DENNIS SCHULTING | Kant’s Radical Subjectivism: Perspectives on the Transcendental Deduction | London: Palgrave Macmillan 2017



This article was first published in Dutch in Tijdschrift voor Filosofie 80 (2) (2018):363–378. The current English translation is a somewhat modified version of the original article. 


By Dennis Schulting

In this paper, I respond to critiques by Karin de Boer, Henny Blomme, Hein van den Berg and Joris Spigt of my book Kant’s Radical Subjectivism: Perspectives on the Transcendental Deduction (London: Palgrave Macmillan, 2017). I address issues that are raised concerning objectivity, the nature of the object, the role of transcendental apperception and the imagination, and idealism. More in particular I respond to an objection against my reading of the necessary existence of things in themselves and their relation to appearances. I also briefly respond to a question that relates to the debate on Kantian nonconceptualism, more in particular, the question whether Kant allows animals objective intentionality. Lastly, I respond to one objection against my reading of Hegel’s critique of Kant. In my reply, I shall proceed thematically, addressing four main themes which are also central to my book: objectivity, the thing in itself, nonconceptual representational content and Hegel’s critique of Kant. This division also neatly corresponds to the focus of the different critics.

Hein van den Berg raises a very interesting question—namely, about the sense in which I suggested that Kant is probably a coherentist (2017:2–3)—which because of lack of space I cannot answer here. In any case, I agree with Van den Berg that Kant cannot be considered a coherentist in a contemporary sense and that Kant is more of a traditional foundationalist. Whether Kant sees truth as correspondence or coherence is a complicated matter, which also relates to the question concerning what Kant calls ‘transcendental truth’ (see 2017, chs 3 and 4); transcendental truth does in some sense concern a correspondence between concept and object or thing (adæquatio intellectus et rei), but one that is fully internalised in the transcendental subject—the metaphysical question of truth and the epistemology of the cognising subject have here merged into one. This is compatible with the view, as Van den Berg suggests, that Kant’s idealism implies that he espouses a correspondence theory of truth, at any rate insofar as the truth of empirical knowledge claims about appearances is concerned. As Van den Berg himself already notes, it is not easy to decide how exactly to interpret Kant’s theory of truth. I want to return to this issue when another opportunity arises. 

1. Transcendental apperception and objectivity 

I confine myself here to the substantial criticisms. First I address the question concerning transcendental apperception, which is broached by Karin de Boer (hereafter cited as De Boer 2018). The central thesis of Kant’s Radical Subjectivism—and this goes against more traditional realist intuitions, which seem ineradicable even among Kantians—is to claim that objects as such do not exist outside the transcendental subject and thus outside transcendental apperception. As De Boer acknowledges, I claim that transcendental apperception is the necessary and formally sufficient condition of objective knowledge (note the adverb ‘formally’!, which De Boer handily fails to do). De Boer considers this claim untenable because it would make Kant into a rationalist. For, as she writes, ‘transcendental apperception [is] a purely intellectual activity’ (De Boer 2018:357)—I disagree with this, by the way, because this would mean that transcendental apperception only concerns the analytical unity of concepts, and not also precisely the synthetic unity of representations in the non-intellectual, sensible intuition, regardless of the question whether this intuition is ‘similar to our own or not’ (B148).

There are two additional subjective sources of knowledge: sensibility and the imagination, as De Boer notes, and the imagination cannot be reduced to the understanding.[1] Obviously, I do not deny that sensibility and the imagination are necessary; what’s more, I dedicate two whole chapters to the imagination and its absolutely central role in the argument of both the A and B Deductions (chs 6 and 7). De Boer suggests she did not consider those two chapters, while the titles of those chapters clearly refer to the threefold synthesis (in the A Deduction) and the figurative synthesis (in the B Deduction), and thus are important for the critique she directs against my reading, especially with respect to the imagination. As said, De Boer also neglects to take note (twice in her text) of important words such as ‘at least formally’. I emphasise again and again that what is at stake is the possibility of objective, empirical knowledge as depending on transcendental apperception, namely insofar as such knowledge can be established a priori as objective. A posteriori sensible impressions must thereby be presupposed as necessarily given. Neither the fact that there are sensible impressions, nor their a posteriori nature, nor the fact that there is objective knowledge, nor the material (contingent) nature of cognitive judgements can be demonstrated in a transcendental proof (see e.g. chs 3, 4 and 7). Hence I wrote that apperception is only formally sufficient for objective, empirical knowledge, insofar namely as it concerns the form and not the matter of such knowledge.[2]

However, one could still object, as De Boer appears to do, that my thesis cannot be right and that transcendental apperception is at the most a necessary and not even a formally sufficient condition of objective knowledge, for transcendental apperception concerns only the principle of self-ascription of representations and does not ground our objective representations, let alone the objects of those representations. De Boer in fact poses the same question as some other commentators from the recent past[3]: why would the necessary and sufficient condition of self-consciousness (transcendental apperception) also be the necessary—let alone sufficient—condition of both the experience of objects and the objects of experience (A111, A158/B197)?

