Abstract:
This thesis examines code variation among Algerian university students and the effects of Arabic and French in contact. It focuses on three main goals:
(1) exploring the linguistic patterns that result from language mixing, (2) investigating the factors of language choice and the extent of language choice predictability, and (3) studying the attitudes towards code-switching. Data from naturally-occurring conversations by 112 students from both sexes and a census questionnaire administered to 248 other ones are analysed to explore the linguistic, the sociolinguistic, and the attitudinal effects and to test the hypotheses related to the main goals.
Two language mixing patterns which include other sub-patterns are identified. Borrowing is performed according to different levels of integration that lead to the production of three sub-patterns: integrated, non-adapted, and non-conventional borrowings. Code-switching can be identified as little or heavier according to the number of items inserted within the Matrix Language.
In addition, the analysis reveals the existence of code-switching between Spoken Algerian Arabic (SAA) and French and between SAA and Modern Standard Arabic (MSA).
Factors of language choice are investigated in the light of Grosjean’s model. Language choice is performed according to specific factors that make it predictable in most cases. However, the results support the hypothesis that there are cases of language choice which are unpredictable. Moreover, female students use language as a communicative strategy more than male students do.
Attitudes towards code-switching are analysed to determine the nature of these attitudes. The findings support the hypothesis that positive attitudes can be associated with code-switching. In addition to the negative attitudes, codeswitchers consider their behaviour positively. These contradictory opinions lead to discrepancies between the speaker’s language behaviour and attitudes.