Pramāṇa · Search · About · API

The Vakataka–Gupta Age (circa 200–550 A.D.) c12: Chapter VIII

The Vākāṭaka–Gupta Age (circa 200–550 A.D.) c12

Language
English

Contents

Chapter VIII

c12.1

THE EXPANSION OF THE GUPTA EMPIRE.

1. Rāma-gupta

c12.s1.1

Until about twenty years ago it was unanimously held that the great emperor Samudra-gupta was succeeded by his son Chandra-gupta II. Since then, the recovery of a few passages of a lost dramatic work, *Devī-Chandra-gupta* by Viśākhadatta, has thrown altogether new light on this question.¹

c12.s1.2

¹ The question has been discussed by a large number of scholars among whom the following deserve special mention : S. Lévi (*JA.* CCIII, pp. 201 ff); R. Sarasvati (*IA.* LII, pp. 181 ff); A. S. Altekar (*JBORS.* XIV, 223 ff; XV, 134 ff); R. D. Banerji (*AIG.* pp. 26 ff) ; Dr. D. R. Bhandarkar (*Malaviya Comm. Vol.* pp. 189 ff); K. P. Jayaswal (*JBORS.* XVIII, 17 ff); Winternitz (*Aiyangar Comm. Vol.* pp. 359 ff); Sten Konow (*JBORS.* XXIII, 444); V. V. Mirashi (*IHQ.* X, 48; *IA.* LXII, 201); N. Das Gupta (*IC.* IV, 216); V. Raghavan (*Benares Hindu University Magazine*, II, 23-54, 307). H. C. Raychaudhuri (*PHAI.* p. 465).

c12.s1.3

By piecing together the scattered evidences contained in this drama and supplementing them by isolated references contained in the *Harsha-charita*, Sanjan and Cambay copper plates, and the *Kāvyamīmāṃsā*, some scholars have reconstructed the story somewhat as follows :—

c12.s1.4

Samudra-gupta was succeeded by his son Rāma-gupta whose wife was Dhruvadevī. In course of a war with the Śaka king he was closely besieged and placed in such a difficult position that, in order to assure the safety of his people,² he agreed to surrender his queen to the Śaka king. His younger brother Chandra-gupta protested against this act of dishonour, and offered to go to the enemy's camp in the disguise of queen Dhruvadevī in order to kill the hated Śaka king. The stratagem succeeded, and Chandra-gupta saved the empire and its honour. The incident must have raised him in the estimation of his subjects as well as of queen Dhruvadevī, and the character and reputation of Rāma-gupta must have suffered a corresponding decline. There was an estrangement between the two brothers, and Chandra-gupta, presumably afraid of his elder brother's design on his own life, pretended madness. Ultimately, by some means which is not known, Chandra-gupta succeeded in killing his elder brother and not only seized his kingdom but also married his widow.

c12.s1.5

² "Prakritindm-dīrāsandya". Some take it to mean "for satisfying the Councillors".

c12.s1.6

The patience, industry and ingenuity of a number of scholars have thus laid bare a fairly complete picture of a momentous but hitherto unknown episode in the history of the imperial Guptas. We have now to decide whether the picture represents actual facts or is based merely on the imagination of dramatists and story-tellers.

c12.s1.7

While the issue is an important one, the judgment is not an easy process. On the one hand we have the two historical names—Chandra-gupta (II) and Dhruvadevi—as the hero and heroine of a drama whose author, even if not a contemporary of them, as some contend, probably flourished not long afterwards¹; and the essential parts of their story, on which the drama was based, are corroborated by two authoritative sources like Bāṇa's *Harsha-charitra* (7th century A.D.) and Rāshṭrakūṭa copper-plates (9th-10th century A. D.) originating from two distant parts of India. On the other hand it has been pointed out that the version of the story given by Bāṇa and his commentators differs from that known to the author of the *Kāvyamīmāṁsā*, and that details not found in the earlier accounts are added in the days of Amoghavarsha and Govinda IV. Further, the story in itself, even apart from such supernatural elements as *Vetāla-sādhanā* (goblin-worship), is not only unusual, almost bordering on the romantic and incredible, but is also *prima facie* so much opposed to our knowledge of facts and belief in practices of the period, that nothing but the strongest evidence should induce us to place any credence in it.

c12.s1.8

¹ S. Lévi places Viśākhadatta some time between the Gupta dynasty and Harsha. Jayaswal, Sten Konow and N. Das Gupta regard him as a contemporary of Chandra-gupta II. Winternitz, who originally held the same view, gave it up on the discovery of *Devi-Chandra-gupta*, and assigned its author to the sixth century A.D. (cf. B. C. Law Volume, p. 50).

c12.s1.9

We have, for instance, a pretty large number of coins and inscriptions of the Gupta period, but they contain no reference to a king Rāma-gupta, who must be presumed, according to the story, to have ruled over the Gupta empire after the death of Samudra-gupta. Again, while the murder of a brother for the sake of kingdom is by no means unusual, the marriage of his brother's widow by the rebel and the regicide clashes with our cherished notions about morality and social custom prevalent in those times.

