Dominic McIver Lopes and I agree that (in his words) “to understand a photograph is at least in part to explain the act of making it, which is to get at what the photographer meant.”1 This agreement is not universal. Many people think that the artist’s intention is either irrelevant or at least not dispositive in determining the meaning of any work of art. Others think that even if the interpretation of, say, a painting must necessarily be an interpretation of what the painter intended, intention cannot play that role in photography. If the famous Cartier-Bresson photograph of Cardinal Pacelli were a painting, for example, there might easily be some question about why the painter depicted the mustached man to the left of the Cardinal looking down but there could be no question at all about the fact that the painter meant to show him looking down. A painter who preferred to show him looking at the Cardinal could just make him look at the Cardinal; Cartier-Bresson, however, had no such power. He may have wished the man were looking at the Cardinal but the direction of the man’s gaze was entirely determined by the depicted man, not by the man doing the depicting. This is why Susan Sontag says that “Photographs don’t seem deeply beholden to the intentions of an artist.”2 It’s what Lopes calls the standard theory of photography.
Lopes thinks the standard theory is misleading and that photographs are not distinctively unintended. Following Elizabeth Anscombe, he argues that all acts are, like the act of taking a photograph, intended under some description (Cartier-Bresson meant to show Pacelli) and unintended under others (he didn’t mean to show three balding men), and hence that “Photographic agency is plain vanilla: no different from the agency of…the painter.” And insofar as my discussion in “I Do What Happens” (the title is a quote from Anscombe) dealt with this view, my point was not to disagree. I think that what Lopes calls his “first deflationary maneuver” (because it deflates the idea that photographic acts are either some philosophically special kind of act or not acts at all) is right.
But I also think it doesn’t actually deflate anything that matters. Lopes’s point is that the unintended aspects of the act are just as much a part of the act as the intended ones, in the way, for example, that the unintended aspects of my typing this response (using electricity) are as much a part of the act as the intended ones (responding to Lopes). But that doesn’t make them a part of its meaning. So if you think that explaining what a photograph means is explaining what the photographer meant by it, the reminder that even if Cartier-Bresson only made the man look down unintentionally, he still made him look down, is beside the point. The question is not only what did Cartier-Bresson do but what did he mean?
Hence, if you want to preserve the force of the first deflation (photographic acts are like every other act), you need a second deflation, one that will make photographic meaning as vanilla as photographic agency. And Lopes tries to get this by returning to what we both agree on—that “to understand a photograph is at least in part to explain…what the photographer meant” but this time insisting on the “in part” part, and arguing that just as agency is not entirely identified with intentionality, meaning isn’t either. There’s a difference between “the semantic content of a vehicle of communication” and “what the communicator uses the vehicle to communicate.” Just as the sentence meaning of a speech act may, for example, be different from the speaker’s meaning so may what the photograph records be different from what the photographer seeks to communicate.
But this second deflation seems to me to work no better and in some respects less well than the first one did. Where the first one didn’t really deflate anything that matters, this one doesn’t deflate anything at all. Suppose the meaning of the photograph is indeed determined by both the photographer’s intention and the relevant “facts about the working of the photographic mechanism.” This obviously doesn’t alter the fact that any interpretation of the photo will be (at least “in part”) an account of the photographer’s intention. And if what’s distinctive about photography are the kinds of issues it raises not about action as such but about action in relation to meaning, those issues are either left untouched or accentuated by the appeal to the mechanism. In the painting of Cardinal Pacelli, for example, we might have an interpretive disagreement about what was meant by the fact that certain figures were shown looking away from the central figure, and that argument would be about what the painter meant by depicting them in this way. But with respect to the photograph, a different sort of argument is not only possible but sometimes unavoidable: did the photographer mean anything at all by depicting them in that way? And this disagreement (which for both Lopes and me is over at least “a part” (the intended part) of what the photo means) is utterly unaffected by the reminder that, whatever the photographer meant, that’s what the camera recorded. We already know that.
