Back to Phil 280 home page

Class Notes for Philosophy of Art

PHIL 280


Introduction (Carroll):


Analytic philosophy of art investigates the concepts and practices of art: e.g., What is an art work? What makes an object an artwork?, etc.

Providing a definition of what constitutes an art work is the main problem in the philosophy of art, since philosophers want to know what it is that makes an art work different from ordinary objects. If a definition can be provided that uniquely identifies, or characterizes, only those objects that are art works, and doesn't leave any art works out, or include objects that are not art works (i.e., a "necessary and sufficient" condition has been provided for art works), then a satisfactory theory of art will have been found (although there could be many such definitions).

Definitions can go awry in one of two ways:

1) "too narrow": the proposed definition leaves out many objects that we believe should be included in the definition. Example: the definition of a "chair" as "a four-legged sitting object" is "too narrow" a definition since it doesn't include all the chairs that have three legs, or two legs., etc.

2) "too broad": the proposed definition includes many objects that we believe should not be included in the definition. Example: the definition of a "chair" as "a four-legged sitting object" is also "too broad" a definition since many people sit on desks, and many desks have four legs, thus the proposed definition also includes desks (which are not properly considered to be chairs, of course).


Imitation Theory of Art (Carroll, chap. 1)


(Plato/Aristotle) Imitation Theory: x is an art work if x is an imitation.

The Imitation theory believes that art imitates life, so art works try to accurately resemble real life objects, persons, events, etc., and this imitation evokes an aesthetic (artistic) response in the observer/audience.

Plato thought that art also aroused emotions in the audience members, and since he believed that emotional behavior/responses tended to override a person's rationality, Plato was thus skeptical of the worth or value of art (since emotionally aroused citizens in a state may act irrationally). Aristotle rejected Plato's misgivings over the value of art: (a) art can cause people to think and reflect on human affairs, as well as their actions and emotions, and this can have an educational effect; and (b) art serves as a "catharsis" of the emotions, such that art helps people to channel out, or purge, raw and powerful emotions—and purging these emotions has a beneficial effect on people.

Problems with the Imitation Theory: Most art does not "imitate" anything (e.g., absolute music, abstract paintings and sculpture, architecture, etc.). So, the definition is "too narrow", since it leaves out many objects that we consider to be art (i.e., it leaves out those art works that don't imitate anything). Also, much art does try to imitate something, but no one knows what the object/event/person looks like (such as God, or a historical figure, etc.); so, how can the art work imitate something that no one knows about?

Representation Theory (Carroll's definition): x represents y (where y ranges over a domain comprised of objects, persons, events and actions) if and only if (1) a sender intends x (e.g., a picture) to stand for y (e.g., a haystack) and (2) the audience realizes that x is intended to stand for y.

The Representation theory tries to get around the problem of art imitating unknown objects (such as God, etc.) by claiming that art works "represent" something, without having to imitate it: that is, a painting of God tries to represent God, but it clearly can't imitate God (since no one knows what God looks like).

Problems with the Representation Theory: the same problem discussed with respect to the Imitation theory also arises for this theory; that is, the Representation theory is also "too narrow". Much art is not representational, such as architecture, design patterns in furniture, abstract music, etc. For example, what does DuChamp's sculpture "Fountain" represent (since it is simply a urinal put on a pedestal)?

Neo-Representational Theory (Carroll's definition): x is an artwork only if x has a subject about which it makes some comment (about which it says something, or expresses some observation or opinion, etc.). In short, art works are "about something". The Neo-Representational theory tries to resolve the problem raised for the Representational theory by allowing art works to simply be about some general subject: therefore, DuChamp's "Fountain" can be understood as making some comment about the overall nature of art (i.e., the urinal makes us reflect on the very nature of art, and art objects, such that we may now look upon ordinary objects as prospective art works, and thus see their previously unrecognized beauty).

Problems with the Neo-Representational Theory: Many art works are not clearly "about anything", such as architecture, purely decorative art, abstract music, etc. (so the definition is, once again, "too narrow"). One might try to state that a particular piece of absolute music is "about sadness" (such as the slow movement of Beethoven's "Hammerklavier" piano sonata), but this seems to be an incorrect judgment: a piece of music can "express" sadness, but it is not "about" sadness.


