1. Is Kant’s Theory of Art a Theory of Modern Art?

It all depends upon what you mean by Kant’s theory of art, and what you mean by ‘modern’.

To start with the latter, ‘modern’ can mean two different things, so ‘modern art’ can likewise mean two different things. On the one hand, deriving ultimately from the Latin hodiernus, ‘of today’, it means ‘pertaining to the present and recent past, as opposed to the remote past’, ‘characteristic of the present and recent times, not old-fashioned and obsolete’, and the like.Footnote ¹ In other words, ‘modern’ means ‘current’ or ‘contemporary’. This is, I believe, the sense of ‘modern’ in ordinary usage. In this usage ‘modern’ is an indexical, indexed to the moment of utterance by some speaker. Modernity in this sense can admit of degrees, depending on the temporal interval between the current moment and the time of the object or event under discussion. I might say, for example, that the architecture of I.M. Pei is modern, but that of Steven Holl is more modern — Pei did his main work half a century ago, Holl has done his in the last two decades, and ‘half a century ago’ and ‘the last two decades’ are both indexed to the moment when I am writing, which will itself recede in time. Thus we can refer to past assertions of modernity that were true when asserted but would not be now: Gothic architecture was modern — in the twelfth century, compared to Romanesque — but no longer is, and Beethoven’s music was modern — at the beginning of the nineteenth century — but no longer is — when compared to that of Philip Glass or John Adams.

But ‘modern’ can also refer to the style of a particular time, which may have been modern in the first sense, that is, current, when it first emerged, but which — now — is of the past, tied to a particular moment that is now gone rather than to the present moment, and which can be superseded by something newer, called ‘contemporaneous’.Footnote ² Beginning with architecture, several arts enjoyed — or suffered — a ‘postmodern’ phase after their ‘modern’ phase, but ‘postmodern’ would not have any sense in the first, ordinary sense of ‘modern’, since what is contemporary is always modern in that sense, so if it used ‘modern’ in the first sense ‘postmodern’ would have had to connote what was then future, not what was then current. ‘Postmodern’ made sense in contrast to the modern art of a previous moment, and in that sense the postmodern could itself be — and at least in the case of architecture has been — superseded and is no longer modern in the first sense, that is, current. Some modern movements in this second sense, descriptive of the style of a particular moment, might last longer than others: for example, modern dance as a style initiated, let’s suppose, by Martha Graham, might still also be contemporary, in a way that the modern architecture that was superseded by postmodernism is not, but it might still become at some future time a style of the past.

We might say that ‘modern’ in the first sense is a term pertaining to McTaggart’s A-series, in which time is represented as moving and past and future are always relative to some present or ‘now’, while ‘modern’ in the second sense pertains to his B-series, in which moments are forever related as earlier than or later than each other.Footnote ³ In the A-series, modern art is the art of the current moment, the always changing present, and is itself (or can be) always changing; in the B-series, modern art is the art of one moment, forever later than, say, classicism or romanticism, but forever earlier than postmodernism.Footnote ⁴ We might also say that ‘modern’ in the first sense is an indexical linked to the current, ever changing moment, while ‘modern’ in the second sense is a descriptive term or ordinary adjective that happens to be linked to some moment in the past, as for example ‘colonial’ in American history refers to conditions and events prior to 1776 no matter when it is used. For sake of simplicity, in what follows I will use the term ‘current’ to convey the first sense of ‘modern’, and ‘modernist’ to convey its second sense. I hope that usage will be more pleasing than neologisms like ‘modern-indexical’ and ‘modern-adjectival’ or ‘modern-now’ and ‘modern-then’.Footnote ⁵

My argument will be that of course Kant has a theory of modern art in the sense of current art, namely his theory of genius; indeed, his theory of genius implies that successful art must be current in some way or other (although also related to classic art in some way or other, which I will not explore here). But he does not have a theory of modern art in the sense of modernist art, although his theory of beauty has caused some to enlist him in the cause of some modernist art, namely, formalist art (especially painting). But then again, it is not clear that at the level of abstraction inhabited by philosophers, and properly so, there is anything such as a theory of modernist art, although various episodes in the history of modernist art have certainly stimulated philosophical theories of art in general, whether modernist or not.

