Critical Response

What Do We Want Photography to Be? A Response to Michael Fried

James Elkins

Although Michael Fried is easy on previous readings of the punctum, it has arguably been one of the two most often misused terms in recent photography theory (the other candidate would be Charles Peirce's idea of indexicality).¹ The punctum is used to speak about viewer's responses that are taken to be idiosyncratic, unpredictable, or essentially incommunicable: yet by citing the punctum to theorize such responses, historians and critics make it public and accessible to other readers, which is, I take it, the exact opposite of what Barthes intended. In effect the punctum becomes an unusual example of the studium, which Barthes disparagingly calls "a kind of education."²

This problem of the punctum is nestled within the problem of what can best be made of Camera Lucida. It is strange that after all the critical writing of the last twenty-five years, Barthes's "little book"—so he called it, reminding us how much is really in it—remains a central text, cited almost by default as a source of insights about photography's "essential features" (CL, p. 3). This is despite the fact that readings by Derrida and others have shown how the text fails to provide the theory it initially promises and how it enacts that failure by contradicting both its claims to universality ("I wanted to learn at all costs what Photography was ‘in itself'") and to privacy ("what I can name cannot really prick me" [CL, pp. 3, 51]).³ For Patrick Maynard, Camera Lucida has no purchase on photography at all, and is instead a meditation on mourning and representation that happens to use images as catalysts.⁴ Nancy Shawcross has argued that Camera Lucida is an experiment in what Barthes called the "third form" between essay and novel, making it unavailable, except by wilful misreadings, as a source of theory.⁵ Either way, it seems that Camera Lucida is of limited value in the history or criticism of photography. These criticisms, I take it, comprise a general consensus: yet at the same time, writers continue to pluck the punctum out of the text in order to speak about private experience. There is perhaps no better evidence of the disarray of contemporary theorizing on photography than the fact that a book as problematic as Camera Lucida is still read and cited as a source of insights about photography.

Michael Fried's "Barthes's Punctum" is the kind of strong reading that Camera Lucida requires if it is going to be used as a source for theorizing about photography, rather than an occasion for reflecting on the impossibility of building theories around personal experiences of certain photographs, or as an opportunity to poach a poetic concept. I expect Fried's reading will put a stop to some of the looser uses of the punctum, not by demonstrating how strange Camera Lucida is (that doesn't seem to have helped), but by making explicit what is entailed in subscribing to the punctum. For theory-building purposes Fried is right to stress that "the detail that strikes as a punctum could not do so had it been intended as such by the photographer" (Michael Fried, "Barthes's Punctum," Critical Inquiry 31 [Spring 2005]: 000), precisely because the point is arguable, and pulls the punctum out of its solipsistic private-language doldrums. Fried links the claim about the absence of intentionality to what he calls the antitheatrical tradition, and, via a reading by Stephen Bann, to a distinction made by Diderot between "seeing" and "being shown." "The punctum," Fried glosses, "is seen by Barthes but not because it has been shown to him by the photographer, for whom, literally, it does not exist." Regarding the second half of Camera Lucida, in which the passage of time is proposed as a punctum, Fried points out that "the sense of something being past, being historical, cannot be perceived by the photographer or indeed by anyone else in the present": hence the punctum understood as a sign of the passage of time is another "a guarantor of antitheatricality" and a parallel instance of seeing without being shown ("BP," p. 000).

"Barthes's Punctum" is not an easy text to critique. It would be unhelpful, I think, to criticize the reading of Barthes for being narrow and selective—Fried knows it is both, and has good reasons. Nor would it be fruitful to characterize the essay as rescue mission directed at just a brief passage ("one short section . . . comprising a single page of print" ["BP," p. 000]) in a text that is otherwise irrecoverable for theory. (How else could it be rescued?) "Barthes's Punctum" is a necessary reading in the specific sense that it is impelled by the thematic of antitheatricality that Fried has explored over the last twenty-five years, and it is supported by examples that have much richer contexts elsewhere.⁶ I take it that his work on the antitheatrical tradition is both fundamental and indispensable for the interpretation of modernism, so it wouldn't be sensible to approach "Barthes's Punctum" as if it could open the question of antitheatricality or its potential applications in the present—those themes are in the books, not in this essay.⁷

