[ternarizator]

Hello,

I'm back with a new specific subject. In june 2015, I proposed my idea of clave-expanding, in order to try and explain the origin of clave-patterns that are often underlying west-african music. The idea was to consider a rhythm made of an odd number of strokes (typically 3 or 5), played twice alternating hands, and keeping only the pattern of the "strong" hand. My hypothesis was that some musicians would play with a stick in the strong hand, the other hand remaining bare, and so the rhythm of the strong hand would emerge.

Later, I decided to go further in my study of Gnawa music, that I had begun fortuitously in the late 80's. My aim was, among others, to study LSM swing in this music (I'll go back on this a bit later).

During my investigations, I stumbled upon videos of "desert Gnawas" (Khamlia, Morocco) who played ganga drums (it's the name of the tbel drum they use) exactly the way I had supposed.

Here are two videos to show this : this one and this one.

While qraqebs play an ostinato (approximately q . q q q . q q q . q q etc.), the "little" drum plays the same rhythm : o . o * o . o * o . o * etc., where "o" is an open stroke on the head of the drum with a bent stick, and " * " a slight stroke of the bare hand at the periphery. The big drum plays the same rhythm, but alternating hands : c . * o * . c * c . * o * . c *, where c is a closed (muffled) stroke in the center of the skin.

To summarize, you obtain :

1 . . . 2 . . . 1 . . . 2 . . .
o . o * o . o * o . o * o . o *
c . * o * . c * c . * o * . c *

Gnawas often play between quaternary and ternary, so it approaches the following :

1 . . 2 . . 1 . . 2 . .
o o * o o * o o * o o *
c * o * c * c * o * c *

So I consider this validates my hypothesis, and I intend to go in greater depth in this direction.
(Gnawa music is only the starting point, because as far I know they don't play clave patterns)

(To be continued)

Vincent.

[ternarizator]

So I decided to deepen the question methodically.

My first task was to explore the first pattern : quaternary o . o * or ternary o o *.

By avoiding the second stroke, you can obtain o . . * or o . *, which is appropriate to "play the downbeat", especially for the ternary one.

You can also play alternately with an open and a closed stroke :

o . . * c . . * or o . * c . *

Or with two open strokes :

o . o * c . . * or o o * c . *

Or, by switching the two beats :

c . . * o . o * or c . * o o *

This led me to discover how to play the famous kenkeni-pattern c .(x). o . o . with the handing of the quaternary passport, which I find rather remarkable (I'm very impressionable...) :

s . . s s . t t / c . . * o . o *
r . . l r . r l / r . . l r . r l

The second thing about it is that the ternary version c . * o o * is much easier to feel in the right place for a beginner than c(x). o o ., which is most of the time understood with bell and second "o" on the downbeat.

Should I add sound samples or is it already enough readable ?

(To be continued)

Vincent

[djembefeeling]

I don't need soundsamples. It's all plain and clear to this point.

[ternarizator]

OK let's go further...

As a variant, you can play alternatively the "tresillo" version and the other (I have chosen to place the tresillo [resp. 3/2] at the beginning) :

1. (quaternary)

1 . . . 2 . . . 3 . . . 4 . . .
o . * o * . o * o . o * o . o *

2. (ternary)

1 . . 2 . . 3 . . 4 . .
o * o * o * o o * o o *

At this point, the conclusion is that you can obtain the well-known pattern :

1 . . . 2 . . . 3 . . . 4 . . .
x . . x . . x . x . x . x . x .

and the corresponding ternary one :

1 . . 2 . . 3 . . 4 . .
x . x . x . x x . x x .

without refering to the concept of clave. (I'll nevertheless revisit it a bit later)

A little modern example of the latter on a zen drum (with LSM swing) (the rhythm stabilizes at about 0:30) : here.

(To be continued)

Vincent

[Carl]

I feel like I might be missing something here...

I'm at work and can't listen to the recordings at the moment.

