It was my sense that a recent series of essays by Walter Benn Michaels that addressed the work of Elizabeth Anscombe, especially her seminal and challenging 1957 volume Intention, in light of artistic phenomenon, brought new light to basic questions around action, agency, and artistic intentionality. We asked five respondents—Joshua Landy, Rob Chodat, Magdalena Ostas, John Schwenkler, and Mathew Abbott—to engage the questions raised by Michaels’ recent writings, and the response to them appears below. We want to extend a special thanks to our original respondent, Dominic McIver Lopes, whose “Making, Meaning, and Meaning by Making” set out some of the basic terms of the debate. The respondents considered four essays by Michaels, all of which can be accessed here via hyperlink: “‘I Do What Happens’: Anscombe and Winogrand,” “Anscombe and Winogrand, Danto and Mapplethorpe: A Reply to Dominic McIver Lopes,” “‘When I Raise My Arm’: Michael Fried’s Theory of Action,” and “Blind Time (Drawing with Anscombe).” My warm thanks to all who contributed to the discussion.—Todd Cronan

Literary Theory

This exchange was organized by Todd Cronan, who felt that literary critical and art historical interest in what Magdalena Ostas describes as the “old topic” of intention is currently being, as Ostas also says, “renewed” and that Elizabeth Anscombe’s Intention is a useful text in helping us think about what might be at stake in this renewal.1 The reason for this is that Anscombe understands the question of intention as a question about action—“what a man actually does” (Intention, 9). By contrast, the debates that were central to literary theory from the late 60s through the early 80s tended to think of intention as essentially internal, a state of the author’s mind, and thus separate from the “text itself,” which separation more or less inevitably produced a series of arguments about whether text and intention could or should be reconnected. For awhile, the focus was on whether trying to connect the text with its author would make criticism more or less objective. Indeed, both “The Intentional Fallacy” (1946) and the first really serious critique of it, E.D. Hirsch’s “Objective Interpretation” (1966), were written to enable what Wimsatt and Beardsley called “objective criticism.” Where they differed was on how to do it. But by the early 70s, the arguments against intention had come to be deployed against rather than for objective interpretation, and, more strongly, against the idea that texts had a single meaning to be interpreted. Intentionalists, however (insofar as there were any), mainly stuck to their guns, continuing to defend the search for intention as a method and also (albeit with less conviction) the idea of a single meaning or at least of a limited number of acceptable meanings, which intentionalism could help you distinguish from the unacceptable ones.

It was for this reason that “Against Theory” (1982)—refusing the separation of intention and text on which the whole debate was founded—criticized both intentionalists (like Hirsch) and anti-intentionalists (the exemplary instance was the deconstructive critic Paul de Man). The argument of “Against Theory” was that intentionalism had absolutely no methodological value (and more generally that there could be no such thing as literary critical method) and that the same text couldn’t possibly have more than one meaning, the one meaning intended by its author. So intentionalists were mistaken to think that intentionalism could provide constraints on interpretation and anti-intentionalists were mistaken to think that the impossibility of producing such constraints somehow meant that the same text could have many meanings, or even more than one.

Neither objectivity nor method comes up in the essays that serve as the occasion for this discussion (and the question of whether works of art have a single meaning is present only indirectly) but they are usefully revived and entertainingly engaged by Joshua Landy’s spirited defense of the argument in Alexander Nehamas’s “The Postulated Author” (1981), a text that was very much a product of those debates in the 70s. Nehamas was skeptical about the pluralist implications of what he understood as deconstructive claims like no work has “a fixed, identifiable meaning” and he thought the “self-evident” fact that “Literary texts are produced by agents and must be understood as such”2 mitigated against those implications. Which sounds a little like “Against Theory” except that when “Against Theory” came out a few months later, he found it (in an endearingly appreciative way) “radical” and “outlandish.”3