In other words, the effort to appeal to a theory of action and then of meaning in order to demonstrate that there’s nothing theoretically distinctive about action and meaning in photography is not so much mistaken as it is beside the point. Why? Because it’s not true that “Getting the nature of action right deflates the difficulties that seem to come with the exercise of photographic agency.” Actually, almost the exact opposite is true. If we can get the nature of action and meaning even a little bit right, then we begin not to deflate the difficulties that seem to (and really do) come with photographic agency but to understand them. It’s really hard to paint a picture of a man looking down without meaning to; it’s really easy to take a photo of a man looking down without meaning to. So in photography the question of what was meant can be shadowed by the question of whether it was meant in a way that it isn’t in painting. That’s a difficulty (also an opportunity) that comes with the exercise of photographic agency.
Putting the point this way, however, my criticism of Lopes might also identify a real area of agreement. Insofar as Lopes’s idea is, first, that because there are no philosophical differences between, say, painting and photography (i.e. they don’t require different theories of action), there are no distinctive philosophical difficulties raised by photographic action, we’re good. And insofar as his idea is, second, that making a painting is nonetheless different from making a photograph (they don’t require different theories but they do require different actions), all that’s needed for us to still be good is the agreement that these different actions may plausibly produce (and have in fact produced) both different kinds of difficulties and different kinds of projects aimed at overcoming them. Michael Fried reads Thomas Demand’s practice of building models and then photographing them as a way of producing pictures from which everything but the intention of the artist has been excluded. For example, every blade of grass in Lawn was first constructed in Demand’s studio, thus turning what would otherwise be a kind of epitome of the unintentional (in an ordinary photograph of a lawn not only would the relations between the blades of grass be unintended by the photographer, they would also of course be unintended by the grass itself) into just the opposite—what Fried calls an image of “sheer artistic intention.”3 Neither the problem (how do I make this picture assert its intentionality?) nor the solution (build a model) would make much sense for a painted Lawn. Rather, since everything depicted in such a painting would already bear the mark of the painter’s intentionality (each blade would only and necessarily look the way it did because the painter had painted it that way), the problem of making something that bore the mark of being made that way would never even arise.
Assuming Lopes and I agree on all this, I still don’t quite see why he thinks of himself as having a philosophy of photography instead of just having his version of an Anscombian theory of action, but I also don’t think it much matters. We both get to talk about why and how the questions of meaning, intention and action might matter differently in photography than in painting (or writing or other arts) but we don’t have to think that photography requires a special theory of agency or of meaning or that it shows the irrelevance of agency or meaning. And we don’t have to deny that it might for some photographers (in my account, Winogrand) become a medium in which to think about some problems in the theory of action. Although here, perhaps, the agreement begins to break down, as it surely does when, in deploying his theoretical distinction between the two different kinds of meaning, Lopes seems to me not to feel the force of the example from Anscombe that is central to my discussion—the ironic hug.
For Lopes hugging someone to express contempt instead of affection shows how “the semantic content of a vehicle of communication is not always what the communicator uses the vehicle to communicate.” The semantic content of the hug—“a big hug signals affection”—is “determined by social conventions, not the hugger’s intentions.” So when you hug someone, the convention says you’re signaling affection. If, however, what you’re actually trying to signal is not affection but contempt (“you silly little twit”), that’s your intention talking. The point for him is that “some meaning” is not “inseparable from intention”; you can “prize” it “apart from intention.”
But how does the distinction between the social convention and the intention “prize some meaning apart from intention?” The ironic hug doesn’t have two meanings, one conventional and one intentional. If we’re trying, in Anscombe’s words, “to give a correct account of the man’s action,” the correct account is just that it’s contemptuous. And, of course, that would be equally (albeit differently) true of a non-ironic hug. It’s not, in other words, as if the non-ironic hug is sincere by convention; it’s sincere because it’s meant sincerely. Which is just to say that the social convention only functions as a vehicle of communication if someone is using it as a vehicle and the act of using the vehicle only means what the user means by it.
My point here is partially that, as long as you’re thinking of meaning as an act, you can’t prize any of it apart from intention. (There’s a reason why Anscombe put “Signing, signaling” on a list of “happening[s]” that “can only be voluntary or intentional” .) Hugging someone contemptuously and hugging him affectionately are two different acts; the ascription of intention is internal to the description of them both and is their interpretation.
