Distinction in Indian Iconography [Part I, 1]

Posted: 02.01.2012
Updated on: 02.07.2015


The paper was published in Bulletin of the Deccan College Research Institute 20 (S.K. De Felicitation Volume), Poona 1960, pp. 164-248.


Distinction in Indian Iconography

Part I: The Phenomena


I. Stereotype Representation in General

Assimilation. [1] Different elements of the representation have often become similar in shape. This is particularly true of “hand-attributes” (§ 1). But cases of assimilation are also found amongst “head-attributes“, vegetable objects, etc. (§ 2).


§ 1. The hand-attributes which are going to be discussed can be divided into two groups and have been reproduced as figs. 17-21 and 22-26.

(Figs. 17-18). On the other walls of the Telī-kā-mandir (Gwalior) there appear 112 figures of small size. About 50 of these wear a jaṭā and carry a trident or a mace in one of their two hands. Whereas the distinction between trident and mace is always maintained, we find besides the normal form of the trident (fig. 91) other forms where this weapon can only with difficulty be distinguished from a spear (fig. 18). But a real spear is carried only by one of the 112 wall-figures, viz. by Skanda (fig. 17).

(Figs. 19-20). Fig. 19 shows a spear which is carried by Skanda (Gwalior Museum) and fig. 20 shows a sacrificial ladle (sruva) which is carried by Brahmā (Jhalrapatan). Again the difference lies only in a scratched line appearing on the second attribute and indicating the cavity of the ladle. One can imagine that the artists have sometimes omitted even this thin line although they meant to show a ladle. In the case of Agni we find this very attribute with line (Agni from Suhania [2]) and without line. If the line is clearly indicated we can call the object a ladle. Otherwise the decision is difficult because the texts mention in their description of Agni spear as well as ladle. In my Khajurāho article I had to leave the matter undecided. The attribute was once described as “gadā (śakti? sruk?)“ (p. 11: Agni on the outer walls of the temple) and once as “śakti (sruk?)“ (p. 23: Agni on the outer walls of the garbhagṛha). [3]

(Figs. 21 and 23). Our drawings show two forms of the lotus - bud, one without calyx (held by an Ambikā) and one with calyx (held by a Sītā). Both specimens belong to the South but similar forms are found in the North. At least the second lotus could be confused with the other attributes in its group.

(Figs. 22 and 24-26). Fig. 22 shows a mace carried by an unexplained male figure. - Fig. 24 looks similar to fig. 23 (note the slight bend!). It is however not a lotus but a stick, carried by Vāyu (from Suhania). The texts prescribe for this god a goad or a stick, never a lotus. The stick is actually only a goad without a hook, and the stick of the Suhania Vāyu corresponds to the goad in the hand of the Vāyu on the outer walls of the Pārśvanātha temple at Khajurāho. [4]


Suhania Vāyu:







Khajurāho Vāyu:








The next attribute (fig. 25) which is held by a Yama from Suhania is also bent, but only very slightly. The texts prescribe unanimously a stick (daṇḍa) for Yama. Therefore our artist must have meant a stick, not a mace, although this is not brought out by the form. The difference which we find between the stick of fig. 25 and the mace of fig. 26 (carried by a Viṣṇu from Padhavli) has only stylistic but no iconographic importance (see § 63). It does not seem that the distinction between stick and mace on which the texts insist was anywhere observed by the artists.

Assimilation of attributes is very common with figures which form a series; but as we have shown there are also many similarities which belong to the style as such: the conventional form of the spear coincides with the conventional form of the sruva, and so forth.

The distinction is rarely lost completely. A well-preserved attribute can normally be identified. But often the difference lies only in a few lines, incised in a core which is otherwise identical for both objects (compare figs. 17 and 18 or 19 and 20).


§ 2. The best example for the assimilation of head-attributes is the transformation of the jaṭā into a mukuṭa (see § 36 and figs. 62-59-60). The mukuṭa-like jaṭā can again be changed into an aggregate of flames (in the case of Agni, see Sivaramamurti, Pl. VIII, A). This does net mean that the god wears an ordinary mukuṭa which is burning, but it indicates a compromise between the normal jaṭāmukuṭa and the normal halo of flames surrounding Agni's head or his whole body (Sivaramamurti, Pl. VIII, B).

Often a figure is surrounded by four vegetable objects appearing in the four corners of the panel. The upper two objects represent a tree, the lower ones represent two plants. Sometimes all four are identical in shape. Our example (fig. 32) is rather primitive, but it shows that this type of assimilation occurs also in the iconography of the Jina in which case it is impossible to call the motif decorative.

