PRINT March 1972

Reflections on the State of Criticism

I DON’T MIND THE POSITIVE work done by formalist critics, but I dislike their interdictory stance—the attitude that tells an artist what he ought not to do, and the spectator what he ought not to see. Preventive esthetics, I call it.

When was it that formalist art criticism first conceived of itself as a prohibitive function? At a certain point we begin to be told that there is only one thing, one alone, to be looked for in art. Thus Baudelaire in his 1863 essay on Delacroix:

A well-drawn figure fills you with a pleasure that is quite alien to the theme. Voluptuous or terrible, this figure owes its charm solely to the arabesque it describes in space. The limbs of a flayed martyr, the body of a swooning nymph, if they are skillfully drawn, connote a type of pleasure in which the theme plays no part, and if you believe otherwise, I shall be forced to think that you are an executioner or a rake.

No one will want to argue the opposite—that you savor a Delacroix for the tang of massacre only. But earlier critics, and I suspect the painter himself, might have allowed one to see a coincidence, a magic identity of pathos and arabesque. Now, at the risk of Baudelaire’s disapproval, this will no longer do: short of confessing that you look to art to gratify prurient or sadistic appetites, you had better see those nude limbs solely as arabesques described in space.

Or compare this from Roger Fry’s Last Lectures:

The sensuality of Indian artists is exceedingly erotic—the leitmotif of much of their sculpture is taken from the more relaxed and abandoned poses of the female figure. A great deal of their art, even their religious art, is definitely pornographic, and although I have no moral prejudices against that form of expression, it generally interferes with esthetic considerations by interposing a strong irrelevant interest which tends to distract both the artist and spectator from the essential purpose of a work of art.1

The provincialism of this passage is characteristic. Indifferent to the intent of the art he condemned, Fry could not see that erotic sculptures on Indian temples are “pornographic” only to the extent that Western images of Crucifixion and martyrdom are “sadistic.” If Fry found the love of Shiva and Parvati more distracting than the agony of Christ, this tells us something of his personal background, but very little about “the essential purpose of art.”

Similar warnings against distracting interference with esthetic form come from most formalist writers. They deplore the tendency artists have of trying to make pictures move you in illegitimate ways—as when Albert C. Barnes says of Michelangelo’s Fall of Man on the Sistine Ceiling that “the effect of movement is vigorous and powerful but tends to be overdramatic.2

An artist who “overdramatizes” his work with expression has wrongheaded values and must pay the price. Serves him right, is the implicit moral of Clement Greenberg’s discovery that Picasso began at one definable moment to lose his certainty as an artist:

A painting done in 1925, the striking Three Dancers, where the will to illustrative expressiveness appears ambitiously for the first time since the Blue Period, is the first evidence of a lessening of this certainty. . . . The Three Dancers goes wrong, not just because it is literary . . . , but because the theatrical placing and rendering of the head and arms of the center figure cause the upper third of the picture to wobble.3

My purpose is not to argue whether the Three Dancers goes wrong. What matters is that an alleged fault of unstable design is blamed upon interference from a “theatrical,” i.e. alien intention—“the will to expressiveness.” This is pure prejudice. Another critic, assuming he found the same fault, might feel that this upper portion had failed through overconcern with formal arrangement. If the artist “went wrong,” who decides which lobe of his Manichaean brain was responsible, the bright half whence pure design is distilled, or that darker side where the will to expressiveness lurks?

Reducing the range of reference has always been the appointed task of formalist thought, but there has been much hard, serious thinking in it. Given the complexity and infinite resonance of works of art, the stripping down of artistic value to the single determinant of formal organization was once — in the 19th century—a remarkable cultural achievement. The attempt was to discipline art criticism in the manner of scientific experiment, through the isolation of a single variable. Art’s “essential purpose”—call it abstract unity of design or whatever prevents buckling and wobbling—was presumed to be abstractable from all works of art. And the whole range of meaning was ruled to be disposable “subject matter,” which at best did no harm but which more commonly burdened the form. In the formalist ethic, the ideal critic remains unmoved by the artist’s expressive intention, uninfluenced by his culture, deaf to his irony or iconography, and proceeds undistracted, programmed like an Orpheus making his way out of hell.

It does not seem to me that the esthetic quality of works of art was ever more than a notional fiction, that it can be experienced as an independent variable, or that it is really isolated by critical judgment. Our experience indicates rather that quality rides the crest of a style, and that when a movement or style as such is resisted, the qualitative differences within that style become unavailable. Ten years ago, all American formalist critics spurned Pop art in toto, and their wholesale rejection admitted no consideration of individual quality or distinction. Whatever merit a Claes Oldenburg may have had remained imperceptible, while the names “Lichtenstein, Rosenquist, Warhol” were run off like the firm name Carson, Pirie, Scott & Co. What interests me here is not to what extent those early denunciations of Pop might need readjustment. The point is that formalist critics could not even confront the question of quality; or were loath to do so lest the exercise of esthetic judgment bestow undue dignity on an aberration. There was to be no voting across party lines.

Contemporary American formalism owes its strength and enormous influence to the professionalism of its approach. It analyzes specific stylistic advances within a linear conception of historic development. Its theoretical justification was furnished by Clement Greenberg whose essay, “Modernist Painting” (1965) reduces the art of a hundred years to an elegant one-dimensional sweep. Following is a brief summary, given as far as possible in the author’s own words.4

“The essence of Modernism lies . . . in the use of the . . . methods of a discipline to criticize the discipline itself—not in order to subvert it, but to entrench it more firmly in its areas of competence.” As Kant used logic to establish the limits of logic, so, argues Greenberg, “modernism criticizes from the inside, through the procedures themselves of that which is being criticized.” How then does this self-criticism proceed? “The task of self-criticism became to eliminate from the effects of each art any . . . effect that might conceivably be borrowed from . . . any other art. Thereby each art would be rendered “pure’” This purity, Greenberg continues, “meant self-definition, and the enterprise of self-criticism in the arts became one of self-definition with a vengeance.” How did this process of self-definition find expression in painting? Pictorial art, Greenberg explains, “criticized and defined itself under Modernism” by “stressing the ineluctable flatness of the support . . . Flatness alone was unique and exclusive to that art . . . and so, Modernist painting oriented itself to flatness as it did to nothing else.”

