1 To appear in Partisan Canons, Anna Brzyski (ed.), Duke University Press. Mere Exposure, Reproduction, and the Impressionist Canon James Cutting Every act of writing or curatorial practice, whenever it gets to the point of naming a name, is participating in a certain level of canon formation, no matter what the intent of its author, no matter whether it represents a challenge to the status quo or a confirmation of it. Russell Ferguson1 It has been said that canons are the “legitimating backbone of cultural and political identity.”2 They are also the bread and butter of what is taught in the humanities, and in art they are what the general public flocks to see. Art museums and art historians feature canonical images. They reproduce them in great quantity such that these images are now seen in greater numbers, and by greater proportions, of people than ever before. Blockbuster exhibits guarantee continuing links among particular images, publicity, and capital—both cultural and otherwise. But what exactly are the contents of a given canon? How do we determine which works are canonical? And how did they attain that status? In an attempt to answer these questions, I will focus on the canon of French Impressionism, and argue that canons are cultural constructs created, in part, through repeated reproduction and exposure. That is, following the epigraph of Russell Ferguson, curator at the Museum of Contemporary Art, Los Angeles, I claim that art historians have been complicitous in the formation and maintenance of the Impressionist canon, and that over decades their wares have fed a sustaining public whose opinions, along with those of art historians themselves, have crystallized and now drag heavily against systematic change.3 There are several reasons for choosing French Impressionism as a case study. First, it is a relatively recent, well documented, and well defined art phenomenon. These facts make it easier to outline its formation, maintenance, and structure. Second, it is sufficiently old that virtually the entire corpus of works is known and accounted for. The artists collectively identified as the Impressionists have been dead for about a century, and about thirty-five percent of their work is currently owned by museums.4 The remainder is in private hands, but as I will demonstrate no canonical images come those holdings. Third, and perhaps the most significant reason for this study, is that French Impressionism was extraordinarily popular across the twentieth century and continues to be so today.5 The Impressionist images bequeathed to the State of France were gathered into the Louvre by the mid-1930s and then sent to the Jeu de Paume in 1947.6 There, overcrowding soon became a problem. In the 1960s through the mid-1980s the Musée du Jeu de Paume was the most heavily trafficked museum per square meter in the world. Significantly, many of those visitors were from the US. In 1986, the contents were then moved to the Musée d'Orsay, which was instantly one of the most visited museums in the world, receiving over four million people annually.7 To be sure, there are also well-known and frequently seen collections of French Impressionist paintings elsewhere in Europe and, in particular, in the US. In addition, French 2 Impressionist paintings have often commanded the highest sale prices at art auctions throughout the twentieth century, and over the course of the 1980s and 1990s they were featured in some of the largest and best-attended exhibitions.8 Moreover, French Impressionism’s high public profile, popularity, and importance for the history of modern art generated, over the course of the twentieth century, a thick texture of literature on the artists and their oeuvres that I will draw upon. Finally, Impressionism is deeply embedded in contemporary American popular culture. Consider two sources of supporting evidence. First, in his 1987 book, Cultural literacy: What every American should know, Edward Hirsch listed thirty-one artists across all of time that he felt US citizens should know. Five are Impressionists.9 Hirsch’s volume, however odd it may appear, can be interpreted as an early and politically charged salvo in the “culture for dummies” offensive. His “requirement” for Americans to learn about Impressionism was surely motivated by its historic visibility in the US. Many of its most avid collectors were American, and those collectors tended to give works to American museums. Indeed, five of the seven museums with the largest, most diverse Impressionist collections are in the United States—the Metropolitan Museum of Art in New York, the National Gallery of Art in Washington, the Philadelphia Museum of Art, the Art Institute of Chicago, and the Museum of Fine Arts in Boston. The works within their galleries—along with those in the Musée d’Orsay and the National Gallery of Art in London—are icons of modernism and are deeply embedded in Western visual culture.10 They even appear on towels, scarves, coasters, t-shirts, refrigerator magnets, and textbook covers. Consider a second source—Impressionism as displayed within contemporary US mailorder catalogs. Among the hundreds of catalogs my household received in 2003, many had images of rooms with furniture, lamps, and rugs for sale, but they also showed books on shelves, coffee tables, and desks. These books were not for sale, but were included gratuitously to create a particular ambience in each room, populating it with appropriate tokens of concretely identifiable