Chapter 4
Déjà vu All Over Again
How Danto's Aesthetics Recapitulates the Philosophy of Mind
As you will see, the tone of this essay is carping and critical throughout. That's because I am eaten up with jealousy. Danto has done something I've been very much wanting to do: namely, reconsider some hard problems in aesthetics in the light of the past 20 years or so of philosophical work on intentionality and representation. What's more – and I do find this hard to forgive – he has done it very well. The lines of investigation he has pointed out are ones which seem to me clearly desirable to pursue. In what follows, I propose to pursue some of them, though not always to the conclusions that Danto prefers. In particular, I'll be concerned with two questions Danto raises in The Transfiguration of the Commonplace: what distinguishes artworks from “mere things”? and what distinguishes art from rhetoric? Both these are, to put it mildly, large issues, and the suggestions I'll be offering are sketchy and preliminary. But not, I hope, entirely misguided.
Wittgenstein famously asked: What more than my arm's rising is there to my raising my arm? (Philosophical Investigations, para. 621). In effect, Wittgenstein imagines twin events both of which are instances of my arm going up and which are identical “to all appearances.” (What, exactly, the caveat amounts to will presently become a little clearer, but only a little.) Wittgenstein asks what it could be that makes one but not the other event an instance of my raising my arm.
In similar spirit, and taking his leaf explicitly from Wittgenstein's book, Danto imagines two objects, indistinguishable to all appearances, one of which is an artwork and the other of which is a “mere thing.” What, Danto asks, could make this difference? Danto thinks that Warhol's Brillo Boxes is an important work precisely because it self-consciously offers an example of this kind. Indeed, Danto thinks that, in doing so, Brillo Boxes signals the convergence of art with its own philosophy; there are now works of art that are about what works of art are in a way that only philosophy papers used to be.
Like many deep philosophical insights, the fact that there can be twin cases with respect to aesthetically interesting properties like being an artwork is perfectly obvious once someone has had the wit to notice it. Consider: there are those who darkly suspect that Hemingway's late novels are self-parodies. Patently, the corresponding suspicion could not arise about Hemingway's early novels. Imagine that Across the River and into the Woods had been Hemingway's first book; imagine that nothing about the text changes except this fact of chronology. The imagined Across the River is, thus, “to all appearances” indistinguishable from its twin; they are tokens of the same text type.1 But they differ in aesthetically interesting ways. Since one but not the other may be a parody, they may belong to different literary genres. They may, even, differ in their aesthetic value in virtue of this difference in genre. Cold Comfort Farm would be an awful book if it hadn't been for Hardy and Lawrence.
So, then: what does the possibility of its having a twin show us about what it is for something to be an artwork? I'm going to suggest an answer that is, I think, rather different from the one that Danto prefers. First, however, a scattering of preliminary remarks.
1 By definition, the possibility of a twin case with respect to property P shows that two things that are, to all appearances, things of the same kind may nevertheless differ in respect of P; and this difference may be “hidden”; it may fail to reveal itself even to close scrutiny. How could this be so? The obvious suggestion is that if there can be twins with respect to P, then P must be a relational property.2 I will take it for granted that Danto's thought-experiment shows that being an artwork is a relational property. The question then arises: which relational property is it?
2 Intentional Realism will be assumed, this being perhaps the major lesson that aesthetics has to learn from the philosophy of mind. Even philosophers who are dubious about the appeal to intentionality in scientific psychology admit that talk about beliefs, desires, intentions, and the like is indispensable for any legitimate purposes outside the strictly scientific. I assume that aesthetics counts as a legitimate purpose, so I'll be unabashed in supposing that people do have beliefs, desires, and intentions, and that their intentional states contribute to the etiologies of their behavior.
3 I have a story to tell about what artworks are. I like this story, but I admit that it may fit some sorts of artworks better than others. Roughly, it works pretty well where it seems unforced to speak of artworks as having intensional content, less well where it doesn't. So, it works better for literature and painting than for music; better for program music and opera than for chamber music; better for “sentimental” composers like Mahler than for “naive” composers like Beethoven. I don't understand why speaking of the content of an artwork is sometimes natural and sometimes not. Perhaps there is more than one kind of artwork.3
4 Not all twin cases are interesting, even philosophically. Wittgenstein's example is philosophically interesting because it bears on the question what makes an event an action;and the notion of an action is involved in all sorts of ways with deep questions about commonsense psychological explanation and about ethical evaluation. By contrast, it's not very interesting that there could be twins one of whom is a banker and the other of whom isn't; or even that there could be twins one of which is a genuine Da Vinci and the other of which isn't. It's obvious that bankers and genuine Da Vinci's are constituted by their relations (to banks in the one case and to Da Vinci in the other); what else is new?
