Do we really perceive 22 notes?

[Manian]

Why we do not have 4 variations in RI,GA,MA,DA,NI
Does really 4 Ris,Gas,Mas,Das and Nis possible in singinging Carnatic Music?


1. Linguistically, the position of the tongue when producing “a” sound is the middle of the tongue and in neutral
position. Thus, the oral cavity can not change if the note “a” with it's frequency components are maintained at
750-950-1100 Hz (Cycles) (Based on Spectrograms).
2. The same is true for “i” production where the frequencies are 190 HS-1900 Hz and 2450 Hz.
Thus, the vowels do not change in their place of articulation or manner.

Then, the only place changes can take place is the vocal chord's frequencies.
By pushing up the vocal chords slightly the fundamental frequency of the
singer can be raised and that is how it is possible to produce variation in the
perceived notes.

3. The biology and the physiology of the ear and it's perception range also dictates that two notes should be distinguished by at least 7 cycles (7 Hz). If two successive notes thus overlap say, 6 Hz range, then the ear has a mirage of either the lower note or the upper note. To prevent this confusion, successive near notes are prohibited (or Sanskrit Varjit) in Carnatic Music. This subtle control of the vocal chord becomes some times violated thus, the listener start talking about this “raga” or that “rage” which is a blind spot.

Why this knowledge is necessary to understand singing is, the physiological changes of the vocal chords and the mouth configuration is not very precise, thus
extremely difficult to control. A very deep and sustained vocal trailing – Sadaka is the only way to achieve this.

[ganesh_mourthy]

It is a controversial topic.

But in my experience, and trying to decipher through the spectogram, it is near to impossible to show this variation. In fact , the opera singers and western musicians( singers) have extra precision over their notes ( but even there , nobody is 100% if we take into account a 5 min or so recording) . We do not so much focus on this much. Many a time it seems like the Sa and Pa is enough and you have a leeway for other notes, give or take. As the speed increases the precision is sacrificed. I have noticed this with all singers and with some it is more. When 12 notes in itself hard to achieve I wonder about 22 notes . It is a huge strain to the brain than to the vocal chords to focus on such . It is all about preparing the mind of the listeners over the years, and then the adjustment of volume and gamakams give such a feeling.

[Manian]

You are right. Now, the Russians have created a spectrogram to read the emotional part of the vocal chord region. I would like to get a hand on it, but unable to get one. Just like language does not allow all sentence types in Aviyal sentence, the Raga statements by the selected notes do produce different musical sentences which are expressed over a natural language structure. Just like a mathematical model (skeleton) is hidden behind a statement and can be perceived only a person who has both perspective (like a doctor who knows the skeleton likes behind the skinned person), the singer does the exposition of the musical sentences. Thus, just trying to isolate and sing is difficult.

[ganesh_mourthy]

It is volume adjustment and make-believe. When I listen to people describing that certain raagas have ma2 lower than usual and in certain raagam Ni lower and closer to Dha, and in some cases g3 even closer to Ma or r1 closer to Sa than usual, It is impossible to really execute it or an ear to capture it. The spectrogram and the theory are far apart. even 12 notes do not fall on where it should be , let alone 22. In the olden days and in the absence of a mic ( if those days are the origin of all these theories ) it is humanly impossible when they had to bellow out. When I listen to ARI and others of those years sing it is more of a pattern than notes. But , still that was a feat to achieve in a temple with all the noises that is not acoustically constructed for singing or music.

