Printed by permission of the author.
Ontology—the study of the nature of being and of beings—is a somewhat esoteric field of inquiry at the best of times, and might seem completely unmotivated when we turn to the kinds of beings that enter into our everyday practices, such as artworks. If asked what kind of thing an artwork is, most people would respond that obviously artworks are the kinds of things you can see hanging on the walls of galleries, or projected on screens in cinemas, or can hear performed in concert halls, or can read in the kind of thing sold in a bookstore. Appreciating an artwork is then a matter of perceptually engaging with it in one of these ways, and artworks are valued because of qualities of the experiences thereby elicited. This view is broadly “empiricist” in holding that artistic appreciation and evaluation require little if any knowledge of features of a work's history of making not determinable from an inspection of either it (in the case of paintings) or one of its instances (in the case of films, musical works, and literature). In this sense, the experiential engagement required to appreciate an artwork is “unmediated.”
One problem for this view arises if we accept a more general principle governing work in the ontology of art. According to the “pragmatic constraint” (see chapter 1 of my Art as Performance (Oxford: Blackwell, 2004)), artworks must be entities that (a) can bear the sorts of properties rightly ascribed to what are termed “works” in our reflective critical and appreciative practice, (b) are individuated in the way such “works” are or would be individuated, and (c) have the modal properties that are reasonably ascribed to such “works” where a work's modal properties pertain to how it might have differed while remaining the same work. The work is the “unit” of criticism, appreciation, and artistic value—it is what our reflective critical, appreciative, and evaluative judgments are judgments about.
The pragmatic constraint presents a challenge to the common-sense view of artworks because it is difficult to reconcile the latter with certain features of our artistic practice. For example, we think it matters whether a painting like The Disciples at Emmaus was painted by Jan Vermeer in the 17th century or by the very clever 20th-century forger Han van Meegeren, even though this presumably makes no relevant difference to our perceptual experience when we look at the canvas. Furthermore, curators often cultivate our interest in the manner in which art objects have been generated, exhibiting earlier sketches next to finished works and informing us about details of the making of a painting only determinable by X-ray photography of the canvas, for example. But it seems that the common-sense view must deny such matters a bearing on the appreciation of the works. And, finally, contemporary art is replete with works the immediately experienceable qualities of whose vehicles seem to bear only obliquely on the appreciation of those works. Does the proper appreciation of Duchamp's Fountain, for example, really consist in an unmediated perceptual engagement with the exhibited urinal? Parallel issues arise in the other arts. Those who subscribe to the common-sense view of appreciation must view our practice as simply mistaken when it treats such aspects of the history of making of art-objects as relevant to their proper appreciation as artworks. Such aspects, they often claim, bear only upon art-historical, sociological, or psychological interests in the products of artistic making.
It will be helpful here to introduce some terminology. Artworks come into existence because something is done in a context where this doing counts as doing something else. For example, an individual applies oil paint to a canvas, and this counts as the production of a painting to which various representational, expressive, and formal properties can be ascribed by receivers. Suppose we term the first doing “manipulating a vehicular medium,” and the product of this activity “the artistic vehicle.” And suppose we term the second doing “articulating an artistic content,” where “artistic content” is a catchall term for the broadly “meaningful” properties that we ascribe to products of artistic activity in our critical engagement with those products. These properties include being a kind of representation, having a particular expressive value, and making manifest some formal property or property of the vehicular medium, plus all of the higher order meanings that we ascribe to the vehicle in virtue of these basic meanings. Manipulating a vehicular medium counts as the articulation of an artistic content because the context in which the former doing takes place, or the context in which the product of the former doing is received, provides interpretive norms or shared understandings that license so taking it. We can term these shared understandings an “artistic medium.” Then the product of an artistic act of making for purposes of appreciation—what we may term the “focus of appreciation”—is an artistic content articulated by an artistic vehicle in virtue of an artistic medium.