But part of the interest of the example—what made it, in my view, particularly relevant to Winogrand and what, I want to suggest now, makes it relevant to how we understand the history of art in the second half of the 20th century more generally—is the fact that the difference between the two hugs may be thought of, Anscombe says, as “a purely interior matter.” What she means by this is that there is no physical difference between them and hence “no outward sign” that one is affectionate and one isn’t. Which might seem to suggest a picture of the act as a physical event that we interpret by reference to some mental state external to the act itself but internal to (in the mind of) the agent. And although this is not in fact what Anscombe means to suggest (her view is just the opposite), in its account of the intention as something mental and of the act as something physical, it does give us a picture of intention (what Stanley Cavell, usefully influenced by Anscombe called a “bad picture”4) that has been foundational for a lot of literary and art critical work in the last half century.
It’s this picture, that animates Wimsatt’s and Beardsley’s “The Intentional Fallacy,” and that Cavell criticized in the new critics, arguing that it was only because they saw the intention as “outside the work” that they could imagine it was irrelevant. It’s this picture (more generally) of any act as, in Arthur Danto’s words, a “movement of the body plus x” and (more particularly) of the “work of art” as “a material object plus y” that produced Danto’s enthusiasm for Warhol’s Brillo Boxes.5 His idea was that because you cannot see the difference between the physical object that is an ordinary Brillo box and the physical object that is a piece by Warhol (they’re “perceptually indiscernible” (61), what the Brillo Boxes show is that you need to look for something outside the object (“y”) to understand which is which. Thus, although Danto was no fan of New Critical anti-intentionalism,6 he nonetheless shared their sense of the relation between the physical object and the intention. The difference was that he thought adding something like the intention was essential rather than irrelevant or impossible or harmful.
And, as I began to suggest in “I Do What Happens,” it’s this subdivision of the work into two separable components—the object and the intention that gives the object its meaning—that comes to function for Winogrand as a problem. Obviously, this apparent separability didn’t count for all artists as a problem (it wasn’t for Warhol7) and, equally obviously, it wasn’t just an issue for photography (again, Warhol). But it’s easy to see how the photograph and especially the snapshot (where is the difference between the physical object that’s a picture of your friends and the one that’s a work of art?) might come to seem a powerful instantiation of the difficulty you could have in making your act count as what you meant it to be. Without, that is, pointing to something outside it—like what was going on inside your head. In this sense, part of the interest of the photograph is that it runs the risk of reducing the artist’s intention to what Anscombe called “a performance in the mind,” what you’re thinking while you take the picture.
The Anscombian response to this worry is that it’s a mistake to break the act down into component parts, a mistake to think of the intention as something that’s outside of the physical act, either as its cause or as a mental state existing either prior to or alongside it. That’s why she says your hug isn’t given its meaning by the words “you silly little twit” “occur[ring]” to you while you embrace your old acquaintance, they have to be “seriously meant.”8 And you could mean the hug to be ironic even if you were thinking only affectionate thoughts at the time you administered it, or thinking nothing at all. The correct answer to the question, “why did you hug him?” would still be, to show my contempt.
But, of course, you might not succeed in showing your contempt. You might hug him ironically but he might fail to see the irony. Obviously all meaningful acts are subject to misinterpretation but part of the point of juxtaposing the ironic hug, the Brillo box and the snapshot has been to suggest that because each of them looks like something it isn’t as much as it looks like something it is, they all raise the question of what they mean in terms which seem to suggest that the answer to that question cannot be found in them. For Danto’s Warhol,9 this counts as the discovery that what makes an object a work of art need not be visible in the object itself, that the question of what it means or even of whether it means demands an answer that you can’t see in it. Because the Brillo Boxes don’t look like art, they lay bare the condition of what it is to be art. They’re both both an instantiation and an allegory of Danto’s analogy between the “bodily act” and the “mere thing” on one side and what he called the “basic action” and the work of art on the other, an allegory both of the essential importance of intention to the work of and of what he conceived as the relative autonomy of intention and work.
