Home | Biodata | Biography | Photo Gallery | Publications | Tributes [Back to Archaeology List] Archaeology

Studies on Mahabalipuram Monuments A Review
Gift Siromoney and M Lockwood
Madras Christian College History Series 75-2
Paper presented at the Historical Society on September 3, 1975
Also in Indological Essays, Commemorative Volume II For Gift Siromoney
edited by Michael  Lockwood, Madras Christian College, 1992

The main questions that are raised in Mahabalipuram studies are, "Who built the monuments and when?" and "What do the monuments and sculptures represent?". Mahabalipuram has more than 14 cave-temples, 9 monolithic shrines called the rathas and 4 relief sculptured rock panels.¹ The chief tools used in Mahabalipuram studies are those of art history. Monuments have been studied from the point of architecture, sculpture and epigraphy. The Stalasayana Perumal temple, the main temple under worship, is usually left out of the purview of Mahabalipuram studies since the structure belongs to the post-Rajasimha period. The main monuments of Mahabalipuram and Saluvankuppam belong to the Rajasimha or the pre-Rajasimha period of the seventh and the eighth centuries A.D.

 The main source material for the study is the collection of monuments at Mahabalipuram. Scholars have tried to relate their knowledge of other monuments far away from Mahabalipuram to the Pallava monuments of Mahabalipuram. Some European sailors and travellers of the 16th, 17th and 18th centuries noted certain broad similarities between monuments of Egypt, Siam, Italy and Greece on the one hand and ² the monuments of Mahabalipuram on the other and postulated theories that attributed foreign influences on the authors of Mahabalipuram monuments. Some others noted the broad similarities between Mahabalipuram monuments and other Indian monuments from Ellora and Amaravati. Such comparisons made by the early visitors to Mahabalipuram are interesting. One must remember that when such comparisons were made there were no photographs of the monuments. 

The rough pyramidal outline of the Shore-Temple reminded the travellers of the Egyptian pyramids. Those who had seen the temples of Siam could see the similarity between the sculptures of Siam and South India but did not realize that the Siamese monuments were directly influenced by the South Indian ones and not the other way around. Amaravati sculptures had traces of the Gandhara art which represented Indian figures in European attire. Certain similarities between the art of Amaravati and that of Mahabalipuram were noticed and thus Mahabalipuram art was linked up with Greek and Roman art. Roman influence on Mahabalipuram art was postulated even as late as the beginning of this century.

Once Indian scholars entered the field they brought to bear upon Mahabalipuram studies their knowledge of the Puranas, the Agamas and Sanskrit and Tamil literature. Accounts of Arjuna in  Kiratarjuna and Mahabharata were used to identify the figures in the main Penance Panel. In addition to tradition, Sanskrit texts on iconography were used to identify the divine figures represented in Mahabalipuram sculptures. Now Tamil translations of many Sanskrit texts are available. Descriptions of Korravai in Silappatikaram were compared with the depiction of Durga at Mahabalipuram. Using Tamil literary sources it was established that Mahabalipuram was the well known port called Mallai or Kadalmallai. Sanskrit sources such as  Avantisundarikatha made reference to Mahamallapura. Mahabalipuram is referred to as Kadalmallai in the inscription of Rajaraja found in the Shore Temple.

Inscriptions found on Pallava monuments outside Mahabalipuram have also been used to settle questions on Mahabalipuram. For instance, the Vayalur inscriptions of Rajasimha give a chronology of Pallava kings. The Copper-plate inscriptions of Pallava monarchs record the achievements of the kings. The discovery of hero-stones with inscriptions in the regnal years of Mahendra and Simhavishnu have led to rethinking on the years of reign of Pallava kings who immediately preceded Mamalla to whom a majority of monuments are attributed.

Local people call the monolithic monuments rathas or chariots. Since there are five grouped together they called them Pancha Pandava rathas. Since Durga also called Draupadi Amman is depicted on one of the monoliths it is called the Draupadi ratha.

Only about a hundred years ago the monoliths were correctly identified as temple vimanas by M.W. Carr (1869). Scholars like Fergusson in the nineteenth century claimed that Dharmaraja ratha was an imitation of a Buddhist vihara. Around 1860, detailed plans of the monuments were prepared with elevations. The Athiranachandesvara temple was cleared of sand more than once to reveal the rock-cut cave. Description of the monuments in terms of architectural features has continued till recent times. At the beginning of this century Dubreuil ³ suggested an evolutionary framework a kind of seriation based mainly on architectural features for all the  South Indian monuments and the scheme is still followed today. He showed for instance that all the pillars of Mahendra's cave temples had square base and a plain corbel. During Mamalla's period the pillars became more slender and the sitting lion motif was introduced. These which started as a simple window in the Pallava period became very elaborate in the Vijayanagar period.

Scholars like K.R. Srinivasan and K.V. Soundararajan used technical terms of the stapatis, the practising sculptors of Tamil Nadu. They freely used both the Sanskrit and Tamil words to describe Mahabalipuram monuments. K.R. Srinivasan has recently brought out a book dealing with a detailed description of Dharmaraja ratha  which succeeds his earlier work on the cave temples of the Pallavas.

