The use of T. S. Kuhn's Structure of Scientific Revolutions as a model for the history of psychoanalysis is addressed. It is shown that his categories have been regularly misapplied in the interest of establishing the scientific status of psychoanalysis, or of proposing a new theoretical structure. It is argued that Kuhn's model may nevertheless be useful for the history of psychoanalysis on condition that its categories are used precisely, and their correspondence to the data is tested rather than assumed.