How artworks modify our perception of the world

Abstract

Many artists, art critics, and poets suggest that an aesthetic appreciation of artworks may modify our perception of the world, including quotidian things and scenes. I call this Art-to-World, AtW. Focusing on visual artworks, in this paper I articulate an empirically-informed account of AtW that is based on content-related views of aesthetic experience, and on Goodman’s and Elgin’s concept of exemplification. An aesthetic encounter with artworks demands paying attention to its aesthetic, expressive, or design properties that realize its purpose. Attention to these properties make percipients better able to spot them in other entities and scenes as well. The upshot is that an aesthetic commerce with artworks enlarges the scope of what we are able to see and has therefore momentous epistemic consequences.

[the artist] disturbs, upsets, enlightens, and he opens ways for a better understanding. Whereas those who are not artists are trying to close the book, he opens it, shows there are still more pages possible.

Robert Henri, The Art Spirit (1923/2007, p. 11)

Introduction

It is sometimes claimed that one way to come to contemplate aesthetically quotidian things and scenes is through the arts (e.g. Goodman, 1984, p. 85; Nanay, 2018; Schwartz, 1985). For instance, Leddy (2013, p. 96, p. 114) suggests that appreciating arts like painting and photography may foster our capacity to experience even junkyards or roadside clutter aesthetically (Leddy, 2008). Indeed, many artists and art historians have drawn attention to the fact that, sometimes at least, contemplating artworks aesthetically bears on our perception of the world. Proust (1913/1981, p. 311) recounts of how Swann fell in love with Odette after noticing her resemblance to Botticelli’s representation of Zipporah in the Sistine chapel. Gombrich suggests that a proper commerce with artworks may disclose to our eyes the beauties hidden in our everyday (1950/2018, p. 59) (cf. also Wilde, 1889/2010). Call this Art-to-World (AtW).¹

I shall investigate AtW focusing mainly on visually-accessible artworks (paintings, etc.). We can formulate AtW as follows:

AtW: In virtue of having looked at an artwork aesthetically, a subject may come to look aesthetically at other things and scenes as well.

It remains to be seen exactly what it means to look at something aesthetically (call this “Aesthetic Looking”), and explain how artworks may affect our perception of the world (call this “Aesthetic Alteration”).²

I will argue that Aesthetic Alteration is accounted for by means of exemplification (Goodman, 1976, pp. 52ff, 1984, pp. 54ff; Elgin, 1996, p. 17; 2002; 2017, pp. 183ff). Artworks embody certain features in a way that makes us better aware of their instantiations in other contexts, including other artworks and our quotidian lives. In this paper I will significantly reinforce and expand this claim by combining it with a content-related account of aesthetic experience (“Aesthetic Looking”), i.e. the claim that an experience counts as aesthetic only if it involves attending with understanding to aesthetic, expressive, or design properties (Carroll, 2000, p. 207; Saito, 2007, p. 9). Furthermore, I acknowledge one important lesson taught by proponents of everyday aesthetics, namely that the palette of aesthetic properties is far broader than what has been acknowledged by more conservative approaches (Leddy, 1995, 2013; Light & Smith, 2005; Saito, 2007, 2017).³ This combination of views will allow me to maintain that attending to artworks aesthetically magnifies not only our capacity to spot elsewhere the features exemplified, but also those properties in the artworks — those aesthetic, expressive, and design properties, broadly construed — that ground the exemplification by scaffolding the percipient’s perceptual flow (Sterelny, 2010).

The claim that artworks thus enable us to notice features we did not notice before in our surroundings has several important implications. Firstly, it is of epistemic significance, as it leads to enlarge a subject’s perceptual uptake of the world. Secondly, philosophers working on everyday aesthetics urge that aesthetic properties have a considerable social, environmental, and moral power, both positive and negative (Mandoki, 2007), that has been hitherto underappreciated in analytic philosophy (cf. Berleant, 2010; Irvin, 2008; Saito, 2017, pp. 196ff). Coming to notice aesthetic, expressive, and design properties in our world may represent one way to make «our lives […] more satisfying to us, even more profound» (Irvin, 2008, p. 41).

A challenge to my view, however, is represented by an alternative, recent account of AtW put forth by Nanay (2018). Nanay defends the following views regarding Aesthetic Looking and Aesthetic Alteration, call them, respectively, N1 and N2:

(N1) In contrast to content-related views of aesthetic experience, Aesthetic Looking is due to a distinctive pattern of attention called aesthetic attention (Nanay, 2018, p. 77).

(N2) Aesthetic Alteration is due to the lingering of aesthetic attention (Nanay, 2018, p. 79).

Clearly, N1 and N2 stand in sharp contrast with the view outlined above. A dialectic engagement with Nanay’s account offers me the opportunity to bolster my own view and show its superiority over Nanay’s account on both conceptual and empirical grounds.

I zoom in on Nanay’s challenge and defend my content-related account of Aesthetic Looking in the first part of the paper (§§2–3); I turn to Aesthetic Alteration, in the second part (§4).

Nanay on attention and aesthetic experience

The content view

Nanay (2018) uses AtW as a foil to assess theories of aesthetic experience.

We can cluster theories of aesthetic experience in two groups.⁴ The two groups disagree about the signature of the aesthetic, i.e. the necessary condition that makes an experience aesthetic. The first group is that of the content-related views. These theories agree that aesthetic experience has its signature in a specific type of experiential content, though they diverge about what further (sufficient) conditions must be met (cf. Carroll, 2012, p. 175). The second group is that of structural views. According to this family of views, the signature of the aesthetic is not the content, but a peculiar structure of experience (e.g. Dewey, 1934/2005).