The big difference, of course, between “Against Theory” intentionalism and Nehamas’s and Landy’s “Postulated Author” is the “postulated.” It’s the difference, as Landy puts it, between the “empirical maker” and the artist we “imagine” when we look at his work. Why do we need to imagine an author? Because, Nehamas says, the actual author, the “historical writer” is the text’s “efficient cause” (145) and is thus irrelevant to the text’s meaning. Which is, as Nehamas notes, exactly what Wimsatt and Beardsley had thought. “A poem does not come into existence by accident,” they wrote; “The words of a poem come out of a head not a hat.”4 Hence the “historical writer” is relevant as the “cause of a poem” but not in determining the poem’s (or picture’s or work’s) meaning.5 The anti-intentionalist picture of intention (the one Cavell called “a bad picture”) is reproduced in Nehamas’s and Landy’s intentionalism.

But where Wimsatt and Beardsley turned from intention to the syntactic and semantic rules of the language for the text’s meaning, Nehamas appeared to recognize (what Hirsch had shown) that the rules were actually just the royal road to the kind of deconstructive pluralism he wished to oppose, and he also recognized that merely postulating an author wouldn’t entirely solve this problem. For, to take a famous example, what happens when the rules change, when To His Coy Mistress’s “vegetable love” begins to sound as much like “erotic cabbage” as “growing passion?” And what happens if the author we postulate—the one we imagine when “we try to assign all the different effects of a text to a given cause” (Landy)—is the one who produced on us the erotic cabbage effect? You’re back with the radical pluralism you invoked intention to oppose in the first place. So Nehamas rescues his postulated author from this mise-en-abime by introducing a “methodological constraint” –that “the postulated author be historically plausible,” which he formulates as “the principle…that a text does not mean what its writer could not, historically, have meant by it” (145). “Vegetable love” does not mean whatever the historical Marvell meant by it (he’s just its cause), but it can’t mean what a 20^th century theorist like René Wellek might have meant by it either—it can only mean something that Marvell could have meant.6

Now, of course, if the historical writer were the author, this principle wouldn’t be necessary; it wouldn’t even be a principle, it would be a tautology. You need it as a principle only because once you’ve separated the author from the historical writer you need some way of making sure that not just anyone can count as the author. As Landy puts it, you don’t want the actual maker but you want a “plausible variant.” But why? Since we’re not interested in what Marvell meant, why should we be interested in what Marvell might have meant? The choice of someone sort of like Marvell is as arbitrary as would be the choice of someone exactly the opposite of Marvell, which would also serve the purpose of providing “methodological constraints,” even though the constraints would be different ones. And (setting the arbitrariness aside), If our idea of the artist’s intention is as the efficient cause of the text, how are we supposed to understand the relevance of our imaginary authors intentions? Why aren’t they just imaginary efficient causes and hence equally (at least!) irrelevant?

This would be, from the standpoint of literary theory, why Anscombe is helpful. What John Schwenkler calls her “fundamental disagreement” with the idea that “we find intentional activity whenever a person…causes something to happen” (italics his) is a fundamental disagreement with the idea that we can think of a person’s intention as the cause of her acts, which is the idea that anti-intentionalists like Wimsatt and Beardsley and intentionalists like Nehamas and Landy have completely in common. So I disagree with Landy both on the utility of the postulated author and the irrelevance of Anscombe.7 Once we think we shouldn’t be interested in what the historical writer meant, no postulated author will do us any good both because there’s no way to choose one over another and because, whatever choice we make, we have no explanation of why imaginary efficient causes should be more relevant to meaning than the real ones are. Indeed, when Landy describes the postulated author as “the result we get when we try to assign all the effects of a text to a single cause,” we realize that the commitment to the postulated author has turned out not to help us interpret the work but to make the whole idea of interpreting it irrelevant. For if we identify the work’s meaning with its effect on us, we’re no longer trying to say what it means, we’re just reporting our response to it. The postulated author might as well be Beardsley’s chimpanzee.

