western time signatures and various nadais/gathis.

arunk · Post by **arunk** » 22 Aug 2010, 21:09

(Mod note: This topic was split from here: http://www.rasikas.org/forums/viewtopic.php?f=9&t=13988 )

On a somewhat related side note: In fact for most people (outside of moderately seasoned carnatic rasikas) even a standard 7-beat or 5-beat (as opposed to the ubiuitous four or even the triplet) is hard to "tap to" perhaps because it is for some reason "unintuitive".

For example, yesterday I was at this party where a gentleman sang the malayalam version of this song (karaoke): http://ww.smashits.com/music/tamil/play ... ndran.html

This is in misra capu (I think) and I think it is Sivaranjani (Ilayaraja again - ah this brings memories - such wonderful music!). The gentleman of course sang beautifully in tune and to the beat but I was silently observing the listeners keep beat. They all did very good for earlier 4-beat songs but I could see them in their silent tapping of feet and hands that they had trouble with this one. They frequently stuttered and slipped. I believe I would have if I had not been exposed to Indian classical music.

Arun

vasanthakokilam · Post by **vasanthakokilam** » 22 Aug 2010, 21:52

Arun, Definitely. Add me to the list of people who will have trouble with keeping to the beat. Chapu's are not easy even for a moderately seasoned carnatic rasika. One way I can increase my success is to keep it at double the speed which to most people is not natural. They instinctively try it at 3.5 for MC but it trips them up due to the uneven beat length.

I think most people feel comfortable keeping thala for ta ka dhi mi ta ka ju nu type alternating stress patterns. And if there is a steady background beat, that helps people. Also, relying on the rhythm clues from the song itself is not foolproof. Mridangam is of no help unlike the steady drum beat in other forms of music since they also play to the layam of the song. You can see people getting tripped up even in fairly common occurrences like 1/2 eduppu where the beat point is silent. Or, applying to even seasoned rasikas, when the singer in the final flourish intentionally goes off beat. The arudhi-like stress is a big savior for me to get in sync.

I think, beyond MC&KC, thala itself is not a determinant of difficulty in thala keeping, it is the layam inherent in the song.
So a 7 beat thisra triputa or 5 beat kanda eka would be OK as long as there is not much vishayam and vishamam in layam. If the major stresses in the song falls on the beat boundary everyone is happy. For example, in Niravadi Sugada, the anupallavi, the first line is difficult but the second line is much easier and natural for many people.

vasanthakokilam · Post by **vasanthakokilam** » 22 Aug 2010, 22:18

Regarding the matra vs akshara issue, I agree the confusion in terminology is quite frustrating. I was on the same bandwidth as you two and I do not mind switching back

. Read this article by Sri. Kalidas in sruthi which "clarifies" this issue, agreeing with the interpretation of matra as the sub-unit: http://www.columbuscarnaticmusic.org/pd ... eaking.pdf
But he also says it is because of how he has heard them used by his gurus. But there is also enough material which use the terms in the opposite manner. Harini uses it in the sense of aksharam being the sub-unit. Hope she reads this and offers some comments.

Same thing with nadai and gati, are they the same or different? Depends on who you ask. And those who say they are different, do not clearly say in what way they are different. They all hedge their answers.

Nick H · Post by **Nick H** » 22 Aug 2010, 22:23

Hah! Can we have a separate thread on my chapu talams are non-intuitive, maybe? It is not just that they are odd numbers, because a 3-beat cycle is perfectly intuitive and foot-tapping. I suspect it is that 7 is actually 3.5 and 5 is actually 2.5 (many people define misra and khanda as 3.5 and 2.5, not as 7 and 5). It is that half-beat feeling of not quite enough or a bit too much that throws me.

I could put misra chapu talam to that film piece (don't know about it if you hadn't told me already, though) but had a persistent feeling that the edipu was actually not at samam, but maybe one beat after it? My layam is just not good enough to have tried that.

Whilst I do not feel that nadai changes, where the patterns used belong to that nadai, are konnakku, I certainly feel that singing the same pattern (even if sarvalaghu) to different nadais is. The excercise in post #10 is not easy: doing the same thing but only using "ta ka di mi" is much harder.

VK. thanks for the link to the Kalidas article. says it vastly better than I ever could! Of course, my small amount of laya theory knowledge comes from the same parampara.

