The Quest for the Pāśupata Weapon

The Gateway of the Mahādeva Temple at Madhyamikā (Nagarī)

In: Indo-Iranian Journal

This article offers a new iconographic reading of the sixth-century architrave of the gateway of the Mahādeva Temple at ancient Madhyamikā (Nagarī). It is argued that the eastern and western face of the architrave should be read in conjunction. The eastern face shows Śiva’s entry as a naked mendicant into the Devadāruvana, while the western face depicts the Kirātārjunīya, a story that is foreshadowed in the panels of the eastern face. The motif that ties both stories together is the Brahmaśiras or ‘Head of Brahmā,’ which is simultaneously the skull that forms Śiva’s begging bowl and the Pāśupata Weapon acquired by Arjuna. The theme of the winning of the Pāśupata Weapon may have had particular resonance for the Aulikara rulers in their troubled times.

Abstract

This article offers a new iconographic reading of the sixth-century architrave of the gateway of the Mahādeva Temple at ancient Madhyamikā (Nagarī). It is argued that the eastern and western face of the architrave should be read in conjunction. The eastern face shows Śiva’s entry as a naked mendicant into the Devadāruvana, while the western face depicts the Kirātārjunīya, a story that is foreshadowed in the panels of the eastern face. The motif that ties both stories together is the Brahmaśiras or ‘Head of Brahmā,’ which is simultaneously the skull that forms Śiva’s begging bowl and the Pāśupata Weapon acquired by Arjuna. The theme of the winning of the Pāśupata Weapon may have had particular resonance for the Aulikara rulers in their troubled times.

Introduction

The village of Nagarī is situated in the Chittorgarh District (Rajasthan), 11 km north of the famous fort. The present archaeological site lies to the south of the village on the right bank of the River Berach. The original citadel along the river extended further northwards and included the space presently occupied by the village (Fig. 1).

d1193850e126Figure 1

The citadel of the old town of MadhyamikāPhoto courtesy Google Earth Pro

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The remains of the Mahādeva Temple are found in the southern part of the archaeological site (Fig. 2).

These remains consist of a reconstructed brick platform surrounded by some stray pieces of architecture: parts of a dhvajastambha (pillar), pillar capitals having bull and lion, a pedestal, an architrave of a toraṇa, a beam with gavākṣas and makara, and a toraṇa post (Fig. 3).

The architrave and post have sculptured panels on both sides, indicating that they must have been part of a free standing gateway, as the following (partial) reconstruction may show (see Figs. 4–5).1

d1193850e157Figure 2

The site of the Mahādeva TemplePhoto courtesy Google Earth Pro

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e164Figure 3

Remains within the precincts of the Mahādeva Temple

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Before we discuss the iconography of the toraṇa, we will give a brief sketch of the historical setting of the site.

d1193850e176Figure 4

Toraṇa east face

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e183Figure 5

Toraṇa west face

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The History of the Nagarī Site

Inscriptions and Coins

Many coins are reported to have come from Nagarī. Bhandarkar found in his excavations sixteen legible punch-marked coins pertaining to the Śibi-janapada. The legend of these coins reads: majhimikāya śibijanapadasa.2 Bhandarkar (op. cit. 123 f.), following an earlier suggestion by Kielhorn, identified this Majhimikā ‘with the Madhyamikā mentioned by Patañjali as having been invested in his time by a Yavana king,’3 and takes it to be the old name of Nagarī. He translates the legend as ‘Coin of the Śibi janapada of the Madhyamikā country.’ This distinguishes the Śibis of Nagarī from those of the Punjab. Bhandarkar (and historians after him) had little doubt that Madhyamikā ‘was the old name of Nagarī and also the district around it.’4

Five inscriptions have been reportedly found in Nagarī and surroundings, one of them is the famous Ghoṣūṇḍī Well Stone Inscription,5 recording the establishment of a stone enclosure around the Nārāyaṇa Vāṭakā, which was dedicated to the two gods Saṃkarṣaṇa and Vāsudeva. The site of this Vāṭakā has been identified by Bhandarkar (1920, 128 ff.) with the so-called Elephant Pen of Akbar (Hāthī-bāḍā), ‘half a mile east of the village [Nagarī].’ Bhandarkar found here, among other things, a stone with grafitto reading: śrīviṣṇupādābhyāṃ in seventh-century characters.6

For our present purpose, the inscription reportedly found ‘not far from the shrine of Mahādeva’ is of greater relevance. It is dated 481 in the Kṛta (= Vikrama) Era, corresponding to AD 424. It records ‘the erection of a temple to Vishnu by the three Baniā brothers,’ Satyasūra, Śrīgauḍha and Dāsa, and thus testifies to a Viṣṇu temple, established at or near the site of the Mahādeva Temple during the reign of emperor Kumāragupta.7 At this time Western Mālava was under control of the feudatory Early Aulikara prince (nṛpati) Viśvavarman, who was ruling from Daśapura (Mandasor), c. 100 km to the south of Nagarī.8

Some fragments belonging to the Gupta period were discovered by Bhandarkar in the area: two capitals of pillars with addorsed lions and bulls (in the Kaṅkalī Mātā shrine in Nagarī village). Other pillars and images were found c. 4 km south-east of Nagarī at the shrine of Sāḍū Mātā.9

One more inscribed stone relevant to our present discussion was found in Chittorgarh in 1959, ‘while clearing debris in the fort area.’ This stone, containing two brief fragmentary, but related inscriptions, was published by Sircar and Gai in Epigraphia Indica 34 (1961–1962). The second inscription attests to the building of a temple, which was, in all likelihood, dedicated to Śiva, since the maṅgala verse speaks of ‘the one who hides the moon in the pile of his curling, tawny matted locks.’10 Its construction seems to have been commissioned by a rājasthānīya or viceroy who was ruling in Daśapura and Madhyamā. The (Śiva) temple is said to be located to the north of the (Viṣṇu) temple of Manorathasvāmin.11

Madhyamā has been identified with Madhyamikā, and the rājasthānīya, son of Viṣṇudatta and grandson of Varāha(dāsa), belonged to the powerful merchant (Naigama) family that kept the office of chief minister or viceroy under the Later Aulikara kings of Daśapura, in particular Prakāśadharman and Yaśodharman (between AD 510 and 533).12 The two Chittorgarh inscriptions belong to this period and record the importance of the Madhyamā/Madhyamikā province within the kingdom of Daśapura in the first four decades of the sixth century. They also testify to shifting religious affiliations amongst regional rulers: the Early Aulikaras, like their Gupta overlords, were Vaiṣṇavas; the Later Aulikaras, who came to power after the turbulent period that saw the fall of the Gupta Empire, were devotees of Śiva. It would seem that the archaeological remains in Nagarī reflect this revolution.

The fort of Chittorgarh dates from the post-Aulikara period, when Daśapura and Madhyamikā had lost their prominence. It is conceivable that the inscribed stone slab came from nearby Nagarī along with other materials, when the newly built fort provided safety to a successor state, possibly that of the Mori Rajputs (Mauryas) in the seventh century.13

Archaeology

Excavations at Nagarī were conducted by D.R. Bhandarkar in the ‘second decade of the [20th] century,’ and after him by K.V. Soundara Rajan in 1962–1963.14 The archaeological situation is summed up by R.P. Sharma in A. Ghosh’s Encyclopaedia of Indian Archaeology, volume II s.v. ‘Nagari.’

[…] Three periods were distinguished, the first two being anterior to the stone fortification and seemingly without baked-brick structures, though limestone structures were known. They had both red and grey wares; the occurrence of the NBPW was negligible, though associated red ware was available. The settlement seems to have originated in c. 400 B.C. […] Period III was marked by the presence of the Red Polished Ware. The other finds of the site include terracotta human and animal figures in Śuṅga and Gupta styles […] The fortification probably originated in Gupta times.

The main site of Bhandarkar’s excavation was around the small Mahādeva Temple. This temple itself is of recent date, but the installed liṅga may be old (Fig. 6).

The structure exposed in the excavations consists of a square brick platform (Fig. 7).

It will be seen from the plan that the east side was unlike the other sides of the platform and shows that it and its superstructure faced that direction. The maximum height preserved of its moulded walls is 4’ [122 cm] and is found in the west wall. This seems to be nearly half of the original height of the platform. Though the upper half of its walls has fallen down, it appears from the terracottas picked up from the débris, to have been covered with decorative tiles of at least three types, some of them probably arranged in string courses. One type is represented by what may be styled bird terracottas. […]

The centre of the platform was originally occupied by a superstructure, very little of which has now survived. It is 43’ 6” [13.30 m] square at the base. Immediately below each side of this square is a foundation wall, 6’ [1.80 m] thick. […] Of the superstructure only the lowermost moulding has been preserved, which is nearly two feet high.

Bhandarkar 1920, 135 f.

The original temple of the Gupta period may have been of brick, but, as observed by Bhandarkar (op. cit. 138), ‘to the second period of additions characterised by the introduction of stone work, or possibly to a period slightly later, belong the remains of a stone toraṇa exhumed in front of the mound.’ It is this toraṇa which is the subject of this article.

d1193850e336Figure 6

Liṅga of Mahādeva Temple

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e343Figure 7

Excavated brick platform in Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The Toraṇa of the Mahādeva Temple

In describing this gateway Joanna Williams (1982, 140 f.) observed that,

The most impressive carvings at Nagarī are the remains of a gigantic gateway or toraṇa that must belong to the early sixth century on the basis of its relationship to works from Mandasor. The decorative side face of the post bears fluted bands and medallions [Fig. 9], which elaborate upon forms found at Mukundara.

