Pramāṇa · Search · About · API

History of India, c. 150 A.D. to 350 A.D.: Naga–Vakataka Imperial Period c16.s3: XIII. Samudra Gupta's Wars in Āryāvarta and in the South

History of India, c. 150 A.D. to 350 A.D.: Nāga–Vākāṭaka Imperial Period c16.s3

Language
English

XIII. Samudra Gupta's Wars in Āryāvarta and in the South

c16.s3.1

§ 131. According to the Allahabad inscription, Samudra Gupta's wars in Āryāvarta were in two parts—one before the Southern expedition and the other after the Southern expedition. The result of these wars was the Gupta Empire as portrayed in the Purāṇas, almost with exactitude, in the shape of its three imperial provinces (§ 125) along with the home province of Anu-Gaṅgā-Prayāga-Sāketa-and-Magadha.

c16.s3.2

132. The first act of Samudra Gupta, which turned the scales of his political fortunes in his favour, was a pitched battle at some place where he defeated two, or rather three, kings—Achyuta, Nāgasena and Gaṇapati Nāga. The immediate result of this was that the Prince of the family of Koṭa (whose name is not given in the verse) was captured by his armies, and Samudra Gupta had the pleasure of re-entering Pushpapura. The verse 7, lines 13 and 14, of the Allahabad pillar inscription describes this as follows:

c16.s3.3

udvelodita-bāhu-viryya-rabhasād-ekena yena kṣaṇād
unmūly-Achyuta-Nāgasena-Ga……
daṇḍair grāhayat-aiva Koṭa-kulajaṁ Pushp-āhvaye krīḍatā,
sūryyene….taṭa………

c16.s3.4

The letters after *Ga* are lost but probably the name of Gaṇapat....was there which is suggested by the remaining letter *ṇa* and the requirement of the metre. This we can gather from the grouping in line 21 which is in prose and which begins the Nāgasena-Achyuta group with Gaṇapati-Nāga:

c16.s3.5

Gaṇapati-Nāga-Nāgasena-Achyuta-Nandi-Balavarma—The most important personality of the group is Gaṇapati Nāga. The great result of the battle being avowedly an easy capture of Pāṭaliputra and of the Prince of the family of Kota, the battle must have been intimately connected with the question of the recovery of Magadha. Samudra Gupta himself did not capture ‘Kota’s descendant’ who was the ruler of Pāṭaliputra at the time. We may take it, therefore, that one army had attacked or laid siege to Pushpapura, and that Samudra Gupta gave battle to Nāgasena and Achyuta and also probably to Gaṇapati at a place other than Pāṭaliputra, and at some distance from Pāṭaliputra. Now, we know from coins and from the *Bhāvaśataka*, a work written under the reign of Gaṇapati Nāga (§ 31) that Gaṇapati Nāga was the ruler of Malwā (*Dhārādhīśa*) with his capital at Padmāvatī and probably a second capital at Dhārā. The name of Achyuta Nandi as set out in full in line 21, and the coin of ‘Achyuta’ found at Ahichhatra which has the same symbols as on the Nāga coins of Padmāvatī and also has the same fabric, suggest that he was a scion of the Nāgas. Nāgasena was probably the son of Kirtisheṇa of Mathurā,¹ father-in-law of Kalyāṇavarman king of Magadha and Pāṭaliputra.² As Kalyāṇavarman who had dispossessed Chaṇḍasena of Pāṭaliputra was related to the king of Mathurā and thus belonged to the confederacy of the Nāga-Vākāṭakas, we find Nāgasena and Achyuta Nandi, in all probability led by Gaṇapati who was a noted soldier and the leader of the Nāgas according to the Bhāvaśataka, meeting Samudra Gupta in a pitched battle. They might have been on their way to the relief of Pāṭaliputra. The convenient place upon which the kings or rulers from Ahichchhatra, Mathurā and Padmāvatī could have converged was Kauśāmbī or Allahabad, more likely the former, as the old royal route to Pāṭaliputra lay through Kauśāmbī. The proclamation of this victory on the Kauśāmbī Pillar seems to convey that meaning. The praśasti was meant to be engraved on this very pillar, as line 30 expressly mentions:

c16.s3.6

¹ This Nāgasena is to be distinguished from the Nāgasena of Padmāvatī, a member of the Nāga family, mentioned by Bāṇa in the *Harsha-Charita*, for the latter had met with his end not on a battle-field, but on account of some political intrigue at Padmāvatī. No coin of his is found. He seems to have been a Gupta subordinate.

