The Brooklyn Rail

OCT 2019

All Issues
OCT 2019 Issue
Editor's Message Guest Critic

Reality by Chance

As one event follows another, we are the products and the causes of chance effects. We live a personal narrative of events we cannot predict.

Portrait of Richard Shiff, pencil on paper by Phong Bui.
Portrait of Richard Shiff, pencil on paper by Phong Bui.


When a second observation connects to a first, it alters the sense. Whatever registers in consciousness, registers a difference—not only its own but also a difference in what came before. Human experience, C. S. Peirce wrote, is “the compulsion ... to think otherwise than we have been thinking.”1 X, if followed by Z, does not remain the same, nor does it convey the same sense that it assumes when followed by Y. Life moves on, changing. Because events continue to occur, their implications and our extended interpretations also change. One experience affects another, as if each were an eruption of reality from within a dreamlike, unchanging order—as if each were invading our dream as an alien force. Reality intervenes. It changes the sense of things.

With or without external stimulation, thoughts are mental events that transform a person’s state of mind, as if the mind were pliant matter and subject to physical change. To put it another way: psychic activity assumes a physical form. The mind is a medium. We can imagine it as a surface that acquires projected images, each of which leaves a trace to be coordinated with all others. A simple mind would be analogous to a drawing composed only of lines or even a single line. The current extension of the line would require adjustments to its previous course to allow the whole to assume a compositional order with a potential for rationality, a hedge against future disruption. Reflective consciousness works to make sense—to dream a coherent meaning, giving order and structure to the projections of reality that continue to accumulate.

Once a cultural order (an ideology, an imaginary) is securely in place, reality becomes a distraction. Intervening, it causes involuntary passage from one sensation or thought to another. We may never know whether a second thought that follows a first strikes us because it feels right as a response to the distraction, or is itself the distraction, the interruption. If the latter, the connection between two observations appears that much more adventitious: first this sensation, now that, with the two associated solely by their sequence, their temporal succession, their chance contiguity. Are rationally entailed thoughts of a different kind? I like to think that they can be as distracting as unexpected occurrences. We never know where our logic leads us. Our most reasonable thoughts are elements of a more convention-bound order than utter randomness, yet still cause disruption. Perhaps logical thoughts are only moderately distracting, whereas abrupt changes of mind (instances of metanoia) are distracting in the extreme. Distraction, like reason, is a matter of degree.

The course of life distracts us. It may be that the connection between two sensory or cognitive moments has no ultimate significance other than articulating the history of the consciousness that experiences it. For this living consciousness, the connection, no matter how evasive it is, defines reality, the moment of greatest awareness: “We are conscious of a dividing instant,” Peirce noted, “with its difference on the two sides.”2 Like a linear boundary, the “dividing instant” between two thoughts or sensations becomes both connection and separation. A connection doubles as a disturbance or break.

Any two successive observations—first this, then that—remain associated in their factual presence, though their existence need acquire no higher sense of regularity or logical entailment. Their sequence and their very existence could have been otherwise. Reality is not a matter of how things must be, but of how they happen to be. “The principle of evolution,” Peirce argued, “requires no extraneous cause, since the tendency to growth can be supposed itself to have grown from an infinitesimal germ accidentally started.”3 Like a distraction, the presence of reality, its interruptive force, comes in degrees: “If all things are continuous”—connected as they pass through their various states—“the universe must be undergoing a continuous growth from non-existence”—undifferentiated chaos—“to existence”—a world of differentiated entities in circumstantial relationship. Peirce concludes: “There is no difficulty in conceiving existence as a matter of degree. The reality of things consists in their persistent forcing themselves upon our recognition. If a thing has no such persistence, it is a mere dream.”4

The real

The observations that drive a logical argument to a conclusion are analogous to those that structure a fictional narrative, directing its course. A narrative must make sense. It is a dream that, in contrast to the kind that interrupts sleep, reaches a satisfying resolution. If well-constructed and efficient, a narrative includes no unproductive elements; each of its descriptive features contributes to the advancement of the narrative order. Its dreamlike logical perfection indicates its fictiveness. Yet each of its incidental details may be plausibly real. As in the case of either material or intellectual abstractions—paintings, sculptures, photographs, texts (including both logical arguments and fictional narratives)—the constituent elements (material marks, linguistic syntagms) possess a high degree of reality, but the composite whole is an imaginary totality, a product of creative fantasy.

In 1968, Roland Barthes composed “L’effet de réel” (“The Reality Effect”), arguing that “the general structure of narrative ... appears as essentially predictive.”5 Retrospectively, a reader perceives how each element prepares for some other, connecting with it. Any detail that fails to connect might justifiably be regarded as an error of composition. Yet Barthes’s “L’effet de réel” is a meditation on the apparent anomalies and disconnections, the disjunctive, superfluous elements—factors of “narrative luxury”—that never acquire a proleptic function in the story as it evolves.6 These elements seem to do nothing, as if they were (to invoke Peirce) accidental starts that went nowhere, beginnings lacking ends. Their only connection is that of placement next to this or that. Each is an interruption, if only minor.

Barthes propelled his argument with a series of implicit questions met by a series of hypotheses. Maintaining a degree of digital order—a structuralist’s analytical economy of yes or no, presence or absence—he avoided speculative byways. Having none of the superfluity it addresses, “L’effet de réel” maintains a disciplined focus (a factor contributing to its influence: readily digested, a little Barthes goes a long way).7 With dispatch, Barthes reached his counterintuitive conclusion: “The very absence of the signified, to the advantage of the referent alone”—the inclusion of a descriptive reference that projects no ulterior meaning or purpose, that, with respect to the narrative, has no independent signification (hence, Barthes’s absent “signified”)—“becomes the very signifier of realism.”8 Stated simply, meaninglessness acquires meaning. The element of no importance lends to everything else the importance of credence.

By denoting a seemingly pointless element of a scene within a narrative, alluding to an object of no apparent narratological function, a writer succeeds in connoting reality. Any stray, random, wasteful reference of this kind—a reference that fails to refer to anything beyond itself—produces what Barthes calls the “reality effect.” Perhaps within reality itself, meaninglessness remains meaningless, like one grain of sand in relation to another. The situation changes only when happenstance intervenes: the grain of sand in your shoe, an irritant, an interruption, becomes different from all the sand on the beach. It may cause you to reflect on sand.

