Development of a Grasp of Transitivity of Length: A Reply to Smedslund

Smedslund's recent article, reporting an 8-year emergence of transitivity of length and criticizing the writer's finding of an earlier age, is answered on two grounds. First, evidence is reviewed indicating that the task requirements were probably not clear to Smedslund's Ss, and that his data are consistent with an earlier age of emergence. Second, Smedslund's contention that simple generalization processes account for the writer's finding of length transitivity in most 5-yearolds is disputed by a control experiment reported-Smedslund's own data on the point seem irrelevant. Further, it is argued that the kind of experimental procedures advocated by the writer yield better support for Piaget's concept of developmental stage than classical Piagetian procedures.