Overt and covert codeswitching in immigrant children from Russia

Abstract

This paper discusses the principles and mechanisms of language attrition in a group of Russian immigrant children in the United States. The discussions and analyses in the study are carried out within the theoretical framework provided by the Matrix Language Frame (MLF) model and its two submodels: the 4-M model and the Abstract Level model. The data are analyzed for the presence of bilingual production, such as codeswitching (CS), convergence, and bare forms. Because different types of morphemes become salient at different levels of the abstract lexical structure, some are less accessible and therefore more vulnerable as language attrition progresses. A structural definition of convergence based on the interaction of the abstract lexical structures of the languages participating in the bilingual speech is developed. The analysis shows that bare forms are produced as a result of both CS and convergence in the abstract structure. Bare forms represent an intermediate stage of language loss, preceded by integrated CS forms and followed by convergence. Overt CS is defined as the presence of morphemes from two varieties; these may be nontarget forms, including bare forms. Overt CS alone is found not to be indicative of language attrition since it does not, by it self, lead to the projection of parts of the abstract structure of the Embedded Language onto the grammatical frame of the Matrix Language (ML). Covert CS (convergence) may occur alone or along with overt CS (nontarget forms, including bare forms). Covert CS is, indeed, an indicator of language attrition in progress since it is a result of a composite Matrix Language.