click here for Kevin Moore's book: Understanding Clave and Clave Changes.

Clave Debate #3: "There is no such thing as 2:3 clave".

In argument 2, we pointed out that those who call rhythm 4 "2:3 clave", while ill-advised about the name, are usually very talented, and that in fact, their ability to intuitively hear music is what leads to the confusion over the name. This is not the case in argument #3. As far as we can tell, the only way to logically conclude that there’s no such thing as 2:3 clave would be to be completely oblivious to the voices, horns, piano, and bass! Nevertheless, we’ll take their argument as far as it will allow:

Suppose you have 4 people in a room. One is playing rhythm 1 on clave. The other 3 are playing timbales, congas, and campana bell, using their most basic patterns and locking to the clave without playing any fills or decorations. Now, suppose you had walked into the room after the music had already started. Which clave is the music in? Are they playing rhythm 1 (2:3 son clave) or rhythm 3 (3:2 son clave)? The answer is that you couldn’t tell. So, the extremist argument goes, "there is really only one type of clave" (they usually say it’s 3:2).

The problem with this argument is that if you had been there when they started, someone would have said ("one, two, one, two, three, four") and they would have started by playing either the 2 side or the 3 side. You would then know that the musicians, in their own minds, were thinking of their rhythm as being 2:3 or 3:2. But if you had walked in after the pattern had begun, you might hear the rhythm starting on either 2 or 3 because you have no point of reference, so your brain creates one for you. The pattern just keeps cycling and it’s up to you to decide where it feels like the beginning is. If you had never heard any type of clave-based music, you might even hear the pattern starting in any of the 14 other possible places!! Each of those is a legitimate rhythm of some sort. It’s just a matter of where the listener perceives the downbeat. If the musicians started on the 2 side and in their own minds are feeling the 2 side as being the beginning of the rhythm, then we would say they’re playing in 2:3 clave. The extremist argument then goes: "well, they should have started on 3 because there is only 3:2 clave". But this fails to take into account that the clave rhythm is also used to accompany songs! And that about half of them begin on the 2 side! For example, "Pare Cochero". [audio example 5]

Pare coche coche - ro
Pare coche

Try singing it while clapping 3:2 and then 2:3 son clave. 2:3 feels much more natural because "re" and "co" fall on the two beats of the two side and "che" and "ro" fall on the second two beats of the 3 side. If you listen to any of the hundreds of recordings of this song, you’ll find that they’re all in 2:3.

If the still doesn’t prove the existence of 2:3 clave, then the problem is again with semantics and we faithful believers in 2:3 need to define our terms better. To do this we’ll play "Twinkle Twinkle Little Star" in both 3:2 and 2:3.

[audio example 6] — Twinkle Twinkle in 3:2 clave

[audio example 7] — Twinkle Twinkle in 2:3 clave

Listen to example 6, in which the nursery rhyme begins right on the downbeat of the 3 side. This is what it means to be "in 3:2 clave". Now listen to example 7, in which the same melody begins on the downbeat of the 2 side. Since Twinkle Twinkle is all quarter notes, it sounds equally dumb in both 3:2 and 2:3 clave, but the point is that if you start singing the song on the 3 side, the listener will perceive the rhythm as 3:2 and if you start singing it on the 2 side, the listener will perceive the rhythm as 2:3.The harmony and melody force you to hear the phrase starting in a specific place, and if the percussion is playing the 2 side, that is what the 2:3 believers call 2:3 clave.

Having satisfactorily proven the existence of 2:3 clave, we can now neatly divide all music in 3:2 and 2:3 and live happily ever after, right? Wrong! Let's look at the song "Por Encima del Nivel", also known as "Sandunguera", by Los Van Van [audio example 8]

Sandunguera
se te va por encima de la cintura
no te muevas más así
que te vas por encima del nivel

Sandunguera,
se te va por encima de la cintura
no te muevas más así
que te vas por encima del nivel

y dicen que
que a esa muchacha no nay quien le ponga el freno que
que qué de qué
que si la dejas se lleva el baile entero
qué facilidad! mírala!
mírala!
mirala!

Sandunguera ...

Start clapping 2:3 clave such that the first clap of the 2 side falls between "gue" and "ra" of "Sandunguera". Hold on stubbornly until you’ve gone through all the words listed and arrived back at "Sandunguera". The section now repeats, note for note. But what happened to your clave? Your 2 side is no longer between "gue" and "ra"! Now your 3 side begins right with "gue". Also, you may have also noticed that the clave began to feel very out of place right after "y dicen que".

Now try dancing to it. At the first "Sandunguera" start dancing a basic step, breaking forward first, and keep dancing until this part comes around again. The second time through you’ll find yourself breaking backwards instead of forwards! Dancing the basic step and clapping one clave both take exactly the same amount of time, 4 beats. But this song has an extra two beats which keeps the clave and basic step from coming out even! What if you have a great song which doesn’t come out evenly? Or, what if you have a great song which sounds better in 2:3 at one point and better in 3:2 at a different point? Let’s see how Los Van Van solved these problems in "Sandunguera".

Try clapping 2:3 clave again, but when you hear "y dicen", leave off the last note of the 3 side and then start the 3 side again (a second time in a row) right on "que". Now everything will come out right and you’ve just experienced what the composer, Juan Formell, refers to as "Clave License".

Note: The "Live in Miami Arena 1999" album has a version of Sandunguera on which Samuel Formell plays clave on the jam block throughout the cuerpo.

Now we’re getting to the really interesting part. Many, if not most, Latin songs, coros, mambos and breaks sound better, sometimes a lot better, when accompanied by one clave rhythm than they do when they’re accompanied by the other. If the rhythm section plays in the other clave pattern underneath, it sounds off, and this is called "cruzado". This is not one of the great clave debates because to'el mundo y su hermano agrees that "cruza’o" is not a good thing! The disgreements revolve around how to best avoid this undesirable state of affairs and this brings us to the term "clave change".

The simplest solution would be to never write music that doesn’t break down into equal four-beat measures, and to never combine musical ideas that sound better in 3:2 with those that sound better in 2:3. This is indeed exactly what happens in more than half of Salsa and Timba arrangements, but there are many great arrangements which don’t. Whenever one of these issues arises the arranger has to do something to reconcile the situation.

The first solution is that the last measure of 2:3 can be cut two beats short so that the 3:2 section can begin on the 3 side. This way, the percussionists don’t have to change, but the emphasis of the new section causes the listener to start hearing the music in 3:2 clave.

The second solution is to let the phrase end naturally and then have the percussionists themselves change to the other clave for the next section. This can be broken down into two types of Clave Change, those where the 2 side is repeated and those where the 3 side is repeated, and each has a distinct personality.

How to name these three types of clave change brings up a whole new semantic can of worms, which we’ll explain in the final section.

click here for Kevin Moore's book: Understanding Clave and Clave Changes.