The Transcendental Deduction of the Categories

The Problem with A Priori Concepts

Kant’s mature philosophical system is notable for the strictures it sets on the kind of cognition and knowledge human beings (or finite rational beings more generally) can achieve through reason alone. Importantly, Kant thinks that pure reason can achieve relatively little on its own. All of our ampliative knowledge that is also necessary and universal consists in what Kant calls ‘synthetic a priori’ judgments or propositions. The central question he then pursues concerns how knowledge of such synthetic a priori propositions is possible. We previously discussed Kant’s arguments concerning the role (the representations of) space and time play in grounding necessary and universal mathematical cognition and knowledge. On Kant’s view, mathematical cognition is synthetic because it goes beyond mere conceptual analysis to deal with the structure of (our representation of) space itself. It is a priori because the structure of (our representation of) space is a priori accessible to us, being merely the form of our intuition and not a real mind independent thing (i.e. a ‘thing in itself’).

However, recall from our discussion of transcendental idealism that, in addition to the representation of space and time, Kant also thinks that possession of a particular privileged set of a priori concepts is necessary for knowledge of the empirical world. But this raises a problem. How could an a priori concept, which is not itself derived from any experience, be nevertheless legitimately applicable to objects of experience? To make things even more difficult, it is not the mere possibility of the application of a priori concepts to objects of experience that worries Kant, for this could just be a matter of pure luck. Kant requires more than the mere possibility of application, for he wants to show that with regard to a privileged set of a priori concepts, they apply necessarily and universally to all objects of experience and do so in a way that we are in a position to know a priori.

Kant’s strategy for demonstrating how this is possible hinges on showing that the experience of objects that he thinks all would agree that we have nevertheless depends on the application of a priori concepts—the categories. Kant makes this clear in his elaboration of his ‘Copernican Turn’ in philosophy:

because I cannot stop with these intuitions, if they are to become cognitions, but must refer them as representations to something as their object and determine this object through them, I can assume either that the concepts through which I bring about this determination also conform to the objects, and then I am once again in the same difficulty about how I could know anything about them a priori, or else I assume that the objects, or what is the same thing, the experience in which alone they can be cognized (as given objects) conforms to those concepts, in which case I immediately see an easier way out of the difficulty, since experience itself is a kind of cognition requiring the understanding, whose rule I have to presuppose in myself before any object is given to me, hence a priori, which rule is expressed in concepts a priori, to which all objects of experience must therefore necessarily conform, and with which they must agree. (CPR Preface, Bxvii)

Kant’s strategy is thus to show how a priori concepts legitimately apply to their objects in virtue of being partly constitutive of the objects of experience, rather than the traditional view in which the objects of experience are the ground of our concepts. Now, what exactly this means is deeply contested, at least partly because it is rather unclear what Kant intends us to understand by his Transcendental Idealism. For example, does Kant intend that the objects of experience are themselves nothing other than representations? This would be a form of phenomenalism similar to that offered by Berkeley. Kant, however, seems to want to deny that his view is similar to Berkeley’s, asserting instead that the objects of experience really exist independently of the mind, and that it is only the way that they are experienced that is mind-dependent.

the fact that I have myself given to this theory of mine the name of transcendental idealism cannot justify anyone in confusing it with the empirical idealism of Descartes (although this idealism was only a problem, whose insolubility left everyone free, in Descartes‘ opinion, to deny the existence of the corporeal world, since the problem could never be answered satisfactorily) or with the mystical and visionary idealism of Berkeley (against which, along with other similar fantasies, our Critique, on the contrary, contains the proper antidote). For what I called idealism did not concern the existence of things (the doubting of which, however, properly constitutes idealism according to the received meaning), for it never came into my mind to doubt that, but only the sensory representation of things, to which space and time above all belong; and about these last, hence in general about all appearances, I have only shown: that they are not things (but mere ways of representing), nor are they determinations that belong to things in themselves. (Prolegomena §13, Note III; 4:293)

I’m mostly going to set this issue to the side in what follows. But the issue of idealism is important to consider, insofar as Kant takes himself to be replying to Hume’s arguments about the nature of the empirical world and the nature and extent of our knowledge of it.

