Tamils - a Trans State Nation..

"To us all towns are one, all men our kin.
Life's good comes not from others' gift, nor ill
Man's pains and pains' relief are from within.
Thus have we seen in visions of the wise !."
Tamil Poem in Purananuru, circa 500 B.C 

Home Whats New  Trans State Nation  One World Unfolding Consciousness Comments Search

Home > Tamil Culture - the Heart of Tamil National Consciousness > Bharata Natyam - Classical Dance of the Ancient Tamils  > The Role of Dance Sculptures in Tamilnad - Padma Subrahmanyam

Bharatha Natyam -
Classical Dance of the Ancient Tamils

The Role of Dance Sculptures in Tamilnad

Padma Subrahmanyam
paper presented at Second International Conference Seminar of Tamil Studies
Chennai, Tamil Nadu  1968

Introduction Dance Sculptures Karana Tanjore Sculptures Kumbakonam Sculptures Chidambaram Sculptures Comparison Dance and Acrobatics Parallels in Tamil and Sanskrit Works Mandalas Evolution of Karanas Conclusion Notes Appendix


Padma Subrahmanyam

The twin arts of dance and sculpture developed together in close spiritual association with the concept of the Divine Being himself as a dancer in Tamilnad. The fusion of these two arts dates back to the dawn of our civilisation. The figure of a dancer unearthed in Mohenja-daro of proto-historic India explains the genetic relationship between the various dance styles of India. The pose of this icon is still found in the varied dances of our country.* Some dances have been mentioned as purely indigenous in Tolkappiyam.(1)We infer from the description that they were rather rustic and had not attained high development or codification.

The first well lighted epoch in the history of the Tamil land is that reflected in the literature of Sangam (the first 3 or 4 centuries A.D.).(2) In this age, the Panar and Viraliyar were said to have been roving bands of musicians and dancers, who preserved the folk songs and dances of an earlier age. Their dance seemed to have also included certain hand gestures as mentioned in Bharata�s Natya Sastra. Thus a conscious attempt to synthesise the indigenous pre-Aryan modes with those of the north resulted in the development of the dance art, as seen in the later work Cilappatikaram.

Dance Sculptures

The dance sculptures not only reveal the origin and evolution of the art in Tamilnad, but also show that the Tamils were free from all linguistic inhibitions in their endless quest for knowledge. They derived inspiration not only from Tamil but also from Sanskrit sources for the development of their culture. Treatises on all arts written in Sanskrit were in the normal course absorbed and preserved in Tamilnad. Translations and abridged editions arose in Tamil. 

The earliest extant literature on dramaturgy is said to be the Natya Sastra by sage Bharata. (3) All the works that came to be written on dance in the post Bharata period had the influence of the Natya Sastra.  Cilappatikaram is no exception to this.(4) The very titles, Bharatasenapatiyam (5) and Panca�bharatiyam (6) mentioned as Tamil works in the commentary of Adiyarkunallar prove the recognition that Bharata enjoyed in Tamil country. All these works deal with the theory of dance; the practical aspect is seen in the sculpture of Tamilnad. 

The association of the various Gods with dance made it necessary for the sculptor to study the Natyasastra before depicting these deities in stone. This knowledge was one of the main factors that contributed to the refinement of sculpture. The presence of an accomplished nartaki - the dancer - attached to the temple induced the sculpture to create dance sculptures. In turn, such sculptures remain as everlasting guides for successive generations of dance enthusiasts. They served to codify and preserve the art for all time. Among such closely inter-related creations, benefiting each other, the most important is the karanam in the field of dance and sculpture. It is a matter of pride for Tamilnad that it has been able to preserve in pristine purity the Kashmiriyan sage, Bharata�s style of dance in the form of sculpture. Though there are dance sculptures all over India, such close adherence to the Bharata tradition cannot be seen anywhere else.


Karanam is a technical term, derived from its Sanskrit route, kr - meaning �to do�. In short, it is a unit of dance which was the basis for concert items in ancient times. The karanm is generally mistaken to be a static pose. As it is a combination of the three elements, namely cari (movement for the legs), nrtta hastan (gesture for hands) and stanam (posture for the body), it is a full movement and not a static concept. Thus a karanam can be compared with the adavu of contemporary dance. Just as many adavus make a tirmanam and many tirmanams an item, according to the number of karanas specified, they were called kalapaka, matrka, bhandaka, sa�ghataka and angahdra. 

