Mere Exposure and Aesthetic Realism A Response to "Perceptual Learning, The Mere Exposure Effect and Aesthetic Anti-Realism" by Bence Nanay 1 MERE EXPOSURE AND AESTHETIC REALISM A R ESPONSE TO “PER CEPTUAL LEARNING, TH E M ER E EXPOSURE EFFECT AND AESTH ETIC ANTI-R EALISM ” BY BENCE NANAY JAMES E. CUTTING Department of Psychology, Cornell University Email: james.cutting@cornell.edu. © ISAST Keywords: aesthetic realism, aesthetic antirealism, canons, mere exposure I admire much of Nanay’s approach to understanding aesthetics, and not simply because I am an author to which much of the article is directed. In attacking aesthetic antirealism, the notion that quality in an artwork does not inhere in the artwork itself, he defends aesthetic realism. Within art history this latter view was perhaps last endorsed by Rosenberg (1968), and then swept away by the spread of semiotics. Nonetheless, it is hard to resist the notion that there is a kernel of something valid in aesthetic realism. But I need to set a number of things straight concerning Nanay’s presentation. To be sure, Nanay is correct in that a general force of Impressionism and its Canon (Cutting, 2006), and more particularly its predecessor (Cutting, 2003), was to show that the repeated, diffuse exposure to artworks (a cultural generalization of the laboratory phenomenon of mere exposure) – even if we can’t remember having seen them – increases our evaluation of their artistic quality. Beyond this Nanay concludes several things. Three are: that I am essentially, if not an “all out,” aesthetic antirealist; that the subsequent results by Meskin, Phelan, Moore and Kieran (2013) limit my conclusions; and that mere exposure has been shown to work for artistic tokens but not types. Let me address each of these ideas. Mere Exposure as One among Many Aesthetic Effects Nanay reports my position: “And what maintains the ‘canons’ of our Artworld is not the quality of the artworks, but the fact that we are exposed to those artworks that are part of the canon, making us like them more, which reinforces their place in the canon.” This is essentially correct for my view of canon maintenance, but not for canon formation. Thus, according to Nanay, I am essentially an aesthetic antirealist. This would mean that I think that there is no such thing as aesthetic realism (and the inherent aesthetic quality) in a painting, and that only our experience with the painting itself guides our appreciation of it. But this isn’t quite correct, and at best incomplete. For example, in the concluding chapter of Impressionism and its Canon (2006, pp. 201-202) I wrote: “That is, it is perfectly legitimate to say: I want to understand A. I know that b is a part of the cause of A, but I really don’t understand b at all. Thus, let me see how far Leonardo Just Accepted MS. © ISAST doi: 10.1162/LEON_a_01081 mailto:james.cutting@cornell.edu� 2 I can get with c, d, e, f, and g— ignoring b completely— in my account of A. And this is precisely the structure of theory I present here. I am interested in canon formation and maintenance—A. I know that the quality of a painting—b—is likely to be important in some way, but I haven’t a clue as to how to assess artistic quality so I am going to ignore it altogether. Instead, then, I will focus on historical coincidence, early accession into a museum and concurrent publicity, curatorial culling, sustained publicity, and mere exposure—c, d, e, f, and g, respectively—and see how far I can get. The procedure also allows for the omission of other factors— perhaps h and i —that are also currently unknown.” Notice two things. First, I didn’t deny the role of quality (aesthetic value) in the evaluation of art. Instead, I deliberately and openly skirted the issue. Indeed, Maynard (2007), in an otherwise very favorable review of my book, chastised me for exactly this tactic. I might add, however, that Nanay hasn’t offered us a clue about assessing artistic quality either. Is it a priori? Are our abilities to assess art inborn? Or are we responding to authority – Kenneth Clark, Ernst Gombrich, Giorgio Vasari, or perhaps Clement Greenberg? The point of my research effort was to help ground discussion of canons (collections of “good” art) in some data, and to show that subtle, personal perceptual experiences can play a role – perhaps even for choosing Manilow over Mozart. Nanay is also off base in thinking my view is that mere exposure is the dominant factor in canon formation and maintenance. The above quotation should make clear that it is but one of several mechanisms. The others for paintings are: the historical coincidences of early purchases by collectors, early accession into a museums following bequests by those collectors, concurrent publicity in public media, curatorial culling after appearance in those museums (a negative factor), and sustained publicity by art historians. Mere exposure is a sixth mechanism, culturally important but certainly not dominant. Notice also that the first five of these lend force to the roles of authority in the historical agglomeration of images into a canon. Surely, such authority would be paramount over mere exposure for canon formation (perhaps as opposed to canon maintenance). But, of course, appeals to authority don’t do much for aesthetic realism either (the Emperor’s New Clothes). Mere Exposure as a Positive, Uniform Process? In showing potential limits to my results, Nanay writes that “one could argue that the mere exposure effect only works for good art.” This is not the place for a detailed critique of Meskin et al. (2013), whose results he cites to support this idea. The Meskin et al. paper is about mere exposure and its possible effect on the aesthetic denigration of “bad” art. Although that paper mirrored much of my methodology in two experiments, and found results roughly similar to mine for a “good” artist (John Everett Millais), it has other methodological and logical problems. In particular, as much as one might like to, one cannot generalize from one “bad” artist (Thomas Kinkade) to all of “bad” art. One needs, on statistical grounds alone, at least two “bad” artists, parallel results for those two, as well as at least two “good” artists, whose works yield results similar to each other but are different than those for the works of the two “bad” artists. And all of these artists should be from the same period in part so that style is controlled – more on this later. Nonetheless, I am actually happy to endorse the general idea of Meskin et al.: Mere exposure might work differently for “good” art than Leonardo Just Accepted MS. © ISAST doi: 10.1162/LEON_a_01081 3 for “bad” art. Overexposure is a well-known laboratory (Bornstein, 1989) and cultural phenomenon. But again from Impressionism and its Canon (pp. 208-209): “All other things being equal, we will prefer paintings reproduced more often. To be sure, it may be relatively rare that all such things are equal, but the bias [of positive effects from mere exposure] is omnipresent. I claim this bias has deep and lasting effects – indeed, sufficiently lasting to help maintain a canon for a century, and more.” (Quoted with clarification and new emphases not in the original passage.) In my methods (Cutting, 2003, 2006) I deliberately chose Impressionism as a venue because of its deep penetration into popular culture (lots of people will have seen lots of Impressionist paintings), I chose eight artists associated with Impressionism, and I chose the method of having viewers compare pairs of paintings by the same artist (Meskin et al. did not), painted at about the same time, with roughly the same content, painted the same general style, and with the same general color palettes. In other words, I was trying to make as many things as equal as I could (ceteris paribus). I find it charming that I am chastened for this, given that Google at this moment returns nearly 600,000 hits for the joint search of “all things being equal” and “philosophy.” Must we embrace ceteris paribus only for thought experiments? Types, Tokens and Mere Exposure Nanay makes a distinction between mere-exposure-effect tokens (MEE-tokens) and MEE-types, suggesting that most exposure experiments are of the MEE-type variety. However, my guess is that almost all mere exposure experiments, per se, are about MEE-tokens, at least as I would define them. Yes, we are exposed to individual people who look different every day, and we like them (mostly), but most experiments test mere exposure to an exactly repeating stimulus. Wollheim (1987) and Wölfflin aside, I think my art exposure experiments are about tokens, particularly since I don’t have to assume that viewers have seen the actual paintings at all. Of course, we might get into a discussion of differences in reproductions – early color reproductions can be quite bad and have often