VṚṢṆIS IN ANCIENT ART AND LITERATURE: AN ADDENDUM

Dr. Vinay Kumar Gupta*

Superintending Archaeologist, Archaeological Survey of India, Jaipur, archaeologist1@gmail.com, 9456053192

Abstract:

This paper is an attempt to study Vṛṣṇi cult in a holistic manner. The author has already worked on the cult in his previously published article “Vṛṣṇis in Ancient Art and Literature”. This article adds to the already published work with three new important findings. The most important finding is the rock painting from Naldah, District Bundi which the author has identified as Pañca-Vṛṣṇi Vīras, the other is important finding of this cult from Pasopa, District Bharatpur and the last is the newly found Vṛṣṇi triad from Berenike, Egypt which has been recently published online by Shailen Bhandare. This article also combines the information on Vṛṣṇi cult in south India which the author had published in a joint paper with Late Mrs. (Dr.) Haripriya Rangarajan.

Keywords: Vṛṣṇis, ancient art, literature, Mahābhārata, Vāsudeva-Kṛṣṇa, Mathurā, lineage, epithets, rulers, restoration.

The Vṛṣṇis constitute one of the most important groups in the saga of the Mahābhārata, and Vṛṣṇi is the clan in which the most famous of the Mahābhārata characters, Vāsudeva-Kṛṣṇa, was born. The Mahābhārata (II.22) mentions that the Vṛṣṇis were part of the rulership of Mathurā, and the town seems to have been their ancestral place.[1] Local power among the Vṛṣṇis is in a way attributed to Vāsudeva-Kṛṣṇa in the epic when it mentions him as Saṃghamukhya (XII.81.25). A series of epithets shows that his identity is closely connected with his lineage; the more frequent epithets or names used for him in the Mahābhārata are: Vārṣṇeya, Vṛṣṇi, Mādhava, Dāśārha, Sātvata, Yadu, Yādava and Śauri. As a scion of the Vṛṣṇis, sometimes only the appellation ‘Vārṣṇeya’ is used to designate him (e.g. Śrīmadbhagavadgītā III.36); sometimes it is associated with him as a sort of paternal surname (e.g. Kṛṣṇa Vārṣṇeya in III.189.54). The Mausalaparva of the Mahābhārata describes the end of the Vṛṣṇis and the rehabilitation of the few remaining Vṛṣṇis by Arjuna at Martikāvata (Bhojarāja’s family) and at Indraprastha and nearby areas (Mbh. XVI.7.67–75)[2]. In the Critical Edition of the Mahābhārata, Vārṣṇeya as an epithet or name of Vāsudeva-Kṛṣṇa appears 119 times, whereas another 45 occurrences are used for others (Brockington 2007: 20). In the Nalopākhyānaparvan of the Mahābhārata (III.52–79), the charioteer of Nala is named Vārṣṇeya (a sūta), although no justification is given for such a name.

Chapter 34 (Harivaṃśa Parva) of the Harivaṃśa Purāṇa (Mahābhārata Khilabhāga) mentions the Vṛṣṇi dynasty in some detail. As per its details the Yadu kula was quite large and a special branch of its ruling people was termed as Vṛṣṇis. Kārtavīrya Sahasrārjuna was the most powerful and famous ruler of the Yadus (also called Haihayavaṃśī), who was killed at the hands of Jamdāgni Paraśurāma, and in his family were born Vṛṣa and his son Madhu, from whom the Vṛṣṇivaṃśa derives its name. In the family of Madhu was born Kroṣṭā, whose son was named Yudhājit. Yudhājit gave birth to a famous ruler named Vṛṣṇi. Another roughly contemporary great man of the Vṛṣṇi was his stepbrother Andhaka, and that’s why the clan of the Mathurā rulers is sometimes called as Andhaka-Vṛṣṇi. From Vṛṣṇi were born Śvaphalka, Citraka and Devamīḍhūṣa. From these three were born many great heroes (Vīras) of the Vṛṣṇi clan, including Akrūra and Śūra. From Śūra was born Vasudeva among other children, and from him were born the greatest Vṛṣṇi heroes Saṃkarṣaṇa/Balarāma and Vāsudeva-Kṛṣṇa, and also Subhadrā (Goddess Ekānaṃśā). In the same clan were born other great Vṛṣṇi heroes including Sātyaki, Sāraṇa, Vidūratha, Uśīnara, Niśaṭha, Satyaketu, Ariṣṭanemi, Anādhṛṣṭi, Ekalavya and Uddhava (a learned man). Pṛthā (Kuntī) was a daughter of Śūra and gave birth to Karṇa, and later, after marrying Pāṇḍu, to the three Pāṇḍava brothers Yudhiṣṭhira, Bhīma and Arjuna. Therefore, four of the most important protagonists of the Mahābhārata were of half-Vṛṣṇi lineage, from their mother’s side. From Vāsudeva-Kṛṣṇa were born the two great Vṛṣṇi heroes Pradyumna and Sāmba, and from Pradyumna was born Aniruddha, another great Vṛṣṇi hero. A similar description of the Vṛṣṇi cult is provided in the Viṣṇu Purāṇa (IV.14–15).

As per the Vedic Index (Macdonell & Keith 1958: 289–90), the Vṛṣṇis are already known in the later Vedic period; their descendants (i.e., Vārṣṇa, Vārṣṇeya, Vārṣṇya) are mentioned in the Taittirīya Brāhmaṇa (III.11.9.3; III.10.9.15), Śatapatha Brāhmaṇa (I.1.1.10; III.1.1.4), Bṛhadāraṇyaka Upaniṣad (IV.1.8) and Jaiminīya Upaniṣad Brāhmaṇa (I.6.1; I.5.4).

Pāṇini in his Aṣṭādhyāyī (4.3.98) mentions the prevalence of the cult of Vāsudeva and Arjuna (vāsudevārjunabhyāṃ vun), and he refers to the Vṛṣṇis in the sūtra ṛṣyandhakavṛṣṇikurubhyaś ca (4.1.114). Another sūtra, jīvikārthe cāpaṇye (5.3.99), is construed, in later commentaries, like the Pāṇinisūtravyākhyā, in a way that images of Vāsudeva, Śiva and Skanda were known in Pāṇini’s time (jīvikārthaṃ yad avikrīyamāṇaṃ tasmin vācye kano lup syāt / vāsudevaḥ / śivaḥ / skandaḥ). Pāṇini (6.2.34) also cites the Vṛṣṇis and Andhakas as Kṣatriya names, rājanyabahuvacanadvandve’ndhaka-vṛṣṇiṣu. Thus, there are clear references about the Vṛṣṇis and also about the possible images of Vāsudeva, Śiva and Skanda. Vṛṣṇi Vīra Vāsudeva is important here for our purpose. Patañjali’s Mahābhāṣya mentions the killing of Kaṃsa by Vāsudeva and the narration and presentation of dramas, presumably in Mathurā, on the life of Kṛṣṇa (Bhandarkar 1874). Even Kauṭilya in his Arthaśāstra (VI.12) mentions a war between the Vṛṣṇis and Dvaipāyana, which means that various episodes related to the Vṛṣṇis were quite popular by the time of Kauṭilya (c. 4th century BCE). Scholars even try to identify Heracles mentioned by Megasthenes as Vāsudeva-Kṛṣṇa, but this identification is quite tentative indeed.

