Friday, June 10, 2011

Time Quotient of Music

Duke Ellington is once reported to have said that there are only two kinds of music: good and bad. This is a wise observation. The fun is trying to sort them out! Musicians and composers are always doing this in their own work. Was that a good sound or not? Is that the right note/chord or not? But listeners do it too. Aesthetic judgments are notoriously subjective, but there are objective aspects as well. For example, over a long period of time, certain music gains in popularity/respect/appreciation, while other music loses. I mention J. S. Bach a lot; the reason is that his music has, over the last nearly three hundred years, proved to be perennially popular with all kinds of listeners. The New York Times music critic embarked on an exercise lately to come up with the list of the ten greatest composers. Here are the results:

Number one on the list is Bach and next is Beethoven. I might be tempted to switch them, but no doubt about their predominance. From there on there is plenty of room for argument. I would certainly have included Shostakovich, for example, probably in place of Bartok. The deeply underlying principle suggested by exercises like this is that the passage of time serves to erode away the more superficial music, leaving only the really worthwhile. Bach, at the time of his death, was, except in very select circles, a rather obscure church musician. His renown has grown without pause over the last 250 years. But during his life, he was overshadowed by more famous musicians like Telemann, Handel, Rameau, Couperin and others. They are also fine composers, but are not as substantial as Bach.

This process of elimination and discovery over time I call the time quotient. In a future post I might try to quantify it a bit. Basically, if a composer is well known now for music written ten years ago, the time quotient is pretty low. But if he is well known for music written a hundred or two hundred years ago, that would be a high time quotient. In the field of poetry, the two poems by Homer, the Iliad and the Odyssey, are nearly three thousand years old and lasting nicely, giving them the highest time quotient ever. If you want a fun exercise, go back to the music of your early life and listen to it again. Some will wear well while others will sound very dated and even others crude. I started really listening in the 60s so here are three examples:

Wow, that's a loooonnnngg 1'52.

There's a few things going on there: a melody, background harmony and a couple of almost interesting chords.

What can I say--after less than fifty years, they stand out. No cliches, very original. Good stuff.

With all the riches of YouTube, you could perform this exercise for almost any decade in the last 200 years.


RG said...

You explain your idea for a time quotient (TQ). You later introducce the idea of a Boring Quotient, BQ.

The BQ is {time in minutes from beginning of performance to moment of first boredom} divided by {time in minutes of performance}. I think the math for a BQ is straighforward.

But there are problems with measurement for a TQ.

First of all, any definition of "known" or "well known" would be problematic. But it should be possible to define a binary opposition by stipulating that, for example, a work is "well/known" in a given year iff in that year that work was performed in such-and-such a sufficiently significant venue or was published in some paper or plastic or electronic. Arbitrary but amenable to consensus. You could, in theory, take a similar approach to "Well" known, by counting various kinds of manifestation of knownness and weighting them.

But this is a "time" quotient. So what time would serve as the dividend (top part of the fraction)? It cannot be the {time from composition to time of obscurity}, because it might begin in obscurity and only become "known" (much) later. Alternatively, you might begin the period of knownness again from an arbitrarily defined moment like first performance in one of those sufficiently significant venues. So the dividend is {time in years from first becoming known to present}. But what if the work only recently became (in our sense technically) "unknown" (a situation rapidly submerging most classical music), but there was a period of 200 years in which the work was "known" every year? You could make it {time from first being known to last being known}. But what if the work was known during two periods of 100 years each, say from 1600 to 1700 and from 1850 to 1950, but not otherwise? It would seem you would have to check out each year and see whether in that year the work was known or not, then sum all those to serve as the dividend. OK, now the divisor (bottom of the fraction). I don't think you can just use {years from composition to present}; because sometimes a work is lost in the nachlass and not discovered for centuries. So, I think, given the dividend measures, you have to measure the divisor as {years from first publication to present}.

So, the TQ would be {years in which known} divided by {years since publication}.

Not impossible (in theory), but more complicated than the BQ. But wait! The BQ divisor {performance time} is unambiguous enough. But the divident mentions "moment of first boredom". What if a 5 min work gets boring at 2:05 but perks up again at 2:25 and stays absorbing right up to the 5:00 mark? Hoo boy...

Bryan Townsend said...

Wow, great comment. Is this a case of "that's all right in practice, but doesn't work in theory"? What I did in the post was think back to the bands that were well known at the time of the first "British Invasion" around 1964 to 1966 and compare three of them. Two did not well survive the test of time, some forty-five years, while one did. Fairly simple, though depending, of course, on a whole set of assumed aesthetic standards that were not expressed. But then I tried to finesse it by throwing in the idea of a "Time Quotient" parallel to my earlier "Boring Quotient" but, as you point out, the whole idea is problematic--as is the BQ as well, depending as it does on the sensitivities of a particular listener.

I think I had in the back of my mind some sort of exercise such as Charles Murray attempted in his book Human Accomplishment which tracks outstanding thinkers and artists from 800 BC to 1950 AD. But as you say, composers go in and out of popularity so that should be taken into consideration.

I still think the exercise is interesting. Pick a decade or two, find out who were the outstanding composers/musicians at the time and compare them now with the perspective of time. In that case you are only dealing with two slices of history.

RG said...

Please forgive my having got carried away with the metrics. It is what I do.

You ask Is this a case of "that's all right in practice, but doesn't work in theory"?

No, it is more "easier said than done".

But I did not want to suggest that they could not be done, just that doing the TQ might be more complicated than the BQ.

Also, far from objecting to the subjective character of the BQ, I think it is an excellent suggestion for sharing and comparing aesthetic judgements. Though, you might improve its applicability by calling it an Attention Quotient or Enjoyment Quotient, so as to keep it neutral and to support such judgements as your (mathematically impossible but rhetorically effective) hyperbolic score for Strawberry Fields of 500!

Bryan Townsend said...

I was mistaken, I posted the Time Quotient one before the Boring Quotient one, even though I wrote the BQ one first.

The BQ expressed the percentage of the piece one can listen to before boredom sets in. A quotient of "1" means you can listen to it once. A quotient of "500" therefore just means you can listen to it 500 times. Perfectly possible.

Back when I used to play a lot of scales I once calculated that I had played G major three octaves roughly 250,000 times. But it was boring!

Marc Puckett said...

I reach the BQ very rapidly when it comes to contemporary popular music. Was born in '57. That was the first time I ever actually listened to the entire HH song (but the memory plays its games), although I recognised it, at least aurally, if that's what I mean.

Bryan Townsend said...

Me too! And that might be the reason that the video is so important now: most of the music is pretty boring.

Christine Lacroix said...

The Dave Clark 5 bit was brutal! The three pieces illustrated your idea perfectly.

Bryan Townsend said...

Don't they just!