Nanay’s strategy is a disjunctive argument, he puts pressure on content-related views and then shows how his structural approach does not suffer the same shortcomings. Since his criticism of content-related views provides the basis for his own account of Aesthetic Looking (N1), we will first have to briefly elaborate on content-related views in order to assay Nanay’s arguments. I zoom in on Carroll’s account for two reasons: first, because it is explicitly targeted by Nanay (2018, p. 76); and second, because it is a minimal or “narrow” (Goldman, 2013) approach that virtually every proponent of content-related views can accept, although some add further conditions for having an aesthetic experience, such as having evaluative responses (Beardsley, 1969; Levinson, 2016).

Carroll maintains that, in so far as every experience has content, aesthetic experiences exhibit a particular type of content:

An aesthetic experience is one that involves design appreciation and/or the detection of aesthetic and expressive properties, and/or attention to the ways in which the formal, aesthetic and expressive properties of the artwork are contrived. (Carroll, 2000, p. 207; cf. also 2001, p. 60).

…an aesthetic experience of a work of art is a matter of attending with understanding to the formal and/or the aesthetic and/or the expressive properties of a work of art. (Carroll, 2012, p. 173).

(cf. also 1999, pp. 156ff; 2000, 2001, 2010, pp. 77ff.; 2012, 2015). Carroll specifies neither the nature of content nor of experience: it can be flexibly understood as either perceptual (seeing, hearing, etc.) or cognitive (remembering, thinking, etc.) content. The content of an aesthetic experience must disjunctively feature aesthetic, expressive, or design (formal) properties (AED-properties for short). Expressive properties refer to emotive expressions, as when we characterize Chopin’s funeral march as “sad” (Argenton, 2008; Noordhof, 2008). Aesthetic properties, like beautiful, graceful, tragic, etc. together with expressive properties, mark the «intensity» of an artwork (Carroll’s term), whereas formal or design properties include properties such as unity, complexity, balance, composition, etc. (Carroll, 2015; Beardsley, 1981); such properties are the «ensemble of choices elected to realize the point or the purpose of the artwork» (Carroll, 2012, p. 173).⁵

Carroll mainly focuses on artworks — though he maintains that his account applies to aesthetic experience of nature as well⁶— and artworks are fundamentally realized or embodied by means of these properties. He maintains (2015) that artworks are embodied purposes, and AED-properties are the ways in which purposes are embodied (§5.1) (cf. also Goodman, 1976, p. 256; Hegel, 1835–1838/2017). To attend to such properties means to attend to how «the work works» (Carroll, 2001, p. 59).

On Carroll’s account, content is not enough to enjoy an aesthetic experience, for a subject may stand in front of an artwork replete with such properties and yet fail to notice them. In order to have an aesthetic experience, Carroll requires the subject to attend with understanding (informed attention) to AED-properties (1999, p. 203; 2010, p. 101; 2015, p. 172). The requirement that attention must be informed mandates that content must be negotiated in terms of «certain strategies and techniques of reception» (2010, p. 104) that are specific to the category the artwork belongs to.

Nanay’s criticism of the content view

Nanay formulates three challenges against content-related views.

The first challenge is framed as a conditional: if AED-properties are construed too narrowly, then it might be difficult to explain the «entire phenomenon of the aesthetic experience of everyday» (2018, p. 78). Call this “Conservatism:”

CONSERVATISM: If we construe AED-properties too narrowly, then it would be difficult to explain the aesthetic experience of the world.

How should we interpret this challenge? One way of doing so, for example, is to espouse a view on which the only aesthetically relevant property is the beautiful (Danto 2007). With this conception in place, it would be difficult to explain AtW on two grounds: first, because “beauty” does not seem an apposite predicate for much of the contemporary artistic production; and second, because “beauty” seems too strong a predicate for many aesthetic experiences of the world (§3.1).

The second challenge is “Different Properties:”

DIFFERENT PROPERTIES: Suppose that at t₁ a subject aesthetically experiences an artwork X. At t₂, she aesthetically experiences the everyday Y in virtue of having looked at X aesthetically (AtW), but X and Y have different properties (Nanay, 2018, pp. 78-79).

The challenge for content-related views would go as follows: If X and Y have different properties it is not clear how one can enjoy an aesthetic experience of artworks and, subsequently, an aesthetic experience of the everyday, given that the two experiences will not share the same content. Of course, it would be trivial to claim that X and Y possess different properties qua numerically distinct entities; instead, X and Y should differ in their relevant properties, properties that type an experience as an aesthetic one, i.e. AED-properties.

Finally, “No-Property-Change:”

NO-PROPERTY-CHANGE: Suppose that at t₁ a subject experiences an everyday object Y. At t₂, she experiences an artwork X aesthetically, and at t₃, she experiences again the very same Y aesthetically (AtW), and none of Y’s relevant properties have changed between t₁ and t₃ (Nanay, 2018, p. 79).

This scenario poses two different challenges for content-related views. First, it is not clear on such views why the subject did not have any experiential uptake of Y’s AED-properties at t₁, but only at t₃. Second, and relatedly, it is not clear what role does X play in eliciting the subject’s aesthetic experience of Y at t₃.

Aesthetic attention

Taking stock of the three challenges, Nanay claims that the signature of aesthetic experience is not content, but a peculiar experiential structure:

(N1) Aesthetic looking is due to aesthetic attention, i.e. the subject’s attention is object-focused and property-distributed (Nanay, 2018, p. 77).

Notice that N1 is target-neutral, we may thus look at an artwork, an everyday item, a face, or scenario through the lens of aesthetic attention. On behalf of Nanay, it is fair to say that he thinks only certain kinds of aesthetic experiences are captured by what he calls aesthetic attention (2015a, b) (§3.1).

The rationale of this view lies in specific and empirically substantiated claims about the nature of attention. It is a well-known fact in cognitive science that the limited amount of time and cognitive resources (e.g. Chun et al., 2011; Franconeri et al., 2013) force the subject’s cognitive system to organize and prioritize the information gathered from the environment. The allocation of cognitive resources is a peculiar task of attention, which can bring into sharper focus only part of the visual scene, feature, or object (focused, or focal attention), with a highly determinate content, or it can be less determinate but more broadly distributed to different items (distributed or diffuse attention) (Nanay, 2010, p. 22; Eriksen & Hoffman, 1972). Attention can also be object-directed — as when we allocate our resources to a specific visual object or scene — or property-directed — as when we allocate our resources to a specific property of the targets (e.g. Dretske, 2010; Nanay, 2010).⁷ What Nanay calls “aesthetic attention” is an attentional configuration that is at once object-focused and property-distributed. This is supposed to shed light on many instances of aesthetic experience (cf. Nanay, 2015a, pp. 27–28).

An important step in Nanay’s dialectic is that the view would be empirically supported by studies on oculomotor correlates of art perception, and in particular by Vogt & Magnussen (2007) (§3.1). Vogt & Magnussen have examined and compared the oculomotor activities of experts and laymen in front of 16 pictures of increasing abstractness. It was found that experts seem to fixate less on familiar objects — such as faces, people, etc. — and tend to stay fixated less on the details than laymen. Furthermore, these studies have shown that experts’ looks covered wider areas of the presented pictures. Nanay interprets these findings as suggesting that art experts tend to deploy an attention pattern that is object-focused (the artwork), and property-distributed (scanning wider areas of the depicted surface).

Defending aesthetic content

Problems with the aesthetic attention account

Before revisiting Nanay’s three-challenges against content-related views, we must first highlight one crucial issue with N1.

Structural views are usually thought to be vulnerable to the following objection: they provide criteria for aesthetic experience that are at once too restrictive and too liberal (e.g. Carroll, 2001, pp. 49ff; Irvin, 2008; Saito, 2017). Too restrictive, for we might easily conceive a case in which the subject’s attention is, for instance, both object and property-focused. Rebecca is aesthetically mesmerized by the suffering expression of Christ in Grünewald’s Crucifixion from the Isenheim’s altarpiece, i.e. her attention is object focused (the painting) as well property-focused (Christ’s expression). But N1 seems also too liberal. A robber points a knife at Rosalind’s chest, in her anguish she focuses on the knife, but her attention is property-distributed, she does not attend to any specific property of the weapon, but to many of them keeping track of every movement of the blade. This case complies with N1, but ascription of an aesthetic experience to Rosalind seems just downright implausible. But let us not jump too quickly to conclusions, there are two possible responses on behalf of N1.

A trivial response is that, however counterintuitive it may seem, Rebecca is not really enjoying an aesthetic experience, while Rosalind is. This strategy calls into question the cogency of our intuitions, but even if we acknowledge that intuitions are not always good guides for doing philosophy, opting for this tack seems unjustifiably too revisionary.

But there is a second response. Earlier, I hastened to add that Nanay’s account is not meant to provide a necessary condition for enjoying all sorts of aesthetic experience. The obvious assumption here is that aesthetic experiences are variegated and may have little in common, perhaps, not even aesthetic attention. Nanay emphasizes this when he cautiously suggests that aesthetic attention captures a few «telling instances» of aesthetic experience, like the following one described by Proust:

But even this ugliness of faces, which of course were mostly familiar to him seemed something new now that their features—instead of being to him symbols of practical utility in the identification of this or that person who until then had represented merely so many pleasures to be pursued, boredoms to be avoided, or courtesies to be acknowledged—rested in the autonomy of their lines, measurable by aesthetic co-ordinates alone. (Proust, 1913/1981, p. 355).

It is not entirely clear to me what Nanay’s “telling instances” really are, but let us assume, as he suggests, that they are typed by the aesthetic attention pattern, i.e. object-focused and property-distributed. If Nanay purports to account for AtW in terms of aesthetic attention, it obviously follows that both aesthetic looking at the artwork and aesthetic looking at the everyday must be somehow subsumed under the same aesthetic-attention kind. However, Nanay’s remarks make room for aesthetic experiences that do not fit squarely within his own aesthetic attention account. This would allow him to concede that, perhaps, Rebecca’s experience is an aesthetic one, after all. Yet, if one makes such a move, one is implicitly suggesting that aesthetic attention may not be the aesthetic signature we were looking for. Clearly, it cannot be attention alone, for otherwise pretty much every experience would be an aesthetic one. In other words, we are back to the issue of finding the necessary hallmark, what I called the “signature,” that makes some experiences “aesthetic.”

We can gain some fresh insights by reading again Proust’s quotation. Swann finds the faces new because he is paying attention to their aesthetic features for the first time, namely their ugliness (Nanay, 2018, pp. 77–78). It is not that these properties did not fall under Swann’s visual field before, but he was not paying attention to them. Elaborating on his aesthetic attention account, Nanay (2015a, p. 29) refers to an illuminating example: suppose you experience aesthetically a Giacometti’s sculpture. Your attention is focused on the object (the sculpture), and property-distributed to its properties. However, Nanay also adds that on this scenario the subject is «admiring [the sculpture’s] composition» (ivi, my emphasis). In other words, the subject is attentively focusing on the sculpture’s design or formal properties (§2.1). Now, suppose a burglar attacks you (Nanay’s example, again). You quickly scan the environment — pausing on the sculpture — looking for a defence weapon. Your attention changes configuration from object-focused and property-distributed (aesthetic attention) to object-distributed and property-focused. Now you look for objects with the right sort of property that may help you to defend against the assault. What makes this experience not an aesthetic one? Perhaps, the fact that in this case you are not attending to the target’s AED-properties trying to appreciate how the «work works», you focus on aesthetically irrelevant properties. So, here is my conjecture: the aesthetic signature can be identified in the way the percipient mobilizes her attentional resources so as to attentively look at the target’s AED-properties. This conjecture, finds considerable empirical support.

Vogt & Magnussen’s results (2007) have been reproduced by several other studies. Pihko et al. (2011) have found that experts tend to deploy more global (rather than local) viewing strategies than non-experts (Zangmeister et al. 1995). This is expressed in the finding that «laypersons concentrate on the details of the picture, experts also examine the spatial construction while evaluating the esthetics of the painting» (Pihko et al., 2011, p. 8; Kapoula & Lestocart, 2006). Experts tend to attend to the pictures’ compositions that are usually not attended to by laymen. The scanpaths reveal that artists and art experts «view familiar objects to a lesser extent than the layman, with an increased preference for viewing more or less abstract, structural features instead» (Vogt & Magnussen, 2007, p. 98; my emphasis). Francuz et al. (2018) have also found that in deploying more global strategies, experts tended to detect the pictures’ balance and structure, while laymen are more attracted to semantic objects such as faces and their expressiveness, people, and other narrative elements. This in turn is correlated with better, more accurate evaluations of the pictures, lending support to Nodine’s conjecture that «visual structure, attention and judgment of compositional design are intimately related» (1982, p. 52). As Vogt & Magnussen suggest, differences in oculomotor movements in experts and laymen mark their different viewing strategies, i.e. the fact that in virtue of their acquired capacities, experts mobilize their perceptual resources in a way different from laymen. The empirical studies converge thus in emphasizing the importance of structural features, of composition, in short: formal or design properties.

We can extrapolate the following lesson: while artists and art experts attend to a greater degree to the pictures’ AED-properties, laymen tend to fixate and see more often familiar or ordinary items. As I will suggest (§3.3) the difference between laymen and experts is more one of degree, rather than of kind; but the important point is that the alleged empirical evidence for the aesthetic attention view actually provides support for content-related views of aesthetic experiences.

Having shifted the balance of empirical support, we must now revisit Nanay’s challenges against content-related views of aesthetic experience.

Nanay’s three challenges revisited

The first objection was Conservatism:

CONSERVATISM: If we construe AED-properties too narrowly, then it would be difficult to explain aesthetic experience of the everyday.

This is a reasonable concern, but I think its threat is greatly exaggerated. Recently, proponents of everyday aesthetics have called attention to a whole array of aesthetic properties that populate our quotidian lives, such as being pretty, cluttered, drab, dirty, sparkled, shiny, etc.⁸ that have been largely neglected by more conservative philosophers of art (Leddy, 1995, 2005, 2013; Mandoki, 2007; Saito, 2015). Notice also that a more conservative aesthetics is also at odds with many uncontroversial examples of contemporary artworks. This is what Danto drives at, when he draws attention to a work like Rauschenberg’s Bed, which embodies an aesthetics based on properties such as grunge and mess, rather than beauty (2007, pp. 123–124); but the point might be easily generalized to many other contemporary artworks as well (Saito, 2017).

The second challenge was Different Properties:

DIFFERENT PROPERTIES: Suppose that at t₁, a subject aesthetically experiences an artwork X. At t₂, she aesthetically experiences the everyday Y in virtue of having looked at X aesthetically (AtW), but X and Y have different properties.

Many contemporary artists recruit material from the everyday and turn it into observationally indiscernible artworks from their everyday counterparts (Danto, 1964):

INDISCERNIBILITY: An artwork X appears identical with an everyday item Y, i.e. they are qualitatively indiscernible in their observable properties.

X might be Beuys’ Filzanzug (1970) — a two-piece, coarse grey felt suit —, or Warhol’s Brillo Box (1964), and Y another qualitatively indistinguishable two-piece, coarse grey felt suit, or a Brillo box. Granting that X and Y will differ in their, say, relational properties, and in the fact that only Xs possess the property of “being an artwork,” Indiscernibility shows that at least as far as observable AED-properties are concerned, X and Y can indeed share AED-properties.⁹ Even in less extreme cases, where X and Y significantly differ, and therefore do not instantiate exactly all the same AED-properties, there is no reason to assume that Y can’t instantiate AED-properties. This admits of different degrees. Products of expert designers — think of Sottsass’ Olivetti Valentine typewriter or Chippendale’s chairs —, or the creations of skilful gardeners are obvious AED-properties bearers. Some AED-properties, of course, might obtain by sheer chance in natural or artificial settings, i.e. even when they do not respond to a particular purpose. Ziff vividly makes this point when he said that: «[g]arbage strewn about is apt to be as delicately variegated in hue and value as the subtlest Monet. Discarded beer cans create striking cubist patterns» (quoted from Leddy, 2008). Furthermore, once we acknowledge a broader palette of aesthetic properties, including negative ones like “being smelly” or “being dirty,” (think of Beuys’ Stuhl mit Fett) I see little cause to deny that X and Y might share AED-properties.

