Escribir el título del artículo


Matters of Taste: Kant’s Epistemological Aesthetics


Zoltán Papp·

Eötvös Loránd University (ELTE), Hungary



This paper is concerned with what I believe is the epistemological mission of Kant’s doctrine of taste. The third Critique inherits two problems from the first. The evident one is that the categorial constitution of nature must be complemented with the notion of purposiveness. The less evident one is that the transcendental theory of experience needs a common sense in order to secure a common objectivity. The judgment of taste, conceived of by Kant as a ‘cognition in general’ not restricted to either the particular subject or the particular object, offers a solution. It turns out to be a judgment that cannot be made without assuming the purposiveness of nature and the uniformity of the cognizing subjects.



Epistemology, Purposiveness, Solipsism, Common Sense


1. Introduction

In §21 of the Critique of the Power of Judgment (CPJ), Immanuel Kant first states that empirical cognitions must “be able to be universally communicated, for otherwise they would have no correspondence with the object”, then concludes that a “common sense […] must be able to be assumed […], as the necessary condition of the universal communicability of our cognition” (5: 238-39).[1] This is puzzling both because Kant unexpectedly adds a new item to the list of the conditions of experience – in the Critique of Pure Reason (CPR) there is not even a hint that common sense has such a role to play –, and because common sense otherwise figures as a principle of taste. There is a similar double puzzle in the two Introductions. The transcendental principle of nature’s purposiveness, though basically identical in its function with what was a regulative idea of reason in the CPR, is still markedly new in that it now belongs to the reflecting power of judgment, itself a newcomer to the elite group of the higher cognitive faculties. And the principle, though originally presented with an eye to securing the coherence of experience, does not remain restricted to this cognitive use but serves as the a priori ground of the judgments of taste, too. Correspondence and coherence being the complementary criteria of empirical truth, the two puzzles might add up to the intuition that Kant’s taste has an epistemological mission.

The relation between aesthetic judging and empirical cognition has become a chief focus of Kant exegesis. But commentators sharply disagree over how exactly his account of the judging of the beautiful as a mental act is to be interpreted in order for it to be(come) a viable account of aesthetic experience. My paper does not take sides in this debate, nor does it offer yet another interpretation.[2] Partly because I could not say anything new, partly because what my paper does offer is more or less neutral with respect to those differences. And chiefly because I have a conjecture that I cannot, of course, prove: most of the difficulties with Kant’s doctrine of taste (such as the so-called ‘everything is beautiful’ problem) stem from its being intended both as a theory of aesthetic judging and as the completion of a twofold epistemological job that remained unfinished in the CPR. What I can do is try to explain why I think that this doctrine can be read as the continuation of an epistemological project (and why I cannot contribute to solving those difficulties as ones of an aesthetic theory proper).

In Section 2, I show how the CPJ’s Introductions transform the principle of nature’s purposiveness, which Kant first tries – but fails – to establish towards the end of the CPR. As a principle of cognition, purposiveness remains regulative. As a principle of taste, however, it gets not only a new function but a new status, too, becoming the sine qua non of judgments of taste, a point I will only indicate at the end of the section. Section 3 is an extended commentary on a perhaps surprising passage of the CPR, which suggests that transcendental idealism cannot avoid a kind of solipsism. I will argue that this can be seen as truly following from Kant’s position: objective truth becomes contingent, because different subjects can ‘make’ objects out of their perceptions in different ways. My main argument comes in Section 4. It is based on the simple idea of taking seriously Kant’s claim that aesthetic judging, as a “cognition in general” (5: 217 and passim), is the condition of objective cognition. The solipsism problem emerging in the CPR calls for an epistemic common sense, which, however, cannot be established within the theory of experience itself. It is only judgments of taste – as involving a “cognition in general” – that, since their very possibility entails the claim to universal assent, make it necessary to assume the existence of a cognitive constitution common to “all subjects” (5: 224) and operative in “all cognition” (5: 219). Analogously, the principle of nature’s purposiveness remains too weak until there appears on the scene a type of judgment, that of taste, which, since its universal validity requires an a priori ground, cannot be made without presupposing that the whole of nature is purposive. I will finish with some critical remarks concerning the notion of “cognition in general” in order to highlight a shortcoming of the doctrine of taste, which I think results from its epistemological commitment, and which Kant himself seems to realize in his doctrine of art.

2. Purposiveness 2.0

Contrary to what its title suggests, the CPR’s Appendix to the Transcendental Dialectic is by no means of secondary importance compared to what has already been settled.


If among the appearances offering themselves to us there were such a great variety – I will not say of form (for they might be similar to one another in that) but of content, i.e., regarding the manifoldness of existing beings – that even the most acute human understanding, through comparison of one with another, could not detect the least similarity (a case which can at least [or well: wohl] be thought), then the logical law of genera would not obtain at all, no concept of a genus, nor any other universal concept, indeed no understanding at all would obtain, since it is the understanding that has to do with such concepts. The logical principle of genera therefore presupposes a transcendental one if it is to be applied to nature […]. According to that principle, sameness of kind is necessarily presupposed in the manifold of a possible experience […], because without it no empirical concepts and hence no experience would be possible. (A653-54/B681-82, my italics)


Experience is at stake again; or still. The Analytic did only half the job by demonstrating that the appearances are “similar” in their “form”, the categories, these pure concepts being the a priori synthetic functions that unite representations in one consciousness. As mere forms, however, they do not determine the particular “content” of experience. Kant’s initial formulations, in the first six pages of the Appendix, suggest that the regulative use of pure reason is aimed at a post factum systematic unity of empirical cognitions. The passage just quoted reveals a more profound dimension. While it is partly about something trivial – empirical concepts presuppose regular similarities among things –, it also sets a daunting task: the Appendix should repeat, mutatis mutandis, the feat of the Analytic, i.e., transform the logical structure of thinking, this time its rational systematicity, into the transcendental structure of nature.

Kant presents what he wants to establish in different but compatible versions. One of them is a triad of principles: “sameness of kind” or “homogeneity”, “variety”/“specification”, and “affinity”/“continuity” (A657-58/B685-86), where the third principle mediates between the first two. Another version is that “the highest systematic unity” of experience can be attained “by means of the idea of the purposive causality of the supreme cause of the world” (A688/B716). Kant first tries to introduce affinity in the A-Deduction, there as a transcendental principle underlying the association of perceptions. The examples he gives, though somewhat fantastic, show that it is basically the stability of natural kinds that needs to be secured. “If cinnabar were now red, now black, now light, now heavy, if a human being were now changed into this animal shape, now into that one, if on the longest day the land were covered now with fruits, now with ice and snow, then my empirical imagination” could not connect these perceptions together (A101-02). Once you identify a thing, on the basis of some its perceivable features, as falling under a certain concept, the rest of its features also must display the marks and obey the rules included in that concept. Kant believes that affinity, as the “objective ground” that makes appearances “associable” (A121-22), follows from the transcendental unity or identity of the apperception (A113-14, A122-24). However, as he himself emphasizes in the same passages, the unity of the thinking ‘I’ is strictly nothing more than numerical. It necessarily manifests itself through the categories, as the formal functions of synthesis, but that is all. Just as the ‘I’ can comprehend “the unity […] of a fable” (B114), so too can it synthesize representations in a nice categorial syntax on the one hand, but in a semantically incoherent manner on the other. Of course, if I witness a fable-like event in what I otherwise believe is a well-ordered nature, I will think twice before I synthesize my perceptions in a public utterance meant to be an objective judgment of experience. But I can do so without falling into transcendental schizophrenia.

