Levels of bilingualism and levels of linguistic awareness.

A framework for relating degree of bilingualism to aspects of linguistic awareness is presented in which metalinguistic tasks are described in terms of their demands for analysis of knowledge or control of processing. Two studies are reported in which children differing in their level of bilingualism were given metalinguistic problems to solve that made demands on either analysis or control. The hypotheses were that all bilingual children would perform better than monolingual children on all metalinguistic tasks requiring high levels of control of processing and that fully bilingual children would perform better than partially bilingual children on tasks requiring high levels of analysis of knowledge. The results were largely consistent with these predictions. The findings are discussed in terms of the implications of bilingualism for cognitive and linguistic development. Research investigating the effects of bilingualism on a variety of academic, linguistic, and intellectual achievements has traditionally led to conflicting results. Many of the early studies that warned of disastrous effects of bilingualism on cognitive development (see Darcy, 1963, for review) were later found to lack proper controls, undermining any interpretation of those findings. Later work revealed a more promising intellectual prognosis for bilingual children. Peal and Lambert (1962), for example, showed how careful selection of subjects in the bilingual population could produce evidence of bilingual superiority on some intelligence tests. The relation between bilingualism and intelligence depended on factors such as social class, degree of language proficiency, and type of bilingualism (Cummins, 1976). A similar debate surrounds the examination of the relation between bilingualism and linguistic awareness. Evidence for a facilitating effect (Ben Zeev, 1977; Cummins, 1978;Ianco-Worrall, 1972), an inhibiting effect (Palmer, 1972), and no effect (Rosenblum & Pinker, 1983) of bilingualism have been reported. Some investigators have found effects in both directions when studying different samples of bilingual children (Ben Zeev, 1977; Cummins, 1978). Just as the early research on the intellectual effects of bilingualism failed to account for relevant factors, so too this problem requires consideration of a wider range of issues. To reconcile the diverse findings a more detailed examination of the two factors, bilingualism and linguistic awareness, is required. The claim proposed in this study is that the relation between bilingualism and linguistic awareness must be stated in terms of the degree and type of bilingualism as well as the degree and type of linguistic awareness.