What do we evaluate when we evaluate music, and for what purpose? Philosophers generally agree that, apart from other value music has, music is composed and performed for the purpose of providing listeners with a valuable experience, most often a pleasurable one. (It seems wrong to describe the tragic shock that one feels at the end of a good performance of Puccini’s Madame Butterfly as a feeling of pleasure. Nonetheless, it is rewarding.) For those with sufficient leisure and training to partake of such experiences, the experiences themselves are an independently valuable end that can only be obtained from music. This value is often identified as music’s intrinsic value. Strictly speaking, however, only the experience possesses intrinsic value, whereas the music is instrumentally valuable for providing that experience. This approach is normally called the aesthetic evaluation of music (Davies 2003; Walton 1993). A piece of music can be evaluated from other points of view, each of which may assign a different level of merit. Evaluated aesthetically, John Lennon and Paul McCartney’s “Love Me Do” is a weak song. Nonetheless, it is of some historical interest as their public debut and its copyright has considerable financial value. In contrast, evaluating it aesthetically involves calculating its capacity to provide pleasurable or otherwise rewarding experiences to appropriately knowledgeable listeners who attend to its musical individuality. Although there is considerable debate about why other factors ought to be excluded, I will begin by focusing on the aesthetic evaluation of music.
Suppose that an inquisitive adolescent music lover decides to consult a range of music criticism in order to identify the greatest individual piece of music ever composed. She intends to fill her life with musical experiences of the highest quality by listening to no other music. Furthermore, she will attempt to listen to it as often as possible. A few days of internet research leads her to conclude that Beethoven’s Ninth Symphony is the work she seeks. Recorded performances allow her to sample this work as played by many orchestras and conductors. She occasionally attends a live performance. Listening during all waking hours, her first 100,000 hearings lead her to conclude that Wilhelm Furtwängler’s 1951 performance at Bayreuth is definitive. Over the course of her life she listens to it 330,000 more times. To maintain her objectivity, she occasionally listens to other performances, live and recorded, repeatedly confirming that this one remains the best.
This behavior seems bizarre, if not deranged, for it appears to frustrate the purposes of listening to music. Evaluating music is not like evaluating sports contenders; it does not aim at identifying a winner (Davies 2003: 196). As such, the scenario invites us to question a standard assumption about music evaluation. Put simply, it is that evaluation prioritizes. Evaluation is a comparative activity leading to a prescriptive ranking; evaluation ranks music in order to direct listeners toward better music and away from inferior music. (For brevity’s sake, this chapter focuses on listeners. With modification, it can be understood to embrace musicians as “listeners” who evaluate their own music-making, as well as composers who “listen” to their own works in progress.) On the standard model, evaluating music is fundamentally aligned with the activity of criticism, a public activity with a prescriptive dimension.
This idea of evaluation as prescriptive criticism is honored by our fictitious music lover. Since it leads our listener astray, we must examine its components. For instance, does the listener’s error stem from a lack of warrant for the evaluation? Yet if a weak warrant is the problem, a better justified evaluation need not recommend different behavior. A stronger justification for the same ranking fails to address the fundamental problem, which is the narrowness of this listener’s musical life.
Looking beyond the problem of justification, the deeper issue is the question of how we profit from listening. In asking this question, we seek an instrumentalist account of value, in which music is evaluated in terms of its capacity as a means to some identifiable valuable end. We have assumed that that end is aesthetic reward (see Chapter 15, “Value,” in this volume).
Let us suppose, for the moment, that justified rankings attain a level of objectivity that makes it plausible to regard them as properly prescriptive. Nonetheless, the criticism model is open to the charge that it puts too much emphasis on publicly articulated evaluations, those with prescriptive force. The process of ranking music and then using the ranking to locate better music might be better understood as secondary activities, offshoots from a more basic evaluative activity. That activity is the operation of musical taste, in which evaluating is an essential element of listening, without which there is minimal reward or pleasure. So it is wrong to regard evaluation as external to – consulted before, or formulated after – listening. If evaluating is internal to listening, then everyone who appreciates music regularly evaluates it. There are relatively fewer occasions that demand construction of an objective ranking of music.
