USER MODELING IN CALL - SOME FUNDAMENTAL ISSUES

There has long been a tug-of-war between, on the one hand, the obvious need to make foreign language teaching courses and materials (especially CALL materials) as relevant as possible to the learner, and, on the other hand, the economic argument that the greater the number of customers for a given product, the more cost-effective (and profitable) the exercise will be. Developments in the resources available to CALL are rapidly shifting the outcome of this argument strongly in favour of tighter learner targeting. But this tendency cannot reach its full potential without considerable adjustments in our implicit assumptions about the nature of learning in general and language acquisition in particular. Such a rethink could well have a profound effect on the design of efficient, effective, efficacious and affordable CALL. Let us take the 'guesswork' out of language learning, both with regard to what the learner wants to know, and what s/he knows already and does not need to be 'taught'. True learner profiling leads to very different CALL learning packages from those currently on offer. And modern technology can now help in ways which were unthinkable a decade ago. But it is we, the teaching profession, who must define the tasks for technology. This paper presents a 'first principles' redefinition of the language learning task from the 'ergative' learner's point of view, and proposes a new role for technology.