I. Introduction
Kant argues that we can distinguish the claim that something is truly beautiful, and should be appreciated by all as such, from the claim that it is merely agreeable, appreciated only by those who have some personal reason for taking pleasure in it. He holds not simply that we have reason to suppose that everyone will agree about whether something is beautiful, but also that we can expect (zumuthen) or demand (fordern) the agreement of all. Though his use of such language leads many commentators to conclude that Kant believes that there is normative force behind the claim to agreement made by a judgement of beauty, I hold that he does not do so. I seek to show that a non-normative reading is both a more plausible interpretation of Kant and a more plausible approach to understanding our responses to beauty.
In Part II, I offer a non-normative interpretation of Kant's account of judgements of beauty. I show that his strongest use of apparently normative language, in the second moment of the Analytic, is in fact an explanation of how the phenomenology and language around judgements of taste support his contention that they are universal. I turn next to the Deduction, arguing that it is meant to legitimate the claim of universal validity described in the Analytic but not to ground a normative demand. I then argue against Henry Allison's normative reading of the Deduction in Part III. In Part IV, I offer a reply to the view that judgements of taste must be normative because beauty is the symbol of morality. In Part V, I consider the position that the normativity of judgements of beauty should shape our understanding of other sorts of reflecting judgements, a view with implications far beyond aesthetics. This view, I hold, assigns such judgements a central role in Kant's theory that they simply cannot occupy. I conclude that normative readings make Kant's account of our experience of the beautiful appear less plausible than it truly is.
II.
Drawing on the distinction made in the Introductions between reflective and determining judgements (First Introduction V, 20:211, Introduction IV, 5:179), Kant first characterizes judgements of the beautiful by differentiating them from other sorts of judgements. Cognitive judgements involve the use of the faculty of understanding to relate an antecedently held concept to an object as we perceive it. We perceive some object, form a representation of it, and interpret that representation in terms of concepts we already possess. When I encounter a visual field, I distinguish shapes, colours, etc. from one another, representing them as separate objects, and I apply concepts, e.g., ‘tree’ or ‘leaf,’ to these representations, thus relating visual phenomena to concepts that pick out other similar objects. Judgements of the beautiful, and judgements of taste more generally, do not subsume objects under concepts in this way, and so are not cognitive. Instead, they relate the perceiving subject's representations to her own feelings of pleasure and displeasure. They are subjective, characterized not by a claim that they make about some feature of the object (e.g., ‘green,’ ‘leaf-shaped’), but rather by a claim that they make about the effect that the object has on the subject itself, i.e., the effect of bringing about pleasure or displeasure.
These judgements are not the only ones related to feelings of pleasure and displeasure, however. Representations of objects judged to be agreeable or good are also related to these feelings, but only insofar as these representations are taken to be of an object in which the perceiving subject takes some interest. In the case of judgements of the good, this interest is in the value that the object of the representation has for us, either insofar as it is good for fulfilling a purpose of ours or insofar as it is good in itself. In the case of the agreeable, this interest is in the object's ability to gratify some physiologically or psychologically conditioned need. Each of these judgements (of the agreeable, the good, and the beautiful) satisfies the subject, but judgements of the beautiful are “the only free satisfaction”Footnote 1 — the only satisfying judgement that is not tied to an interest in the object represented (§5, esp. 5:210).
Having concerned himself in the first moment of the Analytic of the Beautiful (§§1-5) with distinguishing judgements of beauty from others and with showing that they are disinterested, Kant goes on to explain that these judgements are universal. Because judgements of the beautiful are without interest, he says in §6, they are the object of a universal satisfaction. The subject “feels himself completely free with regard to the satisfaction that he devotes to the object, he cannot discover as grounds of the satisfaction any private conditions, pertaining to his subject alone, and must therefore regard it as grounded in those he can also presuppose in everyone else; consequently he must believe himself to have grounds for expecting a similar pleasure of everyone” (5:211, italics mine). Kant takes himself to have the grounds to show in §6, then, that, insofar as judgements of beauty are disinterested, they must also be universal. Because the pleasure involved in such judgements is disinterested, no “private conditions” serve as the basis for the subject's taking pleasure in the representation. So, only conditions present in every other subject must serve as the basis for such judgements, and they are therefore universal.
The aim of §7 is to show why the universality claimed in a judgement of beauty is not common to the other kinds of judgements that relate a sensation of pleasure to some representation of an object. So, §7 is putatively devoted to comparing judgements of the good, agreeable, and beautiful with respect to their universality. Most of this section, however, is a comparative description of our experience in making the two sorts of aesthetic judgements — those of the agreeable and of the beautiful. Much of it is devoted to explaining how the claim that judgements of beauty are universal while judgements of agreeableness are not accords with familiar cases in which we find something beautiful. In describing such judgements of beauty, Kant uses blatantly normative language for which he does not seem to have offered any previous justification.
I contend that this kind of language is not an attempt to show that judgements of taste are normative, but is instead an attempt to show that his distinction between judgements of beauty and agreeableness, respectively, tracks the familiar experience of judging objects to be beautiful. In the text leading up to the passage in which this strongly normative language occurs, Kant discusses familiar instances of aesthetic judgement. One calls something beautiful only when she expects that others will also believe that it is so, and because she believes that everyone will reach the same judgement, she “speaks of beauty as if it were a property of things” (5:212). Kant, of course, does not hold that beauty is a property of things, but is explaining why it is natural for people to speak as though it is, and how this familiar way of talking about the beautiful, though incorrect, supports the claim that judgements of beauty are universal. Drawing on this description of standard cases of judgements of beauty, Kant goes on to say that, because one thinks of the beautiful in this way, one “says that the thing is beautiful, and does not count on the agreement of others with his judgment of satisfaction because he has frequently found them to be agreeable with his own, but rather demands it from them,” and that one “rebukes” those who disagree “and denies that they have taste, though he nevertheless requires (verlangt) that they ought to have it” (5:213). The fact that the aim of §7 is to explain the relationship between distinctions that Kant has already made, rather than to advance new claims, suggests that this strong language does not represent his own view. Kant is continuing to discuss the way that everyday speech and experience support the claim that judgements of the beautiful are universal, without endorsing the way in which this claim is expressed in these contexts. He does not point out that this language is excessive or imprecise, but neither does he do so in the case of the everyday claim that beauty is a property of things — a claim that he clearly rejects. In each case, Kant accounts for, but does not endorse, familiar parlance.
