365bet娱乐

Some Notes and Observations on the Puranic Account of the Imperial Guptas (guptavamsa) [guptavamsavisayakapauranikavivaranasya] / By Dr. S. N. Roy; Deptt. of Ancient Indian History, Culture and Archaeology, Allahabad University. / 267-285

[ yuktisahitena vivecanenatra lekhe nirdharyate yat gupta rajavamsavarnane purananam pramanikatvangikarane nasti kacit vipratipattih | etatsambandhe vayubrahmandayorekasya slokasya pajitara mahodayena samsodhitah pathoh lekhaka mahodayenopanyastah tasya anyebhyah puranebhyah pathabhedamca pradarsya tesam sambandhe puragapramanyamalocitam | sa ca slokastadyatha- 'anuganga prayaganca saketam magadhamstatha | etan jadapadan sarvan bhoksyante guptavamsajah || ' prasange'smin majumadara gangali - pajitara- vilsana- elana- dasarathasarmaprabhrtividusam matanyapyatra vidhivatsamalocitani | tadanantaram da0 daksarathasarmano vicaramanusrtya 'anugangamaprayagam gupta bhoksyanti medinim ' iti patha evam suddha iti vidusa lekhakena nirdharitam | evam prakarena ekaikapuranasya vartamanapathasya suddhimaulikata va tasya vividhahastalekhesu samupalabdhapathantaranam samvadena, puranantaresu suraksitapathasamanjasyena ca nirdharayitum susaketyabhimatam lekhakamahodayasya | manyate casau yadyapi puranagataitihasika prasanganam varnanam paravattikale pathasamsodhanavasad aitihasika mahattvadhayakam, sraddhopapadakamasamdigdham va prayena naiva tisthati, tathapi tadrrapathanam yathavat maulikatvopapadanartham pathanirdharanakrame tesamaitihasika mahattvam laksyikrtya krtam punarnirmanam ( samsodhanam ) sarvatha samicinam sambhavam bahupakarakam va bhavitumarhati | ityetatsarvam saragarbhitam vivecanamatra prastutam | ] Eversince Mr. Pargiter presented his monumental finds regarding the dynasties of the Kali Age culled out from the passages of the early Puranas, fashion has been in vogue among the scholars to examine the state of historical affairs as can be gleaned from them. Among such passages as have been subjected to much frequent comments mention can specially be made of the one concerning the account of the territories ruled over by the Gupta Kings. The wording and formation displayed by it, as would be subsequently shown, seem to be so very peculiar in their own way and it is so much inconsonantly preserved in

266 puranam - PURANA [Vol. XII, No. 2 different Puranic versions that the conclusion worthy of proper history can hardly be arrived at on the basis of too apparent an analysis without testing its merit in a proper perspective. The revisionary role of the Pauranikas, on the other hand, led to the distortion of the early passages to such an extent that the meaning intended originally in the passage under question as well as many others of its kind, is difficult to gather unless the persistent trend operative in their formation is not taken into full account. In view of these considerations it is proposed here to find out as to how far the original form of the passage has survived in the concerned texts and see as to what extent it can prove to be competent for the reconstruction of the Gupta history, if its available form is emendated, restored and interpreted in consonance with the corroborative evidences. Before examining in detail the points at issue, it seems worthwhile to reproduce the varied readings of the passage as found in the different texts and manuscripts. The reading of the Vayu and Brahmanda Puranas has been emendated by Mr. Pargiter¹ as follows: 'Anuganga Prayagam ca Saketam Magadamstatha / Etanjanapadan sarvan Bhoksyante Guptavamsajah' // Some of the significant variants noticed by the same scholar in the manuscripts available to him are Anugangam for Anuganga (MSS. C and J), and Madhyaga (MS. B) and Makhaga (MS. E) for Magadha. We shall have an occasion to discuss below that these variant forms cannot be ignored, while drawing conclusion out of the aforementioned verse. The copies of Visnu Purana have preserved in them the condensed form of the passage. After piecing together the readings of the various versions of this text Mr. Pargiter has restored the passage in Visnu Purana as under: 'Anuganga Prayagam Magadhah Guptasca Bhoksyanti'. The essential variants of the passage which have been brought out by Pargiter in respect of Visnu Purana's manuscripts are the following ones; use of Magadhan (MSS. H and L) and Magadha Suhma (MS. J) for Magadha; addition of Magadhan (MS. B) and omission of Guptasca (MS. A). It is noteworthy that 1. D.K.A. (second edition), p. 53. 2. Ibid, p. 54, Fn. 10.

