Article
Author: Douglas C. Wadle (University of Southern California)
In this paper I argue that composer John Cage’s so-called ‘silent piece’, 4’33”, is music. I first defend it against the charge that it does not involve the organization of sound, which has been taken to be a necessary feature of music. I then argue that 4’33” satisfies the only other condition that must be met for it to be music: it bears the right socio-historical connections to its predecessors within its tradition (Western art music). I argue further that one cannot understand the organized sound condition and the socio-historical condition separately and that understanding their interaction has theoretical benefits—not least of which is providing a groundwork for a more culturally inclusive philosophy of music. Finally, I consider a number of outstanding questions concerning the content of the organized sound condition for Western art music in the wake of 4’33”.
Keywords:
How to Cite: Wadle, D. C. (2023) “Organized Sound, Sounds Heard, and Silence”, Ergo an Open Access Journal of Philosophy. 10(0). doi: https://doi.org/10.3998/ergo.4632
Composer John Cage’s 4’33” (1952) is a work in three movements, lasting a total of 4’33”, during which the performer remains silent.1 Philosophical discussion of 4’33” has focused on its (in)ability to satisfy the condition—taken as necessary for any instance of music—that music involves the organization of sounds (Davies 1997; Kania 2010; Levinson 2011; Dodd 2018). I begin by arguing that 4’33” does satisfy the condition, when it is correctly specified. However, the organized sound condition, alone, will not pick out all and only instances of music. I briefly survey some candidates for additional conditions and argue that there is only one that is promising; namely, Davies’s (2012) appeal to socio-historical connections that a musical work or practice bears to its forebears. I argue that 4’33” satisfies this condition as well, thereby completing my argument that 4’33” is music.
I argue further that one cannot understand the two conditions independently—there is no interesting tradition-independent version of the organized sound constraint and no complete version of the socio-historical constraint that doesn’t address how sound is organized in the given tradition. Once one does understand the interaction of the two conditions, several theoretical benefits emerge: It helps us avoid parochialism about the organization of sound and its relationship to music, pointing the way toward a more inclusive philosophy of music—one that can engage with music from any tradition on its own terms rather than shoe-horning it into a set of conceptual categories used to describe works by Bach, Beethoven, and Brahms. It also clarifies the relationship between the conditions a work must meet to be an instance of a given tradition and the conditions it must meet, with respect to the organization of sound, to be a good instance of that tradition. This will illuminate the musical values of the tradition, particularly with respect to how innovations are introduced and why they are accepted as such, rather than as defective instances of the tradition. It will also illuminate why experimental works and works in unfamiliar traditions are so often dismissed as bad music on a first encounter but come to be accepted as good music after a more thorough engagement.
I conclude by addressing a number of questions concerning the content of the organized sound condition for Western art music that are not directly relevant to the consideration of 4’33” but are suggested by it and other innovative works in that tradition—including (unsurprisingly) other works by Cage. I offer a roadmap for specifying the contents of tradition-specific organized sound conditions and show how this can illuminate these lingering questions for Western art music. Along the way, the organized sound condition is revealed to be a productive point of contact for research in the philosophy of music, music theory, and (ethno)musicology.
Discussions of 4’33” have frequently invoked, but never fully explained, what is taken to be a necessary condition on music; namely, that it involve the organization of sound. Though the content of the organized sound condition hasn’t been explicitly stated, two questions regarding the requirements it imposes seem to be relevant to the evaluation of 4’33” with respect to the condition:
Can a work be entirely silent? In particular, could a wholly silent work evince a sufficient amount of organization to satisfy the organized sound condition?
How much organization—particularly intentionally directed internal organization—is enough for a work to satisfy the organized sound condition?
I do not address (1) directly. Rather, I argue that 4’33” is not silent and so the answer doesn’t actually bear on the question of 4’33”. Regarding (2) I argue that the demands of intentional internal organization never rise to a level that disqualifies 4’33”.
Can a silent work—as 4’33” is often taken to be—satisfy the organized sound condition? There are two ways that a silent work might fail to meet the condition. On the first, silences are not among the primary materials that a composer/performer organizes—that is, they aren’t in the extension of ‘sound’ on the proper interpretation of the organized sound condition.2 Rather, silences are devices that aid in the structuring of the primary materials of a musical work (non-silent sounds).3 If that is correct, silence cannot, by itself, be organized sound. But this view is wrong.
Take, for example, Beethoven’s Symphony, No. 5: In my undergraduate days, I had a music professor who placed great emphasis on the fact that a proper performance of the fifth symphony must make clear that there is an initial eighth-note rest before the famous first three notes of the symphony’s opening. That is, the listener must hear the silence as occupying the first half of the first beat—as occupying a particular place within the overarching metrical structure of the piece.4 This is not a matter of separating, or otherwise structuring, the primary materials of the piece (non-silent sounds). The silence is to be heard as occupying the first half of the first beat before any non-silent sounds have been heard. It is the metrical structure, then, that functions as a structural device, not the silence. But then silences are structured, in at least some musical works/performances, in just the way non-silent sounds are.
On the second way that a silent work might fail to be organized sounds, silences are among the primary materials but are not, by themselves, apt for organization. Add a silence to another and you don’t have a two-part silence. You just have a longer, single silence. And so the fact that silences are part of the primary materials organized in a musical work does not vindicate 4’33”: The three-movement ‘structure’ is not, on this view, actually a structure of three sequential silences. It is just a single (unstructured) silence.
Of course, this only follows if silences are not apt for organization and 4’33” is silent. In this section, I argue that 4’33” is not silent.5 Since no sounds are intentionally produced by the performer of 4’33”, the question of whether 4’33” is silent hangs on whether or not 4’33” appropriates the sounds heard during its performance. The musicological consensus—following Cage’s own comments6—is that there are sounds to be heard during a performance of 4’33” and that the point of the piece was to direct listeners’ attention to them (Nyman 1974: 22–23; Gann 2010: 11). However, this does not obviously entail that these sounds are part of the performance. We need some principle for determining when, if at all, a performance of a piece appropriates ambient sounds. Dodd (2018) stands out for his sustained discussion of this issue. He offers two arguments for the view that 4’33” does not appropriate ambient sounds.
Dodd’s first argument relies on the principle that “a work-token comprising sounds is a performance of that work only if the sounds are performed in the sense just introduced [i.e., they are produced, combined, or somehow structured by the performers in the course of following the instructions laid out in the work’s score]” (Dodd 2018: 636). Since the performers of 4’33” do not produce, combine, or structure the ambient sounds heard during a performance, according to Dodd, the ambient sounds don’t satisfy the principle and are not part of the performance of the work. By contrast, the silences, which are causally attributable to the performers, are part of the work.7 But Dodd’s principle is faulty. Sounds do not need to be ‘produced, combined, or somehow structured by the performers’ for them to count as part of the work performed—at least not as Dodd understands the requirement:8
Suppose that the composer has written a piece for solo cello, in traditional notation and style. However, the composer has also specified that a microphone be placed outside the concert hall, with the sounds it picks up being played over speakers in the hall. These external ambient sounds would undoubtedly count as part of the work—after all, the composer included instructions for the placement, etc. of the microphone in the score—though they are not produced by the performers, the composer, or even the sound engineer.9 Nor are they manipulated in any way to combine or impose structure upon them. The exterior sounds are merely made audible by way of the microphone and PA system.
In light of this, we could modify Dodd’s principle to state that the performers/composer must produce, combine, structure, or otherwise make audible—by following instructions in the score, for notated works—those sounds (and silences) that count as part of the performance of the work. But this won’t rule out 4’33”: One way to make ambient noises audible is to put the audience in a situation in which they are expected to be quiet while the performers, who are expected to make sounds (which would ordinarily mask the ambient sounds), also remain quiet. But this modified version of the principle cannot be right: not every sound heard during a rest of conventional music counts as part of the performance, as the principle implies.