Many other passages suggest that Kant does not take the universality of judgements of taste to be normative in character. Earlier, in the context of distinguishing judgements of the agreeable from those of the good, he tells us that “an obligation to enjoyment is a patent absurdity” (5:209). Late in §8, after his use of apparently normative language, he tells us that “[t]he judgment of taste does not itself postulate the accord of everyone (only a logically universal judgment can do that, since it can adduce grounds); it only ascribes this agreement to everyone, as a case of the rule with regard to which it expects confirmation not from concepts but only from the consent of others” (5:216). Kant is clarifying his position on the normative status of universality. After referring to the “demand” made by judgements of beauty in both §7 and the early part of §8, he tells us that when making such judgements one “believes oneself to have a universal voice, and lays claim to the assent of everyone,” but that this voice is “only an idea” (Ibid.). He states that the universal claim of a judgement of beauty cannot “adduce grounds” like a logical judgement, but “only ascribes … agreement to everyone” (Ibid.), then defers full explanation of the character of the aforementioned demand, which is not fully fleshed out until the fourth moment of the Analytic and the Deduction of Pure Aesthetic Judgements.
While the second moment focused on the universality of judgements of taste, the fourth is devoted to explaining their necessity. In a pure judgement of taste, one's pleasure in the representation of an object is necessary in the sense that it is “regarded as an example of a universal rule,” yet this is a very unusual kind of necessity, because it derives from a rule “that one cannot produce” (§18, 5:237). The second and fourth moments, then, yield two “peculiarities” differentiating judgements of taste from other a priori judgements — the second tells us that (1) in taste, there is universal validity a priori, but of a singular judgement only, and the fourth tells us that (2) in taste, there is necessity, but a necessity that does not depend on grounds of proof (§31, 5:281). The Deduction attempts to show how, despite these peculiarities, judgements of taste are possible, are based on an a priori principle of taste, and can be expected of all persons. Like Paul Guyer,Footnote 2 I hold that this expectation is not a normative demand, but I contend that, even if we accept reconstructions of the Deduction that differ substantially from Guyer's, the Deduction does not establish the normativity of taste.
In the barest terms, the Deduction seeks to establish that normally constituted human beings have the ability to make pure judgements of taste because the conditions that give rise to judgements of taste are necessary for cognition. The power of judgement, when it is in a state that could give rise to cognition but has no matter (i.e., no concept), is directed only to the “subjective element that one can presuppose in all human beings” (§38, 5:290). This state is “requisite for possible cognitions in general,” meaning that one must be capable of achieving this state if one is able to engage in cognition at all. All cognitive subjects are the same insofar as they are capable of achieving this state. A judgement resulting solely from the relationship of such a state to a given representation “must be assumed to be valid for everyone a priori” and “can rightly be expected of everyone” (Ibid.).
The argument of the Deduction differs from an argument establishing a normative demand because it does not enable us to determine the ground of judgements of taste, but only to determine that there is a ground, and that it is based in capacities shared by all subjects. One indication that Kant takes this position is that the language of demand and requirement is conspicuously absent from the section where the Deduction proper takes place and from the Remark that follows; if the Deduction is meant to offer a normative ground for pure judgements of taste, it is difficult to see why Kant would eschew such language. Even more importantly, the line of argument simply does not point to a normatively grounded requirement. The latter must include the ability to offer a basis for insisting on the agreement of others, and this is impossible in the case of judgements of taste (see esp. §7, 5:216, §33, 5:284-285). The nature of the ground of a judgement of taste cannot be articulated and is in principle undiscoverable. At the same time, the argument of the Deduction enables us to conclude that all human beings in full possession of their faculties will be able to make pure judgements of taste in the same way. This provides us with a basis for criticism, since we can point to conditions — such as the presence of an interest, influence, or psychological quirk — that we believe have prevented someone from making a pure judgement of taste, and we can also point to aspects of an object that seem to please in a way that is free from such interfering factors. So, we can point to features of our judgements that we believe indicate their conformity to the criteria for pure judgements of taste. Yet these criteria concern the judging subject rather than the representation eliciting the judgement. My judgement that this clematis flower is beautiful must be universal, I might argue, because I can judge it without appeal to my preferences or cultural associations, or because the appearance of the stamens against the delicate petals appeals to a sense of balance that can be appreciated by all. Yet a judgement's seeming not to depend on private interests does not promise actual independence. It may be that my judgement is tainted by a personal interest of which I am unaware (see, e.g., §8, 5:216, §19, 5:237). Or it may be that the flower is truly beautiful and that my pleasure arises from the free play of the faculties, but that I have failed to correctly identify the formal features that are the cause of the free play, so that I have accurately judged that the flower is beautiful, yet it is not beautiful for the reasons that I have proposed. The claim to universality, then, does not depend on an account of the features of this particular object, but on the nature of the judger's pleasure. An explanation of this object's beauty “in accordance with determinate rules” (§33, 5:284) is in principle impossible for Kant. However, in the case of judgements with normative force, such as moral or logical judgements, we can offer just such an explanation.