In ancient India, the Vṛṣṇis were worshipped either as individual deities (heroes) or in groups (triads, tetrads or pentads). As an individual deity, any of the above-mentioned Vīras may have found his following, but Balarāma, Vāsudeva and Ekānaṃśā were more popular. Balarāma and Vāsudeva might have been worshipped together as doubles, but no such image has survived except the one late medieval example at Pasopa, District Bharatpur. Balarāma, Vāsudeva and Ekānaṃśā are also represented in a group, which is identified as Vṛṣṇi triad. In the tetrads may be included Caturvyūha images, which have the representation of four Vṛṣṇi Vīras. For the pentads, the Mora inscription makes important mention of Vṛṣṇināṃ Pañcavīrānāṃ, who are identified as the Five Great Heroes of the Vṛṣṇis – Saṃkarṣaṇa, Vāsudeva, Pradyumna, Sāmba and Aniruddha. Banerjea (1942: 65–68) confirmed these names with a passage in the Vāyu Purāṇa (97.1–2), which reads Manuṣyaprakṛtīn devān kīrtyamānān nibhodhata saṃkarṣaṇo vāsudevaḥ pradyumnaḥ sāmba eva ca / aniruddhaśca pañcaite vaṃśavīrāḥ prakīrtitāḥ //. The Śrīmadbhāgavata Purāṇa (X.16.45) also mentions four Vṛṣṇi heroes by the names of Saṃkarṣaṇa, Kṛṣṇa, Pradyumna and Aniruddha, as belonging to the Sātvata cult. The Mahābhārata mentions two different groupings of the four Vṛṣṇi Vīras: Saṃkarṣaṇa, Vāsudeva, Pradyumna and Aniruddha are enumerated in one passage (XIII.158.39), while Mādhava, Sāmba, Pradyumna and Aniruddha appear in another passage (XVI.3.44–45). Earlier, Lüders (1937–38: 196–98) had identified them as the “five great heroes” mentioned in Jaina texts as Baladeva, Akrūra, Anādhṛṣṭi, Sāraṇa and Vidūratha, among whom Akrūra was the commander. Quintanilla (2007: 212–13) proposed to accept Lüders’ identification. As far as the names of these heroes are concerned, they definitely indicate the historicity of the characters, but considering them to be of Jaina affiliation is untenable. Although Vāsudeva and Baladeva find some importance with Neminātha, there is no such indication that separate temples of minor local divinities were erected by Jainas before the Christian era. Besides, the site of Mora has not shown Jaina association of any nature till now. The names of five Vṛṣṇi heroes as identified by Lüders cannot be outrightly rejected, as they show some glimpse of historicity in them. But then, Banerjea’s identification seems to be more correct, as the popularity of these characters as close family members of Vāsudeva-Kṛṣṇa indicates. Besides this identification, Rosenfield had identified them as five ancient legendary kings of Mathurā, in this supporting Banerjea (Rosenfield 1967: 151–52). The Pañca-Vīra concept of Vaiṣṇavism had near-parallels in other sects: the Pañca-Brahmas of Śaivism, the Pañca-Buddhas of Buddhism, the Pañca-Tīrthaṅkaras of Jainism, and, more importantly, the Pañca-Vīras of Yakṣism. However, we know little of them to say how they are linked with the principles of several religious systems (Aravamuthan 1969: 71).

After a perusal of the literary evidence, we now come to various archaeological pieces of evidence related to the Vṛṣṇis, which will be discussed site-wise in the following.

Tikla

What may be considered the earliest example of the depiction of the Vṛṣṇis in the form of a triad (Fig. 1) is found in a rock shelter at Tikla situated about 70 km south of Gwalior, in District Shivpuri. V.S. Wakankar first recorded it (2005: 145–46 [based on his PhD thesis, Pune 1973]). He refers to the figures as “presumably a king and queen with an attendant holding an umbrella” (ibid.: 365). This rock painting may be dated within the time bracket of the late 3rd and the 2nd century BCE, on the basis of a short Brāhmī inscription written above the figural depictions on the rock surface. It reads dambukena kāritaṃ odanakita (cf. Neumayer 1992–93: 54; last word: author’s reading). It is quite clear that either the artist who painted the lines on the rock surface was named Dambuka, or that Dambuka was the person who got the figures painted by an artist, but the meaning of the third word is not clear. The painting depicts three human figures. The first figure from the left is a male who holds a halāyudha in his right hand. In his left hand he is shown holding a peculiar object looking like a barbed harpoon with a possible snake hood on top. He is wearing a conspicuous headgear with two circular loops. His dress seems to have been inspired by the tribal/rock art of the period. This figure should be identified as a representation of Balarāma.

The second male, who is shown in the centre and is deliberately made slightly taller than the others as an afterthought, should be identified as Vāsudeva-Kṛṣṇa. His left hand is stretched out and probably holds a disc or a flower, while his right hand holds a slightly similar object as the first male holds in his left hand, but with a longer handle. The most important feature of this figure is the headgear, which has been shown with what seems to resemble a solar disc (Gupta 2013: 205–06). The third figure, which is smaller in proportions than the other figures (as females are generally represented), seems to hold a tall object like a parasol in her right hand. However, a proper contact between the umbrella’s handle and her hand is not visible, and the latter appears to be touching the object held in the left hand of the central figure. The left hand of this figure seems to be empty. It looks a bit strange that the female deity is shown holding a parasol, but that may be meant to represent her as a bestower of sovereignty. Or else, the painter tried to draw a canopy as is found over the head of Ekānaṃśā and other goddesses during the Kushan period. A related evidence may be seen in the Mahiṣāsuramardinī sculptures of the Kushan period, where the goddess is shown in self-coronation after her victory over Mahiṣa. Such a concept might be behind the association of Ekānaṃśā at Tikla with a royal parasol.

The dresses of all three figures are slightly different, and it appears as if the third figure is shown wearing a skirt and a blouse-like clothing. The most striking feature of this triad is that it is the only known example of such type where the proper genealogical hierarchy of the three deities in accordance with Harivaṃśa Purāṇa has been followed. Interestingly, Vāsudeva-Kṛṣṇa is shown as the tallest figure, indicating that he had already attained a superior status. It should also be kept in mind that the Tikla rock shelter site is a part of the Greater Braj region and lies on the route linking Mathura with Tumain and Ujjain.

Naldah, Bundi

There is a long series of rock shelters along the course of River Chambal and its various tributaries extending for few kilometres. Many of these shelters also contain rock paintings. An archaeology enthusiast Kukki Om Prakash of Bundi reported many of these shelters for the first time. Along a small seasonal tributary Mangli which merges with Mez River of Chambal system, there are few important rock shelters containing rock paintings which are situated near village Kewariya. The particular rock shelter which is of interest to us is named Naldah (25°14'38.4"N 75°29'02.4"E). Here again, like the rock art of Tikla, the most interesting and one and only depiction related to Pañca-Vṛṣṇi Vīras is made where the five Vṛṣṇi heroes are shown standing side by side (Fig. 2). The two figures on the extreme left and right are properly shown with haloes and can be identified as Vāsudeva-Kṛṣṇa and Balarāma. Most of the figures appear with their hands held akimbo. They all are shown connected by a long piece of cloth, running like a rope. The one to the extreme right either possibly holds a cakra in his left hand or is having it tied to the rope, can be identified as Vāsudeva-Kṛṣṇa whereas the one to the extreme left either possibly holds a hala (plough-share) or is having it tied to the rope but the artist seems to have taken some liberty by stylizing the depiction in a rectangular form. The three male figures in between these two deities can be identified as Pradyumna, Sāmba and Aniruddha with no differentiating or identifying marks except for the fact that some stylization is done over their heads, possibly indicating their divine nature. Here, one can argue their identification as divinities of some significance in want of proper identifying features. But, the most striking aspect is the representation of divinity with the help of haloes on the two figures shown on the extreme right and left. The number five can either correspond to Pañca-Vṛṣṇi Vīras or the Pañca-Pāṇḍavas. The possibility of latter identification is negated because of the presence of haloes over the heads of two extreme right and left figures as well as their lacking anything related to Pāṇḍavas. The prominence given to only two of the five again indicates their enhanced importance than the rest three, thus helping in the identification of two as Vāsudeva-Kṛṣṇa and Balarāma. The painting on the basis of its style and overall treatment can be dated to not later than third century BCE, thus making it the earliest representation of Vṛṣṇis in any form of art. The location of the rock shelter lies on the ancient trade route linking Mathura with Ujjain. Either some trader got these drawn there or someone from the local community who got to know of the five great heroes drew them on the rock surface. The scenery in front of the rock shelter is simply serene and incredible. The rivulet water and the presence of big and very old jambu trees in a short-specified area would have been a treat for the travellers in the past too, thus resulting in depiction of such an important subject in such an interior area.

The Heliodorus Garuḍa Pillar Inscription, Besnagar

The famous Besnagar Garuḍa pillar inscription mentions the erection of this Garuḍadhvaja of Devadeva Vāsudeva by Bhāgavata Diyaputra Heliodorus, who was an ambassador of the Greek king Antialkidas of Takṣaśilā to the court of Kāśīputra Bhāgabhadra (Sircar 1965: 88; Vogel 1908–09: 126). The second part of the inscription mentions the three fruits of this deed – svarga, dama and apramāda.

