We summarise our research into four issues: children's early hypotheses about how print represents speech, levels of phonemic awareness in children at the beginning of their school careers, the effectiveness of preschool training in phonemic awareness in promoting subsequent literacy acquisition, and characteristics of children who are at risk for reading difficulties because of a family history of literacy problems. All four research questions are cast in the framework of the nature‐nurture debate by focusing on what resources the child brings to the task of learning to read. We conclude that these resources are limited, and that the nature of these limitations does not support those theorists, associated with the “whole language” approach to reading instruction, who hold that learning to read is governed by the same processes as learning to speak. We note some educational implications of our findings, in particular the importance of directly teaching children about the segmental structure of the speech s