We generally think of normative activities as activities that are governed by a universal law or set of rules that are binding for everyone. Further, we are capable of deciding whether to follow these rules or ignore them; this differentiates normative laws from the natural laws that we cannot help but follow. In the case of aesthetic judgements, there can “be no rule in accordance with which someone could be compelled to acknowledge a thing as beautiful” (5:215), so even though we claim that others should agree with our judgements of beauty, these judgements differ from familiar normative judgements (e.g., logical and moral judgements) in important respects: we cannot offer a ground for our judgements of beauty or justify them by explaining how they accord with a relevant rule, so we can only offer negative reasons for believing that they are correct, such as an explanation as to why we think that private interests could not have infected our thinking.
In moral judgement, too, we cannot be sure that we are acting out of respect for the law, because we cannot be sure that we are motivated solely by duty. In judgements of the beautiful, though, we cannot be sure that we are acting in accordance with the correct rule not simply because we do not have complete insight into our own noumenal intentions, but because it is in principle impossible to know what the correct rule could be. We know that the basis for the pure judgement of taste is a distinctive kind of pleasure, and we know that it is a pleasure free from interest, based on a feeling that arises solely from the free play of the faculties. Beyond characterizing the feeling that serves as a basis for pure judgements of taste as a distinctive kind of pleasure that cannot be fully specified, we can only characterize it in terms of what it lacks: it is disinterested and not based on concepts. This differentiates pure judgements of taste from (normatively binding) logical or moral judgements. Though I cannot be sure that my moral judgement is pure, I can understand what it would mean for the judgement to be pure, and how impure judgements fall short. This is not possible in the case of judgements of taste.
Normative accounts of Kantian judgements of taste must show how an undiscoverable rule that people cannot be persuaded to follow through logical or rational appeals can be the sort of thing that one can nevertheless decide to follow. I hope to have shown that a non-normative reading of Kantian judgements of taste is plausible, and to show in what follows that the three prominent normative readings are less faithful to Kant's text and to the idea of a truly free satisfaction in the beautiful.
III.
Allison holds that the central aim of the Deduction is to ground a normative principle for taste. What the Deduction adds to the account of the Analytic, he contends, is a clear connection between the harmony of the faculties and the characteristic activity of the faculty of judgement, namely subsumption. The question of whether judgements of taste are possible can be reformulated as the question of whether it is possible to subsume a representation without concepts, and the Deduction shows that this is possible because it shows how we can subsume representations “under the faculty of judgment itself”Footnote 3 in its capacity for cognition without actually using a concept or engaging in cognition. All that is presupposed in pure judgements of beauty is “the conditions of the possibility of … [the power of judgement's] successful activity.”Footnote 4 Allison claims that this establishes that the claim to agreement in judgements of taste is normative because in his view “normativity in general for Kant is rooted in the conditions of the successful or coherent activity of the faculty in question.”Footnote 5 Allison offers two examples: “the categorical imperative expresses the conditions of the coherent legislation of practical reason in choosing justifiable maxims,” and “rules for the determination of the unity of apperception express the conditions of the possible use of the understanding.”Footnote 6 So, the categorical imperative is normative because, if we fail to employ it, we are not using the faculty of pure practical reason in a successful or coherent way; the categories are normative because, if we fail to judge in accordance with them, we are not using the faculty of the understanding in a successful or coherent way.
Even if this controversial understanding of normativity is correct, however, we cannot extend this reasoning to the principle of taste in its relation to the faculty of judgement. If the case were parallel, the principle of taste would be normative because, if we fail to employ it, we must not be using the faculty of judgement in a successful or coherent way. This, however, is surely not the kind of ‘demand’ made by judgements of the beautiful. Someone might fail to attend to beautiful objects and still be capable of using the faculty of judgement coherently and successfully, whereas this is not possible in the other examples offered by Allison: if I fail to recognize the specific injunctions of the categorical imperative, there is a fundamental flaw that prevents me from properly using my practical reason. On the supposition that we are in rough agreement with Allison's understanding of normativity, we can say that the categories and the categorical imperative are normative because if we did not make use of them we would be unable to meaningfully engage in the kind of activity requisite for being part of humanity; if someone fails to exercise the faculty of the understanding in accordance with the categories, or fails to act within the moral constraints issued by the categorical imperative, we cannot interact with her as we interact with a competent, rational being. This is not true of an individual who fails to attend to something beautiful in the manner requisite for a judgement of taste. Even if I never use my capacity to judge what counts as beautiful, I can make coherent use of the faculty of judgement.
The categorical imperative and the categories demand that (1) we accept the rules they impose on the operation of the respective faculties, and that (2) we apply them correctly in particular cases. The claim to agreement made by judgements of taste is not based on rules and it is impossible to determine its ground in any given case. Instead of attempting to show how such a claim to agreement can meet his stated standard for normativity, Allison suggests (drawing on the analysis of Béatrice Longuenesse) that the capacity to “schematize without a concept” is what is necessary for the coherent activity of judgement.Footnote 7 This would make the capacity to schematize without a concept normative for the coherent employment of the faculty of judgement. According to Allison, this implies the normativity of taste because the Deduction establishes that judgements of taste, if pure, arise solely from a priori conditions necessary for schematism without a concept. So, if Allison's view is correct, schematism without a concept is normative for the faculty of judgement and judgements of taste are one kind of schematism without a concept. His view, then, implies that, while “taste is grounded indirectly in the conditions of cognition,” it is not itself such a condition.Footnote 8 This undermines Allison's comparison between taste and other kinds of normative demands.
