SUPPLEMENT TWO
THE CHRONOLOGICAL AND ADMINISTRATIVE BEARING OF THE KAUTĪLIYA ĀRTHAŚASTRA
Now that the main chronological problems connected with our thesis have been tackled, a determination of the age of the political treatise going by the name of Arthaśāstra and purporting to have been written by Kautilya, alias Chānakya and Vishnugupta, whom tradition regards as the chief minister of Chandragupta Maury a, can be attempted. An estimate related to our dating of Chandragupta I instead of the first Maurya to the time of Megasthenes must be outlined and justified in order to give a finishing touch to this substitution. At the end the polity of Megasthenes's India in its characteristic features has to be compared with the administration under the Mauryas and the Guptas.
Years ago, it used to be a frequent practice to note the concordances between the account by Megasthenes and the information contained in the Arthaśāstra. If this practice could still be approved, there might be some argument to place the first Maurya in the Greek ambassador's period on the basis of that treatise's claim of authorship by this monarch's chief minister. We say "some argument" rather than "convincing proof because the claim would still lack indubitable historical support. Unless the authorship were fully established, the concordances noted would only point to a date for the book in the last quarter of the fourth century B.C. without implying that Chandragupta Maurya was on the Magadhan throne at the time. But today the interpretative situation has strikingly changed.
In 1970 A. L. Basham1 observed about the Arthasatra: "Since its publication certain Indian scholars have been waging a losing battle in favour of its authenticity as a production of Kautilya. In Europe this view is now almost universally abandoned." Among the "Indian scholars", however, we can hardly count the leading lights. "There are grave doubts," says Raychaudhuri,2 "as to whether in its present shape the famous book is as old as the time
1.The Oxford History of India by Vincent A. Smith, 3rd Ed., edited by Percival Spear (1970), "Authorities", p. 115.
2."Note on the Date of Arthaśāstra", The Age of Imperial Unity, p. 286.
Page 546
of the first Maurya..." Mookerji1 also writes: "Many scholars... regard the present text as of a much later date. It is doubtful, therefore, how far we may regard the system of administration depicted in it as applicable to the Maurya period." In the context of the current chronology, what Raychaudhuri and Mookerji mean is both that the book carries signs of an epoch subsequent to Sandrocottus's and that it fails to match sufficiently the account left by Magasthenes whom they make a contemporary of Chandragupta Maurya. Beni Prasad2 comments: "Many scholars now refuse to accept the view that the work was really composed by Kautilya or any statesman of the type, and they regard it as of a much later date. Some bring it down to the third century A.D., though others would prefer a date three or four centuries earlier." Barua3 pronounces: "The prose treatise of the Arthaśāstra, as we now have it, is not only post-Aśokan but post-Śunga in date."
Of course, we must not jump to the extreme of thinking that Megasthenes and the Arthaśāstra disagree toto coelo: there are points of similarity, some of which are fairly sharp but several are of a very general nature and none of them absolutely crucial. The proponents of a late dating of the book urge, as M.A. Mehendale4 marks, that in matters of essential detail its author and Megasthenes entirely differ and that the rules of government laid down by the Arthaśāstra pertain to a small-sized state and not to a large kingdom like that of Sandrocottus or of an empire such as Chandragupta's. The very concordances by which the Sandrocottus-Chandragupta equation seemed strengthened and confirmed are not adequate. Allan5 pertinently points out: "Great ingenuity has been displayed, but with little real success, in finding in the Arthaśāstra passages to prove its Maurya date by comparisons with statements of Megasthenes. Coincidences indeed occur, but many of them are repeated in other ages of Indian history, and the differences are much more striking. Megasthenes finds more corroboration in Manu than in Kautilya."
The precise view taken of the Arthaśāstra at present may best be gathered from Smith and Raychaudhuri combined, with one or
1.Ibid., p. 66.
2.Ibid., pp. 303-04.
