So Kant maintained that we are justified in applying the concepts of the understanding to the world as we know it by making a priori determinations of the nature of any possible experience. In order to see how this works in greater detail, let’s concentrate on the concepts of relation, which govern how we understand the world in time. As applied in the Analogies of Experience, each concept of relation establishes one of the preconditions of experience under one of the modes of time: duration, succession, and simultaneity.
1. Substance: The experience of any change requires not only the perception of the altered qualities that constitute the change but also the concept of an underlying substance which persists through this alteration. (E.g., in order to know by experience that the classroom wall has changed in color from blue to yellow, I must not only perceive the different colors—blue then, yellow now—but also suppose that the wall itself has endured from then until now.) Thus, Kant supposed that the philosophical concept of substance (reflected in the scientific assumption of an external world of material objects) is an a priori condition for our experience.
2. Cause: What is more, the experience of events requires not only awareness of their intrinsic features but also that they be regarded as occurring one after another, in an invariable regularity determined by the concept of causality. (E.g., in order to experience the flowering of this azalea as an event, I must not only perceive the blossoms as they now appear but must also regard them as merely the present consequence of a succession of prior organic developments.) Thus, Kant responded to Hume’s skepticism by maintaining that the concept of cause is one of the synthetic conditions we determine for ourselves prior to all experience.
3. Community: Finally, the experience of a world of coexisting things requires not only the experiences of each individually but also the presumption of their mutual interaction. (E.g., in order believe that the Sun, Earth, and Moon coexist in a common solar system, I must not only make some estimate of the mass of each but must also take into account the reciprocity of the gravitational forces between them.) Thus, on Kant’s view, the notion of the natural world as a closed system of reciprocal forces is another a priori condition for the intelligibility of experience.
Notice again that these features of nature are not generalized from anything we have already experienced; they are regulative principles that we impose in advance on everything we can experience. We are justified in doing so, Kant believed, because only the pure concepts of the understanding can provide the required connections to establish synthetic a priori judgments. Unless these concepts are systematically applied to the sensory manifold, the unity of apperception cannot be achieved, and no experience can be made intelligible.
Having seen Kant‘s transcendental deduction of the categories as pure concepts of the understanding applicable a priori to every possible experience, we might naturally wish to ask the further question whether these regulative principles are really true. Are there substances? Does every event have a cause? Do all things interact? Given that we must suppose them in order to have any experience, do they obtain in the world itself? To these further questions, Kant firmly refused to offer any answer.
According to Kant, it is vital always to distinguish between the distinct realms of phenomena and noumena. Phenomena are the appearances, which constitute the our experience; noumena are the (presumed) things themselves, which constitute reality. All of our synthetic a priori judgments apply only to the phenomenal realm, not the noumenal. (It is only at this level, with respect to what we can experience, that we are justified in imposing the structure of our concepts onto the objects of our knowledge.) Since the thing in itself (Ding an sich) would by definition be entirely independent of our experience of it, we are utterly ignorant of the noumenal realm.
Thus, on Kant’s view, the most fundamental laws of nature, like the truths of mathematics, are knowable precisely because they make no effort to describe the world as it really is but rather prescribe the structure of the world as we experience it. By applying the pure forms of sensible intuition and the pure concepts of the understanding, we achieve a systematic view of the phenomenal realm but learn nothing of the noumenal realm. Math and science are certainly true of the phenomena; only metaphysics claims to instruct us about the noumena.
Although our knowledge of mathematics and natural science yield easily to a Kantian analysis, the synthetic a priori judgments of metaphysics are much more difficult to explain. Here the forms of intuition and concepts of understanding are useless, since they find application only in the realm of our experience, while metaphysics seeks to transcend experience completely, in order to discover the nature of reality itself as comprehended under pure reason.
Metaphysical speculation properly begins with the same method as the “Aesthetic” and “Analytic,” Kant supposed, but it invariably ends up in a “Dialectic.” The transcendental arguments we employ in metaphysics need not restrict their determination to the phenomenal realm alone, since their aim is genuine knowledge of the noumena. Synthetic a priori judgments in metaphysics must be grounded upon truly transcendental ideas, which are regarded as applicable to things in themselves independently of our experience of them.
