Manuscripts in the possession of Śivakumar, a descendant of Tanjōre Quartette

Our music was propagated by two routes – oral and textual. Though we have a textual history of approximately 150 years recording the compositions of prominent composers, the corpus of compositions recorded by this way cannot said to be complete. Also, many compositions exist only in paper as they are not extant in the oral tradition. The converse is also true. Despite this extensive recording, many compositions have not seen the light and remain only in manuscripts and are yet to be published.

Tanjōre Quartette or Tanjai Nālvar as they are fondly called, hail from a family of rich musical heritage with their father and grandfather adorning the court of Maraṭṭa Kings. Cinnaiah (1802), Ponniah (1804), Śivānandam (1808) and Vațivēlu (1810) were born to Subbarāya Naṭṭuvanār, who was delegated to perform musical rites in Tanjāvūr Bŗhadīsvara temple. They were prodigious even at their young age and learnt the basics from their father and grandfather Gaṅgaimuttu Naṭṭuvanār.  Later they had their advanced training from Muddusvāmy Dīkṣitar for a period of 7.5 years under ‘gurukulavāsam’.

We do not have exact details regarding the period of their stay with Dīkṣitar. But it can be presumed, these events could have happened during 1810-1820. Nālvar being exceptional musicians and related to a family having a hoary tradition related to classical dance, turned their focus towards Sadir (as it was called) and created a mārgaṃ, which is still followed. They have authored innumerable kṛti-s, padam-s, varņam-s, jāvaỊi-s, rāgamālikā and tillanā-s. Their compositional style for kṛti-s considerably differs from their dance compositions. It is said Nālvar has recorded their compositions and uruppaḍi-s they have learnt from Dīkṣitar in palm-leaf and paper manuscripts.

This family has given us illustrious musician-composers like Sri K Ponniah Pillai, Veena Vidvan Sri KP Śivanandam, who belong to the sixth and seventh descendant respectively from Gaṅgaimuttu Naṭṭuvanār, through the lineage of Śivanandam (of Tanjai Nālvar). These members are not only involved in the transmission and propagation of the compositions of Nālvar, but also involved in the preservation of these manuscripts.

These manuscripts are now, in the possession of Sri Śivakumār, an eight generation descendant and a proficient Veena and Violin vidvān. It is due to the persevering effort of this family, some of the unpublished compositions of Nālvar saw the light.


Paper manuscripts

Śivakumar has, in his possession several bundles of paper and palm leaf manuscripts. Though the palm-leaf manuscripts are under good condition, paper manuscripts require immediate attention.

Of the paper manuscripts available, a segment of a manuscript replete with the kṛti-s of Tanjai Nālvar and Muddusvāmy Dīkṣitar are considered now. Though, the report cannot be considered as complete, this can definitely give us an idea about the repertoire of Nālvar.

As with any other manuscripts written before the advent of standardized notations, notational style is primitive; lacks a mark to identify sthāyi, anya svaram and ending of an individual āvartanam. Also, these notations do not indicate about second and third speed. Rāga names too was not mentioned for many kṛti-s. Savingly, svarasthāna and the parent mēla of the rāga are given clearly alongside the notations.

The available material can be divided into three segments based on the composer:

  1. Kṛti-s of Nālvar
  2. Kṛti-s of Muddusvāmy Dīkṣitar
  3. Others
  1. Kṛti-s of Nālvar

In the section analyzed, Guru-navaratnamālika kṛti-s are seen with notation. This set of 9 compositions was composed by Nālvar as a Guru stuti. This cannot be considered as a regular Guru stuti. Nālvar invoke their Lord Bŗhadīsvara and they are not paeans composed on their Guru.  Very few direct references to their Guru or his personality can be seen. These are to be compared and contrasted against the Guru kṛti-s composed by Vālājāpeṭṭai Vēṅkataramaṇa Bhāgavatar and/or other disciples of Tyāgarāja Svāmigaḷ.

Navaratnamālika of Nālvar

The following kṛti-s are held at high esteem due to the reasons mentioned above:

Māyātīta svarūpiņi – MāyāmālavagauỊa

Śrī guruguha mūrti – Bhinnaṣaḍjam

Jewel box made of Ivory gifted by Mahārājā Svāti Tirunāḷ to Vaṭivēlu Naṭṭuvanār

Sāṭilēni guruguha mūrtini – Nāța

Śrī karambu – Kanakāmbari

Sārekuni – Cāmaram

Śrī rājarājēsvari – Ramāmanōhari

Paramapāvani – VarāỊi

Sārasākși – Śailadēsākși

Nīdu pādamē – PantuvarāỊi

Two interesting observations can be made from this list. First, the rāga of the kṛti-s sāṭilēni and śrīkarambu is different from the present renditions. Now they are sung in the rāgam PūrvikaỊyāņi and Kāmbhojī respectively.  Second, all the kŗtis-s are set in the “Rāgāṅga rāgā-s” (a term equivalent to the term mēḷakarta, usually referred to the scales in the asaṃpūrṇa mēḷa system). Pantuvarāli is specifically mentioned as a rāgam with sādhāraņa gāndhāra. This is in line with the old practice of calling the present day Śubhapantuvarāli as Pantuvarāli. This was remarked by Subbarāma Dīkṣitar too in his Prathamābhyāsa Pustakamu.