This inscription which belongs to the 2nd century BCE confirms the popularity of the Bhāgavata religion as far as Takṣaśilā and its popularity even among the Indo-Greeks. It is quite probable that the find-spot of the pillar (locally called ‘Khambābā’) might have been a popular local shrine of the Bhāgavatas, and that was why Heliodorus made such an offering there. It is also to be kept in mind that from the same site and also from its vicinity various pillar capitals like the palm capital (Fig. 3) and the kalpa-vṛkṣa capital have come to light, which belong to the Balarāma and Śrī-Lakṣmī (Rukmiṇī) cult, respectively. The former presence of the dhvajastambhas of other Vṛṣṇi deities is quite probable there. The finding of a dhvajastambha of Śrī-Lakṣmī is quite significant, because it confirms that a female goddess either as the sister or the wife of Vāsudeva-Kṛṣṇa was worshipped at the Vṛṣṇi shrines. Besides, there is also an evidence of a palm capital from Pawaya, which is now housed in the Archaeological Museum, Gwalior. Joshi (1979: 23–24) has associated the worship of a lion pillar on a corner upright (State Museum Lucknow, no. J 268) and the lion-plough capital from Mathura now housed in the Victoria and Albert Museum, London, to the Saṃkarṣaṇa cult.[3]

The Malhar Image

Malhar in Bilaspur District of Chhattisgarh has provided one of the earliest known images of the Bhāgavata religion (Sircar 1971: 84–85). The image (Fig. 4) was in worship outside a pond at the site. It represents a four-armed deity holding a gadā and a cakra in his extra upper hands and holding a conch in his normal hands, reminiscent of añjali mudrā. A long sword running from near the head down to the soil is held by the deity vertically, with a support at his right shoulder. The sword has an inscription in Brāhmī characters of the 2nd century BCE. The inscription, sword, mace and wheel as well as the conch Pāñcajanya (if it is so) point to the identity of the deity as a Vṛṣṇi Vīra, here in this case, Vāsudeva-Kṛṣṇa. The other features of the image are slightly like a local (tribal) warrior, viz. the lower dress made up of leaves, the appearance of the face, the treatment of the legs and body, and the ornamentation. But the presence of the Vṛṣṇi cult in a far-off place with such a local touch clearly indicates the popularity of the cult by then. This sculpture may be considered the earliest Vṛṣṇi image in stone.

The Ghosundi (Madhyamikā, Rajasthan) Inscription

An important archaeological evidence regarding the Bhāgavata (Vṛṣṇi) cult was found at the site of Ghosundi (Nagarī) in Chittorgarh District of Rajasthan, in the early 20th century, which comes under the ancient Madhyamikā country (Sircar 1965: 90). The inscription of the reign of King Sarvatāta mentions the construction of a pūjāśilāprākāro nārāyaṇavāṭakā in the name of Saṃkarṣaṇa and Vāsudeva during the performance of an aśvamedha sacrifice. The first mention of the older brother confirms its Vṛṣṇi connection, and the use of term Nārāyaṇa Vāṭakā for a shrine of Vṛṣṇi deities confirms the merger of various aspects of the Vṛṣṇi cult and the later Vedic god Nārāyaṇa by the time of this inscription, which is dated to 2nd/1st century BCE. It is quite interesting to note that the area around Udaipur, which later on became one of the most important areas for the Bhāgavata cult because of the shifting of the Śrīnāthajī idol from Govardhan to Kankrauli-Nathdwara was already important for the same cult during the 2nd/1st century BCE.

The Nanaghat Inscription

The Nanaghat Cave inscription of Sātavāhana Queen Nāgaṃnikā mentions Saṃkarṣaṇa and Vāsudeva along with other Brāhmaṇical deities like Indra and the Lokapālas (Sircar 1965: 192–93). Here again, the elder brother, in spite of his lesser importance, is mentioned before the more important Vāsudeva. Thus, this record should also be taken as connected with the Vṛṣṇi lineage. This inscription is generally dated around the 1st century BCE/early 1st century CE and confirms the proliferation of the cult of the Vṛṣṇis in a far-off place like Nasik in Maharashtra.

The Panel from Kondamotu

An important sculptural panel (Fig. 5) found at Kondamotu in Guntur District of Andhra Pradesh is now housed in the State Museum, Hyderabad (Nigam 2000–01: 114). In this panel, all the five Vṛṣṇi Vīras are carved in their proper genealogical order. Carved from left to right is Saṃkarṣaṇa holding a mace in the right hand and a ploughshare topped by a lion face in the left hand. Next to him is Vāsudeva-Kṛṣṇa with his right hand in abhaya mudrā, and the left hand holding a conch-shell and placed on his hip. The next figure is of Narasiṃha, which clearly suggests the merger of the Sātvata cult with the Pañca Vṛṣṇi Vīra cult. The following figure is of Pradyumna, who holds a bow in his left hand and an arrow in his right hand. To his left is the figure of Sāmba, who holds a wine goblet in his right hand, whereas his left hand is placed on his waist. The last figure is that of Aniruddha, who is holding a sword in his right hand and a shield in his left hand. This sculpture is dated to about late 3rd/early 4th century CE and confirms the popularity of the Sātvata-Pañca Vṛṣṇi Vīra cult in Andhra Pradesh by that time.[4]

Chilas

Chilas is an ancient town on the Silk Route in Gilgit-Baltistan of Pakistan Occupied Kashmir, which at some point of time formed part of the ancient Kamboja kingdom. A large number of petroglyphs and inscriptions have come to light from this area of the Karakoram mountains. At the site of Chilas II, A.H. Dani (1983: 101–20) reported the finding of two pairs of Vāsudeva and Balarāma images as petroglyphs. The Kharoṣṭhī inscription mentioning Baladebo and Vasudebo at one place and Rama-Kṛṣṇasa Dhamaputrasa at the other place confirmed the identification of the painted characters. Dani (ibid.: 106 and pl. 79) attributed two different headgears to both the deities, Balarāma has a crown and Kṛṣṇa a crescent-topped headdress. Both of them are wearing dhotī. The left hands of both of them are akimbo. Balarāma is holding a spear ending with a plough at its lower end, whereas Kṛṣṇa is holding a discus in his right hand. In the other representation are shown two standing human figures with a flowing dress covering their body (ibid.: pl. 95). The Balarāma figure has a crown on his head, whereas Kṛṣṇa has a crescent-topped headgear. Both of them are holding a club in their right hand. Balarāma is holding a plough-topped banner in his left hand, whereas Kṛṣṇa is having a discus in his left hand. Dani (ibid.: 106) considered the first depiction as the earliest representation of these two deities with iconographic detail and name. It may correspond to a 1st century BCE date. The other representation may be dated a century or two later. These petroglyphs confirm the popularity of the Vṛṣṇis in the northwestern part of India, particularly along the trade route​ ​before the Common era.

Newal

This site situated in Unnao District of Uttar Pradesh is identified with the ancient ‘Navadevakula’ (Garg 1950: 175). There is a famous shrine of the Pañc-Pīr at Newal, which, according to V.S. Agrawala, should be connected with the five Vīras – either with the Vṛṣṇis or the Pāṇḍavas. It again illustrates the fact that Pañca-Vīras were very popular in ancient times and that the concept later on got merged with other sects, and that this development would explain the change in name to ‘Pañc-Pīr’. There is a possibility that many other Pīr shrines in the Vraja region or in northern India might actually have been dedicated to Vīras (possibly Vṛṣṇis). I also came to know from a person of the Jayaswal caste that they perform a certain ritual five times in the name of Pīr Bābā, possibly on the occasion of the Holī festival. It might indicate that this ritual was earlier performed for the Pañca-Vīras and later transferred to the Pīr. These Pañca-Vīras should in all probability be the Pañca Vṛṣṇi Vīras.

The Vraja Region

Mathura and its surrounding area known as the Vraja region is believed to be the most sacred and significant area on earth for the presence of the Vṛṣṇis and the Bhāgavata cult. Hitherto, a proper emphasis on this aspect in archaeological and iconographical terms is somewhat lacking, but the present analysis would try to fill up some of the gaps.