What Is a Transcendental Deduction?

In order to prove that there are a priori concepts which legitimately apply to the objects of experience, Kant articulates a special sort of argument, which he calls a ‘transcendental deduction’. As Henrich (1989) points out, the notion of a ‘deduction’ that Kant uses, is a legal one intended to provide a historical justification for the legitimacy of a property claim. In Kant’s case however, it is transformed into a justification of the applicability (in the sense discussed above) of the a priori concepts Kant calls the ‘categories’. These are the concepts given in Kant’s ‘Table of Categories’ (A80/B106); they are Unity, Plurality, and Totality (the Categories of Quantity); Reality, Negation, and Limitation (the Categories of Quality); Inherence and Subsistence, Causality and Dependence, and Community (the Categories of Relation), and Possibility-Impossibility, Existence-Nonexistence, Necessity-Contingency (the Categories of Modality).

With regard to each category, Kant’s aim is to show that it has ‘objective validity’—i.e. legitimate applicability to the objects of experience. Disputes concerning various members of the table should be familiar. Hume famously disputes the legitimacy of our concept cause. Hume thinks (in Kant’s terms) that no ‘empirical deduction’ of the concept is possible—i.e. that the concept cannot be traced to the occurrence of a corresponding impression. Kant concurs with Hume on this point—that there can be no empirical deduction—but argues that this doesn’t show that the concept is illegitimate, for it may be both a priori and legitimately applied in virtue of being a necessary condition of experience.

For Kant then a ‘transcendental deduction’ starts from a premise concerning some feature of human experience, a premise which reasonable interlocutors might be expected to endorse, and then argues to a substantive philosophical conclusion concerning the presuppositions or necessary conditions of the truth of that premise. Since Kant’s concern here is the a priori categories, his aim is to show that a presupposition or necessary condition of some relatively uncontroversial feature of experience is the applicability (or successful application) of the a priori concept(s) in question to the objects of experience. Below we’ll concern ourselves with Kant’s answer to Hume regarding the objective legitimacy of the concept cause.

Experience & Cognition

Kant is concerned with how we might explain and justify the application of a priori concepts to objects of experience. But this question, as it stands, requires still sharper focus. We’ve seen what Kant means in his question concerning the legitimacy of concepts in their application to the objects of experience. But what does Kant mean by “experience“ [Erfahrung]? Our problem is that the notion of “experience,” both in German and in English, is ambiguous and can mean any number of a variety of things—anything from the mere occurrence of sensation all the way to empirical judgment.1

For Kant, “experience” is a technical term that is closely related to empirical cognition [empirische Erkenntniss]. He makes this clear at several points, both in the Deduction and elsewhere in the first Critique. However, in what exactly the relation between the two notions consists is somewhat problematic. For example, Kant says that

Such cognitions [i.e. those that are independent of any sense impression] are called a priori cognitions; they are distinguished from empirical cognitions, whose sources are a posteriori, namely, in experience [Erfahrung] (Introduction, B2; my emphasis).

Here we see Kant saying that the source of empirical cognition is experience. However, Kant also quite explicitly identifies empirical cognition with experience (or treats the terms synonymously).

Empirical cognition, however, is experience (B166).

Experience is an empirical cognition, i.e., a cognition that determines an object through perceptions (A176/B218).

Therefore experience itself—i.e., empirical cognition of appearances—is possible only inasmuch as we subject the succession of appearances, and hence all change, to the law of causality (A189/B234).

it [the presentation “I am”] is not yet a cognition of that subject, and hence is also no empirical cognition – i.e., experience – of it (B277).

