Experimental Philosophy , Conceptual Analysis , and Metasemantics ∗

Empirically informed philosophy is nothing new. Despite aspersions about armchairs, the best philosophers in every era have made serious efforts to produce philosophy informed by the science of their day. In the last decade and a half, however, a new form of empirically informed philosophy has arisen: experimental philosophy, which probes folk judgments about philosophical concepts. Why probe folk judgments? The basic idea runs something like this: When we theorize, we use thought experiments. In these experiments we make snap judgments about philosophically interesting concepts. These judgments are supposed to provide a sort of pre-theoretical constraint on theorizing; if a proponent of a counterexampled theory rejects the counterexample by appeal to the theory itself, he’s not doing philosophy right. But if the judgments are supposed to be pre-theoretical, then they had best not be subtly influenced by other theoretical commitments. Since philosophers have had long exposure to the relevant theory, maybe their judgments shouldn’t be relied on either. Better instead to rely on ’the folk’, people with no prior philosophical exposure to taint their judgments. Foes of experimental philosophy often object to this line of thought as follows: “If we probe the folk about, say, free will, we get judgments about what the folk think free will is like. But why do we care about this at all? We want to know what free will really is — not just what people think it is. After all, we wouldn’t dream of probing folk judgments about simultaneity in a quest to determine the truth of special relativity. The theory of special relativity is a theory about what time is really like, and its truth or falsehood doesn’t depend on what ordinary people think. Why should free will be any different?” Experimental philosophers have often responded tu quoque: If, like time, the nature of (say) knowledge or free will isn’t constrained by pretheoretical judgments, why do philosophers persist in appealing to their own judgments about cases while theorizing? But while the tu quoque may be dialectically effective, it is philosophically unsatisfying. It would be nice to have some sense as to why pretheoretical judgments have probative force when it comes to free will but not when it comes to time. My aim here is to sketch a picture of philosophical theorizing that can answer that question. The idea, roughly, is that what free will or time really is just is whatever content our concepts of free will or time actually pick out. And what these