A snapshot is obviously different from a Brillo box but not necessarily because it looks more like art. Indeed, in 1965, in Photography, A Middle-brow Art, Pierre Bourdieu remarked that “The realization of the artistic intention is particularly difficult in photography, probably because, fundamentally, it is only with difficulty that photographic practice can escape the [domestic, journalistic, etc.] functions to which it owes its existence” (71). For our purposes and for Winogrand’s practice, the difficulty is doubled—first for the Bourdieusian reason that it’s hard to make the snapshot look like it’s intended as art and, second, because, given how much the camera does on its own and how little control the street photographer has of his subjects, it can be hard to make it count as meaningful, hard to make visible the difference between it and a “mere thing. ”And this difficulty, as I’ve already suggested, can be understood to instantiate the separation of the intention and the object that for Danto makes it necessary to look beyond the object.10 Thus, if we think of Warhol as concerned to identify whatever it was about an object that made it a work of art (what Danto in 1981 called The Transfiguration of the Commonplace), we can think of Winogrand as at least as concerned to describe what he (also in 1981) called the “transformation” of “banality” that enabled the “artist” to turn the photograph into something other than a source of “information.”11
But if we look at some of the work Winogrand was producing in 1964 (the year of the Brillo Boxes and “The Art World”), we can see that even though the ambition to make art out of a “mere thing” is something Danto’s Warhol and Winogrand have in common, their ways of going about it were very different. The point of the Brillo Boxes, as Danto understood it, was that they were “indiscernible” from the real thing, indeed, from Danto’s standpoint, there was no reason not to “use the real thing” since the difference between it and the work of art was not in the object itself but in the mind of the artist. What makes the thing art is the artist’s claim that it’s art, itself an invisible gesture brilliantly captured by one of Guy Davenport’s illustrations for Hugh Kenner’s The Counterfeiters: “Mr. Andy Warhol Fetches a Work of Art Through a Metaphysical Barrier.”12 Making that claim visible is a function of what happens after, of the object being exhibited in galleries, bought and sold as art, shown in museums.
Winogrand’s photos were also destined for museums; in fact, the one above was shown at MoMA in 1967. But the photograph’s claim to be art is meant to be visible in it. One way to put this might be to say that its passage through the metaphysical barrier is meant to be visible in it. Another way might be to say that the act of making the snapshot be art cannot quite be understood on the model of Danto’s metaphysics because it’s susceptible to failure in a way that metaphysical fetching is not.
What I mean by this is just that no one could say of the Brillo Boxes that they fail because they don’t look like art. But one could say this of Winogrand’s snapshots and, indeed, years later (in his 1995 book on Robert Mapplethorpe), Danto himself did, describing them (specifically the pictures of women) as providing the information Winogrand explicitly sought to transcend; they “document” “how women dressed in the Sixties; how they wore their hair and made up their faces,” and they also “document” Winogrand’s “yearning” for the women and his “sexual desperation.”13 Mapplethorpe, by contrast, made art. Indeed, Danto describes Mapplethorpe in terms that replicate his interest in Warhol and the entire argument of The Transfiguration of the Commonplace, only here it’s not the “mere object” that poses the challenge and supplies the opportunity for art but “mere pornography”: in a sentence that uses a phrase from Mapplethorpe himself and that gives the book its title, he describes Mapplethorpe as “play[ing] with the edge that separates art and mere pornography.” His ambition, Danto writes: “as to create work that really was pornographic by the criteria of sexual excitement, and really was art. And it incorporated the ‘edge.’”
For Danto, it’s Winogrand—producing documents of his own desire—who ends up in the role of the mere pornographer. The “images are extremely aggressive towards the women for whom [he] hungers” and the camera is tilted “in such a way that it is impossible to suppress the thought that [it] offered [him] a way of looking down bosoms” (27). In fact, taking seriously Mapplethorpe’s (absolutely on target) characterization of his own work as “the opposite of Garry Winogrand’s” (25), Danto’s descriptions of all the things Mapplethorpe does are inversely mirrored by his descriptions of all the things Winogrand does. Mapplethorpe is described as dedicated to a “formalist approach” that requires a kind of “control”—“the people he photographed had to be doing what he wanted them to.” So he realizes, first, that “sex was not really meant to be photographed” (because people having sex mainly do what they want to do) and, second, he understands in order to be able to tell his subjects what to do, he had to have their “consent” and in order to be able to win their consent he needed to win their trust. Danto makes this point many times: “In order to be in control, Mapplethorpe required his subject’s agreement, their knowledge and, as emphasized, their trust” (79). And he goes on to say, this “moral relationship between subject and artists was a condition for the artistic form the images took.” Thus, for example, “As if in formal acknowledgement of the agreement between subject and artists, the images in Some Women are extremely well composed” (31) and it’s at least in part the fact that his “pornography” is “beautifully composed” (90) that brings it “up into the plane of art” (91). Control, consent, composition make the simultaneity of pornography and art possible, make possible work that “incorporates the ‘edge.’”
By contrast, Winogrand’s photos of women “seem uncomposed” (25) just as the women themselves seem “uncomposed.” The women seem uncomposed because their pictures are taken without their consent (“invading” their lives, “if only momentarily,” Winogrand “violated” their “right to privacy” ); the pictures seem uncomposed because in his “aesthetic,” “accident was assigned a role altogether at odds with the firm imposition of artistic will central to Mapplethorpe’s composition” ). Hence, even though they are “fully clothed,” Winogrand’s women are “seen as female flesh” (27)—the “victims” of his “obscene attention” (29). Without consent, control and composition, failing to incorporate the edge that will “transcend” the position of the pornographer, Winogrand fails to bring his pictures “up into the plane of art.”
Winogrand, however, had a different account of the edge, or, perhaps more accurately a similar account of the edge but a different practice for incorporating it. The similarity is that for him, as for Danto’s Mapplethorpe (and, of course, Danto’s Warhol), it involved the transformation of something that wasn’t art into art, “how the fact of putting four edges around a collection of information or facts transforms it. A photograph is not what was photographed, it’s something else.” For the Brillo Boxes, that transformation is invisible (purely conceptual); for Mapplethorpe, it’s the “formalism” that keeps the work from being pornography: “Without the formalism,” Danto says, “the work goes over the edge” (79). The difference is that for Winogrand, the edge is both more literal and more conceptual. For example, the 1964 picture above (Los Angeles 1964) shows one of the “anonymous women,” involved in one of the “private” activities (i.e. not posing for the camera) that Danto rightly says are characteristic of Winogrand’s photographs, and the fact that it’s shot through a car window makes the identification of the photographer with the “voyeur” that Danto insists on even more vivid. But the window plays a role in relation to the four edges of the photograph that’s different from the one it plays in relation to the photograph’s subject. On the one hand, it frames our view of the subject; on the other hand, because it’s tilted at an angle to the literal edges of the photograph, it mobilizes them as an element in (not just the site of) what Danto might have called the picture’s composition. That is, the four edges that any photograph automatically puts around the “information” have been made to function as a structural element that’s internal to the picture, that turns information into form. Because the internal frame of the window creates a space outside it as well as inside it, in front of it as well as behind it, the elements that don’t belong to the view (the inside of the car, the bits of the garden and the house and even of someone else gardening) are asserted instead as belonging to the picture. The picture is disarticulated from the view.
The view, in other words, is of the woman but the photograph isn’t. It includes her; which is to say, it includes the view without being reducible to the view, and, precisely because of the non-identity of the view and the literal frame of the picture, it gives the literal frame conceptual weight. It’s not just where the picture ends; it’s how the picture establishes its difference from the (voyeuristic) view of the woman, from the information the photograph conveys—about the world and even about the desires of the photographer for the things of the world. In this respect, although what we might (following Kenner and Davenport) call Winogrand’s metaphysical ambitions have a lot in common with Warhol’s, his formal ambitions have a lot more in common with someone like Kenneth Noland’s. Or are, at least, more responsive to what Michael Fried (in 1966) described as the new (since “shortly before 1960”) discovery of the importance of the relation between depicted and literal shape, and more responsive especially to the sense that depicted shape could be called upon “to establish the authority of the shape of the support.”14 Obviously, these are terms that derive from painting, and from abstract painting to boot. But the desire to make the straight up rectangle (i.e. the physical fact) of the print itself signify—first by setting it against the tilted rectangle of the viewfinder and hence of the four edges inside the picture and, second, by making it function not just to delimit but to incorporate what exceeds the internal frame—is hard to explain except as the effort not exactly to declare that snapshots can be art but to make one that is.
This is, if one likes, a kind of formalism (the Fried essay quoted above was called “Shape as Form”) and, of course, the women in Winogrand are as much the literal material out of which he made his photographs as is the paint in Noland. But Danto’s worry about Winogrand’s “obscene attention” to his subjects finds no color field application. The relevant parallel and alternative would be Mapplethorpe, whose own “formalism” (as we have seen) Danto praises as the reason for “the moral relationship” he established “between subject and artist.” Because, even when he’s taking pictures of a model’s breasts, Mapplethorpe treats her “as an end and not as a means,” there’s “not the slightest sense of obscenity.” Rather, this treatment is the essential “condition for the artistic form [his] images took” (79).
And here again both Mapplethorpe’s morality and his idea of art truly are “the opposite of Garry Winogrand’s.” For Winogrand’s ambition was always and only to use those women as means not ends, and what worried him about his pictures of “attractive” women was that they didn’t always fulfill this ambition. Indeed, he thought Women Are Beautiful was “not as good” as his other books because the question the photos raised—“when the woman is attractive, is it an interesting picture, or is it the woman?”—didn’t always get answered in the right way. Too often, it was the woman. But, whether in success or failure, the relevant thing is that making an “interesting” picture required not so much a “formalist approach” to a subject as it did making something that had form.
It’s in this respect that Winogrand may be said to have had a theory of intention (or of action) that was different from Danto’s Warhol and his Mapplethorpe. With respect to Mapplethorpe, we can see it in Danto’s enthusiasm for the idea, conveyed to him by Dimitri Levas, that (where Winogrand famously claimed not to know what the photograph would look like until after he’d taken it), “all the shots” in Mapplethorpe’s “contact sheets looked pretty much like the ones selected for the final image, as if the photographer knew precisely what the outcome would be. He did not count on the fortuitous fall of a garment, an unanticipated ripple of muscle” (31). The “firm imposition of artistic will” that makes “mere pornography” into art valorizes what’s inside Mapplethorpe’s head in the same way that the transformation of a mere object into a work of art valorizes what’s inside Warhol’s. Both are resistant to the accidents that infect Winogrand.
But if Mapplethorpe is Winogrand’s opposite as an artist, Warhol is his opposite as a theorist. In Mapplethorpe what’s in your head is prior to the act and making the picture is the effort to make what’s inside your head visible. In Warhol—in the very idea of the indiscernible—the whole point is what’s not visible. That’s why Danto insists on the way in which “the relationship between an artwork and a thing just like it is…analogous to the difference between a basic action and a bodily movement just like it, to all outward appearances” (48). In effect, the Brillo box allegorizes what it takes to turn a bodily movement (standardly,15 the raising of your arm) into an act (instead of, say, “a spasm”)—the appeal to an intention you can’t see in it because it isn’t there; it’s in your head. For Danto, in other words, the transfiguration that takes place when the object is made into art by being offered as or declared to be art emblemizes the transfiguration that makes a bodily movement into an act.
But the point in Anscombe is that neither your intention nor anything else makes a bodily movement into an act. Beginning with her idea that “Intentional actions…are the ones to which the question ‘Why’ is given application” (24), she means not that we have to add what’s in the person’s mind to the act but just the opposite: “we do not add anything to the action…by describing it as intentional. To call it intentional is to assign it to the class of intentional acts and so to indicate that we should consider the question ‘why’ relevant” (28) in a way that it wouldn’t be if, when asked why you were raising your arm, you responded, “I was not aware I was doing that.” And confronted with the “mere thing” turned into a work of art, we can see her point, albeit from a slightly different angle. “The difference between a basic action and a mere bodily movement,” we have seen Danto say, “is paralleled in many ways by the difference between an artwork and a mere thing.” (5) So how does a mere thing get to be an artwork? The artist “declares” (3) it to be one: “Duchamp declared a snowshovel to be one, and it was one.”16 But a bodily movement doesn’t become an action when the agent declares it to be one. My raising my arm to signify, say, my desire to speak, is not an act of signification because I declare it to be one or because of anything I add to the bodily movement—it’s what the bodily movement is from the start.
And once, on the model of the analogy between the mere bodily movement and the mere thing, you understand making the art as adding the intentionality, you’re confronted with the further question of what exactly that adding consists in. In other words, the double reification—of the intention on one hand and the thing on the other—makes the question of what the artist does a puzzling one. That’s really the point of Davenport’s brilliant illustration—what kind of act is fetching a mere thing through a metaphysical barrier? It’s not that you can’t come up with something—buying the thing, taking it to the gallery, putting a price on it, etc. And it’s not that the thing isn’t—because of what you’ve done—a work of art. It’s that its interest as a work of art is the way it allegorizes the idea of the autonomy of the intention as an internal state, an Anscombian “bombination in a vacuum.”