However, looking at the notation, are you implying that what they are playing is somewhere between the two notations? or are you theorizing that the two patterns are results of different interpretations of an underlying "clave"?
I could see a relationship in either interpretation.
In my observations, the "swing" / micro-timing of the

Code: Select all

1 . . 
    ^

position is a common point of inflection. Often used to add a subdivision ambiguity.

I also wonder how people from the Gnawa culture think of the ambiguity? how do they teach the pattern and what degree of variation of micro-timing is common?

Anyway interested in hearing more. (like, what region are the Gnawa from?)

Thanks,
Carl

[ternarizator]

I had planed to talk about the feeling a bit later, but I feel it's the moment to explain a little.

Gnawas play either a true quaternary pattern (known in western poetry as "dactyl"), in case of low tempos :

1 . . . 1 . . . 1 . . . 1 . . .
q . q q q . q q q . q q q . q q

or a "semi-partially ternarized" one, especially at fast tempos, where every second stroke (in red in the following) is slightly delayed, its position between flat ternary (33.33 %) and "binary" (50 %), so about 40 %. The third stroke is approximately in place (66.66 %), resulting in a LSM swing :

1 . . 1 . . 1 . . 1 . .
q q q q q q q q q q q q
. → . . → . . → . . → .

here a little sound sample (made with Percussion Studio).

One can modelize the phenomenon by writing it by groups of 12 (not an underlying subpulse !):

1 . . . . . . . . . . . 1 . . . . . . . . . . .
q . . . . . q . . q . . q . . . . . q . . q . . (quaternary)
q . . . . q . . q . . . q . . . . q . . q . . . (semi-partially ternarized)
q . . . q . . . q . . . q . . . q . . . q . . . (flat ternary)

Only one stroke is displaced in comparison with the pure ternary, whereas two strokes seem to be translated (geometric sense) in comparison with the pure quaternary, with a constant interval of 25 %.

[This is only a model, and of course this pattern is likely to be very flexible (see the work of R. Jankowsky about Stambali, the tunisian "Gnawa" ritual)].

Gnawa musicians consider these two patterns as one and the same, but with a different feeling. As they play qraqebs with no volume nuances, a really flat ternary pattern q q q q q q q q q q q q would absolutely not express the downbeat, whereas using this trick of delaying the second stroke, the downbeat is easily perceivable : ternary with advantages of quaternary...

Most of Gnawa musicians use alternancy, playing the qraqebs this way :

r . l r l . r l r . l r l . etc.,

but the desert Gnawas I am refering to use the other way :

r . r l r . r l r etc.

Gnawas' ancestors come from ancient "western Sudan", and were for example Bambaras or Hausas, among others. I once had a video of Hausa musicians who played with the same feeling, but no way to put my hand on it again today.

There is another rhythmic pattern used by Gnawas (and Hausas too), but I prefer to discuss it later.

I will not have much time these days, for I have to correct a lot of exam papers (baccalauréat) in a rather little period, so don't be surprised if I can't react very soon.

(To be continued)

Vincent

[ternarizator]

(Back to the main subject)

I then tried to experiment every simple manner of playing the basic pattern of the qraqebs with the ganga drums (I also built 3 ganga drums using drumshells, mounted like dununs). My cogitations led me to the following :

1 . . . 1 . . . 1 . . . 1 . . .
o . * o o . * o o . * o o . * o

and of course the "ternary" version :

1 . . 1 . . 1 . . 1 . .
o * o o * o o * o o * o (feeling LSM of course)

And also to :

1 . . . 1 . . . 1 . . . 1 . . .
* . o o * . o o * . o o * . o o

and its "ternary" counterpart :

1 . . 1 . . 1 . . 1 . .
* o o * o o * o o * o o (LSM)

The first one can then be modified, for example into :

1 . . . 2 . . . 1 . . . 2 . . .
o . * o o . * . o . * o o . * .

The "ternary" version happens to be very interesting :

1 . . 2 . . 1 . . 2 . .
o * o o * . o * o o * . (LSM)

Indeed, one can easily parallel it with e.g. b s t t s .,

b and t in place of "o" and s instead of " * ", or see an ambiguity as in a dunumba rhythm :

1 . . 2 . . 1 . . 2 . .
* . o * o o * . o * o o (SML)

The second one can be modified the following ways :

1 . . . 2 . . . 1 . . . 2 . . .
* . o o * . c . * . o o * . c .

