Consciousness, Literature and the Arts

Archive

Volume 10 Number 2, August 2009

___________________________________________________________________

The Intentional-Attributive Definition of Art[1]

by

Alexey Aliev

Moscow State University

Introduction

In the Topics, Aristotle advanced the thesis that a definition was correct only if it contained information on natural, essential, and not accidental properties of a thing. He wrote, in particular,

If a definition is an expression signifying the essence of the thing [logos ho toti ên einai sêmainei] and the predicates contained therein ought also to be the only ones which are predicated of the thing in the category of essence; and genera and differentiae are so predicated in that category: it is obvious that if one were to get an admission that so and so are the only at tributes predicated in that category, the expression containing so and so would of necessity be a definition.[2]

Soon afterwards this thesis became canonical. Aristotle's thesis may be the reason why so many aestheticians sought to expose the nature of art[3] in their definitions. Aristotle defined art as mimesis, or the reproduction of the world in images; Alexander Baumgarten, as a locus of perfection in its own right; Johann Gottsched, as a medium for conveying moral truth; Johann Winckelmann, as the imitation of beauty in nature; Kant, as a mode of representation which is intrinsically final; Schopenhauer, as the manifestation of the pure Platonic ideas; Leo Tolstoy, as the expression of a particular feeling experienced by an artist; Roger Fry and Clive Bell, as a significant form.

However, all of these essentialist[4] definitions either exclude some artworks from the extension of the concept of art, or include foreign items into it, or do both concurrently. For example, the definition of an artwork as a thing that expresses emotion is, on the one hand, too inclusive, because tears and grimaces express emotions, but do not belong to art; on the other hand, this same definition is too exclusive, because there are artworks that do not express emotions – for example, Joseph Kosuth's installation One and Three Chairs, Robert Rauschenberg's painting Erased De Kooning, John Cage's musical composition 4'33'', and even Johann Sebastian Bach's fugue G-dur BWV 541.

The traditional project of finding the distinctive nature of art collapsed after Marcel Duchamp presented In Advance of the Broken Arm (1915) and other ready-mades (Bicycle Wheel (1916), Comb (1916), Traveler's Folding Item (1916), Apolinere Enameled (1916-1917), and Fountain (1917)) to the New York public. When whatever one likes can become art – a shovel, a wheel, a comb, a cover, an advertisement, a urinal – the very idea that artworks might share common properties seems to be odd and dubious. What, for example, is common to such artworks as Hamlet, Campbell's Soup Cans, The Last Supper, Pierrot Lunaire (op. 15), the statue of David, Fountain, the Parthenon, Black Square, The Divine Comedy, Scaramouche, The Flanders Road, and Beethoven's Symphony No. 5 in C minor (op. 67)? These works have nothing in common, except perhaps the property of having been created by someone, but even this is not inherent in all artworks, as there is nonartificial art, for example – driftwood art[5]. No wonder that, in the aesthetic thought of 1940-50s, doubt arises as to whether the word ‘art’ can be defined at all. This doubt was formalized into a theory by Morris Weitz. In the article ‘The Role of Theory in Aesthetics’ he claims that it is logically impossible to define the concept of art and proposes a non-definitional theory of art.

            In this paper, I will expand on Weitz’s theory and present arguments that refute his claim that it is logically impossible to define the concept of art. I will then review the most debated theories in contemporary aesthetics – the institutional, the historical, and the aesthetic – and show that all of these have weaknesses and none stands up completely to criticism. In the final part of the essay, I will propose a new, intentional-attributive approach to defining art.

            1. Weitz's Theory: Art as an Open Concept

            If Weitz is correct – if there is no property which belongs to all artworks – it follows that necessary and sufficient conditions for defining art do not exist. However, this reasoning is unlikely to persuade those who think philosophically. If we cannot identify necessary and sufficient conditions empirically, this does not mean that the conditions do not exist. Weitz therefore suggests a different, logical substantiation of his view. Each definition ‘closes’[6] the concept of art – setting strict norms and thus hindering new objects from being counted as art. However, art develops progressively: artworks that break rules, habits, and stereotypes periodically emerge. Therefore, the concept of art is open[7], free from any regulation, and, consequently, unable to be defined; even if we discover a property that is at the moment common to all artworks and only artworks, we cannot regard this property as a necessary and sufficient condition, inasmuch as an artwork which does not have this property can always appear.

            Why then one thing is called ‘art’, while the other is not? To answer this question, Weitz adverts to Wittgenstein's family resemblance idea, according to which some objects are grouped under one term not because they possess a universal property, but because they ‘form a family the members of which have family likeness. Some of them have the same nose, others the same eyebrows and others again the same way of walking; and these likenesses overlap’[8]. Weitz applies this idea to the concept of art: different things, he says, are taken to be art because of their sufficient similarity to each other and to paradigm cases of art. Thus, artworks are connected by a chain of family resemblances: A is sufficiently similar to B, B to C, C to D, and D to A; at the same time, it is not obligatory that A is sufficiently similar to C, and D to B. 

            Weitz considers a number of artworks and highlights paradigm artworks that possess properties that are sufficient to call them ‘art’. Among these works are, for example, Mona Lisa, the statue of Venus de Milo, A Midsummer Night's Dream, the Coliseum, and The Brandenburg Concertos. Sufficiency, however, neither presupposes nor implies necessity; therefore, these properties cannot be used to define art, but rather, to work out criteria for selecting paradigm artworks.