There is a key difference between the silences of 4’33” and those of rests in conventional works, though. As the musicological consensus has it, 4’33” directs the listener’s attention to the ambient sounds in a way that a rest in a conventional work does not; namely, the silence of the performer exploits conventional concert etiquette such that the audience’s attention is directed to the ambient sounds (more on this below). Such direction of attention is sufficient for appropriating ambient sounds.
To illustrate, consider a piece that is to be played in a space where a sustained, pitched, background noise will be heard (e.g., the hum of an air conditioner). The score directs the performer to transpose the traditionally notated phrases within the score so that they harmonize with the background noise in particular, indicated ways. The resulting pitch relations draw the audience’s attention, thereby including the background sound into the scope of the attended sounds, treating it as a drone. But then the background sound is part of the performance, though the performer does not produce it or structure it and, insofar as she combines it with the other sounds of the piece, she does so by directing the audience’s attention to it.10
This suggests that we should modify Dodd’s principle to state that performers/composer must produce, combine, structure, or otherwise direct listeners’ attention to—by following instructions in the score, for notated works—those sounds (and silences) that are part of the performance of the work. But, again, performances of 4’33” satisfy this principle.11
Perhaps one could seize on the phrase “following the specifications of the composer’s score” to argue that the ambient noises heard during a performance of 4’33” don’t satisfy the principle and, hence, are not part of the performance. After all, the score for the ambient drone piece explicitly calls for actions that direct the listeners’ attention to the appropriated sound. Cage’s score for 4’33” gives no such explicit direction. However, many aspects of a musical performance are not explicitly indicated by the work’s composer without impacting whether or not we take the realization of these features to be features of the performance of the work. If, for instance, the composer leaves instrumentation open and offers no explicit directions for choosing which instruments will be used, it does not follow that the particular timbres of the instruments actually used are not part of the attributes of the performance of the work.12
Granted, other aspects of the sounds of the work (e.g., pitch) might be explicitly specified by the indeterminate score, but they needn’t be. Many works of graphic notation specify neither the instrumentation nor how the notations are to be translated into sonic attributes—Cornelius Cardew’s Treatise (1963–1967) being, perhaps, the most famous (and monumental) example (see Fig. 1). Nevertheless, the sounds resulting from musicians’ engagement with the score in a performance setting are considered a part of performances of that work. Whatever conventions obtain for interpreting graphic scores will be the sole source of guidance for the performance of Treatise, and this guidance will be implicit. Given the fact that the resulting sounds are considered a part of the performance of the work, Dodd’s principle will have to allow such implicit guidance. In the case of 4’33”, then, the performer’s silence—given the conventions of the concert etiquette—is enough to appropriate ambient noises on a properly modified version of Dodd’s principle.
Dodd’s second argument is that interpreting 4’33” as silent better realizes Cage’s artistic intentions than interpreting it as appropriating ambient noises, so we should interpret 4’33” as silent. But why does Dodd think that the silent interpretation better realizes Cage’s intentions? Consider the following comment from Cage:
I saw art not as something that consisted of communication from the artist to an audience but rather as an activity of sounds in which the artist found a way to let the sounds be themselves. And in their being themselves to open the minds of the people who made them or listened to them to other possibilities than they had previously considered. To widen their experience, particularly to undermine the making of value judgments. (Kostelanetz 2003: 44)
Dodd—like many other commentators on 4’33”—takes this to indicate that Cage’s intention was to get his listeners to appreciate ambient noises as they are. So far, so good. He goes on to argue that appropriating these sounds to the work would interfere with Cage’s intentions because it would make it impossible for listeners to experience sounds as they are rather than as part of an artistic enterprise (2018: 636). This echoes similar comments from Davies (1997:453). And this is where the problems begin.
The idea is that, in hearing the sounds as part of the work, one listens to them for how they function within that work rather than listening to the sounds as they are (independently of their relationships to other sounds or any artistic purpose), as Cage intends.13 But Cage was not merely trying to get listeners to drop this sort of hearing-as-part-of-a-musical-enterprise with respect to the ambient noises heard in 4’33”. He wanted audiences to drop this attitude with respect to all sounds, including traditional musical sounds. His intention was to establish a ‘new listening’ that is guided neither by our preconceptions about music, its history, and theory nor by the standard way of listening to ambient sounds, where we are focused on what these sounds reveal about their sources rather than how they sound (if we attend to them at all). This is why Cage said that his goal, from 4’33” onward, was to write music the performance of which did not interrupt (but was appropriated by) and was not interrupted by (but appropriated) ambient sounds (Kostelanetz 2003: 86).14
If successful, then Cage has found a way to make works for which hearing a sound as part of the work does not interfere with the ability to hear it as it is (because the work is not heard as an artistic enterprise, in the relevant sense). But if this is his intention, then it does no harm to appropriate ambient sounds to 4’33”. Rather, it helps it along: The traditional staging of the work—musician on stage with an instrument—holds the audience in expectation of a traditional musical performance. The performer might make a sound at any time. And so the audience remains attentive to the sonic environment. As no expected sounds are performed, this attention is transferred to ambient sounds.15 But ambient sounds are not amenable to treatment as instances of traditional concert music. They are not subject to the expectations of the listeners’ historically conditioned ears. Nor is the audience in a circumstance amenable to interacting with the sources of these sounds, which ordinarily guides our attention to ambient sounds. So the audience is led to listen to the sounds without expectations, as they are. If the performance successfully brings about this kind of attention to ambient sounds, then a new form of listening has been brought into the concert hall, one that is applied to ambient sounds and could be applied—as Cage hoped it would with respect to his post-4’33” works—to sounds produced by performers.16
This interpretation does a better job of capturing Cage’s intentions than Dodd’s, on which 4’33” is a work of conceptual art about music (2018: 638). We can see why by highlighting another aspect of Cage’s thought, which Dodd has overlooked:
New music: new listening. Not an attempt to understand something that is being said, for, if something were being said, the sounds would be given the shape of words. Just attention to the activity of sounds. (Cage 1961: 10)
Given his eschewal of art-as-communication, it is contrary to Cage’s intentions to interpret his work as being about anything, including music. If we should interpret the work in the way that best fits Cage’s intentions, as Dodd suggests, then we should adopt my interpretation over his: 4’33” does contain sounds other than silences.
To summarize: On the proper understanding of what it takes for a sound to be included in a performance of a work, 4’33” is not silent. And a principle of charity with respect to the composer’s intentions poses no difficulty for this result. In fact, it demands it. While this doesn’t directly answer the question of the possibility of a wholly silent musical work, it does show that—despite first appearances—the answer has no bearing on discussions of 4’33”.
The preceding section established that 4’33” is not silent. It appropriates ambient sounds. There remains the question of whether these appropriated sounds—the only sounds belonging to a performance of 4’33”—can satisfy the organized sound condition. According to Dodd, for instance, mere appropriation falls short of the requisite organization: “Organizing sounds is a matter of either producing a structured sound sequence (as performers of traditional works of music do), or else giving some structure to the appropriated sounds [. . .]” (2018: 637). In what follows I will argue that 4’33” involves more than mere appropriation.
To illustrate the point that mere appropriation is not enough to satisfy the organized sound condition, Dodd imagines a piece, Simple Imaginary Landscape, in which a single performer turns on a radio that is automatically set to some station and volume setting.17 After a specified duration, she turns it off. According to Dodd, the sounds emitted by the radio are appropriated, but merely turning on the radio for a given duration does not amount to giving structure to the appropriated sounds. Therefore, Simple Imaginary Landscape doesn’t satisfy the organized sound condition and is not music. He goes on to claim that, even if it appropriates ambient sounds (which Dodd denies), 4’33” fails to organize these sounds for exactly the same reasons that Simple Imaginary Landscape does; namely, the appropriated sounds are given no structure by the performer. So, again, 4’33” is not music.