social aspirations. The presence of such books is not accidental. They are almost always about food, travel, or art—three aspects of conspicuous American consumption. The books on single artists are the most interesting. After taking care to exclude duplicate images, I focused on seventeen catalogs. In those, I found books on just nineteen different painters, excluding a few from the mid- and late-twentieth century. Van Gogh was the most common—six different catalogs had different images of books with the title Van Gogh on its cover or spine; Cézanne was second (5); next were Picasso (4), Rembrandt (4), Leonardo (3), Michelangelo (3), Mondrian (2), and Sargent (2). Artists singly featured were Breugel, Cassatt, Duchamp, Gauguin, Goya, Monet, Renoir, Seurat, Toulouse-Lautrec, Velázquez, Vermeer, and Whistler. French Impressionists (4) and artists associated with broader late-nineteenth and early-twentieth century French art (8 more) dominated in these portrayals of upscale American domesticity. Anecdotal as it may be, such a survey clearly demonstrates that certain high art is part of popular culture. It is linked to the commerce of goods catering to upwardly mobile, middle class tastes, where demand for high quality is met and reinforced through association with canonical art. Mere Exposure, Preference, and Value Mere exposure is a key term in psychology, my own field of research. Robert Zajonc discovered the phenomenon in late 1960s. Among other things, he demonstrated that the more 3 times a nonsense word like dilikli was repeated, the more likely a listener would later think it meant something good rather than bad. Since then, laboratory studies with many kinds of materials—unfamiliar graphic characters, nonsense geometric constructions, photographs of unfamiliar people, as well as melodies and musical passages—have confirmed the validity of the mere exposure effect.11 The basic finding is that, other things being equal, we tend to like things we have seen before, indeed we tend to value them more.12 My own research has shown that this also holds true for art. Repeated exposure to particular images creates and reinforces preferences. I conducted a set of experiments on the effects of mere exposure to paintings and pastels. I presented pairs of French Impressionist images to undergraduates (only 17 percent of whom had previously taken an art history course), older adults (graduate students and faculty in the Cornell psychology department), and children ages 6 to 10.13 I chose images widely, but they were generally by Paul Cézanne, Edgar Degas, Edouard Manet, Claude Monet, Camille Pissarro, Pierre-Auguste Renoir, and Alfred Sisley.14 I selected sixty-six pairs of paintings. The images in each pair were by the same artist, produced roughly in the same year, and fell within the same genre—portrait, group portrait, landscape, cityscape, seascape, still life, nude. Some images are part of the Impressionist canon, as I will discuss later, but most are seldom reproduced and are familiar only to the specialist. Because I was interested in mere exposure, I needed a gauge for how likely and often individuals might have seen each image before. The one I chose was indirect, but I think compelling: I counted every occurrence of each of the 132 images in all the books that Cornell University owns.15 I found these images in almost a thousand different books, published between 1901 and 2002, mostly in English and French, some in German, and a few in Italian, Dutch, Spanish, Danish, and Japanese. Results of those tallies ranged from two reproductions (one of which was always from a catalogue raisonné) to almost three hundred (an image included in over a quarter of the books). I do not presume that any of my subjects spent time thumbing through the art books in Cornell’s libraries, but I do assume that the relative reproduction frequency of images well approximates the relative likelihood of their having been seen before. I asked observers to indicate their preference between the two images. The result, consistent with the mere exposure thesis, was that individuals preferred the more frequently reproduced image in each pair about sixty percent of the time. Moreover, the greater the discrepancy in relative frequency of reproduction, the more likely viewers preferred the more reproduced image.16 Interestingly, few participants in this study recognized particular paintings, and recognition was unrelated to preferences. This latter result was not a surprise. The effects of mere exposure are not a rational response to one’s surroundings. Research has shown we often cannot express the reasons for what we like, but all evidence here points to the fact that what we prefer, we have likely seen before and seen more often.17 There is, of course, the thorny issue of quality. One might claim that paintings that are reproduced more often are “better” paintings and that people respond to quality in art. This latter idea, associated with Kantian aesthetics, was occasionally embraced by modernist art history, but it is not likely to be true.18 In another experiment I presented the same images to students in my undergraduate course on perception. I randomly interspersed them among the slides in my PowerPoint presentations on various scientific topics. I presented them as nonsequitors for a few 4 seconds, each without comment. Across two dozen lectures I showed my class four times the images that were less frequently reproduced in each pair. I presented