Whether it matters that artworks can have twins depends, inter alia, on how centrally the property of being an artwork figures in aesthetics. I have two reasons for supposing it may not be very central, one of which I will tell you now and one of which I will keep till the end.
It may be, as Danto suggests, that artwork is an evaluative concept; the making of mere things into artworks is the “transfiguration” that Danto's title has in mind. Still, it's arguable that being an artwork, all by itself, doesn't buy you much in the way of kudos. Thus something can be indisputably an artwork and all the same be, for example, uninspired, tiresome, vulgar, crude, tedious, superficial, meretricious . . . indeed, quite without aesthetic value. Someone who is perfectly prepared to agree that Brillo Boxes is an artwork might thus consistently characterize it as any or all of the preceding. Indeed, he might be right to do so.
Conversely, all sorts of aesthetic properties can coherently be ascribed to things that plausibly are not artworks (as opposed, say, to aesthetically gratifying artifacts). I suppose Greek pots aren't artworks, but many of them are wonderful objects and we're right to honor the artisans who made them. It is, moreover, perfectly appropriate that they should be displayed in museums of art, since what curators care about isn't primarily whether an artifact is an artwork but rather what aesthetic responses it evokes. I'm inclined to think that Chartres isn't an artwork either; to so describe it would be to evince a sort of aestheticism that quite misunderstands the building. But Chartres is, of course, among the greatest of human creations, and intense aesthetic response to it is entirely appropriate. One can imagine a Chartres twin that clearly is an artwork; it helps to do this if you imagine a Chartres twin that wasn't built to be a cathedral. But it isn't at all clear that Chartres being an artwork would improve it; in this case, it might be a failure of sensibility to prefer the artwork to the twin.4
So much for the time being about why and whether being an artwork matters; but, as will presently become apparent, I think there's more and deeper to be said. It is a main goal of this essay to propose an answer to the following question: “If artwork isn't the central notion in aesthetics, what is?” OK, end of asides and caveats; now to the main business.
It was illuminating of Wittgenstein to ask what makes a mere event – a mere movement of one's arm – an action. But he gave the wrong answer; at least he did as I read the text. (It's notorious that with Wittgenstein you can hardly ever tell how to read the text. I don't propose to argue for my exegesis.) Apparently, Wittgenstein's view is that to understand a movement as an action, rather than a mere happening, is to understand it as contextualized in a certain way. What way? Well may you ask. Here are some of the sorts of things that the Investigations says:
Why do I want to tell him about an intention too, as well as telling him what I did? – Not because the intention was also something which was going on at that time. But because I want to tell him something about myself, which goes beyond what happened at that time. (para. 659)
Or again:
“For a moment I meant to . . .” That is, I had a particular feeling, an inner experience; and I remember it. And now remember quite precisely! Then the “inner experience” of intending seems to vanish again. Instead one remembers thoughts, feelings, movements, and also connections with earlier situations. (para. 645)
Or again:
If I say “I meant him” very likely a picture comes to my mind . . . but the picture is only like an illustration to a story. From it alone it would mostly be impossible to conclude anything at all; only when one knows the story does one know the significance of the picture. (para. 662)
This is all pretty opaque. But maybe the view is that it's the context that makes the difference between a motion and an action; perhaps in the way that the context might make the difference between kinds of actions (between, for example, action twins, one of which is the making of a bid and the other of which is the hailing of a cab). To understand an arm's rising as someone raising his arm is to set it in a context that may include, but needn't be exhausted by, the agent's past and future actions and intentions. No doubt this account would be question-begging if a reductive analysis were the goal; but surely, in Wittgenstein, it's not.
If this (putatively) Wittgensteinian story is right, then you can see straight off why there can be action twins. Being set in a context is a relational property par excellence; hence it's the kind of property that a thing may have and yet not reveal to the closest scrutiny – so long, that is, as the scrutiny is confined to the thing's non-relational properties.
So much for Wittgenstein; however, in detail, his story about action is supposed to go, I suppose that it's intended to exclude what I'll call the “Cartesian” story about action. A first approximation to the Cartesian story is this: in the typical case, what makes a motion an action is that it is caused, in the right sort of way, by the agent's intentions. In the typical case, for example, what makes a motion an act of F-ing is that it is caused, in the right sort of way, by an intention to F. (What makes a rising of an arm an arm raising is that it's caused, in the right sort of way, by an agent's intention that his arm should rise.)5
I don't, actually, have much doubt that the Cartesian story about action has the stick by the right end, but I'm not going to argue for that here. Suffice it that the Cartesian story is, at worst, in the running; and that if it's true, it too would explain why there can be action twins. Having the causal history it does is itself a relational property of an event, hence it's a property that may distinguish events that are “indistinguishable to all appearances.” Differences in etiology, and, in particular, differences in intentional etiology, are therefore candidates for answering questions about what distinguishes action twins. Well, to come to the point at last, this option also suggests itself in the case of artwork twins. A relatively unilluminating version of the Cartesian proposal might be that what makes something an artwork is that it was intended as an artwork by whoever made it. In which case, it could distinguish between an artwork and a mere thing by saying that the latter but not the former was made with the intention of providing a container for Brillo pads.
