Chapter 2 Art and expression (pp. 59-106)
Part I Art as expression (pp. 59-78)
The expression theory of art (pp. 59-66)
Expression theory of art (p. 65)
x is a work of art if and only if x is (1) an intended (2) transmission to an audience (3) of the self-same (type-identical) (4) individualized (5) feeling state (emotion) (6) that the artist experienced (himself/herself) (7) and clarified (8) by means of lines, shapes, colors, sounds, actions and/or words.
The reasons for each individual condition in the expression theory of art as defined on p. 65 (pp. 62-65)
Carroll argues that representationalism rendered art as akin to science. But as science became more and more successful, and as photography was developed, this seemed to move out art. But interestingly, along with the rise of science came Romanticism in art, which stressed the role of art in allowing the artist to express emotions. This helped free music from its role as program music to theater. In addition, once the expressive role was made explicit artists became very good at distorting representations to convey greater emotions.
Tolstoy reflected on these developments and came up with something like the following proto-theory:
x is art if and only if the artist transmits the self-same feeling state that the artist experiences to the audience.
This is very romantic. The artist experiences powerful and refined emotions and manages to share those with the audience. It captures clauses (2), (3), (5), and (6) of the above definition. So we need to see why Carroll adds (1) an intended, (4) individualized, (7) and clarified, and (8) by means of lines, shapes, colors, sounds, actions and/or words.
(1) an intended
Carroll’s story of the sad person (p. 62) who lost his job that makes you sad when you are around him satisfies the original definition, but is not art. So we add (1)
Schlock like greeting cards would satisfy clauses (1), (2), (3), (5), and (6). Carroll (pp. 62-63) suggests that these things are not art because the emotion is too general, not individualized enough.
(7) and clarified
Carroll’s story of the painter who splatters red paint on the wall while cursing, with the intention of hurting his landlord, and his curses all focus on his landlord individually. We see this and are infected with the ire, also getting mad at the landlord. Again, this would count as art by clauses (1), (2), (3), (4), (5), and (6). But if we add the condition that the emotion be clarified, it would not. Art starts with a vague feeling, but by working with the medium in question, the artist is able to experiment until the emotion is clarified and gotten right. This is the difference between mere venting of emotion and art, a controlled activity, not an outburst.
(8) by means of lines, shapes, colors, sounds, actions and/or words.
Finally, art needs to be not just clarified by also externalized.
Carroll discusses dropping the second condition, which would change this transmission theory to a solo expression theory of art.
Objections to the expression theory of art (pp. 66-79)
The criticism of each individual condition in the expression theory of art as defined on p. 65 (pp. 66-79)
(1) an intended (2) transmission to an audience
Carroll discusses these two clauses at length on pp. 67-69, considering artists like Kafka and Emily Dickinson who did not want their art shown to others. Carroll argues two things: (1) the choice of using a public medium shows some intent to transmit to the audience, so at best Kafka and Dickenson had contradictory desires, and (2) the work has to be accessible at least to the artist herself. Carroll does consider slightly changing the clauses from “(1) an intended (2) transmission to an audience” to “(1) at least in principle, an intended, (2) transmission to an audience.” Thus the artist might in face intend the work to not be seen by anyone, but the very nature of art forces the work to be such that it could be seen by other people. Carroll harnesses Wittgeinsteinian private language concerns to make this point.
(3) of the self-same (type-identical)
Carroll calls this the identity condition and discusses it on pp. 69-70. The problem is that artists might be in completely different emotional states than the ones they intend their art to provoke. He gives examples of actors who wouldn’t be able to act if they were really in the emotional state they are displaying. He also considers cynical artists such as sometimes Beethoven, and genre standards such as suspense novels that use formulas to evoke the correct emotion in the reader. It’s far more important that the writer know the formula than to experience any of the emotions.
Carroll discusses this on pp. 73-75. He notes that while individual voice is a goodmaking feature of much art, some even canonical art such as Christian painting and the Indian epics paint emotions in very broad brush strokes. And in any case, bad art is still art. Finally, Carroll wonders if the whole concept of a individualized emotion is incoherent.
