June 3, 2007
Envisioning Abstraction: The Simultaneity of Robert Delaunay’s First Disk
By Hughes, Gordon
Of all of the movements lucky enough to have been caught in the tangles of Alfred Barr's spider-web chart of modern art (Fig. 1), only one, Orphism-Guillaume Apollinaire's misbegotten term and attempt to unite the post-Cubist abstraction of Robert Delaunay, Marcel Duchamp, Fernand Lger, and Francis Picabia-goes nowhere.1 That the diagrammatic arrow leading from Cubism to Orphism ends in what Barr sees as the sole cul-de-sac in twentieth-century art is surprising given that the relation between Cubism and abstract art was the ostensible motive for Barr's flow chart. Though Orphism is important enough to warrant mapping, the exact nature of its significance is unclear within the logic of Barr's schema. For importance, as conceived by Barr, is clearly determined by flow; from movement to movement, one into the next, twentieth-century modernism progresses smoothly and logically down into the twin funnels of "geometrical" and "nongeometrical" abstract art. Everything leads to something else. Everything, that is, but Orphism, which just sits there, an apparent clog in the pipes. Yet this is a clog that cannot be cleanly removed or ignored. Delaunay's abstraction in particular is too much of a modernist milestone-too much of a first-to be left off the chart. Despite its lack of flow- despite the fact that it goes nowhere-Orphism is important. It's just not clear why.Part of the problem, of course, is the lack of fit among the artists assembled under the Orphic umbrella. All had undergone their artistic formation within Cubism, and all shared a tendency toward abstraction as they in turn broke with Cubism in 1912. But otherwise these artists had little of substance in common. Indeed, the divergent nature of those who were muscled into Orphism is apparent by the very fact that most of these artists did slide their way into the channels of modernist flow: arrows can be drawn from Duchamp and Picabia to Dada and Surrealism, and from Lger to the postwar machine aesthetic of Purism. In part, it is these different tensions in direction that prevent Orphism from flowing as it should. The real problem, however, is Delaunay. While the art historical tracks of influence have long been laid for the majority of those who passed through Orphism, Delaunay and his First Disk (Fig. 2) present twentieth-century modernism with a stubborn radicality that it doesn't know what to do with. While others moved on, Delaunay was left, more ebb than flow, to clog things up.
If Barr's chart stands out as the first art historical account to find itself at a loss in how to deal with Delaunay's abstraction, it was not to be the last. More recently Yve-Alain Bois has acknowledged the radicality of Delaunay's First Disk, while simultaneously characterizing it as a "fluke."2 Thierry de Duve has similarly described The First Disk as "a moment of surprise that was without epistemological consequences."3 Pierre Francastel, in his introduction to Delaunay's writings, Du cubisme l'art abstrait, cites The First Disk (along with the Window series) as both a "historical landmark" and "an isolated study."4 The difficulty of The First Disk for art historians is likewise reflected in the scant attention it receives in Sherry A. Buckberrough's 1982 book Robert Delaunay: The Discovery of Simultaneity, which devotes only 3 of its 243 pages to The First Disk, most of which are purely descriptive. Especially surprising is the near-total absence of Delaunay from Clement Greenberg's writings. In one of his few references to Delaunay, Greenberg writes that "abstract art itself may have been born amid the painterliness of Analytical Cubism, Lger, Delaunay, and Kandinsky."5 Unlike the other names on the list, though, Delaunay is discussed only once in Greenberg's collected writings (for a total of two paragraphs), in a 1949 exhibition review. Describing Delaunay in this review as "an enterprising painter whose influence is perhaps more important than his art, fine as it is," the influence ascribed to Delaunay, for Greenberg as for art history in general, is duly noted but never substantiated.6
In large part it is the obdurate singularity of The First Disk that frustrates interpretation-a singularity foregrounded by the fact that the paintings leading up to it are neatly bundled into discrete series: the Saint-Svrin series (1909-10); the City series (1909-11); the Eiffel Tower series (1909-12); the City of Paris series (1911-12); the Window series (1912-14); the Cardiff Team series (1913); the Circular Forms series (1913); and then, suddenly and seemingly out of nowhere, The First Disk (1913). Against the backdrop of its preceding work, The First Disk appears as a one-off that, as de Duve describes it, "bursts out violently as something without real precedent in Delaunay's work."7
Far from being what Bois calls "a unicum in his oeuvre"-coming out of (and going) nowhere-The First Disk must be situated within Delaunay's larger project to ground painting in a new optical model. This model was first put forward in the 1912 Window series, developed further in the 1912 Circular Forms series, and culminated in 1913 with the singular statement of The First Disk. It is only through establishing this developmental logic that The First Disk can properly resist its current characterization as an art historically significant, if otherwise inexplicable, "fluke."
