[ternarizator]

Hello guys

As promised in july, I propose one hypothesis about how some djembé patterns may have been generated.

Let's consider this well-known pattern (kuku e.g.) b . t t . . s ..

There are some other patterns very similar to it :

b . t t b . s . (Moribayassa)

b . t t b . s s (Djolé)

b . t t . . s s (Dalah)

b . t t . b s . (Djagbé)

My idea is that all these patterns can be seen as derived from one "ancestor" b b t t b b s s, following the rules below :

[*] basses almost never go in pairs, and may even disappear
[*] tones almost always go in pairs
[*] slaps can go in pair or not

The "ancestor" has a ternary counterpart : b t t b s s (Siwé e.g.),
from which one can derive b t t b s . (Soli des Manians)

Another "family" can be observed with :

t t s s . . s .|t t . s b . s . (Sofa)

t t s s . . s s|t t s s b . s s (Kassa)

t t s s . s s .|t t s s b . s . (Fakoli)

t . s s . b s t|t . s s . b s t (Sunun)

Here the "ancestor" would be t t s s b b s s.

One could object that in the version of Fakoli shown above, the third slap is not at the right place. The same is true for the second tone in Sunun (the one just before the middle "|" ). We can solve the problem assuming that sometimes one stroke can be changed in another, for a better effect.

In Sunun, we re-obtain a pair of tones by changing a slap into a tone :

t . s s . b s s|t etc. becomes t . s s . b s T|t etc.

In (this version of) Fakoli, one gets more energy by changing a bass into a slap :

t t s s . b s . becomes t t s s . S s .

There is still much to say on this subject, but my post is already long enough for now...( not so long in fact, but writing in english is a real job for me...)

Vincent

[djembefeeling]

nice!

Two things come spontaneoulsy to my mind: all those accompaniments are tresillo. And I remember a theory of Daniel (afoba) about basses in djembe playing. It is very inconvenient to play those while you stand and they could have been completely ignored originally...

[ternarizator]

Two things come spontaneoulsy to my mind: all those accompaniments are tresillo.

I think all those patterns contain tresillo, is that what you mean ?

If you want to build nearly tresillo patterns from these "raw materials", but following the "pairs of tones" rule, you get :

b . t t . . s . from the first one I mentionned, and t t . s . . s . from the second one.

Notice that the 2nd one corresponds exactly to the kensedeni for Kassa :

x x . x . x x .
o o . ø . . ø .

And I remember a theory of Daniel (afoba) about basses in djembe playing. It is very inconvenient to play those while you stand and they could have been completely originally...

Is it me ? Your sentence seems not to be complete...

[off topic : I met Daniel last october in Basel (Gbada dance workshop). He does agree with the beat-shift concept, but tends to see it the other way round : while I consider the 5th pulse of djaa corresponding to the first of dunungbé, he tends to see the 9th of dunungbé similar to the first of djaa, that's to say dunungbé and family as ancestors and djaa family as descendants. For me, it's incompatible with the concept of ternarization. This question has to be deepened...]

OK, the job must go on :

The "t t s s b b s s" complete pattern has (of course) a ternary counterpart :

t s s b s s

I see at least three known patterns that can be seen as derived from this one :

t s s . s .|t s . b s . (MK's accompaniment for Marakadon)

t s s . s s|t s s b s s (one phrase of FK's solo for Gidamba)

t s . b s . (Djembé 1 in Djabara)

If anyone sees another pattern likely to fit into this model...

(To be continued)

[djembefeeling]

I think all those patterns contain tresillo, is that what you mean ?

o.k., you can say so. they all are an embodiment of tresillo.

Is it me ? Your sentence seems not to be complete...

It is incomplete. I intended to write: It is very inconvenient to play those while you stand and they could have been left out completely originally..." (isn't it funny the I left out "left out"?)
that would mean that at least in djembe music original patterns have no basses.

Notice that the 2nd one corresponds exactly to the kensedeni for Kassa :

x x . x . x x .
o o . ø . . ø .

not a good example, because it's a doubtful kensedeni pattern. in Hamana, they wouldn't play it that way. simple beat and no bell. but t t . s . . s . is the djembe accompaniment for Sogoninkun.

he tends to see the 9th of dunungbé similar to the first of djaa

I think so, too. But I don't understand why it implies to say that dunungbé and family is the ancestor and the djaa family the descendant. From the dissemination of the dja family rhythms in large parts of Africa we have to believe that this family is the original, I think.