Allison can still claim, though, that because judgements of taste are one kind of schematism without a concept and such schematism is normative for the faculty of judgement, a requirement to judge correctly in matters of taste is derived from the normative demand to schematize without a concept. This duty is parallel to the duty to correctly apply the categorical imperative to specific cases. Two claims must be established if we follow this route to establishing normativity: (1) that schematism without a concept is normative for the faculty of judgement, and (2) that judgements of taste are an instance of schematism without a concept. The evidence for both claims comes in §35, where Kant states that in a judgement of taste “the freedom of the imagination consists precisely in the fact that it schematizes without a concept” (5:287). Schematism is a means of connecting pure concepts with appearances, and this is Kant's only reference to schematism without a concept. He may even be making some sort of analogy rather than noting a distinct sort of schematism not explained in the first Critique, because schematism without a concept is prima facie impossible.
But, even if we were to accept Allison's understanding of normativity, his reconstruction of the argument of the Deduction, and his claim that judgements of taste are a kind of schematism without a concept, all of which are controversial, we still cannot establish that the capacity to schematize without a concept is a requisite condition for the use of the faculty of judgement and is therefore normative. If this were the case, we would need to be able to schematize without a concept in order to use the power of judgement correctly; this kind of schematization would be essential to the operation of the faculty. Whatever interpretation of schematism without a concept is most plausible, it is mentioned only once in the Kantian corpus with no attendant explanation. It is implausible to attribute to it such a fundamental role.
IV.
Rejecting Allison's account, though, does not show that judgements of taste are non-normative. There are passages in which Kant refers to a ‘demand’ wherein he is clearly not referring to the way that judgements of beauty are made in everyday life. Kant refers again in §8 (5:214, 5:216) to the “demand” that judgements of beauty place on us, and in other places claims that judgements of beauty “require” (ansinnen) our assent (e.g., §37, 5:289, §39, 5:293). We are told in §40 that the feeling of pleasure in a judgement of taste “is expected of everyone as if it were a duty” (5:296). Finally, §59 may seem the most troubling for my reading of the text. There, Kant states:
[T]he beautiful is the symbol of the morally good, and … only in this respect (that of a relation that is natural to everyone, and that is also expected of everyone else as a duty) does it please with a claim to the assent of everyone else, in which the mind is at the same time aware of a certain ennoblement and elevation above the mere receptivity for a pleasure from sensible impressions, and also esteems the value of others in accordance with a similar maxim of their power of judgment. (5:353)
Some commentators claim that a connection between morality and judgements of beauty, including a shared normativity, is key to understanding the universally binding claim made by judgements of taste. On this view, since agreement about judgements of taste is expected of everyone else as “as a duty” and this is the case insofar as beauty symbolizes morality, the requirement to make correct aesthetic judgements derives from the moral law.Footnote 9 I believe that Kant's explanation of the way in which beauty symbolizes morality does not bear out this view, and in fact supports the conclusion that Kant did not hold that aesthetic judgements issue a normative demand.Footnote 10
The nature of the similarity between judgements of beauty and the morally good is obscure, and a subject of continued debate. It is sufficient for my purposes to establish two claims concerning that similarity: (1) Kant is not claiming that the subjective universality of judgements of beauty is the same as or is dependent on the normative, objective universality of moral judgements, and (2) there are more plausible alternative ways to interpret Kant's statement in the quoted passage.
It is first necessary to discuss the claim made in the passage. Kant is saying here that only insofar as the judgement of taste serves as a symbol for moral judgement does it claim the assent of others “as a duty” or involve assessment of others’ value. In order to understand this claim, we should place it in context by briefly elucidating Kant's account of symbolism, an indirect way of presenting concepts.Footnote 11 He holds that “to demonstrate the reality of our concepts, intuitions are always required” (§59, 5:351). That is, concepts must somehow be connected with objects of intuition in order for us to hold that they are actually instantiated. Empirical concepts are presented through example, and pure concepts of the understanding are presented through schematism. Concepts of reason are supersensible and do not admit of direct presentation through example or schematism. He begins §59 by reminding us of this, and of the impossibility of showing the reality of concepts of reason (i.e., ideas, such as God and freedom) directly through intuitions. “[N]o intuition adequate to them can be given at all” (§59, 5:351) because such concepts, as he puts it in the first Critique, “contain a certain completeness that no possible empirical cognition ever achieves” (A568/B596).Footnote 12 Because no empirical cognition is capable of fully representing ideas or concepts of reason, the actual existence of their objects can be neither verified nor disproven. But Kant goes on to explain that there is another way in which we can receive confirmation of their objective reality — through symbolism, a mode of presentation in which we have intuitions that “demonstrate the reality of” concepts by “presenting them indirectly rather than directly” (§59, 5:351). Symbolic presentation involves a similarity between the concept and the symbol in the “rule for reflecting on both and their causality” (Ibid.), not necessarily a similarity in their content. In symbolism, intuitions of objects that are presented to us directly can indirectly present another concept, even a supersensible concept, “by means of analogy” (§59, 5:352), offering a kind of confirmation of the objective reality of the concept. Symbolism is an indirect presentation of a concept in which there is an analogy between (1) the causes, consequences, or method of reflection of a concept, and (2) the causes, consequences, or method of reflection of an object of intuition, even though (1) does not have ‘marks’ corresponding to (2). A hand mill symbolizes a despotic state because, even though the two are not similar to each other, they both operate by exerting an overwhelming, grinding force on that which is within their grasp until it is converted into an amenable form (5:352). Similarly, a beautiful object indirectly presents the concept ‘good,’ because there is formal or structural similarity between the way that we reflect on the good and the beautiful. As Andrew Chignell puts it, symbolism gives us a sense for whether a supersensible concept such as the good is possible “by drawing an analogy between its relationship to something we know to be really possible, and the relationship between two other things that we know to be really possible.”Footnote 13
So, there is something analogous between the way we judge taste and the way we judge the morally good. But this does not show that judgements of beauty are normatively binding in the way that moral judgements are. When Kant explains the symbolism between beauty and morality, he says that both the principle for making judgements of the beautiful and the principle for making moral judgements are universal in the sense that they are “valid for everyone” and “knowable for all subjects” (§59, 5:354). When giving a more detailed explanation of the analogy, then, he makes no reference to a duty’ or ‘demand,’ instead emphasizing that they are both universal. This is because, rather than claiming that the act of judging aesthetically is itself a normative assessment, Kant is claiming that beauty and morality share a claim to universality, and that only when they are considered as a symbol of the morally good do judgements of beauty make a demand on others “as a duty.”
