polyrhythms

[Watto]

herodotus mentioned it in the welcome speech...

It's one area of music which I'd love to explore more - I've got a drummer mate called Ken Edie (google him if you could be bothered - he is an absolute monster) who has said he'll teach me, but we never seem to find the right moment to do it.

Reckon you geniuses could run me through some basics, and then run me through some harder stuff, and then let me lie down for a while?

[herodotus]

OK

The most basic form of a polyrhythm is the hemiola, which the Oxford Dictionary of Music defines as 'a rhythmic device that consists of superimposing 2 notes in the time of three or three in the time of two'.

for instance:



This basic idea is at the core of all polyrhythms:

x units with a metrical value of y and y units with a metrical value of x superimposed on each other.


And so here is 3 against 4 in standard notation:



And in TUBS:



And 4 against 5 in standard notation:



And in TUBS:



Now this is the most basic type of articulation of a polyrhythm. There are much more elaborate ways of employing these ideas.

But we will leave those for part 2

[xauf.alsia]

not to get ahead of anyone, and simply because i'm not sure if this is a polyrhythmic thing there's also the idea of "rhythmic phase". this is sort of like manual looping where the loops are designed to fall into and out of sync while staying sensible.

as an example, you'd take a 4/4 measure of quarter notes and lay it over a 5/4 measure of quarter notes, they will meet up every 20 measures.

[herodotus]

not to get ahead of anyone, and simply because i'm not sure if this is a polyrhythmic thing there's also the idea of "rhythmic phase". this is sort of like manual looping where the loops are designed to fall into and out of sync while staying sensible.

as an example, you'd take a 4/4 measure of quarter notes and lay it over a 5/4 measure of quarter notes, they will meet up every 20 measures.

Yep.

Same phenomenon, different name.

The fact is that musical lexicographers are kind of behind the times on this stuff.

The thing you are talking about has the same structure, but with a more elaborate articulation than the examples given so far.

I will be covering all this thoroughly in the next few days.

And hello!

[Sascha Franck]

I already said so in the time signature thread, possibly *the* most famous polyrhythm in our common music styles has got to be a 3/16 pattern over an otherwise even beat. This can easily be expanded (to 5, 7, 9) and shifted around. Endless possibilities.
It's getting quite more tough for the performer when things such as the mentioned 5/4 pattern over a 4/4 beat are involved. But in that case, to learn those and get familiar with them, computers really are your best friend.
It's getting even easier with Logic, in which I can just shorten a looping part and it'll continue looping but be shortened all throughout. Instant polyrhythmic madness.

[VicDiesel]

The most basic form of a polyrhythm is the hemiola, which the Oxford Dictionary of Music defines as 'a rhythmic device that consists of superimposing 2 notes in the time of three or three in the time of two'.

I hate to argue with anythting with Oxford in the name, but this is not the definition I'm used to. The one I know is in wikipedia:

"In modern musical parlance, a hemiola is a metrical pattern in which two bars in triple time (3/2 or 3/4 for example) are articulated as if they were three bars in duple time (2/2 or 2/4)."

So it's a sort of polyrhythm, but the notes don't change speed: it's more like a change in phrasing. You start with a rhythm in three "One! two three One! two three", and you phrase it as "One! two Three! one Two! three". (See the Mozart example in Wikipedia.) But the speed of the beat does not change.

Victor.

[VicDiesel]

Same phenomenon, different name.

Hm. Not really the same. "Sort of" the same.

You were desribing 1/ take a measure 2/ normally divided in 4 3/ divide it also in 5.

This is more like 1/ take a measure, normally divided in 4 beat 2/ take 5 of those giving 20 beats in 5 groups of 4; 3/ now divide that 20 in 4 groups of 5.

The difference is that in your first example you hear a group of X and a group of Y, with only the first note coinciding. In the second case X and Y are the same (for instance 20) but contain differeent internal accents.

Victor.

[herodotus]

Same phenomenon, different name.

Hm. Not really the same. "Sort of" the same.

You were desribing 1/ take a measure 2/ normally divided in 4 3/ divide it also in 5.

This is more like 1/ take a measure, normally divided in 4 beat 2/ take 5 of those giving 20 beats in 5 groups of 4; 3/ now divide that 20 in 4 groups of 5.

The difference is that in your first example you hear a group of X and a group of Y, with only the first note coinciding. In the second case X and Y are the same (for instance 20) but contain differeent internal accents.

Victor.

Actually, that is what I meant when I said that they were different articulations of the same polyrhythm.

If you were to sound only the first beats of each group in your hypothetical 20 beat phrase you would get the exact same thing as the pictured rhythm.

I use the jargon term 'articulation' to mean 'the actual sonic realization of the polyrhythm idea'. These sonic realizations can sound very different from each other, just like a C major chord can sound very different if it is played, for example, first by a symphony orchestra and then by a ukulele.

But the C major chord has an intelligible reality that informs all of these actual sounding C major chords, but is separate from them.

I was using 'polyrhythm' in a sense precisely analogous to this use of 'chord'.

[VicDiesel]

If you were to sound only the first beats of each group in your hypothetical 20 beat phrase you would get the exact same thing as the pictured rhythm.

Right. But the original point was about different lenght phrases joining up every so many measures. Reducing that phrase to a single note is rather a special case.

But yeah, there are two ways of coming to the same concept.

Victor.

[herodotus]

But yeah, there are two ways of coming to the same concept.

Victor.

There are myriad ways of making a sound event out of a polyrhythm idea.

One can do it not only with percussive sounds, but with pitches, with tone color, with panning in a stereo field; one could make a polyrhythm using nothing but a single held note of a single string patch routed through 2 LFOs synced to different but related tempos; or with any combination of the above.

I admit that there is not much in the way of established terminology, here. But whatever you want to call it, this idea:

x units with a metrical value of y superimposed over y units with a metrical value of x, where x and y are natural numbers

is the structural kernel at the core of almost everything that gets called a polyrhythm.

[Watto]

So can you give us some things to practise? I'm mostly talking about playing jazz, and I guess it gets more complicated when you swing (you can do more intricate polyrhythms).

Cheers!

Watto

[xauf.alsia]

yeah, i have to say that a computer is an invaluable aid to learning polyrhythms. the only thing i wish i could find is a sequencer that could do a "flexible quantize" that might be able to quantize per phrase. is this what "groove quantize" does? i've always just entered the notes using step record by doing each line seperately and then starting a new track as a comparison....

[VicDiesel]

the only thing i wish i could find is a sequencer that could do a "flexible quantize" that might be able to quantize per phrase.

Doesn't Logic do that? Every track, and every pattern can have its own quantization.

Victor.

[rachmiel]

pop quiz: what famous song by what famous rock group has 4/4 playing against 3/4? hints: k, lz. (no fair looking it up on the web!) :-)

[funkadil]

One thing that is much fun is to, instead of doing them within its own signature, like 3 against 4 or 7 against 8. Instead of using base 2 notes it is fun to use a slicer or triplet quantization to achieve the same sounds but against the normal pulse, rather than within.