Canadian Journal of Psychoanalysis. I, Number 2, 1993.: Popper, Grünbaum, and Induction. John Allison O'Neil. Pp. 105-130.

The author understands Grünbaum's philosophy to be a (misguided) reaction to Popper's criterion of falsifiability. This calls for a review of induction. Popper agrees with Bacon and Mill up to a point: whatever generative and enumerative induction might produce, if anything, is futile if not subjected to post hoc competitive selection. However, for the theory to be empirical in any logical sense, falsifiability must be possible in principle: we can never know if the surviving hypothesis, which we have good reason for believing to be closer to the truth than its refuted competitors, is in fact true. According to Popper, so-called observation statements are theory-laden and thus hypothetical. And to talk about “raw clinical data,” or the “clinical observation,” is pre-Darwinian. Further, all meaning originates in myth, group myth, or personal myth (fantasy), and the logic of research can say only “No.” As science, and in contrast to behaviorism, which is incorrigibly unscientific, psychoanalysis is logically possible but not yet refutable because it fails to predict overt behavior, and because it makes ready use of polar opposites.

Turning to Grünbaum's contributions, the author first points to a major inherent contradiction in his approach to Freud. On the one hand, in defending himself against the attack of imposing on Freud's clinical theory an extraneous methodological purism, Grünbaum first recalls that Freud's criteria for theory validation were those of hypothetico-deductive inductivism; but on the other hand, he then goes on to state that his application of Freud's criterion of scientific rationality is not what he himself (Grünbaum) considers a valid criterion of demarcation between science and nonscience. As O'Neil justifiably signals, here Grünbaum declines to embrace the philosophy of science he attributes to Freud, but nevertheless states that his entire