Kant‘s exposition of the transcendental ideas begins once again from the logical distinction among categorical, hypothetical, and disjunctive syllogisms. From this distinction, as we have seen, the understanding derives the concepts of substance, cause, and community, which provide the basis for rules that obtain as natural laws within our experience. Now, from the same distinction, the reason must carry things further in order derive the transcendental ideas of the complete subject, the complete series of conditions, and the complete complex of what is possible. Thus, the “completion” of metaphysical reasoning requires transcendental ideas of three sorts, but Kant argued that each leads to its characteristic irresolvable difficulty.
The Psychological Idea is the concept of the soul as a permanent substance which lives forever. It is entirely natural to reason (as in Descartes’s cogito) from knowledge that “I think” to my real existence as one and the same thinking thing through all time, but Kant held that our efforts to reach such conclusions are “Paralogisms,” with only illusory validity. It is true that thought presupposes the unity of apperception and that every change presupposes an underlying substance, but these rules apply only to the phenomena we experience. Since substantial unity and immortality are supposed to be noumenal features of the soul as a thing in itself, Kant held, legitimate a priorijudgments can never prove them, and the effort to transcend in this case fails.
The Cosmological Idea is the concept of a complete determination of the nature of the world as it must be constituted in itself. In this case, Kant held, the difficulty is not that we can conclude too little but rather that we can prove too much. From the structure of our experience of the world, it is easy to deduce contradictory particular claims about reality: finitude vs. infinity; simplicity vs. complexity; freedom vs. determinism; necessity vs. contingency. These “Antinomies” of Pure Reason can be avoided only when we recognize that one or both of the contradictory proofs in each antinomy holds only for the phenomenal realm. Once again, it is the effort to achieve transcendental knowledge of noumena that necessarily fails.
The Theological Idea is the concept of an absolutely perfect and most real being (or god). Again it is natural to move from our recognition of dependence within the phenomenal realm to the notion of a perfectly independent noumenal being, the “Transcendental Ideal.” But traditional attempts to prove that god really exists, founded as they are on what we experience, cannot establish the reality of a being necessarily beyond all experience.
The general point of the Transcendental Dialectic should by now be clear: metaphysical speculation about the ultimate nature of reality invariably fails. The synthetic a priori judgments which properly serve as regulative principles governing our experience can never be shown to have any force as constitutive of the real nature of the world. Pure reason inevitably reaches for what it cannot grasp.
Now that we’ve seen Kant‘s answers to all three parts of the Prolegomena‘s “Main Transcendental Question” and have traced their sources in the Critique of Pure Reason, we are in a position to appreciate his careful delineation of what is possible in metaphysical thought and what is not.
What most clearly is not possible is any legitimate synthetic a priori judgment about things in themselves. The only thing that justifies the application of regulative principles in mathematics and natural science is their limitation to phenomena. Both sensible intuition and the understanding deal with the conditions under which experience is possible. But the whole point of speculative metaphysics is to transcend experience entirely in order to achieve knowledge of the noumenal realm. Here, only the faculty of reason is relevant, but its most crucial speculative conclusions, its deepest convictions about the self, the world, and god, are all drawn illegitimately.
What is possible—indeed, according to Kant what we are bound by our very nature as rational beings to do—is to think of the noumenal realm as if the speculative principles were true (whether or not they are). By the nature of reason itself, we are required to suppose our own existence as substantial beings, the possibility of our free action in a world of causal regularity, and the existence of god. The absence of any formal justification for these notions makes it impossible for us to claim that we know them to be true, but it can in no way diminish the depth fo our belief that they are.
According to Kant, then, the rational human faculties lead us to the very boundaries of what can be known, by clarifying the conditions under which experience of the world as we know it is possible. But beyond those boundaries our faculties are useless. The shape of the boundary itself, as evidenced in the Paralogisms and Antinomies, naturally impels us to postulate that the unknown does indeed have certain features, but these further speculations are inherently unjustifiable.
The only legitimate, “scientific” metaphysics that the future may hold, Kant therefore held, would be a thoroughly critical, non-speculative examination of the bounds of pure reason, a careful description of what we can know accompanied by a clear recognition that our transcendental concepts (however useful they may seem) are entirely unreliable as guides to the nature of reality. It is this task, of course, that Kant himself had pursued in the First Critique.