Skip to content Skip to sidebar Skip to footer

“I Do What Happens”: Anscombe and Winogrand

The title of this paper, “I do what happens,” is what Elizabeth Anscombe calls a “formula”1 in her book Intention, widely regarded as the most important modern contribution to the philosophy of action and recently invoked as an important text for what is sometimes called the philosophy of photography. The purpose of the formula is to offer an alternative to those philosophers who, she says, have thought that “if we wish to understand what intention is, we must be investigating something whose existence is purely in the sphere of the mind.” “Even though intention issues in actions,” for these philosophers, she says, “what physically takes place, i.e. what a man actually does, is the very last thing we need consider in our enquiry. Whereas I wish to say that it is the first” (9). Hence the formula—“I do what happens,” the idea that to understand intention is to understand intentional action, and that to understand intentional action is to understand what actually, physically takes place.

But, right from the start, I do what happens is a complicated—or complicating—formula, since Anscombe announces it as something she used to say but that everyone who heard found “extremely paradoxical and obscure” (53). And she herself explains this response by acknowledging that the idea that I do what happens conflicts with another idea to which she’s equally committed, the idea that an intentional action (like, say, moving my arm) is something that I know I’m doing without observation—I don’t have to look at my arm moving to know that I’m moving it. The reason this is important is because intentional actions, she argues, are performed for reasons (they are always open to the question, why?). But if you only knew your arm was raised because you happened to see it sticking up there, you could not be said to have (or understand yourself to have) raised it for a reason. If someone asked you why you were raising your arm, and you replied, “I was not aware I was doing that,” that would be “a rejection of the question, ‘Why?’” (14). To do something intentionally, then, is to know you’re doing it without having to look to see if it’s being done.

So the conflict is that what you do (moving your arm), you do without knowing by observation but what happens (your arm moves) can only be known by observation. And although Anscombe is completely committed to the importance of what happens (if you think what happens doesn’t matter and intention is “a performance in the mind” [49] all you’ve got, she says, is “a bombination in a vaccum” [52]), she is at least equally committed to the irrelevance of your being able to see what happens because the whole logic of intentional action—the idea that it is structured by answers to the question why, by reasons rather than causes—depends upon your being able to know what you’re doing without observation, a point that Anscombe immediately reinforces by producing what she calls a “very clear and interesting case” (53).

What do I say I am doing, she asks, if “I shut my eyes and write something?” “I can say what I am writing”; that is, I can say what words I’m writing even though I can’t see those words, and the words that I say I am writing “will almost always in fact appear on the paper.” So I can say, for example, I’m writing “I do what happens” and the words “I do what happens” are appearing—what I say I’m doing is what’s happening. But, she acknowledges, it probably won’t keep happening for long—the words, without my being able to see them will start to become illegible, I might go on writing after my pen ran out of ink, etc. But still, she wants to say, “the essential thing” (writing “I do what happens”) has nothing to do with observation—my eyes are useful in making sure that what I write is written legibly, but I’m writing what I’m writing (and I know what I’m doing) without them.

This is a difficult moment in Anscombe’s text, and although it comes just before an announcement that she is about to “make an end of raising difficulties” and to begin providing solutions, the difficulty of those solutions is marked by the return of the example close to the end of Intention when she imagines herself, eyes shut, writing “I am a fool” on the blackboard while something “goes wrong with the chalk” and no words appear (82). How can I be doing what happens if what I believe I’m doing isn’t happening? But it’s also a useful moment, both for thinking about the idea of what some philosophers have called the philosophy of photography and, as I’ll suggest later, about what the title of our conference (albeit in inverted form) calls philosophy and photography. It’s useful with respect to the philosophy of photography because, in the work especially of Diarmuid Costello and Dominic McIver Lopes, Anscombe’s thought coupled with Donald Davidson’s, has been invoked as addressing (and assuaging) worries about “photographic agency,” worries in particular that in the act of making a photograph, photographic agency is “compromise[d]” because the kind of control exercised by, say a painter, cannot be exercised in the same way by the photographer.2 The idea here is that too much of what the photographer does is necessarily unintentional and therefore in some sense not done by him—an idea that’s been around since the famous Kodak ad (you press the button, we do the rest—in effect, you do it we’ll make it happen). But, Costello and Lopes think, this worry can be pretty much laid to rest by a proper understanding of Anscombe.

