Pramāṇa · Search · About · API

The Colas c35.s1: Sculpture

The Cōḷas c35.s1

Language
English

Contents

Sculpture

c35.s1.1

The Pallava sculptors had won the palm in the art of bas-relief. The primal simplicity of conception and the sureness of grouping that is seen in the depiction of Kṛṣṇa as *govardhana-dhara*, or of Durgā as fighting the Mahiṣāsura, the more complicated but not less convincing ‘tapestry on rock’ known as the descent of Gaṅgā (or Arjuna’s penance?)—all at Māmallapuram are works of such merit as to command a place in any considerable history of world art. The portraiture of Simhaviṣṇu and Mahendravarman I and of their queens in the Ādi Varāha cave also at Māmallapuram deserves in turn high praise for its unusual excellence. The Cōḷa sculptors did not attempt the ‘large frescoes in stone’ as the Pallava bas-reliefs have been called. They took more to sculpture in high relief, nearly in the round, and they concentrated more on the rich iconography that had come up in their time as the result of the growth of the hagiology of the *nāyanārs* and *āḻvārs*, and the tremendous vogue that their popular hymns gave to *purāṇic* mythology in the Tamil country. When they attempted narrative sculpture in stone, it was generally in small panels, sometimes as small as six inches by four as at Nāgēśvara in Kumbakonam, and seldom more than two feet by one foot—which is roughly the measure of the Rāmāyaṇa panels of the Kampahareśvara in Tribhuvanam. They did not fail even to excel the Pallava sculptor in portraiture when they attempted it as at Śrīnivāsanallūr and Kumbakonam (Nāgēśvara); but for some reason not now clearly traceable they did not continue their work in this line, though they attained unexampled skill in the casting of metal icons of very considerable size.

c35.s1.2

Except where metal was used, sculpture was generally subsidiary to architecture but did not, on that account, cease to belong technically to the plastic or glyptic form of art. It was employed as decoration on the walls, pillars, plinths, roofs and other convenient spots in temples. The Cōḷa sculptor, however, appreciated the value of plain spaces on walls and did not crowd them with too many figures; nor did he adopt the method of the ivory carver and goldsmith and seek to produce filigree work in his ornamentation. He generally worked on hard rock and depended on bold strokes and flowing lines for his effects, and the metal images of the time partake of the same qualities, though here the details are naturally more and more elaborately wrought. Portraits, icons, and decorative sculpture are the three main classes of Cōḷa sculpture. Portraits are few and early; they soon fall out of use and often the line between a portrait and an icon is not clear particularly in the case of saints and other religious persons. Several sculptures, obviously portraits, cannot now be identified as we lack the data for it. Among icons which form the most important part of Cōḷa sculpture, Śaiva figures predominate as the Cōḷas were ardent Śaivas, though very fine Vaiṣṇava and Jaina images are not unknown. The iconography is by no means rigidly fixed,⁴⁶ and the art is still fairly free from the domination of text-book rules; there is little that is mechanical and hide-bound about Cōḷa sculpture almost to the end. Decorative sculpture takes many forms; architectural motifs, floral and vegetal patterns, friezes of animals, birds, dancing figures with musical attendants, and of legendary and purāṇic stories. It will not be possible to review here the enormous mass of material to be gathered from the numberless temples, but present only a few select and typical instances of each class. It must also be noted that many of the fine bronzes of the Tanjore temple described in detail in its numerous inscriptions of the reign of Rājarāja I are not now forthcoming.

