On the development of gender in simultaneous and successive bilingual acquisition of French: Evidence for AOA and input effects

This study investigates the effects of Age of Onset of Acquisition (AOA) and the quality and quantity of input on the longitudinal development of gender in the acquisition of French by simultaneous (2L1) and successive bilingual children (cL2). There's an on-going debate regarding the main factors underlying linguistic development in bilingual children. Two independent variables, AOA and input, have been particularly emphasized and sometimes contrasted (Unsworth, 2013). Proponents of maturational accounts (e.g. Meisel, 2009) have argued that AOA is the main factor underlying language acquisition, but more recent research suggest that different aspects of language are more or less influenced by AOA and input respectively. Whereas lexical development seem to be particularly influenced by input properties (Thordardottir, 2011), the question is more open with respect to different grammatical features. One view is that quantity and quality of input is particularly important in the acquisition of “late”, “harder” or “more opaque” grammatical structures (Gathercole, 2007; Unsworth, 2014), structures typically involving interface relationships of some kind (Sorace, 2005, Tsimpli, 2014). However, previous research has typically drawn on comparisons of different grammatical features in different languages which make controlled comparisons difficult.Our contribution to the debate lies in the combined analysis of three different exponents of a single underlying linguistic construct in a single langue, namely French gender, an opaque feature according to Corbett (1991). The development of gender in the child's grammar involves different linguistic levels, including the abstract system level, the lexical level (attribution) and the syntactic level (concord). We studied the discovery of the abstract GENDER feature, gender assignment and gender concord in a longitudinal multiple case study including successive (cL2) Swedish-French bilingual children (n=4), simultaneous (2L1) Swedish-French bilingual children (n=4) and monolingual French children (n=4). Recordings containing a variety of tasks took place in the age span 3;5 to 10 years, but no child was followed for more than three years. Two of the L2 children have an AOA below 4 years and two of them have an AOA above 4 years (cf. Meisel, 2009). All children attend the same French-speaking school in Sweden. Input was measured for each bilingual child individually through the establishment of an input profile (see Ågren, Granfeldt & Thomas, 2014).The findings show that the amount and quality of input correlate with the rate of acquisition of gender attribution, but most clearly in the 2L1 group. However, the discovery of the abstract GENDER features does not seem to be influenced by external input conditions. Group-level analyses of gender concord suggest that input is less important than AOA. In fact, only the cL2 group performed worst on concord than on attribution. The results are also discussed in relation to previous studies on the development of other morphosyntactic structures in the same children. We conclude that input is most important for the development of the lexical aspects of the opaque gender feature in French (attribution), in particular in the 2L1 group, but that AOA seem to play a role with respect to its syntactic aspects.