Sapta-Tala system
Chandar Venkataraman
Hi all:
I understand the classification of the ragas under the 72 melakarta
scheme. But, till this day I have never understood the mathematics
of the sapta-tala classification. Are those the only seven basic
talas you could have? Could we not have more of these basic types?
Also, is it correct to say that there are totally 7 x 5 x 5 talas
taking into account the "jaathi" and "nadai"?
Could someone out there explain the mathematics of this, please?
Thanks,
Chander.
Srinivasan Pichumani
>I understand the classification of the ragas under the 72 melakarta
>scheme. But, till this day I have never understood the mathematics
>of the sapta-tala classification. Are those the only seven basic
>talas you could have? Could we not have more of these basic types?
Yes, there are more of these basic types under the 108 tALa system,
but you don't see much use of them in Carnatic music compositions.
If you hear traditional "tiruppugazh" renditions, you
will notice some of these different-from-sulAdi-type tALas.
There are demonstrations on Simhanandana tALa and other such
esoteric/long time-measure tALas in various music conferences but
not much else goes on in modern Carnatic music that uses the 108 tALa system.
>Also, is it correct to say that there are totally 7 x 5 x 5 talas
>taking into account the "jaathi" and "nadai"?
Yes, if you want to put it that way. This is more a case of nomenclature
than anything else.
>Could someone out there explain the mathematics of this, please?
>Thanks,
>Chander.
Here is a repost of an old article that I wrote for this newsgroup.
It is an elementary write-up on the tALa system in Carnatic music.
----------------------------------------
SulAdi Sapta TAla system
-------------------------
Although there are various tALa systems in Indian music, like the
108 tALa system, the sulAdi sapta tALas (7 different taalas) have been
most popular in Carnatic music since the time of Purandaradasa (1484-1564).
Hence I will try to explain this tALa system, in which >75% of contemporary
Carnatic music is set.
A convenient way to describe this tALa system is to enumerate the various
terms encountered in this framework, and to explain each one.
First of all, the name or type of the tALa itself
-- in the sulAdi sapta (7) tALa scheme, the names are
eka, rupaka, triputa, matya, jhampa, ata, dhruva
Secondly, the constituent units of the tALa, known as the 'anga' (lit. limbs);
-- in the sapta tALa system, there are three types of 'anga', namely,
laghu, drutam, anudrutam;
Of these 3 'angas' that constitute a tALa, only the laghu is variant
in the sense that it can take up a time-measure of 3 beats or 4 beats
or 5 beats or 7 beats or 9 beats.
The other two constituent units are invariant in the sense that the
drutam always takes up a time-measure of 2 beats and the anudrutam
always takes up 1 beat.
Thirdly,the amount of beats that a laghu can take, is indicated by the 'jAti' of
the tALam.
-- Tisra jAti means it can take up 3 beats, chatusra jaati --> 4,
khanda jAti --> 5, mishra jaati --> 7, sankIrNa jaati --> 9.
Using the symbols 1, 0, and ~ to represent the laghu, drutam, and anudrutam
respectively, the formal definition of the 7 tALa types are as follows:
Eka 1 [i.e it consists only of a laghu]
Rupaka 01 [a drutam, followed by a laghu]
Triputa 100 [a laghu, followed by 2 drutams]
Matya 101 [a laghu, then a drutam, and then a laghu]
Jhampa 1~0 [a laghu, an anudrutam, a drutam]
Ata 1100 [two laghus, and then two drutams]
Dhruva 1011 [laghu, drutam, laghu, laghu]
Multiplying each of these tALa types, with the "jAti" of the tALa, we
arrive at 35 different tALams.
Associated with every constituent unit of a tALam, there is an action or
'kriyA' which can either be silent (nishabda kriyA) or be sounded (sushabda
kriyA) with a tap on the thigh or a clap -- some people might view this as
noise ;-)
For the laghu, the first action is always a tap (known in Tamil as taTTu or
as tAli in Hindi) which is followed by fingertip counting;
As an example, if the tALa has a chatusra jAti laghu, i.e the laghu takes
4 beats, there will be a tap followed by 3 finger counts, starting with the
twinky (little finger).
For the drutam, which is invariant as described above, the action is always
a tap and a wave (thattu and veecchu in Tamil; tAli and kAli in Hindi) --
each accounting for one beat duration, thereby accounting for a total of 2
beats.
For the anudrutam, which is also invariant, the action is always a tap.
*** When you see the action or "kriyA", whether it is a tap on the thigh,
*** or a wave, or a finger count without any sound, it TYPICALLY indicates the
*** start of the beat interval.
*** However, some musicians tend to indicate these actions according to their
*** whims, which can confuse the listener.
---------------------------------------
Adi tALam, the most common tALa encountered in composed Carnatic music,
has the following technical name ---> 'chatusra jAti triputa tALam',
Hence, this tALam is indicated by
a tap, followed by 3 finger counts (accounting for the chatusra jAti
laghu which is a total 4 beat duration)
a tap, followed by a wave (the first drutam) (2 beats)
and another tap, followed by a wave (the second drutam). (2 beats)
Thus, totally, you get a 8 beat tALam with stress on the 1st, 5th, and 7th
beat. Then the cycle repeats all over again.
----------------------------------------
In addition to features like type or name, jAti, anga, etc -- there are
other terms like kAlam or tempo, nadai or gati or gait, graha or eduppu
or starting point in the rhythmic cycle. These are as follows:
------------------------------------
KAlam or Tempo -- refers to the time interval between the beats; you can also
look at it as the total time taken up by one full rhythmic cycle. Generally,
speaking there are 3 different tempos -- vilambita, madhyama, and durita
i.e slow, medium, and fast. You can probably add a very slow and very fast
tempo too to this list.
[oru kalai chowkkam, rettai kalai chowkkam are how madhyama and vilambita
tempos are referred to in Tamil]
If you translate this to absolute time units, and for simplicity sake consider
the "Adi tALam", -- vilambita tempo Adi tALa will take up about 12-16
seconds, madhyama tempo Adi tALam will take 8 seconds, and durita tempo Adi
tALam will take 4-6 seconds, for the full cycle of 8 beats.
[The variance from 12 to 16 seconds is needed here, since this tempo is a very
subjective thing and varies from musician to musician; no one uses a metronome
with a fixed tempo setting]
----------------------------------------------------------------------
Nadai or Gati -- as in Tisra Gati, Chatusra Nadai, etc refers to the
division of every beat in the tALam into 4 or 5 or 6 or 7 or 9
'virtual' 'subdivisions'.
By listening closely to the words in a composition, and observing the length of
the syllables, you can figure out the gait. Similarly, the percussionists's
syllables/strokes, particularly when he plays a solo, are also useful in
figuring out the gait.
[Compositions like 'Biraana Varaa Lichi' of Syaaama Sastry in Kalyaani
or popular songs like 'Jayati Jayati Bhaarata Maataa' in Khamaas,
Humko Man ki Shakti Deynaa, are all in tisra gati; I believe the equivalent
term in Western music is the waltz gait]
From my personal experience, I have found that anything other than
'chatusra gati' tends to draw your attention more towards the rhythmic aspect
of the composition. And if you are trying to keep the tALam, you have to be
more careful in these other gaits.
The Chatusra gati seems like a very natural and simple gait to most people,
if not all.
When a musician improvises using solfa syllables, i.e kalpanaa svaras, if he
does this to the 'chatusra gati' or 'quad gait', without bringing in any
elaborate arithmetic/calculation/"kanakku", a specific term called
"sarvalaghu" is used to refer to his method.
------------------------------------------------------------------------
Graha or Eduppu -- refers to the point in the tALa cycle where the
the words of the composition starts, be it the the pallavi or refrain,
or the anupallavi, or the charanam.
The starting point of the tALa cycle is called the samam. The composition
can start on the samam, or before the samam, or after the samam. Tyagaraja's
compositions best illustrate these differences.
Consider a composition like the famous Pancharatna kriti "JagadAnanda kAraka"
It is set to the Adi tALam; and the refrain begins on the samam. The
anupallavi, "gaganAdhipa sat kulaja ...", begins a 1/2 beat after the samam.
All the various charanams in this kriti begin on the samam.
