Freedom, Fact, and Fiction

One of the most puzzling aspects of Kant's philosophy is his attitude towards freedom. Kant denies that we can know that we're free, in any strong metaphysical sense. But he maintains that we are entitled to assume that we're free. We, as it were, announce our freedom, and this announcement has priority over any empirical science. The priority seems to be a transcendental priority: if we weren't free, we couldn't do science, either, since it is the same faculty -- that of reason -- which is deployed in science or in morality.

If empirical science could show that we're not truly free, it would also, at the same time, show that we're not really capable of science. It would self-destruct.

There's a similar argument, I think, as to why we're entitled to accept responsibility for our actions. We cannot know that we're responsible, in the sense that we cannot control what our actions do in the world. Foucault is completely Kantian when he says:

People know what they do, and they frequently know why they do what they do, but what they don't know is what what they do does. (in Foucault, Dreyfus and Rabinow; cited as "pers. comm.")

Exactly: we don't know -- and, Kant would say, we cannot know -- what what we do does. In the face of this massive and incomprehensible ignorance, we must announce that we are responsible, act as if we are responsible, and precisely through acting as if we are -- become responsible.

Vaihinger was right: Kant's philosophy is a philosophy of the "as if." It is perhaps in this respect that comes closest to anticipating not only Nietzsche but also Deleuze's theory that there is something fictional about concepts. Not that concepts are fictions, in any simplistic way, but that concepts can open up new ways of being, much as works of art do, rather than as simply fitting "how things really are."