The Metaphysical Deduction of the Categories

Sections 9-12 of the first chapter of the Analytic of Concepts form what is typically called the “Metaphysical Deduction”—a name which Kant applies only later in the Critique (see B159). The purpose of this section is to argue that of the a priori concepts we have available to us some are more fundamental than others. These are the concepts Kant calls the “categories.” Thus, the basic question that Kant seeks to answer is the question of which a priori concepts serve as the fundamental concepts of metaphysical theorizing. The MD provides what Kant calls a “clue” (or more literally a “Leitfaden” or “guiding thread”) to the discovery of these concepts. The MD has three sections. These consist of (i) an introduction where the notion of a “logical use” of the faculty of concepts or the “understanding” is set out; (ii) an elaboration of the different logical forms of judgment and a corresponding table of those forms; (iii) an explication of the categories as a priori concepts that correspond to the various logical forms of the table as set out in (ii).

Kant’s strategy in the ME is to argue that the categories are the most fundamental concepts in our thought of objects because they are the concepts that are directly linked to the various basic operations of which the understanding (the faculty of concepts) is capable. Such operations are all forms of “judgment”, and Kant construes such forms as providing the basis for determining the categories. In order that we properly appreciate why Kant chooses this strategy for butressing his choice of categories it helps to look at a prior attempt Kant makes for determining the basic concepts of metaphysics.

The Pre-Critical Strategy

In Kant’s “Inaugural Dissertation” of 1770 (whose formal title is “On the Forms and Principles of the Sensible and Intelligible World”) he lays out a theory of the manner in which the structure of the rational human mind determines what it can know of reality. In many ways this work anticipates positions and arguments Kant provides in the Critique of Pure Reason. And like that latter work, in the ID Kant seeks to delimit the fundamental concepts of metaphysics (though he does not in the ID call them categories) via an analysis of the faculty of understanding (or “intellect” more broadly). He says,

the concepts met with in metaphysics are not to be sought in the senses but in the very nature of the pure understanding, and that not as innate concepts but as concepts abstracted from the laws inherent in the mind (by attending to its actions on the occasion of an experience), and therefore as acquired concepts. (ID §8, 2:395; see also ID Corollary, 2:406 and ID §23, 2:411)

Here Kant articulates (or perhaps “gestures at” is more accurate) his view of how the concepts of metaphysics, through which we think of the intelligible world, are acquired. They are acquired by attending to the actions of the mind in course of its having experience. Here it is important for Kant’s project in the Dissertation, as it will be for his project in the Critique of Pure Reason, that his account of the acquisition of the concepts of metaphysics not run afoul of his characterization of those concepts as “intellectual” rather than “empirical” or “sensitive.”1 For this reason they cannot be concepts abstracted from sensible intuition, in the manner that our concepts of space and time are. If they were abstracted from experience such concepts would be (merely) empirical or “sensitive on account of their genesis” (ID §5, 2:393). Instead, the concepts of metaphysics are supposed to be “given in a fundamental fashion by the pure understanding itself” (ID §23, 2:411). There is, however, some question as to whether, in the Dissertation or even in lectures immediately subsequent to its publication, Kant in fact succeeds in providing an account of the acquisition of the fundamental concepts of metaphysics consistent with his stricture on their intellectual provenance.

The account of the Dissertation requires that Kant provide some story as to how we can come to be aware of the mind and its acts, and on what basis this awareness constitutes the acquisition of a non-sensory concept. At this point in his career, Kant has not yet articulated his doctrine of “pure” apperception as distinct from inner sense (more on this in the next set of notes). So it looks as if, given his distinction between intuition and concept and their related faculties, that it must be an inner intuition that accounts for the acquisition of the concepts of metaphysics. However, if it is inner intuition that is the basis for such acquisition, it looks like Kant will run afoul of his requirement that the concepts of metaphysics be intellectual as opposed to empirical (for even the “pure” concepts of space and time, though free of sensation, are nevertheless sensitive on account of being derived from intuition (see, e.g., ID §5, 2:393)).