Fig. 52 shows a very curious case of assimilation: the three layers of the seat of the Jina (from top to bottom: cushion with tassels, snake-coils, lotus-seat) have become so similar that at first sight they seem to be identical.


§ 3. Reduction. Here we shall discuss those cases where the distinctive feature of an object (or the most characteristic element of a composition) has become indistinct, either due to a reduction in size, in boldness of relief or in general scale, or due to a shift from the centre to the periphery.

  1. Reduction of the size. The plaits of hair which distinguish the first Jina Ṛṣabha from the other Jinas are sometimes extremely short. The normal size is shown by fig. 35, whereas figs. 33 f. show the plaits in reduced size. Similarly the hump of the bull (which is Ṛṣabha's cihna) is sometimes hardly indicated, so that the animal cannot be distinguished from other quadrupeds. See also the next §.
  2. Incision. Sometimes not only a single feature of an object (§ 1) but the whole object is incised. Fig. 31 shows a tortoise (vāhana of the river-goddess Yamunā); head and neck are still modelled in the round but the rest is merely incised (unfortunately the drawing has not made this point very clear). This treatment is not uncommon in the case of the vāhanas of the river-goddesses (Gagā and Yamunā). The cihnas of the Jinas are also very often incised. This is natural in the case of abstract symbols like the svastika, but in the case of the animal cihnas a representation in relief would have been better. Occasionally the cihnas are drawn very, artistically but as a rule cihnas in relief are clearer.
  3. Reduction of the scale. Small size does not affect the clearness of the representation if the art of miniature - carving is fully developed and suitable stone available. But often subsidiary figures of an image (e.g. the yakṣas and yakṣīs of a Jina) or even all the figures of a composition (e.g. on a theriomorph Varāha) are executed in small size, although the artistic technique and the condition of the stone did not recommend this method.
  4. Shift of central elements to the periphery. See the next §.


§ 4. We shall now describe a few compositions which are typical examples of reduction and its various aspects.

Reduction of the individualizing elements in epic compositions can be demonstrated by a comparison of the Jetavana panel on the northern gate at Sāñcī with the representation of the same motif at Bodh Gayā (E. Waldschmidt, Buddhistische Kunst in Indien, Berlin 1932, Pl. 69 and 26). At Bodh Gayā the coins offered for the purchase of the garden (and which are essential for the identification of the scene) cover almost one half of the panel. But at Sāñcī they are arranged in two inconspicuous rows in the left lower corner of the composition. The difference in the representation which is to some extent typical of the difference between the art of Bodh Gayā (or Bhārhut) and the art of Sāñcī has been discussed op. cit. p. 50.

Good examples of the reduction of the individualizing elements in iconoplastic art (i.e. in non-epic compositions) are found amongst the images of Sūrya and of the cakravartin. Almost all Sūrya-images of the later time show his chariot and his horses in reduced form; see for example Stella Kramrisch, The Art of India, Pl. 74: Sūrya from Alampur; Rao I, II, Pl. XCI: Sūrya from Haveri. In the first case chariot and horses are shown in an unusual way behind the god and they do not reach above the height of his knees. In the second case the horses are only represented by an inconspicuous carving on the pedestal, i.e. they appear below the god and even below the panel proper. - We mentioned already in the Introduction that the attributes of the Jaina cakravartin are sometimes reduced to such an extent that he looks like a Jina. We shall now compare such Jaina images with the unreduced rendering of the motif at Jagayyapea (Waldschmidt, op. cit. Pl. 22). In the latter case the human ratnas are about half the size of the central figure, whereas on a Jaina image at Deogarh [5] their size is not more than a tenth of the size of the cakravartin Bharata. At Jagayyapeṭa the ratnas surround the main figure from all sides, whereas at Deogarh they are confined to the pedestal (and to the portion immediately above the pedestal). Only one of the cakravartin-images at Deogarh does not show this extreme reduction, but stands in the treatment of the ratnas midway between the specimen from Jagayyapeṭa and the other specimens from Deogarh. It is, however, of inferior quality and late in date. [6]


§ 5. In the previous §§ we have practically neglected the context of the elements compared. But it goes without saying that the impression of similarity is intensified if the different elements take corresponding places in formally related compositions. We can call this process substitution: different compositions follow the same scheme, and one and the same place (Stelle) of the scheme is occupied in turn by different occupants (Stellenbesetzern). Sometimes the different occupants are related through the place alone (e.g. female chowrie-bearer: Ambikā; §13), sometimes also through the form (e.g. bhāmaṇḍala: jaṭāmaṇḍala; § 6), and sometimes through the form and through the function (e.g. lotus: parasol; § 7).