The reader of De Boer’s piece could be forgiven for thinking that I failed to take this question into account and that her critique is justified, but nothing could be further from the truth. It is a central problematic in both my first book and the book critiqued here, and it was again discussed in an exchange between Anil Gomes and myself in Kantian Review (see here and here). If one denies that transcendental apperception is also a (formally) sufficient condition of objective knowledge (the claim namely that Kant makes in B138, the passage I discuss in detail),[4] a so-called GAP results in Kant’s argument in the Deduction (sometimes this analytic jargon comes in really handy!).

My solution to the GAP is, put succinctly, to say that (1) transcendental apperception is not a psychological principle of representation or consciousness, (2) the transcendental unity of apperception is an objective unity, as Kant says himself (B139), and (3) transcendental apperception in the guise of the imagination in sensibility (i.e. as synthesis of the apprehension, cf. §26) is responsible for the necessary combination of spatiotemporal representations, such that we can speak of objective knowledge (or experience, with which Kant identifies knowledge [B148]), assuming of course that there are objects and those objects sensibly affect us (the factualness of experience and of a world of objects is never put to doubt by Kant). Again, it is purely the formal aspect of our objective knowledge which is at issue here, objective knowledge qua objective, not its material aspect.[5] De Boer overlooks this crucial aspect.

That which leads to the GAP is the mistaken assumption that Kant talks about the material aspects of our objective knowledge (knowledge facts) and, crucially, that transcendental apperception is merely the principle of the self-ascription of all possible representations (whether or not interpreted psychologically). The traditional reading of transcendental apperception is modally ‘excessive’, which I explain ad nauseam in both my books. If De Boer were right in her critique, it would directly lead to a multiple GAP (the same holds mutatis mutandis for the critique that Henny Blomme presents; see below). For she is not able to explain the a priori, internal relation between the various formally distinguishable necessary elements of sensibility, synthesis of apprehension, imagination and apperception: what connects them, and how can Kant demonstrate that it is the imagination that is the sufficient condition of objective knowledge?[6] Is it again another faculty which on a more fundamental level connects everything, a kind of ‘schminagination’, as Robert Hanna has put it very aptly?[7]

De Boer’s way of presenting it leads to the philosophical problem of an infinite regress in the explanatory grounds. And the original synthetic unity of apperception was precisely designed by Kant to avoid such regresses in the analysis of possible objective experience. My thesis of Kant’s radical subjectivism bypasses this potential infinite regress in the explanatory grounds and shows how all formally distinguishable explanatory elements necessarily hang together internally (including the categories). Kant must not be read as if he were still thinking along the lines of the definition and classification urge of his predecessors. Rather, he should be read more in the fashion of a dynamical interpretation such as Hegel’s.[8]

As I explicate in detail in the book, there is nothing outside the transcendental apperception of our representations insofar as it concerns objectively valid experience. That is to say, nothing outside our representations simpliciter can function as a justifying ground of our knowledge claims. On what does objective knowledge rest? Not on sensible facts, not on objects out there, but upon our capacity to apprehend, in virtue of an original synthesis, our representations as of a sensible, spatiotemporally intuited object. Transcendental apperception is that original synthesis, and apperception is not merely ‘intellectual’, as De Boer thinks. The ‘intellectual synthesis’ of which Kant speaks at B151 is a formal abstraction in the analysis, which shows that intellectual synthesis is the necessary condition of objectively valid knowledge, whereas figurative synthesis is the sufficient condition; but both syntheses are aspects of one and the same principle of transcendental apperception, manifestations of the understanding, as Kant clearly says in §24 of the B Deduction (see also B153, where Kant speaks of the understanding ‘under the designation of a transcendental synthesis of the imagination’). Transcendental apperception thus establishes the bridge between the purely rational (the combining of concepts) and the sensible, of course by means of the imagination; but the imagination is not another, distinct capacity or faculty other than transcendental apperception itself as the act of the understanding in its manifestation in the manifold of an empirical intuition. The distinctions are formally required for the analysis, but in an actual empirical judgement which yields knowledge transcendental apperception always is the imagination which as synthesis of the apprehension does its synthesising work (see §§24 and 26 of the B Deduction).