We may take it for granted that in this system all narrative and symbolic content had to drain out of painting because that kind of content was held in common with literature. The depiction of solid forms was abandoned because “three-dimensionality is the province of sculpture, and for the sake of its own autonomy painting has had above all to divest itself of everything it might share with sculpture.” Recognizable entities had to go because they “exist in three-dimensional space and the barest suggestion of a recognizable entity suffices to call up associations of that kind of space . . . and by doing so, alienates pictorial space from the two-dimensionality which is the guarantee of painting’s independence as an art.”

Whatever else one may think of Greenberg’s construction, its overwhelming effect is to put all painting in series. The progressive flattening of the pictorial stage since Manet “until its backdrop had become the same as its curtain”5 —the approximation of the depicted field to the plane of its material support—this was the great Kantian process of self-definition in which all serious Modernist painting was willy-nilly engaged. The one thing which painting can call its own is color coincident with the flat ground, and its drive towards independence demands withdrawal from anything outside itself and single-minded insistence on its unique property. Even now, 200 years after Kant, any striving for other goals becomes deviationist. Despite the continual emergence in our culture of cross-border disciplines (ecology, cybernetics, psycho-linguistics, biochemical engineering, etc.) the self-definition of advanced painting is still said to require retreat. It is surely cause for suspicion when the drift of third-quarter 20th-century American painting is made to depend on 18th-century German epistemology. Are there no contractionist impulses nearer at hand? Was it Kantian self-definition which led the American woman into what Betty Friedan calls the “Feminine Mystique,” wherein “the only commitment for women is the fulfillment of their own femininity?”6

A graver objection concerns Greenberg’s management of pre-modern art, and this needs discussion because Greenberg’s Modernism defines itself in opposition to the Old Masters. If that opposition becomes unstable, Modernism may have to be redefined by other criteria.

The problem, it seems, hinges on the illusionism of Old Master paintings—the supposed intent of their art to deceive and dissemble. Now, there can be no doubt that there are, and that there have always been, people who look at realistic images as though they were real—but what kind of people? On August 13, 1971, the cover of Life Magazine featured a nude Eve by Albrecht Dürer side by side with the photograph of a modern young woman in dungarees. In the weeks following, close to 3,000 Middle American readers canceled their subscriptions to Life, protesting the shamelessness of the nude. Many took her for real and thought she had stripped for the photographer. But these people, whatever their moral standards, are not the definers of art.

Yet Greenberg’s contrasting definition of Old Master art relies on just this sort of reading. “Realistic, illusionist art had dissembled the medium, using art to conceal art”; whereas “Modernism uses art to call attention to art.”7 It is as though we were told that modern poetry for the first time draws attention to its own process, whereas Dante, Shakespeare, and Keats had merely used meter and rhyme to tell stories. Has Greenberg been taken in by the illusionism of the Old Masters? Obviously not, for he has a remarkably good eye for painting. And in fact his actual observations constantly overturn the polarity he seeks to establish. Thus: “The Old Masters always took into account the tension between surface and illusion, between the physical facts of the medium and its figurative content—but in their need to conceal art with art, the last thing they had wanted was to make an explicit point of this tension.”8 The defining contrast then is not a matter of essence, but only of emphasis; the Old Masters do acknowledge the physical facts of the medium—but not “explicitly.” On closer inspection the difference between their goals and those of Modernist painting becomes even more elusive:

The Old Masters had sensed that it was necessary to preserve what is called the integrity cf the picture plane: that is, to signify the enduring presence of flatness under the most vivid illusion of three-dimensional space. The apparent contradiction involved—the dialectical tension, to use a fashionable but apt phrase—was essential to the success of their art, as it is indeed to the success of all pictorial art. The Modernists have neither avoided nor resolved this contradiction; rather, they have reversed its terms. One is made aware of the flatness of their pictures before, instead of after, being made aware of what the flatness contains. Whereas one tends to see what is in an Old Master painting before seeing it as a picture, one sees a Modernist painting as a picture first. This is, of course, the best way of seeing any kind of picture, Old Master or Modernist, but Modernism imposes it as the only and necessary way, and Modernism’s success in doing so is a success of self-criticism.9

Are we still on firm factual ground? The “objective” difference between Old Master and Modernist reduces itself to subjective tendencies in the viewer. It is he who in looking at Old Master paintings tends to see the illusion “before seeing it as a picture.” But what if he doesn’t? What if he sees a Giotto, a Poussin, or a Fragonard as a picture first, habitually screening out the deep space indications until he has seen the surface disposition of its formal elements? Does an Old Master painting forego its Old Master status if it is seen in primary flatness and only secondly as a vivid illusion? Consider that typical Old Master expression, the rapid sketch (Fig. 1). Does Rembrandt’s drawing become Modernist if its pen strokes and bister washes emerge for us before, or along with, the old lady’s image? It seems to me that the last thing this draftsman wants is to dissemble his medium, or conceal his art; what he wants, and gets, is precisely a tension, made fully explicit, between the figure evoked and the physicality of paper, penstroke, and ink. And yet, in terms of style, such a sketch as this is integral to Old Master art. It merely dramatizes the quality that enables Baudelaire to see a Delacroix as nothing but arabesques.

