This paper argues for a shift of research and scholarly interest from the study of death to the study of the fatally ill child's illness experience. The issue of “what should a child be told” is discussed as an example of the narrowness in approach which results from the emphasis on the child's prognosis. Some examples from ongoing longitudinal research with leukemic children are given to illustrate the complexity of the problem and the difference in perspective which can be gained through study of the child's illness experience.