Pramāṇa · Search · About · API

The Development of Hindu Iconography c12: Chapter V: Deities and their Emblems on Early Indian Seals

The Development of Hindu Iconography c12

Language
English

Contents

Chapter V: Deities and their Emblems on Early Indian Seals

c12.2.1

Like the numismatic remains of ancient India, her glyptic ones also throw a flood of light on the mode of representing her divinities in different periods. The innumerable varieties of seals and similar objects that have been unearthed in various parts of Northern India and that can be dated from the third or fourth millenium B.C. to the late Gupta period and afterwards contain numerous figures, many of which have been assumed with a great deal of justification to stand for various divinities in their anthropomorphic, theriomorphic and sometimes therio-anthropomorphic forms. On many seals of the Kushan and the Gupta periods, most of these gods and goddesses as also their emblems can be definitely recognised as belonging to one or other of the different religious creeds that were current in the period when they were manufactured. I have already drawn the attention of my readers, in the previous chapter, to the Sirkap bronze seal of Śivarakṣita, that gave me the necessary clue for the identification of Śiva in certain coin-devices of Maues. Mention has also been made by me there of a few other metal and terracotta seals of the Kushan and the Gupta periods, which supply us with characteristic representations of such Hindu gods as Viṣṇu and Śiva, as well as a few of their emblems.

c12.2.2

I shall presently draw the attention of my readers to a good many seals of the Gupta period (a few amongst them going back to the Kushan age), that were unearthed at such old sites of India as Bhita, Basarh, Rajghat, etc. But before I begin a systematic study of some of these seals and seal-matrices, from the iconographic point of view, it will be necessary for me to refer briefly to the many hundreds of sealings that were discovered in the course of excavations at the pre-historic Indus Valley sites of Mohenjodaro and Harappa. Whatever might have been the particular purpose that was served by them, there is little doubt that the figures which very frequently appear on their surface had some connection with the religion that was practised by these pre-historic Indians. The very interesting seal unearthed at Mohenjo-daro, which bears a three-faced horned figure ‘seated on a low Indian throne in a typical attitude of Yoga, with legs bent double beneath him, heel to heel, and toes turned downwards’, has previously been noticed by me. This particular sitting posture clearly corresponds to the Yogic āsana known as kūrmāsana, where the heels are placed under the gluteals in a manner exactly similar to the mode described above. It will be of interest here to give a fuller account of the device, so carefully studied by Sir John Marshall. The two arms of the figure, which are covered with bangles, are outstretched, and his hands, with thumbs to front, rest on his knees;

c12.2.3

on his neck and breast is placed a series of necklaces or torques in a manner similar to that of the graiveyaka ornament placed on the neck and breast of the Yakṣa figures of the Śuṅga and the post-Śuṅga period; the lower limbs seem to be bare and the figure appears to be ithyphallic; his head is crowned by a pair of horns meeting in a tall head-dress. To either side of the god are four animals, an elephant and a tiger on his proper right, a rhinoceros and a buffalo on his left. Beneath the throne are two deer standing with heads regardant and horns turned to the centre. Just below the trunk of the elephant on the top left corner and above the tiger is the crude outline of a human figure (Pl. VII, fig. 1). Marshall is justifiably sure about the divine character of the figure, and from its peculiarly distinctive attributes, such as three faces, the Yogic āsana, its association with animals, as many as five or six in number, its deer-throne and its horns, he concludes that the figure is a prototype of the historic Śiva-Paśupati.¹ The seal just noticed at length is the same as No. 420 in Mackay's list of seals discovered by him at Mohenjo-daro. Two other seals (Nos. 222 and 235) in the same list, contain different representations of apparently the same deity, though many of the details of Mackay's No. 420 are omitted. The figure on seal No. 235 bears only one face, and the head, adorned with a pig-tail hanging down on one side, is shown in profile.

c12.2.4

The head-dresses of the figures in these two seals (Nos. 222 and 235) are very simliar, ‘ but surmounted by a plant motif with three branches in the one case and only a single branch on the other.’ Mackay remarks about this head-dress, ‘The larger figure on seal 420 lacks this spray of foliage, but has instead the fan-shaped ornament commonly associated with the pottery female figurines.’¹

c12.3

¹ Marshall, M.I.C., Vol. I, pp. 52-6, pl. XII, 17. We miss, in the assembly of animals by the side of the god, Śiva's bull Nandi. Marshall has very rightly referred to the association of deer with the historic Śiva. As regards the horns, there is no need to assume that they took the form of the triśūla or trident in later

c12.4

Marshall refers to two seals found at Mohenjo-daro, which contain figures of a god seated in *yoga* posture, on whose either side kneels a half-human half-animal form of a

c12.5

days, and in that guise continued to be a special attribute of Śiva’; for the horns as such were also associated with Śiva, as is evident not only from the epic passage which reads: *Svargāduttuṅgamamalaṃ viṣāṇaṃ yatra śūlinaḥ | Svamātmavihitaṃ dṛṣṭvā marttyaḥ śivapuraṃ vrajet* (Mahābhārata, Vanaparva, ch. 88, v. 8), but also from the fact that the horn as an instrument of music is very often placed in one of the hands of the popular representations of Śiva in Bengal. Hopkins thinks that the horn in the epic passage just quoted may refer to the crest of the image o’ Śiva (*Epic Mythology*, p. 73).

c12.6

Saletore attempted to identify the figure as Agni, in *New Review*, 55, X, 1939; but his grounds of objection to Marshall's view were refuted by Moraes in a subsequent issue of the same journal. In one of the issues of J.R.A.S.B., the problem of the identity of the figure has been thoroughly discussed from the ethnological point of view by A. Aiyappan who has fully endorsed Marshall's identification (Letters, Vol. V, pp. 401-06).