Perhaps Kant thinks we have a special intellectual intuition of the mind and its acts? One recent commentator (Dyck 2016) has argued that Kant is committed in the Dissertation to our having intellectual intuitions of our own minds and that he only subsequently, in the later 70s, rejects this position in favor of one according to which it is inner sense—and thus empirical psychology—that provides the requisite basis for acquiring metaphysical concepts. Dyck’s argument concerning Kant’s position in the Dissertation hinges on two claims. First, that there are clear texts showing that Kant denies that sensible laws (i.e. of time and space) apply to immaterial substances, including the mind/soul (see Dyck 2016, 330). Second, that Kant’s conception of the acquisition of the pure concepts of metaphysics depends on the existence of intellectual intuition of the mind or soul itself (see Dyck 2016, 330–1). However, neither of these interpretive points are particularly compelling. Against the first, it is relatively clear from the context of Kant’s statements regarding immaterial substance that he denies that the principles of the corporeal world apply to such substances (ID §27, 2:414). That would mean that principles belonging to substances understood as bodies do not apply to the mind. But this is compatible with the conception of the mind as nevertheless governed by other sensible principles, and in particular, by time. Against Dyck’s second point, in the Dissertation Kant explicitly states that finite beings lack intellectual intuition (ID §10, 2:396-7) and that “the accidents which are not included in the relations of space, such as the thoughts of the mind” are in time (ID Corollary, 2:406). It thus seems unlikely that Kant holds in the Dissertation that metaphysical concepts are acquired via intellectual intuition, on pain of explicitly contradicting himself at multiple points.2

But even if Kant does not directly contradict himself in the Dissertation, Dyck’s discussion helps point us to a clear tension and perhaps fatal ambiguity in Kant’s pre-critical view. One might, therefore, have hope that Kant’s remarks in lectures and notes subsequent to the publication of the Dissertation clarify his position. However, the lecture texts we have from the 1770s only make matters more complicated. For example, in lectures shortly after the Dissertation Kant says, “We have no intuition in the whole world except the intuition of our self; all other things are appearances” (Anthropologie Collins 25:15 (1772-3)). Kant’s point here is not that we have only one intuition—viz. of ourselves. Rather, it is that all of our outer intuitions are of appearances, while our inner (and non-intellectual!) intuitions present ourselves as substantial subjects. In a reflexion also from the early-to-mid 1770s Kant says that “The I is the intuition of a substance” (R4493, 17:571 (1772-5)). In the Metaphysics L1 lectures, from roughly the same period, Kant contrasts consciousness of external objects with consciousness of the self. One intuits oneself immediately, but the same is not true of external objects (28:206-7, 224). The self (as intelligence) so intuited is substantial, simple, and immaterial (28:224-5).

In a note (R4674) from the Duisberg Nachlass of 1775-6, Kant writes that an (external) object may be represented only “according to its relations”, i.e., only according to the properties, and relations between those properties, that are presented in sense experience.3 And, as he does in the lecture material cited above, Kant contrasts our position with respect to intuition of external objects with the special access we have to ourselves. The inner intuition of oneself is of an object whose properties are not presented, as outer things are, merely in terms of relations.4 The language Kant uses in R4674 is also echoed in other texts from the mid-70s, For example, Kant states that the ‘I’ is the “original concept” of substance which we “borrow” for use in our conception of other substances (Metaphysics L1, 28:225-6). Relatedly, he remarks that “the I expresses the substantial; for that substrate in which all accidents inhere is the substantial. This is the only case where we can immediately intuit the substance” (Pölitz Metaphysik 28:226 (1777-80)).

The view that emerges from these texts, and that may in fact already be present in Kant’s Dissertation, is one according to which inner sense provides a privileged epistemic relation to oneself. In inner sense one is presented not merely as a set of relations, as in outer sense, but as a subject of properties, as what Kant sometimes describes as an intelligence or “thinking substance” (e.g. Metaphysics L1 28:224-5). The self is thus the “original of all objects” in the sense that, as Allison Laywine puts it, “we somehow transfer our representations of the one true subject and apply it derivatively or by analogy to our thought of anything else.”5