§ 6. Substitution of elements of the same form. Well-known are the various substitutes for the halo. In the list given below we have included, this time, pieces of different time and provenance in order to render the survey as complete as possible. But the specimens are treated and arranged as usual without reference to their mutual historical relation (see § 67). Besides the simple disc (plain or with decorative patterns) and besides the round of snake - hoods we find the following circular objects:


Fig. 42:

circle of ordinary hair.

Fig. 43:


Fig. 44:

appendix of the mukuṭa (cf. fig. 60 f. and Sivaramamurti, Pl. XVIII, D). [7]

Fig. 45:

cakra with spokes. [8]

Fig. 46:

flames. [9]

Fig. 47:

śiraścakra (in the form of an āmalaka) representing the Sudarśanacakra. [10]

Fig. 48:



The next example is taken from Siva's Dakiāmūrti. Fig. 49 shows its normal form; here the hair of the god is represented as a bulky mass which projects to the left and to the right of the head (jaābhāra). It is rather strange that in one specimen (fig. 50) the tree which is often found with the Dakṣiṇāmūrti appears on the very place where we expect the monstrous jaṭābhāra, while the jaṭās are reduced to a few inconspicuous plaits.

In fig. 88 we have reproduced one out of three panels which belong together and show different aspects of Narasiṁha. The god appears in a pillar just as Śiva does in his so-called Lingodbhavamūrti (fig. 89). There is certainly some similarity between the Narasiṁha-myth (N. coming forth from a pillar) and the Śiva-myth (Śiva manifesting himself in a liṅga), but this alone cannot account for the close relation between the two images. Our Narasiṁha-image which seems to be unique must have been modelled from the type of the Lingodbhavamūrti.

Sometimes a curved or outstretched animal (or man) is held like a rod in the two uppermost arms of a god. [11] Thus we find a curved snake, elephant, or man which form, together with the arms, a semi-circle above the head; [12]in one case we are told that the snake has the head of an elephant [13] and this would be a good example for the overlapping of stereotype representation with amalgamation. Other animals are held above the head like a straight stick: snake, alligator, elephant (or rather the upper edge of the elephant's skin). [14] A Cakreśvarī from Deogarh [15] holds even a cakra in the two (!) uppermost hands - an unusual and unnatural arrangement which must have been inspired by the. other cases mentioned.

A slab from Deogarh [16] shows 24 Jinas in small size. Below some of the Jinas a simple lotus is shown, (stylistic transformation of the padmāsana), while we find below the remaining Jinas a minute pedestal (in low relief) with a lotus-flower or a cakra on it. Were not the cakra shown from the edge and the lotus-flower from the face, we could easily mistake one for the other.

Substitution also takes place between a certain type of vegetable object and a peculiar form of the scarf. The vegetable object which we are referring to appears in figs. 27, 29, and 84, and is probably different from the “padmapāi-motif“ mentioned in the Khajurāho article (ibid., pp. 28 ff.; figs. 32 and 87 of this article). The ends of the scarf have often been stylized into heavy round objects. [17] The peculiar stylization which we find in Central India (figs. 28, 30, 31) - scarf-end resting on the ground and rolled up to form a cylindrical object - resembles the horizontally fluted mace which is found with early Viṣṇu-images, [18] but the question whether both forms are historically connected or not remains open. The rolled up scarf does not seem to be much earlier than the medieval period.

Both objects (plant and scarf) are similar in shape and occupy about the same place in the composition. They are found either on one or on both sides of the figure. The first case is demonstrated by two panels from the Telī-kā-mandir (fig. 29: fig. 30), the second case by two panels from temple No. 12 at Deogarh (fig. 27: fig. 28). It is possible that the scarf-ends in fig. 28 did not receive the finishing touch of the sculptor; the outlines of the object are however clear.