To compare it with the components of language, the elements of karanas are alphabets, the karanas are words and the rest are phrases and sentences.

Bharata�s Natya Sasstra is the earliest extant literature giving details about these karanas. Bharata has described 108 karanas in his fourth chapter. In Adiyarkunallar�s commentary on Cilappatikdram a reference to karanam is found. While describing the requisite qualities of the dance master, llango says 

" he is supposed to know the rules pertaining to the two types of dances". (7

These types are explained as santi kuttu and vin�da kuttu by Adiyarkunallar. Santi k�ttu is of four types, namely sokkam, mey k�ttu, avinaya kuttu and natakam. The explanation for sokkam is given as ' it is made up of 108 karanas'. It is also called Suddha nrittam or abstract dance. Thus the 108 karanas were commonly in practice in Tamilnad.

Tanjore Sculptures

Apart from the literary evidence for the popularity of Bharata�s karanams, the dance sculptures in the temples of Tamilnad prove beyond doubt that the Tamils took great pains in preserving Bharata�s style.

Just as the earliest extant literature on karanas is the Natya Sastra, the earliest extant visual representation of these are found in the Brhadiswara temple at Tanjore. The credit of identifying them as Bharata�s karanas goes to Padmabhushan Dr. T. N. Ramachandran, the eminent archeologist. When the Chola king Rajaraja built the Tanjore temple in the beginning of the 11th century, dance art enjoyed such a high status in society that he had the karana figures chiselled as sculptures in the first tier of the Vimana.

Natya Sastra was already about a 1000 years old during Rajaraja�s time. It is quite possible that the karanas were developing on new lines, some of them even becoming obsolete; it was his genius that gave immortality to Bharata�s karanas by such sculptural codification. Thus we owe a deep debt of gratitude to the Chola Emperor for this service in the cause of dance art.

The karana figures in Tanjore are about two feet in height and are found one after the other in a serial order as prescribed by Bharata. Starting from talapuspaputam, there are only 81 figures found. Slabs for the rest are found left incomplete. But it is beyond doubt that each sculpture has been carved after a deep understanding of the description of the relevant karanas as found in Natya Sastra as well as its commentary the Abhinavabharati, written by Abhinavagupta.

Kumbakonam Sculptures

Chronologically speaking, next to the Tanjore representation, the karanas are found in the S�rangapani temple at Kumbakonam. These belong to a century and a half later than those at Tanjore. Here, though all the 108 were carved, they are not in Bharata�s serial order, as we see them to-day. But the most interesting feature is that under each figure, the name of the respective karanam has been inscribed in Tamil Grantha script.

Chidambaram Sculptures

The Nataraja temple at Chidambaram marks the next phase in such sculptural codification. The four gopurams were built during the course of three centuries, 13th to 16th. All the 108 figures are beautifully carved in the entrance of these gopurams. The eastern and western gopurams arc particularly important as they have the inscriptions in Tamil Grantha script which are transliterations of Bharata�s text pertaining to each karanam. This was the first and perhaps the last time too that the karanas were carved with their s�tras inscribed in full. It is evident that these were not mere architectural embellish�ments, but they were to guide the dance enthusiasts with regard to Bharata�s work. It is really amazing that the work of the scholarly sage of Kashmir had been transplanted in Tamil soil so effectively without any linguistic, political or geographic barrier. This clearly shows the spirit of assimilation of the Tamils to imbibe and foster all that is best.


A comparison of the karana figures in the three temples is an interesting study. In Tanjore, there is no inscriptional explanation; in Kumbakonam, we see the respective names of karanas below each figure. The high water mark is reached in Chidambaram, where the whole s�tras of Bharata have been inscribed with the corresponding figures. The Tanjore figures depict Siva with four arms; in Kumbakonam the figures are those of Sutraddra as can be inferred from the label inscribed therein. The sculptures here are said to be older than the present gopuram wherein they might have been fixed subsequently. In Chidambaram, we find female figures in the form of ardhasirpas, strictly following the serial order of Bharata�s work.

