Definition Of Art - Past Paper

DEFINITION OP ART 137 the social, historical, inst...

DEFINITION OP ART 137 the social, historical, institutional, or intentional characteristics of art. A theory of art will address several of these issues, display the connections among them, and some- CHAPTER ] times, but only sometimes, attempt to formulate a definition either of art or of art- istic value, or both on the basis of some of these other artistic properties This chapter will survey the main trends that mark the history of the project of defining art in the twentieth century before discussing the most important efforts in DEFINITION the past thirty years_ OF ART I. HISTORICAL BACKGROUND Even before turning to the twentieth century, something should be said about the ROBERT STECKER historical roots of the attempt to define art. It is sometimes supposed that the earli- est definitions of art are to be found in the writings of ancient philosophers such as Plato and Aristotle. In fact, one will not find, in these writers, a definition of art, in the sense of an item belonging to the fine arts or of art in its current sense, if that departs from the concept of the fine arts. It is now widely accepted that the former concept was not fully in place until some time in the eighteenth century, and hence it seems implausible that the ancients would think in terms of, or try to define, art in that sense- What is true is that they wrote about such things as poetry, painting, 'Axe is most often used to refer to a set of forms, practices or institutions. However, music, and architecture, which came to be classified as fine arts and saw some com- when we ask: 'Is that art?' we are usually asking whether an individual item is a work mon threads among them. Plato was very interested in the fact that poetry, like of art. The project of defining art most commonly consists in the attempt to find painting, was a representation or imitation of various objects and features of the necessary conditions and sufficient conditions for the truth of the statement that an world, including human beings and their actions, and that it had a powerful effect item is an artwork. That is, the goal is normally to find a principle for classifying all on the emotions. Aristotle also emphasized the idea of poetry as imitation and artworks together while distinguishing them from all non-artworks. Sometimes the characterized other arts, such as music, in those terms. goal is set higher. Some look for a 'real' definition: that is, one in terms of necessary This way of thinking of the arts wielded enormous influence in the Renaissance conditions that are jointly sufficient for being an artwork. Sometimes the aim is to and Enlightenment, and so when the concept of the fine arts solidified the first def- identify a metaphysical essence that all artworks have in common. initions of art were cast in terms of representation, by such important figures as A definition of art should be distinguished from a philosophical theory of art, Hutcheson, Batteux, and Kant. It is not necessary to set out the exact content of all which is invariably a broader project with vaguer boundaries. Such a theory may of these definitions here, since in the later period in which we are interested they touch on many issues other that the issue of definition, or may even studiously avoid were superseded by other approaches. Of those earlier definitions, Kant's is the one that issue in favour of others. A theory of art will typically concern itself centrally with that has had truly lasting influence. Fine art, according to Kant, is one of two questions of value, for example whether there is some unique value that only art- 'aesthetic arts', i.e. arts of representation where 'the feeling of pleasure is what is works offer. In any case, it will attempt to identify the valuable properties of art that immediately in view: The end of agreeable art is pleasurable sensation. The pleas- are responsible for its great importance in most, if not all, cultures. It may give atten- ure afforded by the representations of fine art, in contrast, is 'one of reflection', tion to cognitive issues, such as what one must know to understand an artwork, and which is to say that it arises from the exercise of our imaginative and cognitive what it is for an interpretation of a work to be good, acceptable, or true. A theory of powers. Fine art is 'a mode of representation which is intrinsically final... and has art may be interested in other sorts of responses or attitudes to artworks, such as the effect of advancing the culture of the mental powers in the interest of social emotional responses. It may focus on the fictionality characteristic of so many works communication' (Kant 1952: 165-6). There are elements in this conception that of art, or on their formal, representational, or expressive properties. It may deal with survive even after the idea that the essence of art is representation is abandoned. 138 ROBERT STECKER DEFINITION OF ART 139 One is a series of contrasts between (fine) art, properly understood, and entertain- Space permits the examination of only one specific proposal to define art in ment (agreeable art). Art makes more demands on the intellect but offers deeper terms of expression. The definition comes from Collingwood's Principles of Art satisfactions. Art is 'intrinsically final', i.e. appreciated for its own sake. Art has some (1938). Collingwood defines art primarily as an activity: that of clarifying an emo- essential connection with communication tion, by which he means identifying the emotion one is feeling not merely as a gen- The struggle to replace the mimetic paradigm t ices place in the nineteenth cen- eral type, such as anger or remorse, but with as much particularity as possible. tury. This occurs on many fronts, just as did the formation of the concept of the fine Collingwood does not deny that one can rephrase this definition in terms of a work arts a century earlier. Artistic movements such as romanticism, impressionism, and of art rather than an activity, but he believes that the work exists primarily in the art-for-art's-sake challenge ideals associated with mimeticism and direct attention to minds of artist and audience, rather than in one of the more usual artistic media. other aspects of art, such as the expression of the artist and the experience of the However, he seems to think of the job of the medium as enabling the communica- audience. Debates among critics in response to these movements raise questions tion of the emotion to the audience who then have the same clarified emotion in about the boundaries of art. The invention of photography challenges the mimetic their minds, which is to say, for Collingwood, the work of art itself. ideal in painting, at least if that is regarded as the increasingly accurate, life-like rep- The definition has well known problems. First, even if expressiveness, in some resentation of what we see. The increasing prestige of purely instrumental music sense, is a widespread phenomenon in the arts, it is far too narrowly circumscribed provides at least one clear example of non-representational art. For some, such by Collingwood. He prescribes a certain process by which a work of art must come music provides a new paradigm captured by Walter Pater's claim that all the arts about, whereas it is in fact a contingent matter whether works are created in the way aspire to the condition of music. In response to all this, new