The only absolute distinction that Kant makes is the one between, on the one hand, the capacity for synthesis, which also grounds our conceptuality as such, and sensibility, on the other. We can never get immediate access, merely in virtue of our conceptuality, to actually existing things. In that sense, sensibility plays a necessary role and provides the materially sufficient ground of our objective knowledge. But objectivity or objective validity, objective knowledge qua objective, is purely a function of our transcendental apperception, naturally in the guise of the imagination to the extent that it concerns spatiotemporal objects. That is my thesis of Kant’s radical subjectivism, which is not a psychological subjectivism, nor an intellectual rationalism, but a transcendental theory of knowledge that shows what it means to be able to say, in general, that we have knowledge of objects and under which conditions it is possible to say so. De Boer rightly points to the role of the categories. While I do talk about that role in the book (in Chapter 2, which De Boer says she did not consider), the real story can be found in the 340 plus pages of my previous book [Kant’s Deduction From Apperception, rev. ed., De Gruyter 2018], in which I discuss in detail the derivation of the categories from apperception and how in this way a relation to an object is established.[9] 

Lastly and crucially, De Boer makes an interpretative mistake in thinking that ‘the concept “real object”’ (De Boer 2018:356) plays a positive role in my interpretation. She is apparently confused about this putative position of mine because it would ‘undermine […] the “idealist” element’ of my interpretation (De Boer 2018:360). However, it is not my position. In the passage that De Boer cites, I am clearly talking about a false view shared by many Kantians with a realist orientation, who think more along the lines of the traditional theory of knowledge, that is, that the categories concern merely our way of experiencing or knowing the object and not the object itself (the ‘real object’) independently of our way of thinking or experiencing. The question that De Boer subsequently poses—‘So is there a “gap” between our pure concepts and the real objects to which they refer after all?’ (De Boer 2018:360)—can confidently be negated. As De Boer says towards the end of §2, but means as a critique of my position: ‘In my view, the term “real object” does not play a role in this context’ (De Boer 2018:361). And nor in my view, and I nowhere suggest this. I make it quite clear—for this is precisely the core of my thesis of Kant’s radical subjectivism—that outside the categories, outside transcendental apperception, no real objects exist (things in themselves do of course so exist, but I will come back to this in my response to Blomme below).[10] 

As with De Boer, Henny Blomme’s view (hereafter cited as Blomme 2018) also runs the risk of an infinite regress in the explanatory grounds when he talks about ‘non-analytic transitions’ (Blomme 2018:370) in the deduction argument. Those transitions are indeed non-analytic in terms of the semantic content of the argument, but at the level of the analysis, i.e. the deduction argument itself, those transitions are very much analytic. There is an ‘analytical connexion’, in the words of P.F. Strawson, in Kant’s transcendental argument, otherwise there is no transcendental-logical sequence of necessarily cohering premisses—the deduction is an a priori argument after all—and a valid conclusion following from those premises. It is the analytic principle of transcendental apperception—from which the steps of the a priori argument proceed—which ‘explains’, as Kant says (B135), the non-analytic (i.e. synthetic a priori) semantic content of the analysis. 

Blomme subsequently refers to ‘weaker’ and ‘stronger’ notions of objective validity, of which I would be making use. But in my perspective there is no such thing as a weaker and stronger form of objective validity. Neither do I see any textual ground in the Deduction that would support such a distinction. Of course, there is—as I make it clear in the book—a clear formal distinction between the concept of an object in general and an actual, concrete judgement about an empirical object, whether this object is real or not. (Naturally, as Blomme says, it can also be an imaginary constructed mathematical object.) However, one must then not transpose this distinction, which plays a formal role in the analysis of particularly the B Deduction, to reality as if some arbitrary epistemic agent literally first could have an objectively valid concept of an object in general, subsequently employs it in a judgement and then drags in an empirical intuition so as to finally have an objectively valid cognition in its so-called ‘strong’ form. As Kant says, concepts are employed only in judgements, not outside of them (A68/B93), and this holds certainly also for the categories, which are nothing but functions of judgement insofar as the underlying intuition is determined (B143),[11] and equally for such an offshoot a priori concept as the concept of ‘an object in general’. The concept of ‘an object in general’ is after all nothing but the objective unity of consciousness or apperception and the objective unity of apperception is nothing but the definition of judgement in its most essential, i.e. objectively valid character.[12] To say that one can have an objectively valid concept of an object in general independently of how such a concept functions in a de facto empirical judgement (whether it concerns concrete, empirical or abstract, mathematical objects) is transcendental-logically nonsensical. One should not try to situate a concept that has a function in an analysis in an ontological context as if it concerned an isolatable moment in concrete experience.

As I explain repeatedly in the book, objective validity is a function of judgement and of judgement only. Judgements need of course not be about literally perceivable objects (one can think of magnetic fields, which are not perceivable with our senses), but objectively valid judgements in the strict sense should always be coupled to possible experience. Blomme appears to make a distinction between, on the one hand, objective validity and, on the other, the objectivity of cognitive judgements or what he calls ‘objective cognitive judgements’ (objectieve kennisoordelen [Blomme 2018:370]). But it is not clear to me what he means by the latter. If he means ‘true judgements’,[13] then I concur, for as I explicate in detail in Chapters 3 and 4, objectively valid judgements need not be true judgements and true judgements need not be objectively valid judgements; put otherwise, objective validity has nothing to do with the truth value of a judgement. I can make a false judgement which is still objectively valid (e.g. a judgement about a certain object that I perceive, but to which I attribute the wrong properties), or I can make a judgement which is not at all objectively valid regardless of the question whether it is true or false (e.g. analytic judgements, which are not about objects).