c12.s1.10

These objections are not unanswerable. It has been pointed out that neither Śāstric injunctions nor social practices prohibited a marriage between a widow and the younger brother of her husband. But we should remember that the Sangali and Cambay plates definitely condemn the act and even describe it as illicit intercourse. As regards the other point, it has been argued that as the official records give the genealogy, and not the succession of kings, the omission of the name of Rāma-gupta need cause no surprise. The absence of coins is a more serious objection. But some have explained it away by assuming a very short rule of the king, while others have attributed to him the gold coins bearing the name Kācha, already referred to above. Dr. Bhandarkar holds that the name Rāma-gupta, which occurs only once in the passages quoted from the Devī-Chandra-gupta, is a misreading of Kācha-gupta, the real name of the king, while Jayaswal regarded Kācha and Rāma as two different names of the same king.

c12.s1.11

These arguments in support of the story are no doubt plausible, but certainly not convincing. There are, however, other objections besides the two mentioned above. It is difficult, for instance, to believe that the inheritor of the mighty empire of Samudra-gupta could be so decisively defeated by a Śaka king that he had no means of saving his army or kingdom save by consenting to an act, which would be regarded as the most ignominious by any king in any age or country, not to speak of the mighty emperor of the golden age of India who had the blood of Samudra-gupta running in his veins. A story preserved in the *Mujmalu-t-Tawārīkh* has been seized upon as a suitable explanation of this enigma. It has been suggested that the king and his retinue were besieged in a hill-fort, and his army having been defeated by the Śaka king he was at the complete mercy of the latter. But even such a situation can hardly condone the utter infamy and disgrace involved in the proposed means of escape from it. It should be remembered that we can not explain it merely as a caprice of a monarch, who might be imbecile or insane, for we are asked to believe that his action had the full approval of the people, even if it was not instigated by them. The code of honour in the golden age of India must be assumed to have been very different from the later and more degenerate days, when, in similar peril women, preferring death to dishonour, are known to have thrown themselves in blazing fire, and men rushed out and fought till they avenged the insult with the last drop of their blood.

c12.s1.12

These considerations stand in the way of accepting as historical the strange episode of Rāma-gupta, until at least the existence of this king is established on unimpeachable grounds. While the story cannot be dismissed off-hand, as altogether a figment of imagination, we must not rush to the other extreme of accepting, *in toto*, plots of drama and popular tales as reliable facts. In other words, we must suspend our judgment upon the historical character of Rāma-gupta, his fight with the Śakas, and the strange event which deprived him of his throne, life and the natural affection and fidelity of his wife.

c12.s1.13

In view of this, it would hardly serve any useful purpose to discuss at length whether the Śaka opponent of Rāma-gupta was the Śaka Satrap of Western India or a Kuṣāṇa king of the Punjab, and whether the battle took place in the Himālayas, or a place called Alipura (in the Punjab), or Nalinapura (near Jelalabad).

c12.s1.14

These problems cannot be solved until further evidence is available, nor is their solution very material for the broad and general historical question that concerns us for the present. We may therefore proceed to discuss the reign of Chandra-gupta II without any further reference to the episode of Rāma-gupta.

2. Chandra-gupta II

c12.s2.1

When Samudra-gupta died, probably in ripe old age, he left many sons and grandsons behind him. Whether Chandra-gupta II was his eldest son is not definitely known. In the conventional Gupta genealogy, as recorded in royal grants and seals, the expression *tatparigrihita* (accepted by him *i.e.* Samudra-gupta) is applied to Chandra-gupta II, whereas the corresponding expression applied to the kings succeeding him is *tat-pādānudhyāta* (meditating on or favoured by the feet of). This has been taken to indicate that Samudra-gupta chose him as his successor out of his many sons. The acceptance of this view would cut at the very root of the theory that Rāma-gupta succeeded Samudra-gupta. But the two phrases indicating the relationship of two successive kings may be taken as merely conventional expressions of good-will and respect without implying anything more about special selection or immediate succession. We should not, therefore, definitely infer, without any corroborative evidence, that Chandra-gupta II "was chosen out of many sons by his father as the best fitted to succeed him", though this is by no means unlikely.

c12.s2.2

Chandra-gupta had a second name Deva and is referred to as Deva-gupta, Deva rāja or Deva-śri. His mother's name is Dattadevi and we know the names of two of his queens Dhruvadevi (or Dhruvasvāminī) and Kuveranāgā. He is styled Parama-Bhāgavata and was evidently a staunch devotee of the Vaishnava faith.

c12.s2.3

Six records (Nos. 5-10) of the time of Chandra-gupta II are known so far. The earliest, found at Mathura (No. 5), is dated in the year 61 of the Gupta era, corresponding to A.D. 380. The inscription also contains his regnal year. Unfortunately this part of the stone is damaged, and the regnal year cannot be read with absolute certainty. Some have read it as *prathame* (first), and others *pañchame* (fifth). The latter reading seems more probable, and accordingly Chandra-gupta's accession would fall in A.D. 376-7. The inscription proves that on this date Mathura formed an integral part of the Gupta empire. It thus stretched beyond the Jumna river which has been provisionally accepted above as the western boundary of the territory directly administered by Samudra-gupta. Whether it indicates further conquests of Chandra-gupta II or whether Mathura had already formed an integral part of the kingdom ruled over by his father, it is difficult to say. In any case it may be taken to prove that the empire did not probably suffer any diminution, towards the west, after the death of Samudra-gupta.