The ternary version goes so :

1 . . 2 . . 1 . . 2 . .
* o o * c . * o o * c . (LSM)

With 2 ganga drums, one can easily re-create the accompaniment of "Soli des Manians" :

b t t b s . (or Moribayassa for the quaternary version)

Another way of modifying is the following :

1 . . . 2 . . . 1 . . . 2 . . .
* . . c * . o o * . . c * . o o

Combining with one found above, you can obtain the following :

1 . . . 2 . . . 1 . . . 2 . . .
o . * o o . * . o . * o o . * .
* . . c * . o o * . . c * . o o

This gives a split quaternary passport. Of course you have a "ternary" version :

1 . . 2 . . 1 . . 2 . .
o * o o * . o * o o * .
* . c * o o * . c * o o

"What is the point ?", you may wonder... My aim is to explore possibilities of Ganga drums, following certain rules inspired by both the Gnawas' practice and my own experience in djembé and dununs music. Next time I'll come to clave(s).

(To be continued)

Vincent

[ternarizator]

Let's come to the heart of the subject : emergence of clave(s).

As I explained before in another thread, my idea is to generate the clave-pattern in the same way as for the tresillo :

(quick reminder) o . o o → o . * o|* . o *.

In this example, a pattern of 3 strokes / 1 beat turns into the tresillo, always 3 strokes, but now on 2 beats.

We can easily generate a 5-strokes pattern starting from the basic patterns I have already discussed above, i.e. o . * o * . o * and o . o * o . o * :

Played together (it's the case, among gnawa musicians, as already seen in my first post), it results in :

1 . . . 2 . . .
o . * o * . o *
o . o * o . o *

A simple listener would essentially hear the following : o . o o o . o .

This pattern, played (twice) with the "stick and hand" technique, turns into :

1 . . . 2 . . . 3 . . . 4 . . .
o . * o * . o . * . o * o . * ., which is the pattern aka "Son clave".

Of course we can do exactly the same with the ternary counterparts :

o * o * o * + o o * o o * = o o o o o .

and then o o o o o . → o * o * o . * o * o * . (compare Gidamba)

All this stuff had already been explained, but it could seem doubtfull at that time. I don't claim it's proven today, but it looks much more conceivable, doesn't it ?

I've been searching for a long time for a similar way to generate the "Rumba clave", but
unsuccessfully : it seems that the "generator", i.e. o . o o o . . o is quite rare or rather unlikely...

But recently I came across an "obvious" fact : "Nature hates vacuums" and there are vacuums in the
o * o * o . * o * o * . pattern (it's quite obvious for the ternary one, but still true for the quaternary one, which is "the same", but played in a more square manner).

So I asked myself what would happen if I filled the gaps in the previous pattern. Here is the answer :

1 . . 2 . . 3 . . 4 . .
o * o * o o * o * o * o

Miracle of the miracles, the "standard pattern" has been "automatically" generated by this way of considering the clave pattern... (Please, if all this was already clear for you, do not scoff)

So what is the connection with (ternary) rumba clave ? I'm coming to that...

Some patterns, e.g. tresillo, are "naturally" connected with a "bell-pattern", in this case cinquillo
x . x x . x x ., or are played in a manner that corresponds. In the french Antilles, they play
"ti bwa" (litterally "little wood", from french "petit bois"), that goes so, with one stick in each hand (blue for right, red for left) :

o . o o . o o . the right hand plays the tresillo, the left hand plays the "natural" complement, and it results in the cinquillo.

If we apply this way of playing to the standard pattern, the result is :

1 . . 2 . . 3 . . 4 . .
o . o . o o . o . o . o, where the right hand now plays the "ternary rumba clave", and the left one plays what becomes now the complement. I believe it's the way cuban musicians conceive the origin of clave pattern.

