Logical empiricism, a powerful and in many ways sanitary philosophical movement which was started by the “Vienna Circle,” propagated in England mainly through Wittgenstein, and in the United States mainly through Carnap, has always been committed to some kind of “linguistic” or “conventionalist” theory of necessary propositions, though it would be dificult to pin down a party line as regards the precise form of such a theory. Such jargons as “the laws of logic are rules of the transformation of symbols,” “all a priori knowledge consists in decisions concerning the use of symbols,” are well known to students of logical empiricism. “Logic formulates rules of language—that is why logic is analytic and empty,” writes a famous logical empiricist. In a sense this theory denies that there is such a thing as a priori knowledge, knowledge of necessary propositions. If, as Schlick wrote, “7 + 5 = 12” is just a rule of symbolic transformation, telling us that we may interchange “12” and “7 + 5” in any context, and a proposition is something that is true or false and that may be believed or disbelieved, then this equation does not express a proposition.