Pramāṇa · Search · About · API

The Chalukyan Architecture of the Kanarese Districts c8: Aihole and its Ancient Temples

The Chālukyan Architecture of the Kanarese Districts c8

Language
English

Contents

Aihole and its Ancient Temples

c8.1

CHANGES in the East have been slow where change has occurred at all. A stroll through the old village of Aihole would leave one under the impression that, at least in this quiet spot, no change, save that of decay, has altered the place or the sleepy life of its inhabitants in any material respect since the times of its prosperity, thirteen hundred years ago. Off the beaten track of modern progress, surrounded by its fields and pastures, it dozes in the sunshine beneath the sandstone crags of the hill of Mēguṭi. Begirt by the remnants of its old cyclopean walls, which have now more gaps than gates, it plods its peaceful way, year in and year out. The village kine, in straggling herds, are daily driven out in the morning to their grazing grounds, and are as regularly driven back again at dusk just as they have been all down through the ages.

c8.2

Within its crumbling walls, which enclose a space little more than five hundred yards across, it has more than thirty ancient stone temples, all jostling one another for room. They are embedded amongst closely-packed dwellings, and are, as a rule, desecrated and ruined, many having been converted into dwellings and cowsheds. Some few have been put to the vilest use imaginable by the villagers themselves, and these once the shrines of their country's gods! The principal streets have at one time been paved, but the flagstones, which have hitherto escaped the pilferer's hands, now, in their broken and dislocated state, rather tend to retard than promote progress. Around the village, without the walls, are many more dilapidated shrines, embosomed in prickly-pear jungle or standing uncared for in outlying fields. In addition to these structural buildings there are two rock-cut temples—one Jaina, another Brahmanical—and one shrine, half excavated and half built, in two storeys.¹

c8.3

¹ For further accounts of some of these temples see the *Archæological Survey of India Annual Report*, 1907-08, p. 205, and the *Report of the First Season's Operations of the Archæological Survey of Western India*, p. 37, et. seq.

c8.4

Aihole¹ has been chosen as the rendezvous and starting point of our pilgrimage round the Chālukyan temples of the Kanarese districts on account of its associations with the first

c8.5

¹ Lat. 16° 50', Long. 75° 57'. The ancient Ayyāvoḷe or Āryapura. See sketch plan, plate II, in which the temples are numbered.

c8.6

Fig. 4.—The old walls of the village of Aihole.

c8.7

settlement of the Chālukyas in the south, and for the fact that it contains the oldest dated structural temple in the country. It is situated in the Bādāmi *talukā* of the Bijāpur District, and about twelve miles due east of Bādāmi. With Paṭṭadakal, in the neighbourhood, these places formed a triad of ancient towns full of archæological wealth and hoary traditions.

The Temple of Mēguṭi

c8.s1.1

The dated temple of Mēguṭi (No. 1 on plan), which stands upon the brow of the hill overshadowing the village on the east, forms an early landmark in the history of the styles we are about to examine (Plates II—XXV). Unfortunately the temple is not complete having lost its tower and much of its lower structure. Although, perhaps, not the oldest temple here, it will be convenient to take it up first, and then arrange the others on either side of it as their peculiarities of construction may seem to demand. Upon its eastern side, engraved upon a fine large slab of stone let into the wall, is an inscription which records its erection in A.D. 634 by a certain Ravikīrti during the reign of the Western Chālukya king, Pulikēśin II.² But, before this, about A.D. 578, the Vaishṇava cave at Bādāmi was excavated by Kirtivarman I., a predecessor of Pulikēśin II. The inscription upon it records the fact that this king excavated the cave, and a grant of land was made for its upkeep upon that date. Though several of the other old temples at Aihole have inscriptions upon them, no others are dated; but the latest possible limit of age for these can be very approximately ascertained from the style of the characters used in the script.

c8.s1.2

² *Indian Antiquary*, VIII. p. 237.

c8.s1.3

The temple of Mēguṭi, as may be seen from the plan (Plate IV), is a long rectangular building consisting of two principal parts—the shrine, with its surroundings, and the forward

c8.s1.4

Fig. 5.—The temple of Mēguṭi from the south-east, Aihole

c8.s1.5

hall. The square shrine stands within a larger compartment whose outer walls completely envelop it, leaving a passage all round between them and the walls of the shrine. This circumscribing passage, which is but dimly lighted by small perforated stone windows in the walls, is divided up into small rooms by cross-walls and pilasters. Out before the shrine is a small hall or antechamber, and, beyond this again, is a larger hall which has been roughly rebuilt in later times upon the plinth of an older one.

c8.s1.6

Mēguṭi is a purely Dravidian structure.¹ The projections and returns, with their flanking pilasters, along the outer faces of the walls, are characteristic of that style. But the most marked feature, which leaves no doubt about it, is the stump of the storey above the shrine, which is the first storey of a Dravidian tower and was used as an upper shrine. The rubble built up upon the roof, seen in the photograph, appears to have been put up in very late times to form a barricade against missiles and to convert it into a small stronghold or watch-tower. Pot-tiles, such as are used for drains, are inserted, in great numbers, in these walls, all sloping outwards and downwards, to fire through at those attacking the place from the courtyard. The spaces between the pillars of the outer hall were also filled in with rubble for the same reason.

c8.s1.7

¹ I keep to the terms "Dravidian," "Northern" and "Chālukyan," in their application to the styles which will come under our consideration as they are defined in Fergusson's History of Indian and Eastern Architecture.

