counterfactual reasoning and mind-reading ability is rather tempting.
After all, the central representational capacity of mind-reading consists in
handling representations which have the structure:Agent – Attitude –
Content. In processing these triadic representations the very same contents
can be embedded in diVerent attitudinal contexts. Thus I can fear what you
hope for; and I may not expect something to happen, but then be disap-
pointed or pleasantly surprised that it does. In particular, people with a
proper representational concept of belief will be able to avoid confusion
between what they believe and the contents of beliefs attributed to others.
Counterfactual reasoning involves something at least partially analogous,
in the way of free-wheeling manipulation of contents accompanied by
some way of marking oVwhat is actually believed by the reasoner from
what is merely being supposed.
Simulationists have urged that the use of suppositions involved in
counterfactual and/or hypothetical reasoningWts their model of simula-
tion. We know that people engage in this sort of reasoning, and so it is
claimed to be an advantage of simulationism that we get an account of how
mind-reading works, essentially for free. Further, they claim that in the
case of the autistic impairment in executive function, what we are witness-
ing are the eVects of a deWcient ability to simulate.
We do not think that theory-theory is at any disadvantage in accounting
for problems with counterfactual thinking, however. Remember that our
hybrid view allows simulation to have a place within the application of
theory of mind. We have explicitly said this in relation toinference-
enrichment(section 3 above). We reject simulationism, but we do not reject
simulation as a cognitive tool. Given that people canWrstdistinguish
between suppositions and their probable/possible consequences, on the
one hand, and what they actually believe, on the other, then they can use
the capacity to generate suppositions in order to simulate. But even if there
is a case for thinking about counterfactual reasoning as a form of mental
simulation, this should not be taken to exclude the sort of self-knowledge
which on the theory-theory requires application of a theory-of-mind to
one’s own mind. For while counterfactual reasoningmayinvolve simula-
tion, it is even clearer that it involves the very sort of monitoring of
psychological states for which a theory-of-mind system is needed – the
ability to keep track of the scope of a supposition, to diVerentiate one’s
actual beliefs from the counterfactual supposition, and to make appro-
priate modiWcations in some of one’s background beliefs (Carruthers,
1996b).
Indeed, our general view about executive function and counterfactual
thinking is that simulationmaybe involved to a greater or lesser degree
(depending on the sort of thinking required for solving diVerent problems),
102 Mind-reading