§ 7. Substitution of elements of the same form and function. We often find as a substitute for a genuine implement some other object, animal, vegetable or lifeless. The snake for example appears as a churning-rope (Vāsuki in the Kūrmāvatāra), as a noose (nāga-pāśa), as a kucabandha (Sivaramamurti, p. 32 and Fig. 16), as an upavīta (Pheru, p. 181: Tumbaru Yakṣa of the Digambaras), as a seat or bed (Viṣṇu on Ananta), as the back of a throne, [19] as a festoon decorating a stūpa (Amarāvatī), and so forth. One and the same implement can be represented by different substitutes: a mountain (Meru or Mandara), a Śiva-liṅga, the trunk of a tree, a lotus-stalk, and similar objects may serve as the churning-stick in the Kūrmāvatāra. [20] In a case like this it may become difficult to decide whether the substitution goes to the credit of the artist or is a feature of the myth. The custom of manufacturing ornaments and utensils in the form of animals or of providing them with animal protomes etc. presents a similar problem. If such objects are represented in art, we may not be able to decide whether some artefact is represented or whether the sculptor himself introduced an animal as a substitute for some implement. Bracelets in the form of a snake were, for example, in actual use.

Still there are many cases where the innovation goes clearly to the credit of the artist. Figs. 40 and 41 show the two parasol - bearers of the two river - goddesses on one of the smaller door - ways of the Tell-kā-mandir. In fig. 40 the attendant figure carries, instead of the parasol pro­per, a lotus. - Whereas the artists could choose between parasol proper and lotus in the case of the river - goddesses, we find on representations of the hybrid Varāha invariably the. lotus - parasol. It appears also above Matsya and Kūrma [21] and we can therefore conclude that it is connected with aquatic motifs.

Fig. 38 shows a crown held by two celestials above a Hindu goddess, and fig. 39 shows a parasol (without stick) held by two celestial garland-bearers above a Jina. We find also an image where a lotus is held by two celestials above the main figure. [22] But cases of this type are rare. Normally the parasol or the lotus above the head of the saint (deity) are flanked, but not touched by celestials. At least in those cases where the object is carried by flying genii we can call it a substitute for the crown (which is rare in Hindu iconography and altogether absent in Digambara iconography).


§ 8. A sound balance between individualizing elements and general elements is not always maintained. Often the general elements become so rampant that the individualizing elements are squeezed out or rendered inconspicuous.

The panel at Sāñcī on which the donation of the Jetavana is depicted (§ 4) shows practically only trees, sanctuaries and worshippers - elements which are not in the least peculiar to the story represented. - The Jinas are often surrounded by a host of smaller figures and it requires some effort on the part of the observer to discover amongst these the yakṣa and yakṣī which help to identify the Jina. The other figures are common to all Jinas (or even common to Jain and Hindu iconography). - Profuse ornamentation can conceal the true character of a hair-dress: the lower halves of the jaṭās of the trimūrti at Elephanta are covered completely with ornaments so that they look nothing like the hair-dress of an ascetic.

Other general elements which occur in iconoplastic art are the various forms of the lotus (§ 14), the two forms of the vegetable object mentioned in § 6, the nāgas (§ 28), the “sacred tree“ (§ 14), etc. The elements mentioned are most misleading as long as they have not yet become “general“ (i.e. common to a large number of different motifs). When they have spread from their original motif only to a few others, they seem still to constitute integral parts within each given composition (nāga in the Varāhāvatāra etc.). Such cases will be discussed in more detail in the chapter on “silent amalgamation“. Here we are only concerned with the negative aspect of the really general elements; we want to show how they distract attention from those elements which are relevant to the identification of the image.


§ 9. Iconographically speaking an inscription which gives the name of the figure represented can be called an element of the composition. In any case there is not much difference between the incision of a cihna (a sort of hieroglyph) and the incision of a few letters. Inscriptions of this type are numerous at Bhārhut but completely absent at Sāñcī. They reappear in Kuṣāṇa art but become again rare in the medieval art of the North. Hardly one percent of the Jina images at Deogarh have got an inscription which mentions the name of the Jina, while every fourth or fifth image shows the name of the donator. The cihnas of the Jinas are also often missing. The incision of the name was however common practice in the case of certain goddesses like the Jain yakṣīs, the Jain vidyādevīs, and the Hindu yoginīs. The yoginīs from Bheraghat and Naresar and the Jain goddesses from Deogarh temple No. 12 have their names inscribed on the pedestals. Often these names are not in agreement with the general tradition. Probably some of the statues were simply misnamed or provided with names which occurred only in the vocabulary of a small community (cf. also § 46).

Share this page on:



Bulletin of the Deccan College Research Institute 20 (1960)

Compiled by PK