c12.s2.4

There are clear indications that Chandra-gupta II emulated his father's military career and went out in a campaign of conquest. A cave in Udayagiri hill, about two miles to the north-west of Bhilsa, was dedicated to Sambhu by Virasena, a 'minister of peace and war' of Chandra-gupta II. It is stated in the short inscription (No. 10), recording this gift, that the minister was an inhabitant of Pāṭaliputra, and had accompanied his royal master to Udayagiri while the latter was 'seeking to conquer the whole world'. This undoubtedly refers to a military campaign undertaken by Chandra-gupta II towards the south-western part of the empire.

c12.s2.5

The great success achieved by the Gupta emperor is indirectly attested by coins. It is a significant fact that the long series of coins testifying to the almost unbroken rule of the Western Kshatrapas for more than three hundred years comes to an end between A.D. 388 and 397 and is replaced by coins of similar design issued by Chandra-gupta II. This leaves no doubt that Chandra-gupta II extinguished the power of the Western Kshatrapas and annexed their dominions. This was undoubtedly the main result of the military campaign which he had undertaken 'to conquer the world', as his minister so characteristically describes it.

c12.s2.6

In addition to the record of the minister Virasena, we have another (No. 6) in the same locality referring to the gift of a Sanakānika Mahārāja, a feudatory of Chandra-gupta II, in the year 82 (=A.D. 401-2). Another inscription (No. 8), dated 93 (=A.D. 412-13), records some donations to the great Buddhist Vihāra at Sanchi by Āmrakārddava, who was an official of Chandra-gupta II, and ‘acquired banners of victory and fame in many battles’.

c12.s2.7

The presence of a feudatory, a minister and a military officer of Chandra-gupta II in the same locality in Eastern Malwa may not unreasonably be connected with the protracted military campaign of Chandra-gupta II which may thus be placed during the early years of the fifth century A.D., though it might have commenced even earlier. The coins issued by Chandra-gupta II in imitation of those of the Western Kshatrapas bear dates in Gupta era of which only the first symbol denoting 90 is clearly legible. These coins must therefore have been issued between A.D. 409 and 415 (when Chandra gupta had ceased to reign). Everything thus indicates that Chandra-gupta's military campaign was planned, and the conquest of the Śaka dominions completed, during the closing decade of the fourth and the first decade of the fifth century A.D.

c12.s2.8

By this brilliant conquest the Gupta emperor not only put an end to the domination of the foreigners, who occupied the soil of India for the longest period, but added the rich provinces of Kathiawar and N. Gujarat to the empire which now extended from the Bay of Bengal to the Arabian Sea. The Gupta empire now controlled a large part of the Indian commerce with the western world and was brought into closer contact with the western civilisation. Chandra-gupta's exploits naturally recalled those of king Vikramāditya of Ujjain who is described in Indian legends as having expelled the first Śaka conquerors of India more than four hundred years before. It is presumably in imitation of this legendary hero that Chandra-gupta, like his father, assumed the title Vikramāditya which gradually came to be regarded as a title of distinction by mighty rulers of India famed for their military exploits. It is also not unlikely that the literary references to Chandra-gupta's wars with the Śaka chief, reviewed in the last section, contain an echo of this great victory.

c12.s2.9

The claims for Chandra-gupta’s almost equally brilliant military campaign in the north-west rest on less solid grounds. A record (No. 67) engraved on the iron pillar which now stands near Kutb Minar at Delhi refers to a king called Chandra, ‘who defeated a confederacy of hostile chiefs in Vaṅga, and having crossed in warfare the seven mouths of the river Sindhu, conquered the Vāhlikas.’ As Vāhlika denotes Bactria, we have to presume that this king, who crossed the Punjab rivers and carried his victorious arms beyond the Hindu Kush mountains, was also in a position to fight, with equal success, against a powerful enemy in Bengal. Even if, as some hold, Vāhlika is located in the Beas valley bordering on Kashmir, the military exploits of king Chandra must be regarded as remarkable.

c12.s2.10

It is, however, difficult to regard as certain the proposed identification of king Chandra with Chandra-gupta II.¹ It is undoubtedly more probable than the proposed identification of Chandra with Chandra-gupta I or Chandra-varman, but in the absence of any corroborative evidence we cannot definitely credit Chandra-gupta II with these brilliant military exploits simply on the basis of the record on the iron pillar at Delhi. The Kushāṇa type of coins, with the name Chandra, however, indicates his supremacy in the N. W. Frontier Province.

c12.s2.11

¹ For the different views on the identification of Chandra and the location of Vāhlika cf. *JRASBL*. IX, 179. In addition to the references contained therein cf. *EI*. XIV, 367 ; *JAHRS*. X, 86 ; *JIH*. XVI, 13.