From the perspective of art historians, my thesis might be considered ‘essentialist’ rather than ‘contextualist’: it maintains that at a sufficient level of abstraction, art has always been doing the same thing, rather than doing essentially (pardon me) different things in different temporal and cultural contexts, and by so doing risks effacing important differences among the practices and products of the arts in different times and places. I have no quarrel with the supposition that it is a vital part of art history to emphasize the latter, but I think the task of philosophy, including philosophy of art, is precisely to distinguish between underlying commonalities and superficial differences. Such an essentialism would not be refuted by the claim that there was no common name for the arts prior to a certain point, say the mid-eighteenth century, as argued by Paul Oskar Kristeller,Footnote ⁶ since people can certainly be operating with a concept without a name for it, especially if a concept just is comprised by some common practices, as argued by Wittgenstein, and does not need a name at all. Philosophy’s work certainly includes digging past differences — or absences — of terminology to discover commonalities of practice. Perhaps it goes even further, to the explication of human potential even prior to the emergence of actual practice; for example, if, as for Kant, the gist of art is the imaginative combination of sensory and intellectual materials, then for a philosopher art has been a human possibility as long as humans have had imagination and intellect.

2. Kant’s Theory of Genius as a Theory of Current Art

Kant’s theory of genius is a theory that successful art is always in some way or other current, novel art. This follows from his conception of genius as ‘exemplary originality’: the work of the artistic genius always involves some departure from the work of his predecessors, or is in that sense original, and is an example of originality and a stimulus to originality to those of his or her followers who themselves have some spark of genius, as opposed to mere ‘copyists’. (Kant writes as if, and no doubt assumes, that the artistic genius is always a biological individual, and a male to boot, but neither of these assumptions is essential to his theory. The artistic genius could certainly be female, or non-binary, or a group.) For this reason, the history of art — any art — is a history of constant ferment rather than of linear, cumulative progress, at least as long as it attracts geniuses rather than plodders. In opposition to Alexander Gerard, who found forms of genius in both art and natural science, Kant reserves the term for the case of successful art only.Footnote ⁷ This difference is not in fact the one that Kant stresses in his explication of the difference between fine art and natural science, but it follows from his conception of genius.

Kant’s theory of genius as a theory of artistic production is the mirror image of his theory of aesthetic judgement as a theory of aesthetic reception. The explanation of our pleasure in beauty and the basis for the justification of our judgements of taste is that contemplation of a beautiful object triggers a free play of imagination and understanding, felt as a pleasurable state that one person can then impute to others through a judgement of taste. This imputation is legitimate because imagination and understanding can and indeed, Kant supposes, must be assumed to work pretty much the same way in everyone, at least under optimal conditions. These conditions are summed up as the freedom of all parties from distortion by the practical and personal interests in the existence of the object of taste that would be connected with a conception of the purpose of the object entailed by its subsumption under some determinate concept. The explanation that our pleasure in beauty is due to the ‘animation of both faculties (the imagination and the understanding) to an activity that is indeterminate but yet, through the stimulus of the given representation, in unison’, is Kant’s ‘key to the critique of taste’ (CPJ, §9, 5: 219, 216), and the argument that everyone’s imagination and understanding must work in a similar fashion in the case of aesthetic judgement because they must do so for the communication of any cognition at all is the basis for his justification of the claim of anyone to speak with a ‘universal voice’ in her judgements of beauty (CPJ, §8, 5: 216) in the form of a presupposition of a ‘common sense’ (CPJ, §21, 5: 238) and the Deduction of Judgments of Taste (CPJ, §38, 5: 289-90).Footnote ⁸ Strictly speaking, this is Kant’s account of judgements of beauty, while his account of the experience and judgement of the sublime is more complicated; but since he also thinks that fine art cannot directly trigger the experience of the sublime, although it can provide beautiful representations of the great vistas and powerful objects in nature that do, we can afford to ignore the sublime in this discussion. (Some recent visual artists, for example, the painter Barnett Newman and the sculptor Richard Serra, the one perhaps a modernist artist and the other a current artist, might be thought to have challenged this claim of Kant’s.)Footnote ⁹

