In a famous passage toward the end of the Critique of Practical Reason, Kant says that two things fill the mind with ever increasing wonder: the starry heavens above and the moral law within. Through Copernicus, Galileo, Newton and others, incredible progress had been made about understanding the "starry heavens above." But skeptical philosophers like David Hume maintained that the scientific success could not be extended to other areas such as theology or morality.
Hume said that what we see in the natural world might have all kinds of causes. Inferring a Creator, let alone a particular kind of Creator, goes beyond the evidence. In fact, the very concept of causality is merely a habit of the mind that we form by constantly seeing two things happening together. If we did not repeatedly observe that a billiard ball moved when hit by another billiard ball, we would not bring the two events in relation to each other. And we certainly would not be able to infer from the movement of the one ball that another ball had struck it, unless we have constantly observed the two together.
Likewise, we cannot infer anything about supposed metaphysical causes by looking at the physical world, because causal relations only form in our minds through repeated experience. But no one in his right mind claims, for instance, that he constantly observes God creating the world. God as a cause is inferred from the effect, but that is precisely what, according to Hume, we cannot do.
These and other thoughts by Hume awoke Kant, as he himself said, out of his "dogmatic slumber." Essentially, Hume raised in him the following questions:
Given what Hume said about causality, how can I affirm the validity of Newton’s physics, which is built on the concept of causality? What makes it possible for natural science and mathematics to give us apparently reliable knowledge of the world?
Why is it so difficult to make similar progress in metaphysical questions about God, the soul, free will, and morality? Is knowledge about these things possible at all? And if so, in what way and to what degree? Clearly, Reason itself compels us to ask these questions, but then Reason seems to be unable to answer them. Why is that?
In Kant’s long and laborious answer, laid out in his Critique of Pure Reason, he essentially harmonizes Hume with Newton. Science and mathematics are so successful, he says, because they only describe reality as it appears to us, not as it is in itself. Science only studies the world that we can touch, feel, see, hear, and smell—the experiences of our senses. Even with the aid of technology such as a telescope or, nowadays, a particle accelerator, we take in all scientific data through our senses. But the way that we experience our senses—the way our mind puts together what we touch, feel, see, hear, and smell—depends on certain forms in which our mind is structured. We cannot help thinking in terms of space and time, cause and effect, quantity and quality, modality and relation. These ways of thinking are what Kant calls a priori, that is, they come before experience. In fact, all coherent experience is dependent on them.
The problem of dogmatic metaphysics is that it pretends we can free ourselves from these necessary preconditions of our experience and get direct, unfiltered knowledge of ultimate reality. Not so, says Kant. Even space and time are not a reality that we perceive without a human filter. On the contrary, it is impossible for us to think of anything without picturing it in time and space. Space and time are, so to speak, mental cookie cutters. They are forms of the human mind that the dough of our experience has to conform to in order to experience anything at all. In Kant’s terminology, they are "forms of intuition" or "forms of sensibility."
Does that mean, then, that space and time are completely subjective and do not exist at all except in our minds? Do we live in some kind of illusory world, like in the Matrix? No, that would be to misunderstand Kant. Space and time are real for Kant. Objects do not merely seem to exist in time and space; they are in time and space. But since space and time are necessary concepts for any experience whatsoever, the concepts only apply to things as they appear to us. In what way and to what degree space and time apply to reality apart from our experience of reality, is unknowable to us. All we can say is they are necessary concepts for experiencing reality. Therefore, while all objects of our experience are in time, not all of reality must necessarily be in time as we understand it. While all objects of our experiences are in space, not all of reality must necessarily be in space as we understand it. On an ultimate level, reality might be quite different from the space-time framework from which we cannot free ourselves.
That is why, Kant thinks, mathematics and geometry have truths that can be definitely proven. They are about the necessary forms of all our experience, which is why they are internally coherent. Geometry takes place in space, and mathematics is done in time. Thus, they must conform to our a priori intuitions of space and time.
