Kant famously argues that space and time do not belong to objects perceived, but rather are transcendental conditions for the perception of any object. He begins by noting, “That in which alone the sensations can be posited and ordered in a certain form, cannot itself be sensation; and therefore, while the matter of all appearance is given to us a posteriori only, its form must lie ready for the sensations a priori in the mind, and so must allow of being considered apart from all sensation” (Critique of Pure Reason, in The Modern European Philosophers, 387). Translated: if there are any conditions required for the perception of some object, then of course those conditions themselves cannot be perceived; they’ll belong to the cognitive structure of the perceiver, and not (at least not knowably) to the perceived.
He then applies this line of reasoning to space: “Space is not an empirical concept which has been derived from outer experiences…On the contrary, this outer experience is itself possible at all only through that representation” (389). This is why “we can never represent to ourselves the absence of space, though we can quite well think it as empty of objects” (389). It is “the condition of the possibility of appearances, and not as a determination dependent on them” (389). The second sentence in this paragraph is key: ingredient to a thing’s perceivability is its possibility of not being perceived; to perceive this oak is to set it against a backdrop of other oaks and beeches and firs and dead leaves and a bare winter sky, all distinct from it. But space, Kant insists, isn’t like that: to imagine perceiving nothing is simply to imagine a void, which is to say, empty space.
Interestingly, I think that Pascal might have anticipated this argument of Kant’s. Consider: “The knowledges of first principles: space, time, movement, numbers, are firmer than anything that reasoning gives us, and it is upon these knowledges of the heart and of instinct that reason must support itself” (Pensees, 512b). At the very least, he can be read as claiming that space and time are not among the deliverances of the senses, but, like the existence of numbers, are given a priori. For Kant, of course, only space and time are subsumed under the “transcendental aesthetic” (TA); our perception of movement and our knowledge of numbers are perhaps “synthetic a priori” judgments, not so logically basic as our perception of extension and duration. But nonetheless, Pascal is moving along the track that leads to Kant.
The TA is at the heart of Kant’s distinction of the phenomenal from the noumenal realms: “It is, therefore, solely from the human standpoint that we can speak of space, of extended things, etc….This predicate can be ascribed to things only in so far as they appear to us” (391). “We cannot judge in regard to the intuitions of other thinking beings, whether they are bound by the same conditions as those which limit our intuition and which for us are universally valid” (391). Space is both “empirically real” and “transcendentally ideal” (391). We do not, chillingly, know the thing in itself; we know it only as it appears to us.
I find Kant’s arguments about our mind’s transcendental construction of the world as perceived both compelling, and useful for undermining really thoroughgoing empiricists of the Humean sort. But I don’t quite feel the force of his use the TA to argue for our ignorance of how things “really” are. It seems to me that someone like Berkeley could freely grant Kant’s conclusion that we can’t readily analogize our perceptions to those of other thinking beings — there’s no reason to think that angels, for instance, perceive the created order as a spatio-temporal manifold in the same way that we do (or, as Thomas Nagel pointed out decades ago, that bats do either, for that matter). But Berkeley could also point out that Kant’s dwelling on the supposed gulf between phenomenal and noumenal still trades on the same tired old thesis that he saw causing such problems in Locke, and which gave way to skepticism in Hume: namely, that we perceive mental “Ideas,” which are merely images of things out in the world (what John O’Callaghan calls the “Third Thing Thesis”). If being simply is being perceived, then there’s no problem that I can see: I perceive a chair — which is a bundle of distinct ideas (hardness, a certain height, color, disposition of parts, etc.) within the spatio-temporal manifold that simply is the “transcendentally ideal” condition of all my perceptions, and you perceive the same chair in perceiving the same bundle of ideas, because it is, by the LORD’s gracious condescension to us, constant across perceptions, meaning subsumable within some ideal scientific description of the world.