“Taste theories,” for example, emphasize that evaluative activity is internal to appreciative listening. Taste theorists argue that musical rewards derive primarily from the active exploration of a musical work’s individuality, which includes evaluating it continuously while listening. A listener experiences perceptual features of the work and, more importantly, various features in interaction with one another; the aesthetic reward arises less from the immediate experience than from the exploratory activity of evaluating that experience while having it. Returning to our overly focused listener, what more is there to evaluate in the same recording after several thousand hearings? Taste theories explain why someone is unlikely to reap aesthetic rewards by limiting the experience of music to a small amount of very good music. After a certain point, there is simply nothing left to evaluate – there is no exercise of taste – and the aesthetic effect becomes that of boredom.
One argument for this position observes that much of our aesthetic terminology is intrinsically evaluative without being particularly descriptive. Were it more descriptive, it would lack the wide range of application that we wish it to have. Consequently, one cannot determine whether a particular musical transition is clumsy without hearing the music and deciding whether it sounds clumsy (see Sibley 2001c). However, this decision involves evaluative assessment. Both localized and overall aesthetic properties of any piece of music are only apparent to those who continuously evaluate it while listening, deciding where it is rewarding and where it is not. A very different argument for the same result begins by noting a difference between receiving pleasure, as when soothed by music, and receiving pleasure in admiring how the music is constructed to have that effect. The latter case, appreciating, requires a second-order response that evaluates the relationship between the musical design and one’s initial felt response. By itself, a mere liking is not evidence of aesthetic merit (Walton 1993). Requiring a second- order response neatly differentiates appreciating music from merely liking it – one can like the sound of Earl Scruggs on banjo without understanding his accomplishment, but one can only appreciate it by recognizing how the pleasure is merited. Furthermore, it makes sense of appreciating music that elicits negative emotions, including sadness, allowing us to find value in what is otherwise unpleasant.
Taste theory and the criticism model are not mutually exclusive. The project of objectively ranking music complements the exercise of musical taste in two distinct ways. Rankings can, as is typically thought, direct listeners toward worthwhile music. But objective rankings have a second function. They are epistemically invaluable for codifying convergences of evaluative judgment and thus providing an external measure of the objectivity of a listener’s musical taste. However, a listener who does not learn how to evaluate musical works independently will not experience the intrinsic rewards that make good music good.
Finally, both the criticism and the taste accounts become more complex upon recognizing that a listener evaluates different musical objects by shifting the range of musical activity to which the music is compared. Even the same piece of music will be evaluated differently, depending on the evaluator’s focus and emphasis. A performance of Beethoven’s Ninth can be evaluated as a musical work, which we may presume is the sequence of sound-types specified by Beethoven as essential to its various performances. Furtwängler’s distinctive contribution should not influence this evaluation, nor should the quality of any of the solo vocalists, for we are only evaluating Beethoven’s accomplishment. However, how can anyone evaluate Beethoven’s Ninth without evaluating different performances of it (Davies 2001: 13–14)? Composers frequently revise works after hearing them performed, so it appears that even composers’ evaluations require perceptual experiences from which emerge the aesthetically valuable features. The problem then arises of determining which properties are due to the work and which are due to the contingencies of its particular realization. Were we in the audience at Bayreuth in 1951, we could evaluate “Furtwängler’s Ninth” (Beethoven’s work as interpreted by a particular conductor). To be warranted, this evaluation must compare his available performances with those of other conductors. Even here, different members of the audience might evaluate it differently – as a particular Furtwängler performance (where the comparison class is other Furtwängler performances, of Beethoven or otherwise), as a Furtwängler Ninth (a much smaller comparison class), or simply as a performance of Beethoven’s Ninth (the comparison class of interest to our overly focused music lover). For twenty-first-century listeners, the experience of that performance is necessarily mediated by its recording, and unless someone regularly listens to older recordings, she is likely to be disappointed that the 1951 recording lacks the sonic range of more recent recordings.
How does an evaluation become warranted? In this section I outline several theories that justify particular evaluations by reference to general principles. In the next section I present objections to these approaches.