c12.7

¹ Mackay, *Further Excavations at Mohenjo-daro*, Vol. I, p. 335; Vol. II, Pl. LXXXVII, figs. 222 and 235, and Pl. XCIV, fig. 420. Mackay is not sure whether there are horns on the head of the figures on seals Nos. 420 and 222; with regard to the latter, he says, ‘The horns, if indeed they are horns, are definitely separate from the head; they are, moreover, represented as fastened to the base of the twig.’ What has been described as a probable *ūrdhvaliṅga* feature of the figure on No. 420 is absent on the figures on the two other seals, where they appear to be wearing a very short piece of loin-cloth comparable, according to Mackay, with *lyaṅgot* (kaupīna?), so frequently worn by yogis and sannyāsis of India.

c12.8

A. Aiyappan has made some useful suggestions with regard to the horned head-dress in J.R.A.S.B., Letters, Vol. V, pp. 401-6.

c12.9

Nāga with hands uplifted in prayer (M.I.C., Vol. III, CXVI, 29 and CXVIII, 11). It is not quite clear, however, from these two seal devices whether the snake-body is attached to the back of the kneeling human votaries of the god; in the early Kushan and subsequent representations of the Nāgas at Mathura and other sites, the whole serpent-body and sometimes only its one or many hoods (five or seven) are invariably attached behind the human body (the latter mode is also adopted in the Śuṅga art of Central India). But on these seals, the technique of showing the Nāgas might have been somewhat similar to the one followed by the Bharhut artist in his presentation of the scene of Elāpatra Nāgarāja's visit to the Buddha; at first Elāpatra is shown in his serpent form, then he is given the human shape with the snake hoods attached behind his head. On these Indus-valley seals, the snakes appear on the far sides while the kneeling human figurines, without any snake hood, on the near sides of the god.

c12.10

Several other composite figures of a more complex form are also found on these seals; human-faced goat or ram, part bull, and part elephant with human countenance,—all these are figured on seals Nos. 378, 380 and 381 (M.I.C., Vol. I, p. 66). These curious composite forms, so clearly reminiscent of the Pramathas or Gaṇas, the attendants of Śiva, of subsequent days, are apparently also represented in the stone images in the round, illustrated in M.I.C., Vol. III, Pl. C, 7 and 9. Marshall remarks, ‘Such stone images can hardly have been other than cult objects intended for worship; on the other hand, the seals, like most of the seals found at Mohenjo-daro and Harappa were almost certainly amulets which were used by the votaries of this curious syncretic form of deity’ (Ibid., Vol. I, pp. 66-7). Mackay’s excavations at Mahenjo-daro brought to light a few more seals with the composite animal figures. The beast on his seals numbering 24 and 494 represents ‘a combination of the usual urus-like animal with two other heads, those of an antelope and a short-horned bull.’ A possible explanation suggested by him about this unusual device is that ‘its owner may have sought the protection or assistance of three separate deities represented by the heads of these three animals.'¹ The same archaeologist also records the discovery of four seals numbering 411, 450, 521 and 636, from the lower levels, having the curious human-faced composite animal similar to that on Marshall's seals numbering 378, 380 and 381, already noticed by me. In this figure, there is a fusion of as many as three, or possibly four, animals,—forelegs of an ox-like animal, the striped hind-quarters and feet of a tiger, short curved horns of a bull or an antelope and the lolling trunk of an elephant and its pair of tusks. Mackay observes that this composite figure perhaps represented a deity that was worshipped at Mohenjo-daro; he is also inclined to think that 'it was perhaps also portrayed in statue form, as the representation of it on the seals shows it to be wearing garlands with which it is likely that its images were adorned.' These chimaera-like creatures distinctly remind us of the human-faced winged bulls and griffins of the early Buddhist art of Central India, whose prototypes have been sought by Grünwedel and others in the similar creatures of imagination portrayed in the early art of Western Asia. I may, however, draw the attention of my readers to one very significant observation of Mackay, in this connection: 'Composite animals are, of course, well-known in ancient art in other parts of the world; they are supposed to have been invented, if we may thus term it, in Sumer and Elam, whence came the later "beast art" of Europe. It is not outside the bounds of possibility that the conception of a composite animal originated in India and spread from there gradually to the west by the land route.'² Reference may be made here to the terracotta sealing (No. 2409), a three-sided tapering prism, unearthed from mound F at Harappa. Each of its three faces contains a standing mythical figure, but the one on the left face is very interesting, it being human above the waist and bovine below. The figures on the right and middle faces also seem to be human above and animal below (Vats, *Excavations at Harappa*, Vol. I, p. 44; Pl. VIII, Fig. 1). I have already mentioned the name of the Gaṇas and the Pramathas, while referring to the human-faced animal forms. The Garuḍas, Gandharvas, Kinnaras, Kumbhāṇḍas and others of the epic and Purāṇic literature and ancient and mediaeval Indian art of the historic period should also be considered in this connection. The base of some of the above is undoubtedly Vedic in character (Garuḍa-Garutman, sun conceived as a bird in the Ṛgveda); but who can doubt that these creatures of imagination owed much for their origin and evolution to the dim memories of the remote past in the minds of the Indians of the age of the Mahābhārata and of the Purāṇas?

c12.11

¹ Mackay, *Further Excavations at Mohenjo-daro*, Vol. I, p. 333.