We also can see other kṛti-s of Nālvar in other rāgāṅga raga-s namely bṛhadīśvara in Gānasāmavarāli and bhakta pālana in Phēnadyuti. This totals to 11 kṛti-s belonging to this category. This makes us to surmise that Nālvar could have composed in all the 72 rāgāṅga rāga-s following the footsteps of their Guru. It is emphasized again that the manuscript referred here represents only a portion of their collection and the entire corpus is to be analyzed to get a definitive conclusion.

Though, an in depth analysis of the version given in this manuscript and the other printed versions is to be done, namely “Tanjai Peruvudaiyān Perisai” and “Saṅgīta Saṃpradāya Pradarśini”, the two authentic texts which give these kṛti-s (either all or a few) in notation, preliminary analysis revealed a significant finding which is worth discussing here. The version given here for the Māyātīta svarūpiṇi is exactly the same as given in Saṃpradāya Pradarśini !! There might be subtle differences which are trivial and some allowances need to be given considering the fact we are dealing with a manuscript.

Another interesting finding is related to the kṛti, “śrī rājarājeśvari” in the rāgam Ramāmanōhari. The version given in this manuscript has phrases like PRRSNN, PNS which are not seen in both the books mentioned though the version given by Subbarāma Dīkṣitar closely follows the manuscript excluding the presence of the mentioned phrases. Though, these phrases appear to be outlandish in Ramāmanōhari, they feature in a gītam notated in Saṅgīta Saṃpradāya Pradarśini. This shows Saṅgīta Saṃpradāya Pradarśini is a veritable source to know the rāga structure of the by-gone centuries. One more noticeable feature seen in these manuscripts is the total absence of ciṭṭa svāra segment for all the kṛti-s, irrespective of the composer involved.

Three other kṛti-s found in this manuscript deserve a special mention – Sarasvati manōhari gauri, Śrī jagadīṣamanōhari and Śrī mahādēvamanohari. Rāgā-s are not marked for these compositions. The kṛti śrī mahādēvamanohari was published in the book “Tanjai Peruvudaiyān Perisai” by the descendants of Tanjai Nālvar with a slight variations in the sāhityam. Whereas their version starts as mahādēvamanohari, the manuscript adds a prefix ‘śrī’ to mahādevamanōhari. Adding ‘śrī’ satisfy the rules of prosody as anupallavi reads as ‘sōmaśekhari’. Dhātu of this composition, as given in this manuscript too give us an interesting finding. Dēvamanōhari described in the treatises belonging to 17-19 CE whose authorship is known always stress the phrase PNNS and a straight forward DNS was never accepted by them. PNNS can be seen only in the version given in these manuscripts.

Rāga of the other two kṛti-s is to be determined. Rāgam of the first kṛti can be presumed to be Gauri as Nālvar had the practice of incorporating the raga mudra in many of their sāhityam. The notation will be analyzed and updated.

Beside these kṛti-s, varṇam-s like viriboṇi and mā mohalāhiri are seen.


        2. Kṛti-s of Muddusvāmy Dīkṣitar

Around 90 compositions can be identified to be that of Dīkṣitar and all are available with notations. Out of these 90, 5 are unpublished. The remaining 85 can all be seen in Saṅgīta Saṃpradāya Pradarśini of Subbarāma Dīkṣitar. As mentioned earlier, the kṛti-s seen in this small portion of the corpus cannot be considered as the complete repertoire of Nālvar. Nevertheless, 85 denotes a significant number and it is to be borne in mind that not even a single composition seen here is outside Saṃpradāya Pradarśini. This shows any kṛti not mentioned in this text is always to be taken with a grain of salt.