Compositions like "etAvunarA" in kalyANI, or the other Pancaratna kriti
"kanakana rucirA" in varALL, are set in the vilambita kAlam or slow tempo
Adi tALam, and their pallavi begins half way during the first beat or in
other words 1/2 beat after the samam -- it is called "arai edam" in Tamil.
The pallavi of "endaro mahAnubhAvulu" in Adi tALam begins one full beat
after the samam.
A composition like "marugelarA" in jayantashrI, which is set in the madhyama
kAlam or middle tempo, begins half way during the second beat -- it is called
"onnarai or ondrai edam" in Tamil.
There are some places where the eduppu is before the commencement of the
tALa cycle. The example that comes to my mind is Baalamurali's varnam in
Nattai. In that composition, which is in Adi tALa, the "eththukkada" pallavi
is "Sapta Svaramulu Naa Jeevamu". The words for this pallavi begin halfway during
the last beat of the 'previous' cycle.
The non-samam eduppus are known as "ateetam" and "anahatam"; I don't
remember exactly which refers to before-the-beat eduppu and which to
after-the-beat eduppu.
I believe that the non-samam eduppu is somewhat similar to "syncopation" in
Western classical music.
In Raagam-Taanam-Pallavis, things can get very hairy. The eduppu might be
after 3/4th of the first beat; or if it is a 4-kalai pallavi, in other words an
extra-slow tempo pallavi, the eduppu can be after 7/8th of the first beat.
When the mrdangist plays his strokes and precisely presents the eduppu to
the main artiste, it is very satisfying aesthetically.
------------------------------------------------------------------
Other than the sapta tALa system, there is the scheme of the 108 tALas;
and there are pallavis set in time-measures of 108 beats, 70 beats. There
are many lec-dems on these at the Music Academy, Madras and other places
that one can attend
Among popular compositions, the "Tiruppugazh" of the medieval saint
Arunagirinadhar, is set to very interesting rhythmic cycles.
--Srini.
ps: mishra cApu (equivalent to the Rupak tALa
of Hindustaani music) can be considered a faster tempo version of
"Tisra jAti Triputa" tALam, which is 7 beats.
The action used to indicate mishra cApu is slightly different from
what I have described above for the sulAdi sapta tALas.
b.chandramouli
This article is in reference to Srini Pichumani's excellent write-up
on the Tala system. I have always had many unanswered questions on
this topic and I thought I will try to give them shape in the form of
questions and hope to get some clarifications. In case, the questions
sound confusing or critical in some instances, it is just my inability to
verbalize my thoughts. Please take it in that stride. Just to give my
tala background, with some difficulty I can identify and keep the beats
for Adi, Rupaka, Misra Chapu and Kanda Chapu talas and on a good day
identify the eduppu for Adi talam.
Now to the questions:
1) On The various Angas:
a) Are these artifacts generated as a result of the effort to
bring the very many talas that were present under one
unified umbrella (by Purandaradasa, I may be wrong here)?
b) Other than the actions (kriyas) that are used to count
the beats, is there any musical significance to them?
Do they denote stress points? For example, a
3 beat laghu followed by two anudrutam is musically different
from a 3 beat laghu followed by one drutam? Or more
interestingly, how about a talam with 5 beat laghu plus a
drutam and a talam with just a 7 beat laghu. If the angas
do represent stress points, then the song should somehow
reflect that too, isn't it? Is that the case with the popular
talas like Adi, Rupakam, Misra Chapu and Kanda Chapu.
c) Do the individual beats of an anga signify any stress points
in the tala? Like, in the Drutam the first beat representing
a major stress point and the second beat representing a minor
stress point?
d) Take the case where one can keep the beats for a song
and can determine the number of beats per cycle. What else
one should look for in the song to determine the structure
of the tala (i.e defining it in terms of the various angas.)
e) Are there any rules (implicit or explicit) followed by composers
with respect to aligning words of a song with the angas? Like
a word (or a syllable) can not cross anga boundary, etc.
f) Are there any rules (implicit or explicit) with respect where a
pause can occur in the tala cycle.
