Art, Definitions of
ART, DEFINITIONS OF
A range of related topics are gathered together under the title "The Definition of Art." These include: (1) metaphysical questions, such as "Is there a set of necessary properties whose possession is conjointly sufficient for a candidate to qualify as an artwork?" and, if so, "What are they?"; and (2) the epistemological issue of how we go about establishing that a candidate is an artwork. Traditionally the default assumption among many philosophers has been that there are necessary and sufficient conditions for classifying things as artworks; that these conditions can be assembled into a real or essential definition of art; and that the application of the aforesaid definition provides us with the means to establish that this or that candidate is an artwork. The trick with this approach is to specify, successfully, the pertinent necessary and sufficient conditions.
Needless to say, this enterprise has turned out to be more challenging than one might have anticipated. And the difficulties encountered in successive attempts to carry off this endeavor have left some philosophers either skeptical or agnostic regarding the prospects of the metaphysical project of defining art. Instead, they have tried more modestly merely to identify the epistemological grounds for classifying candidates as artworks without resorting to real definitions.
The search for a definition of art was not something that taxed ancient philosophers like Plato and Aristotle. For Aristotle, art was skill with respect to any practice or craft. There was an art of poetry and an art of painting, but also the art of medicine, navigation, warfare, and so on. Though Plato, Aristotle, and Horace compared poetry and painting, they did not presume an overarching framework that groups certain arts (in their sense) together in the category that we now call the fine arts or beaux arts, or maybe more simply just Art with a capital A —roughly, poetry (literature), painting, sculpture, music, theater, dance, architecture, and, nowadays, photography, film, and video.
The system of the arts was not stably consolidated until the eighteenth century (Kristeller 1992). Thus, it comes as no surprise that Aristotle felt no inclination toward defining the conditions for membership in the category, though he did analyze the nature of some of the things—like tragedy—that would later be subsumed under the concept of (fine) art. For the ancients, there were arts that were tied to certain functions—quite often religious, political, or otherwise social ones—and these art forms were defined and evaluated in light of that function. For example, Aristotle maintained that the function of tragedy was to educate the emotions by eliciting the catharsis or clarification of pity and fear.
When Aristotle and Plato single out mimesis or imitation as a necessary feature of drama—both tragedy and comedy—and painting respectively, it is immensely unlikely that they were attempting to isolate the essence of art in our sense of fine art. It is more plausible to suppose that they were merely singling out a necessary condition of the relevant art forms that is particularly revelatory of the point and purpose of these practices. If one wants to understand what poetry and painting are about, or wants to know what is appropriate to expect from them, the concept of imitation is central. However, when Plato speaks of mimesis in poetry and painting, he is not offering an analysis or definition of what we mean by (fine) art or even a real or essential definition of poetry or painting. Rather, he is merely pointing to a general feature of these art forms that is especially useful to have in mind, if one hopes to comprehend them and gauge their value.
The pressure to define Art (with a capital A ) does not arrive on the scene, until the subset of arts mentioned above are separated from the rest and treated as an exclusive confraternity. Perhaps the reason for the emergence of this grouping has to do with the rise of the bourgeoisie who, with leisure time on their hands, turned to these particular arts to fill their hours and days. But, of course, once this grouping took hold, a question arises concerning what property or properties a prospective member needs in order to join the category.
At least initially, it seems that the first gambit for answering this question was that a candidate for membership in the fine arts had, harkening back to Plato and Aristotle, to be an imitation, but, more specifically, an imitation of the beautiful in nature. This view is explicitly advanced in Abbé Charles Batteux's 1746 treatise The Fine Arts Reduced to the Same Principle in which the eponymous principle is none other than the imitation of beautiful nature (Beardsley 1966). For something to count as art, then, in the relevant sense of that which the eighteenth century called the fine arts or the beaux arts (and what we now simply call Art with a capital A ), something had to be the imitation of the beautiful, though it seems that sometimes this requirement was slackened to no more than that the art form in question had to be representational. If the art form in question was representational, then a work made in accordance with this propensity of the pertinent art form was an artwork. That is, a painting that is a picture is, all things being equal, a work of art.
This definition of art—often called the representational theory of art (Carroll 1999, Chap. 1)—was certainly ill-suited for the developments in the arts to come, for example: An abstract expressionist painting is not a representation of anything and especially not an imitation of something beautiful in nature (Carroll 1999). So the definition was fated to be incessantly accosted by counterexamples in the future. But, perhaps more to the point, the representational theory of art was not even viable in its own day.