Expression Theory of Art (Carroll, chap. 2)

The "Transmission" Theory (a version of the Expression Theory, Carroll's definition):

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 (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. In short, the Expression theory holds that an artwork is an object that transmits an emotional state, or is the expression of an emotion, to an audience.

What do the criteria accomplish?: (1) excludes unintended or accidental art works (e.g., a painter accidentally spilling paint on a canvas); (2) guarantees that there is an audience; (3) requires that the emotion transmitted to the audience is the same as that intended by the artist; (4) requires that the emotional message be personal, and not general or generic (as in a greeting card or TV commercial); (5) demands that the artist express an emotion in an artwork; (6) guarantees that the artist really has the emotion, and didn't fake it; (7) requires that the emotion be clear and unambiguous; and (8) the expression of emotion should utilize the normal artistic media.

Problem with criterion (2): Many great works of art were never performed, read, seen, etc., for many years, but it seems crazy to declare that they were not artworks during that time period. Therefore, if we drop criterion (2), we end up with the "Solo Expression" Theory (as a version of the Expression Theory), which has all the other criteria (except (2), of course). On this theory, if an art work uses a normal artistic media, then it assumes an audience either (i) in principle (i.e., hypothetically), or (ii) the artist is himself/herself the audience (since he/she critically examines it while it is under construction).

Problems with the Expression theory:

Criterion (1) is "too narrow": Why should an object be disqualified as an art work just because it has been accidentally or unintentionally created? In fact, many ancient artifacts, such as religious icons/symbols, are appreciated in an aesthetic manner, but they were not intended to be art works—so, are we mistaken when we enjoy these objects aesthetically? Similar "too narrow" problems can be raised for criteria (3)-(7): (3) Why should an unintended emotion disqualify the object as an art work (i.e., bad, or faulty, art is still art)? (4) Why can't some art works be generic (such as postcards)?: i.e., one can view a postcard from an aesthetic point of view. (6) Many artists express emotions in their work that they may have not been experienced (such as extreme fear in battle, etc.). (7) Some of the best works of art express emotions which are very ambiguous, but often such works are more admired and emotionally powerful for that very reason (e.g., absolute music, such as symphonies, fugues, etc.)! Also, this theory seems to rule out all computer-generated art, or chance art, since a human isn't involved in the actual construction of these works—yet, we do consider these objects to be art works.

The whole Expression theory also seem "too broad" since many things that we don't consider to be art works would seem to satisfy all of the criteria. For example, an angry letter in the editorial page of a newspaper would seem to meet all of the criteria (since it was intended to express an emotion to an audience, is the same emotion as experienced by the author, is not generic, was not a faked emotion, is clearly expressed, and uses a normal artistic medium—i.e., words).

One of the main problems, of course, with the Expressionist theory is that even if we accept that an art work must express something to an audience, why must it be only emotions? Many art works are more properly viewed as expressing "ideas", or "concepts" (such as DuChamp's "Fountain"), with the emotions being very secondary to the artist's main intention.

Also, how do art works "express" emotion? Art works are not people, and thus they cannot possess mental properties (since emotions are mental properties). Yet, we do attribute such properties to art works (e.g., "the fugue is sad", "the painting is happy"). How?—Carroll argues that art works possess "configurational properties" (e.g., rhythms, speeds, shapes, lines, etc.—which some of our authors have dubbed "form"), and we have, anthropomorphically, assigned emotional properties to these configurational properties. Does this theory imply that a different "artistic" culture could assign a different set of emotional properties to the very same work?


Formalist Theory of Art (Carroll, chapter 3)

Definition (p.115): x is a work of art if and only if x is designed primarily in order to possess and to exhibit significant form (where "significant form" can be conceived as the complex structure or arrangement of the parts of the object: e.g., in the case of sculpture, significant form is the arrangement and interrelationship of lines, shapes, contours, three-dimensionality, etc., to form a whole that is greater than the sum of its parts). The Formalist theory focuses upon the form or structure of art objects, such that art objects are those objects which posses the appropriate type of formal or structural features.

