2008) notion of code copying. The term copy simply refers to the fact that a feature in code A is copied (entirely or partially) into code B, independently from the question of perceived similarity. In convergence, the copy takes place because the speakers (learners) perceive a similarity between a feature in code A and in code B. Second, I refer to congruence as a static notion opposed to convergence as a dynamic one. The paper will then present a case study (section 3): the perfect marker (f)in/’n in Mauritian and in Seychelles Creoles. Undoubtedly the “construction material” is French because it is possible to distinguish the French verbal periphrasis ‘finir de’. However, the grammaticalization of (f)in/’n in Mauritian and in Seychelles Creoles is very advanced because the marker belongs to the closed set of the core TMA markers. This is not the case in the French creoles of the Caribbean where the verbal periphrasis ‘finir de’ left its traces but did not reach an advanced degree of grammaticalization. Contrary to Kriegel et al. (2019), which studies a case of convergence (the body reflexive) by focusing on the continuity with the French dialects spoken by the colonizers, the present contribution deals with a case of convergence (the perfect marker (f)in/’n) by insisting on the crucial role of the languages spoken by the enslaved populations in the Indian Ocean. I will show that these languages have played an essential catalytic role in the grammaticalization of (f)in/’n. This interaction between the languages in contact is all the more noteworthy if we consider that they belong to very different language families: on the one hand we have Eastern Bantu languages, on the other Malagasy, an Austronesian language. Concerning the expression of the perfect (or completivity) speakers of Malagasy, then speakers of Eastern Bantu languages would have perceived similarities (Jarvis & Pavlenko, 2008) between their first language and the target language, the French variety on its way to becoming a Creole during the 18th century. As to the evolution of the marker (f)in/’n, I suggest speaking of a double convergence during the evolution from French to Mauritian Creole, a convergence with Malagasy and with the Eastern Bantu languages spoken by the slaves.
