The Grammar of Jina Iconography II [Part 6]

Posted: 29.02.2012
Updated on: 02.07.2015

The essay was published in Berliner Indologische Studien No. 13/14. 2000, pp. 273-337.


§ 6. The Real Pantheon

Jaina gods and goddesses are an important element of Jaina iconography. Since we have invalidated the traditional system of YY, we are now forced to supply another frame for the Jaina deities. Hence our concept of the “real pantheon“ which will be described below.

The “real pantheon“ flourished mainly in the 11th through 13th centuries. We have already discussed the Jaina couple (§ 2), Ambikā and Kubera (either combined or separate: § 3-5), and Cakreśvarī (§§ 4-5). Sarasvatī (infra) and the Jaina Tārā (infra) are further members of our pantheon concept.

Not a “real pantheon“ but an extended pantheon was already in the minds of the ancient authors. Various deities appear in supplements to the old pantheon lists. In the Oghaniryukti manuscript of A.D. 1161 we have, in addition to the sixteen mahāvidyās, various other deities: Ambikā, Lakṣmī, Sarasvatī, Brahmaśānti Yakṣa and Kapardi Yakṣa (Chandra We: figs. 17-42, esp. figs. 33-37, pp. 29-30). The Nirvāṇakalikā describes after the 48 YY on folio 37b the Śrutadevatā (Sarasvatī) and the Śāntidevatā (Shah Mi: pp. 281ff., esp. p. 283). The manuscript Ka (fig. 12) describes after the YY the deities Sarasvatī and Brahmayakṣa. In Cave 7 on the Khandagiri the series of seven Jinas (with seven yakṣīs) is preceded by Gaṇeśa (Mitra Śā: 128 and pl. 1A).

From the point of view of Jaina philosophy, all the 24 Jinas have the same rank. However, in the Paryuaā Kalpasūtra a difference is made between nos. 1 and 22-24 and the remaining 20 Jinas (JRM: 83). The group of four is treated in “prose“ (long or short biographies), whereas the block of twenty is treated in tabular form. In art we also find special emphasis on certain Jinas (Ṛṣabha, etc.), but there is no fixed group of two or three or four (etc.) “important“ Jinas. There was, however, a late tradition that the Jinas of the “Kalpasūtra tetrad“ (and the four associated yakṣīs of the system) were especially important. The Rūpamaṇḍana (15th cent.; Hingorani Rū: 9) assigns a special position to the said Jina tetrad and to the corresponding yakṣī tetrad: Cakreśvarī, Ambikā, Padmāvatī, and Siddhāyikā. They are pūjyā viśeata (Rūpamaṇḍana 625-26). See Hingorani Rū: 26-27 and Shah Am: 166. The four goddesses are treated together in JRM: chapter 10. There is no tetrad in art, however (supra). The status of Cakreśvarī and Ambikā requires no comment. The position of Padmāvatī is not clear (§ 4); Siddhāyikā occupies no place, at least no easily detectable place in Jaina iconography. We do not know whether there are historical links between the Kalpasūtra and the Rūpamaṇḍana or not.

The present section considers only Ambikā (independent), Cakreśvarī and Sarasvatī. The couple, the K-and-A constellation and Kubera (independent) have already been completed (§ 2 and § 3); the Jaina Tārā will be included in Grammar III/IV (§ 1: Preview). The splitting of the treatment of Ambikā, Cakreśvarī and Sarasvatī (§§ 2-5: § 6) was unavoidable and in keeping with the general pattern of our article (§ 3 supra).

Gomukha (§ 4), Brahmaśānti Yakṣa (Shah Br, Tiwari Br), Kapardi Yakṣa (Shah Br) and Gaṇeśa (supra; Cort Ga; Bhattacharyya Ga [photo unclear]; Tiwari El: figs. 43-45 and pp. 110-112; JRM: 63) are not counted by us as members of the “real pantheon“.

As mentioned already (§ 5), the three goddesses Cakreśvarī, Ambikā and Sarasvatī appear on the outer walls of Deogarh Temple No. 12. Since this fact is of general interest, we have reproduced them for a second time (figs. 15-17). However, the three representations do not contribute substantially to our knowledge of the history of Cakreśvarī, Ambikā and Sarasvatī respectively. We refer the reader also to Cakreśvarī and Ambikā as represented in the two early Khandagiri series. See Mitra Śā: pls. 1A and 2B (Cave 7), 3A and 5A (Cave 8).