A. Majority of the kṛti-s in the majority 85 belong to the clan of ”Rāgāṅga rāga-s”. Kṛti-s of Dīkṣitar can be seen in all the rāgāṅga rāgā-s except for ten. They include Toḍi (8), Bhinnaṣaḍjam (9), Māyamālavagaula (15), Varāli (39),  Śivapantuvarāli (45), Ramāmanōhari (52), Cāmaram (56), Niṣada (60), Gītapriyā (63), Caturaṅgiṇi (66), Kōsalam (71). It is to be remembered here that Saṅgīta Saṃpradāya Pradarśini too didn’t furnish the kṛti-s of Dīkṣitar in the rāga-s 9, 45, 52 and 56. Of these four, a kṛti of Ponniah (of Nālvar) was given for three rāga-s – 9, 52 and 56. Śivapantuvarāli was not awarded with any kṛti. Same pattern was followed in this manuscript too. Kṛti-s were given in order of the rāgāṅga raga. After the rāgāṅga rāga 7, we find the kṛti of Nālvar in the rāgam Bhinnașaḍjam (śrī guruguha mūrti) followed by a kṛti of Dīkṣitar viśvanātham bhajēhaṃ in the rāgāṅga rāgam Naṭābharaṇam (10). This pattern is being followed for the rest too [after Bhavānī (44), Kāśirāmakriyā (51) and Śyāmaḷā (55) we find a kṛti of Ponniah in 45, 52 and 56 followed by a kṛti of Dīkṣitar]. Blessed is Śivapantuvarāli to have a kṛti of Nālvar in this manuscript. This raises a doubt on the authenticity of the Dīkṣitar kṛti-s presently prevalent in the rāga-s 9, 45, 52 and 56.

It is to be accepted that we don’t find a kṛti of Dīkṣitar in others rāgāṅga rāga-s namely 15, 60, 63, 66, 70 and 71. Excluding 15 and 39, the rāga-s preceding and succeeding these left–outs do not occur in sequence. They occur haphazardly; perhaps they might have been written separately and those pages are lost. 15 is an exception here as it is seen in sequence succeeding Vasantabhairavī (14) and preceding Vegavāhini (16). Reason for māyātīta svarūpiṇi replacing śrīnāthādi is not clear. But, it could have been separately written and lost. We have another example to support this view – the kṛti bhajarē citta in Kaḷyāṇi (65) is found separately and not after Bhūṣāvati (64). We find only one kṛti in Kamalāmbā navāvaraṇam (śri kamalāmbikayā in Śaṅkarābharaṇam) and three in Navagraḥa series, namely divākaratanujam, bṛhaspate and sūryamūrte. Reason for not seeing any entry in 39 is an enigma.

B. It can be noticed, after the rāgāṅga raga 7, we see a kṛti of Ponniah in the rāga 9. Rāga 8, Tōḍi does not have any entry. Can we presume Kamalāmbike was the only kṛti composed by  Dīkyṣitar in Tōḍi before and/or during his stay in Tanjōre and due to some reasons  that  was not notated ? Either that was not known to Nālvar or that was composed by Dīkṣitar after his stay in Tanjōre ? Alternatively, was that notated separately and yet to be identified ? But not seeing a composition in such a major rāga is strange.

C. Regarding grouping a rāga under a mēla, this manuscript conforms with the grouping system followed by Subbarāma Dīkṣitar. Āndāḷi is given under mēḷa 28 and Sāma under 29. The only exception to this is Saurāṣtram; considered as a janya of Vēgavāhini in this manuscript. This is understandable due the presence of anya svaram in this this rāgam.

D. Four kṛti-s belonging to Guruguha vibhakti kṛti-s are seen – śri guruguha mūrtē in Udayaravicandrikā, śri guruguhasya dasōham in Pūrvi, guruguhādanyam in Balahaṃsa and guruguhāya in Sama. Bhānumati, though a rāgāṅga rāgam is represented only by the kṛti ‘bṛhadambā madambā’ and not ‘guruguha svāmini’.

E. None of the kṛti-s belonging to Tyāgarāja vibhakti group can be seen. Does it mean these kṛti-s were composed after his stay in Tanjōre ?

F. Almost all the kṛti-s addressing Bṛhannayaki or Bṛhadīśvarar, notated in Saṅgīta Saṃpradāya Pradarśini are seen here.

G. Mīnākṣi mēmudham dēhi is seen in this manuscript suggesting this kṛti must have been composed when he visited Madurai before his stay in Tanjōre.

H. Minority 5 is much more interesting. We see these compositions for the first time. They appear to be a part of Nirūpaṇam than kṛti-s. They include:

Jaya jaya gauri manōhari – 22 janyam (to be identified)

Kāmakṣi namōstute – Pāḍi

Śaranu kāmākṣi – Mēgarañjani

Manōnmaṇi bhavatutē maṅgaḷam – Mēcabauli

Śaranu śaranu mahēśa śaṅkari – Ārabhī

Of these, the first three has been mentioned by Dr Rīta Rājan in her thesis.