It seems like I'm now far away from the Gnawas, but even if I told they don't play claves (at least on the Ganga drums as far as I know), they play the standard pattern and also the "short bell"
x . x . x . x x . x . x, and of course x . x . x . x x . x x . on the Guembri, e.g. here...

Next time I'll come to the way of generating a lot of well-known rhythms of the west-african tradition, starting from the patterns described above.

However, I can also discuss a little bit about the connection that can be found to some extent between the standard pattern (or simply the clave) and the musical scale.

Any reaction ?

(To be continued)

Vincent

[Carl]

Very interesting...

I will try to review this later, no time these days...

I find this interesting and will give you my thoughts once I get around to organizing them.

thanks!
Carl

[djembefeeling]

Any reaction ?

Of course. First some minor points, to get warm...

compare Gidamba

Daniel refers to the rhythm as Dyidanba, and though I know it is commonly known as Gidamba from CDs and workshops and books, I think it is not to late to correct the old way of spelling. Dyi is the Malinké word for water, dan means to get sth. and ba can denote big (or great), mother, but also river. So probably it means the great getting of water or getting water from the river and probably stands for a ceremony before the wedding, when the prospect wife is symbolically washing a piece of cloth for the prospect husband to show everybody that she will be a good wife.

o * o * o * + o o * o o * = o o o o o .
and then o o o o o . → o * o * o . * o * o * .

You omit the * on position 6 in both lines and thus interpret is as a blank in the addition. Only by this minor change you can arrive at the wanted son-clave, otherwise the handing would only allow for two tressilos.

"Nature hates vacuums" and there are vacuums in the
o * o * o . * o * o * . pattern (it's quite obvious for the ternary one, but still true for the quaternary one, which is "the same", but played in a more square manner).

I thought our cosmos is amost nothing but vacuum
But to be serious, I wouldn't base it on physics. Timelines need to avoid more than one pulse pauses because humans have different personal timings and it is getting harder to accord with one anothers timing when there are intervals of more than one pulse.
Why do you say that those vacuums are obvious in the ternary clave and still true in the quaternary? I'd say that they are much more obvious in the quarternary.

My main critique is on your method in general. I can follow every step that you are doing and can see how you arrive at the claves from your basic assumptions. What I cannot see is any necessity. You could generate different patterns from the ones you assume, but why should it be like that? At almost every step there are different choices you could make but you always choose the ones that fit into the direction you want to go. Do you think it has likely developed that way historically or that is an exigency of formal thought?

I think you did read Eugene Novotney: The 3:2 Relationship as the Foundation of Timelines in West African Music. His account has more formal necessity IMO.

From my own account I think of a measure of four beats as a unit of two different parts, like call and response, or rather like fadenya and badenya which are constantly struggling for a balance. Your model strives for a simpler account where the difference of the two sides of a measure just results from alternate handings (which some do and some don't do). Although simplicity is a scientific value in model building, it does not always catch the essence of the object under scrutiny. I think the basic cultural values do reflect in the music of the Malinké.

[ternarizator]

OK for Dyidamba, I'm not at all a specialist in Malinké language.

ternarizatorwrote:
o * o * o * + o o * o o * = o o o o o .
and then o o o o o . → o * o * o . * o * o * .

You omit the * on position 6 in both lines and thus interpret is as a blank in the addition. Only by this minor change you can arrive at the wanted son-clave, otherwise the handing would only allow for two tressilos.

A few lines above (for the quaternary), I had written "A simple listener would essentially hear the following :", so for me there is a back and forth between playing and listening. The " * ", essential in this way of playing, are much less perceivable than the "o".

ternarizator wrote:
"Nature hates vacuums" and there are vacuums in the
o * o * o . * o * o * . pattern (it's quite obvious for the ternary one, but still true for the quaternary one, which is "the same", but played in a more square manner).

I thought our cosmos is amost nothing but vacuum
But to be serious, I wouldn't base it on physics. Timelines need to avoid more than one pulse pauses because humans have different personal timings and it is getting harder to accord with one anothers timing when there are intervals of more than one pulse.
Why do you say that those vacuums are obvious in the ternary clave and still true in the quaternary? I'd say that they are much more obvious in the quarternary.