Passing through the temple gateway is entering a new world. The toraṇa as such epitomizes the rite of passage and the symbolism of this rite informs the iconography of the gateway (below, p. 255). It begins with the gatekeeper and his trident at the bottom of the gate post, eastern side, at the entry of the temple compound, which is oriented toward the east (Fig. 8).15

Just as Bhandarkar, Williams assumes that the fragment with candraśālā arches lying near the architrave is part of a second tier of the toraṇa (Fig. 10). She observed that ‘the top half of the double-candraśālā arches have diagonal projections from their lower corners, a detail mentioned as characteristic of pieces found in Deogarh.’16

The original Mahādeva Temple in Nagarī may have resembled the architectural structures that are depicted next to the candraśālās (Fig. 11).17 An āmalaka (1’ 9” high), like the one depicted, has been found in Nagarī village (Fig. 12).18

We shall approach the gateway from the east.

d1193850e400Figure 8

Gatekeeper with trident facing east

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e407Figure 9

Side face of toraṇa post

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e414Figure 10

Candraśālā arches in the second tier of toraṇa

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e421Figure 11

Miniature temple

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e429Figure 12

Āmalaka found in Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e436Figure 13

Architrave of the gateway, facing east

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e443Figure 14

Reconstruction of the architrave of the gateway, facing eastDrawing by Véronique Degroot

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The Eastern Face (E) of the Gateway Architrave

The full length of the architrave must have measured about 4 metres, but the southern-most panel has broken off at one-third and the concluding sculpture of a garland bearer is missing at that end (Fig. 13–14). The whole architrave consisted, in our reconstruction, of two garland bearers on either end and six panels in between. The middle of the architrave must have been between the third and the fourth panel, which we deduce from the turn of the torsion in the garland at the bottom of the crossbar in between the third and fourth panels. The height of the architrave is 65 cm, its width 38 cm. The architrave is broken at two places: in the first panel at the northern end and in the middle, between the third and fourth panels.19

The iconography of the eastern face of the architrave, seen by visitors when they entered the temple complex, has been discussed in Bakker 2004, 131–134, in particular its panel at the northern end, which, it was argued, contained a Dakṣiṇāmūrti. This interpretation still stands, but the person who is instructed in this mūrti has to be reconsidered.

At the time Bakker worked from a photograph kindly sent to him by Joanna Williams, which he re-published.20 Williams (1982, 141) had observed that: ‘The subjects of the reverse [i.e. eastern face] of this same crossbar remain to be identified.’ In a letter to Bakker she added to this that she wondered ‘whether the scene at the left end of this face might not represent the destruction of Dakṣa’s sacrifice.’21 Bakker accepted this suggestion and interpreted the Dakṣiṇāmūrti scene as representing the initiation of King Dakṣa in the Pāśupatavrata (Bakker 2004, 132 ff.).

The present revision of this interpretation has been prompted by a visit to the excavation site at Nagarī by Bakker and Bisschop and a study of the architrave in the field (15 January 2016). For it turned out that the panel at the southern end (E 1), although incomplete since it has broken off at one-third, actually contains significant information that had been invisible in the above-mentioned photograph. This evidence was spotted by Bisschop and debated in the field. It resulted in the acknowledgement that the earlier interpretation of the panels of the eastern face as depicting the Dakṣa myth needed to be reconsidered.

The incomplete panel at the southern end (E 1) deserves a full discussion (Fig. 15).

d1193850e486Figure 15

The first, incomplete panel of the eastern face of the architrave (E 1)

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Two features of this sculpture catch the eye and suggest a new identification. The pronounced ithyphallic character and the brush with peacock feathers, which is held in the ascetic’s left hand. It invites comparison to an image found at the bottom of the seventh-century doorpost built into the Deur Temple in Malhar, Chhattisgarh (Fig. 16).22

As emerges from this and other panels of the same doorjamb, the Malhar panel depicts Śiva’s entry as a naked ascetic into the Devadāruvana. The earliest account of this myth is found in the Skandapurāṇa. The description of his appearance there matches the present panel:

d1193850e506Figure 16

Lower-most panel in Malhar doorpostPhoto courtesy Natasja Bosma

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The Trident-bearer once entered that forest for the sake of alms. The sages saw the Lord of Gods there with his phallus erect.23

Just as in the Nagarī panel, Śiva carries the ascetic attribute of a brush and is ithyphallic, while his left hand holds a begging bowl ready to receive alms.24 These features do not figure in the Dakṣa myth.

In the Nagarī panel E 1 the protagonist carries something in his raised right hand. The elevated position suggests the object is significant, auspicious it would seem. It is held in such a way as if it is being shown to an audience that populated the missing two-thirds of the panel. This interpretation of the gesture is reinforced by the two following Nagarī panels (E 2 & 3), in which apparently the same object is raised, but in the left hand this time, in order to hold it away from the attackers, whereas the right hand now holds the ascetic brush. Moreover, the object seems to contain something in the first panel, whereas it appears empty in the second and third, a difference that might have significance and which we will discuss below. A deer in the incomplete first panel (E 1) and a clearly visible tree in the third one (E 3) suggest a forest setting (Figs. 15, 18).

d1193850e540Figure 17

Second panel from the left (E 2)

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e550Figure 18

Third panel from the left (E 3)

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

That this forest in the Nagarī panel is also the Devadāruvana, follows from another significant difference between the first panel on the one hand and the second and third panels on the other: the protagonist in the latter two is no longer ithyphallic, or phallic at all. This is not due to damage, but seems to be an intrinsic part of the sculpture. This feature proves, in our view, that the myth depicted on the eastern face of the Nagarī architrave is, like its Malhar counterpart, that of Śiva in the Pine Forest:

Out of envy the bewildered sages felled the great liṅga of the god, O Vyāsa, in the Pine Forest hermitage.25

A problem remains: what is the object that Śiva holds first in his raised right and then in his left hand? Erosion of the object in all three panels hampers its identification. The earlier view that saw in these panels the depiction of the Dakṣa myth naturally interpreted this object as the sacrificial cake (puroḍāś), seized from the sacrifice, but this could apply to the second and third panels only (E 2 & 3), not to the first one (E 1), in which the object is shown by Śiva.26

The hand gesture in this panel recalls images of goddesses of life and natural fertility holding a vase or pot with amṛta, the elixir of life, such as, for instance, the maiden on the stūpa railing found in Bhūteśvara (Mathurā),27 or the terracotta image of Gaṅgā found in Ahichchhatra (Figs. 19–20).28 This pūrṇakalaśa carried in the raised palm of the hand symbolizes vitality and natural growth. And although the Devadāruvana myth is all about life, growth and natural reproduction, depictions of this myth in which Śiva carries a vase are not known to us.29

d1193850e604Figure 19

Fertility goddess, Mathurā

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e611Figure 20

Gaṅgā, Ahichchhatra

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Another terracotta piece found in Ahichchhatra may be considered in this connection. It has been described by Agrawala and Banerjea as a panel depicting a Śiva Dakṣiṇāmūrti (Fig. 21),30 an identification that is, however, untenable in our understanding of a Dakṣiṇāmūrti (Bakker 2004). This panel shows a four-armed deity represented as an ascetic (matted hair, rosary), holding a pūrṇakalaśa with foliage in his back left hand. Apart from the general characteristics of the ascetic, none of Śiva’s usual iconographic markers is apparently present and the question is justified whether we are really concerned with an image of this god.31

We will return to the unidentified object in Śiva’s raised right/left hand later, but first we should have a look at the remaining panels of the architrave.

d1193850e633Figure 21

Terracotta panel, AhichchhatraPhoto courtesy Laxshmi Greaves

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The second and third panels of the eastern face of the cross-beam correspond closely to the second and third panels from below in the Malhar doorpost and thus reinforce the Devadāruvana interpretation (Figs. 22–25).

In both cases Śiva is attacked by furious sages and in both cases the latter realize their mistake, fall to his feet and cry forgiveness:

Thereupon the gods, beginning with Brahmā, and the sages striving for liberation realized that the terrible calamity was all caused by Śaṃkara. After praising him with various hymns, they propitiated Śaṃkara.32

d1193850e650Figure 22

Second panel from left (E 2), Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e660Figure 23

Second panel from bottom, MalharPhoto courtesy Natasja Bosma

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e667Figure 24

Third panel from left (E 3), Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e677Figure 25

Third panel from bottom, MalharPhoto courtesy Natasja Bosma

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

In the fourth and fifth panels the Nagarī and Malhar stories go separate ways. The Malhar panels refer to the Devadāruvana myth as told in the Saromāhātmya, in which Śiva in the form of an elephant plays a central role in re-installing the liṅga.33 We will focus on the Nagarī story.

When we pass from the third (E 3) to the fourth panel (E 4) we cross the middle of the architrave and therewith, it seems, we leave behind the Devadāruvana myth proper. The right half of the architrave is rather concerned with the consequences of the events told in the myth: Śiva is recognized as supreme god, who alone is capable of bestowing power (siddhi), grace (prasāda) and deliverance from all suffering (duḥkhānta). This is what the begging sages in the third panel (E 3) realize; the fourth panel (E 4) shows this new insight. God is depicted sitting under a tree, deep in concentration, as the master of yoga, Yogeśvara (Fig. 26).34

d1193850e724Figure 26

Fourth panel from the left (E 4), Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The reason that his Yogeśvara form and not the liṅga is chosen to represent him may have its origin in the fact that the designer of the architrave wanted to tell another story. Realization of one’s aim in this world and the acquisition of (yogic) power is achieved by following the path (sādhana) that he contemplates and is ready to communicate to the world of men. The four acolytes around Yogeśvara may intimate the spread of this sādhana over the earth, since they evoke the image of the four disciples:

And in the present age as well, when the Supreme Lord had seen the suffering in the world, He emitted four men (puruṣa) from His four faces and spoke to them, since it was His wish to bestow grace upon mankind: ‘You four should go to earth and become twice-born ascetics; O Masters of Yoga, return to me after you have led the brahmins to the highest station.’35

These puruṣas, the Skandapurāṇa (SP) continues, will become the four disciples (śiṣya) of Śiva, whose own incarnation on earth is known as Lāguḍi in the SP or Lakulīśa in later literature. This incarnation, characterised by his attribute the club (lakula), is not depicted on the architrave, but any guru pertaining to the four paraṃparās inaugurated by him, may be imagined to be implicitly involved in the last panel of the eastern face, since these ācāryas are believed to impersonate Śiva when they initiate and instruct their pupils.36 The four acolytes around Yogeśvara perform different acts of veneration and they do not seem all to be of the same standing. The one at the viewer’s lower right may be a householder, indicating that not only ascetics, but the whole world had come to recognize Śiva as supreme god.