These passages all identify experience with empirical cognition. But if they are synonymous notions then how could experience be the source of empirical cognition, as is said in B2? A relatively simple resolution of this problem is to distinguish two notions of “experience”—between “experience” as the sensory result of affection by external objects (via outer sense) and oneself (via inner sense) and “experience” as the result of structuring sense perceptions via the categories. Call the latter, categorically structured state, “complex experience” (or “C-experience”) and the former state “simple experience” (or S-experience). We can then read B2 as saying that empirical cognition, which is identical with C-experience, has its source in S-experience, the material (sensation, intuition, perception) out of which complex experience is constructed.

This also helps us make sense of B1, where Kant says,

There is no doubt whatever that all our cognition begins with experience; for how else should the cognitive faculty be awakened into exercise if not through objects that stimulate our senses and in part themselves produce representations, in part bring the activity of our understanding into motion to compare these, to connect or separate them, and thus to work up the raw material of sensible impressions into a cognition of objects that is called experience? (my emphasis)

All cognition begins with simple experience, which is “worked up” into cognition of objects (i.e. empirical cognition) or (as I’ve called it) complex experience.

In §17 of the B-deduction Kant specifies that a cognition consists in the “determinate relation of given representations to an object” (B137). I take this to mean that S-experience—the contents of a sensory intuition—provides consciousness of ‘appearances’ [Erscheinungen] or ‘undetermined objects’ [unbestimmte Gegestande] (A20/B34)—which is then made ‘determinate’ via the activity of the intellect. The result is empirical cognition of an object. And it is this empirical cognition of objects (and ultimately all cognition of objects whatsoever) that it is the aim of Kant’s Deduction to legitimate, for it is this cognition that depends on the categories.

all empirical cognition of objects necessarily conforms to such concepts [i.e. the categories], because nothing is possible as object of experience unless these concepts are presupposed A93/B126).

Hence, to show that the categories are “objectively valid” or legitimate in their application to any and all objects of possible experience, is for Kant to show that the categories are necessary for empirical cognition of objects, or complex experience. One thing to think about as we move through the argument of the Deduction is whether Kant’s conception of C-experience corresponds to anything in the positions of empiricists like Locke and Hume. Do empiricists need to concede the existence of the kind of experience for which Kant thinks the categories are necessary?2

Kant’s demonstration of the validity of the categories attempts to wind its way between two poles which he terms the enthusiasm of Locke and the skepticism of Hume (A94/B127). Locke’s enthusiasm lay in the attempted derivation of all our concepts from the structure of appearances (whether we interpret these appearances as objects or merely as ideas). But this derivation is, according to Kant, inconsistent with the conditions for the manifestation of appearances and led us to improperly apply our concepts beyond their sphere.

In contrast to Locke, Hume denies that the a priori concepts have any application whatsoever, and thus Hume presents a sort of skeptical alternative to Locke. Kant’s third way argues that the structure of C-experience and the structure of our concepts are interconnected. His strategy for showing this is to show that the structure of C-experience and the structure of propositional judgment have a common root—viz. the unifying activity of the understanding. The Transcendental Deduction is thus an attempt “to try to find out whether we cannot provide for human reason safe passage between these two cliffs, assign to it determinate bounds, and yet keep open for it the entire realm of its appropriate activity” (A95/B128). How exactly Kant does this requires explication of the first step of the Deduction, to which we’ll now turn.

Structure of the Deduction

I take the structure of the deduction to consist of two steps; the first starts at §15 and concludes with §20 (§21 being essentially a recap of the preceding argument). The second step begins with §22 and concludes with §26 (with §27 also being a recapitulation). This “two-step” interpretation is not new, though interpretation of the content of each step and their relations to one another differs widely.3

The first step of the Deduction (§§15-21) argues that combination of representations cannot be given to the subject via the senses, but is rather something that the subject does. In order that the combination of representations result in cognition, Kant then argues that there must be a unitary subject, or combiner of those representations. The unity condition is satisfied by what Kant calls, §16, the “original synthetic unity of apperception.” In the next three sections (§§17-19) Kant argues that this unity of apperception requires a kind of activity, and that this activity is fundamentally that of the functions of the understanding or the “categories.” He concludes in §20 that (i) since all complex unities require combination; (ii) combination requires the original synthetic unity of apperception; and (iii) the unity of apperception is or essentially requires a kind of categorial activity; that (iv) all (complex) unity, including intuition, depends on, or “stands under” (B143) the categories.