c35.s1.3

There is a singular paucity of properly authenticated portraits in Cōḷa sculpture, as to some extent in all Indian sculpture. The underlying cause is seen from a statement in *Śukranītisāra*,⁴⁷ doubtless a late work, but probably embodying an authentic old tradition: 'A divine image, though ugly, does good to men; while a human image, however excellent, does them no good.' Again the natural instinct to reproduce the features of the subject from life was often restrained by a tendency to merge the individual in a type. There were some conspicuous exceptions in early Cōḷa art, but we have now no clue to the identity of these exceptionally realistic portraits. There are three of them, two women and a man (XII, figs. 21-23), all sadly mutilated, on the walls of the Koranganātha temple at Śrīnivāsanallūr; and several others, men and women (figs. 24-29), in the Nāgēśvara at Kumbakonam, which are very well preserved and nearly life-size. Writing of the Kumbakonam sculptures Ajit Ghose observes:⁴⁸ ‘Here for the first time the Cōḷa artist stands in sharp contrast with his Pallava predecessors and the latter’s severely abstract, ideal, and schematic vision. There is no difference in outward bearing between a Pallava king and a god, between a goddess and a queen. But a new and attractive conception of life and beauty had dawned on this Cōḷa sculptor. These Cōḷa ladies are picturesque and realistic human figures, full of feminine grace and the joy of life. This intensely human quality may be said to distinguish every one of the statues in the niches of this shrine. This art, so unconventional, is thus, refreshingly original in conception and spirit. This humanism is the Cōḷa’s principal contribution to South Indian Art.’ In their free and easy poses, in the divergence of their features and forms which impart a pronounced individuality to each figure, in the excellence of their modelling and in the discerning treatment of facial expression particularly of the eyes and mouth, these sculptures have no parallel in any other phase of South Indian art, not even in the Cōḷa period. Though not modelled in the round, being undetached from the blocks from which they are carved, their forms stand out clearly owing to the technical skill of the execution. The figures are generally posed in profile, and the exception here (as also the figures of Śrīnivāsanallūr), where the traditional law of frontality is followed, serves only to emphasise the superiority of the profile presentation. The treatment of dress, coiffures of women, and ornament of both men and women reveal much of life in the high society. The more poignant is our regret that we can only guess that these were perhaps representations of royal donors or members of the royal family. A close scrutiny does indeed reveal a few weak spots in the treatment of the fingers of the hand in one female figure, of the bust in another, and of the feet in all; but the general superiority of these statues is so pronounced as to enforce a revision of the common view that Cōḷa sculpture attained its high-water mark under Rājarāja I and Rājēndra; these sculptures of Śrīnivāsanallūr and Kumbakonam precede the accession of Rājarāja I by more than a century. But we have nothing else like them either before or after.

c35.s1.4

The earliest definitely dated portrait of the Cōḷa period is that of Tirukkarṟaḷi-piccan (fig. 30), who is said to have built (A.D. 932) the stone temple of Tiruvāḍutuṟai.⁴⁹ His figure is cut in relief, about a foot in height, on the south wall of the central shrine of that temple; by his side, in another panel separated by a narrow vertical band, stands the portrait of Iḷaiya Tirunāvukkaraiyan, a devotee of the Lord of the stone temple.⁵⁰ Though the figures are worn by the lapse of time, still the features and postures of these two persons give evidence of the realistic reproduction of facial traits practised by the artists of the tenth century. Piccan has a bag of sacred ashes slung on his left arm while the other devotee has set it on a stand near by.⁵¹

c35.s1.5

A beautiful bronze of a woman 'whose beauty is disturbing' (XV, fig. 31) is preserved in the Freer Art Gallery, Washington D.C. and published by A. K. Coomaraswamy who considers that it may represent Lakṣmī or Pārvatī, or even be the portrait of a queen, say Śembiyan-mahādēvī;⁵² its pleasing proportions and workmanship certainly make it an early Cōḷa bronze. We shall notice presently other instances that hover between being a portrait and an icon.

c35.s1.6

The only portrait known of the reign of Rājēndra I is an excellent bronze statue in the Kāḷahasti temple representing Cōḷamadēvi, the queen of Rājarāja I (XV, fig. 32). The age and identity of the statue are alike most satisfactorily attested by an inscription on its pedestal saying that it was cast under the orders of Rājēndra-Cōḷa-dēva by Niccapaṭṭālagan, obviously the sculptor.⁵³ We have here the first definitely dated South Indian metal image—an excellent portrait and fine specimen of the art of the time.

c35.s1.7

An inscribed metal statuette (fig. 33) of a boy with a dagger in his right hand, the left being held in *kaṭaka-mudrā* at about chest height, is a figure of Kulōttunga III, a gift of Uḍaiyanambi to the temple of Kāḷatti-Uḍaiyār in Kāḷahasti;⁵⁴ the image may have been made about the time of his accession; the figure wears many ornaments and the face is expressive of youthful energy and eagerness. The image is important as perhaps the only authentic contemporary portrait of a Cōḷa monarch so far known, and its date is most probably somewhere about 1180.

