KANT'S CLASSIFICATION

Definitions:

Analytic statements are true because of the meaning of the terms.
Their denial would be self-contradictory. (Hume’s “relations of ideas.”)

Synthetic statements bring together concepts that are not logically connected. They could be false. (Hume’s “matters of fact.”)

Something is known a priori if it can be known in advance of actual experience. Thus all analytic statements are a priori.

Something is known a posteriori if it can be known only after experience. Most ordinary statements of fact are a posteriori.

Question is whether there are synthetic a priori statements. They would be known in advance of any particular experience, but would say something about the nature of things in a way that analytic statements do not. An example is “the same causes always have the same effect.”

In general, rationalists assert that there are synthetic a priori statements (Descartes’s “innate ideas”), and empiricists deny that there are.

Kant argues that each side has a problem. The rationalist can’t explain how we happen to know the synthetic a priori; empiricists can’t explain the world without them, and thus fall into skepticism.

Kant is looking for a way of resolving this problem. Since the mere denial of the synthetic a priori cannot account for our knowledge and there is no point in merely saying that we know these things, we must ask a different question: How is synthetic a priori knowledge possible?