It's obviously an analogy, not the basis of my analysis... However, in their basic patterns, Gnawa musicians tend to follow as closely as possible the pattern of the qraqebs, so filling the gaps makes sense for me. The quaternary clave + " * " shows more gaps, due to the "forbidden" positions (the "2" of each beat), but there are only two allowable positions left for additionnal strokes, resulting in a "quaternary standard pattern", already mentionned by David Peñalosa.

1 . . . 2 . . . 3 . . . 4 . . .
o(.)* o *(.)o . *(.)o * o(.)* .
o . * o * . o o * . o * o . * o

My main critique is on your method in general. I can follow every step that you are doing and can see how you arrive at the claves from your basic assumptions. What I cannot see is any necessity. You could generate different patterns from the ones you assume, but why should it be like that? At almost every step there are different choices you could make but you always choose the ones that fit into the direction you want to go. Do you think it has likely developed that way historically or that is an exigency of formal thought?

I think you did read Eugene Novotney: The 3:2 Relationship as the Foundation of Timelines in West African Music. His account has more formal necessity IMO.

From my own account I think of a measure of four beats as a unit of two different parts, like call and response, or rather like fadenya and badenya which are constantly struggling for a balance. Your model strives for a simpler account where the difference of the two sides of a measure just results from alternate handings (which some do and some don't do). Although simplicity is a scientific value in model building, it does not always catch the essence of the object under scrutiny. I think the basic cultural values do reflect in the music of the Malinké.

I don't try to show necessity. If there were necessity inthere, people would play the same music on the whole planet, which is obviously not the case. As you already know, my goal is to find a simple (the simpliest ?) way to (re)build the fondamentals of african music, starting from nearly nothing, in order to make appear a theory with a maximum of coherency. It may be the historical way it was built, or not. It could also be interesting for teaching (I'll probably test it next year).

I don't see the problem with call and response, or fadenya and badenya : my first examples where precisely based on this o . o . o . o o . o o . , and the way I make clave appear is also an example of "call and response", the response being the same as the call, but played the other way. I think this approach can complete yours, with which I do agree, as you know.

As I already mentionned in another post (long time away), the same phenomenon seems to have taken place in north-african music (west Algeria) : they play s o o s o . on a drum called Guellal, and at the same time, someone plays on the Tbila the corresponding clave :
o . o . o . . o . o . ..
The other rhythm (quaternary) goes so s o . o s . o ., and someone plays on the Tbila the corresponding "clave" : o . . o . . o . . o . . o . . .. I find it hard to believe in a coincidence.

I did read Eugene Novotney, but I wasn't totally convinced. Maybe I should read it again.

OK, time to have lunch...

(To be continued)

Vincent

[djembefeeling]

The " * ", essential in this way of playing, are much less perceivable than the "o".

Admitted.

I don't try to show necessity. If there were necessity in there, people would play the same music on the whole planet, which is obviously not the case.

Nay, I don't mean that kind of necessity, but a necessity that the results of your theory follow with rigor from your basic principles.

Though I do wonder if I understand what you are trying to do. Perhaps I did made up the wrong alternative. I thought you are trying to show how the music might have developed historically from easy patterns to the more complex ones of today, quite like linguists try to show how the difference in languages trace back to some historical origins in more ancient languages.
Then I thought you try to show in a theory how all the different patterns of West African drumming do logically follow from basic first patterns and some formal rules of building things up.
Now my impression is that you rather want to show just a practical way of building complex patterns from easy ones - perhaps for the purpose of easy teaching?

my goal is to find a simple (the simpliest ?) way to (re)build the fundamentals of african music, starting from nearly nothing, in order to make appear a theory with a maximum of coherency. It may be the historical way it was built, or not. It could also be interesting for teaching (I'll probably test it next year).