And, contrariwise, what if the viewer tends to see Modernist paintings as spatial abstractions of landscape? The sculptor Don Judd complains that New York School paintings of the 1950s keep him intensely aware of what their flatness contains—“airiness” and “illusionistic space.” He said recently:

Rothko’s whole way of working depended on a good deal of illusionism. It’s very aerial. The whole thing is about’ areas floating in space. Compared to Newman there is distinctly a certain depth. But I finally thought that all painting was spatially illusionistic.10

Where does this leave us? The difference between Old Master and Modernist is not, after all, between illusion and flatness; it turns out that both are present in each. But if the difference is in the order in which these two presences are perceived, then do the subjective approaches of Baudelaire and Judd reverse the distinction between historic and Modern art?

Greenberg is fully conscious of the airy illusionism observed by Judd in Modernist painting. But though open atmospheric effects, such as are found in Rothko or Jules Olitski, clearly deny and dissemble the picture’s material surface, he nevertheless finds them congruent with painting’s self-definition because the illusion conveyed is “visual,” rather than tactile or kinesthetic. And visual art should, to conform with Kantian self-criticism and scientific consistency, “confine itself exclusively to what is given in visual experience.”

Where the Old Masters created an illusion of space into which one could imagine oneself walking, the illusion created by a Modernist is one into which one can look, can travel through, only with the eye.11

The difference, then, reduces itself to distinct kinds of spatial illusion, but this last saving distinction is one which defines “Modernism” by preindustrial standards of locomotion. How, in what kind of painted space, do you let yourself roam? Greenberg apparently can imagine himself trudging through a Rembrandtesque gloom, but he cannot conceive journeying through an Olitski. Do we need to be reminded that in an age of space travel a pictorial semblance of open void is just as inviting to imaginary penetration as the pictorial semblance of a receding landscape was formerly to a man on foot? Are we now to define Modernist painting against a Kantian concept of transportation? Greenberg’s theoretical schema keeps breaking down because it insists on defining modern art without acknowledgment of its content, and historical art without recognizing its formal self-consciousness.

All major painting, at least of the last 600 years, has assiduously “called attention to art.” Except for tour de force demonstrations and special effects, and before their tradition collapsed in 19th-century academicism, the Old Masters always took pains to neutralize the effect of reality, presenting their make-believe worlds, as it were, between quotation marks. The means they chose were, of course, those of their day, not of ours; and often their careful controls are annulled by our habit of lifting a partial work from its setting—transposing a detached fresco or predella panel into the category of easel painting. But a dramatic narrative painted by Giotto resembles neither a late 19th-century easel painting nor a movie screen. When it is not wrenched from its context (as in most art history books), it works within a wall system, each wall supporting multiple scenes set between elaborate framing bands, within which, in turn, other scenes on different scales are descried. You are never allowed to see one illusion, one sky, or one fictive horizon alone. You are shown simultaneous and incompatible systems whose juxtaposition cancels or checks the illusion. Similarly, the Sistine Ceiling when seen in its entirety: the work is a battleground for local illusion, counterillusion, and emphasized architectural surface—art turning constantly back on itself.

This is a functional multiplicity, and it occurs even in apparently single works. Look at those cornered, diminutive prophets in our Fig. 2, leaning out of the picture plane: their eager demonstrative gestures towards the Christ turn the illusionist scene of the Crucifixion back into a picture of it—complete with its own patterned frame. Or take an engraved baroque portrait (Fig. 3). The depicted man is a paper paste-up on a dog-eared sheet on a flat wall, and the whole stratigraphy of it is exhibited in the changing density of horizontal striations. The artists here do exactly what Greenberg admires as a significant find in a crucial Cubist picture by Braque: “[Braque] discovered that trompe-l’oeil Could be used to undeceive as well as to deceive the eye. It could be used, that is, to declare as well as to deny the actual surface.”12

Where the Old Masters seem to dissolve the picture plane to gain an unambiguous illusion of depth, they usually have a special objective in mind, an objective understood and shared by the viewer. Michelangelo’s Last Judgment, unlike the Ceiling, obliterates the supporting wall plane so that the vision of a Christ “come to judge the quick and the dead” gives immediate urgency to the words of the Creed. Caravaggio’s pictures, whether erotic or religious in their address, were similarly intended to induce a penetrating experience. But their relentless, surface-dissolving illusionism was largely repudiated by the Old Masters. Until the 19th century, the kind of painting which utterly broke the consistency of the surface remained a special, even exceptional resource of Old Master art.

The more realistic the art of the Old Masters became, the more they raised internal safeguards against illusion, ensuring at every point that attention would remain focused upon the art.

They did it by radical color economies, or by eerie proportional attenuation; by multiplication of detail, or by preternatural beauty. They did it—as modern films do with spliced footage taken from older movies—by quotations and references to other art; quotation being a surefire means of shunting the ostensible realism of a depicted scene back into art.

They did it by abrupt internal changes of scale, or by shifting reality levels—as when Raphael’s Expulsion of Heliodorus inserts a group of contemporaries in modern dress as observers of the Biblical scene; or by overlapping reality levels, as when a frescoed battle scene on a Vatican wall curls up at the edges to become a fake tapestry—two self-confounding illusions which call both into question and recall both to art.

The “recall to art” may be engineered by the subject matter itself. In a Dutch interior, the backview of a personage who draws a curtain aside to look at a painting on the far wall acts as my alter ego, doing what I am doing and reminding me (in case I missed the point of the picture’s immense ebony frame) that I too am looking at a flat object. Better still, such 17th-century interiors as Velázquez’ Ladies in Waiting often juxtapose a doorway or window view with a framed painting, and, next to that, a mirror filled with a reflection. These three kinds of image serve as an inventory of the three possible roles assignable to a picture plane. The windowpane or proscenium effect refers to what lies behind it, the looking glass’ refers to what lies before, while the pigmented surface asserts itself; and all three are paraded in sequence. Such pictures soliloquize about the capacities of the surface and the nature of illusion itself.

