logical preponderances, as we may call them, by saying that not only praxeology but parts of thymology are a priori, or instead that some things we thought were parts of thymology must be reassigned to praxeology, or else that praxeology and thymology turn out to overlap, is perhaps a matter of philosophical taste. (I lean toward the third option: “Different concepts touch here and coincide over a stretch. But you need not think all lines are circles.”21)
In addition to logical preponderances, there is also a source of boundary-blurring – among all three realms this time – in Wittgenstein’s solution to the problem of thymological opacity. Wittgenstein argues, on the one hand, that a person does not count as having a mental state unless she has a tendency to express it in her actions, and on the other hand, that a person does not possess the concept of such a mental state unless she is reasonably reliable at identifying its expressions in the actions of others. This means that any subject capable of forming judgments about others’ mental states is going to be right most of the time – in which case thymology is logically guaranteed to be mostly reliable. (Indeed, the reliability of thymology turns out to be itself an instance of logical preponderance.)
This might be taken to imply that Wittgenstein thinks we recognise other people’s mental states by making deductive inferences of the form “S is exhibiting overt behaviour X; therefore S is more likely than not to be in mental state Y.” In fact Wittgenstein thinks no inference is ordinarily involved, and that in standard cases we simply see the mental state in the behavior; that is, we ordinarily perceive actions – indissoluble unities of inner and outer – and not mere physical movements.
Again, I’ve discussed Wittgenstein’s case for all these claims in detail elsewhere and so won’t review it now. But the crucial point to notice is that Wittgenstein’s account seems to deprive the inner realm of the privacy that for Frege was so essential to distinguishing it from the other two realms. The privacy is not really abolished, however; it is merely transfigured. Recall that
21 Ludwig Wittgenstein, Philosophical Investigations, 3rd ed. (New York, Macmillan, 1963), p. 192.
R. T. Long – Wittgenstein, Praxeology, and Frege’s Three Realms – p. 14