Now we can begin the big argument. We have finished the preparation, technical but necessary for understanding what is at stake. We are now going to see Kant's answer to the question of how a priori synthetic judgments are possible, by working through a riddle.
I am going to, in real time, predict the future. I am going to predict something that has not yet occurred. Tonight, when I go home and watch the six o'clock news on Bay News 9, specifically Weather on the Nines with Chief Meteorologist Mike Clay, he will be wearing a gray shirt. I am not sure of the exact shade, but some shade of gray. And I am predicting this with absolute certitude. I am not saying it is likely. I am not saying it is probable. I am saying I know it as a fact, even though it has not yet transpired. The riddle is: how can I know this?
Is it analytic? No: it is not part of the definition of a shirt to be gray. It is contingent; shirts can be any number of colors. Because it is non-analytic, it must be synthetic. It is informative, bringing a property, grayness, that is not inherent to the subject. In Kant's terms, it is synthetic. Hume would therefore say I need experience to know it. But the event is in the future, so I have no direct empirical data about it. How, then, could I know it?
Could it be that Mike Clay has worn gray shirts every day this week, this month, or throughout his career? No. Hume has taught us that a past pattern does not give us certitude about future predictions. We cannot simply assume the future will resemble the past, and there is no way to prove that it will. That gives us no certitude. Could I have bribed him, slipped him fifty dollars to wear a gray shirt? No. Aside from his being far too honest for that, he could spill coffee on the shirt at any moment and be forced to change. And yet I know it.
The solution is that I have a black and white TV. Everything on the TV is some shade of gray, somewhere between white and black. Notice how this works. If I only have access to information through one medium, and that medium has a specific effect on everything that comes through it, turning it some shade of gray, then I can know for certain that everything that comes through that medium will have that particular property.
If I learn what property the medium imposes on everything that passes through it, then I know that property will apply to everything I receive through it. And I know this prior to turning on the TV and seeing what is on it. In other words, I know it a priori, relative to the TV. I do not need to see what is actually on the TV, that would be empirical. I know from the TV itself, before I see what is on it, that it will impose this property.
And yet this knowledge is synthetic, because it is not part of the definition of a shirt to be some shade of gray. There is still genuine content; there is still something being synthesized. This gives me a remarkable result: the limitation is precisely what gives me necessary, universal knowledge. Everything on the TV has to be gray. I know this as a law governing everything I will ever watch on it.
The limitation, in other words, is not merely a restriction; it is the very source of the power. Because the medium uniformly transforms everything that passes through it, I can make an unconditional, universal claim about all of its content before I have seen any of it.
We have to be careful here. The whole riddle turns on the ambiguity of what I mean when I say I see his shirt. What shirt am I actually seeing? I am seeing the shirt on my TV. The shirt that I know about, the shirt that I have necessary universal knowledge of, the shirt that I predicted, that is the shirt on the TV, not the shirt actually on Chief Meteorologist Mike Clay's back. I have no idea what the shirt on his back looks like, but I know for certain, ahead of time, independently of checking, what color the shirt on my TV will appear.
This is the distinction that Kant makes between what he calls things for us and things in themselves. There is the way the world itself is, the way the world really is on its own, independently of the way that we look at it, independently of any effect we have on it. In this analogy, imperfect as it is, that would correspond to the shirt on Mike Clay's back: what color it really, truly is.
But there is also the color that I see through the only means available to me. I can only see that shirt on my TV, the only way I can access it. And so there is Chief Meteorologist Mike Clay's shirt for me, as it appears on my TV. That is what I have knowledge of. That is what I predicted. That is what I have necessary, universal information about. That is what I have a priori synthetic judgments about.
Kant's answer to the central riddle of the Critique, how are a priori synthetic judgments possible, is that our minds are like the black-and-white television. Our minds are not purely passive, perfect replications of the world. When we take in sensory data, we do not simply register it like photographic film, or a mirror, or a tabula rasa. The human mind actively takes in that data and organizes it, imposing structure on the information it receives.
We are bombarded with light waves, sound waves, smells, and countless other streams of information, and in their raw form this data is chaotic and unnavigable. In order to move through the world, to avoid the rain, to inhabit an orderly, structured reality in which we can encounter other people and objects, we must impose a certain order on experience, a structure amenable to human comprehension. If we can identify what that structure is, then we know in advance what everything we will ever experience must look like.