EDIT: see here for definitions that suggest that aksharas-in-matra mighyt be of Northern origin?

arunk · Post by **arunk** » 22 Aug 2010, 22:55

Yes I dont know why 5 and 7 are more non-intuitive than 3 (which BTW Nick could also be viewed as 1.5 and thus your hypothesis about 2.5 and 3.5 would have to account for that).

vk - my point was even when there is a steady 7 beat like in that song - people can struggle. And I think perhaps has to do with the 7 being "stretched out" to a long enough interval AND with the stress points being of "alternate" duration (i.e. 1.5 vs. 1 and 1 for 7 and 1.5 + 1 for 2) - i.e. in many ways the stress points alternate in time obscuring any easy perception of a steady beat(?). But like in TSK case the 7 is done very quickly (like in that example), we perhaps do not perceive the internal patterns as stress points - i.e. we perhaps perceive the 7 as an atomic group and it is easier to perceive a beat (to tap to) (?)

And yes cm melodies (and percussive accompaniment) do not stick to any steady beat like in popular forms. I was recently discussing tihs with a friend who is into the multi-layered rhythms of Africa. I told him "in cm and perhaps in hm as well, the concept of steady beat isn there - the beat as in percussion follows the melody" - I dont know how fair that is but basically I was trying to say in some respect in popular music, the melody in many aspects follows/adheres/is-subservient to the steady beat. But in contrast, in CM while the concept of time and cycle etc. is there, within that cycle the rhythm follows the melody which can follow various patterns.

It was impossible to demonstrate a 7-beat melody to them using a CM song because there is no apparent steady 7-beat

!! The best I could do was do misra chapu talam as 1-2-3 4-5 6-7 with stress on 1,2,4 and 6. But then to make them see that a certain cm song in misra-capu adheres to this framework was quite hard. In fact that part isnt intuitive at all. And one could actually say the same thing even for many Adi songs in catusra gati.

And

vasanthakokilam · Post by **vasanthakokilam** » 22 Aug 2010, 23:20

Arun, that film song is not easy though it has a steady 7 beat. I agree with you that the alternate duration is the biggest problem.
I am not at all surprised that it poses trouble. That is definitely not intuitive to most people. I sort of view MC&KC as mixed nadai thalams. ( to coin another terminology to an already confusing terminology zoo: Misra Nadai thalam

ooch.. it hurts )

>But like in TSK case the 7 is done very quickly (like in that example), we perhaps do not perceive the
>internal patterns as stress points - i.e. we perhaps perceive the 7 as an atomic group and it is easier to
>perceive a beat (to tap to) (?)

Yes and also I think it is because the sense of heavy stress at the beat boundaries is much clearer ( at least prior to all this bEdam ) and so the distinction between beats and sub-beats is more clearly established. I think that is a prerequisite to bEdam of any kind. Establish a reference clearly and then vary it and come back to the established reference.

For a steady 7ish beat, how about sArasadalanayana in chatusra Jampa. It has a neat and steady rhythm with a strong arudhi stress point. I tend to think it would be much easier for a lot of people than many miscra chapu fast songs.

http://sangeethamshare.org/tvg/UPLOADS- ... hithar.mp3

( The click on the above link may not work. Search in sangeethapriya for "2-sArasadalanayana-kamas-dikshithar.mp3" )

arunk · Post by **arunk** » 22 Aug 2010, 23:50

I wouldnt have called that a 7ish beat at all

- in fact we would have a hard time convincing anyone outside carnatic circle it is so. They would view it as a 4x4 beat - but that there are 7 such ones in a line of a song - but that part isnt perhaps as significant to outside Indian classical music.

There are songs (even ones I know now that I recall!) which follow a steadier "smaller" 7-based melody e.g. cUDarE celulArA (pantuvarali, T in nauka caritramu). There is also an aTANa one in naukacharitramu that is a fast khanda cApu and where the melody more or less follows it.

Also in the notation typesetter software "internally" MC and KC are implemented as mixed-naDai only

- so your view isnt that crazy!

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 00:15

OK, if that is the case, let us go to the basics. How about the Triputa thala geethams like Padumanabha and Keraya Neeranu. That would be notated as 7/4.

In western notation, the non-chatusra nadai is notated using a different scheme for triplets, pentuplets etc. They still use the quarter note (or 1/8th note) but in the context of a triplet it is really not a quarter note.

Nick H · Post by **Nick H** » 23 Aug 2010, 00:20

Yes, only misra/khanda chapu has that feeling of half-and-half-and... feeling.