Pasopa

Pasopa village is situated near Adi-badri, an important landmark of ancient Vraja in Dig tehsil of District Bharatpur of Rajasthan. On top of a small hillock in the village, there is a shrine dedicated to Pañca-Vīras near a modern temple dedicated to Rādhā and Kṛṣṇa. The local legend believes that there were five great warriors in the army of the king of Kaman around the early medieval period and three of them got killed in a war with Muslim invaders and two of them survived. In the memory of those five great warriors, the shrine at Pasopa was built and at present the two surviving great warriors are worshipped in the form of stone idols (Fig. 6). In fact, this local legend confirms the prevalence of the cult of Pañcavṛṣṇi Vīras in Vraja region but with the passage of time the cult lost its importance and the people forgot the old (historical) tradition. It is difficult to say that there was the presence of a sizeable physical shrine or sculptures in Pasopa dedicated to Pañcavṛṣṇi Vīras during the ancient times but there must have been some kind of dedicatory platform or some sort of shrine with uncarved stones kept there to offer worship to Pañcavṛṣṇi Vīras. In due course of time, out of the five Pañcavṛṣṇi Vīras, only Vāsudeva-Kṛṣṇa and Balarāma remained significant and popular. When the local folk forgot about Pañcavṛṣṇi Vīras, they developed a new story and forgot that the two great warriors they are worshipping as Pañca Vīras are in fact Vāsudeva-Kṛṣṇa and Balarāma. But, this surviving evidence from Pasopa, a part of Braj 84 kos parikramā is very significant as it confirms the prevalence of the Pañcavṛṣṇi cult in northern India in general and Vraja region in particular.

Mora

A very important site connected with the Vṛṣṇi cult is Mora, which is a village situated to the south of Mathura city, at a distance of about 10 km. The name of the village might have been derived from legendary king Moradhvaja (Gupta 2013: 294). Another important place of worship for Vraja people including the Varshneys,[5] near River Gaṅgā at Narora is also called ‘Morra’ and is a part of the famous Belavana with its presiding goddess of high reverence. Nowadays, this goddess is identified as Śrī-Lakṣmī, but in antiquity it might have been either related with Vṛṣṇi Ekānaṃśā as sister of Vāsudeva-Kṛṣṇa or Śrī-Lakṣmī as wife of Vāsudeva-Kṛṣṇa. Mora village became one of the most famous of Mathura sites, because of some epigraphic and sculptural findings dating back to the first half of the 1st century CE, thus providing the earliest evidence of the worship of the Pañcavṛṣṇi Vīras. Some evidence like the inscribed brick of the period of Bṛhasvatimitra even dates back to 2nd century BCE. Four sculptural finds (Figs. 7–10) and a door-jamb have been reported from this site, along with the famous Mora well inscription (Government Museum, Mathura [hereafter G.M.M.], no. Q 1) that was found by Sir Alexander Cunningham (Vogel 1911–12: 127). This slab containing the inscription was being used as a part of the terrace of the well, possibly displaced from its original location. The inscription was not legible even to Vogel, because of its deteriorated state. The inscription, which was read by Heinrich Lüders, is actually based on Cunningham’s information. Presently, this inscription has become illegible.

Lüders’ (1937–38: 194 [ṛ, ṣ and c adjusted]) reading of the Mora well inscription is as follows:

1  [siddhaṃ] mahakṣatrapasa Rājūvulasa putrasa svāmi………

2  bhagavatāṁ Vṛṣṇīnā[ṁ] paṁchavīrāṇāṁ pratimā[ḥ] śailadevagṛi …

3  ya[s]=To[ṣā]yāḥ śailaṃ śṛīmad=gṛhaṃ=atulam=udadhasamadhāra ... 

4  ārcādeśāṁ śailāṁ paṁca jvalata iva paramavapuṣā………………….

His reading of the inscription on the pedestal of the female statue from Mora (Fig. 7) is as follows (ibid.: 201):

1  ………sya Kan(i)[ṣ]ka[sya]……..[r]…………[m]…

2  ……………etasya purvaye M[ā]thuri kalavaḍ[ā] o[ḍakh]i…

3  ………………..ye Tośāye patimā…………t………….

One point which is clear from these inscriptions is that these belong to two different rulers – one probably to the Śaka Kṣatrapa Śoḍāsa, son of Mahākṣatrapa Rājūvula, and the other to the Kushan ruler Kaniṣka, both separated by a margin of a few decades. The first inscription clearly refers to the popularity of the Bhāgavata cult of the five Vṛṣṇi heroes in this part of the country, and also to the tradition of worshipping their images in a shrine. “Toṣa/Tośa” is significant as occurring in both of them, and it should refer to the same person or place, whatever may be the case.

An important question is the identification of ‘Toṣa’ or ‘Toṣā/Tośā’. Lüders (1937–38: 194–202) writes: “at first sight one would obviously understand ‘the house of Toṣā’ as a shrine dedicated to a goddess called Toṣā, but I am not aware that there ever existed a goddess of that name. Under these circumstances, Toṣā can only be taken as the name of the lady who caused the shrine to be built.” He further adds: “Toṣā does not sound like an Indian name. It is quite probable that Toṣā was of Iranian extraction.” Rosenfield (1967: 151) writes about it: “Although it is difficult to reconcile a Tosha of the time of Śoḍāsa with one of the time of Kanishka, the coincidence makes it most likely that the original Mōrā well inscription referred to the mound in question.” Doris M. Srinivasan (1997: 213) also supports the view of Rosenfield.

In accepting Lüders’ identification of Toṣā, there is a genuine problem that there is no deity known until now by the name of Toṣā, neither in India nor in the regions to the North-west of India. Secondly, if it is taken to be a female by the name of Toṣā who erected the shrine, then how will one explain the gap of a few decades between the two inscriptions? Now, there is a need to look at all this information in the present context. In the past, scholars missed the very important fact that the nearby village of Mora (about 2 km distance as crow flies) is known as Tosh. The village of Tosh is very well linked with the saga of the Vṛṣṇi Vīras. It is believed that this was the name of a sakhā or friend of Kṛṣṇa, who had taught him the art of playing flute, i.e. in a sense his musical guru. Besides, one of his favourite gopikās, Indīvarā, who is one of the famous 108 sakhīs of Kṛṣṇa, is also believed to have resided in the village. No Purāṇic reference could be noted for it, but Growse (1883: 57) has mentioned the mate of Kṛṣṇa by the name of Toṣa during the killing of the demon Aghāsura by him. He has also mentioned that this village was named so because of this particular Toṣa.

I would like to reiterate my suggestion that Toṣa of the inscription is the same Toṣa mentioned in Vraja folklores (Gupta 2013: 294–301). In that case To[ṣā]yāḥ śailaṃ śṛīmad=gṛhaṃ might mean that this great shrine of the Bhāgavata heroes existed as the temple of Toṣa, which might be the name of the village Tosh or a person who was considered the musical guru of Kṛṣṇa. There is certainly some doubt if it is actually Toṣāya, as it might be the caturthī form of the word Toṣā as a female, but neither is this reading a full proof, nor should Toṣāya be considered only as the fourth case in feminine gender (caturthī stṛīliṅga). The name of the village Tosh should be considered of utmost importance in interpreting the word Toṣāya, and in that case it must be in the masculine gender (puṃliṅga vibhakti). Another point which needs to be looked at is that if we consider the person mentioned in the inscription as a female named Toṣā, then the village would be called ‘Toṣāgrāma’, and it would have developed to ‘Toṣāgāma’ or ‘Toṣām’ over a period of time. But it has not been the case with the name of this village, and it was only known as Tosh, with grāma as a normal suffix. This is the reason that the name of the village has not changed as per the rules of local Braj dialect. Even if the name of the village is not considered to be related with the Kṛṣṇa legend, then also the word Toṣa cannot be considered to be of outside origin, as it is a Sanskrit word which means ‘satisfaction’ and is quite popular with the prefix sam as saṃtoṣa, and also with the āśu prefix as Āśutoṣa for Śiva.

During explorations, it was observed that the original inscription, which was found on the outskirts of Mora by Cunningham, in the terrace of a well, might have reached there from a site at some distance from village Mora. Later on, this area (Fig. 11) was taken under protection by the Archaeological Survey of India. Not much information is available about this excavated site, except for the existence of some circular shrine considered an ancient one by Pt. Radha Krishna (Vogel 1911–12: 127). Now, when we take a look at this site, it is observed that this small site, about 80 x 60 m in dimension, is situated at an approachable distance from five important early historical sites, and these are Mora, Tosh, Ganeshra, Fenchari and Sakna. Local people informed that till some 40 years back, people from all these villages came to worship a deity at that site. One important point which became quite evident was that the deities to whom Varshneys offer worship by putting a cow dung-made cake on fire in the morning of the Holī festival are none other than the great ancestors (Pañcavṛṣṇi Vīras), because it was only on the day of the Holī festival that all these people from surrounding villages came to offer their worship at this site of Vṛṣṇi Vīras. In the Vraja region, on the day of Holī festival, people of the area go to worship the great ancestors at ancient mounds or sites. Besides, there is a continuing tradition of great ancestors’ worship at the Mora site in the form of a shrine, which the villagers call by the name of Bhūmiyā Bābā. Recently, a temple has been constructed on this site in which the deity is represented as a horse-riding warrior with a sword in hand (Fig. 12). Generally, bhūmiyā and jakhaiyyā are the local terms used for Yakṣa devatā. However, in this case the deity is not represented as a Yakṣa, but as a Vṛṣṇi Vīra, which confirms some kind of continuity of the past Vṛṣṇi cult at this site.