c8.s1.8

The interior of the temple is very dark. Within the shrine, seated upon a throne against the back wall, is a colossal Jina, while, lying in the passage round the shrine, on the west side, is a huge slab containing the image of a *devī* (Plate IV). This *devī* is, no doubt, the same as we find in Jaina temples in North Gujarāt and Rājputāna under the names of Ambikā, Ambā Bhavanī, Ambādevī, and Ambājī.¹ She is described in inscriptions as seated upon a lion vehicle and having two children, one upon each lap. In this image the children are held by attendants, one on either side. It is also a favourite image in the Jaina caves at Eḷūrā. The block, upon which the shrine image is carved, seems too large to have been brought in through the temple doorways: it was possibly placed in position before the shrine walls were built. The shrine doorway, as well as those of the antechamber, is very plain, having simple vertical mouldings up the sides, and a row of small *chaitya*-window ornaments along over the top.

c8.s1.9

¹ *List of Antiquarian Remains in the Bombay Presidency*, p. 355. *Epigraphia Indica*, IX, p. 156; *Ras Mala*, 1878 ed., Vol. I, p. 328; and *Journ. Bom. Br. R. A. S.*, XVIII, p. 26.

c8.s1.10

The inner shrine walls which rise through the roof, form the walls of an upper shrine, above which the *śikhara* or tower rose, just as may be seen in the Jaina temple at Paṭṭadakal (Plate LI). This upper shrine was a distinguishing feature in these early Jaina temples.

c8.s1.11

The outer walls are severely plain, being relieved by alternate square projections and recesses with pilastered corners. It was evidently intended to decorate the projecting panels with sculpture, for the faces of the central blocks at the corners have been left rough for this purpose, and the intermediate ones have shallow niches in which to fix loose sculptures. Two of the latter lie on the west side of the temple upon the ground. One of these represents two standing figures—a man and a woman. The man holds an offering which seems to be in a pot, while the woman stands upon his right. In the other, the man also holds an offering, but the woman is on his left. Placed in the two niches on either side or the back of the temple, they would be symmetrically arranged, that is, both looking towards or from the centre. A band of small figures, in panels, with some arabesque, runs round the plinth. Many of the little *chaitya*-window ornaments have been left in block, in an unfinished state. There is no doubt the temple was never completed, and, probably, the tower was not built.

c8.s1.12

‘Mēguṭi,’ or ‘Myaguṭi,’ is a corruption of *mēguḍi*, meaning ‘the upper temple,’ or ‘the temple that is above,’ and has been so named on account of its elevated position upon the hill. Enclosing a court around the temple is a fortified wall, following the edge of the hill round the north, west and south sides. The temple faces the north, the Jains not being particular upon this point.

c8.s1.13

Just below the brow of the hill, close under Mēguṭi, is a curious two-storied temple which is partly structural and partly excavated in the rock (Fig. 6). It consists of two long verandahs, one above the other, with a frontage of four heavy square pillars and two pilasters. Off the verandah of the upper storey is a long room and three shrines, cut in the rock; off the lower is the beginning of a shrine. In relief upon the centre of the upper verandah ceiling, in front of the shrine door, is the figure of a small seated *clothed* Jina with a triple umbrella above him.

c8.s1.14

Fig. 6.—The two-storeyed Jaina temple, Aihole

The Temple of Lāḍ Khān

c8.s2.1

The oldest temple at Aihole is probably that of Lāḍ Khān (Plates V—VIII). There is an inscription upon the front wall in characters of the 8th or 9th century A.D., which records a grant by a certain man to the Five-Hundred, the great body of the Chaturvēdis of the excellent capital of Āryapura.¹ The inscription, therefore, does not seem to have any connection with the temple other than that it was a convenient and permanent place to publish the grant.

c8.s2.2

¹ *Indian Antiquary*, VIII, 287.

c8.s2.3

There is no temple at Aihole, nor elsewhere that I know of, which impresses one so much with its cave-like character. Its general massiveness, the simplicity of its construction, and its plan and details, have much more in common with cave architecture than with that of later mediæval temples—and with cave architecture not of the latest. It is peculiarly wooden-looking in its construction, making allowance for the more massive nature of stone-work. This might seem a contradiction of the previous statement did we not know that cave architecture is itself, in great measure, derived from wooden forms. The walls are not walls in the ordinary sense of stone masonry, but are composed of posts, at intervals, joined up by screens and lattice windows. The flat roof, and its want of elevation, are cave-like characteristics. But, more than with anything else, one is struck with the great massive pillars, with their roll bracket-capitals, which evince a simpler and more dignified style than many of those more-decorated ones in Cave III at Bādāmi; and they, certainly, have the appearance of greater age than the latter.

c8.s2.4

The very unusual position of the shrine, which is placed *within* the great hall, against the back wall, has a very primitive air about it. At first sight it would seem that the building was simply a hall or *maṭha* in which, by an afterthought, a shrine was clumsily inserted to convert it into a temple. That this was not so, is clearly shown by the fact that in the similar temples of the Kont-Guḍi group, the beam, from pillar to pillar before the shrine, has been placed on a higher level in the original construction in order to admit of the lofty doorway being seen to its full height, and lion brackets project under the raised beam, one on each side, to further decorate the entrance to the shrine. (See Plate VIII.) After very close examination of these temples I am fully persuaded that these peculiar shrines are original. Moreover, there are, on the north and south sides of the temple, three perforated windows, the central one occupying the central bay of the walling and the other two the adjoining ones; but, in the back wall, which has the same arrangement of bays, between the pilasters, there are only two panels with windows (unfortunately not shown in the plan), those on either side of the central bay which was left unpierced on account of the shrine within. Taking all these points into consideration, and noting the total absence of anything like a *śikhara* or tower, I feel constrained to place this temple at an earlier date than that of Mēguṭi, but how much earlier it is not possible to say with any approximation to accuracy. The architecture of Cave III at Bādāmi is a distinct advance upon this in general style.