"Barthes's Punctum" is a part of a work in progress on photography, and I imagine that when the book appears, much of the reaction will center on the jump in Fried's interests from painting to photography. It's not just that Fried hasn't written much on photography (mainly a page-long footnote that hangs, anomalously, from a meditation on realism in Courbet's Realism), it's that modernist criticism has long been identified with claims about the specificity of media that would apparently prohibit the move in "Barthes's Punctum."⁸ I do not think either of these points should be worrisome. The footnote in Courbet's Realism contains several of the arguments in "Barthes's Punctum," including the parallel with Chardin and the crucial stress on Barthes's idea that the photographer can "not not photograph the partial object at the same time as the total object." The note is appended to a consideration of the properties of Realism in Courbet's painting, and the passage leading up to the note concludes: "the starkness of the opposition between Realism and photography points to their rootedness in the same historical conjuncture." Thus the genealogical tree that could present photography as a modernist art form "entangled with a problem of theatricality" was already in place in Courbet's Realism. The second point, concerning the specificity of media, may seem troublesome because in "Barthes's Punctum" Fried applies several of the same criteria to photography as he has applied to painting, apparently breaching the medium-specificity that has been central to modernist criticism since Greenberg. But it is one thing to claim that some recent "ambitious photography increasingly has claimed for itself the scale and so to speak the address of abstract painting" ("BP," p. 000)—a claim I'll consider at the end of this response—and another to try "to learn at all costs what Photography [is] ‘in itself,'" as Barthes says (CL, p. 3). Fried doesn't write about photography because it is "faced with the task of defeating theater in and through the punctum" ("BP," p. 000) but, I take it, in order to justify the importance of some contemporary photographic practices by demonstrating their connections with themes that, as he says in the footnote in Courbet's Realism, were first articulated "around the middle of the eighteenth century." If this appears as a betrayal of modernist faith in media-specificity, I wonder if that isn't because modernist criticism has a structural inability to determine what constitutes the specificity of a medium. Medium- specificity is either presented as a given—an inherent set of properties comprising "all that [is] unique in the nature"⁹ of each medium—or else as an historical fable, now jettisoned in the "age of the post-medium condition."¹⁰ "Barthes's Punctum" steps around that inbuilt and unproductive choice by paying attention to the pressure exerted on the present by the historically specific forms media have taken, while at the same time acknowledging the possibility that media co-opt properties from one another, thereby rearranging, blurring, or simply switching their historical roles.

Given all that, it seems to me that the most interesting questions to be asked about "Barthes's Punctum" only appear when its reading of the punctum is accepted. What I want to know then is: What kind of photography does the newly theorized punctum give us? And—a separate question—what kind of photography does "Barthes's Punctum" give us? Here I'll propose features and kinds of photographs that are compatible with the punctum as it is read in "Barthes's Punctum," but are not countenanced in the essay. These new features and examples bring Fried's reading into areas that are, I take it, not of interest to him—areas that, as we know from Camera Lucida, were also of no interest to Barthes. The point here is to ask how strictly the reading in "Barthes's Punctum" constrains the punctum, and where antitheatricality and the punctum can go when it comes to current photography. The answer to the latter question is: much further than either Fried or Barthes wants them to go.

For both Camera Lucida and "Barthes's Punctum," much depends on what is made of phenomenology. Toward the end of a series of acknowledgements that Barthes's approach "is nothing if not personal," Fried remarks that Barthes's sense of phenomenology "is one that, unlike classical phenomenology, attaches primary importance to desire and mourning" ("BP," pp. 000, 000). Barthes only mentions phenomenology twice in Camera Lucida: once in a passage Fried quotes, in which Barthes acknowledges that his phenomenology is "vague, casual, even cynical," and again in section 14, in the course of expositing photographic "shock."

"`Shock'" (always, in section 14, in quotation marks), Barthes says, is "quite different from the punctum" in that "`shock'" is less about "traumatizing" than revealing what had been hidden. "`Shock'" comes in five flavors, which Barthes calls "`surprises,'" also in quotation marks.¹¹ The third "`surprise'" is "prowess": "For fifty years, Harold D. Edgerton has photographed the explosion of a drop of milk, to the millionth of a second."¹² The only other comment Barthes has about prowess is in a parenthesis appended to this sentence: "(little need to admit that this kind of photography neither touches nor even interests me: I am too much of a phenomenologist to like anything but appearances to my own measure)" (CL, p. 33). In that one remark Barthes compresses a massive rejection—so much of photography has to do with appearances incommensurate with human measure—with a significant distortion of the concept of phenomenology. This is not "vague" or "casual" phenomenology, if only because it could be defended by appealing to Merleau-Ponty's own rejection of scientific epistemology and his interest in embodied knowledge of the world. I assume Barthes would not want to follow that line of argument because it is also the case that a photograph of milk droplets can, in a reading wholly dependent on Merleau-Ponty, elicit a strongly embodied reaction. How, in a phenomenological account, could a milk drop fail to be seen as if it were human-scaled? Indeed, what can be apprehended—in Kant's sense of that term, in which it is opposed to what can be comprehended—without being taken as an image made to our own measure?

