But the last line suggests a question: does Wittgenstein’s de-psychologising of psychology, by blurring the boundary between the inner and third realms, also blurs the boundary between thymology and praxeology, inasmuch as logical constraints are extended so as to constrain not merely the possible forms of action but also their possible contents?
Mises might perhaps say no, since he had all along taken praxeology to function as a logical constraint on permissible thymological attributions, and had not inferred from this any blurring of the boundary. But Wittgenstein makes a still stronger assault on the boundary. Mises tends to trear praxeological truths as universal and necessary, and thymological facts as variable and contingent; Wittgenstein, however, considers a third category: those features of human action that are not universal, but are logically guaranteed to hold for the most part. Nothing guarantees, for example, that any given attempt at a chess move will be in accordance with the rules; there’s always a chance that the player will (accidentally or deliberately) start moving the rook diagonally, or fail to notice that she has moved her king into check. But inasmuch as the game of chess is defined by the system of practices constituting it, there is no danger that, throughout the history of chess, most of the chess moves (or attempted chess moves) that have been made were in violation of the rules; mistaken chess moves can only be the exception rather than the norm. And what guarantees this is nothing mysterious – not a magical force, say, that prevents players from deviating too often from the rules just as God prevents the Pope from pronouncing false doctrines ex cathedra – but rather because an upper limit on mistakes is necessary for the players to so much as count as playing chess. Hence a question like “how often do chess players make mistakes?” – which Mises would presumably have considered purely thymological and a posteriori – turns out to have an a priori dimension. Whether we should respond to these
20 Stanley Cavell, Must We Mean What We Say? (New York: Charles Scribner’s Sons, 1969), p. 91.
R. T. Long – Wittgenstein, Praxeology, and Frege’s Three Realms – p. 13