Within the fictions of representation, meaninglessness acquires meaning even without the intervention of chance. Meaninglessness is chance, for if chance conveyed meaning, it would seem purposeful and intended, hardly a matter of happenstance (perhaps it would be chance to a lesser degree). Yet every discrete element, whatever its character, establishes a referential ground: “reality” is its default function. The element that makes no difference in the developing plot or moral of a story is the one that anchors the entirety of the text in the real. This is its rhetorical effect: “l’effet de réel.” Analyzing literary language from the inside out—allowing language to “write itself” (as opposed to writing out thoughts that somehow exist elsewhere)—Barthes realized that language generates its effects by its own play of differences, and what differs from an ordered fantasy or dream is the isolated bit of reality, the grain of sand severed from its connection to the order of the beach.9 By the digital logic of yes and no, if there is meaning, there must also be no-meaning: “The pure and simple ‘representation’ of the ‘real,’ the naked relation of ‘what is’ (or has been) thus appears as a resistance to meaning.”10 Meaning is a dream. Reality is chance and has no meaning.11 It remains in want of its dream.

Barthes drew his initial example from the opening of Gustave Flaubert’s story “Un coeur simple,” where a barometer is identified among the disparate contents of a room, as if it merited attention: “An old piano supported, under a barometer, a pyramidal heap of boxes and cartons.”12 Barthes comments: “No purpose seems to justify reference to the barometer, an object neither incongruous nor significant, and therefore not participating, at first glance, in the order of the notable.”13 Details that fit the narrative, as well as details that pointedly do not, would be equally “notable,” attracting and meriting attention. But here, with the barometer, what the writer notes appears unworthy of being noted for either of the two reasons offered (congruity and a contrary incongruity). Barthes is less concerned to ask why an author as painstaking as Flaubert mentions the presence of an insignificant object, and more concerned to wonder what effect its appearance has on a reader. Several pages onward in Barthes’s essay, a reader has gained a degree of self-awareness, benefitting from the critic’s second pass: “Eliminated from the realist speech-act as a signified of denotation, the ‘real’ returns to it as a signified of connotation; for just when these details are reputed to denote the real directly, all that they do—without saying so—is signify it; Flaubert’s barometer [belongs to] the category of ‘the real’ (and not its contingent contents) which is then signified.”14

By metonymy, Flaubert’s arbitrarily selected bit of reality infects every other detail of the narrative. The superfluous element, the signifier of the real, a thing of no particular consequence, stands in for any other thing and all other things. Its reality is not only infectious but continuous: it remains in place, as if its meaning were still to develop. The sole property of the barometer, relevant to the narrative context, is its existence. As Peirce had claimed, “existence [is] a matter of degree. The reality of things consists in their persistent forcing themselves upon our recognition.”15 To read Flaubert is to become aware of the barometer; yet, according to Barthes’s analysis, the singular barometer in the room central to Flaubert’s story—ironically, an object designed to index external conditions—represents real existence in its utter neutrality, its presence to itself. The barometer tends to pass unnoticed because it lacks all import. Perhaps its message (“this is real, and so is the rest”) remains subliminal—connected but not connected.

Insignificant stretches

Observations connect to observations. Using Flaubert as his initial example, Barthes, by the end of his brief essay, has established the cause of the “reality effect.” What brings logical entailment to his argument? What constitutes connection? Along the way, he asks: “Is everything in narrative significant, and if not, if insignificant stretches subsist in the narrative syntagm, what is ultimately, so to speak, the significance of this insignificance?”16 Readers familiar with Barthes will recognize the form of this question as consistent with his characteristic train of thought. The negative of a positive—insignificance, as opposed to significance—is being investigated under suspicion that it conveys meaning (it signifies) beyond simple negation, that it extends the range of possible meanings rather than marking a termination at the null point of meaning, its degree-zero of existence. If the negative of a positive acquires the character of a positive, can oppositional dualities ever generate reliable discursive stability? If positives no longer eliminate their negative antitheses from consideration, anything is possible. In this respect, Barthes’s analysis undermines the dualistic, yes-or-no foundation of much of his structuralist method. A creative thinker follows language where it leads, and here chance returns triumphant. A digital yes-or-no, presence-or-absence, may not be the absolute division it purports to be. Barthes seems to acknowledge that yes and no—as well as any other polar dualities—are somehow connected and capable of exchanging properties. With regard to the magnetic poles of the physical planet, reversal would be traumatic. But oppositional extremes commonly reverse in the course of thinking, sensing, and feeling. The form of the exchange is often chiasmic: a sense of pain becomes pleasure, just as pleasure becomes pain. As with more “logical” patterns of thought, sometimes reversal hardly interrupts the flow of consciousness. And life moves on.

Barthes was well prepared to consider the paradox of the “reality effect.” Several years previously, he identified what he called a “photographic paradox.” To explain it, he developed a similar rhetoric of exchange, reversal, and inversion: “The paradox is that the connoted (or coded) message develops here [in a photographic image] from a message without a code.”17 In other words, the connotation of photographic realism depends on the presumed denotative perfection of this medium. The image that lacks a code, that requires no translation, is nevertheless coded as “this is reality.” Because our initial thought is “denotation, 100 per cent, connotation, zero,” denotation is itself connoted by any ordinary photograph—all the more so, by the press photographs that Barthes used as his primary examples, since these were often the products of anonymous agents who had little incentive to aestheticize or otherwise doctor the image.18 The connotative value of denotation corresponds to the significance of insignificance. To include superfluous, insignificant details of description within a narrative causes it to seem photographic, for an untouched photograph eliminates nothing from its view. These analytical observations are very Barthes-like.

“Insignificant stretches”: I take this phrase from a translation readily available for Anglophone readers.19 Barthes’s French word for stretches is plages—he writes of “quelques plages insignifiantes.”20 Plage refers to an undifferentiated but limited period of time or an extent of surface area, including any area of land (plage as zone délimitée), and, in common usage, a stretch of exposed, open land along a body of water: in English, a beach; in French, une plage (alternatively, une grève). To translate plages as beaches would disturb the contextual order of “L’effet de réel.” It would convert Barthes’s general application of this term (undifferentiated extent) to too specific a usage. The conversion would introduce metaphor, or perhaps catachresis (if the word plage signifies “beach” only because an expanse of space is sufficiently beach-like to transfer the name for a second proper use).