I am not fond of this parenthesis of Barthes's, because the lack of argument on a point so crucial to the book's axial theme of embodied experience can only function, it seems to me, as a sign that a region of photography is being hastily and arbitrarily closed off. Photography is domestic and domesticated in Camera Lucida, because it is identified with what is called vernacular photography: Little Italy, Idiot Children in an Institution, Savorgnan de Brazza (CL, pp. 46, 50, 52). Barthes is attracted to pictures of race, of mental debilitation, of romantically lost places and people, and above all to pictures of what he thinks are unusual costumes, demeanours, and faces.¹³ But what if even vernacular photography included something less human, less immediately freighted with national, social, ethnic, and familial significance, less perfectly suited to Barthes's own family history? What if the concerted search for personal engagement that impels Barthes in Camera Lucida is better described as an elaborate way of failing to find a more difficult sense of photography?

Consider this thought experiment: imagine the Winter Garden photograph—as good an exemplar of vernacular photography as any, especially since it exists only in the collective imagination of Barthes's readers—and take your eyes off the central figures. Look instead, in your mind's eye, at the things that surround the children. You will see almost nothing. A bit of railing on a "little wooden bridge" and a "glassed-in conservatory" is all the picture contains, provided your imagination does not add anything Barthes doesn't mention (CL, p. 67). (When I tried this, I found my memory added some details of their clothing, and drooping plants on either side.) The absence of visual incident makes sense, because for Barthes the photograph exists only as a way to think about his mother; and by extension, in Barthes's account photographs are opportunities to meditate on such things as the passage of time and the modulations of memory, loss, and pain.

If I perform this same exercise with any actual family snapshot, something quite different happens: I become aware of half-occluded pieces of furniture, I notice a mess of foliage outside a window, I see the overexposed glare of a white wall—all the particular matter of the world that was not the point of the photograph. Such details can be hard to look at because they will not adhere to my thoughts, which remain bent on the photograph's subject, the one the photograph was meant to pluck out of the matrix in which it is, in fact, embedded. Those nearly unseeable pieces and forms, shapes and parts are the on-and-on of the world, its apparently unending supply of usually dull and sometimes uninterpretable stuff, and for me they are proof of a difference between whatever photography is and the agendas of vernacular photography in particular.

Or take an example reproduced in Camera Lucida, Alexander Gardner's Portrait of Lewis Payne, the one of whom Barthes says "the punctum is: he is going to die"

Figure 1. Alexander Gardner, Lewis Payne. (Apr. 1865). From civilwarphotos.net

(fig. 1; CL, p. 96). All Barthes says of the background is that Gardner photographed Lewis "in his cell." The wall is apparently two iron sheets, welded together with enormous rivets. The photograph was taken not in Lewis's cell, but in the Navy Yard in Washington, so it is possible Payne was posed on front of a ship: but it goes without saying that even discovering the exact location would not remove the mass of apparently unimportant detail that is the photograph, apart from the small portion that depicts the "handsome" boy (CL, p. 96).¹⁴ This is—just to be literal about it—an image of scratches and scrapes on iron sheets, with a figure interposed.