definitions of art appear, he recognizes. Not unexpectedly, the definition rules out many items normally especially expression theories, formalist theories, and aesthetic theories. accepted as art works, including some of the greatest in the Western tradition, such What all these theories have in common with each other, as with mixneticism, is that as the plays of Shakespeare, which by Collingwood's lights are entertainment rather they each identify a single valuable property or function of art, and assert that it is this than art. The definition assumes that the emotion expressed in a work is always the property that qualifies something as art. Call these simple fisnctionalist theories Such artist's emotion, but it is not at all clear why a work cannot express, or be express- theories dominate the attempt to define art right through the middle of the twentieth ive of, an emotion not felt by the artist when creating the work. In recent years, the century. Although they now no longer dominate, they are still regularly put forward. idea that art expresses an actual person's emotion has given way to the idea that art Those cited at the end of the last paragraph have been the most important and influ- is expressive of emotion in virtue of possessing expressive properties, such as the ential examples of this type of theory Each deserves attention in some detail. property of being sad, joyful, or anxious, however such properties are analysed. Such properties can be perceived in the work, and their presence in a work does not require any specific process of creation. 2. ART AS EXPRESSION Traditional expression theories like Collingwood's have been widely rejected, even if some still believe they point towards one of the central functions of art. However, the idea that art is expression, qualified by a number of additional conditions, lives The ostensible difference between expression and representation is that, while the on in work of Arthur Danto. Though properly regarded as an expression theory of latter looks outward and attempts to re-present nature, society and human form art, I would claim that Danto's version of this theory arises within a sufficiently and action, the former looks inward in an attempt to convey moods, emotions, different intellectual and artistic context as to be best treated at a later stage of this or attitudes. We seem to find instances of expressive art where representation is discussion. So, putting it on hold for now, we turn to other simple functionalist de-emphasized or absent. It is very common to think of instrumental music, or at conceptions of art. least many pieces of music, in these terms. As the visual arts moved towards greater abstraction, they too often seem to de-emphasize, or abandon representation for the sake of expression. One can even extend this to literature, which pursued express- ivist goals from the advent of romantic poetry through the invention of `stream of consciousness' and other techniques to express interiority. So it might seem that one 3. FORMALISM could find art without representation but not without expression. This might encourage the further thought, independently encouraged by various romantic and Developing alongside expression theories of art were formalist theories. If one stops expressivist movements in the nineteenth and twentieth centuries, that, even when thinking that art is all about representation, a natural further thought, if one is expression and representation co-occur, the real business of art is expression. thinking in simple functionalist terms, is that what art is all about is form rather 140 ROBERT STECKER DEFINITION OF ART 141 than representational content. This thought gained support from various develop- abstract sense are 'built' out of line and colour. A more straightforward way to item- ments in the arts during the time of high modernism, along, exciting period roughly ize the formal properties of a bowl would be colour, three-dimensional shape, and between 188o and 1960. Though many artforms contain modernist masterpieces, the the patterns, if any, that mark its surfaces. Notice that any three-dimensional object work of painters were the paradigm and inspiration for many of the most influen- has formal properties so characterized, and those that have significant form are tial formalist theories. Cezanne in particular was the darling of the early formalists a subclass of those that have form. Essentially the same is true in the cases of build- Clive Bell (1914) and Roger Fry (1920). Cezanne's paintings contain perfectly trad- ings and sculptures, though these are typically far more complex in having many itional representational subjects—landscape, portraiture, still life—but his innova- parts or sub-forms that interact with each other and with a wider environment. But tions could be seen as formal, with virtually no concern, furthermore, to express a similar complexity can be found in many three-dimensional objects, both manu- anything inner other than Cezanne's eye making features of visual reality salient. factured and natural. These innovations involved the use of an wide-ranging palette, a handling of line, In the case of pictures in general, and paintings in particular, which is the sort of and an interest in the three-dimensional geometry of his subjects, which give his visual art in which Bell was most interested, speaking of form as arising from line figures a 'solidity' not found in his impressionist predecessors, while at the same and colour is, if anything, more urtilluminating because all sorts of its properties, time 'flattening' the planes of the pictorial surface. Taking such formal features as the including the representational properties so arise. Further, it gives no indication of raison d'être for these paintings became the typical formalist strategy for under- the complexity of the concept as it applies to a two-dimensional medium capable standing the increasingly abstract works of twentieth-century modernism, as well as of depicting three dimensions. The fact is that the form of a painting includes, hut for reconceiving the history of art. Like the other simple functionalist theories under is hardly confined to, the two-dimensional array of lines and colour patches that discussion here, formalism is not just an attempt to define art. It is a philosophical mark its surface. As Malcom Budd (1995) has pointed out in one of the most sens- theory of art in the sense indicated above. It also attempts to identify the value of art, itive treatments of the topic, it also includes the way objects, abstractly conceived, and what needs to be understood in order to appreciate an artwork are laid out in the represented three-dimensional space of the work and the inter- A formalist attempt to define art faces several initial tasks. They all have to do action of these two- and three-dimensional aspects. with figuring out how to deploy the notion of form in a definition. One can't just If we can pin down the sense of form as it applies across the various art media, say: art is form or art is what has form, because everything has form in some sense. can we then go on to assert that something is an artwork just in case it has signifi- The first task is thus to identify a relevant sense of 'form' or, in other words, to iden - cant form? Bell's definition hinges on his ability to identify not just form, but tify which properties give a work form. Second, if objects other than artworks can significant form, and