Odd (or 10 or 14) beat talams with a chatusra structure do not. They just feel like a line of fours

arunk · Post by **arunk** » 23 Aug 2010, 00:28

I am saying there CAN be a difference between misra nadai/gait and tisra-triputa and catusra jhampa. You can most certainly do tisra-triputa in catusra gati as well as tisra gati, Is it still a 7-based beat in both case? Certainly there is a "seven base" there but the gait as in steady rhythm (which is what a non-carnatic listener would perceive easily) isn't 7 based - is it?

But if you compress the tisra triputa (say sing the gitam in moderate speed) it would be easier to be perceive as 7-based gait i.e. misra naDai - i.e. the inner divisions either disappear or are so compressed they perhaps dont play a role in beat perception - this is how cUDarE celulArA is. The TSK misra nadai is perhaps the same.

But you stretch the tisra triputa out - it would be more difficult to see this. And if fill it in with ta-ka-dhi-mi it becomes tisra-triputa in a 4-based gait. You fill with ta-ki-ta - it becomes tisra triputa in a 3-based gait.

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 00:43

I think you are right, especially about the performance speed, which masks the underlying structure --- but I am fast getting out of my depth here :$

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 00:45

>I am saying there CAN be a difference between misra nadai/gait and tisra-triputa and catusra jhampa.

Yes. In fact I will go even to the extent of saying that if a MC song does not have the mixed nadai feel, it is really tisra triputa ( given the purported origins of chapu thalas from folk traditions where non-chathusra nadai is commonplace ). So this may all be a terminology issue with respect to this whole mArga vs dEsi origins.

I listened to the cUDarE celulArA just now. It definitely has the mixed nadai feel to it.

Back to the notation scheme, a tisra triputa should be notated as 7/4 ( or 7/8 it does not matter ). For misra chapu. the proper notation is ( I think ) 3/4, the first note is represented as a triplet of of 1/8 notes and the other two notes with a pair of 1/8 notes each. Give that to drummers and ask them to play it, they will play a neat MC pattern.

Now, how about Chathusra Eka, Misra nadai. The time signature is 4/4. Each of the notes will have the above 3+2+2 1/8th note structure ( triplet 1/8th note, duple 1/8th note and duple 1/8th note ) . Whether they provide a separate emphasis for the measure boundary or not is a different matter.

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 00:58

And we will have this issue as long as we do not clearly feel the sub-beat and beat stresses in the song. Without that, we can not reasonably discuss which category the rhythm belongs. Western notation clearly distinguishes how to notate them depending on the rhythmic feel one wants to convey.

Nick H · Post by **Nick H** » 23 Aug 2010, 01:04

Is it pertinent that western music is much more notation-based?

Although... Do you recall that video of a talk by Evelyn Glennie that was posted on this site recently? She showed how she could play the music as it was written --- and then how she could play it.

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 01:11

>Odd (or 10 or 14) beat talams with a chatusra structure do not. They just feel like a line of fours

Yes, and that is how it should be. As I stated above, in western notation also, the three concepts exist: Sub-beat ( duple, triplet, pentuplet ), beat ( the denominator, in relation to the BPM expressed as the number of quarter notes in a minute ) and the number of beats in a measure ( the numerator of the time signature ).

CM has three other markers: kaLai, arudhi and anga . The musical significance of these three have been discussed numerous times without much of a convergence in opinion. Other musical systems may not recognize these super-rhythmic principles, so be it.

Personally I think they are to be considered as super-rhythmic structures or layers that are built on top of the three primitives and not devoid of laya meaning. kaLai is a grouping of beats ( and NOT sub-beats ) at a sub-measure level, arudhi is a rhythmic anchor inside a measure and anga is a higher level grouping of the kaLai groups. Hence the concept of layers or super-rhythms. The rhythmic significance of anga in current times is not well established and there are several schools of thought but they seem to have originated from poetical meters. So anga may be significant for the interface between text and music.

arunk · Post by **arunk** » 23 Aug 2010, 02:23

I have tried understanding the numerator and denominator of western time signature and I have given up

- But from the wiki page, 3/4 applies to what we could take as tisra-gati. Whether it would apply to misra-capu style - I dont know.

Arun

arunk · Post by **arunk** » 23 Aug 2010, 02:47

The primary problem for me is the denominator.