The four arms found in the Tanjore figures are often used for symmetry; it is also wonderful to follow the way in which the animation of the movement is shown through them. The first two arms show the beginning while the other two, its course or end. *

Dance and Acrobatics

There are slight variations in the interpretation of karanas in the three temples, but it is clear that the 108 karanas are the same for both the sexes. There are no tandava and lasya karanas independently. Some are certainly more graceful than a few acrobatic ones. But even these gymnastic movements seem to be common for men and women. The karanas such as sakatasyam, cakramandalam and gangdvataranam which are highly acrobatic in nature are meant as exercise for the body to keep the dancer trim. It might also be due to the merging of acrobatics and the art of dance which has been natural during the course of history of dance in any part of the world. For example, Curt Sachs, in his World History of Dance says, 

�It is significant that in the old Egyptian, the same word �hbj� is used to designate the ordinary dance and also the gymnastic exercise known as the �bridge� which is frequently depicted on the monuments of the middle and new kingdom.� (9)

 The figure described here is the very same as the karanam, cakramandalam as depicted in Chidambaram and Kumbakonam.

Parallels in Tamil and Sanskrit Works

Coming to the practical aspect of the karanas, it is interesting to note certain parallels in Tamil and Sanskrit works. Natyasastra describes stanam (a static posture for body) as a constituent element of karana. The Tamil works describe the same as nilai. Apart from Bharatasendpatiyam and Pancabharatiyam referred to in Adiyarkunallar�s commentary on Cilappatikaram, Dr. U. V. Swaminatha Iyer often follows the Suddhananda Prakasam which is an early Tamil work on dance. Wherever the complete quotations are not available, this work has been utilised to throw more light on the subject. While describing the rules for the various kuttus, Adiyarkunallar quotes the following from an early work:

"Aruvakai nilaiyum aivakai patamum
irenvakaiya vankak kiriyaiyum
varuttanai nankum niruttakkai muppatum
attaku tolila vaku menpa�

The six nilai are the same as the six stana of Bharata�s work. Suddhanandaprakasam also gives the same names; they are: samam,* vaishnavam,* vaisakam,*  mandalam,* alidam,*and pratyadilamam.*

The five padam or feet variations given in the Tamil work are: samam,* udghatitam,* kancitam,* kuncitam * and sancaram.* The terms, kancitam and sancaram must be the same as ancitam and agratala saitcaram respectively of Bharata�s work.

Of the sixteen angakriyas mentioned in the Tamil and Sanskrit works, some are as follows:

The normal posture of feet is samakali in Tamil and samam in Sanskrit.* Moving the feet in sama posture without lifting is sarikai in Tamil and sarika in Sanskrit;*

Stamping with the heel, toe or sole is kurttanam in Tamil and kuttanam in Sanskrit.*

Crossing the feet is suvattikam in Tamil and svastikam in Sanskrit;*

Encircling with one leg, the other foot which is steady is vettanam in Tamil and vestanam in Sanskrit.

The four varttanai mentioned in the Tamill works are the same as the four hasta karanas of Sanskrit; they are: apavestitam,* upavestitam,* vyavartitam *and parivartitam.*

The 30 nrtta hastas are all exactly the same as in the Natyasastra.

Adiyarkunallar explains mey k�ttu as concerning love themes; they are also called ahamargam are of three kinds, viz., desii, vadagu and si�galam. The same term Ahamarga (11) is seen in the Sanskrit work. Bharatarnava of Nandikesvara.


While describing the dance that was performed by Madavi during her Arangetram - maiden performance - Ilanko says:

�Kottiran tutaiyat�r mantila makak
kattiya mantilam patinonru p�kki�.(12)

Here a reference to mandalam is found. This has been explained by Adiyarkunallar as:

�Pancatala pirapantamakak kattappatta teciyottai oru talattirku irantu parraka pattum tirvu onrumaka patinonru parrale teci kuttai ati mutittu enka.�

The pancatala prabandam is a musical form which consisted of five talas. Just as we have ragamalikas in our contemporary music, there were talamalika compositions in ancient times, which seem to have been used for dance. Madavi�s d�si dance appears to have consisted of eleven parts, including the final tirmanam ( tirvu) which may be compared with the �coda� of western music. For each of the five talas, and danced two parru or groups of movements which may be compared to the nandi of our contemporary dance. It is also said that the dance was in the form of mandalas. Again there is a reference to mw.z4aIam in the same chapter; Ilanko says, 

"..atanai aitu mantilattal" (14)

After describing the musical form, he says that it was danced with the help of mandalas The term aitu is explained as 'beautiful�, thus being used as an adjective to the mandala. But my humble submission is that it is quite likely that the word aitu means the numeral five, thus indicating that the piece was performed in the form of five mandalas.