In this definition, the criterion "designed primarily" is supposed to eliminate those objects which are not artworks but possess a complex arrangement of its parts, such as mathematical equations or road signs. Also, the requirement for a "designed" object prevents accidental objects from being considered as artworks (such as if a painter accidentally spilled some paint on a canvas, such that the spill now exhibits the appropriate form).

Formalism has the advantage (over our previously examined theories) of accounting for modern abstract art. To count as an art work, the object only needs to have a significant structure and arrangement of its parts—thus, Duchamp's "Fountain" qualifies as an artwork, since it is clearly the case that even a urinal has a complex arrangement of lines, curves, shapes, etc.!

Problems with the Formalist Theory:

(1) Some objects that we consider to be artworks, such as ancient religious sculptured figures (to ward of evil spirits, bring good luck, etc), were not "designed primarily" to exhibit significant form. (Furthermore, road signs are often used as artworks, as a sort of wall decoration.) If the formalist tries to account for this problem (of being "too narrow") by dropping the "designed primarily" criterion, then it seems that all objects that possess a complex structure now qualify as artworks (such as road signs and mathematical equations), Unfortunately, a theory that allows everything to count as an artwork is in serious trouble, since we do believe that there is a real "aesthetic/artistic" difference between, say, the nutrition label on a candy bar and a Shakespeare play! (Now the definition is "too broad".)

(2) What counts as significant form? All things have some type of form, but what makes it a "significant form"? What type of structural complexity must an object possess to qualify as having significant form (and how much)? There are real difficulties here, since some artworks do not seem to possess any coherent structure, at all: e.g., John Cage's musical piece "four minutes and thirty-three seconds" (where a pianist simply sits before a piano for the designated time period without playing or doing anything); or Robert Morris' sculptures, which are simply piles of junk. Cage and Morris's work are considered to be artworks, but it is hard to find any form in them. If the Formalist declares that these objects do possess a significant form, then they are left in the embarrassing position of bestowing upon all piles of junk and all silent moments the status of an artwork. (Once again, this is the "too broad" problem.)

(3) As mentioned above, there seems to be many objects that were "designed primarily to exhibit significant form", but we don't consider them to be artworks (such as mathematical equations, chess games, athletic performances, etc., which is the "too broad" problem).

(4) In the Formalist theory, the content of the artwork has no role. That is, what the artwork, and the parts that comprise that object, represents or signifies is irrelevant to significant form (since significant form is only concerned with the structural complexity of the work). For example, the Formalists believe that the content of a painting (what it represents or "stands for") is not important in analyzing and qualifying artworks—this is one of the reasons why the Formalists can allow modern abstract art to qualify as art, since the parts of these works (like the colors on a Pollack painting) really do not represent anything. Unfortunately, this belief, that the content of an artwork is irrelevant to its "aesthetic" qualifications or value, seems to be quite wrong. The aesthetic value of many artworks, if not the majority, depends greatly on understanding what the parts of a work "stand for". For example, the whole aesthetic point or intention of Bruegel's painting (depicting the fall of Icarus, p. 124) depends on the viewer recognizing the leg in the water as representing Icarus' leg, and not simply in viewing the leg as a particular shape, color, etc., within a sophisticated structure of other shapes, lines, etc. Thus, the Formalist theory seems to fail in fully characterizing our aesthetic experiences ("too narrow").

Neo-Formalist Theory of Art:     

Definition (p.136): x is an artwork if and only if (1) x had content (2) x has form and (3) the form and the content of x are related to each other in a satisfyingly appropriate manner.

This version of the Formalist theory tries to account for problem (4) above by allowing the work, and its constitutive parts, to have content (to represent something), and to allow the content of the whole work to pick the form that is appropriate to expressing that content (as in Breugel's "Icarus").


(A) Many artworks do not have content (i.e., they don't represent anything), such as architecture, decorative patterns on furniture, etc. (so the definition is "too narrow").

