Theories of the Unconscious and Theories of the Self: Edited by Raphael Stern. Hillsdale, N.J. and London: The Analytic Press. 1987. Pp. 282 + xiii.

This volume undertakes the task of obtaining a more unified view of the unconscious, the theory of the self, and the emotions. Why do it? Do the theories have any relevance to each other? What can be learned in the attempt? If anything not previously known can be learned, is that useful to psychoanalysts? The philosophical study of psychoanalysis is a topic of current interest to psychoanalysts. At a meeting of the American Psychoanalytic Association in 1987, an afternoon and the entire next day were devoted to the topic. Hundreds of analysts attended, discussion was intense. Many analysts care and at least some care passionately about whether analysis is a scientific discipline and what rules could make its methodology scientific. Therefore, this volume is a welcome one.

An essay by Otto Kernberg deals with psychoanalytic theory as modified for use with patients Kernberg considers to be borderline. He asserts that these patients' pathology derives from earlier developmental stages than the pathology of neurotics. He believes that early relationships are expressed in alternating activation of contradictory conscious ego/id states. The deepest, crudest, and developmentally earlier wishes appear in consciousness in the borderline patient. Conscious states are the criteria he uses for pathology. Kernberg looks at consciousness as a clinical concept, dependent on three criteria: (1) perception, (2) sense of differentiation, and (3) distinction between present and past perceptions. On the basis of this definition Kernberg here makes a case for the following hypotheses: there is a first stage of consciousness within which symbolic thinking occurs during peak affective states. In pleasurable affective states the images of mother and infant are undifferentiated just as they are in painful states. In this stage, which corresponds to Mahler's symbiotic stage, pleasurable experiences would be stored separately from unpleasurable ones. At the next stage, good