[Manian]

Ganesh: Personal opinion and scientific verification are two different things. It is not volume adjustment - opening the mouth more which increases the volume of the mouth and amplifies versus manipulating the vocal chords to increase the perceived pitch or sruiti. Like a Yogi can do many poses - twisting his body and muscles but an ordinary person without years of training can not do it. The point is, when teaching, the teacher produces those notes but does not tell how he produced those notes. Self awareness of what strands or bands in his vocal folds are manipulated can be seen only with an endoscope. Such research is done in the west. In India, in general neither we have such a facility nor the musician interact with medical profession to do this type of research. There is no motivation or monitory benefit to a Music researcher - no permanent job. Should Music college teach this? Yes, they need a very dedicated Physics teacher with deep interest in Sound, Linguistics, Neurology etc. Of course such a person will not get a job nor a wife!. So, these discussion should have some scientific evidence. Do I have it? Yes!. This reply is not to silence you or prove that you can not have your own views, just a clarification only

[shankarank]

There are 12 svarasthAnams. Existence of 22 Srutis - I take it as a postulation , based on vAdi/samvAdi relationships that are not realized as a particular sustained svarams, but rather in the context of a rAga.

Vrinda Acharya has an explanation here in this video: https://www.youtube.com/watch?v=BXXTmrrmgvw

But somehow she sticks to empiricism. To quote : "Sruti is the smallest / finest interval that a human year can comprehend or an instrument or voice can produce". Relying on empiricism will lead to actually studying humans and gets into which humans, trained and untrained, trained under what conditions ( like for e.g the Veena Dhanammal audience) etc. However she makes an interesting point that a Sruti could be realized involuntarily in the context of a rAgA exposition.

I always found people specifying intervals like cents, ( or the 6 or 7 Hz in the opening post) and the capabilities of human ear to distinguish etc., to be missing something here.

We have to ask how do we cognize the 12 svarasthanas to begin with. Various tuning systems have been discussed in this forum and before in earlier forums: natural, just and the geometric interval (12th root of 2) lead to equi-tempered also in Western systems.

The fundamental principle is still consonance (vAdi / samvAdi). All of them are just various derivatives of consonance and equi-tempered only came with the 12 because consonance is the basis of 12 to start with!

Hence, theoretically if consonance can allow us an initial 12, we can ask what else it can allow next. And by extension the next potential is 22.

The 22 cannot be addressed based on finest intervals that a human ear can distinguish, but a Sruti not in the original 12 could be realized from some other distant navigation ( vAdi/samvAdi) and not necessarily from any neighbouring spots.

And some legends doing the rounds claim Pt. Bhimshen Joshi can reach 51 spots - 51 is a nice auspicious number.

Existence of the graha bEdam ( Sruti bEdam) exercise in Carnatic tradition - and this may be a possible empirical evidence - where a musician goes from tODi to kalyANi , certainly opens the possibility of additional Srutis coming into play when the musician is sustaining the kalyANi in the backdrop of a tODi Alapana.

If I may end with a catchy phrase - 12 is given by Smriti ( the initial rules that a teacher would use) and Sruti ( the inner and external hearing) could lead to 22 or more. 22 here actually means infinite - 22 being a Smriti-fication for convenience - if we remove the philatelistic noise and look at it philosophically. It is the first iteration of a recursive process.

[Manian]

Good point. Having said, the 7 Hz perception window is well documented in Western research Papers. What we have is a grammar and tradition. If you take Panini and Tolkappiyam the rules were formulated at that time but have undergone changes through ages. A new Sanskrit Grammar(unfortunately we can still only follow the old one as the total population that speak Sanskrit is very limited) and a modern Tamil Grammar (just to give respect Kannadasan, imaginging him write a new Tamil Grammar) would be different. The two vowels coming together in historical grammar is seldom spoken or written today. Anyway, when we cling to old ways, we use the "dogma' but refuse to move forward with more scientific investigation. That is progress. But we do not have any such fully scientific investigating foundation in India to investigate how our Carnatic music is perceived, processed in our neural circuits in the brain. Therefore, whether we perceive 12 or 17 or 22 is subjective impression only. Fine, what is the point is, we have a huge treasure of knowledge and like the British destroyed out system, but follow their curriculum and University structure, we are doing the same in teaching Music. Now, how do I know? There was an effort at Tata Institute of Fundamental research,(TIFR) in Mumbai and one man on his own investigated the Physics of music and his paper" Simulation of Instrumental Music, 1970, Computer Society of India has details. He was after synthesizing Madurai Mani Iyer, but started with simulating wind, string and struck instruments that he had never heard off or used but was able to simulate it. He produced both Veena, Oboe, Jalarathangam. He also gave an invited talk on "Computer's in Arts" a two day seminar organized by IBM and West German Embassy in 1972. He was the only speaker. Kumudam published a cartoon titled “ Mr. Computer here is our offer coconut etc; Statesman Calcutta published a wrong editorial stating that IIT Kanpur Prof. N. Ramasubramanian is going to get ride of all musical teachers. Unfortunately, he did not try any thing like that. So, scientific investigation should be the basis of our views and not “dogma”. But, in the absence of all such information, we are fighting rather than clarifying things for the younger generation. With AI and deep learning Madurai Mani Iyer will arise via a computer program in the future.This is not an attack on your views, rather only a clarification.