A reconstructed version of the Śaraṇu daru – ‘śaraṇu śaraṇu’ in the rāgam Ārabhī can heard here

I. Though, an in-depth comparison is to be done with the version given by Subbarāma Dīkṣitar, at the outset, can be confidently said not much difference exist between the two.


       3. Others

Other than the works of Dīkṣitar and Nālvar, we also find  padam-s of Kṣetrayya and some other kṛti-s of unknown authorship. Sri Śivakumar also possess another paper manuscript having around 300 gītam in notation. Examination of a sample showed that they are the replica of the gītam-s notated in Saṅgraha Chūḍāmaṇi. This could been written by some other member in the family.



This inventory is not complete and highlights only some important findings seen in a section of a major collection. It is believed these findings will be more helpful to the researchers and musicians alike to get an idea about the Dīkṣitar kṛti-s learnt by Nālvar. When these kṛti-s are compared with the versions given by Subbarāma Dīkṣitar, we can get an overall image about the melodic structure of Dīkṣitar kṛti-s in general. This might be of some help In clearing the controversies revolving around these kṛti-s. Some other points in identifying the ‘real’ Dīkṣitar kṛti too is highlighted so that these findings can be applied or recollected when we progress further and get some additional material.



I profusely thank Sri KPS Śivakumar, an eighth generation descendant belonging to the family of Nālvar and the son of Sangīta Kaḷānidhi Sri KP Śivānandam for sharing these valuable manuscripts.




Varadarāja ninnu kori and Svarabhūṣaṇi – Few insights

Out of 600 or 700 compositions of Saint Tyāgarājā available to us, a significant fraction was composed in vinta or apūrva rāgā-s. Tyāgarājā was the first to use these rāgā-s and the source of these rāgā-s remain obscure. Saint didn’t reveal the name of these rāgā-s to his disciples. Thus, they remain a source of confusion as many kṛti-s composed in these rāgā-s has multiple lakṣaṇā-s, as transmitted by different disciple lineage. Hence, it becomes essential at least, at this point of time to collect and analyze the present available evidences, to know the lakṣaṇaṃ seen in the older versions transmitted by authentic sources. In this post, we are going to discuss few issues related to a kṛti composed in one such vinta rāgāṃ. Before going to the topic proper, a few facts are provided which are helpful in studying the kṛti-s composed in these vinta rāgā-s.

Fact 1 : Generally, rāgā-s handled by this composer can be broadly divided into three categories:

  1. Rāgā-s mentioned in the earlier musical treatises and popular during his time like Nāṭa
  2. Rāgā-s not mentioned in the earlier musical treatises but popular during his time like Begaḍa.
  3. Rāgā-s seen in relatively later treatises (like Saṅgīta Sarvārtha Sāraṃ, Saṅgraha Chūḍāmaṇi etc) or created by him like Kāpi nārāyaṇi.

Fact 2 : Tyāgarājā didn’t reveal the name of these apūrva rāgā-s to his disciples. This is an important fact as the name that we hear today or see today in various texts were named either by his disciples or by musicians of the gone century. 1

Fact 3 : When the composer himself has not revealed the name of these rāgā-s , it is illogical to say that Tyāgarājā has composed in the rāgā-s seen in the treatise Saṅgraha Chūḍāmaṇi of Gōvinda. This point will be emphasized in future posts too.

Fact 4 : The main difference between the earlier musical treatises (treatises composed till Sangīta Sārāmṛtā, dated approximately to 1735, like Sangīta Sudhā , Catuṛdanḍi Prakāśika etc) and the later ones (like Saṅgīta Sarvārtha Sāraṃ (SSS), Saṅgraha Chūḍāmaṇi (SC) etc) lies in the way in which a particular rāgā was handled. Whereas in the former treatises, each rāgā was explained by the phrases they take, latter treatises explain by giving a scale – ārohaṇa and avarōhaṇa. In some, we find a lakśaṇa gītaṃ. Hence, a rāgaṃ is visualized as an synthetic entity which strictly obeys its scale by the proponents of the later treatises; whereas the proponents of the earlier treatises view these rāgā-s as an organic structure which cannot be explained by a scale always.

Fact 5 : Rāgā-s that we come to know by SSS and/or SC is not a complete list; they are just a sample. We have got many manuscripts preserved carefully in various libraries waiting to confuse us. The point that this author tries to establish by quoting this point is, a rāgā can have multiple scales, depending on the author who writes the treatise. A rāgā which is placed under a particular mēḷā could have been placed under a different mēḷā by a different author. Also, a rāgā with a similar set of svarā-s could have been called by a different name by various authors.