Of course, it hasn’t been a standard requirement for artists (any more than for anyone else) that they have a good or even a bad theory of action. People do what they do. But the minute the meaning of the work is imbricated in the question of what the work is, the question of how the work was made (what the artist did) ceases to be a merely causal one and becomes instead an element of that meaning. The work that produces a theory of itself (say, that its claim to being an artwork is invisible) produces also a theory of action (the intention that makes it a work of art is in the artist’s head). More generally we might say that the resemblance between the work of art and the thing comes to matter precisely because it provides an opportunity to think about what an action is, and thus photography becomes increasingly important not despite but because it raises questions about action and intention. If, in other words, the worry about whether you have enough control is forestalled by the Brillo Boxes (since what gives them their meaning is in Warhol’s head, you don’t need any control), that same worry is addressed in the same way by the transfigured pornography: the actual picture is an epiphenomenon of what was already in Mapplethorpe’s head.
Winogrand, who would in the last years of his life take hundreds of thousands of pictures that no one ever saw, was hardly immune to worrying about control. If we think that he no longer needed to see the film developed in order to know what he’d done, his previsualization seems even more complete than Mapplethorpe’s, and if we think that what he was doing no longer involved developing the film, then the act of actually taking the picture begins to look almost as mysterious as the act of fetching the work of art through its metaphysical barrier. And, of course, insofar as the alternative to these two different ways of imagining the primacy of intention might seem to be its irrelevance, or its consignment to a merely causal role in the production of the work of art, what you would get is the photograph as thing, which is to say as information—about the world, or about the photographer or even about the beholder. But Winogrand’s commitment to form instead of information on the one side and to form instead of a “formalist approach” on the other required him to refuse this double reduction. It’s not exactly that he had a different theory of action than Danto; it’s that his idea of art entailed a different theory of action, and his idea of a successful work of art (an “interesting picture”) required him to make that theory visible.
All acts are intentional under some description (and under some other description not intentional); this is uncontroversial. The interpretation of the meaning of every work of art is an account of what the artist intended. This is controversial (thus, for example, even Lopes and I, who begin by declaring agreement with each other turn out really to disagree), but if it’s true it was always true, as true in the 11th century as in the 21st, and true even of works created by artists who don’t believe it or who never thought about it.
But many artists have thought about it; in fact, it might almost be a definition of modernism (and, depending on where their thoughts led them, postmodernism too) to say that insofar as thinking about what kind of object a work of art is involves thinking about the relation between the conditions of its production and its meaning, all modernists and postmodernists have thought about it. It’s for just this reason, as Michael Fried and I have both argued, that photography began to assume a new importance in the second half of the 20th century. With respect to the question of intentionality, for example, it’s because so many things about the photograph could seem to be unintended (in the ways described in 1 above) that many artists and critics could turn to it as a way of refusing intentionality or of demonstrating the ways in which the meaning of the work necessarily exceeded the artist’s intentions. Indeed, once you understand the artist’s intention as the mental state that gets added to the physical fact of the work (the thought you have while taking your picture), you can’t help but start to wonder why it should make any privileged difference to the work’s meaning. Thus a certain defense of intentionalism and a certain critique of it are significantly compatible, and precisely because making a photograph can be thought simultaneously to internalize the intention and externalize the work, photography becomes a central site for trying to do both.
And a central site for trying to do neither. Thus Winogrand imagines the difference between a photography in which the photographer’s intention plays a merely causal role and one in which the intention both is and is visible in the picture itself by distinguishing between the pictures he himself can make (as a “craftsman”) and the pictures he’s trying to make instead. The “good craftsman” has “a particular intention”: “let’s say, I want a photograph that’s going to push a certain button in an audience, to make them laugh or love, feel warm or hate or what—I know how to do this. It’s the easiest thing in the world to do that, to make successful photographs. It’s a bore.”17 What craft means here is obviously not that there is no intention but that the beholder doesn’t have to understand the intention in order to experience the effect. That’s the force of “press a button”—if you ask the craftsman the Anscombian question of why he took the picture from that angle, he has an answer—to make you “feel warm.” So taking the picture was an act. But you don’t need to ask the question or know the answer to experience the effect—from the standpoint of the beholder, the fact that it is an act is irrelevant. In other words, the craftsman seeks to produce an effect but he doesn’t seek to produce that effect by having the beholder recognize his intention to produce that effect. The beholder doesn’t need to interpret his act.