3.Aśoka and His Inscriptions, Part II, p. 42.
4."Language and Literature", The Age of Imperial Unity, p. 275.
5.The Cambridge Shorter History of India (1943), p. 40.
Page 547
two details collected from elsewhere. The post-Megasthenes character of the book is deduced from a number of facts. "Reference to Chīnapatta, Chinese silk [Bk. II, Ch. 11], which, be it remembered, occurs frequently in classical Sanskrit literature, points to a late date...1 The reference must be to the great country of the Far East (cf. 'China which produces silk,' Kosmas Indikopleustes, McCrindle's Ancient India, p. 162)"2 and not to "the small state of Ch'in which later gave its name to the whole of China" nor to "early representatives of modern hill tribes (Shinas of the Himalaya or Chins of Burma)".3 But "the name 'China' applied to the famous land can hardly be anterior to the first emperor of the Ch'in Dynasty (249-210 B.C.)".4 A post-Megasthenes date is also suggested by "the reference to parapets of brick instead of wooden ramparts (II.3) in connection with the royal seat"5 - the Pātaliputra described by Megasthenes whose report has been borne out by Dr. Spooner's excavations which disclosed heavy defence with wooden palisades.6 The Arthaśāstra turns out to be post-Maurya by mentioning "the use of Sanskrit as the official language [Bk. II, Ch. 10], a feature not characteristic of the Maurya epoch".7 Further, "the imperial title Chakravarti (IX. 1) is not met with in inscriptions before Khāravela"8 (second half of the 1st century B.C.) and "the official designations samādhartri and sannidhātri find mention in epigraphs of a still later age".9
"A date as late as the Gupta period is, however, precluded by the absence of a reference to the Denarius in the sections dealing with the weights and coins"1" (Bk. II, Chs. 12 and 19). A tairly earlier period for the prevalence of the study of Arthavidyā is "proved by the Junāgadh Rock Inscription of Rudradāman I [currently dated 150 A.D.], and the existence of treatises on Arthaśāstra is rendered probable by the mention of technical terms like 'Pranaya,' 'Vishti,' etc. It is interesting to note that the
1.The Political History..., p. 277.
2.Ibid., p. 10.
3.The Oxford History..., p. 95, fn.
4.The Political History..., p. 10.
5.Ibid.
6.Chandragupta Maurya and His Times, p. 112.
7.The Political History..., p. 277.
8.Ibid., p. 10.
9.Ibid.
10. Ibid., p. 277.
Page 548
Kautilīya, which purports to be a compendium of pre-existing Arthaśāstras does not quote the views of previous Achāryas or teachers in the chapter on 'Pranaya' (Bk. V, Ch. 2). It is, therefore, not unlikely that Rudradaman I, who claims to have studied the Arthavidyā, learnt the use of the term from the Kautilīya itself if not from a pre-Kautilīyan treatise. In this connection it is interesting to note that the Junāgadh epigraphs show a special acquaintance with the Arthaśāstra literature. The Junāgadh Inscription of Skanda-gupta, for instance, refers to the testing of officials by upadhās..., possessed of a mind that (has been tried and) is (found to be) pure by all the tests of honesty.1 The verse 'Who is capable both in the lawful acquisition of wealth, and also in the preservation of it, when acquired, and further in causing the increase of it, when protected, (and able) to dispense it on worthy objects, when it has been increased' (Fleet), reminds us of Kaut., I. 1 - 'The science of government: it is-a means to make acquisitions, to increase what is protected and to distribute among the worthy what has been increased.' "2
On the strength of all the main facts Raychaudhuri3 declares the Arthaśāstra "assignable to the period 249 B.C. to c. 100 A.D."