I have observed many early religious sites at some distance from the ancient habitation areas of villages, almost on a regular basis. But the important point about this site along with its exceptional finds of Vṛṣṇi Vīras is its location in the midst of some important habitation mounds in an area that was called as jhīl because of a huge depression. This is probably the largest area of depression in the whole cis-Yamunā section, covering more than 16 villages, and it might be one of the reasons for the high density of ancient sites in this area. During the monsoons the whole area between Jikhangaon, Khamini to Satoha and Ganeshra took the shape of a lake, which provided very fertile land for agricultural purpose. The situation has changed only after the construction of the Govardhan drain in the area, which possibly follows some old course of a stream. The trans-Yamunā section along with many pockets in the western section was fertile enough for agricultural production. This in addition to the presence of a number of early settlements in the cis-Yamunā region makes it clear that during ancient times Mathurā was a self-sufficient agricultural centre and needed not to depend on trade, at least not for survival purpose. This is in total contradiction to the view expressed by R.S. Sharma (1989: 31) about ancient Mathurā’s food security.

On closer examination of the site of Mora, a broken sculpture (Fig. 13; for another view see Gupta 2013: fig. 13.3) was found lying there, which is worshipped only on special occasions. This sculpture definitely belongs to the same group of sculpture as the broken lower half of a female figure of the time of Kaniṣka (Fig. 7). It has been sculpted from a red sandstone slab. Whether the other three sculptures of this site belonged to the Śaka-Kṣatrapa period or not is yet not clear, especially after this find.

In conclusion, there are two headless torsos (Figs. 8–9) and the lower leg portion of a person (Fig. 10) that were already known from Mora. We are still clueless about their dates, as well as the date of the recently noted piece (Fig. 13).

The Chamunda Tila Pillar Capital

Another very significant evidence of the presence of the Vṛṣṇi cult in Mathura comes from Chamunda Tila, situated on the outskirts of ancient Mathurā city. It is in the form of a unique pillar capital, now housed in the Government Museum, Mathura (Fig. 14). This pillar capital was found in the year 1972 and since then it has somehow escaped the attention of scholars for its actual importance. What is most amazing in this pillar capital is that it represents five symbols visible from both sides. These are a lion (on each end), a palm leaf (in the centre, on both sides), makara (on each end), a beautifully adorned woman (on both ends) and a Yakṣa (on top in the centre, on both sides). It is quite probable that the topmost symbol in the centre of the pillar-capital is missing. According to Banerjea’s (1937: 13–20) interpretation based on the Viṣṇudharmottara Purāṇa, tāladhvaja belongs to Saṃkarṣaṇa/Balarāma, garuḍadhvaja to Vāsudeva-Kṛṣṇa, makaradvaja to Pradyumna and ṛṣyadhvaja to Aniruddha. In the pillar capital under discussion, two of the deities Saṃkarṣaṇa/Balarāma and Pradyumna are represented by their respective dhvajas. The lion is clearly related to the concept of the Sātvatas. It is related to Narasiṃha as well as to Saṃkarṣaṇa/Balarāma. Yakṣas play a very important role in the evolution and development of the Vīra cult. Many features of early Vṛṣṇi Vīra images like the Malhar image and the Birabai torso (to be discussed next) definitely seem to be inspired by Yakṣa iconography and cult. Therefore, overall, it appears that this pillar capital is a capital where the merger of various doctrines like the Vṛṣṇis, Sātvatas and Yakṣas is visualised. Besides, the continuing tradition of Vṛṣṇi worship in the disguise of a Yakṣa at the Mora site also confirms some kind of association between the Yakṣa and Vṛṣṇi cults. No clear explanation can be given at this stage for the presence of a female figure on both sides of the capital. This capital is ascribable to the second-first century BCE.

The Birabai Torso

Birabai is a village situated near Noh in District Bharatpur, and is about 35 km south of Mathura city. This site has provided a torso of a Vīra (possibly a Vṛṣṇi), which is now housed in the Bharatpur Museum (Fig. 15). This torso is quite similar to the torso of a monumental Yakṣa image from Noh in its overall appearance, but a sword hanging by its side is indicative of its identity as a Vīra. Besides, the continuing tradition of calling the village by the name of a Vīra also confirms its identity as a Vīra. Although, the Vīra concept is applied to some other deities as well, in the Vraja region it is quite likely that this Vīra was a Vṛṣṇi Vīra. Birabai torso is datable to second century BCE.

Vrindavan

Another very important evidence for the Vṛṣṇi cult comes from the most sacred of the Bhāgavata sites, Vrindavan. At the site of the Madan Mohan temple that belongs to the 16th century, during the year 1979, some ancient structure having two inscribed bricks was exposed along the river front of the Yamunā (Mishra 1983). This site is related with the episode of subduing of Nāga Kāliya by Kṛṣṇa during his childhood and is locally known as Dwadash Aditya Tila, also indicating its solar connection. In the ancient structure (Fig. 16), various bricks measuring 52 x 26 x 8 cm, 39 x 29 x 7 cm, 42 x 28 x 7 cm, 36 x 26 x 6 cm, etc. have been used. The larger brick size of 52 cm length corresponds to the Mauryan brick size and is comparable to various Mauryan bricks found at Saheth-Maheth, Sonkh, etc., whereas the 42–36 cm length bricks are comparable to the bricks of the Śuṅga, Kushan and Gupta periods. Besides, a number of reused bricks (broken) and smaller Post-Gupta bricks are also visible at various positions in the upper portion of the structure. The Sultanate period lākhorī bricks can be observed in the upper part of this brick structure. Various bricks used in this structure confirm the point that the structural edifice at the site dates back to as early as the Mauryan period and continued during the succeeding periods of Śuṅga, Kushan, Gupta, Post-Gupta and medieval periods, before the Madan Mohan temple was constructed atop the mound in the 16th century. Since no regular habitation deposit is observed at or near this site, it is quite probable that it served as a religious edifice, and this point was confirmed by the finding of two inscribed bricks in the structure. These two bricks contained a similar inscription on their face, mentioning “…thena bhāgavatena naṭaken kāritaṃ” which might be slightly incomplete (Upadhyay 1988: 36). Mishra (1983: 20) gives a variant reading as “Thāna Bhāgavata Nanaṭakena Kā(ritam)”. This inscription consists of Brāhmī characters of the 3rd/2nd century BCE, and the use of the word “Bhāgavatena” makes it clear that this religious edifice was dedicated to the Bhāgavatas (Vṛṣṇis). Now, the important repercussions of this discovery are: firstly that the town of Vrindavan dates back to at least as early as third/second century BCE for its religious importance; secondly, the most famous of modern Bhāgavata sites was a Bhāgavata spot during that early time as well; and thirdly, this structural evidence would be the earliest evidence of such an ancient religious structure (Brāhmaṇical) in the whole of north India. The same site has continued to be an important religious place since then, as the finding of various temple remains of the Gupta and Post-Gupta period suggests.

The Mathura Inscription of Vasu

The inscription is engraved on the side of a carved door-jamb dug out of an old well in the Mathura Cantonments in 1913, and is now housed in the Mathura Museum. Lüders (1937–38: 208 [ṣ and c adjusted]) reads:

             1  ..[v]…………………………

             2  sa[s]ya……………………….

             3  [vas]…………………………

             4  [p]………[ṇa]…Śiva..………

             5  ṣapu[t]r[e]ṇa.…Kauśi……….

