Understanding Language in Language Minority Education: Toward a Theory of Language in Contact

Jeff MacSwan

Language minority children (LMs) make up about 7.4% of U.S. students, with numbers as high as 25% in parts of the southwest (Macías, Nishikawa & Venegas, 1998); these figures are projected to increase by 96% by 2020, with highest increases among Spanish-background children (Campbell, 1994). Furthermore, of Latino children born outside the U.S. who dropped out of U.S. schools, about 80% were found to speak English with only limited proficiency (Kaufman, Kwon & Klein, 1999). The size and vulnerability of this population of students make continued research in language minority education a matter of great social importance. Most scholars and educators in the field of language minority education now believe that many language contact phenomena constitute evidence of language attrition, and that this presumed deterioration of linguistic ability explains school failure among LMs. As a result of my study, I hope many scholars and educators will see such phenomena as a consequence of interactions naturally permitted by the intact linguistic system of bilinguals, leading to a re-examination of theories of achievement differences in LMs. Over the last several decades, linguists have developed an explicit theory of the structure of language in the mind/brain. Linguists concerned with language structure and acquisition study syntax, or word order, along with morphology (word structure) and phonology (sound patterns). Most recently, syntactic theory, aimed at explaining variation in word order among the world’s languages, has focused analy¬sis on the lexicon, the mental list of vocabulary available to individual speakers. Some lexical items, called functional categories, host parameters, which act as mental switches defining the narrow hypothesis space children entertain when ac¬quiring a first language. Thus, linguistic variation results from the particular array of morphological properties in the lexicon (Chomsky, 1995). Specifically, the model proposes that an op¬eration, called Select, picks items from the lexicon and introduces them into a numeration, an assembled sub¬set of the lexicon used to construct a linguistic expression. Another operation, Merge, takes items from the numera¬tion and forms syntactic objects. The operation Move applies to syntactic objects so formed to build new structures. Movements of elements within the structure are driven by feature checking of case, number, person, and gender. In earlier work on language contact, and following recent developments in syntax, MacSwan (1999, 200b) proposed a theory of codeswitching which postulated that Select may place items from either lexicon in the numeration, with feature checking proceeding just as in monolingual syntax. In this approach, constraints on language mixing are seen as epiphenomenal, resulting from interaction naturally permitted by the linguistic system. The present study will seek to extend the framework just described to another kind of language contact phenomenon known as calquing. Calques involve the vocabulary of one language and the syntax of another (for example, "Eso es lo que estoy pensando en” [‘This is what I am thinking about’]). Like codeswitching, such utterances are often regarded as evidence of deteriorating language ability. In the present study, calque data will be analyzed in terms of current linguistic theory, posit¬ing as an initial hypothesis that calquing is a kind of “codeswitching” at the level of lexical features. If successful, the analysis will show that calques result from interactions naturally permitted by a bilingual’s intact language system. The study of bilingualism is of great importance for educational research and practice. At a time when the language rights of LMs are under intense political attack, the field often seems ill-prepared to provide a sensible rationale for its policy recommendations. Debate and substantive engagement of language holds great promise for the development of a rich theoretical context and more focused inquiry. More narrowly, the results of the study proposed here will inform our evaluation of specific theories in bilingual education – notably, the Threshold Hypothesis – since it speaks to the question of deteriorating native language ability in late sequential bilinguals. The study will additionally inform language assessment design and policy, identification of language impairments in bilingual children, and the over-representation of LMs in special education, and will provide a knowledge base for teachers and teacher educators whose (mis)understanding of the nature of language in LMs forms the foundation for student-level decisions regarding curriculum and instruction.

About Jeff MacSwan

Jeff MacSwan is Assistant Professor of Language and Literacy in the Division of Curriculum and Instruction at Arizona State University. After teaching in Los Angeles pubic schools for a few years, MacSwan obtained his Ph.D. in Education from UCLA in 1997. He has served as the associated editor of the Bilingual Research Journal, and was invited to chair the scientific organizing committee of the 4th International Symposium on Bilingualism, held at Arizona State University in April, 2003. He has published numerous articles and book chapters on bilingualism and the education of language minority students. MacSwan’s dissertation was published in Garland’s Outstanding Dissertations in Linguistics series as A Minimalist Approach to Intrasentential Code Switching (Garland, 1999). A forthcoming edited volume, Grammatical Theory and Bilingual Codeswitching (Jeff MacSwan, editor; MIT Press), explores additional empirical consequences of this research.