In the context of typical issues of textual analysis, it would be misleading to claim that passages of gratuitous description have properties specific to beaches. These descriptive elements are merely stretches—of nothing in particular and between no particular locations and no particular moments. Their “reality effect” can erupt at any stage of the narrative. Because a stretch of intervening reality can extend for any distance or any time, it does not “stretch” in the physical sense of making an effort, extending itself. At least initially, it is more of a lapse than a crux. Its action seems passive, unwilled. Gratuitous effects of the real merely arise. Regarded in this manner, there may yet be something beach-like about these representational markers of reality. Passages of insignificant text, units of descriptive language that fail to advance or modify a narrative, resemble the constitution of beaches, which are stretches (étendues) of land that contain within themselves undifferentiated stretches—the innumerable elements of sand, gravel, or pebble, sometimes known geologically as shingle. If we were to divide a beach into segments, each part would look essentially the same as the whole as well as like the other parts. This is why the parts of a beach assume no meaning; they make no difference. The English word beach has a metonymic derivation, from constituent part to general whole: it first designated the shingle that covers a shoreline area; eventually, it referred to the area itself, a more abstract entity.21 The etymology of the English word appears to invert the direction of its French counterpart, in which a specific instance of a general type, plage, borrows its name from the general abstraction used to designate all such forms of extension, including arrays as abstract as ranges of numbers.

Barthes notes that “‘useless details’ ... seem inevitable: every narrative, at least every Western narrative of the ordinary sort nowadays, possesses a certain number.”22 A single insignificant detail affects the entirety of a narrative, ensuring that we never escape reality in our fictions. The meaning of a detail—its potential to function in some respect, even though structurally and narratologically “useless”—falls to the judgment of a reader or viewer. This fact raises the possibility of indeterminate cases of significance and insignificance over which critics would fail to agree: hence, the practice of literary explication, commentary, and criticism itself. Whether a particular detail does no more than serve the reality effect of a representation is hard to specify. An interpreter can always imagine an obscure narrative function for each insignificant element, as if to protect the fictive dream from interventions of reality. As imaginative as he was analytical, Barthes might insist on pursuing the dream in the opposite way: the only function of certain anomalous elements in the narrative order would be to validate the literary dream by connecting it to bits of secure reality.

Pea stone

An intriguing instance of narrative indeterminacy appears in The Man Who Wasn’t There, a film written and directed by Joel and Ethan Coen, released in 2001. Among its quirky features, The Man Who Wasn’t There presents a trial lawyer who decides that the Heisenberg uncertainty principle (or principle of indeterminacy) can be used as a legal defense, designed to raise “reasonable doubt” in the minds of a jury as they evaluate witness accounts and physical evidence. The lawyer epitomizes the uncertainty principle this way: “Your looking changes things ... The more you look, the less you know.” 23 To paraphrase: judgments of the real rest only on chance; the crucial fact will remain outside whatever order you attempt to impose. But this ironic link to Barthes’s “reality effect” is not the motivation for my introducing The Man Who Wasn’t There.

The pictorial aspect of a technologically conventional film like The Man Who Wasn’t There is thoroughly photographic. This work neither resorts to hand-drawn or computer-drawn animation, nor have its directors admitted to any digital manipulation of imagery, though the cinematography is highly sophisticated (certain scenes appear to be composites of several tracking shots). According to Barthes, the photographic “analogon”—the point-to-point correspondence between a photograph and its model (analog, not digital)—presents a direct representation of reality even if, as in the case of filmic photography, the encompassing cinematic narrative is a fantasy: “In the photograph, since the denoted message is absolutely analogical—i.e., deprived of any recourse to a code, i.e., continuous—there is no need to look for the signifying units of the [primary] message.”24 The identification of what appears is obvious. It is as it looks.

Barthes elaborates on his notion, stating that analysis proves unnecessary if only because of its impossibility: “In front of a photograph, the feeling of ‘denotation,’ or if you prefer, of analogical plenitude, is so powerful that the description of a photograph à la lettre is impossible.”25 A word by word description would fail, because the separate bits would code the image, converting analog to digital and mitigating the sense of plenitude, which can have no code. A photograph is an image of all connection, no separation. A footnote to Barthes’s text distinguishes the selectiveness of hand-drawing from the automatism of photography: “To describe a [representational] drawing is easier, since it ultimately involves describing an already connoted structure, one worked up with a view to a coded signification.”26 Presumably, any representational drawing applies a code—characterizing, interpreting, and modifying its model. To the contrary, even when selectively foregrounding aspects of the image, photography captures so many denotative features that it persists in connoting reality-in-general.

If Barthes were alive in 2001 to view The Man Who Wasn’t There, he would appreciate its narrative complexity. Through several of its scenes, the film brings passive attention—and ultimately, an uncannily delayed notice—to a short stretch of insignificant land, a plage (in the general, abstract sense of this word). I refer to a residential driveway that leads from a public street to nowhere in particular, only to an unused garage. The driveway itself provides parking space for the homeowner’s automobile. Pictorially, it remains superfluous. As characters approach or leave the bungalow where essential segments of the narrative occur, they drive or walk over this piece of ground: this is only natural; this is reality. The several driveway scenes—all but the final one—are brief and merely transitional. Throughout most of the film, a viewer would surmise that the driveway is a manifestation of Barthes’s “analogical plenitude,” a feature of the unedited reality of the site chosen for filming. Residential streets in the United States typically have individual driveways of the type seen in the film. By the terms of the screenplay, the location is Santa Rosa, a town in California, stage-set as 1949. The property belongs to the primary characters in the narrative, Ed and Doris Crane. Their driveway is represented simply for what it is, as it is. To remove it from the picture would lend it significance by its absence, for its presence is expected. To retain it for the various scenes is to preserve its inherent insignificance.

But the specific composition of the driveway eventually assumes an unexpectedly active role. This tempts me to evaluate it for what Barthes regards as the “predictive” function of narrative details. The driveway consists of pea stone, a smooth, rounded type of gravel, occasionally found in a natural state on beaches. In mid-twentieth-century America, pea stone was commonly used for residential areas requiring some kind of paving; it substituted for more permanent but more expensive methods. The scale of filmic representation renders indeterminate the individual components of a pea stone driveway. Like grains of sand or gravel on a beach, the stones are too small to register within the broad scope of the cinematic view. Yet they register by other means. As is customary, this narrative film has a sound track. When a person walks over pea stone, or when a car passes over it, its surface generates a distinctive sound unlike that of coarser types of gravel. As an aspect of The Man Who Wasn’t There, this sound is faint, but definitely audible. An individual previously unfamiliar with pea stone will come to recognize its memorable sound from experiencing the film. Having no narrative significance, the inclusion of this specific auditory feature is analogous to the reality effect of Flaubert’s barometer, a thing among more significant things within a residential setting.

By comparison with the relatively spare, linear order of a literary narrative, photography—and even film, despite its strong narrative component—may seem to display an abundance of superfluous references to the real. Everything is included, nothing excluded from the camera’s view. Yet, one of the conceits of narrative cinema is to maximize the significance of incidental elements for the sake not only of the richness of the story but also the economy of its unfolding plot. General locations, specific background elements, the characters’ items of clothing: anything can be a candidate for retrospective significance in the course of the analogon-like recording of fictively composed scenes. As a film runs, just as in narrative literature, any particular cause of the “reality effect” is subject to become an active agent in the unfolding imaginative fiction—like a clue in film noir, initially unnoticed both by the characters and by viewers.

