Bulletin of the Psychoanalytic Research Society, Volume I, Number 1, Spring, 1992
Fifteen years ago, Roger Greenberg and I attempted in our book, The Scientific Credibility of Freud's Theories and Therapy, to appraise the empirical soundness of some of Freud's major ideas. We reported good support for a number of his fundamental concepts (e.g., oedipal theory, character types) and disappointing outcomes for others (e.g., dream theory, superego for nation). Efforts to check out Freud scientifically have met with stiff opposition at one extreme, by experimental purists who do not consider psychoanalytic ideas sufficiency precise or falsifiable to be evaluated empirically; and, at the other extremes, by the hermeneutic block that regards a scientific paradigm as inappropriate for coping with psychoanalytic lore. However, despite such energetic efforts to prove that it can not be done, the last fifteen years have witnessed an efflorescence of research directly or indirectly aimed at testing psychoanalytic formulations. Indeed, it is the very richness of this work that convinced Roger Greenberg and me to undertake a new book to pull it all together.
As one explores the new mountain of research reports, one catches interesting glimpses of trends and paradoxes. Let me briefly mention a few that have particularly impressed me. First of all, it is clear that the more current research is not only imaginative and often ingenious, but also of a higher technical quality. Experimental designs are more adequately complex and signal a more intricate grasp of what Freud was thinking. It is also an interesting paradox to see investigators employing the strictest of "brass instrument" approaches to test the most esoterically imaginative of Freud's notions. I would cite as an example a doctoral dissertation by Rosalind Jones, under the direction of Joseph Masling at SUNY (Buffalo) that recently found (on the basis of a beautifully controlled subliminal input procedure) solid support for Freud's oedipal penis-baby equation construct.
The conclusions appearing in the new wave of research publications support the position, taken by some of us, that Freud's theories do not represent a monolithic structure. Although his formulations are loosely reticulate, they can be sensibly partitioned into a series of mini-theories. Thus, the validity of the dream theory is a matter largely separate from the validity of the Oedipal schema, or the soundness of the homosexuality-paranoia construct or the integrity of the theory of depression. The new research literature repeatedly presents us with instances in which "chunks" of Freud's ideas are well supported, whereas others are not.
Another impression I have is that some of the most exciting and challenging new data pertinent to psychoanalysis are coming, not from clinically-based researchers, but rather from others like social or cognitive psychologists and anthropologists. By way of illustration, consider that cognitive psychologists are describing experiments devised on the basis of cognitive schemas that lend a new class of solidity to a presumption that unconscious factors influence perception and thinking. Note too that social psychologists are in the forefront of those who are systematically clarifying the dynamics of depression, and one should underscore their revelations concerning depressive realism and their demonstrations of the protective value of optimistic illusions. The work on illusions stands as a possible serious challenge to some of Freud's assumptions about the importance of realistic insight for maintaining psychological equilibrium. I must also mention the intriguing anthropological accounts of New Guinea cultures that open novel vistas on the socialization experiences that encourage or discourage homosexuality and point up the plasticity of the boundaries between homosexual and heterosexual behavior.
Although I have advocated that all psychoanalytic propositions be tested scientifically, I do not think it is sensible to legislate a rigid boundary between the "scientific" and the "clinical." Not only are clinical observations the mother-load of hypothesis-construction, but also often the initial staging area for checking the likelihood that a newly conceived concept is even remotely reasonable. Indeed, I would accept clinical observations as having scientific value if they are repeatable and conform to at least certain bookkeeping requirements.