This proposal is, as I say, relatively unilluminating as it stands; it wants elaboration. (What is it to be caused by an intention “in the right way,” for example?) But unelaborated doesn't mean empty; the Cartesian analysis has striking and tendentious consequences even as it stands. For example, it implies that nothing could be an artwork if it is made in a period, or in a society, or by a species, to which the concept of an artwork is unavailable. There could thus be no artworks made by chimpanzees or (current) computers, though chimpanzees and computers could, of course, conceivably make aesthetically gratifying artifacts. It might also turn out, according to the Cartesian analysis, that there couldn't have been any Paleolithic artworks; that would depend on facts about what the minds of Paleolithic artisans were like – in particular, on facts about what intentions they entertained with respect to the artifacts they created. And to these facts we are, de facto, not privy.6 I take these consequences of the Cartesian proposal to be certainly substantive and, intuitively, reasonably plausible.
But though it has these substantive and plausible consequences, I admit that the Cartesian proposal isn't much help as it stands. This isn't because it's etiological or because it's intentional; it's because intending to make an artwork wants explication in a way that, say, intending one's arm to rise does not. For, of course, the notion of an artwork is unclear in the way that the notion of an arm's rising isn't. We know perfectly well what it is that one intends when one intends that one's arm should rise; but it's a lot less clear what it is that one intends when one intends that something should be an artwork. It is easy, at this point, to suppose that the Cartesian proposal is not just not very helpful but that it's empty after all. Doesn't it just take the original problem and push it inside the head? No, in fact, it doesn't. On the contrary, it tells us that “artwork” is an etiological concept (thereby explaining how there can be artwork twins); and it connects the intensionality of artworks (their aboutness) with the intensionality of mental states. Both of these things that the Cartesian theory tells us are substantive; indeed, given the current climate in Anglo-American (to say nothing of French) philosophizing, they both count as tendentious.
I therefore assume that the Cartesian story is substantive on the face of it, and that it may even be true, but that it is, as things stand so far, too under-described to argue much about. Most of what follows will be devoted to unpacking a version of the intentional etiology story about artworks for the reader to consider. First, however, I want to ask, briefly and in an exegetical frame of mind, how much – if at all – a Cartesian account of what makes something an artwork would conflict with the views that Danto sets out in Transfiguration.
Danto is clearer about the importance he attaches to twin cases than about what he thinks the metaphysics of artworks actually is. However, some of the things he says suggest a story about artworks that is recognizably kin to Wittgenstein's story about actions; except that, in Danto's case, it's the historical context (the “artworld” context) rather than biographical context that bears most of the metaphysical burden. Discussing twin drawings, one of which is an artwork and the other of which is a graph, Danto says:
it may very well be under the concept of style itself that we might look at our various visually indiscernible curves: the graph . . . lacks any stylistic characterization just because it is a graph, whereas . . . the [indiscernible] drawing by Hiroshige is perhaps . . . disciplined and controlled . . . [I]t may be objected that it is only because we know the histories of these curves that we ascribe stylistic predicates to works that are perceptually not to be told apart. But if it is in terms of differential histories that we discriminate between those things in principlesusceptible to stylistic ascription but not to be told apart by immediate perception, it is far from plain that differential histories may not be the sort of tool we are seeking.7
“Differential histories” apparently doesn't mean (or doesn't mean just) different intentional etiologies. Danto is explicit that (some?) “works are in part constituted by their location in the history of literature as well as by their relationships to their authors.”8 Perhaps the right thing to say is that Danto's view neither excludes Cartesianism nor insists upon it. Danto sees that, since there can be artwork twins, the essence of artworks must be (inter alia) relational. I'm proposing a Cartesian account of what the relational essence of an artwork is: an artwork is so constituted by its intentional etiology, by its being intended to be an artwork.
Well, so be it; but what is it to intend a thing to be an artwork? For reasons set out above, unless something can be said about this, one might reasonably view the ground that the Cartesian story gains as minimal even if the Cartesian story is true.
Here's a first installment: the intention that a thing be an artwork is in part the intention that the thing should have an audience. I think, in fact, that nothing can be an artwork unless it is intended to have an audience; hence that the concept of an artwork can't predate the concept of an audience.9 And, whereas (see below) I'm not entirely clear that artwork is a very important aesthetic category, I'm morally certain that audience is.