Dance, for instance, belonged to the system of the fine arts; it had its own muse, Terpsichore. However, not all dance, even in the eighteenth century, was representational. Much dance involved no more than cadenced steps, gracefully executed. In fact, in order to legitimatize a place for dance in the newly anointed system of the arts, choreographers, like Georges Noverre, had to invent the ballet d'action —the ballet that told a story. But in cases like this, the definition of art as a matter of representation, in fact, functioned prescriptively rather than descriptively.
But an even greater embarrassment for the representational theory of art than dance was the emergence of absolute music—that is to say, pure orchestral music. When opera and song were the dominant forms of music, music could be counted as implicated in representation because the words that accompanied the notes referred. But once absolute music took pride of place in the order of Calliope, it became very strained to think of the imitation of nature as the essence of art status. Indeed, as absolute music came in the nineteenth century to be praised for its possession of a condition to which all the other arts aspired, it became less and less credible that imitation was a necessary condition for entry into the citadel of art. Though some swatches of Beethoven's Pastorale symphony are imitative, most of the rest of his purely musical oeuvre is not. If for no other reason than the ascendancy of absolute music, the representational theory or definition of art was clearly inadequate. Another approach was needed.
Consonant with the reigning artistic movement of the day, Romanticism, one alternative approach to the representational theory of art was the expression theory—the view that something is an artwork only if it is the expression of an emotion or a feeling (Carroll 1999). Variations of this view have been defended by figures such as Leo Tolstoy (Tolstoy 1996), and R. G. Collingwood (Collingwood 1938).
If the representational theory of art emphasized the representation of the outer world, the expression theory of art stressed art as the presentation of the inner world of the affective life. William Wordsworth asserted that poetry is "the spontaneous overflow of powerful feelings," and this was also thought to be applicable to the other arts. It certainly appeared to fit the absolute music that is now called Romantic. Is that not why it is called Romantic? Moreover, the expression theory seemed to resist counterexamples insofar as it might be supposed that any human artifact would unavoidably carry an expressive trace of the affects of its maker.
Nevertheless, counterexamples appeared in droves starting in the early twentieth century. One source of these counterexamples were various sorts of aleatoric art; the Dadaist Tristan Tzara composed poems by cutting out words from a newspaper, placing them in a hat, and drawing them out randomly—thus thwarting the possibility of any causal connection with what he was feeling. Related chance techniques were mobilized by the surrealists and artists like John Cage and Merce Cunningham. Another kind of counterexample to the expression theory derived from found artworks an ordinary comb presented as an artwork by the likes of Marcel Duchamp, which projects no expressive properties, let alone the trace of anything felt by Duchamp.
Nor could these counterexamples be blocked by appealing to the idea that every human product bears an emotive residue from its maker, for the preceding strategies incontrovertibly sever the emotional link between the artist and the art object. Moreover, the expression theory of art would not only be challenged by the artists of the twentieth century. The theory was undermined by certain forms of art already in existence in the heyday of Romanticism, including art that aspired simply to beauty, as in the case of decorative art, perhaps some absolute music, and art that aimed only to represent the look of the world objectively.
Defenders of the expression theory might attempt to fend off these examples by invoking the claim that there is an inevitable and manifest emotive tie between any artifact and its creator. However, not only does this appear controversial, but if it were so, then the theory would be far too broad to be a satisfactory definition of art because it would fail to differentiate an artwork from any other artifact.
Around the same time that expression theories begin to make their appearance, so, too, do two alternative accounts of art derived from Immanuel Kant's Critique of Judgment. These theories can respectively be called the formalist theory of art and the aesthetic theory of art (Carroll 1999). Formalism, as presented by someone such as Clive Bell (1914), maintains that something is an artwork if and only if it is designed primarily to possess a formal design (called significant form) that is worthy of contemplation for its own sake. That is, the form of the work is intended, first and foremost, to afford an aesthetic experience, (which is sometimes called an experience of disinterested pleasure pursuant to contemplating the work's design).
The aesthetic theory of art (Beardsley 1983) is like formalism except that it leaves the object of experience unspecified by making no reference to the form of the work. On this view, something is an artwork if, and only if, it is made primarily with the intention to support an appreciable amount of aesthetic experience (in other words, experience valued for its own sake). Both the formalist theory of art and the aesthetic theory of art make essential reference to intentions in order to differentiate artworks from natural scenes that might give rise to aesthetic experience. With their emphasis on experiences valued for their own sake, both these views may actually articulate the motive behind the modern category of art as a grouping of the things suitable for leisured contemplation and/or diversion.
Neither formalism nor the aesthetic theory of art provides necessary conditions for classifying candidates as artworks. For it is implausible to suppose that most religious artworks were created with the primary intention of abetting experiences valuable for their own sake. Rather, like so many other premodern artworks, they were produced to perform a function. They were created with the primary intention of advancing religious purposes. Paintings of Christ's crucifixion were intended to instill reverence; they were not meant to be occasions for intrinsically valuable experiences of painterly form. And the designs on the shields of the Sepik warriors of New Guinea were not drawn in order to engender experiences valued for their own sake, but with the instrumental aim of frightening the enemy. Nor is experience valued for its own sake a sufficient condition for art status. Games of chess may be said to promote experiences valued for their own sake, but games of chess are not artworks, not even performance artworks.
The successive failures of attempts to define art disposed many philosophers to skepticism about the very venture itself. By the mid-twentieth century, the suspicion, generally encouraged by the writings of Ludwig Wittgenstein, that art could not be defined became popular. Philosophers like Morris Weitz (1956) argued that because art making is an arena in which experimentation, innovation, and novelty are prized, the notion of defining art is incompatible with the practice of art making. For to define art in terms of necessary and sufficient conditions would putatively somehow shackle the essential openness of art to invention and creativity. Philosophers of this ilk, often called neo-Wittgensteinians, maintained that to define art was to contradict the concept of art as that which contained the permanent possibility of art to expand its horizons in new directions. Consequently, neo-Wittgensteinians rejected the metaphysical project of identifying for artworks a set of necessary conditions that were conjointly sufficient. Moreover, with respect to the epistemological question of how it is established that something is an artwork, they suggested that it was a matter of family resemblance; something A is an artwork when it resembles artwork B in some respects, artwork C in other respects, and so on for further paradigmatic artworks.
Though initially quite influential, the spell of the neo-Wittgensteinian brief began to wane by the 1970s. On the one hand, the argument that specifying the conditions according to which a candidate counted as an artwork is inconsistent with the innovative nature of art could be seen to rest on an equivocation. For even if the practice of art is always, in principle, open to innovation and, therefore, supposedly inhospitable to definition, it is not clear why this would stand in the way of defining the concept of an artwork because individual artworks are not typically open to the permanent possibility of change. It just does not follow that if art (in the sense of the practice of art) is an open concept, then art (in the sense of an individual artwork) is an open concept. Moreover, this open concept argument, as it was called, was also challenged by the appearance of definitions of art by people like Arthur Danto (1981) and especially George Dickie (1974), which, though stated in terms of necessary conditions, provided more than ample room for artistic invention, accommodating the entire gallery of works of Dada and its legacy.
Finally, the epistemological wing of neo-Wittgensteinianism also came under fire. Because it relied upon similarity to establish art status and because everything is like everything else in some respect, by means of the family resemblance method one could in fairly short order establish that any candidate is an artwork. For example, Auguste Rodin's Gate of Hell and an I-beam about to be shipped from a steel mill are both physical objects, metallic, shaped by human designs, weigh more than 100 pounds, over two feet long, and so on. But all these similarities and more are not enough to warrant calling the I-beam an artwork. Though it may be that in the wake of the found artwork anything can be art, it is not the case that everything is art. Nevertheless, the family resemblance method for classifying artworks would appear to force us to conclude that everything is art now.
A common failing of the theories of art as representation, as expression, as form, as well as the family resemblance model for identifying art is that, in each case, art status rests upon some discernible or manifest feature of the object—such as the possession of anthropomorphic or expressive properties, significant form, or similarities with antecedently acknowledged artworks. Perhaps, it was suggested, by Danto and others, that art status rested in some property of art that the eye could not descry. Duchamp's In Advance of a Broken Arm and an ordinary snow shovel are putatively indiscernible. Thus, a theory of art that relies on discernible features of artworks cannot hope to cut the difference between them. Rather, the property (or properties) that are constitutive of art status is something perceptually indiscernible.