That, on Kant's account, is precisely what mathematics and natural science are: the study of the structures that the mind imposes on experience. Just as the only world we ever reach is the world as it comes through our senses and through this structuring function of the mind, so if we can determine what the mind is doing to everything that enters it, we will know something about everything that comes out of it. We will know facts, synthetic, substantive facts, but facts that hold universally and necessarily, about the world as we live in it.
You have to realize what Kant is doing here, and what he is doing is stunning. On the one hand, he takes Hume's empiricism and says, "Yes, the only way we get any knowledge of the world is through sensory data. You are right, Hume. Empiricism has a deep, profound insight. The way I know the world exists, the way I know anything about the world, is by receiving input, otherwise there is nothing. You are right." But you are wrong to think that is all knowledge is, that everything we receive comes simply through the senses.
To the rationalists, to Descartes, Kant says: you are also right that there are things we know independently of experience. It is just that you are wrong to think that such knowledge can reach beyond experience. The things I can know a priori, necessarily and universally, are not derived from experience, but they are only true of experience. On one side, we now have genuine knowledge of the world we see, the world we live in, the world we inhabit, true knowledge, universal knowledge, knowledge we can rely on. Not the weak substitute that Kant thinks Hume left us with.
But at the same time, and for the very same reason, we know nothing of the world as it is in itself. The same idea that gives us the kind of knowledge we want of the world as it appears to us simultaneously denies us knowledge of the world as it is in itself. Kant guarantees physics and forbids metaphysics at one and the same time, with one and the same idea. Kant giveth and Kant taketh away.
He accomplishes this by taking two radically divergent schools of thought that had moved farther and farther apart, and bringing them together as though they were made for each other. He synthesizes them into a harmonious unity, as if they belonged together rather than being incompatible, within one overarching system. The man is a horrendous writer, but his thinking dances. His thinking is graceful and beautiful.
Here is the core argument: the human mind is active in its structuring and organizing of experience. It is not a passive reception. It receives, yes, but it is not purely passive. It is not a pure reproduction; it is a production. The mind has an effect on what comes in. And as long as we can understand what that effect is, we can understand something about everything we will ever experience. If we understand what the faculties of our mind are doing, we will know something about every single thing we will ever encounter throughout our entire lives, with no exceptions.
And that, Kant thinks, is all we really care about. It may feel as though we are giving up something enormous: "I will never know what the world really is in itself." But Kant's response is that we do not actually care about that. We have never encountered the world in itself. Everything we have ever encountered is something already organized and structured, something already turned into a world for us. That is the only place we have ever been. That is where we keep all our things, where we meet all our friends. That is all we actually care about.
We are therefore giving up nothing real, nothing we genuinely care about, and gaining everything important, all through this one change in perspective. This is Kant's answer to the question: how are a priori synthetic judgments possible?
For the rest of the book, Kant takes this central argument and applies it to the three faculties of the mind. He proposes that the mind has three separate faculties, each of which organizes experience in a different way. It is quintessentially Kantian in its structure: three faculties, each contributing something distinct, each generating its own body of knowledge. He will work through each of them in turn, applying the same argument three times across these different domains.
The first faculty of the mind is what Kant calls intuition, and this, it must be said, is a very unfortunate translation.
The word intuition was translated that way early on and it simply stuck, but it has nothing to do with the way we ordinarily use the word. It is not a hunch. It is not a feeling. It is simply the senses, the way we directly take sensory data into our minds. This faculty of the mind is what gives us mathematics. Each faculty gives us a different body of knowledge, and each faculty imparts a different structure to experience.
The structure that intuition gives us comes in two forms, which Kant calls space and time. Sensory data comes at us in a chaotic and overwhelming flood, and the mind organizes all of this information into relationships. To allow us to find our way around the world, the mind arranges incoming information into relations such as "to the left of," "to the right of," "above," and "below." According to Kant, space is nothing more than these relationships, a way of bringing things together into a certain type of order. He inherited this idea from his rationalist predecessor Leibniz.
The same holds for time. Time is simply before, after, and simultaneous with — the ways that events relate to one another. There was in fact a famous debate between Leibniz and Newton on precisely this point, because Newton believed that space and time were a kind of great container, something that would persist as an empty box even if nothing existed within it. Leibniz argued that space and time are purely relational, and that if nothing existed, there would be no space and time at all. Einstein ultimately came down on Leibniz's side, a victory for the philosophers.