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 267 these variants, too, are as significant as those found in the copies of the Vayu and Brahmanda Puranas. The corresponding text of the Bhagavata reads this passage as 'Anugangama Prayagam Guptam Bhoksyati Medinim'. The only variant reading which seems to be significant in case of this Purana is Gupta in place of Guptam³. While testing the merit of the present passage scholars have been, in some cases, skeptic about its authenticity for the history of the Gupta rulers. In this connection Dr. R. C. Majumdar draws our attention to the 'descrepancies' of tho Purana texts, which render it a difficult task to reconstruct from the Purana passage the extent of territory ruled over by the Guptas. He also observes the vagueness contained in the word Anuganga and finally concludes that 'it is hardly justified to assign too great importance to the Puranic passage, and far less to rely upon it in deducting the extent of territory ruled over by Chandragupta I". Almost a similar remark has been made by Dr. D. C. Ganguly who, however, has examined the problem in his article more elaborately than Dr. Majumdar. The article is concluded with the following words 'It will thus follow that even if the above statements of Puranas are taken to have contained genuine historical facts, they cannot in any way be assumed to have referred to the political condition of the country on the eve of the rise of Samudragupta. As a matter of fact no importance should be attached to them until they are corroborated by authentic evidence'.5 The point whether the Puranic account is genuine or not had already been considered by Wilson. While translating this passage as it occurs in the Visnu Purana, he did not feel any hesitation in suggesting that 'the account is the most explicit and probably most accurate of all'.6 The view-points of Allan', Pargiter aad Dr. Dashrath Sharma are, in their essential aspects, akin to that of Wilson. It has been 3. Variant cited by Dr. Dasharath Sharma, I.H.Q., Vol. 30, 1954; p. 375. 4. Vakataka-Gupta Age (1967), p. 135. 5. D.C. Ganguly, I.H.Q. Vol. 21, June, 1945. 6. Visnu Purana's translation, (1961), p. 385, Fn. 70. 7. Allan, Catalogue of the Coins of the Gupta dynasty, p. XIX. 8. D.K.A. (second edition); p. XII. 9. In Journal of Ganga Nath Jha Research Institute, Vol. 7, 1949 and in Indian Historical Quarterly, Vol. 30, 1954. 9

268 puranam - PURANA [Vol. XII, No. 2 pointed out that the Puranic passage appears as a good evidence for the political condition of India in C. 330 A.D. Of these scholars Dr. Sharma has analysed the passage in historical set-up twice. He is inclined to believe that 'the discrepancy in the Puranic statements is more apparent than real'. Another noteworthy remark which he has made in his illuminating analysis of the said passage is that 'we have to take the texts as they stand and give them their liberal meaning, laying aside, no doubt all preconceived notions and partiality for interpretations that have so far held the field'. The point which deserves prime consideration in this context is to find out how far and in which particular respects the relevance of the passage especially in regard to the preservation of the original matter can be justified in its extant form. It would be seen that the passage as such is a little dislocated from grammatical regularity and syntactical consistency. This is mainly due to the employment of the term Anuganga the accuracy of which, in respect of meaning can hardly be ascertained unless it is supposed to be a clerical error for Anugangam. But the possibilities of Anugangam having been the original form and Anuganga its latet substitute is held in doubt for two apparent reasons. In the first place the term Anuganga occurs almost in all Purana-texts and their manuscripts containing the passage, whereas Anugangam is found only in one manuscript. Secondly, the term Anugangam naturally serves as an epithet of Prayagam. This will seriously affect the consonance of the passage with general mode of the Puranic description of Prayaga as of the confluence of Ganga and Yamuna, wherever the location of Prayaga is intended. Scholars, who have so far examined this passage generally take the term Anuganga in the sense of 'along the Ganges'. Moreover, how the term should be related to Their (Etan) janapadan, has not been explained in clear terms. treatment, however, tends to indicate that in the passage the term Anuganga stands conspicuously separated from the rest of the territories enumerated subsequent to it. The term as far its precise connotation is concerned, is enigmatic to say the least. The natural question which arises here is whether the passage should be analysed in accordance with the literal meanings of the words contained in it or with a view to bringing out the total sense as a whole. It would be seen that the verse-form of the

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 269 passage has in it the second line as under: 'Etan Janapadan Sarvan Bhoksyante Guptavamsajah'. Now if we proceed to analyse the verse in its literal sense, the present line would appear meaningless because in the preceding one only one Janapada viz. Magadha/ Magadha is enumerated. Other place-names detailed in the list are Prayaga and Saketa, which enjoyed the rank of capital and premier cities of the Janapadas. Independently these are never enumerated as separate Janapadas in the available lists of the texts dealing with this topic. It is, therefore, quite evident that instead of taking into account each word in the literal sense emphasis is to be laid on the proper interpretation of the passage as a whole in case we are to gather historical information from it. Problem of right interpretation of the passage obviously leads to the necessity of its emendation and restoration in relation to the possible original form and its commodious adjustment to other evidences sheding light on the history of the Gupta rulers. But the fact of matters is that even the emendated and restored form of the passage, as has been done by Pargiter, does not help us very much in removing the riddle with which it is apparently wrapt in the text. The external evidences, on the other hand, can be utilized in such contexts only when the genuineness of the passages is not disproved and its originality remains unchallenged. One significant fact to which adequate attention has not been paid by the scholars so far is that the style which has been employed by the Purana-compiler in the present passage does not accord so much with the dynastic portion as with the Janapada list occurring in the Bhuvanakosa section of these texts. Except for the employment of the term 'Bhoksyante' which ensures the accommodable consistency of the passage in the setting of the dynastic portion, the style followed by the compiler in its essential aspects is more akin to the one enumerating place-names in the section of Bhuvanakosa. Within a reasonable limit Pargiter rightly pointed out that the original dynastic list of the Puranas concerns itself only upto the period of the Andhras. 10 His analyses, however, do See Sircar, D.C., Studies in the Geography of Ancient and Medieval India, p. 17 ff, and Ali, S.M., Geography of the Puranas, p. 175. 9a. 10. Pargiter, ibid, p. XII; His supposition, however, that these matters were borrowed from the original Bhavisya Purana at two stages is not accepted; see, Hazra, Puranic Records on Hindu Rites and Customs, p. 27.

