Kant's philosophy has been called a synthesis of rationalism and empiricism. From rationalism he takes the idea that we can have a priori knowledge of significant truths, but rejects the idea that we can have a priori metaphysical knowledge about the nature of things in themselves, God, or the soul. From empiricism he takes the idea that knowledge is essentially knowledge of experience, but rejects the idea that we cannot learn any necessary truths about experience, and in doing so he rejects Hume's skepticism.
He is able to create this synthesis largely thanks to a radical reconception of the nature of knowledge that comes from from experience. Though empiricists and rationalists may have disagreed about the value or certainty of knowledge from experience, they both generally thought of the mind as a neutral receptor: knowledge from experience was simply the report of the senses. Kant points out that our knowledge of experience extends far beyond what the senses can report. Our senses can report sensations, but they cannot give these sensations a structure in space and time, or organize them according to cause and effect. According to Kant, our faculties or sensibility and understanding are largely responsible for what we think of as "knowledge from experience."
By giving our mind this complex internal structure, Kant makes room for a great deal of a priori knowledge. Though the sensations that are the basis for all our experience come from things in themselves, any regularity or structure we find in these sensations comes from our mental concepts. Thus, while Kant does not slip into the idealist position of saying that reality is all a matter of perception, he does claim that the laws of nature are the laws of our mental faculties. For something to be an objective law, it must be synthetic and it must be a priori, and Kant identifies the possibility of synthetic a priori knowledge within the structure of our mental faculties.
If our sense of order and regularity is not something we find in experience, but something we impose upon experience, the study of this order and regularity is a study of our own faculties rather than a study of experience. Kant reconceives the purpose of metaphysics as being one of critique: we must seek to understand how knowledge is structured, and consequently how the various concepts of our mental faculties are organized. This is an important step for philosophy: after Kant published his work, there was been less interest in making extravagant claims about the nature of the universe, and greater emphasis on determining what we can know and on what grounds we can claim to know it.
This kind of scrutiny, which Kant advocates, has led to some serious attacks on his work. The German Idealists—notably Hegel—were the first to call into question Kant's concept of the thing in itself. Kant insists that while all we can perceive are appearances, these appearances are caused by things in themselves that are outside the realm of experience. Because they are outside the realm of experience, they are also outside the realm of space and time and any of the regularities we perceive in nature. A number of questions can be raised about what kind of relation things in themselves can have with appearances if categories such as time, causation, and even existence do not apply to them. The response of the German Idealists was to abandon the concept of things in themselves and assert that only appearances exist.
Analytic philosophy also got its start by criticizing Kant. This movement criticized in particular his category of the synthetic a priori. Frege was the first to point out that geometry is not synthetic a priori. Pure geometry—which consists only of deductive inferences—is analytic, and empirical geometry—which deals with what space is like in the real world—is known a posteriori. This position was given a boost by Einstein's relativity, which shows that space is very different from what we had assumed and our understanding of it is certainly not a priori.
Frege also complains that Kant's definition of analytic and synthetic judgments rests on the subject-predicate form of grammar, which is not a necessary part of the logical structure of language. Efforts to define and classify analytic and synthetic judgments have been a major pre-occupation of analytic philosophy, especially in the first half of the twentieth century.
Though many of Kant's doctrines have fallen into question, his exhortation toward critical philosophy remains with us. Perhaps his most lasting contribution is the setting up of a new standard for rigor and circumspection in philosophical investigations.