Not long ago, I was writing about serialism and made an all too common mistake.
I was trying to explain—to people who don’t know much about music—how serial writing got popular among composers in the ’50s. But which composers? “Advanced composers,” I wrote innocently enough, meaning only to say that not all composers got turned on by serialism, but only those who were writing…well, what should we call it? Adventurous music? But what I wrote was a mistake, for two reasons. First, I was saying that serial composers were advanced, which is very different from saying that they weren’t conservative, or weren’t mainstream, or weren’t ordinary, or however else I could have phrased this. “Advanced” would mean very special indeed—in fact, out in front of everybody else which could easily mean “better.” (Of course, even if I’d meant to say this I would have been wrong, since John Cage and Morton Feldman were also “advanced” and didn’t touch serialism. But I’ll return to that.)
Second, and more crucially, I was assuming that there’s such a thing as progress in music, or in other words that it’s actually possible to be ahead. That, of course, is a notion we take almost for granted. Often enough we’ll talk about “avant-garde” art. This means (quite literally, if you translate the words from French) art made by people who are leading the way. I know we can use the term loosely, to mean nothing more than stuff that’s so new it’s outside most people’s experience. But somewhere in the back of our minds most of us nourish the idea that art moves forward, and that some artists march ahead faster than others. The “avant-garde” (I’m keeping the term in quotation marks, so we don’t take it too seriously) would be just what the French words imply, the artists who move forward fastest, who move toward the future more quickly than the rest of us.
But does music (or any other art) really move forward? Yes, it changes, as time moves on. But can we really call those changes progress? What would progress be, anyway? Which aspect of art would be progressing?
The language in which musical ideas are expressed in tones parallels the language which expresses feelings or thoughts in words, in that its vocabulary must be proportionate to the intellect which it addresses….
Or, to put this more simply, some people are smarter than others. To smart people, you speak with a more complex vocabulary than you do to people who aren’t so smart. And this, Schoenberg says, is just as true in music as it is in words. As he goes on to say:
It is obvious that one would not discuss the splitting of atoms with a person who does not know what an atom is. On the other hand, one cannot talk to a trained mind in Mother Goose fashion or in the style of what Hollywoodians call “lyrics.” In the sphere of art-music, the author respects his audience. He is afraid to offend it by repeating over and over what can be understood at one single hearing, even if it is new, and let alone if it is stale old trash….
To demonstrate what he means, Schoenberg quotes two musical examples, the opening bars from The Blue Danube Waltz and “Di quella pira,” the famous tenor aria with the big high Cs from Verdi‘s Il Trovatore. (Curiously, Schoenberg gets the Verdi passage wrong. He starts with the first four bars, and then, with no acknowledgement that he’s doing this, skips four bars and continues with the middle section of the aria. He wanted that middle section, I think, to show that the aria had slightly more complex harmony than the Strauss waltz, but did he remember the music incorrectly? Did he know that he’d left something out?)
Both these passages are full of repetition. The waltz, for instance, repeats almost the same thing six times in a row. (And the aria would be even more repetitive if Schoenberg hadn’t left out the second four bars, which are nearly a literal repeat of the first four.) Schoenberg makes it clear that he doesn’t think these pieces are bad. He even grants that the waltz, apart from its repetitions, is beautiful. But the repetition does mean, he says, that neither of these pieces is music for “alert and well-trained” minds:
Repeatedly hearing things which one likes is pleasant and need not be ridiculed. There is a subconscious desire to understand better and realize more details of the beauty. But an alert and well-trained mind will demand to be told the more remote matters, the more remote consequences of the simple matters that he has already comprehended. An alert and well-trained mind refuses to listen to baby-talk and requests strongly to be spoken to in a brief and straightforward language.
Progress in music consists in the development of methods of presentation which correspond to the conditions just discussed.
Or, to put this differently, progress in music means finding ways to present more—and apparently more complex—musical ideas in any given span of time.
But here I think Schoenberg runs into trouble. He wrote this essay to show that Brahms was progressive, or in other words one of the artists who looks forward and makes art progress, rather than simply accepting things as he found them and standing still. This contradicts the old notion that Brahms was a backward-facing conservative, and that Wagner was the only true radical of that age. If this is true, then—assuming we accept Schoenberg’s idea of progress in music—Brahms found new ways to “draw remote consequences” from musical ideas. Not only that: For this to be progress, he must have found more of these ways than any composer before him and thus drawn more remote consequences than had ever before been possible.
But if that’s true, then why won’t sophisticates lose interest in anything earlier? Why won’t Mozart sound too simple, once you’ve heard Brahms? Why won’t Brahms himself sound too simple after we’ve heard Schoenberg?
Something’s wrong with Schoenberg’s logic. I suppose—trying to give him every benefit of the doubt—that I can think of reasons why, even under Schoenberg’s notion of progress, earlier music might still be acceptable. At some point, let’s imagine, music began to progress. Of course this happened many centuries ago. And after progressing for a while, music finally advanced enough to satisfy even the “alert and well-trained” minds of today. Even if later styles advanced still further, the earlier ones still are good enough.