Why? Because in Anscombe’s account of an intentional act, the intentionality does not exhaust the description of the act. There can be and always are lots of other descriptions. For instance, “I’m writing” might equally be “I’m displaying my fine penmanship” or “using up my last piece of blank paper,” both of which might be true but neither of which I might mean to be doing or even know I am doing. We don’t think my act of writing is less my act because there are many descriptions under which it is unintended by me; why then should we be so concerned to provide what McIver Lopes calls a “strong linkage of photographic agency to intentionality?”3 About this photograph, for example (Henri Cartier-Bresson’s of the future Pope Pius XII, Cardinal Pacelli), we can say, according to McIver Lopes, that Cartier-Bresson “intended to capture the cardinal enthronged, he intended the bracketing techniques of photojournalism…he intended to trip the shutter” (862). But he did not intend, Lopes says, “that the man in the upper left, with the unfortunate mustache, be shown as looking away from the cardinal; his taking the picture was not intentional under that description.”

Henri Cartier-Bresson, Cardinal Pacelli, Montmartre, Paris, 1938

Which no more produces some kind of distinctive problem about photographic agency than, say, the unintended look of a poet’s penmanship or the fact that she’s about to run out of paper compromises her agency as a poet. The point is that all acts are unintentional under some description (right now, I am interesting some people and boring others; the boring is unintentional but I’m doing it [sorry]). So it’s a mistake to try to tie photography too closely to the photographer’s inability to control everything, and, more generally, it’s a mistake to try to tie any art too closely to the artist’s intentions. After all, as Lopes puts it, “Many artistically relevant features of paintings are unintended—from patterns of craqueleure to the retrospective impact of future works” (863). So why should the standards for photography be higher?

As attractive as this deflationism is, however, none of these examples really supports it. For one thing, it’s not at all obvious that we should feel more confident that the man looking down and away from the cardinal is less intended than, say, the woman looking up and at him. Why shouldn’t we think instead that the kind of rhyme between the two (actually three) bald men looking down was part of what made the photograph work for Cartier-Bresson? And to me it seems crucial that not everyone is looking at Pacelli. But my point here is not so much that Lopes is wrong about this; it is instead that he might be right. In a painting, a poem or a sculpture, the question of whether the direction of the mustached man’s gaze was intended could not arise. If the painter didn’t want him looking down, she wouldn’t have made him look down. It’s thus a distinctive fact about this photograph that we can ask not only what the artist meant by positioning one of its central figures in this way but whether he meant anything at all (i.e. that we can plausibly doubt and the photographer could plausibly deny there was any answer to the question, “why do you have him looking down?”). And it’s a distinctive fact about the snapshot more generally that this question about whether he meant anything by it can plausibly arise not because it’s raised about an element of the depiction that’s marginal or not fully articulated but even though it’s absolutely central and powerfully articulated.

Indeed (to turn the point around a little), the ability of the photographer to give us the mustached man without meaning to may be understood as one of the ways in which photography makes it possible to think about what meaning—as a kind of intentional action—is. Imagine, Anscombe says (a few pages before she introduces “I do what happens”), imagine a “contemptuous thought” entering “a man’s mind” so that he means “his polite and affectionate behavior to someone on a particular occasion only ironically, without there being any outward sign of this” (48). What he does is, say, give the someone an affectionate hug while “the thought in his mind” is “‘you silly little twit!’” (49). So, he does what happens, but what’s happened? What Anscombe calls the “forms of affection” (the hug) have been observed but has there been an expression of affection or—without anyone (anyone other than the speaker anyway) realizing it—has there been an insult?

It depends, Anscombe thinks, on the status of the “thought” (“you silly little twit”) in the speaker’s mind. It’s “not enough,” she says, that “these words should occur to him.” It’s not enough, in other words, that he be thinking them, “He has to mean them” (49). That is, there’s a difference between thinking of something while you’re performing some physical act (hugging, writing, speaking) and meaning something by performing the same physical act. The difference need not be visible—the whole point of this example is that it isn’t—but the fact that you can’t see it doesn’t mean it isn’t there. And you can’t have what she calls “a correct account of the man’s action” without knowing not exactly what he was thinking but rather what he was meaning.