the more frequently reproduced images only once. At the end of the semester, I performed the same experiment with this class and found that I had reversed preferences. The students now slightly preferred the images that they had seen more often in class.19 Which Images Are in the Impressionist Canon? Let me offer a thesis contrary to that of Walter Benjamin in “The Work of Art in the Age of Mechanical Reproduction.” Canonical images—the images that hold the greatest prestige and that are considered to be the most significant, and therefore most valued by experts and the public alike—are those images that have been reproduced most often.20 That is, the aura of art in the age of reproduction does not wither but, as Michael Camille also pointed out, is reinforced.21 To determine which images are canonical for French Impressionism, I consulted a second sample, which included all books on Impressionism in the Cornell University libraries—a total of nearly one hundred publications covering the breadth of the twentieth century. “As Harry Abrams once said … about such publications, nobody reads the text anyway; it’s all about the … reproductions.”22 Prompted by the epigraph of Russell Ferguson, I tabulated all images by title and by artist.23 But first, how many French Impressionist paintings are there? If one counts images in catalogues raisonnés of Cézanne, Degas, Manet, Monet, Pissarro, Renoir, and Sisley, the answer is about 9,000. If one includes others—Frédéric Bazille, Gustave Caillebotte, Mary Cassatt, Eva Gonzalès, Armand Guillaumin, and Berthe Morisot—the answer is closer to 12,000.24 How many of these 12,000 appeared in the books included in my sample? The answer is just over 2500. More interestingly, about 1400 images appeared only once, about 975 between two and nine times, and only 138 appeared ten times or more. Figure 1, which presents these data in a graphic form, reveals a veritable iceberg of Impressionism. It has a vast number of unreproduced images at the bottom, floating beneath the surface of visibility. The uppermost spire represents the most frequently reproduced images—the core canon of Impressionism. Notice that it is miniscule in proportion to the entire Impressionist corpus. The spire also demonstrates the extreme consensus with which scholars reproduce images to tell the story of Impressionism. For the purpose of further analysis, I will define the core canon of French Impressionism as the fifty most frequently reproduced images.25 Who Originally Owned the Core Canon? The bulk of the core canon was owned by a very small number of people. The most important of these was Gustave Caillebotte, whose story has been told many times.26 Inheriting his father’s wealth, Caillebotte was a millionaire and a painter at age 28. He was invited by Renoir and by Henri Rouart to join the second Impressionist exhibition. He participated in, organized, and bankrolled it and four others. More importantly here, he also bought his friends’ paintings. In 1894 Caillebotte died suddenly of stroke, and left his collection to the French state. His only condition was that the collection should be exhibited intact. Haggling went on for years taxing Renoir, the executor of Caillebotte’s will. Eventually thirty-nine artworks were accepted, plus two of Caillebotte’s own paintings donated by the family. This part of his collection was installed in the Musée du Luxembourg in Paris in 1897. Strikingly, Caillebotte owned eight images now in the core canon, all in the Musée d’Orsay—Degas’ Femmes à la terrasse d’un 5 Figure 1: A graphical depiction of all of the images of Impressionism by their frequency of appearance in ninety-five books, a veritable iceberg. café, le soir (1877), Manet’s Le balcon (1869), Monet’s Régates à Argenteuil (1872) and his La gare Saint-Lazare (1877), Pissarro’s Toits rouges (1877) and his Printemps à Pontoise, potager et arbres en fleurs (1879), and Renoir’s Bal au Moulin de la Galette (1876) and his Balançoire (1876).27 Only one other collector comes close to Caillebotte, the financier Isaac de Camondo. Camondo was a leading banker in fin de siècle France and a quiet member of an important Turkish-immigrant family. He began buying Impressionist works in 1893 and collected until his death in 1911. By his will, its endowment, and his considerable political clout, his collection went straight to the Louvre.28 Camondo owned six core canon images, all now in the Orsay—Cézanne’s La maison du pendu (1873), Degas’ Absinthe (1876) and his Repasseuses (1884), Manet’s Le fifre (1866), the most popular of Monet’s Cathédrales de Rouen series (1891, he owned four), and the most popular of Sisley’s l’Inondation à Port-Marly series (1876, he owned two).29 In England two collectors and benefactors promoted Impressionism. The first was Hugh Lane, an Irishman and successful London art dealer. In 1908 he established Municipal Gallery of Modern Art in Dublin, now known as the Hugh Lane Gallery. Lane had a small but remarkably important Impressionist collection. He died on board the Lusitania in 1915, and by his official will his collection was given to the National Gallery of Art. But in an unwitnessed codicil Lane stated that the paintings should go to the new gallery in Dublin. The court did not honor the amendment and the paintings went to London. Years of furor and negotiation followed. Finally, an agreement was reached in 1959 and, at least ostensibly, the works are now shared between the two museums. Three Lane