             The paradigm art constitutes a core. The periphery is occupied by the artworks that can not so obviously be classified as art: Fountain, Straight lines in four directions and all their possible combinations, Artist's shit, and many other non-classics. Weitz believes that these and similar items are counted as art because they are sufficiently similar to the paradigm artworks. It being known that similarity with at least one paradigm artwork is necessary, the following statement is correct: x is art, if x is sufficiently similar to y – where x represents a thing and y represents a paradigm artwork.

            2. Critique of Weitz's Theory

            Weitz claims that it is logically impossible to define art in terms of necessary and sufficient conditions; however, if we examine his theory more closely, we can, in fact, discover such a definition: x is art, if and only if either x is a paradigm artwork; or, x is sufficiently similar to paradigm artworks. Weitz might counter this by saying that we have missed the spirit of his argument: one cannot make up a list of necessary and sufficient conditions that once and for all distinguishes art from non-art. Since Weitz's theory does not provide a list of this kind, it does not contain a definition in the traditional sense, but it can still be formally presented as having necessary and sufficient conditions.

            Further, Weitz believes that it is impossible to define art because of its openness: if we introduce necessary and sufficient conditions, then we impose a ban on creativity. But why not introduce the very idea of creativity into the definition?[9]

            It is also not clear why Weitz regards the conditions as confining creativity because sometimes these conditions make creativity possible. For instance, playing chess presupposes abiding by certain rules; however, these rules do not prevent chess players from making original, innovative, and non-standard moves.

            If (1) the fact that revolutionary artworks are created in the artworld does not prove the principled indefinability of art, then (2) the point about the openness of the concept of art appears in Weitz's reasoning as a presupposition. In other words, if (1) is correct – and it seems to be – then Weitz has given no further reason to accept the openness of the concept of art, and so he simply assumes, or presupposes, that openness in his essay without providing any support for it. But why not include another presupposition, according to which there are necessary and sufficient conditions? To resolve the dilemma, we can refer to psychological research and try to establish how human beings classify things. According to Eleanor Rosch's theory[10], people classify things not by necessary and sufficient conditions, but by prototypes: the most typical representatives of a category. Thus, if we want to understand whether x is a member of the bird classification, we select the most typical bird (a dove, an eagle, or a thrush for example) as a prototype and, reasoning from the presence or the absence of resemblance between x and the prototype, we classify x correspondingly. However, the method of family resemblance upon which Rosch's theory is based is not always acceptable, as substantiated by an experiment conducted by Susan Gelman and Ellen Markman[11]. They presented three-year-old children with a picture of a black cat with a white stripe on its back; a picture of a white cat; a picture of a black skunk with a white stripe on its back; and, a picture of a white dinosaur and asked the children to mark the animals that resembled each another. The children marked the cats and the scientists concluded that perceptual similarities are not the only (or most significant) factor used to classify things; if they were, the children would have marked objects that were perceptually most similar to each other: the black cat with the white stripe on its back and the black skunk with the white stripe on its back; and the white cat and the white dinosaur. As a result, the Gelman-Markman experiment casts doubts on the prototype theory.

            Even if Weitz's presupposition is correct, his theory is still challenged by a number of problems. He does not specify how an object must be similar to a paradigm artwork[12]. Meanwhile, all things are similar to each other in some respects; for instance, the moon and a glowworm are both discerned by the unaided eye, shine in the dark, and have two ‘o’ letters in their names.

            Weitz's theory holds true when a thing is sufficiently similar to a paradigm artwork and less similar to everything else. For example, the sculpture Venus de Milo with Drawers is an artwork because it looks like the statue of Venus and does not look like a lamppost, a concrete slab, a dustbin, or any other object of non-art. But what if an artwork resembles an object of non-art more than an object of art? Some contemporary sculptures of metal[13] bear more resemblance to scrap metal than to classic marble sculptures. Fountain looks like an ordinary urinal and has very few common features with paradigm artworks. According to Weitz's theory, Fountain should be referred to as a class of urinals or to a broader class of sanitary engineering objects, but not to a class of artworks. To overcome this difficulty, we need a selection criterion by means of which it is possible to establish similarity only with artworks. However, if we point out such a criterion, then we will certainly arrive at a definition of art in terms of necessary and sufficient conditions. As a result, we have no choice but to postulate the necessity of similarity with artworks. This inevitably leads to a vicious circle in the theory: it is possible to establish whether there is a sufficient similarity between x and the paradigm artworks only if it is known beforehand that x is an artwork.

            An advocate of Weitz's theory may try to overcome this difficulty by referring to the following peculiarity of human thinking. Let us imagine that somebody writes ‘1, 2, 3, 4, 5’ and then asks what number she should write after 5. It is reasonable to answer 6, since the series ‘1, 2, 3, 4, 5’ implies a certain (mathematical) rule. But can not this rule change after ‘5’? Of course, it can, therefore the answer ‘10’ is as logical as the answer ‘6’. Then why do we answer ‘6’ and not ‘10’? Wittgenstein says that in the end there may be no explanation, it is just a brute fact about human beings that they all follow rules in the same way.

            Using Wittgenstein's approach, advocates of Weitz's theory can substantiate the application of the idea of sufficient similarity. Until some moment, a human is capable of indicating a reason why a given thing is sufficiently similar to the other thing. But when she is asked why this and not that aspect is essential for similarity, she has the right to answer that attaching essentiality to this and not that aspect is a fundamental fact of human nature, and so any further elucidation is impossible.

