The tradition of notation is not as firmly entrenched in Carnatic music as it is in, say, Western music across genres. There are two diametrically opposing views on the nature and use of notation in Carnatic music among practitioners. One view - the predominant view- among Carnatic musicians is that Carnatic music cannot be notated as it is an oral tradition and that no useful purpose is served by any type of notation. A slightly modified version of this view is a scornful rejoinder, “busy professionals do not have the time or the inclination to learn from notation and they prefer the Ipod as a teaching device”. The minority view is, of course, the pursuit of honing notational skills to improve the status of notation in Carnatic music (Subbarama Dikshitar, Ranga Ramanuja Ayyangar, Vidya Shankar, Govinda Rao and many others). As in many things Indian, each system of notation assumes that there is no history or (relevant) literature and assumes that theirs is the best system without any discussion.
The former view, in my opinion, should not be dismissed as a prejudice arising out of musical illiteracy. I shall endeavour, in this paper, to show that this view is not without some sound logic to back it. Although I myself belong to the small group which espouses the cause of notation (and my learning, after the first few years through the face- to- face mode from my mother, has been entirely from notation), I shall argue that the art form of Carnatic music is basically an oral genre. It is now recognized in world literature that there exist two, entirely distinct traditions, namely oral and written. The oral tradition, by its very nature, is informed by distinct traits like the lack of fixed texts, the freedom to interpret and the disregard for acknowledging a borrowed idea/metaphor etc. Repetition and/or borrowing from (an) unacknowledged source(s) do not imply lack of originality. In the oral tradition, the object created is, in principle, an ephemeral object; the act of creating on the spur of the moment shapes an art object which ceases to exist the moment the act of creating the object is over. Any reproduction of the art object ceases to be of primary interest. In this sense, oral art is like spoken, day to day, language discourse. The unrehearsed, spontaneous flow of language is always ‘new’ and ‘creative’ but is not really worthy of word for word reproduction. Getting back to Carnatic music, if the aalaapanai, taanam, niraval and swarakalpanai are like the free flowing discourse in day to day life, the art object, the composition is comparable to epic stories/lyrics etc in the oral tradition which differ in every telling but is essentially the same in each narration.
It is in the nature of the oral tradition that no two events are exactly alike. A sentence whispered around a group of people individually will always end up very different from what was said to the first person. Similarly no two renderings of a composition are exact replicas. Over time this becomes even more obvious. For example, if one examines the rendering of a Muttuswami Dikshitar composition even in the same lineage, one finds that the individual traditions have become distinct in the present renderings. I take the example of the notation in Subbarama Dikhsitar’s Sangita Sampradaya Pradarshini (henceforth SSP) and the renderings by the veena vidushi Kalpagam Swaminathan and the renowned vocalist, late D.K.Pattamal of the composition “Kamalambike” in the ragam Todi (the former a private recording and the latter a recording of an All India Radio broadcast) . Both artistes trace their musical lineage to the Dikshitar parampara/tradition. Yet their renderings of the pallavi and anupallavi of the compositions differ from the SSP notation and from each other in subtle ways. Without going into notational details (to make the presentation more accessible to the reader), when we listen to the two renderings we find a few major changes from the SSP notation. Firstly, whereas each line of the pallavi and the anupallavi in SSP starts on the beat, in both renderings, the second line of the anupallavi starts two moras after the beat (arai idam). Secondly, both renderings have different sangatis (vatiations) on different lines of the compositions whereas SSP does not indicate any sangati for any line. Finally, the musical phrasing in D.K.Pattamal’s rendering is more rhythmically oriented. For instance, the sequence ‘ambikee’ becomes rhythmically interesting because the starting point is one mora (kal idam) after the beat. A more detailed examination will, I am sure, reveal more deviations from the SSP notation and also between the two renderings. Finally, the way the notes are rendered by the two artistes, not withstanding the diacritic markings in SSP, specially the versatile notes of the ragam, namely E flat and B flat must be apparent to any discerning listener. The point I wish to drive home is that, notation not withstanding, there is no such thing as hundred percent authenticity; there cannot be. As nature abhors carbon copies (or Xerox copies in modern parlance), so in the oral tradition too, there can be no such thing as a hundred percent authentic rendering.