images are in the core canon—Manet’s La musique aux Tuileries (1862), Renoir’s Les parapluies (1881-1886), and Morisot’s Eté (1879)—but several more are not far behind.30 The other English patron was Samuel Courtauld, the silk magnate. He is important here for two reasons. First, impressed by Lane’s collection, he established a fund for the National 6 Gallery and the Tate to purchase modern works, and he oversaw that fund. Between 1923 and 1926, the Courtauld Fund purchased ten Impressionist paintings, and other more recent works. Second, he also established his own museum in what is now the Courtauld Institute. After the death of his wife in 1931 he gave his paintings, the lease to his house, and an endowment to the University of London.31 Combining those purchased through his fund and those from his private collection, Courtauld accounts for three images in the core canon—Manet’s Le bar aux FoliesBergère (1881, Courtauld Institute), Renoir’s La loge (1874, Courtauld Institute), and Monet’s La plage à Trouville (1870, National Gallery, London). Finally, the Havemeyers were the most important US benefactors of Impressionism— with bequests to the Metropolitan Museum of Art and, through their children, to other museums. In particular, Louisine Elder Havemeyer was active in the women’s suffrage movement, a close friend of Mary Cassatt, and intensely interested in Impressionism. The Havemeyers owned over one hundred fifty Impressionist works, and more than sixty went to the Met. Three Havemeyer paintings are in the core canon—Monet’s La Grenouillère (1869, Metropolitan), Manet’s En bateau (1874, Metropolitan), and his Le chemin de fer (1873, National Gallery of Art, Washington).32 So few hands controlled so much of the French Impressionist canon. Five collectors— Gustave Caillebotte, Isaac de Camondo, Hugh Lane, Samuel Courtauld, and Louisine Havemeyer—account for almost half of it. Interestingly, other collectors who left important bequests to various museums—Etienne Moreau-Nélaton (mostly in 1906) to the state of France; the Palmers to the Art Institute of Chicago (1922), Antonin Personnaz to the Louvre (1937); Paul Gachet fils to the Louvre (through the 1950s); the Tysons to the Philadelphia Museum of Art (1956); and Chester Dale (1963), Paul Mellon (1970-1999) and Ailsa Mellon Bruce (1970) to the National Gallery Washington—together account for only three images in the core canon. These are Manet’s Le déjeuner sur l’herbe (1863, Moreau-Nélaton bequest to the French state), Cézanne’s pastiche, Une moderne Olympia (1872, Gachet bequest to the Louvre), and Renoir’s Les grandes baigneuses (1887, Tyson bequest to the Philadelphia Museum of Art). Most of these donations generally came later than those discussed earlier. In the case of Moreau-Nélaton’s bequest, it languished for thirty years neither in the Luxembourg or the Louvre, and in the case of the Palmers’ it simply achieved less publicity.33 Moreover, as one can see in Figure 2, the accrual of canonical images in museums peaked early, in the decades of the 1910s and 1920s. As Impressionism became accepted, even mainstream, fewer of the images given to museums would achieve canonical status. Who Sold the Core Canon? Dealers handled fewer than half of these core images. This contrasts with the larger corpus of 12,000 images, where nearly two-thirds were controlled at one time or another by sixteen different dealers.34 Accounting for half of the core canon images are friends who bought directly from the artists (like Caillebotte and Paul Gachet père), the families of the artists (like Marc Bazille, nephew of Frédéric, for Atelier de l’artiste, rue Condamine (1869) and Réunion de famille (1867); the De Gas family for La famille Bellelli, 1858-1861, held back from the fourth Degas estate sale; and the Pontillon family of Morisot’s sister for Le berceau, 1872) and the families of friends (like the Dihau family, featuring the oboist in Degas’ Orchestre à l’Opéra, 1870-71). A few others were direct museum purchases (Degas’ Portraits dans un bureau, 7 Figure 2: A graphical depiction of the museum accrual by decade of what would later be the fifty Impressionist images in the core canon. Nouvelle-Orléans, 1873, Musée des Beaux-Arts, Pau), or gifts by subscription (e.g. for Manet’s Olympia, led by Monet and John Singer Sargent to support Manet’s widow and to keep the painting from going to the US). Thus, after one omits those paintings not handled by dealers, only half remain. Despite this, one third of these fifty images—or two-thirds of those remaining—were handled by Paul Durand-Ruel. From 1871 until his death in 1922 Durand-Ruel made his living by selling both the works of Impressionists and the more acceptable Salon painters. For sales of works now in the core canon, the next most important dealers were the Galeries Bernheim-Jeune, Paul Rosenberg, Georges Petit, Ambrose Vollard, and Wildenstein in that order, but the sum of these five is only half the total of Durand-Ruel.35 Perhaps Durand-Ruel had extraordinary taste, guiding his buyers to the best images and allowing him to play a huge role in forging the Impressionist canon? This is possible, but not likely. Without denying his importance for the dissemination of Impressionism,36 there is little evidence that he specifically controlled which paintings would be canonical. Yes, he dealt with a third of these fifty images, but he also dealt with one third of all the images produced by the Impressionist artists. Over a period