History of Art
Yale University
New Haven CT 06520

The issue is motivation. As Eric Gans demonstrated in his 1981 essay “Differences,” it is motivation, or rather the lack thereof, that compromised Derridean deconstruction from the start.(1) For, while Derrida sought to re-historicize language by turning away from the temporally impoverished sameness of speech toward the temporally rich différance of writing, his insistence that différance itself remain “absolutely arbitrary” kept his account from becoming truly historical. As a principle that was “always already” valid, Derridean différance often verged on an a priori truth, assuming precisely the metaphysical properties that Derrida had sought to avoid. The refinement of Generative Anthropology [GA] over the past three decades has to a large extent been dedicated to correcting this, deconstruction’s structural(ist) flaw.

The crux of this corrective endeavor is the “originary hypothesis,” GA’s account of the first sign. Adapted from René Girard’s theory of mimetic desire, the originary hypothesis supplements Derrida’s project by providing différance an anthropological motivation, namely the deferral of the appetitive violence that is incited by the presence of the referent. By identifying the aborted gesture of appropriation as the first sign, the originary hypothesis allows GA to anchor différance in an historical event—indeed, in the first historical event from which all culture is derived—thereby freeing deconstruction from the last vestiges of the metaphysical tradition and fulfilling the historical ambition that Derrida himself prematurely abandoned.

And yet, as compelling as the originary hypothesis is, the mechanism of deferral that drives it has remained something of a stumbling block. The problem is, in my view, narratological and raises again the issue of motivation which, for all the elegance of the originary hypothesis, remains one of GA’s more complex features. The complexity is due to the fact that in the originary hypothesis not one but two sources of motivation are implied: the animal desire for the appropriation of the referent and the human desire for the cultivation of the sign. The failure to attend adequately to the distinction between these two sources of motivation and to the decisiveness of the transition from the first to the second continues to trouble GA, most of all in discussions of deferral, for it is during the interval of deferral that that transition is made.

We get some sense of this from the pathos that informs Gans’s 2011 assurance that “the deferral of action is not a mere postponement.”(2) That this bears repeating some thirty years after GA’s initial formulation suggests that something in our thinking about deferral is out of joint. That something is, I think, the lingering assumption that the interval of deferral is suspended evenly between animal and human sources of motivation, tending toward neither. Indeed, only such an assumption could leave open the possibility that the appropriation of the referent could still be at least as desirable as the cultivation of the sign, a possibility that is necessary for the suspicion that deferral could be likened to anything resembling postponement. Although much of GA refutes this assumption, it nevertheless recurs,(3) perhaps most notably in GA’s claims to scientific rigor. Here too the issue is motivation, for as Charles Taylor has explained, scientific rigor is defined by its being limited to “weak evaluations” in which the constitutive role of motivation is denied rather than “strong evaluations” in which that role is admitted.(4) Whereas weak evaluations are concerned merely with outcomes, strong evaluations are concerned with what Taylor calls the “worth of different desires.” That GA deals in strong rather than weak evaluations is clear, for it is precisely the recognition of the greater worth of the desire for the sign that not only constitutes GA’s object of study (man), but also furnishes the means by which GA and all other cultural undertakings are pursued (language).

Now, while some theoretical projects are able to continue unaffected by their conviction that they deal in weak rather than strong evaluations GA is not among them. As we saw a moment ago, failure to attend adequately to the distinctions between animal and human motivations (read “desires”) and to the decisiveness of the transition from one to the other poses a problem particularly when it comes to deferral. The problem becomes still more apparent when we zoom out and consider GA as a whole. For, whereas in the case of deferral the suspension between animal and human motivation leaves open the possibility of mere postponement, a possibility to which we can respond with arguments as needed, the suspension of GA itself between weak and strong evaluations, that is, between the natural and the human sciences, deprives us of the ability to recommend GA as an interpretive approach—that is to say, it deprives us of the ability to treat it as not only a descriptive but also as a prescriptive part of our disciplinary repertoire. For, once the fundamentals of the originary hypothesis have been established and culture has been shown to be the outcome of the originary scene, the rhetorical reserves of weak evaluation are at that point exhausted leaving GA in its current guise little with which to pursue its role as a protagonist in the ongoing story of culture, a story to which it is so obviously committed but which it is forced to abandon as a fait accompli. The possibility that deferral may be mere postponement returns but on a far larger scale as the possibility that human history itself may have been so much lag time, simply a prelude to an ever more vivid climactic episode of appetitive violence which we, at best, had managed to put off.

Here we arrive at what I suggested a moment ago was the narratological root of the problem of deferral. By narratological I mean simply having to do with that species of motivation peculiar to narrative: what Frank Kermode called “the sense of an ending”—that is, the sense of narrative orientation or directed-ness that allows actions to become emplotted either implicitly or explicitly in something like a schema of rising action, climax and denouement.(5) While deferral serves well enough as a beginning of culture, it has proved less well suited as an end for culture in Kermode’s sense, because, as we have seen, it leaves open the possibility of postponement. In Kermode’s terms, when offered as an end for culture, deferral as it stands now in the literature on GA allows what are in fact two ends—the appropriation of the referent and the cultivation of the sign—to vie with one another, sapping GA’s narrative strength.

The task, then, seems to me to be to narratively re-charge the interval of deferral by reframing it unambiguously as tending toward one and only one end: the cultivation of the sign. This will require our being able to consider the cultivation of the sign as a punctual moment analogous (but opposed) to the appropriation of the referent, for in order for deferral to serve as a narrative end for culture we must posit a narrative end in deferral.