At a minimum, an artwork is something that is intended to have an audience; that's how it can be that Brillo Boxes is an artwork though Brillo boxes aren't. Whereas Brillo Boxes is intended to be shown, to be exhibited, Brillo boxes are intended merely as boxes for Brillo. What it is for something to be an audience is, of course, itself a complicated business. But I take it as plausible that audiences are also constituted, largely, by their intensional states. In fact, I take it that you can't be (or be in) an audience unless you know and expect all sorts of things about artists, performers, works of art, and the like. (So my point isn't to break out of the aesthetic circle; just to enlarge it a bit.) The audience for a novel is constituted, in part, by its knowing about the difference between writing fact and writing fiction. The audience for an opera is constituted, in part, by its expectation that the actors will sing their lines. (Parlando makes its effect by flouting this expectation.) The audience for Brillo Boxes is constituted, in part, by its knowledge that it would be inappropriate to attempt to use Brillo Boxes as a container for Brillo. Or, in any event, that to do so would itself be a sort of critical gesture. Putting Brillo pads in Brillo Boxes would count as art criticism, even though it is, “to all appearances” indistinguishable from things that happen routinely at the Brillo plant that don't count as art criticism. This is itself in part because one gesture but not the other is intended to have an audience.
Conversely, Greek pots aren't artworks because they were intended to put (the Greek equivalent of) Brillo in. This is not a judgment of aesthetic value. I admire many Greek pots enormously, and I don't admire Brillo boxes at all.
I suppose that an audience is something that is intended to be affected in certain ways by an artwork; and an artwork is something that is supposed to have certain sorts of effects on an audience. Hence the conceptual connection between artwork and audience. Hence too, however, the conceptual connection between art and rhetoric, to which Danto insightfully draws our attention: “it is the function of rhetoric to cause the audience of a discourse to take a certain attitude toward the subject of that discourse: to be caused to see that subject in a certain light.”10 Correspondingly, “it may just be one of the main offices of art less to represent the world than to represent it in such a way as to cause us to view it with a certain attitude and with a special vision.”11 These remarks are, I think, distinctly pregnant; it's this direction in Danto's thought – rather than the ideas about artworks being constituted by their historical context – that I propose to pursue.
If artworks and rhetoric are both (partially) constituted by their intentional etiology – they are both intended to have effects on audiences – then the question arises how they differ. This can itself be viewed as a sort of twin problem. Antony's funeral oration for Caesar is, I suppose, beyond question an artwork of some accomplishment. If the historical Antony had given that same speech on the historical occasion of Caesar's funeral, then the text he produced would have been a twin of Shakespeare's text (modulo translation). But Antony's text would have been not an artwork but a rhetorical exercise. Both texts would have been intended to affect their audiences. So, what makes one twin art and the other artful rhetoric?
There are at least three lines one might explore for an answer to this question. One possibility would be to insist on the context. For better or worse, however, my project is to try to reconstruct the required distinction within a Cartesian aesthetics; one which explicates aesthetic categories in terms of intentional etiologies. So let's, for purposes of pursuing the argument, put contextualism to one side.
A second possibility would be to distinguish between the kinds of effects that art and rhetoric are intended to produce in their respective audiences. Quite a lot of aesthetics has taken this line (compare the traditional idea that artworks are intended to evoke specifically aesthetic emotions) but not profitably in my view. My guess is that just about any effect upon an audience that rhetoric may be intended to have, artworks may be intended to have too. A longer treatment would need to argue this case; for present purposes I shall simply assume it. Suffice it that the intention to work on the passions or the convictions of an audience informs lots of art, including lots of the highest art. Wagner provides notorious examples: consider the first few minutes of Lohengrin and the last few minutes of Meistersinger.
I think that the important difference between art and rhetoric is not in the effects they aim at, but in the means that they employ to make their effects. I think Danto thinks this too, but at this point I find I lose the thread of his discussion. As far as I can make out, Danto holds that the clue to distinguishing art from rhetoric is in an intrinsic connection between art and metaphor. “To understand the artwork is to grasp the metaphor that is, I think, always there.”12 To that extent, it would indeed be the intended means rather than the intended effect, that matters. But Danto doesn't always hew to this line; there is a lot about the artist making us see things the way he did;13 and about “the enactment of a metaphoric transformation with oneself as subject; you [the reader] are what the work [Anna Karenina] ultimately is about, a commonplace person transfigured into an amazing woman.”14 I don't know what to make of this aspect of Danto's book; I'm pretty sure that there isn't any useful sense in which Anna Karenina is about me, ultimately or otherwise. Certain passages in Winnie the Pooh, perhaps, but definitely not Anna Karenina. Since I don't really understand Transfiguration's account of the art/rhetoric distinction, I propose to strike out on my own.