(5) feeling state (emotion)
(6) that the artist experienced (himself/herself)
Carroll calls this the “experience condition” and discusses it on pp. 70-71. He’s already (see (3) above) argued against the idea that the emotion expressed by the artwork must have been the same one the artist is experiencing. He adds to this by arguing that that the artist needn’t have any feelings at all. Think of the suspense writer who is just following a formula.
(7) and clarified
This is discussed on pp. 71-73. First, some art such as Beat poetry and Punk Rock gets its whole point by rebelling against art that seems too mannered. For these genres, part of the point is to express raw emotions. Carroll gives a variety of examples of artworks where the artist is not clarifying an emotion: Symbolist Art, Exquisite Corpse writing, Cunningham and Cage’s use of the I Ching and computers to randomly generate choreography and music (Aleatoric works of art).
(8) by means of lines, shapes, colors, sounds, actions and/or words.
Carroll discusses this on pp. 77-78, noting that Conceptual Art should be understood as rebelling against that very condition. He then argues that even if you think Conceptual Art is not art, it would still be premature to try to list all of the possible public media by which art can be made. Who would have predicted video games as an artistic media? Who knows what technology will invent next that artists will appropriate?
The discussion on pages 75 and 76 consider trying to clean up the theory in response to these criticisms, presenting something like the following:
x is a work of art if and only if x is (1) at least in principle, an intended (2) transmission to an audience (3) of some feeling state (emotion) (4) by means of some public medium.
But Carroll argues on pp. 75 and 76 that even this won’t provide necessary conditions for art. Much traditional art (think Navajo rugs, Hopi vases) is just beautiful and merely exists to produce aesthetic pleasure, which is not really an emotion. In addition much contemporary art exists to provoke thoughts rather than emotions. Andy Warhol, MC Escher, and Frank Stella instantiate this.
Finally, Carroll argues on pp. 78 and 79 that neither the original nor the stripped down theory provides sufficient conditions either. All sorts of everyday behavior involves transmitting these kinds of emotions. Carroll gives the example of an ex lover’s bitter letter.
One thing I think we should get on the table. While Carroll is certainly correct to argue that expression of an emotion is not a necessary condition for all works, we can still ask whether the condition is historically, or genealogically, necessary in the sense that without a substantive body of work satisfying the definition, the institution of art would not exist. Most of the cases of radical art are actually parasitic on art that expresses emotions because the radical art in some manner comments on earlier forms of art that do satisfy the expression theory.
As far as I can tell the only possible counterexample to the historical necessity of the stripped down expression theory of art is the art of the beautiful. However, the examples he gives might just establish that theories that capture the art of the beautiful are also historically necessary. Our next two theories, formalism and the aesthetic theory, will take these works as paradigmatic of art. And when we get to historical accounts we will see that the notion of historically necessary conditions can come to play a central role.
Part II Theories of expression (pp. 79-105)
What is expression (79-86)
The “Common View” of expression (p. 81-82)
An artist expresses (manifests, embodies, projects, objectifies) x (some human quality) if and only if (1) the artist has been moved by a feeling or an experience of x to compose his artwork (or part therof); (2) the artist has imbued his artwork (or some part of it) with x (some human quality); and (3) the artwork (or the relevant part) has the capacity to give the artist the feeling or experience of x when he or she reads, listens to and/or sees it again, and, consequently, to impart the same feeling or experience of x to other readers, listeners, and/or viewers.
The criticism of each condition in the common view (pp. 82-86)
(1) the artist has been moved by a feeling or an experience of x to compose his artwork (or part therof);
Carroll calls this “the sincerity condition” and critiques it on pp. 82-84. The first problem is that it’s false of lots of art (we saw this same argument with respect to the transmission version of the expression theory of art, above). Then Carroll explains why it seems plausible because ordinary cases of people expressing emotions almost always work this way. But not always! A person can have a sad face without in fact being sad.
Also, Carroll differentiates between the expression of a given emotion (when the expression is evidence that the person has the emotion) and when a given expression is expressive of a given emotion. This is the sense in which a face can be sad because of its configuration or a weeping, willow tree can be sad.