To return to the Appendix, the invocation of the “case” that Kant says “can well be thought” – nature as a totality of radically individual entities without any generalizable similarities whatsoever – is misleading. This extreme chaos, but only this one, would indeed make all experience impossible. The Appendix should be concerned with and about the more moderate (and easier-to-think) case in which the order of nature and the stability of empirical concepts are merely relative – in which there are experiences, in plural, but there is no unified experience. The Analytic demonstrates that the correspondence criterion of empirical truth, “the agreement of cognition with its object” (A58/B82), cannot be met without “transcendental truth, which precedes all empirical truth and makes it possible” (A146/B185), and which takes shape in the categories and the synthetic a priori principles built on them. The Appendix adds to this a second and distinct criterion, that of coherence: “the systematic connection that reason can give to the empirical use of the understanding […] guarantees its correctness” (A680/B708). A judgment of experience is true if its conceptual components express a right perceptual and verbal identification of the intuitive content and if it fits into the coherent body of empirical knowledge (for the difference between words and concepts, see A728/B756). Ideally (and normally), the two criteria are fulfilled at once. But the categories being mere forms, the contentual order of nature and the consequent overall coherence of experience are not already given with, but complementary to, the legislation of the pure understanding. Where can they be derived from?

Unsurprisingly, the Appendix does not use the apperception argument, which did not work in the A-Deduction. Nor does it have any better argument, however. Kant begins by claiming that what is needed is “a transcendental principle of reason” that “make[s] systematic unity not merely something subjectively and logically necessary, as method, but objectively necessary” (A648/B676). But he then implements, overtly contradicting himself, a string of restrictions that drastically diminish the efficacy of the principle. He calls it a “merely regulative principle or maxim”, a “principium vagum” (A680/B708), a “subjective principle” (A666/B694), a “method” (A668/B696), and he famously uses an “as if” as a leitmotif in the second part of the Appendix: “the speculative interest of reason makes it necessary to regard every ordinance in the world as if it had sprouted from the intention of a highest reason” (A686/B714). The dilemma is clear and unresolvable. On the one hand, only a constitutive principle would be strong enough to impose systematic order on nature, yet it falls under a critical ban, as it would either require the speculative certainty of the existence of God – which Kant has just annulled –, or, in terms of the triad of principles, it would have to rest on the knowledge that all possible objects of experience are interconnected in a universal kinship that makes all manifoldness the diversification of an original unity; from the human point of view of discursive cognition, this would translate into a pyramidal order, with the super-concept of ‘something’ at the tip and the individuals at the bottom, the interim layers representing the species, genera, families, etc. On the other hand, a regulative principle, while critically tenable, is simply too weak to ensure order in nature. In the Ideal chapter, which already contains much of what the Appendix develops in more detail, Kant even calls the “as if” principle a “merely heuristic” one (A616/B644). A heuresis does not entail that it will prove true when applied.[3]

Of course, Kant can say that what makes an experience an experience is that it fits into a system. But, with God in brackets, and the unity of apperception being such that it does not exclude the possibility of logically incoherent empirical syntheses, he cannot transfer this conception of experience onto “the appearances offering themselves to us”. It is not just that logically incoherent judgments of experience are possible. ‘Normal’ ones can always be made, too, without having to rely on the principle that the whole of possible experience is coherent. Yet the Appendix indicates, negatively, a solution. Reason, by its very nature, strives for the systematic unity of empirical cognitions, but, in contrast with the categories, its essential systematicity does not count as an a priori condition under which alone empirical judgments are possible, because it belongs to a faculty which is not involved in ‘primary experience’, in the very act of bringing intuitions to concepts: “pure reason leaves to the understanding everything that relates directly to objects of intuition or rather to their synthesis in imagination” (A326/B382-83). All I have said is not, of course, meant to question the utility of the regulative maxims in natural science. But this is a far cry from their transcendental validity. The latter would require a kind of ‘primary experience’ that cannot be made without a principle of systematicity.

“[F]or all things in nature empirically determinate concepts can be found”: this the first – and most compact – version of “the principle of reflection on given objects of nature” in the CPJ (20: 211). Remarkably, the Introductions do not say a single word about how the new project relates to the old one. Is it a supplement? An alternative? A replacement? I take it to be the third, although Kant mentions the regulative principles and their empirical use in the short Preface (5: 167-68). What is certain is that the two projects are very similar and very different at the same time. Kant adds a long footnote to that first formulation, stressing that while the principle may seem “to be tautological and to belong to mere logic”, it in fact is “a synthetic and transcendental proposition”, for logic as such “teaches us nothing” about


whether for each object nature has many others to put forth as objects of comparison, which have much in common with the first in their form […]; rather, this condition of the possibility of the application of logic to nature is a principle of the representation of nature as a system for our power of judgment, in which the manifold, divided into genera and species, makes it possible to bring all the natural forms that are forthcoming to concepts (of greater or lesser generality) through comparison. (20: 211-12n)


As can be seen (from several other passages, too), the core problem has not changed. Concepts, rules, laws require comparability, which, in turn, requires the pyramidal order of nature: “the comprehensibility of nature and the unity in its division into genera and species [is that] by means of which alone empirical concepts are possible” (5: 187), possible in the ‘strong’ sense that they meet the criterion of coherence. Objects can “be causes in infinitely many ways”, but “each of these ways must […] have its rule” or “law” (5:183): an empirical judgment referring to whatever particular causal occurrence must also represent a rule or law and must thereby meet the criterion of coherence. What has changed, however, is that Kant has apparently realized that nature is to be alienated from logic before it can be reconquered transcendentally. In a similar spirit, he keeps emphasizing that experience begins with intuitions and perceptions (see, e.g., 20: 213, 5: 186, and see as well the splendid description, in §77 of the Teleology, of how discursive cognition works amidst contingency). Also, Kant writes more explicitly and at much more length about what is otherwise trivial, namely the insufficiency of the categories to determine the contentual order of nature. And while he imports various terms from the Appendix of the CPR (such as specification and affinity), the new approach is a full-blown theory under the heading of purposiveness. As these and other points have already been commented on extensively, I do not go further into the details here.[4]

The most spectacular change is coupled with a no less evident continuity, however. The most obvious explanation for why purposiveness becomes a transcendental principle of the reflecting power of judgment is that this faculty is directly involved in empirical cognition, whereas reason is not. But this explanation is not yet sufficient regarding the fact that the Introductions repeat the old restrictions. Kant defines a “transcendental principle” as one that represents “the universal a priori condition under which alone things can become objects of our cognition at all” (5: 181). However, the principle remains “subjective” (5: 185 and passim), “regulative” (20: 151, 5: 197), a “maxim” (20: 205, 5: 184), “a heuristic principle” (20: 205); the second Introduction brings back the theological “as if’” motif (5: 180), and Kant even adds that the principle stems from a human “need” (20: 214, 5: 186, and passim), which is the equivalent of what was “the interest reason” in the CPR.

Does all this mean that the Introductions simply substitute one faculty for another, but otherwise leave the validity and the potency of the purposiveness principle unchanged? It seems that they are compelled to do so. After all, the most important factor also has not changed. “The concept of purposiveness is not a constitutive concept of experience at all […]; for it is not a category” (20: 219-20). If so, however, then it has not gained anything in status. Indeed, it is not even clear what the point is in formulating and especially in calling transcendental a principle which the power of judgment “prescribes […], not to nature (as autonomy), but to itself (as heautonomy) for reflection on nature” (5: 185). Unlike reason, the reflecting power of judgment is the chief agent of discursive cognition. Why does discursive cognition have to order itself to operate according to its own structure?