The following, penultimate panel (E 5, Fig. 27) shows someone set on the path towards instruction in the doctrine, which is the subject of the last panel (E 6, Fig. 28).

The interpretation that conceived of these east-facing panels as depicting the myth of the destruction of Dakṣa’s sacrifice obviously saw in the central figure King Dakṣa (Bakker 2004). In our present understanding of the iconographic programme, however, this is no longer evident.

Of the three figures in a row, the one at the viewer’s right seems to stand within a architectural structure of which roof and pillars are still partly visible. In his left hand he carries something that looks like a quiver. If this is the case, we could expect that there would be a bow somewhere, but the panel is too much eroded to make it out. Twisted locks of hair fall from under his crown-like headgear. This feature recurs in the next panel in the kneeling figure and we infer that both depict the same person.

d1193850e800Figure 27

Fifth panel from the left (E 5), Nagarī

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The figure at the viewer’s left seems to be of a different standing than the two persons to the right. He is naked and of plump build. His headdress seems plain and his right hand makes the ‘do-not-fear’ gesture. His left hand is something of a puzzle. It would seem that this hand is raised and holds an object at head height, if this is not the raised right arm of the central figure.

The tallest figure in the middle is clearly a senior person. His left hand holds the knot of his dhotī. His right arm, if it is not raised, appears to lean on the left arm of his companion to the right, while his hand seems to hold something; a stick? It is clear that he wears a high crown with an ornament.

As a working hypothesis we conjecture that the figure carrying the quiver and who reappears in the next panel is Arjuna, who is being dispatched by his elder brother Yudhiṣṭhira, the figure in the middle, on the advise of Vyāsa, who stands on the (viewer’s) left side.

Mahābhārata 3.37–38 tells the story.37 Vyāsa instructs Yudhiṣṭhira to pass on to Arjuna a spell, a vidyā named Pratismṛti, also referred to as yogavidyā or brahman, which will enable him to acquire the necessary weapons from Indra, Rudra, Varuṇa, Kubera and Dharma.38 Yudhiṣṭhira explains to Arjuna that their enemies possess mastery in the use of all sorts of weapons and therefore, for the Pāṇḍavas to win the war, it is necessary to obtain superior weapons. He will initiate him into this secret knowledge (Upaniṣad) received from Vyāsa, thanks to which the whole world will become visible to him and through which he should seek the grace of the gods.39

After these words the lord King Dharma (i.e. Yudhiṣṭhira) taught him (i.e. Arjuna) that magic, when he was ritually consecrated and controlled in word, body, and thought; then the elder brother told his heroic brother to depart.40

In the Mahābhārata Indra is the first person that Arjuna visits on his quest, but the Śaiva world view, which conceived this temple and architrave, shifted the Vyāsa–Yudhiṣṭhira line of initiation onto Śiva, as we will see in the next panel. It is Śiva’s help that is sought first, since only instruction in the Pāśupata path will enable Arjuna to reach his goal. This is the lesson that the seers on earth had learned in the Devadāruvana, illustrated in the first three panels. We move on to the last one (E 6, Fig. 28).

This panel shows Śiva facing east, with Pārvatī to his left. The necklace and headdress are the same as that of Yogeśvara in the fourth panel (E 4). And the twisted locks of hair falling from under a crown-like headgear with ornament identifies the figure to Śiva’s right (dakṣiṇā), towards whom he slightly bends, as the same princely figure that we have tentatively identified as Arjuna in the preceding panel (E 5). Arjuna is on his knees now, facing north, receiving instruction. The scene is set on a mountain, in a forest. A powerful gaṇa-type of figure stands behind him, bending the hero’s arms behind his back by his front hands, or so it seems; his right back hand sticks up holding something above his head, whereas the contour of his raised left back arm is only vaguely visible. We will discuss below who this four-armed helper of Śiva could be.

d1193850e873Figure 28

Sixth panel from the left (E 6): a Dakṣiṇāmūrti

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

As we have argued earlier, this panel (E 6) shows a Dakṣiṇāmūrti in the original meaning of the term (Bakker 2004, 132 f.). The neophyte who is instructed or initiated in the Pāśupata doctrine is not Dakṣa, according to our present understanding, but the princely figure depicted in the penultimate panel E 5, who is there about to be dispatched by his elder brother and a saintly advisor.

The Devadāruvana myth that is told in the first three panels (E 1–3) does not immediately suggest a figure that would be the recipient of Śiva’s instruction, let alone the figure of Arjuna. Our tentative interpretation of this panel E 6, which sees in it Arjuna’s instruction in the Śaiva sādhana, needs further underpinning. This will be furnished by the iconographic programme of the western face of the architrave, which will make clear that we should read both sides of the architrave as one continuous story.

d1193850e902Figure 29

Architrave of the gateway, facing west

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e909Figure 30

Reconstruction of the architrave of the gateway, facing westDrawing by Véronique Degroot

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The Western Face (W) of the Gateway Architrave

The panels of the reverse side of the architrave, which faced west and were seen by visitors leaving the temple complex, have been indentified by Joanna Williams (1982, 141); they depict scenes from the Kirātārjunīya myth (Figs. 29–30). This interpretation has been accepted by Bakker 2004 and still stands. It accords well with the fact that the poet Bhāravi probably wrote his famous Kirātārjunīya at the court of the Later Aulikara king Yaśodharman, also known as Viṣṇuvardhana.41 Either he or, more likely, his father, King Prakāśadharman, was responsible for the building of this temple and its gateway.

The first or northern-most panel (W 1, Fig. 31) shows, according to Williams, ‘Nara and Nārāyaṇa seated in the wilderness.’ She compares it with the great Deogarh relief (Fig. 32).

The scene interpreted in this way may represent the dialogue between both seers as given in the Mahābhārata.42 The function of this piece of Vaiṣṇava philosophy, however, within the story told in the Nagarī architrave remains entirely obscure. In the Deogarh relief, the scene is overseen by Brahmā, who had sent both seers to earth to fight demons.

d1193850e945Figure 31

First panel on the western face (W 1)

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e955Figure 32

Deogarh, panel on the eastern face of the templePhoto courtesy Michael Willis

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

In the Nagarī panel there appears in the viewer’s right top corner a third figure. His left hand is visible and holds something rounded. Nothing points to Brahmā. And also unlike the Deogarh relief, Nara or Arjuna sits to the right of the ascetic figure with whom he has an argument and who seems to be two-armed. The scene is set in the mountains, where the rest of the Kirātārjunīya myth takes place. A lion is visible in a mountain cave at the foot.

Despite the striking similarities between the Deogarh and Nagarī panels, we are not convinced that the figure to the left of Arjuna is indeed Nārāyaṇa. Instead we think it more likely that the person Arjuna is speaking to is ‘the ascetic (tapasvin), blazing with the lustre of brahman, tawny, with matted hair and lean,’43 who is no other than Indra in disguise. This brahmin asks Arjuna why he has come in full armour and requests him to leave behind his bow (MBh 3.38.32–34). The bow may indeed be visible in the present panel to Arjuna’s right side.

After this brahmin failed to have Arjuna give up his resolve, he reveals his true identity and offers a boon. Arjuna replies:

I do not want wishes of worlds, or divinity, still less happiness, nor the sovereignty over all the gods, overlord of the Thirty! If I leave my brothers in the wilderness without avenging the feud, I shall find infamy in all the worlds for time without end.44

These are the proud words of the Kṣatriya, the quintessence of the Kirātārjunīya. Confronted with so much self-confidence, Indra can think of nothing better than to refer Arjuna to ‘the three-eyed, trident-bearing Lord of Beings, Śiva.’45

The similarity of the Nagarī panel with the one in Deogarh and other Nara–Nārāyaṇa representations is, however, not coincidence, but serves as a reminder to the onlooker that Arjuna is Nara. In Bhāravi’s Kirātārjunīya (12.33) Śiva explains to the seers that Arjuna is in fact an incarnation of Nara, a part of the Primaeval Man (Ādipuruṣa). And, he says, there is Acyuta. Nara and Acyuta have been asked by Brahmā to go among men to protect creation by killing demons,46 but ‘Nārāyaṇa’ as such does not figure in Bhāravi’s work.

Nara is said to have been created by Deva Nīlalohita in Skandapurāṇa (SP) 6. He had issued from the mirror image of Viṣṇu, who was reflected in the blood that the latter had himself donated to Nīlalohita’s begging bowl. This begging bowl is said in SP 6.1 to be Brahmā’s Head (brahmaṇaḥ śiraḥ), that is, his fifth one, which had been chopped off in SP 5 and was then used by Śiva/Nīlalohita on his rounds for alms.47

After Deva (i.e. Nīlalohita) has seen (Viṣṇu’s) reflection in the liquid (rasa) within the skull (kapāla), he issues forth (a) man (puruṣa) resembling Viṣṇu in strength. He says to him: You (shall) be immortal, exempt from old age and decay and invincible on the battle field; Viṣṇu will be your best friend, and you (shall) live with him performing divine tasks. Because you are born from the ‘waters’ (nārā) that rose from Viṣṇu’s body, you shall be called Nara (i.e. ‘man’) and be dear to him.48

Evidently these two sixth-century Śaiva, i.e. Pāśupata, texts play down Viṣṇu’s role; the story told in the Nagarī architrave might do just the same.

With Arjuna we move from Mount Indrakīla, where the meeting with Indra had taken place (MBh 3.38.30), to Mount Himavat (MBh 3.39.11), where the extreme tapas will take place, illustrated in the second panel (W 2, Fig. 33).

d1193850e1048Figure 33

Second panel from left (W 2): Arjuna’s penance

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The overall idea of this panel W 2 is clear: it shows Arjuna’s severe asceticism by standing on one leg in order to propitiate Śiva.49 Nārāyaṇa does not belong to this context, nor has Indra a role to play any longer. It is not immediately clear, however, who the figure to his left could be.

If we follow the Mahābhārata story, the seers, afraid of Arjuna’s ascetic powers, go to Mahādeva, who acknowledges Arjuna’s mortifications and gives his approval:

The Great Lord said: ‘Swiftly return in joyous spirit and unwearied whence you have come. I do know the intention that is lodged in his mind. He does not desire heaven, nor sovereignty, nor long life; this very day I shall accomplish what he desires.’50

We tentatively propose that it is this important moment of Śiva’s consent (and his resolve to put him to the test first) that has been depicted in this panel. It is also conceivable that the figure of Śiva here is a representation of Arjuna’s thought.

The figure seems to be four-armed, though only three are visible; his left back hand, if that is what it is, is raised and carries an object not unlike the mysterious object in the first three panels of the eastern face of the architrave (E 1–3). We postpone the discussion of this object and move on to the third panel (W 3), in which the test starts off.

The scene in this third panel from the left is without problems (Fig. 34).

d1193850e1091Figure 34

Third panel from left (W 3): who shot the boar?

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

It shows the boar and the Kirāta with his wife, a disguise that Śiva and Pārvatī had taken on. Arjuna and the hunter argue about whose arrow pierced the boar.51 The Kirāta is about to pull out the arrow, which enrages Arjuna and a fight is inevitable (Fig. 35).

Shouting again and again, they bored each other with arrows like poisonous snakes. Arjuna shot at the mountain man (Kirāta) a shower of arrows and Śaṃkara received them with a tranquil mind.

MBh 3.40.25–26; tr. van Buitenen
d1193850e1112Figure 35

Fourth panel from the left (W 4). Arjuna’s bow-and-arrow fight with the Kirāta.

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e1122Figure 36

Fifth panel from the left (W 5). Arjuna assails the Kirāta.

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Then, in the following panel (Fig. 36):

My arrows are gone. Who is this man who devours all my arrows? I shall attack him with the nock of my bow, as one attacks an elephant with the point of a spear, and send him to the domain of staff-bearing Yama!

MBh 3.40.37–38; tr. van Buitenen

We arrive at the dénouement in the sixth panel (W 6, Fig. 37), at the southern end of the western face of the architrave.

d1193850e1144Figure 37

Sixth panel from the left (W 6, southern end). Arjuna receives the Pāśupata Weapon

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

The Kirāta had been a form adopted by Śiva to test his devotee Arjuna. The latter is allowed a vision of the Great God. In the Mahābhārata version of the story Mahādeva reveals himself as the god carrying the trident, dwelling in the mountains together with Devī; in the Kirātārjunīya he assumes ‘his own form,’ smeared with ashes and ornamented with the crescent.52 Neither of these are visible to us, since, like its counterpart E 1, the panel is broken off at one-third. But we do see Arjuna, no longer as an ascetic, though, but as a princely figure, wearing his ornamental crown from which his locks fall down, very similar to his representation in the two panels at the end of the eastern face of the architrave (E 5 & 6). This is one of the elements that convinces us that the eastern and western faces of the architrave tell a continuous story, the story of Arjuna and his quest for the Pāśupata Weapon. The other element is this weapon itself.

Just as in the first panel of the obverse (E 1), we cannot but speculate on what has been depicted in the missing two-thirds of the reverse side (W 6). The motive and theme of the Kirātārjunīya story is Arjuna’s quest for the Pāśupata astra, the weapon which alone is powerful enough to counteract all other weapons and which leads its owner to victory over his foes. Arjuna asks Śiva for it:

If it pleases you to grant me my wish, Bull-bannered God, then I wish that divine weapon (astra), the dreadful Pāśupata Weapon, my lord, which is called Brahmā’s Head (brahmaśiras), gruesome (raudra), of terrible power, which at the horrible end of the Eon will destroy the entire world. With it I may burn down in battle the Dānavas and the Rākṣasas, the evil spirits and Piśācas, Gandharvas, and Snakes. From its mouth (yataḥ), when properly spelt (anumantritāḥ), issues forth thousand of tridents, awful-looking, clubs, and missiles like venomous snakes. With it I shall embattle Bhīṣma, Drona, and Kṛpa, and the always rough-spoken son of the sūta (Karṇa). This is my wish, my lord, who took Bhaga’s eyes, so by your grace I may go forth competent!

MBh 3.41.7–12; tr. van Buitenen

Śiva gives this powerful weapon to Arjuna,53 but warns him to use it cautiously, since launched against a person of insufficient strength it may have apocalyptic consequences. If the panels of the architrave are to make any sense, this gift of the Pāśupata Weapon must have been represented somehow. Aside from final release (duḥkhānta), it is the ultimate boon for the Māheśvaras.

The Pāśupata Weapon

If we work on the assumption that this boon must have been represented in the narrative of the architrave, we should answer the question how it has been depicted in its iconographic programme. The weapon (astra) is called ‘belonging to Rudra’ or ‘fierce’ (raudra), but more significantly: ‘Head of Brahmā.’ This may refer, as we have seen above (p. 243), to one of the Pāśupata key myths, namely the cutting off of the fifth head of Brahmā by Śiva/Nīlalohita. This myth was known in the last phase of the Mahābhārata composition, where Śiva is called Brahmaśiropaharta (sic).54

The full story is first told in the Skandapurāṇa.

Then, ordered by Parameśvara, this lord Nīlalohita, his (matted) hair coiled into a top-knot (kapardin), took the Head of Brahmā. After he had taken that shining head, he assumed a disguise, entered a playful state of yoga and started going around begging.55

This ‘Head of Brahmā’ is Śiva’s begging bowl and, according to the Mahābhārata, the Pāśupata Weapon seems to be just that, the Holy Grail of Saivism.56 We should therefore look for a (begging) bowl, and this leads us to the mysterious object that we encountered in the first three panels of the eastern face (E 1–3) and in the second one of the western (W 2).

We return to E 1, the first, incomplete panel of the eastern face. What we see in Śiva’s right, raised hand could agree to the shape of a begging bowl or skull. The function of the bowl/skull shown here is not so much the collecting of alms, rather than exposing it to an audience. This makes sense only if the begging bowl is more than just that. We consider it a rebus, representing the word brahmaśiras in visual (iconographic) form.

If our analysis is correct, it represents the ne plus ultra, the power that devotees who pursue the Pāśupata sādhana aspire to obtain in this world: the Pāśupata Weapon or ‘Head of Brahmā,’ the Śaiva equivalent of the pūrṇakalaśa, the source of life (nara) and death (antaka).57 This equivalence is further expressed by the fact that in the first panel the bowl seemed to have been filled, whereas the bowl appears empty in the following two panels, when it is withdrawn.

The token that was held out at the beginning in the first relief (E 1) was thus finally obtained by Arjuna in the addorsed panel at the end of the architrave’s narrative (W 6). It is the kṣatra or fighting spirit of Arjuna and his unswerving devotion to Śiva that won him this award at long last, the Pāśupata Weapon, and therewith the power to defeat all enemies. The Mahābhārata again describes the scene.

Hearing this, the Pārtha (Arjuna) hurriedly and attentively purified himself; and when he embraced the feet of the lord of the universe, the God said to him, ‘Now learn!’ Then he taught the best of the Pāṇḍavas about the missile (astra), along with the secrets of its return, this missile that is Death incarnate (mūrtimantam ivāntakam). […]

When the moment came, there was an outcry of conches, drums, and kettledrums by the thousands, and a huge quake occurred. The Gods and the Dānavas witnessed how that fiercely burning dreadful missile stood bodily deployed (mūrtimat) at the side of the boundlessly lustrous Pāṇḍava.58

And we believe the visitors to the Mahādeva Temple in Nagarī witnessed it too. Arjuna’s pose in the final panel (W 6) is one of vacillation. The ends of his cloth flutter. His knee is bent, as if he is taken aback. In front of him, we speculate, the Pāśupata Weapon may have stood in bodily form, next to Śiva in all his glory. What the weapon may have looked like, we know from another sculpture telling the same story.

Joanna Williams discusses ‘a pair of pillars found at Rajaona, 60 km east (and slightly north) of Rajgir’ (Bihar). Its subject matter, according to her, ‘resembles that of the Nagarī lintel.’ ‘The remaining face shows Śiva seated with Pārvatī on his mountain, whilst in front Arjuna kneels before the chubby four-armed personification of the Pāśupatāstra, the weapon that has been his goal throughout the story.’59 (Fig. 38).

d1193850e1332Figure 38

Rajaona, column: the presentation of the Pāśupata WeaponPhoto courtesy American Institute of Indian Studies. This fragment is presently located at Calcutta, Indian Museum: A 25106.

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Arjuna, however, is not on his knees in Nagarī; in that way he had been depicted when he was brought to Śiva for instruction by a four-armed figure in the last panel of the eastern face (E 6). We now conjecture that the four-armed figure in that panel E 6 may be the Pāśupata Weapon personified, not waiting on the Pāṇḍava hero, but conducing him to deference. The vigorous way in which this seems to be done could point to the forceful nature of the allegorical figure. The object that he seems to hold above his head may have been the skull, although the sculpture is too much worn to be certain about it. But if so, it would make this four-armed ‘chubby figure’ in the Dakṣiṇāmūrti panel a true āyudhapuruṣa.

A similar allegorical figure must, if our theory is correct, have been depicted in the missing part of the final panel (W 6), this time, however, ‘waiting on the great hero’ (upatasthe mahātmānaṃ).60

The Pāśupata Weapon thus seems to be the true subject of the Nagarī architrave. It is represented as a begging bowl made of Brahmā’s fifth head (Brahmaśiras), and in personified form as āyudhapuruṣa. In either form it appears throughout the iconographic programme: in the eastern-face panels E 1–3, and E 6 (embodied), and in the western-face panels W 2 and W 6 (embodiment conjectured). A close comparison may illustrate this point (Fig. 39).

