Theoretical psychology in its present state must try to develop a system of concepts which shows all the characteristics of a Gestalt, in which any part depends upon every other part. As we do not yet have the knowledge of facts which really suffices to determine this system of concepts and as, on the other hand, this knowledge of "facts" cannot be acquired without developing this system of concepts, there seems to be only one way open: to proceed slowly by tentative steps, to make decisions rather reluctantly, to keep in view always the whole field of psychology, and to stay in closest contact with the actual work of psychological research.

p. viii. - Principles of topological psychology, 1936