Unfortunately, Kant never clarifies how this position concerning the privileged and peculiar role of inner sense is consistent with his general doctrine of sensible intuition, or how it is consistent with the Dissertation’s position that no pure intellectual concept can be derived from the content of sensible intuition. The overall impression of Kant’s position based on his remarks in the Dissertation, and subsequently in the lectures, is that of a steadfast commitment to roughly the following position. Introspection provides a form of intuitive acquaintance with the self as a metaphysical subject, and it is via this acquaintance that we can then form by analogy the representations of objects (construed as metaphysical subjects of properties) distinct from us. But how this position is ultimately supposed to cohere with Kant’s other commitments in the 70s is, at best, unclear. It in facts seems a reasonable suspicion that Kant’s various commitments in the 70s don’t cohere, and that this resulting tension in his view is part of what pushes him towards the critical distinction between inner intuition and pure apperception. We’ll talk about this distinction further when we discuss the argument of the Transcendental Deduction. In the next section I talk about Kant’s revision of his Dissertation strategy in the Critique for the argument of the MD.

The Critical Strategy

In the Dissertation Kant distinguishes between what he terms the “logical” and the “real” uses of the understanding (ID §23, 2:410–11). He says,

in pure philosophy, such as metaphysics, the use of the understanding in dealing with principles is real; that is to say, the fundamental concepts of things and of relations, and the axioms themselves, are given in a fundamental fashion by the pure understanding itself; and, since they are not intuitions, they are not immune to error.

In the first Critique Kant maintains the view that the concepts of metaphysics come from the understanding. However, he significantly alters the method by which these concepts are derived. Specifically he now closely links the categories with the logical forms of judgment. The latter are not just a “clue” to the organization of the table of categories, they are the categories, at least in their “logical” use. The central texts making this plain are as follows.

The same understanding, therefore, and indeed by means of the very same actions through which it brings the logical form of a judgment into concepts…also brings a transcendental content into its representations by means of the synthetic unity of the manifold in intuition in general…(MD §10, B105)

[The categories] are concepts of an object in general, by means of which its intuition is regarded as determined with regard to one of the logical functions for judgments. (TD Transition, B128)

That action of the understanding…through which the manifold of given representations (whether they be intuitions or concepts) is brought under an apperception in general, is the logical function of judgments. … But now the categories are nothing other than these very functions for judging, insofar as the manifold of a given intuition is determined with regard to them (TD §20, B143)

In the metaphysical deduction the origin of the a priori categories in general was established through their complete coincidence [völlige Zusammentreffung] with the universal logical functions of thinking (TD §26, B159)

the pure concepts of the understanding are, of themselves, nothing but logical functions, but that as such they do not constitute the least concept of an object in itself but rather need sensory intuition as a basis, and even then they serve only to determine empirical judgments (Pr §39, 4:324)

These texts communicate two important points. First, they indicate Kant’s pursuit of a strategy for explaining how the pure categories, or fundamental concepts of metaphysics, arise from the intellect. This strategy avoids commitment to any form of content nativism, and it avoids the problem that we saw plagued Kant’s account in the Dissertation—viz. explaining how the pure categories are “pure” in the sense of being traceable only to the intellect, without thereby implicating either intellectual intuition, which he denies, or sensible (inner) intuition, which would undermine their claim to purity.

Pursuit of this strategy means that, second, the very logical functions for combining representations (concepts) in judgment, and which constitute the basis of study in what Kant calls “pure general logic”, are also the functions for combining representations (intuitions) in our experience of objects. Thus for the understanding as a faculty of judging (A69/B94), each logical function of thinking (e.g. categorical judgment), when applied to a multiplicity of intuitions, results in a distinctive way of relating to, or “experiencing” in Kant’s technical sense, an object e.g. a substance.

Thus, Kant has a much more coherent strategy than he did in the Dissertation for explaining how we come to grasp the a priori concepts necessary for doing metaphysics. The notion that we can analyze our logic and determine which are the basic logical forms is plausible and does not require appealing to some privileged access we might have to ourselves in introspection (i.e. inner sense).