Art and rhetoric – art and advertising, to take a case – aim at affecting an audience; and, as I've been saying, one can imagine examples in which the two aim at much the same effect. So what about the intentional etiology of a mere thing could make it an artwork rather than (or, perhaps, as well as) an advertisement? I want to bring to the discussion of this question two ideas developed in other areas of philosophy where intentional activities are analyzed. These are: the distinction between the primary and secondary intention of an act; and a certain “Gricean” picture of what goes on in communication. I hope to convince you that these ideas can be interweaved in ways that illuminate both the art/rhetoric issue and the metaphysical issues that are raised by the possibility of artwork twins.
Let's start with the Gricean story about communication (or rather, with a version of the Gricean story that's adapted to our present purposes).15
Short of telepathy, communication is invariably mediated by a vehicle of communication; artworks, in our case, but also, for example, utterances in the case of speech, flashing lights in the case of stop signs, and so forth. When communication is the goal, the vehicle is produced with the intention of causing certain (typically psychological) effects on an audience. For example, in standard cases, the English sentence type “It's raining” is tokened with the intention of causing the audience to believe that it's raining in the vicinity of the speaker.
So then, nothing is a vehicle of communication unless it is intended to have an effect upon an audience. But, according to the Gricean analysis, there is a further condition that communicative gestures must also satisfy. Not only are they intended to have an effect upon an audience, but they must be intended to have their effect in a certain way; that is, at least in part in virtue of the audience's recognition of the speaker's intention that the communicative gesture of tokening the vehicle should have the specified effect. Call this the reflexive condition for something being a vehicle of communication. Another way of putting the reflexive condition is this: a communicative gesture is ipso facto intended (1) to be interpreted by its audience in a certain way; (2) to be recognized as intended to be interpreted by its audience in that way; and (3) to have its effect on its audience in part in virtue of the audience's recognizing that the gesture was intended to be so interpreted.
I don't care much whether artworks satisfy all the conditions for being vehicles of communication; but I do care, and am going to assume, that they satisfy this reflexive condition. Artworks are intended to affect audiences in certain ways, and to do so partly in consequence of the audience's recognition of this intention. If, for example, Danto is right about Brillo Boxes, then it is intended to be recognized as intended to cause its audience to consider the question what distinguishes artworks from mere things. And similarly, so far, with advertising. A sign that says “Buy Brillo” is intended to get its audience to buy Brillo, and to get its audience to do so, in part, in virtue of its recognition that the sign exhorts its audience to buy Brillo.
So then, I take it that both artworks and ads satisfy the reflexive condition for being a communicative gesture. This is, perhaps, of some intrinsic interest, but of course it doesn't do what we want, which is to make clear how artworks and ads are different. To do that, we need to throw in the distinction between the primary and the secondary intention(s) of an action.
Here's a rough and ready introduction to that distinction: a state of affairs is the (or a) primary (intentional) object of an action if the action counts as a success in every world where it brings about that state of affairs and a failure in every world where it doesn't. A state of affairs is the (or a) secondary object of an action if and only if it is, for the agent, an object of the action but not a primary object of the action. States of affairs that agents secondarily intend may be seen as means to the satisfaction of primary intentions. Or they can be foreseen but not desired consequences of satisfying primary intentions (and there may be other sorts of cases too). No doubt the primary/secondary distinction could do with a lot of working up, but that is not my present concern. I'll settle for showing how drawing this sort of distinction might be of use to a Cartesian aesthetics.
So, now, what distinguishes ads from artworks?
I want to suggest that, in the ad case, the intention is primarily just that they should have their effect upon the audience. The reflexive intention (viz., the intention that the effect be brought about by the audience's recognition that the ad is intended to bring it about) is itself merely secondary. Let's suppose that encounters with the “Buy Brillo” ad do, indeed, cause the audience to buy Brillo, but that they do so by some peculiar psychological mechanism, viz., not by virtue of the audience's recognition that the ad exhorts it to buy Brillo. We might imagine, for example, that the whole business is subliminal and associative. The point is: I suppose the advertiser would be quite content with this. As it were, he wills the ad's causing the sale of Brillo unconditionally; but he wills the ad's satisfaction of the reflexive condition only conditionally; that is, only as a means to the Brillo getting sold. Hence, the logical possibility, in advertising, of “hidden persuasion.”
But now consider Brillo Boxes. Suppose it does succeed in causing its audience to consider the question what distinguishes artworks from mere things, but fails to do so in virtue of the audience interpreting it as so intended. Suppose, in fact, that it too works purely subliminally and associatively; it just turns out that subliminal exposures to Brillo Boxes stirs, in its audience, an impulse to consider the question in what way artworks differ from mere things, just as subliminal exposures to the “Buy Brillo” ad stirs, in its audience, an impulse to buy Brillo. Then Brillo Boxes has the right effect (it causes the audience to be in the intended psychological state) but not by the right means; not in the way that constitutes Brillo Boxes as an artwork. Artworks qua artworks can be intended to persuade; but they can't, qua artworks, be intended to be hidden persuaders. This is a fact about artworks that a theory of their metaphysical constitution ought to capture.
