It is one of the truisms of science that progress generally comes from simplifying the phenomena under study, and then proceeding step-by-step from the simple to the complex cases. The idea that complex behaviour could be understood as the outcome of many simple, experimentally isolable processes was a central tenet of behaviourism. More recently, psycholinguistics has provided a counter example to these beliefs. While our knowledge of such supposedly simple and basic processes as the learning of simple associations between words has scarcely advanced at all, the last 15 years have seen an exciting revival of research into the acquisition of language itself. This is undoubtedly one of the success stories of developmental psychology, since it has yielded a substantial body of observations which are proving to be reliably replicable in subsequent studies of different children, learning different and unrelated languages. Perhaps most impressive of all, many observations are even proving replicable with different observers. This chapter aims to tell something of the history of this research and the theories that have guided it, and to suggest some likely future directions that it may take.