The theory just described is offered by Kant as a general theory of beauty and aesthetic judgement, applying to beauty and the judgement of it in both nature and art. Kant applies it in the first instance to the case of natural beauty, for example the beauty of flowers, birds, and mollusks, and then to cases of decorative art, such as ‘foliage for borders or on wallpaper’ and ‘music without a text’, which ‘signify nothing by themselves’ and ‘do not represent anything, no object under a determinate concept, and are free beauties’ (CPJ, §16, 5: 229). Paradigmatic cases of fine art, however, rather than decorative art and music without text, are more complicated: they do involve the representation of content, namely ‘ideas of reason’, for Kant chiefly moral ideas such as the ideas of good and evil or heaven and hell, although these ideas can never be completely represented in the sensory forms employed by art and thus count as indeterminate.Footnote ¹⁰ The indeterminate sensory or imaginative representation of great ideas of reason is what Kant calls an ‘aesthetic idea’:

or other artistic medium — although for Kant literature, in his terminology poetry (Dichtkunst), which provides ‘nourishment to the understanding in play’ and gives ‘life to its concepts through the imagination’ (CPJ, §51, 5: 321) ‘claims the highest rank of all’ the arts (CPJ, §53, 5: 325), precisely because among all the arts it is best at playing imaginatively with great intellectual content. Its free play with intellectual content thus constitutes the great difference between natural beauty and the beauty of merely decorative art, on the one hand, and the beauty of fine art, on the other, although once having introduced the concept of aesthetic ideas Kant actually asserts that all ‘[b]eauty (whether it be beauty of nature or of art) can in general be called the expression of aesthetic ideas’ (CPJ, §51, 5: 320). Since ordinary beautiful objects in nature such as flowers, mollusks, and birds do not, as far as we know, have any ideas to express, nor, again as far as we know, are they speech-acts of some author of nature (because for Kant although the idea of an author of nature is natural to us, the idea of God too is only an idea of our own reason),Footnote ¹¹ this assertion must itself be an expression of our tendency to treat nature as if it were art by reading intellectual content into it — an intellectual version of what John Ruskin called the ‘pathetic fallacy’.Footnote ¹²

Be that as it may, the core of Kant’s theory of genius is then that it is the ability to create and communicate a free play with great intellectual or rational content, that is, to invent imaginative forms for the communication of such ideas — a special form of communication, however, which because of the requirement that the recipient or audience for fine art must itself experience free play in order to experience pleasure, must be the expression of free play in the mind of the artist that in turn triggers free play and not mere replication in the mind of the audience. ‘The mental powers, then, whose union (in a certain relation) constitutes genius, are imagination and understanding’; in more detail, Kant says of genius:

To be sure, there are ’mechanical’ rules that artists in particular media or genres must follow, for example, in oil painting slow-drying paint can be applied over fast-drying paint but not vice versa if crackling is not to result (though maybe in some case an artist might want that result) or that in iambic pentameter each line must have five beats (although maybe sometimes a poet will want to disrupt that expectation) or that a waltz must be in 3/4 time. But following such rules is merely a necessary condition for artistic success, not a sufficient condition for artistic success. That requires genius, not rules, which is why Kant thinks that even great scientific success is not a matter of genius — for contrary to Gerard, Kant assumes that even the greatest scientific discoveries can not only be followed but also discovered by following rules (though this certainly seems implausible).

That difference is the import of the first of the four clauses in Kant’s definition of genius. The second clause states that an artist must, obviously enough, have a concept of the kind of object she is trying to produce — a painting, or more particularly a portrait or an historical painting; a poem, but more particularly a lyric or an epic; and so on — but that this concept will not fully determine the ‘material’, the imaginative object that is to be produced, that is, the particular painting or poem that will be a public object, accessible to an audience for the work. (‘Material’ here should be taken broadly, to include the words of a poem or notes of a piece of music as well as the pigments of a painting or the clay of a sculpture.) The third clause specifies that the content of the work, in contrast to its material, will be an aesthetic idea, that is, the imaginative expression of an idea of reason. Artists do not invent the rational ideas that are the contents of their aesthetic ideas: no particular artist invented the ideas of good or evil, or of the pride that goes before a fall, but artists do invent the imaginative ways in which to represent such abstract ideas — Dickens certainly invented Pip as his vehicle for exploring the idea of the pride that goes before a fall in Great Expectations, and if Milton did not invent the idea of Satan, he invented the form that Satan takes in Paradise Lost. Finally, the genius puts aesthetic form and aesthetic content together in the free play of her own imagination and understanding, although in the form of the ‘unsought and unintentional subjective purposiveness in the free correspondence of the imagination to the lawfulness of the understanding’ and through the material work of art (second clause) communicates this state to her audience — which, as already suggested, must consist in triggering a free play in the minds of the audience rather than just replicating her own state of mind, although the states of mind in the audience must somehow be relevantly related to the work of the artist, and not just day-dreaming while in its presence.