Again and again, in so-called illusionist art, it is illusionism that is under discussion, the art “calling attention to art” in perfect self-critical consciousness. And this is why the Old Masters are forever inventing interferences with spatial recession. They do not merely “take account” of the tension between surface and depth, as if for the sake of decorative coherence, while reserving their thrust for the depiction of depth. Rather they maintain an explicit, controlled, ever-visible dualism. 15th-century perspective was not a surface-denying illusion of space, but the symbolic form of space as an intelligible coordinate surface pattern. Good illusionist painting not only anchors depth to the plane; it is almost never without built-in devices designed to suspend the illusion, and the potency of these devices depends—like the appreciation of counterpoint or of puns—on the spectator’s ability to register two things in concert, to receive both the illusion and the means of illusion at once.

Some of the Old Masters overruled the apparent perspective by dispersing identical color patches as an allover carpet spread (Pieter Bruegel, for instance). Some worked with chromatic dissonances to weave a continuous surface shimmer like mother-of-pearl. Many—from Titian onward—insured their art against realism by the obtrusive calligraphy of the brush—laying a welter of brushstrokes upon the surface to call attention to process. Some contrived implausible contradictions within the field, as when the swelling bulk of a foreshortened form is collapsed and denied the spatial ambience to house it. All of them counted on elaborate framing as an integral part of the work (“advertising the literal nature of the support,” as Greenberg says of collage)—so that the picture, no matter how deep its illusionism, turned back into a thing mounted there like a gem. It was Michelangelo himself who designed the frame of the Doni Madonna, an element essential to the precious-mirror effect of its surface (Fig. 4).

Greenberg wants all Old Master and Modernist painters to reduce their differences to a single criterion, and that criterion as mechanistic as possible—either illusionistic or flat. But what significant art is that simple? Have you ever asked how deep the thrones of the Sistine Prophets and Sibyls are? Perfectly shallow if you glance across the whole sequence; but, as all the early copies reveal, they run more than ten feet deep as soon as you focus on one alone (Figs. 5, 6, 7). Perspective illusionism and anatomic foreshortening sustain a ceaseless optical oscillation.

“The abiding effect is a constant shuttling between surface and depth, in which the depicted flatness is ’infected’ by the undepicted. Rather than being deceived, the eye is puzzled; instead of seeing objects in space, it sees nothing more than—a picture.”13 These words, in which Greenberg describes Cubist collage, apply equally well to Michelangelo’s Ceiling and to thousands of Old Master works. They describe the effect of a not untypical 14th-century manuscript page (Fig. 8): Missus est Gabriel angelus. Three reality levels oscillate in, and compete for, that capital “M”: an arcade opening on a bed chamber; a trellis for ivy ornament; and a letter at the head of a word. All three at once. The eye is puzzled; instead of seeing objects in space it sees a picture.14

The notion that Old Master paintings in contrast to modern dissemble the medium, conceal the art, deny the surface, deceive the eye, etc., is only true for a viewer who looks at the art like those ex-subscribers to Life Magazine. The distinction a critic makes between Modernist self-analytical and Old Master-representational refers less to the works compared than to his own chosen stance—to be analytic about the one and polemically naive about the other.

It is poor practice, when modern art is under discussion, to present the Old Masters as naively concerned with eyefooling trickery, while reserving for modern art both the superior honesty of dealing with the flat plane of painting and the maturer intellectual discipline of self-analysis. All important art, at least since the Trecento, is preoccupied with self-criticism. Whatever else it may be about, all art is about art. All original art searches its limits, and the difference between the old and the modernist is not in the fact of self-definition but in the direction which this self-definition takes. This direction being part of the content.

At this point Greenberg might answer that self-definition does not deserve its name unless it aims at purity, and that this in turn requires stripping painting down to its irreducible essence, i.e. the coincidence of flattened color with its material support. I reply that this mistakes a special case for a necessity. The process of painting’s self-realization can go either way. For Jan van Eyck, for example, the self-realization of painting is not reductive but expansive. He turns to the sculptor and says, “Anything you can do I can do better”; then to the goldsmith—“Anything you can do I can do better”; and so to the architect. He redesigns everything in the flat and even banishes metallic gold to create the effect of it—like Manet—in pure color and light. Anything anybody can do, painting does better—and that’s where, for van Eyck, painting realizes itself—discovering its autonomy literally in its ability to do without external aid.

Art’s perpetual need to redefine the area of its competence by testing its limits takes many forms. Not always does it probe in the same direction. Jacques Louis David’s ambition to make art a force of national moral leadership is as surely a challenge to the limits of art as is Matisse’s elimination of tonal values. At one historical moment painters get interested in finding out just how much their art can annex, into how much nonart it can venture and still remain art. At other times they explore the opposite end to discover how much they can renounce and still stay in business. What is constant is art’s concern with itself, the interest painters have in questioning their operation. It is a provincialism to make the self-critical turn of mind the sufficient distinction of Modernism; and once it is understood as not its peculiar distinction, then the specific look of contemporary abstract art—its object quality, its blankness and secrecy, its impersonal or industrial look, its simplicity and tendency to project a stark minimum of decisions, its radiance and power and scale—these become recognizable as a kind of content—expressive, communicative, and eloquent in their own way.

The Corporate Model of Developing Art.

It is astonishing how often recent abstract American painting is defined and described almost exclusively in terms of internal problem solving. As though the strength of a particular artist expressed itself only in his choice to conform with a set of existent professional needs and his inventiveness in producing the answers. The dominant formalist critics today tend to treat modern painting as an evolving technology wherein at any one moment specific tasks require solution—tasks set for the artist as problems are set for researchers in the big corporations. The artist as engineer and research technician becomes important insofar as he comes up with solutions to the right problem. How the choice of that problem coincides with personal impulse, psychological predisposition, or social ideal is immaterial; the solution matters because it answers a problem set forth by a governing technocracy.