But Dodd overlooks a crucial difference between his Simple Imaginary Landscape and 4’33”: 4’33” is a work in three movements, which are marked by actions of the performer. (We may take as canonical—as subsequent performers have—David Tudor’s premiere performance of 4’33”, in which he marked the movements by noiselessly opening and closing the keyboard cover of the piano.18) The actions of the performer thereby impose an internal temporal structure on the appropriated sounds—organizing them into three distinct units (movements). This scuttles the analogy with Simple Imaginary Landscape, on which Dodd’s argument relies, and shows that there is more organization in a performance of 4’33” than mere appropriation of sounds. One might worry, though, that these gestures are not sufficient as they impose no audible organization on the appropriated sounds. But this can’t be Dodd’s position.
Consider Cage’s Imaginary Landscape, No. 4 (1951), which both Dodd and Davies—the primary proponents of the view that 4’33” is not organized sound—accept as satisfying the organized sound condition (Davies 1997: 458; Dodd 2018: 637).19 This work is written for 12 radios, each operated by two performers. The score gives indications for volume, tuning, and tone (equalization) settings for each radio over the duration of the piece. However, this durational structure is not directly audible. It can only be heard through the impact of the setting changes on the appropriated sounds. But any audible change could equally well be the result of a change in radio settings or a change in what is being broadcast (e.g., a recording of Imaginary Landscape, No. 4), and adjustments to radio settings during a silence will make no audible difference. Insofar as the imposed durational structure is made evident, it is made evident by the performer’s visible gestures, just as with 4’33”.20 Therefore, the organized sound condition can’t require that the organization imposed on appropriated materials be audible. Perhaps there is some other requirement that Imaginary Landscape, No. 4 satisfies but 4’33” does not.
Davies offers one possibility: For appropriated sounds to count as adequately organized, they must exclude some sounds that could be heard in the performance environment during the performance. Davies further reasons that, since all ambient noises are appropriated in 4’33”, no sounds that could be heard in the performance environment are excluded. Therefore, 4’33” doesn’t satisfy the organized sound condition (1997: 458–59). However, as Kania convincingly argues, 4’33” does meet Davies’s condition: intentional sounds produced by the performer are excluded (2010: 346–47). So this condition won’t secure the conclusion that 4’33” is not organized sound.
Another possibility is that at least one sound needs to be produced by a performer (i.e., not appropriated). But, again, Imaginary Landscape, No. 4 serves as a counter-example. The sounds of the piece are just those sounds that are broadcast over the twelve radios during the performance. These sounds are not produced by the performers. So the performers don’t need to produce any sounds to satisfy the organized sound condition.
If merely appropriating sounds is not enough to satisfy the organized sound condition, but the performer needn’t produce any sounds, then some action of the performer (e.g., the manipulations of radio settings in Imaginary Landscape, No. 4) must—at least potentially—impact the appropriated sounds (where these are the only sounds belonging to the performance). But as we’ve seen (§1.1), the actions of the performer during a performance of 4’33” do impact the way that sounds are experienced by exploiting the audience’s knowledge of standard performance etiquette to direct their attention to the ambient sounds. So 4’33” seems to satisfy this requirement.
Granted, we can draw a distinction between manipulations of the physical features of sounds, themselves, and attentional effects impacting the way those sounds are experienced. Perhaps a work that meets the organized sound condition while including only appropriated sounds in its performances must involve at least some intentional (potential) modification of the physical basis of at least one of the appropriated sounds. After all, one salient difference between 4’33” and Imaginary Landscape, No. 4 is that the latter includes instructions for adjustment to the tuning, volume, and equalization settings on the radios by which the sounds are appropriated. These adjustments impact the qualities of the appropriated sounds. But in the case of 4’33”, the performers’ actions do not modify the physical basis of the sounds heard during the course of the performance. If one wants to drive a wedge between 4’33” and Imaginary Landscape, No. 4 with respect to the organized sound condition, this is the place to do it.
Notice that we are considering a requirement that there be some intentional potential modification of the appropriated sounds. The qualification, ‘potential’, is needed because, in a given performance, the adjustments to the radios indicated in the score of Imaginary Landscape, No. 4 might have no effect, given what (if anything) is being broadcast on the relevant stations at that time. Indeed, a performance of Imaginary Landscape, No. 4 might be entirely silent, and yet it satisfies the organized sound condition.21 But 4’33” does not rule out such potential impacts. A performer could mark the beginning of each movement by opening a door in the performance space and the endings by closing the door. This will remove an impediment to external ambient noises that would be heard during the performance, thereby potentially modifying the physical basis of the sounds heard by the audience.22 And so 4’33” satisfies the requirement that the performer’s actions (potentially) modify the physical basis of the appropriated sounds.
Of course, a correct performance of Imaginary Landscape, No. 4 must include the radio adjustments whereas a correct performance of 4’33” needn’t include the door opening/closing.23 If that is the relevant difference between the two pieces, then a piece, 4’33” with Open Door, that is just like 4’33” except that it requires the door opening/closing, will satisfy the organized sound condition. I doubt that those who think that 4’33” does not satisfy the condition will think that 4’33” with Open Door does. If that is right, they must have a more stringent requirement on the (potential) modification of sounds in mind—one that opening/closing doors does not meet. But then they owe us an explanation (and defense) of these more stringent requirements.
Furthermore, there is reason to think that (potential) modifications of the physical basis of appropriated sounds are not necessary in any form—even when these are the performance’s only sounds. Compare with a case of appropriation in the visual arts: Duchamp’s ‘unassisted readymades’. These are works, such as ‘Bottle Rack’ (Fig. 2), that underwent no internal modifications to their physical structure.24 Any organization involved was merely a matter of choosing how they would be presented within a gallery setting. And yet these works are widely accepted as instances of visual art.25 So, assuming—plausibly—that visual art involves similar organizational requirements with respect to its physical materials, parity of reasoning suggests that musical appropriations do not require (potential) modifications of the physical structure of elements of the work to satisfy the organized sound condition.
Therefore, the performer’s modifying the physical basis of the appropriated sounds is not necessary for those sounds to satisfy the organized sound condition, even where those are the only sounds of the piece. So this cannot be the basis of any difference in 4’33” and Imaginary Landscape, No.4’s abilities to satisfy the organized sound condition.
Those who think that 4’33” is not music might respond by accepting the foregoing while objecting that, in a performance of 4’33”, the performer’s actions do not give rise to an organization of sounds that is sufficiently concretely intended by the composer/performer. For instance, while the performer is responsible for the sounds heard during 4’33” being heard as they are (in virtue of directing listeners’ attention to them), the performer is not intending any particular sound(s) to be heard—let alone to be heard as standing in any particular relation to any others. The sounds may well be organized, but not (primarily) by the performer. To illustrate, contrast 4’33” with Duchamp’s readymades. When Duchamp selects a readymade, he knows exactly what form the appropriated work will take—namely, the form it already has when he selects it. But in 4’33” neither Cage nor the performer knows precisely what form the appropriated sounds will take.
To answer this worry, we can turn once more to Imaginary Landscape, No. 4. Cage says, with respect to the sounds resulting from the manipulations of the radio settings, “whatever happens [is] acceptable (station, static, silence)” (Cage 1961: 58). Insofar as there is any intentional organization of the resulting sounds on the part of Cage, it is several steps removed from explicit choice.26 Similarly, the performers do not know what sounds will result from their actions. Any sound could be transmitted over the airwaves. Any sound could work its way into the piece. A performance of the work could even be completely silent. And so there is nothing that the performers are intending to do with respect to the sounds themselves. Rather, they aim at executing the instructions and letting what happens happen.27
But, again, Imaginary Landscape, No. 4 is widely—and correctly (see §2.3)—accepted as satisfying the organized sound condition. Therefore the condition cannot require direct intentional organization (including the production and modification) of any attributes of sounds—pitch, timbre (sound quality), intensity (loudness), and duration. Duration is the only attribute that might give us pause here, but the intentional durational structure is specified for adjustments to the radio settings, not for the appropriated sounds themselves. The durational structure is, therefore, more like the three movement structure of 4’33” than it is like traditional rhythmic notation.
