Abstract
Why, for such a long time, has there been no Kantian point of view among the most influential theories about thought experiments? The primary historical reason the main trends in the philosophy of science have always rejected the existence of a priori knowledge fits a theoretical reason. Kant oscillated between two very different views about the a priori: on the one hand, he attributed to it a particular content, whereas on the other hand he insisted on its purely formal character. The first interpretation of the a priori was one-sidedly developed by Ørsted and led to his untenable notion of thought experiment. The other interpretation, however, can be further developed into new Kantian theories of thought experiments in the natural sciences and in philosophy.