In America this corporate model of artistic evolution appears full-blown by the mid-1920s. It inhabits the formalist doctrine that Painting aspires towards an ever-tightening synthesis of its design elements. The theory in its beginnings was fairly simple. Suppose a given painting represents a reclining nude; and suppose the figure outlined with a perceptible contour. Within that contour lies a distinct shape. That shape is of a certain color, and the color—modulated from light to dark or from warm to cool—reflects a specific quantity or kind of light. We have then four formal elements—line, shape, color, and light—which can be experienced and thought of as separate and distinct. Now, it is argued, the test of significantly advanced painting will be the progressive obliteration of these distinctions. The most successful picture will so synthesize the means of design that line will be no longer separable from shape, nor shape from color, nor color from light. A working criterion, easily memorized and applied. It tells you not necessarily which picture is best, but which is in line to promote the overall aspiration of Painting—this alignment being a sine qua non of historic importance. By this criterion, the painter of the Sistine Ceiling is, with due respect, relegated to one of the byways of Painting since his inventions, for all their immediate interest, do not ultimately promote the direction in which painting must go; Michelangelo’s forms are “realized in a sculptural rather than a pictorial manner.”15 Indeed, the elements of Michelangelo’s depictions are remarkable for separability—specific shapes sharply delineated by bounding lines, tinted by local color, modulated by chiaroscuro. Though Michelangelo will, I am convinced, be emerging within the next several years as one of the most original colorists of all time, by the criteria enunciated above he fails to contribute—as did Titian’s coloristic diffusion—to the synthesis of the means of design. For the critic-collector Albert C. Barnes he remains a dead end, whereas the course initiated by Titian leads irresistibly to its culmination in Renoir and William Glackens.

This single criterion for important progressive art, moving as by predestination towards utter homogeneity of the elements of design is still with us, now considerably more analytical, more prestigious than ever, and celebrating its latest historical denouement in the triumph of color field painting.

In formalist criticism, the criterion for significant progress remains a kind of design technology subject to one compulsive direction: the treatment of “the whole surface as a single undifferentiated field of interest.” The goal is to merge figure with ground, integrate shape and field, eliminate foreground-background discontinuities; to restrict pattern to those elements (horizontals or verticals) that suggest a symbiotic relationship of image and frame; to collapse painting and drawing in a single gesture, and equate design and process (as Pollock’s drip paintings do, or Morris Louis’ veils); in short, to achieve the synthesis of all separable elements of painting, preferably—but this is a secondary consideration—without that loss of incident or detail which diminishes visual interest.

There is, it seems to me, a more thoroughgoing kind of synthesis involved in this set of descriptions—the leveling of end and means. In the criticism of the relevant paintings there is rarely a hint of expressive purpose, nor recognition that pictures function in human experience. The painter’s industry is a closed loop. The search for the holistic design is self-justified and self-perpetuating. Whether this search is still the exalted Kantian process of self-criticism seems questionable; the claim strikes me rather as a remote intellectual analogy. And other analogies suggest themselves, less intellectual, but closer to home. It is probably no chance coincidence that the descriptive terms which have dominated American formalist criticism these past 50 years, run parallel to the contemporaneous evolution of the Detroit automobile. Its ever-increasing symbiosis of parts—the ingestion of doors, running boards, wheels, fenders, spare tires, signals, etc., in a one-piece fuselage—suggests, with no need for Kant, a similar drift towards synthesizing its design elements. It is not that the cars look like the paintings. What I am saying here relates less to the pictures themselves than to the critical apparatus that deals with them. Pollock, Louis, and Noland are vastly different from each other; but the reductive terms of discussion that continually run them in series are remarkably close to the ideals that govern the packaging of the all-American engine. It is the critics’ criterion far more than the painters’ work which is ruled by a streamlined efficiency image.

But the reference to industrial ideals can serve to focus on certain distinctions within art itself. If, for instance, we question the work of the three painters just mentioned from the viewpoint of expressive content, they immediately separate out. There is obviously no affinity for industrialism in Pollock or Louis, but it does characterize an important aspect of the younger man’s work. His 30-foot-long stripe paintings, consisting of parallel color bands, embody, beyond the subtlety of their color, principles of efficiency, speed, and machine-tooled precision which, in the imagination to which they appeal, tend to associate themselves with the output of industry more than of art (Fig. 9). Noland’s pictures of the late ’60s are the fastest I know.

The painter Vlaminck used to say that he wished to make pictures which would be readable to a motorist speeding by in an automobile. But Vlaminck’s belated expressionism could never realize such an ideal. His palette-knifed snowscapes lacked every access to his ostensible goal. They possessed neither the scale, the format, the color radiance, nor even the appropriate subject matter: good motorists look for signals and signs, not at messages from a painter’s easel. Vlaminck’s statement remains naive because it is essentially idle. But there is nothing naive in Noland’s determination to produce, as he put it, “‘one-shot’ paintings perceptible at a single glance.” I quote from a recent article by Barbara Rose, who continues: “To achieve maximum immediacy, Noland was ready to jettison anything interfering with the most instantaneous communication of the image.”16

Noland’s stated objective during the 1960s confirms what his pictures reveal—an idealization of efficient speed and, implicitly, a conception of the humanity at whom his “one-shots” are aimed. The instantaneity which his pictures convey implies a different psychic orientation, a revised relationship with the spectator. Like all art that ostensibly thinks only about itself, it creates its own viewer, projects its peculiar conception of who, what, and where he is.

