Home > Cultural Informatics > Visvarupa > The Visvarupa Iconographic Traditions - The Archaeological Reports of Professor T. S. Maxwell


[ Previous Page | Next Page ]

Vaikuṇṭha-Viśvarūpa Vol. IV


Stein: 78 X43 X 23

urspunglich 4 Arme

Bilakesvara-Tempel, Bilavar (Billaur), Jammu Division, Kashmir (Siwalik-Gebirge, Nordlich der Daman-i-Koh-Ebene)


The image was briefly noticed in a report on a tour of inspection made by Ram Chandra Kak from Basohli to Jammu published in 1933 ("Antiquities of Basohli and Ramnagar (Jammu and Kashmir State)", Indian Art and Letters NS. VII.2, 1933: 65-91; p.75). No further information on this sculpture has been published since 1933 to my knowledge, nor has a photograph of it appeared.

Ancient Vallapura

In his report Kak refers to Bilavar as Ballaur and to its ancient name as Ballapura. Aurel Stein (RT.2 Chapter IV Section 1, Political Topography, Frontiers of Ancient Kashmir: 432) names it more accurately as Vallapura, the modern version of which he transcribes as Ballavar, which accords fairly well with the local pronunciation of the present day (March 1994), Stein (loc. cit.) reports on the region as follows:

"To the west of Camba and south of Bhadravakasa lay the old chieftainship of VALLAPURA, the modern Ballavar. It rulers are repeatedly referred to in Kalhana's narrative. They retained their independence as petty hill-chiefs till the rise of the Jammu family early in this century, Ballavar was known also to Alberuni. / Of the political organization of the hill territories between Vallapura in the south-east and Rajapuri in the north-west we have no distinct information. The Hindu inhabitants of this tract including Ballavar call themselves now Dogras and their country Dugar. This name is traditionally derived from Skr. *Dvigarta. But this term is nowhere found in our historical texts and probably has been concocted in analogy of the ancient name Trigarta. The original form of the name seems to be Durgara. / It is very probably that the region of the lower and middle hills between the limits indicated was already in old times divided into a number of small chiefships. Of these some eleven seem to have existed up to the extension of the Sikh power into the Panjab Kohistan. They were all absorbed in the growing state of Jammu which was originally one of them. / ...these small hill-chiefs of limited territory but ancient descent [Thakkuras] ..."

In Kalhana's text, the Hill-Rajas or Thakurs of Vallapura, and of the other hill-territories in the region, are presented as playing a significant role in Kashmiri politics from the 11th century onward. They commanded the difficult terrain which lay between the major powers of the north Indian plain and of kashmir; Alberuni mentions Vallapura as situated on the route leading specifically from Kannauj to kashmir (India I: 205). They were thus in the position to levy charges ("blackmail": Stein, loc. cit.) on travellers from India to Kashmir; moreover, those located on the frontiers of Kashmir appear habitually to have profited from payments made by both sides. Stein (RT 2:432) refers specifically in this connexion to the Khasa rulers of the Visalata region on the south side of the Banihal Pass in the Pir Panjal range; "temporarily the chiefs of the hills immediately south of the Pir Pantsal Range may have acknowledged the suzerainty of strong Kasmir rulers. But during the greater part of the period which is known to us from historical sources, they appear to have held their own and rather to have levied subsidies, i.e. blackmail, from the Kasmir rulers" (loc.cit., and see ibid., RT.VIII.2283 n.). Vallapura, farther to the south than Visalata, therefore lay well beyond the historical southern frontier of Kashmir.

