The Pleasure of Patterns in Art

The Pleasure of Patterns in Art The interplay between repetition and variation is central to how we perceive structure, rhythm, and depth across mediums. By: Samuel Jay Keyser A↑ A↓ Off Bright Dark Blues Gray BeeLine Reader uses subtle color gradients to help you read more efficiently. Made at the high point of Kline, de Kooning, and Pollock, Andy Warhol’s “Campbell’s Soup Cans” was a poke in the eye of abstract expressionism. Not only was it blatantly mimetic, but it was being blatantly mimetic with a mundane commercial product found in every supermarket and corner grocery store in America. When people think of repetition in painting, they probably think first of these iconic soup cans. This article is adapted from Samuel Jay Keyser’s book “Play It Again, Sam.” An open access edition of the book is available here. But not all repetition is as in-your-face or as disruptive as “Campbell’s Soup Cans.” One painting from the Impressionist period is particularly pertinent. I am thinking of “Paris Street; Rainy Day” by Gustave Caillebotte. Currently housed in the Art Institute of Chicago, it was originally exhibited at the Third Impressionist Exhibition in Paris in 1877. It is probably Caillebotte’s best-known work. I consider it a masterpiece and regret that I have never seen the real thing. Even so, it never ceases to bowl me over. Discussions of it typically focus on the incredible verisimilitude of the painting, the sense that it is photographic in its vivid capture of an ordinary moment. Thus, art critic Sebastian Smee observes in an article in The Washington Post dated January 20, 2021: Caillebotte compressed different sensations of time and movement into the same picture. A stroll to the farthest visible point could chew up half an hour. But this current predicament — a potential pedestrian collision—will play out in seconds. Do we veer left or right? Our instinctive hesitation is complicated by the man coming into the frame from the right. The space is simply too tight. And all these umbrellas aren’t helping! Smee’s comments are directed not at the form of Caillebotte’s masterpiece, but at its content. That is, of course, not unusual. It is reminiscent of critics of poetry recording their personal evocations for public consumption. But is it helpful? Well, in some cases, when a scene from history or mythology is portrayed, it is nice to know what is going on. Think of “Judith with the Head of Holofernes” by Botticelli, the bodiless head held dangling by its hair from Judith’s left hand, a sword in her right. Explaining what is going on is the stuff of iconography, but it doesn’t help much in shedding light on why the painting is deemed a masterpiece. Gustave Caillebotte’s “Paris Street, Rainy Day” (1877) is designed to elicit visual rhyme as a source of pleasure. Source: Wikimedia Commons When the French painter Paul Delaroche first saw a daguerreotype in 1839, he is supposed to have said, “From today, painting is dead.” He must have been coming from the view that the image was all. From that starting place, it is not unreasonable to ask, What is the point of going to all the trouble of making an image on canvas when a camera can do it much more accurately and much more efficiently? Obviously, Delaroche’s five-word obituary for painting was, like reports of Mark Twain’s demise, premature. But why? What is it about painting that has kept photography from slaying it? I will try to shed some light on that. I begin by asking, Where does the pleasure come from? Smee’s description implies that he finds the painting a source of pleasure. I do as well. But describing the scene isn’t much help. Anyone can do that, though perhaps not as eloquently as Smee. When the French painter Paul Delaroche first saw a daguerreotype in 1839, he is supposed to have said, “From today, painting is dead.” The first thing to observe about this painting is that it is wholly about faces, places, and bodies — three categories our brains are specially wired to recognize. The fusiform face area (FFA) is dedicated to face recognition. Its neurons fire if the subject is looking at a face. But if the subject is shown a tree, or a car, or anything other than a face, the FFA stays quiet. The nearby parahippocampal place area (PPA), meanwhile, responds to environmental scenes like landscapes. Damage to the PPA (for example, due to stroke) often leads to a syndrome in which patients cannot visually recognize scenes even though they can recognize the individual objects in the scenes (such as people, furniture, etc.). A third area is also relevant: the extrastriate body area (EBA), which selectively responds to images of human bodies and body parts (excluding faces). As you can imagine, all three areas are going to light up like the sky on the Fourth of July in the brain of anyone standing in front of Rainy Day. In that regard, it is precisely like the eight centuries of Western painting that went before it, from Cimabue’s “Crucifix” (1288) through Meissonier’s “Campaign of France” (1864) and Caillebotte. But that doesn’t get us closer to why Caillebotte’s painting is so pleasant to look at. Or does it? Art critics regularly bring to a painting history, culture, schools of painting, and so on. But what’s often left out is how the act of viewing shapes mental structures in the brain — how certain