c12.s2.12

No political event of the reign of Chandra-gupta II, except his conquest of the Śaka territory, is known with certainty.² But some of his matrimonial alliances might have some political significance. He married Kuberanāgā, a daughter of the Nāga family, and the issue of this marriage, his daughter Prabhāvatī, was married to the Vākāṭaka king Rudra-sena II. The Nāgas, as noted above, were a powerful ruling clan and a marriage alliance with them might have been of great use to Chandra-gupta in consolidating the newly established imperial position of the Guptas. As regards the Vākāṭaka king, it has been rightly pointed out by V. A. Smith¹ that the geographical position of his kingdom was such that "he could be of much service or disservice to the northern invader of the dominions of the Śaka satraps of Gujarat and Surāṣṭra." The assistance which Chandra-gupta II possibly derived from the Vākāṭakas and his influence over the kingdom have already been discussed above (Ch. V). It is not, therefore, an unreasonable assumption that these matrimonial alliances were deliberately made with a political motive. If we remember how marriage of Chandra-gupta I with a Lichchhavi princess had enabled the Guptas to rise to a position of supremacy, and Samudra-gupta regarded the offer of a daughter as a normal feature of his feudatories' relationship with himself, we may well believe that 'the matrimonial alliance played no insignificant part in the foreign policy of the Guptas'.² An inscription of the Kadamba ruler Kākustha-varman of Kuntala (Kanarese country in the Bombay Presidency) tells us that his daughters were married to the Gupta and other kings. It has been suggested that Chandra-gupta II arranged a marriage between his son and the daughter of Kākustha-varman, the most powerful ruler of the family.³ Although we cannot be sure whether the alliance was arranged by Chandra-gupta II or his successor, it also proves the traditional policy of the Guptas to form matrimonial alliances with the most powerful and distinguished royal families in different parts of India.

c12.s2.13

² Mr. J. Ratnakar describes (*IHQ*. III, 719) a stone horse found at a village named Nagawa in the south-east corner of Benares. The short record on it is read by him as Chandramgu (whom he identifies with Chandra-gupta II). But the published fascimile does not support the reading, and we cannot, therefore, credit Chandra-gupta II with the performance of an Aśvamedha sacrifice.

c12.s2.14

¹ JRAS. 1914, p. 324.

c12.s2.15

² PHAI.⁴ p. 466.

c12.s2.16

³ JBORS XII, 462. It may be noted in this connection that certain mediaeval chiefs of the Kanarese country claimed descent from Chandra-gupta, and according to some literary traditions Vikramāditya sent the poet Kālidāsa as an ambassador to a Kuntala king. The value of these, and their connection with Chandra-gupta II are, however, uncertain (cf. PHAI.⁴ p. 475, fn. 2) cf. Chs. V, XII.

c12.s2.17

Chandra-gupta II introduced a currency in silver and copper. His gold coins, like those of his father, reflect the pomp, power and grandeur of the empire, and to some extent also his striking personality. Some of his coin-types resemble those of his father, but the difference is significant. Thus in one type the king is represented as slaying a lion (Pl. III, 2) instead of a tiger, as on his father's coin. The legends on these coins refer to him as *Simha-vikrama* (having the prowess of a lion) and *‘Narendra-chandra’* or best of kings, unconquered in the world. The substitution of lion for tiger probably represents his conquest of Gujarat where lions are available. The Couch-type of coins resemble the lyre-type of his father, but instead of a *viṇā* or lyre he holds a flower in uplifted right hand and the legend *‘rūpākriti’* probably emphasises his intellectual and physical eminence, or his artistic sense. In a new type of coin introduced by Chandra-gupta II the king is represented as standing with left hand on sword-hilt, while a dwarf-attendant holds a parasol over his head. Here the umbrella is no doubt the insignia of universal sovereignty. In another new type the king rides on a fully caparisoned horse and holds a bow or a sword. This type, which was used extensively, and the lion-slayer type, with its numerous varieties, probably reflect the personal habits of the king, but the couch-type shows that, like his father, his martial spirits were not incompatible with an artistic and intellectual temperament. It is perhaps not without significance that in his coins, the figure of a throned goddess, derived from foreign coinage, was finally replaced by the purely Indian type of a goddess.

c12.s2.18

As noted above, Chandra-gupta II assumed the title Vikramāditya, which, along with Vikrama and Vikramāṅka, occurs in his coin legends. It is held by many scholars that he is the original of the legendary king Vikramāditya who is said to have defeated the Śakas, ruled at Ujjayini, and founded an era (Vikrama Saṃvat) in 58 B.C. Chandra-gupta II defeated the Śaka Satraps and his association with Ujjayinī is rendered probable by his long stay in Malwa in connection with his fight against them. Without entering into the debatable question whether there was a king Vikramāditya in 58 B.C., it may be regarded as probable that one cycle of the legends concerning him refers to Chandra-gupta II, and in this category we may include the presence of the famous poet Kālidāsa in his court. It should be remembered, however, that the title Vikramāditya was also assumed, probably by Samudra-gupta, and certainly by at least two successors of Chandra-gupta II viz., Skanda-gupta ard Pūru-gupta (or Budha-gupta). It is, therefore, equally likely that the cycle of legends, referred to above, reflects the Gupta age as a whole rather than the reign of an individual Gupta king.

c12.s2.19

Reference may be made to some feudatories of Chandra-gupta II. One of them is Mahārāja Trikamala, known from an inscription, dated year 64, engraved on the image of a Bodhisatva at Gaya.¹ Another feudatory was Svāmidāsa, ruler of Valkha, probably situated somewhere in Central India, who issued a land-grant in the year 67.² Mahārāja Śrī Viśvāmitra-Svāmī, whose name occurs on a seal found at Besnagar, was also probably a feudatory chief of Chandra-gupta II.³

c12.s2.20

¹ ASI. 1922-3, p. 169.

c12.s2.21

² EI. XV, 289. But this is doubtful, cf. ABORI. XXV, 159.

c12.s2.22

³ ASI. 1914-15, p. 81.