The basis for the universality of moral judgements allows us to make normative judgements in a way that the basis for the universality of aesthetic judgements does not. There are many clear disanalogies between the two kinds of universality, several of which are explicitly pointed out by Kant. Most obviously, the universal claim of moral judgements is objective, while judgements of beauty are subjective; moral judgements, and their corresponding claims on us, are based on concepts of reason, while judgements of the beautiful are not based on concepts. In order to issue moral judgements, which have normative force, we must hold that concepts of reason correspond to something objectively real — we must hold, for example, that there are things that are truly good, and that we are truly free. Aesthetic judgements, however, rest on the feeling arising from the free play of the faculties of imagination and understanding alone, not on concepts that are postulated as corresponding to objects. Moral judgements are based on the employment of a completely different faculty (the faculty of reason) and are related to concepts, so the two sorts of judgement are independent from one another and rest on wholly different bases.
Yet Kant does say both that morality can improve our aesthetic sense and that experiences of the beautiful and the sublime can contribute to our moral sense; if we take seriously his explanation of symbolism, we must conclude that there is at least one significant structural similarity between judgements of taste and of beauty. The appreciation of beauty requires a separation from merely private concerns and relationships to objects that is akin to the separation from merely private desires necessary for making a moral judgement — only in the first case, this separation is effected on the basis of a feeling, and in the second, it is made on the basis of reason and its universal laws. The appreciation of beauty requires the ability to recognize and set aside our interests and desires, and to take on a universal standpoint, as does the recognition of the morally good. Both Guyer and Allison suggest that Kant is proposing that, because of this structural similarity, beauty can serve as a propaedeutic to morality.Footnote 14 That is, beauty can lead us toward a fully moral outlook. They emphasize that it would be impossible for taste to prepare the way for morality in this manner if beauty were not autonomous from morality; if the universality of taste depended on having a moral outlook, then it could not help us to gain that moral outlook in the first place.
It is possible, then, to reconcile Kant's strong efforts to distinguish judgements of taste from other forms of judgement, including moral judgement, with his claim that beauty is the symbol of morality and with the other relationships he describes between beauty and morality. Beauty and morality preserve their independence from one another, yet the structure of judgements of beauty, which rely solely on an interaction between faculties of mind that is attained in the same way by all and that is independent of private ends or preferences, bears a resemblance to the structure of judgements of morality that allows them to serve as the symbol for those judgements. This reading offers further reason to hold that Kant's apparently normative language throughout the third Critique does not amount to a claim of normativity for judgements of beauty. If the experience of judging objects to be beautiful resembles that of judging morally in important respects and can serve as a propaedeutic to morality, then it makes sense that Kant sometimes tells us we are justified in demanding proper judgements of taste from others, even though these judgements cannot be seen as a precondition of moral reasoning or vice versa. Having taste is not a precondition for moral agency, nor is taste merely derived from moral foundations.
V.
Passages in which Kant refers to a demand made by judgements of taste or a duty to make proper judgements of taste, then, are inadequate to show that judgements of the beautiful are normative on his view. Some argue, however, that the normativity of taste is fundamental to Kant's broader account of judgement and cognition. Hannah Ginsborg, the chief proponent of this view, argues that, if I judge something to be beautiful, “I take it that everyone ought to respond imaginatively to the object as I do.”Footnote 15 Luigi Caranti claims that “the principle of taste can be seen as being at the basis even of the cognitive applications of reflective judgment, most importantly aimed at the formation of empirical concepts.”Footnote 16 In judgements of beauty, one takes “a normative attitude toward her mental activity,”Footnote 17 Ginsborg says, holding that others ought to experience the same play of imagination and understanding upon encountering the object. The content of the normative standard is unknowable because the object is not seen as falling under a specified concept. But because we think of our mental activity as being as it ought to be when making such judgements, we judge that others’ mental activity ought to conform to ours, and thus we make a normative universal judgement.
If this view is right, Kant's account of judgements of taste in the third Critique can be taken as an account of how our faculty of judgement, in making normative claims not bound by any particular rule, enables the formulation of all empirical judgements. In short, a judgement of the beautiful “consists in the fulfillment of a condition that is in some sense required for all empirical cognition.”Footnote 18 This is because, as Caranti puts it, “the particular harmony of the faculties that grounds the normativity of judgments of taste is … at the basis of any application of reflective judgment to nature.”Footnote 19 Judgements of beauty show how a subject can take her own “blindly habitual” associations between representations to “manifest conformity to a normative standard applicable to everyone,” i.e., how she can take such subjective associations to be universal.Footnote 20 If this kind of subjective universal claim can be made in judgements that are not matters of taste, then one can use these associations as the basis for the creation of empirical concepts.
