A Strategy for a Philosophy of Art

Citation preview

A Strategy for a Philosophy of Art Joseph Margolis The Journal of Aesthetics and Art Criticism, Vol. 37, No. 4. (Summer, 1979), pp. 445-454. Stable URL: http://links.jstor.org/sici?sici=0021-8529%28197922%2937%3A4%3C445%3AASFAPO%3E2.0.CO%3B2-3 The Journal of Aesthetics and Art Criticism is currently published by The American Society for Aesthetics.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/tasfa.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Wed Aug 1 00:04:55 2007

JOSEPH MARGOLIS

A Strategy for a Philosophy of Art

IN THEORIZING about the fine arts, one is invariably fascinated by local details. Just recently, for example, an unusual reportpossibly the first such report-has been published of an autistic girl who produced an extraordinary body of drawings by the age of six.1 T h e psychiatrists and teachers who worked with her finally managed to train her to speak a few words; but as a result of their success, she simply stopped the amazing flow of fresh creations that had apparentl y replaced speech. Many have theorized that the kind of genius certain very young prodigies exhibit must exclude the skills of the plastic arts, since the maturation of these skills-unlike, say, mathematical and musical gifts-has been thought to be peculiarly dependent on cultural influences. Karl Biihler, in fact, is reported by Nigel Dennis (who reviewed the original study of Nadia, the prodigy) to have held that the early learning of speech comes to dominate early graphic ability and eventually "swallow[s] it up completely."2 Thinking of Nadia, we cannot fail to ask .ourselves about the difference between what we take to be ordinary perception and the gifted perception of those who, in an apparently effortless way, transform what they see into what they draw. An enormous number of similar cases arrest our attention, demanding to be explained. Confronting the evidence, we realize, of course, that we know remarkably little about the acquisition of language and culture by the young; although we see the success of the process manifested in pertinent behavior. All the artistic creations that fascinate our eye and ear quite baffle the understanding-Blake's illustrations as much MARGOLISis professor T e m p l e University.

JOSEPH

of

philosophy

at

as those of our autistic child, Couperin's choral music as much as the chants of children. We certainly cannot claim to have analyzed the work of the creative mind. Hence, especially in the arts, we rummage among its collecting deposits. But in theorizing about the arts, as in so much else, the power of a conception lies not so much in illuminating selected details -though it must do that much-as it does in defining closer, convincing unities among large domains of inquiry previously felt to be linked in spite of the incapacity of prevailing schemes of thought to match our intuitions. What, for example, is the connection among artworks, speech, human history, and the deliberate actions of intelligent persons? One is inclined to say that they are all culturally significant, culturally emergent phenomena. T o grant the point seems easy enough. And yet, this concession implies, in part, the much less obvious idea that a theory of art will be compelling to the extent that the domain it organizes may be seen to be systematically related to all the other salient features of the cultural life of man. With the promise of such a theory, it is even possible to trade academically reinforced internal plausibilities for potential gains in linking art with the rest of human culture, provided we allow our intuitions sufficient time to adjust to findings and to discount, in stages, any excessive distortion of the relationship between art and other cultural phenomena. Conceptual change works in this way; but to say so is not to deny the necessary overlap, diachronically, between earlier and later phases of the language in which even discontinuous theories are formuIated.3 I n effect, this means that the proposal of a new conception of the arts must be integrated with a running dispute

MARGOLIS

of the most compelling alternative conceptions. Doubtless, there are many ways in which to press such a theory-some of which will not seem too much of an academic exercise. T h e debates of aestheticians will eventually be measured against the interests of the rest of the art community and even the world at large. But ultimately, our theories must be appraised by other informed professional theorists. Therefore, it is just as well to meet the question head on. An apparent, but only apparent, asid? will suggest the force of emphasizing the common cuItural status of art, human history, action, language, and persons. In his well-known Individuals, a book which seemed bold and fresh when it first appeared, coming after a sustained avoidance of ontology in analytic philosophical circIes, P. F. Strawson identified both persons and physical bodies as what he called "basic particulars," that is, a "distinguishable class or category of particulars such that, as things are, it would not be possible to make all the identifying references which we do make to particulars of other classes, unless we make identifying references to particulars of that class, whereas it would be possible to make all the identifying references we do make to particulars of that class without making identifying reference to particulars of other classes."4 As is known, Strawson refused to countenance a Cartesian dualism regarding persons and insisted that both persons and bodies were basic. In Strawson's view, therefore, on pain of contradiction, persons and bodies cbuld not be treated as one and the same, and physical bodies could not form the proper parts of persons. Persons were said to be distinguished from physical bodies in that both P- and M("personal" and "material") attributes could be ascribed to them, whereas only Mattributes could be ascribed to physical bodies. Strawson's account was confronted bv certain problems which he has apparently not yet resolved. For instance, his "persons" prove to be merely sentient creatures, hardly persons in any full-blooded sense; he has never explained the relationship between persons and bodies, in virtue of which