            However, I am skeptical about such a hypothesis because everyday experiences contradict it. People often disagree about things, including which aspect is sufficient for similarity. Weitz also sets aside the question about the means of application of the family resemblance method. If we suppose that A has become art because of its sufficient similarity to A1, A1 because of its sufficient similarity to A2, and An because of its sufficient similarity to An+1, then an infinite regression arises. It is impossible to arrive at the first artwork, and consequently, no art can exist at all. If we assume that there is only one paradigm artwork, and that the rest of the artworks have become artworks because they are sufficiently similar to it, then, anyhow, we can not explain why the paradigm artwork is regarded as a work of art.[14]

            The idea of sufficient similarity between artworks leads to a recursive definition. At first it is postulated that an object (a) is an artwork, and then a procedure that generates all the other artworks is specified: 1) a (this concrete thing) is art; 2) if x is art, then y is art if y is sufficiently similar to x; and, 3) nothing else is art. However, this definition will remain incomplete until we explain why x is art.

            Finally, the most powerful objection to Weitz’s theory concerns the nature of properties against which a comparison is conducted. When Weitz speaks of similarity, he means similarity about perceptual properties, that is, properties that are perceived[15]. For example, in Weitz’s ‘driftwood art’ example, a piece of driftwood is an artwork because it is similar in form to surrealistic sculptures. Therefore, Weitz’s theory would be more accurate if he had said that it is impossible to define art as a result of its perceived properties. This, however, does not imply that the same can not be done with non-perceptual properties[16]. It turns out that Weitz did not refute the possibility of defining art in terms of non-perceptual necessary and sufficient conditions; a definition of this kind was proposed by George Dickie in his article ‘The New Institutional Theory of Art’.

            3. The Institutional Theory of Art

            Dickie criticizes Weitz's theory as being incomplete because Weitz does not identify why paradigm works are art.  Dickie emphasizes one property of the paradigm art that cannot easily be disputed – paradigm art is created. In other words, each paradigm artwork is an artifact, or a thing produced by a human being. However, a natural object that is used in a certain way often resembles an artifact. For instance, if somebody takes a stick and uses it to make a pot, then the stick is an implement. Dickie extends this thought, saying that an object gains its status by means of the way it is used. Therefore, can we not say that objects attributed as art are used by us in a special manner? In this case, Fountain is an artwork because it is used as an artwork. This logic may seem strange and even absurd, but Dickie accepts it. While creating Fountain, he says, the artist used the urinal ‘as artistic media in the way that pigments, marble, and the like are used to make more conventional works of art’[17]. However, this kind of usage, according to Dickie, is possible only within a peculiar structure that he calls ‘the artworld’.

The artworld is a social practice of creating and presenting artworks. Those who are engaged in this practice play either the role of an artist or the role of a public. Dickie establishes two rules for each of the roles. The artist must realize that he is creating an artwork and nothing else; and the artist must possess talents in order to paint, sculpture, compose, write, etc. The public must understand that the presented objects are artworks and nothing else; and the public must have the ability to perceive paintings, sculptures, music, poetry, and so on.

            Dickie further identifies two rules, the adherence to which results in producing an artwork. First, something must be created; and second, the creation must be able to be presented to the public. Thus, Dickie arrives at his definition of art: ‘A work of art is an artifact of a kind created to be presented to an artworld public’[18].

            4. Critique of Dickie's Theory

            The institutional theory raises a number of important objections. First, non-artistic objects can arise within the artworld. As an example, playbills fall under the definition of art because they are created in order to be presented to an artworld public; however, Dickie believes that the playbills are not art because ‘such things are… parasitic or secondary to works of art’[19]. Dickie attempts to eliminate the problem by introducing exclusions that he claims do not detract from the definition's validity. In my opinion, Dickie is evading (rather than solving) the problem by means of rhetorical tricks. Someone might claim, for example, that he has worked out a theory according to which all numbers are even. When he is challenged and asked ‘What about the number 5?’ he answers, ‘The number 5 is parasitic concerning other numbers’.

            The second objection to Dickie’s theory is that it is a vicious circle. Dickie admits the existence of this circle, but he does not think it is vicious, because the interdependence of concepts enables us to conceive how the whole system works. This, in its turn, allows us to conclude that any artifact placed in the described system automatically becomes a work of art. Consequently, Dickie defines art not so much with the help of the concepts themselves, as with the help of causal and other relationships constituted by these concepts[20]. If such understanding is correct, the institutional theory does not go in a vicious circle.

            But does the theory represent the facts? To answer this question, we need to confront the problem of the absence of conditions sufficient for defining art. The counterexample of playbills has already been mentioned. Another counterexample is a dog show, at which specially trained dogs are presented to a professional public, a jury. In this example, we have (formally) the same system as that of the artworld; therefore, we are obliged to regard the dogs as artworks. It turns out that the institutional theory does not provide sufficient conditions, because the conditions would be sufficient only if any object placed in the system belonged to art. Nevertheless, I think that Dickie could overcome this difficulty if he recognizes the dog-show dogs as artworks. The fact that almost no one shares this view is not so important for the theory.

