For much of the twentieth century, philosophers generally have taken care to avoid the "genetic fallacy" and the fallacy of "psychologism": i.e. to follow the example set by Frege in the philosophy of mathematics, of distinguishing the logical issues appropriate to philosophy from all empirical issues, whether in history or in psychology. (Hence much of the force, for instance, of Hans Reichenbach's distinction between the "context of discovery" and the "context of justification", which held sway for so long in the philosophy of science.) In recent years, there has been some relaxation of the barriers between the history of science and the philosophy of science; but the separation of epistemology and developmental psychology is still generally insisted on. It is one thing to analyze and characterize the rational procedures constitutive of "mature, adult thought", which are the presumed destination of mental development; but it is quite another thing (it is said) to investigate the various pathways by which children come to achieve a grasp of those procedures. The purpose of the present paper is to reopen this question. Far from the issues available for study in epistemology, on the one hand, and in psycholinguistics and developmental psychology on the other, being absolutely distinct and separable, they are (I shall argue) in certain respects crucially interdependent. True: it is widely taken for granted today that we can define a particular constellation of rational skills and procedures as constituting the goal of intellectual development-the skills and procedures associated with a grasp of "formal operations", of the "deep structure" of any human