2) Regarding KAlam and tempo: Should a song be composed in a particular
KAlam or it is the performer who chooses the KAlam. The latter makes
sense to me but from what I know, songs are usually set to Rettai Kalai
Chowkkam etc and rarely they are sung in other tempos (except Trichur
Ramachandran :=)).
3) This sounds like a stupid question but let me ask this anyway.
Assuming that a song is composed in rettai kalai Adi, each of the 8 beats
can be felt to consist of two separate sub-beats. Does this have anything
to do with the Nadai or Gati? (since Nadai is also a subdivision of a beat
into subbeats).
4) If I consider a rettai kalai cycle to consist of two normal tempo cycles,
what am I doing wrong musically, if at all? Do the "angas" play a role in
determining if the tala cycle is 8 beats in rettai kalai or 16 beats in
normal tempo.
5) I somehow suspect that my lack of understanding of the significance of
the angas make me ask this question on "eduppu"s: What is
the difference (musically, that is) between a song that starts
in the middle of the 1st beat (arai eduppu) and a song that starts
in the middle of the 3rd beat (both in Adi). Similarly, between two
songs that start in the middle of the 2nd beat (onnarai eduppu)
and the middle of the 8th beat.
6) On Nadai, do the subbeats (virtual beats as Srini calls them) signify
real stress points in the song or Nadai has just to do with
how the syllables of the song are organized. I strongly suspect
that there is a musical significance to the nadai (and not just
with the words of the song) since one can clearly sense,
in the case of a non-chathusra Nadai Adi, it is not a vanilla Adi.
How can one go about finding the Nadai of a song? If the subbeats
give the illusion of proper beat points, an uninformed and uninitiated
person like me, may be tricked into counting 24 beats for the 3
subbeat Adi and conclude that is is actually a rupaka talam with
8 beat cycles.
That is it for now. Again, please do not take these questions as combative
in nature. These are just attempts to get a better understanding of talas.
Thanks in advance for any answers/discussions/enlightenments/flames.
B. Chandramouli
att!acpfs!chandra
Srinivasan Pichumani
>
> 1) On The various Angas:
> a) Are these artifacts generated as a result of the effort to
> bring the very many talas that were present under one
> unified umbrella (by Purandaradasa, I may be wrong here)?
No, these are not artifacts. They do have a rhythmic function;
they split the total tAla cycle into distinct parts. For example,
in the 8 beat Adi tAla cycle, the particular arrangement of laghu
and drutam accounts for a 4+2+2 split, while in the 8 bit Khanda
Jhampa tAla, the angas account for a 5+1+2 split.
If you don't have any use (of the form of poetical stress or otherwise)
for such a split, and only want a rhythm which repeats after say 5 beats,
you can always use the Eka tAla, which has a stress & sounded action
(sashabda kriyA) at the beginning of the cycle and none after that till
the beginning of the next cycle.
This may sound like an aposteriori observation coming after the acceptance
of the SulAdi sapta-tAla system. But even if this particular system and
its postulates were ignored, other older/newer tAla systems have
use for angas of one variety or other. Calling these as druta/laghu/anudruta
or guru/pluta/kAkapada is just a nomenclature issue -- not very significant;
in a similar vein, the actions associated with these various angas are not
the crucial issue.
[There are analogues to this in poetry, as you would expect. In Sanskrit,
the Gayatri metre - defined as consisting of 24 syllables, has 3 feet of 8
syllables each; and the metre with 32 syllables, has 4 quatrains of 8
syllables each]
[In ancient tamil the prosody was not based on akshara/mAtra, but on other
similar units like "acai" and "cIr" (?). There nevertheless was some manner
in which an overall metre like the kuRaL veNpA was split into smaller units
for the purposes of poetic composition]
> b) Other than the actions (kriyas) that are used to count the beats,
> is there any musical significance to them?
> Do they denote stress points? For example, a
> 3 beat laghu followed by two anudrutam is musically different
> from a 3 beat laghu followed by one drutam? Or more
> interestingly, how about a talam with 5 beat laghu plus a
> drutam and a talam with just a 7 beat laghu. If the angas
> do represent stress points, then the song should somehow
> reflect that too, isn't it? Is that the case with the popular
> talas like Adi, Rupakam, Misra Chapu and Kanda Chapu.