Fact 6 : Unless, we see the notation, it is not advisable to get carried away by the rāgā name alone (see Fact 5).

With this basic understanding, we shall move to the post “Varadaraja ninnu kori”.

This is a relatively rare kṛti composed on the Lord Varadarājā of Kāñcipuram. This is believed to have been composed by the Saint during his sojourn to holy places like Kāñcipuram, Tirupati etc. Much about this composition has been mentioned in another relevant article in this site. This article will focus on the history of this rāgaṃ with a special emphasis on Vālājāpet notations.

Svarabhūṣaṇi in treatises and texts

Svarabhūṣaṇi belongs to the third category in the classification mentioned above. Strangely, it is not mentioned in SSS or SC. Hence, it must be in some treatise which is yet to be discovered or it can be a creation of the Saint itself.

It is one kṛti of the Saint which is not frequently seen in the texts published in the last century. First text to link the rāgaṃ with this kṛti is “Oriental Music in European Notation”, published by Sri AM Chinnasvāmy Mudaliyār (AMC) in 18932 (see figure 1) . He tried to collect and record the authentic versions and kṛti-s of Tyāgarājā and hence approached one of his direct disciple, Vālājāpet Kṛṣṇasvāmy Bhāgavatar (VKB). His versions were cross checked with other disciples of the Saint and what we see today is the version approved by more than musician excluding VKB. Though, this kṛti is not notated here, we clearly see for the first time, the rāgā for this kṛti is mentioned as Svarabhūsaṇi, a janya of mēḷā 22. Later this rāgaṃ placed under mēḷa 22 can be seen in various texts including Nathamuni Panḍitar’s Saṅgīta Svara Prastāra Sāgaraṃ published in 1914.

It is to be pointed here we are really clueless on who named this rāgaṃ as it is not seen in any treatises that are presently available to us. But, it can be safely said that the rāgaṃ of this composition is a janyam of mēḷā 22 and is much different from its allied rāga Dēvamanōhari. The musicians who worked with AMC and AMC were well aware of Dēvamanōhari. Listing of few kṛti-s of the Saint under Dēvamanōhari and notating a composition of Gōpāla Krṣṇa Bhārati in Dēvamanōhari in the same book proves the same.

From what we have seen till now, it can be summarized Tyāgarājā has not revealed the name of any of the apūrva rāgā used by him. Some unknown musician has named it as Svarabhūṣani. AMC, who was in search of the authentic compositions and versions of the Saint, accepted this as such.


Fig1 : This shows the index of kritis published in Oriental Music in European Notation (1893) by AM Chinnasamy Mudaliyar. Varadaraja ninnu kori can be seen here with the ragam mentioned as Svarabhushani, a janya of mela 22.

Svarabhūṣaṇi and Varadarāja ninnu kori in manuscripts

Though, efforts have been made from late 1800s to record our music in the form of printed texts, several material remain unknown in manuscripts and they exist as a private collection. A study of these manuscripts is a must as they give a broader picture of the issue in hand.

It is quite rare to find this kṛti in manuscripts too. This shows that this kṛti was not learnt by many disciples and this should have been in the repertoire of only very few. Vālājāpet Vēṅkaṭaramaṇa Bhāgavatar was one amongst them to learn this directly from the Saint.

Let us now see few manuscripts which make a mention about this kṛti.

Manuscript 1

Dr V Rāghavan, in a paper published in the Journal of Music Academy mentioned about the discrepancies in allotting a particular rāgā name to a particular kṛti (of Tyāgarājā). He has presented a paper based on a palm leaf manuscript which he had in his possession. This kṛti find its presence there and the rāgā of this kṛti is mentioned as Śāradhābharaṇaṃ, a janya of mēḷa 34, Vāgadhīṣvari. We are totally unaware of the musical structure as notation was not provided in the paper. 3

Manuscript 2

A manuscript by one Bālasubraḥmaṇya Ayyar, written in the year 1922 says the rāgaṃ of this kṛti as Svarabhūṣaṇi. Notation is provided.

 Manuscript 3

A granta manuscript in the collection of Late, Srivanchiyam Sri Ramachandra Ayyar says the rāgaṃ of this kṛti as Śāradhābharaṇaṃ. Again, notation is not provided.

Manuscript 4

A manuscript written by Vīṇa Kuppaier mentions this kṛti. Unfortunately, rāgā name was not mentioned and notation too was not provided.

Manuscript 5

Vālājāpet notations mention as Svarabhūṣani.