I’ll spend the remainder of this paper trying to do just that. As I am an art historian and deal in pictures, I’ll do so by sketching a (needless to say, abridged and episodic) history of viewing in the Western tradition, one which I will set in the long-durée of deferral while nevertheless giving it the directed-ness proper to narrative, framing it as tending away from a concern with the appropriation of the referent and toward a concern with the cultivation of the sign whose status as an end punctuating the actual experience of looking I will signal using the neologism “aspect acquisition.” In true modernist fashion, that narrative will play out on either side of the picture surface, with the momentum of attention shifting from a fixation on referents felt to be behind the plane to aspects felt to be in front of it.


Cy Twombly, Untitled (1967). Oil and wax crayon on canvas [200.7 x 264.2 cm].
Yale University Art Gallery, New Haven

To anticipate things considerably and to wear my strong evaluations on my sleeve I’ll take this opportunity to introduce one of if not the end toward which I myself will be tending: Cy Twombly’s Untitled from 1967. White rectangles paired with the odd notation rendered in wax crayon and oil posing as chalk on slate, the work is a veritable master class in the Derridean trace, a site of pure différance without presence.(6) More importantly for my purposes, it serves as something like a hermeneutic proof, demonstrating how the cultivation of the sign is intuitively regarded as an end punctuating the interval of deferral which here becomes coterminous with the interval of viewing, allowing it to be experienced narratively. While the painting begins as a series of rectangles haphazardly placed, immanent to the picture surface and compositionally undifferentiated, a ripple soon courses through it announcing the start of an optical drama. On the painting’s middle-right edge the largest rectangle sheds its derivative association with the painting’s frame as it comes into focus as the nearest face of a three-dimensional prism the other faces of which are formed by newly recruited rectangles above and to the right. Or again, on the painting’s top-left corner a stray line is joined to an abutting strip to form a passage of staggering richness and subtlety as one volumetric bar seems to slip backward almost cinematically, in this case fracturing again into planes as if it were casting off rather than taking on dimensionality before finally achieving the sheer flatness of the square behind. If, encouraged by these instances we then expand our field of vision to take in the composition as a whole we notice still more varieties of visual incident like its lilting arc or its triple-banded layering.

What is crucial in all of this is that each time our attention comes to rest on this or that facet we acquire an aspect of the painting—that is, we notice something about it that is of interest to us despite its having nothing to do with reference of any kind. The volumetric prisms, lilting arcs and triple-layers are, after all, synthesized by us and strictly speaking belong neither to the painting itself nor to any other object for that matter. Moreover, and more importantly, in each case we experience the moment of aspect acquisition as one that is narratively satisfying, that is to say, one that brings to a conclusion—to an end—a stretch of viewing that in retrospect would have been infuriating had it continued indefinitely (the sound of the narrative irresolution that is caused by the failure to acquire an aspect is familiar enough—indeed, it can be heard ringing in the echo-chambers of that are today’s galleries and museums: the exasperated sigh “I just don’t get it” that marks not so much the end of a particular stretch of viewing but rather the pained admission that that stretch will, at least for a time, go end-less).

The success of Twombly’s painting in supplying aspects as ends—or, perhaps more specifically, the success of the mode of viewing that we exercise while standing in front of it, accepting the acquisition of aspects as an end—should not be taken for granted. Although we are able to look at pictures as opportunities for aspect acquisition, and although that ability belongs, as GA has shown, to our deepest anthropological inheritance, the history of viewing in the West overflows with works the narrative ends of which remain, in one form or another not the acquisition of aspects but rather the appropriation of referents. Examples might include John Constable’s Cottage in a Cornfield from circa 1833, or Giovanni Tiepolo’s Apotheosis of the Spanish Monarchy from 1764. What is remarkable about these works is how keenly we feel distance in them, how keenly we feel that the cottage in the landscape is “a ways away,” that the attendants of Maria Amalia look at us “from afar.” The result is a mode of viewing very different from that which we experienced a moment ago with Untitled. Standing before these works, time is no longer studded by the satisfactions of aspect acquisition. Appreciation of the handling of paint or the novelty of palette does not make up for the fact that we are interrupted by the gate that blocks our path, jealous of the putti who, unlike us, are free to kiss the attendants’ necks. I’ll call this time un-studded by the end of aspect acquisition “bad deferral,” deferral motivated by the desire to appropriate the referent, deferral as postponement, deferral without end.

Giovanni Battista Tiepolo
The Apotheosis of the Spanish Monarchy
Study [81.6 x 66.4 cm] The Metropolitan Museum of Art, New York
Image © The Metropolitan
Museum of Art

John Constable
The Cottage in a Cornfield (1833)
Oil on canvas [62 x 51.5 cm] Victoria & Albert Museum, LondonPhoto Credit: V & A Images, London /Art Resource, NY

Before delving into the history I promised—the narrative dimension of which will, I hope, offer some relief from this bad deferral by swapping out the end of referent appropriation and swapping in the end of aspect acquisition—I’ll pause here to define my terms beginning with “referent.” In addition to the conventional semiotic understanding of the referent as that to which the sign points, I follow proponents of GA in recognizing the referent as being the focus of an appetitive desire that was originally directed toward a material object in the real world. For those of us raised in the Western tradition of pictorial illusionism, this understanding of the referent has unique consequences insofar as we, our claims to specular sophistication aside, nevertheless see depicted referents when “realistically represented” as somehow coupled to their painted doubles. It is this credulity that allows distance to be felt so keenly in the Constable and the Tiepolo.