By contrast, if we extend our reading of Los Angeles, 1964, we are immediately confronted with a question of description that isn’t easy and is completely, in the relevant sense, interpretive. For we might say that the picture is structured not just by the inner and outer frame described above but by a third framing device: first, the four edges of the photograph itself, second, the car window and third, a series of vertical lines—on the right, the tree and then another tree behind it and aligned with it and on the left, the fence and the house next door—which produce a recessional effect that seems to locate the frames in deep space. With the result that, if we look at it this way, the actual (literal) edges of the picture begin to seem nearer to us—as if they were inside the picture (and no longer literal). And thus what I described above as the desire for the view (what it’s a photo of) to be subsumed by the picture is matched by a desire for the physical object that is the photo to be subsumed by the picture; the picture is disarticulated not only from the view but also from the photograph.
But should we look at it this way? This is a question about how Winogrand means it to be seen, which is to say, a question about what the photograph is about. Whereas the craftsman’s photographs, as Winogrand presents them, aren’t about anything. In other words, this photograph can only have the right effect on us if we understand what Winogrand was trying to do whereas we can be made to feel warm (the craftsman’s desired effect) without being the slightest bit interested in whether he meant us to. The two photographs are equally intended, but only Los Angeles asks to be understood. And only Los Angeles, with its (if I am right) elaborate enactment of its ambition to be understood, is also about what it means for it to be intended, about its own demand to be understood. Its claim to be made by an artist instead of a really good craftsman is its effort to make us see that demand.
Of course, the Brillo Boxes also demand to be understood (they are about something), but that demand is not visible in them. It’s made instead by what Warhol does with them—by his taking them to the gallery to be displayed and by the gallery’s offering them for sale. It’s as if, in other words, Danto’s Warhol recognized that declaring a mere object to be a work of art required something more than simply thinking about it as if it were a work of art or saying it was a work of art; you had also to treat it like a work of art. (You had to seriously mean it.) And the way to treat it like a work of art was to offer it for display and for sale. Which, in the case of the Brillo Boxes, as Blake Stimson points out, involved taking the kind of commodity you could buy in a supermarket and elevating it to “a higher exchange value” by placing it “in the boutique context of an art gallery.”18 Stimson makes this point following and revising Danto; where Danto thought the Brillo Boxes were about art keeping the work “from collapsing into the real object which it is,”19 Stimson says just the opposite: they don’t keep the work from collapsing into the object, they collapse it, only the object is a commodity, and the collapse is of the “once interiorized category of art into the exterior category of exchange.”
For our purposes, however, we don’t really need to choose sides. For insofar as the way to turn the mere object into a work of art is to treat it like a commodity and insofar as the commodity, as Nick Brown argues, is crucially distinguished from the work of art precisely by what it has in common with the mere object (its meaning is not determined by its maker’s intentions, which is to say, like the mere object, it has no meaning), then the Warhol-style transformation of the thing into the work (call it postmodernism) is really the transformation of a thing into another thing. It’s precisely in this context—confronted by what Brown has called the possibility of “its real subsumption under capital”—that the work of art finds it necessary to assert its status as a work of art instead of or rather as well as a commodity. For we may well take the point that since the beginning of modernism (whenever that might have been), works of art have necessarily been commodities but only when it begins to seem possible (the beginning of postmodernism20) that they can be nothing but commodities does it become desirable to assert the ways in which they’re not just commodities. By the same token, works of art have always meant what they were intended to mean but it’s only when it began to seem possible for their meaning to be irrelevant or to be subsumed by their use value that it became desirable for them to be about the fact that they are intended, for their meaning to include the fact that they are meaningful. It’s for this reason that, called upon to produce its theory of itself, Winogrand’s photo produces an account of what it was meant to do and thus a theory of action.