However, for this strategy to ultimately succeed, Kant must provide substantial arguments for several controversial claims. First, he must show that he has successfully given a complete analysis of the logical forms of judgment; second, he must show that the metaphysical concepts derived from these logical forms (i.e. the categories) really provide the complete basis for doing metaphysics; third, he must show that the content of the categories does not include material absent from the purely logical forms—if there were such “extra” content in the categories, then it is not clear how they are supposed to be “identical” with the logical forms, as Kant sometimes indicates that they are.

Unfortunately Kant fails to meet any of these requirements in the MD. There is no formal demonstration of the completeness of either the table of judgment or, correspondingly, the categories. There is likewise no demonstration provided that shows that all and only these categories provide the basic materials for metaphysics. Finally, it is not obvious that the content of the categories is simply “identical” with that of the forms of judgment.

To take just one example, Kant construes the category of substance as related to the form of categorical (i.e. subject/predicate) judgment. The purely logical or grammatical notion is thus of a term or concept that can only occupy subject position and never predicate position. However, this notion does not adequately capture central ways in which Kant uses the concept . More specifically, Kant construes a substance as substanding and subsisting. This means that (i) substance is a being in which things “inhere” and (ii) substance does not itself inhere in anything else (in this circumscribed sense substance is an independent being). Kant construes inherence as a real (that is, non-logical) asymmetric dependence relation between a subject of inherence (the substance) and its properties or modes (i.e. the way or ways in which the subject exists). The purely logical or grammatical conception of substance as a subject term that is never a predicate term fails to properly capture this dependence relation, for it fails to show how a predicate term might asymmetrically depend on a subject term in a manner that models the metaphysical relation between substance and mode.

In the end then, while Kant’s strategy in the MD is a clear improvement over his prior attempt in the Dissertation, it is not at all clear that he successfully defends his claim to have provided a principled and a priori basis for exhaustively determining which are the fundamental concepts. Kant wants an argument that,

has not arisen rhapsodically from a haphazard search for pure concepts, of the completeness of which one could never be certain, since one would only infer it through induction, without reflecting that in this way one would never see why just these and not other concepts should inhabit the pure understanding. (A81/B106-7)

Kant does not provide convincing proof of this position in the MD. Whether such a proof could be given is, perhaps, another story.6


Allison, Henry E. 2004. “The Intellectual Conditions of Human Cognition: Kant’s Metaphysical Deduction.” In Kant’s Transcendental Idealism: Revised and Enlarged, 133–56. New Haven: Yale University Press.

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

Brandt, Reinhard. 1995. The Table of Judgments: Critique of Pure Reason A67-76; B92-101. Edited by Eric Watkins. Atascadero, CA: Ridgeview.

Carl, Wolfgang. 1989a. Der Schweigende Kant: Die Entwürfe Zu Einer Deduktion Der Kategorien Vor 1781. Vol. 182. Göttingen: Vandenhoeck & Ruprecht.

———. 1989b. “Kant’s First Drafts of the Deduction of the Categories.” In Kant’s Transcendental Deductions, edited by Eckart Förster, 3–20. Stanford: Stanford University Press.

Dyck, Corey W. 2016. “The Scope of Inner Sense: The Development of Kant’s Psychology in the Silent Decade.” Con-Textos Kantianos 3: 1–19.

Engstrom, Stephen. n.d. “The Category of Substance.” Logical Analysis and History of Philosophy 21.

Gardner, Sebastian. 1999. Kant and the Critique of Pure Reason. London: Routledge.

Greenberg, Robert. 1994. “The Content of Kant’s Logical Functions of Judgment.” History of Philosophy Quarterly 11 (4): 375–92.

Guyer, Paul. 1987. Kant and the Claims of Knowledge. Cambridge: Cambridge University Press.

———. 2014. Kant. London: Routledge.

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

Klemme, Heiner F. 1996. Kants Philosophie Des Subjekts: Systematische Und Entwicklungsgeschichtliche Untersuchungen Zum Verhältnis von Selbstbewusstsein Und Selbsterkenntnis. Hamburg: Felix Meiner Verlag.

Langton, Rae. 1998. Kantian Humility: Our Ignorance of Things in Themselves. Oxford University Press.

———. 2006. “Kant’s Phenomena: Extrinsic or Relational Properties? A Reply to Allais.” Philosophy and Phenomenological Research 73 (1): 170–85.