The second term I need to define is “aspect.” I choose to speak of “aspects” rather than of “signs” not only because the former term lends itself more readily to visual analysis, but also because I think it captures more accurately the dynamics of abortive appropriation that are at the heart of GA. After all, aspects are more efficient than signs in carrying out symbolically the crucial distributive rite of sparagmos since they, unlike signs, are inherently partitive. As Wittgenstein pointed out in his famous discussions of the “duck-rabbit” illusion, we can see the duck or the rabbit, but never the duck and the rabbit.(7) In other words, we have signs for things but aspects of things. Moreover, if it is necessary, as I think it is, that we posit an end in deferral that is analogous but opposed to the appropriation of the referent, the aspect seems most up to the task. While one can possess an aspect as if it were a referent, feeling the full degree of satisfaction such possession assumes, one cannot possess a sign in quite the same way. Promising at most substitutability for the referent, the sign will always be shadowed by its twin, the object to which it refers but which it is not. For these reasons I suspect that in the originary scene it was the satisfaction of aspect acquisition that marked (indeed allowed) the abortion of appropriation. It was not only fear of reprisal that coursed through the body of the first man as his fingers flexed (or, perhaps, slackened) in ostension. A new species of satiation was also felt, one that must have occurred the instant before ostension when the first aspect dawned on him. Now, on to history.


We begin in medias res, that is to say, amidst the things strewn about the foreground of Veronese’s 1563 Marriage at Cana. The painting is absolutely filled with figures pouring wine, playing instruments and pulling cloaks. While the scene depicted is a story—the story of Jesus’s first miracle, recounted in the second chapter of the Gospel of John—it is not the story depicted in the painting that concerns me but rather that of the history to which the painting belongs. As I promised to frame that history narratively, and as every good narrative deserves a villain, I’ll introduce one now: the figure in orange standing there at the balustrade gazing at us contemptuously over his shoulder, his hand gripping the nearby jug. What makes him a villain is his conspiratorial role in the painting’s logic, a logic driven by the appropriation of referents rather than by the acquisition of aspects. His villainy can be summed up by the fact that despite the proliferation of goods miraculously produced by Christ that line the tables—water turned into wine, loaf turned into loaves—not to mention the proliferation of painterly incident that Veronese everywhere puts on breathtaking display, the figure still operates according to conditions of scarcity proper to the animal desire for the referent in the originary sense. Indeed, once we have locked eyes with him we too are tempted to look past the bounty in the foreground toward something further off, something just behind the balustrade on which he leans which suddenly becomes synonymous with the picture surface itself. For a brief but excruciating moment the specter of postponement returns again to haunt the interval of deferral as the ability of sheer viewing to satisfy is thrown into question. Standing before Marriage at Cana, in the ochre light of high culture, we somehow seem on the verge of relapsing into animality.

Paolo Veronese, Marriage at Cana (1563). Oil on canvas [666 x 990 cm].
Musée du Louvre, Paris
Photo credit: Erich Lessing / Art Resource, NY

If we want to know how this temptation to look past the picture surface developed such that the prospect of relapse could linger on and more importantly how we might condition our analytic and descriptive efforts such that that relapse might be avoided, we must turn our attention from Veronese and the figure in orange to a point some two thousand years earlier in the fifth century B.C.E. when the fork in the road split for the Western tradition, breaking off into two paths, one leading to a mode of viewing oriented toward referents and the other to a mode of viewing oriented toward aspects. The two paths correspond to two early Greek genres, skenographia and skiagraphia, and depart from the picture surface, the one leading to the space of referents behind it, the other the space of aspects in front of it.(8)

Skenographia, a mature example of which can be found in the so-called “Room of the Masks” at the House of Augustus, is derived from the Greek skene, which translates most accurately as “tent” but which also has broader associations surviving in our words “scene” and “scenario.” As a painterly genre it likely emerged sometime in the fifth century B.C.E. when it was used to render the backdrops of tragic plays illusionistically as a series of receding planes. In the “Room of the Masks,” for example, there appear to be nearly a dozen planes, beginning with the faux-plaster ledge then leading on to further planes of marble, wooden architecture, more plaster, and finally a trompe l’oeil curtain decorated with a pastoral scene behind which, it is suggested, the recession may or may not continue. Crucially, the receding planes created by skenographia give way to a space on the far side of the picture surface—a space which, as the mask-filled niches suggest, we feel is capable of accommodating referents.

Dioscurides of Samos, “Street Musicians” (Scene from a Comedy) from Cicero’s Villa,
Pompeii (first century B.C.E.). Mosaic [41 x 43 cm].
Museo Archeologico Nazionale, Naples
Photo credit Scala / Art Resource, NY

With skiagraphia, things are reversed. From the Greek skia or “shadow,” this painterly genre, which also emerged sometime in the fifth century B.C.E., created a space on the near side of the picture surface that was devoted not to referents but to aspects. A mature example can be found in a mosaic from a comedic cycle by Dioscurides of Samos now known as “Street Musicians.” Here, rather than a succession of receding planes a single white ground is established as flush with picture surface. Against this figures cast their shadows as if they were projecting into our space. Even where a concession to recessive illusionism must be made in order to provide a ground plane on which the figures might stand, this is immediately converted into what we might call projective illusionism as the knee, thigh, torso, arm and drum held by the rightmost figure are shown to be further forward than his right foot, the toes of which hang ever so slightly over the lower ledge. Here, action takes place on the near side of the picture surface rather than on the far side as before. We do not desire to breach that surface. Instead we are content to imagine—to project—the tones of the pipes and symbols, to take pleasure in the play of light as it dances across the folds of the fabric.(9)

And so our narrative begins to takes shape with good and evil, hero and villain, assigned their respective territories on either side of the picture surface. On one side, a space of aspectival plenty, on the other, of referential scarcity. The interval of deferral also begins to take shape as the end of aspect acquisition comes into view.

