It is at last clear that Schoenberg’s error was to ignore the inherently hierarchical nature, not just of Western tonality, but of pitched sound itself. It is hierarchy that makes the lower relations among overtones perceived as “consonant,” while the upper partials are heard as “dissonant.”¹ It is hierarchy that puts a melodic line in the foreground and its accompaniment in the background. It is hierarchy that links pitch to duration, making it impossible, outside of pure abstraction, to consider a pitched sound separate from how long it endures in time. It is possible, of course, simply to assert outright that hierarchy is of no importance. It is also possible, in designing aircraft, to assert the notion that airfoil is an arbitrary construct, and to design aircraft that ignore the principles of airfoil, and to contemplate these designs in purely abstract fashion. But if these designs were materially realized, the resultant aircraft would not fly. Similarly, music designed along principles that contradict the ontic realities of pitch hierarchy and the perceptual requirements of human psychology can be contemplated abstractly, but the sounds thus produced will not “fly” – that is, it will not be perceived as music, and if it is not perceived as music, it is quite possible that it is not music. This observation is controversial only because its application is potentially conservative.

Consider the following three sets of English-language words:

My cat Luna has thick black fur and a cold wet nose.

Has cat a fur and thick Luna nose wet black my cold

Luna My fur has thick black cat nose and cold wet a

The first group of words comprise a standard English sentence conveying information about a cat with the proper name, “Luna.” From this we know that Luna belongs to me, and that she has black fur and a nose that is cold and wet. The second arrangement of the same twelve words has been deliberately re-ordered so as create zero-to-minimal syntactical sense. It is a word-row, devised with the end in mind of treating each word as a thing-unto-itself. Syntactical relations between any two consecutive words were avoided, though “thick/Luna/nose” come dangerously close to forming a semantic trio. The third set was ordered via a process of tossing a coin (a quarter). I divided the initial sentence into two parts of six words each: “my cat Luna has thick black” and “fur and a cold wet nose.” I flipped a coin to determine which half the first word in the new sequence would come from: heads for the first set and tails for the second. I narrowed the choice down further with each successive flip until the sequence reproduced here emerged.

It is fair to say that the second and third sets are very much alike, as neither can be construed as exhibiting syntax of any sort, and therefore neither conveys information. An English speaker reading the initial sentence would have no trouble connecting noun to verb to modifier, etc. But in both the second and third sets, while the reader will recognize parts of speech – “cat” as a noun, “my” as a possessive, etc. – the words cannot, because of their lack of syntactical order, be made to fit into any syntactical scheme. What Schoenberg did with pitches was similar to what I have done with the second set of words above. True, Schoenberg chose his rows from among a chromatic scale that was already, in and of itself, a flat collection of materials, but his dodecaphonic/serial method took special care that this flatness should be maintained as the dozen were deployed in the non-sequential order called a series or row. This does not occur outside a specific set of imposed rules. If, for example, I were to choose from among the twelve chromatic pitches, without reference to any rules, a series of pitches using only my ear as guide, there is a good likelihood that the same intuition which makes humans sing certain groups of notes together but not others would lead me to pick tones that suggested a triad, perhaps, or pyramiding fifths, or some other grouping related to the overtone series. Even within the constraint of having to play all twelve tones before playing any of them twice, it is possible to choose pitches that suggest tonal hierarchy, which is why Schoenberg cautioned serial composers to avoid thirds and sixths in their rows, and how Alban Berg, Schoenberg’s student, managed to compose the only serial works that have entered the repertoire precisely because they contradict that edict. (The row governing Berg’s Violin Concerto is so tonal-friendly that it allows the interpolation of a Bach Chorale.)

In the third set, the method has changed, but the result would be equally frustrating for any English-speaker trying to make sense of it. The process of choosing words at random, using a decision-generating device like a coin flip, would seem to be the opposite strategy from carefully selecting words according to a set of rules. And yet, the results are surprisingly similar. Written words are visual representations of the sounds in spoken language, which must, to be language, incorporate syntax. When they are used outside that definition, it is arguable that they are no longer words-as-such. A coffee mug used to plant a violet is a flower pot. Mathematical symbols used to adorn a shower curtain without regard to their function are decoration. And words arranged as patterns – randomly or consciously designed – are, perhaps, objects for some kind of contemplation. But they are not words. There is nothing at all wrong, of course, in converting a coffee mug to use as a flower pot, and mathematical symbols may look elegant on a shower curtain. At issue is the idea of continuing to call the mug a mug and, even more pertinently, the claim of doing mathematics by arranging them for strictly visual purposes. If putting cosigns and tangents willy-nilly on a piece of cloth is not mathematics, how is it that playing notes without regard to their function in music is music? All serialist and most aleatory composers employ pitches that were originally intended to form musical scales, the salient feature of which is the tension created by the hierarchy of relative weights given to those pitches (the tonic, the dominant, the subdominant, and so on down the line), to the end of creating sonic patterns that ignore the hierarchical pitch arrangement that originated the pitches-as-such. (This should remind us of Wagner’s use of tonality to suggest that tonality is invalid, mentioned earlier.) It would seem, then, that serial and aleatory musics are more accurately called “anti-music.” But again, this approach is controversial, as it results in a conservative, even reactionary aesthetic. If we are allowed to say that serialism is not “really” music, why not “that hip-hop crap ain’t music”? This is the usual form taken by the fear that too-precise a definition of an art will narrowly restrict artistic freedom.

Our third example above is there to illustrate the eventual product of Schoenberg’s thesis: aleatory music. The two processes – Schoenberg’s careful plotting of tone-rows and the tossing of coins associated with John Cage – could not, prima facie, be further apart. Yet they each result in the use of pitched sound as something other than pitched sound-as-such, just as the second and third arrangements of words in the verbal examples above used words other than as words. Cage acknowledged Schoenberg as his master, and with good reason. A brief historical sketch of Western art music post-Schoenberg to Cage illustrates the relationship:

Serialism reached its peak in the 1940s and ‘50s, as the music of Anton Webern, Schoenberg’s pupil, achieved a startling level of chic. Every composer who wanted to be au courant tossed aside Stravinskian neoclassicism for the sonic pointillism of Webern – even Stravinsky, whose “conversion” to serialism in the late 1950s was taken as the final fall of all musics but the serial.

By the time of Stravinsky’s “conversion,” plain serialism was already old hat. In 1951, a young Parisian named Pierre Boulez had proclaimed “Schoenberg is dead” (he had in fact just died, his death possibly linked to an anxiety attack over the digits of his age [76] that year – 7 and 6 – adding up to 13; Schoenberg, the inventor of 12-tone music, suffered from extreme triskaidekaphobia) and proceeded to move “beyond” serial procedures. Boulez’s advance took the form, however, not of rejecting serialism, but of expanding it to cover all aspects of music. Serialism ordered only the twelve pitch classes, but Boulez’s “integral serialism” commandeered the ordering of music’s other parameters as well. For instance, a dynamic “row” might be assembled thusly: f, ppp, mp, ff, p, mf, fff, etc.. So, if the piece being composed began forte (f), its next dynamic would have to be pianississimo (ppp), and the next one mezzo-piano (mp), and so on. A score composed for, say, violin, trumpet, vibraphone and double bass would likewise have the pitches of its tone-row performed by a parallel series of timbres: If the bowed violin played the first note of the series, then the second note might be played by the trumpet, the third by the double bass pizzicato, the fourth by the violin pizzicato, the fifth by the vibraphone, the sixth by the double bass bowed, etc. And of course each one of these was at a different dynamic as per the dynamic row we have described. So if the first hexachord of the pitch row was, say, D–A#–A–D#–B–C#, the dynamic row might be f-mp-ff-pp-ppp-fff, and the timbre row violin pizz./trumpet/vibraphone/double bass bowed/violin bowed/double bass pizz.

