Distinction in Indian Iconography [Part I, 7]

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


VII. Related Phenomena (1)

§ 40. Multiplication. We use this expression if the same form is repeated on one and the same piece. Taken in its strict sense it means that the form is repeated without any change (§ 40). Multiplication in the wider sense means that different figures in a composite scheme (§ 41) or in a series (§ 42) become similar through the repetition of certain attributes.

As is well known we find on the surfaces of theriomorph Varāhas, of stūpas, pillars, liṅgas, etc. very often a great number of identical Buddha-, Jina-, or Ṛṣi- figures or (in the case of liṅgas) of miniature liṅgas. We come also across smaller rows of five, six, seven or more figures, and in this case the row may also consist of small Śivas, Viṣṇus, Sūryas, Gaṇeśas, etc. (cf. for example MASI, 23, Pl. XXXV, a). The decision whether such a series has a dogmatical background or not, is sometimes difficult. We know that there are 24 Jinas in all (in our avasarpiṇī) and that any other number (except perhaps the integral multiples) serves only aesthetic requirements. But to decide if a series of five Gaṇeśas or twelve Viṣṇus exists in the dogma we need an intimate knowledge of Hindu literature. Identical figures appear also in composite schemes. In this case symmetrical arrangement is combined with hieratic scaling: the biggest figure is seated in the centre and the smaller ones appear on the periphery. Such schemes are preferably used for the representation of Buddhas and Jinas.


§ 41. The central figure of an image may share certain attributes with the two smaller figures at its sides. The Gupta Śiva at Mandasor (Sivaramamurti, Pl. XIX, B) is flanked by two smaller statues, both of which wear a jaṭā and carry a trident and a lotus in their two hands. This concords with the iconography of the main figure (Śiva's lotus is broken but can be restored); only the ūrdhvaliga and the third eye of Śiva are absent in the case of the two minor figures.

A many-armed standing deity at Rajshahi (Banerjea, Pl. XXVI, 2) brandishes a sword in one hand. This is the most conspicuous feature of the god and it is repeated in two small seated figures to the lower left and right. But this example must be taken for what it is worth. Perhaps we have to imagine a third and fourth subsidiary figure with a sword to the upper left and right where the slab is now broken (cf. A. Grünwedel, Buddhistische Kunst in Indien, Berlin 1919, p. 175: Bodhisattva Mañjuśrī). If the figure is actually related to Bodhisattva Mañjuśrī the multiplication may have a literary background (cf. Grünwedel, loc. cit. p. 176). Chowrie-bearers to the left and right of Pārśva often show a single snake-hood; the two figures on the sides of Sūrya often carry in one of their two hands a lotus of the type found as attribute of Sūrya himself.

Descriptions of attendants or subordinates who are similar to their masters are numerous in iconographic literature. But in some cases (Pārśva!) it can be said at once that the similarity was produced by the artist.


§ 42. Members of a group, however different in origin, may become so similar that they look like repetitions of the same archetype. Rao himself states that sometimes all the mātkās (of the Saptamātkā- or Aṣṭamātkā-group) are provided with a child although this is not prescribed by the texts (Rao, I, II, p. 389). Here the sculptures are probably more faithful to the original tradition than the texts available to Rao [1]. In other cases, however, we are no doubt concerned with multiplication by the artist and not with multiplication by ancient mythologists. All the lower arms of the Saptamātṛkās at Kumbhakonam (Rao, I, II, Pl. CXIX) show the same characterization: the proper right hand displays the abhayamudrā and the proper left rests on the knee. This is clearly a later innovation.


§ 43. Multiplication of hand-attributes is to some extent analogous to the multiplication of figures (§ 40-42). According to Nirvāṇakalikā (p. 37/2), the vidyādevī Apraticakrā carries in all her four arms a cakra. This description is confirmed by the representation of the goddess on the outer walls of temple No. 12 at Deogarh (Photo No. 1123). It is also possible that different hand-attributes of the same figure or of the same group of figures are provided with similar appendices. All the attributes carried by the Saptamātṛkās at Belur (Rao, I, II, Pl. CXVIII, 2) consist in a handle (simple or twofold) and the object proper. Like the prongs of the trident, the conch is attached to a handle, and as a consequence all the attributes look more or less alike.