c8.s2.5

The pillars are the most characteristic feature of the temple, being remarkable for their great massiveness, which are more suited to support the heavy rock in a cave cutting than the lighter roof of a structural temple. The shafts are single heavy square blocks, without bases, from the floor to the bracket-capital, the last being a separate stone. The central four pillar shafts, without the bracket-capitals, are single stones measuring, each, 9′ 7″ by 2′ 5″ square. The roll brackets are thoroughly cave-like in character.

c8.s2.6

Though the decorative details upon this temple are spare, they are vigorous and expressive; they are suited to their position, and are not so crowded and meaningless as in many later buildings. The great lattice windows in the north and south sides are very chaste and effective, introduced, as they are, into otherwise severely plain walling with which they agree most admirably. In the west or back, and the front walls are pairs of circular windows, set in square frames, in which are radiating fish, forming as it were, the spokes of a wheel. This same fish design is found in the ceiling of Cave II at Bādāmi. But the most decorated part of the temple is the front porch, the pillars of which have life-sized images upon them in bold relief (Plate VII). On the extreme south pillar of the façade is a female figure standing on a tortoise, which is intended to represent the river goddess Yamunā. Probably, upon the corresponding pillar at the north end will be found Gaṅgā on her *makara*, but this has been obscured by the wall of an adjacent house which is built against it. Between the pillars is a low parapet wall having a seat running round on the inside. The outside of this wall is panelled and decorated with ornamental waterpots and a complicated knotted design. The ceilings are very plain. In the central bay of the hall ceiling is a small *Nāga* or serpent figure, with his tail rolled twice around him.

c8.s2.7

A remarkable feature is the absence of a tower or *śikhara*, and there is no indication of any intention to build one. This, again, points to a cave prototype, where it was impossible to have one. Over the central bay of the hall ceiling and roof rises a small square shrine facing east, and the roof of this shrine is covered with flat slabs which do not appear to have had a *śikhara* over them, in fact, a *śikhara* over this little shrine would have looked ridiculous, and it would not have been above the main shrine, below, where it would be required by the canons which directed the construction of later temples.

c8.s2.8

As to the dedication of the temple originally, there can be little doubt. Upon the dedicatory block, over the shrine door, is Garuḍa, the vehicle of Viṣṇu, and the doorway is flanked, upon either side, by a two-armed *dvārapāla* or doorkeeper, with a club. Upon the three sides of the small shrine, upon the roof, are images, two of which are but partly finished. That on the north side is a female under a tree; on the south is a four-armed Viṣṇu with his *chakra*, *śaṅkha* and *mālā* while, on the west or back of the shrine, is Sūrya. This last points to the probability of this small shrine having been dedicated to this deity, being placed upon the roof so that the rays of the rising sun could shine straight into the cell and on the image within, unimpeded by the intervening houses of the village. The temple, therefore, was Vaiṣṇava. We know that the early kings of the western Chālukya dynasty were of the Vaiṣṇava cult, since mostly all of their grants open with an invocation to Viṣṇu, and have his boar upon their seals. Within the shrine has been placed, in later times, a *liṅga*, and before it, in the centre of the hall, a large Nandi or bull.

c8.s2.9

The waterpot ornament, which is seen on the front of the porch, is a very favourite device in the decorative details of many of these old temples. That the idea of its use, in this way, is at all connected with the traditional origin of the Chālukyas from the waterpot is very doubtful.¹ It was not confined to Chālukyan work, for it is found repeatedly used in the Rāṣṭrakūṭa caves at Elūrā. Again, it is seen, in rather classic guise, with draperies, at the tops of the several very fine memorial stones at Borivali in Sālsette island near Bombay (see Plates CLIII and CLIV). Nor is it confined to any particular religious sect, for we find it both on Brahmanical and Jaina temples. The finials of temples are but waterpots—the *kalaśa*.

c8.s2.10

¹ For an account of this see *The Bombay Gazetteer*, Vol. I, Part II, page 339.

c8.s2.11

The images of Gaṅgā (the Ganges) and Yamunā (the Jumna) were also favourite objects in the earlier temples, especially in Northern India, being generally placed on either side of the shrine door. They first appear in very early temples at the top of the door jambs, and from there they were brought down to the bottom and finally disappeared altogether in later work.² Gaṅgā and Yamunā are frequently mentioned in old Kanarese inscriptions. Vijayāditya’s inscriptions, for instance, say that he conquered some great king of Northern India when he acquired the *pālidhvaja* banner and also the insignia of the signs of the rivers Gaṅgā and Yamunā. These were also Rāṣṭrakūṭa insignia.³

c8.s2.12

² There is a special shrine of the three river goddesses, Gaṅgā, Yamuna and Sarasvati, excavated in the rock in the courtyard of Kailāsa at Elūrā, in which were three very fine life-sized images, now much mutilated. See *History of Indian and Eastern Architecture*, Vol. I, p. 345.

c8.s2.13

³ *Bombay Gazetteer*, Vol. I, Part II, pp. 358, 368, 371, 387, 396 and 416.

c8.s2.14

The name " Lāḍ Khān," by which this temple is known, is merely the name of a Musalmān, who, not long ago, occupied the building as a residence. In the same way other old deserted buildings in the town are known only by the names of the parties who have been lately living in them. They have been so long in disuse as shrines that the very names of the deities, to which they were dedicated, have been forgotten. Most have been appropriated by the Liṅgāyats, who have introduced the liṅga and Nandi; but even these have been deserted.