many have questioned whether he is able to do this in a non- have form in the relevant sense, one has to find something special about the way circular fashion. His most explicit attempts on this score are plainly circular or artworks possess such form. empty, involving the interdefmition of two technical terms, significant form being The best known and most explicit formalist definition of art is Clive Bell's. what and only what produces the aesthetic emotion, and the aesthetic emotion According to Bell, art is what has significant form. Significant form is form that being what is produced by and only by significant form. Others (Gould 1994), imbues what possesses it with a special sort of value that consists in the affect pro- however, have claimed that a substantive understanding of when form is signific- duced in those who perceive it. Bell rails the affect 'the aesthetic emotion', though, ant can be recovered from formalist descriptions of artworks purportedly in pos- as Carol Gould (1994) has pointed out, this is probably a misnomer since what he session of it. has in mind is more likely a positive, pleasurable reaction to a perceptual experi - Even if Bell can successfully identify significant form, his definition is not satis- epee. So Bell performs the second task mentioned above by claiming that what is factory. It misfires in a number (if respects that are typical of the simple function- special about form in art is that it is valuable in a special way. alist approach. First, it rules out the possibility of bad art, since significant form is However, until Bell dispatches the first of the tasks mentioned above, i.e. until we always something to be valued highly. Perhaps there can be degrees of it, but it is know what he means by form, his claims about significant form are unilluminat- not something that can occur to a very small degree unless one can say that a work ing. Unfortunately, regarding this task, Bell is remarkably cavalier. Being concerned has negligible significant form. Second, it displays the common vice of picking out primarily with the visual arts, he sometimes suggests that the building blocks one important property for which we value art, while ignoring others at the cost of of form are line and colour combined in a certain way. But this is not adequate to excluding not just bad works but many great works. Thus, someone who defines art his examples, which include: St Sophia, the windows at Chartres, Mexican sculp- as significant form has little use for artists like Breughel whose paintings, many of ture, a Persian bowl, Chinese carpets, and the masterpieces of Poussin. Perhaps which teem with vast numbers of tiny human figures, give a rich sense of many even three-dimensional works such as buildings, bowls, and sculptures in some aspects of human life but lack art's defining feature as Bell would understand it. 142 ROBERT STECKER DEFINITION OF ART 143 Perhaps there is a better way to deploy the notion of form or formal value in a def- Despite the fact that the notion of the aesthetic better serves the simple func- inition of art. This is a possibility that, whatever its merits, has gone largely unex- tionalist than the notions of representation, expression, or form, such definitions plored. Instead, those who remained attached to the simple functionalist model are still are far from satisfactory. To bring this out, consider two basic requirements turned to an alternative approach using a more flexible concept, that of the aesthetic. on the definition of any kind (class, property, concept) K: (i) that it provide neces- So, rather than exploring hypothetical formalisms, we turn to this new approach. sary conditions for belonging to (being, falling under) K, and (ii) that they provide sufficient conditions for belonging to (being, falling under) K. To be an artwork, is it necessary that it provide aesthetic experience or even that it be made with the intention that it satisfy an interest in such experience? Many have thought not. Those who deny it are impressed with art movements like Dadaism, conceptual art, 4. AESTHETIC DEFINITIONS and performance art. These movements are concerned, in one way or another, with conveying ideas seemingly stripped of aesthetic interest. Dadaist works, such as The concept of the aesthetic is both ambiguous and contested, but there are other Duchamp's readyrnades, appear to be precisely aimed at questioning the necessary chapters in this volume devoted to the explication of those issues, and so little will connection between art and the aesthetic by selecting objects with little or no aes- be said about them here. For our purposes, we can stipulate that the aesthetic refers thetic interest, such as urinals, snow shovels, and bottle racks. Some instances of in the first instance to intrinsically valuable experience that results from dose atten- performance art appear to be based on the premiss that political ideas can be con- tion to the sensuous features of an object or to an imaginary world it projects. veyed more effectively without the veneer of aesthetic interest. Conceptual works Aesthetic properties of objects are those that have inherent value in virtue of the seem to forgo or sideline sensory embodiment entirely. aesthetic experience they afford. Aesthetic interest is an interest in such experiences Defenders of aesthetic definitions take two approaches to replying to this objection and properties. Aesthetic definitions—attempts to define art in terms of such experi- Some (Beardsley 1983) attempt to deny that the apparent counter-examples are ences, properties, or interest—have been, with only a few exceptions, the definitions artworks, but this seems to be a losing battle as the number of ostensible counter- of choice among those pursuing the simple functionalist project during the last examples increase and gain critical and popular acceptance as artworks. What has thirty years. The brief exposition above of definitions of art in terms of representa- recently come to be the more common tack in replying to the objection is to claim that tional, expressive, and formal value suggests why this is the case. Each of the previ- the apparent counter-examples do have aesthetic properties (Lind 1992.). The ready- ous attempts to define art do so by picking out a valuable feature of art and mades, for example, have such properties on more than one leveL Simply regarded as claiming that all and only things that have that feature are artworks. One of the objects, they have features that to a greater or lesser degree reward contemplation. objections to each of the definitions was that they excluded some works of art, even As artworks, they powerfully express Ducharnp's ironic posture towards art. some possessing considerable value, but not in virtue of the feature preferred by the Can we deploy the notion of the aesthetic to provide a sufficient condition for definition. Hence such definitions are not extensionally adequate. being an artwork? As the previous paragraph already begins to suggest, any object By contrast, aesthetic definitions seem, at first glance, to be free of this problem. has the potential to be of aesthetic interest, and so providing aesthetic experience is Form and representation can both afford intrinsically valuable experience, and, hardly unique to art Beardsley's definition rules out natural objects, since they are typically, such experience does not exclude one aspect in favour of the other. The not made with the