Composer In The Machine

Indulge me in a thought-experiment.

You’re sitting on a plane waiting to take off. In the seat next to you is a composer, an eminence grise with a robust list of professional accomplishments (even though you probably haven’t heard of him). You ask him—as people who sit next to composers on planes often do—what kind of music he writes, and he divulges that what he really does is write programs that write music; moreover, he bestowed upon one of these programs a human name: Emily Howell.

You might inch away from him in your coach-class seat, smiling reassuringly. You might wonder if he also names his guns, his CB radio equipment, or his erotic devices. But would you question whether the music that “Emily Howell” composed really belonged to him?

The effed-up thing about David Cope isn’t that he’s developed software that produces music; as Cope notes in his very measured retort to Noah Stern Weber’s torch-and-pitchfork criticism, the history of algorithmic music goes back further even than Lejaren Hiller’s 1956 ILLIAC Suite. Cope wrote a program that writes music, but he could just as well have written a program that writes programs that write music, or written a program that writes programs that write programs that write music: Any way you slice it, Cope’s executed the labor necessary to produce pieces of concert music, and there’s no serious qualitative argument to be made that his cyborganization of certain parts of his creative process has led him across the Rubicon that separates “writing music” from “no longer writing music.” Rather, what’s so unsettling about “Emily Howell” is that Cope gave it a person’s name.

“Emily Howell,” as Weber astutely notes, is not a person. (Corollary: Ensembles that program a piece by “Emily Howell” can’t claim to be advocating a female composer.) “Emmy” was bad enough, but a surname? Super-weird. Cope didn’t have to tell people he was using his software to write music (although I’m sure that research served him well in front of various committees at UCSC); he really didn’t have to give it a name like “Emily Howell,” which sounds a bit like the name of a Victorian scullery maid—cheap help. Part of what incenses people about Cope’s work, I suspect, is that he insists on doing a lot of the anthropomorphizing for you, just in case it hadn’t already occurred to you that COMPUTERS ARE TAKING OVER.

How can you not get a kick out of the stories about listeners freaking out when they learn that the piece they just heard was produced with the assistance of software? It stings when the myth of genius is shown to be false. Audiences who haven’t read as much Barthes as we have expect to encounter a unique and expressive subjectivity, and when they learn that a piece they really dug was churned out by a machine—perhaps they envision a dot-matrix printer chirping out the score minutes before the concert—they feel like suckers. But of course they haven’t been duped at all: “Emily Howell” is nothing more or less than a part of David Cope. When you hear a piece she’s generated, you’re hearing a piece of him.

For the same reason, I find Weber’s contention that an ensemble interested in a piece by Aaron Copland but unwilling to rent the parts might just feed Rodeo and Appalachian Spring into the ol’ computron a bit hard to swallow: Anyone with the proper training could produce Coplandian material armed only with a pencil and some manuscript paper; I could, Cope could, Weber could. Leaving aside the “document in the history of thought” aspect of musical works—an aspect whose criteria no ersatz Copland score produced by man or machine can satisfy except as such—many characteristics of a piece of music that contribute to a genuine aesthetic experience aren’t apparent at the level of the phrase or the section. I distinctly remember my first encounter with the Mozart knockoffs produced by Emmy, Cope’s first maid: They sounded a lot like Mozart. They could have (indeed, did) fool lots of listeners. But they lacked the “perverse flashes of insight,” to borrow from Ferneyhough—the complex play of semiosis, middle-ground structural articulation, and occasional batshit crazy digression that put Wolfie in his own league.

One respect in which Cope hasn’t gotten his due is that nobody seems to be taking Ms. Howell’s output seriously. It’s a nice-sounding post-minimalist paste, two parts Ravel to one part Pärt. Non-specialists think it’s beautiful, but that’s not surprising: It sounds an awful lot like a whole bunch of music that is widely agreed to be beautiful. If Cope is going to make us call this digitized splinter of his artistic psyche by its own name, the least he can do is keep pushing her to produce compelling results.

Leave a Reply

Your email address will not be published. Required fields are marked *

Conversation and respectful debate is vital to the NewMusicBox community. However, please remember to keep comments constructive and on-topic. Avoid personal attacks and defamatory language. We reserve the right to remove any comment that the community reports as abusive or that the staff determines is inappropriate.