1912: Delaunay contra Cubism
This happened in 1912. Cubism was in full force. I made paintings that seemed like prisms compared to the Cubism my fellow artists were producing. I was the heretic of Cubism. I had great arguments with my comrades who banned color from their palette, depriving it of all elemental mobility. I was accused of returning to Impressionism, of making decorative paintings, etc.... I felt I had almost reached my goal.-Robert Delaunay, "First Notebook," 1939(8)
Nineteen twelve was a watershed year for Robert Delaunay. On March 13 his first major exhibition in Paris closed to great acclaim after two weeks at the Galerie Barbazanges.9 Comprising forty-six works, the exhibition spanned his career to date: his early, self- taught Impressionist works;10 his 1905-6 Neo-Impressionist period; a single painting from his 1909-10 Saint-Svrin series (Saint-Svrin No. 1); a large number of Parisian cityscapes produced between 1909 and 1911; and the series of Cubist Eiffel Tower paintings from 1909-11. Apollinaire, who was to live briefly with Delaunay and his wife Sonia from November to mid-December of 1912, praised these works in his review of the exhibition, portraying Delaunay as "an artist who has a monumental vision of the world. . . . Robert Delaunay has already come to occupy an important place among the artists of his generation."11 Two weeks later, he singled out Delaunay's La ville de Paris in his review of the Salon des Indpendants. "Decidedly, the picture by Robert Delaunay is the most important of this salon," Apollinaire effused. "La ville de Paris is more than an artistic manifestation. ... He sums up, without any pomp, the entire effort of modern, painting."12
Flagged by Apollinaire as a new force on the Parisian artistic landscape, Delaunay was also making steady strides elsewhere in Europe, most notably in Germany, Switzerland, and Russia. Participating in the first Blaue Reiter exhibition in Munich, Delaunay sold four of the five works on view, including the now- lost La ville No. 1, to the painter Alexei von Jawlensky.13 More significant than sales, Delaunay's paintings prompted an enthusiastic response within the Blaue Reiter, leading to active correspondence with Wassily Kandinsky, August Macke, and Franz Marc.14 Delaunay's Blaue Reiter connections in turn led to Erwin Ritter von Busse's article "Robert Delaunay's Methods of Composition," which appeared in the 1912 Blaue Reiter Almanac, alongside Roger Allard's description of Delaunay, in his essay "The Signs of Renewal in Painting," as a painter "who has conquered the arabesques of the picture plane and who shows the rhythm of great, indefinite depths."15 Delaunay went on to exhibit that February in the second Blaue Reiter exhibition in Munich and in the Valet de Carreau.exhibition in Moscow. In March he exhibited in the first Der Sturm exhibition in Berlin and, at the invitation of Hans Arp, in July at the Moderner Bund exhibition Zweite Ausstellung in Zurich. Among the many painters in Germany to come under the sway of Delaunay's influence was the Swiss artist Paul Klee. After visiting Delaunay at his Paris studio in 1912, Klee translated the Frenchman's 1912 essay "Light" into German, and it appeared in the January 1913 issue of Der Sturm.16
Delaunay's critical triumphs in 1912 would have been little more than art historical footnotes, however, were it not for his Window series. Begun in all likelihood in La Madelaine in the Chevreuse Valley, where the Delaunays were vacationing for the summer, the twenty-two-painting Window series marks Delaunay's self-described moment of artistic maturity and break with Cubism. "They were my true aesthetic departure for modern art in reaction to the academicism and confusion of early Cubism," Delaunay writes of the Windows in an undated letter to his friend the Cubist painter Albert Gleizes.17 Delaunay relates the substance of \his break with Cubism- that "which truly began my life as an artist"18-in a 1939 notebook: "At this moment, about 1912-1913, I had the idea for a kind of painting that would depend only on color and its contrast but would develop over time, simultaneously perceived at a single moment. I used Chevreul's scientific words: simultaneous contrast.... I called them Windows."19 It was this series of remarkable paintings, unabashed-flaunting even-in their use of color, that ended Delaunay's otherwise unremarkable apprenticeship within Cubism.20 It was evident to all who saw his paintings that Delaunay had broken ranks; he quickly became, as he stated in his notebook, "the heretic of Cubism [l'hrsiarque du cubisme]."
To reference Delaunay's self-proclaimed heresy, or to characterize him as "breaking ranks," is not, I should make clear, to ascribe an overall aesthetic or conceptual coherence to the Cubism that Delaunay came to oppose. As many scholars correctly insist, the Cubism of Pablo Picasso and Georges Braque is wholly distinct from the so-called Salon Cubism of Gleizes, Jean Metzinger, Henri Le Fauconnier, and others. Differences likewise abound within the critical reception of Cubism. Yet despite the myriad internal tensions and contradictions gathered under the rubric "Cubism," Delaunay and others understood his break in opposition to a very specific group of Cubist painters, and in opposition to a very specific set of ideas associated with those painters. For all their various formal and intellectual differences, the Salon Cubists strategically represented themselves as a more-or-less cohesive movement, downplaying disparity in favor of a unifying common ground. Indeed, prior to his break with Cubism, Delaunay himself played a central role in the decision, made with Gleizes, Metzinger, and Le Fauconnier, to display their work collectively in what would come to be the first public group manifestation of Cubism: the famous Salle 41 at the 1911 Salon des Indpendants (hence "Salon Cubism"). Gleizes notes in his memoirs that it was important for these Cubist painters to appear unified: "Metzinger, Le Fauconnier, Delaunay, and I decided to send work to the next Salon des Indpendants. . . . But we must be grouped, that was the opinion of all."21 Salon Cubist Georges Ribemont-Dessaignes similarly recalls how Gleizes and Metzinger aimed "to establish a kind of legislation of the Cubist movement."22
The first published suggestion that Delaunay had broken with this group of Cubists appeared in the March 23, 1912, issue of L'Assiette au Beurre, in James Burkley's review of that year's Salon des Indpendants. Commenting on entry number 868, Delaunay's La ville de Pans, Burkley wrote, "The Cubists, who occupy only a room, have multiplied. Their leaders, Picasso and Braque, have not participated in their grouping, and Delaunay, commonly labeled a Cubist, has wished to isolate himself and declare that he has nothing in common with Metzinger or Le Fauconnier."23 In an open letter to the Cubist critic Louis Vauxcelles, published in an editorial in Gil Bias on October 28, 1912, Delaunay further differentiated himself from the Salon Cubists. Responding to Olivier-Hourcade's claim, printed in Paris-Journal eight days earlier, that "it was, however, these four painters [Metzinger, Gleizes, Le Fauconnier, and Lger] who, with Delaunay, in 1910 and above all at the [Salon des] Indpendants in 1911, created-and truly are-Cubism,"24 Delaunay retorted:
I don't support the opinion, inaccurately put forward by Mr. Hourcade, that proclaims me the creator of Cubism with four colleagues and friends. Unbeknownst to me several young painters have made use of my early studies. Lately they have exhibited canvases they call Cubist. I don't exhibit. Only some friends, artists, and critics know the direction that my art has taken.... It is necessary to set the record straight.25
Given this explicit and public disavowal of Cubism-a disavowal already picked up on by critics-what was it exactly, vis--vis Cubism, that Delaunay was disassociating himself from? In part, Delaunay's break with Cubism marked a refusal to have his increasingly evident individualism suppressed through Salon Cubism's aspirations for unity. More to the point, however, it was the conceptual means by which the Salon Cubists sought to establish this unity that Delaunay rejected. As the Salon Cubists would have it, Cubism's cohesion as a group was determined by a set of aesthetic and theoretical concerns common to all forms of Cubism-a set of unifying concerns whose discursive power overrode individual formal and stylistic differences. Pascal Rousseau conceives the conceptual stakes underlying the Salon Cubists' efforts toward unity:
It was a matter of uniting the modern movement around a solid critical discourse that would at once validate the inscription of Cubism within a classical tradition " la franaise" (the refusal of Impressionist sensuality in favor of a more cerebral art), privilege structure through essentialist decisions (the permanent harmony of line versus the too-loose and fleeting character of light), and, more generally, translate visually the subjective character of representation through a neo-Kantian interpretation of optical synthesis (multiple views of the object and the "Cubism of conception").26
In short, the substance of Delaunay's break with Cubism centered on rejection of its central orthodoxy: the suppression or elimination of superficial visual sensation-color being the worst offender in this regard-in favor of the invariable and essential ground of conception. This position is voiced in the criticism of Olivier-Hourcade, among others, who claimed that Cubism depicts what we know of the represented object-what we know of its, physical form- as opposed to what we see. As Hourcade writes:
The painter, when he has to draw a round cup, knows very well that the opening of the cup is a circle. When he draws an ellipse, therefore, he is making a concession to the lies of optics and perspective, he is telling a deliberate lie. Gleizes, on the contrary will try to show things in their sensible truth.27
While Salon Cubism and its advocates turned to conception as the basis for a new realism-what Gleizes and Metzinger termed, in their 1912 book Du Cubisme, the "profound realism" of the mind, in opposition to the "superficial realism" of the eye28-Delaunay stood alone in his attempt to develop a new, nonsuperficial model of vision for painting. He first posed this new optical model in the Windows, where he sought to combine Cubism's emphasis on conception with Impressionism's emphasis on optical sensation. In so doing Delaunay not only reconciled the seemingly irreconcilable-Cubism and Impressionism-he also posited a pictorial model of vision that was fully informed by modern optical theory.