Is he a man in a hurry? Is he at rest or in motion? Is he one who construes or one who reacts? Is he a man alone—or a crowd? Is he a human being at all—or a function, a specialized function or instrumentality, such as the one to which Rauschenberg’s Chairs (1968) reduced the human agent. (A room-size transparent screen whose illumination was electronically activated by sound; the visibility of the chairs which constituted the image depending on the noises made by the spectator—his footsteps when entering, his coughing or speaking voice. One felt reduced to the commodity of a switch.) I suspect that all works of art or stylistic cycles are definable by their built-in idea of the spectator. Thus, returning once more to the Pollock-Louis-Noland procession, the younger man, who separates himself from his elders by the criterion of industrial affinity, parts from them again by his distinct view of the viewer.

Considerations of “human interest” belong in the criticism of modernist art not because we are incurably sentimental about humanity, but because it is art we are talking about. And it appears to me that even such professional technicalities as “orientation to flatness” yield to other criteria as soon as the picture is questioned not for its internal coherence, but for its orientation to human posture.

What is “pictorial flatness” about? Obviously it does not refer to the zero curvature of the physical plane—a cat walking over pictures by Tie-polo and Barnett Newman gets the same support from each one. What is meant of course is an ideated flatness, the sensation of flatness experienced in imagination. But if that’s what is meant, is there anything flatter than the Olympia (1950) of Dubuffet (Fig. 10)? If flatness in painting indicates an imaginative experience, then the pressed-leaf effect, the graffito effect, the scratched-gravel or fossil-impression effect of Dubuffet’s image dramatizes the sensation of flatness far beyond the capacity, or the intention, of most color field painting. But in fact, these different “flatnesses” are not even comparable. And the word “flat” is too stale and remote for the respective sensations touched off by the visionary color veils of Morris Louis (Fig. 11) and the bedrock pictographs of Dubuffet. Nor need flatness be an end product at all—as Jasper Johns demonstrated in the mid-1950s, when his first Flags and Targets relegated the whole maintenance problem of flatness to “subject matter.” However atmospheric his brushwork or play of tonalities, the depicted subject ensured that the image stayed flat So then one discovers that there are recognizable entities, from flags even to female nudes, which can actually promote the sensation of flatness.

This discovery is still fairly recent, and it is not intelligible in terms of design technology. It demands consideration of subject and content, and, above all, of how the artist’s pictorial surface tilts into the space of the viewer’s imagination.

The Flatbed Picture Plane.

I borrow the term from the flatbed printing press—“a horizontal bed on which a horizontal printing surface rests” (Webster). And I propose to use the word to describe the characteristic picture plane of the 1960s—a pictorial surface whose angulation with respect to the human posture is the precondition of its changed content.

It was suggested earlier that the Old Masters had three ways of interpreting the picture plane. But one axiom was common to all three interpretations, and it remained operative in the succeeding centuries, even through Cubism and Abstract Expressionism: the conception of the picture as representing a world, some sort of worldspace which reads on the picture plane in correspondence with the erect human posture. The top of the picture corresponds to where we hold our heads aloft; while its lower edge gravitates to where we place our feet. Even in Picasso’s Cubist collages, where the Renaissance worldspace concept almost breaks down, there is still a harking back to implied acts of vision, to something that was once actually seen.

A picture that harks back to the natural world evokes sense data which are experienced in the normal erect posture. Therefore the Renaissance picture plane affirms verticality as its essential condition. And the concept of the picture plane as an upright surface survives the most drastic changes of style. Pictures by Rothko, Still, Newman, de Kooning, and Kline are still addressed to us head to foot—as are those of Matisse and Miró). They are revelations to which we relate visually as from the top of a columnar body; and this applies no less to Pollock’s drip paintings and the poured veils and unfurls of Morris Louis. Pollock indeed poured and dripped his pigment upon canvases laid on the ground, but this was an expedient. After the first color skeins had gone down, he would tack the canvas on to a wall—to get acquainted with it, he used to say, to see where it wanted to go. He lived with the painting in its uprighted state, as with a world confronting his human posture. It is in this sense, I think, that the Abstract Expressionists were still nature painters. Pollock’s drip paintings cannot escape being read as thickets; Louis’ veils acknowledge the same gravitational force to which our being in nature is subject.17

But something happened in painting around 1950—most conspicuously (at least within my experience) in the work of Robert Rauschenberg and Dubuffet. We can still hang their pictures—just as we tack up maps and architectural plans, or nail a horseshoe to the wall for good luck. Yet these pictures no longer simulate vertical fields but opaque flatbed horizontals. They no more depend on a head-to-toe correspondence with human posture than a newspaper does. The flatbed picture plane makes its symbolic allusion to hard surfaces such as tabletops, studio floors, charts, bulletin boards—any receptor surface on which objects are scattered, on which data is entered, on which information may be received, printed, impressed—whether coherently or in confusion. The pictures of the last 15 to 20 years insist on a radically new orientation, in which the painted surface is no longer the analogue of a visual experience of nature but of operational processes.

To repeat: it is not the actual physical placement of the image that counts. There is no law against hanging a rug on a wall, or reproducing a narrative picture as a mosaic floor. What I have in mind is the psychic address of the image, its special mode of imaginative confrontation, and I tend to regard the tilt of the picture plane from vertical to horizontal as expressive of the most radical shift in the subject matter of art, the shift from nature to culture.

A shift of such magnitude does not come overnight, nor as the feat of one artist alone. Portents and antecedents become increasingly recognizable in retrospect—Monet’s Nymphéas or Mondrian’s transmutation of sea and sky into signs plus and minus. And the picture planes of a synthetic Cubist still life or a Schwitters collage suggest like-minded reorientations. But these last were small objects, the “thingness” of them was appropriate to their size. Whereas the event of the 1950s was the expansion of the work-surface picture plane to the man-sized environmental scale of Abstract Expressionism. Perhaps Duchamp was the most vital source. His Large Glass begun in 1915, his Tu m’ of 1918, is no longer the analogue of a world perceived from an upright position, but a matrix of information conveniently placed in a vertical situation. And one detects a sense of the significance of a 90-degree shift in relation to a man’s posture even in some of those Duchamp “works“ that once seemed no more than provocative gestures: the Hatrack laid on the ground, and the famous Urinal tilted up like a monument.

