48

The a priori / empirical borderline has weakened in many recent theories of knowledge. This weakening can be taken in various (and even opposite) senses. This paper tries to draw clear distinctions between thoses senses, to connect them with different conceptions of a priori and then proceed to an analysis of the wittgensteinian notion of “form of representation” and its implications, trying to show that changes in the forms of knowledge cannot be reduced to grammatical (in the wittgensteinian sense of the word) or semantical changes. The end of the paper is a reflexion on the conditions that an historic point of view on the form of knowledge has to fullfill in order not to compromise itself with cognitive relativism.