From the drum-beat examples for in the wiki page for time signature (http://en.wikipedia.org/wiki/Time_signature), I see 12/8 (blues) and 9/8 are ones which to us are in tisra gati - but there is a clear emphasis on top of that so that in the first case the whole rhythm beat repeats for every 4 such triplets for 12/8 and three such ones for 9/8. So I can see how the numerator 12 and 9 are got. But why 8 for denominator? 8 implies eight-note and its duration 1/8th the duration of a whole note. So each measure in the first case is 1-1/2 whole notes, and in the second case 1-1/8? I having a hard time grasping that

One way I can see 1/4 and 1/2 in tisra-gati is if you think a ta-ki-Ta is 3 "beats". Then if you double the speed (ta-ka ta-ri ki-Ta) you can see each part is 1/2. If you double that you will have 12 and each would be 1/4. Double that and you have 24 and thus you have 1/8 for each part. But I dont think that is how the denominator works here or I am not seeing how the same principle is applied to the above.

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 03:12

arunk wrote:3/4 applies to what we could take as tisra-gati.

3/4, I think is rupaka talam, not tisra nadai, because the three beats is an entire cycle.

A waltz does not takitatakitatakitatakita: it is relaxed 1--2--3-- thing

Confusingly, though, it is where we tend to hear that 123 rhythm in carnatic music

arunk · Post by **arunk** » 23 Aug 2010, 03:22

Nick,

That is what I thought originally too - but the page says 3/4 is for waltzes as well. It looks like 3/4, 6/8, 12/8 and 9/8 are all tisra naDai based.

Waltz to me clearly seems the same as ta ki Ta

It is ta ki Ta ta ki Ta ta ki Ta

Albeit drawn out where a emphasis is sometimes on Ta. But to me this seems clearly tisra naDai.

Arun

arunk · Post by **arunk** » 23 Aug 2010, 03:28

But interestingly the drum sample for 3/4 is more in line with rupaka

- 12 34 56 with emphasis on 3 and 5.

it is all quite confusing! I am beginning to think that the time signature (the numerator) perhaps does not indicate gait.

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 03:41

Arun:
3/4 is not tisra gathi. Waltz is not tisra gathi, it is tisra Eka. I used to be confused about this also.

Our use of gathi/nadai as to how a beat is subdivided is NOT represented in the time signature.

Time signature consists of basically three things.

BPM - It is really not beats per minute, but number of quarter notes per minute ( quarter note is sort of the default, but other note durations can be specified also ). This specifies the overall tempo of the song.
X/Y -
Y beat interval. Whether the unit of the time signature is a quarter note, 1/8th note etc. Y in confunction with BPM determines the absolute time of the beat interval.
X is the number of Y note durations in a measure

That is all there is.

Another way of stating the time signature X/Y is, Y is the number of swaras per measure and X is the number of beats per measure. ( leave it if it is more confusing than helping

)

The reason the nadai ( subdivision ) is not part of the time signature is it can potentially vary from beat to beat. So it is considered part of the composition and not part of the overall time signature. We see that in CM with speed changes. Speeds in terms of changes in multiples of 2 happens all the time. Since multiples of 3, 5, and 7 are similar, it is notated in the composition itself. There is a separate notation for triplet ( tisram ), pentuplets ( kandam ) etc.

arunk · Post by **arunk** » 23 Aug 2010, 03:55

tisra eka => tisra naDai is exactly the same correlation between tisra-tripuTa and misra cApu and misra naDai. You got to be careful. Depending on the pace and intervening notes, you can make tisra eka into tisra naDai. A giTam in tisra eka (1 swara per 3) with words in triplets will have a tisra naDai feel.

Unless I am missing something - doubling/quadrupling notes within a beat does not change gait i.e. gati is the "base" if you will:
Catusra => 1 2 4 8 16 ..
tisra => 1 3 6 12 ..
khaNDa => 1 5 10 ...
etc.

Y still seems quite nebulous or atleast I cannot see the logic. Quarter /eigthnote implies quarter/eight of something. What is that something when notes are grouped in triple (say in a tisra gait) and pentuple (a khanda gait)?

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 04:00

>Quarter /eigthnote implies quarter/eight of something. What is that something when notes are grouped in triple (say in a tisra gait) and pentuple (a khanda gait)?

Excellent question, that is the crux of the matter. In western notation, they are relative to the BPM since that is specified independent of what Y is. If BPM is 80 quarter notes per minute, a time signature of 4/8 refers to 4 beats per measure, with 160 beats per minute. How such a beat is sub-divided is in the notation of the composition. Double speed will be notated with 2 1/16th notes joined together. triplet will be notated with 3 1/16th notes joined together. ( even though it is not a 1/16th note anymore in such triplet context but they do not use yet another symbol for each note of the triplet )

arunk · Post by **arunk** » 23 Aug 2010, 04:11

Thanks. That was clarifying. I now remember that this is what one of my colleagues mentioned when he tried to explain although he wasnt precise and thus I didnt get it.

Btw if you just google on say 6/8 you see many people confused about it and many people giving confusing answers

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 04:17

I know.

Here is an even sadder part. If you search for CM laya related things, our numerous discussions here at rasikas.org show up often at the top. We have managed to pollute the internet

Future generations, when they debate CM laya, are going to quote us to back up their claims... ha..ha..

arunk · Post by **arunk** » 23 Aug 2010, 04:19

- I have noticed the same too!

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 15:58

VK, I think you make a little too much of Beats per minute, as, in many instances this is not specified at all: it is optional for the composer to specify the exact tempo of the piece.

A quarter of what? A quarter of a whole note
2nd speed is double what? It is double first speed

It is, as they say, all relative!

In neither instance, western or carnatic, is the actual duration of a musical unit or the actual tempo specified or even implied. It is a case of the number you first thought of!

The carnatic waters are muddied on this by the "definition" of one akshara as one second in some sources, which I do not believe was ever meant to be more than an analagous indication of consistency, or maybe a practical suggestion as a starting point. Were it true, tala would have to be put at the same speed by every performer for every song, which is patently silly.

arunk · Post by **arunk** » 23 Aug 2010, 16:43

Nick. Of course is relative. The question is how can you add up quarter notes and eight notes to arrive at a whole note in a triplet/pentuplet based rhythm.

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 20:16

Arun, I freely admit that such a calculation is beyond me! I suppose it is relative to the individual unit, but I never learnt western music. I've been searching for my dictionary of music, which may have something interesting to say. I don't think it has run out on me after forty years, but I can't find it!

arunk · Post by **arunk** » 23 Aug 2010, 20:58

I like what this site says http://www.studybass.com/lessons/readin ... ignatures/

Again, the rhythms in each bar can be anything as long as they add to 3 quarter notes. This is where time signatures start to seem illogical and students often get confused. How can 3 quarter notes add up to a whole measure? You have to remember that all of our rhythmic terminology is based on 4/4 time since it is the most common. Youâ€™ll just have to accept the fact that music has some weird conventions just as any language. Think of all the illogical ways similarly spelled English words are pronounced.

:lol!! That was pretty weak.

Now if it were this way, it would seem more logical (to me, and probably only me

)

3/4: 12 12 12 => 3/2 i.e. 3 beats per measure with each having 2 sub-beats
6/8: 123 123 => 3/3 i.e. 3 beats per measure with each having 3 sub-beats
12/8: 123 123 123 123 (which some stresses in between making the 12 the repeating pattern) => 4/3
9/8: 123 123 123 => 3/3

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 21:12

3/4: 12 12 12 => 3/2 i.e. 3 beats per measure with each having 2 sub-beats...

But each beat doesn't have two sub beats! Surely it has as many as the composer wishes to give it?

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 21:26

Nick: Here is the reason the BPM is important conceptually. The BPM, whether specified or not, is the absolute and more significantly, my main point is, it is NOT beats per minute but number of quarter notes ( or any other specified note ) per minute. That is, it is specified independent of the denominator of the time signature. I think that is very critical in understanding what is relative and what is absolute.

There is no requirement that a measure needs to add up to a full note... The quarter note is just a name, do not attach too much English meaning to it. You can call it 'Buffalo note' and 'Sheep note' ( two sheep notes to one buffalo notes ).

Tempo is expressed as 80 Buffalo Notes/minute.
Time Signature is 3/Sheep note

Each Sheep note can then be sub-divided into 2, 3, 4 or 5 notes etc.

arunk · Post by **arunk** » 23 Aug 2010, 21:27

Sure but then why is 3/4 more appropriate where 4 => a quarter note. You got the same problem. I think I read it implies "a quarter note gets a beat" whatever the heck that means.

I have read some discussion of this. Most guys in the western music find the term 1/3rd a note completely perplexing (in context of "irrational time signatures" and almost a non-starter. This - even though they are quite familiar with tisra gati - they are somehow able to clearly map quarter/eight note for these - I just dont know how/why and why they dont see "three quarter notes for a triplet" also implies there fourth such notes make a whole note (which would never occur if the song maintains a triplet gait). So I on the other hand find the use of quarter and eighth note to describe triplet/pentuplet gait as a non-starter - however hard I try, mathematically it doesnt add up and so it isnt a starter.

