596
PARTISAN REVIEW
inclined to stress the difference between two "granunars" or "language
games" that are at issue, the one involving knowledge of efficient causa–
tion, the other, pursuit of teleological intelligibility. A cause, Wittgenstein
argues, is something that cannot be known but only conjectured, and that
can only be established by observation of empirical correlations between a
cause and its purported effects. A person's recognition or avowal is irrele–
vant to establishing the causes of his action; it does, however, play an
important role in establishing reasons.
Freud generally portrayed himself as a scientist pursuing causes of psy–
chological phenomena. Wittgenstein suggests, however, that his actual
methods are far more akin to those of an esthetic investigator seeking
agreement as to the aptness of certain comparisons. The shared features of,
say, infantile behavior and adult sexual activity, or of a given manifest
dream and a purported underlying wish, cannot be established by experi–
mental or rigorously empirical methods of investigation; rather, the
phenomena are placed side by side so as to elicit (an eventual) acknowl–
edgment by the person being interpreted-a method that may be
appropriate for establishing reasons but not for validating causal hypothe–
ses. As Bouveresse points out, Wittgenstein does not deny a certain kinship
between the language games of giving causes and giving reasons; he nev–
ertheless believes that Freud's tendency to conflate these language games
constitutes a serious granunatical "muddle" that has created an "abom–
inable mess" of theoretical confusion.
Bouveresse also mentions a more cynical interpretation of Freud's con–
fusion of reasons and causes. The philosopher Frank Cioffi argues that (what
he sees as) the illegitimate psychoanalytic tendency to speak of reasons as
if
they were causes has actually been "ingeniously exploited" by Freud and his
disciples in the interest of protecting psychoanalytic theory from any possi–
ble source of refutation. It allows Freudians to treat a person's
acknowledgment of a hypothesized motive as counting in favor of the
hypothesis (since it
is
a kind of reason), without forcing the Freudians to treat
refusal as counting
against
the hypothesis (since the hypothesized motive is
also a cause, thus independent of possible acknowledgment) . And this, insists
Cioffi, smacks less of grammatical muddle than of a certain grammatical flair,
a flair for inununizing oneself against all possibility of epistemological risk.
As I have suggested, Bouveresse's book is at its best when he lays out
Wittgenstein's critical views concerning the pseudo-scientific aspects of
psychoanalysis and its affini ties with metaphysical philosophy. It is less
effective in conveying the complex set of affinities and divergences of sen–
sibility and ambition that made Wittgenstein feel, at the same time, both
profoundly attracted to and repulsed by the Freudian perspective. As
Wittgenstein himself realized, sometimes to his own despair ("I destroy, I