What the mind does in imposing spatial and temporal relations on experience is not something you are conscious of. It is autonomic, like the way the brain regulates your heartbeat or digestion, something you do without awareness or control. The consequence is that everything you will ever be aware of is already spatialized and already temporalized. You cannot access the raw data before it has been organized into spatial and temporal relationships, because the only way anything can become present to you is after it has passed through this process.
This means that the world in itself is neither spatial nor temporal. Do not try to imagine what that is like. It is impossible, because our imagination is itself structured by space and time. You can think abstractly that the world in itself lacks space and time, but you cannot form any sense of what that would actually be like, because we can only conceive of a world like the one we inhabit. We are barred from the thing-in-itself forever by the very structure that opens up the world to us at all. That structure opens our world to us while simultaneously closing off the world as it is in itself.
Space and time are also what give us mathematics. Space gives us geometry: geometry is simply the set of rules governing the way intuition organizes incoming sensory data, nothing more. Everything you ever experience will be geometric and will obey the laws of geometry, because you make it so. You cannot encounter anything that fails to be geometric, because to encounter something at all is to pass it through the form of space, which means it will necessarily conform to the laws of geometry. Time, in a somewhat less obvious way, gives us arithmetic.
The second faculty Kant calls the understanding. Instead of forms, as the intuition has, the understanding has concepts. And instead of two, it has twelve of them, though we will only discuss the one that is most interesting: causality. Causality is what gives us science. Just as Hume said, science is the study of what causes what, and Kant agrees entirely.
Hume argued that if science were entirely empirical, it would be contingent, and we could not make predictions with any confidence. Kant draws the same conclusion: science cannot be purely empirical. We know that the laws of nature will remain the same, which is precisely what Hume said we cannot know. We know the future will resemble the past, but the reason we know this is that we make it resemble the past, because our understanding always imposes its structure, its concepts, onto incoming data.
Just as a black-and-white television will always render images in black and white, we will always find the same causal relationships in experience. Not because those relationships are simply out there in the world, for Hume would be right about that. If causal patterns were just features of the external world that we derived empirically, our knowledge of them would be contingent and could cease at any moment. But Hume never considered the other possibility: that we are not deriving these patterns from experience, but imposing them on experience.
That is why everything we ever experience will, necessarily and universally, conform to those patterns. And there we have the laws of physics, what Hume declared impossible. We have laws of physics as a priori synthetic judgments, grounded in the nature of the mind itself. Because we are structuring and organizing experience, we have the capacity for exactly the kind of knowledge Hume thought was unattainable.
The thing that seemed so difficult to explain, the existence of a priori synthetic judgments, necessary and universal knowledge that is substantively informative about the world, turns out to be the very clue that helps us understand the mind. The first two parts of Kant's project explain why we can, and do, possess such judgments, knowledge we can guarantee will continue to hold. Hume said it was impossible. This is why it is not. This is how.
The question now is whether metaphysics also becomes legitimate. Metaphysics claims to be a priori synthetic as well — it claims to have substantive facts about the world, and those facts are not based on experience. This brings us to the third faculty of the mind, which Kant calls reason. Reason is what attempts to give us metaphysics, and it operates through three ideas. Here again we encounter that word, and again Kant means something entirely different from the other philosophers who use it. Ideas are simply reason's version of concepts or forms — the structures internal to reason that it uses to organize experience.
Kant thinks reason is fundamentally different from the other two faculties. Reason has what we might call the demand for completeness. When reason looks at the knowledge produced by intuition in mathematics, or by understanding in science, it says: "That is not good enough. You are only giving me part of the picture. I want it all. I want a total view." When intuition gives us space, reason says that the only satisfying knowledge of space would be a map of its entirety — which is precisely what astronomers pursue when they attempt to chart the whole universe. When intuition gives us time, reason demands to know time all the way back to the very first instant, which is what cosmologists study when they trace the history of the universe back to the Big Bang.
When the understanding gives us causality, reason says: "I want to know all of causality — what causes everything, all the way back, without exception. Only when I possess that totality can I finally be content." Until then, reason keeps driving intuition and understanding to produce more and more. But here lies the problem. The reason reason is never satisfied is not due to any weakness of our minds, not because we have failed to work hard enough or think carefully enough. It is due to the intrinsic nature of the forms of intuition and the concepts of understanding themselves.
Notice that the examples here — space, time, and causality — are not artificially chosen. These are precisely the structures that organize science and mathematics. Kant is not forcing anything. But his point is this: look carefully at these structures, and you will see that it is internal to the very nature of time, space, and causality to be incomplete. Completeness is not merely difficult to achieve — it is impossible for these forms, and here is why.