The showing of the begging bowl or skull in the second panel of the western face (W 2) makes sense. It is not offered to Arjuna yet—as it had been to the ascetics in the first panel of the eastern face (E 1), who, however, rejected it in their ignorance—but raising it in the back left hand means that it is displayed as a boon that will be awarded, if Arjuna stands the test.

Having reached this point in our analysis, we venture the suggestion that within the first panel of the western face (W 1) the figure of which only a head and a left forearm are visible (to the viewer’s right, above the ascetic who is Indra in disguise), may be the Pāśupata Weapon, presenting itself in iconic, that is bodily, and in aniconic form, as the begging bowl in his left hand (Fig. 40).61

Concluding Observations

Like the composers of Epic and Puranic lore, the architect of the Mahādeva Temple and the designer of the iconographic programme of the toraṇa shared in the universe of myths. Like textual composers, the designer made his selection from this universe to compose his narrative. And like written texts, his visual narrative is inevitably an incomplete rendering of the stock of myths that circulated in his days. Like a textual author he extracted from this stock, adapted it to his needs and in this way contributed to its evolution. A textual source that tells exactly the same narrative as the architrave and that would thus be considered to be its source is not likely ever to turn up. The narrative of the architrave is the text, and as such it shows instances of intertextuality, thanks to which we can try to understand its message.

d1193850e1405Figure 39

Begging bowls in panels 1, 2, 3 (east), and in 2 (west)

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

d1193850e1412Figure 40

W 1 (detail of Fig. 31): the Pāśupata Weapon?

Citation: Indo-Iranian Journal 59, 3 (2016) ; 10.1163/15728536-05903011

Without doubt the designer was well acquainted with the Mahābhārata, whose story of the Kirātārjunīya was at the basis of his work.62 He could not yet know the Skandapurāṇa, since we believe that its composition only began in the second half of the sixth century, but the universe of Śaiva myths had substantially expanded from the days that the composition of the great Epic was coming to a close. The myth of Brahmā’s decapitation by Śiva, for instance, is only referred to through an epithet of the Great God in an Appendix passage of the epic, but the full story must have been around in Pāśupata circles before it was put down in the Skandapurāṇa. The same is true of the Devadāruvana myth. Familiarity with this lore is to be presupposed, not only for the designer, but for the general visitor at the time.

Arjuna’s quest for the divine Pāśupata Weapon and the Devadāruvana story are both mythical in nature and as such convey general, timeless truths. As all myths, however, they may serve as templates for human action and their depiction in art may function as historical allegory.

It is appealing therefore to read the architrave as a metaphor and to speculate on its connection with the rājasthānīya, the Aulikara viceroy who ruled in Madhyamā/Madhyamikā under Prakāśadharman and Yaśodharman (between AD 510 and 533), and who had, according to the Chittorgarh inscription (above, p. 221), commissioned the building of a Mahādeva temple. The educated contemporary may have seen in its iconographic programme evidence of his governor or king embracing the Śaiva religion by being instructed in its observances; other visitors may have seen only the template, the myth, an ambiguity inherent in Indian—and not only Indian—plastic art in general.

We have discovered the central theme of the Nagarī architrave, the one that connects both faces. It appears to be the recovery of the Pāśupata Weapon, which alone was believed to secure victory in difficult times. It helped the Pāṇḍavas to win back their kingdom, and it may have been thought to bring victory to the Later Aulikara kings Prakāśadharman and Yaśodharman of Daśapura and their allies in Madhyamikā, when they were locked in a deadly conflict with their archenemies, Toramāṇa and Mihirakula, who were commanding the dreaded armies of the Huns.

This theme also links the Devadāruvana myth to the Kirātārjunīya. In addition to Śiva’s stark naked and ithyphallic appearance, misunderstood and causing offence, it is the bowl made out of Brahmā’s head and symbolizing the Great God’s benevolence, that the seers failed to recognize in the first panel (E 1). Standing alone and being despised is, after all, the fundamental attitude of the Pāśupata ascetic.63 When the seers attack him in the next panel (E 2), he keeps it away from them.

In the earliest written version of the Devadāruvana myth that we possess,64 Skandapurāṇa 167.72–80, the ‘audience’ are the Vaikhānasa hermits. It is quite possible that only these or similar sages were depicted in the missing part of the first panel. The role of the wives of the sages may have come more to the fore in later versions, such as the narrative on the doorpost in Malhar, when the connotations of ‘Brahmaśiras’ either were no longer recognized, as ‘Brahmā’s Head’ had subsided into just a bowl to collect alms, or had come into bad grace due to its association with the Kāpālikas.

The exclusive focus in the Pine Forest myth on the liṅga may therefore have been the result of a development which gradually replaced the idea that underlies the Nagarī architrave. This would explain that the liṅga as object of worship does not feature in the architrave. Instead, it was the supreme good that only Śiva can bestow, object of his meditation in E 4, and symbolized in his Brahmaśiras attribute, that took centre stage.

The irony that will not have been lost on the designer of the architrave and the educated Pāśupata visitor of the temple thus seems to be that, if we follow the Skandapurāṇa, Arjuna at the end of his quest recovers the fons et origo from which he had once emerged, the Head of Brahmā. This cycle illustrates the superiority of the high god of the age, Mahādeva.

When they passed through the temple gateway, the king and his subjects were reminded of the major realities of Śiva’s World—asceticism, His benevolence, His revelation of the doctrine. After worshipping God, they saw, while they were leaving through the gate, the path which leads to His grace and which holds out to them the prospect of the acquisition of the highest good in this world, the Pāśupata Weapon—through steadfastness, self-control and bravery.

It was this faith, embraced by the Later Aulikara kings and their court, that was imparted to the visitors of the Mahādeva Temple in Madhyamikā, at the time that they came to be involved in the powerful World of the Great God.

Bibliographic Abbreviations

  1. KABhāravi
  2. CII IIICorpus Inscriptionum Indicarum
  3. EITAEncyclopedia of Indian Temple Architecture
  4. EIEpigraphia Indica
  5. EIAGhosh, A.
  6. IARIndian Archaeology—A Review
  7. MBhMahābhārata
  8. HCIMajumdar, R.C. et al. (eds.)
  9. PPāṇini
  10. PSPāśupatasūtra
  11. VāmPSaromāhātmya
  12. SISircar, D.C.
  13. SPSkandapurāṇa
  14. VDhPViṣṇudharmottarapurāṇa

References

Agrawala, R.C. 1987. Early Vaiṣṇava Icons from Rajasthan. In Parimoo, Ratan (ed.), Vaiṣṇavism in Indian Arts and Culture. New Delhi. pp. 328–334.

Agrawala, V.S. 1947–1948. Terracotta Figurines of Ahichchhatrā, District Bareilly, U.P. in: Ancient India 4 (1947–1948), 104–179.

Allan, John. 1936. Catalogue of the Coins of Ancient India. The British Museum, London (reprint 1967).

Bakker, Hans T. 2004. At the Right Side of the Teacher: Imagination, Imagery, and Image in Vedic and Śaiva Initiation. In Phyllis Granoff & Koichi Shinohara (eds.), Images in Asian Religions. Toronto 2004. pp. 117–148.

Bakker, Hans T. 2014. The World of the Skandapurāṇa. Northern India in the Sixth and Seventh Centuries. Brill, Leiden/Boston. Supplement to Groningen Oriental Studies.

Banerjea, J.N. 1956. The Development of Hindu Iconography. Delhi.

Bhandarkar, D.R. 1920. The Archæological Remains and Excavations at Nagari. In Archæological Survey of India, Memoirs 4, pp. 117–141.

Bhāravi (KA). 1933. The Kirātārjunīya of Bhāravi with the commentary (Ghaṇṭāpatha) of Mallinātha and various readings. Ed. Paṇḍit Durgāprasād and Kāśīnāth Pāṇḍurang Parab, revised by Wāsudev Laxmaṇ Śāstrī Paṇśīkar. Bombay.

Bisschop, Peter C. 2006. Early Śaivism and the Skandapurāṇa. Sects and Centres. Groningen. Groningen Oriental Studies XXI.

Corpus Inscriptionum Indicarum (CII III) 1888. Inscriptions of the Early Gupta Kings and their Successors. Edit. by J.F. Fleet, Calcutta.

Corpus Inscriptionum Indicarum (CII III) 1981. Inscriptions of the Early Gupta Kings. Revised by Devadatta Ramakrishna Bhandarkar. Edit. by Bahadurchand Chhabra & Govind Swamirao Gai. New Delhi.

Cecil, Elizabeth Ann. 2016. Mapping the Pāśupata Landscape. The Skandapurāṇa, Lakulīśa, and the Śaiva Imaginary in Early Medieval North India (6th–10th Century CE). Brown University, Providence (thesis).

Dalal, Chimanlal D. 1920. Gaṇakārikā of Ācārya Bhāsarvajña. With four appendices including the Kāravaṇa-Māhātmya. Edit. by Ch.D. Dalal. Baroda (reprint 1966). Gaekwad’s Oriental Series No. 15.

De Simini, Florinda. 2013. Ritual Aspects of Manuscript Transmission in Premodern India: a Historical Study of Vidyādāna Through Textual Sources. With a First Critical Edition and English Translation of Śivadharmottara’s Chapter Two ‘On The Gift of Knowledge’. Università degli Studi di Torino (thesis).

Donaldson, Thomas E. 1986. Bhikṣāṭanamūrti Images from Orissa. In Artibus Asiae 47 (1986), 51–66.

Encyclopedia of Indian Temple Architecture (EITA). 1988. North India. Foundations of North Indian Style (c. 250 B.C.A.D. 1100). Edited by M.W. Meister & M.A. Dhaky & Krishna Deva. Vol. II.1–2. Princeton/New Delhi.

Epigraphia Indica (EI). 1892–1969. A collection of inscriptions supplementary to the Corpus Inscriptionum Indicarum of the Archaeological Survey, translated by several oriental scholars; ed. by Jas. Burgess [et al.], Calcutta/Bombay. Vol. 1–37.