Explicit in his understanding of the shift from a premodern to a modern conception of perception, Delaunay remarked on the consequences of this shift for painting in an entry to his 1939-40 notebook: "Historically there really was a change in understanding in modes of seeing, and thus in [pictorial] technique."29 Virginia Spate, who begins her 1979 book on Orphism with this quote, takes this shift in visual understanding not as a literal change enacted by modern optical theory but as a general response to the perceptual conditions of modernity. For Spate, in other words, Delaunay's "Perceptual Orphism" stands in relation to historical changes external to the viewing subject, which in turn affect the overall mental conditions of perception. Delaunay's engagement with issues of perception is thus conceived through the "profound aspects of modern life and of a new form of consciousness: Simultaneous consciousness."30 I want to argue something quite different. Delaunay, I believe, should be taken at his word: that his work stands in response not simply to an altered mode of seeing wrought by modernity but to a historical change in the actual understanding of perception-a change, that is, in the understanding of the internal, psychophysiological mechanics of perception. How, then, to formulate this change in understanding, and with it Delaunay's attempt to salvage vision as a viable ground for painting? In order to grasp Delaunay's reformed visual realism, it is first necessary to comprehend the structure of vision as reformed by modern optical theory. It is necessary, that is, to comprehend what Delaunay describes as the "change in understanding in modes of seeing."
In Techniques of the Observer: On Vision and Modernity in the Nineteenth Century, Jonathan Crary posits the beginning of the nineteenth century as a fundamental break from what he terms "classical models of vision." For Crary, one of the primary models that founds and supports the idea of classical vision is the monocular paradigm of the camera obscura. The enclosed, darkened space of the camera obscura creates an inverted image of the external world as light passes through a small opening. While the effects of this simple imaging device have been likened to human vision since antiquity, it was only in the period from the late 1500s to the end of the 1700s that it became the dominant model for visual perception. The ascendance of the camera obscura model effectively brought to an end the prior debates over extramission- the theory that the eye emits as well as receives light.31 Stripped of its active function of emission, the eye became instead a totally passive and transparent receptor of light and the optical information it carried.
In addition to positing a stable visual field, the camera obscura serves as a model of the subject in several important ways. On a structural level, it separates the observer from others, which has the effect of individuating the viewer. This simultaneously supports the viewer as a free and sovereign individual and universa\lizes the observer as an interchangeable position openly available to anyone. At the same time that the mechanism of the camera obscura separates the user from others, however, it also separates the user from the external world. The camera obscura thus became a kind of technological analog to the Cartesian separation of the viewing subject (res cogitans) from the world (res extenso). Constitutive of this interior-exterior divide is a fundamental stability. The camera obscura asserts a coherent and consistently unified visual field "from which any inconsistencies and irregularities are banished to insure the formation of a homogenous, unified and fully legible space."32 And, finally, the model of the camera obscura severs the eye from the body of the viewer. This decisive separation functions to "sunder the act of seeing from the physical body of the observer, to decorporealize vision.. .. The body then is a problem the camera could never solve except by marginalizing it into a phantom to establish the space of reason."33
The ascendance of this model of vision, and the subject position that it supports, came to an abrupt end in the early nineteenth century. In its place developed a modern and heterogeneous regime of vision, one that is grounded, above all else, by the insertion of the body into optical discourses. Within this newly formed optical paradigm, a split emerges between the study of light and color as independent phenomena within the physical world (prismatic light) and of light and color as they are experienced subjectively through the physiological and cognitive processes of the body. The early nineteenth century thus marks a break between optics as a branch of physics (the study of light and its constituent properties) and physiological optics as a branch of perceptual and cognitive theory.
The insertion of the body into various theories of perceptual physiology is perhaps nowhere more evident than in the sudden centrality assumed by physiologically produced chromatic effects in the early nineteenth century. These include such phenomena as visual afterimages; colors that mix in the retina of the viewer; and the experience of light and color from causes such as pressure on the optic nerve, certain narcotics, and so on. As Crary points out, the centrality of internally produced chromatic effects cuts the supposed bond between optical sensation and real-world referent, and in so doing breaks radically from the camera obscura model of classical vision. The experience of light or color is thus no longer dependent on any external light or color. Centralizing the physiology of the body in the optical process creates an epistemological rupture at the heart of visual perception. As Hermann von Helmholtz stated in 1867 in "On the Recent Progress of the Theory of Vision":
We have already seen enough to answer the question whether it is possible to maintain the natural and innate conviction that the quality of our sensations, and especially our sensations of sight, give us a true impression of corresponding qualities of the outer world. It is clear they do not.... Pressure upon the eyeball, a feeble current of electricity passing through it, a narcotic drug carried to the retina by the blood, are capable of exciting the sensation of light just as well as sunbeams. The most complete difference offered by our various sensations ... does not, as we now see, at all depend upon the nature of the external object, but solely upon the central connections of the nerves which are affected.34
No longer grounded in a unified, stable field, visual perception becomes irrefutably conditioned by the body. This break with classical models of vision actively participates in the construction of a new visual subject. Once the physiological intervention of the body is foregrounded in the perceptual process, the previous stability of a clearly demarcated "inside" (the projected image inside the camera obscura, the res cogitans) and "outside" (the world, res extenso) becomes untenable.35 As a result, color and light lose their prior bond to an externally stable and unified visual field.