Our task is to take stock of the information and evaluate it not within the current time-scheme which begins the Imperial Guptas in 320 A.D. but in the light of our transposition of their start to 315 B.C. At once we are led to consider the treatise as preceding the latter date and hence Megasthenes in at least a part of it, while succeeding the Maiiryas and, as Barua insists, even the Sungas. Is there any part that must be counted as post-Gupta? The argument from the imperial title "Chakravarti", which is said to be epi-graphically earliest in Kharavela's Hāthigumphā Inscription, has no longer any force since we catch "Chakravartin" among the Imperial Gupta titles (Fleet's No. 39).4 Evidently samāhartri and sannadhātri are post-Khāravela but pre-100 A.D. If they have no occurrence in Gupta records they must indicate an interpolation which is for us post-Gupta. But we cannot be certain of it in the
1.Cf. Raychaudhuri's account, on pp. 283-4 of the Arthaśāstra's, Amatyas, officers, variously tried and purified. Among the upadhds are mentioned the religious test, the money-test, the love-test and the fear-test.
2.Ibid., p. 9, fn. continued on p. 10.
3.Ibid., p. 9.
4.Mookerji. The Gupta Empire, p. 155.
Page 549
case of the reference to Chīna, for the alternative of a Himālaya tribe is not so absurd as Smith and Raychaudhuri make out. Mookerji1 writes: "V.R.R. Dikshitar proposes to identify Chīna with Shina, a Gilgit tribe known for its manufacture of silk [Mauryan Polity, p. 7]." From Gupta records we know that silk-weaving was practised in India and Chinese silk was not the sole kind used in India: a Mandasor Inscription (No. 18 of Fleet) was made "when Kumāra Gupta was ruling over the earth"2 and it speaks of a guild of silk-weavers renovating a Sun-temple. That silk other than the Chinese should be current in ancient India is absolutely in the fitness of things, for the archaeologist H.D. Sankalia3 reports from Mahārāshtrian Nevasa a necklace of copper beads "strung on cotton and silk threads" - threads of natural silk - datable by Carbon-14 to as early as 1250±110 B.C. Nothing in Gupta records points to the China of the Far East. We cannot assume knowledge of it by either the Arthaśāstra or the Guptas merely because the Arthaśāstra (11.11.13) alludes to Kambu as well, which signifies Cambodia,4 a country of South-east Asia, and Samudragupta's Allāhābād Pillar Inscription includes not only Simhala (Ceylon) but also "other islands" as states under his suzerainty.5
R. C. Majumdar6 tells us: "The inclusion of 'all islands' in addition to Simhala... is worthy of note. Although none is specifically named, it very likely refers, in a general way, to the Hindu colonies in Malay Peninsula, Java, Sumatra and other islands in Indian archipelago." Majumdar7 further says: "One of the oldest Hindu kingdoms in this region was situated in Cambodia, and comprised nearly the whole of it along with Cochin-China. The Chinese call it Fu-nan and have preserved many details of its early history." But the Chinese annals of "states in Malay Peninsula, Java, Cambodia and Annam, with rulers bearing Indian names"
1.Chandragupta Maurya..., p. 339.
2.The Gupta Empire, p. 112.
3.Prehistory and Protohistory of India and Pakistan, pp. 486-87, 565 col. 2.
4.Raychaudhuri, "Note on the Date of Arthaśāstra", The Age of Imperial-Unity, p. 287.
5.Majumdar, "The Foundation of the Gupta Empire", A New History..., pp. 147-48.
6.Ibid., p. 151.
7."Colonial and Cultural Expansion", ibid., p. 310.