The system includes Ambikā, Gomukha, and Cakreśvarī as the only deities who exist in art. As we know, Ambikā appears much earlier than the YY. Gomukha (Jaina Gomukha) and Cakreśvarī surface at about the same time. Cakreśvarī is identical with the mahāvidyā no. 5 (we use the name Cakreśvarī for both goddesses); see Shah Ma: 119. Padmāvatī and Dharaṇendra belong to the system but have no clear identity in art (§ 4). - Refer for Ambikā, Cakreśvarī and other YY to Misra Ya: 126-131 (general archaeological evidence).

We do not consider the evidence of the door-lintels. A more systematic study of the real pantheon (and possibly of the system) has also to consider lintels, pillars, etc. For example, compare a door-lintel from Khajuraho with Ambikā, Cakreśvarī and Padmāvatī: Tiwari Ma: fig. 4. - We now present Ambikā (independent), Cakreśvarī and Sarasvatī:

[A: Ambikā] In the cases of Ambikā and Cakreśvarī, the subsidiary forms have already been discussed (§§ 3-5 and §§ 4-5). The semi-independent forms of Ambikā at Gwalior (fig. 6), Ellora (Zimmer Ar: pl. 243 and PJA: figs. 160/164), and Karnataka have also been considered; see § 3. Refer for Ellora Ambikās in general to Pereira Mo: 120, 124, 128, etc.

Ambikās differ in character even in the same region (refer in a general manner to Tiwari Am and Tiwari Ut). A short list of independent Ambikā images may nevertheless be instructive.

We start with a few special images in Northern and Central India. An early Ambikā image is pal Ex: 35 (“6th cent.“). It is surprising that this goddess is four-armed, Ambikās with four arms being otherwise hardly known before the medieval period. Other early Ambikās (both two-armed) are reproduced in Pal Ex: 176, Shah Ak: 14 (bronze) and in fig. 20. An Ambikā from Mathurā has Gaṇeśa and Kubera to her right and left as attendant deities; the image also has a socle frieze (Tiwari Am: fig. 37 = Cort Ga: 84). The Ambikā from Bihar published in JAA (pl. 91A) has a Jina in her hair-dress (cf. Pal Ex: 178).

A well-known four-armed Ambikā is the goddess from Dhar (A.D. 1034). K. Mankodi has shown that this goddess is Ambikā, and not Sarasvatī as thought previously (Mankodi Am). Our legend below JID: fig. 254 must be corrected accordingly.

The general list follows (stone images if not mentioned otherwise). TiwariAm: fig. 48 (Mandasor); Dhaky Śā: pl. 17 (Huca), Shah Su: fig. 38 (Bengal, bronze); Pal Ex: 177 (Orissa); Shah Su: fig. 37 (Mudbidri, bronze); Tiwari Am: fig. 1 (Meguti temple, Aihole, blurred); PJA: fig. 75 (Kalugumalai, rock-cut); PJA: fig. 95 (Chitaral, rock-cut); Pal Ex: 178 (South Arcot Dt, bronze). A special development are the later Ambikā images: Tiwari Am: fig. 13 (Mt. Abu, Vimalavasahī), Tiwari Am: fig. 67 (South Kanara), Tiwari Am: fig. 71 (Chikmagalur). The images mentioned demonstrate the Ambikā iconography of a particular region (locality) and/or peculiarities of individual pieces. The following images depict the goddess without seated child: PJA: figs. 75 and 95; Shah Su: fig. 37; Tiwari Am: figs. 67 and 71.

The independent images open a new dimension: partial motifs, such as Ambikā's children, become motifs in their own right. The variety in the matter of children is very pronounced: there are one, two or three children, and one child (a boy) may be naked. In the K-and-A context we find goddesses without child where we expect Ambikās. The situation is too complex to be considered here.

[B: Cakreśvarī] Instructive examples of the independent Cakreśvarī are the following (stone images if not mentioned otherwise). Shah Ca: fig. 5 (Mt. Abu); Shah Ca: fig. 38 (Deogarh); JID: fig. 250 (Deogarh); JID: fig. 227 (Golakot); Shah Ca: fig. 20 (Gwalior, late rock-cut sculpture); Tiwari El: fig. 25 (Mathurā, medieval); Shah Ca: fig. 35 (Khandagiri Cave 8 or Bārabhujī, rock-cut; see also Mitra Sā: pl. 6A), Pereira Mo: 118, 120, 126, 150 (plans only); Pal Ex: 181 (Karnatak, bronze); Shah Ca: fig. 21 (Kambadahalli); Shah Ca: fig. 25 (Jinanathpur); Settar Ca: fig. 3 (Andhra Pradesh) and figs. 2, 4-6 (Karnatak). See also Misra Mo: 144.

