Code-switching in a Turkish secondary school John Eldridge English language teachers who teach in monolingual environments have for a very long time been concerned about reducing or even abolishing student use of the mother tongue in the language classroom. The reason for this is presumably to maximize the amount of time spent using the target code, and thus improve learning efficiency. This study will describe and analyse the code-switching of young learners in a Turkish secondary school. It will show that there is no empirical evidence to support the notion that restricting mother tongue use would necessarily improve learning efficiency, and that the majority of code-switching in the class- room is highly purposeful, and related to pedagogical goals. The issue of how we treat language alternation in the classroom is of central metho- dological importance, and one, it will be argued, that has enormous im- plications for practising language teachers. It is therefore vital that we understand precisely its causes, motivations, and effects, and that until that point we avoid making rash, censorial judgements on its classroom manifestations. Introduction Code-switching, which may be briefly defined as the alternation between two (or more) languages. has been receiving growing attention in recent years. Far from being viewed as a random phenomenon, it has come to be seen as a highly purposeful activity. However, the majority of extant studies have been conducted in authentic bilingual speech communities, rather than in the language classroom, which will be the focus of this study. Language attitudes Teachers and researchers in English as a second language have. on the whole, been concerned to minimize code-switching in the classroom, taking it that the switching either indicates a failure to learn the target language or an unwillingness to do so. Willis (1981: xiv), for instance, suggests that ‘If the students start speaking in their own language without your permission. . . it generally means that something is wrong with the lesson.’ Cummins and Swain (1986: 105) similarly contend that progress in the second language is facilitated if only one code is used in the classroom, asserting that the teacher’s exclusive use of the target code will counteract the ‘pull’ towards the native code. There also seems to be a feeling that languages should be kept strictly demarcated-this despite the fact that code-switching is employed ‘in the repertoires of most bilingual people and in most bilingual communities’ (Romaine: 1989: 2). In the case of monolingual native speakers, the concern is perhaps compounded by the fact that they are often unable to determine why the switching is taking place. ELT Journal Volume 50/4 October 1996 © Oxford University Press 1996 303