c12.12

² Mackay, op. cit., Vol. I, p. 333; Vol. II, Pls. LXXXIII, XCIV, XCV, XCVI, XCVIII. 'These composite animals appear only on the seals of the earlier occupations,' as is evident from the fact that the latter are found only in the lower level, but the fact of their not being found in the upper strata should not be stressed too far. For such composite animals, see Pl. VIII, Figs. 1-3.

c12.13.1

Some of these seals also contain elaborate representations of scenes which seem to illustrate mythological stories current among the pre-historic people of this region. These seal devices can very well be compared with the iconographic presentation of various myths associated with different religious creeds of India in the subsequent period. A reference to a few such seal devices will not be out of place here. Vats describes a triangular prism sealing of terracotta with a blurred legendary scene on each side. One face of this seal shows a god in a standing posture; his right arm is profusely decorated, but the left one is indistinct. Its second face shows a tall stalwart man engaged in fighting a bison which has been firmly caught by the horns. Vats observes, ‘The scene may be a representation of Ea-bani fighting a bison in a jungle.’ The third face shows to left a human figure, most presumably a deity, seated in a typical attitude of Yoga with another figure to right seated on its haunches.¹ The same author fully describes an oblong terracotta sealing which contains legendary scenes on both its faces; the order of depiction on each face probably runs from left to right. It is so very interesting for the purpose of our present study that I cannot but fully note his description of the devices on both the sides. The obverse shows first of all a man attacking a tiger from a *māchān* (scaffolding) erected on an acacia tree.

c12.13.2

The deity is next shown seated on a low Indian throne in the well-known Yogic posture; from behind his head-dress there is a long tassel-like appendage to right, which reminds us of a similar object on the head of a similar figure on some Mohenjo-daro seals described by Mackay (*cf.* Nos. 222 and 235 already noted by me). Of the animals to his right, the one in the enclosure may be a goat, that below the projection, a hare or kid, and that above it, an indistinct animal with a long body. The reverse side of it shows from left to right a humpless bull standing by a trident-headed post, with his head bent down a little, then a standing figure, possibly a god, in front of a two-storied structure, followed by three pictograms at the right end. The structure seems to be of wood and is of unusual interest. It 'looks like a combined side elevation and perspective of a double-storied room preceded by a porch—both of open work in front, but seemingly the two-storied room is closed by lattice-work on the rear side and crowned at the corners by somewhat conical finials.' It is not certain what the bifurcated object apparently hanging down from a projection in front of the terrace stands for; just below it, however, is placed a domical something over the porch.

c12.13.3

Vats remarks, "The structure is probably of a sacred character, and in view of the trident post and bull, which are peculiarly associated with Śiva whose prototype has been found at Mohenjo-daro, the possibility of the standing figure being ultimately identified as another form of the same god may not be ruled out."' The scene of what appears to be a tiger-hunt is comparatively familiar in Mohenjo-daro and Harappa. Marshall suggests that ‘such seals may have been used as protective amulets against tigers or other jungle animals’ (*M.I.C.*, Vol. I, p. 71). This explanation holds good as regards the obverse device, but the reverse one partially reminds me of the reverse device of certain Ujjayinī coins, which I have reproduced in Pl. I, fig. 15; the animals are no doubt absent, but here too is some sort of a structure with conical projections (?), as well as a trident-headed post which, however, is held by the right hand of the standing figure on the right side (in the previous chapter, I have suggested the possibility of this figure representing the votary; it may as well be the cult deity in his human form). As regards ‘the domical object over the porch’ on the Harappa seal, it might be the same as the realistic phallus which appears on the obverse of the same types of coins (*cf.* Pl. I, fig. 14).

c12.13.4

It may be noted *in passim*, that the figure standing by the humpless bull on the Harappa seal seems to hold a long staff in his left hand and a water-vessel-like object in his right one, just reminiscent of similar figures on certain punch-marked coins, which I have tentatively identified as Śiva in the previous chapter (*cf.* Pl. I, fig. 4). The devices on the two seals, Nos. 279 and 510 of Mackay’s book, are of great interest for our study. The former depicts a buffalo with its head so represented as to show both the rugged horns, below which is placed an apparently partitioned feeding-trough; in the extreme left corner is shown a man with his foot upon the buffalo’s nose, grasping a horn with one hand and with the other about to thrust a spear with a barbed point into the animal’s back; there was a pictogram on the top right, only one letter being preserved, the others being broken off. The same scene also appears on two other sealings unearthed at Mohenjo-daro. Mackay remarks, with a great deal of diffidence, that this scene ‘may represent a belief not unlike the legend of Dundubhi, the buffalo demon, whom Śiva and other gods attacked with tridents; though their weapons proved powerless against the animal, they eventually killed it by means of incantations.'¹ The parallelism noticed by Mackay is no doubt very interesting;

c12.13.5

I remember one passage of the Durgā-saptaśatī, which, while describing the fight between the goddess Durgā and the evil incarnate in the shape of the buffalo-demon, says, ‘sāruḍhā taṃ mahāsuram pādenākramya kaṇṭhe ca śūlenainamatāḍayat,’ i.e., ‘(the goddess jumped) to climb upon the great demon, attacked him with her leg and struck at his neck with her śūla’ (it may be a trident or a barbed spear). The Purāṇic description of this fight may also be a close parallel, but the human figure in Mohenjo-daro seals seems to be a male one and the different forms of plastic representations of Durgā as Mahiṣamarddinī have very little similarity to the scene on the Indus seals just described. Mackay’s seal No. 510 shows a buffalo which seems to have attacked a number of people who are lying on the ground in every conceivable position. Without excluding the possibility of its depicting ‘an episode that actually occurred to some of the inhabitants of Mohenjo-daro,’ Mackay observes, ‘we may perhaps see in this scene a god, or the emblem of a god, attacking his enemies, a parallel to the well-known scene on the slate palettes of the First Dynasty of Egypt, where the king himself in his character of a “Strong Bull” gores a prostrate enemy.’²

c12.14

¹ *Excavations at Harappa*. Vol. I, p. 129; Vol. II, Pl. XCIII, 310. The tentative explanation of the scene depicted on the first face may be correct; but the scene depicted on the third face is undoubtedly Indian in nature.