But on the New York art scene the great shift came in Rauschenberg’s work of the early 1950s. Even as Abstract Expressionism was celebrating its triumphs, he proposed the flatbed or work-surface picture plane as the foundation of an artistic language that would deal with a different order of experience. The earliest work which Rauschenberg admits into his canon—White Painting with Numbers (Fig. 12)—was painted in 1949 in a life class at the Art Students’ League, the young painter turning his back on the model. Rauschenberg’s picture, with its cryptic meander of lines and numbers, is a work surface that cannot be construed into anything else. Up and down are as subtly confounded as positive-negative space or figure-ground differential. You cannot read it as masonry, nor as a system of chains or quoins, and the written cyphers read every way. Scratched into wet paint, the picture ends up as a verification of its own opaque surface.

In the year following, Rauschenberg began to experiment with objects placed on blueprint paper and exposed to sunlight. Already then he was involved with the physical material of plans; and in the early 50s used newsprint to prime his canvases—to activate the ground, as he put it—so that his first brushstroke upon it took place in a gray map of words.

In retrospect the most clownish of Rauschenberg’s youthful pranks take on a kind of stylistic consistency. Back in the ’50s, he was invited to participate in an exhibition on the nostalgic subject of “nature in art”—the organizers hoping perhaps to promote an alternative to the new abstract painting. Rauschenberg’s entry was a square patch of growing grass held down with chicken wire, placed in a box suitable for framing and hung on the wall. The artist visited the show periodically to water his piece—a transposition from nature to culture through a shift of 90 degrees. When he erased a de Kooning drawing, exhibiting it as “Drawing by Willem de. Kooning erased by Robert Rauschenberg,” he was making more than a multifaceted psychological gesture; he was changing—for the viewer no less than for himself—the angle of imaginative confrontation; tilting de Kooning’s evocation of a world-space into a thing produced by pressing down on a desk.

The paintings he made towards the end of that decade included intrusive nonart attachments: a pillow suspended horizontally from the lower frame (Canyon, 1959); a grounded ladder inserted between the painted panels which made up the picture (Winter Pool, 1959–60; Fig. 13); a chair standing against a wall but ingrown with the painting behind (Pilgrim, 1960). Though they hung on the wall, the pictures kept referring back to the horizontals on which we walk and sit, work, and sleep.

When in the early 1960s he worked with photographic transfers, the images—each in itself illusionistic—kept interfering with one another, intimations of spatial meaning forever canceling out to subside in a kind of optical noise. The waste and detritus of communication—like radio transmission with interference; noise and meaning on the same wavelength, visually on the same flatbed plane.

The picture plane, as in the enormous canvas called Overdraw (1963), could look like some garbled conflation of controls system and cityscape, suggesting the ceaseless flow of urban message, stimulus, and impediment. To hold all this together, Rauschenberg’s picture plane had to become a surface to which anything reachable-thinkable would adhere. It had to be whatever a billboard or dashboard is, and everything a projection screen is, with further affinities for anything that is flat and worked over—palimpsest, canceled plate, printer’s proof, trial blank, chart, map, aerial view. Any flat documentary surface that tabulates information is a relevant analogue of his picture plane—radically different from the transparent projection plane with its optical correspondence to man’s visual field. And it seemed at times that Rauschenberg’s work surface stood for the mind itself—dump, reservoir, switching center, abundant with concrete references freely associated as in an internal monologue—the outward symbol of the mind as a running transformer of the external world, constantly ingesting incoming unprocessed data to be mapped in an overcharged field.

To cope with his symbolic program, the available types of pictorial surface seemed inadequate; they were too exclusive and too homogeneous. Rauschenberg found that his imagery needed bedrock as hard and tolerant as a workbench. If some collage element, such as a pasted-down photograph, threatened to evoke a topical illusion of depth, the surface was casually stained or smeared with paint to recall its irreducible flatness. The “integrity of the picture plane”—once the accomplishment of good design—was to become that which is given. The picture’s “flatness” was to be no more of a problem than the flatness of a disordered desk or an unswept floor. Against Rauschenberg’s picture plane you can pin or project any image because it will not work as the glimpse of a world but as a scrap of printed material. And you can attach any object, so long as it beds itself down on the work surface. The old clock in Rauschenberg’s 1961 Third Time Painting (Fig. 14) lies with the number 12 on the left, because the clock face properly up-righted would have illusionized the whole system into a real vertical plane—like the wall of a room, part of the given world. Or, in the same picture, the flattened shirt with its sleeves outstretched—not like wash on a line, but with paint stains and drips holding it down—like laundry laid out for pressing. The consistent horizontality is called upon to maintain a symbolic continuum of litter, workbench, and data-ingesting mind.

Perhaps Rauschenberg’s profoundest symbolic gesture came in 1955 when he seized his own bed, smeared paint on its pillow and quilt coverlet, and uprighted it against the wall. There, in the vertical posture of “art,” it continues to work in the imagination as the eternal companion of our other resource, our horizontality, the flat bedding in which we do our begetting, conceiving, and dreaming. The horizontality of the bed relates to “making” as the vertical of the Renaissance picture plane related to seeing.

I once heard Jasper Johns say that Rauschenberg was the man who in this century had invented the most since Picasso. What he invented above all was, I think, a pictorial surface that let the world in again. Not the world of the Renaissance man who looked for his weather clues out of the window; but the world of men who turn knobs to hear a taped message, “precipitation probability ten percent tonight,” electronically transmitted from some windowless booth. Rauschenberg’s picture plane is for the consciousness immersed in the brain of the city.