arrangements of forms can trigger deep perceptual satisfaction. Let’s look, then, at Caillebotte’s “Paris Street; Rainy Day” not as a recognizable street scene but as an arrangement of geometric objects as depicted in figure 1 below. The first thing that pops out is the extent to which triangles dominate the canvas. The foreground and midground contain five umbrellas. The umbrellas are themselves rounded distortions of a triangle. But notice that the umbrellas are made up of smaller triangles within a triangle. Here is a painting that luxuriates in representations that “reflect the similarity between two like shapes while also preserving the differences.” It is chock-full of visual rhymes. The triangle motif does not end with the umbrellas. Notice the three figures to the left. They make up the points of a triangle. Figure 1: Triangle and rectangle superimposition on “Paris Street; Rainy Day” The triangle motif is also picked up by the buildings. The building to the left of the dominant couple is triangular. Inside the outlines of the building are more triangles defined by the long rows of balconies that run along the facades. Now look at the cupolas on the two buildings to the right. They are each triangular, and taken together they form three points of another triangle. Repetition of the same/except variety appears elsewhere, most notably in the cobblestones. By “same/except,” I mean a visual relationship in which forms are clearly similar, yet slightly varied, so the brain perceives both sameness and difference at once. In other words, objects share a recognizable pattern but are never exactly identical. This interplay between repetition and variation is central to how we perceive structure, rhythm, and depth across mediums (more on that later). With the cobblestones, for example, the closer they get to the lower left-hand edge of the painting, the more elongated they become, thereby lending a ramp effect to that portion of the painting, inviting the viewer to step in and enhancing the sense of depth that the painting exudes. But repetition appears elsewhere as well. Notice the facades of the visible buildings. Their windows, parapets, and balconies are repeated over and over again. This is obvious in the original image. There is another triangle, more subtle than the ones we have just seen, but just as real. In the original painting the wall on the right is reddish-brown. That color is repeated across the boulevard — seven major avenues meet at the Place de Dublin — and if you draw a line to connect the two reddish-brown walls, you have one side of a triangle. If you connect both ends of that line with the most distant single figure, to the left, you have another triangle that encloses the heads of the two main figures on the right, thereby focusing on their line of sight. We are not finished. There is another triangle made up of the three figures on the right about to collide. And this triangle is paralleled by the triangle noted earlier of the figures walking in the street. The Kanizsa triangle illusion And now we can conjecture. The psychologist Elizabeth Margulis has shown us that human beings find repetition pleasurable. Caillebotte’s painting is filled with objects that provide the viewer with the opportunity to construct triangles. That humans can do this is demonstrated by the famous Kanizsa triangle illustrated to the right. Looking at this figure, you cannot help but construct a triangle and interpret the illusion as a white triangle sitting on top of three black discs. In reality, there is no triangle. There are simply three Pac-Man-like objects placed in such a way that you see a triangle. Caillebotte has done much the same thing. He has painted objects that induce the viewer to construct triangles and he has made the triangles in same/except pairs. The discovery of this visual same/except relationship is the source of the painting’s pleasure no less than rhyme is a source of pleasure in a poem. Indeed, Caillebotte’s “Paris Street; Rainy Day” is designed to elicit visual rhyme as a source of pleasure. The same is true of the rectangular and square motifs in the painting. It is noteworthy that as strikingly photographic, familiar, dramatic — what have you — the painting is, one source of its pleasure has nothing to do with the content of the image but its shape, a shape that forces the viewer to find the rhymes. There is one more property of Rainy Day that I would like to draw attention to, one that has nothing to do with triangles. It has to do with parerga. An interesting line of inquiry in aesthetic theory concerns the parergon. What is a parergon? Kant gives one definition in his “Critique of Aesthetic Judgment”: Even what is called ornamentation (parerga), i.e., what is only an adjunct and not an intrinsic constituent in the complete representation of the object, in augmenting the delight of taste does so only by means of its form. Thus it is with the frames of pictures or the drapery on statues, or the colonnades of palaces. But if the ornamentation does not itself enter into the composition of the beautiful form — if it is introduced like a gold frame merely to win approval for the picture by means of its charm — it is then called finery [i.e., decoration] and takes away from the genuine beauty. So Kant distinguishes between ornamentation, something that is related to