            There is another way to show the absence of sufficient conditions in the institutional definition. Suppose that anthropologist A has arrived in a country about which A does not know anything, in order to find out whether the inhabitants of this country have art or not. A sees that people (x) create objects and then show them to other people (y), who react in one way or another. For instance, they might exclaim ‘Oh!’ or take the object and carry it away to their homes. A does not have enough evidence to conclude that these objects are artworks because the objects may actually be the displays of a weapon exhibition, a fishing-tackle exhibition, or something else. It is obvious that the institutional theory does not give sufficient conditions for being art.

            With respect to the theory’s necessary conditions, does every artwork exist within the system? Let us imagine a woman who lives in a village outside the artworld, and who likes to set ‘an assemblage of egg shells and white glue down on the corner of a table’[21]. According to the theory, the result of her creative work is not related to art, because she is unaware that she is an artist. However, the majority of people would agree that the result of her creative work is an artwork. I am, therefore, inclined to state that the institutional definition does not contain necessary conditions. Dickie could have raised an objection, saying that the woman created her work with the intention to show them to other people, and that therefore the artworld implicitly existed in either case. However, if intentions dictate whether something is art or not, then it is unclear why the theory should introduce the artworld.

            5. The Historical Theory of Art

            The historical theory by Jerrold Levinson resembles the institutional theory in many respects. Levinson suggests that x can be regarded as an artwork if, and only if, x is an object that a person intends for regard-as-a-work-of-art, i.e. regard in any way (or ways) in which prior artworks are or were correctly regarded. This light version of the historical definition is presented by Levinson in the beginning of his article ‘Defining Art Historically’. Levinson subsequently introduces various amendments, but they do not change the main idea.

            Levinson comes to his definition by subjecting other theories to criticism and refuting them: the institutional theory leaves the isolated artist problem unsolved; theories that treat a thing as an artwork if it has been created with the intention of giving it the capacity to excite aesthetic experience are wrong because not all artworks give rise to an aesthetic experience. Many traditional theories define art by pointing out some intrinsic properties; however, because artworks have all kinds of different intrinsic properties, we are unable to pick out of them even one that would be necessary and sufficient. Levinson concludes that the historical approach to defining art is the only possible approach.

            Levinson believes that there are three different ways to define art in terms of how it relates to previous works of art. First, one may specify what is outwardly similar to the artworks that already exist. Levinson rejects this approach because everything is similar in some sense to everything else. Second, one may specify what has been created by means of the intention to afford the same emotions or the same pleasure as that which the existing artworks afford. Levinson opposes this method also, because: 1) there is no unique pleasure or other unique emotions that would be stirred up through the artwork's perception; and, 2) it is easy to imagine a drug that yields the same experience as that which the artworks yield. Third, one may specify what has been made through its intention to evoke the same regard as that which the existent artworks evoke. This is the approach that Levinson chooses.

            As it has already been mentioned, Levinson introduces amendments into the initial version of the definition. Two of them should be mentioned. Under the first amendment, an author must have an appropriate proprietary right over an object that is going to become an artwork. Thus, if an artist does not possess a hat, he may not transform it into an artwork by turning inside out and stating that from now on it should be regarded in the same manner as preceding artworks.

            Under the second amendment, the author's intentions must be more or less stable and can differ, depending on whether the author (1) knows how the existent artworks are regarded; (2) does not know how the existent artworks are regarded, but understands what art is; and (3) has not the slightest idea of art. It should be stressed that when (3) takes place, the author has the art-unconscious intention, and so the definition in this situation is also possible.

            Taking these amendments into consideration, Levinson’s revised, final version of the historical definition is:

X is an artwork at t = df X is an object of which it is true at t that some person or persons, having the appropriate proprietary right over X, nonpass-ingly intends (or intended) X for regard-as-a-work-of-art - i.e., regard in any way (or ways) in which objects in the extension of 'artwork' prior to t are or were correctly (or standardly) regarded.[22]

            Although the word ‘art’ appears in the definiens of the historical definition, Levinson denies that his theory involves a vicious circle. He says that we need not know the preceding artworks, since the matter concerns not them, but how they are regarded. Levinson is right, I think, in saying this; the historical definition holds even if we change the phrase ‘artworks’ to the phrase ‘these things’: x is art if x is an object that its author intends for regard in any way (or ways) in which these things are or were correctly regarded, where by ‘these things’ we mean the actual things that we can point to in past history as examples of artworks.

            It follows from the historical theory that an object can become art many years after it was created. Suppose that F wants the result of his work to be regarded in a way in which nobody has ever regarded artworks. Under the historical definition, the result of his work is not art. This, however, does not mean that it is going to be so forever. It may well be that the required way of regard will appear 200 years later, and it will turn out that F created nothing but an artwork.

            The historical theory confronts us with the same problems as the institutional theory. I will not examine these in order not to repeat myself and will proceed to the new, remaining problems introduced by Levinson’s theory.

            6. Critique of Levinson's Theory

            The historical theory claims that the concept of art (x) depends on its history. This means that if something had happened differently from the way it actually happened, we would have had another, different concept of art (y). In order for y to be meaningful, however, it is necessary for us to identify it with a general concept (z) that unites both x and y. But what is z? There is no answer to this question within the historical theory.[23]

            This objection can be described more figuratively. Imagine that the art of isolated, native American-Indians develops concurrently with the art of Europeans. The American- Indian art, certainly, has an absolutely different history. What allows us to call the American-Indian art ‘art’? Levinson does not provide an appropriate criterion, and this is evidence of the falsity of the very historical principle upon which the historical theory is based.