Imagine that we are trying to map all of space. We send probes out in every direction — faster than light, if we like — and we reach the very edge. We have it. We have the entirety of space. But the moment we arrive at that edge, one question immediately presents itself: what is on the other side? That is simply the nature of space. For any point, there is a point to its left and to its right, above it and below it. There cannot be a final point in space, because it is the nature of every point in space to point beyond itself. There is always a next to. Space, by its very nature, perpetually drives beyond itself and can never be complete.
The same problem arises with time. Suppose we want to reach the very first moment. When we arrive at that first moment, we immediately ask: what happened before it? Because every moment in time has a before and an after — that is how time works. The very notion of a first moment in time is therefore a contradiction in terms. You need a first moment if you are to have a totality of time, yet you cannot have a first moment if it is genuinely to be a moment of time. This is what philosophers call an infinite regress: the chain of befores keeps extending backward forever. Reason demands an end to the regress, but the nature of time makes that end impossible. A moment with no before would be a four-sided triangle — a moment that is not a moment of time at all.
The same structure appears when the understanding tries to grasp causality. To understand any event A, I must understand what caused it — call that cause B. But B is itself an event, and every event must have a cause. So I trace the cause back to C, and then C back to something further, and so on without end. What reason demands is a first cause: a cause that initiates the entire chain but was itself caused by nothing. An uncaused cause. And we have a name for that: God. God is the uncaused creator of the universe, the being who was never itself created. That is metaphysics.
Kant took Hume's notion of causality — which is precisely what science is about — and showed how it intrinsically leads to an infinite regress, driving us, through reason itself, to seek a beginning point. And that beginning point is exactly how people have always talked about God, which is what metaphysics is. Physics drives us to metaphysics. Metaphysics exists not because there have been a lot of peculiar people with some strange appetite for it; it is built into the structure of the mind. The nature of time drives us back and back and back, and forward and forward and forward. The nature of space drives us outward and outward and outward. The nature of causality drives us back and back and back. And what we arrive at are the totality of time and space — a metaphysical idea — and God, also a metaphysical idea. There is a third, the immortality of the soul, which we will not address here.
Metaphysics is an organic, natural growth — the culmination and conclusion of physics itself. Our minds are such that we automatically construct a mathematical, scientific universe, and we yearn for a metaphysical one. The mind itself is oriented toward this, and there is no way to stop it. There is no way to stop that yearning.
But that yearning does not constitute knowledge. There is no first cause. There is no first moment. There is no outer boundary of space. These are impossible to establish. Kant is not saying there is no God — that point must be made very clearly. Kant himself was a deeply religious man. What he is saying is that we can never know whether God exists. Belief in God is just that: belief. It is a matter of faith, not a matter of knowledge. It is not a matter of proof; it is a matter of belief. We must abandon the project of trying to prove it or to know it, because it is simply not that kind of thing. This is something you believe or you do not believe.
In fact, if you were to prove it, you could no longer believe it — because belief, by its very nature, means that you do not have proof. So Descartes was wrong to claim that we could know God exists, and Hume was wrong to claim that we could know God does not exist, because the question is not one of knowledge at all. Both commit the same error: they treat it as a piece of knowledge. It belongs to an entirely different category.
Science has nothing to do with it. Science can never prove God's existence, and it can never disprove it. If science ever begins invoking God as an explanation, that is the end of science — not because Kant has anything against God, but because you stop asking good questions and stop getting substantive explanations. Everything collapses into: why does rain fall? Because God wills it. Why is there lightning? Because God wills it. And all genuinely informative explanation comes to a halt.
Kant argues that we have these ideas of reason built into our minds — we cannot get rid of them. The nature of space, time, and causality is to push beyond themselves, to drive us toward certain conclusions. But there are two different ways we can treat this. He introduces what he calls a transcendental illusion. Bertrand Russell, a later philosopher, used a helpful analogy here: imagine you are wearing pink sunglasses and you look at a white wall. The wall would appear pink — not because the wall is pink, but because you are looking at it through those lenses. The illusion is that the pinkness seems to belong to the wall rather than to the glasses.