Ghosh, A. (EIA). 1989. An Encyclopaedia of Indian archaeology. Vol. I: Subjects. Vol. II: A Gazetteer of Explored and Excavated Sites in India. Delhi.

Greaves, Laxshmi Rose. 2015. Brick Foundations: North Indian Brick Temple Architecture and Terracotta Art of the Fourth to Sixth Centuries CE. Cardiff University, Cardiff (thesis).

Hara, Minoru. 2002. Pāśupata Studies. Edit. by Jun Takashima. Vienna.

Indian Archaeology—A Review (IAR). Archaeological Survey of India, New Delhi.

Lutzker, Mary-Ann. 1984. The Celebration of Arjuna. The Kiratarjuniya and the Arjunawiwaha in South and Southeast Asian Art. University of California, Berkeley (thesis).

Mahābhārata (MBh). 1927–1959. The Mahābhārata. For the first time critically edited by V.S. Suktankar and others. Poona. 19 vols.

Mahābhārata (MBh). 1975. The Mahābhārata. Translated and Edited by J.A.B. van Buitenen. Volume 2: 2 The Book of the Assembly Hall; 3 The Book of the Forest. Chicago and London.

Majumdar, R.C. et al. (eds.) (HCI). 1951–1977. The History and Culture of the Indian People. Bombay. 11 vols. Bharatiya Vidya Bhavan.

Pāṇini (P). 1962–1972. Śrībhagavatpatañjaliviracitaṃ Vyākaraṇamahābhāṣyam. The Vyākaraṇa-Mahābhāṣya of Patañjali, edited by F. Kielhorn. 3rd edition, revised and furnished with additional readings, references and select critical notes by K.V. Abhyankar. Poona. 5 vols.

Parlier-Renault, Edith. 2007. Temples de l’ Inde méridionale. La mise en scène des mythes. Paris.

Pāśupatasūtra (PS). The Pāśupatasūtra with the Pañcārthabhāṣya of Kauṇḍinya. Edit. by R. Ananthakrishna Sastri. Trivandrum 1940. Trivandrum Sanskrit Series CXLIII.

Salomon, Richard. 1989. New Inscriptional Evidence for the History of the Aulikaras of Mandasor. In IIJ 32 (1989), 1–36.

Saromāhātmya (VāmP). 1967. The Vāmana purāṇa. Crit. edit. by Anand Swarup Gupta. All-India Kashiraj Trust, Varanasi. pp. 164–292.

Sircar, D.C. (SI). Select Inscriptions bearing on Indian History and Civilization. Volume I, Calcutta 1942 [2nd ed.: Delhi/Madras 1965]. Volume II, Delhi. 1983.

Sircar, D.C. & G.S. Gai 1961–1962. Fragmentary Inscriptions from Chitorgarh. in: EI 34 (1961–1962), 53–58.

Skandapurāṇa (SP)

The Skandapurāṇa, Volume I. (SP I). 1998. Adhyāyas 1–25. Critically Edited with Prolegomena and English Synopsis by R. Adriaensen, H.T. Bakker & H. Isaacson. Groningen. Supplement to Groningen Oriental Studies.

The Skandapurāṇa, Volume II A. (SP II A). 2004. Adhyāyas 26–31.14: The Vārāṇasī Cycle. Critical Edition with an Introduction, English Synopsis & Philological and Historical Commentary by Hans T. Bakker & Harunaga Isaacson. Groningen. Supplement to Groningen Oriental Studies.

The Skandapurāṇa, Volume II B. (SP II B). 2014. Adhyāyas 31–52: The Vāhana and Naraka Cycles. Critical Edition with an Introduction and Annotated English Synopsis by Hans T. Bakker, Peter C. Bisschop, Yuko Yokochi, in cooperation with Nina Mirnig and Judit Törzsök. Brill, Leiden. Supplement to Groningen Oriental Studies.

The Skandapurāṇa, Volume III. (SP III). 2013. Adhyāyas 34.1–61, 53–69: The Vindhyavāsinī Cycle. Critical Edition with an Introduction & Annotated English Synopsis by Yuko Yokochi. Groningen/Leiden. Supplement to Groningen Oriental Studies.

[For edition of Adhyāya 167 see Bisschop 2006.] (SP S 167).

Skandapurāṇasya Ambikākhaṇḍaḥ, saṃpādakaḥ Kṛṣṇaprasāda Bhaṭṭarāī. Kathmandu 1988. Mahendraratnagranthamālā 2. (SP Bh ).

Viṣṇudharmottarapurāṇa (VDhP). Viṣṇudharmottarapurāṇa. [edit. by] Kṣemarāja Śrīkṛṣṇadāsa. Venkatesvara Press, Bombay V.S. 1969. [1912].

Viṣṇudharmottarapurāṇa (VDhP). 1994–1998 Viṣṇudharmottarapurāṇa. Third Khaṇḍa. Vol. 1: Text, Critical Notes etc. Critically edited by Priyabala Shah. Vol. 2: Introduction, Appendixes, Indexes etc. A study on a Sanskrit Text and Ancient Indian Arts. Vadodara. 2 vols. Gaekwad’s Oriental Series 130, 137.

Warder, A.K. 1989–1992. Indian Kāvya Literature. Delhi. 6 vols.