I take the moral to be that the intention that the reflexive condition be satisfied is primary in the case of the artwork but only secondary in the case of the ad. In so far as a thing is not primarily intended to satisfy the reflexive condition, it is not intended to be an artwork. And, according to the Cartesian view, in so far as a thing is not intended to be an artwork, it isn't one. So that's what distinguishes artworks from ads (and, mutatis mutandis, from other sorts of rhetorical gestures).
I want to call your attention to a further aspect of a Gricean treatment of communication that may have special significance for aesthetics. A communicative gesture is ipso facto intended to have a certain effect on its audience and to do so in virtue of the audience's recognition of the gesture as intended to have that effect. I take it that, when the communicative gesture is the production of an artwork, it is part of the primary intention (it is a condition of the gesture being successful) that it be recognized as so intended. Now, normally, in art and elsewhere, a vehicle has its communicative effect by causally interacting with an audience. It is because you see the sign that says “Buy Brillo” that you come to recognize that you have been enjoined to buy Brillo. And, indeed, it is part of the intention with which the sign is displayed that you should come to recognize that you are so enjoined in consequence of your seeing the sign (specifically, in consequence of your reading the sign). However, the question may be raised whether the intention that there be this causal interaction between the vehicle and its audience is part of the primary intention of the communicative gesture, or whether it is only secondarily intended.
Here too, I think, artworks and ads come out differently. It may be that I am caused to buy Brillo by coming to recognize that I am enjoined to buy Brillo by the Brillo sign without my coming into causal interaction with the Brillo sign. For example, someone may tell me that there is this enormous sign “Buy Brillo” and that may cause me to recognize that I have been enjoined to buy Brillo, and that, in turn, may cause me to buy some. In which case, I suppose, the Brillo people would again be perfectly satisfied. But, though you can have “word of mouth” advertising, you can't, I think, have word of mouth art. So, imagine that it works out this way with Brillo Boxes: imagine that I never do get to see the thing, but Brillo Boxes causes Arthur Danto to write a book in which he says that Brillo Boxes is intended to cause its audience to consider the question what distinguishes an artwork from a mere thing; and I read the book, and am thereby caused to consider the question what distinguishes an artwork from a mere thing. I suppose that this would not count as a case where Brillo Boxes has the intended effect that constitutes it as an artwork; that is because, though it has had an intended effect on me, it hasn't had its effect “in the right way”; viz., in consequence of a causal interaction between audience and artwork. If this is right, then its having its effect on its audience in consequence of its causal interactions with its audience is part of the primary intention that constitutes Brillo Boxes as an artwork. It follows that I would not count as part of the audience for Brillo Boxes if it had its effect on me only via Danto's mediation. “Presence” is, I think, an essential condition for aesthetic success; an object that doesn't need to be encountered to make its effect – an object that would have much the same effect if you were just told about it – either isn't intended as an artwork, or, if it is so intended, then it's a failed artwork.16 I suspect Brillo Boxes of being a failed artwork in just that way. I also suspect that conceptual art (to which, if Danto's interpretation is right, Brillo Boxes is near kin) typically fails in this way.
Well, so much for the difference between art and advertising. I now want to say something about the most familiar line of objection to the Cartesian kind of aesthetics I've been promoting. This will bring us to reconsider the question I tabled at the beginning: whether artwork is, after all, the category of objects that aesthetics primarily cares about. And then we'll be through.
“Look,” someone might say, “according to the present proposal, artworks are constituted by their intentional etiologies. But surely we don't have to know what the intentions of the artist were in order properly to appreciate a work of art. Surely, much of the time we simply don't care what the artist's intentions were; sometimes, indeed, we might actively prefer not to know. Suppose it turned out (against all likelihood) that Weston never noticed that his photographs of peppers suggest the shapes and textures of the female nude. Would that really matter to how we do – or should – see the photographs? Or, to take a less unlikely case, do you really want to know what Wagner was up to in the last few minutes of Meistersinger? Isn't it distinctly plausible, in this case, that knowing its intentional etiology would get in the way of responding to the opera?”
Now, strictly speaking, I could admit all that while continuing to hew to my main thesis. Even if its intentional etiology is (or may be) irrelevant to appreciating, or understanding, an artwork, it wouldn't follow that its intentional etiology is irrelevant to whether something is an artwork. The (roughly, normative) questions about aesthetic appreciation might get answered in quite different terms than the (roughly, metaphysical) questions about what constitutes something as an artwork. I admit, however, that this line smacks of legalism, and I won't insist upon it. I'm inclined, on the contrary, to agree that we sometimes don't really much care about the intentional etiology of a work, and that this attitude is, often enough, perfectly appropriate. It looks like the Cartesian picture suggests (even if it doesn't quite entail) that we always ought to care, and to that extent the Cartesian story would seem to be in trouble. In so far as there is something more to the current anti-Cartesian consensus in aesthetics than a crazy epistemological skepticism about intentionality, I suppose this observation is what it rests on.