Le singe peintre

Indeed it was entirely because of his commitment to intention as cause that Beardsley thought the beholder’s response to a painting would not be “invalidated” by the “discovery” that it had been “painted by a child or a chimpanzee or a machine.”8 Thus, although Matthew Abbott thinks my description of the chimp’s work as “paint randomly applied to the canvas” is unfair because it misses the way the chimp may well “have meant something, may even have sought to express something,” for Beardsley, the question of whether the chimp might actually have meant something was irrelevant.9 Indeed, for anyone who’s committed to the artist’s intentions as causes (and hence to their irrelevance) all artists might as well be chimps. Poems may come out of heads not hats (otherwise there wouldn’t be so many of them) but they wouldn’t mean anything different if they did somehow come out of a hat (or were inscribed on the sand by the waves washing against the shore). If, as theorists, we’re committed to a causal theory of action, we’re stuck forever with identifying the effects some marks or sounds have on people with their meaning and trying to invent some principle by which we can distinguish between the effects we’re prepared to accept as interpretations and the ones we’re not. But if we’re not committed to intentions as causes we don’t have to be theorists any more.

If, however, we turn our attention from literary theory to photography we can see what it’s like to imagine not exactly that the artist is a chimp but that the camera has turned him into one. Thus, although Abbott thinks I am “confusing the issue” when in The Beauty of a Social Problem, I identify “the refusal of intentionality” as part of the attraction of photography, it’s hard to know how else to take Rosalind Krauss’s reminder that “It is the order of the natural world that imprints itself on the photographic emulsion and subsequently on the photographic print.”10 Or Jacques Rancière’s idea that because a photograph like this one by Walker Evans, “tells us neither what the person who laid the planks and cutlery in this manner had in mind nor what the photographer wanted to do,” because it doesn’t even “say whether it is art or not,” it “perfectly illustrates what Kant designated under the name of the aesthetic idea.”11

For both Krauss and Rancière, “what the photographer wanted to do” is in some way negated—rendered irrelevant or unknowable—by the technology that makes it possible, and they understand this negation as part of the interest of photography. Which would seem to suggest, as Chodat says, that Anscombe, with her interest in practical reasoning and “skillful know-how” (what a man does and the intention with which he does it) might not be so relevant to “the sorts of artists and artworks that interest Michaels.” Why care about skill at a moment in the history of art which in important ways is a product of the revolution decisively influenced by the very phenomenon of de-skilling? Why look to Anscombe, Aristotle and practical knowledge when the great interest of the photograph (and really in all the art made under the quite different signs of Duchamp and Pollock) is that it seems to make practical knowledge irrelevant?12

But in fact, it was when the question of the artist’s skill began to look irrelevant that the question of what the artist was doing began most intensely to matter. It’s no accident that one of the first forms of painting that seemed to refuse skill (think De Kooning or Pollock’s drip paintings) was quickly described as Action Painting. Even if we don’t accept a single word of Harold Rosenberg’s account of Pollock, de Kooning et al, we can hardly fail to recognize that what looked like the deemphasizing of a certain kind of technical virtuosity placed a new pressure precisely on the question of the act. What kind of act was involved in the production of a readymade? Why was the physical motion of dripping paint onto a canvas the act of painting? The abstraction of the painting was a way of thinking about action in the abstract—what is an action?13

Thus Pollock was crucial to Robert Morris because he made what he was doing—his “process,” the movements of his body, the properties of his material (“how paint behaves under the conditions of gravity”) “visible in the work.”14 What mattered was what Morris called the “the making behavior” or (even more vividly, more like Anscombe), “the ends-means hook-up” (83). And, of course, once you’re interested in “making behavior” and in making that behavior visible, the question of what an action is has become imbricated in the question of what a work of art is. Hence what I describe as Morris’s effort in Blind Time (Drawing with Davidson) to make visible an essentially causal account of action and the refusal of “form” that he thought was entailed by it. And hence also, the efforts of many of the artists I write about whose interest has been not in refusing form on behalf of process but in making form that hangs on to process, an interest that entails a refusal of event causality. Which is not to say that any of these artists has been interested in the philosophy of action. Actually, even the fact that Morris was a remarkably sophisticated reader of Davidson and a very smart guy is by no means necessary to my account of his work.15 What is necessary is the way the work addresses the question of what it is by addressing the question of what an action is.