Arun

arunk · Post by **arunk** » 23 Aug 2010, 21:29

posts crossed. Yes it does seem quarter/eighth are just mere terms and map to some absolute duration. But I do see some folks being troubled by the mathematical inconsistencies in applying to triplet/pentuplets.

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 21:46

Arun, yes that mathematical inconsistency troubles a lot of people and I was confused about that for quite a while until I talked to my nephew who is a drummer and a colleague of mine who is a guitarist.

Both of them were perfectly comfortable with triples and pentuples etc. In fact, I tried the exercise in Post #10 ( various splits with in a beat ) with my nephew and he breezed past that without any difficulty and he sensed that my timing on 5 and 7 are not quite correct even though I managed to squeeze them into one beat.

Keeping it at the simplest level I mentioned and not get confused on things about which the notation system is not, is quite a sane approach

Nick H · Post by **Nick H** » 23 Aug 2010, 21:58

VK, no... we cannot rename the notes, because, just like numbers, they only have a point if there are other numbers, and two will always be twice four, and four will always be a half of eight. They relate exactly to each other, as sheep and buffalo do not. I believe the notation system allows for 1/32nd and 1/64th (even 1/128th?) notes as well as double (and quadruple?) notes.

So why is the system (sometimes) based on quarter notes? Why is the tempo (BPM), when it is specified, given for the quarter note, and not the whole note.

The reason, I guess is the same that, counting adi tala, we do not take first speed as | ta ka dhi mi || ta ka ju no | which would be our semibreve speed. Crotchet speed, | thakadimi thakajuno thakadimi thakajuno || thakadimi thakajuno thakadimi thakajuno | is much more practical.

The note values are here. By the way, isn't it confusing that a whole note is called a semibreve and a double note is called a breve?!

Sure but then why is 3/4 more appropriate where 4 => a quarter note. You got the same problem. I think I read it implies "a quarter note gets a beat" whatever the heck that means.

It defines the beat for this passage: it says, whenever a quarter note is written, that is beat, and there are three of them in a bar.

6/8 says, whenever an eighth note is written, that is a beat, and there are eight of them in a bar.

Of course, you may not see an actual crotchet or quaver in an example of 3/4 or 6/8, but that is only saying that a song might have no single note that is one whole akshara in length.

I don't see why there is a mathematical inconsistency with triplets, etc --- but maybe that's because I'm no good at maths. If one can cope with the idea of carnatic nadais, which we all can, then western triplets, etc, work in exactly the same way. No problem with dividing that unit (of three, for example) into 1+1/2 1/2 1, or even into 1/4 1+1/2 1/4 1 1/4 1/4 or whatever.

arunk · Post by **arunk** » 23 Aug 2010, 22:02

They certainly know about fitting 3 or 5 or 7 in the space of 4 - because that is precisely how the notation scheme seems to work (i.e. everything in terms of standard/normal time).

Actually from what I read so far even those who claim for a need for irrational notes (1/3) they think it only in terms of a switch in gait

- For example: http://secretsociety.typepad.com/darcy_ ... bitte.html

Of course, these time signatures only make sense in the context of a more standard meter. You can't start in 5/6 -- you have to be chugging merrily along in 4/4 or whatever, and then switch gears to go into 5/6 for a measure, and then go back to a "rational" time signature.

I wonder why? You cant have a song that is fully in a 3,5 or 7 gait? Atleast this clearly isnt the case for all rhythms over the world.

Note however that
1. With the TNS example above (i.e. using a khanda gati pattern keeping sub-beat duration same) you are clearly talking about grouping 5 "notes" which originally were grouped in 4 (loosely speaking). So the western time signature interpretation makes perfect sense.
2. Even if you had TSK case (first sample - switch from catusra to khanda gati by shrinking sub-beat duartion), I believe they would represent by a shift in tempo (5/4 => 5:4) AND then do the same pentuple grouping . So this also makes sense.

But if the piece is entirely in 3-based or 5-based of 7-based is where it sort of becomes nebulous because there is no clear 4-based reference point (except for bpm). Note that bpm is not notated in stuff like bach etc. It is more generic - like moderate etc.?

Arun

arunk · Post by **arunk** » 23 Aug 2010, 22:09

I don't see why there is a mathematical inconsistency with triplets,

Inconsistency only when you express them using quarter and eighth notes.

For example take three equal parts so that they add to 1 but express each part using any combination of 1, 1/2 1/4, 1/8 etc. for each?

Arun

Nick H · Post by **Nick H** » 23 Aug 2010, 22:12

BPM is really a red herring, and is only to specify tempo. There is nothing in western musical notation that fails to work without a specified BMP.