requisite intention, but it seems to rule in many artefacts that same is true for the experience afforded by the expressive properties of works. All are not artworks, but are made with aesthetically pleasing features. such experience can be regarded under the umbrella of aesthetic experience. There are three ways in which a defender of aesthetic definitions of art might try Aesthetic definitions of art are numerous and new ones are constantly on offer. to cope with the pervasiveness of the aesthetic outside of art per se. One way is to I mention here a few of the better known or better constructed definitions. redefine what counts as art as any artefact with aesthetic interest. (Zangwill 2000 suggests this approach.) The problem with this move is that it just changes the An artwork is something produced with the intention of giving it the capacity subject from an attempt to figure out why we dassify objects as art to a mere to satisfy aesthetic interest (Beardsley 1983). stipulation that something is art if it is an aesthetic object. A definition that A work of art is an artefact which under standard conditions provides its includes doughnut boxes, ceiling fans, and toasters, even when not put forward as percipient with aesthetic experience (Schlesinger 1979). readymades, is simply not a definition of art in a sense others have attempted to An 'artwork' is any creative arrangement of one or more media whose principal capture. Second, one can attempt to rule out non-art artefacts by claiming that art- function is to communicate a significant aesthetic object (Lind 1992). works have a `significant' aesthetic interest that distinguishes them from the `mere' 144 ROBERT STECKER DEFINITION OF ART 145 aesthetic interest possessed by other artefacts (see Lind 1992). But this line is equally Each of these suggestions, while proposing that the concept of art is best cap- unlikely to succeed. The more one requires such 'significance', the less likely it is that tured by something other than a definition, in fact lays the ground for new all artworks will possess it, for we have seen that many recent works are not con- approaches to defining art. The family resemblance view claims that the concept of cerned primarily with creating a rich aesthetic experience. The last strategy is to art is formed by a network of similarities. But which ones accomplish this? If none claim that, despite intuitions to the contrary, aesthetic experience is something that are specified then the view is empty, since everything bears a similarity to every- is either uniquely or primarily provided by art. This strategy faces the daunting task thing else. In fact, Ziff suggested that the relevant domain of similarities will be of specifying an experience common to all artworks, and one that art uniquely or social or functional in nature, though, in the case of the latter, not in the way sim- primarily provides, but without making essential reference to the concept of art. ple functionalists had hoped for. As for the cluster concept view, if the set of con- Though some, such as Beardsley (1969), have attempted such a specification, the ditions sufficient for being an artwork are finite and enumerable, it is already consensus is that no proposal has been successful. equivalent to a definition of art, viz, a disjunctive definition. While attempting to demonstrate that art cannot be defined, anti-essentialism actually resulted in a whole new crop of definitions, most of which look completely different from their simple functionalist predecessors and rivals. 5. ANTI-ESSENTIALISM Although aesthetic definitions of art continue to have adherents, the dominant 6. DANTO AND DICKIE trend within this topic since the 195os has been to reject simple functionalism in all of its forms. This rejection began with the more sweeping thought that the attempt to define art is misguided because necescary and sufficient conditions do not exist In a highly influential article, Maurice Mandelbaum (1965) was among the first to capable of supporting a real definition of art. The most influential proponents point out that the appeal to family resemblance does not preclude, but rather of this anti-essentialism were Morris Weitz (1956) and Paul Ziff (1953). Guided invites, definition. It may be true that when we look at the resembling features by Wittgenstein's philosophy of language, they claimed that it was atypical for within a literal family, we may find no one exhibited likeness that they all have in ordinary language empirical concepts to operate on the basis of such conditions. common. However, Mandelbaum observes, family resemblance is no more satis- Rather, as Weitz put it, most such concepts were 'open-textured', meaning that the factorily explicated in terms of an open-ended set of similarities differentially criteria by which we apply the concept do not determine its application in every shared among the family's members; for people outside the family may also possess possible situation. While the concept of art is by no means unique in being open- the exhibited features without these thereby bearing a family resemblance to the textured for Weitz and Ziff, the concept still stands apart from many other empir- original set of people. Rather, what is needed to capture the idea of family resem- ical concepts in one respect. For many empirical concepts, open texture merely blance is a non-exhibited relation, namely that of resemblance among those with a creates a theoretical possibility that situations may arise in which criteria no longer common ancestry. Without proposing a specific definition, Mandelbaum suggested guide us, and a new decision is needed whether the concept applies. Weitz and Ziff that in attempting to define art we may fill in the gap left to us by the family resem- conceived of art as requiring such decisions on a regular basis as new art move- blance view by appealing to some non-exhibited relational property—perhaps one ments continually create novel works. This novelty provides a constant source of involving intention, use, or origin. counter-examples to simple functionalist definitions. Among the first to explore the possibility of defining art in these terms, and cer- Instead of being classified by necessary and sufficient conditions, claimed Weitz and tainly the most influential proponents of this approach, were Arthur Danto and Ziff works are classified as art in virtue of 'family resemblances', or sets of similarities George Dickie. In part because both cast their thought about art in terms of 'the art- based on multiple paradigms. So one work is art in virtue of one set of similarities to world', in part because Danto was not explicit about his proposed definition, for some other works, while another is art in virtue of a different set of similarities. An alterna- time it was thought that they were advancing similar definitions of art. However, it tive approach, also Wittgensteinian in spirit, is that art is a duster concept (see Gaut is now understood that each was developing quite different theories, Danto's being z000). This means that we can discern several different sets of properties the posses- historical and functional and Dickie's, radically afunctional and institutional. sion of any of which suffices for an object to achieve art status, but no one of which is In some early papers, Danto (1964,1973) outlines desiderata to which a definition by itself necessary for such status. of art must conform without yet setting