[linux-audio-dev] ragas and talas

Paul Davis paul at linuxaudiosystems.com
Mon Dec 16 22:35:54 UTC 2002


>Among the talas which are in common use, the sixteen-beat tintal 
>(or trital: 4+4+4+4) is perhaps the most popular today [33].
>Other common talas are:
> 
> dadra - six counts: 3+3
> rupak - seven counts: 3+2
> kaharva - eight counts: 4+4
> jhaptal - ten counts: 2+3+3+3
> ektal and chautal - twelve counts: 2+2+2+2+2+2
> dhamar - fourteen counts: 5+2+3+4
> dipchandi - fourteen counts: 3+4+3+4
> addha tintal or sitarkhani: sixteen counts: 4+4+4+4

don't let this fool you into thinking that the players count:

      1-2-3-4-1-2-3-4-1-2-3-4-1-2-3-4

when playing tintal. they don't. its logically divided into 4 parts:

     da-dhin-dhin-da
     da-dhin-dhin-da
     da-tin-tin-na
     na-dhin-dhin-da

but nobody counts them in 4 sections. actually, nobody really counts
them at all! this idea that classical indian music even counts beats
its a bit of a misnomer to start with. they conceive of the tala as a
cycle, in fact literally a circle in some instructional works. rather
than lay out rythmn on a line the way we do in the west, its laid out
in a circle, and divided in various ways. 

however, the musicians do need to keep time, and even though this is
done partly by a keen awareness of the underlying structure of the
tala, they will also tap. note that its rare to ever hear tintal
played exactly as i wrote it out above. the musicians are free to
improvise on the "theme" of the tala, which they do by subdividing the
beats in various ways. they could take a single "dhin" and play

    da-tere-kete-da-tere-kete-da-tere-kete-da

instead, as one tiny example. in this case, the new "beats" are
clearly traditional western subdivisions (1/16ths, 32nds etc). but
start from jhaptal, which is based on a 10 count cycle, and subdivide
that, and you end up with strange "beat note values" that we never use
in the west. 

finally, although the classical tradition rarely mixes talas within a
single piece, the newer breed of indian musicians, blending ideas from
many different world traditions with their own skills, do this. thats
how people like shankar and zakir hussain end up playing some really
odd rythmnic games. its akin to what people like ornette coleman,
miles davis and coltrane did when they started linking jazz
improvisation with ideas drawn from classical traditions (and actually
very close in many ways to the modal jazz of the "kind of blue" era).

i studied tabla for 3 months - long enough to know i wasn't any good
at it. but dammit, i'm a good computer programmer, and i'm determined
to give zakir a run for his money one day :) for that, i need a
sequencer that can handle these ideas!

>[33] Over eighty-five percent of the ragas on the CDs have been
>performed in tintal.

this is also probably true of most pieces performed live. tintal is
the 4/4 of indian classical music, with ektal or jhaptal playing the
role of the waltz :)

notice, though, that in indian classical music, the raga and the tala
are not inextricably tied together. whereas it would sound plain wrong
to try to play the blue danube in 4 (its probably been tried!), its
not uncommon to find indian musicians playing a raga with different
talas. this is mostly because the raga is more like a scale than a
specific melody, and so the player(s) are free to work with the
rythmic structure. 

--p



More information about the Linux-audio-dev mailing list