01/01/2014 Linguistics Psychology
DOI: 10.1177/0267658313511774 SemanticScholar ID: 145676602 MAG: 2107318803

Implications for child bilingual acquisition, optionality and transfer

Publication Summary

Amaral & Roeper’s (this issue; henceforth A&R) Multiple Grammars proposal (henceforth MG) offers an appealingly simple way of thinking about the linguistic representations of bilingual speakers. In the following I will focus on some implications for child bilingual acquisition. MG develops some of the themes that were originally put forward by Roeper in his Universal Bilingualism proposal (Roeper, 1999), and takes the idea that any speaker is bilingual to its logical consequences by investigating the nature of mental representations of actual bilingual speakers. The central tenet of Universal Bilingualism, further developed by MG, is that the child learner does not set out with a ‘uniform set of rules that captures all of the information’ (p. 5). This is neither possible, given the number of potential exceptions to the rule, nor desirable from a minimalist point of view. Rather the idea is that the child will start by positing the simplest possible UG-constrained rule on the basis of the evidence in the input. Whenever new and contradictory evidence is encountered, the child will add a new ad-hoc rule to deal with this new set of data. Rules do not change on the basis of new linguistic evidence; new rules are created and co-exist alongside the old ones. Somewhat surprisingly, A&R’s views on the process of first language acquisition coalesce into a considerably more inductive and bottom-up approach than one would expect from researchers coming from a generative grammar background. Their underlying assumption is still, of course, that whatever rule the child posits must be ultimately constrained by UG; in this sense hypothesis-building varies within pre-ordained parameters. Nevertheless, A&R credit the child with a much more active role than is traditionally assumed in generative approaches to acquisition; the learner proceeds on the basis of the input that he or she encounters, and rules emerge on a rather ad hoc basis and are

CAER Authors

Avatar Image for Ludovica Serratrice

Prof. Ludovica Serratrice

Reading University - Professor of multilingualism

Share this

Next publication

2009 Psychology

The Dynamics of Category Conjunctions

R. Hutter, R. Crisp, G. Humphreys, Gillian. M. Waters + 1 more