In a tradition that stretches back to nineteenth-century music critic and aesthetician Eduard Hanslick, an objective evaluation of a work must be defended by reasons, which in turn requires reference to what can be heard in a performance of that work. Attribution of beauty to a particular Chopin nocturne can be dismissed as subjective unless the listener understands how that beauty emerges from the particularity of the musical work (Hanslick 1986: 58–9). In the twentieth century, several philosophers developed this insight by articulating evaluative principles that use general criteria to support overall evaluations (e.g. “This music is very good”). In one of the most influential theories of this sort, aesthetic success is reduced to the interplay of three features that are always desirable in an aesthetic experience: unity, diversity, and intensity (Beardsley 1981: 454–89). However, these reasons reflect overall impressions that tell us nothing about a work’s particularity. If two Chopin nocturnes are beautiful, then each will have unity, diversity, and intensity, and so these very general criteria bring us no closer to knowing why the nocturnes are musically good than when we merely attribute beauty to both of them. Consequently, overly general criteria are inadequate as explanatory reasons. We need more specificity in the aesthetic attributions that warrant the overall evaluation of the music, describing how the music impresses knowledgeable listeners who attend to its sonic elements unfolding in time. Locating such criteria, we arrive at myriad principles for aesthetic evaluation (Dickie 1988; Sibley 2001a).
On this model, evaluation takes notice of the music’s lowest-order perceptual properties – in the case of a musical work, its lowest-order perceptual property-types, and in the case of performance, of the actual sounds of the performance – in order to attend to aesthetic properties arising from them, such as the foreboding quality of the opening of Beethoven’s Ninth and the irreverence of Varèse’s Ionisation (Levinson 2001). In other words, evaluation is directed at the perceptual appearances that arise from the arrangement of the music’s lowest-order properties, together with the affective responses that are typically reported by qualified listeners. Evaluative principles codify perceptual and affective features that regularly reward an intrinsic concern for music. Given a sufficient store of such principles, we can determine which listeners offer cogent reasons for their overall evaluations of particular works.
There is considerable disagreement about whether appropriate aesthetic attributions must be evaluatively neutral descriptions. Some attributions, such as “beautiful” and “maudlin,” are irreducibly evaluative. However, irreducibly evaluative attributions are generally rejected as an inadequate basis for an overall evaluation. Such “reasons” cannot be used to justify an evaluation, Jerrold Levinson argues, unless their descriptive content can be separated from their evaluative aspect (Levinson 2001). To function as reasons that can be accepted by others, evaluative labels must be replaced with evaluatively neutralized descriptions of the underlying aesthetic properties. Where their descriptive content cannot be separated out, the criteria beg the question by failing to specify just what a knowledgeable listener ought to be able to hear in the music in order to find it rewarding.
On this approach, evaluation proceeds by assembling an evaluatively neutral description of the music, to which we apply many principles of the following sort:
Music rewards intrinsic concern in so far as it is P.
Music frustrates intrinsic concern in so far as it is Q.
P and Q are placeholders for evaluatively neutral aesthetic attributions, and these are either affective or perceptually emergent characteristics (e.g. “cheerful” and “balanced,” respectively). An example would be the claim that Varèse’s Ionisation rewards intrinsic concern in so far as it is irreverent. Because music is good when it rewards intrinsic concern, the music’s irreverence counts in favor of its being good.
However, even if such principles serve as limited indicators of value, there remains the concern that they are insufficient to tell us which music is good. Overall evaluations do not arise from isolated elements, but from taking everything into account. For example, suppose that a composer compiles a list of principles indicating merit and a list of those indicating deficiency. A work could be created that possesses multiple properties that are merit-qualities and avoids all properties that are deficit-qualities. One moment might express foreboding and, in so far as it expresses foreboding, it has merit. The next moment is irreverent, and, in so far as it is, it has merit. The next moment is intensely joyous, and so on. Every moment has merit in so far as it has the property it has in that moment. Nonetheless, the piece might be a hodgepodge of merit-qualities that lack internal connection to one another. (Such music might be composed by appropriating snippets from a range of familiar compositions, or by juxtaposing fragmentary pastiches, as in They Might Be Giants’ track “Fingertips.”) As Levinson observes, one might say “I like how it sounds” at any given moment, but the music will not reward an aesthetic interest in it unless we also like “how it goes,” that is, how it progresses from moment to moment and passage to passage while presenting its various merit-qualities (Levinson 2006: 197–8).
Consequently, Levinson argues that these lower-level principles must be supplemented by an interest in the overall construction of the music as music. Minimally, he thinks that an overall evaluation must proceed from consideration of two dimensions of the music’s designed progression: as configurational form and as expressive gesture. In turn, these two aspects must be evaluated for their “specific fusion of human content and audible form” (Levinson 2006: 201). A musical work is good in so far as it is rewarding to follow its tonal process, it is good in so far as it is rewarding to respond to what it conveys, and it is good in so far as it is rewarding to experience how what it conveys is embodied in its particular tonal process (Levinson 2006: 203). This strategy of identifying universally valuable dimensions of music is reminiscent of Beardsley’s postulation of unity, diversity, and intensity as the general criteria of aesthetic value. Both grant, for instance, that a high degree of reward in one dimension will generally reduce attention to one or both of the other two. However, Levinson argues that his model is more informative than Beardsley’s, for Beardsley sought criteria that apply to every art form, whereas Levinson offers principles that are specific to music.
