Informing the age-of-acquisition debate: L3 as a litmus test

Abstract

Following Cabrelli et al. (What the start of L3 tells us about the end of L2: N-drop in L2 and L3 Portuguese, BUCLD, 2008), Iverson (Competing SLA hypotheses assessed: Comparing heritage and successive Spanish bilinguals of L3 Brazilian Portuguese, Mouton de Gruyter, 2009) and others, I argue that the L3 initial state is an important tool in testing claims made with respect to ultimate attainment in adult language acquisition. Given the appropriate language combinations, testing knowledge of an L3 at the initial state hints at the possibility (or not) of acquisition of new features and parameter resetting in adulthood. Herein I re-examine two studies, Guijarro-Fuentes et al. (Non-convergence at advanced levels, learnability and the preemption problem in L2 semantics: DP and Bare nominal interpretations in L2 Portuguese, University of Southampton, 2008) and Iverson (Competing SLA hypotheses assessed: Comparing heritage and successive Spanish bilinguals of L3 Brazilian Portuguese, Mouton de Gruyter, 2009) which look at the acquisition of L3 Brazilian Portuguese (BP) by learners with L1 English/adult L2 Spanish. Spanish and BP share many phenomena that English lacks; therefore, any L3 initial-state knowledge of these shared phenomena must come via L2 Spanish. Although not the original goal of these works, they implicitly indicate new feature acquisition in adulthood and highlight how L3 studies in general contribute to the age of acquisition debate.