Browsing by Subject "Bilingualism--Georgia"
Now showing 1 - 1 of 1
- Results Per Page
- Sort Options
Item Patterns of variation in Spanish/English bilingualism in northeast Georgia(2002) Smith, Daniel James; Koike, Dale AprilThis dissertation examines Spanish/English bilingual patterns in a Northeast Georgia community (Habersham, Banks, and Stephens Counties) of Hispanics recently immigrated from Mexico, South and Central America, and the Caribbean. Data from naturally-occurring conversations by 56 children and adults of both sexes are analyzed within Myers-Scotton’s Matrix Language Frame model. Eight language patterns are identified, including monolingual Spanish and English turns, codeswitched turns, and turns showing convergence (morphemes from one language with abstract grammatical structure from the other). Tokens of each type (per sentence or short conversational turn) were counted per informant. Analysis revealed that percentages of monolingual and codeswitched utterances pattern in relation to percentages of utterances showing convergence, indicating that informants’ Spanish does not begin to converge toward English until fewer than 75% of their utterances are monolingual Spanish and that both codeswitching and convergence are mechanisms of language shift from dominance in one language to another. Children of either gender and females of both age groups report higher English use and Spanish influenced by English than older or male informants in general. Spanish utterances converging toward English are more frequent for those who speak English over those who speak Spanish mostly with friends. Non-Mexican Hispanics’ more frequent use of English with friends may be related to their higher convergence rates as compared to Mexicans. Higher preference for English among Mexicans may be related to their more frequent use of codeswitching. Weaknesses of the models used are identified. Myers-Scotton’s Matrix Language Frame model overemphasizes identification of the matrix language (ML) at the intrasentential level, evidenced from data that cannot be characterized within the model. Myers-Scotton’s Markedness model does not emphasize the crucial relation between the ML and the embedded language (EL) at the discourse and community levels. In response, this study analyzes the data in terms of the ML/EL dichotomy at the intrasentential, interpersonal, and community levels. Examples from the data illustrate how the ML/EL dichotomy can be extended to the inter-turn or conversational level. The resulting application of the ML/EL distinction shows the important relation between the ML and the EL at micro to macro levels.