Yes, they do denote stress points. So the first beat of a laghu, or the first
beat of a drutam, are stress points. And a composition will typically reflect
this. Almost all of Purandaradasa's compositions, many shyamA shAstri kritis
like the Bhairavi svarajati etc are "self-evident" as far as the tAla is
concerned -- Mishra Capu in most cases.
However, I will not contend this too strongly, since in a typical kriti
set to Adi tAla, you can normally "hear" the stress on the first and fifth
beats, i.e the first beat of the laghu and the first beat of the first drutam;
but you don't "hear" the stress much at the start of the second drutam i.e on
the 7th beat. Maybe the reason for this is that most kritis have a well defined
syllabic attack at the 5th beat of the Adi tAlam, but not so well defined on
the 7th beat.
[You may contend that the reason one "hears" the attack on the 5th beat is
due to the fact that, that is where the singer slaps his thigh into pulp :-)
But the stress is nevertheless there in the poetic flow of the composition
and in the rhythm accompaniment]
In pallavis, there is very good structure, with a clear emphasis on the
"aruDi" or the midpoint of the pallavi -- most often this coincides with
the start of the first drutam.
> d) Take the case where one can keep the beats for a song
> and can determine the number of beats per cycle. What else
> one should look for in the song to determine the structure
> of the tala (i.e defining it in terms of the various angas.)
You should look for the syllabic attacks in the composition. As I said earlier,
it is easy to find this in the case of a pallavi. However, it can get quite
fuzzy when a musician renders a kriti -- due to his enunciation or whatever.
If you play close attention to the mrdangam, you can pick this up easier.
[Just as an example -- the mrdangist might play a sollu (bOl) which requires
the use of both the hands at certain crucial points in the tAla, like the
start of a drutam]
> e) Are there any rules (implicit or explicit) followed by composers
> with respect to aligning words of a song with the angas? Like
> a word (or a syllable) can not cross anga boundary, etc.
Yes, composers like Purandaradasa, shyAmA shAstri, dIkshitar etc pay close
attention to this. However, TyAgarAja krtis have a lot of "cross-rhythms";
this perception may be due to the fact that many of TyAgarAja's
madhyama kAla kritis are sung in Adi tAla today with a eduppu or graha
0.5 or 1.5 beats into the cycle, rather than the desAdi tAlam in which they
were originally set. I am not very clear on this issue though.
> f) Are there any rules (implicit or explicit) with respect where a
> pause can occur in the tala cycle.
I can see an implicit rule like the following: A pause or a kArvai(in the
form of an elongated vowel) typically seems to occur after the syllabic
attack coincident with the start of the first drutam.
> 2) Regarding KAlam and tempo: Should a song be composed in a particular
> KAlam or it is the performer who chooses the KAlam. The latter makes
> sense to me but from what I know, songs are usually set to Rettai Kalai
> Chowkkam etc and rarely they are sung in other tempos (except Trichur
> Ramachandran :=)).
Well, if you are performing someone else's composition, you should stick to
the tempo they envisioned, subject to your voice constraints and laya
abilities. Which means that it is madhyama kAla/vilambha kAla in almost all
cases. But, concert practice over the past 100 years or so has led to many
changes, as one can imagine, and a lot of "speed-up" has occured.
However, many of these changes seem to have been the result of exigencies,
which is kind of regrettable. What I mean is that -- if a musician is capable
of handling all kinds of tempi well, and does changes to the original tempo
of a song, it can somehow be rationalized (IMHO). However, most often it was
the case of a musician not capable of the demands of vilamba kAla changing
an original slow tempo kriti to a medium/fast tempo kriti.
This process has mauled the beauty of many a composition.
> 3) This sounds like a stupid question but let me ask this anyway.
> Assuming that a song is composed in rettai kalai Adi, each of the 8 beats
> can be felt to consist of two separate sub-beats. Does this have anything
> to do with the Nadai or Gati? (since Nadai is also a subdivision of a beat
> into subbeats).
Well, if your feeling that there are two sub-beats is correct,
then the song is in the "chatusra" nadai or gait automatically.