You might like to have a look at this from which I see I should have been calling them "tuplets" all my life! I cannot get far into it at all without severe brain pain :$

For example take three equal parts so that they add to 1 but express each part using any combination of 1, 1/2 1/4, 1/8 etc. for each?

Isn't that what I just did? although my brain does not support smaller than 1/4 units!

arunk · Post by **arunk** » 23 Aug 2010, 22:15

You did 1+1/2 1/2 1 and 1/4 1+1/2 1/4 1. They are not equal (?)

Arun

vasanthakokilam · Post by **vasanthakokilam** » 23 Aug 2010, 22:15

1. Nick, my buffalo/sheep thing is just to make the point to answer the thorny question of 'quarter note is a quarter note of what'. The answer that it is a quarter note of a whole note is not useful since the whole note itself is not a fundamental concept. But I do agree that a 1/8th note is half the duration of the 1/4th note, in that sense, 1/8th note is meaningful. So, let us leave out the sheep, define 1/8th note as half the duration of buffalo note, 1/16th note as half the duration of the 1/8th note etc.... We will be in business. ( if buffalos caused more confusion than helping, sorry )

2. Regarding triple, pentuple etc.. the key point is, they are not part of the time signature fraction, they are notated in the composition.

3. The mathematical inconsistency definitely exists with triples and pentuple etc. If double speed of a quarter note beat is expressed as two 1/8th notes. a triplet is expressed as 3 1/8th notes.. which does not add up properly but that IS part of the notation system, you have to accept it and learn to use it that way..

4. I agree BPM is not needed to understand all this but for the point that the tempo is defined or understood without any reference to the denominator of the time signature. They are independent. If they happen to be the same, fine. So a medium tempo can be understood as 80 quarter notes per minute and the composition itself can be in 4/8 .

5. If an entire piece is in tisra gathi, chathursra Eka thala, you do have the overhead of specifying the time signature as 4/4 and then for each of the four beats of the measure, express it as three 1/8th notes. As per item 2, tiples etc. are not part of the time signature, hence this overhead in the western notation scheme.

6. Without changing gathi, if you just want to group notes of multiple measures differently, there is a separate notation for that. With that notation syntax, you can take 5 measures of 4/8 and group them into 4 groups of 5 notes each.

arunk · Post by **arunk** » 23 Aug 2010, 22:39

I think a note that is "1/3rd" long, instead of being notated as 3 1/8ths would perhaps be 1/4 + a dot (=> 1/4 + 1/2*1/4 => 1/8+1/8+1/8) - as the latter conveys the idea of one note rather than the former.

A note that is 1/6th would perhaps be 1/8 + dot (?)

Arun

arunk · Post by **arunk** » 23 Aug 2010, 22:45

Mods: I think from post #14 onwards (except suresh's question and my clarification - which are in relation to post #13) should perhaps be moved to a separate thread about western time signatures and various nadais/gathis.

Thanks
Arun

vasanthakokilam · Post by **vasanthakokilam** » 24 Aug 2010, 01:12

I think the dot notation can be coaxed to work for triplet etc the way you are thinking of, but I am not sure how common it is for representing triples etc.

Here is the notation diagram my colleague made to help explain how triples are usually notated.

Each group is a quarter note in duration:

â€¢ A triplet of 8th notes. Where normally 2 eighth notes make a quarter, here 3 are fitting within the duration of a quarter note.
â€¢ Triplet 16th notes. Each set of triplets occupies the duration of an eighth note. The two groups are tied together to represent a duration of a quarter note.
â€¢ A quintuplet of 16th notes, again totaling a duration of a quarter note (each note being slightly shorter than a regular 16th note).
â€¢ A quarter note.

(The notes arenâ€™t strictly eighth or sixteenth, but theyâ€™re the closest thing and the N tells you how many to fit into that space of time.)

Splits like 9 are possible but not common. It is sometimes used in transcription, especially guitarists who try to slam a lot of notes into a beat. Here is one such representation

Nick H · Post by **Nick H** » 24 Aug 2010, 02:00

vk... you are making me think that, although everything is defined, the definitions are meaningless without reference to the whole.

I agree BPM is not needed to understand all this but for the point that the tempo is defined or understood without any reference to the denominator of the time signature. They are independent. If they happen to be the same, fine. So a medium tempo can be understood as 80 quarter notes per minute and the composition itself can be in 4/8 .