(2) the artist has imbued his artwork (or some part of it) with x (some human quality); and
Carroll actually supports this clause (p. 86), and the rest of the chapter is concerned with what it is to imbue an artwork with human qualities.
(3) the artwork (or the relevant part) has the capacity to give the artist the feeling or experience of x when he or she reads, listens to and/or sees it again, and, consequently, to impart the same feeling or experience of x to other readers, listeners, and/or viewers.
Carroll calls this “the arousal condition” and discusses it on pp. 84-86. It fails for three reasons. (1) Often the emotion expressed are different from the ones aroused! Raskolnikov is remorseful, but the reader of Dostoevsky’s Crime and Punishment feels pity. (2) Some of the expressed properties are not even emotions, but character traits. But character traits are not the kind of things that can be aroused. Sometimes when artists share personal information we refer to the artwork as courageous, but this kind of art does not make the members of the audience courageous. (3) Some artworks systematically lack the resources to arouse the very emotions those artworks expressed. An angry piece of music that has no lyrics probably won’t make listeners angry, because anger needs to be directed at something and the lack of lyrics prevent the listener from having any idea about who should be the object of anger. And when there is no object of anger specified, it’s just hard to have one’s anger aroused.
In addition the arousal condition is not sufficient (p. 86) for expression. Lots of unexpressive things can arouse emotions, such as a dry news report of some disaster.
Expression, exemplification, and metaphor (86-92)
This section is dreadfully confusing, Carroll starts with the following view of exemplification, which is a fancy way to denote something being an example of some property.
x exemplification y if and only if (1) x refers to y, and (2) x possesses y,
but then his actual examples are not of this definition, but rather a definition he doesn’t give, namely the following:
x expresses y if and only if (1) x refers to y, and (2) x possesses properties relevant to being y.
His examples: An artwork, call it x, expresses mourning if and only if (1) it refers to mourning, and (2) “it shows for the relevant properties of mourning by being sad” (p. 88). He gives the example of a song that exemplifies stateliness and notes that the composer, “gives the music a slow, deliberate pace, rather than a fast, congested one. In this she tries to capture a characteristic appearance of stateliness, it’s slow, deliberate cadence.” In this manner the song refers to stateliness and possesses relevant properties of being stately (having a slow, deliberate pace).
Then, the rot in his definition of expression infects his definition of the metaphor view of expression (p. 89).
x expresses y if and only if (1) x refers to y, and (2) x possesses y (3) metaphorically.
Again, the second clause of this should be “x possesses properties relevant to being y.” But the problem is, once one makes that change, his argument that we should add the third clause seems to vanish. For his argument that we should add the third clause depends upon the thought that an artwork can’t literally possess an emotion or character trait, so therefore it must do so metaphorically. But his examples of exemplification don’t require the artwork to possess the trait at all, but rather just to possess properties relevant to the possession of that trait.
There’s a lot of stuff here on the nature of metaphor. On the view Carroll explicates, metaphors involve mapping systems of concepts (e.g. hot-tepid-cold) onto types of things to which they don’t literally apply (e.g. styles of jazz) where relations between the two systems are preserved by the mappings. So if I make a metaphorical judgment that x is a (John Coltraine’s style is hot) there must be some other y and b such that y is b (Miles Davis’ style is cool). The homology can be expressed by saying x is to a as y is to b.
Some problems with the theory of metaphorical exemplification (93-95)
Unfortunately most artistic expression doesn’t neatly map homologies such as temperature relations and relative speeds of jazz music. Carroll’s examples (p. 93) “late nineteenth-century, fin-de-siècle, bohemian, anarchistic despair and loathing” and “mid-century American brashness, naïveté, and sentimentality.” The latter is exemplified by the original movie King Kong. But there is no system of relations analogous to the temperature/jazz homology. Or consider the claim that “Patriot Games is very John Waynesque.” We know exactly what this means without a system of homology (again, had Carroll just adopted the definition of exemplification I suggested, there would be no problem with saying a work exemplifies this property).
In fact, lots of metaphors don’t involve such systems, e.g. the moon being “a ghostly galleon.”
