Commentaries

1There is a certain force to Wolff's arguments. They are nested within a growing body of critical proposals aiming to carve out a scientifically rigorous method and data base for psychoanalysis. These proposals, by and large, are welcome and important contributions challenging analysts to approach their theorizing more carefully. Wolff's contribution to this effort highlights some possible directions psychoanalysis might take in evaluating the theoretical yield of infant research. For example, inter-domain translation rules, what Wolff calls formal and theory-derived decision rules, are urgently needed. These potentially provide for interesting, equitable, and mutually enriching commerce between data derived from different methods. This is the case not only for the boundary between infant research and psychoanalysis, but also for that between many other cognate sciences and psychoanalysis. Reasoning by prediction rather than postdiction, avoiding the post hoc, ergo propter hoc fallacy, taking care to control for the seductive error of what is called “affirming the consequent”—or at least being acutely aware of each of these as a potential source of error—must impel future theory construction. Perhaps Wolff's hope for the future of nonlinear systems dynamics, as exemplified by the promise of chaos theory, holds the key to such directions.

Then, too, as Wolff notes, similarity as a basis for attributing conceptual overlap and causality constitutes such a startlingly naive stance toward higher-level theory construction that it makes one wince. Wolff's argument here is important. This should be exposed as flawed method. Attributing causality between items in any perceived set merely because they possess some pattern match was known and warned against centuries ago, yet it tenaciously persists. In psychoanalytic theorizing, similarities can easily be found between two classes of items on virtually any psychological axis, but this does not tell investigators whether they a