This is a marvelous game and entertaining for the composer who plays it. The result can yield an idea or two that might not have occurred to the composer in the normal (non-serial) process of composing music. But its expressive potential is extremely narrow – in fact, the whole point of the process is to turn the usual (in Western terms) expressive voice of the composer over to the process itself. The composer chooses the various rows, but then the interlocking rows are left on their own to manufacture the relationships (in the new, positivist, Schoenbergian sense) of one pitch/timbre/dynamic to another pitch/timbre/dynamic. Control no longer belongs to the composer, who is but a sort of manager of the process.

But our example is missing one important factor. For, just as dodecaphony abolished tonal pitch relations, integral serialism abolished the beat. Analogous to Wagner’s blurring of key centers, various composers in the early 20th century had ruptured the familiar beat patterns of two, three, and four counts with complex patterns that shifted, perhaps, from five to two to seven to one-and-a-half to nine. The beat was “all over the place.” Why not simply get rid of it? The positive aspect of rhythm is simple duration. Therefore, it’s possible to posit durations as related only to each other, and not to the hierarchic arrangement of metered beats. For our set of pitches-with-timbres-and-dynamics above, let’s add durations in the mode of fully integral serialism. I choose for my duration series: an eighth note, a half note, a dotted eighth-note, a quarter-note, a whole note, and a sixteenth note. This actually leaves some room for personal decision-making, as the durations may overlap.

Integral serialism’s aesthetic domination was brief and its pure practitioners were few. Boulez and the American Milton Babbitt were its most prominent figures, and a few years in the 1950s were its heyday. Waiting in the wings was a young American musical artist grown unsatisfied with his own innovations, which were primarily related to timbre. John Cage’s invention of the prepared piano – a piano with erasers, bolts, wooden dowels, etc., stuck between its strings – made possible the presentation of varied colors on an acoustic piano that normally would have required electronics. Cage achieved a certain fame with this and with his whimsical early compositions, but he was restless to push back what he felt was a Western aesthetic bound up with ego and with fake “self-expression.” Cage wanted the subject out of the picture altogether, and when he encountered integral serialism, the path became clear. In integral serialism, the order of things is pre-determined by rows set in place by the composer, but then “let loose,” as it were, to generate music by means of a neutral playing-out of the material (as illustrated above). Why not then also remove control of the originating rows, as well? Instead of shaping a row of twelve tones and other rows of different dynamics, etc., why not simply generate sounds by chance? It was a brilliant move, and while Cage probably did not think of it this way, it is possible to look at it as, in essence, calling modernism’s bluff. You want notes unconnected to each other save by their mutual proximity? Then throw some dice and let it go at that.

Boulez and Cage were musical allies for something like two minutes. As soon as Boulez understood what Cage was about, he withdrew his approval. Composerly control was apparently important, after all, though it is difficult to see why, given the premise that isolated pitch/duration/timbre/dynamics were the ideal. The difference between the isolation created by conscious row manipulation and that created by pure chance was illustrated in the manipulations (above) of the sentence about my cat. For all intents and purposes, there is none.

The line of thought from Wagner to Schoenberg to Boulez to Cage can now be neatly sketched:

Wagner: It is possible to obscure the sense of key center, making the listener unsure of where she is, tonally.

Schoenberg: Then let’s be rid of tonal relations between pitches altogether and create a music without the background presence of hierarchic connections.

Boulez: If it is acceptable to abandon tonal pitch-relationships, then let’s go a step further and free ourselves from the supposed expressive relationships involving color and dynamics. Most importantly, let’s do unto duration something analogous to what Schoenberg hath done unto pitch. Down with the measured count! Let us free duration from the hierarchy of the beat.

Cage: You’re right! There exist no inherent relationships among tones, durations, or anything else. Nor are pitch and duration, etc., even necessary to what we might call “music.” Music is sound listened to inside a frame, nothing else. Listen to traffic and frame it as music, and it is music.