My previous reference to “an uncannily delayed notice” hinted that attention returns to the pea stone of the Crane driveway. Because its distinctive sound repeats in each scene in which the driveway becomes a factor, the sound itself has the potential to gain symbolic value, like a chime announcing a time of day or night. This sound announces that we, as attentive viewers, are returning to or leaving the Crane residence. It would be natural to expect repeated allusions to any feature or quality within the filmic presentation to join the main line of the narrative. At the least, the sound of pea stone ought to refer back to itself, altering the absoluteness of its presence—a first instance being modified by a second, a second by a third, a third by a fourth. When Barthes argued that “the general structure of narrative [is] predictive,” he contrasted realistic description: “It has no predictive mark; ‘analogical,’ its structure is purely summatory.”27 He was not referring to the coded description of pictures by words, but to verbal description at its most primitive and undeveloped, as if it were a list of things with no logical connectives, only those of position or sequence. Such description is its own conclusion. But it joins in a narrative fiction by producing the effect of the real.

The pea stone in The Man Who Wasn’t There—like the passive protagonist of the film, Ed Crane—is neither here nor there. Like Ed, the pea stone is buffeted by chance. Although designed for the driveway, it fails to remain in proper order. Individual pebbles stray into adjacent areas of lawn and walkway, scattered by automobile tires and shoes. This is the functional disadvantage of the material, its behavior under actual conditions. It becomes an issue in a crucial flashback within the film narrative—a flashback for the viewer of the film, whereas for the fictive character Ed, it is his dream as he regains consciousness after a serious automobile accident. The scene closes as Ed and Doris (in contrast to Ed, a proactive personality) silently face the knowledge of a homicide to which each has inadvertently contributed. In the spirit of Barthes, however, let me leave the homicide and the overarching plot developments aside, while pursuing the possible function of seemingly insignificant details. Barthes had pondered the “insignificant stretches” that appear within a text. What can the presence of a stretch of pea stone accomplish within the filmic narrative of The Man Who Wasn’t There? At the least, it motivates the scene in question, which amounts to Ed’s dreamlike recollection of his failure to communicate openly with Doris.

“Pain in the neck,” a door-to-door paving salesman says, as he enumerates the problems that pea stone causes the average homeowner. He addresses Ed, who—in the dreamlike flashback—is sitting on the front porch of his bungalow, awaiting Doris’s return. It is an odd hour, 4:40 in the afternoon, a subtle indication that discovery of the homicide has interfered with the normal work day. The salesman remarks that a piece of pea stone in the lawn, churned up by a lawn mower, is like “the odd sock”; and he feigns distress and frustration at the thought. What does an odd sock signify? A pair of socks is our everyday reality; a single, odd sock wants to tell the story of how it lost its connection to its mate. The odd sock loses more than its double; it loses its place in the dreamlike order of the world, becoming an anomaly, an annoyance. But which is more “real”—the single sock or the pair? When Ed, who lives a most ordinary life, is eventually tried for a second murder, one he did not commit, his lawyer defends him on the grounds that he has lost his place in the order of the universe. If he is guilty of anything, it is only of living in a world that refuses to accommodate his being. Yet it is Ed who represents reality. He embodies “Modern Man,” his lawyer says, as if invoking the comic ironies of Charlie Chaplin, the social theories of American liberals, and the European existentialism of 1949. Ed is guilty of failing to conform to the fantasy of flawless human behavior.

Perhaps the reasonable question to ask of the odd sock and the pair is this: Which will appear more “real,” attaining l’effet de réel? Is the situation of the out-of-order element, relentlessly subjected to chance, the typical reality? Or is it the situation of the in-order element—a reflection of history, society, and culture, of human affairs as they ought to be? The answer is indeterminate. Flaubert’s barometer was suited to its place in the home but had no place in the story. A pea stone driveway is analogously ordinary and perhaps has the potential to interrupt, just as reality so often does. If a person’s attention turns to the sound of pea stone, it can be distracting. Pea stone in the lawn is a different kind of interruption: equally ordinary in its way, the salesman would say, because things always go wrong with this material (“pain in the neck”). A homeowner finds pieces of pea stone where they should not be, but have every material reason to be. Chance puts them there. Yet Ed, in his passivity, will respond to the importuning salesman: “Never bothered me.” Retrospectively, we realize that we too were never “bothered” by the pea stone, despite hearing its sound several times in the course of the film. This incursion of reality created no distraction. In the flashback, however, the audible signs of the physical existence of this material resound, as it becomes—after the fact—predictive of what we eventually come to know. As Peirce might observe, its degree of reality has increased.

Meaning from elsewhere

In the previous section, I ventured a generalization: A photograph is an image of all connection, no separation. In many senses, both technical and interpretive, this is not the case. An image generated by conventional photographic emulsion breaks into discrete particles when sufficiently magnified by enlargement in the development stage or by projection; there is a physical limit to its resolution.28 And when interpreting a photograph, yes, we can say that it represents the scene of a particular event just as it appeared. But most likely, we recognize the event because we discern a certain number of identifying elements—signifying elements—from within a mass of indecipherable or inconsequential (insignificant) bits of tonal information.29 Yet, for Barthes, the remarkable feature of still photography and of other photographic media, such as film, is the capacity to connote connectedness, an “analogical plenitude,” which renders any photographic representation an image of all connection, even if not fully connected in a technical sense.

If devoid of “insignificant stretches,” a narrative (textual or filmic) would amount to a digital code in which the divisions between the elements operated only as connectives, never as separations. Visualize a video screen: its pixels abut one another. Like members of a crowd vying for attention, the units of code jostle and bump. In Flaubert’s “Un coeur simple,” the barometer is not inactive, but less aggressively active than its neighboring units. It still signifies, generically rather than particularly. It is reality, and perhaps also a delay in the message, a bit of textual filler. Every digit in a coded representation changes whatever meaning is otherwise already present; no meaning remains fixed, regardless of the order in which we consider the elements. To regard a text in this fashion—to stress its analog rather than digital nature—is to analyze it as if it were a picture. It may be more accurate to claim that a text has the properties of a picture than to make the more common claim that a picture has the properties of a text. This is why Barthes, a critical writer, analyzes press photography, and why he accepts the challenge of analyzing the pictorial art of Cy Twombly. Analyzing pictures is chancy, because the location of the significant elements within the image as a whole remains to be determined.