of fifty years he sold more than four thousand paintings by thirteen Impressionist artists. By this account he was an enormously successful salesman for Impressionism, but he had no special impact on its core canon. Who Publicized The Core Canon? The Impressionist canon grew incrementally. Beginning just after the turn into the twentieth century, scholars began to reproducing images in the core canon in their books. In order to determine the history of the Impressionist canon’s reproduction, I analyzed a third sample of twenty five books published in French, English, and German between 1901 and 1949 8 Figure 3. A scatter plot of ninety-five books on Impressionism published between 1901 and 2001, and the number of images they included that are now in the core canon. that reproduced any Impressionist images at all.37 On average, each book introduced one or two new images that would later become canonical. But two books clearly stand out – Charles Borgmeyer’s The Master Impressionists (1913),38 an otherwise minor work that introduced nine new canonical images, all from the Caillebotte and Camondo legacies; and John Rewald’s History of Impressionism (1946), which introduced ten. Moreover, Rewald’s perspicacity extended beyond the presentation of new images. He also republished two dozen core images that appeared in earlier books. This means that he published two-thirds of the core canon. Before him even Borgmeyer had published only one third. Shown in Figure 3 are the numbers of now core canon images that appeared in all ninety-five books. Notice the pattern—a gradual coalescence of agreement about what is in the canon across the twentieth century, and Rewald led the way. His accomplishment was not eclipsed in a single volume for thirty-five years.39 Clearly, Rewald played an extraordinary role in shaping the canon. What he published, several generations of art historians who followed also published. Indeed, no Impressionist scholar of last half-century could ignore what Rewald presented. Has the Core Canon Changed? Over the course of the twentieth century the discipline of art history may have changed as much or more than any other. It began the twentieth century with biographies and connoisseurship, it ended it with sophisticated social histories and theoretically driven analyses. Did the reproduction of canonical images change as well? No, or at least very little. Analyses of my databases show that eighty-five percent of the images that were in the core canon in the books published in the first half of the century are still there today. Thus one could argue that even though the arguments are relatively fresh in the field of art history, the images are not. 9 To be sure, some images are published a bit less often. Renoir’s Bal au Moulin de la Galette was by far and away the most reproduced Impressionist image in the first half of the twentieth century. It is now merely among the top four. And Degas’ Etoile, danseuse sur la scène (1876-78, also from the Caillebotte bequest) was once among the top dozen images, but has subsequently fallen out of the top fifty. Many other images are also now published more often. For example, those of Bazille, Caillebotte, Cassatt, Gonzalès, and Morisot now appear more frequently, but they are largely not in the core canon. But consider two other images whose reproduction has burgeoned. Since it was given to the Musée Marmottan in 1957 Monet’s Impression, soliel levant (1873, Donop de Monchy bequest) has become the most reproduced Impressionist image, even outstripping in this context Manet’s Olympia (1863) and his Déjeuner sur l’herbe (1863). This is in large part due to the retelling of the semi-apocryphal tale—first popularized by John Rewald—of this image, the art critic Louis Leroy, his derisive review of the first Impressionist exhibition, and his supposed naming of the group.40 In addition, as a centennial purchase for the Met in 1967 Monet’s Terrasse à Sainte-Adresse (1867) has also leapt into the core canon. Nonetheless, such images are exceptions. The Impressionist canon, perhaps like all others, is relatively fossilized. Scholars continue to display images once owned by Caillebotte, Camondo, Lane, Courtauld, and the Havemeyers, and they continue to follow Rewald. Interestingly, the art promoter least encumbered by this hegemonic uniformity, showing most often and most consistently noncanonical images, is Sister Wendy Beckett.41 Why Reproduction Matters? Let me return to mere exposure. There is another group that plays an important role in maintaining the canon —the general public, the “audience” of art. Members of the public may acquire knowledge about the canon through overt study of individual paintings, but more broadly and more importantly they are influenced by mere exposure. All other things being equal, the more often they see or hear something—so long as this is distributed across time, rather than massed in a small amount of time—the more they will tend to like it. This mostly unconscious exposure shapes their preferences and, as noted by Patricia Mainardi, this “audience exerts a profound influence on the kind of art history we produce.”42 Equally importantly, scholars are no different from the public in the effects of mere exposure. From childhood through college and throughout adulthood, we are all exposed to hundreds of thousands of images. Some are representations of art; others, as during a museum visit, are the artworks themselves. We don’t remember each occurrence