Retour vers liste

Détail de la contribution

Auteur: Alain KIHM

Co-Auteur(s): Fabiola HENRI /CNRS-LLF/Paris7, France

The morphosyntactic status of TMA markers in two Romance-based Creoles

Abstract/Résumé: Miller & Monachesi (2010) present robust arguments in support of the view that the Romance clitic pronouns ought to be analysed as inflectional affixes (agreement markers) proceeding from the grammaticalization of former simple clitics or free pronouns. It follows from this view that constructs such as French Je l’ai embrassée or European Portuguese Beijei-a ‘I kissed her/it’ – signs in SBCG (Sag 2007) – have a double identity as sentences and morphological words. Under the latter identity, they can be defined by paradigm functions such as the following, where embras- and beij- stand for the stems of the transitive verbal lexemes EMBRASSER and BEIJAR (Stump 2001). Periphrastic tenses are assumed to enter paradigms on a par with synthetic tenses (Bonami & Webelhuth 2012). French as well as European Portuguese fill a cell in the respective paradigms of EMBRASSER and BEIJAR. The aim of the present paper is to extend this analysis to French-lexifier and Portuguese-lexifier creoles. Like their lexifiers, FLC’s and PLC’s show clitic pronouns in subject and object function, calling for the same analysis. Seemingly unlike their lexifiers, they express most TMA features with “markers”. The two following examples from Mauritian (MA) and Guinea-Bissau Kriyol (GB) illustrate: (1) Mo’nn ba li. (MA) 1SG PAST kiss 3SG I kissed her/him/it. (2) N na da -l besu. (GB) 1SG PROG give 3SG kiss I will kiss her/him/it. TMA markers such as MA (i)nn and GB na have been variously analysed. Henri & Laurens (2011) about MA demonstrate they cannot be considered verbs of the auxiliary or raising type as proposed by some authors (e.g. DeGraff 1992). They deny them the status of affixes because of their unselectiveness as to the host’s category. Yet GB does not evidence such a property: TMA markers must be left-adjacent to verbs. In MA, there seems to be a gradient pou, ti > (i)nn > pe with pou (MOOD) and ti (TENSE) unselective, (i)nn (ASPECT) marginally unselective, and pe (ASPECT) strictly selective of verbs. The case for TMA markers as inflecting clitics (Spencer 2005) therefore remains a strong one. It does not call for a yes-no answer, though, but it should be embedded within a typology of marking ranging from functors to prefixes.