O.k., here we go, I read it again and you just say it. Maximum of coherency is what I meant with necessity. But I still don't get it. Please pardon my stupidity here, I think this is my missing link that makes it hard for me to get what you try to show. I am thinking out loud here.

What exactly do you think are the fundamentals of African music? The complex patterns like claves, timelines, and prevailing accompaniments and solo pattern? Or the |o.oo|/|*.*o|/|o.o*| and other variants of this pattern? Are you trying to show how the molecules of African pattern based music are all build from the same atoms? What is the profit of your theory?

I think this approach can complete yours

That does sound promising

As I already mentionned in another post (long time away), the same phenomenon seems to have taken place in north-african music (west Algeria) : they play s o o s o . on a drum called Guellal, and at the same time, someone plays on the Tbila the corresponding clave :
o . o . o . . o . o . ..
The other rhythm (quaternary) goes so s o . o s . o ., and someone plays on the Tbila the corresponding "clave" : o . . o . . o . . o . . o . . .. I find it hard to believe in a coincidence.

I also find it hard to believe in coincidence. Is the corresonding clave o . . o . . o . . o . . o . . . a typo? Did you mean the son clave or really this 3 versus 4 movement?

[ternarizator]

Though I do wonder if I understand what you are trying to do. Perhaps I did made up the wrong alternative. I thought you are trying to show how the music might have developed historically from easy patterns to the more complex ones of today, quite like linguists try to show how the difference in languages trace back to some historical origins in more ancient languages.
Then I thought you try to show in a theory how all the different patterns of West African drumming do logically follow from basic first patterns and some formal rules of building things up.
Now my impression is that you rather want to show just a practical way of building complex patterns from easy ones - perhaps for the purpose of easy teaching?

The problem is that my brain tries to pursue these 3 goals at the same time, resulting in a superposition of states, just like Schrödinger's cat, and this brings ambiguity in my speech...

As you know, the "hand and stick" phenomenon was entirely theoretical for me a few months ago, and my research about Gnawas led to the finding that, at least for the tresillo / 3 against 2, it was a reality. Stimulated by this confirmation of my guesses, I began to explore the field of possibilities that it opens, by practicing this way of playing. I think I have respected a certain number of rules in my exploration, but it is not so easy to write them down to the paper. For example, there can be one or two successive "o", and never two successive " * ". Also the " c " can't be in pairs. And obviously no stroke at the forbidden place in quaternary, for my aim is to build ternarizable patterns.

By practicing, I discovered that the polyrhythms are much easier to set up, especially for the quaternary form (the ternary is a bit difficult when your friends doesn't perfectly master LSM feeling), because everyone plays nearly exactly the same underlying pattern |o . o o |, but with different strokes. This could really be a good approach for teaching. One can e.g. work on the bell technique (bell polyrhythm, I guess you see what I mean) at a wholly separate time, and put it together later, I think it would be worth trying.

An advantage of this technique is that you can play quaternary version of well known ternary ones without it appearing artificial (try to play a belle pattern which fits with the following : you'll be bound to insert a "xxx" o . * c * . o o * . c . c . * o)

Another purpose is to confirm that african rhythms can actually emerge from a very simple pattern which concretizes the beat, and to "reduce" (scientific sense) their apparent complexity.

What exactly do you think are the fundamentals of African music? The complex patterns like claves, timelines, and prevailing accompaniments and solo pattern? Or the |o.oo|/|*.*o|/|o.o*| and other variants of this pattern? Are you trying to show how the molecules of African pattern based music are all build from the same atoms? What is the profit of your theory?

Just like Rainer Polak I guess, I actually see |o . o | (LS) and |o . o o | (LSS) as "atoms" that underlie the molecules of african and apparented music, and I am focusing on the latter in my research. Of course, complex patterns are at the heart of my investigation too, since I try to build convincing (not so much it seems) bridges between the atoms and the molecules. The profit of my theory is to suggest an explanation in a very few steps of how to move up from "the beat" to e.g. "the standard pattern" (or whatever you name it), which has already caused rivers of ink to flow, through the concept of clave, that could be a key subject in this field of investigation.

