The Copernican revolution

We ended on the puzzle: how is synthetic a priori knowledge possible — how can the mind know, in advance and with necessity, what every object of experience must be like? Kant’s solution is a single reversal so audacious he compares it to the one that remade astronomy. He calls it his Copernican revolution (Preface to the second edition, 1787).

The old assumption. For two thousand years philosophers assumed our knowledge must conform to objects — the mind succeeds when it accurately mirrors a ready-made world “out there.” But this picture cannot explain a priori knowledge. If knowledge is just the world stamping itself on a passive mind, then everything we know must come after the stamping — a posteriori. How could we ever know something about objects before encountering them? On the copy-theory, we couldn’t. Synthetic a priori knowledge becomes a flat impossibility — which is exactly Hume’s conclusion.

The reversal. Kant proposes the opposite: objects must conform to our faculty of knowledge. Suppose the mind is not a blank slate passively imprinted, but an active power that structures whatever it experiences, according to its own built-in forms. Then some features of every object are contributed by the knowing subject — and those features we can know in advance, with necessity, because we bring them to the object rather than reading them off it.

What the mind contributes, layer one: the forms of intuition. Kant argues that space and time are not things we discover among objects. They are the a priori forms of intuition — the built-in format of all sensing. Every outer experience is necessarily spatial; every experience whatever is necessarily temporal. You cannot perceive an object and set aside its being in space and time, because space and time are the very “screen” onto which sensibility projects anything at all. He argues this in the The Critique of Pure Reason Transcendental Aesthetic.

Layer two: the categories of the understanding. Sensibility hands over a raw, unordered manifold; but a heap of sensations isn’t yet the experience of an object. To think a manifold as a unified object requires the categories — twelve pure concepts of the understanding, the a priori rules by which the mind synthesizes sensations into a law-governed world. Among them, the one Hume attacked: causality (also substance, unity, plurality, and the rest). The categories are to thought what space and time are to sensing — the necessary form.

Now the payoff — why synthetic a priori knowledge is possible. Here is the argument in one line. Every possible object of experience must be given through the forms of intuition (space, time) and thought through the categories (causality, substance…). Therefore we can know in advance, with necessity, that every object we could ever encounter will conform to those forms — because an object that didn’t couldn’t be experienced by us at all. Mathematics (synthetic a priori) works because it unfolds the structure of space and time itself; “every event has a cause” is guaranteed because causality is a condition of experiencing anything as an objective event.

And this is the answer to Hume. Hume looked inside experience for an impression of necessary connection between cause and effect, found none, and concluded causation is mere habit read off constant conjunction. Kant agrees you’ll never find causal necessity among the objects — but that’s because causality is not in the scene; it’s on the lens. It is a condition the understanding must impose for there to be any coherent, objective experience at all. So “every event has a cause” is not a psychological habit and not a perceived feature of things — it is a necessary rule of any possible experience. Hume searched in the right place and drew the wrong map.