Using Concepts

Using a concept involves applying it to the real world. Since possessing a concept involves being able to use it, it follows that the possession of a concept commits us to applying that concept in various ways, and that these applications must be generally reliable and accurate in order for us to possess the concept at all. And from this it follows that one must assent to certain factual propositions employing the concept in order to count as possessing it in the first place, so that no "analytic' use of a concept is intelligible unless it is embedded in a network of "synthetic" uses of that same concept. Hence "propositions of the form of empirical propositions, and not only propositions of logic, form the foundation of all operating with thoughts (with language)” (Wittgenstein). But in this case it no longer makes sense to ask whether conceptual truths are "analytic" or "synthetic." The analytic/synthetic distinction itself presupposes a separability of concept from application that cannot be sustained.

Our conceptual truths are usable only on the assumption that various empirical statements hold. These empirical statements are not themselves conceptual truths, but if they were not to hold, we would not be able to employ our concepts. It is not as though the falsity of the empirical statements would falsify our conceptual truths; that would make the conceptual truths themselves into empirical statements, which they precisely are not. The denial of a conceptual truth employs the constituent concepts of that truth just as much as its assertion does; a situation in which our concepts are disabled is one in which the associated conceptual truths can be neither asserted nor denied....

--Roderick T. Long, "Anti-Psychologism in Economics: Wittgenstein and Mises"  
The Review of Austrian Economics, 2004 (citations omitted)