Considerable variation exists in the number of arms (great number of arms in JID: fig. 250). The four-armed Cakreśvarī may show cakra, gadā, śaṅkha or fewer (one to two) Viṣṇu attributes. In some cases (independent and subsidiary Cakreśvarīs), the cakra is the dominating attribute, as the name Cakreśvarī suggests. Often we find four arms and four cakra.s: fig. 15, Tiwari El: fig. 25 (Mathura), Nagar Wo: pl. 78 (Bhopal State Museum), Mitra Śā: pl. 1B (Khandagiri Cave 7); Mitra Śā: pl. 3A (Khandagiri Cave 8). Garuḍa is shown in most, but not in all cases.

The number of Cakreśvarī images (subsidiary and independent) is fewer than the number of Ambikā images. There are, however, numerous Cakreśvarī images at Deogarh; see Tiwari Ch.

[C: Sarasvatī] Contrary to Ambikā and Cakreśvarī, the Jaina Sarasvatī is not integrated into the system. Nevertheless, she appears under different names in various dogmatic texts, starting with texts of the Jaina canon (Shah Sa: 195-197). The Kuṣāṇa Sarasvatī is a rare instance of a non-Jina shown in the early period of Jaina iconography (Pal Ex: 26,171; Lohuizen Sc: pl. 59, pp. 286-288; Mitterwallner Sc: 36-37, 158). The date given in the inscription is either 154 in the era of Kaniṣka I (omitted hundreds) or 54 in the era of Kaniṣka II (Williams Om). The relevant phrase of the inscription runs “eka Sarasvatī pratīṣṭhāvitā“. The goddess holds a book, the common denominator of all Sarasvatīs (“iconographic constant“: Pal Ex: 172a). The next stage presents two Western Indian bronze Sarasvatīs, also with book (Shah Ak: figs. 18-19), dated by U.P. Shah “ca. A.D. 600-620“ and “ca. A.D. 700“ respectively. Further early specimens are a rock-cut Sarasvatī at Ellora (Shah Sa: fig. 18); and the Sarasvatī depicted in fig. 17.

Good examples of the later period are the following (stone images if not mentioned otherwise). Shah Sa: fig. 14 (Dabhoi, bronze); Pal Ex: figs. 57A (Madhya Pradesh 1061) and 57B (Gujarat 1153); Shah Sa: fig. 19 (Mt. Abu, Vimalavasahī); Shah Sa: fig. 36 (Mt. Abu, Vimalavasahī, ceiling no. 32); JID: figs. 244A-246 (Pallu); JID: fig. 244 (p. 205: inscription kāritā bhaktyā Sarasvatī); Pal Ex: 173 (Madhya Pradesh, A.D. 1061); Pal Ex: 171 (Rajnapur Khinkini, bronze); Shah Sa: fig. 5 (Central Museum, Nagpur, bronze); Shah Su: fig. 40 (“probably from Karnatak“, bronze); Ramachandran Ti: pl. 35.2 (p. 234: “almost modem looking“). See also Misra Mo: 92-96 (general archaeological evidence). [End of A-C]

In the case of differences such as “standing/seated“ and “one, or more than one, pair(s) of arms“ we use the expression “basic variables“. Constancy is rare in this connection: the couple and K-and-A are always two-armed, but this is probably due to the fact that these types are basically restricted to the early period. Later sculptures of Kubera (K. alone) may show the god four-armed (Misra Ya: fig. 15). Early Ambikās are almost always two-armed, but one of the earliest Ambikās is four-armed as mentioned above. Later Ambikās are often four-armed. On the whole, there is the tendency to increase the number of arms in the course of time, but this is no hard and fast rule. Furthermore, Kubera and/or Ambikā are depicted with or without vāhana.s. The case of Ambikā's children has already been mentioned, and although the children are sometimes not very prominent this is an instructive example of basic variables (independent Ambikās and Ambikās in the K-and-A context).

Basic variables are often not concomitant with regional or chronological divisions, or the concomitance is not clear at first sight. This makes general (“All-India“) descriptions difficult. Here is the problem. As a consequence, descriptions of variables as found in current literature are in very many cases only descriptions of the variables of single or closely related images, without any general account of basic variables. The basic variables had also to be omitted in our own descriptions of Ambikā, Cakreśvarī and Sarasvatī (“economy“ of variable accounts).

Minor variables such as the different formulas for the drummer above the triple parasol of the Jina are not relevant to the present discussion. They are not less important, but, unlike the basic variables, they are not part and parcel of comprehensive descriptions as we expect them in general surveys.




Share this page on: