THE RISE AND FALL OF THE HINDU GOD OF WAR
by Pratapaditya Pal
August 01, 2012
text and photos © Pratapaditya Pal and image copyright holders
(click on the small image for full screen image with captions)
The Rise of Mahāsena: The Transformation of Skanda-Kārttikeya in North India from the Kuṣāṇa to Gupta Empires, by Richard T. Mann (Brill: Leiden. Boston 2012. pp. XIV and 282. Figs 43), is a study of the early history or development and decline of the god of war in the Brahmanical/ Hindu pantheon. Despite the chronological parameters indicated in the long subtitle, the book, in fact, is more ambitious. As the author explains in the Introduction (p.1) the study covers the period from the fourth century BCE to the 7th century CE, whereas the Kushan period or the Kushan Empire is dated from the first century CE to the end of the fourth century CE (p.1, n.1). Although Mann prefers to use the spatial expression "Kushan empire" rather than the temporal "period", the extent of the empire is nowhere defined, nor for that matter the Gupta Empire or period. While there is less certainty about the beginning of the Kushan Empire, at least the Gupta period is known to have begun in 320 CE. It is also known that the Kushan empire, by and large, extended from the present day western Uttar Pradesh to Afghanistan in the north expanding from north to south, the indigenous Gupta empire spread from the present day Bihar/Bengal in the east and at its height expanded across much of northern India to the Arabian Sea in the west.
Fig. 1The reader must know a great deal of the background to understand the complex narrative of the origins and early development of a Hindu deity that is known by myriad names, the most familiar being Skanda, Kumāra and Kārttikeya, and, in south India, Murugan or Murukan in Tamil. Less well-known are Guha, Mahāsena, Śākha, Viśākha etc., among which Mahāsena has been selected for the title of the book because it occurs in Kushan historical sources, such as epigraphs, as a popular name (Mahāsena literally means the great warrior though the more correct form should have been Mahasenāpati or the Great Generalissimo). These are only a few of the deity's numerous names, each of which may well have been a separate personality just as the epithets themselves have very different meanings. This however is the case with all major Hindu divinities, each of whom has a thousand or more names.
Fig. 2In a brief Introduction Richard A. Mann who is a religious historian succinctly outlines his thesis as follows. Delving deeply into the early literature, the late Vedic texts and the epics Mahābhārata and Rāmāyaṇa, mostly he demonstrates that Skanda Kumāra began his career as the leader of a group of popular deities identified as graha (literally grasper or seizer), a term used to denote both planets and “a class of dangerous spirits that possessed pregnant women and young children,” generally infants. In due course these seizers were transformed by propitiation into protectors and assimilated into one conceptual deity with different names. Using both literary and material evidence he proposes that these protective spirits were transformed into the auspicious general of the gods and of the Brahamanical (later Hindu) pantheon and mythologized into the son of Shiva during the Kushan period. While there is nothing new in this proposition his arguments are the most detailed and his marshalling of the literary evidence is impressive for its depth and breadth.
He then argues that "the transformation is brought about by two forces, which, while they rarely openly acknowledge the existence of the other, intersect in a dynamic process that changes not just Skanda, but Hinduism as a whole." (p.1)
One of these forces he avers is primarily religious, the attempts by Brahmanical authorities to assimilate the presumably non-Brahmanical malevolent spirits into a mega-benevolent composite deity as narrated not only in the late Vedic texts and the epics but also in early Ayurvedic and astrological texts.
Fig. 3The second force that radically transformed the nature of this rising but assimilated cultive figure, Mann argues forcefully, is political to personify the exigencies and demands of an empire – that of the polyglot Kushan empire. Being intruders from beyond the Indian borders themselves, the Kushans had to exert authority over a complex society of indigenous and other foreign groups who had preceded them into the subcontinent. A new deity with a syncretic personality that incorporated both extraneous and indigenous concepts and traits became a desideratum. As Mann writes (p.2), "I argue that the Kushans and the cultural heterogeneity found in their realm create important changes in early Hinduism, changes that are encapsulated in the history of Kārttikeya during that reign."
Here again the thesis is not altogether novel but the learned author makes his case cogently with copious evidence both literary and materialistic. Where he does deviate from previous scholars is in his claim that having reached its peak in the Kushan period when the moniker Mahāsena or great general was preferred, the decline of the deity began in the following Gupta period. Undoubtedly, the conventional wisdom had been validated by the composition of the great poetic opus called Kumārasambhava or The Birth of Kumāra by the greatest Sanskrit poet Kālidāsa (c. 400 CE), the popularity of the names of the divinity with the Gupta emperors (witness Kumāragupta and Skandagupta) and the redaction of the voluminous Skandapurāṇa. While some scholars may not agree with Mann, there is no doubt that he presents his case diligently and persuasively in chapters 8 and 9 of the book using literature, epigraphy, numismatics as well as the plastic arts, as he does in his earlier chapters (5-7) for the Kushans and the Yaudheyas. Here also he differs from most scholars as to whether the Yaudheyas (who are known to have flourished in a large swath of land in today's Haryana, Punjab, Uttar Pradesh and Rajasthan between the second century BCE and the fourth century CE) formed one or more "tribal republics", preferring the terms "state" or "group", (I personally feel the expression "polity" is the most appropriate). In any event, the evidence for the popularity of the cult of Kumāra among the Yaudheyas is based primarily on numismatics and as I am not a numismatist the validity of Mann's arguments are best left to future scholars. What is clear, however, is that the six-headed form of Kumāra holding a spear indicating his martial aspect appears for the first time in the earliest Yaudheya coins characterized by the great numismatist John Allan (even though some scholars doubt his suggested dates) as class three coins. Yet again in Yaudheya coins we encounter still another moniker for the divinity in "Brahmaṇyadeva," clearly indicating Brahmanization of the deity, though a son of Shiva, probably by the Brahmanical priests of the Yaudheyas.
Fig. 4A particularly agreeable aspect of Mann’s study is the ample use of the visual evidence with as much aplomb as the literary, unlike most historians of religion. However, it is unfortunate that he considered it more important to provide the source of the photograph, which is of lesser use to the reader, than the date of the object (even in the list of illustrations he does not provide any chronological indicator of the artworks he uses). I am also uncertain whether all statues of the Kushana period from both Mathura (a place name in modern usage to which pedantically and unnecessarily he adds a long vowel sign as some Western scholars do even while this is not done in the name “Mathura Government Museum”) and Gandhāra (does the ordinary reader really know the difference in pronunciation of the two other “a”s in this word even with the sign above the second “a”?) can be identified as Mahāsena. Mann identifies the spear-bearing, turbaned figures when accompanying the Mātris or Mother Goddesses as only Skanda (pp. 262-264) (see fig. 1) but when alone the deity is designated as Mahāsena (see Fig. 2). In my opinion the identification of a figure as Mahāsena should be confined to those pieces where inscriptions specifically use the moniker. Noteworthy is that in all instances at Kushana Mathura Kumāra’s only identifying attribute is the spear while in Gandhāra in addition to the spear he also holds the cock, which, incidentally is an occasional iconographical identifier of the deity in Kushana coins and appears in the class six coins of the Yaudheyas which belong to the post-Kushana period (pp.269-270) (see Fig. 3). On the early Yaudheya issues, however, where he is identified as Brahmaṇyadeva (p.260), he is consistently portrayed with six heads (as is Shashti the goddess of the sixth). To date no stone or metal image of a six-headed Kumāra has been found from the Kushana period. Curiously, the Gandhāra iconographic convention of portraying Kumāra with a cock in his left hand while skipping the intervening swath of land is found in a monumental Kushana period image at Mitawali in the Morena district of Madhya Pradesh (p. 271) (see fig. 4), at Samlaji (5th century) in Rajasthan (p.278) and at Elephanta and Ellora (note that none of these modern place names – correctly in my opinion – is provided with the long vowel sign on the last a; pp. 279-280). Why the Mitawali and Samlaji figures are identified as Skanda-Mahāsena and the Elephanta and Ellora representations only as Skanda is not clear. What is interesting, however, is that with one or two exceptions the god is represented as a young adult wearing turbans in the Kushana period and subsequently a tiara. However, in the Udaygiri cave images (p.274 and 277, fig. 37) (see fig. 5) the hairstyle is that of an adolescent boy which is also clear in a few Kushana period presentations from both Mathura (p.264) and Gandhāra but not on the Kushana coins. Only in one small statue from Gandhāra is the deity given the long coat of mail of a foreign warrior; otherwise he is clad in a dhoti and turban with unshod feet in the Indian mode. It is the only work where he is at least dressed as a great warrior (mahāsena).
Fig. 5Mann has correctly interpreted the political significance of the warrior god Mahāsena for the Kushana rulers. The royal cult also seems to have had a brief revival during the Gupta period as it is clear from the names and coins of the Gupta emperors Kumāragupta and Skandagupta and the poet Kālidāsa’s Kumārasambhava, perhaps written to celebrate the birth of the terrestrial prince. If Mann is correct in his assertion that the cult of Skanda-Kumāra was already on the wane during the Gupta period, curiously, it continued to play a role in regal mythology, if not in the need to legitimization, among the Kadamba rulers of Vanavāsi (4th – 6th centuries) on the western Deccan and the Chalukyas of Badami who overthrew the Kadambas. Significantly the moniker Mahāsena remains popular in the royal inscriptions of both these dynasties which would corroborate its military significance for the Kushanas.
The magnificent and monumental reliefs of Skanda both at the sixth century Elephanta and later at Ellora do not entirely denote that his importance had substantially diminished during the Gupta period. In fact, his identity with the Tamil Murukan mentioned in the Sangam literature (ca. 4th C. BCE- 4th C. CE) must have been completed by this time, a subject to which our author is probably turning his attention even as we pen this review. In a footnote to his Introduction he states, “This current study limits itself to the northern Sanskrit tradition and does not engage the extensive Tamil tradition of Murukaṇ, a topic that I intend to treat elsewhere.” (pp. 2-3) We eagerly await the result of that engagement and treatment.
Notes on illustrations
Stylistically, the Mitawali Skanda-Mahāsena (fig. 4, Mann's fig. 27) may well be the earliest and most monumental lithic representation of the deity found to date in the subcontinent. It is certainly akin formally to the colossal statues of Yakshas from central India (cf. early B.C.E. freestanding examples from Uttar Pradesh and Rajasthan in Sonya Quintanilla, History of Early Stone Scultpure at Mathura, c. 150 B.C.E.-100 C.E. [Leiden, Boston, Brill, 2007, for numerous examples]. It would seem therefore that the iconography for the deity was wellknown long before the appearance of the Kushans. Can the Mitawali figure be identified as Mahāsena?
Fig. 6Another noteworthy feature is that in both the Mathura reliefs (fig. 1, Mann's fig. 13) and the freestanding figure (fig. 2, Mann's fig. 14) Kumāra does not have the cock or rooster but only the spear, whereas both attributes appeared earlier in the Mitawali representation (fig. 4, Mann's fig. 27). Moreover, in both Mathura depictions he is a child, emphasized in the statue by his face and hairstyle, though the body is of an adult. In the Udaygiri representation (fig. 5, Mann's fig. 32) which follows the Mitawali figure for the attributes, the god is portrayed as a young adult with heroic proportions. In the Gandhāra relief (fig. 3, Mann's fig. 17), however, he is an older but young warrior which is emphasized further by the long coat of mail, a feature obviously rejected in the tradition further south, where he is more a prince than a soldier. His Indianess in the Gandhāra reliefs is emphasized by his unshod feet and the turban. Apart from demonstrating the iconographical fluidity in the early period, clearly some artists were conscious of the literary tradition that describes Kumāra's instant transformation into the divine general soon after his birth. It may further be noted that the artists of Nepal in the post-Gupta age remained more steadfast than their Indian counterparts in portraying him as a child god rather than even a young adult, as seen in a splendid stone image of the Licchavi period now in the Alsdorf collection in Chicago (fig. 6).