The first key to seeing how the new project works is to understand what it means that it takes place in a new critique. “Philosophy can be divided into only two parts, the theoretical and the practical”, and within this system “everything that we might have to say about the proper principles of the power of judgment must be counted as belonging to the theoretical part” (5: 179). The critical system, however, reflects Kant’s triadic conception of the mental household. The first two Critiques having laid down the “a priori principles” for “the faculty of cognition” and “the faculty of desire”, respectively,


there remains among the properties of mind in general an intermediate faculty or receptivity, namely the feeling of pleasure and displeasure, just as there remains among the higher faculties of cognition an intermediate one, the power of judgment. What is more natural than to suspect that the latter will also contain a priori principles for the former? (20: 207-08, cf. 5: 176-79)


The intermediate position of both the pleasure and the power of judgment is a somewhat arbitrary idea of Kant, but this is not so important here. His argument for wedding them is that both are subjective:


while in the division of faculties of cognition through concepts understanding and reason relate their representations to objects, in order to acquire concepts of them, the power of judgment is related solely to the subject and does not produce any concepts of objects for itself alone [für sich allein]. Likewise, […] the feeling of pleasure and displeasure is only the receptivity of a determination of the subject, so that if the power of judgment is to determine anything for itself alone, it could not be anything other than the feeling of pleasure. (20: 208)


This ‘itselfness’ of the reflecting power of judgment, namely, that “for itself alone” it “does not produce any concepts of objects”, is a crucial motif. The Introductions mostly describe it as a faculty that “seeks concepts to empirical representations” (20: 212n) and applies a principle for the sake of this search. But it turns out that this whole cognition business is not an adequate expression of its true nature. According to its true nature, “the power of judgment” is a “faculty” which “serves only for connecting and which hence cannot provide any cognition of its own [für sich]” (20: 246). And the most basic form of this “connecting” links exactly those two faculties that were left without the direct support of pure reason in the CPR. “The power of judgment […] considered by itself […], as a separate power of cognition, considers only two faculties, imagination and understanding”, not in an act of conceptual object-cognition, to be sure – that would be contrary to its true nature –, but “as in relation in a representation prior to any concept” (20: 233), i.e., in a pre-conceptual aesthetic judging.

Thus the par excellence mental act that the power of judgment performs or orchestrates “for itself alone” and without “produc[ing] any concepts” is nothing else than its working as taste. And the pleasure it “is to determine” is the pleasure in the beautiful. From this it follows that it is in its purest form, as taste, that the power of judgment can join the exclusive club of the higher cognitive faculties: “by the aesthetic power of judgment as a special faculty necessarily nothing else can be meant than the reflecting power of judgment (20: 249). For most of the Introductions, the a priori legislation of the power of judgment appears to consist in applying purposiveness as a principle of cognition (PPC) to nature; this is the best developed idea in both texts. But it turns out that with PPC alone the power of judgment does not yet qualify as a higher faculty. “[W]hat [viz. the critique of the power of judgment] cannot enter into the division of philosophy can nevertheless enter as a major part into the critique of the pure faculty of cognition in general if, namely, it contains principles that are for themselves fit neither for theoretical nor for practical use” (5: 176, my italics). PPC is perfectly fit for theoretical use; indeed, it is unfit for anything else. The twist is that the purposiveness principle has a higher function as well, serving as a principle of taste (PPT). A judgment of taste


wins a claim to universality and necessity, as merely reflective judgment, through the relation of the subjective purposiveness of the given representation for the power of judgment to that a priori principle of the power of judgment, of the purposiveness of nature in its empirical lawfulness in general, and thus an aesthetic reflecting judgment can be regarded as resting on a principle a priori (although it is not determining), and the power of judgment in it can be justified in finding a place in the critique of the higher pure faculties of cognition. (20: 243)


These higher “faculties are called pure because they are legislative a priori (5: 179). So whereas PPC as such could belong to the theoretical part of philosophy, PPT endows purposiveness with the dignity of a truly a priori principle, by becoming – as I will show in more detail in Section 4 – the sine qua non of a certain type of judgment, or, if I may use the expression in an extended sense, of a kind of ‘primary experience’: aesthetic experience.

3. Touch/stone

In the Canon of Pure Reason, Kant makes a distinction between conviction and persuasion. The latter is merely subjective.


Truth, however, rests upon agreement with the object, with regard to which, consequently, the judgments of every understanding must agree […]. The touchstone of whether taking something to be true [des Fürwahrhaltens] is conviction or mere persuasion is therefore, externally, the possibility of communicating it and finding it to be valid for the reason of every human being to take it to be true; for in that case there is at least a presumption that the ground of the agreement of all judgments, regardless of the difference among the subjects, rests on a common ground, namely the object. (A820-21/B848-49)


The second sentence would be a bit more streamlined if there were a single word in English for Fürwahrhalten. But even so, it sounds overcomplicated. Why does Kant not simply say that the external “touchstone” of whether a judgment is true is the object itself, and that the agreement of the subjects rests on this “common ground”? Why does he (have to) say that even the “presumption” of there being a common object requires the consensus “of every human being”? Transcendental idealism (TI) seems to have an unpleasant side effect: a kind of solipsism.

Far be it from me to pretend that I know what TI exactly is. Nor do I want to deny that what the above passage implies, namely, that objective truth becomes a matter of intersubjective consensus, is not exactly a commonplace of critical philosophy. But the passage is there, and it is too elaborate to be dismissed as a momentary confusion. So I would like to point out that it is not impossible to read it as Kant’s response to a problem he can think of as inherent in TI.

As far as I know, Kant nowhere uses the word ‘solipsism’ in epistemological context. But he does use a synonym term: ‘egoism’. Although, and this might be of some interest, the Anthropology from a Pragmatic Point of View is his single published work in which he mentions egoism, otherwise he deals with it only in his university lectures. In the Metaphysik Mrongovius, dating from 1782–83, he characterizes it as follows:


Egoism is when someone maintains that there is nothing present outside him, but rather everything that we see is mere illusion; and whoever maintains this is an egoist. Egoism can be dogmatic or skeptical. Many have maintained skepticism in earnest, and that is feasible if one maintains namely that all grounds to the contrary are not yet adequate. The egoist says: in dreaming I also imagine a world, and am in it, and nevertheless it is not so. Can it not also be the same with me when awake? But against this is that dreams do not connect with each other, rather I now dream this, now that, but when awake appearances are connected according to general rules. […] I cannot refute the egoist by experience, for this instructs us immediately only of our own existence. We do experience mediately that other things are there through the senses; but the egoist says that in these senses there lies only the ground by which we would become aware of appearances. But they would be nothing in themselves. (Kant, 1997, 29: 927)


Apparently, TI and egoism overlap in a crucial respect. That what we “become aware of” are “appearances”, “nothing in themselves” for us, is something on which Kant agrees with the egoist: “objects […] are nothing in themselves without [the] subjective conditions” of intuition, space and time, “in relation to which therefore all objects are mere appearances” (A48-9/B66).[5]

An important distinction must be made with respect to what kind of solipsism TI leads to. On the one hand, Kant’s argument that self-consciousness entails the consciousness of things existing outside the mind – presented first in the Criticism of the Fourth Paralogism, then, in the B-edition, as the Refutation of Idealism – makes TI the opposite of solipsism. The “transcendental idealist” is “a dualist”, because he is “an empirical realist” at the same time (A370). On the other hand, however, the necessary corollary of self-consciousness is not, properly speaking, a collection of objects, a ready-cooked objectivity, but only “something persistent in perception” (B275), “something real in space […], or the material of all objects of outer intuition”, or simply “Something”; and “it is perception through which the material must first be given for thinking objects of sensible intuition” (A373-75). Objects of cognition are a next step, so to speak, they are made out of this perceptual material. But how does this making happen if “space itself with all its appearances, as representations, is only in me” (A375)?


Now cognition of objects can be generated from perceptions, either through a mere play of imagination or by means of experience. And then of course there can arise deceptive representations, to which objects do not correspond, and where the deception is sometimes to be attributed to a semblance of the imagination (in dreams), sometimes to a false step of judgment (in the case of so-called sense-deceptions). In order to avoid the false illusion here, one proceeds according to the rule: Whatever is connected with a perception according to empirical laws, is actual. (A376)


This is one version of the dream argument, which appears in the Mrongovius passage, too, and which Kant repeats several times (see, e.g., A112, A492/B520-21, and Proleg. [Kant 2002] 4: 290). But it does not work. How do I know in advance what is in accord with “empirical laws”? All I know is that certain empirical laws have been valid up to now. Am I entitled to say that something which I perceive is not “actual” simply because it contradicts what I have got accustomed to?

In the proof of the Second Analogy, Kant explains how the subjective succession of perceptions becomes transformed into an objective causal relation:


One quickly sees that, since the agreement of cognition with the object is truth, only the formal conditions of empirical truth can be inquired after here, and appearance, in contradistinction to the representations of apprehension, can thereby only be represented as the object that is distinct from them [im Gegenverhältnis (…) könne vorgestellt werden] if it stands under a rule that distinguishes it from every other apprehension, and makes one way of combining the manifold necessary. That in the appearance which contains the condition of this necessary rule of apprehension is the object. (A191/B236)


This passage is extremely important for two (related) reasons. First, it describes the birth of the object, the Gegenstand, in a way which could reconcile the two poles that make TI so difficult to understand, namely, that the appearances are both outside and inside the mind: in objectivizing my perceptions, I externalize my representations, oppose them to myself (in the sense of obicio). Second, this operation promises to create a common object, one which is “distinguishe[d] from every other apprehension”, too. But Kant’s line of thought is circular again. The “object” is supposed to be the result. How can it already serve as the “necessary rule of apprehension”? Or does Kant mean something like a “formal” object?

In the Prolegomena, Kant develops a complete doctrine, that of judgments of perception and judgments of experience, in order to show how the (dynamical) categories elevate the subjective associations of perceptions to objective syntheses. Whereas the former “do not require a pure concept of the understanding, but only the logical connection of perceptions in a thinking subject”, the latter


always demand […] special concepts originally generated in the understanding, which are precisely what make the judgment of experience objectively valid.

All of our judgments are at first mere judgments of perception; they hold only for us, i.e., for our subject, and only afterwards do we give them a new relation, namely to an object, and intend that the judgment should also be valid at all times for us and for everyone else; for if a judgment agrees with an object, then all judgments of the same object must also agree with one another […]. (Proleg. 4: 298)


It can be seen, first, that Kant is definitely concerned with the change from subjective to objective and common validity. But it can also be seen, second, that he fails to give a sufficient account of it. The requirement that there be, in the case of judgments of perception, a “logical connection of perceptions in a thinking subject” begs part of the question, because it presupposes that the subjective association of perceptions kindly anticipates a ‘normal’ conceptual relation. This makes the change from subjective to objective validity a smooth transition. But it is not clear what the categories add to experience if their use must be preceded by the “logical connection of perceptions”, or if that which a “judgment of experience” goes “beyond” is “the sensory intuition and its logical connection (in accordance with which the intuition has been rendered universal through comparison in a judgment)” (Proleg. 4: 304). And, more importantly here, it is just as unclear what makes this “logical connection” and universalization possible, and not simply possible, but possible in a way such that the resulting object is the same for everyone.

Is it the “material of all objects of outer intuition”? If so, then this material must already be conceived of as a structured and binding objectivity, which would amount to a massive realism hardly consistent with TI.


[C]onsider the following: If the sun shines on the stone, it becomes warm. This judgment is a mere judgment of perception and contains no necessity, however often I and others also have perceived this; the perceptions are only usually found so conjoined. But if I say: the sun warms the stone, then beyond the perception is added the understanding’s concept of cause, which connects necessarily the concept of sunshine with that of heat, and the synthetic judgment becomes necessarily universally valid, hence objective, and changes from a perception into experience. (Proleg. 4: 301n)


But what if I do not perceive that the stone “becomes warm”, while someone else does? What could rule out this possibility and the ensuing headache of conflicting judgments of experience? Both of us rely on our respective perceptions, both of us are able to objectivize them in accordance with the numerical unity of the apperception, and neither of us is able to touch the stone (let alone the sun) as a thing in itself, independently of its being perceptually represented.

What the first person plural in phrases like “representations in us” (A371) refers to is not a single collective ego, a super-subject, but a multitude of individual subjects. While the categories delineate the universal structure of the human understanding as such (or so Kant believes), sensing and perceiving individuate “us”. One of the more than few necessities not included in the formal legislation of the pure understanding is that anything that shows some traits of being a stone must get warm in sunshine. A “stone”, say, a “granite […] might differ in its internal constitution from every other stone which nevertheless look[s] just like it” (20: 216n). As is obvious, the problems discussed in the preceding and the present section are closely linked. What I called a kind of solipsism would diminish to a merely practical difficulty if TI included the certainty of the contentual regularity of the appearances. There would still be conflicting judgments, but disputes could be settled based on constant “empirical laws”; it would be possible to determine which judgment is correct, which perception is veridical, and which object is “actual”. “[S]pace is nothing other than a mere representation, hence only what is represented in it can count as real, and conversely, what is given in it, i.e., represented through perception, is also real in it” (A375-76). This could serve as a motto for TI. But, due to the equation of being real with being perceptually represented, neither the process nor the theory of the making of objects can circumvent the individual minds and their ability to objectivize the “Something” they perceive in different ways.

It is this problem that the passage quoted above from the Canon chapter seems to respond to: by making objective truth a function of intersubjectivity. The “object” ceases to be the external “touchstone” of Fürwahrhalten, the “common ground” that determines how everyone must perceive and judge it. Everyone has their own object, because what they posit as the object and what their judgment refers to is nothing but the objectivization of their representations. In the absence of a commonly accessible object as such, the truth of an empirical judgment becomes dependent on “the possibility of communicating it and finding it to be valid for the reason of every human being”; i.e., that which counts as object is the common reference, not of a common cognition, but of a preceding universal consensus. That is why, strictly speaking, this consensus yields no more than just “a presumption that the ground of the agreement of all judgments, regardless of the difference among the subjects, rests on a common ground, namely the object”.

But, alas, even the “presumption” is hopelessly distant. Suppose that I somehow revise my perception and come to agree with my fellow judger that the sun warms the stone. In order to be able to at least presume that our agreement rests on the object, we should reach out to “every human being”. Quite a task, not least because some human beings are already dead, while others are yet to be born. And even if we managed to poll all of them and find a truly universal consensus concerning the stone, this is just one out of, well, many objects (or object-wannabes). The “possibility” of actually communicating empirical judgments to everyone is actually an impossibility. Critical philosophy needs to find a solution more viable than that doubly endless poll. Perhaps it can try to relocate the “common ground” that used to be the object into the subjects.


The universal communicability of the sensation (of satisfaction or dissatisfaction), and indeed one that occurs without concepts, the unanimity, so far as possible, of all times and peoples about this feeling in the representation of certain objects: although weak and hardly sufficient for conjecture, this is the empirical criterion of the derivation of a taste, confirmed by examples, from the common ground, deeply buried in all human beings, of unanimity in the judging of the forms [der Formen] under which objects are given to them. (5: 232-33, the definite article before “forms” is my insertion)


Though not the easiest sentence ever written by Kant, it seems to say that the aesthetic unanimity that can be detected empirically by looking at the “examples” does not as such prove the existence of “the common ground”. And also, that this ground has a role broader than just facilitating aesthetic consensus: it provides for the “unanimity in the judging of the forms under which objects are given”. Not “certain objects”, not beautiful things – objects. §17, which the quote is from, is the last section of the Third Moment. Still, its “common ground” probably has something in common with that of the Fourth Moment: in making a judgment of taste, “one solicits assent from everyone else because one has a ground for it that is common to all” (5: 237). This common ground, here already called common sense, is the medium of the universal communicability of a tasteful mental state which, in turn, is

4. “Suited to cognition in general”

“Cognition in general” is a central notion of the doctrine taste (‘doctrine of taste’ refers to §§ 1-22 and 30-40 as well as to the relevant sections of the Introductions). Either this term or one of its derivatives appears almost every time when Kant attempts to describe aesthetic judging. I say ‘attempts’ because his formulations are perplexingly vague and diverse; they sound like variations on a missing theme. “Cognition in general” first occurs in §9, but its entrée is less than amazing. If a judgment of taste is to be universally valid, it must have to do with cognition, but if its “universal communicability […] is to be conceived of merely subjectively, namely without a concept of the object”, it must express “a state of mind that is encountered in the relation of the powers of representation to each other insofar as they relate a given representation to cognition in general” (5: 217). Why equate subjectivity with generality? Why equate a mental act that does not yield a conceptual determination of its particular content and does not make it an object with the lack of particularity?[6] Kant goes on to add that the imagination and the understanding enter a “subjective relation suited to cognition in general”, and “any determinate cognition […] always rests on that relation as its subjective condition” (5: 218).

As an important (though isolated) observation in §18 shows, Kant is aware that taste has no principle proper. A judgment of taste is the “example of a universal rule that one cannot produce [die nicht gegeben werden kann]” (5: 237). Aesthetic judging cannot be formalized, it can only be grasped in its particular instances (the same pertains to the power of judgment: this is why general logic could not give it precepts in the CPR, A132-35/B171-74). But the critique of taste is written “from a transcendental point of view [or with a transcendental intention: in transzendentaler Absicht]” (5: 170), which means that it must transcend the particular instances and “produce” a rule. Kant finds a way out of this situation by making the instance the rule: by making aesthetic judging a “cognition in general”. The passage that perhaps best (or worst) illustrates this can be found in the First Introduction. Kant claims that aesthetic judging involves a comparison between the “actual” relation of the imagination and the understanding and “the relation […] in which they must stand in the power of judgment in general”, but the alleged comparison results in a bizarre identification that transforms the representation of the beautiful into the representation of something as something: “the apprehension of [a] manifold in the imagination agrees with the presentation of a concept of the understanding (though which concept be undetermined [unbestimmt welches Begriffs])” (20: 220-21).

What I called the derivatives of this notion are as follows: (a) “faculties of cognition in general” (5: 286, translation corrected to plural), (b) “a judgment in general” (5: 287), (c) “the power of judgment in general” (5: 286 and passim), (d) a reflection of this faculty “by means of which it strives to rise from intuitions to concepts in general” (20: 249), and (e) “the lawfulness of the understanding in general” (5: 241). Unsurprisingly, it is in the Deduction that the transcendental inquiry reaches the maximum of abstraction and formalization. In search of a “justifying ground [Rechtsgrund]”, the Deduction follows the “guideline” of considering “only the formal peculiarities”, the “logical form” of the judgment of taste (5: 287), or of “abstract[ing] from all its content” (5: 281). But what begins as a methodological abstraction soon turns into the event of aesthetic judging: the latter itself appears as a completely content-neutral form. The only phrase missing from the above list is ‘imagination in general’. But only the phrase is missing. “Taste, as a subjective power of judgment, contains a principle of subsumption, not of intuitions under concepts, but of the faculty of intuitions or presentations (i.e., of the imagination) under the faculty of concepts (i.e., the understanding)” (5: 287). This principle of connecting the faculties qua faculties is dubious, to say the least. The experience of the beautiful seems to be become the judging of whatever: (f) “the judging of a sensible object in general” (5: 290). And it is a judging by whomever, to be sure: judgers of taste represent “mankind in general” (5: 301).

The thesis that “cognition general” is the condition of determinate cognition also has different versions, but each of them strengthens the impression that what happens in aesthetic judging somehow underlies all cognition of objects. Kant mentions the “universality of the subjective conditions of the judging of objects” (5: 218); “the subjective condition of cognizing” (5: 238); “the subjective condition of all judgments”, which turns out to be “the faculty of judging itself, or the power of judgment” (5: 287); “the subjective conditions of the use of the power of judgment in general” (5:290); “the subjective, merely sensitive condition of the objective use of the power of judgment in general (namely the agreement of those two faculties [the imagination and the understanding] with each other)” (20: 223-24).

The generality and the condition status of aesthetic judging do not necessarily imply that everything is beautiful.[7] The correspondence of the imagination and the understanding “prior to any concept” (5: 289) is a structural moment which must be present in ordinary cognition, as a point of equalization between the two faculties, but which normally remains unnoticed, whereas in aesthetic judging the subject specifically becomes aware of it. In this perspective, the judging of the beautiful, as a “cognition in general”, is not identical with, but eminently representative of, what takes place in all cognition, and the beautiful is specific in that it occasions the realization of that pre-conceptual harmony as such. This (or something like this) is what Kant seems to mean at the end of §9, where he emphasizes that the representation of the beautiful can by itself, “without comparison to others”, be in “agreement with the conditions of the universality that constitutes the business of the understanding in general” and bring “the faculties of cognition into the well-proportioned disposition that we require for all cognition” (5: 219), then in §21, where he says that there is a proportion of the mental disposition which is “optimal” for the activity of the faculties “with respect to cognition (of given objects) in general” (5: 238), as well as in the first paragraph of the General Remark after the Exposition (5: 240-41).

What is important for me here is that aesthetic judging involves a mental constellation that is not restricted to the judging of particular objects but extends to “all cognition”, even though it need not always (and usually does not) make itself noticeable or felt. According to the standard explanation – which is identical with the one Kant gives in §9 –, aesthetic judging must be a “cognition in general” because this is how it can aspire to subjectively universal validity. I will argue for a different approach: Kant needs aesthetic judging as a subjectively yet universally valid “cognition in general” because this is how aesthetics can lend a hand to epistemology.

I began by remarking that the introduction of common sense as a condition of cognition is an unexpected move by Kant in §21, since the CPR does not know of common sense as such a condition. But it is perhaps even more unexpected in the light of the Prolegomena. In §40 of the CPJ, Kant identifies taste with common sense (meaning that taste becomes its own principle, which might sound odd, but in fact accords with the above observation concerning the peculiar exemplarity of its judgments). More precisely, he identifies it with “a kind of sensus communis”; the other kind is the “common human understanding” (5: 293). Now it is this common sense that appears in the Prolegomena, where it has another name, too, gesunder Menschenverstand (“sound common sense” in the English translation),[8] and where Kant fiercely criticizes its advocates, the philosophers of the Scottish School of Common Sense, saying that what they claim to be immediately certain knowledge lacks universality and necessity, and cannot be used to beat off David Hume’s skepticism. Transcendental philosophy alone can save the concept of causality and other a priori concepts (see Proleg. 4: 369-71 and 257-60).[9]

The CPJ’s newly found aesthetic common sense is thus a superior version of something which Kant seven years earlier declared insufficient as a weapon against the late colleague who had awakened him from his dogmatic slumber. But Hume’s ghost is still haunting. Recall, in the Metaphysik Mrongovius Kant says that egoism can be “skeptical”, and admits that he “cannot refute the egoist by experience”. According to the first sentence of §21, “Cognitions and judgments must, together with the conviction that accompanies them, be able to be universally communicated, for otherwise they would have no correspondence with the object: they would all be a merely subjective play of the powers of representation, just as skepticism insists” (5: 238). I cannot analyze here in detail the argument of §21, nor would that make much sense, because I could not say anything new about it. Just in a nutshell, the universal communicability of cognitions presupposes that of “the mental state, i.e., the disposition of the cognitive powers for a cognition in general”; “this disposition cannot be determined except through the feeling (not by concepts)”; therefore, the feeling itself must be universally communicable, too; this, in turn, “presupposes a common sense”, so “the latter must be able to be assumed, and indeed without appeal to psychological observations, as the necessary condition of the universal communicability of our cognition, which is assumed in every logic and every principle of cognition that is not skeptical” (5: 238-39).

