Rumba Clave - An Illustrated Analysis

[davidpenalosa]

"Rhythmic ratios" are just another way of saying keeping time... right?

Hi Congamyk,
Not exactly. These rhythms are generated from rhythmic ratios, the most fundamental being three cross-beats over two main beats. There are three even beats per two even beats, or three for every two (3:2).

X.X.X. three cross-beats
X..X.. two main beats

(if you put this in size 12 Courier font the parts will align properly)

two cells of 3:2 = one primary cycle

X.X.X.X.X.X. cross (six) beats
1..2..3..4.. main (four) beats

You can see that the cross-beats are generated by playing every-other subdivision (stroke-rest-stroke-rest-ect). In clave-based music patterns appear in their prime position and also in displaced positions. Notice that the cross-beats connect with main beats 1 and 3.

The rests within the six beat cycle constitute an offbeat six cycle: a displaced six cycle diametrically opposed to the one shown above.

.X.X.X.X.X.X off-beat six cycle
1..2..3..4.. main beats

The off-beat six cycle connects with main beats 2 and 4.

What happens if we strike three of the on-beat six, and then shift to the off-beat six?

X.X.XX.X.X.X
1..2..3..4..

We get the standard bell pattern. Rumba clave is a derivative of that pattern. Rumba clave has two on-beat six beats and three off-beat six beats.

It all begins with the ratio 3:2. There’s lots more to this of course, but when James cites the ratios 3:2 and 3:4 as a way to measure the displaced strokes, I couldn’t help but mention that’s an element related to clave’s genesis.

I’m not ready to say "what it means" in terms of displaced strokes, I’m just pointing out the significance of the ratios. Those ratios were first described by the ancient Greek philosopher/mathematician Pythagoras, but the proportions are well documented as laws of nature.

In harmony, the perfect fifth is generated by the ratio 3:2 and the perfect fourth, by 3:4.
Geekilly yours,
-David

[taikonoatama]

While the variations in the tracks may seem slight, both in terms of deviation from the 4/4 written marks in a given track and in track-to-track comparisons, I think those slight variations add up to something bigger in the context of a whole measure of clave, and then consecutive measures in a rhythm. As a whole, to *my ear*, the music-versions have a lot more in common with one another than any one of them does with the 4/4 written-clave version. Why is that? I think it's related to intervals. Humans are able to discern relative time intervals quite well, especially when they're played right after one another, as with the 3rd, 4th, and 5th notes of clave, and the intervals between notes 3 and 4, and then 4 and 5. That's where the music tracks all really differ from the 4/4 written version.

I did the numbers for each track. As a comparison, keep in mind that the 3-4:4-5 Ratio for written 4/4 clave is 3:2 (1.5:1), and for 12/8 it's 1:1. And by "3-4:4-5 Ratio" I mean the elapsed time between the beginning of the 3rd note of clave and the beginning of the 4th, compared to the elapsed time between the 4th and 5th.

Isaac Delgado track at 98bpm - 3-4:4-5 Ratio: 1.22:1
Rumberos_de Cuba track at 100bpm - 3-4:4-5 Ratio: 1.23:1
Rumboleros track at 102bpm - 3-4:4-5 Ratio: 1.17:1
Carlos Embale track at 116bpm - 3-4:4-5 Ratio: 1.14:1
Afrekete track at 122bpm - 3-4:4-5 Ratio: 1.17:1
Los Parragueños track at 122bpm - 3-4:4-5 Ratio: 1.11:1
Los Munequitos track at 138bpm - 3-4:4-5 Ratio: 1.08:1
Pancho Quinto track at 144bpm - 3-4:4-5 Ratio: 1.18:1
Clave y Guaguanco track at 156bpm - 3-4:4-5 Ratio: 1.08:1
Ecue Tumba at 168bpm - 3-4:4-5 Ratio: 1.18:1

Obviously it would be great to have multiple measures of a given track illustrated, to see how much variation there is in a track and maybe average it out, but I don't have a track well-suited to that purpose yet. In spite of that potential issue, all 10 tracks/measures do fall within a fairly narrow range, and none of these are even close to the 1.5:1 ratio of straight 4/4 written clave (though several are getting pretty damn close to a 1:1 12/8-like ratio). For me, it's another piece of the puzzle falling into place.

Addendum: If the deviations from 4/4 written clave we see in a given track are there purely due to the natural variance we might expect in human playing (a person trying to play pure 4/4 straight clave but failing slightly), then we'd expect to see some ratios here greater than 1.5:1. The fact that we do not see anything even remotely close to 1.5:1 tells me that's clearly not happening.