Unless there is a basis for empirical concepts that is not a rule, it seems that we cannot acquire empirical concepts by seeing instances of objects that fall under them, because recognizing which objects can be subsumed under particular concepts requires prior possession of those concepts.Footnote 21 These concepts are, of course, not a priori, so we have no such possession of them.Footnote 22 On this reading, judgements of taste and empirical concepts are created via a similar “self-referential judgment”: one feels pleasure in the seemingly appropriate operation of the faculties of imagination and understanding as they represent objects as related to each other via some concept.Footnote 23 Although guided in the creation of these associations by habit,Footnote 24 the subject views her mental state as universally valid. Habit guides her to make certain associations and is thereby the source of a feeling that the faculties of imagination and understanding are operating correctly, despite the fact that the subject is not guided in her use of the faculties by a rule or explanation for why this is so.Footnote 25
While this account solves a problem for Kant's theory of concepts, it strays from the text and from familiar experiences of the beautiful. Ginsborg cites §65 (5:211) and §18 (5:237) as support for the claim that beauty's subjective universality is normative. In both of these passages, however, Kant does not make any use of specifically normative language, and in the second he even contrasts the necessity in the universality of judgements of beauty with the (clearly normative) universality of judgements of practical necessity. Nor does Ginsborg offer significant evidence to suggest that Kant made a link between judgements of beauty and those that underlie our process of empirical concept formation.Footnote 26 She argues, though, that two statements in §21 show that we are “entitled” to take our mental activity as normative: (1) without universal communicability of cognitions and judgements, these “would all be a merely subjective play of the powers of representation, just as skepticism insists” (5:238), and (2) since the universal communicability of a feeling presupposes a common sense, the latter “must be able to be assumed with good reason … as the necessary condition of the universal communicability of our cognition, which is assumed in every logic and every principle of cognitions that is not skeptical” (5:239).Footnote 27
§21 is meant to show that we have good reason to presuppose a common sense, which is a condition of the necessity of judgements of taste. It seeks to show that cognitions and judgements and the dispositions of the cognitive powers “suitable for making a cognition out of a representation” (5:238) require a common sense. The first passage claims that if it were impossible for us to communicate judgements and acts of cognition, as well as the basis for our mental activity of each sort, then they would involve merely private and fundamentally intransmissible responses to objects or to the interactions between our cognitive faculties themselves. Were this so, we would be unable to subsume particulars under concepts or engage in the cognitive activity that forms the basis for a judgement, for concepts and rules for judgement are, most fundamentally, criteria that are supposed to apply to our representations in all cases. Were they not such criteria, we could not even be assured that we were utilizing the same standards in all cases of judgement and concept application. We would have to hold, then, that there is no way to determine whether our own responses to objects, both cognitive acts and other acts of judgement, represent those objects in the appropriate way, and would consequently be forced to accept scepticism.
Further, though, Kant says, universal communication must apply in the case in which the faculties of imagination and understanding are disposed not toward cognition, but rather toward the “animation” of the faculties of imagination and understanding that is determined by “feeling (not by concepts)” (5:238–5:239). Here he can only be referring to the free play and the feeling of pleasure that occur in judgements of beauty. Ginsborg takes Kant to be suggesting here that the act of reflective judgement that takes place in the formation of empirical concepts is also of this kind, because in this case we also have a general attitude toward our cognition and no way of supporting it through concepts. So, Ginsborg claims, Kant is here justifying our entitlement to the supposition of our own mental activity's universal validity in these reflecting judgements as well as in reflecting judgements of beauty.Footnote 28 Because Kant does not state here that empirical concept formation and judgements of beauty are analogous in this way, this passage could only provide such justification if he had elsewhere suggested that they are analogous or had elsewhere characterized the act of reflecting judgement involved in empirical concept formation as one resting on animation of the faculties determined by a feeling of this kind. This is something he does not seem to have done.
At the end of §21, where the second passage quoted by Ginsborg appears, Kant is indeed saying that, because universal communicability presupposes a common sense and we must assume universal communicability in order to make both cognitive judgements and those of the beautiful, we have good reason to presuppose a common sense. Kant does not suggest, however, that the mental activity giving rise to this sense is the same in judgements of beauty as in other kinds of judgement or cognition. In order to interpret the passage as Ginsborg does, we must assume that Kant is making two important unstated claims: (1) that the character of the universality in judgements that lead to the creation of empirical concepts and the character of the universality of judgements of beauty is the same, and (2) that this universality is normative. Neither of these is suggested by this section, which seeks only to establish that the animation of the powers that underlies cognitive judgements and judgements of taste presupposes a common sense. The next section goes on to say that the judgement of taste is offered “as an example” of the common sense and “could demand universal assent just like an objective [judgement] — if only one were certain of having correctly subsumed” under a principle or rule for everyone (5:239). Yet it is impossible to be certain that we're judging correctly. This is true not only because we cannot be certain that we are applying a rule correctly, but more fundamentally, because the rule or principle is in principle undiscoverable in the case of judgements of taste. This marks a fundamental difference between judgements of taste and normatively binding judgements, such as those of logic and morality.
Like Ginsborg, Caranti holds that we can resolve the puzzle of empirical concept formation by appeal to judgements of taste. He suggests that the Deduction shows the principle of taste itself is the basis from which all other judgements, including those that form empirical concepts, are derived. As he rightly points out, in §35, Kant claims that judgements of taste are like other judgements in that they involve the harmony of the faculties, but unlike other judgements in that they harmonize without a concept. This, Caranti claims, means that they involve only a kind of paradigmatic harmony and must therefore be the basis for all harmony of the faculties. But the fact that judgements of taste rest only on the sensation of the reciprocal animation of the imagination and understanding and thus only on the conditions for the operation of the power of judgement (5:287) does not show that the animation of the powers without a concept is the source of all other operations of the power of judgement. The animation of the powers without a concept, involved in judgements of taste, and the animation of the powers relating to a concept, involved in cognitive judgements, could easily be independent functions of the power of judgement, neither depending on the other. We can imagine a scenario in which someone is incapable of making pure judgements of taste, yet entirely capable of making use of the power of judgement in other ways. One might, for example, be incapable of separating her private interests from her aesthetic judgements, so that her judgements of taste are all impure. There is no reason to think that such a person would have difficulty making other uses of the faculty, for example, recognizing that this object instantiates an empirical concept, or even forming an empirical concept. There is no reason that one who is poor at recognizing the difference between things that please her for idiosyncratic reasons and things that please her because of (shared) features of her faculties will also have difficulty engaging in reflective judgement involving concepts.