The Cartesian intentional/etiological story about artworks, like the Wittgensteinian contextual story, explains how there could be artwork twins, and does so by calling our attention to the artwork's relational properties; to properties of the artwork that, as it were, lie outside its frame. But both stories have the defects of this very virtue; they both seem to direct our attention to what is often intuitively peripheral to the artwork's aesthetic effect. The one suggests that we should care more than perhaps we do about the intentional/causal history of a work, the other suggests that we should care more than perhaps we do about its historical context. It's an interesting question when we should care and when we shouldn't; but, at a minimum, an aesthetics that is responsive to our intuitions should leave room for both kinds of cases. So what, then, ought a Cartesian say about (what's called in the trade) the “intentional fallacy?”
Here's what I'm inclined to think (though only in briefest outline). I'm inclined to think that, though it isn't very important what intentional etiology a thing has, it is very important what intentional etiology it might have had. (Equivalently, it's not very important whether a thing is an artwork, but it's very important whether it could be.) This is to say that what really matters aesthetically is whether a work has, within its frame, the sort of structure that would be coherent with the sort of intentional etiology by which, according to the preceding discussion, artworks are constituted.
Correspondingly, much of the business of interpreting art is the construction of, as one might say, “virtual” intentional etiologies in terms of which the within-frame features of a work are intelligible. Roughly, to interpret a work is to exhibit its within-frame features as compatible with its having been made with the primary intention that it produce a certain effect on its audience, and that it be recognized by its audience as intended to produce that effect. “Compatible with” doesn't, of course, mean logically compatible with; that would be much too weak. I suppose it means something more like: compatible modulo our intuitions about the psychology of our conspecifics; modulo, if you like, our implicit folk theory about what kinds of intentional states people can get into, and how the intentional states that people are in can show up in the properties of the things that they make. For example, given the facts of folk psychology, and quite independent of knowledge of its actual causal history, it is wildly unlikely that Brillo Boxes could have been intended to be recognized as intended to cause its audience to reflect upon the spiritual affinity between death and birth; or, mutatis mutandis, that The Burial of Count Orgaz could have been intended to be recognized as intended to cause its audience to reflect upon the metaphysical relation between works of art and Brillo boxes. Folk psychology tells us, inter alia, what it is reasonable to expect reasonable people to expect. And nobody could reasonably expect those artworks to produce those effects.
Much of one's pleasure in a work consists in being able to see it as the possible product of a plausible virtual intentional etiology. No doubt, one is often professionally, or sentimentally, or anecdotally, or financially, also interested in whether the actual causal history of a work corresponds to one's favorite virtual etiology. But, by and large, specifically aesthetic concerns abstract from the contingencies of a work's production. It is, I think, the internal connection between aesthetic value and virtual etiology that we acknowledge when we speak of important artworks as “timeless,” “universal,” and the like. The moral of the twin cases was that things are artworks in virtue of their actual intentional etiologies; but the universality of an artwork consists precisely in its independence from its historical context, its ability to speak from, and to, the human condition as such.
If that analysis is right, then, strictly speaking, it's interpretable artifacts, rather than artworks, that we really care about. Since, strictly speaking, artworks are constituted by their actual causal histories, artwork is a historian's category, not a critic's or an audience's, or an aesthetician's. But why is it just interpretable artifacts that we care about? Why not interpretable things (sermons in stones, books in running brooks, and so on)? Partly I think this is bedrock; we aren't so interested in made things; specifically, we are interested in things that are made by us. But partly it's a matter of putting bounds to the game. If it's a human artifact, then folk psychology17 limits the possibilities of interpretation; if it's not, then anything goes. The trouble with, for example, trying to interpret the world as God's artwork is: who knows what He might have had in mind in making it?
Well, there is a lot more to be said about this, but not here. Suffice it to stress the main point: a Cartesian can think that the vocabulary of intentional etiology is the right one for reconstructing aesthetic distinctions without also having to think that knowing – or caring – about what the artist's intentions actually were is at the core of aesthetic appreciation. The failure to notice this possibility is, I think, epidemic in current discussions of aesthetics.
