The Rhythm of Life Sciences

Matthew Cain, enthusiastic presenter of the Channel 4 series What Makes a Masterpiece?, has been asking what makes music tick (episode 2). 

Good question.

So he spoke to some scientists: one who had designed a computer which could “recognise” a hit song and another who stuck wires on people’s heads to see how music affected their brains.

The first bit of science begs the question, can a computer write a hit song?

In theory, maybe. The computer identified common traits of hit songs and grouped them in “clusters”. A new song could be tested and, if necessary, tweaked so that it came closer to one of the clusters, thus increasing it's hit potential. But could the computer create a song from scratch? The question was never asked directly but I would have thought it theoretically possible. Although given the fact that computers still can’t hold a 5-minute conversation, "I'm Dreaming of a Megabyte Christmas" may be a way off yet.

(It’s also worth pointing out that this computer only analysed music and, really, a hit song is the successful pairing of a particular bit of music with particular words which, presumably, would multiply the variables around a zillion-fold. So good luck clustering that. Then again, computers are good with zillions in a way that we’re not.)

The second bit of science - the wires and the brain - begs the question, who needs music?

If music is essentially a neuro-chemical effect on the brain and if this could be replicated by some simpler method of brain stimulation, then why bother with the tediously traditional and messy method of composing music which, ultimately, may not have the desired effect for the listener? Again, the programme didn’t explore this issue but, it seems to me, the possibility is there (although still a million miles away given the current technology).

So what does the future hold?

OK, I find it difficult to imagine a time when music is replaced by wires attached to your head. On the other hand I can imagine computers playing an increasingly significant role. In such a future perhaps music will split between “computer-generated” and “natural” in the same way that food has, in recent times, split between genetically-modified and organic. That is, the computer stuff will be cheaper, shinier and more voluminous but a few purists will still pay twice as much for what they like to think of as the "real thing". Most likely, as with food, music will probably be a hybrid and it will be increasingly hard to tell the difference.

For some this will sound reductionist alarm bells. Can the beauty of a Mozart aria be reduced to a computer programme? Can the majestic power of Beethoven’s 9^th ever be explained in terms of bio-chemistry? What about Peter Andre?

The bells have a point but shouldn't be so alarmed; in the end, music is irreducible. When a composer composes they don’t experience the act of composing as a robot churning out data. When a music listener listens they don’t experience the emotions generated as mere neuro-chemistry. Part of the experience is the feeling that something more is going on. So, even if the we had a full scientific understanding of how music works, our experience of music would still point us towards something beyond that understanding. If it didn't, then it wouldn't be music.


"To feel The Rhythm of Life
To feel the powerful beat
To feel the tingle in your fingers
To feel the tingle in your feet..."

Now that's a masterpiece. Spread the religion, Daddy.