Paolo Veronese, Marriage at Cana

Dioscurides of Samos, “Street Musicians”

Nevertheless, following the early Greek moment of parity, the balance shifted and villains ruled the day. The coup was Renaissance perspective. First demonstrated around 1413 by Filippo Brunelleschi, so-called “single-point” perspective was a means of systematically rendering pictorial space such that the objects depicted appeared to recede with geometric regularity away from a single viewing point toward a single vanishing point along a sight line extending from one to the other. Brunelleschi conducted his demonstration at the Piazza del Duomo, painting the Florentine Baptistery on the reverse side of a wooden panel through which he drilled a hole. Holding the panel up to his or her face, a viewer would look through that hole at a mirror held in the other hand, seeing the reflection of the painting as if it were the Baptistery itself glimpsed from the precise spot where Brunelleschi had stood, complete with that day’s sky reflected against a ground of burnished silver.(10)

What is remarkable about Brunelleschi’s demonstration is the degree to which it sought to deny pictoriality. Instead of a fictional landscape, it represented a real piazza, one with which all Florentines would have been familiar. What was seen, the panel claimed, was the thing itself, the baptistery on that very day, at that very moment, flecked by rain or lit by sun as the case may be. Moreover, given the posture required by the demonstration which called for the viewer to hold the mirror at arm’s length, that thing was no longer within walking distance but was now tantalizingly, infuriatingly, within reach(11)—a sensation that was not lost on Antonio Manetti, Brunelleschi’s contemporary and biographer, who concluded his recounting of the demonstration by stating, “…it seemed as if the real thing was seen: and I have had it in my hand [e io l’ho avuto in mano…].”(12) That these two convictions—that of seeming and of holding—cancel each other out suggests the cruelty of Renaissance perspective as a mode of viewing motivated by referent appropriation.(13) While within reach, gripped in both hands, the baptistery remains forever on the far side of an interval of endless deferral, the object of a promised possession that is, per force, never fulfilled.

The extent of this cruelty is further suggested by the work of another of Brunelleschi’s contemporaries, the painter Paolo Uccello. Perhaps best known for an anecdote recorded by Giorgio Vasari in which he responded to his wife’s pleading that he return to bed by exclaiming, “Oh, what a lovely thing perspective is!,” Uccello was even more driven by Renaissance perspective’s promise of possession, leading to such unsettling works as The Hunt in the Forest (1470).(14)

Paolo Uccello, The Hunt in the Forest (c. 1470).
Tempera on wood [65 x 165 cm].
Ashmolean Museum, Oxford
Photo credit: Erich Lessing / Art Resource, NY

As pitched as Uccello’s picture is, the sense that a referent lurked somewhere behind the picture surface grew only more acute as the Renaissance progressed. For example, when Leon Battista Alberti published his treatiseDella pittura in which he further systematized single-point perspective, he likened the picture surface to a velo or veil hanging between viewer and referent.(15) In a woodcut from his 1538 “Handbook for Draughtsmen” Albrecht Dürer literalized the metaphor, stationing buxom reclining women on one side of the reticulated net, a transfixed draughtsman on the other.

Albrecht Dürer, Man Drawing a Reclining Woman. From the
“Handbook for Draughtsmen” (Underweisung der Mesung) (1538).
Woodcut [7.5 x 21.5 cm]. The Metropolitan Museum of Art

This, then, is the pre-history of Veronese’s picture. The balustrade is the velo, the figure in orange the ammonitore, another of Alberti’s theoretical inventions, this one tasked with warning us of infringing on whatever lies behind, “either beckon[ing] [us] with his hand to look, or with ferocious expression and forbidding glance, challeng[ing] us not to come near”—a model-rival of which Girard would be proud.(16) Momentum seems to have shifted once and for all in favor of the figure in orange. The narrative seems to be fixed in its orientation toward the end of referent appropriation, fated to play out during an interminable stretch of bad deferral.


But that is not our narrative, not our end. The aesthetic pleasure we experience standing before works like Twombly’s Untitled proves as much. And Twombly knew it. In his 1970 painting Treatise on the Veil (Second Version) he takes issue with the entire tradition I’ve just recounted. Beginning with its title, which I take to be among other things a jab at Albertian theory, the work undoes patiently and with easy elegance the temporal armatures of referent appropriation, replacing them with the pleasures of aspect acquisition that are proper to speculative art making and viewing.(17) The interval of deferral is now studded with new ends, ends that have nothing to do with material possession, ends that are instead marked by the drafting of “treatises” and the iteration of “versions.” We no longer seek to breach the picture surface but are instead content to drift laterally along it, following the narrow rectangles as they disappear from view, slipping not behind the work’s panels but rather off their edges into the half-optical, half-notional space of aspectivity.

It is the same space of aspectivity to which we were earlier introduced in Untitled, a work that if anything renounces referent appropriation even more emphatically by reversing the relationship between volume and plane, its prisms bulging on the near rather than the far side of the picture surface. But if the choice of aspects over referents was clear for Twombly, it was by no means idle. In the 1960’s, at the height of American modernism’s supposed sophistication, a relapse to reference was occurring, making the choice divisive and stark.

In a series of essays begun in 1960, for example, the critic and historian Michael Fried was forced to reestablish the category of the artistic medium so as to ensure that artworks themselves would not be confused for referents.(18) All artworks, Fried argued, begin as neutral “shapes” and became either “mediums” or “objects” depending on the degree to which they are able to address and overcome their sheer materiality, becoming what he calls “conventions” capable of compelling “conviction.” An artwork that failed to do so and that threatened to establish what Fried considered to be a dangerous precedent was Tony Smith’s Die (1962). Not only did that work make no claim to be anything other than six cubic feet of steel, more importantly it laid an enormous temporal burden on the viewer. Whereas works-as-mediums transcended materiality, producing the sensation of what Fried called “presentness” in which the totality of culture as such seemed available to be experienced, works-as-objects languished in a “presence” that Fried describes as “endless” in that it disallowed any meaning beyond the sheer mass of the work while ensuring that that mass could be possessed by neither eye nor hand. All referent, no aspect, we are only ever able to see a fraction of Die at any given moment and as a result are left circling a work that, though lifeless, is never dead.