It’s a short intellectual distance from step two to step four, and it was a brief historical distance as well. Schoenberg died in 1951. The following year, Cage conceived his iconic piece, 4’33”, which calls for the performer(s) (any instrument or group of instruments) to remain quiet for four minutes and thirty-three seconds, while the ambient sounds of the audience and the hall produce the work of art. 4’33” is consistently referred to as a piece of music, and is even published and available for sale at $5 and change. (The idea of copyrighting a set of instructions without any determinate content is the concept of intellectual property at its most audacious.) With it, we have arrived at a place in which, not only are pieces that operate on premises opposite to how we hear pitched sound considered legitimate and even historically necessary, but sound itself is considered equivalent to music, provided it is labeled as such. It is nominalism unchecked, and it is taken today by mainstream academic and popular commentators as common understanding. In Music, Language and the Brain, Anirrudh Patel defines music as “sound organized in time, intended for, or perceived as, an aesthetic experience.”² With the possible exception of objections to the word “organized,” which to a Cageian smacks of egocentric control, it’s hard to imagine anyone taking serious exception to that definition, even though it literally means that, if I cough and burp, organizing said sounds by initiating the cough and controlling the rate of the burp’s emission, and this subsequent experience provides me what I or another person perceive to be an aesthetic experience, then I have made music. Lest the reader think I am setting up a straw man, Patel also says, bluntly and with the full force of seeming authority: “(I)t is quite clear that there are no sonic universals in music, other than the trivial one that it must involve sound in some way.”³ Noise is only noise when called noise; the same sounds called music are music. At the conclusion of his book, Thomson puts it another way, a reframing of Patel’s commonplace that exposes its absurdity. The ultimate utterance about music, given the legitimacy of the line of thought from Schoenberg to the avant-garde, Thomson observes, is simply: “There are sounds.”⁴ Thus are nominalism and materialism complicit in the same dead end.

But are there even sounds? If music doesn’t exist qua music, but is merely sound framed as music, how can the skeptic be sure that sound itself exists? Pure skeptics, indeed, find a belief in the existence of sound to be quite ridiculous. Sextus Empiricus propounded this in the Second Century CE. The argument, in his “Against the Musicians,” amounts to saying that just because vibrations in the air are registered by ears as sound does not lend existential status to sound as such; sound remains merely an experience, not an entity.⁵ This is the same argument used by those who say that music does not exist as a category separate from sound: “Just because certain sounds are registered by ears as music, over and above sound, does not lend special existential status to music as such.” Once a concept is denied as having any status over and above its material ground, then any other concept attached to that material ground is immediately suspect as well. A vibration in the air is measurable by means other than ears. Therefore, this line of argument goes, “hearing” as such, and “sound” as such, are superfluous; their phenomenal existence apart from the strictly material fact of measurable vibration is an empty concept. Consistent materialism, this illustrates, must result in the erasure of the subject entirely, or the subject is opened to the possibility of qualifying material facts in terms of its experience, thus giving the lie to materialism as such. Most materialists, however, ignore the fact that sound is not a positive fact; that it is not strictly material, but always already an experience. Both true skeptics and metaphysicians recognize this, and come down on opposite sides of it. The skeptic says, in essence: Sound is already an experience, and as such lacks real existence. The metaphysician says: Sound is an already an experience; therefore, let us start with that. Music, too, is an experience, and the experience of music-as-such is separate and different from the experience of sound-as-such. The materialist cannot claim that “music is just sound” and at the same time resist the observation that “sound is just vibration.” Ears are not required for the material phenomenon of frequency to occur, and therefore the materialist who claims that music is nothing more than sound must also claim that sound is nothing more than disturbances of the air. Neither music nor sound can be said to exist as-such, because to make either claim is to introduce a subject, and as soon as a subject is introduced, non-material axiology necessarily shows up. There is no subject without the affect of value. To rid philosophy of one is to rid it of the other.