the subject but is not central to it, and finery (decoration), something that detracts from the beauty of the object. The notion of parergon has a long history. Strabo discusses it in his “Geography” in a story about the Greek painter Protogenes (fourth century BCE), who painted a satyr leaning against a column playing a flute. A partridge sat atop the column. The partridge was so well drawn that admirers of the painting ignored the satyr. So Protogenes expunged the partridge because it was not central to his painting. It was a parergon. As Paul Duro notes in his very helpful historical survey, Jacques Derrida resurrected the idea of the parergon: From the time Jacques Derrida introduced the term into contemporary critical theory in “The Truth in Painting,” its character and function have been largely understood as referencing a threshold or boundary — in particular, that of the border of the artwork. Yet a review of the term’s long history suggests a meaning that differs in significant ways from its current near-univocal characterization as a synonym for a frame. In 1533 Lucas Cranach the Elder painted a portrait of the Roman matron Lucretia’s tragic suicide. Derrida discusses the painting from the point of view of the parergon. It is interesting to read his struggle with the term: Where does a parergon begin and end. . . . For example, Cranach’s Lucretia holds only a light band of transparent veil in front of her sex: where is the parergon? . . . A parergon, the necklace that she wears around her neck? . . . If any parergon is only added on by virtue of an internal lack in the system to which it is added . . . what is it that is lacking in the representation of the body so that the garment should come and supplement it. And what would art have to do with this! I suppose there might be some value in categorizing the contents of a painting in terms of what is central and what is peripheral, for example, the ornate armrest in the “Madonna della Sedia.” But I think the notion contains within it a certain danger, namely, not seeing the forest for the trees. Look again at Cranach’s “Lucretia.” There are two horizontal lines in the painting. One is Lucretia’s necklace. The other is the transparent veil just beneath her mons pubis. These strike me as important markers in the painting — the one underscoring her face, the second underscoring her sex. They constitute a frame within a frame; that is, they embody the current meaning of parergon. They confine what is important in the image, the dagger aimed at Lucretia’s heart because of Tarquin’s abomination. This frame within a frame says these are the important places in the painting. If this is true, then what seem to be candidates for parerga, the veil and the necklace, are not. “Lucretia,” painting (1533) by Lucas Cranach, the Elder This brings us back to “Paris Street; Rainy Day” and the lamppost. Is the lamppost an integral part of the painting or is it, like Protogenes’s partridge, a parergon, a subsidiary ornament that can be done away with or, at the least, downplayed? To begin with, it cuts the painting in two. Indeed, as figure 2 shows, the painting is made up of two separate paintings. Each half can easily stand alone. Now look at the doctored version in figure 3. What’s missing is the bisecting lamppost. Taking out the lamppost flattens the painting. The dimension that Smee talks about, the sense of distance between the dominating couple in the foreground and the spot at the painting’s central vanishing point, disappears. The foreground fades into the background. But insert the lamppost and the sense of depth pops out. Figure 2: “Paris Street; Rainy Day” split into two separate paintings along the line of the lamppost Figure 3: “Paris Street; Rainy Day” without lamppost This strongly suggests that the lamppost, on first viewing a reasonable candidate for a parergon, is in fact the opposite. It is an essential part of the painting. Evidence for this is that when Caillebotte painted Rainy Day, there was no lamppost in that spot. There couldn’t be. That style of lamppost had been removed from the Place de Dublin years before. So Caillebotte put it in for a reason, namely, to enhance the painting’s sense of depth. The overarching point is that we need to have a coherent view of what makes a painting work before we can begin to decide what is a parergon and what is not. I do not think that it is stretching the point to suggest that Caillebotte’s “Paris Street; Rainy Day,” Poe’s “The Raven,” and Duke Ellington’s “Satin Doll” are works of art from three distinct genres each appealing to the same cognitive function: namely, the ability to make same/except judgments establishing repetitive patterns, the detection of which, per Margulis, elevates one’s aesthetic enjoyment. Let’s turn to some photographic examples. The introduction to the eighth edition of the well-known “Janson’s History of Art: The Western Tradition” discusses in detail a photograph by Lee Friedlander, “Albuquerque, New Mexico” (1972), coincidentally also in the collection of the Art Institute of Chicago. Presumably the authors had two goals in mind when they included this photograph in their introduction. The first was to underscore that photography is an art form worthy of a place alongside painting. The second was to demonstrate how to think about a work of art in general