Linguistic Approaches to Bilingualism - Volume 1, Issue 2, 2011

This paper argues against connectionist models of language acquisition. It examines knowledge of the properties of subjects and objects in Spanish, particularly in impersonal passives and inchoatives. In both of these structures, the reflexive clitic se is obligatorily present and the linear order of elements is the same, namely [se V NP], with agreement between the verb and the noun phrase. In other words, the input is identical in both cases (se quemaron los libros ‘the books burned/were burned’ is ambiguous between both structures). However, the NP in the impersonal passive exhibits some of the properties of objects while the NP in the inchoatives behaves exclusively like a subject. An empirical study shows that three groups of speakers of Spanish, a monolingual Spanish group, an early English/Spanish bilingual group and a group of late learners of Spanish, are aware of these differences. As the input is ambiguous, it is argued that a model such as that provided by connectionist theories, which argues that the input is sufficient for acquisition, cannot account for the data.

This paper examines the effects of onset age of exposure to a dominant language (English) with respect to Korean heritage speakers’ knowledge of unaccusativity involving quantifier float. In an attempt to see how the acquisition of heritage grammar is affected by this factor, the present study compares two groups of heritage speakers: early (US-born, n = 13) and late (Korean-born, n = 14) bilinguals. The results show that compared to the late bilinguals, the early bilinguals did not give differential ratings to unergative and unaccusative verbs, which confirms the widely noted observation that the earlier onset age of exposure to English is, the more likely heritage speakers’ linguistic knowledge of the heritage language is incomplete. In addition, the results show that incomplete knowledge was also found with the late bilinguals (mean onset age 9), in that they did not accept unaccusative verbs as strongly as the native controls did.

In this study we consider the role of cross-linguistic influence in the domain of wh-movement and subject-verb inversion in children simultaneously acquiring Dutch and French, two typologically different languages. Wh-questions were elicited in Dutch by means of an elicited production task. The participants consisted of 5- and 7-year-old Dutch-French bilingual children, and two control groups of monolingual Dutch children and adults (N = 46). Target-like wh-fronted questions with subject-verb inversion formed the majority of responses. However, two qualitatively different structures were produced as a result of transfer from French: wh-in-situ questions and wh-fronted questions without inversion. Structural overlap approaches to transfer can predict cross-linguistic influence from the language with more structural options (French) to the one with only one interrogative construction (Dutch). However, we argue that a complexity-based theory of transfer provides a better account for the presence of the attested structures than a structural overlap approach.