By Dennis Schulting

In this notice, I want to address a remark that Anil Gomes (2018) makes, in an excellent critique of my earlier book Kant’s Radical Subjectivism (Schulting 2017), with respect to the modal nature of the claim about the application of categories to objects, namely the belief—in my account—that ‘the destination claim [is] one about the objects of experience necessarily exemplifying the categories’. Gomes writes that this

seems too strong since, on the face of it, it looks like there can be categorial illusions: cases where the objects of judgement, experience, or perception seem to exemplify some category or other but actually fail to do so. (2018:101)

This concerns my central claim—made in Kant’s Radical Subjectivism—that objective validity is not a merely logical condition of a judgement, that is, its being true or false, but concerns what Kant calls the transcendental truth of our cognition (cf. B185/A146). The application of the categories expresses this aspect of objective validity. My judgement about an object is objectively valid in the sense that the categories instantiated in it are not only necessary for my judgement being so, but also sufficient for it. Gomes says:

If this is a point about the subject side of the gap—witness the claim about objective experience in the previous quote—then I am sympathetic to Schulting’s point. But if this is meant to be a claim about the object side—that necessarily the categories are exemplified by the objects of our judgement, experience, or perception—then it would be interesting to hear whether and how this is supposed to be compatible with the possibility of categorial illusions. (2018:101)

As I argued in Kant’s Radical Subjectivism (Chapter 4), the application of the categories to objective experience is an application to objective cognition in the sense of Kant’s Erkenntnis, and does not concern our mere experience in the colloquial (English) sense of the term, as if the experience of the object were separable from the object of experience, of which I can have knowledge (in the standard English sense of ‘knowledge’) (see also Schulting 2018b, ch. 10). Thus, if in an actual judgement about an object the categories must be instantiated in the judgement, that is, in experience (i.e. Erkenntnis), then necessarily, they are exemplified by the object of that judgement. There is no possibility of an actual judgement about an object exemplifying the categories whilst the categories are not instantiated in the object. This is what transcendental truth tells us. Transcendental truth concerns the correspondence between the necessary a priori conditions of cognition and its object. These necessary conditions are the possibility conditions for an object to be an object. They either apply or they do not. It cannot be the case that the categories apply from ‘the subject side’, as it were, while at the same time they do not apply to ‘the object side’, namely the unified whole of representations that defines the object. To argue thus would be to misunderstand what it means to unify one’s representations in virtue of the categories so as to first have a conception of what an object is. This is not something one can be mistaken about in cases in which categories are applied, i.e. in judgements of the type S is P. Of course, categories are not applied in lots of other cases: opinions, beliefs, wishes, any kind of assertion or statement that does not count as an objectively valid judgement in Kant’s strict sense are not categorially governed assertions or statement, for they do not amount to cognition or knowledge in Kant’s emphatic sense (Erkenntnis). But of course those cases are not cases of illusory knowledge; they are just not cognitions in Kant’s strong sense.

I am not sure what Gomes means by categorial illusion. Kant does of course talk about illusion (Schein), e.g. illusion of the senses (cf. Anth §24, 7:161) and especially also, in the Dialectic of the Critique of Pure Reason, which is a logic of illusion, about transcendental illusion—but the latter is not at issue in the account of experience in the Analytic. I suppose cases of hallucination or imaginary experience could be regarded as types of cognition in which categories are supposedly applied, but in a way which does not lead to genuine cognition, or knowledge. These could then be called cases of categorial illusion: one believes to have applied the categories or any one category in particular, but in fact one has not done so, that is to say, one’s assertion or statement does not yield the categorially governed cognition that one believed to have. Take this case: One might believe to have seen a tea mug containing tea which in fact turned out to be a tea mug containing coffee. My belief about the tea mug of which I had a visual representation and about which I judged on the basis of that visual information that it contained tea did not yield the objective cognition that it indeed was a mug containing tea. This is a case in which I was deceived by my visual sense into judging that I saw a tea mug with tea in it. But is this indeed a case of categorial illusion? I do not think so. Kant writes in a well-known passage at the start of the Transcendental Dialectic:

[T]ruth and illusion [Schein] are not in the object, insofar as it is intuited, but in the judgment about it insofar as it is thought. Thus it is correctly said that the senses do not err; yet not because they always judge correctly, but because they do no judge at all. Hence truth, as much as error, and thus also illusion as leading to [die Verleitung zum] the latter, are to be found only in judgment, i.e., only in the relation of the object to our understanding. In a cognition [Erkenntnis] that thoroughly agrees with the laws of the understanding there is no error. In a representation of sense (because it contains no judgment at all) there is also no error. No force of nature can of itself depart from its own laws. Hence neither the understanding by itself (without the influence of another cause), nor the senses by themselves, can err; the first cannot, because while it acts merely according to its own laws, its effect (the judgment) must necessarily agree with these laws. But the formal aspect of all truth [das Formale aller Wahrheit] consists in agreement with the laws of the understanding. In the senses there is no judgment at all, neither a true nor a false one. Now because we have no other sources of cognition besides these two, it follows that error is effected only through the unnoticed influence of sensibility on understanding, through which it happens that the subjective grounds of the judgment join with the objective ones, and make the latter deviate from their destination […]. (B350–1/A293–4, trans. emended and my underlining)

In cases where the senses do have a misleading influence on the understanding we speak of empirical illusion (empirischen Scheine), as Kant indicates (B351/A295), in contrast to transcendental illusion, which is the topic that Kant goes on to discuss in the ensuing sections. However, categorial illusion is not a term that Kant employs, and for a reason. As Kant writes in the above-quoted passage, the understanding cannot err ‘because while it acts merely according to its own laws, its effect (the judgment) must necessarily agree with these laws’ (emphasis added). This means that, necessarily, if the understanding employs the categories in that a judgement about an object is made in virtue of them, then the categories apply. One could not make the judgement in virtue of the categories and at the same be in error as to the application of the categories. In other words, there cannot be a case of categorial illusion. Either there is agreement with the laws of the understanding or there is not: in the former case categories apply necessarily, in the latter case they do not apply and no judgement is made as a result. Hence, Kant speaks of das Formale aller Wahrheit. All determinative judgements about objects are formally true in this sense by definition: their transcendental truth is an incontrovertible element of any such judgement—this excludes non-determinative judgements such as aesthetic judgements, analytic judgements etc.

But what then is it that makes one judge in error that, say, the large cup in front of me is a tea mug containing tea? And in what sense does the false judgement ‘this large cup is a mug of tea’ still constitute a transcendentally true judgement? In a critique of Kant’s Radical Subjectivism, Alexandra Newton (2018) suggested that on my account the transcendental truth and empirical truth would appear to come apart in any actual empirical judgement about a given object, or that in such a judgement the ‘relation (between substance and accident) could exist without the empirical relata existing’. Far from it. In a true empirical judgement transcendental truth and empirical truth coincide. However, transcendental truth and empirical truth are two separate issues that concern, respectively, the necessary features of any true or false judgement about an object, and the contingent features of a judgement that make it factually true or false. While it is of course trivial that a false empirical judgement about an object is not factually true, nonetheless transcendental truth does necessarily apply to it as it does to a true empirical judgement about an object. To take Newton’s own example of a brown table which I falsely judge to be green: in contrast to what Newton suggests, I do not misapply the categories ‘substance’ and ‘accident’ when I falsely judge the brown table to be green, but I falsely attribute a certain accident, namely the empirical concept <green>, to the substance that is a brown table. A false judgement about a given object x is still a judgement about x, and so necessarily involves, as in the above case, the categories ‘accident’ and ‘substance’. Without the application of these categories I could not even formulate an assertion such as “The brown table is green”. So categories are not misapplied in a false empirical judgement about some given object because the judgement is false.

