French-English bilingual children’s acquisition of determiners.

This paper investigates the way two French-English bilingual children acquire determiners in language-specific ways in their two languages. In English and French, determiners are obligatory and their choice is closely linked to definiteness as a marker of information status (Hickmann 2003). Both languages use definite (le, la, les / the) and indefinite (un, une / a, an) determiners. However, English also uses the null determiner with [- count sing], [+ count plur] and proper nouns in generic context as well as with [+ count plur] and abstract nouns in specific context. Evidence that bilingual children produce non-target productions at the discourse-pragmatics interface as a result of cross-linguistic influence between overlapping structures of their languages has been intensively reported in the literature (see Serratrice 2012 for an overview). Extensive research has been carried out on bilinguals’ acquisition of subject and object pronouns in null- and non null-languages (Serratrice et al 2004; Sorace et al. 2009). But to date, very little is still known about how bilingual children acquire information structure knowledge that regulates the encoding of local markers in French and English, especially the extent to which the regular use of two languages may affect the acquisition of determiners. In this study, we examine the longitudinal corpus of two French-English bilingual children: Anne aged 2;04 to 3;04 (exposure to English: 55%) and Sophie aged 2;06 to 3;06 (exposure to English: 58%) in conversation with their carers. The girls, both living in the UK, were video-recorded for one hour in French and one hour in English every month over a year. Referential expressions in argument position were coded for morpho-syntax (i.e. Definite Phrase, Indefinite NP, bare nouns), discourse-pragmatic status (new vs. given), and their ambiguity (i.e. whether the referent was easily recoverable from the discourse or the extra-linguistic context). The data show a very small amount of non-target determiners in the bilinguals’ two languages. Pragmatically inappropriate definites were not observed significantly more than indefinites in English but they were slightly in French. Sophie has a 1% error rate to mark definites and indefinites in object position referential expressions and no error rate in subject positions in her English while Anne has a slightly higher mean rate of non-target definites than indefinites (subject position: .08% vs. .07%; object position: 2.49% vs .61%). In French, Anne did not produce any non-target definites or indefinites while Sophie did produce slightly more non-target definites than indefinites (subject position: none; object position: 4% vs. 0%). In contrast to the pragmatically appropriate use of definite and indefinite articles, we found a high proportion of ungrammatical bare nouns (e.g. singular count noun as in I want apple) in both English and French (see table 1). The high percentage of illicit bare nouns in French shows evidence of cross-linguistic influence from English to French. This paper confirms a late acquisition of the null determiner as opposed to definites and indefinites in English and reports new evidence of cross-linguistic influence from English to French in the use of bare nouns.