This study compares four experiments on the acquisition of binding, two conducted with Italian-speaking children and two with English-speaking children. The task used in these experiments was the truth value judgment task, which I argue permits fairly direct access to children's syntactic competence. The results from these experiments suggest that Italian-speaking children do abide by Binding Conditions A and B. But some explanation must be given for the finding that English-speaking children's mastery of pronominal binding lags behind their mastery of binding for anaphors and R-expressions, Emphasizing that the explanation must (at least be able to) account for such cross-linguistic differences, I argue against appeals to maturation or to the learning of pragmatic constraints. I also consider proposals that can accommodate the cross-linguistic differences, namely a lexical misclassification proposal and Varela's (1989) suggestion about children's early hypotheses of governing category.