Pramāṇa · Search · About · API

The Colas c18: Chapter X. Rājēndra—(A.D. 1012-1044)

The Cōḷas c18

Language
English

Contents

Chapter X. Rājēndra—(A.D. 1012-1044)

c18.2

Parakēsari-varman Rājēndra-cōḷadēva I was declared heir apparent and formally associated with his father in the administration of the Empire in the closing years of his rule. Accordingly he counts his regnal years from some date between 27 March and 7 July, A.D. 1012¹ and we find clear evidence of joint rule in the third year of Rājēndra. This year is mentioned in Rājarāja's Tanjore inscriptions of the twenty-ninth year;² Rājarāja is also stated to have made a gift in the third year of his son's rule.³ The star of Rājēndra's nativity was Ārdrā.⁴

c18.3

Rājēndra inherited from his father an extensive empire comprising the whole of the modern Madras and Andhra and parts of Mysore and the island of Ceylon. The administration had been carefully organised and a fairly powerful bureaucracy brought into existence which, while it scrupulously respected the 'liberties,' feudal and corporate, of the various magnates and associations that studded the land, successfully maintained the king's peace, and enforced all civil rights. The army was a strong and tried body of men, well able to defend the wide land frontier and to keep down any threatened outbreaks in areas newly subjected to the empire, and ready for aggressive warfare abroad. The hold on Ceylon and some other islands like the Maldives was securely maintained by a powerful naval force which also served to protect the considerable overseas trade of the empire with the islands of the East Indies and with China. During the thirty-three years of his reign, Rājēndra turned these initial advantages to the best possible use and succeeded in raising the Cōḷa empire to the position of the most extensive and most respected Hindu state of his time, and one which possessed though perhaps only for a time a not inconsiderable dominion over the Malay peninsula and the Eastern Archipelago. The history of Rājēndra's reign is very largely the history of the extensive wars and conquests he undertook in the first half of his reign. Like his father, he has left behind in his stone inscriptions and in the Tiruvālangāḍu and Karandai (Tanjore) copper-plates a trustworthy account of the military and naval transactions of his reign which receive confirmation at all points at which we have the means of controlling it by other evidence.

c18.4

The most common form of Rājēndra's Tamil *praśasti* is that which begins *tiru manni vaḻara*. We come across this form as early as the third year, though it is more generally employed only from the fifth. This introduction progressively increases in length by the narration of fresh conquests until the thirteenth year, after which date it becomes stereotyped. By tracing its growth we are able to fix the internal chronology of the reign with a precision not often attained in the annals of Ancient India. Another Tamil *praśasti* known so far from only one inscription⁵ of the tenth year, gives some particulars about Pāṇḍyan affairs which are confirmed by the Tiruvālangāḍu plates. In another inscription dated in the twenty-fourth year,⁶ the Tamil *praśasti* follows the usual form up to the conquest of Takkaṇalāḍam, and then copies the *praśasti* of another king; most probably a mistake occurred in the re-engraving and two inscriptions got mixed up.⁷ Lastly, the account given of Rājēndra's military achievements in the Sanskrit part of the Tiruvālangāḍu plates was written, quite obviously,⁸ after all conquests were over, and by the side of indications gained from the different stages of the Tamil *praśasti* in different years, this Sanskrit account must be held to be of decidedly inferior import to the chronology of the reign. But the poet Nārāyaṇa who composed the long Sanskrit *praśasti* is entitled to our admiration not only for his high literary quality but for the attention he gives to the facts of his patron's reign which is fuller than is usual with court-poets. In fact, his account forms, at several points, a valuable supplement to the Tamil *praśasti*.

c18.5

Very early in Rājēndra's reign, he appointed his son Rājādhirāja as *yuvarāja* to assist him in the work of the state. Calculations made from Rājādhirāja's records have led to the conclusion that this happened on some day between 15

c18.6

March and 3 December a.d. 1018,⁹ most probably in the early months of the seventh year of Rājēndra himself. For over twenty-five years from that date, father and son, Parakēsari and Rājāhkēsari ruled together and shared the burdens of empire. The inscriptions of Rājādhirāja, with the introduction beginning *tingalērtaru*, give an account of the part played by this prince in the campaigns of his father, and up to about the twenty-sixth year of Rājādhirāja's reign, his records must be understood to supplement those of Rājēndra by giving an alternative description of more or less the same transactions from the standpoint of Rājādhirāja's part in them.¹⁰ In an inscription of his twenty-sixth year from Tirumalavāḍi which adopts the shorter form of his introduction,¹¹ Rājādhirāja's umbrella of state is said to have functioned as if it were the shadow of the white umbrella of his father who conquered with his army the Ganges in the North, Ceylon in the South, Mahōdai in the West and Kaḍāram in the East. This clear statement that the son ruled in full regal state in the life-time of his father, and that for as many as twenty-six years, provides the key for the proper understanding of an important aspect of Cōḷa history. Adopted in the first instance, possibly, as a device to obviate disputed succession, the system of choosing a successor in the life-time of the ruling king and associating him, after a formal installation, in the discharge of important public duties pointed the way, under the stress of empire, to a more deliberate and extensive application of the principle to the administrative arrangements of the empire. The princes of the blood royal who had come of age were appointed to positions of authority in the different provinces of the empire, care being taken to give to each the position suited to his capacity and talent. The person chosen as heir apparent was distinguished from the rest by a formal installation, and by his higher status; in the case of Rājādhirāja, who was not the eldest son of Rājēndra and must have been chosen as *yuvarāja* for his outstanding ability, his status is clearly implied by his separate historical introductions and by his titles even during his father's life-time. The contemporary Cōḷa-Pāṇḍya viceroys, also Cōḷa princes, either use no historical introductions in their inscriptions, or when they do so, adopt that of the ruling sovereign, Rājēndra. In the early years of his reign Rājēndra seems to have been assisted only by one such subordinate ruler with jurisdiction over the Pāṇḍya and the Kēraḷa countries, besides the crown prince Rājādhirāja, though it is possible that in later years others were similarly recognized elsewhere. It is probable that when chosen for such employment, the princes underwent an investiture of some sort which, among other things, conferred on them the right to wear coronets of their own, suited to their relative positions in the state. Rājādhirāja and his successor are said, in their inscriptions, also to have continued these arrangements. This wise system initiated by Rājēndra I, by finding suitable occupation for the energies of restless princes of the royal family, doubtless allayed their discontent, diminished the chances of palace intrigues and revolutions, and at the same time brought new strength to the administration of an over-grown empire which was called upon to face many difficult problems, domestic and foreign.

c18.7

We shall now trace the progress of Rājēndra's conquests as recorded in the sources above mentioned, accepting as our basis the precise indications of chronology in the *tiru manni vaḷara* introduction. From this introduction we learn that up to the third year of his reign he had conquered Iḍiturai-nāḍu, Banavāse (Vanavāsi) encircled by a continuous hedge of forest, Koḷḷippākkai whose ramparts were surrounded by *śuḷḷi* trees, and Maṇṇaikkaḍakkam whose fortifications were unapproachable. Whether this account is a retrospective version of Rājēndra's invasion of Raṭṭapāḍi of a.d. 1003-6 so vividly described in the Hoṭṭūr record of 1007, or recounts a second campaign necessitated by the recovery of Satyāśraya, cannot be decided with certainty. In any event, the campaign took place before a.d. 1008 because it was directed against Satyāśraya himself, who was succeeded by Vikramāditya V about that year. Iḍiturai-nāḍu, Eḍedore 2000, 'was a stretch of country between the rivers Kṛṣṇā on the north and Tungabhadrā on the south, comprising a large part of the present Raichur district.'¹² Koḷḷippākkai was, doubtless, Kulpak, about 45 miles north-east of Hyderabad, 'invariably termed Koḷḷipāka in the inscriptions,'¹³ and the centre of a 7000 district which was later on, at the beginning of the next century, ruled for, some time by Sōmēśvara III, as viceroy under his father Vikramāditya VI. Kulpak was still the centre of a province ruled by a governor under the Kākatīyas in the thirteenth century. In Maṇṇaikkaḍakkam with its strong fortifications one may easily recognise the Mānyakhēṭa which according to the Kanyākumārī inscription¹⁴ became the sporting ground of Rājēndra's forces in battle. Mānyakhēṭa (Mālkhed) must have suffered greatly in this invasion of Rājēndra. In the closing years of Rāṣṭrakūṭa domination, it had been once sacked and plundered by the Paramāra ruler of Mālwa; and now, forty years later, under the Cāḷukyas, a worse fate overtook the unfortunate city. This must have weighed greatly with the Cāḷukyas who, soon after, transferred the capital to Kalyāṇi or Kalyāṇapura, 48 miles to the N. E. of Mālkhed.¹⁵ An inscription of Rājēndra from the Trichinopoly district, dated in his third year,¹⁶ mentions an incident in the fight with Satyāśraya before Maṇṇaikkaḍakkam. A certain Śrutimān Nakkan Candran, a native of Ūṟṟattūr, was ordered directly by the king (*perumāḷ tiruvāyāl moḻiya*) to attack the enemy's elephant and apparently lost his life, and for his benefit a gift was made to the temple of Mahādēva in his native village of Ūṟṟattūr. In this campaign, therefore, Rājēndra crossed the Tungabhadrā, carried the war into the heart of the Cāḷukya country and attacked their very capital. It is not easy to see how these operations across the Tungabhadrā could have been coupled with an attack on Banavāse, long celebrated as the centre of Kadamba power, and then part of the W. Cāḷukya kingdom. Banavāse lies much outside, and too far to the west of the line of advance indicated by the rest of this campaign;¹⁷ and yet the language of the inscription seems to render it obligatory to treat the whole as one campaign, and the names mentioned as the different stages, though not in that order, of a single expedition. This and the date of the Hoṭṭūr inscription go far to suggest that the data furnished by Rājēndra's inscriptions also belong to the war which was waged about A.D. 1004 against Satyāśraya by Rājarāja, and that in this war the advance of Rājēndra against the Cāḷukyas started from somewhere in the N.W. of Mysore, took a generally north-eastern line along the course of the Tungabhadrā until the Raichūr doab was reached, when a more northerly course was struck, and Kulpak and Mālkhed became the objects of attack. A *maṇḍapa* at Tiruvorriyūr called Maṇṇaikoṇḍa-Śōḻa was among the mementos of this campaign.¹⁸

c18.8

The conquest of the whole of Īḻa-maṇḍalam (Ceylon) is the next achievement mentioned. As some of the inscriptions of the fifth year¹⁹ do not include this conquest, while others²⁰ do so, we may be sure that the war against Ceylon was undertaken in the fifth year, a.d. 1017-18. The *Mahāvaṁsa* places the completion of the Cōḷa conquest of Ceylon in the 36th year of Mahinda V which falls also in a.d. 1017 according to the latest scheme of Ceylonese chronology established by Geiger.²¹ About twelve years before this date, Rājarāja, taking advantage of the confusion in the island kingdom which arose from a military revolt against Mahinda V, had brought it under his power with the exception of its remoter parts which were still held by the Sinhalese. As a result of his expedition Rājēndra claims to have captured the crown of the kings of Ceylon²² the exceedingly beautiful crowns of their queens, the fine crown and the garland of Indra which the Pāṇḍya had previously deposited with them (the kings of Ceylon), and the whole Īḻa-maṇḍalam on the transparent sea. The Karandai (Tanjore) plates (vv. 58-9) say that Rājēndra conquered the king of Ceylon with a fierce army and seized his territory, his crown, his queen and her crown, his daughter, all his wealth, his transports, and the spotless garland of Indra and crown of the Pāṇḍya left in his charge; after having lost the battle, and being shorn of his queen, son and other belongings, the king of Ceylon, out of fear, came and sought the two feet of Rājēndra as shelter. The *Mahāvaṁsa* does not mince matters and gives a straight account which confirms the claims made by Rājēndra in his inscriptions :²³

c18.9

‘In the six and thirtieth year of the king’s (Mahinda V’s) reign the Cōḷas seized the mahēsi, the jewels, the diadem that he had inherited, the whole of the (royal) ornaments, the priceless diamond bracelet, a gift of the gods, the unbreakable sword and the relic of the torn strip of cloth.²⁴ But the Ruler himself, who had fled in fear to the jungle, they captured alive, with the pretence of making a treaty. Thereupon they sent the Monarch

c18.10

and all the treasures which had fallen into their hands at once to the Cōḷa Monarch. In the three fraternities and in all Laṅkā (breaking open) the relic chambers, (they carried away) many costly images of gold etc., and while they violently destroyed here and there all the monasteries, like blood-sucking yakkhas, they took all the treasures of Laṅkā for themselves. With Pulatthinagara as base, the Cōḷas held sway over Rājarattha as far as the locality known as Rakkhapāsāṇakaṇṭha. . . . King Mahinda dwelt twelve years in the Cōḷa land and entered into heaven in the forty-eighth year (from his ascent of the throne.)'

c18.11

Rājēndra thus succeeded in getting hold of the Pāṇḍyan regalia left behind by Rājasimha, which Parāntaka I had sought and failed to secure. The Cōḷa inscriptions are silent on the details of the conquest and draw a veil over the pillage of Laṅkā so vividly chronicled in the Ceylonese annals, though one inscription apparently mentions Mahinda's submission to the Cōḷa king after he was transported to the mainland.²⁵ Rājēndra's success was complete, and the whole of the island became a Cōḷa province. We have some inscriptions of Rājēndra at Polonnaruwa and in the Colombo museum with the *tiru manni vaḷara* introduction;²⁶ but these are in a very damaged condition, and valuable only as furnishing clear epigraphical confirmation of Rājēndra's conquest and rule in Ceylon. Several Hindu temples, Dēvāles devoted to Śiva and to Viṣṇu, have been discovered in the vicinity of Polonnaruwa; these are built of stone and in the Tamil Cōḷa style of architecture, and all of them must have been constructed in this period of Cōḷa rule in the island. The *Mahāvaṃsa* adds that twelve years after Rājēndra's invasion, possibly also after the death of Mahinda V, his son Kassapa who had been brought up in secret by the Sinhalese for fear of the Cōḷas, became the centre of a national resistance against the Tamil power, and that, after a six months' war in which a great number of Damiḷas were killed by the Sinhalese forces, he succeeded in making Rohaṇa once more independent of the Tamil province and ruled it as Vikkamabāhu I for a period of twelve years (a.d. 1029-41). The events of Vikkamabāhu's reign belong to a later stage of Rājēndra's rule and may be reserved for further consideration at the proper place.

c18.12

In his sixth year,²⁷ A.D. 1018, Rājēndra seized the heirloom of the Kēraḷa monarch including the crown praised by many and rightfully worn by him, and the garland emitting red rays.²⁸ He also took the 'many ancient islands, whose old, great guard was the ocean which makes the conches resound.' In the next year, he captured the crown of pure gold worthy of Tiru (Lakṣmī) that had been deposited in Sāndimattīvu, in view of its strong fortress, by Paraśurāma who, roused in war, had uprooted the kings (of the world) twenty-one times.²⁹ Some light is thrown on these rather obscure transactions by the account of Rājēndra's southern campaign given by the Tiruvālangāḍu plates and by a single stone inscription of the tenth year³⁰ which gives an account of Rājēndra's policy in the Pāṇḍya country, not found in the other stone records of the reign. The Tiruvālangāḍu plates³¹ say:

c18.13

'This famous and heroic king, possessed of a powerful army and bent upon the performance of meritorious deeds with heaps of money acquired by his own arm, then set his heart upon a *digvijaya*.

c18.14

Accordingly after arranging for the protection (in his absence) of his own capital, the unrivalled king Uttama Cōḷa first started in the direction marked by Triśaṅku (the south) desirous of conquering the Pāṇḍya king.

c18.15

The commander (*daṇḍanātha*) of the ornament (*tilaka*) to the Solar race (Rājēndra) (thereupon) struck the Pāṇḍyan king who had a great force; and the Pāṇḍya abandoned his home in fright and fled for refuge to the Malaya mountain, the abode of Agastya.

c18.16

Rājarāja's son, the master of policy, took possession of the bright spotless pearls, the seeds of the spotless fame of the Pāṇḍya kings.

c18.17

After establishing there his own son, Śrī Cōḷa-Pāṇḍya, for the protection of the Pāṇḍya country, the light of the Solar race then proceeded to the conquest of the west.

c18.18

Having heard of the ignominy sustained by kings at the hands of Bhārgava in battle, this proud king, not

c18.19

'finding him (Bhārgava) on the face of the earth, developed a desire to conquer the land created by him.

Who else, other than this supreme lord (Paramēśvara) can entertain the thought in his mind of subjugating (*lit.* humiliating) that ancient land protected by the glory of the ornament of the Bhṛgukula and free from the inroads of enemies?

The fearless Madhurāntaka crossed the Sahya (mountain) and forthwith set upon the Kēraḷa in great force and there ensued a fierce battle which brought ruin upon kings.

After having (thus) conquered the Kēraḷa kings and harrowed the land guarded by the austerities of the lord of the Bhṛgus, the prince returned to his capital, the abode of prosperity.'

c18.20

It may be doubted if by this campaign Rājēndra added any new territory to his dominions. The Pāṇḍya and Kēraḷa countries had been conquered by Rājarāja very early in his reign, and 'the many ancient islands' towards its end. These islands were the Maldives; the obscure Śāndimattīvu of legendary fame is apparently to be sought also among the islands of the Arabian sea. The effective hold retained by Rājarāja on his southern conquests becomes clear from his numerous records in the Pāṇḍya country and from an inscription of the third year of Rājēndra³² mentioning an endowment in Tiruviśalūr by the queen of a Pāṇḍya king called Śrīvaḷḷuvar. While the Tamil *praśasti* of Rājēndra mentions some new achievements of the king in Ceylon and Kēraḷa, the most considerable among them being the seizure of the regalia of the different kingdoms of the south, it has nothing to say on Pāṇḍyan affairs. The vague statements of the Tiruvālaṅgāḍu plates that the Pāṇḍyan king fled to the Agastya hills and that Rājēndra took his pearls are too conventional to be accepted literally; a stone inscription of the tenth year,³³ however, confirms the statement of the copper-plates that the king established his own son in Madura as his viceroy with the title Cōḷa-Pāṇḍya and also states that Rājēndra built a palace in Madura 'by whose weight the earth became unsteady.' The same inscription implies that Rājēndra repeated his father’s performance at Kāndaḷūr-śālai. This Pāṇḍyan viceroy, who was soon after placed in charge of the Kēraḷa country as well, was Jaṭāvarman Sundara Cōḷa-Pāṇḍya to whose time belong the largest number of the Cōḷa-Pāṇḍya inscriptions so far known.

c18.21

An inscription³⁴ of Rājēndra informs us that in his twenty-fourth regnal year the emperor made a grant of land to the temple built at Mannārkōvil (Tinnevelly) by the Cēra king Rājasimha and called Rājēndra-Śōḷaviṇṇagar, and that this grant was to take effect from the fifteenth year of Jaṭāvarman Sundara Cōḷa-Pāṇḍya. Assuming that the fifteenth year of the viceroy fell either in the twenty-fourth year of the emperor or possibly a little earlier, Jaṭāvarman Sundara Cōḷa-Pāṇḍya must have been appointed viceroy about the sixth or seventh year of Rājēndra, a date which fits in with the indications furnished by the Tiruvālangāḍu plates taken along with the Tamil *praśasti*. The Mannārkōvil inscription is also valuable in other respects. It shows that the viceroys enjoyed almost regal status and were allowed officially to issue orders dated in their own years of office. It also indicates the close contact maintained between the headquarters of the empire,—Rājēndra was residing in his palace at Kāñcīpuram when this gift was made,—and the viceregal courts. Lastly, the jurisdiction of the Cōḷa-Pāṇḍya viceroy over the Cēra country is clearly seen in the Cēra king building a temple in the Pāṇḍya country and naming it after the Cōḷa emperor. From his inscriptions we see that Jaṭāvarman Sundara Cōḷa-Pāṇḍya ruled as viceroy for at least twenty-three years, up to about a.d. 1040. One of the latest³⁵ of these inscriptions opens with the *praśasti* of Rājēndra himself, (*tiru manni vaḷara*): Another states that Śucīndram in Nāñjināḍ (South Travancore) came to be called Sundara-śōḷa-caturvēdimangalam after the viceroy.³⁶ One curious fact not easy to explain is the record of a grant at Kōṭṭār (near Nagercoil) in south Travancore by an Eastern Cāḷukya prince who called himself Sarvalōkāśraya Śrī Viṣṇuvardhana Mahārāja *alias* Cāḷukya Vijayāditya Vikkiyaṇṇa. The inscription is dated in the 11th year of Sundara Cōḷa-Pāṇḍya³⁷ *i.e.*, about a.d. 1029. Kōṭṭār was a strong fortress of strategic importance in those days and the Cōḷas maintained a strong garrison there; it is quite possible that an Eastern Cāḷukya prince who held an important place in the army lived in Kōṭṭār for some years. But his identity is uncertain in the extreme.

c18.22

In the years A.D. 1021 and 1022, Rājēndra resumed the war against the Western Cāḷukya power. Jayasimha, who had succeeded his brother Vikramāditya V about A.D. 1016, was displaying unwonted energy in recovering territory lost to the Cōḷas in previous wars. The Beḷagāmve inscription of 1019 describes him as defeating the Cōḷas and the Cēras,³⁸ and this is borne out by the presence of his inscriptions about this time in Bellary and the N. W. of Mysore.³⁹ Rājēndra's war against Jayasimha is described in his Tamil *praśasti* in the following terms:

c18.23

'(He captured) the seven and a half lakhs of Raṭṭapāḍi (which was) strong by nature, and vast quantities of treasure, together with the inestimable reputation of Jayasimha who, out of fear and to his great disrepute, turned his back at Muśangi and hid himself.'⁴⁰

c18.24

The assertion that the whole of Raṭṭapāḍi fell into the hands of Rājēndra is, of course, a gross exaggeration. In fact, in the Cōḷa inscriptions of this period, this is no more than a conventional way of recording some temporary advantage gained in the field against the Cāḷukyas. The rest of the statements in the *praśasti* seem to be true. There was an engagement at Muśangi or Muyangi, as it is spelt variously in the inscriptions, and Jayasimha certainly did not get the best of it. Muśangi has been identified with Uccangidrug in the Bellary district; but more likely it was Maski.⁴¹ To a description of this campaign the Tiruvālaṅgāḍu plates devote a considerable number of well-turned verses in the best Kāvya style, and though there are ten such verses,⁴² we learn on the whole less about the actual occurrences in the war from these than from the few lines of the Tamil *praśasti* translated above. The king started from Kāñcīpuram on his march against the Raṭṭa country, there was fierce battle between the forces of the Cōḷa king and those of Jayasimha, the latter fled to the forests and Rājēndra returned to his capital with much booty. Typical of the turns of thought of the composer and suggestive of the date of composition is the verse:⁴³

c18.25

'It may be no wonder that the fire of his anger burst into a flame as it came into contact with the descendant

c18.26

of Taila. This however, is strange that, having crossed the waters of all the oceans, it (the fire of his anger) consumed the enemy fuel (*dviṣadindhana*).'

c18.27

Despite his defeat at Muśangi, and the consequent boast of Rājēndra that he captured Raṭṭapāḍi, Jayasimha was successful in retaining his hold on his territory up to the Tungabhadrā, if not beyond. The Mīraj grant dated in A.D. 1024 shows that Jayasimha was then in possession of Eḍedōre 2000 and affirms that he had regained it after driving out the strong Cōḷa, lord of the five Drāviḍas.⁴⁴ As a rule the Cōḷa-Cāḷukya wars in this period were waged on two fronts—the western front in which Mānyakhēṭa and Kalyāṇi were the objectives of the Cōḷas and the Tungabhadrā the natural frontier between the two powers, and the eastern front which centred round Vēngī the possession of which was coveted by both the parties. Of the occurrences in the east during the war we derive little direct information from the Tamil *praśasti* of Rājendra I which omits them and details only the expedition to the Ganges as the next great event of the reign; but to understand this expedition in its proper setting we must consider the affairs of Vēngī.

c18.28

There Vimalāditya who had succeeded his brother Śaktivarman I died or retired in 1019; Jayasimha availed himself of the opportunity to obstruct the accession of Rājarāja (Narendra), the son of Vimalāditya by Kundavai, by supporting his stepbrother Viṣṇuvardhana Vijayāditya VII. Rājarāja could not celebrate his coronation and appealed for help to his maternal uncle Rājēndra I. A number of short but interesting inscriptions in Kannaḍa and Tamil from Koṭṭaśivaram (Anantapur district)⁴⁵ give us the clue to what happened. One of these is dated in the tenth year of Rājēndra. They mention a Cōḷa general Araiyan Rājarājan *alias* Vikrama-Cōḷa Cōḷiya-varaiyan who by his exploits in the Cāḷukya and Vēngī wars earned such high titles as Nālmaḍi Bhīma, Cōḷana-cakra, Sāmantābharaṇam, Vīrabhūṣaṇam, Edirttavarkālan (Tam. death to opponents) or Ahitaroṭṭaḷivan (Kan.), and Jayasingakula-kāla. There is a Tamil verse among these records stating that he fought with Kalingas, Oḍḍas and Teluṅgas. Another undated inscription⁴⁶ records the flight of the Vēngī king when he heard of the advance of this general ordered by the Cōḷa monarch. The king who fled must have been Vijayāditya, and this flight must have taken place roughly about the same time as the battle of Maṣki. The rulers of Kaliṅga and Oḍḍa appear also to have joined Jayasiṁha II and his protégé Vijayāditya, and the Cōḷa commander had to deal with them also. And the expedition undertaken for their chastisement naturally developed into the celebrated march to the Ganges and the fetching of the holy waters to the new capital that Rājēndra was building. This is clearly implied by a verse in the Chārăla plates of Vīrarājendra.⁴⁷

c18.29

We may now turn to the second phase of the campaign—the march of the army through Kaliṅga and Oḍḍa to the Ganges, and the advance of Rājēndra himself to the Gōdāvari and beyond to protect the rear of the Ganges expeditionary force which was threatened by the sullen rulers of Kaliṅga and Oḍḍa who were acting under instructions from Cāḷukya Jayasiṁha II.⁴⁸

c18.30

In the words of the Tiruvālaṅgāḍu plates:⁴⁹

c18.31

'The light of the Solar race (Rājēndra), mocking Bhagīratha who by the force of his austerities caused the descent of the Gaṅgā, set out to sanctify his own land with the waters of that stream brought (thither) by the strength of his arm.'⁵⁰

c18.32

Judging from its duration, this campaign which lasted less than two years,⁵¹ in which so many kingdoms of the north are stated to have felt the strength of Rājēndra's troops, could hardly have been more than a hurried raid across a vast stretch of country. And the Tiruvālaṅgāḍu plates state explicitly that the expedition was led by one of the king's generals and that Rājēndra met him on his return somewhere on the banks of the Gōdāvari.⁵² The events of the campaign as narrated in the same source may be summed up as follows: After crossing many streams by making the elephants in his army serve as bridges across them, the commander of Vikrama Cōḷa's⁵³ forces first fell upon the strong army of Indraratha and took possession of the territory of that ornament to the Lunar race of kings; then he captured the vast treasures of Raṇaśūra and entered the land of Dharmapāla and subdued him also; thereby, he reached the Ganges and caused the water of the river to be brought by the conquered kings to his sovereign lord Madhurāntaka, whom he met on the banks of the Gōdāvari after conquering Mahīpāla and taking away his fame together with many precious jewels. Then the valorous Rājēndra struck the evil-minded Oṭṭa king and his younger brother and forced from him a tribute of rutting elephants.⁵⁴ Thereupon, after himself killing an elephant that charged him while he was seated on the back of another, the king returned to his own splendid capital.

c18.33

The Tamil *praśasti* records the same transactions, almost in the same order, but with much more detail, as follows:⁵⁵

c18.34

'(He seized) Śakkarakkōṭṭam, whose warriors were brave; Madura-maṇḍalam destroyed in a trice,⁵⁶ the prosperous city of Nā-maṇaik-kōṇam with its dense groves,⁵⁷ Pañcap-paḷḷi whose warriors (bore) cruel bows,⁵⁸ Māśuṇidēśa with its green fields⁵⁹; a large heap of family-treasures together with many (other) treasures (which he carried away), after having captured Indraratha of the ancient race of the moon, together with (his) family, in a fight which took place (at) Ādinagar,⁶⁰ (a city) whose great fame knew no decline⁶¹; Oḍḍa-viṣaya which was difficult of approach on account of its dense forest defence⁶²; the good Kōśalai-nāḍu, where Brahmins assembled; Taṇḍabutti, in whose gardens bees abounded, (land which he acquired) after having destroyed Dharmapāla (in) a hot battle; Takkaṇalāḍam, whose fame reached (all) directions, (and which he occupied) after having forcibly attacked Raṇaśūra; Vangāḷa-dēśā, where the rain water never stopped, (and from which) Gōvindacandra fled, having descended (from his) male elephant; elephants of rare strength, women and treasure, (which he seized) after having been pleased to put to flight in a hot battlefield the strong Mahīpāla⁶³ by the sound of a conch from the deep sea; Uttiralāḍam (on the shore of) the expansive ocean (producing) pearls; and the Gangā whose waters bearing fragrant flowers ⁶⁴ dashed against the bathing places (*tīrtha*).'

c18.35

The facts that Śakkarakkōṭṭam was the first place taken by Rājēndra's army in the course of this campaign and that the king met his victorious general on the banks of the Gōdāvari, on his way back at the end of the campaign imply that the Vēṅgī kingdom had come back to the same relation of close subordinate alliance with the Cōḷa empire which it had held in Rājarāja's time.⁶⁵

c18.36

Śakkarakkōṭṭam has been identified with Cakrakōṭya which finds mention in a Nāgavamśi copper-plate grant from Bastar dated A.D. 1065, and its modern representative is probably Citrakūṭa or Citrakoṭa, 8 miles from Rājāpura where the copper-plates were found. Rājāpura, the capital of Bastar, is itself 22 miles north-west of Jagadālpur, on the bank of the Indrāvati river.⁶⁶ Śakkarakkōṭṭam and the places that follow up to Māśuṇi-dēśam have thus to be sought in the territory contiguous to the Vēṅgī kingdom to the north-west of it. Māśuṇi-dēśam literally means the land of the snakes; the kings of the Chindaka family represented by the Rājāpura plates, called themselves Nāga-vamśo-dbhava (born of the Cobra race), and Bhōgavatī-pura-varēśvara (lord of Bhōgavatī, the best of cities); in a later stone inscription of Śaka 1140, one of them is called Śrī-bhujagavara-bhūṣaṇa-mahārājulu,⁶⁷ the māharāja who was the ornament of the race of the best of serpents. It is perfectly reasonable to suppose that by Māśuṇi-dēśam is meant the land ruled by these kings. On this assumption, Maduraimaṇḍalam, Nāmaṇaikkōṇam and Pañcappaḷḷi must be sought in the same region and held to be parts of Māśuṇi-dēśam. It may be noted that Cakrakōṭa is itself called a maṇḍala⁶⁸ like Madurai-maṇḍalam, and that the donor of the Rājāpura plates is called Madhurāntaka.

c18.37

Of Indraratha of the lunar race, whose defeat at Ādinagar led to the surrender of the Oḍḍa (Orissa) country and the (southern) Kōsala, nothing can be added to Kielborn's suggestion⁶⁹ that he might be the same as the opponent of Bhōja of Dhārā mentioned in the Udaipūr inscription.⁷⁰

c18.38

The Tamil inscription says that after the capture of Kōśalai-nāḍu, the Cōḷa general attacked and overthrew in order Dharmapāla of Daṇḍabhukti, Raṇaśūra of southern Lāḍa and Gōvindacandra of Vaṅgāḷa before he fought with Mahīpāla of Uttara-lāḍa and reached the Ganges. The Tiruvālaṅgāḍu plates, on the other hand, state that the attack on Raṇaśūra preceded that on Dharmapāla, and that the overthrow of Dharmapāla led the Cōḷa general to the banks of the Ganges.

c18.39

They also imply that the conquest of Mahīpāla was achieved on the return march. Obviously, both these accounts cannot be true, and as a choice has to be made, the Tamil *praśasti* which was recorded almost immediately after the campaign must be accepted as the more authentic. On this basis, 'most probably Daṇḍabhukti was the march-land between Orissa and Bengal'⁷¹ and its ruler Dharmapāla, of whom we know nothing more than his name, may have been a relative of Mahīpāla, the powerful Pāla ruler of Bengal at the time. The language of the Tamil inscription appears to suggest,⁷² what seems likely even otherwise, that Mahīpāla had a sort of supremacy over the other chiefs named in this context and that the overthrow of Dharmapāla, Raṇaśūra⁷³ and Gōvindacandra led to the final struggle with Mahīpāla. Lāḍa (Raḍhā) was the ancient name of a part of Bengal, which was bounded on the north by the Ganges, the divisions of Bengal across the river being known as Mithilā and Varēndra.⁷⁴ The conquest of Vangāḷa apparently deflected the course of the Cōḷa army a little to the east, and for the rest of it, its march was due north from the land of Southern Kōsala.⁷⁵

c18.40

There is nothing incredible in this record of an audacious raid into the northern countries ordered by Rājēndra and carried out so thoroughly by his *daṇḍanātha*. It is possible that small successes were magnified into great victories and that any reverses sustained were glozed over; it is likely that the statement of the Tiruvālangāḍu plates⁷⁶ that the water of the Ganges was carried to Rājēndra by the defeated kings of the north at the bidding of the Cōḷa general is a boast without foundation. But of the substantial correctness of the story in its essentials we can entertain no doubt whatever. Partly on account of his imperfect knowledge of the political geography of the period, and more on account of the embellishments introduced into the story by his own imagination, Venkayya greatly underrated the veracity of the inscriptions of Rājēndra, and held that the expedition was nothing more than a pilgrimage to the Ganges.⁷⁷ Though the fetching of the water of the Ganges was perhaps present from the beginning as the object of the expedition,⁷⁸ the motive behind it was undoubtedly an exhibition of the power of the Cōḷa empire and a demonstration of its strength to the rulers of Northern India. Such *digvijayas* were undertaken by all powerful monarchs in India and were enjoined upon them by the political code of the country. The aim of the expedition was then not merely getting down the water of the Ganges to the Cōḷa capital, but doing so after establishing a right of way, so to say, across territories outside the empire by a strong show of force. This becomes clear from the statement that at the end of the expedition Rājēndra erected a 'liquid pillar of victory' (*gaṅgā-jalamayam jayastambham*) in his capital with the waters of the Ganges in the form of the tank Cōḷaganga.⁷⁹ 'The invasion of the great southern conqueror Rājēndra Cōḷa I,' says R. D. Banerji, 'seems to have left some permanent marks in Bengal.... Some obscure Karṇāṭa chief seems to have followed Rājēndra Cōḷa I and settled in western Bengal... From him was descended Sāmantasēna, who is generally taken to be the founder of the Sēna dynasty.'⁸⁰ The Karṇāṭas of Mithilā probably had a similar origin. A commentary, of uncertain date, to the *Siddhāntasārāvaḷi* of Trilōcana Śivācārya mentions the fact that Rājēndra imported Śaivas from the banks of the Ganges into his own kingdom and established them in Kāñcīpura and in the Cōḷa country.⁸¹

c18.41

After he met his victorious general on his return from the Ganges on the banks of the Gōdāvari and after punishing the kings of Kaliṅga and Oḍḍa for their hostile demonstration, Rājēndra enabled his nephew Rājarāja Narendra to perform his long-delayed coronation (16 August, 1922), and perhaps also gave him his daughter Ammaṅgā in marriage at the same time. But Rājarāja was seldom free from trouble during his long reign of forty-one years. He had to flee the country more than once; his half-brother Vijayāditya, though vanquished in fight, never gave up his designs upon the throne and was unceasing in his efforts to bring about his downfall with the aid of the Western Cāḷukyas. Vijayāditya succeeded in driving Rājarāja out and crowning himself king of Vēṅgī on 27 June, A.D. 1031⁸² under the title Viṣṇuvardhana-Vijayāditya. It was probably on this occasion that Cāvaṇarasa, the W. Cāḷukya general, invaded Vēṅgī in force, captured the Vijayavāḍa fort and occupied most of the country.⁸³ Rājarāja once more appealed for Cōḷa help, and the sequel seems to be revealed by the undated Kalidiṇḍi plates of Rājarāja Narendra.⁸⁴ Rājēndra sent a powerful Cōḻa army under the Brāh-man general Rājarāja Brahma Mahārāja and two other officers Uttama Cōḻa Milāḍuḍaiyān and Uttama Cōḻa-Cōḻakōn; in the hotly contested battle of Kalidiṇḍi in the neighbourhood of Vēṅgī all the three Cōḻa commanders laid down their lives, and later Rājarāja built memorial temples dedicated to each of them. But the mission of the Cōḻa army seems to have been successfully accomplished, and we see Rājarāja established on his throne about A.D. 1035.⁸⁵ But this was by no means the end of Rājarāja's troubles. Towards the end of Rājēndra's reign, about 1042, a new ruler, Sōmēśvara I, of Kalyāṇi renewed the aggression, and Rājarāja once again appealed to his Cōḻa uncle and father-in-law; Rājēndra, too old to undertake the task himself, sent his son Rājādhirāja I to deal with the new situation of Vēṅgī; and once more there was a Cōḻa-Cāḷukya war on two fronts. But before we turn to the details of this struggle with Sōmēśvara, other events must receive attention.

c18.42

Rājēndra's overseas expedition against Kaḍāram is mentioned for the first time in his inscriptions of the fourteenth year.⁸⁶ While the Tiru-vālangāḍu plates dismiss this achievement in a half verse which merely records that the king conquered Kaṭāha with his powerful troops that had crossed the ocean,⁸⁷ the Tamil *praśasti* gives a detailed narrative of the expedition and its course in the following words:⁸⁸

c18.43

‘(Who) having despatched many ships in the midst of the rolling sea and having caught Saṅgrāma-vijayōt-tuṅga-varman, the king of Kaḍāram, together with the elephants in his glorious army,⁸⁹ (took) the large heap of treasures, which (that king) had rightfully accumulated; (captured) with noise the (arch called) Vidyādhara-tōraṇa at the “war-gate” of his extensive city;⁹⁰ Śrī Vija-ya⁹¹ with the “jewelled wicket-gate”⁹² adorned with great splendour and the “gate of large jewels”; Paṇṇai with water in its bathing ghats; the ancient Malaiyūr with the strong mountain for its rampart;⁹³ Māyiruḍiṅgam, surrounded by the deep sea (as) by a moat; Ilangāśōka (i.e., Lan-kāśōka) undaunted (in) fierce battles; Māpappāḷam having abundant (deep) water as defence; Mēviḷimbangam having fine walls as defence; Vaḷaippandūru having Viḷap-pandūru (?);⁹⁴ Talaittakkōlam praised by great men,

c18.44

MAP TO ILLUSTRATE THE KAḌĀRAM CAMPAIGN OF RĀJĒNDRA

c18.45

(versed in) the sciences; Mādamālingam, firm in great and fierce battles; Ilāmuridēśam, whose fierce strength rose in war;⁹⁵ Mānakkaṿāram, in whose extensive flower gardens honey was collecting;⁹⁶ and Kaḍāram, of fierce strength, which was protected by the deep sea.’⁹⁷

c18.46

No clearer measure can be required of the progress made in our knowledge of South Indian history than the difference between what was known of this expedition before and what we make of it now. The text of Rājēndra's inscription was recovered and published in 1891⁹⁸ by Hultzsch. The larger Leyden grant had been known already for some years, and Hultzsch recognised at once in Saṅgrāma-vijayōttunga-varman of Rājēndra's inscription, a successor of Māra-vijayōttunga-varman of Kaṭāha or Kaḍāram of the Leyden grant. But his search for this place extended no further than the southern districts of the Madras Presidency, and strangely enough, as it now appears, he overlooked the facts that Rājēndra’s expedition was a naval war and that the Pāṇḍya country had been conquered and subjected to the Cōḷa sway several years before the date of this expedition; and he identified Kaḍāram with the ‘headquarters of a talluqa of the Rāmnād zamindari in the Madura district.’⁹⁹ Even as late as 1903, though a great advance had been made by him from his original position, Hultzsch was still far from the mark when he said:¹⁰⁰ ‘Of the numerous places which are mentioned in connection with this expedition, Mr. Venkayya has identified two, viz., Nakkavāram and Pappālam. The former is the Tamil name of the Nicobar islands,¹⁰¹ and according to the *Mahāvaṁsa* (lxxvi, 63) Papphāla was a port in Ramañña, i.e, the Talaing country of Burma. Hence Kaḍāram will have to be looked for in farther India.’ For some years thereafter, Rājēndra’s expedition was held to have been directed against the kingdom of Pegu, and the archaeologists of Burma even announced their discovery of two octagonal granite pillars near Pegu, which were identified by them ‘with the Jayastambha or pillars of victory set up by Rājēndra Cōḷa who overran Pegu in A.D. 1025-27.’¹⁰² It was only in 1918 that Coedès brought together in his cogent and lucid paper *Le Royaume de Śrī Vijaya*¹⁰³ evidence accumulated along various lines by several years of study on the part of many scholars, discussed fully the identification of the places mentioned in connection with Rājēndra's campaign, and laid the basis for an intelligible account of it. The Archaeological Department of Burma, though at first inclined to be rather critical of Coedès' scheme,¹⁰⁴ later acknowledged its substantial accuracy by removing the celebrated granite pillars from the list of the protected monuments of Burma.¹⁰⁵

c18.47

One fact to which Hultzsch himself drew pointed attention has sometimes escaped the notice of later authors who have discussed this campaign. It is that the inscription clearly implies that all the places named were taken from the king of Kaḍāram and in the course of a single campaign. In the words of Coedès:¹⁰⁶ 'The text says in effect that Rājēndra Cōḷa, I, after having vanquished the king of Kaḍāram, seized his treasures, then a certain number of countries and lastly Kaḍāram. It is a question, then, of one and the same campaign, and it is *a priori* infinitely probable that the different countries enumerated must have been either vassal states of the king of Kaḍāram, or even simply the different towns or provinces of his kingdom.' Once this is recognised, the identification of the different places mentioned would be rendered easier by that of Kaḍāram and Śrī Vijaya, the two places ruled by the same king in the reign of Rājarāja, and conquered by Rājēndra from Saṅgrāmavijayōttunga-varman.

c18.48

'Now, the annals of the Song (dynasty of China) mention, in 1003 and 1008, two embassies from the country of San-fo-tsi, the first sent by the king Sseu-li-tchoulo-wou-ni-fo-ma-tiao-houa and the second by the king Sseu-li-ma-lo-pi. It is not necessary to be a sinologue to recognise in the first name a magnificent transcription of Śrī-Cūḷāmaṇi-varmadēva, and in the second the transcription of the first syllables of *Śrī-Māravi* jayōttunga-varman'¹⁰⁷ (Coedès). As these two monarchs are exactly those mentioned in the larger Leyden grant, we may conclude that the kings of the San-fo-tsi of the Chinese annals were the rulers of Kaḍāram and Śrī Vijaya. San-fo-tsi was first used by the Chinese writers of the Song period for the place called Che-li-fo-che or Fo-che in the earlier literature of China; all Chinese writers have identified this name with Palembang, on the eastern coast of Sumatra. And Coedès has shown good reason for restoring the name San-fo-tsi, Che-li-fo-che, into Śrī Vijaya, rather than the usual but meaningless form Śrībhōja.¹⁰⁸ It thus becomes clear that Śrī Vijaya, which is the first among the places taken by Rājēndra from the king of Kaḍāram, is the name of the kingdom of Palembang in Sumatra. The great part played by this kingdom from about the eighth to the thirteenth century a.d., in the affairs of the Malay peninsula and Archipelago, and the relations of Southern India with this important kingdom still await full elucidation. The epigraphs of Rājēndra's reign which narrate his invasion of Kaḍāram and Śrī Vijaya furnish much welcome information on the affairs of the kingdom at the beginning of the eleventh century. Writing towards the close of the twelfth century a.d. or the beginning of the thirteenth,¹⁰⁹ Chau Ju-kua gives a list of fifteen *chou* (provinces or towns) over which the rule of San-fo-tsi extended;¹¹⁰ and as Coedès has observed, there is a partial coincidence between this list and that of Rājēndra's inscriptions.¹¹¹ The identification of Kaḍāram presents more difficulty; this may be discussed after we have dealt with the other places.

c18.49.1

In the *praśasti* of Rājēndra the name mentioned after Śrī Vijaya is Paṇṇai which has been identified with Pani or Panei on the east coast of Sumatra.¹¹² Ancient Malaiyūr was a principality 'at the southern end of the Malay peninsula, and precisely on the northern shore of the Old Singapore Strait where, besides the Malāyu river, time-worn traditions of a Malaya country and people confront the enquirer.'¹¹³ As for Māyiruḍingam which had the deep sea for its moat, this place is quite obviously the same as *Ji-lo-ting* mentioned by Chau Ju-kua among the dependencies of Śrī Vijaya. The same author also states that Ji-lo-ting and Kia-lo-hi 'are of the same kind' as Tan-ma-ling.¹¹⁴ Coedès has proved by decisive epigraphical evidence that Kia-lo-hi is the same place as Grahi at Jaiya and that consequently Ji-lo-ting (Yi-ru-ḍingam) which formed one of the northern dependencies of Śrī Vijaya must be sought somewhere in the region of Jaiya towards the centre of the Malay peninsula.¹¹⁵ Ilaṅgāśōkam has been very properly identified with Ling-ya-sseu-kia of Chau Ju-kua's list of dependencies, and its locality was to the south of the state of Kedah in the Malay peninsula.¹¹⁶ Māpappāḷam, as was shown by Venkayya, is mentioned in the *Mahāvaṁsa*¹¹⁷ under the name Papphāḷama,¹¹⁸ as the place where the Tamil general Ādicca landed when he was sent on an expedition against Rāmaññadēśa by Parākramabāhu I of Ceylon about a.d. 1165.

c18.49.2

From this Venkayya concluded that Māpappāḷam must be a place in the Talaing country of Lower Burma, and he has been followed by other writers who have proceeded to make other identifications on this basis.¹¹⁹ In fact it seems at first sight that this mention of Pappāḷam in an expedition against Rāmaññadēśa violently contradicts the assumption that all the places captured by Rājēndra were dependent on Palembang and within easy reach of it. Coedès however, draws attention to the fact that the long list of the grievances which Parākrambāhu had against the ruler of Rāmañña ends with his capture by force of a Sinhalese princess whom the ruler of Laṅkā had sent to the Kāmbhōja country;¹²⁰ and suggests that 'as it is infinitely probable that the messengers going from Ceylon to Kāmbhōja passed by the isthmus of Kra, it is in this region that the abduction (of the princess) must have been committed, and consequently, the authority of the king of Pagan might have extended so far.'¹²¹ In the beginning of the eleventh century, however, the suzerainty of Palembang extended up to the Bay of Bandon, and there is no difficulty therefore in assuming that Māpappāḷam was a locality in the region of the isthmus of Kra, though its exact identity cannot now be made out.

c18.49.3

In any event, the presence, among the conquests of Rājēndra Cōḷa I, of a locality which became part of Pegu in the 12th century, is not sufficient to invalidate the identification of these conquests with the vassal states of Palembang. Mēvilimbangam and Vaḷaippandūru¹²² do not lend themselves to any identification at present. Talaittakkōlam, most probably the same place as Takkōla of the *Milinda-Pañha* and Takkōla of Ptolemy, is localised by Gerini in the modern Takuapa district south of the isthmus of Kra and identified with its chief town, also called Takuapa.¹²³ Others are inclined to locate it somewhat higher up, in the isthmus itself; in any case, there is general agreement that it is a place on the West Coast of the Malay peninsula. Mā-Damālingam, firm in battle, can easily be recognised in the name Tan-maling which figures in Chau Ju-kua's list of the dependencies of San-fo-tsi. The same authority says:¹²⁴ Ling-ya-sseu-kia (Ilaṅgāśōkam) 'can be reached from Tan-ma-ling by sailing six days and nights;

First 50 of 261 passages. Read on

Linked texts (18)