The third challenge is No-Property-Change:

NO-PROPERTY-CHANGE: Suppose that at t₁, a subject experiences an everyday object Y. At t₂, she experiences an artwork X aesthetically, and at t₃, she experiences again the very same Y aesthetically (AtW), and none of Y’s relevant properties have changed between t₁ and t₃.

As we have seen, this poses two challenges. I should anticipate that these challenges pertain to Aesthetic Alteration, and that they will be properly examined in §4. For now, let us observe that — in virtue of its second “attention” requirement — the Content View does have the same resources of Nanay’s attention-based account. Accordingly (first challenge), one could say that at t₃ the subject redirects her attention (voluntarily or not) to Y’s AED-properties in virtue of having experienced an artwork aesthetically with similar properties. Thus, on these terms the aesthetic attention view does not offer an explanatory advantage over the Content View. Yet, it is not clear why the subject comes to notice AED-properties at t₃, and (second challenge) what is the role of X in bringing about this change. Before I postpone further thoughts on this to §4, however, I would like to pinpoint once again that in so far as both Carroll’s Content View and Nanay’s view refer to attention they are explanatory on a par about this specific issue.

Looking aesthetically: Content plus attention

So far, I have casted doubt on Nanay’s aesthetic attention view, calling into question (i) its empirical support; and (ii) its plausibility as signature of aesthetic experience. Then, I have defended Carroll’s Content View on the ground that (iii) it squares better with the empirical evidence Nanay’s credit to support his own view; (iv) that it is not vulnerable to the shortcomings that Nanay discusses in his three challenges; and (v) that, as it stands, it already makes attention a necessary ingredient of aesthetic experience, thus having the same explanatory advantages of Nanay’s own view without its shortcomings. I hasten to clarify that my defence of Carroll’s Content View does not shield it against Goldman’s charge of “narrowness,” nor is it my intention to do so here.¹⁰ Perhaps, additional ingredients are required to turn an experience into an aesthetic one.

Given my focus on mainly visually-accessible artworks (and everyday), we have enough to build an account of Aesthetic Looking. The aesthetic literature contains plenty of references to an aesthetic mode of perception (e.g. Levinson, 2016, p. 39; Tomas, 1959). Reflecting on the importance of attention in shaping our perceptual acquaintance with artworks, we can say that, when the subject looks aesthetically at a given target her attention structures «her mental life so that a state of seeing that thing [AED-properties, in our case] is prioritized» (Watzl, 2017, p. 45; cf. also Kalderon, 2018, pp. 163ff). Basing on the foregoing considerations, we can put forth the following:

AESTHETIC LOOKING: A state of looking is an aesthetic one, iff:

• (a) Its content disjunctively exhibits at least some AED-properties; and

• (b) The subject attends with understanding to such properties (informed attention).

The second clause (b) dictates that attention must be deployed in a way informed by the subject’s relevant background knowledge or understanding, i.e. in terms of «certain strategies and techniques of reception» (Carroll, 2010, p. 104) specific to the category the artwork belongs to (§2.1); as shown by the experimental studies (§3.1).¹¹ My account of Aesthetic Looking is easily contrasted with Nanay’s N1, for on his account aesthetic attention only (i.e. object-focused and property-distributed attention) provides the aesthetic signature, whereas the Content View I recommend reinstates the importance of AED-properties as well. Three caveats are in order.

First, my notion of aesthetic looking is target-neutral, but the reader may doubt the application of (b) to the everyday and in general non-art things and scenes. While in the case of artworks informed attention will structure the perceptual process as to grant «priority of processing» (Carrasco et al., 2004) to AED-properties in conformity with the artwork’s purpose, quite often — that is, leaving aside products of design, gardens, etc. — quotidian things and scenes do not have any “purpose,” and may thus simply lack design properties. Think again of Ziff’s case: garbage strewn about may be as delicately variegated as a Monet, but of course this is likely due to sheer chance. In such a case, the subject may simply acknowledge the randomness of such properties, thereby guiding attention accordingly; or she might recruit the strategies and techniques of reception of some specific art-domains, transposing them to the everyday target. One, however, may also come to appreciate the instantiation of AED-properties in relation to the function (if any) of the object (cf. Parsons & Carlson, 2008).¹²

Second, my view might account for the gradualness of such experiences. It seems plausible that two experts may dwell on different aspects (AED-properties) of the same target, for instance because of their differing degrees of understanding of the relevant subject matter; in this case, they will perceptually prioritize different features of the target. It also allows the difference between laymen and experts to be one of degree, rather than kind. Recall the finding that «artistically trained participants view familiar objects to a lesser extent than the laymen, with an increased preference for viewing more or less abstract, structural features» (Vogt & Magnussen, 2007, p. 99; my emphasis). In both cases, in light of contemporary vision science, it is hardly surprising that subjects with different degrees of expertise will tend to look at (and overlook) different features of the target (Schwartz, 1985, p. 712). Nanay (2015a) appropriately cites studies on inattentional blindness as a way of comparison (e.g. Simons and Chabris, 1999). Such studies illustrate that unattended properties or objects of a scene are either non-consciously processed (e.g. Rensink et al., 1997) or do not gain access to systems responsible for cognitive broadcasting (e.g. Block, 2007). We need not take a stance on this controversy here, it suffices to point out that attention to some feature enables cognitive reportability, consequently affecting aesthetic judgment and evaluation (Nodine, 1982).