forth a definition that satisfactorily meets 146 ROBERT STECKER DEFINITION OF ART 147 the desiderata. The first point, illustrated by the readyrdes as well as by such status, even if it does so through informal procedures. Increasingly, however, he works as Warhol's Brillo Boxes, is that art and non-art can be perceptually indistin- came to view it differently, as one geared to the production of a class of artefacts guishable and so cannot be marked off from each other by 'exhibited' properties. and to their presentation to a public (A corollary to this is that one artwork cannot always be distinguished from another As might be guessed from his changing understanding of the institution of art, by appeal to exhibited properties.) Second, an artwork always exists in an art histor- Dickie has proposed two distinct institutional definitions of art, the second being ical context, and this is a crucial condition for it to be art. Art historical context relates based on his own rejection of the first. Both, however, have received a great deal of a given work to the history of art. It also provides 'an atmosphere of artistic theory', attention and exercised considerable influence, so each deserves some discussion art being 'the kind of thing that depends for its existence on theories' (Danto 1981: here. The first definition goes as follows: 135). Third, 'Nothing is an artwork without an interpretation which constitutes it as Something is a work of art if and only if (1) it is an artifact, and (2) a set of aspects of which such' (p. 135). Every work of art is about something, but, equally, invariably expresses has had conferred upon it the status of candidate for appreciation by some person or an attitude of the artist towards the work's subject or 'way of seeing' the same. An persons acting on behalf of the Artworld. (Dickie 1974: 34) interpretation, then, tells us what the work is about and how it is seen by its maker; Notice that the status conferred that makes some artefact an artwork is the sta- further, it expresses the artist's intention on this score. tus not of being art (at least, not straightforwardly that), but of being a candidate Danto's most important work in the philosophy of art, and his most sustained for appreciation, and this status is conferred on a set of aspects of the item rather attempt to discern the essence of art, is his book The Transfiguration of the than on the item itself. Dickie's definition itself does not tell us who in the artworld Commonplace (1981), in which he elaborates on the considerations stated above and typically confers status. One might think it would he people like critics, art gallery adds others. However, it was left to commentators to fashion an explicitly stated owners, or museum directors, because they are the ones who select and make definition of art from this material. The best statement, and one endorsed by salient to a broader public aspects of a work for appreciation. However, Dickie's Danto, is provided by Noel Carroll (1993: 8o) as follows. X is a work of art if and commentary on the definition makes dear that he thinks artists are the exclusive only if (a) X has a subject (b) about which X projects an attitude or point of view agents of status conferral. Since conferring would seem to be an action, one might (c) by means of rhetorical (usually metaphorical) ellipsis (d), which ellipsis requires wonder what an artist does to bring it about_ It can't just be making something with audience participation to fill in what is missing (interpretation) (e), where both the properties capable of being considered for appreciation. Stephen Davies (1991: 85) work and the interpretation require an art-historical context. has suggested that conferral consists in someone with the appropriate authority To a considerable extent, this definition follows the pattern of traditional simple making, or putting forward, such an object. functionalist definitions of art. Basically, conditions (a) and (b) give to art the func- For many, the crucial idea that makes this definition of art institutional is that tion of projecting a point of view or attitude of the artist about a subject, and this being an artwork consists of possessing a status conferred on it by someone with puts it in the broad class of attempts to define art in terms of expression. That this the authority to do so. However, this is precisely the idea that Dickie eventually function is accomplished in a special way (c), and requires a certain response from rejected. Rightly or wrongly, he came to view status conferral as implying a formal the audience (d), are not uncommon features of expression theories. If anything process, but felt that no such process need occur—nor, typically, does it occur—in sets Danto's definition apart from other simple functionalist proposals, it is the bringing artworks into existence. final condition, (e), which requires that a work and its interpretation stand in a Dickie's second definition of art is part of a set of five definitions that present the historical relation to other artworks. 'leanest possible description of the essential framework of art': It is this last feature that has made Danto's definition influential, but it is not dear that it helps very much to save it from the fate of other simple functionalist defin- An artist is a person who participates with understanding in making a work of art. itions. Many believe that there are works of art that fail to meet all of the first four A work of art is an artifact of a kind created to be presented to an artworld public. conditions. For example, aren't many works of music, architecture, or ceramics, A public is a set of persons whose members are prepared in some degree to understand and even some abstract or decorative works, which are arguably not about anything, an object that is presented to them. The artworld is the totality of all artworld systems. nevertheless instances of works of art? An artworld system is a framework for the presentation of a work of art by an artist to George Dickie's artworld is different from Danto's. Rather than consisting in his- an artworld public. (Dickie 1984: 8o—i) torically related works, styles, and theories, it is an institution. In attempting to define art in terms of an institution, Dickie abandons the attempt to offer a defin- The basic idea here is that the status of being art is not something that is conferred ition not only in terms of exhibited features, but in terms of functions of any sort. by some agent's authority, but instead derives from a work being properly situated in Dickie originally conceived of this institution as one that exists to confer an official a system of relations. Pre-eminent in this system is the relation of the work to the artist 148 ROBERT STECKER DEFINITION OF ART 149 and to an artworld public It is the work's being created by the artist against the 'back- systems that Dicicie gestures towards might be defined historically. Walton's sug- ground of the artworld' (Dicicie 1984: n) that establishes it as an artefact of a kind cre- gestion is that the artworld consists of a limited number of proto-systems plus ated to be presented to an artworld public, i.e. an artwork. other systems that develop historically from these in a certain manner (1977: 98). If we abstract from the particulars of Dickie's two definitions, one can discern a Dicltie (1984: 76) has pointed out that this leaves unsettled the issue of why the common strategy that gives rise to a