(B) As mentioned above (problem 2), many artworks don't have a significant form (and if they don't have content, it also follows that they don't have form, since form and content are co-dependent in this theory). ("too narrow")

(C) Bad artworks would not qualify as art since the form wasn't appropriately tied to the content. But, we do consider bad art to be art (so the definition is "too narrow" once more).

(D) Many non-artworks are intended to possess a significant conjunction of their form and content, such as a Brillo pad box or stop sign (i.e., the shape and color of the stop sign are conjoined in a significant form to represent the information "stop"). Thus, since the theory allows non-artworks to qualify as artworks, it is also "too broad".


Art and Aesthetic Experience (Carroll, chapter 4):

Definition: x is an artwork if and only if (1) x is produced with the intention that it possess a certain capacity, namely (2) the capacity of affording an aesthetic experience.

The Aesthetic theory tries to identify artworks as those objects which cause, or bring about, an "aesthetic" (artistic) experience in the observer. Of course the problem then becomes, "What is an aesthetic experience? How does it differ from other experiences, and what causes it? There are two versions of this theory, which attempt to answer these questions:

(A) Content-oriented definition: x is an artwork if and only if x is intended to present unities, diversities and/or intensities for apprehension (i.e., the work has "formal" features which cause aesthetic experiences in people).

Problems: The criterion is "too broad" because many non-artworks, such as stop signs, billboards, etc., are "intended to present unities, diversities and/or intensities for apprehension." If one tries to rule out these potential objects by insisting that only artistically or aesthetically "relevant" features can qualify something as an artwork, then the definition becomes "circular"—that is, the theory was supposed to identify the aesthetic feature of an object, but now it seems that we need to know what those aesthetic feature are before we use the theory!

(B) Affect-oriented definition: x is an artwork if and only if x is intentionally produced with the capacity to afford the disinterested and sympathetic attention and contemplation of x for its own sake. Problem: same as above, "too broad", since many non-artworks can be seen to "afford the disinterested and sympathetic attention and contemplation of an object for its own sake"—e.g., car designs, cutlery, etc.

One of the main problems with the two definitions provided above is that they accept the view that an aesthetic experience is only "pure", or valid, if the observer is completely "disinterested" (or at least to the greatest degree possible), the reason being that a person who has a certain non-aesthetic reason for wanting to experience an artwork may be biased in their experience of that work. Yet, why must we require that a person be disinterested? (And, is it even possible?) It seems that we can have "interested" reasons for wanting to acquire an aesthetic experience; such as for a class, or to write a review, etc. It would appear that the crucial requirement is "attention": someone must "attend to" the object in order to have an aesthetic experience, but their motivation seems irrelevant (which is Dickie's argument).


Notes for the Final Exam Start Here


Objectivity/Subjectivity of Aesthetic Properties: 

What is the status of aesthetic properties? Are they subjective (entirely relative, or reducible, to some other features of the world, such as human emotions), or are they objective (not entirely relative, or reducible, to some other features of the world)? Furthermore, how do we detect, or project, aesthetic properties? Are aesthetic properties like the properties of taste, color, or other bodily sensations?

Objectivist Theory: aesthetic properties are objective. Just as people agree on colors, most people agree on aesthetic properties (e.g., "the piece of music is sad"). For instance, there is inter-subjective agreement on color properties among all people even though we know that colors don't exist in the world completely independent of our minds (why?: because colors are sensations, and although objects cause those sensations in us, those sensations only exist in the mind—the same holds true for pains, tastes, feelings of hot and cold, etc.). In the same manner, aesthetic properties are objective because there is an inter-subjective agreement on the mental properties (e.g., sad, happy, etc.) produced by art objects.