c12.15

¹ M. S. Vats, *op. cit.*, pp. 129-30, Pl. XCIII, 303. Both these terracotta seals were discovered in Mound F, belonging to Stratum No. III.

c12.16

¹ Mackay, op. cit., Vol. I, p. 336; Vol. II, Pl. LXXXVIII, fig. 279, Pls. XCI, 4a, XCII, 11b. He quotes, as his authority for the Dundubhi legend, Oppert's Oriental (evidently a misprint for Original) Inhabitants of India, pp. 473-74. In the 9th chapter of the Avantikṣetramāhātmyam of the Avantya-Khaṇḍam of the Skandapurāṇa, we find the story of the buffalo-demon named Hālāhala being killed by the Gaṇas of Śiva as well as the other gods assembled in the Rudrakṣetra near Avanti.

c12.17

² Mackay, op. cit., Vol. I. pp. 336-7; Vol. II, plate XCVI, fig. 510.

c12.18

It is time now to refer to a few more early Indus Valley seals and incidentally to other objects of a somewhat similar nature, that seem to prove the existence of the Mother-goddess cult among the people of Mohenjo-daro and Harappa. Marshall has observed that though there is no direct proof about the existence of Śaktism in this region, yet there is enough indirect evidence in the shape of phalli, baetylic stones and ring-stones. The same author drew the attention of scholars to numbers of female figurines of terracotta, etc., that were discovered not only in this part of India but also in Baluchistan, though the ones discovered in the latter place differ from those of the Indus Valley in that they are not full-length images.¹ The great majority of these female figurines appear as ‘a standing and almost nude female, wearing a band or girdle about her loins, with elaborate head-dress and collar, and occasionally with ornamental cheek cones and a long necklace.’ Mackay remarks (op. cit., Vol. I, p. 265), ‘In fact, what are generally regarded as images of an Earth or Mother-goddess are practically always nude, save for quantities of jewellery, a wide girdle and their remarkable head-dresses.’ Now, an oblong terracotta seal with scenes depicted on both sides, that was unearthed at Harappa, most probably contains a representation of the same goddess with some additional traits. The right side of the obverse face is occupied by a nude female figure shown upside down with legs wide apart, and ‘with a plant issuing from her womb; her arms are shown in the same position in which those of the prototype of Śiva-Paśupati are shown; at her left side are shown a pair of tigers standing facing each other (these are regarded by Marshall as two genii, animal ministrants of the deity). The left part of the reverse side of this seal contains two human figures, one male and the other female; the latter seated, with her hair dishevelled, raises her hands in supplication to the male who stands in front of her in a threatening attitude with a shield-like thing and a sickle-shaped object in his left and right hands respectively. Marshall suggests that the ‘scene is intended to portray a human sacrifice connected with the Earth Goddess depicted on the other side, with whom we must also associate the two genii.’ This striking and unique representation of the goddess with a plant issuing from her womb is compared by Marshall with a terracotta relief of the early Gupta age from Bhita on which the goddess is shown with her legs in much the same position, but with a lotus issuing from her neck instead of from her womb.¹ One of the most interesting seals bearing the representation of a goddess, this time a tree-goddess or spirit, was discovered at Mohenjo-daro. The tree, an *aśvattha* as recognisable from its leaves shown on the top right corner, is represented by its two branches only springing from a circle on the ground; between the two branches stands the nude deity having long hair, a pair of horns with probably a spring of foliage in between, or *triśūla* horns, and armlets; in front of the tree appears a half-kneeling worshipper, also with long hair, armlets and horns (between a pair of these horns a leaf-spray or plume is recognisable), behind whom stands a goat with human face; in the register below are seven ministrants or votaries, each dressed in short kilt and wearing long pig tails with a spray of leaves or a feather in the hair; beyond the foot of the tree on the right is a square partitioned receptacle very similar in conception to the pottery dishes found in Mohenjo-daro (Pl. VII, Fig. 2). Marshall is of opinion that the whole scene represents the epiphany of the tree-goddess, taking the composite human-headed animal figure as a protecting local divinity of a minor type accompanying the suppliant into the presence of the tree-goddess.²

c12.19

¹ Marshall, M.I.C., Vol. I, p. 48 ff. Marshall refers to the wide belt of the ancient world from the Indus to the Nile, in which these figurines have been found; he is sure that they are ‘effigies of the great Mother-Goddess or of one or other of her local manifestations.’

c12.20

¹ Marshall, M. I. C., Vol. I, p. 52, pl. XII, fig. 12. M. S. Vats, *op. cit.*, Vol. I, p. 42, Vol. II, pl. XCIII, 304. Marshall, after comparing the two animal "genii" on this sealing with those hailing from the Aegean area and Mesopotamia, remarks: "That the conception of these animal genii arose independently in Greece, Mesopotamia and India is hardly conceivable, but whether it originated in the East or West has yet to be determined."

c12.21

² Marshall, *op. cit.*, Vol. I, pp. 63-5, pl. XII, fig. 18. Mackay, *op. cit.*, Vol. I, pp. 337-8; Vol. II, pl. XCIV, fig. 430, pl. XCIX, A. The goddess standing between the branches of the tree is reminiscent of one of the variants of the goddess Lakṣmī, in which she is made to stand on the pericarp of a lotus flower, with lotus flowers and leaves on long stalks spreading on her either side; *cf. H. I. I. A.*, pl. XIV, fig. 52.