c16.s3.7

² *Kaumudī-mahotsava*, Act IV.

c16.s3.8

bāhurayam-uchchhhritaḥ stambhaḥ.

c16.s3.9

All the three rulers or sub-kings were killed in one day (kṣaṇāt) on the battlefield.

c16.s3.10

133. This war may be dated in or about 344-45 A.D. immediately after the death of Pravarasena I, the Vākāṭaka Emperor. This war gave the large tract of the Gangetic valley to Samudra Gupta. From Oudh which already belonged to him and had been his base, his territory extended up to Hardwar and the Siwalik, and to the east, from Allahabad up to Bhagalpur at least, if not up to Bengal which seems to be included by the Purāṇic reference to Pauṇḍra. Leaving probably the valley of the Yamunā for the time being, Samudra Gupta consolidated his power in Magadha and decided upon attacking the south-eastern end of the Vākāṭaka empire. It was far from the centre of the Vākāṭaka seat of power which, up to that time, was in the Kilakilā region. To Samudra Gupta it was nearer from Chota Nagpur. The Vākāṭakas evidently governed their provinces of Kosalā-Mekalā through and from the Central Provinces. Samudra Gupta, therefore, could cause successfully, apart from other military considerations, not only confusion but almost helplessness to the Vākāṭaka Emperor by attacking the Vākāṭakas in Kosalā, Mekalā and Andhra. The Pallavas who occupied an important position in the South at the time were in subordinate alliance with the Vākāṭaka Emperor, being their branch. The four Aśvamedhas of the late Vākāṭaka samrāṭ had given him dominion over all the four quarters of India. In the South Samudra Gupta had a policy more of conciliation than of aggression. He captured rulers and released them, and except for the territories which were integral parts of the Vākāṭaka Empire—Kosalā and Mekalā—he did not annex any southern territory. In Kaliṅga he set up a new feudatory. His progress, consequently, in the South must have been swift. At the same time it was very profitable. The whole of Northern India was soon flooded with gold presumably imported from the South. Samudra Gupta coined only gold money, and at his aśvamedha at a later stage, he struck gold coins to such a volume that he could distribute them to an unprecedented extent.

c16.s3.11

§ 134. It cannot be entertained that the enumeration of the names of the Southern kings and chiefs in the Allahabad record is made at random. The writer Harisheṇa, who was one of Samudra Gupta's marshals and a man intimately associated with the emperor and who held the portfolio of the Minister of Peace and War, must be expected to follow an accurate record of his master's conquests. He was composing history which was intended to be published on an Aśokan pillar for all ages to come. He divided the conquests and submission of All India into Southern, Northern, Western and North-West groups, where he was following a geographical plan with accuracy. The string of names could not have been put in by haphazard. Further, we may assume that the composition must have found approval of the Emperor who was alive when the record was published.¹ Kāñchī, Avamukta, Veṅgī and Palakka are one division. Palakka as Palakkada appears several times in Pallava inscriptions² which refer to grants in the Guntur District, and also to Veṅgorāshṭra which here corresponds with Samudra Gupta's Veṅgī between the Godāvarī and the Krishṇā.

c16.s3.12

¹ Ante, p. 121, n. 2. See Bühler's opinion, J.R.A.S., 1898, p. 386, with which I fully agree.

c16.s3.13

² I.A., Vol. V, pp. 51, 52, 155, consult also E.I., Vol. VIII, p. 159 [kada='place', p. 161.]