and photographs in particular. Lee Friedlander, “Albuquerque, New Mexico” (1972) © Lee Friedlander. Courtesy of Fraenkel Gallery, San Francisco and Luhring Augustine, New York Take a look at the photograph in question above. And now consider a sample of commentary from the book: In “Albuquerque,” Friedlander portrays a modern America that is vacuous and lifeless, which he suggests is due to technology. How does he do this? The picture has a haunting emptiness. It has no people, and it is filled with strange empty spaces of walkway and street that appear between the numerous objects that pop up everywhere. . . . Everywhere, nature has been cemented over, and besides a few scraggly trees in the middle ground and distance, only the weeds surrounding the hydrant thrive. In this brilliant print, Friedlander captures his view of the essence of modern America: the way in which technology, a love of the artificial and a fast fragmented lifestyle were spawning alienation and a disconnection with nature and spirituality. Like Sebastian Smee’s account of “Paris Street; Rainy Day,” this commentary focuses on the content of the photograph and not on its form. To that extent it is telling us about how the authors feel about the image rather than how the image works as art. In other words, it is an evocation. Thus, they are “disturbed” by various features in the photograph, and shadows appear to be “mysterious” (p. xxix): Disturbing features appear throughout the composition. The street sign — which cannot be seen because it is cropped at the top of the print — casts a mysterious shadow on the wall. A pole visually cuts the dog in two, and the dog has been separated from his attribute, the fire hydrant, as well as from his absent owner. Criticism that talks of images as being “disturbing” or “mysterious” has its place. One might think of it as evocative reportage. My interest lies elsewhere, namely, in explaining why this superficially mundane photograph works so well. Let’s take a closer look. A dog is sitting on the sidewalk. It is the only living thing in the photograph, although its collar hints at an owner waiting in the wings. The eye is drawn to the living dog. But it is not the only dog in the photograph. At virtually the same latitude, a larger shadow dog is mimicking the real dog (see figure 4). Both dogs have open mouths and extended tongues. The shadows of the shadow dog and the shape of the concrete wall make it easy to read back, body, and behind. The shadow dog is created by the sun shining behind a street sign atop the pole that bifurcates the real dog but is out of sight at the top of the photograph. It is the street sign that connects the two images. The pole cuts the real dog in half only to make him whole again in the shadow that it casts. Looked at this way, the “mysterious” shadow loses its mystery. Figure 4: “Albuquerque, New Mexico” with real dog and shadow dog encircled The calculated nature of this photograph was not lost on the authors: Friedlander did not just find this composition. He very carefully selected it and he very carefully made it. He not only needed the sun, he had to wait until it was in the right position (otherwise, the shadow of the fire hydrant would not align with the street). I agree with all that, except it wasn’t the shadow of the fire hydrant that had caught Friedlander’s eye. It was in fact the dog and shadow dog described above. The calculated character of the photograph is supported by this simple operation. Cut the photograph in two. Now in your mind’s eye superimpose one half over the other. Shadow dog (not hydrant) and real dog are in precisely the same spot. In other words, the right half is a copy of the left half with respect to the dog. They are the same, except one dog is real and one dog is a shadow. The photograph is another example of a visual rhyme, only it was not constructed à la Caillebotte with paint. It was constructed à la Friedlander with a camera. And we have seen that same/except repetition elicits pleasure. During an online presentation at the Fifth New York Institute on December 9, 2023, I showed Albuquerque and asked 37 viewers if they saw the shadow dog in it. Twenty-seven (63%) did, while 10 (37%) did not. Although a small, spontaneous poll, this split likely reflects a real difference in perception. There are photographs, of which “Albuquerque” is one, that exploit the entire spectrum of gray shades from black to white. In such photos the ends of the spectrum, pure black and pure white, tend to be the first focal point. As it happens, in Albuquerque, the real dog is the purest black and the area around the shadow dog, the column with the shadow dog image, is the whitest. I suggest that seeing the shadow dog depends on which focal point you view first. If the first place you look at is the whitest area, then your eye will encounter the image on the column before it encounters the dog. In that case, you are far more likely to see the shadow as the number 7 rather than as a shadow reflection of a dog. I think this is crucial — the order in which you encounter the shadow and the dog. If you encounter the real dog first, then you are more likely to see the image on the column as a shadow dog. Why? Because the number 7 does not prime the figure of a dog, but the image of a dog primes its shape in a shadow. The reason why the majority