faded with time – but I think this is another issue. To circumvent any idea that tokens might masquerade as a type, Nanay creates a distinction between MEE-superdeterminate vs. MEE-determinable exposures, where this distinction, I think, is being rallied to shore up the MEE-token vs. MEE-type distinction. If so, fine. However, Monahan, Murphy and Zajonc (2000) showed that certain MEE-superdeterminate and a MEE-determinable effects might be found within the same experiment. That is, those authors found larger effects for liking a specific Chinese character or polygon through mere exposure; and a smaller, general effect for other, previously unseen Chinese characters or polygons as compared against neutral, previously unseen tokens of previously unseen types. Next, Nanay argues that my experiments are MEE-superdeterminate, not MEE-determinable, and that only the latter could be used to support an antirealist position. Fine. This type of experiment might, for example, present many landscapes by Sisley and test for an increase in the appreciation of other Sisley landscapes, perhaps as opposed to those of Pissarro. This is a good idea. Leonardo Just Accepted MS. © ISAST doi: 10.1162/LEON_a_01081 4 But since – by the arguments I presented above – I don’t think that I am an antirealist, this shouldn’t have any force against my claims. I claim, ceteris paribus, that mere exposure can play a role in canon maintenance through the building up of our (the public’s) professed liking of particular pictures. The intended force of my work was to show that aesthetic realism isn’t enough; not that aesthetic antirealism is all. In Conclusion I confess that I don’t think of myself as either an aesthetic antirealist or an aesthetic realist. I simply don’t know where – beyond the avenues that I have explored – the perceived quality of an artwork can generally be said to come from. I thus find the only defensible position to be a hybrid view that endorses perceptual learning from mere exposure combined with cultural education on the one hand while maintaining an agnostic eye towards aesthetic realism on the other. This may be messy, but perhaps a psychologist is used to tolerating more mess than a philosopher. Personally, when reflecting on any artwork that has deeply affected me, I find that my perception of the virtuosity of the artist is very important. This reflection is general to all the arts that I enjoy – graphic and sculptural arts, dance, music, literature and movies – and it goes for sports, too. Nonetheless, I recognize that virtuosity without a “something else” can be pretty empty. For me, some of the “something else” has to do vaguely with form and my emotional response – the work has to speak to me. But I can’t volunteer much further insight beyond that. If anyone could convince me about how the inherent quality of an artwork might be assessed, it would be fun to try to test the idea. References Bornstein, Robert F. (1989). “Exposure and affect: Overview and meta-analysis of research, 1968- 1987,” Psychological Bulletin, 106, 265-289. Cutting, James E. (2003). “Gustave Caillebotte, French Impressionism, and mere exposure,” Psychonomic Bulletin and Review, 10, 319-343. Cutting, James E. (2006). Impressionism and its canon. Washington, DC: University Press of America. Maynard, Patrick (2007). Review of Impressionism and its canon. Journal of Aesthetics and Art Criticism, 65, 246-248. Meskin, Aaron, Mark Phelan, Margaret Moore & Matthew Kieran (2013). “Mere exposure to bad art,” British Journal of Aesthetics, 53, 139-164. Monahan, Jennifer L., Sheila T. Murphy, & Robert B. Zajonc (2000). “Subliminal mere exposure: Specific, general, and diffuse effects,” Psychological Science, 11, 462-466. Nanay, Bence (2014). “Perceptual learning, the mere exposure effect and aesthetic antirealism,” forthcoming in Leonardo. Rosenberg, Jakob (1968). On quality in art. London: Phaidon Press. Wollheim, Richard (1987). Painting as an Art. London: Thames and Hudson. Leonardo Just Accepted MS. © ISAST doi: 10.1162/LEON_a_01081 5 M anuscript received 19 September 2014. James E. Cutting is a Susan Linn Sage Professor of Psychology at Cornell University and author of Perception with an Eye for Motion (MIT Press, 1986), Impressionism and its Canon (University Press of America, 2006) and over 100 scientific articles on various aspects of perception. Leonardo Just Accepted MS. © ISAST doi: 10.1162/LEON_a_01081