Blomme uncouples the objective validity of the categories from judgement, which in my view is not justified because categories are in fact functions of judgement. I suspect he conflates objectively valid and true (factually objective) judgements, but if that is true, it is clearly mistaken, for as said not all objectively valid judgements are true or factually objective and not all true judgements are objectively valid—i.e. analytic judgements, which are actually not real judgements in the sense of Kant’s definition in the Deduction in §19, but rather statements or propositions that have a surface grammar similar to proper judgements. Uncoupling objective validity from judgement only leads to such confusion.[14] Blomme’s analysis takes place on too abstract a level and he confuses the ‘event’, as it were, of judgement with the formal analysis of its various constituent elements. 

The difference between types of object does not make a difference to the nature of objective validity, which is purely and only a function of the unity of apperception, by which judgement is fundamentally characterised, and thus of the categories. An objectively valid judgement is a judgement in which the categories are instantiated, independently of the question of whether the judgement is true or false. Empirical intuition is a necessary condition of concrete empirical judgements, but it does not furnish anything in terms of a justifying ground which is not already provided completely by the unity of apperception, which happens in the form of a figurative synthesis in the manifold of an empirical intuition (cf. B151–2). Kant’s theory of knowledge is not an epistemic externalism, but a thoroughgoing internalism as far as justification is concerned. Therefore, I do not see a reason to make a distinction between ‘weaker’ and ‘stronger’ forms of objective validity.

What empirical intuition does uniquely provide (and this is probably what Blomme means), is real possibility. It is real possibility which makes the difference between judgements about transcendent objects that we cannot experience and objectively valid judgements about objects of our possible experience—which is the real topic of the Deduction. What Kant thus demonstrates in the so-called ‘second step’ of the B Deduction is to show that the concept of an object in general that is conceived in the ‘first step’ corresponds to a spatiotemporal really possible object in empirical intuition. In a certain sense—and this is of course the actual goal of the Deduction—it is only in the second step that the objective validity of the categories in a judgement about an empirical object is first demonstrated. But this does not detract from the fact that objective validity lies in the nature of judgement as such, because it is—and this is the core of my account about Kant’s radical subjectivism—the transcendental, judging, apperceiving subject which determines the objective validity and nothing else. (I fully concur in this with Robert Pippin’s position.) In other words, it is not as if with the empirical intuition a little more objective validity were added to the concept of an object as it is employed in judgement.

In any case, Blomme is mistaken to believe that I use ‘two notions of “objective validity’ of the categories’ (Blomme 2018:370) and I do also not go along with his proposal to distinguish between the objective validity and ‘the objective realisation’ of the categories (Blomme 2018:370–2). I see neither the textual support nor the interpretative requirement for such a move,[15] let alone for what he calls the ‘subjective validity’ and ‘objective validity of judgements’. Blomme probably means by ‘subjective validity’ of judgements the judgements of perception from the Prolegomena which are only subjectively valid. But Blomme could have known from reading the book that I concur with Konstantin Pollok’s thesis that Kant abandoned the theory of judgements of perception before the B Deduction.

As far as his complaint is concerned that I do not sufficiently differentiate between perception, judgement of perception, experience etc., I believe that Blomme tries to impose his own ‘layer cake’ model on my reading, a model that following James Conant[16] I reject. Blomme thereby completely ignores my account of the modality of Kant’s argument regarding perception in e.g. §26 of the B Deduction (see Chapters 5 and 7). What he subsequently says about my reference to ‘limitation’ in Chapter 9 is unwarranted. Blomme disregards the metaphysical context of the discussion in that chapter. For an exhaustive delineation of the function of the categories of quantity and quality in the deduction itself, I should like to refer him to Chapters 8 and 9 of my previous book.

2. The Thing In Itself

Blomme puts forward more interesting points of criticism in his discussion of the thing in itself and how I construe it. He wonders if the existence of things in themselves independently of our cognitive capacity is at all compatible with my thesis of radical subjectivism. However, it is precisely peculiar to radical subjectivism that it is not a psychological or empirical idealism, which denies the independent existence of things in themselves. I assert many times that the things in themselves and their Sachheit, their reality (cf. A143/B182), are entirely independent of the transcendental subject. Put briefly, Being is as such not dependent on possible experience, hence things in themselves are as such not dependent on possible experience. 

Unfortunately, some of my claims are taken out of context by Blomme, which he then sometimes also interprets wrongly. An example. Blomme writes: 

On p. 42, Schulting talks explicitly about the existence of the Ding an sich as a mere possibility instead of a certainty: ‘We do represent, by way of its appearance(s), the real thing in itself, which exists outside us (if it exists)’ (my italics). (Blomme 2018:375–6, Blomme’s emphasis, quoting Schulting 2017:42) 

It is not at all the case that I talk here about the mere possibility of the existence of things in themselves. The quoted phrase indicates that we can say that an appearance exists on the grounds of a thing in itself existing only if that thing in itself actually exists so that it can appear. What is important here is to take into account the modality of our judgements: if we judge that an empirical object x is an appearance E of y, whereby y is a thing in itself, then the judgement that x is E is true if and only if x exists, x is E, and y exists. (And this can be considered independently of the question wether the relation between x and y is one of numerical identity, i.e. independently of what the status of E is with respect to the relation between x and y.)