My claims that Kant’s theory of genius is the mirror image of his theory of aesthetic response, and that the free play in the mind of the artist must trigger free play rather than mere replication or repetition in the mind of the audience, are inferences from Kant’s premises, not what he explicitly says. But he does explicitly discuss the relationship between the genius and subsequent artists, and here is where he implies what I am calling his theory of current art, that is, a theory of constant ferment in change but not linear progress in any art. Kant actually discusses this relationship in the several sections preceding the introduction of the concept of aesthetic ideas in §49. Thus, in §46 he precedes the quadripartite definition of genius of §49 with this tripartite definition of genius, in which he characterizes it as exemplary originality: Genius is

The first sentence continues Kant’s polemic with Gerard, his assertion that only art and not science requires genius; the second sentence echoes Kant’s general claim that judgements of taste, now including judgements about art, must be intersubjectively valid, but then turns to the relation between the artist and other artists, specifically followers or students, rather than to the relation between artist and audience more generally; and this topic is continued in the third sentence. Kant argues in the last clause of 3) that the artistic genius cannot communicate to his followers or students precepts by which they could produce works as great or successful as his own, because he has not produced them by following any rules (other than the ‘mechanical’ rules for proper use of the relevant media, which of course can be taught) — not just because the artist does not know what rules he might have followed in producing the work, but because there are no such rules, and there cannot be if the followers or students are themselves to produce free products of their own imaginations, as they must if they are to be geniuses and not mere plodders. The statement in sentence 2) that the exemplarity of the work of genius makes it an object of ‘imitation’ (Nachahmung) for others might seem to conflict with this. However, in the next section Kant clarifies his point by means of a distinction between ‘imitation’ (Nachahmung) and mere ‘copying’ (Nachmachung):

So by ‘imitation’ — perhaps not the best translation of Nachahmung — Kant does not mean merely replicating what someone else has done, but rather being inspired by the originality of the predecessor or teacher to do something original of one’s own, although no doubt inspired in some way by the specifics of the work of the former to create something that stands in some relevant relation to that work — as say Manet may have been inspired by Velasquez,Footnote ¹³ or Mozart by studying the counterpoint of Bach, which he certainly did not replicate. Of course, people do merely copy the works of others, or even, as Kant calls it, ‘ape’ them, ‘cop[ying] everything, even down to that which the genius had to leave in, as a deformity, only because it could not easily have been removed without weakening the idea.’ Sometimes people do this with no pretense of originality, such as copying old masters in museums just to learn their technique, or merely for their own pleasure; sometimes they make forgeries with the intent to deceive; but sometimes they copy the works of a genius, no doubt with some variation, because of lack of their own imagination. One failure in art is thus ‘aping’ (Nachäffung), copying what has already been done thoughtlessly and without inspiration. An alternative failure is what Kant calls ‘mannerism’ (Manieren), meaning by this not the specific movement in the history of Italian painting after the High Renaissance, but rather ‘another sort of aping, namely that of mere individuality (originality) in general, in order to distance oneself as far as possible from imitators, yet without having the talent thereby to be exemplary at the same time’ (CPJ, §49, 5: 318). Exemplarity without originality, or aping, and originality without exemplarity, or mannerism, are both artistic failures: a work of artistic genius must speak with a universal voice, that is, be accessible to all, at least under optimal conditions (not only must all involved be disinterested, but they must also know the relevant languages or other features of the media, perhaps relevant mythology, and no doubt more), but in a way that stimulates the free play of imagination and understanding for all; and then specifically for further artists, the work of a previous genius must stimulate their own originality, but also the use of it in an exemplary way if their work is to be artistically successful.