            The theory poses other difficulties as well. If the art being created must be regarded analogously to the art which already exists, then how is a novel art possible? Levinson divides this question into two sub-questions: (1) how is an original art possible; and, (2) how is a revolutionary art possible? The original art, he says, is just a new art that differs markedly in its style from the preceding art, but can still be regarded in the same manner as the preceding art.    Revolutionary art requires an unprecedented manner of regard in order to exist. However, in compliance with the historical definition, it is then excluded from the extension of the concept of art. Levinson proposes two ways out of this difficulty. First, artists presuppose that, while creating the revolutionary art, percipients will at first commit several unsuccessful attempts to regard their work in the same way as the existing works and afterwards find an absolutely new approach to their regard. The second way lies in inclining the percipients to another manner of perception (cardinally different from the rest) by making a contrasting comparison of the revolutionary art with the other works.

            Levinson's attempt to substantiate the possibility of revolutionary art is not very convincing. The claim that we regard the revolutionary artworks in a radically new manner is weak; on the contrary, it seems to me that we react to them in the same manner as to the preceding artworks: we find the experience that we get contains ends in itself[24]. Even if we assume that everything is exactly as Levinson proposes, his theory can not avoid other difficulties.

            Why does revolutionary art necessarily belong to art? Suppose that someone suggests that we regard a blank sheet of the A4 paper in one of the ways in which existent artworks are regarded. I doubt whether it can be done, and consequently, the current manner of regard is to be replaced with a new one. But why should this new manner of regard have any relation to art? Levinson does not sufficiently explain what the manners of the artworks' regard are. He himself gives an example that illustrates this difficulty: if people started using Renaissance portraits as insulation, then we would have an incorrect manner of regard[25]. But how is it possible to distinguish the correct manner of regard from the incorrect one?  

            Levinson notes that the author must have an appropriate proprietary right on the object that will become a work of art at the end of the creation process. According to the historical definition, if an artist takes somebody else's paints and depicts something on someone else's canvas, the result of his activity (his picture) will not be art – even if it is acknowledged to be a masterpiece.

            According to Levinson's amendment, an author can have an art-unconscious intention to create x so that x is regarded in any way (or ways) in which objects in the extension of artwork prior to t are or were correctly regarded. But this view seems strange. If the intention is art-unconscious, what gives us right to speak of its presence? What is the criterion by which the intention can be revealed? If we arrogate this art-unconscious intention arbitrarily, depending on our ideas, this will not solve the ‘isolated artist’ problem and will make the whole theory vacuous or meaningless.

            Levinson remarks that the first artwork conditions all the rest artworks. However, he does not explain why the first artwork has become an artwork. Instead, he simply takes it that the first artwork is, in fact, an artwork. So his historical definition becomes recursive: 1) a (this concrete thing) is an artwork; 2) if x is an artwork, then y is an artwork if y is an object that a person intends for regard as-a-work-of-art (i.e. regard in a way in which x is or was correctly regarded); and, 3) nothing else is an artwork. Such a definition does not help us understand why an object is regarded as a work of art and not as (for example) insulation, fuel, or something else.

            Levinson suggests that we can identify the original artworks simply by tracing back from current artworks to the things that were first regarded in the way that current artworks are to be regarded. This tracing-back will indeed pick out the actual objects that happen to be the first to be regarded in that way, but it does not explain why those objects (and not other objects) are so regarded. And it will not help us to decide whether objects in some different culture, which have no historical connections with these objects, are or are not art – for example, objects in American-Indian culture. The historical theory, as well as the institutional one, gives rise to more problems than it solves. I believe that the aesthetic theory of Monroe Beardsley is closer to the truth.

            7. The Aesthetic Theory of Art

            In the article ‘An Aesthetic Definition of Art’ Beardsley specifies three conditions that a work of art meets. First, it is a physical thing, with qualities that can be perceived, that is created by someone. The word ‘create’ is used here in a broad sense and refers to such actions as making, remaking, joining, collecting, and also to some kinds of movements. Second, the author had certain intentions while creating the artwork. Third, the artwork is aimed at satisfying the aesthetic interest. The concept of the aesthetic interest implies the need for an aesthetic experience that involves ‘a sense of freedom from concern about matters outside the thing received, an intense affect that is nevertheless detached from practical ends, the exhilarating sense of exercising powers of discovery, integration of the self and its experiences’[26]. Using these conditions, Beardsley proposes that ‘an artwork is something produced with the intention of giving it the capacity to satisfy the aesthetic interest’[27]. It follows from this that we should understand the author's intention in order to establish whether x is art or not. If the intention has consisted in an attempt to give x the capacity to satisfy the aesthetic interest, then x belongs to art and there is no need to establish whether x actually gives rise to the aesthetic experience or not.

            8. Critique of Beardsley's Theory

            In the final part of his essay, Beardsley considers five possible objections to the definition. First, according to the theory, even children can create artworks. Beardsley says that it seems to him ‘invidious to deny children this capability, especially on rather a priori grounds’[28]. Second, the aesthetic definition allows forgeries to be regarded as art. Beardsley does not argue against this: if a forgery is created with the intention of giving it the capacity to satisfy the aesthetic interest, then it is an artwork. Third, in compliance with the aesthetic theory, if a thing has become art, it will never stop being art. In other words, if x is art, then x must be art. Meanwhile, a work of art can change into something else under certain circumstances; for instance, Fountain can change into an ordinary urinal. However, Beardsley does not take artworks such as Fountain seriously; he calls them ‘kidding art’. Fourth, it is impossible to fail in creating an artwork. Beardsley admits that the only way to fail to create an artwork is to not do anything. Finally, the fifth objection points out that the aesthetic theory ranks vulgar and needless things as artworks. Beardsley reminds us that his theory is a value-neutral, rather than an evaluative, descriptive definition of art.