Having seen where things led, from the first sounding of the Tristan chord in 1868 to silence/noise-as-music less than a century later, we are at last in a place to consider Nietzsche’s famous rejection of Wagner for what it truly was: A rejection of the very future we have outlined, a future Nietzsche saw coming. Modernism, as he said repeatedly, would be the death of Man, save Man’s rescue by the Overman. Only, what was modernism? What idea or philosophical approach distinctive to 19th-century Europe was so powerfully destructive that it stood to bring Western civilization itself to an end? Nietzsche notoriously swung wide at every figure in sight, from Plato to Christ, but modernism was not one of these easy targets. His attempt to pin down modernism in music takes this potent form in Human, All Too Human:

The artistic objective pursued by modern music in what is now, in a strong but nonetheless obscure phrase, designated “endless melody” can be made clear by imagining one is going into the sea, gradually relinquishing a firm tread on the bottom and finally surrendering unconditionally to the watery element: one is supposed to swim. Earlier music constrained one – with a delicate or solemn or fiery movement back and forth, faster and slower – to dance: in pursuit of which the needful preservation of orderly measure compelled the soul of the listener to a continual self-possession: it was upon the reflection of the cooler air produced by this self-possession and the warm breath of musical enthusiasm that the charm of this music rested…. [Endless melody] endeavors to break up all mathematical symmetry of tempo and force and sometimes even to mock it; and he is abundantly inventive in the production of effects which to the ear of earlier times sound like rhythmic paradoxes and blasphemies. What he fears is petrification, crystallization, the transition of music into the architectonic – and thus with a two-four rhythm he will juxtapose a three-four rhythm, often introduce bars in five-four and seven-four rhythm, immediately repeat a phrase but expanded to two or three times its original length.⁶

And then comes the prescient intuition of what may come from all this, the sort of foresight that earns Nietzsche the label of cultural prophet: “A complacent imitation of such an art as this can be a great danger to music: close beside such an over-ripeness of the feeling for rhythm there has always lain in wait the brutalization and decay of rhythm itself.”⁷ (Emphasis mine.)

There it is, from 1878, a vision of integral serialism’s arrival seven decades hence. One wonders why Nietzsche did not apply his observation to harmony as well as to rhythm, and indeed one can substitute the word “harmony” for “rhythm” in the sentence above and it is just as accurate a prophecy. Nietzsche had begun his intellectual life six years prior as Wagner’s champion in his first book, The Birth of Tragedy. He had perceived the dramatic impact of the older man’s music as Dionysian antidote to the Apollonian “petrification” of the music of the day – something Nietzsche, too, feared. (And rightly so. Mid-19th century academic music was a frigid landscape.) But now, Nietzsche saw, Wagner’s way out of things was capable of producing something much worse than a temporary freezing up of music’s creative urges: it had the potential of leading to the dissolution of music itself.

In this light, Nietzsche’s rejection of Wagner in favor of Bizet can be understood, not as some kind of angry filial punishment of the father figure, nor as the frantic grabbing for a life-preserver in the middle of a churning Wagnerian sea, but as the conscious turn of a sharp musical mind from swimming-into-musical-nihilism to standing on solid ground. Nietzsche chose to turn his back on Wagner’s innovations because he saw in them the seeds of a future in which music itself would die. Was he really so wrong?

Endnotes

1 Schoenberg put this distinction to one side, re-positioning consonance and dissonance along a continuum, with consonance more and dissonance less “comprehensible.” Yet this does nothing other than restate the distinction. It relieves the distinction of bifurcation, but only by arranging for consonance and dissonance as hierarchy in another form. Even so, Schoenberg writes as if the consonant-dissonant continuum eliminates hierarchy itself.

2 Aniruddh D. Patel, Music, Language and the Brain, Oxford University Press EBook (2008), 2.2.1.

3 Ibid., 2.2.1.

4 William Thomson, Schoenberg’s Error, p. 196.

5 Sextus Empiricus, Against the Musicians, trans. By Denise Davidson Greaves, Lincoln: University of Nebraska Press, 1986.

6 Friedrich Nietzsche, Human, All Too Human. Trans. R.J. Hollingdale, (1986) Cambridge University Press pp. 275–276.

7 Ibid., p. 276.