These ordinarily unnoticed forms can prick me, as the punctum is supposed to do. But more often they thrive in my peripheral vision like an infestation. They resist interpretation not so much because they are irrelevant to the production and dissemination of photographs, and certainly not because they are likely to be fragmentary and therefore illegible, but mainly because they tend to be boring: they are only available to be seen because the photograph has placed them there. In Gardner's photograph I find the scratches—including those on the print itself—more absorbing than the "handsome" boy, more "wounding" and "bruising" (to use two of Barthes's words) than his shiny manacles or his prison-issue woollen shirt and pants, and certainly more "poignant" than his fixed, off-center stare (CL, p. 27). What is this stuff if not the texture of antitheatrical meaning in vernacular photography, seldom "intended as such by the photographer" and rarely even noticed by viewers?¹⁵

1I would argue two things about uses of the index and indexicality in photography theory. First, such readings have made use of a very selective reading of Peirce's semeiotic, ignoring for example the interdependence of all three kinds of signs; their division into trichotomies according to function (what Peirce calls firsts, seconds, and thirds); the fact that icon, index, and symbol are taken in relation to objects and that two other divisions name signs in relations to themselves and to what Peirce calls interpretants; and their ramification into divisions and even 59,049 cases. (In short: such readings are so abbreviated that it becomes unclear in what sense they are citations of Peirce's semeiotic at all.) Second, uses of the index in photography theory have tended to identify the indexicality with cause and effect, so that the work indexicality has been made to do could often have been done without any reference to Peirce. These points are discussed in my "What Does Peirce's Sign System Have to Say to Art History?" Culture, Theory, and Critique 44, no. 1 (2003): 5–22.
2 Roland Barthes, Camera Lucida, trans. Richard Howard (New York, 1981), p. 28; hereafter abbreviated CL.
3See Jacques Derrida, "The Deaths of Roland Barthes," The Work of Mourning, ed. Pascale-Anne Brault and Michael Naas (Chicago, 2001), pp. 31–67; Margaret Olin, "Touching Photographs: Roland Barthes's ‘Mistaken' Identification," Representations, no. 80 (Fall 2002): 99–118; and Graham Allen, Roland Barthes (London, 2003), chap. 9, "Camera Lucida: The Impossible Text," pp. 125–32. Allen argues very directly that Camera Lucida "blends the discourse or language of method (theory) with a wholly personal discourse (of mourning) and thus unsettles and disturbs the very results it seems to present" (Allen, Roland Barthes, pp. 125–26). Fried cites Olin's essay, noting that she doubts the existence of the Winter Garden photograph, but does not comment on her argument that Barthes's desire overwhelmed his theory, compelling him to construct the photograph out of parts of existing photographs. I take it the essential point is not that the photograph must decisively never have existed but that "Barthes," the author of Camera Lucida, needs to "use photography to satisfy his desire to possess or commune with his mother," and that the desire displaces the punctum, "like an alibi" (Olin, "Touching Photographs," pp. 115, 112). It seems to me that in Olin's essay the punctum in Camera Lucida is too unreliable to contribute to a theory of photography. Another text that reads Barthes's book as an exercise in self-subversion is Stamos Metzidakis, "Barthesian Discourse: Having Your Cake and Eating It Too," Romanic Review, 91, no. 3 (2000): 335–47.
4Maynard says Camera Lucida is not "a sustained account of photographs" but is "actually reductive to the subjects photographed, taken substantively: usually people or details of them and their attire" (Patrick Maynard, The Engine of Visualization: Thinking through Photography [Ithaca, N.Y., 1997], p. 13). A similar argument regarding Barthes's use of photography to make unrelated points is made in Jean-Michel Rabaté, introduction, to Writing the Image after Roland Barthes, ed. Rabaté (Philadelphia, Pa., 1997), pp. 1–16.
5Nancy Shawcross, Roland Barthes on Photography: The Critical Tradition in Perspective (Gainesville, Fla., 1997), pp. 67–85. Barthes introduces the "third form" in The Rustle of Language, trans. Richard Howard (New York, 1986), p. 281.
6Let me mention and dispense with what I think may be an objection to this equation of the claim that the punctum is unnoticed at the time of the making of the photograph, and the function of "being shown" in the antitheatrical tradition. The case of photography, so it might be said, is different from painting, where the signs of the antitheatrical thematic—such as, in Fried's examples, the open drawer in Chardin's The Card Castle or the torn jacket in Soap Bubbles ("BP," p. 000 n. 17)—are placed in the paintings by the painters. When a photographer inadvertently includes a feature that will figure, for some future viewer, as a punctum, it is merely because the photographer "cannot not" photograph that feature. But Fried intends only a parallel of the appearance of not having been shown, and he emphasizes that Barthes "goes well beyond anything to be found in Diderot or for that matter any eighteenth- or nineteenth-century critic or theorist" by insisting that the photograph "carry within it a kind of ontological guarantee that it was not intended to be [antitheatrical] by the photographer" ("BP," p. 000).
