The fourth lesson focused first on the epistemological objections to conceptual analysis and some possible responses. Among the objections: the approach is useless (science would make all the necessary reductions ). It does not pay enough attention to real cases. It relies on a definition in terms of necessary and sufficient conditions, which has its limits. We have pointed out some possible answers, without underestimating certain persistent difficulties (reminded, in particular, by Susan Carey's analyses in her recent book on the origin of concepts). We also evoked a partial but unsatisfactory response in the form of fictionalism or neutralism (Norman Malcolm, Wittgenstein), and suggested new, more fruitful avenues that would be based, on the one hand, on the useful distinction between various forms of necessity and, on the other, on the maintenance (contested by some good arguments in the twentieth century from Quine, Kripke, Putnam) of a distinction between a priori and a posteriori , whose respective domains should be better delimited.
We then went on to examine the more directly metaphysical objections, stressing that the logical possible is not a good consequence of the real possible. But we also recalled the importance of metaphysical treatment of the possible (Duns Scotus; J.E. Lowe) and the need to maintain a link between a priori and a posteriori (without reducing the former to the latter).