To settle question of authorship of Mahabalipuram monuments which were created within a short period of less than a hundred years one has to establish the evolutionary changes in architectural features that took place within a hundred years. The main problem is the fact that old features persisted side by side with new ones. There are some features which are treated as new by some scholars but not so by others.

Take for instance the appearance of sculptured stone panels at the back of the sanctum. None of Mahendra's temple has a stone panel at the back even though they probably had a painting or a wooden panel at the back. Shrines, such the Kailasanatha temple, Kanchipuram, Panamalai temple and the Shore Temple which are attributed to king Rajasimha have Somaskanda stone panels on the back walls. What happened during the intervening period of about 70 years? Ganesa ratha and Dharmaraja mandapa, both attributable to Paramesvaravarman, do not have panels at the back. Arjuna ratha, Sahadeva ratha and the Varaha mandapa do not have any.⁶ on the other hand, Draupadi ratha. Dharmaraja ratha (top floor), Trimurthi cave, Mahishamardini mandapa and Ramanuja mandapa have stone figures on the back wall. The reclining Vishnu figure, though not a panel, is placed at the sanctum. The temple depicted in bold relief at the Penance Panel has a figure of Vishnu carved at the back. Where do we fit in this feature? The old feature of a wooden panel fixed at the back probably existed for some time, side by side with the new feature of stone images. To the observer it would not have made any difference since both would have been painted in bright colours.

The process of evolution of architectural features was not acceptable to T.N. Subramanian and R. Nagaswamy and they attributed the monuments to Rajasimha on epigraphical  grounds.^6,7

Those working on architectural features showed that the earliest forms of the temple gopura was to be found at the Shore Temple and at the Kailasanatha temple of Kanchipuram.

Those scholars whose main interest was in the field of iconography identified the different features on the basis of weapons, ornaments as well as the mudras or gestures. They described the figures in terms of the stance or bhanga. The Tamil terminology ner~ner~ner for samabahanga has not so far been used to describe features in the straight posture. In the early period many figures were identified on the basis of tradition. Later Sanskrit texts and Gopinatha Rao's Hindu lconography written at the beginning of this century, were made use of. About a hundred years ago A. Hunter, at a meeting at the Evangelical Hall (part of the Old Christian College), showed photographs of Mahabalipuram monuments and interpreted the Somaskanda panel as the scene depicting Buddha taking leave of his wife. Today the picture has completely changed and we are no longer in doubt about the identity of most of the deities represented in Mahabalipuram monuments.

Regarding the identification of human figures there is a long history of disputes among scholars. Who do the figures on the Arjuna ratha represent? Who is the ascetic figure in the Penance Panel? Is it King Mahendra I or his grandson who is represented in the Adivaraha cave? It is not clear whether such disputes can ever be settled with absolute certainty.

We shall first take up the controversy over the Penance Panel. If the central figure doing penance represents Arjuna then the panel can be interpreted as Arjuna's penance. Hunter identified the panel as depicting a Buddhist scene. More recently Mayilai K. Venkataswami has interpreted the panel as depicting the Jain legend of Sagara and his sons. In 1914 one V. Goloubew put forward the interpretation of the descent of the Ganges. Dubreuil was also of the same view and identified the central figure as Bhagiratha. To support the Arjuna's penance theory, the Mahabharatha and the Kiratarjuniya are  brought in as evidence. This puts people like T.N. Ramachandran on Arjuna's side and Nilakanta Sastri on Bhagiratha's. The controversy is likely to go on till some new epigraphical or literary evidence comes to light.

Let us consider the dispute over the identity of the royal figures in Adivaraha cave in which label inscriptions were discovered in 1925. Panel I has a royal figure seated on a throne attended by two women. It has the label inscription Simhavinna potradhirajan. Panel II has the label Mahendra potradhirajan and has a standing figure with two women. The women in the panel are identified as queens. Many scholars have identified the king in Panel I as Simhavishnu, some as Narasimha and others as Rajasimha. Panel II has been interpreted as a royal portrait of  Mahendra I by some, as Mahendra II by others, and as Mahendra III by some others.

There are nine possible combinations of royal pairs but some of which may be ruled out. We present here in Table I, the different solutions put forward by scholars.

TABLE I
IDENTIFICATION OF ROYAL FIGURES OF ADIVARAHA CAVE BY DIFFERENT SCHOLARS

 We have rejected the possibility of the two panels representing the same king since the labels are different and they are roughly coeval with the panels. The identification, to a large extent, is dependent on the question, "Who built the Adivaraha cave and when?" For instance if one believes that the builder was Simha Vishnu, then the sitting figure cannot represent Narasimha or Rajasimha who came decades later than Simha Vishnu. One of the peculiar arguments used in this controversy is whether a son, even if he were to be the reigning monarch, could be represented in a temple seated on a throne when the father is represented standing in a panel opposite to his son's figure. This is probably the reason why K. R. Srinivasan changes his position between 1958 and 1964 since he feels that the standing Mahendra could be only the son of sitting Narasimha and not his father. No one seems to have raised the question whether Lakshmi could be represented sitting and bathing in the same cave temple where Vishnu, Siva and Brahma are represented standing.