Problems with the Objectivist theory:


(1) The first criticism can be simply stated: There is inter-subjective agreement on color properties, but not for aesthetic properties. People will generally agree that something "looks blue", or "tastes bitter", or "feels hot"; but they often disagree on aesthetic properties, such as, "this movie was funny, or deeply moving, etc." In response, the Objectivist may claim that all people experience the same aesthetic property, but just interpret that experience differently (so there really is still an aspect of aesthetic experience that is common to all people). Unfortunately, this is not a very plausible reply, since how do you separate the aesthetic experience from the interpretation of that experience? That is, the interpretation of the aesthetic experience seems to be part of the aesthetic experience (so how can you pull them apart?). Moreover, if you brought our non-Western tribesman to a concert hall (who has never heard Western music before) it is not clear that they would hear, say, Beethoven's 5th Symphony as "powerful" (in fact, it is entirely unclear just what kind of experience they would have at all).

(2) Our second criticism states: unlike color properties, Aesthetic properties are socially and culturally conditioned, or constructed, and thus are not objective in any way. In response to this allegation, the Objectivist may respond as follows: since disagreement presupposes a certain amount of agreement (i.e., you need to agree on many things before you can begin to disagree), it must follow that the different uses and applications of aesthetic properties among cultures, societies, etc., presuppose a good deal of agreement on the objective aspects of aesthetic experience. However, this response is faulty because (even if we accept the argument) the shared cultural properties required to guarantee disagreement need not be aesthetic properties at all! Rather, the properties that are common to both cultures may be entirely non-aesthetic objective properties, such as the rhythms, tone, pitch, etc., of a musical piece. Therefore, two cultures may experience completely different aesthetic experiences (without any similarities) given the same physical (formal) properties of the artwork.

Ed's view (which I think is Hume’s view): While I agree with the previous two criticisms, it doesn't rule out the case that there can be objective aesthetic properties within a culture, society, etc. That is, if we are willing to allow a slightly "weaker" notion of objectivity for aesthetic properties, then it can be an objective fact that all people judge Beethoven's 5th Symphony to be "powerful", but only within my culture, of course.


The Neo-Wittgensteinian/Institutional Theories of Art (Carroll, Chapter 5)   

Neo-Wittgensteinian Theory of Art

This theory is largely based on the views of the great Austrian philosopher, Ludwig Wittgenstein. The theory holds the following views:

(1) Art is an "open" concept: it is impossible to provide an exhaustive list of all the properties that an object must possess to qualify as an "artwork". That is, it is not possible to provide an "essential " definition (in term of "necessary and sufficient conditions", see notes to Introduction). Previous theories of art thought that you could provide such a complete list of all the conditions that an object must meet to qualify as an artwork, but this was a great mistake, as the history of art has demonstrated (e.g., at one time, “imitation” was thought to be the only important criterion, but that changed, as did the other beliefs about what constitutes an artwork). Hence, our concept of art is constantly changing an adapting over time, which is what it means to call "art" an open concept.

(2) There is no single property that is common to all artworks; rather, there is a complex series of overlapping properties and resemblances. In short, a loose collection of properties is what is common to all artworks, but it is not the case that there is a single property in this loose collection which is possessed by every individual artwork. All that is required is that the object have a large number of those properties, but it might not be the case that one single property is possessed by all of them. This way of conceiving properties is called a "family resemblance" theory; for, just as in a family resemblance, it might be the case that there is not one feature (nose, eyes, hair, etc.) that all of the people who are judged to resemble one another possess; but, each person has some of these same properties, which is enough to guarantee the resemblance among them. The advantages of this theory is that it allows many different objects, none of which possess the same identical feature, to all count as artworks.

(3) An "art practice" is the means by which the open concept of art changes over time (and also the means by which certain objects are judged to meet the family resemblance concept of what qualifies as an artwork). The art practice is simply the social activity (of artists, critics, audiences, etc.) that deems objects to be artworks.


(A) The "open" concept idea of art is "too broad", since it seemingly allows for anything to qualify as an artwork. In other words, since the definition of artwork can change over time, and can change in any way, it seems that anything can be an artwork, which is absurd. (Whether or not this criticism is effective is left for the reader to decide.)

(B) The "family resemblance" theory of artworks is also "too broad": since everything bears some similarity to everything else, everything can qualify as an artwork. For example, DuChamp's "Fountain" resembles all urinals much more than it resembles a Beethoven symphony; so. if "Fountain" is an artwork, then is seems that all urinals are artworks! In short, defining artworks by a collection of properties or resemblances, rather than a single property, will inevitably open the door for many non-artworks to qualify as artworks.

