Sunday, January 3, 2010

The world in six embarrassing mistakes

If you told someone in the year 1950, say, that the nonfiction bestseller lists in the 2000s would be full of books about music cognition, you would probably have been laughed out of town. Yet anyone with the slightest interest in music has now read either Musicophilia or This is Your Brain on Music, usually both. Music teachers, always anxious for new ways to justify their existence to poker-faced administrators, were filled with renewed vigour: if popular science books are for us, who can be against us? Yet, despite my admiration for author Oliver Sacks, a particularly gifted essayist, I've managed to enter 2010 without reading either of these music cognition books, in which happy state I would have persisted indefinitely had I not received The World in Six Songs as a Christmas gift.

The World in Six Songs is by Daniel Levitin, and is a sequel of sorts to his This is Your Brain on Music; where the earlier book tried to give the lay reader an account of the relationship between music and the brain, this new book purports to describe all of musical expression as manifestations of six song archetypes, each of which fulfills its own evolutionary role. A book whose stated goal is to "explain" aesthetic experience as an epiphenomenon of neo-Darwinian evolutionary processes sets off all the Neon Arrow alarm bells; furthermore, the first few chapters of the book seem to be designed for the explicit purpose of annoying me. Does the first chapter set out the axiom that classical and popular musics are really at bottom the same thing, that no distinction can be made between them except personal preference, and that anyone who disagrees is ignorant and bigoted? Check. (Has anyone noticed that the level of invective directed against anti-pop recusants increases in inverse proportion to their actual numbers?) Does the book make the nonsensical and historically illiterate equation of orally transmitted folk musics with electrically transmitted pop musics? Check. Is it full of proofreading mistakes? Oh, my friends, is it ever full of proofreading mistakes. Levitin thinks "opus" is the name of a musical genre, falls all over himself in a mercifully brief allusion to Gregorian chant, and cites the Song of Songs with Handel's Messiah as an archetypical example of religious music, even though none of its musical settings are particularly well-known and the poem is religious only by analogy. Anecdotes and facts that originally appeared earlier in the book are spattered through later chapters more or less at random, each time presented as though the reader was encountering this story for the first time. So sorely does this text need a good proofreader that one wonders if Levitin's publisher has some kind of grudge against him; indeed, early in the book, we find that someone at Viking has inserted "[sic]" into the middle of Levitin's body text.

Enough whining. We've established that I don't like the writer's style or his ideological presuppositions - what about the argument?

The book attempts to subdivide the world's musical repertory into six song types. Some speak to particular emotional states (songs of joy, songs of comfort); others express the bonds between two individuals, or between members of a society (songs of love, songs of friendship). Still others are intended to accompany ceremonial display (songs of religion), or are used as teaching tools to pass on the knowledge of the community (songs of knowledge).

This sounds quite plausible, but I don't like it at all.

One problem is that the model is hard to apply to music of any particular length or complexity; the pieces that I find most satisfying blur the edges of several categories, or seem to invent new categories of their own. Levitin realizes this, and so most of his examples are cherry-picked to express a single mood in uncomplicated terms, including car safety jingles and songs by James Brown "I feel good/I knew that I would" and The Turtles "I can't see me lovin' nobody but you". But what happens to works that express more complicated emotions, or to large-scale song cycles or albums that touch on a wide variety of emotions over the course of an hour, or for that matter to the entire corpus of instrumental music? Levitin can't force any of these into his scheme because their worth has always been based on traditional concepts of aesthetic expression - that they help us to exercise an imaginative perception, or that they exhibit objective beauty of a sort that the listener can apprehend.

The chasm that separates Levitin and myself, in this case, is that his conception of music is instrumental rather than aesthetic. For Levitin, there can be no objective aesthetic standards because our preference for music is the result of evolutionary selection over several millenia. We are attracted to certain combinations of sounds as musical, and consider frequencies an octave apart to be equivalent, because a random mutation caused our ancestors to develop these traits and to thereby attain some competitive advantage over other early hominids. Daniel Dennett is approvingly quoted as saying that we find babies cute not because they are "objectively" cute, but because mothers that find their babies cute care for them more lovingly, which is to the benefit of the species. The spectre of intentionality and free will thus safely banished, we can return to our real task: unmasking the instrumental purpose our attraction to music originally served.

This is all very well, of course, but Levitin can't give a clear account of what advantage that random, musical mutation would have given to that first hominid ape. It's obvious, of course, that social practices like music have evolutionary benefits when they're fully developed - a society that uses music to communicate has many advantages in terms of social cohesion, knowledge transfer, and general cultural sophistication - but random base-pair mutations occur in individuals, not simultaneously in entire societies. It's unclear how possessing an aesthetic sense would give a single individual a competitive advantage in a community where no-one else was capable of perceiving beauty. After all, a meaningful musical experience cannot exist without a community of performers and, still more, a community of listeners who give collective meaning to the musical sounds they create. One would think that a sudden, incomprehensible attraction to meaningless sound combinations would be more of a disadvantage than otherwise, estranging the individual from his peers and turning him into a sort of Pleiocene village idiot. Still less can I understand how random base-pair mutation can account for the totality of aesthetic experience: we perceive the different means of aesthetic communication not as separate activities involving different sense modalities (and thus different parts of the brain), but as an interlinked nexus of activities ("artistic expression") which collectively express our society's imagination and view of itself. Neo-Darwinism has no convincing answer to any of these questions, and I would argue that it cannot provide one, and indeed that it is overstepping its boundaries in trying to do so.

Yet even after all of this it's possible that some neo-Darwinian might be able to plausibly model a way that the interlocking aesthetic phenomena I've described might have served some instrumental purpose at each stage in their evolution. It wouldn't matter, though, because our differences are philosophical, not scientific. If I were merely offering traditional aesthetics as a sort of Occam's Razor solution to the limitations of current knowledge, I could be refuted by the next discovery in neuroscience; however, I'm making the stronger claim that Levitin's project will never succeed, because it's based in a materialist philosophy that is essentially incoherent. An account of our musical perception is an account of consciousness, and a satisfactory account of consciousness can never emerge from a neo-Darwinian analysis. Neo-Darwinism treats all lifeforms as automata without thoughts, feelings, or ideas - it cannot take into account free will or goal-directed thought, because those things cannot be accounted for in material terms. If you insist on making neo-Darwinian processes the sole basis of reality, you have only two options: you can either assert that free will and consciousness do not really exist, like Paul Churchland, or you can sweep them under the rug and hope no-one notices, like Levitin.

With better editing, and with a more anthropological focus, Levitin's "six stories" idea could have made a good short book, or a long academic paper. Embarrassingly, however, he insists on trying to explain all of musical perception in neo-Darwinian terms, and the attempt ends in a messy train wreck.

But I'm still glad of the gift; who ever learned anything from a book they agreed with?

No comments: