Researching 'practiced language policies': insights from conversation analysis

In language policy research, 'policy' has traditionally been conceptualised as a notion separate from that of 'practice'. In fact, language practices were usually analysed with a view to evaluate whether a policy is being implemented or resisted to. Recently, however, Spolsky in (Language policy. Cambridge University press, Cambridge, 2004; Work Pap Educ Linguist 22(1):1-14, 2007; Encyclopedia of language and education. Springer, New York, 2008) has argued that policy and practice need not be seen as distinct and that there is a policy within language practices themselves. In this paper, I propose to call the policy found at the level of language practices a 'practiced language policy'. The aim of this paper is to explore further this new conceptualisation of language policy and to propose an approach to research it. I argue that Conversation Analysis in its broad sense (that is, including Sequential Analysis and Membership Categorisation Analysis) can be an efficient tool to discover practiced language policies and give an illustration of this argument drawing on a case study of an induction classroom for newly-arrived immigrant children in France.