Abstract

The contemporary interpretations on the a priori in the philosophy of science have been dominated mainly by a semantic and a pragmatic reading. Although the contribution of these conceptions has been valuable, it is my point of view that they have rejected without justification the fundamental meaning of the original Kantian notion, that is, the idea of the constitution of the experience. In this paper I intend to do two things: first, after a presentation of the argument between Friedman and Richardson, both supporters of the semantic and pragmatic readings, I will show the historical reasons that one can use to reject the semantic interpretation of the a priori. Then, I will sketch a conception of the constitutive a priori which, tracking the proposals of philosophers of the 19th century, will insist on the idea that the experience is dominated by a formal function which consists in the production of “invariants”.