In 1979, Barthes published two rather similar essays on Twombly, and—to judge only by appearances—one either draws from the other, or each draws from its counterpart, exchanging references and phrasing, to and fro. To say that these two essays are similar is to note that Barthes presents the same understanding of Twombly twice over. The writings resort to the same examples and terminology, and a number of the sentences are identical. The final paragraphs of each differ by no more than a few words. Yet the two essays are dissimilar, differing rhetorically. One is quite expository; and the other, at least with respect to its fragmented, elliptical structure, is more poetic. In his capacity as theorist as well as writer, Barthes might qualify this distinction between exposition and poetry, just as he would wish to undermine a distinction between science and literature.30 He might claim that one essay was merely indulging in the poetry of, the rhetoric of ... exposition. Unadulterated exposition would use colorless language, that is, a rhetoric of the reality effect, the connotation of denotation (as in photography). Though there are degrees of color, even white is a color. With regard to rhetoric, even transparency colors a text.

The difference in tone in Barthes’s two essays on Twombly probably reflects the different destinations intended for his analysis. The more expository essay, known in English as “The Wisdom of Art,” was commissioned for Twombly’s New York retrospective at the Whitney Museum of American Art; this writing would have had to meet standards for accessibility by a general American readership.31 The more poetic and more elliptical essay is known in English by a colorless title that seems ironic in retrospect: “Cy Twombly: Works on Paper.” This title eliminates the troublesome Latin quotation used to identify the original French version, “Non multa sed multum” (not many but much), which refers to the contrast of quality to quantity.32 The substitution of everyday English for rarefied Latin is in this instance immaterial, because the Latin phrasing figures prominently in the text, whether English or French.

In fact, “Non multa sed multum,” the more poetic essay, was composed in 1976, having been commissioned by the editor of the comprehensive catalogue of Twombly drawings, Yvon Lambert, who had previously held two exhibitions of the artist at his Paris gallery. Barthes wrote the more expository—and, it seems, derivative—version for the Whitney Museum at the very end of 1978.33 The presence of Barthes within a catalogue, whether destined for a general audience or a select one, contributed the prestige of his name. Barthes himself was sensitive to names. In the Whitney essay, he addressed Twombly by his proper surname. But in the more playful, more rhetorical essay, the artist becomes TW, leaving it to the reader to decide how to interpret this odd sign, the combination of upper-case T, upper-case W. It might be two consonants or a single compound consonant, sounded as a variety of diphthong, a sound not heard in French as tw, but more or less in words such as trois (three) and toit (roof), and toi (you). As in the case of pea stone, the sound of this sign might signify more than its look.

Cy Twombly, Hero and Leandro, Part I, 1981–1984. Oil, crayon, and graphite on canvas. © Cy Twombly Foundation.

Academics familiar with Barthes will probably find nothing that he discerns in Twombly’s art to be out of character with the writer’s established interests and attitude toward interpretive reading and viewing. Consider Twombly’s painting Hero and Leandro, completed in 1984.34 It happens to date a few years later than the time of Barthes’s two essays, but its mythological theme accords with the writer’s emphasis on Mediterranean culture. This is the cultural-history aspect of Barthes’s analysis; in the more expository essay, he links Twombly not only to the Italian culture that this American expatriate had adopted, but also to the classicism of the ancient Mediterranean and Roman world and of the French baroque painter Nicolas Poussin, who installed himself in Italy for the sake of his art. In the more poetic essay, Barthes plays to a greater extent with the tension between pictorial marks and words. Hero and Leandro, like many of Twombly’s works, sets an idiosyncratically hand-written name against a relatively amorphous image that comes into greater focus because of the suggestiveness of the name.35

Handwriting has its own version of a reality effect. When irregular and idiosyncratic—in the way that those who write by hand often cultivate their cursive script—handwriting represents the degeneration of the signifying mark (a legible sign) into a condition of materiality that signifies little other than the trace of the individual hand (legible perhaps as an index of the hand, but less reliable as a symbolic sign). In view of the title as well as the name “Leandro,” cursively scrawled across the canvas, Twombly’s smears of paint become convincing as a watery wave, despite their incongruous coloration, red and green. The mythical Leandro (Leander) drowned in the sea. Even the lumbering slant of Twombly’s handwriting, as he inscribes the name, seems coordinated with the putative wave, which bears the weight of the name in its flow, as if this name—this physical imprint of lettering—were the body, the corpse, of Leandro. Barthes was sensitive to the way that our recognition of the name—as an announcement of a theme, the story of Hero and Leandro—brings a connotation of sea-water to the applications of red and green paint, which, as denotation, are merely abstract, material smears. In turn, the wave-like paint brings connotation to the denotative handwriting. The paint, along with the coarse character of the handwriting, poeticizes this inscription all the more, by incorporating it as an active pictorial element. The script becomes more than just the abstract designator of the person Leandro. It seems to float on the sea and risks being submerged in the wave. This is the nature of a constructed picture: every element becomes something more than what it is in itself, even the written name, which becomes more than a title. Materially, not merely semiotically, the inscription becomes a metonymic substitution for the body, borne by the wave.

Here I am, an interpreter after the fact, and neither Barthes nor Twombly. My interest centers in the way that Barthes views Twombly—whether Twombly’s art differs from, or is analogous to, both literary forms and photographic images. My experience with Twombly’s art is as longstanding as with Barthes’s writing, probably longer. So I come to this interpretive project lacking innocence. My mental attitude might actually correspond with the attitude that Barthes’s writing seems to project. More often than not, he was working on topics requested by others, claiming (according to one biographer) that he “rarely wrote an article without being asked to do so.”36 This was true in the case of his Twombly assignment. With a certain passivity—like the fictional Ed Crane—Barthes was opening himself to a chance sequence of critical practice (his own writings) in which the effect of one on the next would be unpredictable because no one was managing the narrative order. He would learn the significance of his actions only as he moved along.

I, too, prefer to receive writing assignments from others. It introduces a welcome factor of chance to my research. I am not the most consistent of scholars. My mind wanders from topic to topic, and if I immerse myself in reading Barthes one day, I might quickly turn to a different type of writer the following day. Roaming of this kind was characteristic of Barthes as well. Among his occupations, he was a journalist, accustomed to changing topics daily, responding to events, following coincidence and the coincidental requests of those who commissioned his work. For all his capacity to analyze a text word by word, à la lettre, when it came to the topic of Twombly, he was willing to interpret without knowing very much, as if he were responding to a deadline or were anxious to keep moving onward.

Cy Twombly, Untitled, 1970. House paint and crayon on canvas, 136 1/4 x 159 1/2 inches. The Menil Collection, Houston. Gift of the artist. © Menil Foundation, Inc. Photo: J. Littkemann.