270 puranam - PURANA [Vol. XII, No. 2 not sufficiently make it clear as to whether uniformity of style in It is the early and subsequent descriptions is maintained or not. quite likely that after the completion of the original account of the dynastic list running upto the period of the Andhras an attempt was made at its enlargement at that stage when the Bhuvanakosaportion was already complete and to an appreciable degree it had also left its popular appeal among the Purana-readers. The testimony to this possibility presents itself, when it is observed that the line 'Etanjanapadan Sarvan Bhoksyante Guptavamsajah' describing the places under the royal possession is quite similar in its expressive grab to the line 'Etanjanapadan Aryan Ganga Bhavayate Subha' which is aimed at giving the description of the places in the Gangetic basin." The fact, which is of special note here is that the words 'Etanjanapadan Sarvan' are repeated in all the passages subsequent to the above one in the dynastic list upto the end of the section. In as much as the wording in a similar set-up is not found in the dynastic section prior to these passages, similarity of which is otherwise suitably established with those found in the Bhuvanakosa-portion the unity of authorship with regard to the former can evidently be held in doubt. Answer to the question as to when and in which particular circumstances these passages were compiled in the Purana-texts necessitates re-assessment of one of the basic problems of the Puranic studies. That problem is which of the two portions Vamsanucarita and Bhuvanakosa has the stamp of earliness on it. It would be seen that in the wellknown Puranic definition, by way of alternative occurrence these are both enumerated as the fifth characteristic of a Purana. But while Vamsanucarita is mentioned in a great majority of the Purana-texts laying down the said definition, Bhuvanakosa occurs in the single passage of only one Purana-text.12 Outside the Puranas we find that Vamsanucarita figures prominently in the Puranic definition given by the lexicon Amarasimha, while Bhuvanakosa as a part of Purana-subjects occurs in the Kadambari of 11. Vayu P., XLVII. 49; Matsya P., CXXI.51. Brahmanda P., II.18.52, however replaces 'Aryan' by 'Manyan'. For the rendering of the Matsya passage see, Agrawala, V.S. Matsya Purana-A Study (All India Kashiraj Trust, Varanasi, 1963), p. 206. 12. Matsya P., II.22, see Pusalker, Studies in the Epics and Puranas, p. XLV (Intro.)

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 271 Bana Bhatta. 13 This shows that among the Purana characteristics Bhuvanakosa was not only less popular than Vamsanucarita, but it seems to have been coined also as a substitute of the latter comparatively at a later stage of Purana-Compilation. In a general way it can be said that the replacement of Vamsanucarita by Bhuvanakosa was effected after the age of Amarasimha , who is known to have flourished in the Gupta period. There is, however, one serious objection that can apparently be raised against the present supposition. The datable limit of the dynastic account can not be extended beyond the early Gupta period, whereas our analysis tends to place the passages under purview in the postGupta period. The weight of this objection is to a great extent reduced by the fact that these passages do not seem to be from the pen of the same compiler who recorded the major bulk of the dynastic account. The style which is followed in them is in conformity with that of the Bhuvanakosa-portion; and if the latter is to be dated in the post-Gupta period, the former can also be placed within the same date-limit. One noteworthy aspect of these passages is that while repeating the words 'Etanjanapadan Sarvan' in a set style, they also refer to such terms in them, whose exact significance has rarely been emphasized by the scholars so far. These terms are as under: Manidhanya/Manidhara, Devaraksita, Mahendra and Guha. These appear to be the names of royal personages ruling over the Janapadas specified in each of the Vayu, Brahmanda , Visnu and Bhagavata Puranas; which, however, have also variant readings in some cases. Commenting on these Dr. Dashratha Sharma is of the opinion that herein we have the 'correct Gupta line of succession'. He is further of the opinion that we also get in these texts 'a good idea of Gupta imperial expansion'. As regards the particular territories mentioned in these texts, the point has been cautiously examined by him. In as much as the present discussion is intended at the exposition of a different issue, it seems needless to reproduce Dr. Sharma's analysis regarding the applicability of these territories in the context of the Gupta history. Nevertheless the other aspect of his conclusion that Devaraksita, 13. Puranamiva Yathavibhagasthapitasakalabhuvanakosam, Kadambari, Purvabhaga. 14. J.O.G.J.R.I.; Vol. 7, 1949, p. 61 ff.

