SUSAN HAACK
561
tainable. The proper response is that, unless and until more evidence is
available, scientists had better suspend judgment - and that the lay public,
philosophers included, should not be too uncritically deferential to scien–
tists' sometimes unwarrantedly confident claims about what they have
discovered. Underdetermination, in this sense, has not the slightest ten–
dency to show that we may legitimately choose to believe whatever
theory suits our political purposes .
In
another version, the appeal to underdetermination is intended,
rather, to rest on the Quinean thesis that there can be incompatible the–
ories with the same observational consequences; theories, therefore, be–
tween which not even all possible evidence could decide. Fortunately it
is not necessary to discuss whether the thesis is proven (though it may be
worth noting that Quine himself at one point suggests that what he
elsewhere describes as empirically equivalent but incompatible theories
would really only be verbal variants of one theory). For in any case, if
the thesis is true, it is true only of the genuinely theoretical (in the sense
of "unobservable in principle"); it is irrelevant, therefore,
to
such ques–
tions as whether men's hunting or women's gathering mainly sustained
prehistoric communities . And if it were relevant to such questions, the
feminists' appeals to it would be self-defeating, since in that case it
would undermine their presumption that we can know what theories
conduce to the interests of women, or what those interests are.
The second line of argument from criticisms of sexism in scientific
theorizing to feminist epistemology urges the untenability of the sup–
posed boundary between science and values, and hence, again, the ap–
propriateness of all owing feminist values to determine theory choice.
In
one version, the argument seems to be that the idea that feminist values
could not constitute evidence with respect to this or that theory rests on
an untenable distinction of descriptive versus normative. This argument is
only as good as the reasons for thinking the required distinction unten–
able. What is at issue is not whether moral or political criticisms of
priorities within science, or of uses of the findings of science, are ever ap–
propriate; not whether some epistemic norms may turn out to be
covertly of a descriptive, means-end character; but whether it is possible
to derive an "is" from an "ought." I can find no argument in the litera–
ture that even purports to show this, and neither can I think of one.
That it is false is manifest as soon as one expresses it plainly: that proposi–
tions about what states of affairs are desirable or deplorable could be
evidence that things are, or are not, so.
In
another version, the argument seems
to
rest on the claim, not
that the distinction between fact and value is in principle untenable, but
that in practice it is impossible entirely to exclude "contextual" (that is,
external, social and political) values from scientific theorizing.
In
this