I also find it hard to believe in coincidence. Is the corresonding clave o . . o . . o . . o . . o . . . a typo? Did you mean the son clave or really this 3 versus 4 movement?

No typo in there ! If you start from the generator :

1 . . . 2 . . .
s o . o s . o . (a typical pattern of the Oran region, west-Algeria. Of course you can find variants, but the skeleton is always this one)

and conserve every other stroke :

1 . . . 2 . . . 3 . . . 4 . . .
s(o). o(s). o .(s)o .(o)s .(o).

you get 33334 :

1 . . . 2 . . . 3 . . . 4 . . .
o . . o . . o . . o . . o . . .

and this is actually what is played on the tbila (a large bowl covered with camel skin, hit with a simple stick).

(To be continued ?)

Vincent

[djembefeeling]

The problem is that my brain tries to pursue these 3 goals at the same time, resulting in a superposition of states, just like Schrödinger's cat

Careful! That would mean your theory is half dead

As you know, the "hand and stick" phenomenon was entirely theoretical for me a few months ago, and my research about Gnawas led to the finding that, at least for the tresillo / 3 against 2, it was a reality. Stimulated by this confirmation of my guesses, I began to explore the field of possibilities that it opens, by practicing this way of playing. I think I have respected a certain number of rules in my exploration, but it is not so easy to write them down to the paper. For example, there can be one or two successive "o", and never two successive " * ". Also the " c " can't be in pairs. And obviously no stroke at the forbidden place in quaternary, for my aim is to build ternarizable patterns.

That clarifies your theory a lot...

One can e.g. work on the bell technique (bell polyrhythm, I guess you see what I mean) at a wholly separate time, and put it together later, I think it would be worth trying.

No, I don't know what you try to say here.

and conserve every other stroke :

1 . . . 2 . . . 3 . . . 4 . . .
s(o). o(s). o .(s)o .(o)s .(o).

you get 33334 :

1 . . . 2 . . . 3 . . . 4 . . .
o . . o . . o . . o . . o . . .

Ah, o.k.

To be continued ?

Of course!

[ternarizator]

vincent wrote:
One can e.g. work on the bell technique (bell polyrhythm, I guess you see what I mean) at a wholly separate time, and put it together later, I think it would be worth trying.

No, I don't know what you try to say here.

For beginners it can be very tricky if one plays x . x x . x x . x x . x while another playsx x . x x . x x . x x ., like in some versions of Sòkò, or in a Sangban & Dunumba chauffe. So I plan to work on the one hand the Ganga drums technique :

first quaternary :

1 . . . 2 . . . 3 . . . 4 . . .
o . o * c . . * o . o * c . . *
o . * o o . * o o . * o o . * o

then ternary (or nearly)

1 . . 2 . . 3 . . 4 . .
o o * c . * o o * c . *
o * o o * o o * o o * o

it's very easy to catch, for " * o o * c " falls exactly with " o o * o o "

Then, separately, you make them work, very slowly, the superposition of the two bell-patterns :

1 . . 2 . . 3 . . 4 . .
x x . x x . x x . x x .
x . x x . x x . x x . x

Finally, if all goes well, your beginners can play it in the Upper Guinea style (and are not beginners anymore).

(Notice that the quaternary form, although interesting, is rather unconceivable in this style, because of the bell technique. That's one of the reasons for which I'm interested in this way of playing.)

Of course, it works for a lot of situations. For example, you want to put together the o . o o . . and the o . o . o . . c . c . . clave (MK's djaa, or a variant in N'gri, as you want), and your beginners don't find the good imbrication. The Ganga drums solve the problem :

1 . . 2 . . 3 . . 4 . .
o * o o * . o * o o * .
o * o * o . * c * c * .

The two musicians play in fact the same pattern !

And again you can work on the bells (if you chose to play the little one with a bell) separately :

1 . . 2 . . 3 . . 4 . .
x . x x . x x . x x . x
x . x . x . x x . x . x

Precisely I had planned to discuss the influence of bells on the way of playing.

(Tbc)

Vincent