A number of philosophers have suggested, against the common-sense view of artistic appreciation, that we can grasp a work's focus of appreciation only if we locate the manifest product—the artistic vehicle—in the context of its history of making. (See, for example, Arthur Danto, The Transfiguration of the Commonplace (Cambridge MA: Harvard, 1981); Jerrold Levinson, “What a Musical Work Is,” Journal of Philosophy 77 (1980), 5–28; Denis Dutton, “Artistic Crimes,” British Journal of Aesthetics 19 (1979), 304–14; Gregory Currie, An Ontology of Art (London: St Martin's, 1989), chapter 2; and Art as Performance, chaps. 2 and 3.) Indeed, this may be necessary even to identify the artistic vehicle—as is the case with much late modern visual art. Suppose that this is right and that, as a result, we must reject the common-sense view of appreciation. Given the pragmatic constraint, which holds ontology of art accountable to our reflective critical and appreciative practice, what should we conclude about the nature of artworks?
The standard answer is some form of “contextualism.” (For variants on contextualism, see the works by Danto, Dutton, and Levinson cited above.) Contextualists view a work as the contextualized product of an act of artistic making, as a focus of appreciation the grasp of whose relevant properties requires knowledge of certain aspects of that act and whose identity, as a work, incorporates that history of making in some way. For the contextualist, a work, qua focus of appreciation, is an artistic vehicle as generated in a particular art-historical context. For example, Jerrold Levinson claims that musical works are not, as the common-sense view might maintain, pure sound (or sound and instrumentation) structures but such structures as “indicated” in such a context. And Arthur Danto proposes a contextualist theory of visual artworks in line with an assumed contextualist theory of literary works.
There are, however, reasons to prefer an alternative and more radical response to the perceived demise of the common-sense view of appreciation. Recognizing the significance of the history of making of its artistic vehicle to the appreciation of an artwork, we might identify artworks of all kinds not with the contextualized products of the generative activities of artists—their acts of making—but with those generative activities themselves as completed by those products. On what may be termed the “performance theory,” the artwork is a particular generative performance consisting in the motivated manipulations of a vehicular medium whereby a particular focus of appreciation is specified. The work, for the performance theorist, is a particular event—the doing of certain things by one or more individuals. It is not a type of generative event that can in principle be multiply instantiated, as has been suggested by Gregory Currie in chapter 3 of his An Ontology of Art. For Currie, an artwork is a type of event consisting in the discovery of a particular structure-type through the execution of a particular kind of strategy. As a type of event, the work admits at least in principle of multiple occurrences. The performance theorist, by contrast, maintains that we should treat works as historically situated individuals, as token events or “doings.” While some works—musical and literary works, perhaps—may admit of multiple vehicles whereby their artistic contents are articulated for our appreciation, the work itself is singular, like all particular actions that we perform.
If the performance theory is to be taken seriously as an alternative to contextualism, two questions need to be addressed. First, what arguments can be given for preferring the performance theory to some form of contextualism? Given the pragmatic constraint that ties what is rightly said in the ontology of art to our artistic practices, why should we think of artworks as generative performances rather than as the products of such performances? Second, how exactly are we to understand the “performances” with which the performance theory proposes to identify artworks? While the initially counter-intuitive idea of artworks as performances might be made to sound more plausible by pointing out problems with contextualism, the devil will surely be in the details. With what “performances” could artworks plausibly be identified?
Let us look first at considerations that might support the idea that artworks are generative performances rather than the contextualized products of those performances. Because of the intimate relations that exist between the properties of any product and the properties of the process whereby it was produced, it is difficult to come up with knockdown arguments in the contest between contextualism and the performance theory. The argument for the latter must therefore be “cumulative” in the sense that individual considerations acquire more weight as it becomes clear how they fit into an overarching theoretical framework for thinking about artistic practice and our discourse about art. The analysis of the fine structure of the focus of appreciation is a crucial first step in developing this kind of argument. In arguing against empiricism, the contextualist points to the complex manner in which the different elements in the focus of appreciation are related both to one another and to the context of agency in which the focus is specified. This undermines the appeal, by proponents of the common-sense view, to a distinction between artistic properties and art-historical properties.
But it also leads us to ask which features of the history of making of an artwork—which “provenential properties” as we might call them—are “artistically relevant” in the sense that they bear upon the proper appreciation of works? Ontological contextualists who rely upon arguments for a contextualist epistemology certainly ascribe artistic relevance to those provenential properties that determine aspects of the work-focus—either the nature of the artistic vehicle or aspects of the artistic content articulated through that vehicle. We may describe such provenential properties as “focus-determining.” For example, where the representational content of a painting depends in part upon features of the art-historical context in which it was painted, as with certain represented gestures in Renaissance paintings, those features will be “focus-determining.”