c16.s3.14

135. This Southern or the ‘Dakshiṇāpatha’ campaign was not undertaken, as generally supposed, as a digvijaya. It was a military move against the Vākāṭaka power, necessitated by the result of Samudra Gupta's first battle in Āryāvarta in which Gaṇapati Nāga, Achyuta Nandi and Nāgasena fell. The second centre of the Vākāṭaka power was rooted in the Andhra country, where, from the capital Daśanapura¹ the junior branch of the Vākāṭakas had been ruling as 'the Pallava emperors' ('Pallavendra'²) of the South, and had penetrated as far south as Kāñchī, the capital of the Cholas—the most important Tamil State. Samudra Gupta's sole objective in the South was the Pallava army. If, to avenge the destruction of the Vākāṭaka military leaders (Gaṇapati Nāga and others) in the North, the Pallavas with their generals and feudatories from the South and Rudrasena from Bundelkhand invaded Bihar, Samudra Gupta would have been placed between two fires. To avoid this junction, Samudra Gupta would have thought of taking them and dealing with them in detail. He descended swiftly via Chota Nagpur, Sambhalpur and Bastar straight into Veṅgi, the original seat of the Pallavas, and reached the battlefield on the Colair Lake. It is the old route which takes one direct into Andhra-deśa. Samudra Gupta did not follow the East-Coast route, as none of the lower Bengal and Orissa towns is mentioned by his secretary Harisheṇa. The Colair Lake in the seventh century again became the arena of a sanguinary fight in the time of Pulakeśin II.³ If we consider the list of the rulers enumerated by Samudra Gupta's secretary and prince-marshal, we see at once that all these rulers belonged to the regions of Andhra and Kaliṅga lying within the range of the Kurāla or Colair Lake. They were all, evidently, assembled together (§ 135 A) and a decisive battle was fought,⁴ and owing to some clever movement and strategy on the part of Samudra Gupta all the leaders were enveloped, a débâcle followed and they surrendered. Samudra Gupta restored them to liberty on terms. Samudra Gupta returned from this place—the region between Bezwada and Rajahmundry. He had no necessity to go to Kāñchī. Nor was he interested at the time in any other Southern state either on the East Coast or the West Coast. He hurried back to Bihar, having defeated and generously and diplomatically won over the Pallava group and having detached them from the Vākāṭaka allegiance. When back, he marched against Rudra Deva who fought bravely as every one of his northern subordinates did, and was killed with them on the battlefield, probably at Eran (§ 137).

c16.s3.15

¹ E.I., I, 397, where it is described as adhishṭhāna, 'capital'; see Fleet, I.A., V, 154. In the latter inscription it is again called 'capital'—'Vijaya-Daśanapura'.

c16.s3.16

² This is the title by which they are designated both by the Gaṅgas and the Kadambas, their feudatories. E.I., XIV, 331; VIII, 32.

c16.s3.17

³ E.I., VI, pp. 3, 6.

c16.s3.18

⁴ The list is (line 19): (1) Kausalaka Mahendra; (2) Mahākāntāraka Vyāghrarāja; (3) Kaurālaka Maṇṭarāja; (4) Pishṭapuraka-Mahendra-

c16.s3.19

135 A. Samudra Gupta passed through Kosala on his Sambhalpur route, and then Mahākāntāra which on the evidence of the Mahā-Bhārata we have identified with Kanker and Bastar. Next he came to Kurāḷa. He must have passed Veṅgī¹ but the ruler of Veṅgī is placed next to the ruler of Piṣṭapura, the capital of Kaliṅga, which was in the Godavari District. This ruler (Svāmidatta) had also two districts in Ganjam round the hill-fortresses of Mahendragiri and Koṭṭūra. Eraṇḍapalli was a town in Kaliṅga, in the district of Ganjam, in the neighbourhood of Kaliṅganagara (Mukhalingam) which is mentioned in the copperplate of Devendra-varman, found at Siddhantam near Chicacole (E.I., XIII, 212). This district must have been under Svāmidatta of Piṣṭapura, and Damana of Eraṇḍapalli must have been a ‘rājā’ or ruler of the status of a district officer. Next to him is Viṣṇugopa, the ruler of Kāñchī, who at the time was the Yuvarāja to his brother Simhavarman I, or probably the guardian of girika-Kauṭṭūraka Svāmidatta; (5) Eraṇḍapallaka Damana; (6) Kāñcheyaka Viṣṇugopa; (7) Avamuktaka Nīla-rāja; (8) Vaiṅgeyaka Hastivarman; (9) Pālakkaka Ugrasena; (10) Daivarāṣṭraka Kubera; (11) Kausthalapuraka Dhanañjaya (prabhṛiti-sarvva-Dakṣiṇāpatha-rāja-, etc.).

c16.s3.20

¹ For its location near Ellore in the Godavari District, see E.I., IX, 56.