For Danto (2000), like G. W. F. Hegel, the relevant feature here is aboutness in a double sense. Something will be an artwork, on this account, only if: (1) It is about something; and (2) its mode of presentation says something about, makes some comment upon, or advances a point of view concerning whatever it is about. However, this formula is, on the one hand, too exclusive—there are artworks that may be about nothing, but which are simply beautiful or delightful to the senses. On the other hand, Danto's theory may be too inclusive. Though Danto means it to tell us the difference between Andy Warhol's artwork Brillo Box and an allegedly indiscernible, though inartistic, one from Proctor and Gamble, surely the ordinary soap pad container in the grocery store meets both of the conditions of Danto's theory of art.
Like Danto, George Dickie is impressed by the thought that the defining features of art might be perceptually indiscernible. This has disposed him to look toward the context that surrounds and frames the work for clues about its status as a work of art. That is, the work does not wear its artistic status on its face; rather, its position in a social framework or institution is the source of its pedigree. This insight has motivated Dickie (1984) to develop a series of what have been referred to as institutional theories of art, the latest version of which he has christened The Art Circle. According to Dickie, our concept of art can be captured by five interlocking definitions:
1) An artist is a person who participates with understanding in the making of a work of art.
2) A work of art is of a kind created to be presented to an art world public.
3) An art world public is a set of persons the members of which are prepared in some degree to understand an object which is presented to them.
4) The art world is the totality of all art world systems.
5) An art world system is a framework for the presentation of a work of art by an artist to an art world public.
Even a cursory examination of the preceding set of definitions reveals that it is circular. One needs the concept of an art world to define what counts as a work of art but the concept of a work of art figures in the definition of an art world system, which, in turn, is an element in the definition of an art world. Dickie is aware of this circularity but claims that it is not problematic. Yet it appears to leave the crucial notion of art undefined, though a definition of art was that at which Dickie was aiming.
Dickie's framework does articulate the structure of any communicative practice with its emphasis on mutual understanding. However, what makes art the very communicative practice it is rather than some other, such as philosophy, has not been clarified by Dickie's analysis. Moreover, some, such as Jerrold Levinson, suspect that the model does not even offer a set of necessary conditions for art status because it does not allow for art made by a solitary artist for himself—for example, some Neolithic wanderer who arranges a pile of colored stones in front of his fire because they are delightful to look at as the flames illuminate them variously (Levinson 1979).
Instead of social context, Levinson locates the defining feature of art in the intention of the artist. On Levinson's view, a candidate is an artwork if, and only if, it is created by a person: (1) who has a proprietary right over the work in question; and (2) who nonpassingly intends the work for regard as a work of art (i.e., in one or more of the ways that artworks have been correctly regarded historically [Levinson 1979]). Like Danto and Dickie, Levinson deploys a non-manifest property of the work—a certain kind of intention—as the crux of his definition. Because this intention must be linked to the history of art, Levinson titles his approach defining art historically.
It is not clear why Levinson feels compelled to require that artists must have a proprietary right over the work in dispute. Surely if Brancusi constructed a sculpture out of stolen materials, there would be no question that he had created a work of art, even if the ownership of the object was in question. Moreover, the second condition of Levinson's definition is also fraught with difficulties. Though it is called a historical definition, it is historically insensitive. It overlooks the possibility that some historical art regards may become obsolete. For example, appreciating the verisimilitude of a picture was an art regard for centuries, but it is arguably no longer decisive, lest many ordinary family snapshots made with the intention to be appreciated integrally and nonpassingly for their accuracy would, counterintuitively, count as artworks. Unfortunately, Levinson makes no provision for anachronistic art regards.
Like Levinson, Robert Stecker (1997) appeals to history in order to define art. He labels his view historical functionalism. It is a disjunctive definition of art. Stecker claims that something is an artwork if, and only if, it is in a central art form at time t and it is made with the intention of fulfilling functions standardly or correctly recognized for that form, or it is an artifact that achieves excellence in fulfilling one of the functions of the central art forms at t.
This definition seems far too inclusive. According to Pierre Bourdieu, one of the functions of our art form is to produce social capital, or status, or identity. Thus, a Cadillac convertible would be a work of art in virtue of the second disjunct in Stecker's formula. The problem here is that Stecker has not limited the functions he countenances to exclusively artistic functions, but, of course, it is not evident that he can do that readily without inviting circularity.
