In what follows, I would like to bring out some features of Kant’s conception of transcendental philosophy in the Critique of Pure Reason. One of the aims of this work was to answer the question raised in the famous letter to Markus Herz of February 21, 1772, about the possibility of a priori cognition’s relating to its object.1 And as we know, the answer elaborated in the Critique restricts the use of this pure cognition to possible experience. Cognition a priori is valid only when applied to appearances, within experience. It leads therefore to a finite knowledge, limited to the phenomenal world.
Now the question is: What is the legitimacy of Kant’s own transcendental discourse when we take account of the restrictions imposed on human knowledge as a result of this very investigation? In other words, the use of a priori cognition being restricted to possible experience, how could the cognitive claim of this philosophical inquiry be justified if it does not itself take place within possible experience? The problem already becomes obvious with the formulation of the main result of the critical investigation, namely that the only valid sphere for a priori cognition is “possible experience.” Yet how are we to understand the use of the word “possible” here? To be sure, possibility is a modal category whose conditions of legitimate application are strictly determined and restricted in the Transcendental Analytic. And the same goes for the category of contingency, which, as we will see, plays an important role in specifying the status of the conditions of the possibility of experience. But then again the question is: Does Kant make an inappropriate or illegitimate use of these categories when, rather than applying them to objects of a possible experience, he uses them to describe the scope of valid a priori knowledge as a whole? By apparently removing the restrictions on their use here, it seems as though Kant is making a transcendent or, as he would say, “transcendental,” employment of the categories of modality.
I will argue, however, that this use of the categories complies, all things considered, with the constraints imposed upon them for their application in experience. We will find that the categories at work in Kant’s transcendental discourse are not employed without caution in their purely intellectual significance, which would lead them to open unto the unconditioned, as in the Transcendental Dialectic. On the contrary, the standpoint from which critical philosophy condemns every attempt to gain knowledge of the unconditioned is not itself unconditioned. I would like to show that transcendental philosophy, even when it circumscribes the limits of human knowledge with the help of the modal categories of possibility and contingency, nevertheless acknowledges its own finitude and refrains from overstepping the “bounds of sense.” In order to achieve this, we will have to attend clearly to and carefully define Kantian expressions found in the first Critique, such as “possible experience,” “transcendental cognition,” “conditions of the possibility,” and “contingency,” which in the end will lead us to establish the modal status of the ultimate presuppositions of critical philosophy.
Possible Experience
Before we inquire into the role played by possible experience in the transcendental philosophy developed in the first Critique, we would be well advised to focus on the meaning of each of these words. “Experience” is a well-known Kantian term designating the cognition of objects that are “given in empirical intuition” or, if one prefers, the “cognition of the objects through perception.” The short definition is thus “empirical cognition,”2 whereby it is understood that this cognition lays claim to objectivity. Experience is in fact objective empirical knowledge.
“Possibility” is a modal category which, from a strictly logical point of view, refers to that which does not contradict itself. To say that a concept is “possible” simply means that its internal components are not in contradiction to each other. This is what Baumgarten in his Metaphysica calls “absolute” possibility3—a characterization for which he is criticized in Kant’s lectures on metaphysics as well as (though implicitly) in the Transcendental Dialectic.4 In fact, Baumgarten uses the adjective “absolute” to designate “intrinsic” possibility, which according to Kant is the “least” that can be said of the concept of an object. But the strong meaning of the word should be maintained. In Kant’s view, absolute possibility properly signifies the “most” that can be said of a concept, namely that its object is possible literally “in all respects” and “without any restriction” whatsoever.5
Kant feels the need to restore the strong sense of the term “absolute” at the beginning of the Transcendental Dialectic because it is synonymous with “unconditioned,” which is the main topic of this second part of the Critique devoted to the logic of illusion. Indeed, a possibility that is absolute in the full sense of the word pertains not to the understanding but to reason,6 and this category gives rise to the same problem as the modal concept of “unconditioned necessity” used by the dogmatic metaphysician in the cosmological proof of the existence of God. That is, both concepts exceed the grasp of the human mind: “the unconditioned necessity, which we need so indispensably as the ultimate sustainer of all things, is for human reason the true abyss.”7 The treatment reserved for the modal categories in the Transcendental Analytic reveals that only “hypothetical” necessity, that is, conditioned necessity, is available to human cognition, and the same goes for possibility, as we learn in Reflexion 4005: “with reason we can cognize only conditioned possibility.”8 The only kind of possibility that can be grasped by a finite understanding is relative possibility, namely that which is possible only in some respects. Accordingly, possibility is cognizable only if it is “restricted to conditions.”9 Let us take for instance Kant’s example of the “invented concepts” of substances and forces supposedly present in experience. It is not enough to say that such concepts are possible because they are not self-contradictory. To be sure, this satisfies the minimal requirement of their logical possibility, but their real possibility also has to be established within experience by showing that these objects can be instantiated according to the known laws of experience.10 This is a clear example of conditioned possibility.