             6  Vasunā bhaga[va](to Vāsude)-

             7  vasya mahāsthāna…………(śai)-

             8  laṁ toraṇaṁ ve(dikā ca prati)-

             9  ṣṭhāpito prīto [bha](gavān Vāsu)-

            10 devaḥ svāmi[sya] (mahākṣatra)-

            11 pasya Śoḍā[sa](sya)…………

            12 saṁvartayatāṁ………………

The inscription mentions the erection of a stone gateway and a railing by a person named Vasu, during the reign of Mahākṣatrapa Śoḍāsa at the great temple of Bhagavat Vāsudeva. This inscription again confirms the presence of a Bhāgavata (possibly Vṛṣṇi) temple at Mathurā in the 1st century BCE. Since the temple already existed and has been called a “great” temple, it confirms the hoary antiquity of the shrine and the Vṛṣṇi cult at Mathura proper. In no way, this inscription could have come from Mora. Whether it was the same temple which is now considered as the birthplace temple of Kṛṣṇa or some other temple is not clear. There should not be any doubts about the existence of more than one important Bhāgavata shrines at a popular centre like Mathura.

Sunrakh

A broken torso (Fig. 17) was found by me at a shrine in Sunrakh village situated on the outskirts of Vrindavan in 2006, which unfortunately no longer exists there. The headless bust of a standing male was quite damaged and seemed to date to about the Kushan period. The left hand of the deity was kept near his waist, holding a sword. His right hand was completely missing, but might have been kept in abhaya mudrā, as was the norm with Kushan images. Possibly, a sword was tied with his kaṭisūtra, which was distinctly visible in this image along with the thick folds of drapery tied at the waist. He was shown wearing a thick necklace like many other Vaiṣṇava images of the Kushan period. Heavy armlets were visible in the remaining part of the left hand of the deity. The uttarīya, which the deity is shown wearing, appeared to be of indigenous nature and was definitely different from the drapery worn by the Kushan royal images. As the sword was a feature well associated with the Vīra cult like that of the Malhar image (Fig. 4) and the Kondamotu relief (Fig. 5), there is a strong possibility of this image representing a Vṛṣṇi Vīra. The total height of the image would have been almost equal to the height of the Mora torsos (possibly Vṛṣṇi Vīras). The finding of this image nearby the Bhāgavata temple at Dwadash Aditya Tila in Vrindavan again confirms the presence of the Vṛṣṇi cult in ancient times in Vrindavan.

Akrur

Akrur is a small hamlet situated on the outskirts of Vrindavan, along River Yamunā, in between Mathura and Vrindavan. The name of the village is derived from Akrūra, the Vṛṣṇi hero. It is believed that at this place Saṃkarṣaṇa/Balarāma and Vāsudeva-Kṛṣṇa had given him the darśanas of their caturbhuja rūpa when Akrūra was taking them to Mathurā from Vraja. Since Akrūra is the great ancestor of the Varshney caste, they consider this spot to be their holy place. There is a mound on which the modern temple of Akrūra is situated. The mound is an isolated one and appears like a thūpa. Inside the modern temple of Akrūra, the presiding deity was Kṛṣṇa along with Rādhā. Some importance was also given to Saṃkarṣaṇa/Balarāma and Akrūra, but recently, new effigies of Kṛṣṇa, Balarāma and Akrūra have been installed inside the temple garbhagṛha. The importance of this site is not only in terms of legends and literary references, but archaeological, too. In the early decades of the 20th century, a female statue was found from the site, which is now housed in the Government Museum, Mathura (Fig. 18). This statue represents a standing female with her right hand in abhaya mudrā and the left holding the sashes of her drapery. There is also a possibility that she might have held a pot in her left hand. The treatment of her lower portion is very much like that of Ekānaṃśā sculptures and also like that of the Mora female deity (Fig. 7). The height of the figure is comparatively smaller (about 2 feet only). On overall stylistic grounds the image belongs to about the early decades of the 1st century CE, and a similar date is suggested by Quintanilla (2007: 208). Now, a few important points require our attention. Firstly, the female has to be regarded as a major deity, since very few female sculptures in samapāda sthānaka pose of that period are known, and that too exhibiting the abhaya pose. Thus, she has to be a prominent deity, as has been noted earlier (ibid.: 211). Secondly, the figure has been found from a locality which is a sacred Vṛṣṇi place. Thirdly, the similarity of this figure with Ekānaṃśā sculptures of the same age (e.g. G.M.M., no. 15.912, as illustrated in Schmid 2010: fig. 3) suggests its identification as Ekānaṃśā. Whether this identification can be confirmed or not, it is for sure that the place which is being considered sacred and important by the people of Vraja is actually an ancient site dating back to the first century CE or earlier, and, moreover, is associated with a female deity of some significance.

Vṛṣṇi Triads

It was another very important theme of the local Mathura tradition, which later on became universal. In this kind of representation, the three most important personalities of the Vṛṣṇi clan, Saṃkarṣaṇa/Balarāma, Ekānaṃśā and Vāsudeva-Kṛṣṇa, are represented standing side by side. Harivaṃśa Purāṇa II.2.48–49 mentions her name as Kauśikī and locates her place on Vindhyagiri. The same text (II.4.48) also mentions that she was worshipped as Ārādhyadeva by all the devatās born in the Yadukula. Joshi (1977: 122–26) has also pointed out the similarity of Ekānaṃśā with Subhadrā and Vindhyavāsinī Devī, the most popular goddess in eastern U.P. Srinivasan (1997: 213–14) has talked of six Vṛṣṇi triads, with four of these belonging to the Mathura region (G.M.M., nos. 67.529; U 45; 15.912, cf. Schmid 2010: figs. 1–3; one relief in the former Robert H. Ellsworth Collection). The fifth one is a weight stone relief from Pakistan, while the sixth triad was actually found as three separate sculptures from Gaya, Bihar (Patna Museum, nos. 11300; 11269 and 11299). Another Vṛṣṇi triad, which was found at Sanghel near Nuh in Haryana, a part of greater Braj region (Kurukshetra Museum, acc.no. 1/395), has been published recently by Devendra Handa (2006: 38–39, pl. 21). A small triad relief from Bangarmau, situated near Newal in Unnao District and belonging to the Gupta period is also reported (Shukla 1986: 122). This finding is even more important for the fact that Newal is known for a Pañc-Pīr shrine, and the finding of a Vṛṣṇi triad confirms its association with the Vṛṣṇi cult, and thus, very probably, the interlinkage of the Pañc-Pīr with the Pañca Vṛṣṇi Vīras. The already mentioned rock painting from Tikla should be considered the earliest representation of a Vṛṣṇi triad till now known in India. Except for the Gaya examples and the Ellsworth relief (Srinivasan 1997: pl. 16.5), the preservation of these triads is not good, however. A significant addition to these existing triads has come from the excavations at the important port site of Berenike in Egypt (Bhandare online 2023). Three standing figures are carved on a stone slab (Fig. 19). The one on the right is holding a cakra in his left hand and a long mace in his right hand, hence is identifiable as Vāsudeva-Kṛṣṇa; the one on the left is holding a ploughshare in his left hand, hence identifiable as Saṃkarṣaṇa/Balarāma whereas the female figure in the centre is of Ekānaṃśā with her left hand held akimbo possibly holding a pot and the right hand in protection granting posture. The overall treatment of the figures is quite crude and the piece may be datable to the early centuries of the Common era based on other contemporary findings. It is very important to note that the cult of Vṛṣṇis had by then become popular as far as Egypt, though possibly through trade connections.

According to the genealogical hierarchy that is followed in the representation of these triads, the eldest brother Saṃkarṣaṇa/Balarāma is shown as the tallest and the first in the row (on the viewer’s left/proper right). The central figure is that of Ekānaṃśā who according to some texts was marginally elder than Vāsudeva-Kṛṣṇa, but Harivaṃśa Purāṇa II.2.35 mentions that she was slightly younger than him (she was born on bhādrapada kṛṣṇa navamī, while Vāsudeva-Kṛṣṇa was born on bhādrapada kṛṣṇa aṣṭamī). Her height is the least, as was a norm for a female representation. To the left of Ekānaṃśā, Vāsudeva-Kṛṣṇa is represented. Generally, the two male deities are shown with four hands, having their usual attributes. Ekānaṃśā is always depicted with two hands in these triads; one of her hands is placed near the waist with or without holding a water pot, while the right hand is generally shown in abhaya mudrā (cf. Couture & Schmid 2001: 181–89). Doris Srinivasan (1981: 132) has tried to establish some correlation between the Vṛṣṇi triads and the performance of śrāddha, on the basis of a passage in Viṣṇudharmottara Purāṇa and the finding of a triad from Gaya, the most important place for the performance of śrāddha in India. One of the most popular of Vaiṣṇava shrines in India, the temple of Jagannāth in Puri, houses the Vṛṣṇi triad as the main cult object of worship, and the antiquity of the shrine as a local one might date back to a time prior to the 7th century, developing into a major shrine around the 10th/11th century.