Blomme consistently speaks of ‘real reality’ (werkelijke werkelijkheid), ‘genuinely real’ (echt werkelijke), but that is at least misleading because the categories Wirklichkeit (actuality) and Realität (reality) must be kept separate in the Kantian context—in contrast to what Blomme thinks, in my book a putative ‘real reality’ (werkelijke werkelijkheid) which would reach further than the actuality determined in virtue of the categories is not at all at issue. There is no further actuality (Wirklichkeit) than the actuality determined in judgement, nor is there an empirical reality that reaches further than the reality determined in judgement. Whatever more real reality might lie beyond knowable reality, i.e. the reality of things in themselves or a putative noumenal reality, lies beyond our ken. 

Whatever the case may be, Blomme believes that my argumentation is potentially circular. Why? According to Blomme, on my view Kant would ‘claim both (1) that Dinge an sich necessarily must exist because they are the cause of Erscheinungen and (2) that those Dinge an sich exist only insofar as they are Erscheinungen’ (2018:375). That is not what I have Kant claim, nowhere; and that’s just as well, for the fact that appearances exist only if things in themselves exist in no way implies that things in themselves exist only if appearances exist! God, for example, is such a thing in itself that can exist without existing as appearance (if he exists). Blomme recounts further: 

Why must we then according to Schulting say that the Ding an sich exists, rather than saying that we are necessitated to say, as a result of the way in which our understanding works, that if we think the concept <Ding an sich> on the basis of the concept <Erscheinung>, we think it [the thing in itself] in the first instance as existing? (2018:375)

Blomme commits an error here. Of course, the mere concept <Erscheinung> does not imply the existence of any specific thing x, y, z, but I nowhere made this claim. What is at issue is to determine that, in general, if there are appearances, then there must also be things in themselves; and there are appearances, therefore, there are things in themselves. (We leave aside here issues of the identity between appearances and things in themselves.) This holds also for a specific given appearance: if there is an appearance x, then there must be a thing in itself (or things in themselves) of which x is the appearance. In contrast to what Blomme suggests, we do not just think the thing in itself (or things in themselves) as existing, rather we must think it as existing if there are appearances or if there is an appearance x—which is the same as saying that the thing in itself of which x is an appearance, exists, because if something must be thought of as existing, then it exists. The existence of the thing in itself is independent of our thinking and it is the necessary condition of thought. If the thing in itself were only a thought-concept, but need not exist realiter, then appearances could not exist. What is more, if the thing in itself (as simple Being, as an ens realissimum, or as a plurality of things in themselves)[17] did not exist realiter, we would neither be able to think it. For the real possibility of our thinking rests on the reality of Being (this is very broadly taken Kant’s argument in the Transcendental Ideal, which is a critical translation of the argument that he already expounded in Der einzig mögliche Beweisgrund des Dasein Gottes).[18]

This does of course not mean that the mere concept of any arbitrary thing in itself T implies the existence of the thing in itself T or that in an arbitrary judgement S is P the x as the substrate of S is posited as existing already with the concepts. But in general it holds that things in themselves or a thing in itself must exist insofar as there is thinking at all. In particular, for any actual judging about an object the modality of the object as existing as a thing in itself is already posited with the judgement. To put this differently, it is not the judging per se that produces the existence, but the judging eo ipso implies the existence of the thing about which is being judged that so and so; that it is not judging itself that produces the existence of the object is manifested in the fact that a judging relies on the modal constraint of sensory intuition for real possibility. At any rate, Blomme cannot say in general that we only think a thing in itself and thereby need not be committed to its eo ipso existence. Let’s call such a claim P:

(P) We only think a thing in itself and thereby need not be eo ipso committed to the existence of things in themselves

Blomme cannot generally deny the existence of things in themselves (that are independent of his thinking) because at the very least he himself, as a thinker, must exist as an existing thing in itself in order to be able to think a thing in itself (i.e. to have a concept thereof), to be able to make the very claim P, or think anything, for that matter. And if he himself exists as thing in itself, then there must be Being, a thing in itself or things in themselves which is/are not dependent on thinking, but rather is/are the necessary condition of the possibility of thinking. 

Again, this does not detract from the fact that in an arbitrary judgement about a particular thing x the thing x cannot yet be proven to exist purely on the basis of the concepts that are employed in the judgement about x: either the thing’s existence cannot be proven empirically because no empirical intuition is or can be given (God, soul, etc.), or there is indeed an empirical intuition, and so there is an appearance that refers to the thing’s existence, and in the latter case the category ‘existence’ is justifiably applied in the judgement. 