The point of this example for Anscombe is that “intention is never a performance in the mind” (49) since the mere fact that the words “you silly little twit” “occur” in your mind doesn’t mean your action was ironic. Meaning something involves an “outward act”; your meaning has to be expressed in that act. But the difference between thinking something (which the outward act doesn’t express) and meaning something (which it does) isn’t just in your mind but nonetheless is in your mind; you can’t know what was expressed—you can’t even know if anything was expressed—without reference to something irreducibly although not exclusively internal—what was meant.

For the philosophy of photography, what’s important here is just the centrality of intention. To give “a correct account” of Cartier-Bresson’s picture of Cardinal Pacelli is to give a correct account of the act of making it, which is to say, a correct account of what Cartier-Bresson meant. In this sense, then, an Anscombe-style theory of action makes the question of what was intended more, rather than as McIver Lopes suggests, less central, and, indeed, makes meaning inseparable from intention. In which case the effort to weaken the connection between the photographic act and the question of intention is doomed, although not for reasons that have anything to do with photography. In fact, if we think back to Lopes’s remark about the “artistic relevance” of “craqueleure,” we can see that it’s not at all obvious why we should understand the unintended things that can happen to a work—the paint cracks, the photograph fades, the light bulb in a Dan Flavin goes out—as having any necessary relation to its meaning. Which makes intention as crucial to any work of art as it is to this photograph, and which, conversely, is why the relative opacity of the photograph with respect to the photographer’s intention can be understood after all to raise a distinctive question about photographic agency.

But the real interest here is not in getting the right general account of meaning and intention. It’s in recognizing that because what the photograph looks like (the mustached man’s downward gaze) can so plausibly be disconnected from the question of what was going on in the photographer’s head (was he meaning something by it? was he just thinking, “you silly twit?” or was he thinking nothing at all?), photography (even or especially in the practice of photographers who understand themselves as just pressing the button) can put as much pressure as Anscombe does on the relation between what I do and what happens. What Garry Winogrand (maybe the exemplary instance of the press the button photographer) did, for example, was go out with his camera, begin taking what might come to dozens of pictures, working very fast but each time raising his camera to his eye and looking through its viewfinder because if you didn’t, he said, you gave up too much “control,”4 then put the undeveloped film away sometimes for a year or two, then make contact sheets, select which photos to print and get someone to print them. So obviously our interest in the intentional structure of this action—Winogrand making a picture—can’t be reducible to an interest in each of the individual physical acts or to what was going on inside his head at any particular moment. The point is rather that they belong to a single intentional structure, each element answerable to the question why, each element explicable in terms of doing what happens—i.e. making the picture.

But, starting with any version of Winogrand’s famous mantra, “I photograph to find out what something will look like photographed,” we can see a certain tension in this narrative. For this formulation—insisting that what he will see in the photograph is not what he saw when making the photograph, that he cannot know what he is doing until he sees what he has done—suggests the necessity of precisely that knowledge by observation that in Anscombe calls the very idea of an intentional act into question. Winogrand’s photographer is almost the opposite of Anscombe’s writer; her eyes are shut but she nonetheless knows what she’s doing; his eyes are open, but he doesn’t, since it’s not how the world looks through his viewfinder but how it looks in the picture that determines what happens. That’s what he meant by insisting that it was something you couldn’t see when you were taking the picture—the frame (the “four edges”5 that you put around the world)—that made a good photo. And in a picture like this one, organized around the alignment of the arm stretched out to give the beggar some change with the horizontal edges of the picture itself, we see what he meant.

Winogrand, Street Beggar Reaching out to Receive a Donation, 1968

The photograph distinguishes the photographer’s or camera’s view of the world (tilted) from the beholder’s view of the picture (straight up) and aligns its own internal structure—that’s the force of the parallel between the arm and the horizontal edge—with an orientation that is neither the photographer’s nor the beggar’s but the beholder’s.6 In other words, it’s when the beholder’s orientation is not the beggar’s or the photographer’s (when the horizontal edge that sets the beggar at an angle sets the beholder straight) that the force of the frame emerges. And insofar as the extraordinary personal appeal of the man begging is thus juxtaposed with a formal appeal that is rigorously distinguished from it, the whole thing functions almost as an allegory of Winogrand’s own sense that the attraction or attractiveness of his subjects (often, famously, good-looking women) was a problem (the problem of making the picture of them, not them, the object of interest), and that that problem is what made his work “interesting.” Indeed, if we remember that the reason he gave for not developing his film until years after taking the pictures was that he liked to wait until he had “virtually no memory of the act of taking an individual photograph” (or of the pleasures involved in that act),7 we can begin to think that it’s almost by repressing the act of taking the picture that he understood himself to be making the picture, as if its intentional structure were almost an aftereffect, and what he meant by it were something at least as external as it was internal.