7Fried's work, I think, is exemplary of modernism and for modernism, which is what I mean when I say that the reading in "Barthes's Punctum" is necessary. I discuss Fried's modernism at length in The Master Narratives and Their Discontents (forthcoming); I have also discussed his art criticism (especially in regard to the crucial difference between having a claim, a position, and a stance) in What Happened to Art Criticism? (Chicago, 2003), pp. 65–77; and I have explored the close relation between his forms of narrative address and the claims he makes in my book Our Beautiful, Dry, and Distant Texts: Art History as Writing (New York, 2000), pp. 246–52.
8See Fried, Courbet's Realism (Chicago, 1990), pp. 282–83. The footnote is anomalous in that the book is formatted with endnotes rather than footnotes, with only three exceptions, of which this is the longest. It is an asterisked footnote, a full page long, less than seven pages before the end of the book—a genuine compositional anomaly. I take that as an indication that even though the logic is consistent between the footnote and the context in Courbet's Realism, the historical continuity between realism (in painting) and photography remains troublesome. I thank Joel Snyder for alerting me to this note, which I'd forgotten.
9Clement Greenberg, "Modernist Painting," Clement Greenberg: The Collected Essays and Criticism, ed. John O'Brian (Chicago, 1993), 4:86.
10See Rosalind Krauss, "A Voyage on the North Sea": Art in the Age of the Post-Medium Condition (London, 1999).
11First surprise: the "rare"—a man "with two heads, woman with three breasts, child with a tail, etc.: all smiling." Second surprise: the "numen of historical painting," where we are shown the moment that "the normal eye cannot arrest": Baron Gros's Plague-House at Jaffa, where "Bonaparte has just touched the plague victims" and his hand withdraws. This second surprise is "habitual to Painting," but a "surprise" when it appears in photography. Fourth surprise: the "contortions of technique: superimpressions, anamorphoses." Fifth surprise: "the trouvaille or lucky find": "an emir in native costume on skis." Barthes does not approve of these "surprises" because they are orchestrated, and therefore, as Fried emphasizes, shown to the viewer instead of lying unseen in the images, waiting to be discovered. Barthes says that relying on "surprise" makes it necessary, "by a familiar reversal," to find the "surprise" in all photography, in photography itself. Instead of searching out "surprises," amateur photographers say that whatever odds and ends they photograph are automatically "notable" (CL, pp. 32–33).
12Barthes copied this from a popular magazine, which reported the facts inaccurately: Edgerton's milk-drop photos were made between 1932 and 1957. See Harold Edgerton, Stopping Time: The Photographs of Harold Edgerton, ed. Gus Kayafas (New York, 1987), p. 126. Kayafas tells me that Edgerton produced about 20,000 negatives of milk drops and destroyed all but two dozen or so; see Kayafas, letter to the author, 1999.
13I use vernacular photography here to denote a set of practices that include portraiture, journalism, street photography, and the snapshot. See Douglas Nickel, "Roland Barthes and the Snapshot," History of Photography 24, no. 3 (2000): 236–39. On Barthes's choice of images, see also Olin, "Touching Photographs."
14The argument I am making here is parallel to one made by Maynard, Engine of Visualization, pp. 29–33, in reference to scratches and doodles in a Walker Evans photograph, except that Maynard is not valorizing photography's incidental marks, but considering it as a "surface-marking" technology (p. 34).
15This is also where Barthes's equation of photographs with reproductions of photographs becomes especially significant. Fried puts this quite accurately: "for Barthes, being alone with a photograph seems above all to have meant being alone with the reproduction of a photograph in a book or magazine" ("BP," p. 000 n. 32). Fried goes on to talk about Ruskin and reading, and his observation about Barthes's reliance on magazines is nearly an argument that for Barthes, looking at photographs is reading. Nothing is lost in reproduction as far as Barthes's theory is concerned. In "Barthes's Punctum" the physical presence of photographs is important, but not such things as the inevitable gloss of a photograph's water-resistant surface, the slight depth of its layers of grain, and the heft of its paper backing (or the translucency and thickness of the plastic support, in the case of a light box). The stuff that comprises photographs gets a bit lost, even though it is not necessarily a sign of theatrical address, and even though it is not irrelevant in large-scale installations like Struth's or Wall's.