The Hill-Rajas allied their houses with the royal house of Kashmir through marriage: in the 12th century the Kashmiri king Sussala (c. AD 1112-1120, 1121-1128) had a wife named Jajjala from Vallapura (RT.1: 287 n.220; RT.VIII. 1444). The Raja of Vallapura is mentioned as one of the hill-chiefs who presented themselves at the court of King Kalaśa of Kashmir in the winter of AD 1087-1088 (RT.VII 588). Intervention in the politics of Kashmir is indicated at RT.VIII.539 and VIII.547 sqq., where the king and yuvaraja of Vallapura join a league of hill-chiefs to support the pretender to the throne of Kashmir, Bhiksacara (r. AD 1120-1121), against the Kashmiri king Sussala (c. AD 1112-1120, 1121-1128; RT.1:287 n.220); another prince of Vallapura, the Rajaputra Brahmajajjala, is said to have supported King Sussala (RT.1:287 n. 220). The kings of Kashmir played a correspondingly powerful role in the local politics of the Hill-Rajas, who were subdued by force: in the 11th century King Ananta of Kashmir (c. AD 1028-1063), "who won victories over various kings, uprooted at Campa [Chamba] King Sala, and placed a new ruler on the throne. The king, while rashly making wanton inroads into foreign territories, often ran into danger. When on an expedition against Kalaśa, the son of Tukka, his troops had become worn out, Haladhara [the prime minister of Kashmir] extricated him from Vallapua by cunning" (RT.VII.218-220). Again, in the following century, King Jayasimha (AD 1128-1149) "uprooting King Vikramaraja at Vallapura put in his place King Gulhana, and did thus with other rulers in other [territories]" (RT.1:287 n.220: RT.VIII.2452). These references all concern the 11th and 12th centuries, but it is to be expected that despite the evident strength and independent spirit of the hill-chiefships a not dissimilar state of affairs had also obtained earlier, during the 9th and 10th centuries, under more powerful Kashmiri regimes. The goodwill or obedience of the Hill-Rajas must always have been a desideratum for the kings of Kashmir, in order to regulate contact between themselves and the powers of the north Indian plains.

The Bilakesvara temple

The temple at Vallapura is today (1994) called the Bilakesvara or Bilvakesvara and is dedicated to Śiva (a Śivaliṅga occupies the centre of the sanctum floor). At the time of Kak's visit in the 1930s, however, it was known as the temple of Harihara.

Kak lists no image of this combined deity in his report on the contents of the temple (unless the confused account of No.5 in his list refers to this and some other piece, which he took to be a single sculpture), but in March 1994 I found a large damaged image of three- (or four-) headed Harihara at the back of the sanctum (see No.06 in this Report) with a Varāha profile emanating from the left side of the head and a Bhairava face from the right, indicating that it is a combined form of multiheaded Śiva and Vaikuṇṭha, an icon that was created in Kashmir.

Other remains found in the sanctum of the Bilakesvara temple show that Vallapura was the centre of a flourishing Viṣṇu cult. The two most indicative are:

1. Bhūdevī, with a fragment of a vanamālā above and Cakrapuruṣa to her left (63 x 60 x 29.5): clearly from the base of an image of Viṣṇu, probably Vaikuṇṭha.

2. Cakrapuruṣa, turned to his right; this is an image of some size (72 x 32 x 29) and clearly belonged to the plinth-figures of a very large Viṣṇu- or Vaikuṇṭha-image.

(Since Kak in his report gives only the briefest of descriptions of the pieces he found, makes speculative identifications, and omits all measurements, it is unfortunately not possible to correlate these two fragments with objects in his list.)

The Vaikuṇṭha image

The royal Vaikuṇṭha cult, which originated in Kashmir in the 9th century under the Utpala king Avantivarman (c. AD 855-883), spread southeastward throughout the hill region. This expansion is represented by consecration of its images in the centres of Thakur power in Chamba, Kulu and elsewhere from the Ravi to the Sutlej valleys. These local cult-images are clearly tokens of the Kashmiri political hold over the Hill-Rajas. The same applies to the present image in the capital of the Vallapura territory, in the valley of the River Ujh (a lower tributary of the Ravi, named by Kak [op. cit.73] as the Bini opposite Ballavar, not mentioned by Stein in his Notes or Ancient Geography).