272 puranam -- PURANA [Vol. XII, No. 2 Mahendra and Guha can stand respectively for Chandragupta II, Kumaragupta and Skandagupta seems eminently instructive especially in regard to the ascertainment of the approximate period of the Purana-passage. The only possible conclusion which can emerge from these analyses is that these passages were incorporated in the Purana-texts much later than the commencement of the reign of the Gupta dynasty, in all probability after the imperial line had ceased to rule, 'when', as it would be clear from the subsequent Purana passages, 'India temporarily lay prostrate before the barbaric invaders'. The above analysis can very well lead us to assume that the whole of the dynastic portion was not compiled in one plan and at one stage, and that the addition of the passages relating to Gupta history coincided with the compilation of the Bhuvanakosa section. For the simple reason that in both these sections placenames had to be specified and emphasis had to be laid on the demarcation of areas, common expressions pointing to the identity of style were characteristically used in them. If the present problem is to be visulized from this consideration, we find that the Bhuvanakosa portion supplies pertinent parallel not only of the expressions 'Etanjanapadan Sarvan' but also of the term Anuganga meaning of which as shown above, is said to be vague. The source of the term Anuganga seems to be Anuga used in a passage of Bhuvanakosa, whose originality can hardly be doubted, as it occurs in all the three Puranas Vayu, Brahmanda and Matsya. The interesting part of the said passage is that while in the scheme of the seven great rivers of the Himahva country the river is called 'Anuga', the names applied to it elsewhere are Bhagirathi and Ganga. In this connection the Puranas also explain the etymological significance of Anuga by narrating that the river is so known because it began to flow in the footsteps of king Bhagiratha to whose ascetic observances its descent on the earth was due. Scrutiny of the two terms from the above angle and the consideration that their construction is based on the same prefix (Anu) and the same root (Gam) would probably establish an indubitable unity in them as far as their meaning is concerned. The replace- 15. Vayu P., XLVII.39, Matsya P., CXI. 41, Brahmanda P., II. 18.41, Brahmanda P. has, however, changed Anuga into Anvagat. This seems to be a later arrangement. 15

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 273 ment of Anuga in the dynastic section points to the readjustment of the term in agreement with the more popular name of the river and also to its provision associated with the metrical consistency in the verse form of the passage. The last expression viz. Sapta Vai' of the previous verse, which in fact precedes Anuganga if the two verses are to be combined and read together, tends to the possibility of a similar conclusion. In two Purana-manuscripts 'Sapta Vai' is transferred into 'Sapta Vi' and 'Saptatami¹s with the result that Sapta Vi/Saptatami Ganga of the dynastic passage assumes the appearance of close similarity with Saptami Anuga of the Bhuvanakosa passage. This is a good evidence of the fact that the source of the dynastic passage, as far as its extant reading is concerned, should be traced in the Bhuvanakosa passage and that the meaning originally intended in it cannot be sought out without its evaluation in a relative set-up. The natural query that may conveniently be put here is how and motivated by which particular factor the compiler selected the term Anuganga, while bringing into relief the territories under the Gupta rulers. As a matter of fact the places mentioned in the verse could well have been specified even without any reference to the river Ganga, because these are too well known to be expressed by their association with the river. It might be, therefore, interesting to find out whether this is a case of deliberate use without which the intended idea could not have been fully expressed in the Attention may here be drawn to the conspicuous description of the territories of Gangetic basin, under the name of Arya Janapada in the geographical section of the Vayu, Brahmanda and Matsya Puranas." Reference to Arya Janapada reminds us of the territories of Aryavarta, the conquest of which was accomplished by the imperial Guptas under Samudragupta before he extended his arms beyond this geographical division as recorded on the well known pillar at Allahabad. The genetic expression Arya Janapada applied to the territories of Gangetic basin is, however, not repeated in the dynastic section. But considering the identity of style found in both these sections and the fact that the word Janapada in the dynastic section is qualified (though not in strict grammatical 16. These variants are listed by Pargiter, Ibid, p. 53, fn. 4. 17. Vayu P., XLVII. 49, Brahmanda P., II. 18.52, Matsya P. CXXI. 51, The text of Brahmanda reads manyan for aryan.

274 puranam - PURANA [Vol. XII, No. 2 sense) by Anuganga, the possibility can not be altogether ruled out. Reference to 'Anuganga......Janapadan', whose equation with 'Aryan Janapadan' is not beyond the limit of logical consideration, shows that in the Puranic verse there is perhaps an echo of the The two placeAryavarta-conquest by the Gupta sovereign names Prayaga and Magadha intervening between Anuganga and Janapadan may point to the assertion that only a considerable part and not the whole of Aryavarta is intended in the Verse. Reference to Saketa is not made in the prose form of the passage occurring in the Visnu Purana and Bhagavata. It was probably incorporated in the Vayu and Brahmanda Puranas out of metrical exegencies. Samudragupta is known to have subdued the rulers of the entire Aryavarta, the conclusion of the scholars that the Puranic description in the present context points to the reign of Chandragupta I is too reasonable to leave any room for doubt. Such assumption can hold good only in respect of the verse Anuganga Prayagam etc, which, however, does not mark the end of the dynastic description in the Puranas. As remarked earlier, to the present verse are appended such verses as reveal the distinct stamp of lateness in their form and meaning. We have noted above the remark of Dr. Dasharatha Sharma relating to the equation of Devaraksita, Mahendra and Guha of the subsequent verses with Chandragupta II, Kumaragupta and Skandagupta of the Gupta dynasty. Unfortunately the suggestive element contained in the conclusion of Dr. Sharma has not been recognized so far, and the general trend of the scholars has been to assess the authenticity and applicability of the Purana-text regarding the Gupta history as far as it has been reconstructed by Pargiter. Thus a note of objection has been put forth against the view point of Dr. Sharma in a recent work on the history of the imperial Guptas But the ground cited in this context is miserably inadequate and misleading for the disproval of the proposed indentification. It is essentially and entirely based upon the single and solitary argument that the 'Puranic statements regarding these rulers are too confused to warrant any such conclusion'.18 The fact, however, remains that a patient and careful analysis with a correct methodical approach may yield data of genuine historical value from what appears as muddled Puranic accounts. The valuable information 18, S.R. Goyal, A History of the Imperial Guptas, pp. 50-51.