Notice, too, that the sounds that are heard are dependent on the time and location of the performance and how that interacts with the specifications for tuning, equalization, and volume appearing in the score. If no signal is picked up from any radio station in the performance venue, then all that will be transmitted is static, regardless of the performers’ adjustments to the tuning dials. If the radio picks up silence, then adjustments to the equalization and volume will make no difference to any attribute of sound. This affords the performers some small amount of control over the sounds that will be heard—a control deriving from their choice of the performance location and time—though they will still be at the mercy of the station managers, who get to decide what will broadcast by those stations that are actually be received at that time and place.
Performers of 4’33” are granted a similar amount of control owing to their selection of performance venue and time.28 In fact, performers of 4’33” arguably have more control over the sorts of sounds that will be heard during the performance, there being no station managers to consider. It looks as though 4’33” is at least as well positioned as Imaginary Landscape, No. 4 with respect to requirements on intentional organization.
In sum: Both pieces exhibit a directly intentional durational structure imposed on appropriated sounds. Both exhibit a highly indirect intentional control of (the attributes of) the appropriated sounds. Furthermore, the direct intentional durational structure of both pieces concerns the durations of groupings of appropriated sounds, not the appropriated sounds, themselves. These are the only aspects of the pieces that are relevant to their ability to satisfy the organized sound condition. Imaginary Landscape, No. 4 satisfies the condition. Therefore, 4’33” does too.
The arguments thus far show that 4’33” satisfies the organized sound condition. But there is reason to think that there are further criteria that instances of music need to satisfy.
The content of the organized sound condition is extremely liberal: It does not require more intentional organization than that which is found in 4’33”. And, given the tremendous variety within musical traditions of the world, it is far from certain that we can require even this much without ruling out some music.29 But a version of the organized sound condition that is liberal enough to apply to all instances of music will also be satisfied by much non-music. To use a well-worn example, Morse code will satisfy such a general organized sound condition. Hence the need for additional criteria in a definition of ‘music’.
Levinson (2011) augments the organized sound condition with an appeal to the purpose of the putative musical performance—namely, that it needs to be intended for the enrichment or intensification of experience (2011: 270–73). Levinson’s proposed purpose is not applicable to all music, though. Satie’s furniture music, for instance, is intended to function as unobtrusive background noise. If this enriches or intensifies experience, it is not through the usual channels and certainly not by “active engagement (e.g., listening, dancing, performing) with sounds regarded primarily, or in significant measure, as sounds” as the rest of Levinson’s definition requires (2011: 273). Indeed, furniture music is intended for inactive engagement, if it is intended for engagement at all.30
Similarly, in the Chinese literati qin tradition, the goal of playing the qin—a long, seven-string plucked zither, played in private or, occasionally, for a very small audience—is to promote one’s moral development, enlightenment, and/or transcendence (according to either Buddhist or Taoist doctrine). It is a stretch to include this under the enrichment/intensification of experience. Even if we allow it as such, this purpose is not pursued by active engagement with sounds regarded primarily/in significant measure as sounds:
[. . .] the literati maintained that it was more important to discern and contemplate the inner meaning and significance of music than to cultivate the outer manifestations of musical sound; the act of playing and the resulting musical sound were but vehicles toward these loftier ends. (Yung 2017: 509)31
The world’s musical traditions are simply too varied with respect to the function music plays within them, and the ways sound is used to perform those functions, to reasonably expect that there will be some variation on Levinson’s functional condition that will, in conjunction with an organized sound condition, pick out all and only music.
We see a similar problem with an additional constraint proposed by Kania: that the sounds of a musical work must either contain or be listened to for musical qualities (2010: 348).32 The problem lies in the interpretation of ‘musical qualities’. Avant garde and experimental artists explicitly seek to introduce innovations, which traditional-minded listeners often reject as ‘noise’. If we are taking a traditionalist stance on ‘musical qualities’, this will illicitly rule out such innovations, which are frequently accepted as music by subsequent generations of traditionalists. As McKeown-Green puts it, Kania’s definition is not ‘future-proof’ (2014: 395). But it also risks ruling out current musical traditions if those traditions include sounds not captured by or to be listened to for musical qualities as understood by the traditionalist. And if we take a broader stance that doesn’t rule out these innovations/traditions, the condition will capture things that are not music—for example, Morse code, given a sufficiently broad understanding of ‘rhythm’ as a musical quality (Davies 2012: 538). For these reasons, I think Kania’s condition should be rejected.
Davies (2012) is on surer footing. He augments the organized sound condition with a socio-historical constraint on which a piece/performance must bear the right sorts of cultural and historical connections to at least some musical tradition to count as music. The socio-historical constraint clearly avoids the problem—seen in the purportedly tradition-neutral requirement of Levinson—of ruling out entire musical traditions from the extension of ‘music’.33 It also allows us—when properly formulated—to evade the problem of innovation encountered by Kania’s definition (see below).
What exactly the relevant socio-historical factors are and how they are to be balanced against one another is a difficult question. So, too, is the question of how a musical tradition is established in the first place. But perhaps an appeal to organized sound can help here. We might say, for instance, that a tradition forms when there comes to be a social practice of organizing sounds but not primarily for conveying semantic content. The point is not to offer an unassailable version of the socio-historical constraint, but merely to show that it is plausible and flexible enough to meet the obvious challenges when combined with the organized sound condition.
Davies draws out an important feature of the socio-historical constraint:
But by contextualizing music historically, we may be able to see the presently diverse array of possibilities as emerging in a regular fashion [. . .] One can see this as change and development in the “language” that the composer had at his or her disposal at any given moment. Awareness of the direction of historical musical change might also allow us to generate more context-sensitive and perspicuous generative principles of musical construction. In effect, rather than trying to define music per se, we might do better by defining it as what could be created as music at a given time in a given tradition. (Davies 2012: 547)
When we focus—as Davies urges—on the organizational principles available to a tradition at a given time, we narrow the scope of the organized sound condition to a manageable extent. Rather than concerning ourselves with a general organized sound condition, we can focus on a more robust organized sound condition that need only apply to, and be conditioned by, instances of the tradition in question. Being explicit about the narrowed scope and tradition-boundedness of such a robust organized sound condition has significant theoretical payoffs.34
It helps us avoid cultural chauvinism: mistakenly taking our own tradition-bound interpretation of the organized sound condition as a tradition-independent interpretation of the condition and applying it to traditions for which it is irrelevant. Similarly, it helps us avoid parochialism: applying too narrow a conception of organized sound, based on the instances of music in a tradition with which one is familiar/comfortable, to instances that are properly part of the tradition but outside the basis of the too narrow conception of organized sound.
Parochialism is often linked to a confusion—encouraged by colloquial talk (‘That’s not music!’)—of evaluative standards for the organization of sounds with the conditions for membership within a particular tradition. Of course the identification of instances of a work in a particular tradition is closely connected to the evaluation of those works (as members of that tradition). Careful attention to the tradition-bound organized sound condition can help us to disentangle those requirements related to the identification of works as instances of the tradition and those further requirements that distinguish good instances of the tradition from the bad or mediocre—at least for those aspects of evaluation based on objective reasons related to how the work fulfills the requirements for being an instance of the tradition and how it relates to prior work in that tradition. By understanding changes to a tradition’s organized sound condition over time, we can learn about the musical values of that tradition, which can be inferred from answers to questions such as: Which deviations from tradition are accepted as innovations and which are rejected as bad music?