In his report of 1933, Ram Chandra Kak briefly describes (without illustration), as his No.1, the sculpture that which I have numbered No.05 as follows:

"Viṣṇu in the round; lower half of his body missing; has three heads (man's, lion's, and boar's); wears an elaborately ornamented tiara, lozenge-shaped jewelled ear ornaments, a jewelled necklace, sacred thread, armlets, waistband, srivatsa and mandaramala (flower garland) - the latter is fragmentary; the halo round his head is partly broken, his forearms are missing. In style the fragment is very similar to the Kashmir sculptures of the ninth and tenth centuries" (loc. cit.).

The similarity should more reasonably be referred to the iconography, which is that of the classic Vaikuṇṭha image of Kashmir, rather than to the sculptural style, which is regional and clearly reflects the distance from the main centres of Kashmiri art at which it was made. That there existed a prolific local workshop in ancient Vallapura, with a strongly pronounced sculptural style of its own, is proved by a dozen more images in the sanctum of the Bilakesvara temple, including a Harihara (No.06 in this Report), also of Kashmir-derived iconography but in the Vallapura style. Avantipura, place of origin of the royal Vaikuṇṭha image, lies some 300 kilometres to the NNW, beyond the hazardous Pir Panjal range. The closest point to Vallapura at which a genuine Kashmiri sculpture has been found, is the extensive temple site of Karimchi (also Krimchi) in the hills above Udhampur, some 90 kilometres to the north-west; the sculpture in question is a damaged image of Vaikuṇṭha (No.08 in this Report), clearly brought there from the Kashmir Valley by way of the southern route over the Banihal Pass and through the Visalata area controlled by the Khasas, and installed in one of the karimchi shrines. This and probably other icons exported directly from Kashmir along the same route would have served as models for the sculptors of Vallapura, since the way from there to Karimchi, skirting the lower foothills and passing though the temple centre at Kaladhera and Manwal (see No.07 in this Report), would have presented none of the physical and political difficulties that a journey to Kashmir involved in the 10th and 11th centuries.

The image at Vallapura / Balavar is broken off below the hips, and all four arms and most of the nimbus on the left and lower right sides are missing. The broken vanamālā, which curves behind the shoulders and loops through the crook of the front arms, is of round cross-section and bears the scale-like representation of leaves which is typical of its representation in the art of Kashmir. The upavita is a plain triple cord which loops just below the floral clasp of the girdle. There is some carving near the centre of the chest (it is displaced somewhat to the left), below the necklace, which evidently represents the srivatsa mark, as suggested by Kak, but it is very indistinct due to the layers of oil which have been applied to the surface; it appears to have the form of a flower. The body ornaments consist chiefly of large round jewels or pearls with superimposed flowers. These form the main motif in the girdle, the keyūras (which consist of double rows of beads or pearls with a frontal blossom-shaped clasp), the large diamond- or lozenge-shaped ear-ornaments, and the broad necklace and the rim of the crown (both with blossoms at intervals). The chief part of the crown consists of the three upright disks (maulicakratraya, RT.V.231) which are characteristic of the crown of Kashmir, also inlaid with round jewels. Above them, against the upper part of the nimbus, as in early Kashmiri Vaikuṇṭha images of the 9th and 10th centuries, there hovers a lotus blossom. The flat plain nimbus, which may have extended downward below shoulder-level, had vertical sides, and the top curves inward to meet in a point at the apex, behind the lotus. The sculpture being placed against the side-wall of the sanctum, it was not possible to discover whether there is any carving (a fourth face, for example) on the back of the nimbus; the local people could provide no information, and indeed the pancayat was hostile to all examination of the sanctum contents.