But if we also remember, as the recent Winogrand show has forcefully reminded us, that when he died he left behind at least 300,000 (either undeveloped or not contact-printed) pictures that he never saw, the project of photographing to find out what something will look like photographed (the identity of what I do with what happens) begins to come apart in the opposite direction. The issue here isn’t the ethical one of whether these pictures should have been printed and some of them displayed; it is instead what Winogrand meant by taking them. Like Anscombe’s writer with her eyes shut, the photographer who never looks at the pictures he has made would seem to be engaged in a project that is not legible as art. Or that if it is legible would be so only as a form of conceptualism, where we might be invited to see the innumerable boxes of undeveloped film either as the detritus left behind by the work (here understood as a kind of solitary performance, a performance without an audience) or as the work itself, as if the point of exposing film were to make not pictures but exposed rolls of film. And if neither of these seems very plausible as an account of what Winogrand was doing, what they have in common and what they point toward in Winogrand is his attraction to the inwardness of the act. As if the project of seeing what the world looks like photographed were one that could be fulfilled without actually seeing it, like writing (however illegibly) with one’s eyes shut.

In this sense, we find in Winogrand a more intense version of the difficulties that Anscombe characterized herself as raising—the difficulties involved in thinking through the idea, on the one hand, that intention is not “a performance in the mind,” that “what a man actually does” is “what physically takes place” and the idea on the other hand that what physically takes place may be determined by the “purely interior” difference between thinking about something and meaning something.8 Intention is central both times. That’s why, along with McIver Lopes, we can say that Anscombe is helpful, although not entirely for the reasons he gives. That is, she doesn’t help us disarticulate agency and intentionality in a way that shows us that intention is not a problem for photography; rather, in working through the difficulties of understanding what it means for all acts to be structured by intention she helps us see why making art by taking photographs might sometimes provide a particularly vivid instance of those difficulties.

Which is to say, she is helpful above all for philosophy and (not of) photography, for seeing the ways in which the practices of some photographers have themselves functioned as efforts to think about intentional acts.

In Winogrand, those efforts center on what I’ve been describing as the tension between what the photographer does in taking the picture and what he does in making the work, two moments that can be understood as part of a single act but that, insofar as he imagined the second as the repression of the first and eventually chose the first over the second, come to have an almost aporetic relation to each other. Of course, there are other psychological explanations for Winogrand’s practice in those last years—all of them essentially focused on what he called his own “hopelessness and helplessness about the world.”9 But hopelessness and helplessness (especially hopelessness) are almost technical terms in Winogrand. The photographer who insists that you cannot know when you take it that you have made a good picture (he once said of a photo by Robert Frank “When he took that photograph, he couldn’t possibly know—he just could not know—that it would work….He knew he probably had a chance”10) has hope about the world written into the very structure of his acts. And the photographer who, giving up that hope, continues to take pictures, has found in the interiorizing of meaning (you silly twit, meant not just thought) an act that cannot fail—even if no one perceives it, you’ve expressed your contempt; even if there is no photograph to see, you’ve acted—instead of you press the button, we do the rest, it’s you press the button, no hope necessary, that’s all there is.

So there is a sense in which Winogrand’s personal crisis expresses a theoretical position, just as there’s an equally important sense in which that theoretical position and the alternatives to it (what it might mean to intend something, and especially what it might mean not to or what it might mean for your reasons to be treated as causes) were becoming at the time of that crisis central to literary theory and aesthetic production.11 Indeed, in what is now an aesthetic and a political as much as a philosophical sense, the structure of intentional action has recently emerged as a crucial issue, as I want to suggest by looking briefly and in closing at a piece I wrote about when we began the process that led to this conference.