c12.s2.23

The last known date of Chandra-gupta II is 93 (=412-3 A.D.), and he could not have ruled much longer, as his son was on the throne in 96 (=415-16 A.D.) He thus enjoyed a fairly long reign of more than thirty-six years. His reign saw the consolidation of the Gupta empire, and if we may credit the stories about Vikramāditya as applicable to him, it also witnessed an outburst of intellectual activity which has made the Gupta age, the classical age or the golden age of India.

c12.s2.24

An idea of the peace and prosperity prevailing in the vast empire may be had from the account of the Chinese pilgrim Fa-hien who travelled through Chandra-gupta's wide dominions for more than six years. Unfortunately Fa-hien does not give any account of the political condition of India,—he does not even mention the name of the great Gupta emperor. Still Fa-hien's book is of great value in forming an estimate of the culture and civilisation in the Gupta age to which reference will be made later. The art and literature of the period will also be dealt with in separate chapters. It is only necessary to emphasise here that the conquests of Samudra-gupta and Chandra-gupta II brought about that imperial peace and prosperity to which we mainly owe the flourishing state of art, literature and the other aspects of civilisation which distinguished the age.

3. Kumāra-gupta I

c12.s3.1

On the death of Chandra-gupta II, his son Kumāra-gupta, born of queen Dhruvadevī, ascended the throne about 414 A.D., and enjoyed a long reign of more than forty years. No less than thirteen records of his reign (Nos. 12-24) have come to light, the largest number that we possess of any Gupta ruler. Although they do not throw much light on the events of his reign, they convey in a general way that he maintained intact the vast empire that he had inherited from his father. This conclusion is also supported by the finds of numerous coins of his in Western India, as far as Ahmadabad and Bhaunagar. The coins further prove that Kumāra-gupta, like his grandfather, performed an Aśvamedha sacrifice. A large hoard of Kumāra-gupta's coins, found at Satara in Bombay, has been taken by some as a possible indication of Gupta influence in the South Western Deccan,¹ though obviously we cannot draw any definite conclusion from this or the find of 13 coins of his at Ellichpur.

c12.s3.2

¹ PHAI. pp. 475 (fn. 2), 480.

c12.s3.3

Kumāra-gupta issued several new types of gold coins, one of which depicts Kārtikeya riding on his peacock on the reverse, and the king feeding a peacock on the obverse. He extended the silver coinage and introduced it for the first time in the central provinces of the empire where the peacock was substituted for Garuḍa on the reverse of the coins. It appears from the legends on his coins that he assumed the title Mahendrāditya, and he is referred to as Śrī-Mahendra, Mahendrasimha, Aśvamedha-Mahendra etc.

c12.s3.4

The inscriptions have preserved the names of some of his feudatories and governors. Ghaṭotkacha-gupta, a member of the royal family, and probably a son or brother of the empreror, was ruling in E. Malwa with jurisdiction over Tumbavana about 50 miles to the north-west of Eran (Ins. No. 17). He was probably the governor of Airikiṇa (Eran) known to be a Gupta province (pradeśa) in the time of Samudra-gupta (Ins. No. 4). Further west, Bandhuvarman, probably also a feudatory chief, was ruling at Daśapura, Mandasor in Western Malwa (Ins. No. 52). The emperor's younger brother Govindagupta was also probably a governor in this region.¹ Another governor, Chirātadatta, was ruling over Puṇḍravardhana or N. Bengal (Ins. Nos. 20-21). These records may be taken to indicate progress in the development of administrative machinery during the reign of Kumāra-gupta I, but as the absence of similar documents for the earlier period may be purely accidental, we may not regard it as a special characteristic of his reign.

c12.s3.5

¹ The cases of Bandhu-varman and Govinda-gupta have been discussed in detail under the reign of Skanda-gupta.

c12.s3.6

Whether the Aśvamedha sacrifice of Kumāra-gupta implies any new conquests on his part we cannot say. But towards the close of his reign the kingdom was certainly disturbed by wars, the exact nature of which it is difficult to determine. The passage in the Bhitari Stone Pillar Inscription (No. 30), which is our sole authority for this notable event, is unfortunately not free from difficulties. According to the reading generally accepted, the adversary of the Gupta emperor was the king of the Pushyamitras, but according to a proposed emendation of the text,² the hostile chiefs are simply referred to as enemies. If we adopt the reading ‘Pushyamitras’, it is not easy to locate the tribe. They are referred to in the *Vishnu Purāṇa* and probably lived in the valley of the Narmadā or near the source of that river. But this is by no means certain.¹

c12.s3.7

² Fleet read the crucial expression as "Pushyamitrāṁś=cha", but noted that the second syllable of the name is damaged (CII. III, 54, 55,

c12.s3.8

¹ IA. XVIII, 228. For other views cf. CDG. xlv-xlvi.