Chodat worries that Anscombe’s examples of means and ends—opening a window, writing on a blackboard, giving someone a contemptuous hug—are not “germane” to the more complex and extended “process of interpretation” that we carry out with works of art. Insofar, however, as some of the most challenging works of art in our period are challenging precisely because they seek to extend or alter or undermine our sense of what it is to make art, Intention seems to me much more relevant than most of the usual suspects. Although not, of course, if you’re looking for a better intentionalist way of interpreting such works, both because we don’t need such a thing and because there is no such thing. As long as our experience of a work is, as Cavell—powerfully because innocuously—puts it, the experience of feeling that “I am meant to notice one thing and not another, that the placement of a note or rhyme or line has a purpose,”16 we don’t need Anscombe’s or anyone else’s intentionalism because we can’t help but be intentionalists. Anscombe helps us explain what we’re doing, she doesn’t provide better resources for doing it.17

But Chodat’s deeper and I think more fundamental worry is that Anscombe’s interest in “purposive behaviors performed either with particular intentions or with particular goals” makes her account of action irrelevant to art because “aesthetic pleasure arises when we grasp not a particular intention but intention as such, not the meaning but meaningfulness.” It seems to me, however, that Anscombe actually helps us see the limits of the very idea of grasping “intention as such.”

We can see a version of intention as such in Rancière’s list of the various ways we might think about what he calls the “harmonious dissymmetry” (13) of Evans’s Alabama Farm Interior. We might think that the photographer was trying “to transpose a completely random or functional layout into an artistic quality.” Or that he was trying to show us “the aesthetic taste of the inhabitants” or “that he simply photographed what was in front of him without any particular intention and that the photograph thus benefits from the beauty of the random.” Or that, “attracted by the simple and brute material, and the art of living and doing transmitted by generations of simple people,” he “wanted to underline a certain beauty of the functional.” There are others we could add: e.g. that the photo makes something aesthetic out of what is less functional than dysfunctional, a product of the Fields’s poverty, and that his sense of them is less as “simple” (Rancière’s word) than as “damaged” (James Agee’s). What Evans is trying to do on this account is to mark a radical difference between the photographer and his subjects with respect to something they have been deprived of (it’s part of the damage): the very idea of a work of art.

But for Rancière the whole point is that we cannot choose among these interpretations because the photo “does not tell us” “what the photographer wanted to do” and “We don’t know what was going through Walker Evans’s mind in framing his photo as he did” (13). In other words, Rancière thinks of photography as creating the conditions in which the work of art produces speculation about what the artist meant while locating whatever the artist actually did mean in his head rather than in his act. And he sees in this what “Kant designated under the name of aesthetic idea: ‘a presentation of the imagination which prompts much thought, but to which no determinate thought whatsoever, i.e., no [determinate] concept can be adequate.’” Here Rancière’s Kant meets Chodat’s “not the meaning but meaningfulness.” Except that if what the artist actually was trying to do is rendered in principle unknowable or irrelevant, our interpretations of what was meant are turned into imaginings of what might have been meant. Basically we find our ourselves (minus the interest in “methodological constraints”) back in the world of the postulated author.

The problem with that world, however, is that no one lives in it. The minute you ask even the most basic question (why did Evans photograph that?) you’ve committed yourself to the possibility of a particular answer and to the idea that one particular answer will be true (“the meaning”) and the others false. Of course, you may not care which one is true but the point here is about the logic of interpretation not its psychology. If Evans did x because y then he didn’t do it because not y. That’s not “meaningfulness”; it’s “the meaning.” And if you suspend the question of what particular thing the artist was trying to do (whatever Evans was trying to do, I see in that picture the dignity of simple people) then you don’t have the meaning but you don’t have meaningfulness either; you just have your response. So Anscombe’s interest in particular intentions and particular goals helps us see the degree to which an interest in what the photographer wanted to do (his particular intention) is built into our relation to the work of art, even and especially (not despite but because of the plausibility with which it can be ignored) to the photograph.