of each image, or where we saw it. We often will not even recognize the image if we see it again, but its trace can influence our future assessments. These are not overt cognitive responses on our part. They are not directly related to the formal part of our education, but they are very much a part of our general and higher education. The effects of mere exposure are quite automatic and independent of what we pay attention to in our day to day activities. They accrue simply as the result of being a member of a culture, experiencing cultural artifacts. If all other things are equal, the more canonical images are published, the more likely everyone—the public as well as the scholar—tends to like them. Walter Benjamin suggested that “reproduction of art changes the reaction of the masses toward art.”43 Indeed, I would argue, it has effects far beyond, and in opposite direction of, those he considered. 10 Endnotes I thank Claudia Lazzaro for many discussions over several years on the topic of canons; and Anna Brzyski for her unrelenting eye for editorial detail, her unsurpassed enthusiasm, and her patience with my own professional deformities. This chapter is based, in part, on my monograph Impressionism and Its Canon (Lanham, MD: University Press of America, 2006). 1 Russell Ferguson, “Can we still use the canon?” The Art Journal, 58 (1999), p.4. 2 Griselda Pollock, Differencing the Canon: Feminist Desire and the Writing of Art’s Histories (London: Routledge, 1999), p. 3. 3 This argument bears some resemblance to ideas in the sociology of art (see, for example, Janet Wolff, The Social Production of Art (New York: New York University Press, 1981), except that it applies not simply to the production of individual images, but to the relationships among all images in the canon as a whole. This approach is also very much against the zeitgeist of work within Impressionist scholarship, which generally regards the corpus of canonical images as given and not shaped by their own efforts, or those of collectors, curators, and chance. Among the hundred books on Impressionism that I have surveyed, few authors mention what governed their choice of images. Michael Howard, in his Encyclopedia of Impressionism (London: Carleton Press, 1997, p. 6) comes closest to what I present here: “The works reproduced in this book are among the best known and best loved paintings … in the world and our appetite for them is boundless… These familiar images have kept their power to enthrall … They are kept before our eyes by a flood of advertising campaigns.” 4 The basis of this assertion comes first from the number of images in museums at the time of a given artist’s catalogue raisonnés, given in those volumes. This number is then modified by amount of time that has passed since their publication. The modification is based on the fact that both Cézanne and Manet have had two catalogues raisonné. With two, one can look at the accession rates into museums between those publications. These relative rates are essentially the same for these two painters and, although different painters are differentially popular, the rates can be applied to the thirteen Impressionists considered here—Bazille, Caillebotte, Cassatt, Cézanne, Degas, Gonzalès, Guillaumin, Manet, Monet, Morisot, Pissarro, Renoir, and Sisley—and extrapolated to 2005. 5 Robert Herbert, in his essay “Impressionism, originality, and laissez-faire” in his From Millet to Léger: Essays in Social Art History (New Haven, CT: Yale University Press, pp. 91-97) expands on the many reasons for Impressionism embedding itself in popular culture. 6 The Jeu de Paume was technically a part of the Louvre and not a separate museum, although 400 meters down the Rue du Rivoli. Part of the rationale for this was that the will of Isaac de Camondo, whom I will discuss later, insisted that his paintings be hung in the Louvre. See, for example, Germain Bazin, Impressionist Paintings in the Louvre, S. Cunliffe-Owen, Trans. (London: Thames & Hudson, 1958). 7 For the volume of visitors to the Jeu de Paume and the Orsay, see Andrea Kupfer Schneider, Creating the Musée d’Orsay: The Politics of Culture in France (University Park, PA: Pennsylvania State University Press, 1998) pp. 12 & 106. 8 On Impressionist art sales, see The Art Newspaper (February 2000, XI, No. 100, p. 61), which reported that six of the twelve most expensive paintings sold at auction in 1999 were Impressionist works—three Cézannes, two Monets, and a Degas. The Art Newspaper (September 2001, XI, No. 106, p. 70, Art Market) also commented on the skyrocketing sales prices of Impressionist art over the decade of the 1990s. In addition, among the four most expensive paintings ever sold (through 2002) one was a Cézanne and another a Renoir. Even more striking is the tally of artists with most works sold at auction for over one million dollars through 2001: First was Picasso with 272, but the next five were Impressionists—Monet (218), Renoir (196), Degas (100), Cézanne (80), and Pissarro (74) (Russell Ash, The Top 10 of Everything, New York: Dorling Kindersley, 2002). On art exhibitions, see The Art Newspaper (February 2001, XII, No. 111, p. 20), which reported that 2000 was the first year since 1994 that there wasn't an Impressionist exhibit in the top ten most frequented exhibitions worldwide. In 1999 there were three in the top ten, and in 1998 there were two. See also, Patricia Mainardi, “Repetition and Novelty: Exhibitions Telling Tales” in Charles W. Haxthausen (Ed.) The Two Art Histories (New Haven, CT: Yale University Press, 2002), pp. 81-86, on the force of Impressionism in museums; and see Gary Tinterow, “Blockbuster, Art History, and the Public: The Case of Origins of Impressionism” in the same volume (pp. 143-153). 