psychology needed to work well in small, tribal groups; and there is no
reason to think that they do not. That makes it unlikely that its principles
are radically incorrect.
But what if Stich turns out to be right? Would that mean that we should
then abandon folk psychology, and agree that there are no such things as
beliefs and desires? That really is more than implausible. The practical
utility of folk psychology falls short of being a proof that it is actually
correct. But it is a most persuasive reason for thinking that it is eVectively
indispensable – as Stich, too, acknowledges. So if the programmatic elim-
inativists turn out to be right, then the best bet about how we would react is
this. We would concede that strictly speaking there are no such things as
beliefs and desires, but most of the time we would not feel the need to speak
strictly. On the contrary, we would need to speak loosely and roughly – so
that we would become, so to speak, pragmatic instrumentalists. That may
sound odd. But it is really much the same as physicists’ attitudes to
Newtonian mechanics. Theoretically it is false. But for most technological
applications it gives results which are accurate enough, and is so much
more convenient to use than relativistic theory.
5 Using folk psychology
Since we take folk psychology to be broadly correct in its major causal-
functional categories, there is no reason for scientiWc psychology to ignore
the intentional states it postulates. ScientiWc psychology tried to do just
that during its behaviourist phase, and the results were not encouraging.
We think that scientiWc psychologists should not be embarrassed about
relying on certain aspects of the psychology of the folk.
It has been argued that folk psychology and scientiWc psychology are
more or less unrelated, however – claiming that the latter can and should
develop independently of the former; and that we need not concern our-
selves about integrating folk psychology with cognitive science and neuro-
science. On this view, what physicalists should assume is just an integration
ofscientiWcpsychology with neuroscience, with folk psychology being left
to the folk. This position has been defended in a number of places by
Wilkes (1978, 1991a, 1991b).
One of Wilkes’ main arguments is that folk psychology is quite a
diVerent kindof enterprise from scientiWc psychology, because it has
diverse diVerent purposes. For example, one needs to use folk psychology
in order to persuade, cajole, threaten, warn, advise, seduce, and console
others. This is certainly true, but we do not see it as a good reason for
supposing that there will be no interconnections between folk psychology
and scientiWc psychology. And in particular, it is very hard to see how folk
46 Folk-psychological commitments