It will be useful to address two common but misguided objections to evaluative principles before proceeding to more serious problems with their claim to securing evaluative objectivity. First, it is sometimes claimed that aesthetic properties are not objective properties of objects. Because aesthetic attributions describe phenomenal characteristics, they do not refer to objective properties, at least not in the way that the year of the debut of Beethoven’s Ninth is a matter of historical fact. Lacking objectivity, these principles confer no prescriptive force. This common objection has been frequently answered. The classic reply is that the same emergent characteristics are recognized by most, if not all, listeners who have considerable experience with that kind of music. The convergence of agreement about these properties is no less than holds when recognizing color distinctions, the presence of a visual pattern, or the sweetness of honey. Therefore aesthetic attributions should be regarded as furnishing appropriately objective descriptions of what is heard by knowledgeable listeners (Sibley 2001b; Levinson 2001). In turn, the fact of convergence can itself be employed to test the objectivity of a critic, and so a music lover who cannot hear that “the opening of Beethoven’s Ninth Symphony is dark and foreboding” is not competent to say what is present in music of this kind (Levinson 2001: 80).
A second baseless criticism holds that aesthetic evaluation reflects the interests of an elite population and that it is based on principles that privilege fine art. Consequently, it improperly undervalues popular, folk, and non-Western music, which are of value for rather different reasons. In response, at least some “low” music succeeds admirably when evaluated in terms of standard aesthetic values (Shusterman 1991). More importantly, it is clear that non-elite and non-Western cultures employ recognizably aesthetic standards for their cultural productions (Dutton 2000), including music (Davies 2001: 268–73). The fact that evaluative criticism is frequently derailed by cultural biases is no evidence that aesthetic evaluation is essentially elitist, or that beauty is an elitist value.
More serious difficulties arise with low-level principles involving particular aesthetic attributes. One problem is that they operate in terms of isolated features, none of which are necessary for a positive overall evaluation. The principle that witty music is good, to the extent that it is witty, does nothing to help evaluate music that lacks wit. Thus, it tells us that Gilbert and Sullivan’s “patter” songs are to some degree good, but tells us nothing about the Adagietto movement of Mahler’s Fifth Symphony. For Mahler’s Adagietto we need another principle, but it will not always be evident which is the most appropriate. And because they are indefinitely many in number, our inventory will never be complete (Beardsley 1981: 509). At best, our present stock of such principles provides a reminder of the wide range of different norms that apply in various cases. Even in the best cases, we cannot be confident that we possess the principles that justify a positive or negative evaluation; the skeptic says we can never be confident.
Another difficulty with low-level principles is that they treat evaluation as an additive process. Guided by principles, we can articulate how many distinct ways a work is good. However, there are no principles for evaluating interactions among the relevant artistic and aesthetic properties. A piece of music might be good for its expressive melancholy (e.g. the country music standard “He Stopped Loving Her Today”). Another might be good because it features frequent inversions of standard musical syntax. Each of these features is normally rewarding and an intrinsic good. Nonetheless, either can lead to an extrinsic deficit, as happens when one intrinsic good interferes with our appreciation of another (Gaut 2007: 62–3). For example, expressive melancholy and disruptions of standard musical syntax tend to interfere with one another when combined in the same piece. Thus, when Haydn writes a melancholy song, such as “She Never Told Her Love,” he eschews the musical playfulness he displays at the close of “The Joke” String Quartet (Op. 33 No. 2). So it appears that evaluative principles will frequently mislead us unless they receive a host of additional qualifications about their extrinsic entanglements. To remain useful, they must take this form:
Music rewards intrinsic concern in so far as it is P, unless it is also Q.
But the list of entanglements is so open-ended that we can never be confident that we have the full list. Because we can only evaluate these interactions by observing them, case by case, our principles do not guide overall evaluations of any music. We gain nothing by incorporating qualifiers about negative interactions with other aesthetic properties. In the end, principles never license an evaluative conclusion stronger than “the music has some aesthetic merit in so far as it is P” (Gaut 2007: 65; see also Dickie 1988: 159–60).