c12.s3.9

But whoever might have been the adversary (or adversaries), he was very powerful and his progress must have constituted a grave menace to the empire.² The inscription expressly states that the enemies ‘had great resources in men and money’, and in course of his fight with them ‘to restore the fallen fortunes of his family’, Skanda-gupta passed a whole night on bare earth. In spite of possible poetic fancies and exaggerations, the statement leaves the impression that the Gupta emperor had met with serious reverses and was threatened with utter ruin, when Kumāra-gupta's son Skanda-gupta turned the scale in his favour by inflicting a crushing defeat upon the enemy. The poet tells us that this heroic achievement of Skanda-gupta was sung in every region ‘by happy men, even down to the children’. The sense of relief echoed in these laudatory songs may be regarded as a proper measure of the apprehended calamity. It is significant that in four successive verses, the poet refers no less than three times to the ‘ruined fortunes of the Gupta family’, and their restoration by Skanda gupta This emphasises the serious nature of the crisis that was averted by Skanda-gupta, but its exact nature still remains unknown.

c12.s3.10

² Dr. H. C. Raychaudhuri infers from the title ‘Vyāghra-parākrama’ that Kumāra-gupta probably invaded the tiger-infested forest territory beyond the Narmadā, and the imperial troops met with disaster, which was retrieved by Skanda-gupta (PHAI.4 p. 480). But neither the assumption of the title nor the find of coins in the Satara district would justify such an inference.

4. Skanda-gupta

c12.s4.1

The aged emperor Kumāra-gupta I died in the year 136 (=A.D. 455-6) before Skanda-gupta returned from his victorious campaign. Indications are not wanting that there were some

c12.s4.2

[^fn. 2]: Mr. H. R. Divekar proposes to read the whole thing as ‘Yudhy-amitrāṁś-cha’ (ABORI. I, 99 ff).

c12.s4.3

troubles over the succession, and the brother and other sons of Kumāra-gupta probably set up rival claims for the throne.¹ But the evidence is so vague and uncertain that no definite conclusions are possible.

c12.s4.4

¹ Arguments in support of this hypothesis are given in *JASB.* NS., XVII, 253 ff. 'They have been criticised in detail in *PHAI.*⁴ pp. 481 ff. The criticism misses the real points of many of the arguments. Thus regarding the omission of the name of Skanda-gupta's mother in the Bhitari pillar Ins., it is merely observed that 'the names of the mothers of kings are sometimes omitted' and 'there was no rule prohibiting the mention of ordinary queens in inscriptions'. These remarks show a lack of appreciation of the main argument, viz., * that the omission of the *Mahādevī* of Kumāra-gupta I, the mother of the reigning king, in striking contrast to the mention of the other *Mahādevīs* of earlier kings in the same record, cannot but be looked upon as significant". It is true that almost all the facts, on which the hypothesis is based, may be explained away in a different manner. If they were not, then the view would not be a mere hypothesis but a definite fact. But nothing has been said to indicate that the proposed view is not a probable and a reasonable inference from the facts before us.

c12.s4.5

According to the royal seals Pūru-gupta was the son of Kumāra-gupta I and his chief queen (*Mahādevī*) Anantadevī. Curiously enough the royal seals ignore Skanda-gupta, and while the genealogical portion, even in the inscription of Skanda-gupta (No. 30), refers to the chief queens, who were mothers of his three predecessors, it is silent about his own mother. This undoubtedly raises a suspicion that his mother did not occupy the status of a *Mahādevī*, though we cannot be quite sure on this point.² It has been suggested that his mother's name was Devakī, but this view rests merely on an analogy which the poet had drawn between his visit to his widowed mother after his victory and that of Kṛiṣhṇa to Devakī.³ This analogy might have been due to similarity of circumstances⁴ rather than similarity of names, and here, again, no definite conclusion is possible.

c12.s4.6

² The analogy of the Banskhera and Madhuban plates does not, as has been suggested (*PHAI.*⁴ p. 483), take away the force of the argument. For here Rājya-vardhana's mother is mentioned and as Harsha-vardhana is said to be his *anuja*, the separate mention of his mother is rendered unnecessary.

c12.s4.7

³ Sewell. *Hist. Ins. of South India*, p. 349; *PHAI.*⁴ p. 480.

c12.s4.8

⁴ *JASB.* NS., XVII, 254.

c12.s4.9

The same uncertainty prevails regarding the early events of Skanda-gupta's reign. Apart from the war of succession, if there were any, there are references in the Junagadh inscription (No. 26) to a struggle with hostile kings including those against the Mlechchhas. This presumably refers to a series of engagements at the very beginning of his reign, though unfortunately no details are preserved. The war with the Mlechchhas probably refers to his fight with the Hūṇas which is specifically referred to in the Bhitari Pillar Inscription. But whether the Mlechchhas are the same as Hūṇas, or were a different tribe, both the records claim that Skanda-gupta completely defeated these enemies.¹ The verse describing the conflict with the Hūṇas, though mutilated, leaves no doubt that it was a severe one. The Hūṇas who appear now for the first time in Indian history were destined to play an important rôle which will be discussed later. For the present, it will suffice to state that they lived in Central Asia on the western border of China as far back as the second century B.C. In course of their migrations to the west one branch (or race) of them, known as the Ephthalites or White Huns, occupied the Oxus valley and conquered Gandhāra. They destroyed this kingdom and set up a king who was cruel and vindictive and practised the most barbarous atrocities. According to the Chinese pilgrim Sung-yun, this took place two generations before his time (520 A.D.).² It is evident, therefore, that not long after his accession to the throne Skanda-gupta found his empire menaced by the onrush of these barbarians who had crossed the Indus, carrying devastation and destruction all around. Where Skanda-gupta met them we cannot say, but the statement in his inscriptions that he thoroughly defeated them seems to be borne out by the fact that we have no evidence of the Hun depredations east of Gandhāra till the close of the fifth or the beginning of the sixth century A.D.¹

c12.s4.10

¹ Allan finds an echo of Skanda-gupta's victory over the Hūṇas in a story of king Vikramāditya preserved in Somadeva's *Kathāsaritsāgara*. According to it Vikramāditya, son of Mahendrāditya king of Ujjain, having succeeded to the throne on his father's abdication, utterly defeated the Mlechchhas who were overrunning the earth (*CGD.* p. xlix, fn. 1).

c12.s4.11

² Beal—*Records* I, p.c.

c12.s4.12

¹ According to a Buddhist text *Chandragarbha-paripṛcchā*, King Mahendrasena, who was born in the country of Kauśāmbi, had a valiant son. After he had passed the age of twelve Mahendra's kingdom was invaded by three foreign powers in concert—Yavanas, Palhikas and Sakunas who took possession of Gandhāra and countries to the north of the Ganges. The young son of Mahendrasena led his father's army of two hundred thousand men against the enemy whose soldiers numbered three hundred thousand. The prince, however, broke the enemy's army and won the battle. On his return his father crowned him king saying "henceforth rule the kingdom", and himself retired to religious life. For twelve years after this, the new king fought these foreign enemies and ultimately captured and executed the three kings. It has been suggested that this story gives an account of the fight between Skanda-gupta and the Hūṇas (IHI. p. 36). But no great reliance can be placed on the details of such stories.

c12.s4.13

If we remember that the cruel devastations of the Huns had spread from the Danube to the Indus, that their leader Attila, who died in 453 A.D., was ‘able to send equal defiance to the courts of Ravenna and Constantinople’, and that thirty years later they overwhelmed Persia and killed its king, we can well realise the value of the great victory of Skanda-gupta over them. All over the vast empire the people must have heaved a sigh of relief at the great deliverance. This heroic achievement that saved his kingdom from the scourge of a cruel barbaric invasion justified the assumption of the title of Vikramāditya by Skanda-gupta which we find on his coins along with Kramāditya.² The continual stress of wars during the reign is also reflected in the coinage. The gold coins of Skanda-gupta are comparatively few and belong mostly to the archer-type. This as well as a depreciation in the purity of gold was possible due to the financial drain caused by the war. His silver coinage was, however, very extensive and presents a variety of types.

c12.s4.14

² According to Ārya-Mañjuśrī-Mūlakalpa, Skanda-gupta was also called Deva-rāja. This, as well as the title of Vikramāditya, was borne by his grandfather Chandra gupta II.

c12.s4.15

One type of Skanda-gupta's gold coins is of more than usual interest. In it the king is represented standing with a bow in one hand and an arrow in the other; in front of him is the Garuḍa standard; beyond it, on the right side, stands a female figure facing the king and holding a lotus in her left, and an uncertain object, probably a fillet, in her right hand. She was formerly identified as the queen of Skanda-gupta, but Allan regards her as Lakshmī, the goddess of sovereignty. Allan rightly associates this picture with the statement in the Junagadh Inscription that the goddess of sovereignty ‘of her own accord selected him as her husband, having in succession discarded all other princes’. It is possible to interpret this coin as a memorial to the fact that Skanda-gupta did not owe the throne to the right of succession but to his own prowess and valour.

c12.s4.16

Skanda-gupta is said to have appointed governors of different provinces (lit. all provinces), almost immediately after his accession. Special reference may be made to the appointment of Parṇadatta as governor of Surāṣṭra. The restoration of the ancient embankment of the great lake or water-reservoir on the Girnar hill, which had burst in the very first year of Skanda-gupta's reign, was a great achievement that redounds to the credit of the governor Parṇadatta and his son Chakra-pālita, the local Magistrate. They saved the country from a great disaster and the poet, echoing the voice of the grateful peoples, lauded up to the skies the virtues and merits of both the father and the son, in a composition which is expressly stated to be the ‘Book on the repair of Sudarśana Lake’ (*Sudarśanataṭāka-saṁskāra-grantha-rachanā*) (Ins. No. 26).

c12.s4.17

More than usual interest attaches to another governor of Skanda-gupta. An inscription found at Mandasor¹ records some constructions by Dattabhaṭa, Commander-in-chief of the forces of king Prabhākara, in the Mālava Saṁvat 524 (=467-68 A.D.). The inscription mentions emperor Chandra-gupta II and his son Govinda-gupta, and we are told that Dattabhaṭa's father Vāyurakshita was the general (*senādhipa*) of Govinda-gupta. The date of the record places it during or immediately after the reign of Skanda-gupta and presumbly Prabhākara was his governor. But it raises several interesting problems.

c12.s4.18

¹ Bhandarkar's List No. 7. The inscription was noticed in *ASI.* 1922-23, p. 187 but has not yet been edited.