The Wealthiest Americans

It’s in this context that I want to register the force of Magdalena Ostas’s comment on the “tension” between my “investment in the category of intention” and what she thinks of as my “refusal…to consider…how we become engaged in and attentive to art as the very people we are.” She’s clearly right that there are “ways of understanding art’s forms of address and appeal that do not rely on a strict contrary between form and human feeling, aesthetic composition and personal responsiveness” and clearly right that I’m very attached to ones that, at least in some sense, do. But my original investment had nothing to do with art. It was an effect of spending many, many hours with Steve Knapp arguing about Derrida’s understanding of intention in “Signature Event Context” and Limited Inc., out of which emerged “Against Theory,” and, necessarily, the claim that since every work of art (indeed every speech act) only and always meant what the speaker meant, then its meaning never had anything to do with the response of the reader. In literary theory, of course, that claim went against what many people thought; in art it also went against what many people were trying to do. That is, the idea (in Douglas Crimp’s words) that “the art object in and of itself” had “a fixed and transhistorical meaning” (19) was identified with a practice (“modernist) as well as a theory, and opposition to it involved a different practice—one that sought to escape or overthrow modernism and that crucially involved an appeal to the role of the beholder in determining the meaning of the work.

My own sense of what was wrong with the theory preceded my sense of what I came to think was wrong with the art, although the last chapter of The Gold Standard and the Logic of Naturalism (“Action and Accident: Photography and Writing”) and the book’s cover (one of James Welling’s Aluminum Foil photographs) turned out to be a gesture in the direction of thinking about intention as an aesthetic question. And actually, that gesture was more about the art (beginning with Welling) that I found compelling than about the art I didn’t. The theoretical point was that the meaning of every work was always determined by the artist’s intention and hence was independent of the beholder’s response; the aesthetic point was that what every work always has—autonomy of meaning — could, redescribed as autonomy of form, be something the work sought to overcome or could become something it sought to reassert. So Ostas is right that the “contrary between form and human feeling” shouldn’t really be strict since the theoretical independence of the one from the other need not be made into aesthetic opposition. But it could be and, for many of the artists in whom I was interested (artists born and raised in the hegemony of the postmodern, the performative and the beholder), it was.

To take an example outside of photography, think of the pre-Batman work of Christopher Nolan and especially Memento, which both insists on and distances itself from its surprise ending, that is, from its focus on producing a certain experience for the viewer. Long before we have the chance to be surprised, the film has imagined for itself an alternative aesthetic in a brief conversation between Leonard and his wife, when he asks her why she’s rereading that book she’s read “like a thousand times” and she responds, “It’s good.” According to Leonard, the “pleasure of reading a book is in wanting to know what happens next,” a pleasure you can’t experience when, having already read it, you know what’s going to happen next. But according to his wife (and, I argue, according to the movie https://electronicbookreview.com/essay/the-death-of-a-beautiful-woman-christopher-nolans-idea-of-form/), what makes the book good is something that must be disarticulated from that surprise. The movie’s form not only exceeds the viewer’s experience but—opposing the pleasure of being surprised to the pleasure (“I enjoy it,” she says) of understanding what your surprise means– depends upon a certain repression of it. Form and feeling are not just different; the one emerges at the expense of the other.

Thus, rather than insisting on a “strict contrary between form and feeling,” it would be better to say that in certain work, the independence of form from feeling gets turned into a contrary, which involves a certain development in the history of art (the turn to the beholder that it both engages and transcends) and in political economy (the neoliberal turn—redescribing the structural opposition between labor and capital as the individualizing exchanges between human capitalists). Anyone interested in that argument can read, as Abbott has, The Beauty of a Social Problem,18 but, as we have already begun to see with his defense of Beardsley’s chimpanzee, his criticism has less to do with our relation to the work of art than with our relation to each other. He thinks I don’t fully recognize the importance of the “recognition of difference” at the heart not only of anti-racism and anti-sexism but also anti-speciesism, and his diagnosis of this failure is that I’m “stuck in (what was once called) economism.”