Edited By taikonoatama on 1201135271

[blango]

I just love the academic nature of this discussion. Seriously advanced study of the science of Clave.

Thanks James, David and others!

If the goal is to play clave like the significant Rumba troupes, one can try the following. It really works and cuts to the chase. Forgive me for a repeat here, but i think its worth mentioning.

play M de M, Los Papines, Clave Y Guaguanco or some top family over your stereo, play clave over it to see if you can make the clave on the track 'disappear'.

When you can, then play the same track on headphones, play clave to it and record yourself. Listen to the recording and make adjustment. Try again till you sound like you groove solidly. Its incredible how your ear can hear when you are not grooving, or playing like a Cuban.
This may take months, it may take weeks, but you wont get it quickly. patience.

Somehow the ear can tell. One would think, if i can hear it when its wrong so easily on a recording of myself, why didnt i play it right the first time!! funny how that works.

ok, after doing that till it clicks...

Take your clave to the street and check it with Cubans from Cuba.

Another thing i do is think of clave in the melody of rumba.

DagadaGadagadaGadagaDagaDagadaga at a minimum for me with feel applied. 16ths.

My guess is that the feel could be covered by writing it out in 64ths, if anyone is into it.

You have to love the use of technology to tackle the idea of 'feel' !

have fun,

Tony



Edited By blango on 1201132485

[blango]

subscript.

there are many 'right' ways to play clave. Like the first post of this string. there are several 'feels' that are commonly acceptable to the master drummers, and thus most everyone else as well.

Keep in mind. There are many acceptable feels, but there are infinite wrong ways to play clave!

These acceptable 'feels' can be found in the tracks of the top Troupes, unless you happen to grow up in the streets of Havana, or have access to master Cuban drummers.

Peace,

Tony

[davidpenalosa]

James,
I think that my previous post, where I digressed on the subject of rhythmic ratios, may be irrelevant. I see that you did not come up with either the 3:2 or the 3:4 ratio. Ooops, sorry. Never mind…

Maybe someone benefited from the ‘lesson", despite its lack of relevance to this topic.

Here’s a bit of info that may have some relevance to the subject at hand - Kevin Moore (timba.com) has midi files of timba drummers playing many consecutive cycles of rumba clave. There is a certain amount of inconsistency in where those strokes fall, cycle to cycle. This is just the natural degree of variation though, proving that people are not as metronomecly as precise as midi machines. Still, the human ear perceives a certain range of preciseness as "dead-on". In other words, that degree of variance recorded in midi is not perceptible by the human ear (brain). It sounds solid.

I think that some of those slightly displaced clave strokes in your graphs may be so close to being dead-on that they are heard as in the "grid". What degree of displacement does the ear detect? I don’t know. I have two books on how the brain perceives music, but they both proved to be of no help on this.

Interestingly, although the midi clave and other midi data (keyboards, etc) Kevin recorded showed considerable metronomic variance, he says that all the strokes were played slightly ahead of the beat. So, in that case, ahead-of-the-beat is the main "feel". I think that’s an aesthetic of popular music that is not always appropriate for folkloric music, as evidenced by your graphics.
-David

[taikonoatama]

Great tips, Tony, thanks.

I think the part about recording yourself is particularly good. Painful, but good.

Naturally I've taken it to the extreme of taking that recording of myself, graphing it as I did on the site, and then comparing that graph to those I pulled from the rumba recordings. I'm a very visual learner, and it's helped me see where I need to make adjustments.

I may enjoy breaking all of this down, but ultimately I'm just trying to get the right feel down and play it without thinking of any of this stuff. We all learn a bit differently. It's gotta be internalized at some point to really come alive and be in touch with everything else going on. Ultimately that's what I'm aiming for.

[taikonoatama]

I think that some of those slightly displaced clave strokes in your graphs may be so close to being dead-on that they are heard as in the "grid". What degree of displacement does the ear detect? I don’t know. I have two books on how the brain perceives music, but they both proved to be of no help on this.

Could you clarify your usage of "grid" here, David?

[davidpenalosa]

By "grid" I mean a precise subdivisions of the beat, as illustrated in your graphics by red (thirds) and blue (quarters) vertical lines. A stroke played precisely on a subdivision would align with the grid, whereas a displaced stroke would not. My question was how offset from the "grid" does a stroke need to be in order to be perceived as displaced?
-David