The torso has an unnaturally symmetrical hour-glass shape, nipped in at the waist which is marked by a transverse line, and a cross centred on the navel was incised very formally, without understanding of its anatomical purpose. The neck is rather too tall and carved with the trirekha. The facial features are somewhat blurred, presumably by rubbing in the course of worship; the eyes are open, and a fringe of long curls, parted in the centre, appear on the brow below the crown rim, as in Kashmir.

Two very large animal profiles appear on the sides of the central head, that of the Lion to the right and of the Boar to the left. They project horizontally, well clear of the shoulders (immediately above the top edge of the vanamālā, extending from there up to the lower rim of the crown), as in Kashmiri Vaikuṇṭha icons, and not at an angle as in north Indian images. (The corresponding heads in the Kashmiri fragment at karimchi, No.08 in this Report, are severely damaged, but the remains clearly show that they were equally large and heavy.) The animal heads are not provided with separate crowns, but tall triangular elements lined with round gems appear above them, curiously inserted in front of the side-disks into the main crown rim, which is made to bulge correspondingly at that point. This is clearly a misunderstanding of the Kashmiri iconography, in which the side-heads wear an element of the triple-peaked crown (as in north India they wear the subsidiary karanadamukuta) coalesced with the royal triple disks, The animal faces are composed of large rounded masses, with slightly bulging eyes (complete with eyebrows), opened mouths revealing rows of teeth, and small pointed ears above those of the central Viṣṇu face. They are clearly differentiated, the Lion-head having bulging pads on the upper lip and a long flat nose, the Boar-head being elongated with a square jaw, raised snout, and long tusk extending up to the corner of the eye. Both have a short fringe of human curls high on the brow below the crown rim, and the long curve around the back of the jaw representing animal hair, as in the Kashmiri sculptures.

The dagger, which is usually worn at the right hip in Kashmiri Vaikuṇṭha icons (as is the case in No.08, the Kashmiri Vaikuṇṭha at Karimchi), is omitted from this Vallapura Version.


The image of Vaikuṇṭha at Vallapura / Bilavar, like its probable Kashmiri archetype at Karimchi, is to be dated late in the 10th century, which accords fairly well with kak's suggested date for the Vallapura temple, ca. AD 1000 (loc.cit., p.76).

Nr. 06: Vallapura (Bilavar)

Bilakesvara-Temple, Bilavar, Jammu, Kashmir


07. D. Ref.: Bilavar 43- # 57 DREI- oder VIERKOPFIGER HARIHARA

Stein: 134 x 56 x 44

ursprunglich 4 Arme

Bilakesvara temple, Bilavar (Billaur), Jammu Division, Kashmir (Siwalik-Gebirge, nordlich der Daman-i-Koh-Ebene)


The image stands against the back wall of the sanctum of the Bilakesvara temple in Vallapura / Bilavar, near the north-west corner. (On the territory of Vallapura and its history, see 06.1 in this Report.) It has suffered considerable damage, the face and the central disk of the crown, the right half of the nimbus, the forearms, most of the vanamālā, and the feet, all being broken off. The broken ends of the legs have been set into a rectangular pitha of Kashmiri type, with moulded sides and a water-spout at one end, which may well be the original pedestal for this image. The nimbus is pierced, forming a wide flat ring around the multiple heads of the deity, and rises to a point at the apex. It is there carved with a crescent supporting a motif, conceivably in imitation of a Buddhist maṇḍala, which consists of four disks or round petals, surrounding a central circle in the shape of a cross, against a square background. From this there curves down across the face of the nimbus a remarkable version of the patralata, almost resembling more an ornamental baldric than a leafy vine. Below it, rising from behind the shoulder of the deity and perhaps also deriving from Buddhist art, is a cloud-like billow of cloth.

The sculpture represents a male figure standing in a light ābhaṅga posture with the hips slightly deflected to the left and the right leg advanced. The anatomy is similar to that of No.05, but carved more decisively in imitation of the Kashmiri style; the belly is rounded and slightly protruberant, the top of the abdominal region defined by a sharp sloping line to either side under the chest. But the elongation, both of the trunk and of the neck, which was noted in No.05, is also in evidence here, contrasting with the heavy shoulders. The figure wears an adhoṃśuka or dhotī, the folds of cloth represented by curving doubled parallel lines on the thighs. Over this garment is slung an animal skin, apparently that of a lion or tiger, the head of the animal appearing on the right thigh. The corresponding covering on the left thigh spangled with six star-like flowers scattered on its surface. A long folded scarf curves down from the waist over these coverings, above the knee, with a subsidiary loop on the left side. A belt, studded with two rows of pearls or other round gems and held by a central floral clasp, is slung low on the hips and tilted to the right to conform to the ābhaṅga posture. The upvita cord loops in the centre, behind the top of the belt-clasp. A broad necklace with attached flower-ornaments, in form very similar to the belt, hangs on the chest, and a longer necklace, consisting of a thin cord or chain, supports a floral pendant hanging on the upper abdomen. The ear ornament on the right is a diamond-shaped plaque edged with pearls, as in No.06, but that left is a solid disk with a raised rim. The figure wore a vanamālā of the kind seen partly preserved on No.06, with scale-like leaves and superimposed blossoms; it does not appear along the top of the shoulders, however, but emerges lower down, from behind the upper arms, to curve briefly forward below the biceps and under the forearms, at which point it is broken off, on both sides, As there are no fragments of this garland on the remaining portions of the legs, it must have looped near the ankles. The face, as noted above, has been broken off; the missing features are crudely replaced today by cup-shaped silver eyes (inlaid with turquoise) and a flat silver nose-plate attached to the stone. The right half of a third eye seems to be clearly visible on the forehead, but rubbing and the application of sindhur make this observation uncertain. Long strands of coiled hair hang from the back of the head behind the ears and spread along the shoulders, some reaching as far as the keyūras on the upper arms. Curls also appear on the brow beneath the rim of the crown, parted at the centre. The crown itself is of the same triple-disk type seen in No.06, with small blossoms at the side of the head, as often in Kashmiri images. A large lotus flower, with finely carved filaments, appears above the damaged central disk of the crown.

Two side-faces emerge laterally from the central head, that on the left being mostly broken off. This damaged face was that of an animal, as can be seen by the small pointed ear, the line of bristling hair curving around the back of the jaw, and the whiskers appearing near the corner of the mouth. In all probability (see below) this was the profile of the Boar or Varāha incarnation of Viṣṇu, whose head appears in this position in the Vaikuṇṭha image (no.06). Here, however, its crown is fully represented, consisting of a jewelled headband supporting a central disk of the kind which appear in the crown of the principal head. The other side-head, on the right, is intact apart from the nose (it is also free of the oil which blackens the rest of the sculpture, showing the material to be a soft, brownish white stone), and clearly represents the Bhairava aspect of Śiva. This is not merely a profile but a complete face, as can be seen when the sculpture is viewed from the side, its demonic features quite obviously copied from Kashmiri iconography, The divided lower lip, the ends of curled moustaches, thick bushy eyebrows arching over furiously bulging eyes and ending in the twin knows of a frown above the root of the nose, are typical of the face of Bhairava, and of Kapila on the reverse of Vaikuṇṭha images, in the sculpture of kashmir. A large third eye divides the low forehead, its tip piercing the central parting in the vertical jatas. The crown consists of a coiled snake (its scales indistinguishable from the leaves of the vanamālā on the right arm below it) whose tail is carved at the side of the head so as to be visible from the front of the image, and whose three heads rise in the shape of a trident above the eye on the forehead, Above this, the chief part of the crown consists of five grinning skulls. This right side of the image, with the head of Bhairava and the lion-skin, represents Śiva, and the left side must represent Viṣṇu with the head of one of his animal-avatāras, most probably of the Varāha. (The difference in shape of the ear-ornaments, noted above, is typical of combined Indian images of this sort; in this case, however, the disk or large ring would normally be associated with Śiva and the diamond-shape with Viṣṇu, as in the Vaikuṇṭha image (No.06). Their reversal in this sculpture can probably be seen as an iconographical error on the part of the Vallapura sculptors, working as they were in relative isolation both from the north Indian plains and the Kashmir valley where the norms were set.) The piece corresponds iconographically to the Harihara from Kashmir in the Museum fur indische Kunst in Berlin (MfiK No.I.5835), consisting of the combined forms of Māheśvara Śiva and Vaikuṇṭha Viṣṇu. The back of the image, where the Berlin sculpture shows the split features of two krodharsis (kapila and Durvasas?), could not be examined for the reasons mentioned above in connexion with No.06.