> 4) If I consider a rettai kalai cycle to consist of two normal tempo cycles,
> what am I doing wrong musically, if at all? Do the "angas" play a role in
> determining if the tala cycle is 8 beats in rettai kalai or 16 beats in
> normal tempo.
You might not be doing anything wrong, but if the mrdangist does this, then
he could be concluding his phrases at mid-cycle rather than the right point.
> 5) I somehow suspect that my lack of understanding of the significance of
> the angas make me ask this question on "eduppu"s: What is
> the difference (musically, that is) between a song that starts
> in the middle of the 1st beat (arai eduppu) and a song that starts
> in the middle of the 3rd beat (both in Adi). Similarly, between two
> songs that start in the middle of the 2nd beat (onnarai eduppu)
> and the middle of the 8th beat.
I am not sure I understand you here, but let us take a song like MarugelarA
in JayantashrI and in Adi tAla with an eduppu/graha/take-off 1.5 beats into
the cycle. [One easy way for you to sing this would be to say
"ta-ka-dhi-mi ta-ka"-Ma-ru gEla rA..." where the ta-ka-dhi-mi takes up the
first beat and ta-ka takes up half of the second beat]
If you sing this instead with an eduppu of .5 beats, then the "overall"
stress produced by the vocalist's and mrdangist's action at the beginning of
the first drutam, and the enunciation of the "ra" syllable in "marugElarA"
would not occur.
A composer typically choses words of appropriate length, with the right mixture
of consonants and vowels, and assigns the right amount of "kArvai" or
elongation to the various vowels, in order to fit it into a particular tAla
framework. If you change this, then various quirks would appear in the flow
of the song.
Just as there are legitimate complaints about the "speeding-up" of many
compositions, there have also been charges relating to kritis like shyAmA
shAstri's "himAcala tanaya" in Anandhabhairavi. The basic complaint has been
that the word structure in this kriti and the kArvais etc makes it obvious
that it should be sung in mishra cApu tAlam. However, the majority of the
people in this century have sung it in Adi tAlam, and in order to do this,
they have do a lot of juggling like starting phrases after a tap indicating
the start of the beat, elongating certain vowels beyond necessity etc.
> 6) On Nadai, do the subbeats (virtual beats as Srini calls them) signify
> real stress points in the song or Nadai has just to do with
> how the syllables of the song are organized. I strongly suspect
> that there is a musical significance to the nadai (and not just
> with the words of the song) since one can clearly sense,
> in the case of a non-chathusra Nadai Adi, it is not a vanilla Adi.
> How can one go about finding the Nadai of a song? If the subbeats
> give the illusion of proper beat points, an uninformed and uninitiated
> person like me, may be tricked into counting 24 beats for the 3
> subbeat Adi and conclude that is is actually a rupaka talam with
> 8 beat cycles.
It seems like you prefer to dichotomize the issue into literal and musical
entities. A composition is much more fused than you might want to accept or
believe, which is probably a reason why there are very few true vAggEyakArAs
or composers.
Anyway, a non-chatusra nadai certainly makes you sit up and take notice while
the standard chatusra nadai or sarvalaghu tends to put you in a sort of
bovine reverie. Maybe it has to do with our internal rhythms or whatever.
The interaction between the words of a song and the gait into
which it is cast is the most apparent way in which the musical significance
of a gait is demonstrated (otherwise, nadais like tisra, khanda etc would
remain mere postulates); the other ways in which this is demonstrated is when
the singer sings kanakku svaras, or when the mrdangist plays his solo.
--Srini.
b.chandramouli
>In article <1992Aug7.2...@cbfsb.cb.att.com> chan...@cbnewsf.cb.att.com (b.chandramouli) writes:
>>
>> 1) On The various Angas:
>> a) Are these artifacts generated as a result of the effort to
>> bring the very many talas that were present under one
>> unified umbrella (by Purandaradasa, I may be wrong here)?
>
>No, these are not artifacts. They do have a rhythmic function;
>they split the total tAla cycle into distinct parts. For example,
Thanks very much for giving a clear insight into many of the
aspects of the tala system that I had wondered about for quite
some time. I will read through your answers again later at
leisure and if I have questions or need for clarifications I
will raise them them. Thanks again.
B. Chandramouli
