Case Systems in Contact: Syntactic and Lexical Case in Bilingual Child Language

ABSTRACT. This study examined effects of language contact and attrition on the Hungarian morphological case system in Hungarian-English bilingual children's speech production. To discover whether lexico-semantic and grammatical features are equally vulnerable in bilingual contact, the study compared patterns of change in case morphemes that express semantic relations with patterns of change in one that expresses a syntactic relation. Drawing on a lexically-based approach to production, an analysis of naturally-occurring data showed clear differences in both relative accuracy and pre-dominant error types between case morphemes that mark lexico-semantic information (local cases) and the case marker of the syntactic object (accusative). Overall, local-cases are less accurate than syntactic case; moreover, they tend to show substitution errors rather than omission, which is the pre-dominant error in application of syntactic case. These asymmetries are explained in terms of (a) how and when various case morphemes are accessed in production, and (b) typological differences in morphological marking of abstract case relations. * INTRODUCTION. Recent years have seen an increased interest in the role of language contact in bilingual first language (L1) development. Considerable attention has been directed toward the early years of bilingual acquisition, addressing how and to what extent the bilingual child's languages interact (e.g. Meisel 1989, Paradis & Genesee 1996, Lanza 1997, 2000) and how cross-linguistic and language-specific factors may explain why various aspects of the weaker language are affected in different ways (e.g. Schlyter 1993, Muller 1998, Sinka & Schelletter 1998, Dopke 2000b, Paradis 2000). However, there is still relatively little empirical research available that looks beyond the primary years of bilingual acquisition to examine effects of language contact on the development of the pre-adolescent bilingual's languages, especially on the development of the weaker L1 in immigrant situations (cf. Kaufman & Aronoff 1991, Pfaff 1991, Bolonyai 1998, Schmitt 2000). Schlyter (1993:289) points out that our knowledge about 'the quality of the weaker language' in bilingual children is particularly limited. By analyzing data from primary school children whose weaker L1 (Hungarian) and stronger L2 (English) are typologically very distinct, this study examined the mechanisms that may determine how change operates in the less stable L1. This study explored the extent to which Hungarian-English bilingual children's L1 displays effects of bilingual contact and developmental attrition as a result of the children's having been raised in an English-dominant social environment. Developmental attrition is used here as a general term, referring to incomplete access to certain aspects of a previously completely or incompletely acquired language in the context of childhood bilingualism (cf. Kaufman & Aronoff 1991, Polinsky 1994, Seliger 1996, Hamers & Blanc 2000). The main goal of the study was to investigate whether lexical (semantic/conceptual) or grammatical (morphosyntactic) features are equally susceptible to change in the unstable L1. To address this question, I examined how the Hungarian morphological case system is affected in bilingual children's production. Hungarian case markers provide a good testing ground for such investigation because some are more lexical and some are more syntactic in character. In specific, the study asked whether all Hungarian case suffixes, those marking syntactic relations (i.e. accusative case marking direct object) and those marking semantic relations (i.e. local cases marking spatial and thematic relations), pattern alike in production, and it aimed to identify key factors that motivate differences in their vulnerability. Working within a lexically-based approach to production (e.g. Levelt 1989, 1999, de Bot 1992, Bock & Levelt 1994), and the 4-M model of morpheme classification (Myers-Scotton & Jake 2000b, 2001), this report will explain the observed distribution of case morphemes on the basis of (a) how the morphemes are accessed in production, and (b) typological differences between Hungarian and English. …