Kont-Guḍi Temple

c8.s3.1

Our next temple is that known as Kont-Guḍi, which is of the same type as the last, and, like it, situated in the village (No. 4). " Kont-Guḍi," again, is not the original name of the temple, but has been applied to it because the last occupant of the building, as a residence, was the man whose privilege it was to carry the konta or triśūla of Śiva to the village boundary at the time of the annual Dasara festival. It was found in the same state in which he left it, the walls and pillars, within, being thickly coated with cowdung plaster which has covered up all the surface carving; and it is thick with soot, grime and cobwebs. It is a smaller building than Lāḍ Khān's, but, except for the pillars, it is hardly less massive in construction. Like the last it is square in plan, but has only four central pillars. The shrine is, if anything, more awkwardly contrived than Lāḍ Khān's, for it is formed against the east wall by screen walls from it to the nearest two pillars, and across between the pillars to form the front, in which is set the door frame. The shrine faces the west, although the entrance doorway to the hall is on the north side. The roof in the middle, above the four pillars, is flat, and it slopes away from this square space, all around, to the four walls. The shrine is, therefore, as at Lāḍ Khān's, under the sloping roof.

c8.s3.2

Upon the flat central portion of the roof have been laid a few courses of a śikhara, the first storey in fact, but this is probably an addition of a rather later period. The flat ceiling immediately under this, inside the temple, is an aṣṭadikpāla ceiling, that is, it is divided into nine panels, in the central one of which is Brahmā, while the surrounding eight contain images of the eight regents of the compass. This is a very favourite design of ceiling in later Chālukyan temples, but does not accord well with the heavy massive style of this temple (see Plate V). It was very probably put in at the same time that the śikhara or tower was commenced.

c8.s3.3

This first storey of the śikhara is about five feet high, and has an image niche on each of its four sides. That, on the north, seems to contain an image of the tāṇḍava of Śiva, but it is damaged; on the west is Varāha, the boar avatāra of Viṣṇu; on the south is Bhairava, while, on the east is the Vāmana avatāra of Viṣṇu. Vaiṣṇava sculptures occupying the east and west niches, together with Garuḍa presiding over the shrine door, point to the temple being, originally, Vaiṣṇava. A great heavy stone ladder, standing against the side of the temple, was very likely set up for the workmen who were engaged upon the building of the śikhara. There was no intention of constructing a shrine upon the roof, for no doorways are left.

c8.s3.4

Fig. 7.—A pillar in Kont-Guḍi, Aihole.

c8.s3.5

The four pillars in the temple are of one pattern, and shew a distinct advance on that of Lāḍ Khān, but are not so far advanced as those in Cave III at Bādāmi. The shafts are of the same type as those in the former temple, having no bases, and being square all the way up; but they are provided with a round squat cushion capital between the shafts and the brackets above. These are rather clumsy, and have not the more graceful proportions of those in the cave.¹ The surface decoration on the bands round them, of bead festoons and lozenge-shaped ornament, has been executed with a firmer and much more certain touch than that of Lāḍ Khān, where the prentice hand is apparent. There seems to have been no figure sculpture upon the original building.

c8.s3.6

¹ See Plate XXV of the First Report of the Archæological Survey of Western India (Burgess)

c8.s3.7

Within the temple, placed against a pillar, is an inscribed slab containing a record of Chāmuṇḍa II. (A.D. 1169), one of the Sinda chiefs. It is much worn, and little of it can be deciphered, but it begins with an invocation to Śiva. This must have been engraved about the time of Basava, the founder of the Liṅgāyat sect, and may possibly be connected with the conversion to their use of some of the old temples here.

c8.s3.8

By the side of Kont-Guḍi is another of these early temples, with the same unusual arrangement of the shrine (Plate VIII). The building is more like a long open verandah, facing east, with three rows of columns in its depth; in fact, it is, in plan, remarkably like a simple early Muḥammadan mosque. The shrine is just where the miḥrāb would have been, being formed by enclosing the space between the central pair of back pillars and the back wall, the doorway being between the pillars. The pillars in this temple are of the same type as those at Lāḍ Khān's, and are decorated with broad bands around the shafts, with medallions above them and vertical strips on the face of each connecting them with medallions below. These have been all richly chased with arabesque.

c8.s3.9

In each of the three central bays of the ceiling, before the shrine, is a finely carved image slab. That in the centre bears a representation of Śiva, that on the north of the centre has Viṣṇu upon Śesha, and in the southern bay is Brahmā (Fig. 8). The images of these three deities occur in the panels of the ceilings of several of the oldest temples here. Upon the two side walls of the shrine, outside, are other images, namely, a standing image of Viṣṇu on the north side, and an image of Bhairava on the south, while presiding upon the door lintel is Garuḍa. Upon the front of one of the pillars, in the front row, is an image of Narasiṃha, and there are dvārapālas upon the two pillars which flank the shrine door.

c8.s3.10

Kont-Guḍi, and this temple, are linked together by an intervening porch, which may or may not be coeval with the temples. Its masonry is not built into that of the temples, and it is not placed on the centre line of either temple. Moreover, there is no entrance into Kont-Guḍi from it; and, as the two temples are not on the same centre line, it is evident that they were erected independently of one another.

c8.s3.11

A few yards to the south of the last is another, rather smaller, built upon the same open verandah plan, and also facing east. The shrine is formed in precisely the same manner against the back wall. But there has been an attempt here to enclose an inner hall by a cross wall built along, from end to end, against the middle row of pillars, leaving a doorway in the middle opposite the shrine. This cuts off a long outer verandah from the rest of the building, and thus provides a porch and closed hall.