It might seem that §21 has nothing to do with aesthetics. It does not even mention taste. But taste does appear here in the disguise of “cognition in general”.[10] And in §22, which is again explicitly about taste, Kant calls “common sense” a “principle of the possibility of experience”, though he leaves it open whether it is “constitutive” or “regulative” (5: 240). The question decisive of the role and significance of §21 is how to read the phrase “for otherwise” in the first sentence. It could be read as referring to faits accomplis: cognitions are universally communicable, because they do correspond with their objects, and there is an epistemic common sense, since the universal communicability of cognitions entails or presupposes its existence. In this case, the section’s argument is a sort of external support for the doctrine of taste: aesthetic reflection being a “cognition in general”, its theory can rely on something that otherwise belongs to epistemology. But even apart from the fact that transcendental philosophy has up to now been completely silent about that epistemic common sense, such a reading disregards the problem spotlighted by the passage quoted above from the CPR’s Canon chapter. What I called there a kind of solipsism is not qualitatively different from the “skepticism” of §21. It too is “a merely subjective play of the powers of representation” in that the subjective objectivizations of perceptions do not produce a common objectivity: because they do not simply reproduce an already given objectivity, which, if it existed, would also be ‘sensed’ commonly. §21 – and the whole encounter of epistemology and aesthetics in the CPJ – must be read the other way round: it is the theory of empirical cognition that gets support from that of taste.

At the end of Section 2, I differentiated between two functions of the purposiveness principle. A similar distinction can be made here between common sense as a principle of cognition (CSC) and common sense as (a principle of) taste (CST). A non-solipsistic or non-skeptical conception of experience obviously requires that there be CSC. It requires “the unanimity in the manner of sensing [Einhelligkeit der Sinnesart]” (5:240), which is a synonym for common sense. This “unanimity” alone can guarantee that the subjects will perceive whatever they perceive in an identical manner, which, in turn, guarantees the ‘commonness’ of the cognitions made of these perceptual contents. With respect to CSC, however, Kant seems to have three equally bad options.

First, he cannot convert to common-sense realism or to any kind of exclusively or even just predominantly realistic ontology and epistemology; that would amount to denying TI and destroying metaphysics. Second, the assumption of CSC cannot rely on what I think Kant means by “psychological observations”, i.e., on empirical, historical evidence, on statistical data showing the agreement among so many people on so many objects; this consensus, however broad, is a far cry from what could only be – and hence cannot be – confirmed by what I above described as a doubly endless poll. Third, and most importantly in the present context, there is a complication that seems impossible to overcome in a transcendental idealist epistemology. While it is true that a non-solipsistic conception of experience requires CSC, it is also true that the practice of experience is possible, at a limited scale (if I may put it so awkwardly), without CSC. It is possible because an empirical judgment as such is always limited. Limited both to some particular content and to some judging persons. The judgment that “the sun warms the stone” is about this particular stone and this particular occasion of sunshine, even though it can also be meant as an instance of universal laws. Anyone who makes this judgment is, or should be, aware – depending on whether or not they have already read the CPR – that the alleged “object” of the judgment is but the objectivization of their actual perceptions. Still, they take what they make as an objective experience – how else? They are happy if they get “confirmation” from “others” (5: 216) who likewise perceive the same stone as becoming warm and make the same judgment of experience. But, again, they all are, or should be, aware that it is not the object itself that this agreement is based on. They know that this is just an intersubjective consensus. And they also know that it would be extremely hard to find out whether “every human being” (5: 219) makes (or has ever made, or will ever make) the same judgment.

To put the third point somewhat differently, CSC as such is not a transcendental condition under which alone experience, in the limited sense outlined here, is possible. If someone asks me what I think guarantees the objective validity of an empirical judgment of mine, including the certainty that others will agree with me, I cannot, as a fan of Kant, reply either by resorting to pure realism or by adducing historical evidence. But, worse, I also cannot say that I am not able to make that judgment, to unite perceptions in my consciousness and objectivize them, without there working in me a principle (CSC) according to which everyone is compelled to perceive and judge the thing in question – let alone everything – in an identical way.

Enter taste to save the day with/as CST. Its judgments are doubly unlimited. First, though always occasioned by a particular object, they decouple from this object both in that they are not about the object, but about how “the subject feels itself as affected by the representation” (5: 204) – this is the first tenet the Analytic of the Beautiful lays down –, and in that this feeling turns out to be that of a “cognition in general”: of a mental state that, regardless of its particular content, is the condition of every determinate objective cognition. In this respect, judgments of taste are about “a sensible object in general”. Second, they are also unlimited in that they demand universal assent: they are made in the name of “mankind in general”. Since, according to that first tenet, judgers of taste cannot even think of trying to base their demand on objective criteria, they have no choice but to look in themselves for something to be declared universally communicable, able to be shared by everyone. And the only thing they can find is “cognition in general”.

In the Metaphysik Mrongovius, the paragraph after the passage quoted above begins with this sentence: “Dualism (pluralism) is opposed to egoism” (Kant, 1997, 29: 928), the word Pluralismus being written above Dualismus in the manuscript. Indeed, both are at place here. Dualism fits here because TI entails the existence of things outside the mind. Pluralism fits here because it means that cognitions and their objects should be common to all. But how to achieve that? “If the judgment of taste must not be counted [gelten muss] as egoistic, but necessarily […] as pluralistic, then it must be grounded in some sort of a priori principle” (5: 278).

According to the Deduction, the “problem of the critique of the [aesthetic] power of judgment belongs under the general problem of transcendental philosophy: How are synthetic a priori judgments possible?” (5: 289) A judgment of taste involves an a priori synthesis. Of course, this apriority does not mean that it could be known in advance, before any (aesthetic) experience, whether or not an object will cause satisfaction. “It is an empirical judgment that I perceive and judge an object with pleasure.” If “it is not the pleasure” itself, then it must be “the universal validity of this pleasure perceived in the mind as connected with the mere judging of an object that is represented in a judgment of taste as a universal rule for the power of judgment, valid for everyone”, and the judgment “that I find it beautiful, i.e., that I may require that satisfaction of everyone as necessary” is already “an a priori judgment” (5: 289). But, again, the pleasure to be declared universally valid is the feeling of a “cognition in general”. Taste


can be directed only to the subjective conditions of the use of the power of judgment in general (which is restricted neither to the particular kind of sense nor to a particular concept of understanding), and thus to that subjective element that one can presuppose in all human beings (as requisite for possible cognitions in general),


so, ultimately, a “judgment of taste […] asserts only that we are justified in presupposing universally in every human being the same subjective conditions of the power of judgment that we find in ourselves”, provided “that we have correctly subsumed the given object under these conditions” (5: 290).

This is not necessarily the first answer that I, even as a fan of Kant, would give if someone asked me ‘What do you think is the only thing that a correct judgment of taste asserts?’ Anyway, what Kant has found is a real transcendental jewel. The difficulty relating to CSC, namely, that the particular instances of experience do not warrant its assumption – because they are doubly limited, and because it is not impossible to make judgments of experience without having to think that everyone else will perceive and objectivize the given content, let alone everything, in the same way ­–, this difficulty is over now. I cannot ever use the predicate ‘beautiful’ without consciously exemplifying CST, i.e., without consciously assuming, through an a priori synthesis, that the working of the cognitive faculties that I feel in myself is the same as they work in everyone else. Common sense, the whole of it, is established, because if CST is established, then – but only then – CSC becomes established, too. If “cognition in general” must be assumed to be universally valid, then the universality of the determinate cognitions conditioned by it must be assumed, too.

Note well, this solution is one which TI permits. Objectivity remains dependent on individual minds. Common sense is not realistic, it does not entail that there is a given objectivity that must be sensed commonly, in the same way by everyone. A judgment of taste only assumes this sameness in the sphere of “cognition in general” – but assumes it as the condition of its own possibility. This is the most that can be achieved within the framework of TI. The jewel Kant has found is a type of judgment that cannot be made other than as a blow to solipsism/skepticism: without necessarily assuming the uniformity of the human constitution that is at work in all cognition. The “difference among the subjects”, which, as a difference among too many subjects capable of producing too many diverse objects, not so long ago threatened to block even the “presumption” of a common world, is not a concern anymore for the transcendental theory of experience. In practical terms, there might still be cases in which the subjects disagree over what the object is. As a matter of principle, however, their difference has been overwritten by “unanimity”.