The most insightful early writing on Twombly is Robert Pincus-Witten’s relatively short review of the artist’s career published in 1968, which the author expanded to a longer essay in 1974.37 During that period, Twombly was making large, rather abstract works in black and white. They look like script that lacks the words, or exercises of the arm and hand in preparation for writing. Barthes did not mention Pincus-Witten by name, but the American’s essay is the source of the quotations from Twombly that Barthes invoked. He also used Pincus-Witten’s information about Twombly’s practice of writing in the dark, the artist’s way of breaking free of habits of refinement. Twombly, Pincus-Witten had said, acquired a look of awkward left-handedness; this notion appears also in Barthes’s account, as does another of Pincus-Witten’s observations—Twombly as an “anti-colorist.”38 Had he not read Pincus-Witten, would Barthes have argued in this manner? Perhaps so, because these paths of interpretation had been applied to other modern artists; they are readily assimilated, given the drift of prevailing discourse. To cite one example, during the 1930s, the art historian and theorist Henri Focillon had made much of the aesthetic conceit of left-handedness.39 But I suspect that the connection between Pincus-Witten’s essay and Barthes’s own was quite specific. It may be that someone familiar with the existing writing on Twombly suggested Pincus-Witten as the most useful guide then available. So the encounter of Barthes with Pincus-Witten could have been an accident of a research process, but it could also have been staged by a third party.

On my part, I have never objected to allowing chance to generate my research. I have been pleased to let someone else pick the topic or ask me to study a body of art in which I might never have developed an interest. I became involved in working this way—perhaps I was acting on an inclination I already had—when I was an undergraduate student with access to the closed stacks of the central research library at Harvard University. I had few intellectual priorities, few hierarchies of intellectual value. I was academically open, naïve. From day to day, I would sit at the research desk of a different absent scholar, most of them candidates for a doctorate. I would skim through the selection of books that an advanced researcher had gathered. Even though I had no previous involvement with the particular topic, and often had to guess at what the topic might actually be, I was motivated by the feeling that I was gaining access to the cutting edge of knowledge, acquiring information that would not become public for at least several years to come. This information was denotative on the surface (often collections of primary documents) but connotative in a deeper respect because it had already been selected and arranged by an active intellect. I needed to guess at what the connotative message might be. From my perspective, each body of research was an odd sock, out of place in the universe. But so was I, not having a particular order to follow.

There was an additional factor of fascination in reading this way. The links between the various topics became spatial and temporal rather than logical. One thing followed another thing, altering its import. My connection to these topics depended on the chance distribution of the location of the individual study carrels, assigned not by subject but by availability. When one doctoral student finished a dissertation and left the university, another would occupy the vacated desk. And, from day to day, my choice in where to read would depend on the temporary absence of the person who had gathered the particular body of research. I was playing a knowledge game for which chance set the rules.

Twombly’s painted surfaces contain arrays of marks that look like nothing other than marks—ciphers, smudges, smears, jottings—little figurations, hard to gather into coherence. Do they follow only by chance? In the more expository of his two essays on the artist, Barthes refers to “the double problem of figuration and signification”—the fact that a sign has a figured or material form located in space and time, as well as a conceptual meaning, possibly timeless (although forever subject to change).40 Twombly makes both sides of this equation obvious, because he produces scrawls of marks, and often assigns a name, such as Bacchus or Orpheus, to the pictorial surface, whether as a title or cursively inscribed across the canvas as the image itself. Sometimes the smear or the scrawl and the name have the same likely referent. Perhaps the game is to make a discourse out of elements that are not consistently discursive, like making a narrative out of elements that advance the story as well as those that delay its denouement.

Barthes’s theoretical point is this: any interpreted meaning is likely to translate into words; otherwise, we would never have thought of the meaning in the first place. And the meaning will translate with an economy of verbal equivalence that figuration alone will rarely be able to match. Artists and critics commonly note that drawing a line is like thinking.41 But if the line fails to create a recognizable image, if it fails even to produce an indexical image, a clear trace of the action that produced the line, what would be the “thinking” of the line? Can a thought be a thought and yet mean nothing?

To take another angle on the question: Would we notice a mark as an independent mark within a visual field if it were not that we had decided it must signify something? We need to have decided that the mark has the look of signification, even if we have no inkling of what significance it might assume. We study the process, not the product. Barthes writes of his effort “quite quickly on slips of paper, to make certain scribbles ... I am imitating [Twombly’s] gesture ... I am not copying the product, but the producing.”42 Twombly’s gesture leaves “the impression of being ‘thrown’ ... to throw is an action in which are simultaneously inscribed an initial decision and a terminal indetermination: by throwing, I know what I am doing but I do not know what I am producing.”43 The meaning is yet to be determined. To me, this resembles a definition of reality. It is also a definition of chance.

Imagine looking at a piece of pea stone in a driveway—if we ever notice such particles individually. Or attempt to focus on a grain of sand on a beach. Can we regard the pebble or grain as a mark or sign of pea stone or sand? Left in their proper places, such elements signify little if anything. In a painting, however, tiny dots of ochre, set into an area next to a rather solid blue, might connote sand—not necessarily the individual grain, but the collective substance with the identity “sand.” This would become the meaning of each mark, given the pictorial context. Even something unidentified can have a verbalized identity, that of the anomaly, the odd sock, the irritating element out of order.

The nub

Concerning the tension between meaningless figuration and meaningful verbalization, Barthes argues: “It is never naïve ... to ask, in front of a canvas, what it represents.”44 We ask because we expect the production of an image to be motivated by a purpose, more than mere chance. We could be clever about an answer, holding that any painting represents the feelings of the artist; but this is unlikely to satisfy those who have never been indoctrinated with theories of aesthetic expression and the psychology of desire. Barthes makes his remark in the context of Twombly, because he realizes that people will wonder what the scrawling and smearing might signify in a straightforward way. Is its meaning aligned with its appearance? Does a smear represent smearing? If a canvas shows the handwritten name “Orpheus,” where is Orpheus? A name is not a portrait. A name can be a conceptual portrait, however; and it can be a representation by metonymy (substitute the abstraction of the name for the appearance of the person). Yet the typical viewer expects a representational, figured portrait to bear some degree of visual resemblance to its model—more metaphor than metonymy. Sooner or later, by one or the other of these rhetorical tropes, the public finds a meaning, even in the case of Twombly’s images of looping lines that appear to write nothing.

To this end, Barthes states a general principle: “Meaning sticks to man: even when he wants to create non-meaning or extra-meaning, he ends by producing the very meaning of non-meaning or of extra-meaning.”45 Here, again, the argument is characteristic Barthes: the negative of a positive becomes a positive. Just as a convincingly denotative photograph (signifying nothing outside itself) can morph into a connotative sign by assuming the look of a lack of coding (referring to “reality”), so a painting—if no other meaning emerges—can demonstrate a rhetorical point about either metaphor or metonymy. To view Twombly’s art is to recognize that we cannot escape its signifying force, even though, like a piece of pea stone in the lawn, it should be too insignificant to have assumed any meaning whatever.

I like to imagine that each time Barthes turned the negative of a positive into a positive, he was just then passing the tipping point between a structural analysis and what we might call a post-structural analysis—right there, for example, in the course of writing the first pages of his essay on press photography in 1961, “The Photographic Message.” He may not have known where this gesture was leading him. In structural analysis, denotation and connotation can be opposed antithetically; but in post-structural analysis, even these structural, rhetorical forms become dynamic and fluid. We can no longer rely on them to maintain their prescribed function.

Barthes states that Twombly’s work “coincides with its appearance,” which is another way of referring to a producing that is not a production.46 “Of writing, TW retains the gesture, not the product.”47 The formulation recalls Barthes’s sense of fully denotative images, that they have no coding and require no interpretation. If Twombly represents writing by writing a name on a canvas, we simply see writing (the lines), not a picture of writing (a text to be interpreted). Yet we wonder whether this is Twombly’s natural way of writing or an artistic, staged manner of writing, one that he created—his performance. Does the question not apply to all handwriting, even to all marking, to all gestures of signification? Are they not connoted by the nature of the gesture (call it “style,” if you will)? Such questions lead back to typical theories of expression, hard to avoid. At the very least, Twombly’s line becomes the metaphor of the literal, the convincing simulation of the real—it has the look of being a line so casually made as to appear virtually by chance. “Never mind if, in fact, the work is the result of precise calculation,” Barthes comments: “What matters is the effect of change or, to put it more subtly ... of inspiration, that creative force which is, in a sense, the euphoria of chance.”48 We are left to make a critical choice, which may not be one: expression or chance.

Twombly’s art gives the appearance of being made of nothing, of coming from nowhere. Like an anonymous photograph, its meaning is reality. It expresses chance. A slight protrusion of fibers within the weave of a piece of canvas—an imperfection, a nub—becomes the instigation for a composition. In other instances, Twombly might begin by noting an area of discoloration on a sheet of drawing paper, or a slight bump of dry pigment within an otherwise smooth application of a ground color. Such aesthetic nonentities are accented by the artist’s attention, each becoming an aesthetic event—a sensory interruption within Twombly’s life of art. He might circle the nub with a pencil line, perhaps adding a spot of color within the circle and several penciled arcs to one side. By this additive process, he generates his own chances, each as real as the last.

Nub is a curious word. It refers to an insignificant imperfection in a fabric as well as to the most significant feature of a situation. Perhaps it could function as metaphor for the insignificant stretch that becomes the significant effect of the real: the nub of “Un coeur simple” is the barometer; the nub of The Man Who Wasn’t There is the pea stone. Barthes referred to knowing what he was doing but not what would result. If this was true of Twombly, things seem to have worked out for the best. Barthes writes of the artist’s “good luck” in dealing with accident—with irregularities, like nubs.49 This is our situation, too, over and again: something insignificant (a nub) becomes significant (the nub).

As one event follows another, we are the products and the causes of chance effects. We live a personal narrative of events we cannot predict. The line we draw, our gesture, produces a figure we would never conceive, had we not drawn it. We sometimes fail to recognize ourselves in photographs: even the most denotative image proves indeterminate and unreliable. Yet life remains rich within this pattern of insecurity. Given all the factors of chance, we have a chance at good luck. So when we conduct research, or write, or merely think, and when we realize how little sense we have of what meanings will emerge—significant or insignificant—we realize as well that we could be lucky.