c8.s3.12

Above the shrine doorway, upon the dedicatory block, is the usual Garuḍa. This doorway is the most elaborate of those in these four old temples, and I am not so sure that the lower frame is not somewhat later work. It partakes very much of the character

c8.s3.13

Fig. 8.—Images in the ceiling of the temple beside Kont-Guḍi, Aihoḷe.

c8.s3.14

of Vijāyanagar work. The dvārapālas, below, are each four-armed, and, by their symbols, represent Śiva and Viṣṇu. Beyond these, on either side in the corner, sits a squat fat figure such as is seen at the bottom of one of the doorways at Kailāsa at Eḷūrā.¹ As in the case of the Kailāsa doorway, this one has Gaja-Lakshmi, not, however, upon the dedicatory block which holds Garuḍa, but away, high up above, on the beam. Both doorways have the heavy roll cornice, which is of an early type, heavier and deeper in this Aihoḷe example. This temple is known by the name of the Sarang-Guḍi after some late occupant.

c8.s3.15

¹ See The Eḷūrā Cave Temples, Burgess, Plate XXV, Fig. 1.

c8.s3.16

In front of the last is yet another old temple, but it is in the occupation of some private individual, and is closed against intruders.

c8.s3.17

These four early temples seem to stand apart by themselves both as regards age and type, and may safely be looked upon as the oldest erections in the place, preceding Mēguṭi, perhaps, by a century and a half.

c8.s3.18

Though they do not help us so much in our study of the development of what is known as the Chālukya style from the early Dravidian, yet we must take up a series of temples of about the same age as those we have been considering, which have been built after the Northern style, at least so far as their towers are concerned. At Aihoḷe and Paṭṭadakal the two styles overlap. Though the two types of towers, used in these early temples, differ entirely from each other, the rest of the buildings differs in no respect in their other features, and they were probably built by the same people at the same time. The builders of those days seem to have had no particular reason for selecting the one tower or the other: certainly, no temple, that we know for certain was originally Jaina, in this district, has a tower in the northern style, but Brahmanical temples, both Śaiva and Vaiṣṇava, are found with both. In the case of Jaina temples, the northern class of tower did not lend itself so readily to the formation of an upper shrine.

The Durga Temple

c8.s4.1

Of temples with northern śikharas there are two particularly fine ones, namely, the Durga temple at Aihoḷe (No. 3) and that of Pāpaṇātha at Paṭṭadakal, the former being, undoubtedly, the finest and most imposing temple in the village (Plates IX—XI). It is most unique in that it is built upon the lines of the apsidal cave chaitya of the Buddhists, the position of the shrine being that of the dāgaba; and, like its prototype, two rows of columns separate the body of the hall into a central nave and two side aisles. Stone, as a building material, at a time when constructive arching was unknown, determined that the roof should be flat and not arched, but sufficient likeness to the chaitya was obtained by making the central roof lofty and that of the side aisles low and sloping, the slope being the nearest approach to the half vault of the chaitya. The entablature, four feet deep, sculptured with friezes of figures, foliage and arabesque, built over the columns on either side in order to raise the height of the nave, reminds one forcibly of the same as seen in the cave chaityas above the pillars, such as that in cave XXVI at Ajanta.¹ Upon plate XI will be found a small scale plan of cave XIX at Ajanta, and it is interesting to compare the two plans. It will be seen that the rock-wall periphery of the cave corresponds to the outer pillared periphery of the temple; and the walls of the temple are just the columns of the cave all linked up. The dāgaba of the cave, upon the face of which is the principal object of worship—the image of Buddha—is represented by the shrine with its image. The structural necessities of the temple required the addition of these extra columns in the hall. There is no doubt that this type of structure descended from a wooden prototype, which was originally provided with a thatched roof. Such a roof has been reproduced, in brick and mortar, over the old Buddhist chaitya at Tēr in the Nizām's dominions.² The temple faces the east.

c8.s4.2

¹ See The Buddhist Cave Temples (Burgess), Plate XXXVI.

c8.s4.3

² See Annual Report, 1902-03, Archæological Survey of India, Plate XXIX.

c8.s4.4

The pillars in this temple are not quite so heavy and massive as they are in Lāḍ Khān's, though they are very simple in their general outline, being square blocks, without bases, surmounted by very plain bracket-capitals. Those in the front of the temple have been decorated with pairs of figures, in full relief, upon their outer sides. Some of the innermost columns in the porch are further enriched with bands and medallions of arabesque (Fig. 9). It will be noticed that all the male figures wear a high crown, and thus differ from that in the temple of Mallikārjuna at Paṭṭadakal, which has a very elaborate coiffure of curls (see Fig. 18). The latter looks more like a portrait statue than these, which are, probably, merely decorative figures.

c8.s4.5

Garuḍa presides above the shrine door, where he grasps, in his two hands, the serpent tails of nāgas whose upturned human bodies are a little way down the jambs on either side. The doorway is much after the style of those of the vihāras at Ajaṇṭā.

c8.s4.6

Though the image groups, representing Narasiṃha, Mahiṣāsura-mardanī, Varāha, Viṣṇu, Ardhanārī, and Śiva, in the niches around the walls in the outer verandah, are an impartial mixture of Śaiva and Vaiṣṇava subjects, it is very likely that the temple was originally dedicated to Vaiṣṇava worship—possibly to Sūrya-Nārāyaṇa. The central niche at the back, which, in a temple, generally contains the most important image next to that in the shrine, and which gives a clue to the principal object of worship, is empty.

c8.s4.7

Fig. 9.—A pillar in the Durga temple, Aihole.

