Saturday, December 10, 2011

Who Killed Harmony?

I've put up a few posts on the problem of harmony. Here, here and here. I'm putting up another one because a friend of mine just sent me an email about a concert she attended of the Del Sol String Quartet playing new music. She commented that there was virtually no harmony, everything was layered counterpoint. Yes, I think that there are two main schools of thought among composers these days. One school is still loosely following the modernist tradition begun by Schoenberg that emancipates dissonance. Without the tension between consonance and dissonance there can be no functional harmony so they relied mainly on counterpoint and canon. Serialism, with its use of a tone-row along with its retrogrades and inversions is the prime example. But even in the absence of that technique composers in this school tend to avoid any suggestion of functional harmony. The other school, often called 'minimalist', unashamedly uses consonance, but as they tend to sit on one harmony most of the time, thus making it by default 'tonic' also don't really have functional harmony. The third post linked to above discusses this in detail. Oh, one other school is the neo-classicists, but they tend merely to imitate traditional forms with "wrong notes" so their harmony isn't functional either. The result in all three cases is to my mind unsatisfactory because music without harmony is a lot less than it could be.

For the last few years I have been exploring how to write truly harmonic music again. I think the first thing composers should shed is their irrational fear of harmony. Don't worry, even if you write functional harmony it isn't going to sound like Beethoven or Brahms anyway--unless you try to parody them. What you need to do is develop a sensitivity to harmonic function and let your instincts go to work. There are two models that I think can be useful. One is Debussy as I talk about here. One thing I am surprised by is that while, on the one hand, everyone talks about the huge influence Debussy had on 20th century music, but on the other hand, everyone avoids being influenced by his music. Dig into it and you will see that underneath the beautiful 'impressionistic' surface, he is using a lot of harmonic function to structure the music. The other model is Shostakovich, anathema though he may be to composers. He is certainly appreciated by audiences and performers. And yes, he very much uses functional harmony, though of a kind that has scarcely been examined by theorists. Another composer who continued to write functional harmony was Benjamin Britten.


The lesson that most composers seem to have gathered from the paroxysm in music that happened in the 20th century is that functional harmony is no longer a possible option. But I think that music with no functional harmony is crippled both in terms of how you can structure it and in terms of the expressive content. So let's rediscover functional harmony. We don't have to follow the old rules, we can make up new ones, or just rely on instinct. Here is another piece by Britten:


7 comments:

Nathan Shirley said...

I think I agree with you here, but I'm confused as to how you define neoclassical. I consider Britten to be neoclassical, if not strictly, at least much of his music leans that way. And Shostakovitch as well. Bartok, Stravinsky (before the later moved to the US and lost his way musically) and Prokofiev also neoclassical. These are some of the TRUE greats, and certainly knew what they were doing harmonically (especially Prokofiev).

Or are you referring to the school which Hindemith belonged (German and some of the French neoclassicists)? I would argue that much of their music was simply less inspired than the Russian camp.

Or maybe my definition is far too broad?

Bryan Townsend said...

Yes, I think you may be drawing the lines too broadly. The neoclassical in music is a self-conscious style, mostly during the 1920s and 30s, which borrowed older forms. The Stravinsky of Pulcinella, the Octet, the Symphony of Psalms. Some Prokofiev, Milhaud, Poulenc. Some even say that there are pieces by Schoenberg and Berg that show neo-classic influence. But people like Britten and Shostakovich are not considered part of the school. What they are doing, I suppose, is better understood as an extension of tonality. Bartok is another approach entirely.

Nathan Shirley said...

So you have a much more literal definition of neoclassical I suppose. But then I'm still confused...

Are you saying Stravinsky's Pulcinella, Prokofiev's Classical Symphony, etc, don't have real harmony?? (functional harmony is another term that can have very fuzzy edges).

Or are you saying the main neoclassical composers were Hindemith and the like? Who do you consider to be typical neoclassical composers?

Bryan Townsend said...

These are interesting questions and I suspect that they are ones that have yet to be fully investigated. My sense is that neo-classicism (the category itself is controversial) was a style that used some of the forms and gestures of tonal music: a minuet perhaps, or a chorale as in L'Histoire du Soldat, but used them either as a box to put things in, or as a kind of ironic quotation. In either case, the tonality is not nearly as functional as it would have been in the original. Stravinsky's Great Chorale is, as one counterpoint text says, "a caricature of a Bach chorale".

The main neo-classic composers are Stravinsky (certain pieces), Milhaud, Poulenc, Prokofiev (some pieces). But as I said above, neoclassicism in music is a much-disputed topic.

I think that Pulcinella, L'Histoire and similar pieces certainly have real harmony, but the functionality has been severely impaired. This is what I sometimes call "wrong-note" music as the composer, to make sure his music isn't mistaken for ordinary tonal music, puts in some jarring dissonances...

Nathan Shirley said...

Now I'm starting to see where you are coming from. But I don't think neoclassicism necessarily = less functional harmony, necessarily. For example Prokofiev often wrote out his music with fairly straightforward harmony, then went in and added 'wrong' notes. But these wrong notes were not arbitrarily chosen, and I would argue they are not 'wrong' notes at all. They actually enhance the harmony, creating new harmonies. By adding leading tones into chords for example it can create extra tension so that the resolution is even more satisfying (even if the resolved chord is not completely settled, but if done well, all the more interesting. I think Shostakovich does this all the time even though he had a different compositional process).

Now, in less inspired hands this can certainly lead to bad results (Poulenc is hit and miss for me, Milhaud is more miss than hit, even Stravinsky has his share of misses). But to me they are simply cases of less GOOD music, not necessarily less functional harmony (again depending on your definition).

Interesting perspective at any rate, I've never given it this much thought!

Bryan Townsend said...

Very good observations, Nathan. I confess that I have not studied Prokofiev to any extent and your remarks make me think that is overdue. Added leading notes is not what I would call a typically neo-classic device, though. After all, this is exactly where the very functional Neapolitan and augmented sixth chords come from. When I think of neo-classic harmonies I think of extraneous notes that do not resolve. But again, I really need to have a look at what Prokofiev is doing.

What we find sometimes is polytonality where there are two different and clashing tonalities present at the same time, in two different layers. I haven't thought a lot about this, but I notice that Britten does this sometimes to good effect.

You are absolutely right, the only really important thing is whether there is a good musical result!

Nathan Shirley said...

>>>"After all, this is exactly where the very functional Neapolitan and augmented sixth chords come from."

Right, I was using a simple example. Some of the more complex (undefinable) chords result from minor seconds added in (as just one example), acting like leading tones (up OR down) pulling and pushing towards chords not necessarily tonic. So things get ridiculously complex if you try to analyse music with so many 'wrong' notes. But after only a few listens your ear learns where it is all going.