c12.s4.19

First, as to the position of Govinda-gupta. He is also known to us from a clay-seal at Vaiśālī which records the name of ‘Mahādevī Śrī Dhruvasvāminī, wife of the Mahārājādhirāja Śrī Chandra-gupta, mother of the Mahārāja Śrī Govinda-gupta’. It has been suggested that Govinda-gupta was the governor of Vaiśālī during his father's reign.¹ In that case it is not very likely that he was alive in the year 467-68 A.D., far less that he held any important position in that year. How are we then to explain the reference to him as a great ruler in the Mandasor record which does not even mention the name of the emperor Skanda-gupta? Dr. Bhandarkar has pointed out that ‘as Indra is represented as being suspicious of Govinda-gupta's power, the latter seems to have been a supreme ruler’. This would mean that he had rebelled, either against his brother Kumāra-gupta or the latter's son Skanda-gupta, a presumption that lends colour to the theory of internal troubles during the closing years of Kumāra-gupta I or the early part of his son's reign. Viewed in this light, the omission of all references to Skanda-gupta in the Mandasor Inscription of 467-8 A.D., becomes significant. This date is the last-known date of Skanda-gupta, and it is just possible that troubles broke out again immediately after his death,—but all these must remain a pure conjecture for the present.

c12.s4.20

¹ CGD. p. xl.

c12.s4.21

We must also consider in this connection another feudatory line ruling in W. Malwa with its capital Daśapura, modern Mandasor, in which place and its neighbourhood four records (Nos. 49-52) of this family have been discovered. The first two rulers of this family, Jaya-varman and his son Siṃha-varman, seem to be independent rulers in the latter half of the fourth century A.D. Nara-varman, the son of Siṃha-varman, was ruling in 404 A.D., and Viśva-varman, the son of Nara-varman, in 423 A.D. Although these dates fall within the reigns of Chandra-gupta II and Kumāra-gupta I, there is nothing in the records of the two rulers of Mandasor to show that they acknowledged the supremacy of the Guptas. The records sing their glories as if they were independent kings, and make no reference to the Guptas. It is interesting to note also that all the records of this family are dated in the Mālava era and not in the Gupta era.

c12.s4.22

Only the last record of Mandasor (No. 52) connected with this family explicitly refer to Kumāra-gupta as the overlord. It is a long record composed in beautiful verses. After referring to Kumāra-gupta as the ruler of the earth, it mentions king Viśva-varman and his son Bandhu-varman. While this Bandhu-varman was ruling over Daśapura a temple of the Sun-god was built by the guild of silk-cloth weavers in the Mālava year 493 (=436 A.D.). In course of time, under other kings, part of this temple fell into disrepair and so in the year 529 (=472 A.D.) the same guild repaired the temple.¹

c12.s4.23

¹ For the long and protracted controversy over the interpretation of this record, and particularly the date when the temple was repaired and the record set up, cf. the references given under Ins. No. 52.

c12.s4.24

The main object of the inscription was thus to record the repair of the temple in 472 A.D., and it must have been composed in or shortly after that date. Under ordinary rules of construction, Kumāra-gupta should be understood to have been the overlord at the time the record was set up *i.e.*, in 472 A.D.² but most of the scholars have taken the reference to Kumāra-gupta in connection with the original construction of the temple. In other words, they hold that in 436 A.D., when the temple was built, Bandhu-varman was the governor of Daśapura, and Kumāra-gupta was his overlord. According to this view, the Gupta suzerainty was established over Mandasor in or before 436 A.D. In that case it becomes significant that the record does not name either the Gupta overlord or the local governor of Mandasor in 472 A.D. *i.e.*, at the time when it was actually set up. On the other hand, it vaguely refers to other kings (the plural number denoting at least three) ruling between 436 and 472 A.D. Whether these refer to the local rulers or Gupta overlords we cannot say, but it gives the impression of some trouble or confusion prevailing in the region between A.D. 436 and 472. The importance of this will appear in the discussion of the history of the Guptas after Skanda-gupta. But whether Bandhu-varman was really a feudatory of Kumāra-gupta or not, it may be reasonably held that Western Malwa had probably already been a feudatory state under Skanda-gupta, as it undoubtedly was in 472.

c12.s4.25

² This is the view of Pannalal (*Hindustan Review*, 1928, p. 31) and Mr. Diskalkar (*JBBRAS*. NS., II, 176) who naturally take this Kumāra-gupta to be Kumāra-gupta II.

c12.s4.26

It is necessary to discuss in this connection the claims put forward by the Vākāṭaka king Narendra-sena that his ‘commands were obeyed by the lords of Kosalā, Mekalā and Mālava’.¹ It is undoubtedly tempting to connect Narendra-sena’s invasion with the early struggles of Skanda-gupta’s reign or the troubled state in Malwa between 436 and 472 A.D. as disclosed by the Mandasor inscription, discussed above, but it is difficult to come to any definite conclusion until the date of Narendra-sena is more definitely known.

c12.s4.27

¹ EI. IX, 271. For the history of Narendra-sena cf. ante. Ch. V.

c12.s4.28

On the whole, so far as the available evidence goes, we may reasonably hold that in spite of the Hūṇa invasion and other troubles, probably at the beginning of his reign, Skanda-gupta maintained till the last his hold over the vast empire that now literally stretched from the Bay of Bengal to the Arabian Sea, and comprised practically the whole of Northern India to the east of the Punjab and Rājputāna. The poet who referred in the year A.D. 460-1 (Ins. No. 27) to the tranquil reign of Skanda-gupta, the lord of hundred kings, was not probably guilty of serious exaggeration. When this great Gupta emperor died about 467 A.D., little did he or anyone else dream that the mighty empire which he left in peace and security would crumble away almost before the eyes of the existing generation.

Cites (8)