The burden of this charge in Abbott is the reminder that “capitalism is vicious not just because it creates wealth inequality but because it exploits and denies what Marx calls our ‘free activity,’ subordinating our capacity for intentional action to the capitalists who buy it as labor power.” Setting aside the parallelism here between “intentional action” and “free activity” (don’t I do what happens even if I do it at the point of a gun? It’s at least not obvious that action that is less free is less intentional), I am happy to grant both that money isn’t everything and that “control of production” is indeed a more radical goal than redistribution of wealth. But I am struck by the historical circumstances in which the desire to remind people that there are problems as important as “wealth inequality” and solutions more radical than redistribution has seemed to become more urgent. Both the above points were made against The Trouble with Diversity,19 which was published in 2006 when the bottom 90% of the American population controlled a little less than 30% of the wealth. That was bad compared to what they had even 15 years before but, as the chart below shows, it’s great compared to what they have now. Still, the charge of economism retains its appeal; the greater the (upward) redistribution the louder the call for recognition, and conservatives, it turns out, are not the only ones who love to tell poor people there are things more important than money.

Which is not to say that ending capitalism wouldn’t be better than redistributing surplus value. (Although here too, after a half century of uncontested redistribution upward, the exhortation to not worry so much about redistribution downward [which we haven’t been doing anything about] and worry instead about taking control of the means of production [which we don’t even have any idea how to do anything about] strikes me as, at best, politically ambiguous.) But, for The Beauty of a Social Problem’s claims about the political economy of form, the more relevant point is that the goal of controlling the means of production is even more tied to the irreducible conflict between labor and capital than is the idea of redistribution. Insofar as I argue that the commitment to formal autonomy is a way of modelling a world produced by the structural antagonisms of class (antagonisms that exist regardless not only of how we see each other but of what we feel we owe one another) the art I’m describing is much more radical than my politics are. The good news is that there’s a place in art for that radicalism; the bad news is that there’s no place for it in our politics.

I am a fool

And one sure sign of my foolishness is the fact that I’m about to disagree with the author of the (really great) Oxford Guide to Intention on a point of interpretation. But I’ll try to rescue myself a bit by making the disagreement as narrow as possible. The disagreement is about a famous passage that raises an objection to “I do what happens” by insisting that even something that isn’t happening can count as what you’re doing. If you write “I am a fool” with your eyes shut and the pen runs off the page or out of ink and the words don’t appear, you are still, Anscombe says, writing “I am a fool.” So the problem is that if what I do is something I must know without observation (even with my eyes shut), I’m writing “I am a fool”; if what I do is what happens, I’m not writing “I am a fool.” In my essay on Anscombe and Winogrand, I treat this as a contradiction in Intention (although what I actually say is something more equivocating– it’s “a difficult moment”) but Schwenkler thinks that’s a misreading because Anscombe is “not putting” the argument that I know what I’m doing even when it’s not happening “in her own voice”; she’s articulating a position she means to reject.

I’m not so convinced by this. After all (as Schwenkler knows better than I, so I’m not imagining I’m calling overlooked passages to his attention but I am imagining they don’t work exactly the way he describes them as working), Anscombe comes up with a version of this problem on several different occasions and says very similar things. Indeed, right after another passage, just as famous but with very little disagreement, the two-step is repeated. The passage is the one that says “I write ‘I am a fool’ on the blackboard with my eyes shut. Now when I said what I wrote, ought I to have said: this is what I am writing, if my intention is getting executed: instead of simply: this is what I am writing?”

