This dissertation analyzes patterns of communication strategy usage in the speech of 41 French immersion students in Toronto, Canada collected from informal student interviews with a native French speaker. The study contributes to a more complete understanding of communication strategies in French immersion by addressing such issues as the relative range and frequency of strategies; the interplay between strategies; the interaction between participants related to strategy usage; and the effect of extralinguistic factors on strategy usage (e.g. students' sex; age/grade; French language media exposure; time in a Francophone environment; stays with a Francophone family; home language).
Strategies were coded and frequency counts obtained. The strategies identified included: L1-based (language switch and foreignization); L2-based (circumlocution, word coinage and approximation); sociopragmatic (appeal for assistance, message abandonment, and mime); and ambiguous/potentially L3-based strategies. While students use a range of strategies, the tendency is to rely on language switch. Rather than risk inaccuracy in the TL, students prefer to be economical and, assuming that the interviewer is bilingual, are confident that she will understand the strategy. Appeals for assistance from the interviewer are also frequent, demonstrating that the presence of an interlocutor plays an essential role in how students deal with lexical problems.
The emergence of a strategy continuum provided support for the notion that some strategies are riskier than others by showing that frequency and perceived level of risk associated with that strategy is related to the amount of follow-up that a strategy receives. Follow-up strategies occur (i) for the purpose of correcting a prior usage; (ii) due to awareness of French/English translation difficulties; (iii) due to uncertainties about TL usage; and/or (iv) as means to treat the interview as a learning experience or to meet expectations of speaking French during the interview.
Statistical analyses revealed that L1-based, sociopragmatic, and ambiguous/potential L3-based strategies are correlated with some of the social factors examined, including extracurricular exposure to French and age/grade, but not students' sex.
It is concluded that in spite of the non-conventional lexical choices in students' TL messages, they are still comprehensible (i.e., meaningful) to the interlocutor and communication goals are achieved successfully.