Historical functionalism is also too exclusive. It cannot assimilate as artworks the initial avant-garde entries of radical art movements, for these works may not belong to a central form of art and they may be designed expressly to repudiate the recognized functions of art at time t. Consider the cases of found objects (Duchamp), found music (Cage), and found movement (Yvonne Rainer and Steve Paxton) when they first emerged. They were not obvious examples of a central form and, in any event, they repudiated the functions correctly associated with the forms to which they were related adversarily. Yet certainly any definition of art at this late date must accommodate works such as these.
Perhaps the historical functionalist will attempt to negotiate this shortfall in the theory by saying that once art movements like Postmodern Dance are successful they become—say at time t+1 —central forms of art with correctly recognized functions; thus, in virtue of the second disjunct of the theory, the originating works of the movement from time t can be reclaimed as art. Yet this gambit comes with costs because it has the avant-garde works in question only becoming artworks due to our appreciation of them long after their creators produced them. But surely a dance such as Satisfyin' Lover or a composition such as 4'33'' were artworks from the very moment of their inception. And it is their actual creators who imbued them with art status and not some other folks at time t+1.
Due to the recurring difficulty with constructing an adequate conceptual analysis of art, some contemporary philosophers are agnostic about the metaphysical prospects of discovering a set of necessary properties that are conjointly sufficient for identifying artworks. Instead they focus their energies upon articulating epistemically satisfactory methods for identifying candidates as artworks which methods are not real definitions. Berys Gaut (2000), mining Wittgenstein again for inspiration, resurrects the notion of a cluster concept, arguing that it is sufficient for classifying a candidate as an artwork that the candidate scores well against the following ten criteria:
- It possesses positive aesthetic properties.
- It expresses emotion.
- It is intellectually challenging.
- It is complex and coherent.
- It has the capacity to express complex meanings.
- It exhibits an individual point of view.
- It is an original exercise of the imagination.
- It is the product of skill.
- It belongs to an established form of art.
- It is made with the intention to be a work of art.
This is not a real or essential definition of art because none of these properties are necessary conditions for art status. Anything that is a work of art will have at least one of these features; a work that has more and more of these features provides us with more and more reasons to categorize it as an artwork. On this view, a cluster account of a concept is true of that concept just in case it isolates properties whose possession by the work in question necessarily counts toward its belonging to that category. However, though Gaut provides this list of the components of the cluster concept, he does not believe that the cluster concept approach to identifying art stands or falls with his particular sketch of it. He asserts that even if problematic cases for his formulation exist, that should not lead us to distrust in general the cluster concept approach to identifying artworks.
But is Gaut's assertion here convincing? Clearly, there are problem cases with respect to his dissection of the putative cluster concept. I see no reason why a delicious meal made by a master chef to express his devotion to his beloved and to recall their life together by means of culinary references could not instantiate every component of Gaut's list save obviously (9). Indeed, since the preparation of food occasionally figures in certain theatrical works, and especially in examples of performance art, maybe a case could even be made that it satisfies (9), generously construed. It should, therefore, count as a work of art, though this is certainly at least a very controversial case and, for many, a decisive counterexample to Gaut's proposal. But if Gaut's proposal is defeated, why believe that there is some other model of the cluster concept of art that will do the job? If it is inadmissible to maintain that the definitional approach to the concept of art will succeed despite the lack of evidence so far, why should one have faith in the cluster concept approach, when the best version of it so far misses the mark?
Another non-definitional approach to answering the epistemic question of how we might establish that a candidate is an artwork is that we do so by employing historical narratives (Carroll 1993 and 2001). According to what we may call narrativism, establishing that a candidate is an artwork involves telling a certain kind of story about the work in question, namely an accurate historical narrative about the way in which the candidate came to be produced as an intelligible response to an antecedently acknowledged art-historical situation. That is, in order to corroborate the claim that something is an artwork, one standardly mobilizes a narrative explanation of how the work emerged coherently from recognized artistic modes of thinking, acting, composing, decision-making, and so forth already familiar to the practice.
Usually the pressure to establish that something is an artwork arises when there is some dispute over its art status, as frequently occurs with works of the avant-garde. The narrativist observes that these imbroglios are typically managed by recounting art historical narratives that demonstrate the connection between the disputed work and some earlier artworks whose membership in the order of art is uncontested.