I see at least three known patterns that can be seen as derived from this one :

t s s . s .|t s . b s . (MK's accompaniment for Marakadon)

t s s . s s|t s s b s s (one phrase of FK's solo for Gidamba)

t s . b s . (Djembé 1 in Djabara)

hmm. those examples are not very convinving, I think. MK's accompaniment is just one way of many to do solo on Maraka, and not the dominant one I think. Drissa Kone does b s s . s .|b s . b s . And the accompaniment on Djabara, how "authentic" is that? Famoudou has invented some stuff while he was stimulated from the systematic approach of Paul Engel. In that fashion he told us "why not play t s . . s . on the dja rhythms?" on a workshop in Bamberg.
Yet another thought: it does not contain "dosillo" ( ). While b s o o s .|b s . b s . is an embodiment of the son clave, all non-"dosillo" pulses are disposible, like in Tiriba: b s o . s .|b s . b s . or or Kakilambe: b . o o s .|b o . o s . or in the old accompaniment for Maraka: b . o o s .|b . o o s .

[ternarizator]

Yet another thought: it does not contain "dosillo" ( ). While b s o o s .|b s . b s . is an embodiment of the son clave, all non-"dosillo" pulses are disposible, like in Tiriba: b s o . s .|b s . b s . or or Kakilambe: b . o o s .|b o . o s . or in the old accompaniment for Maraka: b . o o s .|b o . o s .

Help, I'm lost with your last sentence ! Could you be more precise or explicit ?

What do you mean by "dosillo" ? It seems that it should be clear to me, but it's not the case...

Yet I see the connexion between this kind of patterns and the ternary son-clave (or its bell) :

x . x . x . x x . x x.
o . o . o . . o . o . .
b s t . s . b s . t s . (FK on Gidamba, good or not ?)

(More propositions about b s t t s . family in my next message)

Vincent

[djembefeeling]

Help, I'm lost with your last sentence ! Could you be more precise or explicit ?

Dang! So many mistakes in my posts now, I should take more attention. I guess it's because I am a bit overworked these days. the sentence should read like this:
While b s o o s .|b s . b s . is an embodiment of the son clave, all non-"dosillo" pulses are disposible, like in Tiriba: b s o . s .|b s . b s . or Kakilambe: b . o o s .|b o . o s . or in the old accompaniment for Maraka: b . o o s .|b . o o s .

"Dosillo" is my half-comic way of talking about the tresillo in the ternary rhythms. if they call it tresillo in the quarternary where it denotes the 3pulse movement against the quaternary pulse structur of the beat, it should be dosillo in the 2pulse movement against the ternary pulse structure of the beat in 12/8.

So, while you put up all these tresillo embodiments in the quarternary examples, the last ternary example (first djembe on djaraba) t s . b s . does not cover "dosillo", that is the first half of the 12/8 son clave.

[ternarizator]

OK, It's a bit more understandable...

However, I think your vocabulary is not relevant in this case : "tresillo" is the spanish word for triplet (de : triole), it normally should mean that there are three equal (in duration) notes (or strokes) instead of 2 (or 4 , or 8...). In fact, the cuban use of this word is already wrong, because in the "quaternary tresillo", the durations are not equal (3-3-2), and for the "ternary tresillo", the (equal) durations are not in conflict with the flow of pulses (3+3 = 2+2+2). But the original meaning that, in my opinion, have to be kept, is "three strokes" (i.e. 3 more or less equal strokes on 2 beats in this case).
It seems obvious if you consider the "cinquillo" (i.e. "5-illo")x . x x . x x . : 5 strokes...

Now, you point out that t s . b s . doesn't contain the "ternary tresillo". So what ?

The "ternary passport" s . t s .(b) doesn't anymore... Perhaps it has to do with the fact that there is no ambiguity about the beat in quaternary rhythms, while ternary rhythms can be more ambiguous (4x3 = 3x4 etc.) : the ternary tresillo x . x . x . doesn't show clearly the beat. Patterns like t s . b s . and s . t s .(b) bring to light, at least, the ternary nature of the rhythm.