It is vital not to mistake the significance of physiological optics within modern optical theory as simply a new form of epistemological skepticism. Throughout the seventeenth and eighteenth centuries the observation of internally experienced physiological effects was frequently cited as evidence for the fallibility of the senses.36 Modern optical psychology does not tell us anything new, or even anything less than obvious, in stating that pressure on the eye produces the internal experience of light. What modern optical theory provides is a conception of "pure" sensory information that is internal to the body and actively produced by the senses. By way of contrast, a pre-nineteenth-century thinker such as David Hume (who relied on the camera obscura model of vision) understood the perceptual image to exist externally to the body and independently of the senses. Hume is typical of seventeenth- and eighteenth-century thinkers in his conception of the perceptual image as a unified, external entity that flows through the senses much as water flows through an inlet: "Nothing can ever be present to the mind but an image or perception, and the senses are only the inlets through which these images are conveyed."37 While a modern optical theorist such as Helmholtz understands the senses as actively producing sensory information, for Hume the senses function as mere conduits for an external perceptual image.
The significance of internally produced chromatic effects for modern optical theory went beyond providing new evidence for the fallibility of the senses: it served to demonstrate that sensory information is produced actively by the senses. This was the crucial step that was excluded from the camera obscura model of vision-that between the eye and the brain, optical information exists in a "pure" state, wholly distinct from the external stimuli that generate it and the final perceptual image that is registered in the brain. As Crary states, it was this middle step)-the moment of pure, internally produced sensory information-that was unthinkable prior to the advent of modern optical psychology: "In the seventeenth and eighteenth centuries this kind of 'primordial' vision could not be thought, even as a hypothetical possibility."38 Svetlana Alpers similarly describes how the seventeenth-century optical theorist Johannes Kepler made a revolutionary distinction between the world outside of the eye (idola, or visual species) and the image formed on the retinal surface (pictura). Despite this distinction, Kepler was unable to conceive of an intermediary step between the retina and perception. As Alpers writes, "The study of optics, so defined, starts with the eye receiving light and ceases with the formation of the picture on the retina. What happens before and after-how the picture so formed, upside down and reversed, was perceived by the viewer-troubled Kepler but was of no concern to him."39
How, then, can "pure optical information"-the optical sensory data that is transmitted from the optic nerve to the brain-be characterized? First, as is commonly known, pure optical information when registered on the retina is inverted in relation to the external world, both upside down and left and right. second, this inverted image on the wall of the retina is not simply a straightforward mirror image of the world turned on its head, for the neatly bounded distinctions of figure and ground that we see within everyday visual perception do not exist within pure optical data. Yve-Alain Bois has pointed out that one of the crucial distinctions between visual perception and pure optical information is precisely the absence of a figure-ground distinction in the latter: "To perceive is first of all to perceive a figure against a ground (this is the basic definition of perception). But the ground is not always given: it is indeed what we must preconsciously construct differently each time we are solicited to perceive."40 Third, optical information is binocular, and this binocularity creates a slight discrepancy in the information that is registered in the different retinas. This discrepancy in the retinal image is crucial to the visual process, as the brain compares the slight differences within the eyes as a means to generate depth perception.41 Finally, pure optical information as it is registered on the concave surface of the retina is completely two-dimensional. The knowledge of optical flatness was indeed so widespread within the critical literature of the mid-nineteenth and early twentieth centuries that Adolf Hildebrand felt only a footnote was necessary to remind the reader of this well-established fact: "The reader need hardly be reminded that our actual impression is two-dimensional, a flat picture on the retina."42 In sum, optical information is understood by modern optical theory to be fundamentally unlike what we actually see.
Granting the specificity of discrete sensory operations, modern optical theory faced the central problem of how dissimilar forms of sensory information combine to generate a coherent, cohesive perceptual array.43 Helmholtz in particular was intensely concerned with the problem of how twodimensional, inverted, binocular optical data, devoid of all spatial perception, is experienced as visual perception with depth and clearly bounded figure-ground distinctions. The problem for Helmholtz and others was how the distinct nerve functions of touch and vision come together to construct a unified and functionally seamless field of vision. Helmholtz thus identified two basic components fundamental to the process of visual cohesion. The first is that the deficiencies of optic and haptic sense functions compensate for each other as they merge cognitively in the mind. As Helmholtz states: "The two senses which real\ly have the same task, though with very different means of accomplishing it, happily supply each other's deficiencies. Touch is a trustworthy and experienced servant, but enjoys only a limited range, while sight rivals the boldest flights of fancy in penetrating unlimited distances."44 For Helmholtz, therefore, touch and vision are made to cohere within cognition in order to produce normative visual perception. "Ordinary vision," as Helmholtz asserts, "is not produced by any anatomical mechanism of sensation, but by a mental act."45
Given the conclusion that touch and sight supplement one another, how then can touch be mobilized when an object is only seen? In answering this problem, Helmholtz proposed that the second component fundamental to visual perception is memory. No longer an innate, fully formed condition present from birth, perception is reconfigured as a process that is learned. Accordingly, Helmholtz analyzes the means by which an infant develops visual perception as it learns to connect the tactile senses of its body with what it sees in order to develop perceptual unity:
A child seizes whatever is presented to it, turns it over and over again, looks at it, touches it, and puts it in its mouth.... After he has looked at such a toy every day for weeks together, he learns at last all the perspective images which it presents.... By this means the child learns to recognize the different views that the same object can afford in connection with the movements which he is constantly giving it. The conception of the shape of any given object, gained in this manner, is the result of associating all these visual images.... All these different views are combined in the judgment we form as to the dimensions and shape of an object.46
We learn, over a period of time, to make sense of what we see and what we know. In so doing our memory brings past sensory experience and knowledge into the present.47 When we see an object we are remembering how our body has interacted with it-how it feels, how it recedes in space, how tall it is, how hard, cold, or sticky it is.48 As Henri Bergson explains in Matter and Memory: "our senses require education. Neither sight nor touch is able at the outset to localize impressions. A series of comparisons and inductions is necessary, whereby we gradually coordinate one impression with another."49 Beginning in the mid-nineteenth century, seeing ceases to be a given- it ceases to be a simple fact of being that we are born into, fully formed and already present. From the mid-nineteenth century on, seeing becomes instead a physiological and cognitive process that we must learn.
Cubism and Antivisuality
Delaunay's attempt to rehabilitate vision as the basis for a new pictorial realism must be placed squarely in the context of the modern optical theory explained above. But just as important, it must also be seen in opposition to the Cubist reception of that same optical theory. For as far as the Cubists were concerned, modern optical theory demonstrated conclusively that vision-now understood to be fundamentally different from pure optical information and subject to the vagaries of the body in which it is enmeshed-is not to be trusted. According to the Cubist reception of modern optical theory, vision had been proven to be wildly erratic and prone to a range of corruptions. This stood in contrast to the mind, which filters out impurities, giving us a true and stable image-giving us perception as an act of cognition. References to modern optical science abound in Cubist criticism. Gleizes and Metzinger refer to the modern optical theory of binocular accommodation in Du Cubisme. "As for visual space we know that it results from the agreement of the sensations of convergence and 'accommodation' in the eye."50 Maurice Raynal likewise invokes Helmholtz in an appeal to Cubist antivisuality: "'The truth is not in the senses,' said MaIebranche, 'but in the mind,' and Helmholtz-and indeed Bousset before him- showed that the senses tell us nothing but our own sensations."51
The most sustained and thoroughgoing use of modern optical theory within Cubist-era criticism is found in DanielHenry Kahnweiler's The Rise of Cubism. In his understanding of the relations between sensation, memory, and cognition, Kahnweiler demonstrates a clear debt to modern optical science. Kahnweiler contends that the process of "seeing" a Cubist painting (Kahnweiler uses the quotation marks) almost identically mirrors the process of visual perception as theorized by Helmholtz. In both cases, a visual stimulus is charged with a specific memory image. The ensemble of visual stimuli and memory images combine within the viewer's mind to produce a final image. "Seeing" a Cubist painting, for Kahnweiler, constitutes more an act of conception-of cognitive assembly-than an act of vision: "When 'real' details are thus introduced the result is a stimulus that carries with it memory images. Combining the 'real' stimulus and the scheme of forms, these images construct the finished object in the mind. Thus, the desired representation comes into being in the spectator's mind."52 For Kahnweiler, as for Helmholtz, the viewer combines the known with the seen, such that, as Kahnweiler describes Cubism, we mediate "the two dimensional 'seen' with the three dimensional 'known.'"53
Not only does Cubist criticism use modern theories of vision against vision, it does so in the name of an essentialist realism. Cubist painting, according to early advocates, provides the oncological blueprints-the essential truth-of its objects, such that, as Olivier-Hourcade claims, "The ruling preoccupation of the [Cubist] artists is with cutting into the essential TRUTH of the thing they wish to represent, and not merely the external and passing aspect of'this truth."54 Likewise, for Jacques Rivire, "The true purpose of painting is to represent objects as they really are; that is to say, differently from the way we see them. It [Cubism] tends to give us objects in their sensible essence, their presence; this is why the image it forms does not reveal their appearance."511 Stripping away "contingent visual and anecdotal elements" from its represented object, "Scientific Cubism," as Apollinaire terms it, "resulted from the fact that the essential reality was depicted there."56
Beginning with the Window series and culminating with The First Disk, Delaunay broke decisively with Cubism's attempted end run of vision into the essentialism of conception. Understanding that modern optical theory does not invalidate vision, as the Cubists argued, but rather reformulates it into a wholly new and modern conception, Delaunay sought to do the same for painting. Engaging rigorously with the structure of perception as claimed through modern optical theory-a structure premised on the double role of cognition and sensation-Delaunay's paintings emphasize the role of the mind in the act of vision. And in so doing, Delaunay's paintings develop a fundamentally new model of visual realism-a visual realism in which painting serves to bridge the body of the viewer with its ground in the world.
The immediate optical impact of the Windows announces Delaunay's commitment to opticality-and thus his break from Cubism-in no uncertain terms. Revisiting Neo-Impressionism's involvement with chromatic retinal mixing, Delaunay abandoned the Divisionist technique of colored dots, or taches, in favor of larger colored planes of "simultaneous contrast," a form of optical mixing first theorized by the nineteenth-century optical theorist Michel-Eugne Chevreul. Chevreul explains this process in the introduction to his 1839 De la loi du contraste simultan des couleurs:
A ray of solar light is composed of an indeterminate number of differently colored rays ... [these] have been distributed into groups which have been given the names red rays, orange rays, yellow rays, green rays, blue rays, indigo rays, and violet rays; but it must not be supposed that all the rays comprising the same group, red for instance, are identical in color. On the contrary, they are generally considered as differing, more or less among themselves, although we recognize the impression they separately produce as comprising that which we ascribe to red.57
Looking onto a field of contiguous colored planes, we perceive what appear to be discrete colors, "we recognize the impression they separately produce as comprising that which we ascribe to red." But as Chevreul points out, these apparently unified colors are in fact composed of "an indeterminate number of differently colored rays," which intermingle and vary according to the proximity, hue, brightness, and surface texture of adjacent colors. Rays of colored light mix with neighboring rays of light to produce the optical mixing of simultaneous contrast. As Biaise Cendrars describes the effect of simultaneous contrast within Delaunay's paintings, "A color isn't a color unto itself. It is only a color in contrast with one or more colors. A blue is only blue in contrast with a red, a green, an orange, a grey and all the other colors."58 No longer is vision construed as a classical piercing of space-a progressive (or diachronic) succession into depth, from the eye through lines of perspective to an ever-receding vanishing point. Rather, vision is reconfigured as a simultaneous (or synchronic) field, distributed across the visual plane."