This helps us answer the concern that the organized sound + socio-historical condition approach is unable to accommodate future musical innovations because the acceptance of such innovations is highly conditioned by arbitrary cultural/historical forces.35 On my way of filling in the details of the interaction between the organized sound condition and the socio-historical condition, there is no need to appeal to arbitrary cultural or historical forces.36 Rather, we identify an aspect of the musical culture—how it handles innovations in the organized sound condition for that tradition—and assess putative changes to the tradition (including future innovations) in light of their conformity to that aspect of the culture. The focus on specific musical traditions makes this far more tractable: we are more likely to find clear social and cultural factors influencing the development and acceptance of music at the level of specific traditions than we are to find general, cross-culturally robust, factors—as should be clear from the discussion of qin music.37 In the next section I will illustrate how such an account of innovation within a tradition can be developed and applied in the case of Western art music.
In addition to its theoretical usefulness, I suspect that any further constraint augmenting the organized sound condition will, in fact, be subsumed under the socio-historical constraint. As with the organized sound condition, there are tradition-specific socio-historical constraints that determine how the general requirement that the work/performance bear the right sort of socio-historical connections to some tradition is to be satisfied for a particular tradition. These will include the conditions under which innovations are accepted as such by the tradition and (perhaps) the functions fulfilled by musical performances/works within the tradition (e.g., disinterested appreciation, improvement of the character, etc.). Putatively general constraints (e.g., Levinson’s functional condition) will only seem plausible as general constraints if we are not adequately attending to the great variety in musical cultures (due to chauvinism or parochialism).
The arguments of §1 can now be seen as arguments that the organized sound condition for Western art music will be satisfied by 4’33”.38 But versions of the organized sound condition for other traditions—or even more restricted subsets of Western art music—will not need to include 4’33”. This does a lot to explain reactions to 4’33”. Those that reject its status as music either apply a narrower conception of organized sound than they ought—for example, parochialists who reject the innovations of experimental composers—or they believe that 4’33” doesn’t have the right socio-historical connection to the Western art music tradition—for example, they think Cage was a fraud.
The first group is guilty of over-generalizing a too-narrow interpretation of the organized sound condition or of confusing evaluative standards (perhaps also over-generalized) for membership conditions. Both are clearly mistakes, provided that the correct interpretation of the (membership determining) organized sound condition for Western art music is as I have claimed it to be. If, however, Cage’s innovations don’t bear the right socio-historical connections to the Western art music tradition to count as innovations within that tradition, then my charge of over-generalization will fail. If that is correct, we cannot appeal to innovations introduced in Imaginary Landscape, No. 4 to vindicate 4’33”. Neither piece will count as Western art music. This just is the second objection, above—that 4’33” doesn’t satisfy the socio-historical constraint. In the remainder of §2, I will argue that 4’33”—and the innovations leading up to it—do bear the right socio-historical connections to the Western art music tradition.
So how are innovations in the scope of ‘organized sounds’ introduced in the Western art music tradition?39 Arnold Schoenberg, with whom Cage studied, provides an excellent illustration.
Schoenberg’s (1911/2011) Harmonielehre (Theory of Harmony) articulates a view of the evolution of harmony in Western art music that enriches our understanding of his own twelve-tone music. Schoenberg sees the increasing complexity of Western harmony as an attempt to approximate successively higher overtone relations in twelve-tone equal temperament. He sees this project as reaching its culmination with impressionist composers, such as Debussy—twelve tone equal temperament is not fine-grained enough to approximate overtones higher than those implied by impressionist harmonies.
The overtone series, which led our ear to it, still contains many problems that will have to be faced. And if for the time being we still manage to escape those problems, it is due to little else than a compromise between the natural intervals and our inability to use them—that compromise which we call the tempered system, which amounts to an indefinitely extended truce. This reduction of the natural relations to manageable ones cannot permanently impede the evolution of music; and the ear will have to attack the problems, because it is so disposed. (Schoenberg 1911: 25)
Schoenberg goes on to propose more finely divided equal-tempered scales and also non-tempered scales (scales conforming to the intervals found in the overtone series). But he claims that, at the time he was writing, the available instruments were insufficient for the performance of music written in such systems.40 Schoenberg’s response to this situation was to abandon the approximation of overtone relations and to fully embrace twelve-tone equal temperament. This required abandoning the hierarchical relations among pitches found in tonal harmony (and the overtone series). But Western harmony had been the primary basis of large-scale structural organization of musical works. So Schoenberg needed new organizational principles.
His solution was the twelve-tone row—an ordered list in which each of the twelve equal tempered pitch classes (a pitch and all its octave transpositions) appears once. Local organization of the work was based on the various forms of the row: the prime (original) form, its retrograde (reverse), its inversion,41 its retrograde inversion, and the transpositions of each. While working through a row form, the composer cannot return to a pitch class in the row, after leaving it, until all the other 11 pitch classes have been used. However, multiple forms of the row can be used at any given time. Large-scale musical structure was based on relationships between the row forms used in the work.
Whatever one’s feelings about Schoenberg’s serial music with respect to other aspects of aesthetic evaluation, the integrity of the approach and the cleverness with which Schoenberg addressed the “problems” posed by the overtone series and the supposed limitations of our instruments cannot be denied. And these are precisely the features that we value with respect to innovations introduced into the form. Furthermore, Schoenberg’s achievements are widely accepted as musical achievements by the institutions of Western art music (concert halls, symphony orchestras, conservatories, etc.). Indeed his serial approach came to dominate avant garde and academic composition in the United States for much of the 20th century. Schoenberg himself taught at UCLA and USC, where Cage studied with him.42
Nor is this pattern of reasoning, with respect to innovative approaches to organizing sound, unique to Schoenberg. Similar experiments were being conducted by Charles Seeger and Carl Ruggles—approximate contemporaries of Schoenberg—in their formulation of ‘dissonant counterpoint’. But one can also see this sort of reasoning in earlier Western music. We see it, for instance, in the shift from the modal polyphony of the Renaissance to the major/minor tonality and the use of harmony as a structural principle in Western music in the 18th century—formalized by Rameau in his Traite de L’Harmonie (Treatise on Harmony, 1722/1971) published in 1722—and in the elaborations of the sonata-allegro form of the classical symphony by Beethoven and his followers in the romantic era. In all these cases, new ways of organizing music were introduced to accommodate shifts in the permissible materials of the tradition at the time of the innovations (and, in some cases, to perceived dead ends with respect to traditional methods/materials).
Furthermore, the emancipation of dissonance advanced by Schoenberg led to further innovations with respect to the sounds permitted within the musical context. The music of the 20th century French-American composer Edgard Varèse provides a notable example of this. Varèse introduced a number of new (non-tonal) sonorities and unusual sounds in his work (including, famously, a siren in his Ameriques [1921]), which had a profound influence on Cage’s early percussion music.43 Other influences in this regard include Henry Cowell’s music and his text, New Musical Resources (1930), which introduced a number of so-called extended techniques: non-traditional techniques applied to traditional musical instruments to create unusual sounds. Cowell’s experiments with inside-the-piano techniques, as heard in his Banshee (1925), were an influence on Cage’s own Sonatas and Interludes for Prepared Piano (1946–1948), a tour de force of extended technique composition calling for pencil erasers, nuts, bolts, and weather stripping to be inserted between the strings of the piano, turning the instrument into a one-player percussion orchestra. This laid the groundwork for Cage’s further incorporation of non-traditional sounds—for example, his use of sounds appropriated from radio broadcasts in Imaginary Landscape, No. 4.