If, for example, the distorted figuration of German Expressionist painting is rejected as art properly so-called on the grounds that it departs from the canons of accurate pictorial representation, the narrativist traces its lineal descent from styles of art, such as that of the medieval artist Matthias Grunewald, where distortion was a strategy for signaling the sentiment of the artist toward his subject. Even if German Expressionist art repudiated prevailing styles of realism, the narrativist argues that there is still reason to count the works in question as art because they harken back to earlier forms of art making, discharging functions, such as the expression of feelings, that are abroad, alive, and acknowledged in their contemporary art world.
One objection to narrativism is that it is circular. However, though circularity is a defect in definitions, it is not clear that it raises any problems for narratives. It is also charged that narrativism confronts the same problem that perplexed the family resemblance approach to identifying art. But this is not the case because narrativists do not merely cite similarities between earlier and later works, but also seek to establish a network of causal relations between them. It is not merely that German Expressionist paintings resemble some medieval art that supports their art status; it is also the case that German Expressionist painting was influenced and inspired by the antecedently recognized medieval art.
Insofar as the narrative approach relies upon tracing lines of descent within historically situated artistic practices, the question arises as to how the narrativist intends to identify artworks in alien traditions. A first response is: by tracing the emergence of later works in that tradition from earlier works. But how can the narrativist identify the first works in alien traditions of art—something he needs to do in order to establish the bona fide origin of subsequent artworks from genuine precedents? Here, the narrativist needs to concede that narrativism is not the only way in which artworks may be identified.
With works in alternative traditions of art making, we frequently need to fix the earliest instances of art in those practices by isolating the works that in that culture are meant to perform the same functions—such as representation, expression, symbolization, decoration, signification, and so forth—that the earliest, already recognized artworks execute in our own culture. This, of course, admits that narration is not the only means of identifying candidates as artworks; sometimes we must depend on functional considerations. Moreover, though historical narration may be sufficient for establishing that a candidate is an artwork, it is not a necessary condition for art status, if only because with certain cases of art, notably from ancient and remote civilizations, it may not be possible to retrieve a narrative account of their provenance.
Beardsley, Monroe. "An Aesthetic Definition of Art." In What Is Art?, edited by Hugh Curtler. New York: Haven, 1983.
Beardsley, Monroe. Aesthetics: from Classical Greece to The Present. New York: MacMillan, 1966.
Bell, Clive. Art. London: Chatto and Windus, 1914.
Carroll, Noël. Beyond Aesthetics. Cambridge, MA: Cambridge University Press, 2001.
Carroll, Noël. "Historical Narratives and The Philosophy of Art." Journal of Aesthetics and Art Criticism 51 (1993): 313–326.
Carroll, Noël. Philosophy of Art: A Contemporary Introduction. London: Routledge, 1999.
Collingwood, R.G. Principles of Art. Oxford: Clarendon Press, 1938.
Danto, Arthur. "Art and Meaning." In Theories of Art Today, edited by Noël Carroll. Madison: University of Wisconsin Press, 2000.
Danto, Arthur. The Transfiguration of The Commonplace. Cambridge, MA: Harvard University Press, 1981.
Davies, Stephen. Definitions of Art. Ithaca, NY: Cornell University Press, 1991.
Dickie, George. Art and The Aesthetic. Ithaca, NY: Cornell University Press, 1974.
Dickie, George. The Art Circle. New York: Havens Press, 1984.
Gaut, Berys. "Art as a Cluster Concept." Theories of Art Today, edited by Noël Carroll. Madison: University of Wisconsin Press, 2000.
Kristeller, Paul Oskar. "The Modern System of The Arts." In Essays on the History of Aesthetics, edited by Peter Kivy. Rochester, NY: University of Rochester Press, 1992.
Levinson, Jerrold. "Defining Art Historically." British Journal of Aesthetics 19 (1979): 232–250.
Stecker, Robert. Artworks. University Park, PA: Pennsylvania University Press, 1997.
Tolstoy, Leo. What Is Art?. Translated by Almyer Maude. Indianapolis, In.: Hackett, 1996.
Weitz, Morris. "The Role of Theory in Aesthetics." Journal of Aesthetics and Art Criticism 15 (1956): 27–35.
Noël Carroll (2005)