From the study of manuscripts, it becomes clear that there was confusion in the rāgā of this kṛti. Two different sources saying the rāgā as Śāradhābharaṇaṃ is an issue to ponder. Also, two different sources ascribing this kṛti to Svarabhūṣaṇi also validates the musical structure, where in the rāgā takes the svarā-s of mēḷa 22. Unless, we get a manuscript or text which gives the version in Śāradhābharaṇaṃ, we cannot come to a conclusion that Śāradhābharaṇaṃ and Svarabhūṣaṇi are two different versions (See fact 4).


Svarabhūṣaṇi – its scale

To the best knowledge of this author, Saṅgīta Candrikai of Māṇikka Mudaliyār, published in the year 1902 is the first printed text to mention the scale of this rāgaṃ as SGMPDNS  SNDPMRS, placing it under the mēḷa 22. The two manuscripts mentioned above (manuscript 2 and 5) give the same scale. Vālājāpet notations give additional information that this takes the notes of Kharaharapriya.

Earlier texts and manuscripts are uniform in their opinion that this is a janyaṃ of Kharaharapriya and the scale can be taken as SGMPDNS   SNDPMRS.

Varadarāja ninnu kōri – Vālājāpet version

Vālājāpet manuscripts form an important source to understand the kṛti-s of Saint Tyāgarājā. These manuscripts were written by Vālājāpet Vēṅkaṭaramaṇa Bhāgavatar (VVB) and his son Vālājāpet Kṛṣṇasvāmy Bhāgavatar. It is even said Tyāgarājā could have seen this as they were recorded during his life time.4 These notations were preserved at Madurai Sourāṣtra Sabha and the transcripts are available in GOML, Chennai. Few of these transcripts can be accessed online here. These transcripts are the main source for this post.

In the absence of first hand records made by Tyāgarājā, these notations form a very valuable and authentic source to understand the version learnt by his prime disciple Vēṅkaṭaramaṇa Bhāgavatar.

In the notations, it is mentioned as Svarabhūṣaṇi with the scale SGMPDNS SNDPMRS. This scale is much adhered to in the version given.

Pallavi starts from dhaivataṃ, reaches madhya ṣaḍjaṃ and goes to gāndhāraṃ as DPMRSGMP. This clearly shows the rāga lakshaṇaṃ without any ambiguity. Anupallavi again starts from dhaivataṃ, but here proceed upwards and reaches tāra ṣaḍjaṃ. From here again reaches tāra gāndhāraṃ. The intelligent use of dhaivataṃ as a graha svaram and careful emphasis on the scale gives a melodic structure much different from Dēvamanōhari. Nowhere we find the phrase NDNS in this version. It is only DNS.

Caraṇaṃ has something interesting to say. It has got an additional line “maruḍu śiggu chē manḍarāḍaṭa”.

This is not seen in any of the versions recorded – either oral or textual. Interestingly, this additional line is seen in the manuscripts of Vīṇa Kuppaier!! Knowing the association between VVB and Vīṇa Kuppaier, this line adds authenticity to this version.

But, in the manuscripts of Vīṇa Kuppaier, there is a slight change in the sāhityaṃ. It reads as “maruḍu  śiggu  chē    munḍararāḍaṭa”.    This was the correction mentioned by Ravi too (See another article on this topic in this site).

Errors like this where there is a replacement of one syllable to another is much common in manuscripts. They are not the printed texts which are proof-read several times before publication (even they are prone to errors!!) What we see now, the transcripts are the genuine duplicates of the manuscripts preserved at Madurai Sabhā. The scribe, when trying to duplicate the contents from manuscripts could have made this error involuntarily. In this case, except that syllable, absolute concordance is seen between the two manuscripts under consideration. An unbiased researcher who is accustomed in reading the manuscripts will never judge the authenticity of the composition or the source which gives this composition based on the errors of this magnitude.


Let us now see the importance of this additional line. Caraṇaṃ with the additional line is represented below:

varagiri vaikuṇṭha maṭa      varṇiṃpa taramukāḍaṭa

maruḍu śiggu chē  man      ḍarāḍaṭa – nir       (munḍararāḍaṭa)

-jarulanu tārakamulalō        candrudai merayuḍu vaṭa

vara tyāgarāja nuta             garuḍa sēva jūḍa srī


‘Ra’ is used as dvitīyākśara prāsaṃ in this caraṇaṃ. When it is sung in rūpaka tāḷaṃ (catusra rūpakaṃ), each tāḷa cycle ends with maṭa, dhaṭa, man, nir, mulalō, vaṭa, nuta and juḍa. Hence each āvartanaṃ starts with a word which has ‘ra´ as its second syllable. Totally, we get 8 tāḷa āvartanaṃ only due to the presence of this additional line. In the commonly heard versions, if sung in rūpakaṃ, runs only for 6 āvartanaṃ!! Also, ‘nir’ is pushed to previous āvartanam to be in accordance with the rules of prosody.