Caturvyūha

It is one of the most important and interesting representations of Brāhmaṇical art of Mathura of the Kushan period (Fig. 20). The sculpture was found as four broken pieces from Saptasamudrī Kūpa in the Museum compound. The total height of the remains is 43.2 cm. The central figure is of Vāsudeva-Kṛṣṇa who is possibly shown four-armed. In his natural left hand, he holds a śaṅkha; in his upper right hand he holds a mace, but no part of this extra hand is visible in the present sculpture. The mace is shown as a palm tree trunk and is positioned just below the body of the second emanation, Saṃkarṣaṇa/Balarāma. The natural right hand of Vāsudeva-Kṛṣṇa is shown in abhaya mudrā, whereas the upper left hand, which usually holds a cakra in Vāsudeva-Kṛṣṇa sculptures, is missing completely. This figure, broken off below the abdomen, wears a short garland of leaves and flowers, armlets and quadruple bracelets, which are incised with a wine pattern, and a plain necklace. He wears a tall flat-topped crown of cylindrical shape, decorated with a repeated rosette pattern. In his ears, two pendants are worn as ear ornaments, but broken now. The eyes of Vāsudeva-Kṛṣṇa are shown looking at the nose (nāsāgradṛṣṭi), indicative of its late Kushan date. The first emanation from the body (from the right shoulder) of Vāsudeva-Kṛṣṇa is of Saṃkarṣaṇa/Balarāma, who is clearly identifiable with his serpent hood as well as the wine goblet held in his left hand. The right hand of Saṃkarṣaṇa/Balarāma, which is now broken, would have been probably raised above the head in abhaya mudrā. The torque (ekāvalī) and the short garland of the Saṃkarṣaṇa/Balarāma figure are quite similar to that of central figure of Vāsudeva-Kṛṣṇa. On the top of Vāsudeva-Kṛṣṇa (from his crown) emerges another emanation, which should be of Pradyumna. Actually, there is an attempt to create an illusion that this emanation is coming out from behind the Vāsudeva-Kṛṣṇa figure. Not much is clear about the features of this figure beyond the fact that it is two-armed. Another emanation would have been issuing out from the left shoulder of Vāsudeva-Kṛṣṇa, but is now completely lost. This figure in all probability would have represented the fourth Pāñcarātra vyūha – Aniruddha, the grandson of Vāsudeva-Kṛṣṇa. A tree has been depicted at the back of this sculpture (Fig. 4.20b), which is related with the concept of vyūhavāda. The main trunk of the tree has been shown on the back of Vāsudeva-Kṛṣṇa figure, from which rise two lateral branches (on the back of Saṃkarṣaṇa/Balarāma and possibly on the back of the broken-off Aniruddha figure) and a central branch (on the back of Pradyumna) towards the top. A slightly similar visualisation of a theological concept is observed on Pañcamukha Śiva Liṅgas.

The vyūhas are named after the elder brother, the son, and the grandson, respectively, of Vāsudeva-Kṛṣṇa, namely Saṃkarṣaṇa/Balarāma, Pradyumna and Aniruddha; and the pair of guṇas connected with these are respectively: jñāna and bala; aiśvarya and vīrya; śakti and tejas (Schrader 1916: 35, as cited by Maxwell 1988: 30; following sentences adapted from ibid.). All six guṇas, before pairing, constitute in aggregate the body of the personal God Vāsudeva-Kṛṣṇa (Ahirbudhnya-saṃhitā 6.25: ṣāḍguṇya-vigrahaṃ devam). These six guṇas are the aprākṛta guṇas, conceived as the ‘attributes’ of the Creator which appear at the first phase of movement in His potential state, in the śuddha-sṛṣṭi, before, or underlying, the manifestations of the triguṇas, whose evolution immediately anticipates the material creation. Vāsudeva is the male aspect of the creative principle quickened into his state of potentiality as the container of the guṇas prior to their projection as pairs in the vyūha stages. Before he is stirred to activity by the waking presence of the female principle (śakti), the source is sleeping potential only, known as Nārāyaṇa; his awoken, actively potential state is God Vāsudeva. Hence in the present Caturvyūha icon, the God Nārāyaṇa, embodying the primeval power, is not represented and the four manifested emanations are the four Vṛṣṇi powers of which Vāsudeva-Kṛṣṇa is the foremost and most important emanation, which is responsible for the three other emanations.

Numismatic Evidence

Various symbols are found on a number of punch-marked coins dating back to the pre-Mauryan and Mauryan periods (c. 5th to 3rd century BCE and earlier). There is a great possibility of some of them belonging to the Bhāgavata (Vṛṣṇi) cult. P.L. Gupta (1989: 127) has proposed the possibility of Balarāma being depicted on a few post-Mauryan punch-marked coins, especially on those from the Mathura region. The fragment of this coin type shows a standing figure holding a plough in his left hand and a long stick (musala) in the right hand. It would be one of the earliest representations of any Vṛṣṇi hero in coinage. There is also a strong possibility of the ‘Mathurā’ symbol identified on these coins by Gupta to be related with Mathura and the Vṛṣṇis. Almost contemporary or slightly later (about early 2nd century BCE) is the unambiguous representation of Vāsudeva and Balarāma on the coins of Agathocles found from Ai-Khanum, Afghanistan (Narain 1976: 73–77). On the obverse – or Greek – side, occurs the figure of Balarāma, holding a club and a plough in his right and left hands, respectively. On the reverse – or Brāhmī – side appears the two-armed Vāsudeva holding the wheel and conch (?). Since the Greek side is more important, the representation of Balarāma indicates his elderly position as compared to Vāsudeva in the Vṛṣṇi lineage. The finding of such coins further confirms the spread and popularity of the Vṛṣṇi cult in northwest India during the early 2nd century BCE. A coin assigned to Vṛṣṇirājanya-gaṇa, c. 1st century BCE, bearing a cakra on the reverse and a pillar surmounted by an animal on the obverse, was probably discovered in northern Punjab (Allan 1936: clv-clix; Sircar 1971: 16). Ten clay sealings of the Vṛṣṇis of the Gupta period from Sunet (Ludhiana, Punjab) are also published by Saraswati (1976), on which musala, gadā, śaṅkha and cakra, the symbols of Balarāma and Vāsudeva, and a composite figure being half elephant and half lion along with the mention of Vṛṣṇirājanya-gaṇa are found.

Vṛṣṇi Cult in South India

Vṛṣṇi Vīra cult in North India is slightly better known than it is in Tamil country or South India. The heroes of Vṛṣṇi dynasty who originated from the environs of Mathura were also quite popular in Tamil country as well from quite early times. A survey of the Sangam literature in its present form indicates that various Tamil concepts of worship besides the totemistic or animistic beliefs, each of five geographical divisions mentioned in it came to be associated with a major deity e.g. Māyōn (Viṣṇu-Kṛṣṇa) was the deity par excellence of mullai (pastoral), Vendan (equated with Indra) of marudam (plains), Varuṇan of neidal (god of coastal region), Seyōn (or Muruga equated with Skanda-Kārttikeya) of kuriñji (hilly) and Korravai (equated with Durgā) of pālai (desert). Apart from these major deities, Vāliyōn (Balarāma) is grouped with other important deities Sūrya, Kāma, Sāma and Śiva in Kalittogai (V. 26). The popularity of Saṁkarṣaṇa/Balarāma as an independent deity is also attested by the presence of independent temples dedicated to him.

[6] The South Indian texts also provide iconographic references of Balarāma which are actually quite similar to what is observed in Balarāma images of Mathura region as well as in references in North Indian texts viz. Purāṇas. So, Balarāma as the elder brother of Kṛṣṇa is mentioned in various works. He was a deity related with agriculture which is indicated by the presence of ploughshare (hala/ nāñjil) as chief attribute in his hand. His complexion is said to be white like milk or conch.[7] He is called as Vāliyōn, i.e., possessor of strength (bala).[8] He is called as the one who wears a single earring.[9] He is said to wear a blue garment[10] and a red garland or sometimes a white kadamba garland.[11] He had the palmyra banner.[12] The story of Kṛṣṇa’s dalliance with Pinnai, which makes the theme of an entire chapter of Śilapaddikāraṁ, mentions the dance of Balarāma along with Māyōn (Kṛṣṇa) and Pinnai (possibly Rādhā of later Hinduism).[13] Various childhood episodes of the life of Kṛṣṇa are mentioned in a number of Tamil works which are part of Sangam literature. There is a belief among some scholars that the Bhāgavata Purāṇa was composed in South India around 9^th century CE.[14] All the important forms of Kṛṣṇa as a Vṛṣṇi Vīra, as a vyūha and an avatāra had following in the Tamil country. Paripāḍal refers to the four vyūhas in the following verse[15]:

Sengaṭkāri karungaṇ vellai

pon kaṭ pachchai paingaṇmā-al

i.e., the dark complexioned Vāsudeva with red eyes, the white complexioned Baladeva with dark eyes, the green complexioned Pradyumna and the soft eyed Aniruddha respectively.