Now when Kant considers possible experience as a whole, rather than just particular objects within it, he maintains a similar restriction on his recourse to the concept of possibility. Experience might well be declared “possible,” but here again only as a conditioned possibility. Hence the central expression for the Critique of Pure Reason: “conditions of the possibility of experience.”11 Experience taken globally will thus be possible solely under a set of conditions. As we can see, Kant remains coherent in his use of the category of possibility: experience’s possibility also depends on conditions, and most importantly on transcendental conditions. We will have to inquire into the nature of this transcendental conditioning, but before going any further we must again focus our attention on Kant’s terminology.
Transcendental Cognition
Let us quote the canonical definition of the term “transcendental” given in the Introduction to the second edition of the Critique: “I call all cognition transcendental that is occupied not so much with objects but rather with our mode of cognition of objects insofar as this is to be possible a priori.”12 Here we clearly have two levels of cognition: a first level concerned with the cognition of objects and a second level dealing with our mode of cognition a priori of these objects, which is precisely what transcendental cognition is about. This means that in the Transcendental Analytic, Kant is concerned primarily with a priori cognition, or better: with our a priori mode of cognition of objects.
However, all a priori cognition is subject to a constraint: it can never reach the object in its actuality. For this, according to the teachings of the Transcendental Analytic, the empirical dimension of the object must be added, at which point the cognition becomes a posteriori. As we can read in the Metaphysical Foundations of Natural Science, “to cognize something a priori means to know it in its mere possibility.”13 Transcendental cognition is no exception here, and so it is no surprise that it is oriented exclusively toward “possible” experience.14 The a priori cognition introduced in the Transcendental Analytic is aimed at grounding the mere possibility of experience, and for this purpose it has to “abstract from everything empirical in the appearances.”15
Furthermore, it must be noted that cognition of the a priori conditions of experience is not just any kind of a priori cognition. It is, as I have already said, a second-level cognition. Its distinctive nature will become clear if we compare it with the first-level kind of a priori cognition such as pure geometry. While geometry is focused exclusively on its objects, which are merely ideal, the a priori conditions of experience, on the other hand, concern our mode of cognition of real objects insofar as it presents their a priori components. It is nonetheless a form of “cognition,” with its own truth claim—not empirical, to be sure, but transcendental. And since, for Kant, truth means adaequatio, that is, the correspondence of cognition with its correlate,16 transcendental truth also involves such a correlate: possible experience.
In order to establish this correlation we can mention the two passages in the Analytic dealing with this specific truth claim. The first one appears, as we know, in the Schematism chapter: “transcendental truth, which precedes all empirical truth and makes it possible, consists in the general relation to this [the entirety of all possible experience].”17 The schemata are those products of the imagination that allow the pure concepts of the understanding to connect with possible experience. These a priori concepts thereby acquire their truth, that is, their objective validity, since from then on they have a correlate that can ground their claim to transcendental truth. The second passage is to be found in the Postulates of Empirical Thinking, to which we will return later on. It concerns the categories of relation, which gain their objective validity by their mere reference to experience in general: “one cognize[s] their objective reality [of the categories of relation], i.e., their transcendental truth, and, to be sure, independently of experience, but yet not independently of all relation to the form of an exper...