As a result, principles themselves do not seem to warrant the rankings that we need on a criticism model of evaluation. The rankings are not straightforward products of the principles.
George Dickie offers a partial solution (Dickie 1988). Suppose two works have a common set of aesthetic features, so that both are subject to exactly the same principles, and one is superior with respect to all of these properties. That one is the better of the two. (Imagine that the two works are two variants of the same folk ballad.) A third work that shares the same properties can then be compared with those two, then a fourth with those three, and so. We can thus plot a matrix of better and worse works. Faced with a work that has a property not yet in our matrix, the work can be ranked against otherwise similar works by imagining an additional work possessing all of these properties, and then ranking all of them in relation to that possibility. By gradually comparing actual and imagined works, we can roughly rank most works into the categories of excellent, good, and poor.
While a system of this type may be our only method for comprehensive ranking, it does not get us far. First, comparisons are made to imagined works, which overlooks the way in which composers can be surprised by their own works when they are realized in performance. Second, it retains the problem that the interaction of two independently valuable properties cannot be calculated by appeal to a principle. One (or both) might be of lesser value due to the presence of the other, and the resulting level of reward in the context of the interaction can only be determined by appeal to the consensus of qualified listeners. Hence, even the best scenario for constructing objective rankings is subject to the complaint that the rankings are really a consensus of taste. As such, the principles are ultimately dispensable.
This problem of interaction is marginally addressed by Levinson’s more general principle about the interaction between tonal process and whatever is conveyed by a work. Unfortunately, as Levinson acknowledges, he has simply reintroduced a variant of the problem, for his interaction principle does not rule out the possibility that “works of a markedly representational character” might be of a sort that “harmfully competes with attention to the configurational” (Levinson 2006: 204). He offers no examples, but racist and misogynistic songs illustrate the problem. However well written, the repugnance of the musical persona might negate any rewards to be had from the way that the musical processes support the hateful message. We are thrown back, each time, to an evaluative decision that receives insufficient warrant from our principles.
A third serious difficulty is that principles only emphasize what is typical, for they are generalizations from a range of examples. As such, it is not clear that they are even correct when restricted to saying that music rewards intrinsic concern in so far as it is P. Considered in isolation from its interaction with other properties, a “universally” desirable property might sometimes make a work unrewarding. For example, consider Beardsley’s proposal that a work is always good in so far as it has intensity. Yet we can imagine cases of intense works that are unrewarding (Sibley 2001a: 113). A variant objection is that the phrase “too P” implies fault, and the modifier can be applied to any property to which our principles assign value. Unless mitigated by its interaction with other features, the intensity of a piece might be too great in its overall effect – an intensely sad work, for instance, might be too sad. Given that it is difficult to test our generalization by locating a work that possesses only a single, isolated property (and, even if we could, one that does not induce boredom), the “too P” problem is difficult to defuse. Hence, principles are merely rough heuristics for evaluating partial aspects of works, and they may fail us altogether when the property in question is an overall characteristic of the work in question. The problem arises equally for low- and high-level principles.
There remains an obvious, frequently raised objection to the philosophical focus on evaluating music aesthetically. Most music, in most of human history, was created as a means to some other purpose. Music created to reward an intrinsic concern for its musical individuality is the exception and not the rule – most music accompanies and supports some other activity, and so on. A few of these purposes include encoding and transmitting histories, myths, and so on, in preliterate cultures, coordinating the movements of groups of people (including, but not limited to, dancing and military maneuvers), frightening enemies, facilitating healing, and indicating and reinforcing social differences. Aesthetic evaluation imposes a distorting perspective on such music, which is not designed to reward an interest in it for itself alone (Merriam 1964: 260–3). Therefore each piece of music should be evaluated functionally, as a means to its culturally intended purpose. If there is no justification for evaluating most music aesthetically, then different ends will separate music according to multiple, incommensurable ranking systems.
Some philosophers counter that there is at least one other common purpose, one to which art traditionally subordinates aesthetic purpose. Art and music express and transmit the values of their originating culture (Scruton 1997: 457– 508; Kaufman 2002). Hence, there is an independent basis for commensurate comparison of different musics. Religious or secular, “art” or not, we can ask how well a particular piece of music embodies the values of the culture in which it functions, and we do not have to endorse those values in making this determination. Unfortunately, this position faces the standard criticisms aimed at ethical relativism, including the problem that it does not give positive value to alterity, the music of “otherness,” nor to any music that subverts the dominant culture (see also Gracyk 2007: 167–75).