[rajeshnat]

Manian
You have joined recently and really appearing as a breath of fresh air. In future Please add whitespaces and have content as more paragraphs for better readability. I am assuming you are a senior citizen and with your references centered on TIFR in 70s , i am assuming you have moved from india to USA and settled there.

I just googled as anything and everything on MMI interests me . I am assuming you are referring Prof Narasimhan of IIT kanpur. More about it please in Madurai mani iyer thread in Vidwans and Vidushis section

[shankarank]

Statesman Calcutta published a wrong editorial stating that IIT Kanpur Prof. N. Ramasubramanian is going to get ride of all musical teachers.

Since you mentioned TIFR, IIT Kanpur and the professor's name, I could connect all three and I know that person and talk to him from time to time. If indeed you are him, posting under the short name : manian, then we will talk soon. Prof. Ramasubramanian of TIFR/IIT Kanpur lives in Dallas area, If you and I are talking about the same person.

As regards AI : You need to teach it , train it, still it doesn't "know" it seems - I am only a beginner there

https://youtu.be/IHZwWFHWa-w?t=907

It could recognize 22 Srutis more, but it could not sing it?

Well , we will see it might one day - that is the hope of humanity - as we cannot fund artistes anymore!

[Manian]

Shankaran:
I wish that I am him. I live here in Bangalore. I met that gentleman at Prof. Narasimhan's place in Banaglore before he died. He gave me a copy his material and the research he had done and claimed to be unique. He felt that the research was not carried out by others because they did not have the multi-disciplinary backgroun that Prof. Ramasubramanian have had. I kept some of these articles Prof.Narasimhan gave and gave contact to Brigadiar Balasubrmaniam who was present at the Delhi IBM_West German Embassy talk. I understand that the professor lives in USA. I asked Prof. Narasimhan how did he recognize that Prof. Ramasubramnian he gave a detailed reply. Thanks to Padma Sri late Prof. Narasimhan for this information.

[Manian]

rajeshnat: Thanks. I do not know the restrictions of writing here. But your thoughts are worth following:
The way I think (an have learned ) is known in mathematics conjecture or hypotheses. I collect articles , read , re edit and digest. I discuss that with a very few selected friends who have no insecurity. That is, they are scholars on their own, thus do not argue, but ask inquisitive questions. So, I have developed interest in many subjects and trying to connect them to understand things around us. This is known as "Domain General Intelligence" as opposed to "Domain Specific Intelligence - which most people have : Physics, Maths, Music, etc.

If you think this domain specific intelligence as the stem of an umbrella, the ribs and the cloth over it along with all assembly is the Umbrella. Another way is to think- a Coconut tree or Palm tree gives lots of benefit, but one can not sit under it when the Sun shine is hot. In Tamil, "தேம்படும் பனையின் திரள் பழத்து ஒரு விதை வானுர ஓங்கி வளம் வளரினும், ஒருவர்க்கு இருக்க நிழலாகாதே என்று சங்கத் தமிழ் சொல்கிறது. Therefore most of the discussions do not educate us, rather dogmatically repeating what we hear but not do the research. By sharing what I have learned I am inquisitive to refine it, rather than being right or all knowing. Sharing is caring. On AI: Google and others have a very large data base to learn speech. I dream if some very really rich person established a foundation to collect music data base so that the “Deep learning” method of AI can bring out the wonders and make us happy. But will any one in India do is a question. I am greedy for knowledge only and my ego is 0.001%(necessary to survive).

