Interlanguage influence and multilingualism : an empirical investigation into typologically similar and dissimilar languages

• Main Author: De Angelis, Gessica Luisa
• Published: Birkbeck (University of London) 2002
• Subjects: 401; Crosslinguistic influence
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.268401

This dissertation investigates interlanguage influence, i.e. the influence of one or more non-native language on the production of a third or additional language. A qualitative and a quantitative study were conducted with speakers of Italian as third or additional language with English, Spanish or French as native or non-native languages. Subjects of Study 1 are a French Ll speaker and an English Ll speaker. Subjects of Study 2 are 238 Spanish L1 speakers and 130 English Ll speakers. Combining a traditional Crosslinguistic Influence (CLI) approach with theories of L 1, L2 and L3 production, we examine the role of crosslinguistic phonological activation during on-line processing and the effect of recency of activation on written production. It is argued that the process of transfer may be constrained by underlying crosslinguistic activation during on-line processing and that the incorporation of non-target lexicon into production may be influenced by the recent activation of a non-native language fonnally dissimilar to the intended target. We also hypothesise the existence of a cognitive process which we refer to as 'system shift' by which learners may transfer a lexical unit from one non-native language to another and then fail to recognize the source of their knowledge in the original linguistic system. We further propose two principles that interact in blocking Ll transfer: perception of correctness and association of foreignness. We also examine the relationship between the incorporation of non-target function words and the rate of subject insertion and omission in written production. Results show the rate of subject insertion to be significantly higher in the texts of English L1 and Spanish LI speakers with knowledge of French as a non-native language. We discuss the implications of our findings for a general and comprehensive theory of CLI