But when, exactly, did music reach that critical mass? With Bach? Further back, with Palestrina? Or still further, with Guillaume Dufay? Or even sometime around 1200 A.D., when Perotin first wrote polyphony? How can we answer this question? How, especially, can we answer it if we have any respect for the alert and educated minds of other centuries (not to mention other cultures)? One problem, I think, is that Schoenberg makes a false analogy between speech and music, or, more precisely, between verbal thought and musical thought. Suppose I read a scientific book from the 18th century. The science in it wouldn’t satisfy my educated mind. But in it, even so, might be reasoning that I’d enjoy. I’d enjoy, in other words, the way the writer thinks. Here we have two kinds of thought—ideas themselves and how they’re handled. The ideas sometimes progress in verbal thought. (They do in science, anyway, if not in ethics or philosophy.) But the process of using them changes relatively little. (Apart, perhaps, from developments in symbolic logic and other things we don’t encounter much in ordinary life.)
In music, it’s the other way around. Changes in ideas don’t matter much; a chromatic phrase from Schoenberg doesn’t in itself improve on any modal figure from a medieval chant. And while the way that these ideas have been presented (and have had further thoughts deduced from them) really has changed greatly, there’s no analogue for that in verbal thought. So the analogy between verbal thought and music doesn’t work. Progress, as we experience it when we talk about ideas, isn’t much like progress as we encounter it in music. Progress in verbal thinking really matters—while I can enjoy the thinking in 18th-century scientific books, I won’t read them very often because their ideas are just too primitive. Music from the 18th century, by contrast, sounds just fine. Its ideas are just as plausible as new ideas, and its logic works.
Schoenberg, in any case, cares about musical logic way too much. He, after all, was the composer who almost scrapped the second theme of his first chamber symphony because he couldn’t derive it from the first, and also the composer who invented the twelve-tone system because he got lost in the unmappable wilderness of free atonal music. If now I might risk a comparison with verbal language, it’s as if Schoenberg feels most comfortable with statements he can verify. He’s less at ease with connotations, with hints, emotions, body language, or with basic, simple truths. But these, I think, are even more important in art than logical ideas. And they may not change as centuries unroll, at least they won’t change much. Or in any case—this is tricky to define—some things stay recognizable. Thus, we can look at a face in a portrait from the 18th century and feel we know the person. That happens in music, too, no matter what new kinds of thematic development Brahms might invent. (Schoenberg, I should add, constantly declared that artists produce art the way apple trees grow apples, and otherwise conducted himself as an artist far more intuitively—he’d compose at white heat, for instance—than his fixation on musical logic might lead us to expect. He was human, in other words, and perfectly capable of contradicting himself.)
But even if Schoenberg’s idea of musical progress doesn’t quite make sense, he was on to something. Between the 13th century and the middle of the 20th, music really did progress, in a certain way. And—though only in this certain way—Schoenberg really did advance it further.
That’s because there really was a line of musical development that began with the invention of polyphony and ended with serialism. This is usually described as if it was mainly about how harmony evolved, and in some ways that’s true. Polyphony made musicians notice chords. Chords, over centuries, were organized into the tonal system. As harmony got more chromatic, chromaticism led to atonality, which in turn got organized into serialism. And, in the normal telling of this story, what went on was more than simply change. It was progress, in the most old-fashioned moral sense. History evolved toward “the emancipation of the dissonance” (as if intervals like minor seconds had been slaves), and therefore music, at least in the serial era when people really believed this, had gotten better—freer, more flexible, and able to do more and better things.
But I’d prefer to tell the story in another way. To me, it’s about the growth of something harder to define. Maybe I can call it the density of musical information. Early polyphonic music, from this point of view, wasn’t dense at all. The chords that separate voices formed were almost random, within limits. All that mattered were the cadences, and even they were random, by later standards, because they didn’t cohere into firm tonal centers. (Because of this, theorists and musicologists have tended to think that pre-tonal music wasn’t as mature as tonal music, a misconception nicely skewered in Susan McClary‘s book Conventional Wisdom.)
When tonality emerged as a formal system, the information music carried got more dense. Every note could be explained as a member of chord (or as an accident, such as a passing note). Every chord could be explained as part of a progression; each progression took its place as an episode within a larger tonal structure.
And, especially with Beethoven, music also grew denser with motifs. More and more of what you heard in any piece turned out to be significant. Less and less was taken from the stock of standard scales, arpeggios, and cadences. By the 20th century, nearly everything could be motivic.
In Stravinsky, hardly any musical detail is innocent. Compare the openings of two pieces in C major, Beethoven’s Waldstein Sonata and Stravinsky’s Symphony in C. In both (or, in the Symphony, a little after the beginning, when a perky oboe theme begins) we hear repeated C major chords. In the Waldstein, C major chords are all they are. The progression that they’re part of turns out to be thematic, but the chords themselves—even though the way they’re spaced is nicely chosen—have no special meaning. They could have come from Sammy’s Chord Shop, if we can imagine composers ordering materials from any place like that.