The emphasis on optical phenomena in the Windows has led art historians to characterize Delaunay as a "retinalist," a term of disparagement coined by the Cubists to denigrate the "superficial realism" of Impressionism. Rosalind Krauss maintains that Delaunay's paintings establish a visual homology between the surface of the canvas and the surface of the retina. For Krauss, this "retinalism" eliminates the role of the mind, stripping vision of its conceptual depth: "the 'arrt la rtine,' the stoppingof the analytic process at the retina ... [became] a kind of self-sufficient or autonomous realm of activity.... This is the logic we hear, for example, in Delaunay's assertions that the laws of simultaneous contrast within the eye and the laws of painting are one and the same...."60 More than just a superficial model of vision lacking "the analytic process," however, retinalism is also premised on speed. Deprived of cognitive function, visuality according to the retinalist model indulges the pure optical stimuli of rapid and kaleidoscopic retinal sensations. Rousseau, for example, argues that Delaunay's paintings attempt to translate the frenzied pace of modern life as it darts across the surface of the eye: "The paintings of Robert Delaunay are entirely motivated by the avid retinalism of the 'painting of modern life': 'Looking to see,' to see more and more quickly, to see too much, sometimes to the point of risking a hypnotic vertigo as the eye is carried away by the gyrating movement of colors."61 Rousseau sees the speed of Delaunay's retinalism as part and parcel of the artist's larger efforts to recover a primitive vision, cleaved from knowledge and experiential memory. In a 1997 essay, Rousseau claims that Delaunay "adopts the 'innocent eye' thesis defended by Ruskin" and quotes a passage from Paul Valry's 1895 Introduction la mthode de Lonard de Vinci:
Most people see with the intellect much more frequently than with the eye. Instead of colored spaces they become aware of concepts. A tall, whitish cube with holes filled with the reflections of glass is immediately a house: the House! A complex concurrence of abstract qualities. When they move they miss the movement of the rows of windows, the transformation of the surfaces continually changing their aspect-for the concept does not change. They see through a dictionary rather than through the retina....62
It is this innocent eye engagement with a world of pure opticality that Rousseau assigns to Delaunay: "The project of simultanism adds to this refutation of a priori knowledge within representation through a claim to a primitive sense of color."63 Along similar lines, Spate, too, asserts that Delaunay attempted to locate "a consciousness found in perceptual experience unclouded by conceptual, learnt experience."64
The view of Delaunay as a "retinalist" who seeks to derive knowledge from the domain of the visual not only is at odds with the experience of his paintings, it also underestimates (or misunderstands) the depth of Delaunay's engagement with modern optical theory.65 Far from expressing a naive retinalism that is optically immersed in the speed of modern life, Delaunay's paintings, beginning with the Window series, are at pains to slow the gaze of the viewer, while at the same time coupling vision with experiential knowledge. Consistent with modern optical theory, Delaunay's concern is to move beyond the two-dimensional surface of the retina and into the depth of visual perception. Knowledge works in tandem with the senses to create spatial perception such that, as Delaunay writes, "We live in depth, we travel in depth. I am there. The senses are there. And the mind!!"66 This equilibrium between the senses and the mind forms the basis of Delaunay's reconstituted visual realism. And, true to modern optical theory, this is a model of vision in which we must literally learn to see.
Seeing in Time
Delaunay's Windows perform the slow process of learning to see described by Helmholtz. As Delaunay suggests, it is only over time that we learn to see his paintings: "I had the idea for a kind of painting that would depend on color and contrast, but would develop over time."67 As discussed above, Helmholtz theorized that young children must learn to cognitively merge two-dimensional optical information with three-dimensional spatial knowledge in order to develop visual perception. Remarkably, this complex process of learning to make sense of what we see is performed experientially in the act of looking at the Delaunay's Windows. We can see this if we begin with what is now generally considered to be the second (though first finished) painting in the series, Simultaneous Windows (1st Part, 2nd Motif, 1st Replica) (Fig. 3).68 Our initial experience of the painting, much like that of an infant, with its undeveloped perceptual acuity, is purely optical; we see a loosely articulated grid of apparently abstract colors with no clear figure-ground distinction, no evident orientation-top to bottom, left to right- and a two-dimensional array of rough, ill-defined chromatic shapes that bleed one into the other. What we see, in other words, replicates the conditions of pure optical information as it is registered on the retina of children prior to perceptual development. Color, in this initial view, both painterly and perceptual, takes precedence over form.
The granting of chromatic priority in the Windows places Delaunay in direct opposition to Cubism's privileging of form (determined through conception) over effects of color (determined through vision). Kahnweiler, for example, claims Cubism's suppression of color serves to separate the "primary qualities" of geometric form and spatial position from the "secondary qualities" of color and tactility:
[The Cubists] distinguish between primary and secondary qualities. They endeavor to represent the primary, or most important qualities, as exactly as possible. In painting these are: the object's form, and its position in space. They merely suggest the secondary qualities such as color and tactile quality, leaving their incorporation into the object in the mind of the spectator.69
Kahnweiler's reliance on optical physiology notwithstanding, the nineteenth-century realization that infantile vision is initially experienced as pure optical sensation prior to the learned perception of form and space supports Delaunay's visual prioritization of color. No longer subservient to preexistent three- dimensional objects, color, in Delaunay's Windows, refuses to function as a coating, applied as a secondary characteristic to a priori forms in space. As Apollinaire maintains, "color is no longer used for just coloring ... color is now itself the form.... Color no longer depends on the three dimensions, for it is color that creates them."70 Walter Benjamin expresses a similar view, which he relates directly to infantile perception, in his 1914 essay "A Child's View of Color." Emphasizing that for young children, "their eyes are not concerned with three-dimensionality, this they perceive through their sense of touch," Benjamin, like Apollinaire, claims the priority of color over form: "for the person who sees with a child's eyes . . . [color] is not something superimposed on matter, as it is for adults. The latter abstract from color, regarding it as a deceptive cloak for individual objects existing in time and space."71
Prior to the perception of form, touch and color exist within separate sensory registers before the mind learns to combine them into a cohesive, three-dimensional view of space. This double separation and cohesion of touch and color is also performed in the experience of viewing the Windows. Avoiding a view of color that is "superimposed on matter," Delaunay strategically engages the retinal mixing of simultaneous contrast to separate color from its material ground. Our experience of color, in looking at the Windows, is necessarily determined through the physiological mixing of color in the eye. This physiological experience of color, determined through the effects of simultaneous contrast, is thus literally stripped of form. At the same time, however, we see the contrasting textures of paint application and the tactile differences in paint created by the wooden ground of the frame and the rough weave of the canvas. The colors we see in Delaunay's Windows are thus located simultaneously and indeterminately between the pure, physiologically produced colors of the eye-devoid of form and internal to the body- and the material colors of the paint, inextricably bound to its tactile form and ground, external to the body. The effect of simultaneous contrast, in other words, allows for a seemingly impossible and paradoxical expression of color, in which it is simultaneously both separated from and bound to tactile form.
This simultaneous separating and reassembling of sight from touch is further articulated in a curious detail in Delaunay's painting. If we look at the rough, conspicuous miters of the painted frame, we see how Delaunay has carefully delimited each miter with paint so that two or more colors meet along the exact seam of the wooden joint. This holds true in all but the upper-right corner of the frame, where we see the light blue cross over the forty-five-degree angle of the miter with an exacting deliberation. It is as if Delaunay were at pains to show how the coarse frame that we can feel, and the color that we see, have been cleaved from one another- separated as distinct forms of sensory information-only to be reassembled again in the rest of the frame. In prying the tactile form of the frame from its corresponding color, it seems that Delaunay is intent on showing not simply the structure of painting but also how it relates to the structure of the viewer's vision.