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 275 which can be gathered from these accounts is apt to be missed in case we are inclined to test their validity with the yardstick of accuracy and precision setting aside other criteria to ensure their genuineness for the study of history. As a matter of fact, the Puranic materials could have hardly been enlisted among the sources of Political history of ancient India in the absence of Pargiter's Purana-text of the Dynasties of Kali Age, the essential structure of which is prepared by way of collation, restoration and emendation of the materials scattered in various Puranas and their manuscripts. Despite the profundity of scholarship displayed in his unique and monumental work, it can never be denied that he could not grasp the real historical import of the passages in question. It is true that in presenting these passages Pargiter has shown the same degree of caution and care as in case of other passages in the section. But the notes of Pargiter on their restored forms as also the meaning which he is inclined to attach these seems too bypothetical to be regarded as sound. In the method which he has employed with reference to the interpretation of these passages the terms Manidhanya/Manidhara, Devaraksita, Mahendra and Guha appear to be the names of ruling families along with the family of the imperial Guptas. 19 In view of the fact that such ruling families are not known from any other source, the approach of Pargiter to these passages can hardly be treated as a case of right interpretation. Dr. Dasharatha Sharma's analysis 20 makes it convincingly clear that the Puranic verses of these contexts allude to the Gupta rulers by such names. This is specially true of Devaraksita and Mahendra, the former being the same as Devagupta of the epigraphic records while the latter can essily be treated as the simplified form of Asvamedha-Mahendra legend alluded to in the commemorative coins of Kumargupta. Employment of the term Guha in place of Skanda should not surprise us in view of the fact that the compiler of Puranic verse could not separate himself from the convention of the age in which the use of synonyms even for expressing proper names was not inadmissible, if the same was required for metrical adjustments. The attempt at the identification of the Puranic terms brought out in these lines would perhaps remain incomplete in 19. Pargiter, Ibid, pp. 73-74. 20. J.O.G.J.R.I., Vol. 7, 1949, p. 61 ff. 10

276 puranam - PURANA [Vol. XII, No. 2 case the mystery centering round Manidhanya/Manidhara is not removed and the term is not explained even to the degree of a satisfactory reservation. The point as to whether its equation can be sought forth with any one of the known Gupta rulers has not been, however, touched upon by Dr. Dasharatha Sharma, the task can, nevertheless, be attempted in consonance with the methodology which has been followed by him with regard to the other terms of the group. Three vital factors have to be considered before the historical interpretation of this term is made. In the first place the name occurs in the verse previous to the one that contains the name of Devaraksita. Secondly, the style followed by the compiler of these passages points to the fact that the kings are referred to in these verses either by such names as were less commonly known or by the synonyms of their well-known names. Thirdly, of the two terms; Manidhanya and Manidhara; only one seems to have been referred to in the original Purana-account the variant being a distinct case of scriptorial oversight. The only Gupta ruler who responds to these three factors is Samudragupta, the immediate predecessor of Chandragupta II, called Devaraksita in the Puranic account. The question now left to be answered is the consideration as to what extent and in which particular respect the name of Samudragupta could be expressed by the term Manidhanya/Manidhara in accordance with the Puranic description . Its equation can by no means be established with the known epithets of Samudragupta, which are so very distinct and so widely circulated in the Gupta records. So, in case the term is at all intended to refer to Samudragupta it can only be considered as a synonym of the latter. Before carrying on the present point of discussion to its possible end, it may be remarked that so far as the synonymic consideration is concerned the word Manidhanya/Manidhara can not stand related to Samudragupta as directly as can be found in case of Guha to the name of the Gupta monarch analysed in the above lines. The fact, however, remains that before making a reference to Devaraksita in the succeeding verse, the compiler must have been inclined to select such word which could evidently express the name of the immediate predecessor of Chandragupta II. Instead of concerning himself with the general and more popular meaning of the word Samudra, he probably took it in view of the

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 277 literal signification. His emphasis seems to have been on the root term Mudra, and consequently he interpreted the word Samudra in the sense of Mudraya Sahitah i.e. one who bears a Mudra. In a more convenient sense it may be stated that the compiler was in the know of the fact that the Rajamudra of Samudragupta was honourably applied in the subjugated territories, 21 and thus the word Samudra was taken by him in the sense of Mudradhara which was capable of imparting the above idea in a more commodious form. Elsewhere the Puranic passage alludes to God Siva as 'Mudramanidharaya Ca'22 i.e. one who is bearer of Mudra and Mani. The dificiation of the kings being a well-established fact by the emergence of the Gupta period, the selecting of the word Manidhara in the Purana appears to be an attempt for bringing out divine prestige in the royal personage. 28 23 The present problem can also be tackled from another angle. It is quite probable that while choosing the term Manidhara for the name of Samudragupta, the compiler had in his mind the name of Kaca, who not only falls in the line of the monarchs of the Gupta dynasty but has also behind him such historical traces worthy of reliance, as can prove that he was a contemporary and rival of Samudragupta 24. Doubt is, sometimes, entertained regarding the separate existence of a monarch bearing the name Kaca, and it is pointed out that Samudragupta himself was probably called by that name at the early stage of his political career. 2 Although the bearing of the Puranic reference on the present point is by no means certain, yet the high probability of the following two broad facts can well be taken into account. The word Mani carries the meaning exactly opposite to that of Kaca, and consequently the two terms are found juxtaposed even in one and the same verse 26 for the sake of imparting adversative impression. In 21. See line 23 of Allahabad Pillar Inscription, C.I.I., III, pp 8 and 14. 22. Namah Priyaya Varadaya Mudramanidharaya Ca, Vayu P. XXIV. 23. 246. It has been suggested that the tendency of attaching divinity to kings can be traced even earlier; S. C. Bhattacharya, Jl., A.U.S. 1963-64, p. 68 ff. 24. Heras, A.B.O.R.I., IX, p. 83 ff. 25. Allan, Op. Cit, p. XXXII. 26. E.g. kacam manim kancan amekasutre grathnasi bale kimidam vicitram/vicaravan Paninirekasutre svanam uvanam Maghavanamahiti.

