It has been claimed that psychoanalysis cannot be given a scientific f
oundation and that psychoanalysis is therefore untenable. One response
that has been given to this claim is that psychoanalysis does not req
uire a scientific foundation, that it is an autonomous enterprise from
the enterprise of science. In this paper, I argue that psychoanalysis
indeed requires a scientific foundation, and I offer a sketch of how
connectionism might hold promise of yielding a new ,,metapsychological
'' theory that could vindicate psychoanalytic clinical theory.