Of course the complete history of art, or the raw material for the history of art, is littered with the works of copyists and mannerists; they fill the storerooms of museums, and in some cases the public spaces, as in the Prado where the galleries of Velasquez and Goya are followed by many rooms of Velasquistas and Goyistas. But the history of art in the sense of the story or narrative we construct out of its raw materials, insofar as it is a history of works of genius, or of the works that stand the test of time,Footnote ¹⁴ is a history of works of exemplary originality inspiring other works of exemplary originality, and so on ad indefinitum. It is in this sense that, according to Kant’s theory of genius, the history of art proper must be a history of ever-new current art: no work of genius can be the same as any previous one, and each work of genius must involve some novelty in both form and content — if not at the level of ideas of reason, which may not change, then at the level of aesthetic ideas, which can and must. Of course, new genres and media, such as photography, cinema, television, computer games, etc., also enter into the history of art, but even apart from that, artistic genius must produce a history of ferment and revolution. Anything less would be a history of mere copying and mannerism.

Now Kant may assume that all aesthetic ideas have some moral idea for their content, and that the moral ideas are fixed, so the content of great art is not itself a source of artistic ferment. We will no doubt differ from him on this point, holding that moral ideas are variable across times and cultures, and also that art may respond to cultural, social, political, etc., ideas and developments that are not strictly moral. Indeed, nature itself and what we mean by ‘nature’ is also undergoing constant change. To the extent that art has intellectual content and responds to change in human affairs and nature more broadly, change in art is guaranteed. So Kant’s theory of art is certainly, at its level of abstraction, a theory of current art. Our remaining question is whether it is also a theory of modernist art, that is, whether Kant’s theory anticipates any the main movements of twentieth-century art known as modern or modernist art, and their accompanying theories (at the level of art history or criticism).

Is Kant’s a Theory of Modernist Art?

My question now is what would be a theory of modernist art, and could Kant’s theory be one? Precisely because Kant’s theory of art insists upon content as well as form, and on free play between them, Kant’s theory of art is certainly not a theory of formalism, if that is what is meant by a theory of modern art. But more generally, at the level of philosophical abstraction, there may be no such thing as a philosophical theory of modernist art at all. I suppose there is a difference between a ‘theory’ of art, or some particular art, and a ‘philosophy’ of art, properly only of art in general; for example, an architectural theory is a program, in harsher terms propaganda for a particular style or method of building, while a philosophy of architecture is an analysis of the nature of building, of the role of buildings in human life, etc. The former kind of theory will certainly be ‘contextualist’, while on my account the latter should be ‘essentialist’. In terms of this contrast, there are theories of art occasioned by modernist art that are meant to defend or advocate the difference between such art and prior art, but philosophical theories of art that have been occasioned by particular works of art in new styles or genres or media, ‘modern’ or modernist in their time, are meant to accommodate such works of art precisely by showing that in spite of their differences from previous art they do what art has always done. At the philosophical level, there may be current theories of art occasioned by modernist art, but they are not theories of modernist art in particular. And then my answer to the general question of whether Kant’s theory of art can be considered to be or is compatible with a theory of modern art is that, sure, it might be a philosophy of modernist art, the art of the recent past, as much as any philosophy of art might be, just because these are all current philosophies of art but not art-historical or critical theories of modernist art in particular.

We have to return to two points in Kant’s initial analysis of aesthetic experience and judgement before we can come back to his philosophy of fine art in particular. One point is that insofar as Kant’s initial analysis of aesthetic judgement is an analysis of judgements of beauty, it suggests that insofar as fine art is subject to aesthetic judgement it will be so because it aims at or achieves beauty; and this of course can seem contrary to the spirit of some modernist art and much of its rhetoric, for example Marcel Duchamp’s polemic and practice against merely ‘retinal’ art, art aimed to look beautiful.Footnote ¹⁵ The other point is that in his exposition of judgements of beauty, Kant explicitly asserts that the beauty of several kinds of art lies in its form, and this has been taken to show that he anticipated a major tendency of modern art, especially modern visual art, namely ‘formalism’; so Kant was understood at least by Clement Greenberg, and by many others.Footnote ¹⁶ Thus, in the ‘Third Moment’ of the Analytic of the Beautiful, Kant goes from saying that the object of beauty — that is, the quality in objects in virtue of which we judge them to be beautiful — is ‘subjective purposiveness’ or ‘merely formal purposiveness’ (CPJ, §12, 5: 222) to saying that what is essential in beautiful art in particular is ‘the purposiveness of the form’ (CPJ, §13, 5: 223). Thus:

This can sound like an advertisement for some of the ‘formalist’ art of the early or mid-twentieth century (modernist art), such as the all white paintings of Kazimir Malevich or the all black paintings of Ad Reinhart, or perhaps the twelve-tone-row compositions of Schoenberg and his school — although Kant’s claim that in music what is essential is the composition and not the agreeable tones of instruments might also apply to some of the scores of J.S. Bach, such as ‘The Art of the Fugue’, which do not specify any particular instrumentation and which have been performed with many different types of instrument (although there is no evidence that Kant knew this work of Bach, or any work of Bach).

But there are several points to notice about Kant’s supposed formalism. First of all, as I have argued for many years and have yet to be convinced otherwise, his inference from the premise that beauty consists in the form of purposiveness to the conclusion that it consists in the purposiveness of form is fallacious; there is nothing to justify it.Footnote ¹⁷ But even leaving that objection aside, another point here is that Kant says that in making his Albertian claim about visual art, for example, that the drawing but not the coloration is ‘essential’, not that it is the only source of pleasure;Footnote ¹⁸ or he says that drawing and composition are what ‘constitute the proper object of the pure judgment of taste’, but not that all our judgement about fine art is pure judgement of taste. Greenberg mistook Kant’s analysis of the pure judgement of taste for his philosophy of art; as Diarmud Costello has written:

Indeed, immediately after making his comments about the drawing and composition, which might be considered to be the beginning of Kant’s transition from the analysis of judgements of natural beauty to the theory of fine art proper, although most of the latter will come only much later in the text of the Critique of the Aesthetic Power of Judgment (CPJ, §§43-53), Kant complicates his picture of judgements about art, and indeed about judgements of natural beauty itself: some judgements about beauty and some art, namely decorative art and music without themes, are ‘free’, not dependent upon any concept of what a thing ought to be or what its purpose is, but others are ‘adherent’, that is, do depend upon a concept of what a thing ought to be or what its purpose is (CPJ, §16, 5: 229). Kant does speak about aesthetic judgement insofar as is dependent upon such a concept as an adherent judgement of beauty, or a judgement of adherent beauty, but such a judgement is not simply a judgement about the graceful or pretty appearance of an object that happens to fall under a concept, but a more abstract judgement about the harmonious relation between the appearance of an object and some relevant concept of it, in particular, some concept of its purpose. And of course, as we noted in the discussion of genius, Kant claims that the spirit of fine art consists in aesthetic ideas, that is, in the play of imagination with great concepts — this is not a theory of retinal art at all. Nor is it a theory that art must be beautiful; although Kant does state that genius is required for beautiful art (CPJ, §48, 5: 311; see also CPJ, §49, 5: 315), for in much of the discussion of aesthetic ideas he emphasizes the free play of the imagination with ideas of reason without emphasizing that what that yields is beauty. The emphasis at least is on imagination rather than beauty as such.

