In lieu of an abstract, here is a brief excerpt of the content:

6 The Critical Period, Access to Universal Grammar in First and Second Language, and Language Attrition So far, the concepts of modularity and the competence-ability distinction have proved useful in a number of ways. We will continue applying them in two areas of language development. First, in this chapter we will apply them to aspects of grammatical development: how child language research has tried to explain simultaneous , sequential, and subtractive bilingualism, L2 learning, and L1 attrition. All of this will lead to the important debate, still current after many years, over critical period effects in L1 and L2. The interesting point is that the critical period question is not the same for L1 and L2. A closely related topic, among generativist-oriented researchers, is “access to Universal Grammar (UG) in L2 acquisition,” also still controversial, but with recent signs of hope regarding a possible convergence around some proposals at least. Second, in chapter 7 we will apply the concepts of modularity and the competence-ability distinction to the development of academic language proficiency and L2 literacy. The same properties associated with modularity will also apply, but in a different way. In regard to grammatical development, bilingualism in children may develop toward a steady state of balanced competence in two languages or toward an imbalanced competence in which one language begins to undergo attrition or early stabilization . For a number of good reasons (and some not as good), the first outcome has attracted more attention from researchers than the second. However, as we saw in chapter 5, it is the unbalanced, out-of-equilibrium condition that sometimes reveals more clearly what the components of language are, and how they interact. In child L2 learning, a similar distinction is often drawn between additive and subtractive bilingualism. The research review in this chapter focuses on the latter, keeping in mind that “attrition” and “subtraction” are not the best metaphors for characterizing the developmental shift toward one primary/dominant language. As this chapter will propose, pathological language development aside, all “attrited languages” in children have an accompanying “replacing language” (RL) that attains completeness in a timely manner. The studies of L1 attrition reviewed here also offer a new way of looking at the two related questions of critical period effects 142 Chapter 6 in L1 and L2 and access to UG in L2 learning. If, as will be proposed, RLs are not diminished versions of fully formed primary languages, language acquisition capacity must remain intact (i.e., not deteriorate) during the period in childhood after L1 acquisition is consolidated, and perhaps even beyond. It is worth emphasizing, one last time, that the research on childhood bilingualism has resolved at least one important long-standing question, correcting previous misconceptions that early bilingual development was a potential source of cognitive confusion, possibly resulting in developmental lags or other dysfunctions. Cutting across theoretical perspectives, the consensus on this point stands as a major conceptual acquisition in the field: that knowledge of two or more languages in early childhood does not contribute to language deficiency (see note 4 in chapter 5 on “semilingualism”). The Faculty of Language (FL) appears to operate equally robustly and flexibly in both monolingual and bilingual development. Despite significant variation in cultural language socialization patterns, and assuming at least a threshold amount of passive or interactive input to the FL, complete acquisition of the child’s primary language core grammar is assured. Primary language acquisition unfolds automatically , deflected or attenuated only by the most extreme degradation of usable primary linguistic data. A number of studies have shown that this capacity extends to bilingualism: that the Language Acquisition Device (LAD) is also equipped to process input from two languages such that simultaneous acquisition of two primary (first) languages is a normal and typical outcome (Genesee 2001). Again, this outcome is obtained despite considerable variation in the young child’s bilingual experience (e.g., codeswitching by parents; unequal distribution of the two languages in different realms of language use in the home). Does the FL, then, include among its innate capacities an “endowment for multilingualism ”? The claim that the LAD is designed to accommodate bilingual and multilingual input just as easily as monolingual input appears to be well- founded; at the same time, it needs to be qualified. The challenge posed to researchers, based on the existing incomplete evidence, is to formulate a provisional working model that can help us explore in which ways the claim can be supported, and...

Share