9 Edward D. Hirsch, Cultural Literacy: What Every American Needs to Know (Boston: Houghton-Mifflin, 1987). In his appendix, which he compiled with collaborators Joseph Kett and James Trefil, he included Cassatt, Cézanne, Degas, Manet, and Renoir, plus Gauguin (who also exhibited at four of the Impressionist exhibitions). Interestingly, he did not include Monet. 11 10 The Philadelphia Museum of Art and the National Gallery London have just under one hundred Impressionist paintings and pastels. The Musée Marmottan in Paris and the Barnes Foundation in Merion, Pennsylvania, have more than this, but these collections are not diversified. The Marmottan has almost 100 Monets but few paintings by other Impressionists, and the Barnes more than 60 Cézannes and more than 40 Renoirs, but again few by others. 11 Robert Zajonc, “Attitudinal Effects of Mere Exposure,” Journal of Personality and Social Psychology, 9 (1968), 1-27. The very many studies on the phenomenon in its first two decades were then summarized by Robert Bornstein, “Exposure and Affect: Overview and Meta-analysis of Research, 1968-1987,” Psychological Bulletin, 106 (1989) 265-289. For recent results with music, see Karl K. Szpunar, E. Glenn Schellenberg, and Patricia Plinar, “Liking and Memory for Musical Stimuli as a Function of Exposure,” Journal of Experimental Psychology: Learning, Memory, and Cognition, 30 (2004), 370-381. 12 Two additional points should be made. First, for mere exposure to affect positively the evaluation of a given item it should be initially perceived as at least neutral. Unpleasant items typically become more unpleasant with exposure. Second, mere exposure works best either when one doesn’t pay much attention to each presentation, or when the intervals between presentations are relatively long. Otherwise, overexposure and boredom can result. 13 These studies are reported in detail in James E. Cutting, “Gustave Caillebotte, French Impressionism, and Mere Exposure,” Psychonomic Bulletin & Review, 10 (2003), 319-343. 14 I recognize that strictly speaking Manet was not an Impressionist, and that arguments against the inclusion of Cézanne can be made as well. Nonetheless, both are very part of the Impressionist story. 15 Cornell University is a major research institution and its libraries hold more than six million volumes. It’s Fine Arts collection is extremely extensive, and is likely to hold all of the major works within any artistic subdiscipline. 16 For example, images that were up to 40% more frequent than their mates were preferred 57% of the time, those about twice as frequent were preferred 60% of the time, and about four times as frequent were preferred 63% of the time. 17 See, for example, Richard E. Nisbett and Lee Ross, Human Inference: Strategies and Shortcomings of Social Judgments (Englewood Cliffs, NJ: Prentice-Hall, 1980). Note also that children did not show the adult preference pattern. To be sure, they enjoy art and have strong preferences, but those do not match rates of reproduction. Children have simply not lived long enough in Western culture to benefit from the effects of mere exposure to art. 18 For statements outside art history concerning the perception of quality in art, see for example Immanuel Kant, Critique of Aesthetic Judgment. J. Meredith, Trans. (Oxford: Oxford University Press, 1952, original work published in 1794); Robert Pirsig, Zen and the Art of Motorcycle Maintenance (New York: Morrow, 1974); and Richard Feynman, “Surely You’re Joking, Mr. Feynman”: Adventures of a Curious Character (New York: Norton, 1985); or for statements within art history, see for example Jakob Rosenberg, On Quality in Art: Criteria of Excellence, Past and Present (Princeton, NJ: Princeton University Press, 1967); and Susan Woodford, Looking at Pictures (Cambridge, UK: Cambridge University Press, 1983). For counters to the notion of artistic quality see Mika Bal and Norman Bryson, “Semiotics and Art History,” Art Bulletin, 73 (1991), 174-208; Mark A. Cheetham, Kant, Art, and Art History (Cambridge, UK: Cambridge University Press, 2002); and Keith Moxey, The Practice of Theory: Poststructuralism, Cultural Politics, and Art History (Ithaca, NY: Cornell University Press 1994). 19 Viewers preferred the less frequently published images 52% of the time. In previous studies the same images in the same pairs were preferred 40% of the time. The difference was statistically highly reliable. 20 Walter Benjamin, “The Work of Art in the Age of Mechanical Reproduction,” in his Illuminations, H. Zohn, Trans. (New York: Schocken Books, 1968), pp. 217-251. 21 Michael Camille, “Rethinking the Canon,” The Art Bulletin, 78 (1996), 198-201. 22 Mainardi, pp. 83-84. 23 A somewhat similar analysis was done by Daniel W. Galenson, Painting Outside the Lines: Patterns of Creativity in Modern Art (Cambridge, MA: Harvard University Press, 2001). Galenson, in part of his study, analyzes all of the images that appear in general art history textbooks dealing with his topic. However, most of his work deals with market forces on the sales of images by 125 modern painters and inferences that might be made about creativity. Robert Jenkins, in his “Measuring canons: Reflections on innovations and the nineteenth-century canon of European Art” (this volume), follows in this tradition. 