Owen Kydd, Marina and the Yucca, 2012.

In Marina and the Yucca (and I’m really sorry Marina Pinsky couldn’t be here; anyone who’s at all interested should absolutely check out her fascinating account of being the model for this piece:, the question of action is introduced by the juxtaposition of the person and the plant, and the difference between the movement of the person and of the plant. The leaves of the yucca occasionally waver (as if there were a breeze), and if we were to imagine that Kydd made this work under the influence of Anscombe (a pure counterfactual), we could juxtapose the utterly causal and external character of the motion with her invoking Wittgenstein’s fantasy of “some leaves blown about by the wind and saying ‘Now I’ll go this way…now I’ll go that way’ as the wind blew them” (6). But we don’t need the Anscombe to feel the difference between those physical movements and the human body’s, which is to say, the degree to which the relation between our physical acts and our intentions is not one of commentary. We may think “now I’m writing” as we write but it’s isn’t those thoughts that make writing an act, and the motion of the cactus stands here as an alternative to what Marina’s body does. Which is not to say that the occasional flutter of her slightly closed eyelids, the brief swaying of her body or (less visible but more crucial) the almost imperceptible but real slump of her body count as intentional acts. On the contrary, if you asked Marina why she was slowly slumping she could convincingly reply “I was not aware I was doing that.” (It would be knowledge by observation.) But the sense that those motions are nonetheless hers and that Marina’s inexpressiveness signifies the presence of an interior life whereas the yucca ’s inexpressiveness is a mark of the way in which its relation to itself is more like its relation to the breeze—utterly exterior—is crucial to the contrast between them. A cactus can wilt but it can’t slump.

At the same time, however, in the moment in which the video begins to repeat and Marina is made, in effect, to sit up, that unintentional act is transformed into a completely intentional one, not Marina’s but the photographer’s. That is, the photographer produces an intentional act out of Marina’s unintentional one. And he does so in a way that, far from being problematized by the automatism identified with the camera, is enabled by it—in fact, is only possible because of it. The local point here would be that in Marina and the Cactus, we see an allegorization of art as the rendering intentional of something that isn’t, as the production of intentionality itself. The more general point would be that this particular project makes sense only in relation to a moment in which, for reasons that have to do with both with the history of art and the history of capitalism, the question of intentionality has become deeply identified with the question of autonomy, thematized here by the way the work produces rather than records the act.

What does it mean to think of the problem of intentionality or the possibility of aesthetic autonomy as political issues? With respect to art produced in the 21st century, that question and an effort to answer it is at the heart of my book, The Beauty of a Social Problem (Chicago, 2015). With respect to Winogrand, however, making pictures before the beginning of the dramatic rise in American inequality that today makes itself felt as both the symptom of a problem and as a problem itself, both the question and the answer will be a little bit different. After all, the photograph of the beggar was made in 1968, the most economically equal year (in terms of income) ever recorded in the U.S. For now, perhaps the main point worth making is just that the project of converting the interest of the subject into the interest of the photograph has a slightly different meaning when the subject is a beggar from the one it has when the subject is a good-looking woman. Each involves a politics, but not the same politics. And the difference between those politics (between, for example, what it means to make art out of exploitation and art out of discrimination) would be, in my view, foundational for the subsequent history of photography.

But the topic of our conference is photography and philosophy not photography and history. And not, as I noted earlier, philosophy of photography. So the final point I want to make is that photography and philosophy is interesting in a way that the philosophy of photography is not because we don’t really need a philosophy of photography, especially in the sense of needing a theory of what photography really is or how photographs mean. Which is fortunate since the theories we’ve been offered are mainly either wrong or (more often) beside the point. Of course, a reading of Anscombe might well lead to a better theory, one that would solve the kinds of problems about intentional action that photography can be understood to raise. But what matters for art (and for politics) is less the act that produced it than the photograph’s understanding of the act that produced it.


