Ted Gracyk, in his The Philosophy of Art Chapter 7 "Aesthetics," discusses the question of the nature of aesthetics with the assumption that aesthetic judgments are a matter of determining how aesthetic properties supervene on non-aesthetic properties. He discusses problems with a couple types of supervenience theory, but never questions the very idea of supervenience. To be fair, in the resources section of the chapter, he does mention two challenges to supervenience theory, one by Marcia Eaton and one by Ben Tilghman. I could add my own objections which are in my book The Extraordinary in the Ordinary pp. 144-149. But why is supervenience theory accepted without question in the body of the chapter? The answer to this question can be seen in how the chapter is set up. We begin with a shocking quote (not apparently shocking to Gracyk) from Alfred North Whitehead that "Nature is a dull affair." We learn that what this means is that nature itself is colorless and in general has no secondary qualities. This dualistic view of nature -- that the material world cannot be beautiful or have any other aesthetic property since such qualities are secondary and based on the mind -- is a basic assumption in the chapter. Mind is one thing, body is another. (I do not intend to imply that Gracyk is assuming the metaphysical theory of mind-body dualism in which the mind and body are distinct substances. The form of dualism found here is what I would call "property dualism" or, better, "functional dualism." Although not metaphysically dualist, supervenience theory works in the same way as metaphysical dualism, and so is functionally dualist.) Pragmatism traditionally rejects dualism and, as a pragmatist, I knew we were off to a bad start with this quote. Moreover, there is something almost immoral in the idea that nature is "a dull affair," as though environmentalism is a waste of time since all of the good qualities of nature are actually just in our separated-from-bodies minds (or are just properties of mental things), and as if humans somehow existed in another realm than nature and thus are not responsible for what happens to nature. (I realize that this may not seem fair to Gracyk who is probably an environmentally sensitive guy. The point is though that a functionally dualist view is assumed by supervenience theorists, and the same problems adhere to that as to metaphysical dualism.)
Gracyk thinks it ironic that when we discover new scientific facts about a painting, facts that are hidden from direct observation, that we change our response to the sensory properties of the painting. This would only seem ironic to someone who accepted the idea that nature is a "dull thing." The woman in Raphael's painting "The Madonna of the Pinks" seems much more graceful after we discover that the painting is actually by Raphael and not by a student. Is it a mistake to see the painting differently? Neither Gracyk nor I think so, but for different reasons, as we shall see.
Gracyk lists six types of judgment concerning paintings: scientific, historical, sensory appearance, interpretive, aesthetic (woman is graceful), and economic. He then launches into the supervenience view. People define this differently but I am sticking with Gracyk here. He understands supervenience as present if "the existence of a property of one type depends on the presence of some property, or arrangement of properties, of another type." (126) In aesthetics, aesthetic properties are said to depend on the "presence of ordinary perceptible properties, such as colors, sounds, and textures." Moreover, you cannot change the aesthetic properties without changing the non-aesthetic properties. Later, he considers adding other sorts of non-aesthetic properties than sensuous properties to the supervenience base. This of course would be needed to explain why it would be justified to see the woman in the Raphael painting as graceful when the sensory properties have not changed. Gracyk observes that a painting will have multiple supervenience relationships, for example the sensory properties of the paint will supervene upon chemical properties as well. So, in general, a family of properties depends on one or many others. He observes that this metaphysical claim is distinct from the epistemological claim that in order to properly judge something as having a certain aesthetic property one must need to know about the supervenience base. Epistemological supervenience is not always necessary, he argues. For example, it is not required to know the chemical supervenience base for certain perceptual properties of paint to know that the paint has those properties. Further, some interpretive judgments depend on properties of the supervenience base that we know and some do not. We know that the painting is of Mary and Jesus and this is because of knowledge we have of provenance, i.e. of Christian traditions.
So far, so good. But why accept supervenience theory? The only support Gracyk gives for the idea that aesthetic properties supervene on non-aesthetic properties is the note that if Raphael had given Mary bulky shoulders she would have been less beautiful. This is an odd example, however, since "bulky" is an aesthetic property and already contains within it an aesthetic judgment. Gracyk of course would reply that the supervenience base is the "shape of the continuous line that forms Mary's shoulder and neck" and not the bulkiness: "bulkiness" is shorthand for a change of this shape that causes us to experience the aesthetic quality of bulkiness. The other support Gracyk provides is that the painting would be less beautiful if it we added fluorescent colors. This is certainly true. However, is this truth enough to support the claim that aesthetic properties must supervene on a non-aesthetic base? Note that, if the fluorescent paints were added, the painting as a whole would be experienced in a different way: it would perhaps now have the property of "aesthetically damaged." The fluorescent paints would be seen as damaging. Could they be separated in our experience from their anti-aesthetic or aesthetically damaging nature? Note that adding fluorescent paints does not always damage a painting. So it is these paints in this context that gives us the experience of aesthetic damage. What does this tell us about the need for talk about a supervenience base for aesthetic properties? For my argument against the adequacy of such arguments for supervenience theory please see the above-mentioned pages in my book. However, very briefly, the argument makes two separate points, first, that Hume was right that value statements cannot be based on factual statements (the "is-ought" problem), and second, and related to this, that anytime someone makes an aesthetic claim and backs it up, that claim must be based on another aesthetic claim using words in their aesthetic meaning.