It is in this context that Twombly chose aspect over referent. He did so by demonstrating time and again the pleasures that aspect acquisition affords. In his 1966 painting Night Watch, for example, a cube not unlike Smith’s is rendered using a handful of spindly lines. Regardless of its fragility, the cube nevertheless seems to bulge as it projects out toward us, as if it were wrapped around a swelling mass. And yet, just as this prospect occurs to us—a prospect altogether unprecedented in Twombly’s oeuvre—we notice again the right angle formed by the dense layer of planes to the left. Suddenly the cube is punctured. As it begins to deflate we abandon it and shift our attention to that layer of planes now delighting in its volume—not the literal volume of a singular referent, but rather the figurative volume of limitless aspects.

Cy Twombly, Night Watch (1966).
House paint and wax crayon on canvas
[190 x 200 cm]. Private CollectionImage © Cy Twombly Foundation
Rembrandt van Rijn, Night Watch (1642).
Oil on canvas [359 x 438 cm].
Rijksmuseum, Amsterdam
Image © National Gallery, London / Art Resource, NY

But Twombly was not content to let each experience of aspect acquisition stand as an isolated event. He wanted instead to ground them in a revised art history that tended away from reference and toward aspectivity, an art history in which they would be cast as triumphs over reference—the art history I’ve been sketching in these last few pages. Hence the title Night Watch, which alludes to Rembrandt’s 1642 work. It is not surprising that Rembrandt should be given a pivotal role for it is he who more than any other seventeenth-century artist broke with the referential conventions of the Renaissance, ushering in the era of non-referential Baroque vision.(19) Consider, for example, his 1632 etching The Raising of Lazarus. Here we as viewers survey the scene from what is surely the least propitious position. Of the four groups of viewers, not only are we farthest away from the action, we are also least able to see Christ, whose face remains hidden from us. And yet, remarkably, we do not suffer for it. We do not envy the views of the others. Although Rembrandt renders the miracle as if seen by four distinct groups, we have no desire to consolidate these vantages. They are partitive aspects of a scene which we have no desire to recompose or make whole.

Rembrandt van Rijn, The Raising of Lazarus:
The Larger Plate
(c. 1632). Etching and
burin on paper [37 x 25.7 cm].
Staatliche Museen, Berlin
Image © RMN-Grand Palais / Art Resource, NY

Night Watch distills all this in the beautifully rendered hand of Frans Banning Cocq. Not only does the shadow cast by that hand mount a defense on behalf of partitive aspects by doubling that hand, transforming it into a gesture of aspective generosity, it also marks the picture surface, stating unequivocally that—as in Twombly’s painting—no referent lurks behind the canvas. The action is taking place on the near side of the picture surface, somewhere between the viewer and the wall.

Cy Twombly, Night Watch Rembrandt van Rijn, Night Watch (Detail)


By seizing Rembrandt’s victory over reference as a point of departure, Twombly reinvigorates the interval of deferral, reimaging it as the interval of art history as such, an interval into which we are ourselves thrown. When looking at his paintings we get the impression that we are cycling back and forth between two temporalities, traveling between two chronotopes—one immediate, all pigment and cotton, the other more diffuse but no less dense, human on the grandest scale. It is for this reason that the lines of Night Watch seem, as is so often the case in Twombly’s work, to be once tactile and tactful, as if they know both less and more than other lines.

This all comes with its share of anxiety. Twombly is asking a lot of us. Most of all, he’s asking us to keep the interval of deferral going. The longer we look, the more convinced we are that Rembrandt’s victory was a fragile one and that Twombly’s watch is, if anything, the more vigilant of the two. That vigilance plays itself out in Twombly’s emphasis on practice. If Twombly’s Night Watch can be said to be a painting of something it is one of “night watch-ing”—that is, of a certain way of looking, a certain practice of vision. And of course it is a painting of something. After all, it is nothing if not mimetic, only what it mimes is not some object out there but the very practice of first pulling chalk (and, indeed, fingers) across a smooth and slightly damp piece of slate and of then standing back to take in the results. That chalk is really wax and that slate is really paint does not bother us because what is on offer is the practice depicted, a practice that Twombly, by mimetically showing it to us, allows us to fully grasp, if only for as long we stand there looking.

It is for this reason that Twombly strikes me as a painter who managed somehow to have gotten at the anthropological root of things, for in his paintings aspect acquisition assumes something like the full dimensions of a referent, a punctual end studding the interval of deferral. When we look at his paintings we aren’t haunted by the specter of postponement the way we were with Constable’s or Tiepolo’s. In them deferral becomes satisfying.  In them deferral becomes good.


But this is a recent development and one that—as I said at the outset—should not be taken for granted. If the narrative I’ve been sketching has gained shape as the centuries have piled up, it has also become more charged. Indeed, with Twombly we’ve come not so much to a stop as to a teetering pause. That, to me, is how it should be. After all, the issue is motivation and motivation is a precarious thing, for to admit motivation is to entertain the prospect of failure. And so, as I began amid things I’ll end it there too, but not among things to have but rather things to do.

While our consideration of Twombly has given some indication of how the interval of deferral can be narratively recharged through the introduction of the moment of aspect acquisition as an end in deferral, his work does little to suggest how urgent that narrative recharging in fact is. Here, a few words on the work of his colleagues Robert Rauschenberg and Jasper Johns might help.

Both Rauschenberg and Johns were, perhaps even more so than Twombly, convinced of the poverty of reference as a motivation for viewing. Indeed, in his Factum I from 1957, Rauschenberg short-circuits reference as such. Adopting the gestural repertoire of Abstract Expressionism, which the previous generation of New York painters had cast as being semiotically without remainder, each confessional stroke achieving—it was claimed—a perfect consubstantiality of signifier and signified, he painted Pollock-eque drips over sheets of newsprint and photographs.(20) Though taken from here and there, these seem, when collected on the canvas, to be pulled gravitationally into the orbit of some referent—stars circling some hidden idea, some interpretive key, perhaps known only to Rauschenberg himself. But this tendency toward reference is promptly—and deservedly—ridiculed by Factum II as naïve and, well, boring. Now, what had been so full of referential promise is not only shown to be iterative but is also exposed as laughable—a lark. While viewers of the pendant works still hung up on reference will leave (or worse, stay) disappointed, those among us unencumbered by reference are free to delight in a cleverness in which we too share. In true scenic fashion, we’re in on it.