            In my view, Beardsley's answers to the objections are quite convincing; however, we need to assess whether his theory stands up to the pressure of additional, stronger criticism.  Above all we should ask ourselves: does the given definition embrace all artworks and only artworks? On the one hand, the definition can be too inclusive, for example, if a thing is found that meets the conditions but is not art. On the other hand, the definition can be too exclusive, for example, if a thing that is acknowledged by everyone as art does not meet the conditions. For example, do adventure sports fall under the definition? It is likely that a rock-climber experiences much the same feelings as a person who perceives an artwork. I think Beardsley could answer this objection by pointing out the condition that the artwork should ‘be created with the intention’; this condition is not met in the rock-climber case and similar cases. The rock-climber does not produce his climb with the intention of giving it the capacity to satisfy the aesthetic interest; he gets experience not as a result of but in the process of climbing. The adventure sports do not fall under the definition, but what about attractions and computer games? Nothing stops us from supposing that attractions and computer games are produced with the intention of giving them the capacity to satisfy the aesthetic interest. Yet, people who visit attractions or play computer games can experience feelings similar to aesthetic ones. That is why both attractions and computer games meet the conditions and, consequently, fall under the definition of an artwork.

            The aesthetic definition excludes Fountain and the other ready-mades from the class of artworks, because it would be absurd to suppose that Duchamp (for example) produced his works with the intention of giving them the capacity to satisfy the aesthetic interest. But Beardsley refuses to recognize things like ready-mades as art:

Many objects exhibited today by the avant-garde evidently do make comments of some kind on art itself, but these objects may or may not be artworks. To classify them as artworks just because they make comments on art would be to  classify a lot of dull and sometimes unintelligible magazine articles and news paper reviews as artworks, and where is the advantage of that?[29]

            However, someone could reply to Beardsley that if the comments are witty and made with the intention of satisfying the aesthetic interest just because of their wit – or because of the grossly awful, comical way in which they appear as dull and unintelligible – they can count as art, even on his definition.

            The aesthetic definition includes the non-art objects (attractions, computer games) in the extension of the concept of art and excludes the art objects (Fountain and the other ready-mades) from the extension of the concept of art. This means that the aesthetic definition is inadequate and cannot be accepted.

            9. The Intentional-Attributive Definition of Art

            9.1 The Relevance of Defining Art

            Having completed a critical survey of the contemporary theories to define art, I will try to formulate my own definition of art. Although some believe that art requires no definition[30], I believe that the project of defining art is worth pursuing. A definition of art will enable philosophers to elucidate many aesthetic problems. With the help of a definition, arts critics may establish, with a high degree of probability, whether a thing relates to their themes, and anthropologists can use the definition to distinguish the art-making activity from political, economic, religious and other kinds of activities. The practical benefit of the definition of art is evident. In addition, I believe that people in general want to know what art is and why some things are attributed to it and some are not.

            9.2 The Definition

            I concur with Weitz that it is impossible to define art in terms of perceptual properties. However, I believe that the rest of Weitz’s theory is misleading. Among modern theories of art, Beardsley's seems to be on the right track, notwithstanding some critical difficulties. Beardsley defines an artwork as a thing for which the author had a special intention during its creation – the intention of giving the work the capacity to satisfy aesthetic interest.

I believe that the main problems of this definition are the narrowness of the term ‘create’ and the vagueness of the term ‘aesthetic interest’. Why should an artwork necessarily be created (made, produced) by an author? It is well known that neither Traveler’s Folding Item (the Underwood Typewriter cover), nor In Advance of the Broken Arm (the snow shovel) were created by Duchamp, though he is considered to be their author.

            Beardsley characterizes the aesthetic interest as a feeling of freedom from concern about matters outside the thing; a feeling of being detached from practical ends, a feeling of exercising powers of discovery, and integration of the self and its experiences. But why are these particular feelings emphasized and not others? Moreover, what is the difference between aesthetic feeling and non-aesthetic feeling?

            It is also not clear why an artwork must be intended to satisfy the aesthetic interest. In 1999 Tracey Emin exhibited a part of her own bedroom in London's Tate: the bed was unmade and different items such as empty bottles, cigarette packs, newspapers, a tube of toothpaste, used underwear, a pair of slippers, condoms, and a dog toy were scattered next to it. Did the artist really want to evoke a feeling of exercising powers of discovery or any other aesthetic feeling?

            Taking into consideration the shortcomings of the aesthetic definition, I would like to suggest a new (although similar to the aesthetic definition in its type) definition:

x is an artwork if the author of x mainly intended to reveal such properties of x, the perception of which would evoke experiences, containing ends in themselves. There is a need to elucidate the notion of the phrase ‘reveal properties’ and the notion of the phrase ‘containing ends in themselves’.

            In most cases, the author actually creates an artwork – she paints with oils, moulds from clay, makes up letters, note signs or creates bodily movements. Other authors are not engaged in creating artworks: Duchamp made no secret of the fact that he bought Bottle Rack at a market and purchased Fountain, an ordinary urinal, from the J.L. Mott Iron Works. Cage ‘composed’ the three-movement composition 4'33''; in the printed music he appointed the performer to do nothing throughout its 4 minute and 33 seconds duration. And not long ago Damien Hirst, one of the Young British Artists, exhibited as an artwork a rotting cow's head with maggots and flies feeding it off (Hirst's installation was called A Thousand Years).