a person may have precisely the same attributes as an associated physical body said not to be part of that person; two distinct entities not parts of one another must, contrary to Strawson's original intention, be admitted to occupy precisely one and the same place; and so on.5 This is not to discount Strawson's striking and reasonable claim that persons cannot be reduced to physical bodies; it is only to say that if his proposal is to be defended at all, it had better be defended in an entirely different way. What is extraordinarily suggestive about Strawson's maneuver, as far as the philosophy of art is concerned, is simply that it is a consequence of Strawson's theory that persons and physical bodies cannot be distinguished from one another i n any purely perceptual way: whatever properties a physical body has that may be discriminated by means of the senses cannot, according to the theory, distinguish that body from a person said (in some way not explained by Strawson, but capable of being explained) to be (non-dualistically) "affiliated" with that body. Once we grasp this point, we see that it is an easy matter to hold-in fact, the thesis may be even more obvious-that words and sentences cannot be identical w i t h the sounds and marks w i t h which they are (in some sense to be explained) "affiliated," though they cannot be distinguished from those sounds and marks by any purely perceptual means. If we entertain the possibility that all cultural phenomena exhibit this characteristic, namely, that they are in some sense (non-dualistically and non-reductively) affiliated with physical bodies or material entities of some sort, then we are directly led to suppose that the pattern is exhibited as well by artworks. This is a remarkably economical way of outflanking a great many of the best-known theories of art in the twentieth century and of welcoming some of the more interesting newer theories that have surfaced in recent years. It is not, of course, a complete argument, only a sketch of the strategy of an argument. The reason for pursuing it is its initial simplicity and its promised scope. It rests on intuitions that are at least worth considering, analogies that are at least plau-

A Strategy for a Philosophy of Art

sible, and a conceptual thread that draws us back to our opening concern with the linkage among the various sub-domains of the cultural world. Much of the charm of our approach lies in its not having to prejudge-except in terms of resisting an extreme reductionism-the actual ontology of art and in its ability to proceed, at least dialectically, against certain claims and admissions made by those who have already formed a focused theory of art. A few specimen cases will show at a stroke the surprising power of this modest beginning. I n one of the most suggestive recent papers bearing on the ontology and appreciation of art, "The Artworld,"6 Arthur Danto offers a number of intriguing claims that he has still to make fully systematic and explicit. These claims include at least the following: (i) artworks are related to, and distinguished from, " real [physical] objects" by "the is of artistic identification,"7 that is, they are not identical with those objects but possess physical properties or incorporate physical parts of such objects in virtue of which they are identified as, and are, the entities they are; (ii) artworks are easily mistaken for real objects when they are (in the sense of "the is of artistic identification") just those real objects;8 (iii) artworks are specimens of entities that "enjoy a double citizenship," apparently belonging to the real world (of physical objects) and to what Danto terms the "artwor1d";g (iv) it is the theory of art, to which we subscribe in some sense, "that takes [an object] up into the world of art, and keeps it from collapsing into the real object which it is (in a sense of is other than that of artistic identification)";lo (v) two different but perceptually indiscernible artworks may be identical with (in the sense of the is of artistic identification) one and the same real physical object, which is thereby "contain[ed] . . . as part of itself";" (vi) construing physical objects, via some theory of the artworld, in terms of the is of artistic identification "constitutes it a work of art";lz (vii) "To see something as art requires something the eye cannot descry-an atmosphere of artistic theory, a knowledge of the history of art: an artw0rld,"~3that is, a knowledge of the

