Teaching the Music that Doesn’t Behave

Now Amy Bauer’s posts have got me thinking about music that adheres to theoretical paradigms versus music that doesn’t.

In theory class, I feel as much as any professor the pull of pieces that behave nicely. It’s so satisfying to pass out the Webern Piano Variations, the D# minor Fugue from WTC Book I, Chopin’s B-flat minor Nocturne, Schubert’s G-flat Impromptu, the Maple Leaf Rag, Clementi’s First Sonatina, and know that the analysis is going to go just the way you think it will. That’s why Beethoven’s First Sonata is in every theory book: not that it’s his most impressive piece, but it exquisitely fits the textbook definition of sonata form you’re about to give them, and convinces them you know what you’re talking about. How inconvenient that K. 545 C-major Mozart Sonata is, with its recap starting in the subdominant! It’s cute and everyone knows it, but you can’t use it until you’ve taught ten other sonatas and throw it in as a wierd exception. Otherwise the students look at you funny.

Every theory teacher, I feel sure, collects over the years a repertoire of pieces for analysis guaranteed not to make him look stupid in class. But if the teacher always looks so smart, doesn’t the music start to look stupid? I have to use those pieces, because sometimes you have to make a specific point in a circumscribed amount of time, but I consciously resist limiting myself to them. I have some pieces I analyze – the bitonal Saudades de Brazil of Darius Milhaud, a symphony (Second) by Martinu, Liszt’s Sposalizio, that just don’t behave well. I give them William Caplin’s rules for the sonata, distilled from Haydn, Mozart, and Beethoven, and then analyze – Dussek!, who never read Caplin’s book. Every other year I wade a group of students into “Emerson” from the Concord Sonata and we try to figure out what the hell Ives had in mind, with perennially sparse results. I have learned to savor that stupid feeling of not being able to explain a piece fully, and not being able to justify its existence any better than pointing out that I like it. Sometimes I bring in pieces I’ve never analyzed, with no idea what we’ll find, and we just start rummaging around. I even teach an entire course that way, my Advanced Analysis Seminar, in which we spend all semester on three works I’ve never analyzed, chosen specifically because I don’t understand how they work. Once we got lost in the Stravinsky Piano Concerto, practically my favorite piece of his, and thought we’d never get out.

Of course, there are repercussions. The students form the dangerous idea that not all music is nice and neat and compact and explainable. They gather that spontaneous inspiration, subjective taste, and irrational will are part of the composer’s arsenal. Of course, discipline, rigor, structure, and foresight are too, but why skew the weight so far in favor of one set of values over the other? Why parade Apollo onstage, and bind and gag Dionysius in the dressing room? Let us admit, at the risk of sacrificing some of our unquestioned authority, that the academic canon of Music Acceptable for Analysis, the music we put forward as The Best Mankind Has to Offer, is actually The Music that Makes Us Look Smart, because it follows the rules we’ve been taught to explicate. Then we can play in class whatever music we most love, and if it works out neatly on paper, or if it doesn’t, those are both lessons. “Let us try for once not to be right” – Tristan Tzara. And maybe young composers – rather than either slavishly follow our directives or guiltily break away from us – will learn that the proportion between freedom and order is one every artist has to work out for himself.

[I’m going – gasp! – offline for a couple of days, and will print your comments when I return. I gotta stop staring at this screen.]


  1. Michael Wittmann says

    How funny this is to read. I’m a physics professor. In class, so much seems factual and Makes Me Look Smart – except that the most basic question isn’t answerable. What IS gravity? I can describe properties, interactions, all sorts of things, but what causes it? Who knows. Same with mass. What is it really? Who knows. And, by the way, what’s the meaning of a question like “what is it really?”?

    Discussing ideas from the perspective of doubt, playfulness, and community exploration is wonderful. I teach non-science-majors a course in quantum physics, and it’s a gas. Ad hoc rules, coherence of thinking out of inconsistent results, and so on. What a blast. It’s nice to know that music theory can be taught the same way, and I love hearing the ideas you present as you write about it. Thank you.

  2. Amy Bauer says

    Kyle I really identify with this post, as I feel like I’ve spent the last 20 years attempting to analyze music that “doesn’t play by the rules.” I’m teaching my favorite course this fall: Theory and Analysis of Post-1945 music, which I am constantly tinkering with (if only I could have a term for each decade. . . .)

    For every Babbitt “Post-Partitions” or brief excerpt from Lutoslawski that virtually analyzes itself there are 10 pieces that defy easy parsing.
    The student comments I received the last time I taught this course cracked me up: “Too much history, not enough theory.” I tried to cram too much in, yes, but I also think that their notion of what constituted “theory and analysis” was somewhat narrower than mine(!)

    I don’t want to bog down what is intended as a more general blog, but if you have any tips for how to approach more recent music (beyond what Ive already cribbed from The Music of Nancarrow), I’m all ears. There’s not enough time in one term to teach an individual analytical method for each work, much as each may deserve it, so I often fall back on charts and graphs.

    And, alas, leave a lot out,while trying to do justice to many different traditions.

    Thanks for responding!

  3. says

    I haven’t been in a classrom in many years (in any capacity) and my remarks should be taken in that light.
    I think students should be told whay a given piece is being analyzed and why analysis is important in the first place. That is, if you believe it is, which I do. And I would never analyze a piece that I didn’t like on some level, no matter what paradigm it seemed to embody.
    Once that paradigm is discovered and laid out through the piece, I would find the elements in the piece that push against the paradigm. I think I would go so far as to say that if there are no such elements in the piece it’s not any good after all.
    In short, I would find the strange in the familiar and the familiar in the strange. I’d extend that to performance, too, but that’s another (not unrelated) issue.

  4. Daniel says

    As a recent graudate of musicology, I was always trying desperately to get my professors to help me (even allow me) to analyze the local “noise music” and “computer music” here in Southern California. Much of the music is algorithmic, improvisational, or both; but as you say in an article in your Music Downtown, analysis of Elliot Carter can hardly be understood by the listener who has not studied the score – what’s on paper isn’t always heard. I came to a similar conclusion in a paper I wrote called “Memory and Composing With Twelve Tones,” based on Schoenberg’s proclamation that what audiences need in order to understand his music is a better memory.
    As an analysis junky, I am at a loss for words when trying to describe some of the greatest sonic experiences I’ve had, from the recently past charm of Jim Tenney’s “For Ann (Rising)” to the multichanneled digital blips of the music my friends are making in LA. What more, I fear there is no work for a musicologist like me, unless I get lucky enough to work for the Village Voice!