c12.22.1

Though the objects now to be noted by me do not really fall in with the category of sealings, still I feel a brief reference to them will be of some use to us in our present study. A large number of ring-stones, ranging from half an inch to nearly four feet in diameter, have been found in the course of excavations at Mohenjo-daro and Harappa; the larger ones are made of stone, while the smaller ones are of different materials such as stone, faience, shell, or imitation carnelian. The most typical of them have their upper and lower surfaces undulating; in others, the lower surface is flat, and the top takes a quatrefoil form (Marshall). Two explanations were suggested by scholars with regard to the nature of these objects; according to some, the larger ones of them served as architectural members, while according to others they were stone money. But Marshall has raised very reasonable objections to both these suggestions, and his original interpretation that these are to be regarded as representations of *yoni*, the female organ of generation, as symbolising motherhood and fertility still appears to be the correct one.

c12.22.2

When they are compared to the numbers of phalli—they are so realistic that they cannot be explained in any other way (*cf.* the realism manifest in the earliest stone phalli of the historic period discovered at Gudimallam, Mathura and other places)—that have been discovered in the same region, there remains very little doubt about the truth of Marshall's explanation. But it must be borne in mind that in the Indus Valley both the phalli and the *yoni* stones appear to have served the purpose of cult objects separately, as seems to have also been the case with the early phalli and the *yoni* stones of subsequent days. In fact, the *lingam* in *arghya* (or *yoni*) design is comparatively late in appearance and even then in the conventional Sivalingas the spout-like projection, from which the *pūjābhāga* of the Sivalinga rises upward and which is taken by the uninitiated as symbolising *yoni*, is really a *praṇālī* or drain for the easy outflow of the volume of water usually poured on the top of the emblem by the numerous devotees of the god.¹ These phalli and the ring stones thus appear to have separately symbolised the principles of virility and fecundity, both of which are highly esteemed by all men in all ages. Marshall has referred to his own discovery of several curious stone discs, three of which were unearthed from the Bhir Mound at Taxila, one from inside the structures uncovered near the foot of Hathial (Taxila), and one at Kosam.

c12.22.3

The Hathial one is described by Marshall, thus: ‘It is of polished sandstone 3¼" in diameter, adorned on the upper surface with concentric bands of cross and cable patterns and with four nude female figures alternating with honey-suckle designs engraved in relief around the central hole’ (A.S.I.A.R., 1927-28, P. 66, Pl. XX, Fig. 7). It will be of interest now to compare with the above Taxila discs a partially broken reddish steatite circular disc, about 2½" in diameter, found at Rajghat, which contains on the outer side of its top surface a very wellcarved decorative design. The decoration consists of a palm tree with a horse by its side, beyond which is a female figure holding a bird in her outstretched right hand (there is an indistinct object beneath her right hand and a taurine symbol near her left shoulder); then follow in successive orders—a long-eared and short-tailed animal, a crane, the goddess again with her hands this time stretched downwards, some object which is broken, a second palm-tree, a bird, a small circular disc, the goddess again with the circular disc near her left shoulder, then a winged mythical animal and lastly a crane with a crab-like object near its legs. The goddess is thrice repeated with the various accessory figures noted above in between her three representations.

c12.22.4

But one thing to be noted here is that, unlike the Taxila disc just described, the device appears here on the top surface instead of on the side of the central depression of the disc, and the hole is not there; the surface near the central hole of this one is filled with a beautiful scroll design. The carving is so very beautifully executed

c12.23

¹ The elaborate pedestal, however, in the conventional Sivalingas of the subsequent period were definitely regarded as illustrating the female principle as the

c12.24

iconographic texts of a comparatively late date, as well as many late Sanskrit works, prove. It must be noted, however, that these elaborate pedestals are usually absent in the phallic emblems of an earlier date.

