ÁñÁ«ÊÓƵ

Stuart Sutherland's review of my Why Freud Was Wrong: Sin, Science and Psychoanalysis, (THES, October 20) confuses Ernst Fleischl with Wilhem Fliess, dismisses by magisterial assertion a complex argument about Freud's messianic identity and castigates me, oddly, for not offering a definition of "religion".

Most surprisingly of all in his rather surprising review he attributes to me a position from which I explicitly dissociate myself. For nowhere do I propose that any alternative theory to psychoanalysis "should be constructed more through our feelings than by science". In fact I specifically warn that any attempt to rely upon "our most impulsive insights" would be "foolish and dangerous", adding that "we still need all the sceptical, systematic and evidence-demanding conventions which (are among) the most valuable elements in our scientific tradition" (pp. 503-4).

It is true that, like the neuroscientist Antonis Damasio, I oppose the dualism of "reason" versus "feeling". It is also true that my book contains an extended critique of rationalism and of the mind-body dualism which has been its main citadel throughout the centuries.

But the purpose of this critique is not to reject science. It is to point out how psychologists (including Freud) have managed to avoid the demands of genuine scientific empiricism by sweeping the larger part of human behaviour under the psychological carpet and building theories on the basis of the mentalistic (or experimental) extract which is left. My plea for theories is not based on feelings. It is for a theory based on evidence which has not been preselected according to spiritualist, creationist or rationalist presuppositions about human nature. Psychoanalysis, I suggest, does not meet these criteria, largely because it was conceived in crypto-theological terms as a theory of "mind".

Stuart Sutherland believes that since previous theories of human nature have been failures, we should abandon the attempt. But pseudo-scientifc theories of human nature - Lacanianism, Levi-Straussian structuralism, sociobiology - continue to proliferate. What I argue in the last part of my book is that bad scientific theories can only be driven out by good scientific theories and that a crucial element is missing from the current Darwinian paradigm.

Sutherland is entitled to disagree. But when he portrays as anti-scientific an argument which criticises as "crypto-creationist" the kind of rationalism he favours but which remains explicitly committed to scientifc empiricism, he only weakens his own position.

RICHARD WEBSTER Southwold, Suffolk