But many of the provenential properties to which we refer in our critical and appreciative discourse about works do not in any obvious way play such a focus-determining role. Examples would include: overpainting visible only through X-ray technology; being generated through the use of devices such as the camera obscura; being prefigured in early drafts or sketches whose design elements are superseded in the finished vehicle; being a product of failed or unrealized semantic intentions on the part of the artist, or of experiments with the medium upon which an artist drew in her manipulations of that medium in a given work; and standing in relation to broadly biographical facts about how the artist acquired various skills, dispositions, or interests bearing upon those manipulations. The challenge for the contextualist is to deal in a principled way with our interest in those aspects of an artist's generative performance just cited that usually leave no lasting mark on the resulting focus. For the performance theorist, on the other hand, such provenential properties are artistically relevant to the extent that they bear upon the appreciation of the generative artistic performance that is the artwork.
A related advantage of the performance theory is that, in distinguishing between the work, as performance, and its focus of appreciation, it allows us to reconcile two apparently inconsistent intuitions that we may have concerning the relevance of the artist's semantic intentions—her intentions as to the artistic content of her work—upon the work's proper appreciation. On the one hand, it seems to matter that an artist intended her work to articulate a particular content, but, on the other hand, we resist the idea that merely intending that a work have a particular content can make this the case. The performance theorist can accommodate both of these intuitions by maintaining that artistic intentions about content do not generally determine the content actually articulated by a work's artistic vehicle, but are always relevant to the work's appreciation because they are among the things motivating those manipulations of the vehicular medium that constitute the work.
A further argument that complements the foregoing considerations provides an independent reason to question the adequacy of contextualism. This argument appeals to structural features of our modal intuitions about artworks—that is, our intuitions as to how a given work might have been different from the way it actually is. Contextualism has difficulty accounting for the fact that aspects of provenance bear upon our modal judgments with a variable force that reflects our overall sense of what is to be appreciated in a given work. In some cases—for example, in the case of Andy Warhol's Brillo Boxes—it is difficult to see how one could have the same work in a different art-historical context. In other cases—for example, the work of a “naive” painter—we have no such reasons to modally constrain the work. Nor can this be explained purely in terms of the kind of artistic content that a work has if, as already argued, non-focus-determining features of a work's provenance can bear upon its proper appreciation. In fact, it can be argued, our modal intuitions about works track our modal intuitions about the particular performances whereby focuses are specified. We ask, “Could this have been done under those circumstances?” where “this” refers not to the product of the artist's activity but to that activity itself, the doing qua performance. A natural way to account for this is to identify the work, about which we have such intuitions, with the relevant performance whereby its focus is specified. To do so is to endorse the performance theory.
The performance theory identifies artworks with performances generative of focuses of artistic appreciation. But, it might be asked, is there a principled way of picking out the generative performance, or “doing,” with which the performance theorist proposes to identify an artwork? After all, it may take a considerable time—even years in some cases—for an artist to execute a particular work. Surely we do not want to count everything that the artist does during that period of time as an element in the work. Even at those times when the artist is at work on a particular piece, surely many of the things she does—ordering a pizza, for example, to sustain her during an all-night session on a painting—plays no role in the identity of the work. A natural proposal here might be as follows. An artwork, qua performance, is a sequence of motivated manipulations of a vehicular medium through which a particular focus of appreciation is specified. These manipulations are to be characterized in terms of the contents that they aim to articulate and the cognitive resources upon which they draw. The doing identified with a work W will comprise those motivated manipulations we take to be instrumental in the articulation of artistic content c through artistic vehicle v, and will incorporate both features of provenance that are directly focus-determining, and manipulations that, while elements in the process generative of v, do not leave enduring marks thereupon. It will also incorporate those features of the social, cultural, and historical situation implicated in what we judge to be a perspicuous representation of those motivated manipulations. The “motives” that are partly constitutive of the work, qua doing, are ones that pertain directly to the manipulation of the vehicular medium in order to articulate a particular artistic statement. Other broader motives of the artist, and other actions she performs in the process of generating the work-focus, will not enter into the identity of the work, though they may well bear on how we explain or value what was done. While this account calls for further elucidation, it seems to offer the promise of a principled way of identifying those performances that are artworks, for the performance theorist, and thereby of accounting for those features of our artistic practice not easily accommodated by contextualist alternatives.