Analysis 60.1, January 2000, pp. 112–23. © Bernhard Weiss
Generalizing brains in vats
Bernhard Weiss
Wright (1994) provides an anti-sceptical argument which highlights the
kernel of Putnam’s argument against such hypotheses as that we are brains
in vats. Wright claims that his formulation is effective against a certain
range of sceptical hypotheses, namely, those which we can understand. I
argue that the argument can be made more generally applicable, providing
a uniform refutation of a class of sceptical hypotheses. To be more precise,
I claim that such a uniform refutation is available on condition that the
original argument holds. There has been some discussion of whether the
restricted applicability of the argument constitutes a failure in its ability to
furnish a refutation of metaphysical realism. The generalization I offer
should suffice to dismiss that worry. However, as I explain, for rather
different reasons the argument still won’t be effective against all forms of
metaphysical realism nor, indeed, against all forms of scepticism.
1. The Wright-Putnam argument
First then to Wright’s formulation of Putnam’s argument. It proceeds as
follows:
(i) My language disquotes.
(ii) In a language, L differing from mine, if at all, only in respects
dictated by the truth of H, ‘F’ does not refer to F’s.
(iii) In my language, ‘F’ is a meaningful expression.
(iv) In my language, ‘F’ refers to F’s – from (i) and (iii).
Hence
(v) My language is not L – from (ii) and (iv).
But
(vi) If H is true, my language is L – from (ii).
So
(vii) H is not true. QED (1994: 236–37)
Wright then goes on to note that the manner in which this argument might
be wielded against metaphysical realism is to note that the argument can
be seen to be effective against any specific description of a state of affairs
whose possibility the metaphysical realist is committed to. So, although the
metaphysical realist would not be formally inconsistent in claiming that
some such state of affairs is possible, she would be committed to something