If we step back from concerns about specific passages in favour of considering the third Critique as a whole, concern about the faithfulness of this kind of reading remains. Even when speaking broadly about the significance of the universality of aesthetic judgements, Kant does not suggest this fundamental role for them. When arguing that these judgements are “something remarkable … for the transcendental philosopher” and reveal “a property of cognition that without this analysis would have remained unknown” (§8, 5:213), Kant makes no indication that we should view the analysis of beauty as offering support or explanation of the foundations underlying our ability to judge conceptually.
Further, Kant's discussion of empirical concepts suggests that they are formed in a manner that does not mirror the formation of judgements of taste. In the published Introduction, Kant says that we make the reflecting judgements that form empirical concepts only after subsuming them under the idea of the purposiveness of nature (5:180). We can only represent nature as unified by a system of empirical laws if we represent it “as if an understanding contained the ground of the unity of the manifold of its empirical laws” (5:181). So, only after considering the objects we encounter as though they were unified by a conceptual structure do we seek concepts under which to subsume them. Empirical concept formation is thus quite different from judgements of the beautiful: in the former, we seek unity through a particular concept, while in the latter, we seek unity independent of any concept.
Ginsborg, however, holds that judgements of taste are grounded in the idea of the purposiveness of nature, so empirical concept formation and judgements of taste share the same kind of claim to universality.Footnote 29
It is easiest to evaluate this position if we briefly consider the idea of the purposiveness of nature and how it arises in Kant's account of judgement. In the first Critique, Kant explains how the concepts of the understanding serve as a priori principles for determining judgement, which applies concepts that we possess to the manifold of intuition. Determining judgement, then, does not have its own corresponding a priori principle, and does not function as an independent faculty (Introduction IV, 5:179). But in reflecting judgement we are faced with a representation and must locate a concept that corresponds to it, so concepts of the understanding cannot guide the application of this sort of judgement. A central task of the third Critique, Kant claims in both Introductions, is to explicate an a priori principle for reflecting judgement. This task is complicated by the fact that Kant identifies two sorts of reflecting judgement (see esp. First Introduction V, 20:211): logical (the location of concepts that unify or systematize groups or classes of objects or representations, which involves the comparison of such objects or representations), and formal (the location of concepts that apply to single objects or representations, which involves reflection on a single object or representation). Judgements of taste are, of course, a species of formal reflecting judgement. Ginsborg suggests that empirical concepts are also formed in an act of formal reflecting judgement — one that takes “what my imagination actually does in the perception of an object,” i.e., a given representation or way of grasping what is given to us in experience, “to serve as a rule determining how that, and other such objects, ought to be perceived.”Footnote 30 When it first feels right to represent an object as exemplifying a particular rule, an empirical concept is created, so that the creation of the concept is also its first application.Footnote 31
Kant states that the principle of the purposiveness of nature (i.e., the principle that the whole of nature is organized by laws that are purposive for our cognitive faculties) is “merely a principle for the logical use of the faculty of judgment” (20:214). This principle must be presupposed for “the cognition of nature, but not merely as nature in general, but rather as nature as determined by a manifold of particular laws” (5:182). Yet Ginsborg claims that it is also a principle for formal reflecting judgements, including aesthetic judgements and those that she believes underlie empirical concepts,Footnote 32 because she holds that (1) judgements of beauty and the formation of empirical concepts are both “due to an exercise of empirical conceptualization”Footnote 33 (in the latter, the application of an empirical concept, and in the former, the free play of imagination and understanding without the application of a determinate concept), and that (2) conceptualization cannot take place unless we assume that nature is organized in a hierarchical fashion that corresponds to a manner of systematization with a certain feeling of ‘rightness’ or ‘fit’ for cognition. The principle legitimizing judgements of taste, then, “must be the same principle which underlies, at the most general level, our capacity to bring objects under empirical concepts: and this is the principle of the systematicity of nature.”Footnote 34
But Ginsborg says little about what it means for a judgement of beauty to be both free from concepts and governed by the principle of nature's purposiveness or systematicity. It is difficult to see how this principle, which is a claim concerning the origins of nature and objects within it, could be a necessary presupposition for judgements that, Kant insists, are grounded solely on the basis of a response to a representation of an object without regard to its origin, structure, or purpose. If judgements of beauty are grounded therein, all acts of reflecting judgement are guided by the principle of the purposiveness of nature. If this is true, in order to find something beautiful, the judging subject must presuppose that nature and the laws that govern it are the product of a will that has specified the forms of objects represented as beautiful in a way that is suitable for bringing about the free and harmonious play of the faculties of imagination and understanding in subjects.
But we need not presuppose that nature or natural objects are purposive for our cognitive or psychological needs in order to experience the free play of the faculties. Indeed, one reason that Kant considers judgements concerning natural objects to be paradigmatic judgements of beauty is that our judgements of such objects are not bound up with our knowledge of their purposive nature. If we make judgements of taste within the context of an assumption that the objects of our judgements are part of a system of hierarchically organized laws made to conform to our cognitive needs, we are assuming much more than Kant intends when he says that judgements of taste are purposive without having purpose or end (§10, 5:220); judgements of taste are presented as though they have ends, in a way that is uniquely pleasing to us without having any identifiable end. Being purposive for our cognition is an identifiable end, and one that is ruled out by Kant's explicit, repeated insistence that in judgements of taste there can be “no representation of an objective end, i.e., of the possibility of the object itself in accordance with principles of purposive connection” (§11, 5:221). While indeterminable, the principle of nature's purposiveness is more conceptually loaded than purposiveness without end, and the latter is clearly all that Kant intends.