Page 550
cover no time earlier than "the second century A.D."' India's contact with these islands precedes that time: "Indian literature, particularly the stories narrated in the Buddhist and Jaina books for purposes of edification, contain frequent references to merchants sailing to the east for purposes of trade. The various islands and other localities mentioned in them cannot be always identified but the stories leave the general impression that the eastern coast of the Bay of Bengal and the islands in the East Indies were regarded in ancient India as the veritable El dorado which constantly allured enterprising traders by promising immense riches to them. This idea is also reflected in the name Suvarnadvīpa or Suvarnabhūmi (Land of Gold) which was used as a general designation for this vast region."2
Pliny,3 after quoting Megasthenes on Ceylon (called "Tapro-bane"), goes on to tell us what "we have learned from the old writers". "The island in former days ... was thought to be twenty days' sail from the country of the Prasii [=Magadha], but the distance came afterwards to be reckoned at seven days' sail, according to the rate of speed of our ships... In making sea-voyages, the Taprobanè mariners make no observations of the stars, and indeed the greater Bear is not visible to them, but they take birds out to sea with them which they let loose from time to time and follow the direction of their flight as they make for land." Pliny's testimony establishes navigation beyond Ceylon in the days of Sandrocottus, king of the Prasii known to Megasthenes. Samudragupta's phrase about "Simhala and other islands" can be kept in countenance within our chronological scheme with the help of Megasthenes no less than ancient Indian literature.
Great antiquity, however, cannot be ascribed to Kambu's interchange with China, nor is there "indisputable evidence of any contact between India and China before the Han period (206 B.C.-A.D. 220)."4 Thus, on the basis of the Arthaśāstra's "Kambu", Dikshitar's proposal to identify Chīna with the Shinas need not be discounted and the Arthaśāstra does not have to be as late as the Nāgarjunikonda inscriptions where, according to
1.Ibid.
2.Ibid., p. 309.
3.The Classical Accounts of India, pp. 347, 348.
4.Raychaudhuri, "Note on the Date of Arthaśāstra", The Age of Imperial Unity, p. 287.
Page 551
Raychaudhuri,1 China recognizably appears for the first time in Indian epigraphy as a foreign country.
What about the Denarius? Raychaudhuri thinks that the Denarius does not make its appearance in epigraphy prior to the Gupta period as currently computed - from 320 A.D. onwards for about two and half centuries more. Barua2 says that the Arthaśāstra "takes no notice of dīnāras that find mention in the Nagarjunlkonda inscriptions". Whatever be the case, we have argued already that the dīnāra of the Guptas has no necessary relation to the Roman denarius aureus. Consequently, this treatise should not be dated with any reference to that foreign coin-name. But, if the Guptas commence in 315 B.C. and the Arthaśāstra ignores the dīnāra, it must antedate them or at least the third Gupta - Chandragupta II - in whose reign we first have the non-Roman dīnāra epigraphically attested.
Where, then, in more precise terms shall we put the treatise? The quotation Raychaudhuri has made from Skandagupta's Junāgadh Inscription affines it to the Gupta mind. There is also the fact emphasized by Mookerji:3 "All the Gupta inscriptions are written in Sanskrit, replacing Prākrit (and Pālī) of the earlier inscriptions." Another Arthaśāstra-touch in the Gupta period is brought out by Raychaudhuri:4 "There is no reference in Kautilīya Arthaśāstra to royal titles characteristic of the Maurya age. On the contrary, Indra-Yama-sthānametat (1.13) cannot fail to recall Dhanada-varun-edrāntaka-sama of the Allāhābād prasasti." Yet the marked differences from Megasthenes as well as the absence of the dīnāra stops us from giving the book a Gupta date. Besides, the Junāgadh epigraph of Rudradāman I, who was before the Guptas and whose time as reckoned by the Śaka Era of 551 B.C. would be (551-72=) 479 B.C., betrays knowledge of the book. Again, "Rudradaman was not only a great conqueror and administrator, but a patron of classical Sanskrit."5 His celebrated epigraph is, of course, in Sanskrit. The Arthaśāstra cannot be far removed from his age.
2.Op. cit., p. 47.
3.The Gupta Empire, p. 141.
4.The Age of Imperial Unity, p. 287.
5.Sircar, "The Śaka Satraps of Western India", The Age of Imperial Unity, p. 185.
