ABSTRACT
The primary finding of the research reported in Goodluck and Behne's chapter, that adult competence in certain control structures is acquired rather late, has significance beyond the simple empirical fact. A two-sided question arises about the nature of the differences between the structures that are learned early and those that are learned late. One side of the question concerns the nature of the adult grammar; the other side of the question concerns how the child grammar approximates the adult grammar. In these comments I want to discuss, somewhat speculatively, these differences.