It is crucial to Delaunay's project that our perceptual experience of the Windows not rest with this initial sensory view of color and tactile surface but that it extend into an acquired perception of form determined through experiential knowledge. This move beyond the sensory is vital, for in the Windows, Delaunay seeks to foreground the perceptual priority of color over form by replicating in the viewer the process of infantile perceptual development. Accordingly, in the process of learning to see-as in the process of learning to see Delaunay's Windows-we combine sensory data with cognitive information inorder to determine spatial depth and decipher images. This occurs with a series of images that are gradually perceived within the chromatic grid. The first, and most easily discerned, is the green elongated triangle of the Eiffel Tower in the center of the canvas, with its lighter, more difficult to see supporting columns below (Fig. 4). The two small windows on a building front follow in the lower part of the frame. Most difficult to determine are two images that have gone unnoticed in past accounts of the Windows and have taken an especially long time to learn to see. The first is a face in the yellow field of the viewer's right-hand side of the painting (Fig. 5). The dark green patch of paint two-thirds of the way down the right-hand side functions as lips, while the quarter-circle of yellow beneath forms the chin. The ear nestles in the right-hand corner of the base of the tower. The jaw and then the neck extend the sloping, fragmented line of the tower a fraction away from the corner of the canvas. This line then continues into the outer edge of the picture through the rough, prominent miter of the frame. The second is a rectangular form that cuts diagonally across the surface of each painting in the series, which represents an aerial view of the Champ de Mars, the field on which the Eiffel Tower sits. Delaunay took the aerial image from a photograph published in 1909 in the journal Comoedia (Fig. 6) and later reworked it in his 1922 lithograph Tour Eiffel et Jardins du Champ de Mars (Fig. 7).72
All of these images make sense. Setting the stage for our seeing, Delaunay tells us his painting is a window. Given Delaunay's renown for cityscapes of Paris, and the Eiffel Tower especially, it stands to reason that we should come to see the tower. The surrounding sky blue pushes back into space, and the two small dashes of green on the bottom of the frame ground the tower in neighboring buildings. By the same token, we know from experience that facing a window, we can see our mirror image reflected back into our line of vision. Drawing from experience, each individual viewer can learn to see his or her face (or is it Delaunay's face, the first viewer of this window?). We also know that Delaunay's past Cubist representations of the tower contained simultaneous and aerial views. Accordingly, there is a visual logic to seeing the Champ de Mars in the diagonal rectangular form. Similarly, we also know from experience how aerial views activate memory in the act of seeing. Looking at the all but unrecognizable sprawl of Paris from the top of the actual tower, we perceive isolated fragments that orient us to the overall structure of the city. Memory works to relate part to whole such that, over time, we begin to form a comprehensible image. As Roland Barthes describes this process of aerial vision from the top of the Eiffel Tower:
Take some view of Paris taken from the Eiffel Tower; here you make out the hill sloping down from Chaillot, there the Bois de Boulogne; but where is the Arc de Triomphe? You don't see it and its absence compels you to inspect the panorama once again, to look for this point which is missing in your structure; your knowledge struggles with your perception, and in a sense, that is what intelligence is: to reconstitute, to make memory and sensation cooperate so as to produce in your mind a simulacrum of Paris.73
In laying bare the structure of vision, the aerial view demands that the mind cooperate with the eye in an act of visual interpretation. Separating and grouping, moving between recognized fragments and their position within the larger structure, the mind must negotiate between what is known and what is seen. Visual perception, as it is experienced and seen from the top of the Eiffel Tower, is thus a process that takes place over time as the mind pieces together all the parts of the visual puzzle, moving between the memory of past experience and what is given in sight.
Duplicating a forty-five-degree rotation of the central canvas almost exactly, the rectangular form of the Champ de Mars twists the structure of painting (the canvas support) into a metaphor for the structure of vision (the aerial view). Indeed, in drawing a connection between visual and pictorial structure, Delaunay radically reformulated the correlation between painting and vision, both of which were classically understood to provide transparent, windowlike views onto the world. Delaunay's window thickens space so that its visual opacity gives way to depth only after negotiation with experiential knowledge. The orthogonal grid of the Windows exemplifies this imbrication of pictorial and visual structure. Simultaneously marking both the flat, literal surface of the painting and the spatial vectors that have traditionally served as the markers of perspective, the grid fluctuates, back and forth, between the literal form of the painting's surface and the grid of perspectival vision. Krauss foregrounds this distinction between the literal material flatness of the grid, on the one hand, and its simultaneous relation to vision and the perception of depth, on the other. "The grid," Krauss writes, "is flattened, geometricized, ordered. It is antinatural, antimimetic, antireal. It is what art looks like when it turns its back on nature."74 At the same time, Krauss stresses that prior to its inscription on the pictorial surface in the twentieth century, the grid was first articulated in the nineteenth century through the Symbolist fascination with the window: "The grid appears in symbolist art in the form of windows, the material presence of their panes expressed by the geometrical intervention of the windows' mullions." Yet what Delaunay's Windows make explicit is that vision, despite appearances to the contrary, is precisely not like a window. In Delaunay's series, the image of the transparent window vies with the literal flatness of the nontransparent screen of the painting's surface. In order to see through this particular window, out onto the world of objects, we must first learn to see through its resolutely two-dimensional and opaque grid.
Given Delaunay's concern with structure, it does not take long for us to realize that in addition to the frontal and aerial view of the Eiffel Tower, there is another view of the tower that we eventually learn to see. Interwoven within the cohesive solidity of this orthogonal structure, we come to find another geometric pattern: a kind of prismatic shattering of the window whereby the rectangular grid splinters into a subset of intersecting triangles. As with the view from the tower, we can also learn to make sense of these interlocking rectangular and triangular forms. If we look, for example, at one of the pen-and-ink studies of the tower done between 1910 and 1912, we see about a third of the way up from the bottom a conspicuous and familiar-looking square that is subdivided into intersecting triangles (Fig. 8). And once we have noticed it, we see it throughout the remainder of the ink drawings in this series as well as in the Cubist series of red towers that Delaunay painted prior to the Windows between 1910 and 1911 (Fig. 9).
The orthogonal and diagonal lattice of the grid presents us with another view of the Eiffel Tower, a view not from straight on or from above but from within-an image of the interlocking grids that form the skeletal structure of the tower itself (Fig. 10). What we see in this close-up image of structure are the pressures and strains-the nuts and bolts-of a structural framework that, once assembled and looked at from a remove, piece together to form another image of the tower.