278 puranam - PURANA [Vol. XII, No. 2 case the compiler of the Purana-passage was guided by this convention, it would evidently follow that the term Manidhara/Manidhanya was employed by him in view of Kaca being an adversary and contemporary of Samudragupta. But the possibility of the other alternative seems to be more justified and suitably relevant in the present context, and as such the use of Manidhara/Manidhanya would be a case of Puranic attempt aimed at giving the stamp of an augustus form to the monarch's name originally being uncouth and awkward in case the term applied to it, was Kaca. Out of the two readings Manidhara and Manidhanya preference seems to be deserved by the former one in view of its approximity to Samudra and in view of the fact that the words in the Puranic passages are, more than often, dislodged from their original and intended form due to the ignorance of the later copysts. Another noteworthy aspect with which the historical interpretation of the Purana-passage has to be aimed at, is to find out the intended syntanctical relationship of the two words Gupta and Magadha in these texts especially in the Visnu Purana, reading of which is so often cited by the scholars on the present issue. It would be observed that these two words constitute the relevant part of the sentence in the Visnu Purana-text, which may be reproduced as under: Magadha Guptasca Bhoksyanti. Despite the sentence being as explicitly simple as can be expected, its syntactical significance has been differently approached by the scholars and consequently two varied meanings have emerged out of it. Wilson 27 in his pioneer work on the Visnu Purana interpreted the word Magadha as an adjective in relation to Gupta, The analysis of the passage, as has been done by Dr. Dasharath Sharma does not deviate from the translation of Wilson 28. The ground which he has put forth in respect of his approach is essentially based on the grammatical order in which the two words Magadha and Gupta are related to each other in the sentence. Dr. Majumdar 29 has taken them in the sence of two independent The verse is quoted by Tikakara Sridharananda Sastry in Laghusiddhanta Kaumudi, on Panini VI, 4, 133; see also K. C. Chatterji, Patanjalis Mahabhasya (Calcutta, 1953) p. 126. 27. Wilson, Ibid, p. 385. 28. I.H.Q., vol. 30, 1954, pp. 374 ff. 29. Vakataka-Gupta Age, p. 134.

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 279 words each being used in the sense of noun. Majumdar's approach has received agreeable response from those who are inclined to find in the Purana-sentence an echo of what is treated, in view of numismatic indication, as the joint-rule of the two powers, the Licchavis and the Guptas 30. The exact meaning of the Purana-sentence can, however, be ascertained if the analysis is based on the following considerations. In the first place, the two words are variedly used in the versions of the other Puranas, and moreover the sentence constituted by them assumes a different form, if it is emendated in view of textual variation even of the Visnu-Purana itself. In other Purana-texts the word Magadha serves the purpose of an object, whereas Gupta (Vamsajah) is subject in its relation. The historical interpretation of the passage would hardly be justified without taking into account their bearing on the point being considered in this context. As regards the textual variation of the VisnuPurana, as noted earlier, one of its manuscripts, has the word Magadhan in addition; and thus the sentence would be restorable as under: Magadha Guptasca Magadhan Bhoksyanti. One should not feel any hesitation in agreeing with the view of Dr. Dasharath Sharma that 'on account of the extreme closeness of the words Magadha and Magadhan in the sentence the later word was inadvertantly dropped by some one' in all probability for avoiding Punaruktidosa of the Sanskrit poetic convention. It would be seen that the above sentence of Visnu Purana does not convey any such meaning as can be different from the other Purana-texts. Secondly, we can not completely ignore the applicability of grammatical rule in the Purana-sentence, to which its meaning is largely due. Thus the particle Ca which is found in 'Magadha Guptasca Bhoksyanti' necessarily implies that the relative position of the words Magadha and Gupta can be respectively established only in the sense of qualifier and qualified words in the sentence. Had it not been so, as remarked by Dr. Dasharath Sharma, 'the particle Ca must probably have been repeated twice, this being the general rule in prose though not in poetry'. It is thus evident that the word Magadha can be interpreted only in the sense of an adjective, while the omission of the word Magadhan simply shows 30. S. R. Goyal, Ibid, pp. 51-52. 31. I.H.Q., Vol., 30, 1954, p. 376.