Blomme does not appear to be wholly sensitive to the subtleties of my modified version of a metaphysical two-aspect reading of Kant’s idealism à la Lucy Allais.[19] An example. Blomme writes, quoting a passage from my book:

On p. 381 Schulting defends both a version of T1 and T2: ‘if we experience determinate appearances, things in themselves as their underlying ground must exist. Strictly speaking, it is not the thing in itself as such that we know exists, but that of which we determine in judgement that it exists. Existence (as a category) is first bestowed upon the thing by the determinative power of the understanding; more accurately, the category of existence is applied to the object of experience, the appearance, not to the thing in itself as such. But that does not mean that the thing so determined does not exist mind-independently, nor that it is not the mind-independent thing in itself only that is the denizen of Being.’ I wonder if the position defended in this citation can be coherent. (Blomme 2018:376, added underlining mine) 

I suppose it is the underlined sentence that caused Blomme’s confusion. Blomme’s reconstruction of my view falters with his premise B3: ‘We do not really know of the Ding an sich that it exists’ (Blomme 2018:376). This is not what I say or suggest. As is clear from the broader context (I talk about the so-called ‘two-aspect’ interpretation in relation to judgement), what I claim is that it is not the thing in itself as such of which we know that it exists, but that thing of which we determine, in an arbitrary judgement, that it exists, namely, as appearance of the same thing that also has an in-itself status. Nevertheless, the thing of which we know in a judgement that it exists, but of which we know only its appearance, exists (even: must exist) for sure as a thing in itself even if we do not know it as such, namely as regards its in-itself status (and so neither can we know its mode of existence as thing in itself). For the conclusion that the thing in itself exists if we determine its appearance as existing (by means of the category of existence) analytically follows from the fact that we determine that the x of our judgement—the thing of which we determine that it exists as appearance and is so and so—exists. One cannot determine the appearance as existing and deny the analytical implication that a thing in itself exists that is the necessary ground of the appearance. That would be a transcendental-logical contradiction, so to speak. <Thing in itself> and <appearance> are conceptual attributions of something that, respectively, necessarily has an existence in itself and an existence for us, that is, in the manner in which it appears (see Allais’s paradigmatic reading in her book). It looks like Blomme believes that only an appearance exists or can exist and that a thing in itself is merely an object of thought. I do not think such a reading plausible nor is it compatible with my reading.[20]

3. Nonconceptual Content

Van den Berg (hereafter cited as Van den Berg 2018) wonders what I, as a moderate conceptualist, think about the possibility that animals have objective perceptions of objects. Sacha Golob posed the same question in a recent discussion of my book. Van den Berg cites Colin McLear’s well-known paper on Kant and animal consciousness.[21] He refers to McLear by saying that ‘the conceptualist can take two positions with respect to non-human animals’.On the one hand, the conceptualist ‘can argue that non-human animals are conscious only of their sensations, which as a result of a lack of conceptual capacities are experienced as a blooming, buzzing confusion’, in the words of William James. ‘Or the conceptualist can argue that non-human animals do not have objective perceptual representations nor have sensory consciousness of objects’ (Van den Berg 2018:384).

In my view, these are not the exhaustive possibilities. For the moderate conceptualist reserves space for the possibility that animals have representations of objects in virtue of the fact that those objects affect their sensible capacities, and have in some sense consciousness of these because each representation as such must have an intensive magnitude which is >0, which corresponds to the intensity of consciousness. Golob formulated his critique of the Kantian conceptualist in such in a way that, because the conceptualist reserves objectivity or object-directedness or intentionality for category-regulated object-directedness—namely, intentionality which is a function of the categories of the understanding—he (the conceptualist) must deny that animals are object-directed.  

This makes it difficult, according to Golob, for the conceptualist to explain how it is possible that animals are in fact rather well directed towards the objects with which they interact. Golob mentions the excellent example of an eagle that, in its swoop and equipped with a highly developed eyesight, is precisely directed at its prey and is able accurately to differentiate it from the rock adjacent to it. According to Golob, this requires a highly accurate capacity for discriminating complex natural relations. But, the reasoning goes, the conceptualist reserves that capacity for discriminating complex natural relations for rational beings who comprehend those complex relations as such in virtue of the application of categories in judgements. Golob also thinks that my view on objective validity as merely a function of judgement runs the risk of reducing this capacity to rational beings, and that on my account animals must be denied the capacity for discriminating complex things. In such a perspective, for animals everything would be a ‘blooming, buzzing confusion’.