The First Disk
Delaunay's Eiffel Tower serves metaphorically to map visual structure onto pictorial structure. In so doing, the metaphoric image emphasizes, as metaphors do, the role of the mind. It is this emphasis on the visual role of the mind and its reliance on metaphor that The First Disk attempts to balance in relation to the body. Without in any way downplaying how we are drawn into the intimate surfaces of the Windows-we feel and literally see ourselves thrust into the heart of the tower-the accent, both metaphoric and experiential, is placed on cognition. The First Disk by contrast, foregrounds not only the impact of embodied vision as it strikes the painted surface but also the impact of the surface as it strikes embodied vision-an impact made explicit by Delaunay in his multiple references to The First Disk as "a punch."75
Like the Windows, The First Disk folds the structure of vision into the structure of painting. The First Disk, like the retinal image, presents no clear sense of orientation, no obvious left or right, top or bottom. And, like the retinal image, The First Disk is resolutely two dimensional. The merging of visual and pictorial structure is most immediately evident, however, in the shape of The First Disk. As in Frank Stella's stripe paintings of the early 1960s (Fig. 11), the depicted shape of the circular bands echoes the literal shape of the canvas. The shape of Delaunay's First Disk, unlike that of Stella's stripe paintings, is additionally structured according to the viewer's vision-to the circular radius of the viewer's optical cone, whose circumference delimits vision prior to its peripheral distortion. Vision is thus inserted as an active term in the play between the painted concentric bands and the material structure of the circular support. Indeed, it is crucial to the viewing of The First Disk that vision can be aligned to the shape of its outer edge and also with the shape of each of the seven painted bands. For, as painted and visual radii coincide in their respective diminishment, The First Disk positions not just the orientation of the viewer's vision but also the physical position of his or her body. With each step closer, the viewer's visual radius realigns with a new painted radius. The \First Disk pulls us in, nearer and nearer toward its central blue and red bull's-eye, only to push us out again, back toward the outer frame of the support.
The diminishing circumference of the painted bands can be seen equally in terms of the body's movement toward The First Disk and The First Disk's movement toward the body. The successively smaller circles can be visualized as increasingly proximate cross-sections within a static visual cone extending from The First Disk to the viewer. Accordingly, the outer edge of the painting corresponds to the most distant point of vision, with the absolute center of the painting being the point at which the tip of the cone touches the viewer's eye.
If the painted bands do function as distinct positions within the viewer's optical cone, collapsed into the flatness and circular shape of the painting, then the central crossing of the horizontal and vertical lines that bisect The First Disk designate the contact point between the eye and the painting. At the same time, these lines also serve to entwine visual and pictorial structure. As vision is aligned along the cross-hairs of the painting, the intersecting lines reflect the structural cross bracing used to reinforce the stretcher of the tondo.
Delaunay further blurs the boundaries between vision and its pictorial object through his strategic use of simultaneous contrast. In the process of viewing The First Disk, the colors that are perceived exist both physiologically within the eye as a result of retinal mixing and as a material property of the pigment, immanent and bound to the painted surface.76 Color butts against color in The First Disk to produce retinal effects of simultaneous contrast that seem to belong at once to the interiority of the eye and to the exteriority of the painted surface. Neither quite inside the eye nor quite outside on the surface of the painting, the location of color is perceived indeterminately between The First Disk and the viewer. It is this visual indeterminacy-an indeterminacy that blurs the distinction between the viewing subject and its painted object-that relates The First Dish back to the optical model first established in the Windows. Yet, while the emphasis in the Windows is on a metaphoric thrust into and from above the Eiffel Tower, The First Disk performs this visual thrust on an experiential level, as the boundaries are broken between the eye and the painting.
The retinal effects of simultaneous contrast do more, though, than confound the neatly prescribed location of color in either the eye or the painting. The chromatic vibrations of retinal mixing also produce the effect of movement-a slow, optical illusion of gentle rotation and pulse. As Delaunay described the physiologically produced perception of movement in The First Disk "The experience was convulsive. No more fruit dish, no more Eiffel Tower, no more street. . . . I tried to cry out to them: I have found it! It turns! But they avoided me."77
Along with the effect of chromatic retinal mixing, the rhythmic movement of simultaneous contrast further obscures the boundary between the eye and the paintingbetween the physiological pulse of movement and the representation of movement as a static image. We see, therefore, the quiet turns of The First Disk as a perceptual effect, produced through the chromatic vibrations of simultaneous contrast. Delaunay characterized this effect of movement as being distinct from The First Disk's surface in a 1926 interview: "[Simultaneous] Colors offer the depth of a penetrating rhythm . . . a movement . . . insolently outside of the noticeable surface."78 But The First Disk also provides a graphic suggestion of movement that is bound to its surface-a representation of movement resembling a spinning propeller or wheel. Hidden in plain sight like Poe's purloined letter, the representational function of The First Disk sits in open view, even while it is concealed within the rings of its own abstraction. Like the colors of simultaneous contrast, the exact location of The First Disk's movement is again indeterminate: both inside the eye of the viewing subject and outside, bound to the form of the painted image. It is unclear in viewing The First Disk whether we perceive movement as a physiological response to its circular color patterns or as a representational form independent of the eye. Unable to clearly separate physiological effect from representational form, the viewer is folded into an indeterminate space that combines the perception of movement with the stasis of the image. As Delaunay claimed: "That's what I tried to realize with my simultanism, that which can properly be called static movement."79
Delaunay's first attempt to represent movement in relation to the effects of simultaneous contrast occurs in the later paintings of the Window series. This is seen, for instance, in Delaunay's first shaped canvas, perhaps the most abstracted work in the series, Les fentres (Fentres ouvertes simultanment 1re partie, 3e motif) Windows (Windows Open Simultaneously, 1st Part, 3rd Motif) (Fig. 12). Here, more than in any other painting in the series, the triangular motif of the Eiffel Tower is obscured in favor of the interchromatic reactions of the orthogonal and triangular grid patterns. All but gone are the iconic images of the Champ de Mars, the tower, and the reflected face whose barest residue can be seen in the yellow forms just to the right of center. If the small green square that sits in the middle of the white broken triangle on the left side of the painting is to be taken as a window, it is by no means obviously so. The retinal force of the chromatic grid and its material assertion of the picture plane have all but subsumed the composition. Within the rectilinear and diagonal grid pattern, however, we see a new form that appears for the first time in the series: the S-shaped patch of white on the right side of the painting.
The appearance of this serpentine form in the otherwise angular network marks Delaunay's first attempt to represent circular movement. Just as the oval shape of the canvas carries the eye around the edge