Cage’s use of these expanded sonic resources influenced the development of his attitudes towards musical organization:
I wanted to find a way of making music that was free of the theory of harmony, of tonality; and so I had to find a way of composing with noise. And I came to the conclusion that the important aspect, or as we would say in the twelve-tone language, the important parameter of sound, is not frequency but rather duration, because duration is open to noise, as well as to what has been called musical. When you take such a point of view, you can shift your allegiance, as it were to Varèse [. . .]. (Kostelanetz 2003: 54)
Cage’s desire to incorporate noises into his music led him to seek new structuring principles, based in duration, that could accommodate both pitched and unpitched sounds. This carries forward the innovations introduced by Varèse with an approach that mirrors Schoenberg’s systematic introduction of new structuring principles in his twelve-tone method.44
Notice that Cage’s search for organizing principles was a search for organization in terms of one or another of the aspects of sounds (pitch, duration, timbre, intensity). In this, he is wholly within the mainstream of Western art music. He is simply reacting to the dominant practice of using pitch—and particularly harmony—as the primary basis of organization in that tradition. Due to his interest in noise, he chooses duration as the primary basis of organization. (Schoenberg’s serial music, by contrast, rejects traditional harmony but is still organized on the basis of pitch.)
The duration-based approach to musical organization was already manifest in Cage’s primarily percussion-based works of the 1940s. Importantly, Cage would link his focus on the durational aspect of sound with his interest in silence as well:
I thought, if I’m not going to have tonality in my music, I’ll need something to make an alternative structure; and that was rhythm. I examined the nature of sound, which has pitch, duration, overtone, and amplitude. Then I examined silence; and of those four things, silence had only duration. (Kostelanetz 2003: 64)
In 4’33” this manifests as a work whose only directly intentionally organized features are durations. However, this durational approach is not the only innovation introduced by Cage in the lead up to 4’33”. In the mid-1940s, Cage began a deep engagement with Zen Buddhism, which would lead to significant changes in his compositional outlook. The progression of his thought is captured in his lecture ‘Composition as Process’ (1958):
Drawing a straight line between this situation and that presented by the later work, the deduction might be made that there is a tendency in my composition means away from ideas of order towards no ideas of order. And though when examined the history would probably not read as a straight line, recent works, beginning with Music of Changes [Cage’s first work using chance procedures, composed in 1951 for solo piano], support the accuracy of this deduction [. . .] For what happened [in Music of Changes] came about only through the tossing of coins. It became clear, therefore, I repeat, that structure was not necessary. (Cage 1961: 20–22)
Here Cage describes a turn away from organization in his music. In particular, Cage was turning away from a robust view of direct intentional internal organization of his works, introducing non-intentional means of selecting materials at some level or other.45 In Music of Changes, Cage generated tables, each containing 64 options (corresponding to hexagrams of the i Ching), for the number of events occurring in a given structural unit (one chart), tempo of the given structural unit (one chart), durations of individual events (eight charts), sounds (eight charts), and dynamics (eight charts). Cage determined which of the 64 elements on a given chart was selected, at a given point in the compositional process, by flipping three coins six times to generate a six-line hexagram according to instructions in the i Ching. (There are 64 hexagrams, each of which is assigned to one of the 64 positions on the chart.) 32 of the elements on each of the eight sound charts are silences. For the eight-chart parameters, coin tosses determined which chart was used for a given structural unit (Cage 1961: 58). The determinations resulting from all these coin flips were then transcribed into the score, in standard notation.
The description should make clear: There is a highest level of organization that is intentionally chosen by the composer. This organization is only organization in terms of duration. (Even here the intention is somewhat undercut by the use of chance procedures to determine tempos.) Within each unit, the elements (sounds and silences) are organized using chance procedures. The specific values for the sonic parameters (duration, pitch, timbre, loudness) were not directly intentionally chosen. They were determined by procedures chosen by the composer, the outcomes of which are highly unpredictable, given the diversity of elements on the charts to which the procedures are applied.
What goes on each chart is intentionally chosen to generate certain effects (e.g., density of sound is controlled by making half the sound elements silences, all twelve pitch classes of the equal-tempered chromatic scale are more or less equally represented). Furthermore, the resulting score is traditionally notated and is engaged with by the performer just as any other piece of traditional notation would be—by following the specifications so as to produce the sounds/perform the actions as indicated. So, while Music of Changes involves indirect intentional organization of sounds on the part of the composer, it retains the usual amount of direct intentional internal organization of sounds on the part of the performer. There is a particular sonic structure indicated in the score that the performer intends to accurately render.
It is worth noting that both the highly systematized approaches to composition and the use of chance procedures evident in Music of Changes have precedents within Western art music and so are not disqualifying viz. the socio-historical constraint. Regarding systematized approaches, we can cite Schoenberg’s serialism and the total serialism of the mid-20th century (e.g., Pierre Boulez’s Structure 1A [1951]), which applies the serial principles developed by Schoenberg to deal with pitch to other musical parameters as well (duration, intensity, timbre, articulation). Regarding chance procedures, we can point to the musical dice games of the 19th century in which precomposed materials were pieced together according to the results of dice rolls, which bear an obvious resemblance to Cage’s use of coin flips and charts of musical indications.
The compositional procedures for Imaginary Landscape, No. 4. (also from 1951), are almost identical to those used for Music of Changes, with modifications to accommodate the fact that this work is written for 12 radios rather than solo piano. The sound charts contain indications for the radio settings to be used in the 32-non-silent places rather than containing single pitches and aggregates. Though Cage is still responsible for creating the charts, the performances derived therefrom are highly variable and depend on the quality of the reception, which stations carry programming in the vicinity of the performance, and what they happen to be airing during the performance. As a result a performance might contain static, silence, speaking, a symphony, or any other sounds on any of the active radios at a given moment. The timbre and loudness of any sounds heard are impacted, but not wholly determined, by indicated adjustments to the equalizer and volume dials. In contrast to Music of Changes, the performers of Imaginary Landscape, No. 4 do not aim at producing particular sounds indicated in the score. Rather, they perform actions that do not directly determine which sounds result (see §1.3.2). So there is only indirect intentional organization of the sounds by the performers.
This demonstrates a progressive relinquishing of control over the internal organization of the work—from the works of the 1940s to Music of Changes to Imaginary Landscape, No. 4—coupled with a retention of Cage’s use of duration as the primary means of intentional organization. 4’33” is the logical next step in this progression. The thinking that leads from Imaginary Landscape, No. 4 to 4’33” is no more radical a departure from precedent than is that which leads from Music of Changes to Imaginary Landscape, No. 4. And neither is any more radical a departure from the thinking leading to Music of Changes from its predecessors. Yet no one has denied that Music of Changes is (Western art) music—nor should they. So it looks like 4’33” and Imaginary Landscape, No. 4 are on solid footing with respect to socio-historical constraints on admissible innovations in the Western art music tradition. Therefore, the innovations to the organized sound condition for Western art music introduced by Imaginary Landscape, No. 4 are legitimate. But we already saw that, if Imaginary Landscape, No. 4 satisfies the organized sound condition, so does 4’33”. Since the organized sound condition and the socio-historical constraint are the only plausible constraints that must be met for a work to be a piece of music, 4’33” is music.
Following Davies, I have suggested that specifications of the content of the organized sound constraint ought to be relativized to a particular musical tradition and time. The discussion of 4’33” offers an illustrative example of how innovations can alter the content of the condition, for a given tradition, over time. In the case of 4’33” the tradition is Western art music. Of course, the history of Western art music did not stop in 1952, with the composition of 4’33”. Other innovative works have raised different questions about the content of the organized sound condition. Unsurprisingly, Cage composed many such works. So there remains work to be done if we are to articulate the current content of the organized sound condition for Western art music.
Take, for instance, the score to Cage’s 0’00” (1962), which reads: “In a situation provided with maximum amplification (no feedback), perform a disciplined action.” (See Figure 3 for the full score, which adds a few restrictions on the disciplined action.) No specification of the sound (or even number of sounds) to be produced is provided. Intensity is given some indirect intentional organization, via “maximum amplification (no feedback)”, akin to that seen in a given durational unit of Imaginary Landscape, No. 4. But notice that Cage provides no directions regarding internal structure—even for intensity—though some internal structure might be evident (depending on the ‘disciplined action’ performed and whether or not it includes interruptions). This contrasts with 4’33” in that 4’33” contains an intentional internal durational structure (i.e., the durations of its three movements).46 If 0’00” is a musical work, there needn’t be any intentional internal structure indicated by the composer.
