I have discussed the role of prediction in relation to theory building in psychoanalysis, with special regard to the theory of psychoanalytic therapy. Though prediction as both a principle and an operational tool plays a central role in the growth and development of most areas of science, its place in our science is much less firmly secured. Freud and many analysts after him have been skeptical of psychoanalysis ever being more than a postdictive science. This position is based largely on the multiplicity of determinants that converge on human behavior as retrospectively unraveled in psychoanalysis, and the wide variety of possible evolutions of human behavior as viewed prospectively from any vantage point in time.

Despite this, as a principle, the predictive process, though usually not made explicit, is basic to all clinical analytic work; every responsible action in analytic diagnosis and treatment actually involves predictions, usually unacknowledged and only