280 puranam - PURANA [Vol. XII, No. 2 that the region of Magadha was original seat of the Gupta power. As a matter of fact what is noted as Magadha Guptasca in the text of Visnu Purana accords, to a satisfactory extent, with the Puranic style often making distinct reference to the original locale of the dynasties and kings by adjectival terms. By way of its example we may quote here references like Magadhanam Barhadrathanam, Nrpan Vaidesikan, Amdhrah sriparvatiyasca³² in all of which words pointing to the original place of the dynasties and kings have their adjectival forms. This possibility is further brought out in clear terms if we take into account the variants of Magadhan found in the MSS of Vayu Purana. Thus MSS 'b' and 'c' cited by Pargiter omit Magadhan and read out respectively Madhyaga and Makhaga in its place Although these two words appear too insignificant to ascertain any distinct status attributable to the Gupta rules, yet their importance in the context of the present discussion can not be totally ignored. As far the grammatical status, these words in the Vayu Purana's passage seem to be substitute for what is referred to as Magadha in the text of Visnu Purana. In as much as these are essentially used in the sense of adjective in their relation to Gupta (Vamsajah), similar status is evidently assignable to their counterpart Magadha in the text of Visnu Purana 33 The fact, which is still more important and whose significance has been much rarely realized by the scholars so far, is that the account of Visnu Purana at least in its extant form can not be preferred to that of Vayu Purana for the purpose of historical investigations The degree of originality and the element of genuineness contained in the Visnu Purana's account seem to be doubtful in view of its far condensed prose form as compared to the elaborate and versified account preserved in the text of Vayu Purana. The relative chronology of these two texts specially with regard to the dynastic account had already been analysed by Mr. Pargiter, who in close agreement with the view point of Prof. R.G. Bhandarkar pointed out the comparative lateness of Visnu Purana 34 It is not improbable that the short account of Visnu Purana has 32. References enlisted by Pargiter, D.K.A. pp. 14, 46 and 49. 33. Ibid, p. 53, Fn. 7. 34. Pargiter, Ibid, Intro. p. XIV, R.G. Bhandarkar, Early History of Deccan (1895), p. 162

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 281 been designed on its own material much of which has been sliced off the original and early form. The point which has not been taken into a proper account by the scholars, who have deduced two different conclusions from the texts of Visnu and Vayu, is that the oft-quoted line of the former does not present its position as clearly as can be necessary for comparative analysis of the problem. As shown above, if the Visnu Purana's text is interpreted in view of the variant reading, which is found in one of its versions, the possibility of meaning deduced out of it as different from the Vayu Purana's text is totally ruled out The high probability of the reference in the Visnu Purana's line to one people only is brought out in clear terms, when it is further observed that the expression Guptasca has been dropped out in one, 35 of its copies, whose reading, nevertheless resembles that of the other versions in respect of the rest of the words. The sentence, which is substituted in it is as under: Magadha Bhoksyanti for Magadha Guptasca Bhoksyanti of the other versions can only denote in this context that the word Magadha with or without Guptasca was intended to carry on the same meaning in both the forms. The compiler of the Purana-copy could not have probably left out the expression Guptasca in case the region of Magadha was not considered to be the original seat of Guptapower and in case the word Magadha denoted the people as distinct from the Guptas. Despite the fact that uniformity in the account of the Visnu and Vayu Puranas is maintainable in view of the aforementioned consideration, attempt has been made to establish their textual discrepancy in another context. As noted previously a version of the Visnu Purana refers also to the region called Suhma in the list of the territories under the Gupta-rule. Dr. D. C. Ganguly makes a distinct note of this variant reading and considers its significance too seriously in touching upon one fundamental problem of the Gupta history. Dr. Ganguly lays stress on the genuineness of Visnu Purana's account in regard to the present reading, which according to him is corroborated by the evidence proving the region of Bengal as the original homeland of the Gupta rulers. The same scholar makes note of the fact that reference to Suhma in the copy of Visnu Purana should not be 35. MS. 'a' of Pargiter's list, D.K.A. p. 54, Fn. 8,

282 puranam -- PURANADAD [Vol. XII, No. 2 regarded as a later interpolation, though the term is not referred to in other copies of the text. 36 The ground on whose basis this remark has been made is however too very inadequate and unless otherwise the authenticity of the evidence cited in its favour is established beyond the range of all doubts it can not prove itself to be worthy of authenticating other evidences on the point. As a matter of fact unless it is admitted that the term Suhma was inserted in one copy of Visnu Purana at a later stage, it remains inexplicable why does the term find no mention in other copies of Visnu Purana on one hand and in the texts of the Puranas of similar value on the other. It is often forgotten that rare reading found in a particular Purana-text can speak well of the fancy of the subsequent compilers aiming at the adherence to the technique of the Purana-compilation which required amendments in the Purana-recording to the changes of the later times. In view of this persistent trend of the Purana compilation the only possible inference deductible out of this additional reading is that insertion of Suhma was effected in the original tcxt for the sake of fashioning it uptodate after the extension of the original Gupta kingdom as far as Bengal in the east. Previously we have stated that the phraseology of the Purana line is far close to the one found in the Bhuvanakosa-portion particularly to to the passage enumerating the names of the places in the basin of river Ganga. The word Suhma finds prominent mention among the places specified in the said section 37 ; and it is quite likely that with a view to making a reference to it, the Purana-sentence in the dynastic portion was retouched sometime during the late Gupta period when the Puranatexts were being considerably modified consequent upon the incorporation of the geographical passages into them. The actuality of the circumstance contributing to the insertion of the term Suhma in the Visnu Purana-copy may also be visualized from the consideration that while Saketa of the other Purana-texts was deleted without the placement of any substitute of it in the various copies of Visnu Purana, it was substituted by the word Suhma for the sake of bringing about conformity between two sections of the Purana-text. In this attempt the copyist seems to have kept himself concerned 36. I.H.Q. Vol. XXI, 1945, p. 142; History of Bengal, p. 69. 37. Suhmottaramsca Vamgamsca Tamraliptamstathaiva Ca, Brahmanda P. II. 18. 51, The texts of Vayu and Matsya read Brahmottaramsca for Suhmottaramsca. The former seems to be the original reading.