c35.s1.8

Three statues, one of stone from Śrīnivāsanallūr, and two of metal from the South Arcot and Tanjore districts are, like the female figure noticed above, on the borderland between portraiture and iconography. The statue from Śrīnivāsanallūr (fig. 35) stands in a niche on the north wall of the Korangonātha temple; the presence of other portrait statues in other niches here makes it likely that this too is a portrait much better preserved than the rest; but its position and the garland betweeen the palms held in *añjali* lead us to think of Ādicaṇḍēśa.⁵⁵ The metallic statue from Tirunāma-nallūr (S. Arcot) believed, to be that of Narasinga Munaiyadaraiyar (fig. 36), the Milāḍa Chief famed as the patron of Sundaramūrti; and another from Kōḍikkarai (Tanjore district) said to represent Gōḷakamaharṣi (fig. 34), the founder of a well-known line of Śaiva Ācāryas, are both more icons of the types of, royal patron and spiritual leader, than portraits proper.

c35.s1.9

We may now turn to the icons, first stone, and then metal. The Śaiva forms predominate in those presented here as in the Cōḷa period generally; Vaiṣṇava icons will not be treated separately, but as far as possible in their chronological order together with Śaiva forms. The reader will be left to gather iconographic details from a study of the reproductions and the specialised treatises on the subject as full details cannot find a place in a general historical account. The relative legends should also be gathered from these treatises.

c35.s1.10

We may start with two images of Vīṇādhara Dakṣiṇāmūrti enshrined in appropriate niches in the temples of Tirukkaṭṭaḷai (fig. 37) and Koḍumbāḷūr (fig. 38) in Pudukkottah; the difference in the shape and disposition of the Vīṇā in the two cases should be noted. It is possible, however, that the Tirukkaṭṭaḷai figure is that of Tripurāntaka, as may appear from a comparison with another figure found along with that of Tripurasundarī at Koḍumbāḷūr (figs. 39, 40). From Koḍumbāḷūr again we get a fine image of Ardhanārīśvara (fig. 41); though somewhat mutilated the excellence of the sculpture and the deft handling of the right male and left female half is apparent. The male half often has two arms, one resting on the bull and the other holding a symbol such as trident, so that the whole image has three arms; here we find, however, only two arms. Krishna Sastri reproduces a drawing of an unusual image from Tiruvadi (Tanjore district)

c35.s1.11

where the right half is the female part, and the left the male part with only one arm held akimbo, the elbow resting on the head of the bull.⁵⁶ From the Koranganātha temple of Śrīnivāsanallūr we have the sadly mutilated but superb renderings of a standing Śiva (fig. 44), probably resting his right foot on the shoulder of a *gaṇa*, and of a seated Dakṣiṇāmūrti (fig. 45) beneath a tree in a finely wrought *toraṇa* pavilion surrounded by pupils and characteristic animals, demi-gods and so on.

c35.s1.12

The Śiva temple at Tiruvālīśvaram (Tinnevelly district) is a veritable museum of superb early Cōḻa iconography of the time before Rājarāja I. In the niches of the *pañjaras* of its *vimāna* and the recesses between them, there are magnificent sculptures of Śiva in his various forms. On the southern side, for instance, there is Naṭarāja in the centre (fig. 48), with Vṛṣabhārūḍha and Gaṅgādhara on his proper left (fig. 47), and Vīrabhadra and Dēvī on the right. The western side has Liṅgodbhava with Viṣṇu and Brahmā on either side in the centre, Kālaharamūrti and Kirātamūrti on its proper left, and Yoga Dakṣiṇāmūrti and Umāsahita on its proper right (fig. 49). The northern side shows Gajārīmūrti in the centre (fig. 50), Caṇḍeśānugraha and Sukhāsanamūrti on its proper right (fig. 51) and Somāskanda and an unidentified sculpture on its proper left. Elsewhere we get an *ardhanārī* of the usual type with three arms, and a group of Somāskanda (?) standing with Nandi and a *gaṇa* sportfully seated near by (fig. 46). Another seated Dakṣiṇāmūrti from Kāvēripākkam (fig. 52) with a book in left fore-arm and the right fore-arm (which was perhaps in *Jñānamudrā*) broken, furnishes instructive comparison with the Yoga Dakṣiṇāmūrti of Tiruvālīśvaram, and the two Viṇādharas from Pudukkotah mentioned earlier.