One could of course misapply, say, the category of substance and accident in the sense that, for example, what one thought was a substance turns out to be an accident of a more fundamental substance. This might then be an illustration of a case in which one could speak of categorial illusion, but only in a secondary sense, in a theoretical discussion about metaphysical problems, say. In standard cases of judgement, however, error of judgement is the result of having the wrong or lacking sufficient empirical evidence for one’s judgement; here, as Kant says, ‘error is effected only through the unnoticed influence of sensibility on understanding’. Or suppose I have a family of concepts about an animal, say, and I try to apply them to something that is a machine, I could then be accused of making a ‘category’ mistake in the sense of applying the wrong set of concepts to my representations. But it would be wrong to call such a mistake categorial illusion in any straightforward Kantian sense, because the concepts here are still empirical ones. Categorial misapplication in a stronger, Kantian or transcendental, sense—where putatively, as Gomes appears to suggest, the belief that one has applied a pure concept such as substance turns out to be mistaken—is not at issue in standard determinative judgements about objects, either true or false ones, because the categories are required first to establish the very transcendental relation to the object of judgement—that is, the object is defined by the rules for synthesis in a judgement that define the categories. If no categories were instantiated in a judgement, no unity of representations would be constituted, and hence no object—as the correlate of one’s conceptual unity among the representations—would be established. This means that categories are instantiated in the object of judgement because the object of judgement just is the combination of one’s representations and categories are nothing but the set of unitary representations that first establish that combination (see further Schulting 2018, ch. 7 on the relation between combination and unity of one’s representations).

However, one might object that not all judgements are even transcendentally true, much less empirically true, for some judgements do not refer to any object at all; that is, no relation to an object is established in such judgements. Newton mentions the fact that so many metaphysicians seem to be mistaken about the claims they make, not in the sense that their claims about objects are false because they are based on misleading empirical evidence, but in the sense that they are not even about any true objects—which presumably makes them transcendentally false. They make assertions about things in themselves that they cannot back up with empirical evidence and which they therefore cannot prove to be objectively valid. So, Newton asks:

What sort of errors do they commit, if they are not errors in the misapplication of categories such as ‘substance’ or ’cause’ to things in themselves? (2018:127)

We have to be careful here. While transcendental truth is related to objective validity—it constitutes the relation to an object—it is of course not sufficient for the truth of a determinative empirical judgement; empirical, spatiotemporal intuition is needed for this. However, this fact does not imply that I cannot legitimately conceive of objects that do not have any grounding in material reality, or for which I do not have any warrant for believing in their grounding in material reality. Kant writes at the start of the A-Deduction:

Once I have pure concepts of the understanding, I can also think up objects that are perhaps impossible, or that are perhaps possible in themselves but cannot be given in any experience […] or perhaps pure concepts of the understanding will be extended further than experience can grasp (the concept of God). (A96)

Thus metaphysicians are not deceived because they misapply the categories to things that reach further than experience, such as God or the soul. But they are mistaken in believing that while employing the categories a veridical cognition of such things is possible without the help of possible experience, i.e. without empirical intuition. The lack of criticism on their part is not due to a failure ‘to grasp [their] own acts of synthesis’ (Newton 2018:127); rather, they fail to see that a legitimate act of intellectual synthesis is objectively real only if it applies to or has an effect on our sensibility (cf. B152). That a particular claim turns out to be illusory, i.e. not based on an empirical intuition of a real object, does not imply that the transcendental apperception necessarily involved in the claim, and thus the instantiation of the categories necessary to form the concept of an object in general, is illusory. The transcendental illusion does not concern the functionality of apperception or judgement, or indeed the conception of a substance, which possesses unity, identity etc. Rather, it concerns the failure to ground our judgement in the sole sufficient condition for the truth of empirical knowledge, i.e. empirical intuition.

