Sunday, May 31, 2020

Additive Rhythm, and Meter

I have been wrestling with some problems with notating rhythms recently. Perhaps I should say I have been exploring rhythmic possibilties? In my String Quartet No. 2, I decided that I wanted a particular effect in the first movement where the last beat of the measure would be slowly expanded to give a sort of limping meter. At first I wanted to do this with just a text instruction: "Repeating the chordal passage sixteen times, expand the last beat of the measure from a quarter note value to a half note value incrementally." My violinist suggested notating this precisely so I ended up with this:

Click to enlarge
That is just part of the passage. Each measure is repeated four times. The last beat of the measure grows from a quarter note, to a quarter tied to a sixteenth, to a quarter tied to an eighth and so on, until the last beat becomes a half note. The latter part of this process is shown above. Now there is no doubt that this is notationally ugly! I soon realized that the composer that had really explored this technique, known as "additive rhythm" was Olivier Messiaen. What inevitably happens if you use additive rhythms is that you destroy the meter; this kind of music is "ametric." I discovered this looking at Messiaen scores and seeing that he simply omits meter entirely. Here is an excerpt from his Catalogue d'oiseaux, "Le Chocard des Alpes."

Click to enlarge
Even where, as in the beginning of the piece, the meter is clearly 2/4, he shows no meter and there are many passages like this where there is no regular meter. Unfortunately, not showing or more importantly, not having a regular meter, is not an option in my music software, so for my most recent piece, I simply notate everything in 4/4, but the bar lines are actually irrelevant. In this passage the violin is doing a "subtractative" rhythm while the guitar does an additive one:

Click to enlarge
Deal with that, copyright bots!

4 comments:

Maury said...

As you know the time signature describes a pattern of strong and weak beats in the measure. Therefore it is not good practice to use a 1 in the numerator because it cannot define such a pattern> It is better to subdiviu=de down to the level you need to resolve the pattern. So for example 16/16 to 17/16 to 18/16. Of course you start with 18/16 ...

Bryan Townsend said...

Yes, that would be another way of writing it, and perhaps a better one. But as I suggested in the post, the use of additive rhythms tends to make the music ametric. In other words you really don't have that pattern of strong and weak beats. Instead you have variable durations.

Maury said...

I am a bit unclear as to what you are really driving at since you through the curveball of ametric music into this. It is very difficult to write ametric music and the string quartet players will be resistant to it. Yes whenever there is no pattern the time signature should be left out and just a tempo indication of some kind left or probably better a time interval for a certain section to be played in as Bartok did.

It is unclear from the example whether you want that passage to be without meter. However even without time signature the players are going to impose one when they practice. Since each short phrase is to be repeated it forms a kind of structure on it from the start. So if it were me I would stick with the first example and use 16 in the denominator.

Bryan Townsend said...

I didn't show the whole passage, but I kind of like the idea of using a 16/16 meter, then 17/16 etc. Not sure if the players will like it, but it will at least the meter will be less unsightly.