India

by John Keay

  On one of these temples, a rather plain construction dedicated to a Jain saint at Aihole, the poet Ravikirti recorded Pulakesin II’s successes. Reminiscent of Samudra-Gupta’s great Allahabad inscription, this record has the bonus of a date, equivalent to 636 in the Christian calendar. It makes the shrine ‘one of the earliest dated temples in India’11 and, as noted earlier, has provided a benchmark for chronological calculations reaching back even to Manu and the Flood. Here too, in presumptuously comparing his literary talents with those of Kalidasa, Ravikirti provides the earliest dated reference to ‘Sanskrit’s Shakespeare’; whenever Kalidasa lived, he must have been well dead by 636.

  Of more relevance to the Chalukyas is the detailed listing of Pulakesin II’s extensive conquests. Since he succeeded to the throne after a period of internal strife, he had first to consolidate his hold on the Badami region, his base, by again subduing the Kadambas, Gangas and other rival kings in Karnataka. It was probably after this feat that he assumed the titles of maharajadhiraja and paramesvara (‘lord of the others’). The west coast (Konkan) from Goa up to and beyond where Bombay now stands was also subjugated while several of its islands, probably including that of Elephanta, were assaulted by the Chalukyan navy. Further north the Malavas of Malwa and the Gurjaras of southern Rajasthan submitted; and a Chalukyan viceroyalty was established in Gujarat. Clearly Chalukyan forces had crossed both the Tapti and Narmada rivers and were therefore threatening Harsha and his confederates.

  Next in the Aihole listing comes Harsha himself. His stature is acknowledged in a well-worn cliché about his lotus feet gleaming with the jewels of those who bowed to his sway. But in identifying this formidable challenger the poet also introduces a neat pun on the word harsa (‘harsha’). Harsa as a noun means ‘joy’, and thus Pulakesin’s victory is signified by a phrase about how ‘the harsa [of his enemy] was melted away by fear’. Another source has it simply that the lord of the Daksinapatha (the ‘South’) routed the Lord of the Uttarapatha (the ‘North’).

  Pulakesin’s circuit of conquest then continued east, flattening more rivals and reaching the Bay of Bengal in Orissa. Most of the rich lands comprising the Kistna and Godavari deltas in what is now Andhra Pradesh were placed in the care of his younger brother, whose descendants would constitute the ‘Eastern Chalukyas’, a dynasty which would survive until the eleventh century when it merged with its Chola allies of Tamil Nadu. From Vengi, as the Eastern Chalukya kingdom would be called, Pulakesin II resumed his victorious progress down the east coast into Pallava territory. Again his champions and their punch-drunk elephants triumphed as the Pallava king was forced to seek safety within the walls of Kanchipuram. There, unwisely in view of the sequel, Pulakesin left the Pallava and continued south. He crossed the Kaveri and completed his circuit by accepting overtures of friendship from the ancient kingdoms of the extreme south – the Cholas of the Kaveri delta, the Pandyas of Madurai and the Cheras of the Kerala coast.

  Now ‘lord of both the eastern and the western seas’ and indisputably master of all India south of the Vindhya hills, Pulakesin II returned to Badami. Hsuan Tsang calls him a ksatriya yet credits him with magnanimity and foresight, qualities rarely accorded to a ‘heretic’ by the devout Buddhist, let alone to an enemy of his beloved Harsha. He is not mentioned in any other Chinese sources but some authorities insist that an Indian mission received by Khusru II of Persia in 625 must have been from the Chalukyan king.

  However far-flung his fame, Pulakesin II’s manoeuvres as listed in the Aihole inscription are of great interest as an illustration of the theory of Indian paramountcy. The assumption is usually made that his triumphs, like those of Samudra-Gupta as recorded on the Allahabad pillar, are organised in chronological sequence. It cannot be proved; but what here is self-evident is that, whether chronological or not, they were certainly logical. Pulakesin was doing the rounds of his neighbours. South, west, north, east, and back to the south, the Chalukyan was circling – or was seen to be circling – a universe of territory, riding its bounds as it were just like Raghu in Kalidasa’s Raghu-vamsa. Both kings were, in Indian terms, defining a raja-mandala, the diagram of concentric ‘circles of kings’ which is discussed at length in Kautilya’s Arthasastra and other works on political theory.

  In Indian cosmology the mandala design commonly serves as a map. At the centre is the sacred Mount Meru, the axis of the world, outside which the innermost circle is divided into four lands (dvipa); one of these four, jambu-dvipa (‘the land of the rose-apple’), is the earth. Outside this, the next circle is the sea, the next more land, then more sea and so on. The seas are filled with, or named after, familiar liquids – obviously salt-water in the case of the first, then treacle, wine, butter and other kitchen ingredients. To literal minds, like that of Thomas Babington Macaulay, minds which had been schooled on the scientific certainties and rational arguments of the European Enlightenment, these ‘seas of treacle and butter’ would seem contemptible absurdities; India’s only hope of advancement lay in forsaking such nonsense, and to this end Macaulay, in a famous minute on Indian education in the 1830s, would issue a damning and still resented indictment of Indian culture as he insisted that India’s schools forsake Sanskrit and adopt a Western-style curriculum.

  A no less exasperated attitude is detectable in many nineteenth-century attempts to reconstruct India’s history. From inscriptions and sasana the flowery epithets about lotus-footed ancestors and star-bright toenails were ruthlessly discarded in an effort to extract some credible nugget of political or genealogical import. The raja-mandala as a useful symbol of political relations suffered the same fate. Crudely, it represented the idea that just as cosmic harmony depended on the hierarchy of gods and men actively participating in the triumph of dharma, so political harmony depended on the triumph of dharma through an ordered hierarchy of kings. But because this earthly hierarchy was constantly under threat from the matsya-nyaya (the ‘big fish eats little fish’ syndrome), it required frequent adjustments.

  The raja-mandala, in which the maharajadhiraja took the place of Mount Meru at the centre or axis, demonstrated the basic principle of these adjustments. Thus the immediate neighbours of the axial ‘king of kings’, those therefore within the first circle, are to be regarded as his natural enemies; those beyond them in the next circle are his potential allies; those in the third circle are his enemies’ potential allies, those in the fourth his allies’ natural allies, and so on. According to Kautilya, this was the basis of all external relations and of any world order.

  Additionally the raja-mandala, when represented as a diagram, was divided by vertical and horizontal radials into four quadrants or quarters. These were seen to correspond to the four dwipa, or lands, of a mandala map. Harsha’s digvijaya, or ‘conquest of the four quarters’, was therefore a bid for universal dominion. In the same way the maharajadhiraja who would be a cakravartin, a ‘wheel-turning’ world-ruler, must as it were weld the rims to the hub by spokes of conquest and alliance and so oblige the kings within each circle of the mandala-raja to acquiesce in and harmonise with his new and, of course, self-centring world order.

  This geography, indeed geometry, of empire was crucial. It presupposed that ‘society of kings’ already mentioned and it necessitated frequent or, in the case of Harsha and Pulakesin, almost continual perambulation of one’s domains. But it also made conflict a largely dynastic affair which, though of great frequency, may have been of low intensity. The troops involved seem to have been professional warriors who, while dependent on local supplies and transport, otherwise left the agricultural classes alone, as in Megasthenes’ day. Acts symbolic of submission were highly prized; so was the acquisition of accumulated wealth, war elephants, musical instruments, jewels and other symbols of sovereignty. On the other hand the heavy casualties and widespread devastation implied by boasts of ‘annihilation’ cannot be substantiated, nor is there evidence of any consequent economic collapse. On the contrary, the ease with which ‘uprooted’ kings again to
ok root suggests an almost ritualised form of warfare not unlike that which survived, even into the twentieth century, amongst another society of Hindu kings – namely that on the Indonesian island of Bali.

  Back in the south India of the seventh century, while Pulakesin II was still celebrating his success in overrunning the rich Pallava country in Tamil Nadu, the Pallava king was ready to take the field again. At Polilur, a place near Kanchipuram where the British would suffer one of their worst defeats in India, the Pallava king claims to have ‘annihilated his enemies’, presumably the Chalukyas, and by 642 he was marching on their capital at Badami. Pallava records claim that Badami was then destroyed and that the Pallava king, Narasimha-varman I, made such a habit of defeating the great Pulakesin II that he fancied he could read the word ‘victory’ engraved on his adversary’s backside as he again took to flight. More certainly Narasimha-varman engraved a record of his success on a rock at Badami and thereafter assumed the title of Vatapi-konda, ‘conqueror of Vatapi [i.e. Badami]’.

  The Chalukyas would return the compliment. Pulakesin seems to have died in the midst of these reverses and the Chalukyan kingdom to have remained in relapse during a succession crisis. But in 655 one of Pulakesin’s sons, Vikramaditya I, claimed the throne, quickly reasserted Chalukyan sovereignty, and was soon hammering again at the Pallavas. This time Kanchi was surrendered. Then once again the Pallavas struck back. With intermissions while the Pallavas dealt with the Pandyas of Madurai to the south or went to the aid of their allies in Sri Lanka, and while the Chalukyas saw off their own rivals, including the first Arab incursion into Gujarat, the ding-dong struggle between the paramount powers of the arid Deccan and lush Tamil coast continued for over a century. Not infrequently it well demonstrated the Kautilyan raja-mandala. The Pandyas, the southern neighbours and so natural enemies of the Pallavas, assisted the Chalukyas, while the Pandyas’ neighbours and natural enemies, the Cheras of Kerala and the kings of Sri Lanka, rendered support to the Pallavas.

  In C740 Vikramaditya II, a Chalukya, again captured Kanchi and this time took the opportunity to leave a record of his success. His inscription in the soft sandstone of one of the pillars of the Pallavas’ just-built Kailasanatha temple is still legible and boasts not only of his conquest but also of his generosity to the city, which he spared, and to the temple, to which he returned the gold that belonged to it. Significantly, as with the Pallava inscription at Badami, no attempt seems to have been made to erase this patronising record when the Pallavas duly recovered their capital.

  Nor does this almost constant warfare with its frequent ‘annihilations’ seem to have inhibited either dynasty in the practice of kingship. The sasanas, from which our knowledge of their struggles is largely derived, continued to be issued; and the great temples, for which both dynasties are now best remembered, continued to be built. Narasimha-varman I, Pulakesin II’s eventual conqueror, was also known as Mahamalla or Mamalla (‘great wrestler’), and after him the Pallavas’ main port at Mamallapuram (Mahabalipuram) was named. There the famous stone-cut temples, or raths, each hewn from a single giant stone, were probably the work of Narasimha-varman II (also known as Rajasimha), who ‘assumed titles galore – about 250 of them’12 – and reigned from C695-C728. He also built the so-called Shore temple at Mamallapuram, and began the Kailasanatha at Kanchi.

  His Chalukya contemporary was Vijayaditya, the grandson of Vikramaditya I and another man of many titles and many temples; most of the structures at Aihole belong to his reign. He also began, but never completed, the first temple at Pattadakal. A level site lying between the twin towns of Badami and Aihole, Pattadakal would under his successors usurp the ceremonial role of both places as the commemorative capital of the Chalukyas. Here during the first half of the eighth century the Chalukyan temples assumed a size and magnificence of ornamentation unsurpassed by anything in contemporary India and rivalled only by the temples of Kanchi. But whereas today the latter are scattered about a large city richly endowed with later architecture, at Pattadakal, always a site rather than a city, the temples now rear up amongst soggy fields of sugarcane where a mud village and a milky cup of tea is the height of modern magnificence.

  Two of these temples, parked side by side like vehicles from another planet, were commissioned by two sisters who were the successive wives of Vikramaditya II, he who left his mark on Kanchi’s Kailasanatha temple. Celebrating this victory, the sisters’ twin temples closely resemble the Kailasanatha and so are indisputably of the so-called Dravida style (which climaxes with the great eleventh-century Chola temple of Tanjore). Others, however, both here and at Aihole, show features like the curvilinear sikhara (tower) which are distinctive of what used to be called the Nagara or northern style of temple (as famously represented by the later Khajuraho temples). There are also examples of the straight-sided pyramidal style of tower later associated with the Orissan temples, especially of Bhuvaneshwar. It seems unlikely that, as once thought, all these variations were developed by the Chalukyas’ architects. Sculpture and iconography show Gupta influences and imply rather that the far-ranging Chalukyas, in their architecture as in their empire, made of the great Deccan divide a bridge between north and south.

  * * *

  THE CHALUKYAS AND THE PALLAVAS The Rival Successions

  * * *

  Culturally their Pallava rivals look to have performed the same bridging role between the Indian subcontinent and the Indic kingdoms of south-east Asia. No region or dynasty of India had a monopoly of south-east Asian contacts. We know that Bengal had regular contacts with both mainland south-east Asia and its archipelago; Fa Hian sailed for Indonesia or Malaya from the Bengali port of Tamralipti, and many Chinese and south-east Asian Buddhists reached the great university of Nalanda in Bihar via the same port. Orissan influences have also been traced in Burma and the Indies; failing any better explanation, it is quite possible that ‘Kling’, the name by which people of Indian origin are still known in Sumatra and parts of Malaysia, derives from ‘Kalinga’, the ancient Orissan kingdom. Likewise Kerala and Gujarat seem to have had regular contacts with south-east Asia, which with the entry of the Arabs into the carrying trade of the Indian Ocean would be greatly increased.

  However, the most pervasive influence in south-east Asia during the fifth to seventh centuries seems to have been that exercised by the Pallavas of Kanchi. In mainland south-east Asia an important new kingdom had begun to emerge in the sixth century. Based in Cambodia, it would soon absorb Funan, the Indic kingdom on the lower Mekong from which it had probably broken away, and would eventually emerge as the great Khmer kingdom of Angkor. Its kings, like many of those of Funan and Champa (another Indic state in Vietnam), almost always bore names ending in ‘-varman’, just like the Pallavas. More significantly, they claimed descent from the union of a local princess with a certain Kambu whose descendants were known as ‘Kambujas’.

  From this word came ‘Cambodia’ and ‘Khmer’. But the Kambujas, as both a people and a place, first occur in the epics and the Puranas where they are located in the extreme north-west of the Indian subcontinent, a good three thousand kilometres from Cambodia. It has already been suggested that the sacred geography of the Sanskrit classics tended to get replicated as new regions became Sanskritised (e.g. Mathura, Madurai, and Madura in Indonesia). Kambuja’s improbable removal from the upper Indus to the lower Mekong looks to be another case in point. Moreover the adoption of Kambu as a common ancestor would seem to show how such transpositions might have come about, with kings as far away as Indo-China laying claim to the legitimacy provided by an adopted Sanskritic forebear. But what is also significant is that this particular myth seems to have been a revision of the story of the brahman Kaundinya and ‘Willow-Leaf’, his ill-clad local queen. And that in its turn ‘shows a certain kinship with the genealogical myth of the Pallavas of Kanchi’,13 indeed ‘is strikingly similar’ to it.14

  Indo-China apart, the Pallavas are known to have become involved in dynastic struggles in Sri Lan
ka, to have developed Mamallapuram as a long-distance trading station, and to have had diplomatic relations with China. No doubt commercial, religious and political factors all played their part in promoting a more direct, if still conjectural, Pallavan influence in the south-east Asian archipelago. An inscription found in Java uses the Pallava script and that island’s earliest surviving Hindu temples, small stone-built shrines scattered across the misty highlands of Dieng and Gedong Songo, show clear affinities with the architecture of Mamallapuram.

  In Indonesia as in Indo-China important political developments were under way. The eighth century saw the emergence from obscurity of Srivijaya, a maritime power and possibly a dynasty, which would control a seaborne empire stretching from Sumatra to Malaya, Thailand, Cambodia and Vietnam. In terms of national psyche the watery imperium of Srivijaya is as important to modern Indonesia, itself ‘a pelagic state’, as is the continental empire of the Mauryas to Indian centralists. Like Cham-pa and Cambodia, Srivijaya was nevertheless a decidedly Indianised polity, although apparently more Buddhist than brahmanical. Its capital, near Palembang in south-eastern Sumatra, looks to have been the place where in the late seventh century I-tsing (I-ching), another Chinese scholar, found a thriving monastic community. From its monks he received preliminary instruction before proceeding on to Bengal and Nalanda. Returning, he lived with the Srivijayan Buddhists for several years as he worked on the translation of texts acquired in India.

  Also in the seventh and eighth centuries there arose in central Java the rival, but eventually joint, kingdoms of the Sailendra and Sanjaya. The origins of these dynasties and their relationship with Srivijaya, let alone India, are subjects of much debate; but to one or both of them must be ascribed the first glorious phase of Javanese temple-building which began C780. As in the Deccan and south India, the temples are all clustered within a small compass, here centred on the city of Jogjakarta. Moreover many conform in all but detail to the norms of layout and elevation found at the Pallavan and Chalukyan sites.