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 283 not so much with the historical cacurrcy of the original description as with maintaining the consonance of the places, as far as possible. spotted out with reference to the Anuga of the Geographical section with those relating to Anuganga to be specified in the dynastic section. Very little justice has been done by certain scholars to the passage of Bhagavata, while taking into account its general bearing on the history of the imperial Guptas. Thus relying on one of the Bhagavata-copies and on the commentary of Sridharasvamin relating to it, Dr. Ganguly 38 has entertained grave doubts regarding the uniformity of the Purana-passage and their authenticity for the history of the Gupta-rulers. The copy of Bhagavata, which he has consulted reads Anugangama Prayagam Guptam Bhoksyati Medinim and the commentary of Sridharaswamin on it runs as under: Gangadvaramarabhya Prayagaparyantam Guptam Palitam Medinim Bhoksyati. In case the present reading is not properly checked; the import of this text would certainly go much against the other texts, which do not have any reference either directly or indirectly to Haridvara as the westernmost boundary of the Gupta kingdom. For want of a properly reconstructed and restored text, the version of Bhagavata has been misjudged by some 39 as a 'decidedly late work'. It is surprising to find out that while making such remark on the text of Bhagavata, no attention has been paid to the significant variant readings of the different copies of Bhagavata, which can otherwise be well utilized for the restoration of its passage in the intended original form. Instead of attaching much importance either to the restored passage of this text quoted in the work of Pargiter or to its construction and interpretation figuring in the commentary of Sridharasvamin, the variant readings should have been collated in view of their competence for making out the order and form of the passage correspondingly similar to that of the other Purana-texts. In this connection Dr. Dasharath Sharma¹0 rightly invites our attention to the passage being misconstructed and misinterpreted by Sridharasvamin, use of which has been made in the paper of Dr. Ganguly without considering its authentic or the otherwise nature 38. I.H.Q., Vol. XXI, 1945, p. 142. 39. S.R. Goyal, Ibid, p. 51, fn. 1. 40. I.H.Q., Vol. XXX, 1954, p. 375. 11 -ad el qujanionso (asonali

284 puranam - PURANA [Vol. XII, No. 2 from a proper angle. The fact that the said passage is open to serious doubts becomes evident not only from the consideration that commentator Sridharasvamin belongs to a fairly late period and as such he was less competent to interprete it in the right channel of history but also from its being thoroughly irrelevant in the context of the general report of the Puranic evidence on the point. In view of the various readings available in the Bhagavatacopies as listed by Pargiter, Dr. Sharma suggests the possible reconstructed form of the passage in the following order 'Anugangama Prayagam Gupta Bhoksyanti Medinim' and this appears to be the passage composed in the original copy of the Bhagavata. The genuineness of the passage of an individual Purana can be established in a correct and coherent form only when its readings scattered in various manuscripts are picked up after taking into due consideration of the extent of their correlationship and identity to the readings which are preserved in the versions of other Purana-texts. There is little doubt about the fact that the information supplied by the Bhagavata-passage reconstructed in th eabove form can be different from the texts of the other Puranas. Noninclusion of the word Magadha in the Bhagavata passage can not be taken for the discrepency of this text in relation to the passage of the other Purana-texts. It is simply an indication of the fact that at the time of the composition of the passage in the Puranatext association of Magadha with the Guptas, as the original seat of their power was too well-known to affect the intended meaning when as a result of the omission of name of the region the passage was made condensed in this text. s The The notes and observations made by us in this humble attempt may probably lead to the following general conclusion: account of the Puranas can, to the extent of original data contained in their passages, be treated as a reliable source of historical information. By their very nature the passages of these texts could not survive, in most cases, in their original and unadulterated forms consequent upon the revisonary fancies and recompilatory trends of the later compilers. The fact, however, remains that without rearranging and readjusting their readings distributed in the various copies any statement worthy of proper historical reconstruction can never be made. Their convergence to one general historical conclusion is sometimes held in doubt because of the

July, 1970] PURANIC ACCOUNT OF THE IMPERIAL GUPTAS 285 contradictory readings available in them, but the same can be easily brought out if these readings are properly checked up in consequence of the restoration of the passages of such Purana-texts as have not been analysed with this end in view. After giving due regard to all these factors it is made evident that the Puranic evidence regarding the Gupta-history is neithhr discrepant in itself nor in any way discomfortable in the context of the other evidences on the topic. a beanwoner ved 1 otel an denial to shine a ads to and ad bloode notre tirwito son 0100