Finally, one might object that my account only works in conjunction with some version of the rich content view of perceptual experience (Siegel, 2010), i.e. if the reach of perceptual (visual) content includes not only low-level properties such as colors, forms, etc. but also higher level properties as well, such as “being a human”, “being a pine tree”, “being expressionist” etc. (Stokes, 2018). The rich content view would square nicely with my account of Aesthetic Looking, but there are also alternatives. For instance, it seems relatively unproblematic to assume that in many cases the cognitive appraisal of AED-properties will be grounded in distinctive patterns of low-level perceptive properties. Hence, it seems plausible to say that token states of Aesthetic Looking single out the relevant patterns of low-level properties that enable higher-order cognitive categorizations. But notice also that some clearly low-level properties may well play the role of AED-properties, for instance, colours (as when one contemplates a Rothko, or a Yves Klein’s blue) and combination of forms.

Another possible option might be to construe AED-properties in nominalistic terms as the application of predicates. In this sense, one might easily allow for certain entities or scenes to metaphorically possess AED-properties; this obtains when a predicate that belongs to a symbolic realm gets transferred to a different realm (e.g. Goodman, 1976, pp. 68ff), as for instance when we apply the predicate “miserable” (realm: human emotions) to the figure of Picasso’s The Old Guitarist. There is no need to work out the details of these different options, for my purposes, it suffices to say that my account of Aesthetic Looking is flexible enough as to accommodate different options about the reach of perceptual content.

Exemplification and aesthetic alteration

From lingering attention to exemplification

We now have an account of Aesthetic Looking, what we need is to examine Aesthetic Alteration. Nanay’s take on this problem is:

(N2) « […] it is the aesthetic way in which we are attending to the world that lingers» (2018, p. 79).

On Nanay's account of AtW, our aesthetic attention response to atrworks lingers on, even when we afterward come to look at the everyday, we still have the same aesthetic attention pattern. Since I have rejected N1, N2 is undermined, it is no longer clear what lingers on; hence, we need a new account of Aesthetic Alteration.

A natural thing to do would be to see whether the Content View offers us some resources to deal with this problem. Earlier (§3.2), we have seen that the No-Property-Change scenario bears on our issue, as it consists of two challenges: to explain why the subject notices AED-properties of Y at t₃ and not at t₁, and what is the role of X in bringing about this change. Neither the Content View nor Nanay’s aesthetic attention view have the resources to cope with this problem as they merely specify the subjective side of the relation between subject and artwork (or any other object), i.e. they merely specify what occurs in the subject when she perceptually engages with an artwork. With this, I suggest that we should look at how artworks are made or embodied in order to grasp how they elicit Aesthetic Alteration.

What underpins Aesthetic Alteration is the capacity of artworks to refer to AED-properties. This aspect may be accounted for by means of one of Goodman’s (1976, pp. 252ff; also 1978a/1988) symptoms of the aesthetic: reference by displaying or exemplification. According to Goodman (1984, pp. 54ff) reference by exemplification exhibits an “object-to-feature” direction (Elgin, 1996, p. 171; 2017, pp. 183ff; Goodman, 1976, pp. 52ff). Think of a colour chip in a paint shop. The chip is a colour sample that is used to make us aware of its colour, enabling us to «recognize that feature when we encounter it in other contexts» (Elgin, 2017, p. 187). The feature conveyed can be a property, a set of properties, a pattern, etc. In general, exemplars are symbols used in specific contexts for some purpose (Elgin, 2017, p. 253).

Goodman said that exemplification requires «possession plus reference» (1976, p. 53; 1981). This is definitely too meagre, we need to flesh out this notion if we want to construct an account of Aesthetic Alteration. I give the following definition, and then spell it out with reference to artworks:

EXEMPLIFICATION: An item O (the exemplar) is a symbol that exemplifies F-ness in a context c where it plays an intended function f iff:

• (a) O possesses (literally or metaphorically)¹³F, and

• (b) O refers to F-ness, i.e.:

• (b.1) O embodies F in such a way as to highlight it, i.e. as to draw S’s informed attention to O’s being F (a); and,

• (b.2) O makes thereby S epistemically aware of F-ness.

Exemplars are always given in some context and play a function that depends on someone’s intentions (as individuals or collective implicit intentions, like social norms). Pollock’s Number One can be taken as exemplar in an art history class or in an investment seminar. Accordingly, its function will vary. In the investment seminar, it might exemplify volatility, whereas in an art history class it may exemplify the paint’s viscosity (Elgin, 1996, p. 175). Even within the same context, the exemplar’s function can vary, and so the F-ness it makes reference to. A Tarbell may be used to exemplify the feature “being impressionistic,” or more specifically, the distinctive traits of American impressionism, and so on.