The term mandalam has been elaborately dealt with in the Natya Sastra. Though there is a mandala stsna or posture mentioned, this term is also explained in a different context. A mandalam is a combination of a group of caris or leg movements. Hence mandalam is a more complex concept than even the karanam. While the karanam is made up of a cari, nrtta hastam and stanam, the mandalam is made up of many caris. Unless the caris and karanams are all perfectly understood, the mandalas cannot be deciphered. Hence, to reconstruct or even to get a vague idea of what the dance of Tamilnad looked like during the age of Cilappatikaram, a thorough knowledge of the karanas has got to be our primary goal. A study of the karanas by the understanding of the literature concerned has got to be guided by the study of dance sculptures.

Evolution of Karanas

The 108 karanas which had formed the foundation of the dance of Tamilnad in ancient times underwent many changes as centuries passed. The culmination of the process of assimilation of Bharata�s work together with indigenous development is seen in Cilappatikaram. While describing Madevi�s dance, Ilanko says,

�Nattiya nannull nankukataip pitittu". (15)

It is most likely that by Nattiya nannull he means only the Natyasastra of Bharata which Matavi is said to have strictly followed. Coming to the Pallava period, we find a sculpture of Siva in catusra posture of Bharata in the Dharmaraja Rata in Maliabalipuram; closely following this, there is adherence to Bharata�s work in Kailasanatha temple at Kanchi. Later even the Saivo Agamas which dealt with the 64 tandavas of Siva had the influence of Bharata�s karanas; for example, the ananda tandava figure of Nataraja is known as bujanagatrdsitam in the Agamas and the Natiyasastra. 

We have already seen the progress of sculptural codification of Bharata�s karanas from the 11th century. Any art, which is dynamic, develops, changes and grows on new lines with the passage of time; in the 13th century, the desi style reached new heights by the merging of folk arts and creative genius of the local artistes. Even this development was the subject of literary codification; in his Sangita ratnakara, Sarangadeva describes 36 desi karanas in the name of utpluti karanas, i.e., karanas with jumps.

Kudumiamalai inscriptions mention the terms, tiru, pattataivu, mey kattataivu, tiruvalatti as dances. The term adaivu literally means combination. The Tamil terminology for karana must have been evolved as adaivu. Gradually the term and concept of karana as propounded by Bharata were fading. During the time of the Nayak kings, innumerable dance sculptures were added to our temples; that often they seem to be repetitions of the same poses. From this we infer that they were meant to be only architectural ornamentation and not meticulous interpretation of theory. During the 18th century, the Maratha king Tulaja in his Sangita saramrta, recorded the adavus of his time; by this time the ataivu had by usage become adavu. 

Tulaja�s work may be termed as the basic text for what is to-day known as Tanjore style of dance. The final phase was reached at the time of the famous Nattuvanats, the Tanjore quartette; by then Bharata�s karanas had almost become a dead concept for all practical purposes. It is a sad fact that even some of the adavus which were prevalent as late as the 18th and 19th centuries are lost to-day. Due to various political, economic and sociological causes, we have lost the capacity to retain with us some of our valuable artistic traditions. But there is no doubt that our art has the invincible capacity to survive. The striking evidence for this is the fact that Bharata�s karanas are still found scattered all over India, the Far East and even in the West.


The explanation of the origin and evolution of dance based on mythology cannot draw a clear line between fiction and fact. Only a rational approach can satisfy those interested in art history. For others who believe in �art for art�s sake� the technique of theory and practice is the main attraction. Thus for all those who are interested in the spiritual, emotional, intellectual, esthetic, historical or any other aspects of our dance art, the sculptures are the primary source. They depict what was really in vogue and are not mere fantasies. By studying them, we will not only regain Bharata�s art in the real sense, but also have the profit of the 2000 years of evolution this art has undergone.

Proper preservation of the dance sculptures in the country is an immediate necessity. Spoiling them by white-washing and covering them by careless constructions should be ruthlessly prohibited. An extensive survey of all the dance sculptures of our country and the Far East will reveal valuable facts. Evidently individuals cannot afford to do this. Educational foundations and universities should come for�ward to undertake such gigantic projects. It is high time that our universities had faculties for dance, giving the art its due place in the academic world. In this connection, I would like to express my grati�tude to the Annamalai University for having given me an opportunity to carry on my humble research.


1. Tolkappiyam ascribed to c. 300 A.D.: K. A. Nilakanta Sastri, History of South India, 2nd ed., Oxford University Press, 1958. For a description of dances, see V. R. Ramachandra Dikshitar, Studies In Tamil Literature and History, Luzac & Co., London, 1930, pp. 191-2: �Vallikkuttu....in honour of Valli...popular among lower classes of society. Kalanilaik�ttu was of high order.. .arranged in honour of a young soldier... .His friends presented his Virakkalal an anklet and indulged in a dance... Purattinaiyial refers to Velan Veriyadal. The chief feature of this dance was to offer bali or animal sacrifice.., in the course of such worship one got possessed with the spirit of God and began to dance... .There was another kind of adal... .It was a custom then that when once the King who led the host fell in battle, to whatever side he might belong, other kings stopped the fight, surrounded the dead body and honoured it by a kind of dancing in which skilled display of swords was a feature.�

2. K. A. NILAKANTA SASTRI, op. cit., p. 100

3. See MANM0HAN Ghosh, The Natyasastra (translation), 2nd ed., Manisha Granthalaya, 1967, Introduction, pp. lxiv-Ixv: �Apart from the fact that Bhasa once mentions the Natyasastra, there is plenty of evidence to show that the dramatist was acquainted with the contents of the work. Under these circumstances the most probable date for Natyasastra becomes about 500 B.C. because it was known to Bhasa.�

4. Ascribed to 600 AD. See K. A. Nilakanta Sastri, op. cit., p. 82. See also ibid., p. 112: �In its present form the work cannot be placed earlier than the fifth century.

5. Dr. U. V. Swaminatha Aiyar, Cilappatikara mulainum... .atiyarkkunllar uraiyum. 5th ed., 1950, pp. 10, 79, 113, 115-6.

6.  Ibid., pp. 9, 231.

7.�Iruvakaik kuittin  ilakkanam arintu �: see Ara�kerru kdtal, line 12 (p. 57).

8. �Atu nurettu karanamutaittu �: see Swaminatha Aiyar, op. cit., p. 80.

9. Curt Sachs, World History of Dance, trans. Bessie Schonberg, Bonanza Books, New York, 1937, p. 221.

10. "Aruvakai nilaiyum aivakai patamum
irenvakaiya vankak kiriyaiyum
varuttanai nankum niruttakkai muppatum
attaku tolila vaku menpa�

See Swaminatha Aiyar, op. cit., p. 81.

11. Dola hastha (a gesture for hands) and kuttanam (movement for feet) are involved in dances pertaining to ahamarga (erotic sentiment). See Nandikesvara, Bharatarnava, ed. and trans. by K. Vasudeva Sastry, Saraswathi Mahal Library, Tanjore, 1957, pp. 34, 138.

12. �Kottiran tutaiyat�r mantila makak
kattiya mantilam patinonru p�kki�.

Arakerruru katai, lines 144-5 (see Swaminatha Aiyar, op. cit., p. 74). 

13. �Pancatala pirapantamakak kattappatta teciyottai oru talattirku irantu parraka pattum tirvu onrumaka patinonru parrale teci kuttai ati mutittu enka.� Ibid., p. 119 (Arankerru katai).

14. "..atanai aituman tilattal", Arakerruru katai, lines 153-2.

15. �Naltiya nannull nankukataip pitittu". Ibid., line 158.

16. Viswanatha Pillai, Tamil-English Dictionary, 7th ed., Madras School Book and Literature Society, 1963.


Details of Demonstrations

* Denotes demonstration during the course of reading the paper.

Demonstration after Reading the Paper:

1. Demonstration of Kararnas along with the projection of the slides of respective sculptures.

(a) Representational: Gajakrsiditakam, Mayuralalitam.

(b) Abstract: Talapuwaputam, Kajiccinnam, Mattafli. (Lasya type); Danpapadam, Aksbiptarecitam (Tandava type).

(c) Demonstration of Karanas evolution into Adavus.

2. Demonstration of Tirmanam choreographed with Karanas weaved into Adavus. Example from �Meenakshi Kalyanam� Dance-drama.

3. Demonstration of the jati (rythmic syllables) inscription. This inscription in Brahmi script at Aricalur has been recently deciphered and is ascribed to the second century.

4. Demonstration of a Thillana which was composed simultaneously when the Karanas wore included in the Choreography.


Mail Us Copyright 1998/2009 All Rights Reserved Home