The Introductions can be interpreted in an analogous manner, with only two notable differences. First, here it is the notion of subjective purposiveness that connects the epistemic and the aesthetic. With respect to what goes on in the mind in (or as) the judging of the beautiful, the descriptions “harmony of the faculties” and “subjective purposiveness” are interchangeable, purposiveness being nothing but an intentionally or quasi intentionally effectuated harmony, as a unity in diversity. As a quality of aesthetic judging, subjective purposiveness refers both to the imaginative, pre-conceptual synthesis of a manifold of intuition and to the agreement of the faculties. The second difference is a shift in perspective. CST had to secure the cognitive uniformity of subjects. What I above called PPT must secure the uniformity of objects in their cognizability. The link between the two projects is, unsurprisingly, the notion of “cognition in general”, which remains central in the Introductions, too, and which, as I have argued, extends to everyone and everything. (This centrality of the notion is why it is not necessarily a mystery of the CPJ that the judgments of taste have two different a priori principles, common sense and nature’s purposiveness, in one and the same book.) The underlying logic being identical, I confine myself to enumerating the essential points.

The problem with PPC is that, while it is required for the coherence of experience, it cannot break out of the status of a regulative principle; the necessity with which it is assumed is that of a “need”. Like in the case of CSC, Kant has three non-options to endow PPC as such with a deeper necessity and to make it more than a nominally transcendental but in fact superfluous principle, which orders discursive cognition to do what it does. First, he cannot return to ontotheology. Second, the sounder-than-heuristic validity of PPC cannot be based on statistical evidence, here on the knowledge that experience has already proven so coherent (or that nature has already shown itself to be so well-ordered). Third, and again most importantly: as it was clear already in the CPR, though systematicity might be a must in the conception of experience and nature, coherence is not an a priori condition of individual judgments of experience. It is not just that incoherent experiences cannot be ruled out. Everybody can say (though probably nobody says) that in making a ‘normal’ judgment of experience they took care to act in accordance with the discursive structure of human cognition, but nobody can say that they would not have been able to make that particular judgment without assuming that the whole of nature is purposive.

With the notion of aesthetic reflection as a subjectively purposive mental state and as a “cognition in general”, however, Kant can easily take the decisive step. “Judgments of taste […] lay claim to necessity and say, not that everyone does so judge […] but that everyone ought to so judge, which is as much as to say that they have an a priori principle for themselves” (20: 238-39). It is their very “possibility” that “presupposes an a priori principle” (5: 191). What makes the step from PPC to PPT so easy is that the cognitive, concept-oriented employment of the reflecting power of judgment is a direct continuation of its aesthetic use. The most compact version of PPC, as “the principle of reflection on given objects of nature”, was “that for all things in nature empirically determinate concepts can be found” (20: 211). The harmony of “the imagination” and “the understanding” satisfies “the subjective, merely sensitive condition of the objective use of the power of judgment in general” (20: 223-24): the elementary condition of conceptualizability.

What could be more obvious than that the judgment of taste is to be based on the principle of purposiveness? As already quoted, it


wins a claim to universality and necessity, as merely reflective judgment, through the relation of the subjective purposiveness of the given representation for the power of judgment to that a priori principle of the power of judgment, of the purposiveness of nature in its empirical lawfulness in general, and thus an aesthetic reflecting judgment can be regarded as resting on a principle a priori. (20: 243)


To complete the analogy, this ‘aestheticization’ amounts to a transubstantiation. Unlike in the case of common sense, here Kant clearly says what it means for the purposiveness principle to gain this aesthetic function. “The concept of purposiveness is not a constitutive concept of experience at all […]; for it is not a category” (20: 219-20). But a “higher faculty” is one which “contains constitutive principles a priori”, and “for the feeling of pleasure and displeasure it is the power of judgment” that has such a principle (5: 196, cf. 5: 197). Judgments of experience (in the ‘primary’ sense) are possible without PPC – judgments of taste are impossible without PPT. Purposiveness is established.

This solution, too, is one which can be given within TI. To PPC as such, the critical restrictions still apply. It remains regulative etc. It is not about what nature is; this kind of validity is reserved for the categories. But taste jumps in to help again. In making a judgment of taste about natural beauty, i.e., in performing an act of “cognition in general”, I cannot but presuppose the purposiveness of nature, this being the only way to demand universal assent: the only way to make such a judgment at all. And I do so all the easier because, as I proceed in the Introductions, the difference between what I might have thought to be two separate attitudes to nature – cognitive and aesthetic – gradually disappears, at least at the level of principles. Towards the end of the published Introduction, i.e., chronologically speaking, towards the end of the CPJ, Kant once for all obliterates my distinction between PPC and PPT:


the aesthetic power of judgment […] alone contains a principle that the power of judgment lays at the basis of its reflection on nature entirely a priori, namely that of a formal purposiveness of nature in terms of [nach] its particular (empirical) laws for our faculty of cognition, without which the understanding could not find itself in it. (5: 193, translation modified)


This is one of the reasons why I think that the epistemological mission prevents Kant’s doctrine of taste from becoming a genuine theory of aesthetic judging. I simply cannot imagine that whenever I make a judgment of natural beauty, I should mobilize in myself a principle according to which “for all things in nature empirically determinate concepts can be found”.

This can be my fault. As a second reason, however, there would be a crucial question that Kant does not answer, indeed, does not even ask. If aesthetic reflection, as a “cognition in general”, is the condition of determinate cognition, why does it not materialize into a determinate cognition? If it is, as all its descriptions suggest, a pro-conceptual act, why does it not lead to cognition? “The beautiful […] requires the representation of a certain quality of the object, which also can be made intelligible and brought to concepts (although in the aesthetic judgment it is not brought to concepts)” (5: 266, translation modified). Why is this quality not brought to concepts if it could be? The pleasure in the beautifulhas a causality in itself, namely that of maintaining the state of the representation of the mind and the occupation of the cognitive powers without a further aim. We linger over the contemplation [Betrachtung] of the beautiful because this contemplation strengthens and reproduces itself” (5: 222, translation modified). This is a very important observation, but the self-maintaining character of pleasure remains a marginal motif. Relatedly, the doctrine of taste does not say anything essential about what it calls (but only calls) the free play of the faculties. I say ‘relatedly’ because play is that autotelic and self-maintaining activity, and if Kant pushed this point, he could hardly maintain his condition thesis, the cornerstone of his epistemologically driven theory of taste. A condition is like a threshold. It must be possible to move on from it to that which depends on it, and the latter must occur if the condition is there. By contrast, play is purposive in and for itself.[11]

Interestingly, however, Kant picks up the question of play in his doctrine of art, where “cognition in general” and its derivatives are completely missing, save for a single mention of “the power of judgment in general” in connection with “taste” (5: 319). But taste has already lost its primacy due to the sharp distinction, at the end of §48, between what is artistically valuable and what is merely tasteful (5: 313). Immediately after that, already in §49, Kant claims that it is an “aesthetic idea” that “sets the mental powers into motion, i.e., into a play that is self-maintaining and even strengthens the powers to that end” (5: 313). Then he gives a definition that reads like a denial of what he wrote some fifty pages earlier: “an aesthetic idea” is “that representation of the imagination that occasions much thinking though without it being possible for any determinate thought, i.e., concept, to be adequate to it, which, consequently, no language fully attains or can make intelligible (5: 314). It might seem a mistake to make a comparison between the two sentences. The latter, which denies conceptualizability, is in the doctrine of art, whereas the former, which asserts the same, is in the doctrine of taste, whose protagonist is fixated on natural beauty and is inherently unable to deal with anything intentionally meaningful, at least as pure taste (which, incidentally, makes it a rather anachronistic entity in 1790). But Kant takes a further step in §51: “Beauty (whether it be beauty of nature or of art) can in general be called the expression of aesthetic ideas”, even though in the first case the judging is “a mere reflection on a given intuition, without a concept of what the object ought to be” (5: 320).