c16.s3.21

his son Simhavarman II of Kāñchī. From Eraṇḍapalli to Kāñchī is a big jump. This can be consistent only on the hypothesis that they were together at one and the same spot. Then comes the ruler of Āvamukta or Avamukta. The Ava country or people had their capital Pithuṇḍa near the Godavari. Ava and Pithuṇḍa are noted in the Hathigumpha inscription.¹ After him comes the ruler of Veṅgī, a tract which Samudra Gupta had already passed on his way from Mahākāntāra to Kurāla. It was not possible for Samudra Gupta to go to Kāñchī, if he went there, without meeting the ruler of Veṅgī on the way: this is another proof that all these fighters were at one place. Palakka, as already pointed out, is the same place wherefrom several grants in the district of Guntur and near about Bezwada were made by the early Pallavas. In the grants the name appears as Palakkada. It was situated in the Andhradeśa nearabout the Kṛṣṇā. Devarāṣṭra, which figures next in the person of its ruler, fixes again the location of all these ‘rājans’ at one and the same place. It was a district (vishaya) in Elmañchi-Kaliṅgadeśa (modern Yellamanchilli) according to a plate of Chālukya Bhīma I² whose another plate was discovered at Bezwada. Kusthala-pura must have been similarly some district-place in the same region, though we have not yet recovered its name in any other record. Probably except the rulers of Kosala and Mahākāntāra, all these military chiefs—from the rank of kings, e.g. Svāmidatta and Viṣṇugopa, to that of district magistrates [against whom no conqueror would take the trouble of marching]—must have been together and must have fought at one and the same battlefield. The order in which they are mentioned probably represents the order of the battle-array or the order of their surrender. Their importance is as fighters and military leaders, not as rulers. They seem to have been grouped under two chief leaders: [the numbers before the names denote their order in the Allahabad inscription. See footnote 4 to § 135, pp. 136-137.]

c16.s3.22

¹ E.I., XX, 79; line 11; J.B.O.R.S., XIV, 151.

c16.s3.23

² Madras Report on Epigraphy, 1909, pp. 108-109.

c16.s3.24

(3) MAṆṬA-RĀJA of KURĀḶA and leading
(4) Svāmidatta and
(5) Damana of Eraṇḍapalli.

c16.s3.25

II
(6) VIṢṆUGOPA of KĀÑCHĪ leading
7. Nilarāja of Avamukta,
8. Hastivarman of Veṅgī,
9. Ugrasena of Palakka,
10. Kubera of Devarāṣṭra, and
11. Dhanañjaya of Kusthalapura.

c16.s3.26

The main army was under Viṣṇugopa, who was supported on flanks by the Kaliṅga forces. The battle might be called the ‘Battle of Kurāḷa’. By this battle there was achieved the conquest of the Vākāṭaka provinces of Kosalā, Mekalā and Andhra. Samudra Gupta returned by the same route of Kosalā, as no other countries are noted by Hariṣeṇa. We may date it about 345-346 A.D. It must have followed soon after the battle of Kauśāmbī (344 A.D.). Like Khāravela, Samudra Gupta's campaigns might average one in two years [344 A.D. to 348 A.D.]. Starting after the rains he could have returned to Patna the same year.¹

c16.s3.27

¹ The easy march according to Kautilya (ch. 130) was one yojana (7 miles) a day for an ordinary army; for a better one, 1½ yojanas, and for the best 2 yojanas a day. [The value of yojana to be 7 miles is well ascertained by Cunningham.] The campaign of Samudra Gupta must have been swift.

c16.s3.28

136. Samudra Gupta on his return now invaded the real Vākāṭaka centre, the home-province of the Vākāṭakas in Āryāvarta between the Yamunā and Vidiśā, or to adopt the modern geographical name—the land of Bundelkhand. This Āryāvarta War gave to Samudra Gupta dominion over ‘the Āṭavī rulers [of Āryāvarta]’, i.e., the Vindhyan principalities of Baghelkhand and Eastern Bundelkhand. The war was waged, therefore, in the Vindhyan regions of Āryāvarta, that is, in or about Bundelkhand. To operate in the hills of Pannā is a difficult venture which military leaders would avoid. To the south-west, Bundelkhand is bounded by the districts of Bhilsa [Vidiśā] [Eastern Malwā]. And Bundelkhand is much more accessible from Eastern Malwā, to which there was and has been a good plain road from the Gangetic valley across the Betwa or the Chambal. Samudra Gupta would have attacked the province of the Kilakilā-Vidiśā from the plain country now lying mostly in the Gwalior territory—the passage-land of the Marathas to Hindustan. The battle seems to have been fought at Eran. The reasons for this conclusion are as follows.