c8.s4.8

There are two or three short inscriptions on and about the temple, but they are of little interest. On a stone in the basement are the words "Śrī-Jin-Alayan"—'Holy temple of Jina' in characters of the same date as Vijayāditya's inscription on a pillar in the porch of the Mahākūṭeśvara temple (A.D. 696—733). As there seems to be not the shadow of a doubt that the temple was originally Vaiṣṇava, this may mean that at that early date the temple was either in disuse, when some Jaina stranger, mistaking it for one of his own shrines scribbled this upon it, or that it had been temporarily in the possession of the Jains. In either case it shows that the temple was then of considerable age and had been deserted. On a pillar is another short inscription in letters of the 8th or 9th century, which reads "Śrī-Basa-mayya, the bhaṭṭa of Kisuvoḷal" (Paṭṭadakal). This may be connected with the time when these temples were restored for Śaiva worship. Yet another on the north wall of the south gateway records a grant to Āditya, a priest of the temple.¹

c8.s4.9

¹ *Indian Antiquary*, VIII, 285.

c8.s4.10

The northern type of tower is the most marked feature on this temple—so like the śikharas of the oldest temples in Orissa.² Unfortunately the upper part has fallen, but we have a more complete example in that of Huchchimalli-Guḍi next to be considered. There is another temple with a northern tower, near Lāḍ Khān's, which is complete with the crowning member and kalaśa or waterpot finial, though, perhaps, of a slightly later date. These śikharas are fully described in Fergusson's *Indian and Eastern Architecture*.³ What might be called the frontispiece of the tower is missing, but it is seen on that of Huchchimalli-Guḍi, the temple on the south of the village, and others, where it has a circular panel of carving representing Śiva dancing the *tāṇḍava*. This sculpture, which is found on the front of most of these old temples, was possibly placed there, in substitution for some other, when Śaiva worship became the fashion under later kings. It is thus found upon the great Śaiva temples of Virūpāksha and Mallikārjuna at Paṭṭadakal. In these earliest temples it certainly seems so, the sculpture being poor compared with that on the other parts. The slab often rests against the tower instead of being built in with it.

c8.s4.11

² See *History of Indian and Eastern Architecture*, II, 96 and 98.

c8.s4.12

³ Vol. I, p. 322.

c8.s4.13

At different points in the walls are perforated stone windows of curious patterns which serve the purpose of ventilation better than that of admitting light. On plate XI are shewn some examples of these, among which will be seen one in the familiar Greek fret or key pattern, but worked into the form of the favourite *śvastika*.

c8.s4.14

The temple stands within the ruins of a fort-like enclosure, which has probably given the present name to the temple—*durga* or 'fort'. It has nothing to do with Durgā, a name of Pārvatī.

Huchchimalli-Guḍi

c8.s5.1

The next five temples to be described are possibly older than the Durga temple, judging by the general style and more cyclopean-looking masonry, and certainly older than that of Mēguṭi. It requires but a glance at the Huchchimalli-Guḍi, which is situated in the fields a short distance to the north of the town, to be convinced that it can claim, on the score of its massive simplicity alone, a somewhat earlier date (No. 14). Here, again, the shrine is contained within the main body or hall of the temple, being placed towards its east end, so that a passage or *pradakshiṇā* is left around it. This arrangement of the shrine within the hall is found at the Elephanta Caves and in the Dumar Lena at Elūrā.

c8.s5.2

As already mentioned, this temple has a spire in the northern style, but its details are more archaic-looking than in that of the Durga temple. It is interesting to compare it with the temple of Paraśurāmēśvara at Bhuvanēśvara in Orissa.¹ There is a great similarity between the two. Dr. Fergusson, in his first edition of his *History of Indian and Eastern Architecture*, was inclined to place the tower and cella of the latter at about A.D. 450, the *maṇḍapa* or hall being a little later. In the revised edition Dr. Burgess has brought the date down to the seventh or eighth century, but I think this too low an estimate. The outline of the tower has an older look than the Aihoḷe one, though in its details, the latter is simpler and heavier-looking. But, in judging the age of these temples, local peculiarities have to be taken into account as much as other matters.

c8.s5.3

¹ *History of Indian and Eastern Architecture*, II, 96.

c8.s5.4

The plan of the hall of Huchchimalli-Guḍi, as in other temples of this class, does not give a clear idea of the arrangement of the interior of the temple. It is practically the same plan as that of hundreds of small mediæval temples which are met with of the eleventh, twelfth and thirteenth centuries, omitting their outer halls, where the four central pillars support a central square compartment of the ceiling, boxed in on all four sides by deep beams above the pillars, the marginal ceiling, outside and around this, being lower. But, in these older ones, instead of taking the four pillars as a central group, the two on either side should be taken as linked up with the end pilasters to form two continuous lines of columns from the front door to the shrine. The whole length of the ceiling, between these, is of the same height, which is made much loftier than the side ceilings by a deep running architrave over the beams above the pillars. This long ceiling is divided into bays by shallow cross-beams above the architraves. The hall is thus divided into a central nave and two side aisles. It will be noticed that there are no pilasters on the side walls opposite the pillars as in later temples such as those shewn on plate LXII. The side ceilings are formed of plain sloping slabs.

c8.s5.5

The interior of the temple is perfectly plain, excepting the shrine door which is decorated, which, like some of the cave doorways, has a band of little panels, with pairs of small figures in them, running down either side of the doorway. Some of these figures are rather suggestive of indecency. The interior of the shrine has been totally wrecked, portions of the *linga* and the paving of the floor lying about in great confusion. This has been due to treasure hunters who expected to find valuables buried beneath the image.