The answer, clearly, is no, an answer that as I wrote it sounded so familiar that I started looking around among the things I’ve been reading and came across, in the exact same context, Adrian Haddock’s “There is agreement that the answer to the first question she raises here is ‘No.” Followed by an equally helpful, “However, opinions diverge as to how we are to make sense of the ensuing remarks.”20 These are the ensuing remarks:

That intention…would not have been executed if something had gone wrong with the chalk or the surface, so that the words did not appear. And my knowledge would have been the same even if this had happened. If then my knowledge is independent of what actually happens, how can it be knowledge of what does happen? Someone might say that it was a funny sort of knowledge even though what it was knowledge of was not the case! On the other hand, Theophrastus’ remark holds good: “the mistake is in the performance, not in the judgment.”

Haddock goes on to ask, “when she says ‘And my knowledge would have been the same even if that had happened,’ is she speaking in the voice of an interlocutor who is making a claim, which the succeeding remarks, said in her own voice, are designed to reject?” And he, like many before (and probably many more to come) works out his own account of who exactly is saying what.

So why should those of us who are interested in reading Anscombe alongside Robert Morris care about this particular thorny patch? What almost immediately follows this last invocation of Theophrastus is a list of actions which, Anscombe says, fit the “form of description ‘intentional actions” (85). For some of them (e.g. Telephoning, Signing, Signaling, marrying,” the “role of intention will be obvious.” Others (e.g. “Kicking [and other descriptions connoting characteristically animal movement])—she feels the need to defend and she does so by saying that what all the descriptions of intentional action have in common is that they “go beyond physics” (86). Any reader who remembers what Is virtually the mission statement at the beginning of the book, will be struck by this. Many things lead us to believe, she writes early on, that if we want to know what someone’s intention is, we should focus on the “contents of his mind” and that “what physically takes place…is the very last thing we need consider”; “Whereas, I wish to say that it is the first” (9). So at the beginning, we need to start with what physically takes place but, at the end, we need to go beyond physics.

My point here is not that this is a contradiction. After all, going “beyond physics” might be the next step after starting with “what physically takes place.” But that would give us a picture of action as the physical motion of the body plus the intention as a mental state we attach to it, which is the opposite of Anscombe’s. Rather what’s suggested by the juxtaposition of starting with what physically takes place and going beyond physics is the idea that what physically takes place can’t be reduced to physics. If what physically takes place is intentional, and what happens is what physically takes place, what happens must be already under the sign of the intended.

At the end of the book, along similar lines,  she distinguishes between Theophrastian “practical” knowledge and “speculative” knowledge by calling the first “the account that one could give of what one was doing, without adverting to observation” (eyes shut: I’m writing “I am a fool”) and the second, “the account of exactly what is happening at a given moment (say) to the material one is working on” (eyes opened: “I am a fool” is not getting written). It’s pretty clear that “what happens” isn’t what “is happening to,” that I do what happens is not I do what is happening to the material I am working on, and that what happens is not physics.

My point here is that while Anscombe clearly recognizes the tension writing with your eyes shut produces between what “I do” (known without observation) and what (I observe) actually “happens,” it’s not all that clear that she wants the truth of the second to obliterate the first. Indeed, the invocation of the difference between mistakes in judgment and mistakes in performance seems to me precisely the expression of her desire to have it both ways, even if, as Richard Moran says, “it is not a good answer to this problem of error and knowledge to advert to Theophrastus and the thought that the mistake here lies in the performance and not in what is said.”21 Because the attraction of the appeal to performance, and this is something Moran also says, is that if you ask what the woman failing to write “I am a fool” is doing it’s “not clear that there is any better answer that could be given to the question” of what intentional act is being performed. And it’s not at all clear you’d want to say that no intentional act is being performed.

We can parse these distinctions in the failed act of writing where I say I’m writing “I am a fool” and you see I’m not. From the standpoint of physics, we can’t actually say I’m failing—all we can give is the causal account of what my arm is doing and what is happening to the chalkboard. It’s only “the agent’s knowledge of what he is doing that gives the descriptions under which what is going on is the execution of an intention” so it’s only the blindfolded agent’s (practical) knowledge that she’s writing “I am a fool” that enables us or her (when she opens her eyes) to say not exactly that she isn’t writing “I am a fool” but that she is failing to write “I am a fool.” And it’s that description Anscombe wants to hang on to. If the place to start is with what physically takes place (if the place to start is with what happens), what’s physically taking place, what’s happening, is that she is failing to write “I am a fool.”