This bit I do not follow. They are measures of different things, so they cannot "happen to be the same". One is a measure of structure, one of speed. How would a tempo of 80 be related to 4/8, or any other time signature?

One thing... After this thread, I'm never again going to boast to my Western friends about the complexities of carnatic music!

arunk · Post by **arunk** » 24 Aug 2010, 02:44

Thanks vk (this is how the wiki tuplet page also describes it) I think I get it now - just have to suspend attaching any meaning to quarter and eighth which would take some discipline

. Atleast I know now not to hit that brick wall !!

Nick - I always by default assume all systems are equally complex and sophisticated albeit sometimes in different aspects. To claim that one I like is superior is to me just a sub-conscious (and hence devious) form of ego-stroking

Our ego does wave that carrot in front of our minds often - many times in subtle disguises - thus is quite hard not to fall for it 100% of the time

It is just many forms of "My tastes are refined and superior. I am that good and above more basal ones. You just wouldnt understand"

Arun

vasanthakokilam · Post by **vasanthakokilam** » 24 Aug 2010, 03:01

>One thing... After this thread, I'm never again going to boast to my Western friends about the complexities of carnatic music!

May be you still can to pop musicians but if you talk to, say a drummer in a marching band, they will see you and then raise you!! I think the CM complexity is in the way the sophistication is put to use as a rule in most concerts than as an exception and that too in the improvisation context. But I do not think the concepts are complicated, it is just that it is not clearly explained. In either case, pulling it off in a live concert is no mean feat.

>They are measures of different things, so they cannot "happen to be the same". One is a measure of structure, one
>of speed. How would a tempo of 80 be related to 4/8, or any other time signature?

I wish I never went there since I have unnecessarily confused you. I think you and I are on the same page, the confusion is in the explanation. At one time, I was confused by all this and that insight helped me understand.

What I meant by "happen to be the same" is, if tempo is defined as N quarter notes per minute and the time signature is 3/4, the beats are quarter note durations and tempo is defined using the same unit. So if you want to know the length of the beat interval, it is pretty straightforward. It is same as the tempo definition, namely N/60 seconds
Keeping that tempo definition the same, for a time signature of 3/8, the beat length is N/120 . All of this is to point out that the nomenclature of "beat" in the "beats per minute" in the tempo definition is NOT the "beat" that is defined by the time signature.

>although everything is defined, the definitions are meaningless without reference to the whole.
I did not get what you are saying... see if you can explain it further given the above, if needed.

Nick H · Post by **Nick H** » 24 Aug 2010, 03:18

Arun, yes you are right about the ego stroking. Best just to enjoy the music!

VK, you are right too, especially about the improvisation. Anyway, I love Carnatic music whatever. But then, I guess I love lots of music!

Got you now on the tempo thing. I think it is a mistake to try to attribute fixed values, as I said about the 1 akshira = 1 second thing. It does confuse!

Thus... what is a quarter not or an eight note? Well, without the bar of which it is a part. along with its time signature, it is meaningless! What is the duration of a tuple? Again we have to see it in context. Hence we need the whole, or at least a big enough unit of it.

By the way, being able to sight-read all this stuff is amazing in itself, but how about being able to read multiple-instrument parts on a page of music, and hear it in your head! One guy who I met who could do this didn't seem to think it was anything really remarkable. Rather like the violinist, who, when I asked, "how on earth do you know where to put your fingers?" looked amazed, and said, "I just put them in the right places!"

vasanthakokilam · Post by **vasanthakokilam** » 24 Aug 2010, 04:10

arunk wrote:I think a note that is "1/3rd" long, instead of being notated as 3 1/8ths would perhaps be 1/4 + a dot (=> 1/4 + 1/2*1/4 => 1/8+1/8+1/8) - as the latter conveys the idea of one note rather than the former.

A note that is 1/6th would perhaps be 1/8 + dot (?)

Arun

I know we are at a good state here in terms of resolution to all issues, so I do not want to spoil that necessarily

Arun, I am not sure if you wrote the above pre-resolution or you still mean to use the dot notation to represent gathi, ( single dot, double dots, triple dots etc. ).... what you need to come up with is a note duration of 1/12 if you want to split a quarter note into 3 for thisram.

Nick H · Post by **Nick H** » 24 Aug 2010, 04:15

A note that is 1/6th would perhaps be 1/8 + dot (?)

It's late bed time, so no time for me to try sums but...

1/8 dotted would be 1/8 + 1/16 ie 3/16

let me try double dotted!

1/8 + 1/16 + 1/32 = 7/32.

Now I'm going to bed