causal and historical conditions relevant to the production of artworks, informed by a theory of how real objects are related (by the is of artistic identification) to artworks. There are, however, certain telltale weaknesses in this intriguing set of claims. First of all, there is a clear equivocation (doubtless intended but nevertheless not entirely resolved) in holding that there is one object, one and the same, that is both a purely physical object and also an artwork. If the "is" of artistic identification is not the "is" of ordinary identity (as must be true), then one and tlze same object cannot possess dual citizenship, as Danto claims. Secondly, if Danto favors the analogy with Strawson's case, which he mentions sympathetically, then he cannot hold that the parts of a physical object can as such be the parts of an artwork; and if he does hold that the parts of the one may be the parts of the other, then he cannot (so it would seem) maintain that the "is" df artistic identification is not the same as the "is" of identity. Finally, if he holds to the parts thesis unequivocally, then it looks as if he cannot hold that the identification of an artwork depends on considerations of a non-perceptual sort. Clearly, we are invited to refine Danto's account -by virtue of the convincing instances that Danto himself adduces: for instance, his analysis of two (fictitious) frescoes, Newton's First Law and Newton's Third Law, which look for all the world like identical white vertical rectangles nearly divided into equal squares by a horizontal black line, deliciously confirms the ease with which perceptual indiscernibility can be demonstrated. It is quite important to bear in mind that Danto wishes to press two quite different but related theses about perceptual distinction: (a) that it is possible to identify as numerically distinct two works of art, having quite different properties, that are perceptually indistinguishable; and (b) that to identify an object as an artwork is to rely on non-perceptual, causal and historical, grounds informed by our theorizing about the artworld. More recently, Danto has sketched a very pretty and amusing account of how to construe

MARGOLIS

the Manhattan telephone directory as both a kind of novel and a kind of sculpture.~4 T h e plausibility of his illustrations and the general drift of his account, together with the existence of an enormous bodv of artworks that require a theory rather along the lines he suggests, undermine completely all versions of the theory that artworks are essentially (or in some alternative sense, characteristically) perceptual objects. T o grasp the force of Danto's thesis-admitting the weaknesses noticed-is to defeat at a stroke such well-known and extremely influential theories as the one defended for so many years by Monroe Beardsley. Let us be clear about the strategy of the argument before applying it in detail. We begin with a general intuition that cultural phenomena-persons, artworks, words and sentences, and actions that form the body of human history-share certain distinctive properties, precisely those that mark such phenomena as culturally emergent. The essential clue lies with the fact that some kind of functional and non-perceptual distinction is required to sort persons from mere bodies, words and sentences from mere marks and sounds, human actions from mere bodily movements, and, in all likelihood, works of art from the physical materials in which they are somehow manifested. Strawson pro&des an incompletely worked-out suggestion along these lines, with respect to persons; hardly anyone disputes the point, with regard to language; and Danto presses just such a thesis with regard to art. The supporting analysis of cultural phenomena is lacking, of c0urse.~5 But the thesis functions dialectically to cast considerable doubt on theories that treat artworks exhaustively, essentially, characteristically, or distinctively in terms of perceptual qualities. Part of the plausibility of pressing the thesis lies within the ready availability of specimens that cannot be suitably sorted in perceptual terms; this obtains among artworks as well as among other kinds of items, and it obtains even among the so-called visual arts (as opposed to, say, literature, the most recalcitrant of the arts in this respect)-for example, Rauschenberg's "Erased DeKooning Draw-

ing" and Duchamp's L.H.O.O.Q. Shaved.16 Its plausibility also derives from not depending on any particular theory of culture, that is, on any theory that might be thought to be merely ad hoc. Of course, the provision of a suitable theory would be decisive; but such a theory would be bound to accommodate or to explain away the importance of just the non-perceptual considerations mentioned. There is, therefore, a certain prima facie weight in favor of the thesis, so that any version of the counterthesis may be judged suspect to the extent that it fails to address itself effectively or at all to the apparent discrepancy within the cultural domain. It is very easy, particularly on the strength of specimen cases, to suppose, for instance, that cultural phenomena must be distinguished in intentional, causal or productive, functional, or historical terms. But of course to admit that much is to admit that artworks must be identified in non-perceptual terms. Thus, for example, Frank Sibley's wellknown theory, that the description of aesthetic qualities is a form of perceptual taste,l7 becomes immediately vulnerable merely by being made to focus, as Sibley intends, on the appreciation of art. Sibley's view of the logical peculiarity of aesthetic concepts is undermined precisely because the thesis requires that the concepts in question be essentially perceptual in nature. Admit distinctions involving aesthetic taste that are not perceptual: you will have admitted distinctions that Sibley's theory cannot accommodate. Similarly, the argument condemning the so-called Intentional Fallacy, developed by William Wimsatt and Monroe Beardsley,l8 may be seen to be particularly contrary to the very nature of art itself; it appears that (if art is admitted to be a cultural phenomenon) it is quite impossible to deny the relevance of intentional considerations. These are mere sketches of arguments, of course. But precisely for that reason, they show the remarkable economy and promise of the strategy being considered. As it happens, that strategy is doubleedged. Not only does it oblige us to disqualify theories that hold artworks to be merely

A Strategy for a Philosophy of Art

perceptual objects; it obliges us to clarify the relationship between physical and cultural phenomena as one other than that of identity. Hence, not only must Sibley's and Beardsley's theories fall; Danto's and Strawson's must fall as well, at least to the extent that each contradicts the very distinction on which it rests. Strawson's does so, by implicitly admitting that two numerically distinct things may occupy the same place, even though occupying a place is central to Strawson's formulation of the very notion of numerical identity. And Danto's does so (more pertinently to our issue), by implicitly claiming that a work of art is both a physical object and an object different from any (mere) physical object. T h e uneasy ambiguity of Danto's position becomes, ironically, more significant precisely insofar as, favoring the spirit of Danto's own account, we review his analysis of other cultural phenomena. Without doubt, his theory of action is the most instructive in this regard. There, he explains, "I want to isolate those bare, neutral actions before they are colored by the sorts of meanings they are shown to have on the Arena walls and in common life . . . basic actions, as I shall term them. . . ."I9 O n the other hand, he also says: "actions we do but not through any distinct thing which we also do . . . I shall call basic, and mediated ones [those in which a man does . . . something through some other thing that he does] are accordingly non-basic."20 Danto treats a basic action, therefore, as a "component" of a mediated action.21 (In this, he differs from Donald Davidson, who is inclined to treat any seemingly mediated action in terms of alternative descriptions of one and the same "primitive action."22) Hence, Danto regards mediated actions as not identical with basic actions. Furthermore, mediated actions are distinguished as "more human and more social, . . . taken up into the fabric of communication, and deposited as part of human history . . . described in human or cultural terms."23 But, says Danto, a basic action "is identical . . . with a physiological series [of movements],"24 which of course reinforces (at a price) his point that basic actions are not identical with medi-

ated actions. Still, one would have expected that the relationship between a basic action and a mediated action would form an appropriate (cultural) analogue of the relationship between a physical object and an artwork-that is, a construction parallel to the is of artistic identification. Such a theory would have been a very trim revision of our original intuition about the common features of different kinds of cultural phenomena. But the parallel could not obtain i n the same sense i n the case of human basic actions, since, on Danto's own theory, it is "I," one and the same person, who performs both kinds of action. Hence, it is quite impossible that a basic action be strictly identical with a set of physiological movements; and, in the sense in which, by parallel to the is of artistic identification, a basic action is (culturally) identical with a series of physiological movements, it cannot be the "bare, neutral," colorless action that Danto originally posited. This follows, of course, directly from the admission that it is one and the same culturally informed agent who performs both sorts of action. This seeming aside is instructive about Danto's theory of art because it helps, precisely, to clarify (and challenge) the thesis that one and the same object, the artwork, enjoys "dual citizenship" as both a physical object and a work of art. We see here, therefore, the need to clarify the ontology of art in a way that explicitly escapes the temptation toward reductionism. Turn, then, to Beardsley. In his comprehensive survey of the field of aesthetics, Beardsley says very little directly about the nature of a work of art. The Index of his Aesthetics shows only a few entries under the heading "work of art"; none under "art" and none related to the definition or ontology of art; the entries he does provide refer, suggestively, to the heading "aesthetic object."Zs Since this important overview of the literature, Beardsley has rarely turned to examine the nature of artworks. When he has, his primary concern has been to oppose two theses about art: (i) that, against Danto, he says, he is "not yet convinced that all works of art must be about something";26 and (ii) that, against T. J. Diffey and

MARCOLIS

George Dickie, the thesis of "art-status-conferrable kinds of object" is circular and would be unconvincing even if it were not.27 Now, the key to understanding Beardsley's view rests within his notion of an aesthetic object. But, in the context of what has gone before, we may approach the issue once again by way of a seeming tangent. First of all, Beardsley presses the point that "we may easily agree . . . that the possession of aesthetic qualities (in Frank Sibley's sense) is normal to artworks. If, therefore, the possession of such qualities could be shown to depend directly on the existence of an institution, art would be essentially institutional by [reference to the condition: 'If the existence of some institution is included among the truth-conditions of "this artwork has property P," where P is a normal property of artworks, then artworks are essentially institutional objects']."28 Beardsley's strategy, therefore, is to hold fast to a version of the thesis that art is essentially perceptual; if, then, the institutional thesis could be made to accommodate the perceptual, Beardsley would be prepared to accede to it. Secondly, Beardsley actually touches on the general problem of cultural emergence. H e takes note of my own notion of the "is" of embodiment;29 acknowledges the analogy between persons and artworks; and draws attention to an intended further analogy between artworks and human actions. Of three descriptions of what is putatively one event or one action (the point is not clearly resolved)-"The man's arm's moving"; "The man's raising his arm"; "The man's signalling"-Beardsley says: "On my view, they all describe the same event. . . . There is no need to talk of 'embodiment'. . . It has not been proved, I think, that the armmovement cannot be the same event as the signalling; it occurs at the same time and place. . . . Granted that apart from human society there is no such act as signalling; but surely there may be such an act as raising one's arm-that is not dependent on society or culture."30 But if signalling is "dependent on . . . culture" and if the raising of the arm is one and the same action as the signalling, then, on the hypothesis Beardsley himself adopts, it is quite impossible that

.

the action in question, raising the arm, is not dependent on culture. Beardsley claims further that the action is identical with the mere physical event of "the man's arm's moving," but he does not show this. Here, we may simply observe that he fails to come to grips with the very difficulty that Danto himself failed to resolve previously; and he compounds the puzzle by neglecting to show that no relationship other than strict identity is required (the "is" of artistic identification, the "is" of embodiment, or some suitable analogue) in order to account for the peculiarities of our discourse about cultural phenomena. I n particular, if Beardsley concedes that signalling is culture-dependent, then he has conceded that that action can be identified only on non-perceptual-that is, on intentional, historical, selected causal-grounds. If, further, he holds that the physical event of the man's arm's moving can be identified on perceptual grounds (which seems entirely reasonable), then it is quite impossible that the action be identical with the event. Consequently, if, as seems to be his intention, Beardsley concedes that the analysis of persons, artworks, and human actions ought to yield to a common rubric i? the relevant respect, then he cannot but have undermined his own theory that works of art (or their appropriate surrogates, in the context of his account) are perceptual objects. Once again, then, we see the advantage of canvassing the issues in a setting larger than that afforded by the philosophy of art. Beardsley's quarrel with Danto, that works of art need not exhibit "aboutness," or intentionality (in Brentano's sense), is entirely fair-but, unfortunately, it misses the point of Danto's claim. It is true that Danto appears, in "The Artworld," to hold that all artworks exhibit intentionality. But his more considered view,31 is that questions of aboutness are always and necessarily pertinent to artworks-as opposed to mere physical objects-even if it is the case that a particular artwork is not "about" anything. In short, the discovery that a given work is not about anything rests on nonperceptual considerations. So Beardsley's objection, ironically, confirms the more important claim. Beardsley's objection against Dickie is also

A Strategy for a Philosophy of Art

double-edged. Although Dickie's thesis, that artworks are what they are by virtue of some sort of institutional conferring of artworkstatus, seems either preposterous or vacuO U S ,Beardsley ~~ does not show that artworks are not what they are by virtue of the institutional life of a culture; he does not even challenge the claim. But if he were to concede it, then, on the foregoing consideration, he would have to concede further that artworks are identified on non-perceptual grounds, for reasons paralleling what has already been said about the distinction between actions and physical events and between persons and physical bodies. Dickie's maneuver, then, is in general accord with the more promising approach to cultural phenomena; but somehow, apart from the weakness of his thesis about an actual conferring act, he utterly neglects to explain the nature either of artifacts or of institutions. He clearly borrows Danto's conception of an artworld, by which he means to stress that the distinguishing marks of art are "nonexhibited propert[ies]" which (in Danto's phrase) the eye "cannot descry."33 He rightly finds the remark to be "in need of elucidation,"34 but he does not supply what is needed. He merely announces that the artworld is "an institution" or "an established practice" that provides "a framework for the presenting of particular works of art."35 However, Dickie does emphasize and argue for the thesis that "theie is no reason to think that there is a special kind of aesthetic consciousness, attention, or perception."36 This is precisely what Beardsley is at pains to resist; but his argument is a little obscure because, in one of his most recent accounts of the aesthetic situation, he substitutes the notion of "experience" for that of "perception."37 In context, however, it is reasonably clear that Beardsley does not wish to deny his earlier thesis that what we are interested in aesthetically is what we perceive as directly possibly (as in speaking of literature) he finds the term "experience" broader and thus more satisfactory than the term "perception." I n fact, his intention is simply to concede that our interest in literature cannot be merely perceptual in the sense proper to the plastic

and musical arts; he does not concede that, in the latter arts, our interest is not restricted to the perceptual. In his most recent version of the issue (his presidential address for the American Philosophical Association, 1978), Beardsley speaks of our interest in "perceptual and intentional" values; but "intentional" is clearly meant to range over what, in the aesthetic experience of literature, is merely the analogue of what is aesthetically perceived in the plastic and musical arts. A few of Beardsley's remarks will make his position clear. "The aesthetic value of an object," he says, "is the value it possesses in virtue of its capacity to provide aesthetic gratification"; and "Gratification is aesthetic, when it is obtained primarily from attention to the formal unity and/or the regional qualities of a complex whole, and when its magnitude is a function of the degree of formal unity and/or the intensity of regional quality."38 The relevant clues to understanding his view of aesthetic values, then, are these: (i) they are objectively found in objects, are possessed by them; (ii) they are discriminated by means of perception or experience, particularly "when correctly experiencedU;39 (iii) they are a function of properties of formal . unity and regional quality, which Beardsley has always construed explicitly as perceptual. In fact, Beardsley holds, in Aesthetics, that: "In our descriptions of aesthetic objects we are interested in the perceivable properties [of such objects], for which we shall reserve the word 'qualities.' Thus when I speak of the regional qualities of a complex, I mean its perceptual regional properties."40 Similarly, he holds that unity applies "to the phenomenally objective presentations in the experience" of an object;41 by "phenomenally objective presentations," he clearly means "all that one is aware of, or conscious of, at a given timeu-as in "visual, auditory, or verbal" phenomena, explicitly excluding "knowledge of . . . causal conditions, . . . the physical basis, the physical processes of creation, and the biographical background (of whoever produced the object).42 Nevertheless, in the very context in which he advances this well-known thesis, Beardsley concedes that "a work of art is certainly

MARGOLIS

something deliberately fashioned by human effort-it is a work, it is the product of art, or skill, at least in the traditional sense of the term."43 This would seem utterly to undermine the restrictions that Beardsley imposes on "aesthetic objects." For, though he does not wish to treat "work of art" and "aesthetic object" as equivalent notions, he does say, that, having so far "sedulously avoided . . . the term 'work of art,"' we might (once admitting that non-art may be aesthetically appreciated), try "to sort out works of art from all perceptual objects that are not works of art."44 So Beardsley seems to be disposed at one and the same time to treat artworks as perceptual objects and to admit that they must be distinguished as art by reason of their having been produced by some culturally informed human effort-which cannot be perceptually discriminated and which is explicitly excluded from the aesthetic domain. Consequently, Beardsley is caught in the dilemma that the very objects he invites us to appreciate aesthetically cannot be identified as such under the conditions which are to restrict our relevant appreciative attention. This is surely a reductio. All the other wellknown difficulties of Beardsley's position bearing on plural "presentations" of "the same aesthetic object" (no longer, puzzlingly, of the same artwork)45 depend on his having overlooked this essential dilemma.46 One final reference may serve to round out our general strategy. Nelson Goodman was greatly responsible for advancing a theory of art in terms of which it proved quite impossible to treat artworks merely as perceptual objects.47 Goodman demonstrated this in detail in his analysis of representation, symbolic function, forgery, and notationality. But Goodman's explanation of the symbolic function of art is noticeably defective; its essential weakness concerns the range of attributes that may be directly ascribed to, or "found in," artworks. Hence, as in our examination of Danto's account, an examination of Goodman's may be expected to reveal some difficulties either with the attributes of art or its conditions of identity or both. On the second count, nothing need be said here, except that Goodman's constraints on the re-identification of a work

of art are neither convincing nor necessary.48 But the adjustment required does not bear decisively on our present issue. I t is rather Goodman's account of the properties of works of art that is instructive. Now, Goodman holds that works of art are "symbol systems . . . of a particular kind."49 H e needs to show, therefore, that every work of art, qua work of art, has some assignable symbolic function. Apparently, works of art always or characteristically possess expressive properties, even if they fail to possess any other kind of symbolic property. Here, his claim is that "what is expressed is metaphorically exemplified"; "whereas almost anything can denote or even represent almost anything else, a thing can express only what belongs but did not originally belong to it."50 We may press Goodman's thesis at two points: (a) why should the possession of expressive qualities entail exemplification? and (b) why should expressive qualities be construed metaphorically? T h e same considerations intrinsic to Goodman's thesis dissolve both questions. Goodman explains exemplification thus: "Exemplification is possession plus reference. T o have without symbolizing is merely to possess . . . [A] swatch [of cloth] exemplifies only those properties that it both has and refers to."sl If, therefore, works of art merely possess properties-in particular, expressive properties-without exemplifying them, we should have provided a strong basis on which to deny that (in Goodman's terms) works of art are symbol systems. T h e argument is straightforward. First of all, it is clear, even in terms of Goodman's example, that an object may possess a great many properties (for instance, perceivable properties) that it does not exemplify; a piece of blue cloth may not be a sample of blue cloth or of a certain blue even though it possesses the color in question. We may not choose to treat it as a sample. Goodman's argument would be entirely unconvincing if it did not maintain that expressive properties must be treated in a distinctive waythat is, metaphorically. T h e second step of the argument, then, is this: even if an expressive property is first ascribed metaphorically, it does not follow that, in possessing that property, an object exemplifies the

A Strategy for a Philosophy of A r t

property; it is logically quite possible, say, that we ascribe "being a toad" to some particularly disagreeable person without claiming, implicitly or explicitly, that that person thereby exemplifies "being a toad." There seems to be no difference here between literal and metaphorical ascription-though there is, also, no reason as yet to construe expressive ascriptions metaphorically. The third step of the argument, then, is this: in order to hold that, in possessing an expressive property, an object must exemplify that property, Goodman is obliged to hold that artworks are of such a sort that they can never merely possess (without exemplifying) such a property. Goodman nowhere provides the argument-though recent theorists of art seem somewhat inclined to follow him in the claim.52 On the face of it, there is only one possible way of construing the objects to which we are to ascribe expressive properties so as to render Goodman's claim manageableeven if not defensible. That adjustment would force us to treat artworks as mere perceptual objects-which of course Goodman resists. Concede, for the sake of argument, that to speak of artworks is to speak of the distinctive functioning of physical objects. Construe certain objects as physical objects possessing perceivable properties, and hold that only by treating them as entities exemplifying expressive properties can they be treated as possessing those propertiesand, therefore, as functioning as artworks. That is, if possessing expressive properties is what justifies us in treating an object as an artwork, then its identification depends on non-perceptual grounds-because, on Goodman's thesis, such possession itself presupposes that the object has a symbolic function, exemplification. Hence, if expressive properties do not depend on exemplification, then, on an ad-justment of good man'^ thesis, they would be ascribable to objects first identifiable in perceptual terms. But then, the conceptual distinction of artworks would be entirely lost. The trouble is that Goodman had already compellingly shown that artworks were not perceptual objects; that is, he had shown that they were not objects that (i) could be properly identified or re-identimfied in purely perceptual terms,

or (ii) could be ascribed relevant properties exclusively of a perceptual sort. If, however, artworks were admitted to be cultural or culturally emergent entities, then there would and could be n o antecedent reason for denying: (1) that artworks literally possess expressive properties, and (2) that artworks possess them without exemplifying them. This would not show that artworks do not possess symbolic functions essentially or characteristically, only that they could not be said to do so on Goodman's grounds. In short, his entire maneuver appears to be self-serving: first, claim that artworks possess a symbolic function; then, assign a superfluous symbolic function (exemplifiication) in order to demonstrate that whatever important (expressive) properties artworks possess, they possess in virtue of the symbolic function assigned. The arbitrariness of the maneuver becomes particularly clear in Goodman's account of style. "Style," he maintains, "has to do exclusively with the symbolic function of a work of art as such"; in fact, it has to do with a work's exemplifying the style it possesses.53 But this is most uneconomical-and entirely unnecessary. Hence, Goodman's account gains much of its seeming plausibility by an implicit appeal to paradigms of actual objects (physical objects, for instance) unsuited to the kind of ascription literally required in the artworld. We may break off our account at this point. It was, after all, intended as a dialectical exercise to demonstrate the philosophical force of testing the most debated theories of art against conceptual uniformities among the salient phenomena of the various subdomains of culture. If the foregoing considerations are indeed reasonably telling, we may claim that the strategy both economically exposes the essential weakness of the most influential theories current in AngloAmerican aesthetics, and directs us to the key puzzles that a more satisfactory philosophy of art may be expected to resolve.

' Lorna Selfe, Nadia: A Case of Extraordinary Ur.nwi,ig .4bility in an Autistic Child (New York, 1978). ' Nigel Dennis, "Portrait of the Artist," Neru Yo1.k Review of Books, XXV ( M a y 4, 1978), 8, 10, 12-15.