Hence, this line must have been an integral part of this kṛti known only to the disciples learnt directly from the composer and singing without this line is an aberration.


Here is the link to Vālājāpet version of this kṛti.


A note on the version by Sri Bālasubraḥmaṇya Ayyar

No detail can be collected about this musician. The version given by him is much in line with the version that we hear starting in tāra saḍjaṃ, though differences exist. A ciṭṭa svara passage is too seen. Additional line seen in the two manuscripts mentioned above is missing. This version too does not sound like Dēvamanōhari. Needless to say, the version given here is much different from that of Vālājāpet version.



The following are “take-home” messages from this post:

Our music is transmitted very well through both textual and oral tradition. In the absence of one, the other is to be taken into consideration. A wise researcher will never neglect an evidence gained through one source when the other one is unaware of the same. Oral renditions and the available texts are only samples to show what was sung in he past. Voice of many musicians were not recorded and the knowledge of many researchers remain unpublished. If we get an additional evidence from unpublished source, that should be analysed and digested. This an only be considered as a true research. In this case, Valajapet versions were in the dark for many years.  When the notations adhere well to the scale, it should be accepted as  an old version. This will be explained more in further posts too.

“Varadarāja ninnu kōri” was composed in a rāgaṃ which takes the svarā-s of mēḷa 22. (till we get an evidence from other authentic source saying it as Śāradhābharaṇaṃ or something else).

It is better to call this rāgaṃ as Svarabhūṣaṇi as it is the name seen in one of the earlier texts published (as gleaned from the available evidence) and no other rāgaṃ exist with that name.

We don’t have any textual tradition to call it as Dēvamanōhari. Even oral traditions call it as Svarabhūṣaṇi, though versions differ. Older version like Vālājāpet notations gives us the real lakṣaṇaṃ of a rāgaṃ like this. Svarabhūṣani had a distinct melody which can be best experienced by listening to Vālājāpet version.

The additional line, seen in Vālājāpet version and manuscript of Vīṇa Kuppaier is integral to this composition. That line is to be included to make this kṛti a complete one.

Vālājāpet notations help us to know about the authentic versions learnt by VVB, directly from the Saint and solve many issues pertaining to the rāga lakṣaṇaṃ of vinta rāgā-s like this.

This example also highlights the importance of collecting and analyzing unpublished manuscripts to understand the rāgā-s handled by the Saint.



I like to thank Sri V Sriram, Secretary, Music Academy for allowing me to peruse the manuscript of Sri Balasubrahmanya Ayyar preserved at Music Academy library.

I thank Srivanchiyam Sri Chandrasekar, son of Srivanchiyam Sri Ramachandra Ayyar for sharing the rare manuscripts collected and preserved by his father.

I thank Sri Ravi Rajagopal for taking efforts to correct the error in sāhityam seen in the additional line .



  1. Subbarāma Dīkṣitulu. Prathamābhyāsa Pustakamu, Pg 129. Vidyā Vilāsini Press, Eṭṭayapuraṃ Subbarāma Samasthānaṃ, 1905.
  2. Chinnasvāmy Mudaliyār. Oriental Music in European Notation. Ave Maria Press, Madras,1893.
  3. Raghavan V. Two manuscript of Tyagaraja Songs. Journal of Music Academy. 1947: Pg 142.
  4. Sāmbamurti P. The Walajapet manuscripts. Journal of Music Academy. 1947: Pg 114-129.


A note on the sāhitya of Vātāpigaṇapatim

Before 1900, most kīrtanams of Śrī Muttusvāmi Dīkṣitar were well-known only within the small core group of śiṣyaparamparā. A few kīrtanams were published in early musical sources such as the works of the Taccuru brothers. It wasn’t until the publication of the Saṅgīta-sampradāya-pradarśini in 1904 that many kīrtanams saw the light of the day. Vātāpi-gaṇapatim in the rāga Hamsadhvani was one kīrtanam which has enjoyed a long history in the performance platform and is seen in early music publications. Accounts of Mahāvaidyanātha Iyer’s embellished and improvised version of Vātāpigaṇapatim with many saṅgatis are well known. Saṅgīta-sarvārthasāra-saṅgrahamu of Vīṇa Rāmānujayya (1857) and the Gāyakapārijatamu of the Taccuru brothers (1877) provide the sāhitya of this kīrtanam.

Recently, an issue with the charaṇa-sāhitya of this kīrtanam was brought to our attention by a samskṛta scholar, Vidvan Brahmaśri Dr.V.Shriramana Sharma.

पुरा-कुम्भ-सम्भव-मुनि-वर-प्रपूजितं त्रिकोण-मध्य-गतम् |
मुरारि-प्रमुखाद्युपासितं मूलाधार-क्षेत्र-स्थितम् |
परादि-चत्वारि-वागात्मकं प्रणव-स्वरूप-वक्र-तुण्डम् |

Here the issue is with the samāsa (compound word) in the phrase “परादि-चत्वारि-वागात्मकम्” (One who is the true import / nature of four-fold speech beginning with Parā – the other three being paśyantī, madhyamā and vaikharī) The correct expression should be “परादि-चतुर्-वागात्मकम्”. The form चत्वारि is a declined form in prathamā vibhakti and in napumsakaliṅga. In a samāsa, only the prātipadika(base) appears as there is lopa (elision) of the सुप् pratyaya that are affixed to the base. Therefore the base form of “चतुर्” is what would occur in this samāsa and not a declined form such as चत्वारि.

In this case, there are two possibilities. One is that there was an error in transmission and the second is that this was how it was composed by the composer. If we examine the first possibility, all existing published versions of the sāhitya and extant pāṭhāntara-s uniformly use the form परादि-चत्वारि-वागात्मकम्. In any case, Śrī Subbarāma Dīkṣitar himself has admitted to Pt.V.N.Bhatkande, when the latter visited him in Eṭṭayapuram, that he was not formally trained in Samskṛta (but knew enough prayoga or usage to compose kīrtanams). This is also corroborated by the fact there are some visargasandhi errors in the 1904 edition in the sāhitya of Bṛhadiśvaro rakṣatu in rāga gānasāmavarāli and the gauḷa kīrtanam Śrīmahāgaṇapatiravatu mām. However the trouble with asserting this viewpoint is that there is a second case of the identical expression used in the Aṭhānarāga kīrtanam on Bṛhaspati, “Bṛhaspate tārāpate. (incidentally this has also appeared in Gāyakalocanam, a 1902 publication).


Considering the second possibility, another Samskṛta scholar Smt.Dr. Sowmya Krishnapur adds that the ‘G r’ svara corresponding to “tvā ri” indicates the intended usage of a svarākṣara here. This svarākṣara occurs in Bṛhaspate as well. Therefore, the possibility that this is an error in transmission could be discounted.  Further, in this case, another explanation is possible to justify the composer’s usage. It could be that the composer had the famous śrutivākya “चत्वारि वाक्परिमिता पदानि तानि विदुर्ब्राह्मणा ये मनीषिणः |” in mind along with the usage in Gaṇapatyatharvaśirṣa where Gaṇeśa’s tattvasvarūpa is expressed as “त्वं चत्वारि वाक्पदानि” and used the expression “चत्वारि”, despite it being in a samāsa. According to Sri.Shriramana Sharma, such a usage can be considered as an ‘anukaraṇa’ wherein the quoted word is not to be analysed grammatically with its prakṛti-pratyaya-vibhaga and ‘artharūpa’ but in terms of ‘śabdarūpa’. The “word-form as-is” that occurs elsewhere is used for the express purpose of highlighting or quoting either for repetition or recall something that is well-attested or mentioned elsewhere. Thus, the initial “परादि” specified for the purpose of clarifying what the four-fold speech is and the words “चत्वारिवाक्” brought in as-is from śrutivākya. In other words, परादि “चत्वारिवाक्” आत्मकम् | In support of this, another example is cited in śāstra. One is the Pāṇini’s sūtra, “प्राग्रीश्वरान्निपाताः” that defines the term nipāta, explained as रीश्वरात् प्राक् निपाता: and discussed by commentary writers. Mahābhāṣyakāra asks why “रीश्वराद्” is used instead of “ईश्वराद्” and the explanation is given that “रीश्वराद्” is used so that “वीश्वराद्” (when the sūtras are read as samhitāpātha) शकि णमुल्कमुलावीश्वरे तोसुन्कसुनौ” (शकिणमुल्कमुलौ + ईश्वरे तो सुन्कसुनौ) which also matches the śabdarūpa specified does not get included here but only that used in another sūtra अधिरीश्वरे.  Shri. Shriramana Sharma also notes in passing that वारणास्यं is often rendered as वारणाश्यं but there is no known usage of आश्यं in the meaning of “face” to give the compound meaning “the elephant faced one”. आश्य is only known in the meaning of “eatable” and hence this could be avoided.

Based on the above, we requested Prof.S.R.Janakiraman to attempt to render the composition with expression “परादि-चतुर्-वागात्मकम्” to understand how this would impact the musical flow of the composition. Given that there is a valid explanation of such a usage, the usage of परादि-चत्वारि-वागात्मकम् can be left as such. With these, we leave the reader to ponder on the issue.