Just as in language where purists frown on change/innovation and believe in a ‘pure’ variety of the language but the ground realities are far removed from the purists’ claim, so too in the predominantly oral tradition of Carnatic music, change is the only certainty. Otherwise how does one reconcile oneself to the fact that even in musicians in the same lineage we find many irreconcilable differences. For instance, Kalpagam Swaminathan’s use of the accidental note E in the ragam Manji not found in either SSP or in D.K.Pattamals’ style of rendering. Many more examples can be cited but this should suffice to support my argument that the oral tradition of Carnatic music cannot be captured forever in any notational system. Just as in the English word spelt ‘knee’ which is true to a much earlier pronunciation when English allowed the cluster /kn/ word initially and the vowel also rhymed with the vowel in the word ‘age’, while the spelling reflects an earlier stage of the language, the language has under gone many changes over time and across groups of speakers.
This brings me to the central point of my thesis which is that, in principle, a writing system for any oral system is only a convention agreed upon by the community and never a true reflection of the physical reality of the oral event, be it language or Carnatic music. Even in orthographic systems which are syllabic, as in Tamil or Devanagari, the actual pronunciation is never reflected in the writing system. While in Tamil one symbol is used to capture two sounds /p/ and /b/, for instance, in Hindi too, what is written as /mahal/ is pronounced with two different vowels and the ‘h’ may also be missing in some styles of speech or the orthographic symbol ‘u’ in words like ‘put’ and ‘but’ in English. Similarly, in Carnatic music too what we call a B flat in the ragam Madhayamavati may admit many, many variations in pitch (pitch manipulations). Even a simple phrase like [G, B flat, B flat] in the charanam of the composition “Palintsu kamakshi” of Shyama Shastry in the line beginning “kantamaku…” may admit different pitch interpretations across iterations and across performers. There have been several attempts in the history of Carnatic music at capturing the exact inflections of pitch movements that take place using fairly elaborate systems of diacritic markers on the notes. I would now like to show that these attempts are, perhaps, misplaced given the nature of Carnatic music. Let me begin with the observation that ‘notation’ is a neutral term that covers a wide spectrum of writing systems. In the linguistic literature, it is customary to distinguish between orthography or writing systems on the one hand and transcriptions on the other. While writing systems are meant only for the initiated (and hence the use of the same symbol ‘u’ in words like ‘put’ and ‘but’ do not create any problems for the user (may be some for learners and people wanting to initiate spelling reform), transcriptions are not meant for the ordinary language user but for specialized, research purposes. Even for language research, the same type of transcription will not suffice across various sub-disciplines. For instance, for examining the sentence patterns of a language what is known as ‘broad transcription’ will do. This type of transcription indicates the broad pronuncitation of the words, their morphological composition e.g., derivational and inflectional markings etc. For undertaking an analysis of how words change their shape/pronunciation in the context of word formation e.g., ‘five ~ fifth’ or the pronunciations of the symbol ‘s’ in the words ‘dogs’ and ‘cats’, we need a transcription which is slightly more detailed than the one required for studying sentence patterns. Further, for studying the systematic difference between say, English and French in words beginning with the symbol ‘b’, we need to indicate more details pertaining to vocal cord vibration, the starting point of vibration, the intensity of the vibration and the duration of the burst when the lips are opened etc. Therefore, the transcription gets more and more detailed depending on the nature and area of research. In principle, while the speech event is a continuum and the transcription is always indicated by discrete symbols, no transcription can capture all the physical details.
