Letters

To the Editor:

I am pleased that Robert Holt devoted so much thought to reviewing the MIT Press edition of my Freud Evaluated (Psychoanalytic Books, 8, 397-410). Although what he says is very critical, he does conclude that it should be read, especially by “the Freudian faithful.” I therefore trust that any such among your readers will find my response also worth reading.

1.1 Holt begins by saying my “methodological principles” derive from an unknown author and are based on the “not very satisfactorily defined term” of characterisation. Actually, my general methodological principles do not derive from O'Neil at all: I use only those of his theses relevant to the ‘reality’ of theoretical terms. One might argue that they are inadequate, or outdated, or that my use of them is misguided, but is it any sort of an argument that O'Neil's work is not known to Holt? Holt's charge of definitional vagueness might also have some substance had I not devoted most of chapter 6 to using characterisation as a basis for a detailed comparison of the ‘reality’ of the mechanisms by which Charcot, Janet, Breuer, and Freud explained symptom formation. Holt understood characterisation well enough there to see that Luria-like experiments might be used to investigate Charcot's proposal. My explicit realist stance and rejection of MacCorquodale's and Meehl's position on theoretical terms should also have warned Holt there is no similarity between operational definition and characterisation.

1.2 In stressing the deductive structure of explanations, it may have been remiss of me not to have mentioned the status of the theoretical postulates. I do not see the omission as damaging, because false explanations are also deductive in form, and not only because the postulates are false.

1.3 When introducing my final remarks about the so-called structural theory, I say that Freud's realism has the consequence of rendering