Home | Biodata | Biography | Photo Gallery | Publications | Tributes
What are the essential features of the Indian systern of iconometry? Do the canonical texts describe a single system or a mixture of two systems? Did the Pallava sculptors follow the canonical proportions prescribed in any of the Silpa Sastras that have come down to us?
To answer these and other interesting questions, measurements were made on the sculptures using anthropometric instruments and the data analysed using a giant electronic computer.
Canons of iconometry follow the ancient "Taalamaana" system in which the basic units are the angula and the taala, and the latter stands for the length of the palm. The angula is either a fixed length or a proportion. We have
found that there are two different systems of iconometry described in the
textthe pure "Taalamaana" system and the derived "Taalamaana" system. Over the years the two systems had got mingled together and the co-existence of these two systems has been over looked by scholars.
According to the Indian system of iconometry, a sculpture could be made according to one of the ten main types of proportions. Each type is further sub-divided into three sub-types. The basic type of proportion is the standard nine face-length and it is called the ''Madyama Nava Taala" in the Sanskrit canonical texts.
In this type, the face is of length one "taala" or 12 "angulas" and the total height is nine "taala" or 108 "angulas". In the pure "taalamaana" system any sculpture made according to the "nava taala" type of proportion would have a height of nine "taala" units and the face length would be of one "taala" unit. In other words, the face will be one-ninths of the total height irrespective of the actual size of the sculpture. There are two other sub-types in the "nava taala" type and one sub-type is longer by four "angulas" and the other shorter by four "angulas".
The Indian system makes use of the fact that persons with disproportionately larger faces appear short and those with disproportionately smaller faces appear tall. Dwarf figures were made by following the four "taala" system where the total height is only four times the face length. Canons of iconometry describe ten types from the "eka taala" or single "taala" to "dasa taala" or ten "taalas". "Dasa taala" proportions were prescribed for images of the gods so that they appear tall and majestic. When we measured human beings, we found that there are people with total height eleven times or even twelve times their face length, but the canons do not approve of such proportions.
We have found that the surviving Silpa texts do not follow the pure "taalamaana" system uniformly. A second system must have got superimposed on this ancient canonical system and today both the systems are found in the surviving texts. In the pure "taalamaana" system a "dasa taala" proportion would mean that the total height is ten times the face length. In the derived "taalamaana" system, however, ''dasa taala" just means that the total height is 120 "angulas" or 10 "taalas" irrespective of the face length. Some proportions labelled "dasa taala" of the second system may be such that the face length is only one-ninth of the total height. Practising sculptors might have wished to make two sculptures in the same proportion but in different heights and such proportions might have been named after the total height such as the nine "taalas" (9 x 12 "angulas") or the ten "taalas" (10 x 12 angulas) without any reference to the face length.
To confirm the finding that there are two systems of proportions we made use of a technique called cluster analysis. The different proportions found in south Indian texts were fed into the computer and different clusters or groups were formed. It was found that in many cases the "nava taala" and the "dasa taala" proportions were grouped together, showing thereby that what was "nava taala" according to the pure system was labelled as "dasa taala" according to the derived system and had found its way into the texts. Thus we were able to confirm our discovery of the existence of two systems with the use of computer methods.
Though there is a lot of variety in measurements between the different texts, we found that there are certain common features . First, the face length is equally divided between the forehead, the nose and nose-to-chin. Secondly, the pubis is the midpoint of the height of a nude figure. Thirdly, deities are prescribed a higher "taala" compared to human figures. Fourthly, children are represented in a lower "taala" such as the four "taalas" of the pure system. This is because the face length, compared to the total height, is larger for children than for adults among humans. There are some exceptions to the general features, but we shall not discuss them here.
With these general features in mind we examined the Pallava sculptures of the Kailasanatha temple in Kanchi which was built around 700 A.D. by King Rajasimha. We took actual measurements with anthropometric instruments and used computer methods of analysis. The sculptures are of extraordinary beauty and delicacy, but the proportions are quite different from the canonical proportions that have come down to us.
The eyes are depicted as looking upwards giving a dreamy quality to the figures, but such a depiction of the eye is prohibited according to the available texts. The nose is disproportionately long, the chin very small and the eyes long and narrow. These characteristics are also seen in the famous painting at Panamalai and in the sculptures of the Shore Temple in Mamallapuram. All these three temples were built by Rajasimha.
Usually the facial proportions are compared only visually, but we have used precise mathematical measures of similarity to compare the facial proportions. From a computer analysis of facial proportions we have found that the sculptures of deities in the Kailasanatha temple are quite different from those found in the Krishna Mandapa in Mamallapuram which belongs to the pre-Narasimha period and the Vaikuntha Perumal temple in Kanchipuram which belongs to the post-Rajasimha period. Many sculptures in the Kailasanatha temple have disproportionately long legs which is a non-canonical feature. In the famous Arjuna's Penance panel (Mamallapuram) which belongs to the pre-Rajasimha period the main ascetic figure, has disproportionately long legs and the proportions do not fulfill the canonical norms.
Could some other canon which has not come down to us been used by Rajasimha's sculptors? If we assume that sculptors using some canon would have strictly adhered to the iconometric rules, they would have made sculptures of the Somaskanda panels of the small shrines in front of the Kailasanatha temple in the uniform pattern. However, there is a good deal of variety in the sizes of the Somaskanda panels, though all the small shrines are of the same height. The Silpa Sastras prescribe the height of the main deity in proportion to the size of the entrance to the sanctum. The sculptors could not have followed any canon of iconometry.
A computer analysis shows that the figures of gods and goddesses, chauri bearers and male worshippers have close resemblance. Such close facial similarity would support the theory that the same artist was responsible for many figures on the main vimana of the Kailasanatha temple.
In conclusion, there are two systems of proportions intermingled in our Silpa texts. The sculptures of Rajasimha Pallava were not made according to any of the canonical proportions that have come down to us. Computer methods are useful in analysing the sculptures and for finding those pieces which resemble each other closely. It is possible to construct a canon based on the Rajasimha sculptures based on the average values of proportions of the different classes of figures, but such a canon will be quite different from the canons found in the existing Silpa Sastra texts.
I am thankful to Mr. M. Bagavandas and Mr. S. Govindaraju for help in writing this article.