1. G.E.M. Anscombe, Intention, 2nd ed. (Cambridge: Harvard University Press, 2000), 52. Subsequent references are in parentheses in the text.
2. Diarmuid Costello and Margaret Iversen, “Photography Between Art History and Philosophy,” Critical Inquiry 38:4 (Summer 2012): 688.
3. Dominic McIver Lopes, “Afterword: Photography and the ‘Picturesque Agent,’” Critical Inquiry 38:4 (Summer 2012): 862. Subsequent references are in parentheses in the text.
4. Mason Resnick, “First-Person: How Garry Winogrand Changed My Life as a Photographer,” First accessed December 2014. The article was originally published in Modern Photography in June 1988.
5. Barbara Diamonstein, “An Interview With Garry Winogrand,” First accessed December 2014. Originally published in Barbara Diamonstein, American Photographers on Photography (New York: Rizzoli, 1982).
6. “I have a picture of a beggar,” Winogrand says, “where there’s an arm coming into the frame from the side. And the arm is parallel to the horizontal age and it makes it work.” Where by “make it work,” he means makes it unlike what was seen or what was there to be seen: “A photograph is not what was photographed, it’s something else” (Diamonstein).
8. Another way to put this with respect to Winogrand would be to note that, whether he meant to develop them or not, his physical actions in taking the photographs were, up to a point, the same: going out on the street, looking through the viewfinder, pressing the button. But for most of his life processing the film, looking at the contact sheets, selecting some images, printing them, etc., were also a part of the action. This is what Jennifer Hornsby means when she says, “Where actions are compound, their components constitute a whole by virtue of the intention that unites them.” (Jennifer Hornsby, “Actions in Their Circumstances,” in Essays on Anscombe’s Intention, ed. by Anton Ford, Jennifer Hornsby and Frederick Stoutland [Cambridge: Harvard University Press, 2011], 120). But should we think of going out on the street, looking through the viewfinder and pressing the button as the same act when the rest (processing the film, etc.) is abjured? In the case of the conceptualist looking to produce boxes of undeveloped film, obviously not; what he means to do would be ruined by the rest of the steps. For him, the intention that unifies the whole involves producing those boxes, not some prints. My point here is not that Winogrand was (or became) that conceptualist. It is instead that the way to explain the existence of such a conceptualist would be by reference to his interest in the relation between act and intention (in, say, the identity or non-identity of the physical movement and the act) and that a way to think about the limitations of his work would be by asking whether the boxes he produced had succeeded in making that interest compellingly legible (in which case he would be no more conceptualist than any other great artist?).
9. Quoted in Garry Winogrand, Figments of the Real World, intro. John Szarkowski (New York: MOMA, 1988), 40. The context of this self-characterization—an answer on a form he was asked to fill out in connection with treatment for the cancer he’d just been diagnosed with to a question about what if anything about him might have helped bring on the disease (!)—makes the psychological point obvious. But the structural relevance of the remark both to Winogrand’s work and to the problem of intentional action should be equally clear. Just before introducing “I do what happens,” Anscombe rejects two ways of thinking about intentional action, the first of which is the idea that it’s “only the intention, or possibly also the bodily movement” (51), and that what actually happens is only the result, that you can, for example, “will” your arm to move but not a matchbox. “But that is a mad account,” she says, since the only meaning of “will” here is something like staring at the matchbox and willing it to move and that won’t work for your arm either. The second way is to think that the intentional act is just what goes on in your head, and if what actually happens “coincides” with what you intend “that is just a grace of fate” (52). This view she identifies with a passage in the Tractatus and with a remark in Wittgenstein’s notebooks: “I am completely powerless.” This “is nonsense too,” she says, for reasons we have already noted above (a bombination in a vacuum). But the point of thinking about Winogrand’s helplessness and hopelessness in in the context of Wittgenstein’s powerlessness is to suggest that although the hopelessness may indeed be psychological the questions of help and hope were, even in happier times, built into his artistic practice, just insofar as that practice involved a continual reflection on the relation between doing and happening.
10. The ASX Team, “Monkeys Make the Problem More Difficult—A Collective Interview with Garry Winogrand (1070),” First accessed November 2014.
11. It’s arguable, however, that the terms of the question were at the same time being transposed, and that thinking about the relation between I do and what happens (in other words, thinking about intention in the context of the theory of action) would take the form (especially in literary theory) of thinking about the properties of the text (“itself”) and of its relation to the writer and especially to the reader.
12. See also my “The Force of a Frame: Owen Kydd’s Durational Photographs,” to which Pinsky is responding.