Against music theory
I am mercifully finished with music theory in grad school and couldn't be happier about it. You may find this surprising. My blog is full of music theory. How could a guy who enjoys thinking about music in analytical terms as much as I do have such a wretched time in my graduate music theory classes? It wasn't the work, I mostly breezed through that. No, it was the grinding Eurocentrism. Common-practice period classical music theory is fine and good, but as presented in a typical theory core, it's dry, tedious, and worst of all, largely useless to a musician like me. The strict rules of eighteenth-century European art music are distantly removed from the knowledge that I need to do anything in the present-day music world (except, I guess, to become a professor of common-practice tonal theory.)
The title of this post is a reference to the Susan Sontag essay, "Against Interpretation." She argues that by ignoring the subjective sensual pleasures of art and instead looking for rigorously logical theories of its inner workings, academics are missing the point. She calls scholarly interpretation "the intellect's revenge upon art." I'm with her. Music theory as practiced at NYU and elsewhere is the intellectual's revenge on music. Sontag's punchline is right on: "[I]n place of a hermeneutics we need an erotics of art." Speak it, sister!
Not everybody shares Sontag's position. Louise Norlie points out that if Sontag is right, then "[i]nterpretation would be little more than a marvelous description of how a novel or a sculpture makes [elite audiences] feel." But that's the whole point. What is a theory of art if not an organized method of understanding the sum total of people’s tastes? Where do the rules of musical correctness come from if not from our collective whims? Theory might include some handy mathematical methods of describing those whims, but do our whims emerge out of the math? I don't believe so.
Understand: I have nothing against music theory of a particular kind, the kind that's descriptive of what people have found to sound good. A worthwhile theory teaches us specific skills: how to listen and understand, how to perform and interpret, how to compose and improvise new works. If you learn how to derive the modes and chords from the major scale, you gain a valuable toolkit for understanding and generating a lot of the sounds that Western people like. Same goes for an understanding of the circle of fifths, or the blues, or the backbeat.
Good music theories are descriptive, rather than prescriptive. They look for patterns and structures that have been found to work. They don't define what's acceptable and unacceptable. Bad theory is the basis for codifying arbitrary value judgments. The classical rule against parallel fifths is nonsensical if you play rock power chords, or traditional music from a variety of world cultures for that matter. The rule against parallel voicings generally would throw out all of rock and a sizeable chunk of jazz. Maybe it's time to stop thinking of the classical rules as rules at all, and instead call them what they are: a guide to the subjective preferences of people in powdered wigs.
Scientific theories are valid only to the extent that they make predictions. But music theories deal with the vagaries of human preferences. We don’t know what we’re going to like until we hear it. Musical innovation is only possible when some inventive composer or improviser plays something “wrong” in a way that sounds right. A theory can be a coral reef for memes, a hospitable ecosystem for the proliferation of new music. A bad theory inhibits such inventiveness.
A lot of popular music is tonal, but a lot of it continually violates the rules of tonality. A quick scan through the songs on radio reveals unresolved tritones, parallel fourths and fifths, major and minor thirds freely intermingled, and so on. Blues and jazz in particular are exceptionally poorly described by the rules of common practice tonality. Yet blues-based music is not “atonal.” There are clear key centers, functional chords with associated scales, more and less dissonant embellishing notes. A good theory of tonality would be able to account for the blues.
Really, a good theory needs to be able to account for everything that people find gratifying. The parallel to scientific theories is necessarily limited. The rules of quantum mechanics or thermodynamics are universal and unvarying. But the rules of music are culturally contextual, and they evolve over time. When one of my theory professors played us an example of “bad” voice leading, moving an entire major chord in parallel up a whole step, the classroom lit up with pleasurable recognition — it sounded exactly like the opening phrase from “I Want To Hold Your Hand” by the Beatles. If a classroom full of musicians likes something that's "incorrect," who's wrong, the musicians or the rule?
There can be an empirical basis to music theory, but we should be aware of the limits of what we can know about human preferences. We may be able to discern trends and patterns in our tastes, but we can never make predictions or models with the precision of physical theories. I like Steve Dillon's "extended aural perception" framework: we should consider the effect of music on the listener, and then examine the effectors that caused it. We can draw a circle around the most generalizable effectors and call them "music theory."
A theory of music emerges from and defines an interpretive community. We are all members of the community that recognizes octave equivalency (along with rhesus monkeys.) We are pretty much all members of the community that likes combinations of the pitches derived from the overtone series. Mozart was in the community that doesn’t dig parallel fifths; Kurt Cobain was in the community that does. Music theory should really be a subdiscipline of ethnomusicology. Then maybe the classes would become a lot more useful.