c35.s1.13

Other interesting early Cōḻa sculptures are the relief of an eight-armed Durgā—on a slab from the Vīraṭṭānēśvara with two devotees kneeling on either side, the one on the left apparently about to offer his head in sacrifice to the goddess, and a group of Viṣṇu and his two consorts from the ruined temple of Viṣṇu at Olagapuram, South Arcot. The walls of the two Bṛhadīśvara temples of Tanjore and Gaṅgaikoṇḍa-cōḻapuram contain numerous icons of large size and forceful execution, of which only a few can be reproduced here. First may be placed an image of Sarasvatī (or Lalitā) from Tanjore, (XXIII, fig. 54), and a magnificent Naṭarāja on the south side of the *garbhagṛha* of Gaṅgaikoṇḍa-cōḻapuram (XXII, fig. 55), a conception of Śiva in cosmic dance of which more will be said later when we consider metal Naṭarāja images. The fine Harihara (fig. 56) on the same side comes next. The Caṇḍēśānugrahamūrti (fig. 57) on the north side is well known as it has often been reproduced and admired by several writers, but cannot on that account be omitted here; it may be compared with the smaller panel of the same subject from Tiruvālīśvaram, (fig. 51) where the bull serves as the footstool for Pārvatī; the later sculpture is grander and perhaps more perfect in its technique. The Kāmāntaka panel on the north wall (fig. 58) is a fine composition whose correct identification is suggested by Pārvatī doing penance on one leg on right top corner, Manmatha and Rati represented to the proper right of the seated central figure of Śiva as flying, obviously after the command of Indra, to go and disturb Śiva's meditation; the panel is thus part icon and part narrative. We must also notice an image of Viṣṇu with his two consorts facing him from the sides, all framed in a typical niche on the wall (fig. 59). There is a curious Sūrya stone in the form of a Kamala-yantra drawn by seven horses on the west face of the *mahāmaṇḍapa* of Gaṅgaikoṇḍa-cōḻapuram temple (fig. 60); it contains all the paraphernalia of Sūrya, but no image of the deity who seems to be represented by the lotus form of the top of the stone. We shall conclude this rapid survey of Cōḻa iconography in stone with a reference to an image of Kaṅkāḷamūrti (fig. 61) and a vigorous representation of Gajahāmūrti (fig. 62) with frightened Umā and Skanda by the side—both from Dārāśuram.

c35.s1.14

'Bronzes' is a convenient and conventional term for metal images irrespective of the exact number or proportion of metals mixed in the amalgam out of which the images are cast. Most of the Cōḻa bronzes were made by the *cire perdu* process, and the Tanjore inscriptions give some indications of the technical details, and mention images which were made solid, hollow, or semi-hollow with thick sides. They also describe many iconographic groups of images representing the stories of the lives of the Śaiva saints, but none of them seems to have survived. But what we now possess of Cōḷa bronzes from different parts of South India is enough by quantity and quality to bear undying testimony to the great height reached by metal-casting in those days. The Naṭarāja image in its various forms naturally holds the first place among the Cōḷa bronzes. Some fine stone images of this form have been mentioned above; the best bronzes of this type are not only spread over the museums of the world, but are still under worship in the live temples of South India. The Naṭarāja image in the Nāgēśvara temple, for instance, which has never been photographed, is one of the largest and finest images known. We have reproduced an image of the usual classical type from Tanjore (fig. 63), another with eight arms in Kālikā-tāṇḍava from Nallūr (fig. 66), and a third in Catura-tāṇḍava from Tiruvarangulam in Pudukkottai (fig. 67). Two other figures belonging to the Madras Museum and got from Tiruvālaṅgāḍu (Chittoor, fig. 65) and Vēlāṅgaṇṇi (Tanjore, fig. 64) are also given. The conception of the Divine Dancer and its cosmic significance, and the excellence of the Cōḷa sculptor's presentation of it have often won the highest praise from discerning critics including great modern sculptors like Rodin. This is what Grousset writes:

c35.s1.15

'Whether he be surrounded or not by the flaming aureole of the *tiruvācci* (*prabhāmaṇḍala*)—the circle of the world which he both fills and oversteps—the King of the Dance is all rhythm and exaltation. The tambourine which he sounds with one of his right hands draws all creatures into this rhythmic motion and they dance in his company. The conventionalized locks of flying hair and the blown scarf tell of the speed of this universal movement, which crystallizes matter and reduces it to powder in turn. One of the left hands holds the fire which animates and devours the worlds in this cosmic whirl. One of the God's feet is crushing a Titan, for "this dance is danced upon the bodies of the dead", yet one of the right hands is making the gesture of reassurance (*abhaya-mudrā*), so true it is that, seen from the cosmic point of view and sub specie aeternitatis, the very cruelty of this

c35.s1.16

universal determinism is kindly, as the generative principle of the future. And, indeed, on more than one of our bronzes, the King of the Dance wears a broad smile. He smiles at death and at life, at pain and at joy alike, or rather, if we may be allowed so to express it, his smile is both death and life, both joy and pain......From this lofty point of view in fact, all things fall into their place, finding their explanation and logical compulsion. Here art is the faithful interpreter of the philosophical concept. The plastic beauty of rhythm is no more than the expression of an ideal rhythm. The very multiplicity of arms, puzzling as it may seem at first sight, is subject in turn to an inward law, each pair remaining a model of elegance in itself so that the whole being of Naṭarāja thrills with a magnificent harmony in his terrible joy, and as though to stress the point that the dance of the divine actor is indeed a sport (*līlā*)—the sport of life and death, the sport of creation and destruction, at once infinite and purposeless—the first of the left hands hangs limply from the arm in the careless gesture of *gajahasta* (hand as the elephant's trunk). And lastly as we look at the back view of the statue, are not the steadiness of these shoulders which uphold the world, and the majesty of this Jove-like torso, as it were a symbol of the stability and immutability of substance while the gyration of the legs, in its dizzy speed would seem to symbolize the "vortex of phenomenon."'

c35.s1.17

We have a fine early Cōḻa Śiva—two armed, and his consort (fig. 68) recently discovered from Tiruveṇkāḍu. A four-armed image of standing Śiva, most probably Vṛṣabhārūḍha though the bull is missing, from Gaṅgaikoṇḍacōḷapuram (fig. 69), and a seated (*sukhāsana*) figure from the Pudukkottah Museum (fig. 70) may be noted. The first is an early Cōḻa image of this great God whose cult was promoted so much by the Cōḻa rulers; the other is perhaps of the late Cōḻa period. The episode in the *Mahābhārata* celebrated in Bhāravi's great poem *Kirātārjunīya* seems to have been very popular in those days. Its incidents are sculptured in stone in several places in stone including perhaps Māmallapuram, and metal images of the Kirātamūrti of Śiva are some of the finest among the South Indian bronzes. Here Śiva is shown standing and with two arms, a bow in one and arrow in another; a really early image with an ovoid *prabhāmaṇḍala* and the left arm raised to it in position to hold the bow comes from Tiruvēṭkaḷam, (fig. 71) near Cidambaram, one of the sites where, as the name shows, the Kirāta incident is localised; though this bronze has been thought to be Pallava,⁵⁷ there is no conclusive reason for placing it earlier than the early Cōḷa period—the period of transition as we have called it. Another slightly later image is a perfectly wrought bronze which comes from Rādhā Narasimhapuram (Tanjore district, fig. 72) and we give two views of this fine bronze showing details of *Jaṭāmakuṭa* and the ornaments of the neck besides the expressive facial features. An *āliṅganamūrti* with a *prabhāmaṇḍala* is a fine piece of early Cōḷa work from Tiruvāḍuturai (Tanjore district, fig. 73); the proportion of the figures, and their stances are very pleasing, and the details of coiffure, ornamentation and drapery worth careful study. Another popular Śaiva legend is that of the deity wandering as a naked beggar (*bhikṣāṭana*) in expiation of the sin of *Brahmahatyā* incurred by his cutting off one of the heads of Brahmā. All the wives of the sages of Dārukavana except Arundhatī, the wife of Vasiṣṭha, are said to have been entranced by the overwhelming beauty of the naked god's form, and this episode has given rise to many sculptures in stone and metal of the Bhikṣāṭanamūrti and the closely allied form of Kaṅkāḷamūrti reproduced above from Dārāśuram. We have a very pretty metallic image, a really early one, from the Śiva temple at Tirunāmanallūr, (fig. 74), which by its boldness and simplicity offers a striking contrast to a later, more ornate image from Tiruccengodu (Salem district, fig. 75) which has a deer to the right of the God and a *gaṇa* bearing the almsbowl on his head and is therefore more complete and equally pleasing with the earlier icon. The Tiruccengodu figure is a Bhikṣāṭana. The earlier image has no symbols in its arms as seen now excepting the adhering skull of Brahmā in the upper left arm. Skanda or Subrahmaṇya, the younger son of Śiva and Pārvatī, is a very popular deity, held by some to be an original Tamil god—Muruga, though, as in the case of many other members of the pantheon, modern criticism has recognised here also a syncretism of Aryan and pre-Aryan elements in the attributes and achievements of the deity; an early Cōḻa bronze of this god with an ovoid *prabhā* comes from Tiruviḍaikkaḻi (Tanjore district, fig. 76); a much later image from Tiruvarangulam, Pudukkottah (not reproduced) with circular *prabhā*, the peacock-mount and the two wives of the god standing on the sides may be compared with advantage for two different representations of the same concept; the earlier image with only one head and four arms, two of them bearing the characteristic attributes, is much more pleasing as a work of art, though devotees of the cult may see many reasons to prefer the later image and others of its class.