But the larger point here is that even the current (or at least recent) philosophy of art that emphasizes form is not a theory specifically of modernist art, but is intended as a theory of what has always been essential in art. The classical statement of twentieth-century formalism was Clive Bell’s Art (Reference Bell1913). That was intended as an apology for the emphasis on form that Bell and his colleague Roger Fry found in ‘post-impressionist painting’, that is, then recent or current painting beginning with the work of Cézanne; but Bell’s argument, although he was a critic, aspired to the level of philosophy in contending that form was what had always been most important in painting. Bell’s claim was that there is a special aesthetic emotion, distinct from ordinary emotions such as love and hate, but also distinct from whatever emotions impel us to apply the epithet ‘beautiful’ to lovely things in nature, such as birds and butterflies — and in this regard his philosophy is different from Kant’s, which takes that application of ‘beautiful’ to be paradigmatic — but also, and especially, from the emotions that impel us to apply this epithet to human beings, above all the feeling of sexual attraction — and in this regard his theory is closer to Kant’s conception of the disinterested pleasure of the beautiful distinct from the mere gratification associated with the merely agreeable. Then Bell asserts that this special emotion is produced by ‘significant form’. In the case of visual art, significant form consists in ‘lines and colours combined in a particular way, certain forms and relations of form’, or ‘forms arranged and combined according to certain unknown and mysterious laws’;Footnote ²⁰ in the case of poetry, ‘it is the formal music that makes the miracle’, while other kinds of literature, relying on their content for their interest, are ‘never pure art’.Footnote ²¹ Bell’s claim that there is a special aesthetic emotion, distinct from ordinary emotions but also not the same sort of pleasure that we take even in natural beauty, is obscure, and his claim that this emotion is produced only by form in the spatial and/or temporal sense could only be empirical. But the crucial point is that in his view the connection between significant form and aesthetic emotion did not just become true beginning with the work of Cézanne, for example, although Cézanne’s painting may have highlighted the significance of form; great art, such as the late antique mosaics of Ravenna or the medieval painting of Cimabue, was always distinguished by its significant form and pleased in virtue of that more than by whatever other features it incidentally had. The modernist painting that Bell favored and for which he advocated may have come closer than earlier art in stripping painting down to what is essential, but otherwise it was not, in his view, doing something radically new. Had Kant’s philosophy of fine art actually been a purely formalist theory, it would have had the same implication: that form had always been the essence of art, whatever the stylistic variations in any art form had been, indeed had to have been, according to his theory of genius. However, Kant’s formalism was his theory of beauty, but far from his whole theory of fine art. But that too — the theory that the spirit of fine art lies in aesthetic ideas — although Kant illustrated it in the first instance with an allusion to some then relatively recent art, namely that of John Milton,Footnote ²² was clearly meant to apply to art of all ages, the art of Sophocles and Shakespeare as well as Milton.Footnote ²³

Several more recent philosophies of art that were occasioned by works of modernist art are current theories of art but are not specifically theories of modernist art. The two foremost general philosophies of art in recent times — the past half-century — have been those of George Dickie and Arthur Danto. The two views have often been confused, because they use a common phrase — the ‘artworld’, perhaps more precisely the ‘artworld’ in Dickie’s case but ‘a world of art’ in Danto’s case — but they are very different theories, although both start from the premise that a work of art need not look different from other kinds of objects and cannot be defined by some particular, purely sensory property, such as visual or temporal form, or beauty, if that is anything specific.Footnote ²⁴ Dickie’s stimulus, or poster-boy, was Duchamp’s Fountain, an ordinary urinal that he signed ‘R. Mutt’ and submitted to an avant-garde art show in 1916, for which however it was too avant-garde to be accepted; Danto’s stimulus was Andy Warhol’s Brillo Boxes, plywood boxes painted to look like the packing cartons for boxes of the eponymous scouring pads, which he saw at an exhibit at the Stable Gallery in New York in 1964. Duchamp’s work drove Dickie to the ‘institutional theory of art’, according to which the art world is ‘the broad social institution in which works of art have their place’ and ‘a work of art in the classificatory sense’, as opposed to in the evaluative sense of a successful work of art, or in Kant’s terminology a work of genius, ‘is (1) an artifact (2) a set of the aspects of which has had conferred upon it the status of candidate for appreciation by some person or persons acting on behalf of a certain social institution (the artworld)’.Footnote ²⁵ Or as Dickie also put it, ‘A work of art is an object of which someone has said, “I christen this a work of art”’.Footnote ²⁶ This is a sociological theory of art, defining art by reference to a certain social institution, rather than what Dickie calls an aesthetic theory of art, defining art by reference to any particular aesthetic quality, such as beauty, or the aim at beauty. It can be regarded as an empirical, extensional version of Kant’s intensional, normative conception of the aesthetic and of art insofar as it is aesthetic: on Dickie’s account, whatever a certain group of people, linked by certain institutions and practices — schools, museums, galleries, journals, etc. -- say is art, is art, and there is no attempt to say what makes such institutions the institutions of the artworld in the first place; while on Kant’s account, works of fine art, insofar as they are intended as objects of taste or aesthetic judgement, aim to please all human beings at least under optimal conditions, but their fates are not determined by the responses or votes of any actual human beings. A true work of art could be created in isolation and remain in isolation, by an entirely self-taught artist who need not even know that she is an artist, while that does not seem possible on Dickie’s account. But once again, Dickie’s theory is not intended to be the theory only of recent modernist art such as Duchamp’s Fountain, although it is designed to accommodate such art; it is supposed to be a theory of what art has always been and will always be. It might be a current or new philosophy of art, insofar as prior to Dickie’s work people might be supposed not to have known what they really meant by calling something a work of art, and had to await Dickie’s work to realize what they had been doing all along, but is not a theory only of modernist art. In this regard it is similar to Kant’s theory, or Kant’s theory is similar to it. We might also observe that Dickie’s theory is not necessarily a theory of current art in the sense of a theory of the necessity of change in art, because there is nothing in the concept of an artworld that requires that it be in constant ferment, that it be constantly churning up new styles of art and even new media of art, although something like that did seem to be a consequence of Kant’s theory of genius. In this regard one might say that Kant’s theory is actually more of a theory of modern art than Dickie’s, that is, a theory of current art, or, we might say, of currency in art.