c16.s3.29

137. As Samudra Gupta built his monuments at Eran which was in the heart of the Vākāṭaka home-province, it is definite that he had a triumphant march into the Vākāṭaka country. In the reign of the next Vākāṭaka king, Pṛithivīsheṇa I, we find Bundelkhand still under the Vākāṭakas. Immediately to the south of Eran and also to its east there were the Republics (§ 145). Samudra Gupta's founding a Vishnu temple at Eran, which was not in the territory under his direct administration, is suggestive. In the Eran inscription, where the king has not yet assumed the title of Mahārājādhirāja, and where the set genealogy is yet unknown, lines 21 to 26, verses 6 and 7, show that after a military victory, Samudra Gupta raised a war memorial, like the one raised later by his grandson at Bhitari. The inscription is earlier than the Allahabad pillar inscription. The word ‘Antaka’ is emphasised in this inscription; a whole lot of kings (pārthiva-gaṇas sakalaḥ) are noted to have been overthrown and deprived of their sovereignty, and the king is said to have taken a consecration¹ and is described to have become one whose valour had become ‘irresistible’—a title afterwards adopted on his coins. His action in war is given prominence, in line 21, which made his enemies fear him even in sleep. To mark the glory, he established (line 26) a foundation, which is evidently the Vishnu Temple, still existing. On the freeze of the temple a funeral-scene is depicted,² which is unusual, and is to be interpreted as the cremation of the Vākāṭaka king, defeated and slain. The town from that day became a direct, personal possession of the Gupta Emperor—which is the real meaning of the term 'svabhoga-nagara'.

c16.s3.30

¹ ‘Abhisheka’.

c16.s3.31

² A.S.R., Vol. X, p. 85.

c16.s3.32

138. Eran is at the gate of Bundelkhand on one side and Malwa on the other. The whole of Malwa, both eastern and western, was under republics who submitted, without any fight, to Samudra Gupta. It seems that at this place, which had been a strategic point and had an ancient fort and which with a vast stretch of plain country is designed by nature to be an excellent battlefield, the action with the Vākāṭaka king was fought. It became a battlefield again in later Gupta times, as we have the memorial given here to a Gupta general (Goparāja) who fought and died here in the time of the Huns, where 'his devoted wife in close companionship' accompanied him on to the funeral pyre.¹

c16.s3.33

¹ Fleet, G. I., p. 92.

c16.s3.34

139. Rudradeva was defeated and killed on the battlefield by Samudra Gupta. Rudra is the only king whose name is coupled with 'deva' in the inscription of Samudra Gupta, and we may take it to be intentional. Rudrasena was the biggest monarch in India at the time, having succeeded his grandfather who had been a true Emperor of All-India. 'Sena', in Rudrasena is really not a part of the name: as pointed out above we may compare the dropping of sena in the Nepal inscriptions where the Lichchhavi king Vasantasena is sometimes described as Vasanta-sena and sometimes as Vasantadeva. Deva is more dignified and stood for full royal dignity. The chronology proposed above makes Rudradeva succeed in 344 A.D., and Samudra Gupta's conquests are unanimously dated about 345 to 350 A.D. The Rudradeva of the inscription is thus contemporaneous with and identical with Rudrasena I (§ 64).

c16.s3.35

140. The Āryāvarta kings defeated by Samudra Gupta were:

c16.s3.36

Rudradeva, Matila, Nāgadatta, Chandravarman, Gaṇapati Nāga, Nāgasena, Achyuta Nandi, and Balavarman.²

c16.s3.37

² G. I., p. 12.