But Stravinsky’s chords don’t come from Sammy’s. It’s a shock, in fact, to listen carefully, or to look at the score, and discover that they’re not even C major chords, even though the oboe tune makes us hear them that way. They’re only minor thirds, E and G, repeated over a bass line that’s also E and G, popping up in a familiarly irregular Stravinsky rhythm. And thus the E and G in that accompaniment turn out to be a three-way pun. In relation to the tune, they say “C major”; by themselves they say “E minor”; while in the larger meaning of the movement they invoke the C major seventh chord, since they’re the two notes that its constituent triads (C major and E minor) have in common.
Plus the oboe’s first notes are B, C, G, and E—the C major seventh chord. And the C major seventh at the movement’s end is weighted in the bass with E and G, densely crushed together, just as they are in the beginning, but even more so. Thus the C major at the start couldn’t come from anybody’s chord shop. The chords that evoke it were created for this piece, and are dense with information about the work’s own special harmony.
Atonal music is still denser, because any musical event—a chord, a rhythm, a sonority, a pitch-class set, a single note—could be motivic. Twelve-tone music has (at least in theory) still more information in it, because every note can be explained, as part of at least one twelve-tone row. And serial pieces from the ’50s carried—at least if you believe the theory—the most information of all, because everything was organized. Play one note on the piano in a piece like Boulez‘s Structures, and you’ve invoked a multitude of meanings. The pitch-class of the note comes from its position in a pitch-class set; its rhythm is established by a rhythmic set; its place in a dynamics set tells you how loud it is…
But of course we know that total serialism didn’t work. It didn’t work for composers; Boulez, for instance, found his own Structures too rigid. Aurally, it didn’t work; you can’t hear the information theoretically encoded in it. As Ligeti pointed out even in the ’50s, pieces that are completely organized sound aleatoric. And maybe serialism didn’t even work theoretically, since its status as coherent language has been challenged, in the ’60s by no less than Claude Lévi-Strauss and later by a variety of people, the music theorist Fred Lerdahl, for instance, or the aesthetic philosopher Roger Scruton.
Mostly, though, serialism didn’t work historically. It came and went. It isn’t with us now, in any serious, widespread way. And its claims to be the culmination of music history now seem silly. When I look at what came after it, that single arc of progress I described—oops, I slipped; I should have said the single arc of musical development, the one that brought us more and more musical density, until by the time Schoenberg wrote his Violin Concerto (to name just one of his late twelve-tone works), musical events jump out, several at a time, from every corner—looks as if it’s over.
After serialism (and the complex of atonal styles that flourished in its wake), the next turn of the wheel produced minimalism. And after that (though also along with it, since historical developments can overlap) we’ve returned to tonality, while we embrace all sorts of sounds from outside classical music. It’s not clear that these evolutions follow any logic. Or if they do, it’s not a purely musical logic. Maybe the arc of density told the story of the rise of western culture, which reached a peak and then degenerated, clinging to what it thought was certainty and logic. Musically, at least, it turned inward, away from the world at large.
And then, in a sense, it collapsed. Or at least it did if you think serialism was its peak. More hopefully, we could also say it came alive again, taking energy from music that wasn’t classical. Steve Reich was influenced by African music; Ravi Shankar inspired Philip Glass.
Maybe this is progress. But only in a global sense, which ultimately doesn’t have much to do with art. Art only reflects it, or maybe anticipates it. Maybe one meaning of the arc density would be that Western culture died between 1914 and 1945, as the arc reached toward its peak. The peak was self-contradictory; the arc had no way to continue. From there we fell to limbo, a period when the old hasn’t fully disappeared, and the new hasn’t yet been born. If that’s true, then we should take Schoenberg very seriously when he said that twelve-tone music would ensure the supremacy of German music for the next thousand years—though we have to turn his statement inside out. What he really told us was that the great European musical tradition couldn’t possibly be saved. Unconsciously, he sensed that. He fought to find a rescue, of course in vain. And the very scope and certainly of what he said—that he hadn’t just invented something useful, but had saved music for 1000 years—almost proves his helplessness.
From this point of view, the twelve-tone system—and, even more, the more tightly organized serialism that followed it—really would be a sign that western music had come to an end. Nor would it be coincidental that serialism arose at the same time as rock & roll, which—in bringing something straight from Africa into the western pop-music mainstream—was a gigantic portent, warning everyone that western culture soon would open to the other cultures of the world.
But here’s one final thought. If we know that history didn’t culminate in serialism, non-serial—and non-atonal—composers of the 20th century seem much more important. It’s been obvious for quite a while that John Cage and Morton Feldman were just as “advanced” as any serialist, and in fact much more so. But what’s fascinating now is to read the present back into the past. One feature of our culture now is irony and self-reference. Nothing seems certain. Everything exists on more than one level, does more than one thing. And we’re all aware of what we’re doing.
If that’s the present—the point where, for the moment, history has come to roost—then the most important composers of the past century ought to be the most ironic. I’d vote for Poulenc and Shostakovich, both of whom can easily say more to us than any serialist who ever lived.