            To accommodate the foregoing, I prefer the wider term ‘reveal’ to the term ‘create’ (‘produce’ or ‘make’). An artist paints a picture in order to reveal properties, the perception of which would evoke experience, containing ends in itself. A poet has the same intentions when writing a verse, a dancer when performing a dance, and a composer when composing music. The revelation of properties, however, does not necessarily occur as a result of a physical action, but can also be achieved by seeing, peering, and observing. In this way, the artist reveals a property in the object that evokes emotions, having ends in themselves. This is the moment when he discovers a new work of art. It should be noted here that the mental act of revealing the object's properties only finds the properties, but does not change them. Therefore, possible reproaches relating to constructivism[31] and corresponding objections are inappropriate in this case.

            Some philosophers characterize experiences that are evoked as a result of the perception of art as valuable for their own sake. Thus, one of the authors of the formalist theory of art (Clive Bell) believes that each artwork has a significant form, the contemplation of which brings about experience valuable for its own sake. I would subscribe to this view if all artworks were really aimed at exciting such experience. However, a series of examples, taken in the area of modern art, is evidence of the opposite. Some minimalist works seem to be aimed at leaving a spectator indifferent. Consider the work Untitled (Six Boxes) by Donald Judd. It is hard to believe that he wanted to evoke delight, or aversion, or any other intense feelings in people, when creating these brass cubes.       It is unlikely that all emotions that are evoked by art are valuable for their own sake. Many of them are dim and meager, and so have no value. Moreover, I suppose that art can be aimed at exciting emotions that a person typically does not want to have or wants to get rid of: disgust, melancholy, horror, boredom, anger. Works such as Mark Rothko's late paintings (Blue, Green and Brown, Untitled [No. 4], Untitled (1968), Brown, Blue, Brown on Blue, Untitled (1953)) convey nothing but the feeling of all-absorbing doom.

            The experiences that are evoked by art differ from the other experiences not in self-value, but in self-endedness. The art experiences withdraw into themselves. To make this clear, let us consider how emotions appear and what they are. According to Jenefer Robinson ‘there is widespread agreement among emotion theorists that emotional responses occur when our wants, goals, and interests, and/or those of our kin or our social group, are perceived to be at stake’. [32] We have fun and are happy when we know that everything is going well. We are sad when we think that we cannot arrive at something. We feel angry when we believe that someone has offended us. We are worried when we wish something not to happen and we are not sure whether or not it will happen. If Robinson is correct, then there should be an objective reason as a result of which our wants, goals, and interests turn out to be affected; namely, an external event. Taking into consideration this remark, an emotion can be characterized as a mental response of a person to a particular situation that is relevant for his wants, goals, and interests.

            Let us examine some basic features of emotions. First of all, it should be noted that an emotion involves focusing attention on an object by which it has been excited. [33] Thus, when I look at Tsuguharu Foujita's Girl in a Field, a feeling of uneasiness is evoked in me, and this feeling rivets my attention to the painting. Emotions prompt us not merely to pay attention to different aspects of the environment, but also to evaluate them as relevant for our wants, goals, and interests. [34] For instance, if I am delighted while listening to Puccini's Tosca, I evaluate this artwork as good or pleasant – as relevant for my wants, goals, and interests.       

            Focusing attention and evaluation are not enough to make an inference about the existence of an emotion. A person who concentrates on a certain aspect of reality and considers it relevant for his wants, goals, and interests does not always have emotions. If I look attentively at an artwork and find it great, I may not feel anything at all. Therefore, additional evidence that someone is in the emotional state is needed. William James proposes that this additional evidence is provided through different physiological reactions. In The Principles of Psychology, he writes:

Particular perceptions do produce wide-spread bodily effects by a sort of immediate physical influence, antecedent to the arousal of an emotion or emotional idea… In listening to poetry, drama, or heroic narrative we are often surprised at the cutaneous shiver which like a sudden wave flows over us, and at the heart-swelling and the lachrymal effusion that unexpectedly catch us at intervals. In listening to music the same is even more strikingly true.[35]

            An emotion is evoked when a person focuses his attention on a certain state of affairs, evaluates it as relevant for his wants, goals, and interests, and undergoes physiological changes in his heart, vessels, and nervous system. These changes (1) make a person ready for subsequent actions and/or (2) intensify attention on what has caused the emotional response.

            Here we come to an important thesis of the intentional-attributive theory; with respect to the art-evoked emotions, (2) – intensifying attention – is kept, whereas (1) – making ready for subsequent actions – is not kept. The emotions further the retention of the percipient's attention to an artwork, but they do not presuppose any actions from his side.