The religious context in which this image was made at Vallapura is indicated by the other remains found in the sanctum. As shown in 06.1, in discussing the Vaikuṇṭha image there, the place was clearly the centre of a thriving Viṣṇu-cult. Further pieces, executed in the same style, show that a Śiva-cult existed alongside it. Kak mentions a "Liṅga, five-headed, very much defaced" (No.3 in his list) which was not there at the time of my examination of the sanctum in March 1994 (unless it is hidden beneath the metal covering of the Liṅga which is under worship), but two seated Ganesas that he describes (Kak Nos.2 and 11) were still to be seen, and the remains of a stone Nandi (Kak No. 13), stood in the forecourt. The Harihara image thus represented the coalescence of these two cults. (The identity of the three "finials", which are also clearly cult-objects, and which Kak describes under Nos.4 to 6, is uncertain.)


The sculpture represents Harihara, combining the two multiheaded forms of Śiva and Viṣṇu, māheśvara and Vaikuṇṭha, in the style of Vallapura but derived from the iconography of kashmir.

Vallapua, 10th century.

Nr. 07: Valapura (Bilavar)

BIlakesvara-Temple, Bilavar, Jammu, Kashmir



Steinrelief: Block 51.5 x 40.5 x 23.5


Wahrscheinlich aus Kaladhera; archaologisches Depot in Manwal (Nr. MWL-44 (12), Jammu Division, Kashmi (Siwalik-Gebirge, nordlich der Daman-i-Ko-Ebene)


This is a large, battered block from a temple wall, It is carved with a shallow niche based on a plain shelf and flanked by two round pilasters with rings, The relief within the niche, severely eroded and damaged, represents a three-headed bust. The main head has a necklace, pendant ear-ornaments, and a high crown, perhaps of the triple-disk type. From the right side of the face there projects a small lion-head with a bulging eye and open mouth, surmounted by a single-peaked crown. On the left, a small protruberance with a round eye and a tilted front evidently represented the Boar-head with a similar crown.

There is no new iconographical information to be gleaned from this battered relic. The fact that in later times, presumably as the cult declined and the iconography lost its importance, the side-heads of Vaikuṇṭha could be greatly reduced in size and deformed in order to fit them into too small a space, is already known from other late images.

It is however of considerable interest to note that Vaikuṇṭha, like multiheaded Śiva, could appear as a bust at least in the role of a subsidiary deity on temple walls, and possibly as a lintel-figure above a temple entrance signifying the identity of the chief deity to be worshipped within. No cult-image of this type has yet been discovered.


Vaikuṇṭha in the form of a bust, occurring as a subsidiary deity in a niche on a temple wall.

Manwal or kaladera, ca, 11th century.

Nr. 08: Kaladhera oder Manwal

Archaologisches Depot, Manwal, Jammu, Kashmir

(Devi Bhagavati-Templebezirk)


[ Previous Page | Next Page ]

 Report Volumes  | Viśvarūpa Multimedia Project

HomeSearchContact usIndex

[ Home | Search  | Contact UsIndex ]

[ Sutradhara | International Dialogue Unit | Multimedia Development ]

Copyright (c) T. S. Maxwell 1994, 1995