Obviously, it would not be easy to associate “much thinking” with the contemplation of a “hummingbird” (5: 290), and, just as obviously, even a much larger and possibly more beautiful bird is not a fatal threat to ornithology. Nevertheless, the very extension of the notion of aesthetic ideas to nature introduces an element to the judging of natural beauty which, on the one hand, should have been part of the doctrine of taste – as the answer to the question that Kant neglects to ask there –, but which, on the other hand, retroactively questions its epistemological mission. Originally, “mere reflection” is the same as “cognition in general”. In §51, however, it must be understood as a self-prolonging contemplation, as a ‘lingering’ which maintains itself because the beautiful, in the very “quality” that makes it beautiful, resists conceptualization and prevents discursive cognition from doing what it normally does, from registering the content of intuition as an instance of a concept.[12] Compared to this play of the imagination and the understanding, which is driven by the “impetus to think more, although in an undeveloped way, than can be comprehended in a concept” (5: 315), conceptual determination is a loss rather than a gain readily anticipated by the intuition. Can you say that this play is that which “any determinate cognition […] always rests on […] as its subjective condition” (5: 218)? If I am not completely wrong, and if this last question signals a real tension within Kant’s aesthetic theory – one between its epistemological mission and its somewhat belated commitment to providing a credible account of the encounter with the beautiful –, then it could at the same time serve as the starting point for another paper, which would be concerned with the moral significance of an aesthetic experience that suspends its subject’s conceptual control over reality, natural or man-made.




Allais L. (2015), Manifest Realism. Kant’s Idealism and his Realism, Oxford University Press, Oxford.

Allison H. E. (2001), Kant’s Theory of Taste. A Reading of the Critique of Aesthetic Judgment, Cambridge University Press, Cambridge.

Avramides A. (2001), Other Minds, Routledge, Abingdon.

Ginsborg H. (2015), The Normativity of Nature. Essays on Kant’s Critique of Judgment, Oxford University Press, Oxford.

Guyer P. (1979), Kant and the Claims of Taste, Cambridge University Press, Cambridge.

Guyer P. (2005), “The Harmony of the Faculties Revisited”, in Guyer, The Values of Beauty, Cambridge University Press, Cambridge, pp. 77-109.

Heidemann D. H. (1998), Kant und das Problem des metaphysischen Idealismus, De Gruyter, Berlin.

Hughes F. (2007), Kant’s Aesthetic Epistemology. Form and World, Edinburgh University Press, Edinburgh.

Kant I. (1997), “Metaphysik Mrongovius”, in Kant, Lectures on Metaphysics, trans. and ed. K. Ameriks and S. Naragon, Cambridge University Press, Cambridge, pp. 107-283.

Kant I. (1998), Critique of Pure Reason, trans. and ed. P. Guyer and A. W. Wood, Cambridge University Press, Cambridge.

Kant I. (2000), Critique of the Power of Judgment, trans. P. Guyer and E. Matthews, ed. P. Guyer, Cambridge University Press, Cambridge.

Kant I. (2002), “Prolegomena to Any Future Metaphysics That Will Be Able to Come Forward as Science”, trans. Gary Hatfield, in Kant, Theoretical Philosophy after 1781, ed. H. E. Allison and P. Heath, Cambridge University Press, Cambridge, pp. 29-169.

Kemp Smith N. (1923 [1918]), A Commentary on Kant’s Critique of Pure Reason, Macmillan, London.

Kuehn M. (1987), Scottish Common Sense in Germany, 1768–1800. A Contribution to the History of Critical Philosophy, McGill-Queen’s University Press, Montreal.

Küplen M. (2015), Beauty, Ugliness and the Free Play of Imagination. An Approach to Kant’s Aesthetics, Springer, Heidelberg.

Longuenesse B. (1998), Kant and the Capacity to Judge. Sensibility and Discursivity in the Transcendental Analytic of the Critique of Pure Reason, trans. Charles T. Wolfe, Princeton University Press, Princeton.

Massimi M. (2017), “What is this Thing Called ‘Scientific Knowledge’? – Kant on Imaginary Standpoints and the Regulative Role of Reason”, in Kant Yearbook, Vol. 9, ed. D. H. Heidemann, De Gruyter, Berlin, pp. 63-84.

Massimi M. and Breitenbach A. (eds.) (2017), Kant and the Laws of Nature, Cambridge University Press, Cambridge.

Matherne S. (2019), “Kant on Aesthetic Autonomy and Common Sense”, Philosophers’ Imprint 19, no. 24, pp. 1-22.

Schulting D. and Verburgt J. (eds) (2011), Kant’s Idealism. New Interpretations of a Controversial Doctrine, Springer, Dordrecht.

Shier D. (1998), “Why Kant finds nothing ugly”, British Journal of Aesthetics 38, no. 4, pp 412-418.

Wachter A. (2006), Das Spiel in der Ästhetik, De Gruyter, Berlin.

Wenzel Ch. (1999), “Kant finds nothing ugly?”, British Journal of Aesthetics 39, no. 4, pp. 416-422.

Zhouhuang Zh. (2016), Der sensus communis bei Kant. Zwischen Erkenntnis, Moralität und Schönheit, De Gruyter, Berlin.

Descripción: kant_ctk_TRANSPARENTE1.png

· Zoltán Papp is docent in the Department of Aesthetics at Eötvös Loránd University (ELTE), Budapest, Hungary, and an editor of Estetika: The European Journal of Aesthetics. E-mail: [email protected]

[1] References beginning with the volume number 5 or 20 are to the third Critique (Kant 2000), those beginning with A and/or B are to the first (Kant 1998).

[2] This is why I make a single reference here to the five books I admire the most: Guyer 1979, Allison, 2001, Hughes 2007, Zuckert 2007, Ginsborg 2015 (the subtitle of my paper is meant as a twisted allusion to the title of Hughes’s book). For a typological overview of interpretations, see Guyer 2005, where he at the same time revises his earlier approach. Küplen (2007, Ch. 3) adds two more categories to the typology.

[3] Kant’s “teaching […] is extremely self-contradictory, wavering between a subjective and an objective interpretation of the Ideas of Reason” (Kemp Smith 1923, p. 547).

[4] For a recent collection of essays dealing with various aspects of Kant’s conception of the lawfulness of nature, see Massimi and Breitenbach (eds) 2017.

[5] Though solipsism is not a hot topic in Kant scholarship, Massimi (2017) also interprets the Canon passage in terms of solipsism. Heidemann (1998) investigates Kant’s idealism in the historical context of egoism. For solipsism in Descartes, Locke, Berkeley, and others, see Avramides 2001, Chs. I-VI. – For the variety of approaches to TI, see Schulting and Verburgt (eds) 2011 as well as Allais 2015, Pt. One.

[6] For Kant, conceptual consciousness and object-consciousness are the same: “object […] is that in the concept of which the manifold of a given intuition is united” (B 137). This means that, strictly speaking, the beautiful is not an object.

[7] For a debate specifically about this problem see Shier 1998 and Wenzel 1999.

[8] To make the picture even more colourful, the Hungarian phrase for this common taste translates as ‘sober reason’. – For a comprehensive analysis of the different meanings and roles of sensus communis in Kant’s writings, see Zhouhuang 2016.

[9] For Kant’s reception of the common-sense philosophy, see Kuehn 1987, pp. 167-207.

[10] This is why I do not think that the question whether the common sense of §21 is an aesthetic or a cognitive one is really a question. Most recently, Matherne (2019) has argued for the second option.

[11] Wachter (2006, pp. 88-120) gives an excellent analysis of this problem.

[12] As Longuenesse puts it (1998, p. 164): “What makes judgments [of taste] merely reflective is that in them, the effort of the activity of judgment to form concepts fails. And it fails because it cannot succeed.”

Enlaces refback

  • No hay ningún enlace refback.

Creative Commons by-nc 3.0 Logo

ISSN: 2386-7655


  Scimago Journal & Country Rankscopus logo


DOAJ LogoErih Plus LogoCitefactor logoredib Logo

LatIndex LogoISOC Logo MIAR Logo
MLA LogoZenodo Logo
ESCI LogoEBSCO LOGOWeb of Science