*This article is the outcome of our joint visit to the village of Nagarī in January 2016. The fieldwork formed part of our research within the ERC Synergy Project, Beyond Boundaries: Religion, Region, Language and the State. We gratefully acknowledge the comments we have received on an earlier version from Elizabeth Cecil and Yuko Yokochi, who both accompanied us on our site visit. Thanks are also due to Arlo Griffiths. We are much obliged to Véronique Degroot for preparing the drawings of the architraves illustrating this article. Unless stated otherwise, the photos were made by Hans Bakker. The authors take sole responsibility for the interpretation offered here.
1For another, well-preserved example of such a gateway from the same period and area, compare the tall post found at Khilchipura, now at Mandasor Fort. Joanna Williams (1982, 142) draws attention to the striking similarity; she considers the Khilchipura piece to be slightly later.
2Bhandarkar 1920, 122; Allan 1936, cxxiii f.
3Patañjali ad P. 3.2.111: aruṇad yavano madhyamikām.
4Bhandarkar 1920, 123 f.
5The Ghoṣūṇḍī village lies c. 6 km NE of Nagarī, but the inscribed stone came originally from the so-called Hāthī-bāḍā at Nagarī. There exist apparently three copies of this inscription, one still in situ. EI 16, 25–27; EI 22, 198–205; Sircar in SI I, 90 f.: ‘2nd half of the 1st cent. BC’.
6Bhandarkar 1920, 129; Agrawala 1987.
7Bhandarkar 1920, 121. The text of this inscription has not been published, although it is referred to in several publications. We only know about the basic contents of this inscription from Bhandarkar’s original description. According to Bhandarkar it was ‘deposited in the Rajputana Museum, Ajmer.’ Although the line of the inscription that mentions the Kṛta Era has been extensively discussed in CII III (1981), 192 ff., and the inscription clearly falls within the category of ‘Inscriptions of the Early Gupta Kings,’ the editors of the revised edition of Volume III of the CII (B. Chhabra & G.S. Gai) refrain from including and editing this inscription in their volume, nor do they give any more information about the inscription itself and its whereabouts. It shows again the limited value of this ‘revised edition.’
8CII III (1880), 72–78; SI I, 399–405; Salomon 1989, 19 f.
9Bhandarkar 1920, Plates XIV (a), XV (b).
10EI 34, 57: āpiṅgabhaṅgurajaṭācayalīnacandram.
11It is impossible to say whether this temple is the one erected by the Baniā brothers in the fifth century.
12Sircar & Gai in EI 34 (1961–1962), 53–58; Salomon 1989, 18.
13Cecil 2016, 116 f. HCI III, 161 f.
14Bhandarkar 1920; IAR 1962–1963, 19.
15That this is the eastern, that is entry side, follows from the fact that it has the figure of a Dvārapāla, and because it fits only the eastern face of the architrave. The latter face is the eastern one, because it contains a Dakṣiṇāmūrti, which would be without meaning if it faced west (Bakker 2004), and an image of Śiva in meditation (Yogeśvara), which should likewise face east. It also seems to follow from the iconographic programme of the architrave itself, which begins, as we will see, in the panels at the entrance, that is the eastern side, and continues with the panels at the exit, that is the western side.
16Williams 1982, 140 f. Cf. Williams 1982, Plate 202 (Deogarh).
17This feature/structure has been described by Michael Meister in EITA II.1, 143: “The upper cross-beam had makara heads at each end with architectural representations on the cross-lintel that show dvibhūma Phāṃsanā structures [i.e. ‘tiered, pyramidal roof-type’] faced with candraśālās and with small, square, stone platforms surporting globular āmalakas represented to either side of the upper bhūmi. Such units have been found at Bhūmarā and Sārnāth, are suggested at Sōndani and earlier at Bilsaḍ, are still utilized on structures in Surāṣṭra in the seventh century, and contribute conceptually to the formation of the Nāgara superstructure, where they become incorporated as bhūmikhaṇḍas.”
18Bhandarkar op. cit. 125, Plate XIV(b) (probably upside-down). Courtesy ASI.
19Bhandarkar 1920, 139 thinks that ‘enough has been recovered to show that each broader side was originally divided into nine panels [including garland bearers].’ Bhandarkar’s reconstruction in Plate XXIII shows that he conjectures that one complete panel plus garland bearer is missing at the southern end. In view of the turn of the torsion between the third and fourth panels (E 3–4, W 3–4), we think this is less likely. An argument against our assumption, as voiced by Véronique Degroot, is that the architrave panels show alternating indentation and protrusion: panels E 2/W 5, E 4/W 3 and (northern-most) E 6/W 1 are protruding and so, assuming symmetry, one might expect a southern-most panel to be protruding as well. Another argument, however, against a missing 7th panel, in addition to the change in torsion, seems to be that the breaking point in the southern-most panel (E 1/W 6) corresponds with the crack in the northern-most panel (E 6/W 1), the positions where the architrave rested on its two posts. The collapse of the toraṇa thus seems to have taken place by a break in the middle and at the two junctures. Six panels on each side makes sense according to the iconographical analysis offered in this article. If a 7th, lost panel was originally part of the architrave, the iconography of panels E 1 and W 6 that we conjecture may have been divided, broadly speaking, over two panels.
20Williams 1982, Plate 216; Bakker 2004, Figure 4.1.
21Letter dated 10 March 2001, quoted in Bakker 2004, 132.
22This Deur Temple in Malhar and its ancient doorposts is discussed in the forthcoming thesis of Natasja Bosma.
23SPS 167.74: bhikṣāhetor vanaṃ tat tu prāviśat kila śūladhṛk| sa dṛṣṭas tatra deveśo munibhiḥ stabdhamehanaḥ‖ For the Devadāruvana myth in the Skandapurāṇa etc. see Bisschop 2006, 195 f.
24Although the ascetic brush (sthalapavitra; picchikā in Jaina texts) is usually associated with Jainism, it was a mark of mendicants in general and of Pāśupatas as well. Cf. Pañcārthabhāṣya ad Pāśupatasūtra 1.6: tathā bhikṣos tridaṇḍamuṇḍakamaṇḍalukāṣāyavāsopajalapavitrasthalapavitrādi liṅgam. The sieve and brush are listed as donations to be given to the śivayogin in Śivadharmaśāstra 12.70 (De Simini 2013, 291). On the adoption of the principle of ahiṃsā by the Pāśupatas, see Hara 2002, 67–76. For Bhikṣāṭanamūrtis with staff (trident) with peacock feathers see Donaldson 1986, 56.
25SPS 167.75: īrṣayā munibhir liṅgaṃ tasya devasya tan mahat| vimūḍhaiḥ pātitaṃ vyāsa devadāruvanāśrame‖
26Cf. below, n. 57 on p. 250.
27Government Museum, Mathurā: Acc. No. 11.151.
28National Museum, Delhi: Acc. No. L. 2. Photo courtesy National Museum. http://www.nationalmuseumindia.gov.in/prodCollections.asp?pid=24&id=2&lk=dp2 (accessed 17-5-2016).
29In depictions of Śiva as a beggar he generally holds the begging bowl in his hand held low, but a strikingly similar sculpture from Alampur shows him displaying a filled object in his raised, left hand and holding a brush in his right. See Parlier-Renault 2007, 135, fig. 89.
30V.S. Agrawala in Ancient India 4, pp. 169–170. Banerjea 1956, 471 Pl. 7: “The four-armed god seated in the ardhaparyaṅka pose holds a rosary in the back right hand and a vase with foliage in the left; the lower right hand (broken) seems to have been either in the jñāna or vyākhyāna pose, the front left hand resting on the thigh. There are two figures on the left of the god, one male and the other female with hands in the namaskāra mudrā.”
31A new interpretation of this panel has recently been proposed by Laxshmi Greaves (2015, 308–313, 450–455). After a comparison of this plaque with the depiction of the sages Nara and Nārāyaṇa on the east face of the Gupta temple at Deogarh and a terracotta plaque from Bhitargaon, Greaves proposes to identify the four-armed figure in the Ahichchhatra panel with Nārāyaṇa. The nymph standing to his left, Greaves argues, is the Apsaras Urvaśī, and the panel depicts the myth told in the Viṣṇudharmottarapurāṇa (1.129.1–19, 3.35.1–18), which describes how the sage Nārāyaṇa created Urvaśī by drawing her outline in mango juice on his thigh: ‘It is apposite then that Sage Nārāyaṇa has been depicted with his left hand on his thigh, perhaps illustrating that the creation of Urvaśī has just taken place’ (Greaves op. cit. p. 454). We consider this interpretation interesting, but conclude that for the moment the Ahichchhatra panel, whether Nārāyaṇa or Śiva, cannot help us to decode the iconographic programme of the Nagarī architrave. One of our doubts concerns the upanayana cord. Could this be a snake, whose head appears at the deity’s left ear? If so, this would point to Śiva, rather than Nārāyaṇa. In the present state of conservation it is impossible to see if the deity is ithyphallic.
32SPS 167.77: brahmādayas tato devās te ca siddhā mumukṣavaḥ| jñātvā śaṃkarajaṃ sarvaṃ tad apāyaṃ sudāruṇam| saṃstutya vividhaiḥ stotraiḥ śaṃkaraṃ paryatoṣayan‖
33VmP Saromāhātmya 23.29–35. Cf. Donaldson 1986, pp. 53–54.
34The iconography of Śiva steeped in meditation underneath a tree is obviously derived from Buddhist examples.
35SPS 167.119–120:vartamāne kalau cāpi jñātvā duḥkhārditaṃ jagat|catvāraḥ puruṣān sṛṣṭvā svasmān mukhacatuṣṭayāt|provāca parameśāno lokānugrahalipsayā‖ 119 ‖yūyaṃ yāta mahīṃ sarve dvijā bhūtvā tapasvinaḥ|mām evaiṣyatha yogīśā nītvā viprān paraṃ padam‖ 120 ‖
36Bakker 2004, 124 f. Kauṇḍinya ad PS 1.1. Dalal 1920, p. 9: Ratnaṭīkā ad Gaṇakārikā 5c.
37Cf. Kirātārjunīya (KA) 3.10–29.
38MBh 3.37.25–28.
39MBh 3.38.9–10: kṛṣṇadvaipāyanāt tāta gṛhītopaniṣan mayā| tayā prayuktayā samyag jagat sarvaṃ prakāśate‖ 9 ‖ tena tvaṃ brahmaṇā tāta saṃyuktaḥ susamāhitaḥ| devatānāṃ yathākālaṃ prasādaṃ pratipālaya‖ 10 ‖
40MBh 3.38.14 (tr. van Buitenen): evam uktvā dharmarājas tam adhyāpayata prabhuḥ| dīkṣitaṃ vidhinā tena yatavākkāyamānasam| anujajñe tato vīraṃ bhrātā bhrātaram agrajaḥ‖ 14 ‖ Cf. Kirātārjunīya (KA) 3.24–29.
41Bakker 2014, 35–37.
42MBh 3.13.37–41 (tr. van Buitenen): ‘After the Pāṇḍava (i.e. Arjuna), who was the very self of Kṛṣṇa, had thus spoken to himself, he fell silent and Janārdana (i.e. Kṛṣṇa) said to the Pārtha (i.e. Arjuna). “You are mine and I am yours, and my people are yours. He who hates you hates me; who follows you follows me. You, invincible hero, are Nara and I am Hari Nārāyaṇa. Nara and Nārāyaṇa, the seers, have come from their world to this world. You are no other than I, Pārtha, I none other than you, Bhārata, no difference can be found between the two of us, bull of the Bharatas.” ’
43MBh 3.38.31: tato ’paśyat savyasācī vṛkṣamūle tapasvinam| brāhmyā śriyā dīpyamānaṃ piṅgalaṃ jaṭilaṃ kṛśam‖ 31 ‖
44MBh 3.38.40cd–41 (tr. van Buitenen): na lokān na punaḥ kāmān na devatvaṃ kutaḥ sukham‖ 40 ‖ na ca sarvāmaraiśvaryaṃ kāmaye tridaśādhipa| bhrātṝṃs tān vipine tyaktvā vairam apratiyātya ca| akīrtiṃ sarvalokeṣu gaccheyaṃ śāśvatīḥ samāḥ‖ 41 ‖
45MBh 3.38.43: bhūteśaṃ tryakṣaṃ śūladharaṃ śivam.
46KA 12.35–36; Warder 1989–1992 III, 206.
47SP 6.2d: bhaikṣāya pracacāra, see below n. 55 on p. 249.
48SP 6.10–12. For Sanskrit text and notes see SP I, 71, 144.
49MBh 3.39.23: caturthe tv atha saṃprāpte māsi pūrṇe tataḥ param| vāyubhakṣo mahābāhur abhavat pāṇḍunandanaḥ| ūrdhvabāhur nirālambaḥ pādāṅguṣṭhāgraviṣṭhitaḥ‖ 23 ‖
50MBh 3.39.28–29; tr. van Buitenen. Cf. Kirātārjunīya canto 12.
51Cf. Bakker 2014, 36 f., in which Bhāravi’s Kirātārjunīya 14.14 is quoted to illustrate the scene.
52MBh 3.40.55–56: tato devaṃ mahādevaṃ giriśaṃ śūlapāṇinam| dadarśa phalgunas tatra saha devyā mahādyutim‖ sa jānubhyāṃ mahīṃ gatvā śirasā praṇipatya ca| prasādayāmāsa haraṃ pārthaḥ parapuraṃjayaḥ‖ Bhāravi gives the following description: atha himaśucibhasmabhūṣitaṃ śirasi virājitam indulekhayā| svavapuḥ atimanoharaṃ haraṃ dadhatam udīkṣya nanāma pāṇḍavaḥ‖ (Kirātārjunīya 18.15).
53In the Kirātārjunīya 18.44 Śiva gives Arjuna the Raudra Weapon, encompassed by flaming fire (jvaladanalaparītaṃ raudram astram).
54MBh 13 App. I, No. 6 l. 45. Yuko Yokochi (personal communication) has questioned van Buitenen’s translation of brahmaśiras with ‘Brahmā’s Head’ in MBh 3.41.8a. Admittedly, there is no reference to the myth of the decapitation of Brahmā here. The Brahmaśiras, like all divine weapons, is a mantra weapon and should be ‘properly spelt’ (anumantrita) to yield the desired result. As such the word may signify ‘the foremost (śiras) of the Vedic mantras (brahman).’ We consider it plausible that the ambiguity was deliberate and was made use of by the designer of the architrave.
55SP 6.1–2 (only in the R and A recensions):tataḥ sa bhagavan̄ devaḥ kapardī nīlalohitaḥ|ājñayā parameśasya jagrāha brahmaṇaḥ śiraḥ‖ 1 ‖tad gṛhītvā śiro dīptaṃ rūpaṃ vikṛtam āsthitaḥ|yogakrīḍāṃ samāsthāya bhaikṣāya pracacāra ha‖ 2 ‖
56MBh 3.41.7–8 (cf. MBh 14.62.15, 133* ll. 3–4):bhagavan dadāsi cen mahyaṃ kāmaṃ prītyā vṛṣadhvaja|kāmaye divyam astraṃ tad ghoraṃ pāśupataṃ prabho‖ 7 ‖yat tad brahmaśiro nāma raudraṃ bhīmaparākramam|yugānte dāruṇe prāpte kṛtsnaṃ saṃharate jagat‖ 8 ‖SPBh 98.7–8 seems to distinguish between the Pāśupata Weapon and the Brahmaśiras when it lists the four weapons that form the four tusks of Varāha:catvāry astrāṇi daṃṣṭrāś ca kṛtāni sumahānti vai|astraṃ pāśupataṃ pūrvaṃ dvitīyaṃ cakram eva ca‖ 7 ‖brahmadaṇḍaṃ tṛtīyaṃ ca caturthaṃ brahmaṇaḥ śiraḥ‖grīvā tasyābhavat somo rudraḥ skandhe ca saṃsthitaḥ‖ 8 ‖
57Above, p. 243; below, p. 250. SPBh 84.3 compares the battle with the sacrifice. The ‘heads’ are the ‘sacrificial cakes’ (puroḍāśāḥ śirāṃsi), the divine weapons are the mantras (mantrāś cāstrāṇi divyāni).
58MBh 3.41.17–22 (tr. van Buitenen). […] athāstraṃ jājvalad ghoraṃ pāṇḍavasyāmitaujasaḥ| mūrtimad viṣṭhitaṃ pārśve dadṛśur devadānavāḥ‖ 22 ‖
59Williams 1982, 151, Plate 240. Lutzker 1984, 36 f.
60MBh 3.41.19: upatasthe mahātmānaṃ yathā tryakṣam umāpatim| pratijagrāha tac cāpi prītimān arjunas tadā‖ 19 ‖.
61An alternative sees in this figure the Yakṣa (rājarājānucara, KA 3.30) who leads Arjuna to Indra in Bhāravi’s Kirātārjunīya.
62We have not found references specific to Bhāravi’s Kirātārjunīya, which was being composed at about the same time or slightly later (above, p. 241; cf. n. 61 on p. 252).
63Pāśupatasūtra 3.3: avamataḥ.
64Bisschop 2006, 195 f. In MBh 13 App. I. No. 4 ll. 66–67 it is only said that Śiva ‘sports with the daughters and wives of the sages, with bristled hair, with a great penis, naked, with distorted eyes.’ No doubt, an idea like this was the source of the Pine Forest myth.
  • 2