“Working blindfolded for an estimated 8 minutes, the hands attempt to blacken the entire page.” Morris’s Blind Time IV drawings are accompanied by handwritten descriptions of what he was doing, characteristically in terms of what he was attempting to do. So the question of failure and success is built into them. But not in the same way it’s built into “I am a fool,” since it’s easy (at least for an English speaker) to see whether Anscombe’s failed or succeeded but, without the descriptions, no one could tell by looking at the partially blackened page below that Morris was trying to blacken the whole page. Which is why he gives us the description.

(The photo is of a page from the beautiful book, Robert Morris Blind Time Drawings, 1973, brilliantly conceived and edited by Jean-Pierre Criqui from the exhibit he curated at the Centro per l’Arte Contemporanea Luigi Pecci in Prato and brilliantly printed by Steidl.)

Each drawing is, in effect, accompanied by a statement of intent, almost as if it were a literalization of the idea that the intention is found somewhere outside the work. But where the idea that the intention is not visible in the work has characteristically led theorists to think of it as irrelevant, Morris wants us to know what it was. He wants us, consistent with his interest in the “end-means hook-up,” to see the drawing as the product of the process by which it was produced. Which means that the Beardsley-style exhortation to ignore the question of the painter’s intentions and to find the meaning in the work itself is rendered impossible. If, for example, we were to start talking about the meaning of the contrast between the darkened top part of the drawing and the lightened bottom third, our interpretation would be utterly short circuited by the text which, after telling us “the hands attempt to blacken the entire page,” goes on to remark, “Depletion of material and unintended variation of motion…occur.” The contrast between dark and light is a record of that depletion. And this is true even if we think of the relation between intention and drawing not as literalized by the separation of the two but as represented by it. After all, the intention is announced within the frame; why shouldn’t it be considered part of the work? The statement of intention thus becomes a way not only of describing how the work was made but of turning the way it was made into its subject.

This was what Morris had praised in Pollock and Morris Louis when in “Anti Form” he claimed that their “recovery of process” had brought them closer to the “physicality” (41) of painting, and suggested that “In some ways Louis was even closer to matter in his use of the container itself” (42) to pour the paint. But how we’re supposed to understand the relation between the process and the painting is very different. One way we might put the difference is that we’re supposed to see what Louis was trying to do in the painting. This means first that we can see the paint as poured so if the paint thins as it runs down the canvas we can see that in (as part of) the painting. And it means, second, that the fact the paint was poured is, as Morris himself said about Pollock’s drip paintings, “part of the end form of the work” (43). In Morris, understanding the action explains why the drawing looks the way it does and forestalls the question of what the drawing means; the statement of intention is his way of trying to make the process that produced the drawing visible without making the drawing meaningful. In Louis it’s the opposite; understanding the action is a way of understanding the painting.

The point here is not to criticize Morris by contrast to Anscombe or even to Louis. Actually in the relation between Morris Louis and Robert Morris, I can’t help but see a kind of chiasmus (I do what happens/what happens is not what I did) that vividly instantiates the logic of “making behavior” at a moment when the question of how to understand that behavior had emerged as central to the idea of the work of art. And it’s in Morris that we see why Anscombe is so committed to “I do what happens” and to the idea that “what happens” is not “what is happening to.” In Morris, the account of what you’re trying to do (“the hands attempt to blacken the entire page”) is understood as turning what happens (the page isn’t entirely blackened) into the result of what you did. What you did was not what happened; what happened was physics. Whereas what Anscombe wants is for the teleological structure of the action to be dispositive whether you succeed or not. She wants what you do with your eyes shut (what you know without observation) to be what happens precisely because, if it isn’t, then what happens (even with your eyes open) becomes, like the Blind Time Drawings, physics.