            Someone may disagree that listening to music can evoke emotions that compel one to behave in one way or another. This belief in the physical power of music is nourished solely by different fictional stories: the Sirens sang travelers to sleep, the ancient Greek poet Terpander pacified the mutineers in Sparta by playing his lyre, a young lady from Venice who heard the music of the Neapolitan violinist Stradella and suddenly fell in love with him. In fact although there is ‘not a single authentic record of any specific change of disposition or intention, or even the inhibition of a practical impulse in any person by the agency of music, this belief in the physical power of the art has come down to modern times’. [36] Listening to music causes changes in the pulse and breath rate,

but beyond evoking impulses to sing, tap, adjust one's step to musical rhythm, perhaps to stare, hold one's breath or take a tense attitude, music does not ordinarily influence behavior. Its somatic influences seem to affect unmusical as well as musical persons… and to be, therefore, functions of sound rather than of music.[37]

            I claim that an artist does not have an intention to reveal in his work such properties, the perception of which would evoke experiences, aimed at inclining to act in one way or another. [38] If a novelist wants to evoke a feeling of indignation with his novel, this feeling is not aimed at inducing a reader to tear up this book or set the shop where it is sold on fire. If a painter seeks to evoke feelings of despair and doom through his painting, these feelings are not aimed at making him commit suicide. If a musician is going to evoke feelings of reverence and fear with his music, these feelings are not aimed at forcing a listener to yield to the God's will or buy a gun. If an architect wishes to evoke feelings of serenity and calm with his building, these feelings are not aimed at making a viewer sleepy.

            I conclude that emotions that are evoked by art are non-utilitarian, non-functional, and contain ends in themselves. Having such experiences does not prepare us for specific subsequent actions – after seeing Picasso’s Guernica we are not prepared to go out immediately and overcome the despair by taking up arms right outside the museum. But having such experiences does make us familiar with these sorts of experiences, and it makes us familiar with the range of experiences that human beings may encounter. Being familiar with these things is valuable for our later ability to act in the world. I suppose there might also be a kind of pleasure or value for its own sake in just having such an experience, just as someone might value the experience of finally seeing some famous object, or of hearing some famous person speak, even if the sight or the hearing is not very pleasant, itself.

            9.3 Objections

            I will now review the objections the aesthetic definition faces with and show that the intentional-attributive definition copes with them. Jerrold Levinson raises two objections against Beardsley's theory. First, he claims that various experiences are evoked while perceiving art. Second, the aesthetic experiences, he says, can be evoked not only by art, but also by a drug. The intentional-attributive theory overcomes these objections. Although it is true that experiences that are evoked by art are quite different, our art experiences have at least one common feature – they contain ends in themselves. The second objection is also inapplicable to the intentional-attributive theory; the drug in Levinson's example is a cause of experiences, whereas artwork is not only a cause, but also an object, of experiences. When I see a painting B, the light comes from its surface and meets my eyes and this stimulates my brain processes; as a result I have an experience of B. This experience is maintained if I continue to focus on B, and this means that B is an object. Therefore, it is correct to claim that an artist intended to reveal in B objective characteristics: properties that could cause experiences, containing ends in themselves.

            The intentional-attributive definition embraces all artworks, including such artworks as Fountain, Artist's Shit, Serie ELA 75/K (Easy, Breezy, Beautiful). There is no doubt that the authors intended to reveal in these artworks properties, the perception of which would evoke emotions having ends in themselves. Unlike the aesthetic definition, the intentional-attributive definition includes religious artworks in the extension of the concept of art. Many icons have been given properties that can evoke reverence, humbleness, trepidation, and other feelings, containing ends in themselves.

            It is important to show that the intentional-attributive definition does not include non-artworks. Attractions and computer games are placed among artworks by the aesthetic definition. I doubt whether the designer of an attraction intended to impart to it properties, the perception of which would bring about emotions containing ends in themselves. The intention of attractions is not to enjoy their appearance. On the other hand, if the designer of the attraction actually intended to impart to it properties, the perception of which would bring about emotions containing ends in themselves, then, I think, the attraction could be regarded as an artwork and, in particular, as architecture. The computer game case is analogous. The games are programmed in order for them to be played. Yet if, while creating a computer game, a programmer intended to impart to it properties, the perception of which would bring about emotions containing ends in themselves, then there are no reasons against regarding the game's characters or sceneries as works of art.

            Two difficulties arise in connection with the intentional-attributive definition. First, not all authors intend to reveal in their artworks properties, the perception of which would evoke experiences containing ends in themselves. Disc Jockeys (DJ) create music with the intention of giving it the capacity to prompt people to dance. It is obvious that the emotions that bring about a wish to do something are not self-ended. I see three ways out of this difficulty. We can say: (1) that if the intention is lacking, it is not art but music-made-to-be-danced-to; (2) that the DJ also has the intention of revealing in the music a property whose perception would evoke experiences containing ends in themselves; or (3) that the audience appropriates the music as ‘art’ by treating it as if it were made with that intention, even if it actually was not.

            The second difficulty can be illustrated with the following example. Imagine that some people dance and some people watch them dancing. There is no doubt that the dance is an artwork, because the dancers intend (or at least it can be supposed that they intend) to reveal in their dance properties that will evoke in the viewers experiences containing ends in themselves. But what will happen if the viewers join the dancers? Will the dance continue to be an artwork? I think it is not so important whether there actually are viewers or not; what is important is that the possibility of viewers exists and also the possibility that the dance is made or performed with the relevant intention. In other words, there could be viewers. Moreover, each of the dancers can imagine how he or she looks from the outside. Therefore, we are not prevented from including art cases when there are no viewers (percipients) in the intentional-attributive definition. 

            9.4 Conclusion

            The outlined theory does not pretend to be fully complete. I realize that some of its features need to be clarified, specified, and corrected. But despite its imperfections, the current version of the theory copes well with the task of defining art.

            Notes