Danto has shown us that what lies outside the art object – the kinds of relations that can distinguish between artwork twins – matters to the metaphysics of artworks. What we now need to do is square this discovery with a proper respect for the artwork's frame. The Cartesian account of the metaphysics of artworks that I've been proposing militates in this direction since, according to that account, it is part of the artwork's conditions for success that its audience should recognize it as intended to produce its effects and should do so in consequence of encounters with the artwork. But what recalls us most to what is in the frame is that it is in large part virtual rather than actual etiology that counts aesthetically. Because this is so, and because with successful artworks the possibilities for virtual etiology can be virtually boundless, questions of aesthetic interpretation are hardly ever settled by purely historical enquiries. And, because artwork twins ipso facto have their virtual (though not, of course, their real) etiologies in common, it is plausible that aesthetic value supervenes largely on what artwork twins ipso facto share.
Here, then, is how art differs from life. Aesthetic appreciation cares about the virtual etiology of artworks, but psychological explanation and moral evaluation care about the actual etiology of behaviors. Aesthetics care about ideal causes, psychology and ethics care about real ones. Which is, I suppose, why Aristotle and Nietzsche thought that art is truer than history or ethics respectively.
Notes
I want to thank my CUNY colleague, David Rosenthal, for his exceptionally insightful comments on an earlier draft of this essay. Thanks also to Peter Kivy, Ernie Lepore, and Bob Matthews for very helpful discussions.
1. You might want to argue that precisely because they are tokens of the same text type, they aren't, strictly speaking, twins. Piles of Brillo boxes aren't tokens of the type Brillo Boxes; perhaps no two objects can be tokens of the same construction (or painting) type (compare etchings, photographs, and cast sculptures). This doesn't, however, affect the main point, which is that, once you've seen that there can be artwork twins, you see that there are plenty of cases where objects that are “to all appearances” indistinguishable can nevertheless differ in aesthetically interesting ways. Like aesthetic differences between twins, aesthetic differences between tokens of the same artwork type provide cases in point.
2. Another possibility is that whether a thing has P is determined by its microstructure. Hence Putnam's widely discussed example in which one but not the other of two superficially indistinguishable substances is actually water. I assume that the artwork case is not of this kind. What determines that this twin is water and that one merely fools-water is that only this twin is H20. But whether a twin is an artwork isn't, surely, a matter of its chemical analysis.
3. I've been feeling much less unhappy about this concession since reading Peter Kivy's article “Is Music an Art?” Journal of Philosophy, 88(10), 1991, 544–54.
4. It's not essential to my purposes that you share these intuitions, though it would be nice if you did. What is essential is that you agree that a theory about what artworks are should leave the question of aesthetic value open; that it should allow for aesthetically gratifying objects that certainly aren't artworks and for objects that certainly are artworks but are no damned good. I shall take that for granted in what follows.
5. For a variety of reasons, including, notably, the relational character of many action descriptions, this Cartesian condition for something being an act of F-ing couldn't be either necessary or sufficient if “F ” is allowed to vary freely. But it might hold for a philosophically interesting set of core cases.
6. The caveat “de facto” is essential. It is a standard objection to Cartesianism in aesthetics (and elsewhere) that intentional etiologies are epistemically inaccessible; so that, in the present case, if Cartesianism is true, then we can never know (or, maybe, can never know “for sure”) whether a thing is an artwork. This objection strikes me as preposterous; it rests on the assumption that we can never know (or know for sure) what intentional state someone is in, and why on earth would one want to grant that? It is, in any event, a Very Bad Idea to confuse metaphysical questions (“What is an artwork?”) with epistemological questions (“How do we know whether something is an artwork?”) Here too it's déjà vu all over again.
7. Danto, The Transfiguration of the Commonplace (Cambridge, 1981), p. 142.
8. Ibid., pp. 35–6.
9. Oh, well, I suppose there might be an artist whose intention it is that some or all of his or her artworks should never see the light. We're told that this was true of such otherwise unlikely a pair as Kafka and Emily Dickinson. (“Je chante pour moi-même” Carmen sings, singing, of course, for José and for us.) I don't think this really matters much, so long as such cases may plausibly be treated as peripheral. Perhaps, in these cases, intending that something be an artwork is intending that it should belong to a kind of which the paradigms are intended to have audiences. If you don't think it's the kind of thing that they would publish in the normal course, why bother telling them to burn it?
10. Transfiguration, p. 164.
11. Ibid., p. 167.
12. Ibid., p. 172.
13. Ibid., p. 164.
14. Ibid., p. 173.
15. What Grice actually had in mind was, of course, a theory of meaning, not a theory of communication. For that reason among others, he really mustn't be blamed for any of what follows.
16. The question what it is to encounter (to interact causally with) an artwork gets a little tricky in cases where the type/token distinction is problematic. Have I “encountered” the Waldstein Sonata if I have just read the score? If I have heard only bad performances? These sorts of questions are perhaps of intrinsic interest, but I take it that the fact that they arise doesn't jeopardize the claim in the text.
17. Or, for that matter real psychology. Consider how Freudian theory extended the possibilities for art interpretation.