This paper is a first report on a longitudinal study of the acquisition of a syntactic rule: the inversion of WH-questions.1 The general aim of this presentation is to show how certain modifications in the theory and practice of linguistics can contribute to the study of acquisition, and to the psychology of language generally. These new developments spring from an approach to linguistic analysis that is based on the observation of speech in every-day contexts, and experiments carried out in every-day situations. In this approach, introspection may be taken as a handy guide to the issues, but not as evidence for the developing theory.