c35.s1.18

Four Vaiṣṇava bronzes, all early Cōḻa, from a ruined temple in Śērmādēvi (Tinnevelly district) next claim our attention.⁵⁸ Here are two figures of Viṣṇu, (fig. 77, 78) one of Rukmiṇī (? fig. 79), according to Ajit Ghose, and the last of Lakṣmī (fig. 80). Both the Viṣṇu images are in *samabhanga* pose, with the body in perfect equilibrium, and both on a *padmāsana* supported by a *bhadrāsana*. The symbols on the hands and gestures are the same in both. The larger image is simpler and bolder and may be of the time of Rājēndra I; it is 3 feet 2 inches high, the largest South Indian bronze of Viṣṇu known. The figure of Rukmiṇī (or is it also Lakṣmī?) is also remarkable for its severe simplicity and must be earlier than the Tanjore stone Sarasvatī mentioned above, or the Lakṣmī bronze now noticed. Both Rukmiṇī and Lakṣmī stand in *tribhanga* pose and, being early images, lack the breast-band (*kucabandha*); the former has her left leg slightly bent, and the latter the right, and there is a corresponding difference in the pose of the rest of the body resulting from the respective postures, Of these figures it has been said: 'Lakṣmī is a slender young goddess; Rukmiṇī is a matron in the full glory of maturity. Both are idealized but not exaggeratedly so.' As regards the dates of the four bronzes, the same writer observes: 'The Rukmiṇī is the oldest, next comes probably the smiling Viṣṇu and then the Lakṣmī, while the large Viṣṇu is later than these. All the images may be dated between A.D. 875 and 1032.' But their order in dates is not so sure. A group of three bronzes of Rāma, Lakṣmaṇa and Sītā from Tirukkaḍaiyūr (Tanjore district) with Hanumān in an attitude of worship (fig. 81) is obviously one of the finest products of the best age of Cōḷa bronze-casting, the time of Rājarāja I and Rājēndra, and is representative of a large number of similar groups many of which luckily are still extant in temples and museums. A unique group representing Pārvatī's marriage with Śiva (fig. 82) is also a recent discovery made at Tiruveṇkāḍu.