Likewise the theory isn’t sufficient for expressiveness, because many metaphorical properties held by artworks are not expressive properties. Carroll (p. 94) gives the example of a work of art being brittle.
Is expression always metaphorical? (pp. 95-104)
The argument that expression is always metaphorical (pp. 95-96).
1. If artworks (and parts of artworks) possess expressive properties, they do so either lieterally or metaphorically, (premise)
2. If artworks (and parts of artworks) possess expressive properties lieterally, they must be the kinds of things that can bear mental properties (premise).
3. Artworks (and parts of artworks) are not the kind of things that can bear mental properties. (premise)
4. Therefore, artworks (and parts thereof) do not possess expressive properties literally.
5. But artworks (and parts of artworks) do possess expressive properties. (premise)
6. Therefore, artworks (and parts thereof) possess expressive properties metaphorically.
The critique of premise 3 (pp. 96-99) of the argument that expression is always metaphorical
Premise 3 states that Artworks (and parts of artworks) are not the kind of things that can bear mental properties. Carroll provides a variety of counterexamples involving parts of artworks: characters of fiction (Mulder and Scully being deadpan in The X Files), properties of human groups (the boys barbarity in Lord of the Flies), and the implied artist’s point of view (the director’s ironic stance in Pulp Fiction, and the implied, often fictional, narrator of lyric poems, e.g. Larkin’s “Church Going”). Carroll considers three objections.
(1) The idea that fictional characters can’t be ascribed mental properties. This is pretty weak. We aren’t being metaphorical when we do this.
(2) The idea that we are only attributing human properties to parts of artworks here. But this is again weak. Generally what we mean when we say a work of art has some property is that one (often the dominant or most salient) or more of its parts do.
(3) Maybe the exemplification theorist is confusing exemplification and representation? Carroll is actually not very clear at all on this point. He notes that within an artwork there can be the difference between mere representation (say the words “this is angry”) and expression (a character doing something angry). But the representation theorist is saying that the artwork represents the expression, and as a result of this the artwork doesn’t itself express anything literally. Carroll doesn’t consider this point.
The critique of premise 2 (pp. 99-102) of the argument that expression is always metaphorical.
The second premise says If artworks (and parts of artworks) possess expressive properties lieterally, they must be the kinds of things that can bear mental properties. Carroll refutes this with the example of how a Saint Bernard’s face can be expressive of sadness even if the Saint Bernard himself is in fact happy. It is expressive of sadness because of its characteristic properties. Droopiness is a property characteristic of being sad, and because of this we call droopy things sad. We do this kind of thing all the time, such as calling a storm furious.
Carroll argues that these things aren’t in fact metaphorical, but it seems like a dead issue to me because the important part is that we apply an anthropomorphic predicate when the things in question bear properties relevant to being an instance of that predicate. That’s the important part, and whether or not someone wants to call it “metaphorical” seems to me to be beside the point. In this light, reconsider my criticism of Carroll above, when he first characterizes exemplification and expression.
Two additional points-
(1) Carroll never notes an interesting point about expression. Just as Danto argues that the representational properties of an artwork are often a function of things that the unaided senses, aimed at the artwork, cannot alone uncover, expressive properties or artworks are similarly externalistically determined.
When we listen to Beethoven’s 9th symphony we hear the joy in overcoming adversity, in performing the alchemist’s trick of turning the muck of life into gold. Part of the reason we hear it this way is because we think of Beethoven’s own life, and abused child who as a deaf adult has managed to wrestle this beauty from the mess of his own life. But if the song had been composed in quite different conditions we would hear it differently. It’s not hard to think up examples.
(2) Carroll notes how an artist’s emotional state can be quite different from what her artwork expresses. One should extend this to note that in lyrical music the emotions expressed by the music itself are often in tension with the emotion expressed by the lyrics if read as a poem. This tension is sometimes part of what makes a great song great. The Toadies song “Possum Kingdom” is an example of this, with pretty peppy music making the actual song all the more creepy, likewise with the Talking Heads “Psycho Killer.” In a happier vein, Weird Al Yankovic’s satire often exploits this tension.