Possession (a) is intrinsic and can be characterized in a literal or metaphorical way (Elgin, 2017, p. 261). Possession secures that the exemplar itself is denoted by the predicate or property it exemplifies. A sample of green must itself be green. But possession alone does not suffice for exemplification. A colour chip possesses many properties, but not all of them will be exemplified. The exemplified property is the one the exemplar makes reference to.

How does a symbol refer to F-ness (b)? Textor (2008) suggests that what singles out F as the property referentially conveyed is grounded in the symbol’s being intentionally used for that function. The intended function might be explicit, or implicit as a social convention. This proposal is largely sympathetic with Goodman (1978b, p. 162), but I doubt that it provides a sufficient ground for reference. To see why, let us consider two cases:

C1: John shows to Emily O, a colour chip, as sample of a shade of red, R. Although O does possess R (condition (a)), R only occupies a tiny fraction of O’s surface, which causes Emily to struggle to single out R.

C2: John fetches another colour chip, O’, which is fully covered with R. However, O’ has a curious torus-like shape S. Basing on contextual information (John is a salesman at the local paint shop),¹⁴ Emily grasps John’s intention to display R. Yet, she struggles with O’ as well, as its S averts her attention from R.

The chips in C1 and C2 are not good exemplars, they do not embody the conveyed features in the right manner. This suggests that, in order to refer, something more than mere intention is needed, something about the way the exemplar conveys F-ness. Elgin urges (2017, pp. 191ff) that the exemplar should downplay other features, and highlight F. But it is not clear what “highlight” means.

On this point, Textor (2008) claims that talk about exemplars highlighting or calling attention to something is absurd, for things do not call attention to anything, people do. He casts doubt on the very idea of an exemplar “highlighting” anything by appealing to Goodman’s alleged dismissal of the idea in his reply to Beardsley’s criticism. I say “alleged,” for in that context Goodman does not actually use the term “highlight.” Goodman (1978a, b, p. 172) talks about laying “emphasis” in terms of conspicuousness. In another passage, Goodman says that «[t]o exemplify is to bring out, call attention to, but not necessarily to stress a feature; a significant feature […] may be quite subtle […]» (Goodman & Elgin, 1988, p. 69; my emphasis). It is clear that Goodman drives a wedge between, on the one hand, highlighting and calling attention to something, as vehicles of exemplification and, on the other, emphasizing, making conspicuous, or stressing. The exemplified feature might be subtle and difficult to spot. Sometimes the prominence or conspicuousness accorded to a feature might be a consequence of the deliberate choice to make some other feature salient. Think of Gauguin’s Le Christ jaune. The painting clearly exhibits a conspicuous amount of yellow, but it can hardly be taken — at least in most art-related contexts — as exemplifying yellowness. Rather, the conspicuousness of yellow is functional to realize the purpose of the painting (§2.1), for instance displaying the isolated life and piety of the peasants.

Elgin (2017, p. 196) says that exemplification must do more than merely call attention to the object’s instance of F-ness if it is to serve any role in understanding. Indeed, displaying F by means of O is just a means through which the exemplar is used to restructure the receiver’s epistemic outlook (Goodman, 1976, p. 65). In using Number One as exemplar of volatile investment, the instructor usually does not mean to discuss that particular instance, but to draw the audience’s attention to investment volatility more generally. In other words, the exemplar must provide epistemic access to F (b.2).

Aesthetic alteration

I believe that, pace Textor (2008), it is possible to make sense of things calling attention to some of their features, and that highlighting a feature does not need to be reduced to conspicuousness. It is at this juncture that this digression on exemplification rejoins with my earlier account of artworks as embodied purposes. I have given a first hint at this move when I specified that the color chips in C1 and C2 do not embody the feature in the right way. But we must now show how an exemplar might embody a feature in such a way as to refer to that feature, rather than merely instantiating it.

Looking at an artwork is a complex activity that involves the interplay of several subjective and objective factors (e.g. Leder et al., 2004; Leder & Nadal, 2014). Among the former factors, the subject must be able to exercise her recognitional dispositions and discriminatory (perceptual) capacities in an informed way mandated by both the context and intended function of the exemplar, and the way the exemplar embodies its purpose. The latter condition does justice to the foregoing consideration that a good sample must embody the feature in an appropriate way. Such an appropriate way, as I will now suggest, in the case of artworks consists in providing some scaffolding for the deployment of the subject’s informed attention (e.g. Clark, 2008; Sterelny, 2010) .

Roughly speaking, the scaffolded thesis suggests that «human cognitive capacities both depend on and have been transformed by environmental resources. Often these resources have been preserved, built or modified precisely because they enhance cognitive capacity» (Sterelny, 2010, p. 472) (cf. also Menary, 2014). In our specific case, my claim is that artworks scaffold our perceptual processes. In order to see how this works, we should refer back to Carroll’s description of artworks as embodied purposes. Artworks have purposes, they sometimes convey a complex message, or draw our attention to some subtle feature from our quotidian lives. Take for example Dorothea Lange’s photograph Migrant Mother. Lopes comments this photograph saying that it is meant to convey the message that we ought to «help the poor, who are noble and do not deserve their poverty» (Lopes, 2005, p. 237). The exemplification of this feature, Lopes urges, is only possible thanks to its «aesthetic aspects — from its composition, its lighting, and printing to the fine balance it strikes between the depiction of pride and resourcefulness on the one hand and despair and vulnerability on the other» (ivi, p. 143). Such aspects, its AED-properties, are contrived in a way that is functional to the expression of its purpose. In order to achieve this function, the artworks provide some instructions, or some scaffolding for its reception. That artworks are structured in a way that (partially) guides the observational flow of its spectators to enable its function is well-known.¹⁵