set of common problems for his approach. In proto-systems belong to the artworld in the first place, and has expressed the belief both definitions, Dicicie set out a structure that is shared with other institutions or that no real explanation is possible. This assessment may again be over-hasty. One practices beyond the 'artworld'. Conferral of status occurs in many settings, and possible place to look for the set of original prow-systems would be the formation even the conferral of the status of candidate of appreciation frequently occurs out- of the system of the fine arts in the eighteenth century, with poetry, painting, sculp- side the artworld (whether or not it occurs within it). For example, an 'official' ture, architecture, and music (possibly confined to vocal music) being the paradig- tourist brochure issued by a tourism board confers the status of candidate for matic proto-artforms. Surely, there is an explanation of why these forms comprised appreciation on some particular place. So does official recognition that a building an important category at this time. This explanation might refer to a common is 'historical'. (Remember that Dickie self-consciously refuses to say what kind of functional property, or, it might itself be historical. A residual problem with this appreciation is conferred by agents of the artworld.) Even advertising might be approach is whether it accounts for all items classified as artworks. The view thought to confer such status, as is certainly its aim. appears to imply that to be art it is necessary and sufficient that it belong to an art- How does Dickie's first definition distinguish between these conferrals of candid- form or art system, and not everyone would accept both parts of that claim acy for appreciation from an-making conferrals? Only by referring to the artworld, (Levinson 1979; Stecker 1997). The view, even rehabilitated along quasi-historical i.e. gesturing towards artfonns and their making, distribution, and presentation, lines, may also fail to account for artworks and artforms from non-western and without explaining what marks these off from other status-conferring practices. earlier western cultures that are conceptually but not historically linked in the right Similarly, regarding the second definition, there are many artefact production and way to the eighteenth-century prototypes. presentations systems outside the artworld. Wherever a product is produced for Stephen Davies is the most important defender of the institutional approach consumers, there is such a system. How does Dickk distinguish artworld systems since Dicicie. Davies does not actually offer a definition of art, but sketches lines from other artefact presentation systems? He does so only by naming the artworld along which it should develop. First, it should reinstate the idea that the artworld is systems 'artworld systems', i.e. by gesturing towards the relevant systems without structured according to roles defined by the authority they give to those who explaining what marks them off. occupy them. Art status is conferred on works by artists in virtue of the authority This strategy gives rise to the problems of circularity and incompleteness (see of the role they occupy. Second, artworld institutions should be understood histor- Walton 1977; Levinson 1987; Davies 1991; Stedcer 1986, 1997). Dicicie acknowledges ically. Davies's discussions of the historical roots of art has come to focus more on that his definitions are circular, but denies that this is a problem. It is clearly a prob- individual artworks than on artworld systems. Consider very early artworks. Did lem, however, when a definition is insufficiently informative to mark off the exten- such works exist in an institutional setting? If so, what gave rise to these institu- sion of what it is attempting to define. Because Dickie's definitions simply gesture tions? Surely, it was even earlier works around which the institutions grew. Davies towards the artworld without marking it off from similar systems, it is incomplete initially attempted to give an institutional analysis to cases like this as well as cases for lack of informativeness. Dicicie (1989) replies that it is ultimately arbitrary of isolated artists whose work is disconnected from art institutions as we know whether or not a system is part of the artworld, but such a claim seems to be an them (Davies 1991). His current view, however, is that the earliest art, the prototypes admission that the definition cannot be completed. from which art and its institutions arose, are to be understood functionally. Such items are art because their aesthetic value is essential to their function. However, once art institutions become established, art can develop in ways that no longer require an aesthetic—or any other—function (Davies 1997, woo). 7. HISTORICAL APPROACHES AND THE In addition to attempts to historicize the institutional approach to defining art, a number of philosophers have explored other forms of historical definition. REVIVAL OF FUNCTIONALISM Jerrold Levinson has proposed that an historical relation holding among the inten- tions of artists and prior artworks is definitive of art (Levinson 1979, 1989, 1993); Others have proposed that the situation is not as hopeless as Dicicie (inadvertently) James Carney claims that the relation is one holding among historically evolving suggests. Kendall Walton (1977) was among the first to suggest that the artworld styles (1991, 1994); while Noel Carroll, though not offering a definition, has put 150 ROBERT STECKER DEFINITION OF ART 151 forward the suggestion that art is identified by historical narratives which link later a readymade. He was not successful, but not for lack of an appropriate intention works to earlier ones (Carroll 1994). Robert Stecker asserts that art is defined in (Carney 1994). A forger of a Rembrandt self-portrait may intend that his work be terms of historically evolving functions (1997). regarded in many ways as the original is correctly regarded, without thereby creating Levinson's proposal is one of the best worked out and most carefully defended. another artwork (Sartwell 1990:157). There are also objections questioning whether It is that 'an artwork is a thing that has been seriously intended for regard-as-a - the definition provides a necessary condition for being art. There can be objects that work-of-art, i.e., regard in any way pre-existing artworks are or were correctly achieve functional success as art, in that they reward a complex set of intrinsic regards, regarded' (Levinson 1989: 21). but lack the required intention- They may spring from an artistic intention based on One wants to know more about what it is to intend a thing for regard-as-a-work- a misunderstanding of earlier works, or from a utilitarian intention that adventitiously of-art, and why this core aspect of Levinson's definition does not make it as tightly results in an object with artistically valuable properties. For example, one might set circular as DicIde's. It turns out there can be two relevant types of intention. On out just to make a vessel that holds water and end up with a remarkably beautiful pot. the 'intrinsic' type, one intends a work for a complex of regards for features found Levinson has replies to all of these counter-examples (see Levinson 1990, 1993). in earlier artworks without having any specific artwork, genre, movement, or Duchamp failed because he lacked the relevant 'proprietary right' to