[Manian]

Look at these incomplete tables on Carnatic Music: Type of articulation
Labial + Vowel alveolar + Vowel Velar + Vowel
===========================================
Friction of air SA
---------------------------------------------------------
Stop flow of ai PA DA GA
--------------------------------------------------------
Nasalization MA NI
--------------------------------------------------------
Trill (Tongue) RI
--------------------------------------------------------
NA Now look at Madurai Mani Iyer's genius
LI
Substituting "la" and "na" he tries to avoid da, ra , ri which are harsh. But when he
sings, La La La, and uses all the vowels which are not used in our learning Music, what he had
discovered is if you sing in lower octaves with melodious vowels, you get intoxicated.
That means, the brain produces "dopamione" a chemical in wine, old rice soaked in water
(5% alcohol). You hear Rajarathan Pillai's Shanmugapriya this dopamine will flow till the
next day when people will be murmuring when they take shower. That is divine music.
But just learning to sing and just the Kritis of others is necessary but not sufficient to intoxicate
you. It is not the language, but the notes and the emotional range they kindle is what counts.
I sing Kanthimathim and almost get lost if you really separate each word and sing in Madyama Kala and
the meaning will appear as a flash:

Sri Kanth Mathim Shankara Yuvathim
Sri Gruruguha Jananai vanthe aham

Next you can join and sing the words. Dishitar wrote it for Venaa thus has to be very subtle in kindling
the emotions.
Two paise worth my thoughts

[Manian]

Sorry for some minor errors and I could not post a right table. Is there a way? My screen shot also could not be posted.

[shankarank]

Just as a recap one of the two publications that Arvindh Krishnaswamy from Stanford published. Especially to note are his comments on Saveri - Rshabha and Gowla Rshabha!

This from production side of things!

https://ccrma.stanford.edu/~arvindh/cmt/smac03.pdf

[Manian]

Thanks Shanaran.. I had that copy too. What is being said is that even though we have a huge historical treasure of knowledge, there is not a single MIT/Stanford where one can carry on such investigation. Also, there is a need for Neurologist, Psychologist, Musician,m Physics, EE in a group and investigate our system in a scientific manner. This is not a job creation method, rather understand and see if it can be applied to help mentally challenged, autistic, stress reduction and an array of heath related fields. West will do this if it only a business.

Secondly I had contacted some really good musician asking them that if Kalyani has three emotions (as claimed in one of the video by a female musician,)she did not sing Pallavi, Anupallavi and Charam in three emotional mode and also selected a lyrics without saying why did she selected that? This is not to question them or demean them, but like tree waiting for a shower, will it not be divine?

Shemmangudi has a UN record exposing the rasas with example: Sringaram Srinidi Nandine Vikarane - The sringram is the sea touching the sky- so when one sings can they not bring that emotion and flood the raskikas in a highly spirited mode? Why can we teach mood , say Mukari can be heard in Oppari in a dead person's house, but can we not imbibe the same emotion in Mukhari. So, we have lots of questions but answers are rare to come. Personal opinions can not substitute scientific investigation and duplication.

[Manian]

Sorry- Shankaran.

[Manian]

Sankaran: Here is one of many of my collection on Fundamental Frequency of emotion in speech which I would like to investigated with our music:

http://www.researchgate.net/publication/229438971
Fundamental frequency analysis for speech emotion processing

[Manian]

Here is IIT-Madras Paper on Raga Pitch
compmusic.upf.edu/system/files/static_files/Ishwar-ISMIR-2013.pdf