The transcendental illusion works the same way. We take qualities that belong to our own minds, to our own faculties, and mistake them for qualities of the world. This is what every philosopher before Kant had done, in his view. When we fall into this illusion with respect to causality, we assume there is a first cause out there waiting to be found, and so much of metaphysics has been devoted to proving the existence of God and determining his nature. Kant considers this wasted effort — not because he is anti-religious, but because he wants religion to remain in the domain of religion rather than science.
The transcendental illusion uses these ideas in what Kant calls a constitutive way. That means treating them as if they refer to something actually out there in the world — as if the wall really were pink. Because reason demands that causality come to a conclusion, we treat the first cause as a real thing we could locate and search for. This leads nowhere. Anything we come up with will be mistaken, a fundamental misunderstanding of how the mind works.
But we cannot simply stop thinking this way, because these ideas are hardwired into our minds. What we can do is look at them in an entirely different way. Instead of constitutive, we can treat them as regulative. Regulative means we know the first cause is not literally out there to be found — we know the wall is not pink. We know we can never actually reach the outer boundary of space, the first moment of time, or the uncaused first cause. And yet we can keep looking.
This sounds strange, because ordinarily when we search for something, we expect to find it. But Kant's point is that we can keep searching for something while knowing we can never reach it — and that this is precisely what drives science forward. Cosmologists have been pushing closer and closer to the Big Bang, reaching unimaginably tiny fractions of a second after it, without ever arriving at it. But by getting closer and closer, they learn. The search is not futile; it is generative.
Because of the way time works, there will always be more information to uncover, finer and finer intervals to examine, and each one will be informative. If we simply gave up on the grounds that we could never reach the beginning, we would learn nothing. If we falsely convinced ourselves that we had reached it, we would either congratulate ourselves prematurely or collapse into frustration. Instead, we say: we know we will never get there, but we will not stop trying. And so we get closer and closer, learning more and more about the universe, precisely because we are striving toward a goal we know is unreachable.
This is what gives science its motor. Treating these ideas as regulative — as guides that orient our search rather than objects we can actually find — is what makes the pursuit productive. And this is how metaphysics is, in Kant's view, both impossible and necessary at the same time.
Kant brings the early modern period to a close by reconciling empiricism and rationalism, and in doing so he opens up the next period of German idealism in the nineteenth century. On one hand, he agrees with Descartes. Descartes observed that when you look at a candle and a star, they appear to be the same brightness — your eyeball, if it could speak, would say they are the same fire, the same intensity. But your brain overrules it, recognizing that the star is incredibly bright and simply very far away, while the candle is much dimmer and much closer. For Descartes, even when we are using our senses, we are using our reason. That is the rationalist view.
Kant accepts this insight: even when we are looking at the world, the mind is adding a priori components — the forms of intuition and the concepts of the understanding. We are not simply taking experience in passively, as the empiricists claim. But Kant then turns to Hume. Hume argues that even when we reason, we are relying on experience, because the only way anyone could know that a second billiard ball would move upon impact is by having seen it happen over and over again. For Descartes, even experience depends on reason; for Hume, even reason depends on experience.
Kant's response is that both are right — but both went too far. Each philosopher possesses a genuine nugget of wisdom, and Kant reaches into their respective systems, extracts it, and brushes away the extremes and misunderstandings. Once freed from those excesses, the two pieces fit together perfectly. We do need experience in order to know, but knowledge is not only experience. We can have knowledge that exceeds any single experience — we can know the future, for instance — yet even that knowledge remains tethered to experience.
The result is that we can have the natural science Hume thought was impossible, but we cannot have the metaphysics Descartes thought was within our reach — not in the same way, and not on the same terms. Asking those metaphysical questions is still enormously valuable, even if definitive answers lie beyond our grasp.
Kant was the great thinker of the Enlightenment. "The Enlightenment," he said, "was the time when humanity grows up." It is the moment we mature, the moment we take our seat at the adult table. And what that means, ultimately, is that we think for ourselves.
Descartes said he would not rely on the education he had received, that he would start from scratch and think everything through on his own — but then he had to rely on God. Hume said he would work through everything without relying on God, but Kant thinks he ended up with very little. Kant's conclusion is that we must rely on ourselves alone: not on God, and not on nature in the way Hume supposed.
We do not have to depend simply on what the world shows us. The order and regularity we find in the world are not discovered there — they are imposed by us. We are autonomous. We are not finding these connections; we are forging them. We are in charge of our epistemological fate.
We are the ones lighting the torch that illuminates the darkness and yields the knowledge we sought. That is Kant's great contribution: the recognition that we are in control of our own understanding of the world.