24 The exact number of Impressionist images is not known, in part, because Renoir’s catalogue raisonné is unfinished. Nonetheless, these totals are based on tallies from the other catalogues and an estimate for Renoir based on his productivity in a comparable period to Cézanne, Degas, and Monet. The basis for claiming that these seven painters were the ‘major’ Impressionists comes from the tally of thirty books on Impressionism published throughout the twentieth century. Works by these seven artists were included in at least twenty-nine of them. Only 12 Cézanne and Sisley were missing from one each. In comparison works by Bazille appeared in nineteen, Caillebotte fourteen, Cassatt twenty-one, and Morisot twenty-five. 25 A cutoff of 50 is completely arbitrary, but the same general patterns as those reported here recur at 25, 100, or even 350 images. 26 For histories, see Marie Bérhaut, Gustave Caillebotte: Catalogue Raisonné des Peintures et Pastels (Paris: La Bibliothèque des Arts, 1994); Michael Marrinan, “Caillebotte as Professional Painter: From Studio to the Public Eye,” in Norma Broude, Ed., Gustave Caillebotte and the Fashioning of Identity in Impressionist Paris (New Brunswick, NJ: Rutgers University Press, 2002), pp. 21-65; Philip Nord, Impressionists and Politics (London: Routledge, 2000); and Kirk Varnedoe, Gustave Caillebotte (New Haven, CT: Yale University Press, 1987). 27 Except where otherwise indicated, all paintings are currently in the Musée d’Orsay. 28 On Camondo, see G. Migeon, P. Jamot, R. Vitry, & C. Dreyfus, La Collection Isaac de Camondo au Musée du Louvre (Paris: Gazette des Beaux-Arts, 1914). Camondo was also a friend of Georges Clemenceau, and his bequest made Degas, Monet, and Renoir the only living artists ever to have works in the Louvre. 29 The full title of the particular Rouen Cathedral image is La cathédrale de Rouen, le portail et la tour SaintRomain, plein soleil, harmonie bleue et or. 30 On Lane, see http://www.hughlane.ie/about/hugh.shtml. Since 1979 Manet’s Portrait d’Eva Gonzalès (1870), Morisot’s Eté (1979), and Renoir’s Parapluies (1881-1886) have been in Dublin, but Degas’ Bain de mer. Petite fille peignée par sa bonne (1868-1869) and Manet’s La musique aux Tuileries (1862) have stayed in London. 31 For data on the Courtauld fund see the website of the National Gallery of Art, London: http://www.nationalgallery.uk.org . On the Courtauld institute, see John Murdock, The Courtauld Gallery at Somerset House (London: Thames and Hudson, 1998). Also, the collections of the National Gallery and the Tate were shuffled in 1996, with all Impressionist images going the National Gallery, the Tate afterwards being reserved for British works. Courtauld images just missing this list include Degas’ Mlle La La au cirque Fernando (1879, National Gallery London) and Manet’s La serveuse de Bocks (1878-1879, National Gallery London). 32 On Louisine Havemeyer, see Alice C. Frelinghuysen, Gary Tinterow, S. A. Stein, G. Wold, and J. Meech, Splendid Legacy: The Havemeyer Collection (New York: The Metropolitan Museum of Art, 1993). 33 The most reproduced of the Palmer bequest is Monet’s Au bords de l’eau, Bennecourt (1868, Art Institute of Chicago); that of Personnaz is the most popular of Monet’s several Pont d’Argenteuil (1874, Musée d’Orsay); that of Dale is Morisot’s Dans la salle à manger (1886, National Gallery, Washington); that of Mellon is Manet’s La prune (1877, National Gallery, Washington); and that of Mellon Bruce is Renoir’s Le pont neuf (1872, National Gallery, Washington). But again, none of these is in the most frequent fifty. Two earlier bequests to the French State with core canon images were those of Marc Bazille in 1904, Frédéric’s nephew (Atelier de l’artiste, rue Condamine,1869, and Réunion de famille, 1867), and Etienne Moreau-Nélaton in 1906 (for Manet’s Le déjeuner sur l’herbe, 1863). 34 This analysis is based on the study of the catalogues raisonnés of the Impressionist painters. In addition, see Harrison White and Cynthia A. White, Canvases and Careers: Institutional Change in the French Painting World (Chicago: University of Chicago Press, 1993, 2nd ed) for an account of the role of dealers in the changing art world of the mid-late nineteenth century. 35 Again, the calculations are based on going through the catalogues raisonné of each artist, or museum catalogues, checking for provenance. 36 See Pierre Assouline, Grâces lui soient rendues, Durand-Ruel le marchand des Impressionistes (Paris: Plon, 2002). 37 The list was compiled largely based on an appendix in John Rewald, The History of Impressionism (New York: Museum of Modern Art, 1946). 38 Charles Borgmeyer, The Master Impressionists (Chicago: Arts Press, 1913). 39 That book was Diane Kelder’s The Great Book of French Impressionism (New York: Abbeville, 1981). 40 The story of Louis Leroy’s naming of Impressionism stems from a widely cited account that Monet gave in an interview in 1900. For an important revisiting of this account see Jane Mayo Roos, Early Impressionism and the French State 1866-1874 (New York: Cambridge University Press, 1996). 41 See Wendy Beckett and Patricia Wright, The Story of Painting (London: Dorling-Kindersley, 1994), and Wendy Beckett and Patricia Wright, Sister Wendy’s 1000 Masterpieces (London: Dorling-Kindersley, 1999). 42 Mainardi, p. 81. 43 Benjamin, p. 234. 
Purchase answer to see full attachment