The author discusses the extent to which psychoanalytic treatments can be tested nomologically. He concludes that nomologically oriented research operates from assumptions for which no empirical foundation is possible. He further claims that findings of naturalistic studies obtained with the nomological conception of science merely suggest that psychoanalytic treatments are effective and that the specificity of these treatments is overlooked in such research. To resolve this dilemma, the author suggests that naturalistic studies be separated from the nomological conception of science and that the structures of the course of treatments are examined and systematized within psychoanalytic treatment theory relative to their outcomes. A conclusion is that psychoanalytic treatments will be successful if the thus articulated theoretical sequences are actually realized in treatments that also take into account the patient's individuality. This approach, however, requires that the treatment theory is based on a psychoanalytic conceptual common ground.