c12.25

on this piece in the collection of the Bharat Kala Bhavan Museum, that it can justifiably be assigned to the same age to which the Taxila, Kosam and other discs belong. The same Museum has in its collection another fragment of a red steatite disc unearthed in course of excavations at Rajghat near Banaras, which is more similar to the Taxila disc. This has a hole through the centre, around which as in the Taxila ones are engraved two nude female figures with their hands stretched downwards with probably a honeysuckle in between them; on the flat surface of the disc between cable designs are two monkey-like animals holding a creeper(?) with a lizard (or an alligator) in between them; there is a partially defaced inscription in early Brāhmī script on its rim, which is illegible. Another partly broken similar disc hailing from Kosam, which was acquired by the aforesaid Museum at Banaras, contains a much damaged though partially legible inscription in Aśokan Brāhmī. The inscription reads,......ma ṃ tha ṃ ka bhā dā ma tha lo ga tara śa a ga la(?) na(ni?) ka ye la ṃ ca le......; it is unfortunate that no sense can be made of it. The ring-stone has two bands of decoration cut in relief on one face around the hole. On one band can be seen a row of alligators below a twisted rope, and on the second band which extends into the hole are carved the nude goddesses between three-pronged trees. The inscription noted above appears on the side of the disc. All the above discs can justifiably be regarded as cult objects comparable with the pre-historic ring stones of the Indus Valley on the one hand and the cakras and the yantras of the Śāktas, the Viṣṇupaṭṭas of the Vaiṣṇavas and the āyāgapaṭas of the Jainas on the other. But their ideological association with the former, viz., the cakras and the yantras of the latter-day Śākta cult, appears to be closer.¹ Marshall observes about the Taxila discs, ‘In these ring-stones, which are quite small and used perhaps as *ex voto* offerings, nude figures of a goddess of fertility are significantly engraved with consummate skill and care inside the central hole, thus indicating in a manner that can hardly be mistaken the connection between them and the female principle.’ The association of the mother-goddess of some of these discs with alligator or alligators is also of unique importance and interest from the point of view of the developed Śakti cult in India. In mediaeval Pārvatī images of Bengal, an alligator (or iguana, *godhā*) is almost invariably shown on the pedestal. *Godhā* plays a prominent part in the mythology of the goddess Caṇḍī and Kālaketu in the mediaeval *maṅgalakāvyas* of Bengal. The animal is also found in many Umā-Maheśvara reliefs of southern India. One of the earliest reliefs in which a *godhā* is shown stretched in two hands of the twelve-armed goddess Mahiṣamarddinī is carved on the ‘Chandragupta cave’ facade at Udayagiri (Bhilsa, Madhya Bharat) can be difinitely dated in the first or second year of the fifth century A.D. These details fully prove that the moorings of the Śakti cult in India go back to many centuries before the Christian era. Mention may be made in this connection of the twenty-one stone-discs that were accidentally discovered from a deep drain in May, 1951, in Murtaziganj Mahalla of the Patna City (a section of ancient Pāṭaliputra). They are of soapstone, and the carvings on many of them are of a very high order; their date probably falls in the Śuṅga period. Five of them contain the figures of the nude mother or the fertility goddesses associated with various animals and birds like lion (some of them are winged), elephant, horse, antelope, stag, ram, goose, peacock and

c12.26

¹ Rai Sahib Krishnadas, the Curator of the Museum, kindly gave me permission to utilise the stone-discs in the collection of the Bharat Kala Bhavan for my book. I may here refer to one cylindrical amulet-like object of red steatite about 1¼" in length and ½" in breadth, found at Rajghat, which is somewhat similar to a few cylindrical seals (amulets?) unearthed at Mohenjo-daro. There are three shallow incuse bands, two on either side and one in the middle, the latter dividing

c12.27

¹ M.I.C., Vol. I, pp. 62-3. In the f.n. No. 1 on page 63, he says, ‘That ring-stones of this type had a wide vogue in ancient India is shown by the discovery of another specimen at Sahet-Mahet (ancient Śrāvastī) in the U.P., and by the fact that they were copied by the Buddhists, though with this difference that the nude figures of the goddess were eliminated.’

c12.28

the small cylinder in two fairly equal sections; in one of them are found, in order, a taurine, a long-eared and short-tailed animal, a two-humped camel and a lion, while the other section bears in succession a taurine, a horse, the long-eared and short-tailed animal and an elephant.

c12.29

parrot. In this strange medley of animals no alligator or iguana is seen, it is true, but the association of lion and other animals and that of the birds with the goddesses is very significant. Profuse plant and vegetation designs, like those of lotus flower, palmyra, date-palm, etc., cluster round the goddess or goddesses, and they may emphasise the vegetation aspect of the deity so prominent in the Purāṇic concept and worship of the goddess Durgā, the divine Mother. They must have been used as cult objects, and did not merely ‘serve the purpose of decorating walls or doors’ as suggested by Shere, and they fall in line with the other stone discs and rings recovered from various sites of India, Indus Valley onwards. A ring-stone similar to the Taxila ring-stone, but with a Mauryan Brāhmī inscription viśākha (Patna Museum Archæological Exhibit No. 8814) was wrongly described by Jayaswal as a toy-wheel; it has, however, no animal, plant or human figures, but has some decorative and geometrical designs in several rows round the central hole. It should be noted incidentally that these cult objects of the historical period were recovered mostly from historical sites such as Taxila, Kosam (Kauśāmbī), Mathura, Banaras (Rajghat) and Patna (for an objective description of the Patna stone discs and ring, cf. J.B.O.R.S., Vol. XXXVII (1951), pp. 178 ff. and plates; see also Pl. XII, Figs. 1-5).