Arthur Danto’s theory is actually much closer to Kant’s in that it does propose some substantive constraint on what counts as a work of art, although not at the specific level of restricting how a work of art should look or sound, and indeed, like Kant’s theory of aesthetic ideas, it does specify that art have intellectual content and present this in an imaginative way. For even when Danto first introduced the concept of the artworld in his 1964 paper of that name responding to Warhol’s work,Footnote ²⁷ he did not mean that a work of art is defined by any institution or group of persons, thus that some such had suddenly given Warhol permission to make his copies of the ordinary Brillo cartons. What he meant by an artworld was rather a realm of possibilities defined by a particular artist’s or perhaps artistic group’s or movement’s conception of what art might be: ‘What in the end makes the difference between a Brillo box and a work of art consisting of a Brillo box is a certain theory of art. It is the theory that takes it up into the world of art, and keeps it from collapsing into the real object which it is’,Footnote ²⁸ that is, in the case of Brillo Box, a plywood box painted to look like a printed cardboard carton. Then, in further work, Danto went on to define what makes a theory a theory of art. In his 1981 book The Transfiguration of the Commonplace, Danto argued that a work of art always has a ‘semantic’ component or a meaning, which he associates with Gottlob Frege’s Sinn and Bedeutung, but further that it ‘expresses something about its content’ in a way that ‘transcends semantic considerations’, which he associates with Frege’s Farbung or ‘coloration’.Footnote ²⁹ A work of art has meaning or ‘representationality’ on the one hand but also exploits ‘rhetoric, style, and expression’ on the other, through which a work of art becomes ‘like an externalization of the artist’s consciousness, as if we could see his way of seeing and not merely what he saw’.Footnote ³⁰ In later writings yet, Danto used a phrase that had earlier been used by the earlier twentieth-century University of Michigan philosopher DeWitt Parker, namely ‘embodied meaning’: a work of art presents some meaning in some form of embodiment, and he recognized the affinity of his conception to the concept of ‘symbolic expression’ employed by Susanne Langer, whom Danto acknowledged as a teacher and friend.Footnote ³¹ What is Danto’s theory of the stylish expression of meaning, or his and Parker’s concept of embodied meaning, or Langer’s theory of symbolic expression, but a version of Kant’s theory of aesthetic ideas, the theory that art, or at least fine art of genius rather than merely decorative art, expresses profound ideas in imaginative ways? And all these theories, the twentieth-century theories as well as Kant’s, were clearly meant to comprehend art of all periods, not just the recent or contemporary art that may have occasioned them. These may be philosophical theories produced at particular times that considered themselves current and that were intended to subsume modernist art, but precisely by showing that in spite of differences in their appearances, new art was doing the same thing that art had always done, as long as that was philosophically, that is to say abstractly, understood. Once again, there is no philosophy of modernist art, but a current or new philosophy of all art; and Kant’s is such a philosophy as much as Danto’s. Indeed, although I am not sure that Danto entirely appreciated my point when first I made it in 1997, by 2007 he had come to recognize that his theory of art is in essence a refined restatement of Kant’s, designed to make explicit that the Kantian theory can accommodate and properly classify art that looks very different from anything that Kant could have imagined — although, again, it is an implication of Kant’s theory of genius that after a couple of centuries much art should have looked very different from anything that Kant himself could have imagined.Footnote ³²