Gracyk now elaborates that supervenience theory does not identify aesthetic and non-aesthetic properties but simply says that "there cannot be differences in supervening properties unless there are differences in base properties" and that a particular base property is sufficient to generate a particular aesthetic property, although that aesthetic property (for example "beauty") could be based on other base properties (as in music). So, on his view, no fixed set of base properties is necessary for any particular aesthetic property, and indeed base properties can interact unpredictably.
Once supervenience theory is accepted based on a commitment to functional dualism we need only choose between two kinds of such theory. So Gracyk proceeds to discuss these two kinds. One, proffered by Nick Zangwill, says that the supervenience base for aesthetic properties must be sensory properties. This is the "weak dependence" theory. The second called "aesthetic empiricism," is represented by Gregory Currie. It holds that aesthetic properties are restricted to sensory properties, and that other properties, for instance historical properties, cannot form part of the supervenience base. This view would imply that a perfect copy of a Rembrandt would be as beautiful as the original. Gracyk believes the second view is so popular that it is taken as common sense. But actually it is not popular among philosophers and is easy to shoot down. It seems obvious that non-sensory art-historical properties do contribute to aesthetic properties. Moreover (Gracyk argues), as Kendall Walton showed, even though piano music in general sounds percussive, you can still have a piano piece that sounds delicate because of its comparative delicacy...but this comparison requires going beyond the sensory qualities of the object itself. Similarly imitators of certain work may seem trite or tired in relation to the original.
More interesting is an issue with Zangwill's weak dependence theory: it would seem to imply that scientific laws and mathematical proofs could not have aesthetic properties. There is no sensory property of a mathematical proof that supports our saying it is beautiful. It also implies that conceptual artworks have no aesthetic properties. (I will question these ideas below.)
The case is difficult for science and math but one wonders how a scientist who gains his data from the world by way of the senses could, in appreciating the beauty of a law, detach himself completely from that fact. Laws are manifested in perceptual phenomena and talk of such laws would be meaningless without such manifestation. This goes for mathematics as well. It is arguable that mathematics is just a special form of natural law, and so, if this is true, the argument above for a sensual base for the beauty of natural laws would also apply to mathematical laws. Admittedly, the case for basing mathematical beauty ultimately on sensory experience is difficult to make, but it is worth trying. Mathematicians do not live in a world separate from the world of experience: they too are live creatures interacting with their environments. Everything they do is based ultimately on that.
Gracyk gives the work of Yoko Ono as an example of conceptual art. The question is whether a work such as Ono's "Earth Piece" in which we confront a piece of paper that has the following typed phrase on it "Earth Piece. Listen to the sound of the earth turning. Spring 1963" has aesthetic qualities. Some would argue that it does not. But, the truth is that Ono's work is apprehended in galleries or in artist's books through our senses. All conceptual art is apprehended through our senses in some way or context. The fact that the words are arranged in a certain way on the page is important. The look of the page is important. The typeface is important (as can be seen in the fact that she chose a different typeface when recreating the work later). To claim that conceptual art has no aesthetic properties is absurd. Gracyk would agree with that point. He argues that Ono's work can have the aesthetic properties of being funny or elegant. Still, I would insist that these properties are based on sensuous properties.
Where I would disagree with the supervenience theorist, and even with the weak supervenience theorist, is that these sensory properties are themselves non-aesthetic. On my view, if an experience is aesthetic, all of the properties within the experience are aesthetic. Gracyk says that "Freshness, wittiness, and elegance are aesthetic properties. Therefore there appear to be aesthetic properties that do not include sensory properties in their supervenience base." This seems wrong in two ways. First, freshness etc. do depend on perception and hence on sensation. Second, there is a problem with the whole idea of there being "sensory properties" i.e. of there being properties that are non-aesthetic even in an aesthetic situation in which the thing being experienced is being experienced aesthetically. To use Dewey's language, when a live creature interacts with its environment it does so by way of its senses. But these senses do not, I suggest, provide a base of properties upon which something else is based r epistemologically. At the same time, I will remain an agnostic concerning whether supervenience works solely on a metaphysical level.