Inscriptions play an important role in Mahabalipuram studies. Goldingham is reported to have taken note of the inscriptions of Dharmaraja ratha as early as 1798. Babington reported the existence of two Sanskrit inscriptions from Athirana Chandeswara temple in 1830 in the Transactions of the Royal Asiatic Society. These two inscriptions have many verses in common and are written in two different scripts, viz. Nagari and Pallava Grantha. Most of the Pallava inscriptions found on Mahabalipuram monuments are in Pallava Grantha scripts. Burnell brought out in 1878 his classical work on South Indian scripts and within a few years the first volume of South Indian Inscriptions was published. During the last two decades the label 'Mahamalla' on the Dharmaraja ratha and the label 'Narapatisimha-Pallavesvara-Grham' on the lintel of the Vishnu temple in the Shore Temple complex were discovered by K.R. Srinivasan. Inscriptions on Dharmaraja ratha are engraved in a script more archaic than those found in the Ganesh ratha and the Dharmaraja mandapa. Identical verses have been found in the last two monuments and these are attributed to Paramesvara by many scholars including K.R.Srinivasan and T.N. Ramachandran.

Nagaswamy and Subramanian who attributed to Rajasimha the common authorship of many monuments used a peculiar kind of argument. Many Pallava Kings had many birudas in common and it is difficult to establish that certain titles were held exclusively by certain kings. Nagaswamy based his theory on the premise that Atyantakama was the exclusive title of Rajasimha. Furthermore just because a certain title is found on a monument, the monument can not be attributed to the king with the title. Succeeding kings also engraved their inscriptions on existing temples which they did not build. The difficulties in the methodology are dealt with separately in Mahabalipuram Studies.

Another peculiar kind of argument was used by reputed scholars like T.N. Ramachandran and Nagaswamy. They would fix the authorship of monuments on the basis of their own interpretation of birudas. To Ramachandran, Mahendra I was vichitra-chitta, a king with a brilliant mind, and many monuments were attributed to Mahendra on that basis. To Nagaswamy, Rajasimha was atyantakama which he interpreted to mean a man of endless desires. Monuments were attributed to Rajasimha on the evidence that he was a man of endless desires.

It is an indisputable fact that letters changed in form from century to century and inscriptions can be dated on the basis of paleography. The problem in Mahabalipuram is to work out an indisputable system of seriation for a span of about 125 years. Since it is not possible to work out such a system disputes cannot be settled solely on the basis of paleography.

A new approach to Mahabalipuram studies was made in the paper on costumes and jewellery published in the Madras Christian College Magazine in 1971. It was found that the dress and ornaments of the pre-Rajasimha period were distinctly different from those of Rajasimha period. Lockwood, Siromoney and Dayanandan developed a method of seriation based on ornaments to classify the Somaskanda motif into pre-Rajasimha and Rajasimha periods. They showed that the horned Dvarapalakas were ayudapurushas. They established the reworking of the Mahishamardini cave from a Vishnu shrine to a Siva shrine with a Somaskanda at the back wall of the central sanctum. On the basis of the study of ornaments they rejected the theory of Nagaswamy that Rajasimha was the sole author of Mahabalipuram monuments.

During the last 100 years scholars from different disciplines have worked on Mahabalipuram monuments. One would expect in the future, work based on computer techniques to grow and develop into an area of computer iconometry.

Foot Notes

1. Michael Lockwood, Gift Siromoney and P. Dayanandan, Mahabalipuram Studies, Madras:  Christian Literature Society, 1974.
2. William Y. Willets, An Illustrated Annotated Annual Bibliography of Mahabalipuram Studies, Kuala Lumpur: Department of Indian Studies, University of Malaya, 1966 . Many of the references in the paper are based on this bibliography.
3. Jouveau-Dubreuil, G. , Dravidian Architecture. Madras: 1917.
4. Srinivasan, K.R., The Dharmaraja Ratha and its Sculptures Mahabalipuram, New Delhi: Abinav Publications, 1975.
5. Srinivasan, K.R., Cave Temples of the Pallavas, New Delhi, 1964.
6. Nagaswamy, R., "New light on Mamallapuram" , Silver Jubilee Volume of the Archaeological Society of South India , 1962, pp. 1-50.
7. Subramanian, T.N., The Pallavas of Kanchi in South-East Asia, Madras: Swadesarnitran Press, 1967.
8. Gift Siromoney, "Mahabalipuram Costumes and Jewellery", Madras Christian College Magazine, XXXIX, 1970, pp. 76-83. The diagrams are reproduced in the revised edition of C. Minakshi's Administration and Social Life Under the Pallavas, Madras: University of Madras, 1977.