the building. tradition in mind. One might intend it for regard for its form, expressiveness, The forger does not create an artwork because, though he intends the forgery to verisimilitude, and so on. Alternatively, there is the 'relational' type of intention, in receive many of the regards correctly directed to the Rembrandt, they are not cor- which one intends an object for regard as some particular artwork, genre, etc. is or rectly directly to his own painting. Levinson seems to admit that there can be art that was correctly regarded. When one fills in these possible regards, in theory, one elim- lacks the intentions he ordinarily requires for arthood, but holds that this points to inates the expression 'as-a-work-of-are, which is the basis of Levinson's defence farther, less central, senses of art. All these replies, as well as the above remarks on against the charge of circularity. first art, add new conditions to, and hence considerably complicate. Levinson's orig- As with some other historical accounts (such as Carney's and Carroll's), inal definition. Sometimes, too many qualifications can kill a proposal. In this case, Levinson's main idea is that something is a work of art because of a relation it bears though, the patient is arguably still alive and attempting to recuperate. to earlier artworks, which are in turn art because of a relation they bear to still Still, at a number of junctures it appears that Levinson might have achieved a earlier works, and so on. Once this is clear, it becomes obvious that, as one moves simpler definition by appealing directly to functions or regards rather than inten- back along the relational chain, one will come across artworks for which there are tions. Robert Steelier (1997) formulates a definition of art that appeals more directly none earlier. These earliest artworks have come to be called 'first are. We need a sep- to an historically evolving set of functions, without completely dispensing with a arate account of what makes first artworks art, and a reason for thinking that this reference to artistic intentions. (For another such attempt, see Graves 1998.) Stecker separate account won't serve to explain why all artworks are art, obviating the need does not define art explicitly in terms of an historical relation linking the art of one for a historical approach. Davies now gives an essentially functional account of first time with the art of an earlier time. Rather, his definition proceeds by reference to art in his historidred institutional approach (1997,2000), and would claim that this time-relative artforms and functions. At any given time, art has a finite set of func- won't explain why all artworks are art because, within an art institution, objects can tions that range from genre-specific values to those widespread representational, acquire art status while lacking the original function of art. expressive, formal, and aesthetic values enshrined in the simple functional defin- Levinson prefers to avoid this straightforwardly functionalist approach to first itions considered earlier. The functions of art at a given time are to be identified art. For him, what makes something first art is that it is 'the ultimate causal source through an understanding of the artforms central to that time. However, that does and intentional reference of later activities we take as paradigmatically art'. not mean that items that don't belong to a central artform are never art. According Furthermore, first art aims at 'many of the same effects and values, that later, para- to Stecker, almost anything can be art, but artefacts outside the central artforms digmatic art has enshrined' (Levinson 1993: 421). These remarks come dose to have to meet a higher standard. This motivates a disjunctive definition of art an a functional approach similar to that of Davies, but substitute causal and inten- item is an artwork at time t where t is not earlier than the time at which the item tional relations to functions for direct reference to the functions themselves. is made, if and only if (a) it is in one of the central artforms at t and is made with There are a number of objections to Levinson's definition. Against taking it as the intention of fulfilling a function that art has at t, or (b) it is an artefact that a sufficient condition for being art, various examples have been offered where the achieves excellence in achieving such a function. requisite intention is purportedly present, but the item in question is arguably not an With this definition too there are various problems. The appearance of circular- artwork. In 1915, Duchamp attempted to transform the Woolworth Building into ity is handled in much the same way as with Levinson's definition: by eliminating 152 ROBERT STECKER DEFINITION OF ART 153 reference to art by enumerating central forms and functions. However, this requires for instance? Should we even continue to assume that we are looking for a single that Stecker provide some account of these items. What makes something a central correct definition, or should we now accept the possibility that there can be sev- artform? How are genuine functions of art distinguished from accidental functions eral equally useful definitions of art, several equally good solutions to the same (e.g. using sculptures as a doorstops or paintings for insulation) and extrinsic func- problem-or perhaps several problems calling for different solutions? tions (e.g. using art an investment)? Further, not every function is appropriate to See also: Value in Art; Ontology of Art; Aesthetics of Popular Art; Aesthetic every candidate artwork, so functions have to be coordinated with their appropri- Experience. ate forms. Finally, there are things that appear to fulfil functions of art to a high degree, but no one would call them artworks. Suppose there were a pill that induced a fine aesthetic experience. The pill is not a work of art even though it appears to fulfil a function of art with excellence. (For replies to these and other objections see Stecker 1997: 51-65.) BIBLIOGRAPHY Views like those of Davies, Levinson, and Stecker suggest that a consensus is Anderson, J. (woo). 'Aesthetic Concepts of Art', in N. Carroll (ed.), Theories of Art Today. developing about how art should be defined (see Stecker z000; Matravers z000). Madison: University of Wisconsin Press, pp. 65-92. Though each at first appears to represent a different approach (institutional, inten- Beardsley, M. (1969). 'Aesthetic Experience Regained'. Journal of Aesthetics and An Criticism tional, functional), the similarities among these views are more striking than the 28: 2-11. differences. All accept Danto's view that art must be defined historically; and all, -(1983). 'An Aesthetic Definition of Art', in H. Curlier (ed.), What is Art? New York in the end, are committed to a definition that consists of a disjunction of suffi- Haven Publications, 15-29. cient conditions rather than a set of necessary conditions that are jointly sufficient Bell, C. (1914). An. London: Chatto & Windus. Reprint New York Capricorn Books, 1958. (so-called real definitions). Further, unlike simple fimctionalist definitions, these Budd, M. (1995). Values of An: Pictures, Poetry, and Music London: The Penguin Press. definitions do not form the kernel of a larger, normatively aimed theory of art, but Carney, J. (1990. 'The Style Theory of Art; Pacific Philosophical Quarterly 72: 273-89. - (1994). 