The second step of the Deduction (§§22-7) discusses both positive and negative points raised by the first step. Negatively, Kant argues that the categories are limited to empirical cognition – i.e. to the cognition of appearances, not things in themselves. This importantly includes even our cognition of ourselves as subjects of consciousness. Positively, Kant argues in §26 that not only does all cognition rest on the application of the categories, but that one and all, appearances are necessarily categorial. He does this be showing that (i) we have a priori cognition of space and time themselves as objects, (ii) that this is only possible via the categories, (iii) and that since all appearances are governed by space and time, any appearance in space and time must also be categorially structured. This means that there is no possibility of an object’s appearing (i.e. being “given) that is not susceptible to determination by the categories. If this argument is successful then Kant will have demonstrated both that if we have cognition (“experience” in Kant’s technical sense) then it depends on the categories, and that we do have such cognition, since mathematical cognition preuspposes that we cognize space and time as objects.


Allison, Henry E. 2004. Kant’s Transcendental Idealism: Revised and Enlarged. New Haven: Yale University Press.

———. 2015. Kant’s Transcendental Deduction: An Analytical-Historical Commentary. New York: Oxford University Press.

Ameriks, Karl. 1978. “Kants Transcendental Deduction as a Regressive Argument.” Kant-Studien 69 (1-4): 273–87.

Beck, Lewis White. 1978. “Did the Sage of Königsberg Have No Dreams.” In Essays on Kant and Hume, 38–60. New Haven: Yale University Press.

Ginsborg, Hannah. 2006. “Kant and the Problem of Experience.” Philosophical Topics 34 (1&2): 59–106.

Guyer, Paul. 2010. “The Deduction of the Categories: The Metaphysical and Transcendental Deductions.” In The Cambridge Companion to Kant’s Critique of Pure Reason, edited by Paul Guyer, 118–50. Cambridge: Cambridge University Press.

Henrich, Dieter. 1969. “The Proof-Structure of Kant’s Transcendental Deduction.” The Review of Metaphysics 22 (4): 640–59.

———. 1989. “Kant’s Notion of a Deduction and the Methodological Background of the First Critique.” In Kant’s Transcendental Deductions: The Three Critiques and the Opus Postumum, edited by Eckart Förster, 29–46. Stanford, CA: Stanford University Press.

Howell, Robert. 1992. Kant’s Transcendental Deduction: An Analysis of Main Themes in His Critical Philosophy. Dordrecht: Springer.

Kitcher, Patricia. 2011. Kant’s Thinker. New York: Oxford University Press.

Longuenesse, Béatrice. 1998. Kant and the Capacity to Judge. Princeton: Princeton University Press.

Pereboom, Derk. 1995. “Self-Understanding in Kant’s Transcendental Deduction.” Synthese 103 (1): 1–42.

———. 2001. “Assessing Kant’s Master Argument.” Kantian Review 5: 90–102.

———. 2006. “Kant’s Metaphysical and Transcendental Deductions.” In A Companion to Kant, edited by Graham Bird, 154–68. Blackwell Publishing.

Van Cleve, James. 1999. Problems from Kant. Oxford: Oxford University Press.

  1. For a helpful discussion of different ways “experience” might be interpreted see (Van Cleve 1999, 73–76). See also the discussion of kinds of experience in (Beck 1978) and the sources cited therein.

  2. For a related worry about triviality see (Ginsborg 2006). [return]
  3. The locus classicus for the two-step interpretation is (Henrich 1969). Note that I am not endorsing any of the specifics of Henrich’s account, e.g., that the second step of the Deduction consists in the comparatively trivial reminder that since human forms of intuition are spatio-temporal, the categories must be know to apply only to spatio-temporal objects.