§ 44. Multiple adoption. The inclusion of a particular figure in different religious systems and the repeated inclusion in one and the same system may be called “multiple adoption“. Brahmā for example appears thrice in the Jain pantheon: as Brahmayakṣa (belonging to the 10th Jina), as Brahmaśāntiyakṣa, and as dikpāla Brahmā. Sometimes only the name is adopted (and used for a different iconographic type), sometimes only the type (and used under a different name), and sometimes both together are adopted.

The yogi-type which occurs in Buddhism, Jainism, and Hinduism has already been mentioned in § 17. Another type is characterized by the water-jar [2] in the proper left hand and the rosary in the proper right hand. It serves not only for the representation of ascetics but appears also in the case of numerous deities:

  1. Representations of the Navagrahas in general. They include several figures which belong to our type.
  2. Śiva on a Śiva-Pārvatī image from Kosam (Banerjea, Pl. XXXVIII, 2; the rosary is not clear in the reproduction).
  3. A “nāga“ from Nālandā (J.Ph. Vogel, Indian Serpent-Lore, Pl. XIV).
  4. Agni from Bihar (Sivaramamurti, Pl. VIII, B).
  5. Central figure of a trimūrti from Padhavli (Banerjea, Pl. XXXIX, 3).
  6. Jain god on a double-image (the other figure on the same slab represents Pārśvanātha). Jain temple at Golakot; Photo No. 425.
  7. Female figure with five snake-hoods on the left outer wall of the Mālādevī temple at Gyaraspur.
  8. A Buddhist bronze from Nepal (Bodhisattva?; Grünwedel, op. cit. p. 164, Fig. 93).

This simple iconography has often been elaborated by adding further arms with attributes, e.g. in the following forms:

A. Nārayaṇa at Deogarh (fig. 54):







B. Goddess from Kota (now in the Gwalior Museum; photo No. 1773 of the Museum):






water – jar.

C. Goddess from Amarkantak (MASI, 23, Pl. LII, a):






water -jar.

Another familiar type shows the hand-attributes varada (proper right) and lotus (proper left). This was adopted in Buddhist iconography (for certain forms of Tārā and of the Bodhisattva), in Hindu iconography (Sivaramamurti, Pl. VII, D: “Pārvatī“), and in Jain iconography (fig. 3; the proper right hand of the figure shows the varadamudrā). In the last case it is fairly certain that the adoption goes to the credit of the artist and that we cannot give any name to the figure. The same is true for the Jain variant of the “couple tutélaire“. The identifications proposed so far are only guess-work: “happy twins of the suamā-era“, “yakṣa and yakṣī“, “parents of Jina“. A textual reference to the motif has not yet been traced and will probably never be found. The figures (Jain Tārā and “couple tutélaire“) are distinguished from their Hindu and Buddhist counterparts by a small Jina carved on top.


§ 45. At a comparatively late date, an iconographically indifferent forms of the couple-tutélaire-motif and of the Ambikā-motif have been employed to represent the parents and the mothers of the Jinas. The parents are represented on a ceiling panel at Kumbhāriā (Shah, Fig. 83). and the mothers on an independent slab in the Luna Vasahī temple at Ābu (Muni Śri Jayantavijayaji, Holy Ābu, Bhāvnagar 1954, p. 99 f.). In both cases each figure is identified by an inscription. [3] Sometimes we can only infer that the type serves as a substitute for something different. In fact, nothing prevents us from explaining the Jain version of the resting-queen-motif (originally probably associated with queen Māyā, the Buddha's mother) as a representation of the mother of a Jina. Even the panel in fig. 65 which has preserved the sāl tree of the Māyā-legend (see fig. 66) must be understood in this way. But actually the legend of the mother of the Jina has instead of the sāl tree a plantain arbour. [4] Besides, the bathing etc. in the plantain arbour takes place after the birth of the child, and both - child and mother - are bathed, not the queen alone as in the Buddhist legend. The other representations in Jain art, where child and tree are missing, can of course be referred to the time before the Jina's birth; but the panel of fig. 65 (where the tree is depicted but the child omitted) cannot be called correct from the point of view of Jain tradition. The resting queen of Jain iconography is nowhere identified by an inscription, but miniature Jinas represented along with her show that the icon is Jain.


§ 46. §§ 44 and 45 have shown that the correlation between the gods of the dogma and the various realizations of the iconographic types is extremely loose. Therefore, the study of Indian icons should be in the first place a study of the ramifications and interrelations of the iconographic types, and only in the second place an attempt should be made to coordinate the archaeological and the literary evidence.