'Defining Art Externally'. British Journal of Aesthetics 34: 114-23. are compatible with many different theories. In particular, these definitions, like Carroll, N. (1993). 'Essence, Expression, and History: Arthur Danto's Philosophy of Art; in Diclde's definitions, distinguish an understanding of what art is from a conception M. Rollins (ed.), Douro and his Critics. Oxford: Basil Blackwell, pp. 79-106. of the value of art. In fact, the disjunctive character of recent definitions suggests (1994) 'Identifying Art', in R. Yanal (ed.), Institutions of Art. University Park, Pa.: not only that there is no one value or function essential to art, but that there is no Pennsylvania State University Press, pp. 3-38. essence of art at all. - (ed.) (woo). Theories of Art Today. Madison: University of Wisconsin Press. Whatever the extent of this consensus, it excludes two parties to the debate. One Collingwood. R. G. (1938). Principles of Art. Oxford: Oxford University Press. comprises those who are still interested in pursuing a simple functionalist defin- Currie, G. (1993). 'Aliens Too'. Analysis 53: 1t6-38. ition, typically in terms of aesthetic experience or properties (see Anderson z000; (woo). 'A Note on Art and Historical Concepts'. British Journal of Aesthetics go: 186-9o. Danto, A. (1964). 'The Artworld'. Journal of Philosophy ex: 571-84- Zangwill z000). The other comprises those who are sceptical of the possibility of._ 39: 1-17.....eoria (1973). ;fitrtworks and Real Things; Th any definition of art (Tilghman 1984; Novitz 1996). (1981). The Transfiguration of the Commonplace. Cambridge, Mass.: Harvard University It is an interesting question just where future work in this area should direct its Press. efforts (see Stecker z000). On one side of the issue, those in the sceptical camp Davies, S. (i991). Definitions of Art. Ithaca, NY: Cornell University Press. might do more to develop their arguments. On the other side of the issue, instead (1997). 'First Art and Art's Definition'. Southern Journal of Philosophy 35: 19-34. of developing more proposals of the sort we have just been considering, it would be (2000). 'Non-Western Art and Art's Definition', in N. Carroll (ed.), Theories of Art worthwhile for the non-sceptical to step back to ask more basic questions. What is Today. Madison: University of Wisconsin Press, pp. 199-216. NY: Cornell it that we are trying to define? Is it the concept of art, the property of being art, or Diclde, G. (1974). Art and the Aesthetic an Institutional Analysis. Ithaca, University Press. a classificatory (or possibly evaluative) social practice, or something else? Suppose (1984). The An Circle. New 'fork Haven Publications. we say we are trying to define a concept. There is an interesting general literature (1989). 'Reply to Stecker', in G. Diclde, R. Sdafani, and It Roblin (eds.), Aesthetics: on this question (Peacocke 1992; Fodor isiog) which it might be useful to bring to a Critical Anthology, znd edn. New York St Martin's Press. the issue of defining art. What should we hope to achieve with such a definition? Fodor, J. (1998). Concepts: Where Cognitive Science Went Wrong. Oxford: Oxford University The traditional goal was to identify the essence of art. If we follow recent definitions Press. in abandoning that goal, what are we doing instead-describing or idealizing, Fry, R. (1920). Vision and Design. London: Chatto & Windus. Reprint New York: Dover, 3998. 154 ROBERT STECKER Gaut, B. (z000). "Art" as a Cluster Concept; in N. Carroll (ed.), Theories of Art Today. Madison: University of Wisconsin Press, 25-44 Gould, C. 0994). 'Clive Bell on Aesthetic Experience and Aesthetic Truth'. British Journal of CHAPTER 8 Aesthetics 34: 124-33. Graves, L. (1998). 'Transgressive Traditions and Art Definitions'. Journal of Aesthetics and Art Criticism 56: 39-48. Kant, I. (1952). The Critique of Judgement, trans'. J. C. Meredith. Oxford: Oxford University Press. Levinson, J. (1979). 'Defining Art Historically'. British Journal of Aesthetics is: 232-30. ONTOLOGY --(1987). Review of The An Circle. Philosophical Review 96: 141-6. (1989). 'Refining Art Historically'. Journal of Aesthetics and An Criticism 47: 2]-33- OF ART (i99o). A Refiner's Fire'. Journal of Aesthetics and An Criticism 48: 231-5. (1993). 'Extending Art Historically'. Journal of Aesthetics and An Criticism 51: 411-24. Lind, R. (1992). 'The Aesthetic Ecsence of Art'. Journal of Aesthetics and An Criticism 50: 117-29. STEPHEN DAVIES Mancielbaum, M. (1965). `Family Resemblances and Generalization Concerning the Arts'. American Philosophical Quarterly 2: 219-28. Matravers, D. (z000). 'The Institutional Theory: a Protean Creature'. British Journal of Aesthetics 4o: 242-30. Novitz, D. (1996). 'Disputes about Art'. Journal of Aesthetics and Art Criticism 34: 153-63. Peacocke, C. (1992). A Study of Concepts Cambridge, Mass.: MIT Press. Sartwell, C. (1990). 'A Counter-Example to Levinson's Historical Theory of Art'. Journal of Aesthetics and An Criticism 48: 157-8. Schlesinger, G. (1979). 'Aesthetic Experience and the Definition of Are. British Journal of Aesthetics 's.: 167-76. Stecicer, It. (1986). 'The End of an Institutional Definition of Art'. British Journal of Aesthetics 26: 124-32. 1. INTRODUCTION (1996). 'Alien Objections to Historical Definitions of Art'. British Journal of Aesthetics 36: 305-8. - (z997). Artworks: Definition, Meaning, Value. University Park, Pa.: Pennsylvania State ONTOLOGY is the study of the kinds of things there are in the world. The ontology University Press. of art considers the matter, form, and mode in which art exists. Works of art are - (z000). 'Is it Reasonable to Attempt to Define Art?' in N. Carroll (ed.), Theories of Art social constructs in the sense that they are not natural kinds but human creations. Today. Madison: University of Wisconsin Press, 45-64- The way we categorize them depends on our interests, and to that extent ontology Tilghman, B. (1984). But IS It Art? Oxford: Blacicwell. is not easily separated from sociology and ideology. Nevertheless, some classifica- Walton, K. (1977). Review of Art and the Aesthetic: An Institutional Analysis. Philosophical Review 86: 97-101. tions and interests are likely to be more revealing of why and how art is created and Weitz, M. (1956). 'The Role of Theory in Aesthetics'. Journal of Aesthetics and Art Criticism appreciated. It is these that our ontology should reflect. 15: 27-35. There are a number of traditional classifications of the arts, for instance in terms Zangwill, N. (z000). 'Aesthetic Functionalism', in E. Brady and J. Levinson (eds.), Aesthetic of their media (stone, words, sounds, paint, etc.), their species (sculpture, literature, Concepts: Essays after Sibley. Oxford: Oxford University Press. music, drama, ballet, etc.), or their styles or contents (tragedy, comedy, surrealism, - (2002). Are There Counterexamples to Aesthetic Theories of Art?'. Journal of Aesthetics impressionism, etc.). The ontology of works of art does not map neatly on to these and An Criticism 6o: 111-18. classifications, however. In the plastic arts, a wide variety of media and structures Ziff, P. (1953). 'The Task of Defining a Work of Are Philosophical Review 62: 466-80. are used In music and drama, not all works are for performance; for instance, tape compositions and theatrical films are not. Not all works of a kind are organized at the same levels, and higher levels cannot generally be analysed in terms of lower ones. Not all literary works are reducible to word sequences, and not all share a