In his pioneering experimental studies on eye movements, Yarbus has shown that «composition is the means whereby the artist to some extent may compel the viewer to perceive what is portrayed in the picture» (1967, p. 193). But artists and art theorists (Kepes, 1944/1995; Klee, 1925, p. 23), art critics, and art psychologists have long been aware of this. The idea is that artists exploit basic visual phenomena that draw attention to particular regions of the targets (Wolfe, 1998). As a way of example, consider the following passage from Kepes:

A color spot generates different experiences of space depending upon whether it is placed in the middle of the picture-plane, to the left or right, or at the top or bottom. […] The introduction of more than one spot increases the sensation of space. The spots move away from or toward each other, receding or advancing, and seem to have weight or a centripetal or centrifugal direction. (Kepes, 1944/1995, p. 24)

The clever use of space and figures in this example elicits an (illusory) impression of movement and space that shape the percipient’s gaze drawing her attention to some features of the painting. Another example comes from studies on the role of obliquity. Consider now Fig. 1.

Fig. 1

Giotto, Il bacio di Giuda (1303–1305) from Cappella degli Scrovegni, Padova

Building on Arnheim’s (1954) work, Argenton (2008, p. 178ff) illustrates how the diagonal lines of the torches and spears that surround the figures of Christ and Judas direct the percipient’s gaze on the depiction of the kiss (cf. also Schapiro, 1983, pp. 37ff). These examples illustrate how artists may assemble AED-properties in a way that contributes to the artwork’s purpose by exerting some control over the beholder’s gaze. The artworks, in other words, provide some scaffolding for the orientation of the beholder’s attention.

The subject needs to mobilize her cognitive capacities to fully appreciate and discern the exemplified feature, and this in turn depends on acknowledging the context and intended function of the exemplar. Within the institutionalized context in which artworks are embedded, a subject being shown a Simone Martini should deploy the cognitive strategies specified by a highly codified body of knowledge about Gothic art (Elgin, 2017, p. 188). This places some constraints on what sort of understanding (§2.1) the subject must mobilize in order to appreciate an artwork.¹⁶

Notice that the subject’s knowledge and expertise as they inform her attention also confer to the subject a degree of freedom from the artworks’ scaffolding. Koide and collaborators (Koide et al., 2015) have shown that while naïve subjects’ eye-patterns are mainly stimulus-driven, experts’ eye-patterns seem stimulus driven only in early fixation. After salient stimuli have grabbed their attention, experts’ fixations seem less stimulus-driven and more guided by top-down factors that enable them to spot compositional properties of the artwork under observation.¹⁷

From this discussion it will be clear that, if we want to find out what an artwork exemplifies, what F it is meant to make us epistemically aware of, we must look at it aesthetically (§3.3). The referential capacity of exemplars lies in the complex dynamic of interaction they contribute to elicit through the way they are structured, providing some scaffolding, and the way we mobilize our informed attention. This vindicates Carroll’s insight (§2.1) that in cases of aesthetic experience (or aesthetic looking) the experienced content must be negotiated in terms of certain strategies and techniques of reception, plus of course the properties of the target observed. Notice also that, since the referential relation is dynamically defined by the interaction, it allows — in agreement with Goodman (1978a, b/1988, pp. 64–65) — for change: an artwork may be devised to exemplify F-ness, but later come to exemplify Q-ness. This is so because artworks are symbols. A stone encountered in our quotidian life is not an artwork and does not exemplify anything, but it may become an artwork if it is invested of a complex symbolic function. Specifying when something counts as an aesthetic symbol (Goodman, 1978a, b/1988, pp. 57ff)¹⁸ is a question that goes beyond the scope of the present investigation; for my purposes, it suffices to acknowledge that there indeed are such symbols.

My claim is that Aesthetic Alteration is explained by the fact that artworks are symbols, and the specific interaction with such symbols demands aesthetic looking. Sometimes, AED-properties may just be the exemplified properties, for example in didactic contexts when someone takes artworks as exemplars for AED-properties. Some artworks may be designed for the purpose of exemplifying a specific aesthetic property often found in our quotidian lives. But in general, if artworks demand aesthetic looking qua symbols, they make us thereby better aware not only of the exemplified property F, but also of the AED-properties that are functional to its conveying F-ness, thereby facilitating our recognition of that feature(s) when we encounter it (or them) in other contexts, such as the everyday. This entails that AtW only enables us to spot the properties we have looked at aesthetically in artworks beforehand.

Conclusion

Time to wrap things up. I have claimed that we look aesthetically (direct our informed attention to an artwork’s AED-properties) because they are symbols that exemplify a certain feature. In doing so, we heighten our sensibility not only toward the exemplified feature but also toward the AED-properties in an artwork that ground the exemplification. This means being better able to recognize instantiations of such properties in our quotidian lives, other artworks, things, and people. We now notice the cubist pattern in the discarded beer cans, or the resemblance of a face to a Botticelli’s figure. These are instances of cultivating our aesthetic sensibility that expand our perceptual commerce with the world and provide a source of «gratification» that may help experience our lives in a «more satisfying» way (Irvin, 2008, p. 41), or better notice its negative aesthetic properties.