    Bhandarkar 1920, 122; Allan 1936, cxxiii f.

  • 4

    Bhandarkar 1920, 123 f.

  • 6

    Bhandarkar 1920, 129; Agrawala 1987.

  • 7

    Bhandarkar 1920, 121. The text of this inscription has not been published, although it is referred to in several publications. We only know about the basic contents of this inscription from Bhandarkar’s original description. According to Bhandarkar it was ‘deposited in the Rajputana Museum, Ajmer.’ Although the line of the inscription that mentions the Kṛta Era has been extensively discussed in CII III (1981), 192 ff., and the inscription clearly falls within the category of ‘Inscriptions of the Early Gupta Kings,’ the editors of the revised edition of Volume III of the CII (B. Chhabra & G.S. Gai) refrain from including and editing this inscription in their volume, nor do they give any more information about the inscription itself and its whereabouts. It shows again the limited value of this ‘revised edition.’

  • 9

    Bhandarkar 1920, Plates XIV (a), XV (b).

  • 13

    Cecil 2016, 116 f. HCI III, 161 f.

  • 16

    Williams 1982, 140 f. Cf. Williams 1982, Plate 202 (Deogarh).

  • 19

    Bhandarkar 1920, 139 thinks that ‘enough has been recovered to show that each broader side was originally divided into nine panels [including garland bearers].’ Bhandarkar’s reconstruction in Plate XXIII shows that he conjectures that one complete panel plus garland bearer is missing at the southern end. In view of the turn of the torsion between the third and fourth panels (E 3–4, W 3–4), we think this is less likely. An argument against our assumption, as voiced by Véronique Degroot, is that the architrave panels show alternating indentation and protrusion: panels E 2/W 5, E 4/W 3 and (northern-most) E 6/W 1 are protruding and so, assuming symmetry, one might expect a southern-most panel to be protruding as well. Another argument, however, against a missing 7th panel, in addition to the change in torsion, seems to be that the breaking point in the southern-most panel (E 1/W 6) corresponds with the crack in the northern-most panel (E 6/W 1), the positions where the architrave rested on its two posts. The collapse of the toraṇa thus seems to have taken place by a break in the middle and at the two junctures. Six panels on each side makes sense according to the iconographical analysis offered in this article. If a 7th, lost panel was originally part of the architrave, the iconography of panels E 1 and W 6 that we conjecture may have been divided, broadly speaking, over two panels.

  • 20

    Williams 1982, Plate 216; Bakker 2004, Figure 4.1.

  • 21

    Letter dated 10 March 2001, quoted in Bakker 2004, 132.

  • 36

    Bakker 2004, 124 f. Kauṇḍinya adPS 1.1. Dalal 1920, p. 9: RatnaṭīkāadGaṇakārikā 5c.

  • 41

    Bakker 2014, 35–37.

  • 51

    Cf. Bakker 2014, 36 f., in which Bhāravi’s Kirātārjunīya 14.14 is quoted to illustrate the scene.

  • 57

    Above, p. 243; below, p. 250. SPBh 84.3 compares the battle with the sacrifice. The ‘heads’ are the ‘sacrificial cakes’ (puroḍāśāḥ śirāṃsi), the divine weapons are the mantras (mantrāś cāstrāṇi divyāni).

  • 59

    Williams 1982, 151, Plate 240. Lutzker 1984, 36 f.

  • 64

    Bisschop 2006, 195 f. In MBh 13 App. I. No. 4 ll. 66–67 it is only said that Śiva ‘sports with the daughters and wives of the sages, with bristled hair, with a great penis, naked, with distorted eyes.’ No doubt, an idea like this was the source of the Pine Forest myth.

If the inline PDF is not rendering correctly, you can download the PDF file here.

  • 2

    Bhandarkar 1920, 122; Allan 1936, cxxiii f.

  • 4

    Bhandarkar 1920, 123 f.

  • 6

    Bhandarkar 1920, 129; Agrawala 1987.

  • 7

    Bhandarkar 1920, 121. The text of this inscription has not been published, although it is referred to in several publications. We only know about the basic contents of this inscription from Bhandarkar’s original description. According to Bhandarkar it was ‘deposited in the Rajputana Museum, Ajmer.’ Although the line of the inscription that mentions the Kṛta Era has been extensively discussed in CII III (1981), 192 ff., and the inscription clearly falls within the category of ‘Inscriptions of the Early Gupta Kings,’ the editors of the revised edition of Volume III of the CII (B. Chhabra & G.S. Gai) refrain from including and editing this inscription in their volume, nor do they give any more information about the inscription itself and its whereabouts. It shows again the limited value of this ‘revised edition.’

  • 9

    Bhandarkar 1920, Plates XIV (a), XV (b).

  • 13

    Cecil 2016, 116 f. HCI III, 161 f.

  • 16

    Williams 1982, 140 f. Cf. Williams 1982, Plate 202 (Deogarh).

  • 19

    Bhandarkar 1920, 139 thinks that ‘enough has been recovered to show that each broader side was originally divided into nine panels [including garland bearers].’ Bhandarkar’s reconstruction in Plate XXIII shows that he conjectures that one complete panel plus garland bearer is missing at the southern end. In view of the turn of the torsion between the third and fourth panels (E 3–4, W 3–4), we think this is less likely. An argument against our assumption, as voiced by Véronique Degroot, is that the architrave panels show alternating indentation and protrusion: panels E 2/W 5, E 4/W 3 and (northern-most) E 6/W 1 are protruding and so, assuming symmetry, one might expect a southern-most panel to be protruding as well. Another argument, however, against a missing 7th panel, in addition to the change in torsion, seems to be that the breaking point in the southern-most panel (E 1/W 6) corresponds with the crack in the northern-most panel (E 6/W 1), the positions where the architrave rested on its two posts. The collapse of the toraṇa thus seems to have taken place by a break in the middle and at the two junctures. Six panels on each side makes sense according to the iconographical analysis offered in this article. If a 7th, lost panel was originally part of the architrave, the iconography of panels E 1 and W 6 that we conjecture may have been divided, broadly speaking, over two panels.

  • 20

    Williams 1982, Plate 216; Bakker 2004, Figure 4.1.

  • 21

    Letter dated 10 March 2001, quoted in Bakker 2004, 132.

  • 36

    Bakker 2004, 124 f. Kauṇḍinya adPS 1.1. Dalal 1920, p. 9: RatnaṭīkāadGaṇakārikā 5c.

  • 41

    Bakker 2014, 35–37.

  • 51

    Cf. Bakker 2014, 36 f., in which Bhāravi’s Kirātārjunīya 14.14 is quoted to illustrate the scene.

  • 57

    Above, p. 243; below, p. 250. SPBh 84.3 compares the battle with the sacrifice. The ‘heads’ are the ‘sacrificial cakes’ (puroḍāśāḥ śirāṃsi), the divine weapons are the mantras (mantrāś cāstrāṇi divyāni).

  • 59

    Williams 1982, 151, Plate 240. Lutzker 1984, 36 f.

  • 64

    Bisschop 2006, 195 f. In MBh 13 App. I. No. 4 ll. 66–67 it is only said that Śiva ‘sports with the daughters and wives of the sages, with bristled hair, with a great penis, naked, with distorted eyes.’ No doubt, an idea like this was the source of the Pine Forest myth.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 138 32 0
Full Text Views 303 157 3
PDF Downloads 26 22 2