Picture a bilingual child. What languages does this child speak? What kind of neighborhood does she live in? What are the educational arrangements that either support or demand bilingualism? Are any of the child's languages spoken in the community outside the home? What were the circumstances that led to her bilingualism? How long has the child been living in the present country? By changing even a single answer to this small sample of questions, the child being described is importantly different from one who would have elicited a different answer. Is there a common experience that unites this diversity of children? Is this common experience reflected in some deeply rooted element of their intellectual development? Does bilingualism in early childhood influence the nature of children's cognitive development?

These questions presuppose a more basic issue: How do we decide who is bilingual? We all know shreds of other languages although we would hesitate to include those imperfect systems as evidence for our bilingualism. Children's knowledge of any language is incomplete compared with that of an adult. At what point does a child have enough command of two languages to be declared bilingual? In part, the answer to that depends on how the two languages were learned and to what purposes they are put. But that does not solve the problem of deciding what is entailed by partial knowledge of one language for a child whose linguistic knowledge of any language is partial at best.