Better late than never–remember that Washington Post experiment where they sent world-famous violinist Joshua Bell out busking?
“No one knew it, but the fiddler standing against a bare wall outside the Metro in an indoor arcade at the top of the escalators was one of the finest classical musicians in the world, playing some of the most elegant music ever written on one of the most valuable violins ever made.”
It’s occurred to me that, while the Post was hung up on cultural reasons for his audience’s disinterest, there may have been something else entirely at work. A piece of music is built out of a series of phrases, and the relationship between whole phrases makes up a major part of how we experience it. Ever been to a concert, in any genre, where the performers showed technical skill, but somehow failed to make a lasting impression? The missing “spark” is hiding in the way the phrasing is structured.
Of course, the question of exactly what kind of structure we enjoy tosses us back into the realm of personal taste. In the case of Bell’s subway experiment, though, we add an interesting objective restriction: if we decide whether we like a piece of music or not based on the structure of the phrasing, then for a piece of music to grab our attention as we pass we must be able to hear a sufficiently-large sample of that phrasing. If you watch the videos included with the Post article, you can see that a single line from, say, Bell’s Bach selection just takes too long to reach a resolution. By the time he’s done, his prospective patron has already left the station.
Here is a selection from the Post’s recording of Bell’s performance:
(direct link)
This line resolves in two phrases and lasts a little over ten seconds. For comparison, here’s a selection from a Peruvian pan-flute band that a friend of mine recorded in the Paris Metro:
(direct link)
It reaches a resolution in one phrase of five seconds. (If you want more context, or just like Peruvian pan flutes,
here’s a longer excerpt.)
It may seem like a very fine distinction, but just think–all else being equal, the second example fully doubles your chances of delivering a satisfying payoff while your target’s still in listening range. The experimenters, I think, just didn’t understand their audience.
“Edna Souza is from Brazil. She’s been shining shoes at L’Enfant Plaza for six years, and she’s had her fill of street musicians there…On her speed dial, she has phone numbers for both the mall cops and the Metro cops. The musicians seldom last long…’He was pretty good, that guy. It was the first time I didn’t call the police.’”
It’s significant that, out of all the Post’s interviewees, Souza’s the only person not mentioned as having some background in classical music who has something nice to say about Bell. And, the whole time, she was stationary.