The work reported in this volume began during my final years as a doctoral student at University of California, Berkeley. At the time the work began, it seemed clear to me what a functionalist position was and what it wasn't, and I felt convinced that as an approach, it provided an excellent account of children's development of language. As I moved away from Berkeley, and as I traveled around to talk about my research, what had once seemed so clear was becoming more fuzzy. It was at that point that I decided to write this book, not so much as an explication of a coherent approach, but more as a way to clarify for myself, what others had meant by functionalism, the relative strengths of such an approach as an account for children's language development, and where further work needed to be done.

In reviewing what others had said about functionalist accounts of language in general, and functionalist accounts of child language in particular, it seemed to me that much of the explicit discussion centered around the issue of what one takes the domain of language to be. That is, most of the work seemed focused on describing a view of language as a nonautonomous system. the first chapter of this book summarizes such research, reviewing the extent to which child language researchers have picked up on trends found in functional linguistic theorizing.

A common theme among the functionalist child language literature is that children's earliest use of grammatical forms is linked to the ways they interact with the world. in chapter 2, I suggest that if we are to take this claim seriously, we must not only focus on the ways languages link forms with particular semantic and pragmatic meanings, but also weigh such knowledge in light of what research in the area of social and cognitive development . . .