||There is little general agreement amongst language teachers about the value of getting students to talk to other students in the language class, apart from the assumption that getting students to speak is important in the same way that practice of any set of skills is intuitively regarded as necessary to reinforce and automatise those skills. Indeed, there are many who see such interlanguage exchanges as potentially harmful in terms of the manipulation of 'deviant' linguistic forms which it appears to encourage (Prabhu 1987). Pedagogic and evaluative tasks designed for oral practice and evaluation range from elaborate simulations, which might involve a whole class over an entire course, to the simpler discrete information-gap and jigsaw types. A common element in the design of such tasks is that they are intended to be performed by the learners themselves, with the teacher as facilitator rather than controller. The resulting performances from this type of peer-group communication task have already been the focus of a considerable amount of research as well as attracting the attention of materials writers. Current interest in task-based approaches to oral practice, self-access methodology and other more learner-centred approaches, have all directly and indirectly promoted the use of language activities which can be managed by the learner without teacher intervention. Other more practical considerations, such as large class sizes and the increasing realisation amongst teaching staff that oral practice in such environments cannot be given the focus it urgently requires, have meant that an increasing responsibility has been placed upon learners themselves.