of viewers in my poll saw the shadow dog is because one’s eye tends to go to the blackest area first. If you encounter the real dog first, then you are more likely to see the image on the column as a shadow dog. That at least is my conjecture. The work of the American-born artist Roni Horn makes an interesting parallel to Warhol’s “Campbell’s Soup Cans,” discussed earlier. Two examples are found in “Becoming a Landscape,” below. Each indicates a recurrent technique in Horn’s work, pairing two images that are the same except for tiny differences. Just as Warhol’s soup cans constitute a form of visual rhyme in painting, Horn’s works do the same in photography. The viewer is invited to inspect the images to tease out their same/except character. There are differences, very small ones; but we humans are hardwired to make those determinations and doing so as an artistic task is pleasurable. I believe Horn is exploiting this aspect of our same/except detection capability. Roni Horn, “Becoming a Landscape” (detail) 1999–2001 © Roni Horn. Courtesy of the artist and Hauser and Wirth Roni Horn, “Becoming a Landscape” (detail) 1999–2001 © Roni Horn. Courtesy of the artist and Hauser and Wirth When you glance from one portrait to the other in the first image, there is no doubt that the expressions are not identical. But it is a matter of great subtlety to pinpoint exactly what the difference is. Are the eyes wider in one? Are the lips slightly different? On a website that carries Horn’s work, the second pair is described like this: The two photos are almost identical, but if one looks at the splash of water, you will see the difference. Horn requires that you look and discover. The “identity” is always different. Or to put it differently, the rhyming images are always the same/except. Calvin Klein mirror image ad It is not surprising to find Roni Horn’s technique exploited in the advertising world. I was reading a New York Times article on my iPhone when, while scrolling, I encountered an ad for Calvin Klein. The ad wants to persuade you to give someone Calvin Klein gifts for Christmas. But the image isn’t just hawking lingerie. It’s an example of the Roni Horn strategy. It is intended to stop you from scrolling past by offering a puzzle. You are invited to determine whether the figures are identical mirror images or the same, except for a small difference. The hope, I suspect, is that you will accept the invitation because it presents a visual same/except puzzle, a potential source of pleasure. Once you have become invested in the image, you will be inclined to investigate the product. One doesn’t have to look hard to find examples of visual rhyme everywhere where photos are the lingua franca. One doesn’t have to look hard to find examples of visual rhyme everywhere where photos are the lingua franca. Let me close this essay the way I opened it, with an image that is saturated with a same/except motif. The photograph is “Girls in the Windows,” the photographer, Ormond Gigli. This photograph from 1960 is among the highest-grossing photographs of all time, perhaps the highest. With prices ranging between $15,000 and $30,000 a print, the sale of 600 copies has added up to $12 million and it is still selling all over the world. The vision to produce an inventory of prints belongs to Gigli’s son, Ogden, a photographer in his own right. In a New York Times article, David Segal (2023) quotes Ogden Gigli as saying, “It was my belief that the appeal of this image would carry on forever.” Segal claims that an important part of the photograph’s appeal starts with the image, of course, which is a brassy, joyful combination of glamour and urban grit with a dash of “Mad Men”–era nostalgia. The building embodies a glorious slab of vanishing New York City, and those women look like they’re ready to break into song. Ormond Gigli, “Girls in the Windows,” 1960 That description could fit several images — for example, Marilyn Monroe’s encounter with a subway vent. But it doesn’t explain the photo’s eternal appeal. For that I think you have to look to its form. “Girls in the Windows,” which predates Andy Warhol’s “Campbell’s Soup Cans” by two years, is a perfect example of visual rhyme. The photograph contains 40 models in 40 windows. What is the same is the windows. What is different is each model in her self-selected gown and pose. Think of the photo as a visual representation of multiple rhyme schemes. In every case, the window will be the same except for the model in it. In other words, the windows are to the Campbell’s soup cans as the models are to the different Campbell’s soup can labels. The delight in these works — from Warhol to Friedlander, Horn to Gigli — comes from the same source. Our eyes trace patterns, spot subtle variations, and construct visual rhymes, taking satisfaction in order amid difference. The satisfaction of solving this visual puzzle is a key part of the pleasure it provides. Samuel Jay Keyser is a theoretical linguist. He is Peter de Florez Emeritus Professor of the Linguistics and Philosophy faculty, and former Associate Provost at MIT. He has authored numerous books, including “The Mental Life of Modernism” and “Play It Again, Sam,” from which this article is adapted, and is Editor-in-Chief of the journal Linguistic Inquiry. An open access edition of “Play It Again, Sam” is available here.