But does this not precisely reinforce the problem raised by Gomes, namely that there can be ‘cases where the objects of judgement, experience, or perception seem to exemplify some category or other but actually fail to do so’? Categories are necessarily instantiated on ‘the subject side of the gap’, as Gomes puts it, but not thereby necessarily on ‘the object side’. But note, first, that my discussion with Gomes circled around the question of a putative gap between the first and second steps of the B-Deduction argument, namely, whether the categories are applicable to all our intuitions (or at any rate the extent in which they are so applicable). But as I argued in my reply to him (Schulting 2018a), any gap that may exist is not a gap between the subject of apperception and the concept of an object, an object in general, and also not a gap between the subject of judgement and the empirical object qua object; rather, the argumentative gap between the first and second step concerns merely the explanation of how the categories are to be seen to be instantiated in an empirical intuition of an object, that is, in virtue of the determination of time and space in contrast to the exposition, in the first step, of the mere concept of an object in general. However, this determination of time and space (explained in the second step) is the same determination by the same understanding that established the unity of one’s representations so as to constitute a concept of an object (in the first step). The differentiation between the concept of an object in general and the determinate intuition of a spatiotemporal object does not reflect the distinction between possible error (the possibility of categorial illusion) and establishing objective reality (via intuition), but two distinct steps in an explanatory exposition of the possibility of an objective cognition that consists of two necessary elements: pure concepts of the understanding and empirical intuition. There is no discrepancy between the object and subject side of the gap for any actual empirical judgement, while of course empirical illusion is still a real possibility as a result of ‘the unnoticed influence of sensibility on the understanding’. But it is not the purpose of the Deduction to exclude that latter possibility. Empirical illusion can’t of course be excluded.

The reason that in the kind of metaphysical claims that are the object of Kant’s critique in the Dialectic there is a real case of (transcendental) illusion, or, a misapplication of one or more of the categories—in other words, a categorial illusion of sorts—is because while employing an in itself objectively valid concept of an object such claims transcend the bounds of experience and thus lack the materially sufficient ground for genuine objective validity that is provided by empirical intuition alone. Categorial illusion—though not a term that Kant uses—could therefore be seen as a characteristic of unlawful metaphysical claims. The categorial illusion consists in the unbridgeable gap between an in itself objectively valid concept of an object (God, soul, etc.) and its instantiation in an empirically real object. Such a gap does not exist between the concept of an object and its instantiation in an intuition in an empirical judgement that relies on the intuition for its empirical reality. This is not because I could not, for whatever reason, be mistaken about the empirical properties of the object of my judgement, e.g. when I falsely attribute the property green to a brown table; empirical illusion is a real possibility, with which the Transcendental Deduction is not in the least concerned. But it is because for any given determinative empirical judgement about a given empirical object categorial illusion is excluded in virtue of the necessary conformity of such a judgement with the necessary conditions of its, i.e. the object’s, possibility, namely the pure concepts of the understanding or the categories.

Acknowledgements: Thanks are due to Christian Onof for comments on an earlier version of this notice.


Gomes, A. (2018), ‘Minding the Gap: Subjectivism and the Deduction’, Kantian Review 23(1): 99–109.

Newton, A. (2018), ‘Comments on Dennis Schulting: Kant’s Radical Subjectivism‘, Kantian Review 23(1): 123–130.

Schulting, D. (2017), Kant’s Radical Subjectivism. Perspectives on the Transcendental Deduction (London/New York: Palgrave Macmillan).

——— (2018a), ‘Gaps, Chasms and Things in Themselves: A Reply to My Critics’, Kantian Review 123(1): 131–43.

——— (2018b), Kant’s Deduction From Apperception. An Essay on the Transcendental Deduction (Berlin/Boston: De Gruyter).

© Dennis Schulting, 2019.

Dennis Schulting is a former Assistant Professor of Philosophy (UD) at the University of Amsterdam. His work concentrates on Kant and German idealism. His most recent journal publications were published in TIJDSCHRIFT VOOR FILOSOFIE, KANTIAN REVIEW, STUDI KANTIANI, KANT YEARBOOK, KANT-STUDIEN, and the PHILOSOPHICAL REVIEW. He is the editor of KANTIAN NONCONCEPTUALISM (Palgrave Macmillan 2016). His most recent book is: KANT’S DEDUCTION FROM APPERCEPTION. AN ESSAY ON THE TRANSCENDENTAL DEDUCTION OF THE CATEGORIES (Walter de Gruyter, 2018). He is currently working on two books on, respectively, Kant’s transcendental logic and apperception in Kant and German Idealism, as well as a monograph on Heidegger’s Schwarze Hefte.