c8.s5.6

Between the two pillars, immediately in front of the shrine, a stone perforated screen has been inserted, with a doorway through it. It stands out 5 feet 8 inches in advance of the shrine and was no doubt intended to form a sort of ante-chamber to it. It is probably not an original feature.

c8.s5.7

The temple faces the west; and in front of the entrance doorway, which is fairly well decorated, stands a porch upon four disengaged pillars. In later temples the porch is built upon two disengaged pillars and two wall-pilasters. On either side of the porch, between the pillars, is a low seat, the outer back of which forms a parapet which is decorated with panels containing urns or waterpots as upon the porch of Lāḍ Khān's. In the porch ceiling is a group of sculpture representing Kārtikēya seated upon his peacock, with a lance in rest in his right hand and a flower in his left. Attendant figures, which surround him, make offerings, while a prostrate figure, with a shield and dagger or short sword, lies beneath the peacock's feet. There can be no mistake about the peacock here. It stands in profile, and thus shews its long tail to advantage, at the end of which the "eyes" are well delineated, while upon its head is the top-knot of feathers. But Kārtikēya, who sits side saddle, has only one head. Sometimes he is represented like this, in other cases with his six heads. Kārtikēya, Svāmi-Mahāsena, Shaḍānana, Skanda, Subramanya, or Shaṇmukha was the god of war, who led the early Chālukyas to victory as well as the Kadambas of Banavāsi.

c8.s5.8

All the pillars in the building are plain, square, heavy shafts with bracket-capitals, those in the hall having octagonal necks near the top of the shafts similar to those in the centre of the hall of Lāḍ Khān's (see Plate VIII).

c8.s5.9

Upon the front of the temple is an old Kanarese inscription which records a grant of oil to the priest of the temple by Vijayāditya in A.D. 708.¹

c8.s5.10

¹ *Indian Antiquary*, VIII, 284.

Temple in the Field to the South of the Village

c8.s6.1

In the field, under revenue No. 270, to the south-west of the village of Aihoḷe, stands another old temple (No. 9) of the same heavy massive style, but having its shrine as a separate compartment from the hall (Plates XIII—XVIII). Otherwise the general characteristics are the same. The outer walls are very similar to those of the last, having the same mouldings and the same plain surface between the plinth and the cornice. The śikhara, however, takes more of the older curve of the Bhuvanēśvara temple, that of Huchchimalli-Guḍi being considerably straighter in outline.

c8.s6.2

An advance upon the last temple are the niches round the shrine walls, one on each face. These contain the Vaiṣṇava sculpture Narasimha, in the back or principal niche, and the Śaiva image of Bhairava in the north niche. The image from the south niche has been removed. The temple faces the east, at which end is the entrance porch, supported upon four massive pillars. It will be seen that the two back pillars are placed against the front wall and touching it, in this respect differing from Lāḍ Khān and Huchchimalli-Guḍi where they stand away from the walls. They are still separate pillars, however, and are not pilasters as in later temples.

c8.s6.3

These pillars are adorned with pairs of figures standing out in relief, as on the pillars of the porch of Lāḍ Khān. On a great slab forming the ceiling of the porch is a group representing the tāṇḍava of Śiva. The great cornice of the porch is formed of a heavy roll moulding, quite a quadrant in section, the underneath side of which is ribbed like the eaves of Kailāsa at Ēlūrā; and, like the latter, the centre and corners are decorated with rich flowing arabesque in low relief.

c8.s6.4

Fig. 10.—Interior of temple No. 9, Aihoḷe.

c8.s6.5

The hall of the temple is divided, longitudinally, into a central nave and two side aisles, as in the last two temples described, the central roof being raised, by a deep sculptured, entablature, considerably above the lower sloping roof of the sides. There are two freestanding pillars on either side, which, with the adjacent pilasters, support the running architrave. Cross slabs, laid over from one entablature to the other, close in the roof, the under surfaces of them being carved. But the three slabs which occupied the three middle bays, one in each, are missing, and the remainder have been pushed up together towards the shrine end. These slabs have been identified, however, with three found in private possession at Dhārwār. Their measurements correspond to the spaces in the respective bays, and their carvings with similar ones in the same positions in two other old temples at Aihoḷe of the same class. The stone is the same, being that of the sandstone of the neighbourhood, in which all these temples are built. It was known that these slabs were carried away, with a fourth one, from the Bādāmi neighbourhood over forty years ago by a judge at Dhārwār, and were set up in his garden there, where I found them. They are illustrated on Plate XVII. At that time, all the old temples in the village, which were not in use as such, were occupied by people as dwellings, and could not well be rifled like this one, lying deserted and ruined in the fields. In the other two similar temples, it is found that the sculpture of Brahmā is in the ceiling bay next the shrine, with Śiva in the central one. This was probably the order here, Viṣṇu on Śeṣa, the narrowest slab, being nearest the entrance doorway.

c8.s6.6

On the face of the entablature and cross beams separating the three bays from one another, are small groups of sculptures. Those in the first bay from the door represent, generally, the avatāras of Viṣṇu, while those in the central bay show the aṣṭadikpālas or ‘Regents of the eight points of the compass’ of which two, Varuṇa and Nirṛiti, are missing, the beam being broken off. Agni and Kuvera have interchanged places. A description of these Regents will be given later on in connection with their frequent occurrence in mediæval temples. There are no such sculptures in the bay nearest the shrine; this was, probably, never finished.

