Créoles et français : Quelques différences dans la valence verbale

Cette contribution montre la différence de fonctionnement entre le français et deux créoles français parlés dans l’océan Indien, les créoles mauricien et seychellois, dans quatre domaines qui concernent la valence verbale : l’expression du réfléchi, la diathèse passive en ganny, le codage des verbes ditransitifs et le codage du mouvement ablatif avec des verbes intransitifs. Pour les quatre évolutions, je proposerai une explication par différentes situations de contact de langues survenues suite aux colonisations européennes dans l’océan Indien.In this contribution, I look at four valency patterns of Creole spoken in the Seychelles and Mauritius, which clearly differ from the corresponding French patterns. In the four cases, morphological material from French was lost during creolization: 1. Following the loss of the reflexive clitic se marking co-reference between the subject and the object, this marker has been replaced, among other techniques, by the possibility to form the reflexive voice using the possessive determiner and a lexical item referring to the human body (son lekor) (see examples (1) and (2) from Seychelles’ Creole). 2. For the expression of the passive voice, the loss of the French passive auxiliary être favored the grammaticalization of the passive marking by ganny (< ‘gagner’) in written varieties of Seychelles’ Creole (see examples (6)-(8)). 3. The loss of the French preposition à to introduce the recipient role in ditransitive constructions resulted in Double-Object constructions (DOC) in which Recipient and Theme are both zero-marked (see Michaelis 2008) (see examples (10)-(12) from Seychelles’ and Mauritian Creole). 4. The loss of the French preposition de caused complex restructuring for the expression of ablative movement with intransitive verbs (simple path constructions). Since the second half of the 19th century, the replacement of de by depi (< ‘depuis’) in ablative contexts has been observed (see Kriegel 2012). These evolutions can only partially be explained by language-internal grammaticalization processes, reinforced for some of them by growing literacy. My hypothesis is that the determining factor for these changes is contact-induced language change during and after creolization. Reflexive marking by son lekor and ablative marking by depi can be explained by convergence: human body reflexives are attested in some varieties of French present in the contact situation leading to creolization. A comparable pattern also exists in Malagasy, one of the main contact languages in the emergence of Mauritian Creole. Speakers may have perceived the similarity between the construction in the two codes and this fact may have triggered a convergence process that led to the existence of body reflexives in Seychelles’ (and Mauritian) Creole. Depi in path marking, more precisely in ablative marking in Indo-Mauritian and written varieties of Mauritian Creole, could be interpreted as a convergence between congruent features in two Indo-European languages, French and Bhojpuri. For the two other patterns studied in this contribution (DOC coding of ditransitive verbs and passive marking by ganny) a French model could not be found. I argue that DOC coding of ditransitive verbs in Mauritian and Seychelles’ Creole and passive marking by ganny in Seychelles’ Creole are the result of simple code copying without convergence: following Michaelis (2008), DOC constructions are due to substrate influence from Eastern Bantu languages. The passive voice marked by ganny may be interpreted as being a copy from English (get-passive) (see Kriegel 1996)