MARGOLIS

Contra Paul Feyerabend, Against Method (London, 1975); and the early views of Thomas S, Kuhn, T h e Structure of Scientific Revolutions (Chicago, 1970). ' P. F. Strawson, Zndividuals (London, 1959), pp. 38-39. For a fuller account of these issues, see Joseph Margolis, Persons and Minds (Dordrecht, 1977). "ournal of Philosophy, LXI (1964), 571-84. ' Ibid., 577. "bid., 575. Ibid., 582. lo Ibid., 581. " Ibid., 578. l 2 I b i d , 579. l3 Ibid., 580. " I n a paper originally mistitled "Pictorial and Artistic Representation" but dealing primarily with the nature of an artwork-presented at a Symposium: "What Is a Painting?" sponsored by the Behavioral Research Directorate, US Army Human Engineering Laboratory, at the Philadelphia Museum of Art, April 17, 1978. Is An attempt to supply it is provided in Persons and Minds; the relevant ontological theory is presented in Joseph Margolis, "The Ontological Peculiarity of Works of Art" Journal of Aesthetics nnd Art Criticism, XXXVI (1977), 45-50. "These are mentioned and interestingly discussed in Timothy Binkley, "Piece: Contra Aesthetics," Journal of Aesthetics aird Art Criticism, XXXV (1977), 265-77. " Cf. Frank Sibley, "Aesthetic Concepts," Philosophical Review, LCVIII (1959), 421-50. I8Cf. William K. Wimsatt, Jr., and Monroe C. Beardsley, "The Intentional Fallacy," in W. K. Wimsatt, Jr., T h e Verbal Icon (Lexington, 1954), Ch. 1. Arthur C. Danto, Analytical Philosophy of Action (Cambridge, 1973), p. ix. 2o Ibid., p. 28. For difficulties in Danto's theory of basic actions, see Myles Brand, "Danto on Basic Actions," A'ous, I1 (1968), 187-90; and Frederick Stoutland, "Basic Actions and Causality," Journal o f Philosophy LXl' (1968), 467-75. 21 Loc. cit. 2 2 Cf. Donald Davidson, "Agency," in Robert Binkley, Richard Bronaugh, and Antonio Marras (eds.), Agent, Action, and Reason (Toronto, 1971). 23 Op. cit., p. X. "Ibid,, p. 115. 2s Monroe C. Beardsley, Aesthetics: Problems in the Philosophy of Criticism (New York, 1958). 28 Monroe C. Beardsley, "Is Art Essentially Institutional" in Lars Aagaard-Mogensen (ed.), Culture and Art (Nyborg and Altantic Highlands, 197fi), p. 203. @

%'Ibid., p. 200. T h e relevant papers are T. J. Diffey, "The Republic of Art," British journal of Aesthetics, IX (1969), 145-56; and George Dickie, "h'hat is Art?" from Art and the Aesthetic (Ithaca, 1974). 28"Is Art Essentially Institutional?" pp. 197, 208. '@Joseph Margolis, "Works of Art as Physically Embodied and Culturally Emergent Entities," British Journal o f Aesthetics, XIV (1974), 187-96; cf. "The Ontological Peculiarity of Works of Art," and Persons and Minds. " Is Art Essentially Institutional?" pp. 205-06. '' Developed in the Behavioral Research Directorate paper, noted above. 32Cf. Joseph Margolis, review of Art and the Aesthetic, Journal of Aesthetics and Art Criticism, XXXIII (1975), 39-42. 3 W p . cit., pp. 27, 29. 341bid., p. 29. 35 Ibid., p. 31. 3RIbid.,p. 40. See further, Chs. 4-6: also, Joseph Margolis, T h e Language of Art and Art Criticism (Detroit, 1965), Ch. 2. '' Monroe C. Beardsley, "The Aesthetic Point of l'iew," in Howard E. Kiefer and Milton K. Munitz (eds.), Contemporary Philosophic Thought, Vol. IV (Albany, 1970). Ibid., pp. 224, 225. 3u Ibid., p. 228. 'O.4esthetirs, p. 83. " Ibid., p. 529. 421bid.,pp. 37, 41, 52, 53. 43 Ibid., p. 59. 'I Ibid. 4 5 Ibid., pp. 46ff. Cf. also, Richard Rudner, "On Seeing What We Shall See," in Richard Rudner and Israel Schemer (eds.), Logic 6 Art (Indianapolis, 1972). '%See further, Joseph Margolis, T h e Language of Art and Art Criticism, Ch. 2; and "Robust Relativism," Journal of Aesthetics and Art Criticism, XXXV (1976), 37-46. " Nelson Goodman, Languages of Art (Indianapolis, 1968). 48See Joseph Margolis, "Numerical Identity and Reference in the Arts," British Journnl of Aesthetics, X (1970), 138-46. 'O Op. cit., p. 40. Goodman, Problems and Projects (Indianapolis, 1972), p. 136. "Languages of Art, pp. 85, 89. S1 Ibid., p. 53. s2 Danto does so, in the Behavioral Research Directorate paper; Marx Wartofsky, in a paper, "Picturing and Representing," presented at the same Symposium, also follows Goodman here. 6a Nelson Goodman, "The Status of Style," Critical Inquiry, I (1975), 808.