c8.s6.7

Running the whole length of the entablature, on either side, are two richly sculptured foliage bands. The lower, with its undulating stem and leaves, has been a very favourite form of ornamental band all down through the centuries of temple building. The upper one, in the form in which we see it here, was only used in very early work and may be seen largely used on the great tope at Sāñchi. There is a very natural look about the unconventional soft vegetable forms of the stalks, leaves and flowers in these bands which is lost in the harder and more geometrically-cut curves of later ornament. Above this runs a frieze of little dwarf figures, such as are seen in Śiva’s gaṇa in the caves, in all sorts of eccentric and grotesque attitudes. They are possessed of great verve and abandon, so different to the stiff stereotyped figures of later times (Plate XVIII).

c8.s6.8

The pillars are much plainer than those of Lāḍ Khān, and more like those in the temple beside the Kont-Guḍi, only that the band and medallions have not been carved with arabesque as they are there. They were possibly left unfinished (Fig. 10). What these could have been, had they been fully decorated, is seen in the case of the two middle pillars in the lower storey of the Tin Thal at Elūrā, which are most richly carved, on the plain square shaft, while all the rest in the cave are simple squared blocks without a line of relief upon them.¹

c8.s6.9

¹ Cave Temples of India, Plate LVIII.

c8.s6.10

Upon the dedicatory block above the shrine door is Garuḍa, who holds the ends of the tails of two Nāga figures which trail down the door mouldings to the bottom of the door on each side, where they reverse upwards again with human bodies, each crested by a three-hooded snake. There is a great dressed block of stone still lying in the otherwise empty shrine which has the appearance of having been part of a square altar or seat on which an image had been placed; it is certainly not part of a *linga*. There is little doubt but that this temple was, like the others that have been described, originally Vaishṇava. There is a *dvārapāla* upon each pilaster on either side of the entrance to the shrine. That on the north side is Śiva, but the one on the south side has been mutilated; it was probably Vishṇu.

c8.s6.11

The method of roofing the halls of these temples is worthy of notice. The central nave is covered over with great flat slabs lying across from one architrave to the other, and the side aisles are covered by similar slabs which slope downward to the outer walls. As the stones are not so closely and smoothly dressed as to make the joints watertight, the joints are covered by long narrow stones. To prevent water running in under these to the joints, they are hollowed below, while the two adjacent edges of the roof slabs are grooved to receive them. The one thus lies over and fits into the other like the half pot tiles in ordinary native roofing.

c8.s6.12

Upon the front of the tower is the usual slab bearing the representation of the *tāṇḍava*, but it just lies loosely against the tower, and, like that of Huchchimalli-Guḍi is, doubtless, a later addition. On the north side of the temple is a short inscription in letters of the ninth or tenth century. On the south side is another in letters of the seventh or eighth century which seems to record the name of an architect. It runs thus: "Hail! There has not been, and there shall not be, in Jambudvīpa, any wise man, proficient in (the art of building) houses and temples equal to Narsoba." This is, possibly, the name of the man who had charge of the work of the restoration of the building when it was first converted to Śaiva worship.

c8.s6.13

Fig. 11.—*Dvārapāla* in temple No. 9, Aihoḷe.

c8.s6.14

There has been a temple on the south side of the last, but all that remains of it now is the shrine doorway and the seat or throne of the image. This building stood at right-angles to the last and faced the north. The doorway, as seen on Plate XIV, is decorated in a very simple fashion, but Gaja-Lakshmi occupies a block the whole width of the doorway above. On either side, below, is a great waterpot with foliage issuing from it and hanging over the sides. On the outer side of each waterpot is an elephant with its trunk coiled round the foliage, while, on the inner, squats a little fat Kuvera-like figure. On the front of the image seat is a well-sculptured peacock, with its crest and long tail, which would denote that the temple was dedicated to Sarasvatī. It thus occupies exactly the same position with regard to the other temple as the later temple of Sarasvatī does to the temple of Trikūṭēśvara at Gadag to be described further on.

Temple in Field No. 268, Near the Temple of Galagnātha

c8.s7.1

Another old temple of the same plan and style as the last, except that it has three pillars on each side of the nave instead of two, is situated in Survey No. 268 near the temple of Galagnātha (No. 10). The śikhara or tower is totally different, being far more archaic looking and decidedly clumsy. It gives one the idea of an early stage in the evolution of the northern style of tower, but this is not so, for, on examining it closely, it is found to be nothing more than a repetition, layer upon layer, diminishing to the top, of the flat roof over the maṇḍapa or hall of these temples. Compare it with the edge of the roofing of Huchchimalli-Guḍi (Plate XIX). The same thing, more elaborated, is seen on the temples, of a later period, behind the Desai's house, in the village (Plate XXV), and on the temples on the north side of the lake at Bādāmi. It became an ordinary style of tower, and a still later example of it is that of the Jaina temple in the fort at Belgaum (Plate CXXXV).

c8.s7.2

The pillars of this temple are simple square shafts, with roll bracket-capitals, the rolls of which are bound together by broad bands or fillets (XX). The flat bands and medallions on the shafts of the pillars have been left uncarved. Above the shrine door is Garuḍa, but, in this case, he holds the hooded ends of the snakes instead of their tails. Down below, on either side, are Gaṅgā and Yamunā, each accompanied by two pairs of figures. Within the shrine is a small dilapidated looking liṅga set flat on the floor. It is certainly not original.

c8.s7.3

The architraves, above the pillars, are decorated with the scroll and lotus patterns, but there is no frieze of little figures as in the last temple. The main bays of the ceiling had, no doubt, the same sculptures, but only one of these, that of Brahmā, in a mutilated condition, next the shrine, now remains. Brahmā is seated, cross-legged, full to the front, upon his goose which is shown in profile,¹ with neck stretched out as if flying or crouching flat under the weight of Brahmā. Beside the pilaster, near the shrine on the south side, is placed a large image of a dvārapāla with four arms.

First 94 of 131 passages. Read on

Linked texts (12)