c12.30

The pre-historic people of the Indus Valley appear to have been great believers in animism also, as is proved by a good many seals discovered at Mohenjo-daro and Harappa. The worship of trees or the tree-spirits is the characteristic manifestation of animistic belief. I have already referred to a seal which seems to unite in its device the worship of the female principle as well as that of the tree-spirit, where the epiphany of the female deity in the tree is portrayed in a half realistic, half conventional manner. Many seals in the Indus Valley sites show the presence of two different forms of tree-worship among the people of the locality: ‘One in which the tree itself is worshipped in its natural form, the other in which the tree spirit is personified and endowed with human shape and human attributes.' On several sealings of the Indus region (M.I.C., Vol. I, Pl. XII, Figs. 16, 20, 21, 25, 26), various sacred trees are represented which the artists have attempted to differentiate one from another. A few of these trees appear to be like those enclosed by walls or railings commonly surrounding the base of the sacred trees (vṛkṣacaityas) depicted in the later reliefs of the historic period. In the fourth chapter of this book, I have drawn the attention of my readers to one of the commonest devices on the early indigenous coins of India, which is 'tree within railing.' These enclosed trees on the Indus seals can very well be compared with the above and can justifiably be taken as distant prototypes of the vṛkṣacaityas and the sthalavṛkṣas represented by the latter. The terracotta seal (No. 2410) found at Harappa has as its obverse device 'a deity wearing a kilt or short tunic and a three-pointed head-dress (or triśūla horns?), standing under an ornamental arch, which appears to be made of the bent bough of a pipal tree. The lower ends of this bough are rounded up to form loops, each enclosing a star. The head of the deity is turned a little towards the right and on both arms he wears a number of armlets' (Vats, ibid., Vol. I, pp. 43-4). The device on one of the sides of a three-sided terracotta prism discovered at Mohenjo-daro, can be described thus: On the extreme right a horned figure with arms adorned with bracelets, standing between two pipal trees; on its left, a sacred goat decorated with garlands, recalling the scene explained by Marshall as the epiphany of the tree-goddess; beyond it a kneeling horned deity, apparently a goddess (cf. the long pig-tail), holding out her hands, a small offering table with something like a bird on it being shown on the extreme left (Mackay, ibid., Vol. I, p. 351; Vol. II, Pl. LXXXII, Nos. 1-c and 2-c). It is no doubt impossible for us at the present state of our knowledge to be sure about the exact significance of this scene, but we shall not be far wrong if we find in it also the representation of a mythology associated with a tree-spirit. The scene on an amulet noticed by Mackay (ibid., Vol. I, p. 355, Pl. XC, 23-b) may be referred to in this connection. Two men are shown, each carrying a tree, torn from the ground, with their roots clearly visible; perhaps, the men are about to transplant the trees for the abode of a spirit who is depicted in between the tree-carrying figures; the leafy nature of the arm of this spirit really represents the armlets of the divinity. Mackay has cited an interesting parallel to this scene in that of the Purāṇic story of the Yamalārjuna trees which were uprooted by the child Kṛṣṇa, thereby releasing the two spirits confined in them. We find its iconographic presentation in reliefs of the late Gupta period and afterwards, and it has been suggested by Mackay that it owed its origin to a similar myth of a much earlier date.¹

c12.31

¹ The two Arjuna trees were really the two sons of the Yakṣa king Kubera, viz., Maṇigrīva and Nala-Kubera, who were cursed by Nārada to be changed into trees. Kṛṣṇa released them from this accursed existence by uprooting the trees. The scene on the Mohenjo-daro amulet is somewhat different from its purāṇic counterpart, inasmuch as, in the former, two persons instead of one are shown with the uprooted trees in their hands.

c12.32

The above survey of a few representative seals of the Indus Valley has partially acquainted us with the nature of the beliefs and practices of the pre-historic people of India in that region. Several conclusions have been drawn about the iconographic presentation of some of their gods and goddesses after a careful study of the devices appearing on the seals and amulets; the nature of these conclusions, however, is still a tentative one to a certain extent. As I have said in the second chapter, the unravelling of the mystery of the script and language of the seals, if it is ever unravelled at all, will shed more definite light on the problem.² Marshall makes the interesting remark about the representation of the Indus Valley divinities of the remote past that ‘the people of Mohenjo-daro had not only reached the stage of anthropomorphising their deities, but were worshipping them in that form as well as in the aniconic’; for, the highly conventionalized type of the image of what he justifiably describes as the prototype of Śiva-Paśupati, ‘its stylized details and the fact that the kindred image portrayed on the faience sealing is being worshipped by the Nāgas clearly point to its being a copy of a cult idol.’¹ The decoration (cf. the armlets, head-dress, etc.), the sitting posture, the mode of showing the hands, the horns on the head, etc., appear also in other figures, some of which may depict the different aspects of the same god. The nude goddess, either in association with a tree or not, with some of the above characteristics, is shown as an object of veneration. Many composite human and animal figures found on the seals and amulets very probably stand for divinities in their theriomorphic or therio-anthropomorphic forms, though many others are to be regarded as mere accessories. Most, if not all, of the above types of figures appear to have been based on actual icons of cult gods which were being worshipped by the people in those days.

c12.33

² In the second chapter of the first edition, I had hesitated to endorse fully the conclusions of Marshall, Mackay and Chanda. But I subsequently went to Harappa and studied the seals and other antiquities on the spot. I could then accept many of their findings.

c12.34

¹ M.I.C., Vol. I, p. 56. Italics are mine.

c12.35

The most interesting fact to be noted in this connection is that the ideology which seems to underlie many of the above divinities corresponds very little to the concept of such Vedic deities as Indra, Mitra, Agni, Varuṇa and others. It is true we cannot describe the former as so many Hindu divinities and their representations as those of so many Hindu gods, yet it can be suggested that they contributed a great deal towards the formation of the concepts underlying some of the later Hindu deities. The apparent reproductions of mythical scenes on these prehistoric objects might also have contained the germs of different mythologies of the later period. It is not suggested, however, that the myths current about many of the Vedic gods and the anthropomorphic conceptions underlying them had nothing to do with the shaping and development of a good many of their epic and purāṇic counterparts. I have already written about the great part which the former had to play in formulating the various god-concepts of later times; this can also be fully demonstrated in any work dealing with the images of the various Hindu gods and goddesses. But what I want to emphasise here is that the Vedic traits of the latter, especially in the case of some of the cult divinities, were really superimposed on their primitive pre-Vedic core. As the Vedic period was far nearer to the epic and purāṇic times and as copious literary data of the former age are available to us, we can trace out the analogies and influences with more certainty. Further researches and excavations in various old sites of India, let us hope, will supply us with more clues and links of the intervening period, that are now missing, which will enable us to connect the Indus Valley evidence with the epic and the purāṇic data with more definiteness. Even the changes in the Vedic beliefs and practices of a date later than that of the early Ṛgvedic hymns, as has been suggested by me in the previous sections of this book, were brought about by the rites and customs of these prehistoric people of India.