Pradyumna is generally identified with Kāma in the Sangam literature. For Pradyumna specifically, references are almost negligible. In Kalittogai and Paripāḍal, Kama is called as ‘Māl Magan’ which means son of Māl, i.e., Viṣṇu. Sāma, the younger brother of Kāma may be identified with Vṛṣṇi Vīra Sāmba where he is grouped with other deities such as Balarāma, Sūrya, Kāma and Śiva.[16] There is no particular reference to five Vṛṣṇi Vīras as such in Sangam literature as we find in a verse of Vāyu Purāṇa, but the Vṛṣṇi Vīras were well known to them and were also worshipped.

Kondamotu panel (Fig. 5) as already mentioned above and found from Guntur District of Andhra Pradesh is the earliest and the most important evidence of Vṛṣṇi cult in South India. The other important evidence of Pañcha-Vṛṣṇi cult in South India in sculptural art is as follows:

Tiruvallikkeni Temple, Chennai

This temple popularly known as Partha-Sarathi Temple is situated at Tiruv-Allika-Keni in Chennai city. In its sanctum sanctorum by the side of main deity Vāsudeva-Kṛṣṇa stand five other stone images (Fig. 21) which are of Rukmiṇī (consort), Saṃkarṣaṇa/Balarāma (brother), Pradyumna (son), Aniruddha (grandson) and Sātyaki (a kinsmen) which become clear as the officiant pronounces their names while performing the temple rituals. All the images including the chief deity are two handed, not so tall and not very richly ornamented except for the chief deity. Details of these are not much clear. As per the current practice, these are always dressed up and their photography is strictly prohibited.

Tiruvallikkeni is referred to in the Tamil scriptures of the Vaiṣṇavas known as Nālāyiraprabhandhaṁ by the saints Peyālvār, Tirumaliśai-Ālvār and Tirumaṅgai-Ālvār.[17] There is no clue to the exact date of actual construction of this temple except for a reference in Tirumaṅgai-Ālvār’s hymn in Periatirumozhi (verse 130) to a Toṇḍaiyar-kon (Pallava) who is said to have built it.[18] From the stone slab inscription found near the sanctum sanctorum, it is presumed that the present temple was built by the Pallava-malla Dantivarman (770-830 CE) and he may be the king who is mentioned by Tirumaṅgai. According to an inscription of 1564 CE belonging to the period of Vijayanagar ruler Sadāśivadeva Mahārāya, images of Kṛṣṇa, Rukmiṇī (Yugamaṇi), Bala (deva), Madana (Pradyumna), Aniruddha (Anuruddiran) and Sātyaki (Sattigai Ālvār) were installed in the temple.[19] It should actually be understood as a reinstallation.

Another significant aspect about this shrine is that it has four other separate shrines dedicated to Varadarāja, Narasiṃha, Rāma and Ranganātha but it is only the shrine of Narasiṁha which has a separate Flag-staff installed in front of the deity. Now, this feature needs to be looked in relation to the Kondamotu relief described above. There also in the center, prominent position is occupied by Narasiṃha and same is the case in Tiruvallikkeni Temple. It suggests that there was a merger of Sātvat cult with the cult of Narasiṃha (which later on became a very important part of Vaikuṇṭha mūrti) in about early centuries of the Common era in the Tamil country. This aspect needs to be studied further.

Nāchchiyār-Kōyil Temple, Kumbhakonam

The other temple where a group of Vṛṣṇi Vīras is enshrined in the garbhagṛha is the Viṣṇu Temple at Tirunaraiyur or Nāchchiyār-Kōyil, about 10 km distant from Kumbhakonam in Thanjavur district. The temple is believed to have been raised by Koch-chengaṇān (Red-eyed king)[20] who is usually assigned to the period immediately after the close of the Sangam age, i.e., fourth-fifth centuries CE or even slight earlier. The present day temple is a renovated one and is dedicated to the Lord Śrīnivāsa or Venkaṭeśa. It is popular as Nāchchiyār-Kōyil (the temple of goddess) who is the enshrined consort of Śrīnivāsa. But the images of Śrīnivāsa and his consort are not the only ones set up in the shrine but four more images - those of Balarāma, Pradyumna, Aniruddha and Sāmba are there. This temple differs from Tiruvallikeni in the inclusion of Sāmba in the group of Vṛṣṇi Vīras in place of Sātyaki. So, this group in Nāchchiyār-Kōyil represents more of a family link of Kṛṣṇa than the Tiruvallikeni Temple.

Temple at Tiruttangal, Sivaganga

At Tiruttangal, 3 km from Sivaganga in Ramanathapuram district is another group of temples where Vṛṣṇi Vīras are found to be in worship. On a hillock are three temples and the one at the top is dedicated to Viṣṇu as Ninra-Nārāyaṇa. Though this temple was renovated in the thirteenth century, it dates from a very early period as the god of the temple has been sung by the Vaiṣṇava Ālvārs of seventh-eighth centuries. The shrine houses a number of images among whom the pride of the place is assigned to Kṛṣṇa (Ninra-Nārāyaṇa) himself. To his right are the images of two of his consorts, one of whom is Jāmbavatī. The line of image breaks, then, to the right and left; on the right are the images of Pradyumna and Garuḍa. On the left are Uṣā, Aniruddha and Bhṛgu.

At the conclusion of this brief study of Vṛṣṇi cult in South India, it can be said with certainty that the cult was widely popular in the region as was the case with the North India. How it spread there, is a matter of further research but it should be believed that it should have been popular by the early centuries of the Common era since we get a number of references related with some Vṛṣṇi heroes. Trade activities and mass movement of people from north to south might be one of the reasons of this cult reaching south. The incorporation of local deities mentioned in Sangam literature in the cult of such prominent deities also requires further research. Whether the Vṛṣṇi cult forms some kind of relation with Northern Mathura and the Southern Mathura (i.e., Madurai) may also be looked at but it is quite interesting to note that there is not a single surviving temple in existence in North India dedicated to Vṛṣṇis whereas in South India we have at least three such temples. There must be various missing links in between which would come up to our knowledge with future research and field work.

Conclusion

The above study confirms the widespread popularity of the Vṛṣṇi cult in a major part of ancient India. Its presence attested during the 3rd/2nd century BCE at far-off places like Afghanistan, Vidisha and Malhar clearly proves that by that time many aspects of the cult had already evolved and developed. In the continuing centuries the cult became popular as far as Berenike in Egypt and in South India. A number of Vṛṣṇi findings in the Vraja region contradict the view of some scholars who consider the origin of this cult outside Vraja. Mathura (Vraja) was the most important centre of the Bhāgavata (Vṛṣṇi) cult and most of its doctrines evolved at Mathura, as the occurrence of almost all the aspects (independent images, triads, Caturvyūha, Pañcavṛṣṇi Vīras) indicate. Various independent images of a four-armed male deity holding discus, conch and mace are identified as Vāsudeva-Kṛṣṇa, whereas a number of representations belong to Saṃkarṣaṇa/Balarāma. Thus, at Mathura, the cult of the Vṛṣṇis was so popular that no other place can think of rivalling it. The continuous presence of people belonging to the Varshney caste in the Vraja region confirms the historicity of the Vṛṣṇis and the importance of Mathura.

This is an updated version of my previously published articles:

“Vṛṣṇis in Ancient Literature and Art” in Indology's Pulse Arts in Context, Doris Meth Srinivasan festschrift volume, Eds. Corinna Wessels Mevissen and Gerd Mevissen with assistance of Vinay Kumar Gupta, Delhi, 2019, Aryan Books International, pp. 69-90.

Rangarajan, Haripriya & Vinay Kumar Gupta “Vṛṣṇi Cult in South India”. History Today XVI. Delhi, 2015, pp.  82–85.

Endnotes