c16.s3.38

This list is divisible into two parts. (1) From Gaṇapati Nāga to Balavarman are the names of the kings already defeated in the first Āryāvarta War; the battle of Kauśāmbī accounting for the three, the last one, Balavarman, might have been the ruler of Pāṭaliputra captured by Samudra Gupta’s army who is left unnamed in verse 7. If so, B a l a v a r m a n, was the second or the abhiṣeka- [the coronation-] name of Kālyāṇavarman. The other group must therefore consist of the kings and rulers defeated in the Second War or in a probable continuation of the Second War.¹ Of these Nāgadatta is to be identified with the Nāgadatta, father of the Mahārāja Maheśvara Nāga—a Nāga sub-king whose seal bearing the lāñchana of a cobra was found at Lahore and has been edited by Fleet in his Gupta Inscriptions, which according to its script belongs to the fourth century A.D. (G. I., page 283). Matila ruled in the district of Bulandshahr, whose seal with another Nāga emblem was discovered.² We do not know who this C h a n d r a v a r m a n of Samudra Gupta’s inscription was,³ but we know that a Yādava dynasty of feudatories had been set up about 250 A.D. at Siṃhapura in the Jālandhara Doab [§§ 78 and 80]. This house must have been feudatory to the Vākāṭakas. Their names ended in ‘varman’. Although in the list of the Simhapura rulers we do not find any name as Chandravarman yet it is possible that he might have been a younger cadet who came to the battlefield to fight for Rudrasena, or Chandravarman was a second name of some king of the line. The sixth king who would have been a contemporary of Samudra Gupta and whose name is Vṛddhivarman, is described as ‘Chandra’

c16.s3.39

¹ Very likely a little later, another expedition was undertaken to the west of Mathurā in the Srughna country and up to Jālandhara.

c16.s3.40

² I.A., XVIII, p. 289. It is the symbol of the Nāga Saṅkhapāla. There is a conch and a serpent. The shape of the serpent is round and rays (ābhā) come out of its body, i.e. from its body rays emanate. A dhyāna of Durgā describes Saṅkhapāla thus : dāhottīrṇa-suvarṇābha,° used as a kaṅkaṇa (bracelet) by the Goddess.

c16.s3.41

³ Once V. Smith proposed to identify the Chandravarman of Samudra Gupta with the Chandravarman of the Susunia inscription (J.R.A.S., 1897, p. 876). But the script of the latter (E.I., XIII, p. 133) is much later.

c16.s3.42

verse 7 of the Lakkhā Maṇḍal inscription [E.I., Vol. I, 13]. Chandra-varman, according to the Allahabad inscription, was a neighbour of Nāgadatta who must have been a ruler beyond Mathurā and whose successor's seal was found at Lahore. There is no room for Nāgadatta between Ahi-chhatra and Mathurā. The grouping—Rudradeva-Matila-Nāgadatta-Chandravarman—is made up in a geographical order: Matila was next to Rudradeva, Nāgadatta was further west and Chandravarman, furthest, in the Eastern Punjab.

c16.s3.43

140 A. Now the question is whether all these three rulers fought on the side of Rudradeva in one battle or separately. Nāgadatta and Chandravarman were not in the neighbourhood of Rudrasena, but we know from Indian History that kings and allies travelled from long distances to fight pitched battles; it is not unlikely that the three feudatories fought along with Rudradeva in one and the same battle, which must have been expected. This must have been the biggest battle of Samudra Gupta, as he notes that all the Āṭavika kings became his servants after his war with these kings, which means that the rulers of Bundelkhand and Baghelkhand took part in this battle and changed allegiance to the Gupta conqueror on the fall of their sovereign lord. But as to the two western kings or rulers it seems more likely that there was a subsequent engagement to the west of Mathurā. The Purāṇic (Va. and Br.) description of Samudra Gupta's empire at the date of the death of Rudrasena [§129], which does not include the Punjab, would also indicate it. There was thus, more likely, a Third War in Āryāvarta, a year or so later.

c16.s3.44

141. The enveloping movement of Samudra Gupta against the Vākāṭaka Empire was practically a continuation of his First Āryāvarta War. The three big battles form really one continuous campaign. The campaign, therefore, must have been carried through quickly. Under a generalship which was so perfect that it never met with a reverse or check, the campaigns would have been finished easily in three fighting seasons—October [Vijayādaśamī] to April each year. Following the chronology adopted above we may date the First Āryāvarta Battle about 344-345 A.D., the Second Āryāvarta War in or about 348 A.D., and the probable Third War c. 349 or 350 A.D.

Linked texts (9)