The flatbed picture plane lends itself to any content that does not evoke a prior optical event. As a criterion of classification it cuts across the terms “abstract” and “representational,” Pop and Modernist. Color field painters such as Ken Noland, Frank Stella, and Ellsworth Kelly, whenever their works suggest a reproducible image, seem to work with the flatbed picture plane, i.e., one which is man-made and stops short at the pigmented surface; whereas Pollock’s and Louis’ pictures remain visionary, and Frankenthaler’s abstractions, for all their immediate modernism, are—as Lawrence Alloway recently put it—“a celebration of human pleasure in what is not man-made.”18

Insofar as the flatbed picture plane accommodates recognizable objects, it presents them as man-made things of universally familiar character. The emblematic images of the early Johns belong in this class; so, I think, does most of Pop art. When Roy Lichtenstein in the early ’60s painted an Air Force officer kissing his girl goodbye, the actual subject matter was the mass-produced comic book image; Ben-day dots and stereotyped drawing ensured that the image was understood as a representation of printed matter. The pathetic humanity that populate Dubuffet’s pictures are rude man-made graffiti, and their reality derives both from the material density of the surface and from the emotional pressure that guided the hand. Claes Oldenburg’s drawing, to quote his own words, “takes on an ‘ugliness’ which is a mimicry of the scrawls and patterns of street graffiti. It celebrates irrationality, disconnection, violence and stunted expression—the damaged life forces of the city street.”19

And about Andy Warhol, David Antin once wrote a paragraph which I wish I had written:

In the Warhol canvases, the image can be said to barely exist. On the one hand this is part of his overriding interest in the “deteriorated image,” the consequence of a series of regressions from some initial image of the real world. Here there is actually a series of images of images, beginning from the translation of the light reflectivity of a human face into the precipitation of silver from a photosensitive emulsion, this negative image developed, re-photographed into a positive image with reversal of light and shadow, and consequent blurring, further translated by telegraphy, engraved on a plate and printed through a crude screen with low-grade ink on newsprint, and this final blurring and silkscreening in an imposed lilac color on canvas. What is left? The sense that there is something out there one recognizes and yet can’t see. Before the Warhol canvases we are trapped in a ghastly embarrassment. This sense of the arbitrary coloring, the nearly obliterated image and the persistently intrusive feeling. Somewhere in the image there is a proposition. It is unclear.20

The picture conceived as the image of an image. It is a conception which guarantees that the presentation will not be directly of a worldspace, and that it will nevertheless admit any experience as the matter of representation. And it readmits the artist in the fullness of his human interests, as well as the artist-technician.

The all-purpose picture plane underlying this post-Modernist painting has made the course of art once again nonlinear and unpredictable. What I have called the flatbed is more than a surface distinction if it is understood as a change within painting that changed the relationship between artist and image, image and viewer. Yet this internal change is no more than a symptom of changes which go far beyond questions of picture planes, or of painting as such. It is part of a shakeup which contaminates all purified categories. The deepening inroads of art into nonart continue to alienate the connoisseur as art defects and departs into strange territories, leaving the old standby criteria to rule an eroding plain.

Leo Steinberg



1. Published posthumously, New York, 1939, p. 150.

2. Albert C. Barnes, The Art in Painting, N.Y., 1925, p. 408.

3. Clement Greenberg, “Picasso at Seventy-Five,” Art and Culture (1961), Boston, 1965, p. 62.

4. Greenberg, “Modernist Painting,” The New Art, ed. Gregory Battcock, New York, 1966, pp. 101ff. The essay first appeared in Art and Literature, Spring, 1965.

5. Greenberg, “Abstract, Representational and so forth,” Art and Culture, p. 136.

6. Betty Friedan, The Feminine Mystique, New York, 1963, p. 37. Cf. her analysis (pp. 29–30) of the ideal American woman as presented in an issue of McCalls, July, 1960—“pared down to pure femininity, unadulterated.”

7. Greenberg, “Modernist Painting,” p. 102.

8. Greenberg, “Cézanne,” Art and Culture, p. 53.

9. Greenberg, “Modernist Painting,” pp. 103-104.

10. Don Judd, Art News, October, 1971, p. 60.

11. Greenberg, “Modernist Painting,” p. 107.

12. Greenberg, “Collage,” Art and Culture, p. 72.

13. Greenberg, “Collage,” pp. 73-74.

14. The deliberate exploitation of surface-to-depth oscillation characterizes all major painting; it is the inexhaustible resource of the art. But the degree to which the resultant duality registers on the viewer’s attention depends on the culture and the set of expectancies he brings to his appreciation. He mistakes the goal of the Old Masters if he imagines them aiming for that near absolute dissolution of the picture plane which distinguishes late 19th-century academic painting.

To take an outstanding example of illusionist Old Master art—Velzázquez’ Menippus: the heavy impasto that molds the scrolls and books in the foreground tells you explicitly where the paint is. But the jug and bench in the “background,” where the raw canvas appears barely stained by a thin wash of pigment, says—“this is where the canvas is.” And the palpable mystery of the painting is the old cynic’s material presence inserted in the non-dimensional film between canvas and paint. No painting was ever more self-defining than this.

15. Barnes, Art in Painting, p. 408. For the author’s insistence on synthesizing the elements of design, see pp. 55, 61, 67f.

16. Barbara Rose, “Quality in Louis,” Artforum, October, 1971, p. 65.

17. The fact that some Louis paintings may actually hang upside down is immaterial. Their space will still .be experienced as gravitational, whether the image suggests falling veils or shooting flames.

18. Lawrence Alloway, “Frankenthaler as Pastoral,” Art News, November, 1971, p. 68.

19. Quoted in Eila Kokkinen’s review of Claes Oldenburg: Drawings and Prints, Arts, November, 1969, p. 12.

20. David Antin, “Warhol: The Silver Tenement,” Art News, Summer, 1966, p. 58.