The belief in self-consistent logical systems of argument must also be accepted on faith, without logical proof. There is no falsifiable theory that proves fals

The belief in self-consistent logical systems of argument must also be accepted on faith, without logical proof. There is no falsifiable theory that proves falsification as the final criterion for distinguishing science from non-science, etc. It's your faith-based attachment to the the falsifiability demarcation theory and to a metaphysically "laden" notion of logical validity that causes the impasse. All of these "defeater"-style logical arguments ultimately rest on a kernel of faith. Foucault, or any other decent theoretician, would recognize in this the expression of power that comes out of a particular historical tradition of knowledge rules. The empowered popular discourse they are rooted in is exactly what Foucault would call an "episteme". (Which again, few thoughtful practicing scientists would claim they "believe" in. Rules are always provisional frameworks for the complicated objectives of science, which themselves exist inside a sort of network or market. Nothing is simple or linear.) See my other response above about the materiality and contingency of linguistic utterances. Like all other physical phenomena, the scientific approach views language as an event, a behavior, a material process that is always context-dependent and, at the "lowest" levels, pervaded by uncertainty, even in our best attempts to model it.