This is also the case in Johns’s False Start from 1959, where both reference and bad deferral are lampooned. Here the Ab-Ex stroke’s claim to semiotic hermeticism is reduced to all but the lowest level of absurdity, made the equivalent of the matching of the word “red” to its corresponding color swatch. I say all but the lowest level because that level is reserved for those who, noticing that word and swatch do not in fact match, wish that they did. Johns’s painting seems to challenge us to leave them to it. With the wind of modernism at our backs, the painting seems to suggest that we leave them to the false starts and perpetual postponements of bad deferral while we get on with the urgent task of aspect acquisition.

Jasper Johns, False Start (1959). Oil on canvas
[170.8 x 137 cm]. Museum Ludwig, Cologne
Photo credit: Erich Lessing / Art Resource, NY Art
© Jasper Johns / Licensed by VAGA, New York, NY

And that task is urgent. For, as I said in my introduction, one of the lingering paradoxes of GA is that, while the animal appetite for the referent can neither be satisfied nor strictly speaking even pursued, a condition which when recalled throws into relief the staggering degree to which the human appetite for the sign—i.e., for the aspect—has been the source of the satisfactions that we do experience, the two appetites nevertheless continue to be confused, not only in GA but also and to a far larger extent in our own sign-saturated commodity culture. Indeed, the commodity itself is the principal site where this confusion occurs. Though an abstraction, as Marx argued, the commodity is alas not nearly abstract enough.(21) A can of Ballantine Ale, for example, satisfies both an animal appetite (it has use value) and a human appetite (it has symbolic value). That the satisfaction of the animal appetite requires the continued production and consumption of can after can while the satisfaction of the human appetite can continue, sustained or even enriched by bronzing just two, is a distinction that is either lost or repressed in our commodity culture, resulting in the social political and environmental jeopardy in which we find ourselves today.

If GA wants to be truly generative as a theoretical project, if it wants to participate in the story of culture to which it is so obviously committed, it might strengthen its strong evaluations by stating its motivations more firmly and allocating some of its considerable interpretive power to clearing up this confusion of appetites. On the societal level that would require a retrieval of the foundations of liberalism similar to that attempted by C. B. Macpherson or Isaiah Berlin.(22) After all, it was Berlin who noted that perhaps the sole consolation offered by the reckless adventurism of the twentieth century was the knowledge, gained at terrible expense, that “ends are made, not discovered.”(23) On the textual level it would require, as I have been arguing, our narratively re-charging the interval of deferral by introducing into it ends toward which it might tend, ends which would clearly state and re-state the virtue of choosing signs over referents. That recharging is necessary, for though our becoming human may have occurred in an instant, it seems that our staying human will require more persistence.


  1. Eric Gans, “Differences” in MLN, Vol. 96, No. 4, French Issue (May, 1981), p. 801. See also Richard van Oort, “Epistemology and Generative Theory: Derrida, Gans, and the Anthropological Subtext of Deconstruction” inAnthropoetics, Vol. I, No. I (June 1995). (back)
  2. Eric Gans, A New Way of Thinking: Generative Anthropology in Religion, Philosophy and Art (Aurora: Davies Group, 2011), p. 267. (back)
  3. Gans comes close to an explicit refutation of the equal desirability of the referent and the sign in the chapter on “The Origin of Fiction” in Originary Thinking: Elements of a Generative Anthropology (Stanford: Stanford University Press, 1993), where he distinguishes between the referential conventions of imperative and declarative sentences. Whereas imperative sentences like “Hammer!” require only that we imagine the referent “hammer” without giving any thought to its scenic qualities—its temporal and spatial surroundings—, declarative sentences like “Hammer—no!” imply a cognitive shift as the scenic dimensions of space and time become vital concerns. Instead of simply imagining the referent “hammer,” we respond to declarative sentences like “Hammer—no!” by imaging its absence, an absence that is mitigated by the newly apparent sign “hammer” that now circulates among the community of language users that make up the scene. No longer simply a stand-in for the referent “hammer,” the word “hammer” now exists as a distinct entity in and of itself. As Gans explains, it is the ability of declarative sentences to cause this shift to scenic thinking that makes them the originary sites of fiction, for unlike imperative sentences which are instantly verifiable, declarative sentences are verifiable only after a delay in which the newly founded community of language users turn from sign to referent, assessing the relationship between the two. It is this delay, Gans argues, that is the origin of narrative time and, insofar as it provides reprieve from appetitive violence, of human history as such. And yet, while the distinction between imperative and declarative sentences hints that the scales of desire have tipped decisively toward the sign, they tip back toward the referent when Gans describes the role of descriptive language in fiction. Having noted the importance of narrative time for the ritual maintenance of the otherness of the sacred, which itself becomes synonymous with the “absent” and the “fictive,” Gans makes the following remark regarding descriptive language: “Description makes real the ‘other scene’ on which God subsists before becoming present to the community. The subsequent action appears more plausible as the result of the description of its divine agent. Description intensifies the desire for presence through deferral of the essential movement toward presence [emphasis added]” (p. 98). It is this “intensification of the desire for presence through deferral”—this intensification of the desire to once again tether sign to referent even in what we would expect to be the referent-transcending realm of fiction—that betrays GA’s ambivalent motivations. (back)
  4. Charles Taylor, “Interpretation and the Sciences of Man” in Philosophy and the Human Sciences (Cambridge; New York: Cambridge University Press, 1985), pp. 15-58. (back)
  5. Frank Kermode, The Sense of an Ending: Studies in the Theory of Fiction (London and New York: Oxford University Press, 1968). (back)
  6. Rosalind Krauss signaled the potential for Derridean readings of Twombly’s works in an article written in anticipation of his 1994 retrospective at the Museum of Modern Art, “Cy was here; Cy’s up” in ArtForum International, Vol. 33, No. 1 (September, 1994), pp. 71-74. She then elaborated her Derridean account in her interpretation of Twombly’s 1957 painting “Olympia” in Yve-Alain Bois and Rosalind Krauss, Formless: A User’s Guide (New York and Cambridge: Zone Books, Distributed by MIT Press, 1997), pp. 147-151. Though an important precedent, Krauss’s approach to both Derrida and Twombly differs fundamentally from my own and stands outside the framework of this paper. (back)
  7. Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M. Anscombe, P. M. S. Hacker and Joachim Schulte (Chichester: Wiley-Blackwell, 2009). (back)
  8. On skenographia and skiagraphia see David Summers, Vision, Reflection and Desire in Western Painting (Chapel Hill: The University of North Carolina Press, 2007), pp. 16-42. Many of my comments in this section are indebted to Summers’s insightful account of these terms. (back)
  9. As E.H. Gombrich notes in “The Heritage of Apelles” in The Heritage of Apelles: Studies in the Art of the Renaissance (Ithaca: Cornell University Press, 1976), pp. 3-18, this play of light has more to do with the position of the viewer than with the properties of the robe. As Gombrich explains, viewer-dependent light came to be known as “splendor” and was distinguished from non-viewer dependent light, which came to be known as “lumens.” The suggestion of richness that survives in our own casual use of the word “splendor” offers, perhaps, some sense of the specular satisfaction early Roman viewers may have experienced when standing before works like “Street Musicians.” Being viewer-dependent that satisfaction is aspectival rather than referential. (back)
  10. For a summary of Brunelleschi’s invention, see Martin Kemp, Science of Art: Optical Themes in Western Art from Brunelleschi to Seurat (New Haven: Yale University Press, 1990), pp. 9-53. (back)
  11. As John White notes, this distance was extended in Brunelleschi’s later study of the Palazzo Vecchio, which required both a larger panel and a larger mirror. See John White, The Birth and Rebirth of Pictorial Space(New York: Harper & Row, 1972), pp. 116-117. (back)
  12. For the English translation of this passage, see White, p. 116. For the original Italian, see Antonio Manetti, Vita di Filippo Brunelleschi; edizione critica di Domenico De Robertis (Milano: Il polifilo, 1976), p.59. (back)
  13. Should my description of Renaissance perspective as “cruel” seem unjustified, consider Manetti’s own tragi-comic novella “The Fat Woodcarver” in An Italian Renaissance Sextet: Six Tales in Historical Context, Luigi Ballerini & Massimo Ciavolella, ed. and trans. Murtha Baca (Toronto, Buffalo and London: University of Toronto Press, 1994), the plot of which turns on the protagonist’s being deceived by an elaborate perspectival scheme orchestrated by Brunelleschi himself. Although the story can be read as chastising those viewers who are naïve enough to take Renaissance perspective’s promise of referential fidelity seriously, that reading would still admit that referential fidelity was aimed at, if only to then be ridiculed by initiates. (back)
  14. Giorgio Vasari, Lives of the Artists: A Selection, Vol. I, trans. George Bull (London: Penguin Books, 1987), p. 104. It should also be noted that it was Uccello’s c. 1432 painting The Battle of San Romano that Michael Baxandall used as an example of what he argues was a habit among Renaissance merchants of using perspectival paintings as opportunities to exercise their ability to accurately gauge three-dimensional volumes, a skill that was essential for trade. A merchant-viewer’s ability to quickly gauge the volume of the hexagonal hat in Uccello’s painting, for example, was in essence the same as his ability to gauge of the amount of grain or spice contained in an oddly shaped barrel as it came off a newly docked ship laden with goods. Needless to say, the plausibility of Baxandall’s argument only confirms my suspicion that early modern vision was steeped in conventions of reference. See Painting & Experience in Fifteenth-Century Italy: A Primer in the Social History of Pictorial Style (Oxford: Clarendon Press, 1972), pp. 89-91. (back)
  15. Leon Battista Alberti, On Painting, trans. Cecil Grayson (London: Penguin Books, 1991), pp. 77-78. For an insightful account of Alberti’s aesthetic system, see Joan Gadol, Leon Battista Alberti: Universal Man of the Early Renaissance (Chicago and London: The University of Chicago, 1969). For an account of the velo and other compositional devices used by early modern draftsmen, see Kemp, pp. 163-221. Oddly, in his Civilization of the Renaissance in Italy, Jacob Burckhardt speaks of the Renaissance as a moment when the “veil” of the Middle Ages was torn away, writing: “In the Middle Ages both sides of human consciousness—that which was turned within and that which was turned without—lay dreaming or half awake beneath a common veil. The veil was woven of faith, illusion and childish prepossessions, through which the world and history were seen clad in strange hues. Man was conscious of himself only as a member of a race, people, party, family, or corporation—only through some general category. In Italy this veil first melted into air; an objective treatment and consideration of the State and of all the things of this world became possible.” See Jacob Burckhardt, The Civilization of the Renaissance in Italy, trans. S. G. C. Middlemore (London and New York: Penguin, 1990), p. 9. Of course, from the vantage adopted in this essay, the Renaissance claim that a “veil” separated the viewer from the referent and Burckhardt’s later claim that only once the “veil” of Medieval illusion had been torn away could the referent be seen clearly differ only in a manner of degree. (back)
  16. Alberti, p. 77. (back)
  17. The notion that Twombly’s painting deals with a physical veil is rendered suspicious by the crossing-out of a photograph of a woman’s veil in a “study” for Treatise on the Veil (Second Version), now at the Menil Collection in Houston. Though not impossible, the linking of Twombly’s painting to a physical veil strikes me as uninteresting, especially given this intervention by Twombly himself. (back)
  18. Michael Fried, Art and Objecthood: Essays and Reviews (Chicago and London: University of Chicago Press, 1998). (back)
  19. My thinking in this section about Night Watch and The Raising of Lazarus have benefited from Michael Podro’s discussion of both works in his Depiction (New Haven and London: Yale University Press, 1998), pp. 61-87. Heinrich Wölfflin’s famous description of Baroque vision as “optical” rather than “haptic” may be read as signaling the degree to which aspects became satisfying ends for viewing in the 17th century. As Wölfflin noticed, the 17th-century pursuit of a pure optical space in which atmospheric conditions and the play of light and shadow were allowed to obscure the objects depicted nevertheless introduced a set of new tactile values as viewers began to delight in the surface appearance of things which they largely synthesized themselves, imaginatively conjuring not only individual objects but also whole scenes from out thick pools of brown and black paint. See Heinrich Wölfflin, Principles of Art History: The Problem of the Development of Style in Later Art, trans. M. D. Hottinger (New York: Dover Publications, 1950). The new tactile values that Wölfflin describes strike me as distinctly aspectival, occurring as they do often at the expense of reference. (back)
  20. The notion that the gestural repertoire of Abstract Expressionism strove to be semiotically without remainder is nicely summed up by Rosalind Krauss’s description of Pollock and de Kooning’s marks as executed in a moment of “pure instantaneity…pure release.” Rosalind Krauss, The Optical Unconscious (Cambridge: MIT Press, 1993), p. 7. (back)
  21. Karl Marx, Capital: A Critique of Political Economy, 3 vols. (Harmondsworth: Penguin Books, in association with the New Left Review, 1976-1981). (back)
  22. See for example C. B. Macpherson, Democratic Theory: Essays in Retrieval (Oxford: Clarendon Press, 1973). (back)
  23. Isaiah Berlin, “The Apotheosis of the Romantic Will” in The Crooked Timber of Humanity (Princeton: Princeton University Press, 1991), p. 237. (back)

Works Cited

Alberti, Leon Battista. On Painting, trans. Cecil Grayson (London: Penguin Books, 1991).

Baxandall, Michael. Painting & Experience in Fifteenth-Century Italy: A Primer in the Social History of Pictorial Style (Oxford: Clarendon Press, 1972).

Burckhardt, Jabob. The Civilization of the Renaissance in Italy, trans. S. G. C. Middlemore (London and New York: Penguin, 1990).

Fried, Michael. Art and Objecthood: Essays and Reviews (Chicago and London: University of Chicago Press, 1998).

Gans, Eric. “Differences” in MLN, Vol. 96, No. 4, French Issue (May, 1981), pp. 792-808.

            . Originary Thinking: Elements of a Generative Anthropology (Stanford: Stanford University Press, 1993).

            . A New Way of Thinking: Generative Anthropology in Religion, Philosophy and Art (Aurora: Davies Group, 2011).

Gombrich, E. H. “The Heritage of Apelles” in The Heritage of Apelles: Studies in the Art of the Renaissance (Ithaca: Cornell University Press, 1976), pp. 3-18.

Kemp, Martin. Science of Art: Optical Themes in Western Art from Brunelleschi to Seurat (New Haven: Yale University Press, 1990).

Kermode, Frank. The Sense of an Ending: Studies in the Theory of Fiction (London and New York: Oxford University Press, 1968).

Krauss, Rosalind. The Optical Unconscious (Cambridge: MIT Press, 1993).

            . “Cy was here; Cy’s up” in ArtForum International, Vol. 33, No. 1 (September, 1994), pp. 71-74.

Krauss, Rosalind and Yve-Alain Bois. Formless: A User’s Guide (New York and Cambridge: Zone Books, Distributed by MIT Press, 1997).

Manetti, Antonio. Vita di Filippo Brunelleschi; edizione critica di Domenico De Robertis (Milano: Il polifilo, 1976).

Marx, Karl. Capital: A Critique of Political Economy, 3 vols. (Harmondsworth: Penguin Books, in association with the New Left Review, 1976-1981).

            . “The Fat Woodcarver” in An Italian Renaissance Sextet: Six Tales in Historical Context, Luigi Ballerini & Massimo Ciavolella, ed. and trans. Murtha Baca (Toronto, Buffalo and London: University of Toronto Press, 1994), pp. 171-213.

Podro, Michael. Depiction (New Haven and London: Yale University Press, 1998).

Summers, David. David Summers, Vision, Reflection and Desire in Western Painting (Chapel Hill: The University of North Carolina Press, 2007).

Taylor, Charles. “Interpretation and the Sciences of Man” in Philosophy and the Human Sciences (Cambridge; New York: Cambridge University Press, 1985), pp. 15-58.

van Oort, Richard. “Epistemology and Generative Theory: Derrida, Gans, and the Anthropological Subtext of Deconstruction” in Anthropoetics, Vol. I, No. I (June 1995).

Vasari, Giorgio. Lives of the Artists: A Selection, Vol. I, trans. George Bull (London: Penguin Books, 1987).

White, John. The Birth and Rebirth of Pictorial Space (New York: Harper & Row, 1972).

Wittgenstein, Ludwig. Philosophical Investigations, trans. G. E. M. Anscombe, P. M. S. Hacker and Joachim Schulte (Chichester: Wiley-Blackwell, 2009).

Wölfflin, Heinrich. Principles of Art History: The Problem of the Development of Style in Later Art, trans. M. D. Hottinger (New York: Dover Publications, 1950).