Of course, the performer of 0’00” selects the disciplined action and, in doing so, will—provided it is a complex action—introduce some intentionally organized internal structure, though it won’t necessarily be audible structure. So perhaps the composer and performer are on different footing with respect to the internal organization requirement. Or perhaps performances of the work are musical performances though the work, itself, is not music because the score provides no internal structure but the performances do. To answer these questions, we have to trace the context leading up to the composition of 0’00” to see what Cage (and other experimental composers) were up to in the ten years since the composition of 4’33”. Were there any innovations that bear on the question of the requirement of internal structure provided by the composer? Or is the fact that directly intentional internal structure (such as the three movements of 4’33”) needn’t be audible enough to justify classifying a work lacking any composer-imposed internal durational structure as organized sound, in the relevant sense?
There are also questions concerning the present content of the organized sound condition for Western art music that did not need to be addressed in the consideration of 4’33” but were suggested by it. Most conspicuously, I did not answer the question of whether or not a musical work can be wholly silent. Given that internal structure needn’t be audible, this question comes apart from the question of internal organization itself. And, given the changeability of the organized sound condition over time, we have to clarify the question. We can ask if a musical work can be wholly silent, given the current content of the organized sound condition (for Western art music) or we can ask if a wholly silent work could, at the present time, be accepted as an innovation leading to a change in the organized sound condition allowing wholly silent works.47
Consider, for instance, a variant of 4’33”—4’33” in Space—which is just like 4’33” except that it is to be performed in space such that no ambient sounds are present.48 Performances of the work will be truly silent. Nevertheless the work does organize its silences into three distinct movements. Whether or not this is sufficient to satisfy the organized sound condition is an open question, though I think the burden is on those who think it cannot, given the fact that silences are part of the materials organized by composers—on an equal footing with (non-silent) sounds (see §1.1)—and the fact that the intentional internal organization need not be audibly demarcated.
Now consider Kania’s Composition 2009 #3—designed to provide an example of a truly silent work. (Kania agrees that 4’33” is not silent.) The score reads, in full: “Indicate a length of silence, using the usual cues with which you would signal the beginning & end of a single movement, song, etc. (The content of this work is the silence you frame, not any ambient noise.)” (Kania 2010: 351). If Kania is correct, his piece is music and so the organized sound condition cannot require even inaudible internal organization by performer or composer (unlike 4’33” and 0’00”). Whether or not he is correct will depend on how one connects his composition to the Western art music tradition in light of the socio-historical constraint.49
And still other questions remain. To give a few: If internal organization is required, must it be intentional—for example, does Dodd’s Simple Imaginary Landscape satisfy the organized sound condition (despite Dodd’s claims to the contrary), given innovations in experimental music since Cage? Does internal organization require some kind of perceivability—for example, would it undermine 4’33”s status as music if the performer didn’t visibly mark the boundaries between movements? Does it require knowability of the internal organization? Can we ever have unintentional external organization delimiting the boundaries of the piece? Can attentive listening provide the requisite organization (see n.52, below)? How does the organized sound condition interact with Cage’s eventual definition of music as ‘sounds heard’?
To answer these questions—and similar questions for other musical traditions—we will need some tools for unpacking the organized sound condition at a given time for a tradition. The simpler aspect of the condition, which sounds are permitted, can usually be sorted out by seeing which musical instruments are employed within the tradition and what the traditional techniques for performing on these instruments are. In this respect Western art music is extremely permissive: We have already referenced extended techniques (the production of unusual sounds by unusual techniques applied to traditional instruments), but Western art music has also incorporated pre-recorded sound and digital sound synthesis. In this situation, any sound that can be produced or recorded can be used. However, there may still be restrictions on how these sounds can be used.
Figure 4 offers a schematic of the types of organization possible in a musical work. For each branch on the diagram, we can ask whether or not the sort of organization represented by that branch is a requirement of the organized sound condition, as currently constituted. The first branches correspond to the hierarchical level of the relevant organization: external organization (the offsetting of the experience of the musical materials from other (auditory) experience) is at the highest level and nested levels of internal organization (the structural organization of the musical materials within the piece/performance) are at lower levels.
An example of nested internal hierarchical organization can be seen in the harmonic organization of Western classical and romantic music, against which Schoenberg was reacting. Take the traditional symphony—a four movement work. The harmonic relations of the keys of the movements is one aspect of the internal organization of the work (typically, the first and last movement are in the same key and the second and third are in closely related keys). At the next level down, we get internal harmonic organization within a movement. For instance, the opening movement is usually divided into a two-part exposition, the first part in the tonic key, the second in the dominant key (a fifth above the tonic). This is followed by the development, which begins on the dominant key but moves through several different key areas before returning to the tonic for the recapitulation—a return to the exposition but now with both parts in the tonic key. At a lower level—say at the level of one part of the exposition—we have more localized harmonic organization supporting the theme of that portion of the exposition.
Within each hierarchical level of organization, there are various materials that can be organized, corresponding to the aspects of sound: pitch, timbre, intensity, and duration. The harmonic organization just discussed is one dimension of organization by pitch. Melody is another. Timbre is typically organized by assigning instruments to specific pitches/melodic lines. The organization of each of these parameters may be provided by the composer (if there is one), the performer(s), or even—potentially—the listener, as indicated in the next level down from the aspects of sound organized in Figure. 4. For each of these potential organizers, the relevant organization may be intentional (as it was for Beethoven in notating his Symphony, No. 5) or it may be unintentional (as it was for Cage with respect to the attributes of the ambient sounds appropriated in a performance of 4’33”). The discussion of 4’33” has shown that intention comes in degrees—the performer of 4’33” might exert some degree of control in the selection of the venue and time of the performance, though this falls far short of the sort of control evinced by Beethoven’s notational choices. So, we might want to think of the distinct boxes for ‘intentional’ and ‘unintentional’ as marking extremes on a continuum, though nothing will hang on that here.
In specifying the content of the organized sound condition for Western art music—or any other tradition—we can begin at the top of the diagram and move downward for each hierarchical level of organization, asking if organization at that level is required by the body of works under consideration as representative samples of the tradition and, if so, of which aspects of sound (or what number of aspects), by whom, and whether the organization must be intentional (and to what degree). This is, essentially, the procedure I have followed in answering the questions posed about the organized sound condition and their implications for 4’33” in §1. I will now briefly recapitulate these results, in the terms introduced in this section, to illustrate (in outline) the systematic examination of a socio-historically situated organized sound condition I am recommending here. This sets up a discussion about the content of the organized sound condition for Western art music in the wake of 4’33” which occupies the remainder of the section.
Concerning the requirement for external organization—the setting apart of a work’s sounds from sounds not included in the performance—the discussion of 4’33” has shown that the performer’s actions can organize sounds that she does not (intentionally) produce—for example, ambient noises.50 In 4’33” these sounds are set apart from any that the performer intentionally produces, which are not a part of the work.51 Therefore, with respect to external organization, the organized sound condition can require, at most, that the performer set apart the sounds of the work by some attention-directing act of framing.
With respect to a requirement for external organization on the part of the composer: Typically, in Western art music, the score delimits which sounds count as part of the work; the sounds resulting from the indicated actions of the score are set apart from those not resulting from the indicated actions. Those sonic attributes that are determined by the indicated actions are intentionally set apart from those (attributes of) sounds not resulting from the indicated actions; those left indeterminate by the indicated actions, but nevertheless resulting from them (e.g., when a composer gives no indications regarding volume), are unintentionally set apart. The discussion of 4’33” (and Cardew’s Treatise) has shown that we need a fairly liberal understanding of ‘the sounds resulting from the indicated actions’—one that includes sounds to which the audience’s attention is directed by the performer.
