Is nativism sufficient?

The past and present state of the empiricism-nativism issue is analysed. Empiricist philosophical doctrine ('no innate ideas') distinguished idea from structure or mechanism. However, Chomsky's conception of innate linguistic universals erased this distinction. The elimination left would-be empiricists without a coherent and defensible position. I argue that the issue remains alive primarily because of tension between two scientific tasks that face students of development. One is to discover what is cognitively and linguistically primitive, a task that encourages nativism. However, nativism is ultimately unsatisfactory because it systematically neglects the other task, which is to account for development, including the emergence of postulated innate primitives. To account for such primitives, it is necessary to relate them to particular central nervous system structures in such a way as to explain how the structure has the particular cognitive effects that define the primitive. That is likely to be difficult, and I show how the study of learning--much neglected in recent years--can help by reducing the number and type of innate primitives whose origin must be explained in that way.