Laywine, Alison. 2005. “Kant on the Self as Model of Experience.” Kantian Review 9 (1): 1–29.

———. 2006. “Kant’s Metaphysical Reflections in the Duisburg Nachlass.” Kant Studien 97 (1).

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

———. 2006. “Kant on a Priori Concepts: The Metaphysical Deduction of the Categories.” In The Cambridge Companion to Kant and Modern Philosophy, edited by Paul Guyer, 129–68.

Lovejoy, Arthur O. 1907. “Kant’s Classification of the Forms of Judgment.” The Philosophical Review 16 (6): 588–603.

McLear, Colin. 2017. “Motion and the affection argument.” Synthese, 1–17.

Mohr, Georg. 1995. “Freedom and the Self: From Introspection to Intersubjectivity: Wolff, Kant, and Fichte.” In The Modern Subject: Conceptions of the Self in Classical German Philosophy, edited by Karl Ameriks and Dieter Sturma, 31–45. Albany, NY: SUNY Press.

Nussbaum, Charles. 2008. “Concepts, Judgments, and Unity in Kant’s Metaphysical Deduction of the Relational Categories.” Journal of the History of Philosophy 28 (1): 89–103.

Pereboom, Derk. 1991. “Kant’s Amphiboly.” Archiv für Geschichte der Philosophie 73 (1): 50–70.

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

Reich, Klaus. 1992. The Completeness of Kant’s Table of Judgments. Stanford University Press.

Serck-Hanssen, Camilla. 2001. “Apperception and Deduction in the Duisburgischer Nachlass.” Kant und die Berliner Aufklärung: Akten des IX. Internationalen Kant-Kongresses 2: 59–68.

Wolff, Michael. 1995. Die Vollständigkeit Der Kantischen Urteilstafel: Mit Einem Essay über Freges Begriffsschrift. Vittorio Klostermann.

Wuerth, Julian. 2014. Kant on Mind, Action, and Ethics. Oxford: Oxford University Press.

  1. What are the “concepts of metaphysics” for Kant? Certainly they include the concept <God>, as well as <soul> and <world>. These all presuppose possession of the concept <substance>. Presumably also on the list are other concepts one would find in, e.g., Baumgarten’s Metaphysica, such as <possibility>, <necessity>, <accident>, and so forth.

  2. For consideration of the problem of self-consciousness in the 1770s more broadly, as well as Kant’s rejection of intellectual intuition, see (Klemme 1996, 118–26). As (Mohr 1995, 32–36) points out, Kant does speak of an intellectual (not inner) intuition of freedom in reflexionen from the early-to-mid-1770s (e.g. R4336 17:509 (1769-75)) But Kant also speaks of apperception as opposed to intuition in the mid-1770s (e.g. R4723 17:688 (1773-5); R6860 19:183 (1776?-91?)). It is clear that he is struggling to express the nature of the awareness we can have, as rational beings, of our own activity (e.g. R4220 17:462 (1769-70)).

  3. For discussion of Kant’s view that the objects of sense consist entirely of relations, or relational properties, see (Pereboom 1991; Langton 1998, 2006; McLear 2017).

  4. There is also indication in the Duisburg Nachlaß, as there was in Kant’s lectures, of his endorsement of the rational psychology he would later come to criticize. For example, in the Nachlaß Kant says that “I would not represent anything as outside of me and thus make appearance into experience (objectively) if the representations did not relate to something that is parallel to my I, through which I refer them from myself to another subject” (R4675, 17:648 (1775); my emphasis). For discussion see (Guyer 1987; Carl 1989b, 1989a; Serck-Hanssen 2001; Laywine 2005, 2006; Kitcher 2011).

  5. (Laywine 2005, 9); see also (Carl 1989a, 182:91–92, 97; Kitcher 2011, 73–74; Wuerth 2014, 104; Dyck 2016, 335–8). For criticism of Carl’s, and to a lesser degree Laywine’s, position see (Allison 2015, 121–30).

  6. There are a variety of prominent attempts to defend Kant’s position here. See (Reich 1992; Brandt 1995; Wolff 1995; Longuenesse 1998).