Once we turn to internal organization, the assessment of the organized sound condition in light of 4’33”’s innovations closely tracks the discussion of §1. There we saw that, at best we can require direct intentional organization, by the composer, of duration—though not necessarily the durations of individual sounds, themselves, but rather the durations of actions of the performer by which the sounds are produced or framed. No other elements are subject to this requirement (if there is such a requirement). Nor are they subject to a requirement that they be directly intentionally organized by the performer. (Recall the contrast between Music of Changes and Imaginary Landscape, No. 4 in this regard.) The performer of Imaginary Landscape, No. 4 or 4’33” is not structuring any attributes of the sounds within each of the directly intentionally organized durations. That is left to chance. So there is a lower hierarchical level at which none of the attributes of the sounds are being intentionally organized.52
Therefore, the most stringent form of the organized sound condition for Western art music, in the wake of 4’33”, can require external organization by both composer and performer and intentional internal durational organization by both composer and performer. It might also require unintentional—or indirectly intentional—organization of other sonic attributes at lower hierarchical levels of internal structure. If the requirement is for indirectly intentional organization, it remains an open question how indirect this can be (though it will need to be pretty indirect). And what of innovations introduced since 4’33”?
If, for example, 0’00” satisfies the historical constraint—and I see no reason why it can’t—and thereby counts as introducing a revision in the organized sound condition for Western art music, then the condition will not require intentional internal organization, by the composer, of any sonic parameter. It might still require indirect intentional external organization of intensity, given the call for “maximum amplification”. No other attributes of sound are in any way intentionally externally organized by the composer, though the score will call for actions that produce sounds and thereby provide (unintentional) external structure. So, the condition could still require unintentional external organization by the composer. 0’00” does not challenge a requirement on intentional organization (internal or external) by the performer.53
A similar analysis can be offered for other putative innovative works of Western art music. Perhaps once we have done this, we will find that the most stringent version of the organized sound condition for Western art music is no more demanding than Cage’s (post-4’33”) claim that “sounds one hears are music” (Cage 1967: 163). On the most liberal interpretation of this statement, absolutely any sound heard counts as music, and the mere fact of hearing is sufficient to impart the requisite organization (which is, in this case, merely external organization by the listener): hearing the sounds sets them apart from sounds not heard.54 This is, moreover, an example of unintentional external organization by the listener: any sound heard is music regardless of the listener’s intention. More likely, Cage had in mind the view that music just is those sounds that one listens to attentively. In this case, sounds are externally organized in virtue of being attended to by the listener and nothing more. On this view, one could choose to attend, for some duration, to as many sounds heard in that duration as one can and, in doing so, impart the necessary external organization for the sounds heard to count as music.55 Alternatively, sounds might be selected for attention due to some feature of the sounds themselves—for example, one might choose to attend to all the quiet sounds in one’s environment for a particular duration. The remaining features of the sounds attended to would be unintentionally organized.
The point is that we can only answer these questions in light of the innovations introduced into the Western art music tradition since 4’33”. If these innovations bear the right socio-historical connections to that tradition, then they will count as introducing a revision in the content of the organized sound condition for that tradition. Furthermore, we can undertake this sort of analysis for any musical tradition, identifying the content of the organized sound condition for that tradition at a given time, thereby deepening our understanding of the music (and musical values) of that tradition. For whatever traditions we choose to study, though, this will require some musicological due diligence. We cannot simply specify the content of organized sound condition on the basis of our intuitions. That approach quickly leads to chauvinism or parochialism.
I have argued that 4’33” satisfies the organized sound condition on (Western art) music. Furthermore, 4’33” satisfies the socio-historical constraint, on which a work must bear the right sort of relationship to some musical tradition to count as music. If I am right that the socio-historical constraint is the only other condition on music, then 4’33” is music. Along the way I have shown how the organized sound constraint and socio-historical constraint interact, particularly with respect to innovation within specific musical traditions, and how we can approach an analysis of the organized sound condition for a given tradition and time.
It is my hope that this will lay the groundwork for a more fruitful interaction between philosophers of music, (ethno)musicologists, and music theorists. For philosophers, understanding the technical innovations introduced into a musical tradition, the historical context, and social values ascribed to music within that tradition will clarify ontological and evaluative questions concerning music—and it will do so in a way that can transcend the Western art music-inspired conceptualization of music that has historically dominated the philosophy of music. For music theorists and musicologists, philosophical advances on the evaluation of innovations will help clarify the value of particular musical interventions beyond what can be inferred from historical influence. It might even help composers and performers pinpoint new avenues for musical exploration.
Thanks to Janet Levin, James Van Cleve, Paul Garofalo, and especially Andrew Stewart for their perceptive comments on various drafts of this paper. Thanks also to the editors and referees at Ergo for their astute suggestions.
1 Cage, John (1960). 4’33”. Henmar Press.
2 Cage, John (1961). Silence: Lectures and Writings. Wesleyan University Press.
3 Cage, John (1962). 0’00”. Henmar Press.
4 Cage, John (1967). A Year from Monday: New Lectures and Writings. Wesleyan University Press.
5 Cowell, Henry (1930). New Musical Resources. Alfred A. Knopf.
6 Davies, Stephen (1997). John Cage’s 4’33”: Is it Music? Australasian Journal of Philosophy, 75(4), 448–62.
7 Davies, Stephen (2012). On Defining Music. The Monist, 95(4), 535–55.
8 Dodd, Julian (2018). What 4’33” Is. Australasian Journal of Philosophy, 96(4), 629–41.
9 Gann, Kyle (2010). No Such Thing as Silence: John Cage’s 4’33”. Yale University Press.
10 Goldsmith, Steven (1983). The Readymades of Marcel Duchamp: The Ambiguities of an Aesthetic Revolution. The Journal of Aesthetics and Art Criticism, 42(2), 197–208.
11 Kania, Andrew (2010). Silent Music. The Journal of Aesthetics and Art Criticism, 68(4), 343–53.
12 Kania, Andrew (2011). Definition. In Theodore Gracyk and Andrew Kania (Eds.), Routledge Companion to Philosophy and Music (3–13). Routledge.
13 Kania, Andrew (2020). Philosophy of Western Music: A Contemporary Introduction. Routledge.
14 Kostelanetz, Richard (Ed.) (2003). Conversing with Cage (2nd ed.). Routledge.
15 Levinson, Jerrold (2011). Music, Art, and Metaphysics. Oxford University Press. (Original work published 1990)
16 McKeown-Green, Jonathan (2014). What Is Music? Is There a Definitive Answer? The Journal of Aesthetics and Art Criticism, 72(4), 393–403.
17 Nyman, Michael (1974). Experimental Music. Schirmer.
18 Rameau, Jean-Philippe (1971). Treatise on Harmony (P. Gossett, Trans.). Dover. (Original work published 1722)
19 Ravasio, Matteo (2019). What 4’33” Also Is: A Response to Dodd. Australasian Journal of Philosophy, 97(2), 395–400.
20 Schoenberg, Arnold (2011). Theory of Harmony (R. E. Carter, Trans.). University of California Press. (Original work published 1911)
21 Shambroom, Donald (2000). Leonardo’s Optics through the Eyes of Duchamp: A Note on the Small Glass. Tout-fait, 1(2). https://www.toutfait.com/leonardos-optics-through-the-eyes-of-duchamp-a-note-on-the-small-glass/
22 Sorensen, Roy (2008). Seeing Dark Things. Oxford University Press.
23 Varèse, Edgard and Chou Wen-Chung (1966). The Liberation of Sound. Perspectives of New Music, 5(1), 11–19.
24 Yung, Bell (2017). An Audience of One: The Private Music of the Chinese Literati. Ethnomusicology, 61(3), 506–39.