We should also not be misled by the artists, who can easily settle the identity of their creations by inscriptions, by cihnas, and even by single attributes which have historically and dogmatically speaking nothing to do with the respective iconographic type. We cannot dismiss such evidence but we have to remember that identifications based on stray characteristics are of little value.


§ 47. Multiplication in literature. In Indian theology every god can be multiplied by what is best described as a “change of the coordinates“. The list given below may be read as an appendix to §§ 40-46.



Change of time:

Different Jinas in different periods.


Change of place:

Different Jinas in different parts of the world.


Change of function:

Śiva as God: Śiva as dikpāla Īśāna.


Change of the sphere of existence:

Buddha : Dhyānibuddha.


Change of dharma:

Vaiṣṇavī in Hinduism: Cakreśvarī in Jainism.


Change of incarnation:

Different avatāras of Viṣṇu.


Change of sex:

Viṣṇu : Vaiṣṇavī.


Change of nature:

Jina as human: as nāga (Pārśva)


Change of rank:

Śiva's iconography partly repeated by the human form of Nandin (§ 22).


Change of iconography:

Viṣṇu's caturviśatimūrtaya (§ 16).


In certain cases the newly created forms are mere shadows of their prototype (Jinas on other continents!), in other cases figures of different origin form a series by virtue of a common factor (e.g. the Viṣṇu-element which recurs in each avatāra). [5]


§ 48. Convergences. Since certain “elements of stylization“ and certain “form-fragments“ are ubiquitous in Indian art unintended similarities between different forms are but natural.

We call the slight bend appearing in many handles and sticks (figs. 24 and 90) an element of stylization. The bend is probably due to the influence of vegetable forms, especially of lotuses. The opposite case is also possible. Curved objects like the kākapakas [6] of Skanda (fig. 83) and various vegetable forms (figs. 82 and 84) are often straightened. - One and the same form-fragment, most probably of vegetable character, is found with the three attributes reproduced in figs. 77-79 (trident, spear, conch).

The spreading of certain elements of stylization is stereotype representation in its least complex form; the spreading of certain form-fragments is amalgamation in its least complex form. It depends always on the peculiarity of the case in question whether these convergences give rise to confusion or not. A weapon with a bent handle can be mistaken for a lotus (fig. 24), but a tuft of hair can never be confused with a chowrie (fig. 83). There is no clear-cut line of demarcation between intentional similarities as discussed in § 1 ff. and convergences, but in the cases described below intentional assimilation is excluded by the chronology or it is otherwise unlikely.

The similarity between the lotus in fig. 74 (placed between the legs of the Jina) and the stylized folds of the lower end of the garment in fig. 75 (appearing between the legs of the Buddha) is evident. It is possible that in Digambara iconography the lotus, originally the āsana, was reduced in size in order to have a substitute (§ 6) for the “dhotī-mark“ (slight indication of the lower end of the dhotī of the seated Jina; occurs in Śvetāmbara iconography and is related to the stylization of the robe of the Buddha as demonstrated by fig. 75). But the rosette shape of the folds cannot have been influenced by the lotus because the lotus was introduced at a much later date. It is the general tendency to lotus-like and cakra-like stylization which determined the shape of the folds. - Ṛṣabha's long hair in fig. 80 is similar to the snake behind Pārśva in fig. 81. Normally the snake-coils behind Pārśva are arranged horizontally and it is indeed possible to show the gradual transition from the normal realistic representation of the horizontal coils to their stylized rendering as seen in fig. 81. But it is also possible that the undulations on the sides of Pārśva do not indicate the hairpin bends of the tail of the snake piled up behind the Jina, but represent the wavy bodies of two snakes which rise vertically from the ground and join above the Jina's head (cf. Lalit Kalā, Nos. 1-2, 1955/56, PI. XII, fig. 11: early medieval Pārśvanātha in bronze from Vasantagah). - The middle portion of the crown of the tree in fig. 86 looks like a ḍamaru, and the lower portion of the creeper to the right in fig. 87 looks like a thunderbolt. Both objects show the same element of stylization, viz. a constriction, which occurs also in fig. 85 (viśvapadma, cf. § 60) and in fig. 87 (lower portion of the creeper to the left).

Share this page on:



Bulletin of the Deccan College Research Institute 20 (1960)

Compiled by PK