c12.36

Seals and seal matrices with devices of an iconographic character on their surface, which can be dated in the Maurya or the Śuṅga period, are very rare. The small stone discs with the figures of the Mother-Goddess (Earth Goddess?) carved around their central hole have already been mentioned by me in connection with the ring-stones discovered in the sites of the Indus Valley; but they cannot be described as so many seals. Numerous terracotta seals, however, with Hindu divinities and their emblems on them, have been unearthed in two of the old sites in Northern India, viz., Basarh and Bhita, which are of great archæological interest; these mostly belong to the early and late Gupta period, a few being of a still earlier date. To these will have to be added the terracotta seals of the Kushan and Gupta periods found at Rajghat near Banaras (a few in this lot even go back to the Śuṅga date, though they do not bear any iconographic device); some of them bear representations of deities and their emblems. Many terracotta seals were also unearthed at Nalanda, some of which are of unique interest from iconographic point of view; they, however, mostly date from the late Gupta period and afterwards. Different purposes were served by these seals : some were attached with a string to letter tablets ; others were royal, official or mercantile guild tokens meant for the use of their servants and followers ; a few of them again were undoubtedly manufactured for the use of the heads of religious establishments and their retainers, and a vast number were also the sealings of private individuals. It has been suggested that as a large number of such seals (over 700) were discovered in one single spot at Basarh, it is likely that the seal matrices were manufactured there ; so many impressions—sometimes double, triple, and multiple—on a single lump of clay denoted that the former were being tested in that way. The finished seals were usually made of clay, perhaps prepared according to one of the processes to be mentioned in connection with the manufacture of terracotta images in the next chapter. Most of them were burnt after they had received the impressions from the particular seal matrices, some being very lightly burnt, while a few others were merely sun-burnt. Many of the above varieties of seals bear the figures of several Brahmanical gods and their emblems, the former being fewer in comparison with the latter. Sometimes, only the name of the cult-deity accompanied with some auspicious symbol is engraved, without any impression of his iconic figure or emblem, while at other times different emblems in varieties of combinations make their appearance. In many cases, there is a characteristic connection between the name of the issuer and the deity or his emblem or emblems reproduced on the seals, as we find on some coins of the Pañcāla series. One thing, however, is quite evident from our study of representative specimens from Bhita and Basarh, *viz.*, that even when the Brahmanical cult-gods were being iconically represented, they were comparatively infrequently used in the terracotta seals, where copious use was made of the varieties of their emblems. Again, it is highly probable that an emblem which, in its association with others, would belong to one particular cult, may, when depicted singly, be connected with another. Thus the conch-shell with wheel and other emblems is undoubtedly Vaiṣṇava in character, but when appearing alone may sometimes denote the *śaṅkhanidhi* of Kubera, a very appropriate symbol for merchant guilds and bankers.

c12.37.1

Siva and his emblems are found depicted on the seals in various ways. I have already referred to the representation of him in his *liṅga* form between two trees with the legend ‘ *pādaśvara* ’ in the field in Gupta characters, which I found in the collection of Dhir Sing Nahar of Calcutta. A pointed oval seal was discovered by T. Bloch at Basarh, which bears on it a Śivaliṅga with a trident-battle-axe symbol (Bloch simply says *triśūla*, but the combined *triśūla-paraśu* is quite clear from his plate), the legend in exergue below being *Āmrātakeśvara*, meaning the lord of Āmrātaka (Pl. X, Fig. 9). Now Āmrātaka is the name of a mountain; Bloch draws our attention to the eight Guhya *liṅgas* mentioned in the *Matsyapurāṇa*, viz., Hariścandra, Āmrātakeśvara, Jaleśvara, Śrīparvata, Mahālaya, Kṛmicaṇḍeśvara, Kedāra and Mahābhairava, which, according to him, were situated in Avimukta, *i.e.*, Banaras (*A.S.I.A.R.*, 1903-04, p. 110, No. 30, Pl. XL, 2). Bloch remarks about it: ‘ The letter to which it was attached must have been sent by the custodians of the temple of Āmrātakeśvara’ (*Ibid.*, p. 104). The oval seal (No. 39) in the same series (*ibid.*, p. 111) simply bears the legend *Nama Paśupateḥ*. The square seal matrix (No. 574) discovered by D. B. Spooner at the same site (Basarh) in 1913-14, and reproduced by him in the *Annual Report* of the year (Pl.

c12.37.2

XLIX) bears three symbols on the top section and the legend *Bañjulaka* in early Gupta characters in the lower one, the sections being separated by two closely parallel horizontal lines. Of the three symbols, the middle one is a *triśūla* with a short handle, that on the right ‘ resembling in shape the early Brāhmī character for *dhu* ’ is nothing but a longish water-vessel as seen in the hands of Siva appearing on the coins of Kanishka and Huvishka, and the other on the left ‘ looking like *ra* ’ is but a short staff as is placed in one of the hands of the same deity appearing on some of Huvishka’s coins. So, these are nothing but the three attributes usually held by Siva. A fragmentary sealing or seal impression of the early Gupta period found by Spooner at Basarh (ibid., pp. 121, 150, Pl. L, No. 672) shows ‘a very roughly sketched bullock running to right with the crescent moon above’ (the suggested reading Māradatta cannot be supported if one refers to the plate, and I can suggest no other reading as the plate is too indistinct); this is, of course, nothing but Śiva with crescent moon (Śaśāṅkaśekhara) in his theriomorphic form (Nandin). The unique seal impression (ibid., p. 129, No. 84, Pl. XLVI) shows on the upper edge of its slightly concave surface a small conventional śaṅkha in outline and a very good humped bull recumbent to left in the middle of the field; the legend is Rudradevasya.

First 47 of 93 passages. Read on

Linked texts (20)