What Is Music?

1.1 Beyond ‘Pure’ Music
In most of this entry, the discussion focuses on ‘pure’ or ‘absolute’ music – instrumental music that has no accompanying non-musical components. Most of the philosophers whose work is discussed below also put the focus here, and there are three reasons to do so. The first is that pure music often presents the most difficult philosophical problems. It is less puzzling how a musical setting of a maudlin text could be expressive of sadness, for instance, than how a piece of music without even a programmatic text could be, since the emotional expression could somehow be transferred to the music from the text. The second reason is that, though the problems are more difficult, the solutions are likely to be more easily evaluated in the pure case. Just as apportioning blame is easier when one person is responsible for a crime than when the blame must be divided between a number of conspirators, the success of a solution to the problem of musical expressiveness will be clearer if it can explain the expressivity of pure music. Thirdly, it is certain that the expressiveness of pure music will play a role in the expressiveness of ‘impure’ music. Though a text may be able to impart some of its expressiveness to a song, for instance, the musical elements of the song must play some role. A maudlin text set to a jauntily upbeat melody in a major key will clearly not have the same overall expressivity as the same text set to a plodding dirge. Though I have used expressivity as an example here, these same points will apply to discussions of musical understanding and value. There may also be interesting questions to be asked about the ontology of ‘impure’ music, but it is not clear they will be of the same kind as those to be asked about expressivity, understanding, and value.

Given the global prevalence of rock music, broadly construed, it is plausible that song is the most common kind of music listened to in the contemporary world. Film and other motion pictures, such as television, are also ubiquitous. There has been some significant work done on the aesthetics of song (Levinson 1987, Gracyk 2001, Bicknell 2005, Bicknell & Fisher forthcoming), music drama (Levinson 1987, Kivy 1988b, 1994, Goehr 1998), and film music (Carroll 1988, 213–225; Levinson 1996c; Kivy 1997a; Smith 1996). (See also the chapters in part V of Gracyk & Kania 2011; on hybrid art forms more generally, see Levinson 1984.) However, it seems that there is plenty of room for further work on the aesthetics of impure music. ‘Muzak’ is another musical phenomenon that is ubiquitous, yet has received little serious attention from aestheticians, being used primarily as an example to elicit disgust. Whether or not there is anything interesting to say about Muzak philosophically, as opposed to psychologically or sociologically, remains to be seen.

1.2 The Definition of ‘Music’
Explications of the concept of music usually begin with the idea that music is organized sound. They go on to note that this characterization is too broad, since there are many examples of organized sound that are not music, such as human speech, and the sounds non-human animals and machines make. There are two further kinds of necessary conditions philosophers have added in attempts to fine tune the initial idea. One is an appeal to ‘tonality’ or essentially musical features such as pitch and rhythm (Scruton 1997, 1–79; Hamilton 2007, 40–65; Kania 2011a). Another is an appeal to aesthetic properties or experience (Levinson 1990b; Scruton 1997, 1–96; Hamilton 2007, 40–65). As these references make clear, one can endorse either of these conditions in isolation, or both together. It should also be noted that only Jerrold Levinson and Andrew Kania attempt definitions in terms of necessary and sufficient conditions. Both Roger Scruton and Andy Hamilton reject the possibility of a definition in terms of necessary and sufficient conditions. Hamilton explicitly asserts that the conditions he defends are ‘salient features’ of an unavoidably vague phenomenon.

The main problem with the first kind of condition is that every sound seems capable of being included in a musical performance, and thus characterizing the essentially musical features of sounds seems hopeless. (We need only consider the variety of ‘untuned’ percussion available to a conservative symphonist, though we could also consider examples of wind machines, typewriters, and toilets, in Ralph Vaughan Williams's Sinfonia Antartica, Leroy Anderson's The Typewriter, and Yoko Ono's “Toilet Piece/Unknown”.) Defenders of such a condition have turned to sophisticated intentional or subjective theories of tonality in order to overcome this problem. If the essentially musical features of a sound are not intrinsic to it, but somehow related to how it is produced or received, we can classify just one of two ‘indiscernible’ sounds as music. The details of one's theory of essentially musical features will determine how much avant-garde ‘sound art’ counts as music.

If one endorses only an aesthetic condition, and not a tonality condition, one still faces the problem of poetry – non-musical aesthetically organized sounds. Levinson, who takes this approach, excludes organized linguistic sounds explicitly (1990b). This raises the question of whether there are further distinctions to be made between arts of sound. Andy Hamilton defends a tripartite distinction, arguing that sound art, as opposed to both music and literature, was established as a significant art form in the twentieth century (2007, 40–65). This is one reason Hamilton endorses both tonal and aesthetic conditions on music; without the former, Levinson is unable to make such a distinction. On the other hand, by endorsing an aesthetic condition, Hamilton is forced to exclude scales and Muzak, for instance, from the realm of music. Kania (forthcoming a) has suggested that it is a mistake to think that music is necessarily an art, any more than language. He argues that we should distinguish music simpliciter from its artistic uses, just as we do in the cases of language and literature, depiction and painting, and so on. By means of a disjunctive condition, Kania also adds a further distinction to Hamilton's set. Kania argues that music is ‘(1) any event intentionally produced or organized (2) to be heard, and (3) either (a) to have some basic musical feature, such as pitch or rhythm, or (b) to be listened to for such features’ (2011a, 12). The latter disjunct allows for two indiscernible works, neither of which strictly possesses basic musical features, yet one of which is music and the other sound art because of the complex way in which the former is intended to be approached. Kania's approach thus draws on some of the machinery of recent definitions of art in place of an appeal to an aesthetic condition.

Having discussed additional conditions, it's worth returning to the basic idea of ‘organized sound’. Most theorists note that music does not consist entirely of sounds. Most obviously, much music includes rests. You might think that silence can function only to organize the sounds of music. One counterargument is that an understanding listener listens to the rests, just as she listens to the sounds (Kania 2010). Another is to provide putative cases of music in which the silences are not structural in the way ordinary rests are. John Cage's 4'33" is frequently discussed, though there is broad agreement that this piece is not silent – its content is rather the ambient sounds that occur during its performance. Anyway, both Stephen Davies (1997a) and Andrew Kania (2010) argue Cage's piece is not music, though on different grounds. Kania considers several other contenders for the label of ‘silent music’, arguing that there are indeed extant examples, most notably Erwin Schulhoff's “In Futurum” from his Fünf Pittoresken, which predates Cage's 4'33" by some 33 years.