c35.s1.19

This brief study of the metal art of the Cōḷas may be concluded by a notice of the many finely executed seals borne on the big copper rings which bind the sheets of the copper-plate grants of the Cōḷas. One of the best preserved is also one of the most recent discoveries, and it is one of the two seals on the Karandai plates from Tanjore (fig. 83).⁵⁹

c35.s1.20.1

The last division of sculpture in stone that remains for consideration is that of decorative motifs. Half-way between portraits and icons on the one side and pure decoration on the other are the *dvārapālaka* figures at the sides of the gateways and entrances to temples and shrines. Perhaps it was the rule of old to make effigies of conquered rulers serve as gate-keepers in palaces and temples as suggested by certain epigraphic references of early date. But soon this gave place to the habit of employing conventional figures of semi-divine beings, at first with two arms, then with four, in that capacity. The symbols borne in the hands of the *dvārapālakas* differed with the cult followed in the shrine or temple. *Dvārapālikās* (women door keepers) also came into vogue especially for shrines of goddesses, sometimes even more generally. A *dvārapāla* from Vijayālaya-Cōḷēśvaran (fig. 84) and a female door-keeper (?) from the Kamparharēśvara of Tribhuvanam (fig. 85) may be seen to mark the features of this type at the beginning and end of the period covered in this survey. Besides figural sculptures like the *dvārapālakas* and icons which adorn niches on the exterior of temple walls, many other features contributed to the decorative effects of the structure.

c35.s1.20.2

There were the mouldings often heavy and impressive in the larger temples broken by string courses of animals or birds, of which a good example of a frieze of alternating elephants, lions and *yāḷis* comes from Nārttāmalai (fig. 86). Other friezes of *yāḷis*, and *gaṇas* and birds below and above cornices have been noted in the descriptions of individual temples given above. The plinth was also often broken by niches (*devago-* ṣṭhas) of varying sizes and designs so as to produce a pleasant relief in the great length of uniform mouldings; three examples of these are reproduced from the last great Cōḻa temple of Tribhuvanam (XXXV, figs. 89-91), and attention may be invited to the absence of the lion pillar in one of them, to its presence in another, and to the presence besides the lion pillars of another larger set of rearing lions on either side of the entire niche in a third—an anticipation of many later developments under Vijayanagar. Flights of steps and the balustrades on their sides were other ornamental features of note. Two balustrade sides—one from Dārāśuram depicting a lion attacking an elephant in the open (fig. 87), and the other from Tribhuvanam showing a more usual pattern of the proboscis of *yāḷi* forming the top of the curved balustrade with sculptured scenes beneath (fig. 88) are fairly typical.

c35.s1.20.3

On the balustrade of the steps leading to the southern entrance to the transept in the Tanjore temple we have panels of Buddha's enlightenment used as decoration, which is interesting in more ways than one. Lastly, there are panels of purāṇic stories figuring as decoration at suitable places on the plinth, especially on the broad flat member which in large temples is about a foot or foot and a half in height, and provides an admirable field in the open circumambulatory for devotees to study them at leisure and be pleased and instructed. The *Rāmāyaṇa*, *Mahābhārata*, Śiva's sports, the lives of the Saints and other edifying topics furnish the subject-matter for this type of sculpture. Ranking very high as skilful works of art, these narrative panels are not without their own interest. One of the Rāmāyaṇa scenes from Tribhuvanam (fig. 97) shows Rāvaṇa with many hands and heads carrying off Sītā in his chariot, and the bird Jaṭāyu which fought Rāvaṇa, was mortally wounded in the fight, and gave up his ghost later. Much smaller panels, in positions not so readily accessible to the devotee unless he cares particularly to look for them, sometimes covering small rectangles of about six inches by four, are tucked away underneath the pilasters of the exterior of the *garbhagṛha* in some temples, e.g. Nāgēśvara, Puḷḷamangai, Puñjai. From the Rāmāyaṇa panels of Nāgēśvara, are reproduced, (1) Agni presenting *pāyasa* to king Daśaratha, (fig.

c35.s1.20.4

92), (2) the King distributing the *pāyasa* to the queens, (fig. 93), (3) birth of Rāma, (fig. 94), (4) his fight with Tāṭakā, (fig. 95) and (5) Hanumān's meeting with Rāvaṇa in Durbar (fig. 96).

c35.s1.21

From the Puñjai temple we get the Varāha-avatāra panel, (fig. 98) and that of Kṛṣṇa sucking milk out of Pūtana's breast along with her life (fig. 99). The Periya-Purāṇam sculptures on the Dārāśuram temple are well known. Above all there are many interesting patterns, floral and vegetal, sometimes even geometrical as in Muslim art, too numerous for detailed treatment here. In fact, Cōḷa architecture and sculpture have remained neglected fields, and comprehensive monographs on the best extant temples are an urgent desideratum.

Cites (5)