The drift of the argument above must have become clear by now. We can now apply the same principle for notating Carnatic music. We start with the theorem that the type of notation to be used depends strictly on the function of that notation. If a notation is meant to jog the memory of the performer, a broad notation is more than sufficient. In fact, given the nature of Carnatic music, performers would find any other type of notation an infringement on their rights. To be told that a note must be held in a certain way with a diacritic marking either gives them information they already have or tells them something they are going to disregard as it may be contrary to their style or their intention for the particular rendering/iteration. Let me illustrate the point with the note F sharp in the ragam Kalyani. In the phrase /E F sharp G/ rendered in normal tempo, any one who is a practitioner of Carnatic music knows that the note F sharp is a pitch movement going upto G, backtracking to F sharp and then going up again to G (labeled an augmented F sharp in Indian music theory). However, a long F sharp at the beginning of a phrase has several options for the ragam Kalyani and it is the sole prerogative of the performer to decide which option to select and no one, including the composer (or his heir/disciple) has the right to dictate to the performer which variant to select. In this context, a notation which includes diacritical marking specifying the variant is totally inappropriate and this may be the reason why established performers rightly decide to ignore notation altogether. However, when notating a composition in a ragam not commonly used, diacritical markings spelling out the idiosyncratic values (if any) of the notes become necessary, even for established artists.
I will now give an instance where diacritical marking may have actually helped a performer learn a composition not widely performed. But what we find is that where one expects to find some useful diacritic marking on a note we see none. Take the notation of the Muttuswamy Dikshitar composition “vina pustakadharini’ in the ragam Vegavahini (popularly known as Chakravaham) in SSP. The note F on the fourth beat of the first line of the pallavi is notated as a bare F which is also long. This F is part of the phrase [F,G,]. In this context it is unusual to find a long, bare F. Ranga Ramanuja Ayyangar (Kritimanimalai, part III) makes a minimal change notating the note as a short one. This certainly gets around the awkwardness in the SSP notation. However, if one looks around, the only performers who included the composition in their repertoire, the late Brinda and Mukta who must have learnt it from their illustrious grandmother, the legendary Veena Dhanam render this F as a deep pitch curve from E going upto G. It is upto us, researchers interested in describing prevailing music practices to decide whether we should go by the notation in SSP or the rendering of Brinda and Mukta (recordings available).
This brings me to the last point I wish to make, namely the scope of notation as a research tool in Carnatic music. Since only a minuscule minority uses the notation in Carnatic music to learn compositions they have never encountered before (the only acid test of notation as a learning tool), since notated texts across practitioners are never constant and there is no one standard notation of any composition, what is the scope of notation as a research tool in contemporary Carnatic music (with or without diacritical markings)? Let me rephrase the question by contextualizing it. For instance, in this day and age, can we say with a great degree of certainty that Muttuswamy Dikshitar must have set the music of a composition of his in one specific, indisputable sequence which can be immutably notated? Since notation was never a teaching tool, we would never know whether his descendants/disciples (including the author of SSP) had absorbed the master’s intention totally. We can only do the next best thing and go back to recordings of renderings as far back as we can and establish the commonalities across renderings. By doing this we may arrive at a reasonable picture of the shared features that constitute the great composer’s style. Of course, such minute attention can and should be paid to other composers too in the interest of musicology.
For undertaking a mammoth project of this kind, of course, a detailed system of notation or rather transcription would be an invaluable tool. Examining renderings across performers spanning diverse musical traditions, schools of music, generations would require a vast array of diacritical machinery which we will have to establish first (the traditional labels we already have may not suffice). Such transcriptions should include, apart from a array of diacritical markings, indicate phrasal boundaries, prominence, emphatic prominence, linking between phrases etc (see Vijayakrishnan (2007) for a sample). Needless to say, such detailed transcriptions must based on verifiable recordings and they will be rather difficult to undertake on a large scale by any single researcher. As in any rigorous scientific research, we need to train a group of researchers in the art and science of such detailed transcriptions of Carnatic music. We would also need to establish empirical tools which can falsify any claim made on the basis of incorrect transcription of recordings etc. But are there any takers for such a project?
References
Subbarama Dikshitar: 1908. Sangita Sampradaya Pradarshini (the English Web Version dated 2008).
Ranga Ramanuja Ayyangar. 1967. Kritimanimalai III (second edition)
Vijayakrishnan, K.G. 2007. The Grammar of Carnatic Music. Mouton de Guyter:Germany.