What Else Transfers?

In his seminal paper on the ‘fundamental difference’ between first and second language (L1 and L2) acquisition, Bley-Vroman (1990) proposed that this disparity is due to the lack of continued access to Universal Grammar and the unavailability of ‘domain-specific learning procedures’ beyond a critical period in child language development. He considered four possible alternative accounts of this difference, arguing that each falls short of providing a convincing explanation: (i) the input hypothesis; (ii) the affect or socialization hypothesis; (iii) competing cognitive systems; and (iv) L1 interference. Whilst various interrelated factors are needed to comprehensively explain the many differences between L1 and L2 acquisition, this paper argues that the importance of (iv) is still significantly underestimated. Despite the increased importance attributed to in syntactic and phonological transfer in recent years (Schwartz and Sprouse, 1994, 1996; Archibald, 1998; Brown, 2000), transfer accounts have largely ignored the acquisition of the lexicon. If lexical representations are not universal but relative, and if the L1 lexicon constitutes the initial state of the interlanguage lexicon (Sprouse, 2006), then L2 lexical acquisition involves the relexification of the entire L1 lexicon, with all the syntactic information it contains. This paper re-examines the role of the L1 lexicon in L2 acquisition, and suggests that the problem of relexification goes a considerable way toward explaining L2 learners’ general lack of convergence on the target language. In Section 2, I review the basic tenets of the Fundamantal Difference Hypothesis, and consider the growing emphasis on the role of the L1 in L2 research. Sections 3 and 4 present two concepts which are pivotal to the understanding of lexical acquisition in a second language: lexical relativity, and lexical transfer. The implications of these two phenomena are considered in Section 5, in terms of four types of L1-L2 lexical mapping problems. Finally, the learnability problem is set in the light of the vastness of the task of vocabulary acquisition.