Furthermore, valuing music for its capacity for cultural integration and solidarity provides no reason for members of one culture or subculture to value the music of another group. In the same way in which one can grasp the value of golf for avid golfers without thereby receiving any reason to golf, one can recognize that opera lovers have reasons to value European opera without therefore receiving a reason to value it. In fact, this strategy provides a reason not to value European opera if it is not part of one’s cultural inheritance, for it will be at the expense of investment in one’s own culture. In short, objective evaluations of cultural products have no prescriptive force unless they are relevant to the life projects of concrete individuals. Far from being a universal language, music appears to be a divisive force.
Happily, this argument overstates the problem. While it is false that music is a universal language, music-making is a universal human activity. Aesthetic rewards are part of the explanation for music’s prominence in diverse cultural activities. Combining music with a cultural activity attaches aesthetic value to that activity, which furnishes an independent incentive to cooperate socially, namely, in order to have access to aesthetically enjoyable music. However, it cannot function as an incentive unless it supplies its own value to a practice or activity. In effect, most “functional” music has very little value unless it also has the potential to become a common bond among individuals who have no other reason to interact. (For example, consider how Haydn’s reputation in England led him to travel there.) Thus, it is not an error to evaluate music for rewarding an aesthetic interest in it. Aesthetic evaluation can be legitimately directed at all music (Gracyk 2007: 41–72).
See also Aesthetic properties (Chapter 14), Authentic performance practice (Chapter 9), and Value (Chapter 15).
Beardsley, M.C. (1981) Aesthetics: Problems in the Philosophy of Criticism, 2nd edn, Indianapolis: Hackett.
Davies, S. (2001) Musical Works and Performances: A Philosophical Exploration, New York: Oxford University Press.
—— (2003) “The Evaluation of Music,” in Themes in the Philosophy of Music, New York: Oxford University Press, pp. 195–212.
Dickie, G. (1988) Evaluating Art, Philadelphia: Temple University Press.
Dutton, D. (2000) “‘But They Don’t Have Our Concept of Art’,” in N. Carroll (ed.) Theories of Art Today, Madison: University of Wisconsin Press, pp. 217–38.
Gaut, B. (2007) Art, Emotion, and Ethics, New York: Oxford University Press.
Gracyk, T. (2007) Listening to Popular Music: Or, How I Learned to Stop Worrying and Love Led Zeppelin, Ann Arbor: University of Michigan Press.
Hanslick, E. (1986 [1891]) On the Musically Beautiful, trans. G. Payzant, Indianapolis: Hackett.
Kaufman, D. (2002) “Normative Criticism and the Objective Value of Artworks,” Journal of Aesthetics and Art Criticism 60: 151–66.
Levinson, J. (2001) “Aesthetic Properties, Evaluative Force, and Differences in Sensibility,” in E. Brady and J. Levinson (eds) Aesthetic Concepts: Essays after Sibley, New York: Oxford University Press, pp. 61–80.
—— (2006) “Evaluating Music,” in Contemplating Art, New York: Oxford University Press, pp. 184–207.
Merriam, A.P. (1964) The Anthropology of Music, Evanston: Northwestern University Press.
Scruton, R. (1997) The Aesthetics of Music, New York: Oxford University Press.
Shusterman, R. (1991) “Form and Funk: The Aesthetic Challenge of Popular Art,” British Journal of Aesthetics 31: 203–13.
Sibley, F. (2001a) “General Criteria and Reasons in Aesthetics,” in Approach to Aesthetics: Collected Papers on Philosophical Aesthetics, ed. J. Benson, H.B. Hildred, and J.R. Cox, New York: Oxford University Press, pp. 104–18.
—— (2001b) “Objectivity and Aesthetics,” in Approach to Aesthetics: Collected Papers on Philosophical Aesthetics, ed. J. Benson, H.B. Hildred, and J.R. Cox, New York: Oxford University Press, pp. 71–87.
—— (2001c) “Particularity, Art, and Evaluation,” in Approach to Aesthetics: Collected Papers on Philosophical Aesthetics, ed. J. Benson, H.B. Hildred, and J.R. Cox, New York: Oxford University Press, pp. 88–103.
Walton, K. (1993) “How Marvelous! Toward a Theory of Aesthetic Value,” Journal of Aesthetics and Art Criticism 51: 499–510.