  1. Charles Sanders Peirce, “Phaneroscopy or the Natural History of Concepts” (c. 1905), Collected Papers, ed. Charles Hartshorne, Paul Weiss, and Arthur W. Burks, 8 vols. (Cambridge, MA: Harvard University Press, 1958-1960), 1:170. I thank Sima Godfrey (University of British Columbia) as well as Jessamine Batario and Jeannie McKetta (University of Texas at Austin) for essential aid in research. A version of my essay appeared under the same title in Kodikas/Code: Ars Semeiotica 37, no. 3/4 (July-December 2014): 191-208.
  2. Peirce, “A Guess at the Riddle” (c. 1890), Collected Papers, 1:206.
  3. Peirce, “Scientific Metaphysics: The Architecture of Theories” (1891), Collected Papers, 6:15.
  4. Peirce, “Fallibilism, Continuity, and Evolution” (c. 1897), Collected Papers, 1:72.
  5. Roland Barthes, “L’effet de réel,” Communications 11 (March 1968): 84-89; “L’effet de réel,” Oeuvres completes, ed. Éric Marty, 5 vols. (Paris: Seuil, 2002), 3:25-32; “The Reality Effect,” The Rustle of Language, trans. Richard Howard (New York: Hill and Wang, 1986), 141-48 (quotation, 142 [emphasis original]).
  6. Barthes, “The Reality Effect,” The Rustle of Language, 141 (emphasis original).
  7. As evidence of this influence, l’effet de Barthes, see, for example, Seymour Chatman, Story and Discourse: Narrative Structure in Fiction and Film (Ithaca: Cornell University Press, 1978), 143-45.
  8. Barthes, “The Reality Effect,” The Rustle of Language, 148.
  9. Roland Barthes, “Science versus Literature,” Times Literary Supplement, no. 3422, 28 September 1967, 898.
  10. Barthes, “The Reality Effect,” The Rustle of Language, 146.
  11. “Chance [spontaneity] is merely the possible discrepancy between the character of the limited experience to which it belongs and the whole course of experience [regularity]”: Peirce, “Uniformity” (1902), Collected Papers, 6:78. The incursion of an anomalous element of reality into a dreamlike fiction amounts to an instance of chance.
  12. Barthes, “The Reality Effect,” The Rustle of Language, 141; Gustave Flaubert, “Un coeur simple,” Trois contes (1877), Les chefs-d’oeuvre de Gustave Flaubert, ed. Maurice Nadeau, 10 vols. (Geneva: Éditions Rencontre, 1970), 9:193: “Un vieux piano supportait, sous un baromètre, un tas pyramidal de boîtes et de cartons.”
  13. Barthes, “The Reality Effect,” The Rustle of Language, 142 (emphasis original). Barthes can imagine somewhat remote connotations that the piano and heap of containers might bear, but discovers no plausible possibility for the barometer, given the terms of Flaubert’s narrative. The disorderly piano might just as well have been located under a clock or a painting.
  14. Barthes, “The Reality Effect,” The Rustle of Language, 148 (emphasis original).
  15. Peirce, “Fallibilism, Continuity, and Evolution,” Collected Papers, 1:72.
  16. Barthes, “The Reality Effect,” The Rustle of Language, 143 (emphasis added).
  17. Roland Barthes, “Le message photographique,” Communications 1 (1961): 130; “Le message photographique,” Oeuvres completes, 1:1123-24; “The Photographic Message,” The Responsibility of Forms, trans. Richard Howard (Berkeley: University of California Press, 1985), 8 (emphasis original).
  18. Writing in 1961, Barthes was reflecting on film photography, not digital.
  19. The translation is Richard Howard’s in The Rustle of Language.
  20. Barthes, “L’effet de réel,” Oeuvres complètes, 3:27.
  21. Oxford English Dictionary, s.v. beach, (accessed 30 June 2014).
  22. Barthes, “The Reality Effect,” The Rustle of Language, 142.
  23. The screenplay for The Man Who Wasn’t There is found at't-there. In several minor details, this version fails to correspond to the film as released. My quotations have been taken from the film itself.
  24. Barthes, “The Photographic Message,” The Responsibility of Forms, 8 (emphasis original).
  25. Barthes, “The Photographic Message,” The Responsibility of Forms, 6-7. I have substituted Barthes’s French term à la lettre for the translator’s word literally to restore the sense of part-by-part analysis, which photographic images resist.
  26. Barthes, “The Photographic Message,” The Responsibility of Forms, 7 (emphasis original).
  27. Barthes, “The Reality Effect,” The Rustle of Language, 142-43 (emphasis original).
  28. See Richard Shiff, “Realism of Low Resolution: Digitisation and Modern Painting,” in Terry Smith, ed., Impossible Presence: Surface and Screen in the Photogenic Era (Chicago: University of Chicago Press, 2001), 124-56.
  29. The presumed innocence of a photograph does not cause us to become innocent viewers. We bring our expectations and previous experience to its reception, as well as being affected by its context of presentation. “Deciding what is being talked about is, of course, a kind of interpretive bet. But the contexts allow us to make this bet less uncertain than a bet on the red or the black of a roulette wheel”: Umberto Eco, “Overinterpreting Texts,” Interpretation and Overinterpretation, ed. Stefan Collini (Cambridge: Cambridge University Press, 1992), 63.
  30. Barthes, “Science versus Literature,” 897-98.
  31. Roland Barthes, “Sagesse de l’art” (“The Wisdom of Art,” trans. Annette Lavers), Cy Twombly: Paintings and Drawings 1954-1977, ed. David Whitney (New York: Whitney Museum of American Art, 1979), 9-22; “Sagesse de l’art,” Oeuvres complètes, 5:688-702; “The Wisdom of Art,” trans. Richard Howard, The Responsibility of Forms, 177-94.
  32. Roland Barthes, “Non Multa Sed Multum” (”Cy Twombly: Works on Paper,” trans. Henry Martin), in Yvon Lambert, Catalogue raisonné des oeuvres sur papier de Cy Twombly, Volume VI, 1973-1976 (Milan: Multhipla, 1979), 7-13; “Cy Twombly ou ‘Non multa sed multum,’” Oeuvres completes, 5:703-20; “Cy Twombly: Works on Paper,” trans. Richard Howard, The Responsibility of Forms, 157-76.
  33. See Richard Leeman, “Roland Barthes et Cy Twombly: le ‘champ allusive de l’écriture’,” Rue Descartes, no. 34 (December 2001), 61. According to Leeman, Lambert also requested an essay from Michel Foucault, who declined.
  34. Hero and Leandro is a four-part work consisting of three separate canvases and one work on paper; the “Leandro” inscription appears on Part I.
  35. “Sometimes I like a title to give me impetus or a direction or a feel for the way [the painting] should go”: Cy Twombly, quoted in David Sylvester, “Cy Twombly” (2000), Interviews with American Artists (London: Chatto & Windus, 2001), 176.
  36. Andy Stafford, Roland Barthes, Phenomenon and Myth: An Intellectual Biography (Edinburgh: Edinburgh University Press, 1998), 17.
  37. Robert Pincus-Witten, “Learning to Write” (1968), reprinted in Nicola del Roscio, ed., Writings on Cy Twombly (Munich: Schirmer/Mosel, 2002), 56-60; Robert Pincus-Witten, “Cy Twombly,” Artforum 12 (April 1974): 60-64. Pincus-Witten based his analysis on conversations with the artist in 1966 (Pincus-Witten, statement to the author, 8 January 2008).
  38. Pincus-Witten, “Learning to Write,” Writings on Cy Twombly, 58-59; Pincus-Witten, “Cy Twombly,” 62-64; Barthes, “Cy Twombly: Works on Paper,” The Responsibility of Forms, 163-66.
  39. “Whatever is ‘gauche’ about the left hand is indispensable to an advanced culture; it keeps us in touch with man’s venerable past, with a time when he was not over-skillful”: Henri Focillon, “In Praise of Hands” (1936), trans. S. L. Faison, Jr., in The Life of Forms in Art, trans. Charles Beecher Hogan and George Kubler (New York: Wittenborn, Schultz, 1948), 67. Barthes himself was left-handed: see Roland Barthes, Roland Barthes par Roland Barthes, trans. Richard Howard (New York: Hill and Wang, 1977 [1975]), 98.
  40. Barthes, “The Wisdom of Art, The Responsibility of Forms, 183.
  41. See, for example, Richard Serra, “About Drawing: An Interview” (interview by Lizzie Borden, 1977), Richard Serra: Writings Interviews (Chicago: University of Chicago Press, 1994), 52.
  42. Barthes, “Cy Twombly: Works on Paper,” The Responsibility of Forms, 171 (emphasis original).
  43. Barthes, “The Wisdom of Art,” The Responsibility of Forms, 181-82 (emphasis original).
  44. Barthes, “The Wisdom of Art,” The Responsibility of Forms, 183 (emphasis original).
  45. Barthes, “The Wisdom of Art,” The Responsibility of Forms, 183.
  46. Barthes, “Cy Twombly: Works on Paper,” The Responsibility of Forms, 157.
  47. Barthes, “Cy Twombly: Works on Paper,” The Responsibility of Forms, 160.
  48. Barthes, “The Wisdom of Art,” The Responsibility of Forms, 181 (emphasis original).
  49. Barthes, “The Wisdom of Art,” The Responsibility of Forms, 181-82.


Richard Shiff

Richard Shiff is Effie Marie Cain Regents Chair at The University of Texas at Austin, and a Consulting Editor at the Rail.


The Brooklyn Rail

OCT 2019

All Issues