Institutional Theory of Art

This theory can be seen as an extension of the general idea of art accepted by the Neo-Wittgensteinians. In particular, the Institutional theory accepts that art is an "open" concept, and so what constitutes an artwork can change over time. However, unlike the Neo-Wittgensteinians, the Institutional theory does hold that certain conditions can be specified that identify objects as artworks (i.e., artworks have an essential property, or, saying the same thing, there is a "necessary and sufficient condition" for an object to qualify as an artwork). The means of identifying objects as artworks is simply the "Institution" of art: that is, the social "practice" that comprises the world of artists, art audiences, etc. It is this practice, conceived as a social entity, that picks the objects that qualify as artworks. Various members of this practice have the ability to select/identify objects as artworks (e.g., artists, art-appreciators, etc.). Consequently, to be an artwork is simply to be the object that functions as an artwork in the artworld practice. 

Definition (Carroll, p. 227): x is an artwork in the classificatory sense if and only if (1) x is an artifact (2) upon which someone acting on behalf of a certain institution (the artworld) confers the status of being a candidate for “appreciation”.


(1) Other social practices have well-defined rules on what can be legitimately done, who can do it, etc.—but, this is not the case with the artworld practice. It is not clear just who has the authority, and how they obtain the authority, to designate objects as artworks. This is not really a major problem for the theory, however, because many other social practices may also have loose parameters for designating the object as having the alleged property, such as Wittgenstein's examples of "games" or “family resemblance”. 

(2) It is unclear if a solitary person, such as a sailor stranded on a deserted island, is a member of the artworld. If not, then Robinson Crusoe (or Tom Hanks) can't make an artwork on his deserted island. But, this seems a false conclusion, since we have no problem with accepting the idea that artworks can be fabricated in complete isolation from other people. This raises the problem of just who is, and who is not, a member of the artworld. An Institutional theory, it seems to me, can accept the notion that an isolated individual can be a member of the artworld (because one can claim that if other people come around, then the alleged artwork can be appreciated).

(3) The definition is "too broad" since many other social practices would seem to fit the definition of the art institution (as given above). For example, the “business practice” could be seen as satisfying the above definition, because business people also produce "artifacts", such as business reports, that are conferred the "status of being a candidate for appreciation" by people within the practice, such as when the business committee "appreciates" the report. This is a serious problem, because it means that the definition of the artworld practice is not capable of singling out just that practice, so we still need to know what specifically makes the art practice different from the business practice, or sports practice, etc. If one tries to resolve this problem by claiming that the art practice alone deals with art works (whereas the business practice deals only with business objects), then one has to identify what an "artwork" is (as opposed to a business object, or sports object)—but, the Institutional theory was supposed to provide this definition. So, if you need to know what an artwork is before you use the theory, then the theory is "circular" (because it was supposed to identify artworks, but now it seems that you need to know what an artwork is before you use the theory).  


 Articles from Aesthetics in Perspective, K. M. Higgins

Higgins provides a nice summary to all of the articles in each chapter of her anthology, so I will not bother to repeat them. Make sure that you know the basic ideas and arguments underlying the following articles (assorted by chapter): therefore, you will not need to know all of the material in these chapters.

Chap. 1: Plato ("The Form of Beauty"), Hume, Kant, Nietzsche

Chap. 3: Plato, Wilde

Chap. 9: Tolstoy, Bell, Langer

Chap. 8: Hanslick

Chap. 4: Bullough, Dickie

Chapters to study for the Final begin here.

Chap. 2: Weitz, Dickie

Chap. 5: Harries, Gass, Kuspit

Chap. 6: Mattick, Young, Stalker & Glymour

Chap. 7: Plato, Kant, Schopenhauer

Chap. 10: Wimsatt and Beardslay, Levinson

Chap. 11: Danto

Chap. 12: Allison, Appiah

Chap. 13: Marx