This makes arithmetic and geometry "synthetic a priori truths." "Synthetic" means that they are truths not merely about the meaning of words, which would be "analytic" truths, but about the real world. At the same time, however, these truths are a priori in that they are not based on our experience of the real world.
Kant’s conception of time and space as they relate to arithmetic and geometry is a good example of why he called his philosophy a new "Copernican Revolution." Just like Copernicus had reverted our understanding of the universe by proposing that it is us who move, even though it looks as if the sun moved, so Kant tried to show that it is really the structure of our minds that "moves" in a certain way and lets reality appear to us according to these "movements." Empiricists such as John Locke had thought that objective reality impressed itself on the blank slate of our minds. Kant proposed that it was the other way round: Our minds impressed themselves on objective reality. All our experiences of the outer world must necessarily conform to the a priori forms of our intuition and fundamental categories of our understanding. If Locke and Hume had been right, all our ideas about the world would arise as a result of our experience. They would come "after" experience, a posteriori. In contrast, Kant thought that a priori knowledge—knowledge prior to experience—was possible, namely the knowledge of the categories of our understanding to which our experiences conform.
Kant divided these categories as follows:
Quantity: unity, plurality, and totality.
Quality: reality, negation, and limitation.
Modality: possibility, existence, and necessity.
Relation: inherence, causality, community, and correlation.
While there was nothing particularly new about this list as such, as Aristotle and the Scholastics had already made similar categorizations, the revolutionary move on Kant’s part was to say that they are a priori: that they precede experience and are necessary conditions for it. In the longest and probably most difficult part of the Critique of Pure Reason, called the "Transcendental Deduction of the Categories," Kant tries to prove that these categories are in fact a priori.
Having, in Kant’s mind, provided this proof, he is now ready to answer his initial questions more explicitly. (1) What makes it possible for natural science and mathematics to give us apparently reliable knowledge of the world? (2) And why is it so difficult to make similar progress in metaphysical questions about God, the soul, free will, and morality? The answer is that science, arithmetic, and geometry stay within the framework of what is possible for us to experience—a framework that is set by the forms of intuition and the categories of the understanding. This is why science is so effective, but it also means that science is limited. It shows us only the “phenomenal” world, not the “noumenal” world. That is, it only shows as the world as it appears to us, not as it is in itself. The world as it is in itself is the realm of metaphysics, and this is why we cannot give scientific answers to metaphysical questions.
The last third of the Critique of Pure Reason attempts to show, as Robert Kane has put it, “how we get into trouble when we try to press the human mind beyond possible experience into the realms of metaphysics.” Take the existence of God, for instance. Trying to prove his existence has failed, Kant says, because such proofs take the categories of the understanding that only apply to the physical world of appearances and impose them on matters that exceed the bounds of our possible experience. We cannot infer what caused the universe to come into existence based on our way of thinking, because our way of thinking is only fitted for inferences about the physical world of appearances.
Or take the soul. We cannot, Kant stresses, know the soul as such, only the self as it appears to us—our “phenomenal” self, not our “noumenal” self. We do not and cannot know what the soul is in itself, apart from the way our sense of self appears to us. Neither can we say anything of scientific certainty about the mortality or immortality of the soul. We may have faith and hope in eternal life based on what Kant calls “practical” Reason, but not scientific certainty.
Nor can we scientifically resolve the conflict between determinism and free will. Science is based on the view that everything works by determined laws, but our understanding of morality is based on the view that we have free will. Without the concept of free will, none of our moral practices would make any sense: no commands or punishments, no blame or excuses, no forgiveness or justice. But how can we reconcile the determined world of science with the free world of morality? Science itself cannot resolve this contradiction, says Kant. It cannot be resolved by “theoretical” Reason, only by “practical” Reason.
And it is this practical Reason that Kant tackles in his second Critique.