Vincent

[djembefeeling]

However, I think your vocabulary is not relevant in this case : "tresillo" is the spanish word for triplet

I thought so. And I shouldn't establish my private terminology. But the thing is: triplet in this context is very confusing. Three equal notes played over two pulses or notes is something very different than over two beats. The movement in he first half of the 12/8 son clave is no triplet, but binary against ternary. For me it is more important that the tresillo challenges the beat by stressing a movement that falls one puls short to the the pulsation of the beat, so 3 against 4 in 4/4 or 2 against 3 in 12/8. This is what creates tension and energy in the first half that waits to be reconciled on the second half.

Now, you point out that t s . b s . doesn't contain the "ternary tresillo". So what ?

It was just remarkable that all your quartenary examples contained tresillo, now this last example in the 12/8 category does not, like the passport accompaniment as you pointed out. for me this is relevant because of the function of these two different movements. the rhythmic challenge of a displacement is the functional counterpart to the beat. And I would be astonished if an "ancestor" would generate both the beat and the challenge. But I guess I have to write my lengthy account of the connection between Mande culture and Mande rhythms to make that more understandable. Yet time is something rare and precious for me these days...

[ternarizator]

And I would be astonished if an "ancestor" would generate both the beat and the challenge.

Look at the following :

b b s s t t s s
↓
b . s s t . s s
↓ (pair of tones)
b . s t t . s s
↓(ternarization)
b s t t s s (full pattern. Can be found in Maraka, with Lss feeling)
↓
b s t t s . (djagbewara a.o.) → b . t t s . (Maraka)
↕
b s t t s . → s .(b)s t t|s .(b)s t t (Db family, shifting the beat by 4 pulses and removing basses)

As you know, b s t t s . challenges the beat, but s . . s t t|s . . s t t highlights it on the contrary. May be a kind of pirouette...

Jürgen if you're overworked, take your time ! Perhaps one answer per week is enough ?

Vincent

[ternarizator]

Some relevant observations of Jürgen make me deepen my reflection. E.g. how can we imagine a way of generating the "quaternary passport" ?

My first opinion was that we can take the s s b b s s t t hypothetical ancestor, remove some strokes to obtain s . . b s . t t (I use it regularly to force students to play the "right handing", that's to say "r . . l r . r l"), and finally decide to turn the "b" into a "s" because it's more powerful.

But we can consider a totally different way of generating the same, by aiming at building an accompaniment which expresses (or embodies)at the same time the beat and the tresillo.

We can e.g. take the following sangban pattern : M . . M . . S . (woima), and realise a cross-over with this kensereni pattern : K . . . K . . ..

The first one could be played s . . s . . t . by the djembé, but the rule of tones coming in pairs leads us to play rather s . . s . . t t (notice that its ternary counterpart is the well-known s . s . t t)

Adding the strokes of kensereni as slaps (high-pitch sound), we obtain the traditionnal quaternary passport s . . s s . t t

Notice that without the rule of tones coming in pair, we would have gotten the habanera rhythm :

s . . s s . t .

So perhaps a good way to search how some - not all of course - djembé patterns have been generated (or can be generated in the future) is to mix my "ancestor" approach (which I haven't fully explained yet) and the concept of tresillo...


[off-topic : can someone confirm the s . t t s . . b djembé pattern in the Bamako-Foli woloso ? This could be a key pattern in my clave-expanding theory...]

[djembefeeling]

Some relevant observations of Jürgen make me deepen my reflection.

I am glad that I did even though I wrote so much rubbish lately

[off-topic : can someone confirm the s . t t s . . b djembé pattern in the Bamako-Foli woloso ? This could be a key pattern in my clave-expanding theory...]

unfortunately, I cannot. The pattern starts as |s.Tt|s..b| (T=tonflam), but when the konkoni starts it is clear that it is |s..b|s.Tt|

more of interest might be the jenbe realbook (1), where Yamadou Dounbia, Drissa Konés master and that of so many others, starts with |s.tt|s.bb|st.t|sbbb|

[ternarizator]

Thanks for your answer !

Surprisingly, it tends to confirm my first hypothesis, with quaternary passport being a changed
s . . b s . t t (the T-flam doubtless being an embellishment), but it also confirms your interpretation, since the pattern I was asking about doesn't contain tresillo...