I do indeed deny that, as Golob sees it, animals are object-directed, which is however characterised by Golob as different from the category-governed directness by which human capacities is characterised. I deny this because object-directedness presupposes an explicit subjectivity, that is, a subjective perspective from which the subject herself relates, reflexively, to an object. If this reflexivity is lacking, such as with animals, then strictly speaking one cannot speak of object-directedness. Nonetheless doing so would in fact be an anthropomorphic projection. Perhaps it sounds plausible to speak of an animal’s perspective, but that is again from the perspective of how we make it intelligible to ourselves how an animal relates to its environment. The animal, by contrast, relates to its environment in an immediate fashion and exactly not in a reflexive relation in which it itself is conscious of its environment as distinct from itself. This absence of reflexivity, or subjectivity, does not prevent an animal from having creature consciousness, nor does it deny the fact that their objective environment sensibly affects them and that as spatiotemporal objects in nature and fully in line with the laws of nature they can perform complex actions and can react to complex structures and other objects. Of course much more could be said about this complex problematic, but unfortunately there is no space to do that here.

4. Hegel’s Critique of Kant

In his clearly formulated and interesting delineation Joris Spigt (hereafter cited as Spigt 2018) connects my account with the modern-sceptical problematic and I believe this can be very fruitful. For reasons of space, I would like to take this up on another occasion. Here, I want to respond to Spigt’s critical comments on my reading of Hegel’s critique of Kant. Spigt and I are more in agreement than he makes it seem. Spigt focuses on my reading of a text passage in Hegel’s Faith and Knowledge in which Hegel criticises Kant on a point that concerns the possibility of rational cognition. Let me quote that passage in its entirety:

…the Kantian philosophy has the merit to be idealism insofar as it shows that neither the concept alone, nor the intuition alone is something, and that intuition in itself is blind and the concept in itself is empty; and that neither the finite identity of both in consciousness, which is called experience, is a rational cognition. (translation mine)

In the quotation in my text, which Spigt cites, I use the phrase ‘rational cognition’ (vernünftige Erkenntnis) from the above-quoted passage in Hegel’s text, not to comment on Hegel’s remark, but to make it clear that Kant characterises the identity which lies in experience, not as rational cognition, but as a cognition by the understanding (Verstandserkenntnis), and that, according to Hegel, Kant is wrong about this. Spigt says that in the above quotation Hegel in fact says exactly the opposite of what I am saying, namely that Kant is quite right not to regard experience as rational cognition. My quotation of that phrase was somewhat taken out of context, i.e. the introductory pages of the section on Kant in Faith and Knowledge. Spigt’s formal criticism however ignores the context in which I thematised the general critique that Hegel raises against Kant in Faith and Knowledge. There, Hegel for sure criticises Kant’s view that the identity that lies in experience cannot be qualified as rational. Put succinctly—for a longer version, refer to my book—Hegel accuses Kant of not following through on his own reasoning with respect to the necessary original synthetic unity that lies in experience, in the connection between concept and intuition, in that he identifies that identity—which according to Hegel is an absolute one—as a mere mode of the understanding, and so demotes it to a relative identity between concept and intuition. 

In other words, rational cognition does not lie, according to Hegel, outside experience, as Kant himself believes, but rather in the relative identity between the understanding and the object, an identity established by the categories—however, not qua ‘finitude’, as Hegel says in the above-quoted passage. It is the absolute identity in the relative identity that points to infinite [rational] cognition. Like Spigt says: ‘an analysis of finite cognition must culminate in in-finite cognition. For this cognition is always already implicitly present in finite cognition’ (Spigt 2018:353). The understanding itself cannot see this and that is where the problem lies: according to Hegel, the understanding always thinks in terms of reflective conceptual oppositional pairs and does not see that those reflective oppositions should not be absolutised, but rather must be reflected back unto themselves. That is what rational cognition alone is capable of achieving. The inconsistency of intellectual cognition (as opposed to rational cognition) is that it does not consistently think through its own arbitrary absolutisation of the reflective oppositions. According to Hegel, Kant did allude to such a rational reflection with his notion of the original synthetic imagination, but as Kant, according to Hegel—and this is a correct interpretation of Kant (which De Boer for example misses)—sees the imagination merely as a function of the understanding, Kant misses the opportunity of seeing the rational (vernünftig) potential of the original synthetic unity of apperception. 

The inconsistency (in Kant’s thought) then lies, according to Hegel, in the fact that Kant disregards the absolute identity in his own reflections with respect to the synthetic a priori as the necessary condition of experience. On the contrary, Kant in fact squanders the implicitly present absolute identity by in the end choosing for the finite understanding as ‘der Pfahl des absoluten Gegensatzes’ (Faith and Knowledge, 4:323). Spigt says that rather than accusing Kant of inconsistency, Hegel shows that a potential is not put to good use. By contrast, I see those two as connected, for the unused potential can be demonstrated only internally, namely by showing the internal inconsistencies (or contradictions) in Kant’s logic regarding the necessary conditions of cognition. 



[1] Heidegger too thinks that the capacity of the imagination must not be reduced to being a function of the understanding. Heidegger’s reading rests, as is well-known, on the A Deduction. However, as I show in Chapter 6, the A Deduction need not be read in the way that Heidegger does. Hegel says mutatis mutandis the same, whose critique I address in detail in Chapter 8 of the book. Though I disagree with Hegel’s criticism, his interpretation of Kant, that is, that the imagination is merely a function of the understanding, is correct. De Boer’s critique of my view rests on her own presupposition that Kant himself makes a sharp, not merely formal distinction between the power of the imagination and the understanding, a reading that Hegel’s criticism of Kant’s identification of the two alone makes implausible. I suspect that on this point De Boer is led by her reading of Heidegger’s Kant interpretation.

[2] See especially Schulting (2017:77–8). My analysis there of the term Erkenntnis and how it relates to our contemporary concept of knowledge directly contradicts De Boers critique that I conflate ‘objective cognition’ and ‘empirical knowledge’, as she contends.

[3] For references see Schulting, Kant’s Deduction From Apperception. An Essay on the Transcendental Deduction of the Categories (Berlin: De Gruyter, 2018), ch. 4.

[4] I follow Henry Allison’s so-called ‘Reciprocity thesis’ for the Deduction. See Schulting (2018).

[5] See also the discussion in Schulting (2017), ch. 7.

[6] See again the discussion between me and Anil Gomes, op. cit.

[7] Robert Hanna, ‘The Togetherness Principle, Kant’s Conceptualism, and Kant’s Non-Conceptualism’, The Stanford Encyclopedia of Philosophy (Fall 2013 edition), ed. Edward N. Zalta,

[8] See e.g. Hegel’s Faith and Knowledge, which I discuss extensively in Schulting (2017), ch. 8. One could say that Hegel in that sense, because of the semblance of a ‘facultative’ classification urge, rightly accuses Kant of psychologism or at least the appearance of psychologism, which he inherits not only from Locke but also from Wolff and Baumgarten. See Schulting (2017:348). See also my reply to Spigt further below.

[9] In his commentary, Blomme says that I ‘make it appear as if [my views] are shared by only a marginal minority, such as the thesis that the categories express the unity of the original apperception and thus can, according to Kant, in one way or the other be derived from it’ (Blomme 2018:365). Well, that last thing is precisely a standpoint that is shared by exactly none of my English speaking interlocutors! I’m pleased with Blomme’s support in this, but I am afraid that this will remain one of those issues which will separate the classical systematic-historical Kant scholarship from the broadly analytical approach. Notice, by the way, that the thesis that all twelve categories are in fact derived in the Transcendental Deduction is also not delineated by important German scholars such as Manfred Baum, Klaus Reich and Michael Wolff, though they are sympathetic to it (this holds certainly for Klaus Reich, whose derivation of the logical functions of judgement was the model for my own interpretation of the derivation of the categories).

[10] De Boer further says that ‘I am rather vague about the question about where exactly our empirical judgements get their objectivity’ (De Boer 2018:360). I find this surprising, since, in Chapters 3 and 4, I extensively discuss the connection between objective validity (objectivity) and judgement. The fact that my reading did not convince her could be due to the fact she has not really captured the radicality of my interpretation of the role of judgement, and what this says about objective validity, truth and correspondence.

[11] That last qualification ‘insofar as the underlying intuition is determined’ does not imply, as Blomme suggests, that the categories and the functions of judgement are not identical.

[12] I cannot concur with Blomme’s view that the concept of an object in general in its ‘minimal sense’ is the ‘unity of a conscious representation’ (Blomme 2018:367–8, 370). That would imply that any arbitrary representation or series of representations that occur is by definition an objective unity of apperception. This is philosophically as well as interpretatively not a defensible position (see Schulting 2017, ch. 4; cf. 2018, ch. 4).

[13] That this is what Blomme appears to mean is suggested by his phrase ‘intersubjectively certain factualness (the being-there)’ (Blomme 2018:368), which he associates with the objectivity of judgements.

[14] This confusion is increased by the fact that Blomme later on speaks of objectively valid ‘judgements about Gegenstände überhaupt […], in which the categories thus necessarily find a certain realisation’, but which are ‘not automatically objective cognitive judgements’ (Blomme 2018:370). Blomme reads my interpretation with glasses on through which I cannot read.

[15] Sometimes a distinction is made between ‘objective validity’ and ‘objective reality’, but Kant often conflates the two terms.

[16] James Conant, ‘Why Kant is Not a Kantian’, Philosophical Topics 44(1) (2016): 75–125.

[17] Kant cannot commit himself to either the plurality or simplicity of the existence of a thing in itself or things in themselves, because he would then go beyond the bounds of possible experience.

[18] See further Schulting (2017), ch. 9.

[19] See Lucy Allais, Manifest Reality. Kant’s Idealism and His Realism (Oxford: OUP, 2015).

[20] Premise B4 in Blomme’s argument (see Blomme 2018:376) is in any case an incorrect interpretation of my reading. On my phenomenalistic account, an appearance can never exist independently of us.

[21] Colin McLear, ‘Kant on Animal Consciousness’, Philosophers’ Imprint vol. 11 (15) (2011).