Konstantin Pollok has also recently claimed that this principle governs judgements of taste, identifying purposiveness without end and the principle of the purposiveness of nature.Footnote 35 He holds that this principle is “a necessary presupposition for our reflecting power of judgment.”Footnote 36 Yet his own account of judgements of taste do not suggest that they must be governed by this principle, stating that they “invoke some kind of purposiveness”Footnote 37 and that it is “hard to understand”Footnote 38 how this principle operates in the case of judgements of taste. Ultimately, Pollok concludes that aesthetic judgement governed by this principle is meant “to celebrate … the aptitude of our higher faculties for an indeterminate reflection on nature (and nature like artifacts).”Footnote 39 It is possible, though, to distinguish between the purposiveness of nature as a law-governed system and a wholly subjective purposiveness that recognizes an object as supremely well suited to our faculties despite the fact that it is not so in virtue of any actual purpose that it fulfills. Kant does not claim that the two are the same, and the thinner understanding of purposiveness is consistent with reflection on, indeed appreciation for, the suitability of our faculties for grasping natural objects.
The suggestion that judgements of taste presuppose the principle of the purposiveness of nature also does not match the phenomenology of aesthetic judgement. When we observe that something is beautiful, we do not engage with it as something produced for our sake, but rather as something that pleases in a very particular way — no matter its purpose or origin, that which is beautiful satisfies those faculties that are capable of cognition without satisfying any particular aim we have in cognition. Beautiful objects satisfy in a way that only something without any discernible purpose can, because a central aspect of what makes the pleasure in them unique is that they seem ideal, yet there is no way to fully explain why. Consider the response when one is viewing a nautilus shell or complex crystal formation: we are struck by its intricacy and delicacy, by its seeming somehow exactly as it is meant to be, though we cannot identify a standard by which we measure it. We need not presuppose anything in order to view such an object as something that pleases in a remarkable way that is suitable for our faculties, and something's suitability for our faculties is not the same as the presupposition that it has the end of being suitable for our faculties.
Yet it is clear that Kant believes that judgements of taste have a foundation in another idea, that of the supersensible substratum of humanity (§57, 5:340). One could argue that this shows that I have an overly restrictive interpretation of Kant's claim that judgements of beauty are free from concepts. Kant claims that the idea of the supersensible substratum of humanity resolves the antinomy of taste by showing how judgements of taste can be free from concepts while remaining grounded in an intersubjectively valid way through a concept that cannot be presented directly. There is controversy about the plausibility of Kant's claim that his account of judgements of taste requires a foundation in the idea of the supersensible.Footnote 40 Whether or not we accept this claim, though, it does not lend credence to the suggestion that judgements of taste are based on the idea of the purposiveness of nature. Unlike the idea of purposiveness, the idea of the supersensible substrate does not dictate an end. Instead, it claims that both judgements of taste and the subjective purposiveness of nature are grounded in something that is an affirmation of the shared underlying capacity for cognition possessed by all of humanity and yet is at the same time itself “indeterminable and unfit for cognition” (§57, 5:340).
VI. Conclusion
I have argued that Kant's text itself does not suggest that judgements of taste make a normative claim on us. Neither the contention that judgements of taste are a species of schematism without a concept, made by Allison, nor the contention that judgements of taste possess normativity that derives from the demand of the moral law, succeeds in locating a textual basis for a normativity claim. Further, the argument for a normative reading of judgements of taste as a way of clarifying Kant's account of empirical concept formation fails to account for the explicit and substantive distinctions separating these two phenomena.
However, a non-normative reading of the text allows us to preserve the idea that we should all judge in the same way in matters of taste, without forcing us into the implausible position of saying that such judgements are binding in the way that those of morality or logic are. In those cases, when others fail to concur with our judgements, we question their ability to reason or to participate in human society in a meaningful way. When others fail to concur with our judgements of taste, we question their ability to recognize which features of objects are pleasing, or wonder whether they are actually paying attention to what they perceive. Such people may be unobservant, or may lack the ability to distinguish between features of their experience that arise from personal preferences and features that are universally projectable. Failure to judge accurately in these cases even raises real concern about a person's ability to accurately separate private concerns from concerns that are basic to our nature as human beings in other contexts (such as the moral context). Yet one's failure to judge accurately in these cases does not lead us to ostracize her from society or label her cognitively disabled. One's capacity to make truly universalizable pure judgements of taste can, then, have implications for one's capacity to make normative judgements, but taste is itself an independent capacity.
This demand for agreement differs in character from normative demands because it cannot adduce grounds and because the capacity to judge in matters of taste is not necessary for the coherent utilization of our faculties or for the realization of our cognitive aims. The demands of the moral law and of logic, however, as Allison well explains, must be met if we are to think and live coherently.
The notion of analyzing beautiful objects does presuppose that we have intuitions about what kinds of pleasurable feelings will be shared with others and what kinds relate only to facts about ourselves and our experience. Kant intends the account of universal subjective validity to capture the former kind of pleasure. The account explains how something can command our attention in a deeply pleasurable and rewarding way in virtue of a quality that does not seem to depend on an aim, goal, or interest of ours. In judging something to be beautiful, we must at least be attempting to take on a “broad-minded way of thinking” that leaves aside private concerns (§40, 5:295). This is what Kant means when he says that we must attempt to make judgements of beauty “from a universal standpoint” (Ibid.). We can only understand Kant's account of our experience in making aesthetic judgements in this way if we abandon a normative reading of his claim that pure judgements of taste are universal. Such an interpretation distorts the text and forces us to take Kant to claim that we ought, indeed that we are required, to take pleasure in certain objects. Kant makes no such assertion. The view that he truly offers us is one that our aesthetic experience should lead us to appreciate and seek to understand fully and sympathetically.
