At least as important as developing a philosophical theory of what to count as a mental representation is a theory of what to count as a representational state – a state expressed in English by a relation between an individual and a proposition, that is best explained in terms of (or just is) a relation between an individual and something like a mental representation of (the meaning of) the proposition. Lillard (1993) notes that, simply in virtue of using a pen to stir my coffee, I do not necessarily represent my pen as a spoon, so using a pen to stir my coffee is not a representational state. Following Josef Perner, she calls using a pen as if it were a spoon an instance of hypothetical (as opposed to symbolic) substitution. The difference between the clearer cases of hypothetical (as opposed to symbolic) substitution are reflected in facts such as that “I am using a pen as if it were a spoon” is not, surface-grammatically, a relation between an individual and a proposition – an independent clause does not follow “as if.” But English could have been different here. We could have said “I am using a pen that it is a spoon;” perhaps that is roughly how locutions of this sort work in other languages. And maybe English does have a relation between an individual and a proposition in the deep structure of sentences reporting hypothetical substitutions.
I guess we could just say that hypothetical substitutions aren’t relations to mental representations because they are extensional, and relations to mental representations aren’t. For example, my intuition is that, if I am not pretending, then I am acting as if I am Clark Kent iff I am acting as if I am Superman iff I am acting as if I am the fictional male lead portrayed as having grown up in Smallville in such television shows as Lois and Clark. But that intuition is kind of weak, and I suspect that I would give it up in the face of a good argument. Do other people share this intuition?