Page 552
A few interpolations, however, are certainly there from other ages: e.g., some official designations and "the use of the word surunga (Greek syrinx)" which suggests "contact with the Greeks over a long time."1 A small number of items may belong even to the very time of Sandrocottus and Megasthenes. Likewise a limited number of them may have been traditional matter inherited from a much earlier age than 479 B.C. If Rudradaman I is dated so far back rather than in 150 A.D., Chandragupta Maurya would recede still further in time and in possibly associating anything of the Arthaśāstra with him we should have to move deeper into antiquity by nearly 5 centuries if we deem Smith justified to whatever extent in some remarks of his. After making many "serious reservations" he2 writes: "the Arthaśāstra of Kautilya may still be used as a general guide to Mauryan polity. Though the work is almost certainly post-Mauryan, it is equally certain that it is pre-Guptan, and the system of government envisaged in the text corresponds more closely to what we know of that of the Maury as than to that of later times. The detailed instructions for the organization of the department of state strongly suggest that the author, though he himself may have been a theorist, had at his disposal the work of a practical politician, whether the great minister of Chandragupta or another, who probably lived in Mauryan times."
To elucidate Smith's word "theorist" as well as to recommend a little caution vis-à-vis his partly pro-Mauryan attitude, we may cite some earlier words3 of his own: "the Arthaśāstra,... though of great value, is often of dubious reliability, as far as the Mauryan period is concerned. In its existing form it is certainly several centuries later than Chandragupta. For instance, it mentions peoples who cannot well have been known to the Indians at this time, and though it recognizes the possibility of a large empire it accepts as the unit of government a comparatively small kingdom. Thus it is not, as some earlier authorities believed, an official manual of instruction for the Mauryan emperor and his court. It must also be remembered that the work, like the relevant portions of the Dharmaśāstras, outlines the views of the author on the best means
1.M.A. Mehendale, "Language and Literature", The Age of Imperial Unity, p. 275.
2.The Oxford History..., p. 96.
3.Ibid., pp. 95-96.
Page 553
of governing the state rather than the actual system of government."
These are some of the sanest opinions, all the more valuable because their author has an amount of sympathy for those who seek to put the Arthaśāstra side by side with Megasthenes. The alleged writer of the treatise, Kautilya, is not acceptable as Chandragupta Maurya's minister nor can the book be dated substantially to the post-Alexandrine period and adjudged the Indian counterpart of the Greek ambassador's information. But what renders Smith's general position rather confused is that his verdict of some affinity in the Arthaśāstra to "Mauryan times" on matters of practical politics is based on his assumption that the patron of Megasthenes was Chandragupta Maurya.
Basham's latest pronouncement1 moves along similar lines. He still refuses to encourage the Indian scholars who "maintain the authenticity of the Arthaśāstra as the genuine work of Kautilya" and he affirms: "Statistical analysis of the text has proved with virtual certainty that the Arthaśāstra is a compilation" - "it is a conflation of at least three earlier texts, composed by different hands" - "in its present form it is post-Mauryan". Yet he echoes Smith by holding that "the Arthaśāstra agrees with what we know of the Mauryan state-system better than with that of any other Indian dynasty", and he believes the compiler of the treatise to have "made use of a document which was composed early in the Mauryan period, or possibly just before it, suggesting the guidelines on which that state should be run". The reason he gives for this belief is, first of all, the Smithian that in these guidelines the Arthaśāstra is comparable in general with "the account of Megasthenes"; but it is, in the second place, that for these guidelines we can draw also upon "the Aśokan inscriptions". Unfortunately, he does not tell us whether the book agrees on the same points with both Megasthenes and Aśoka or whether the agreement is on different points. The second kind of agreement would derive strength only from the conventional identification of Chandragupta Maurya with Sandrocottus; the first would seem to support, in whatever limited measure, the identification. We say "seem" because in fact the identification is already assumed when Basham declared that the state-system of no other Indian dynasty agrees so
1. Foreword to Kautilya and the Arthaśāstra by Somnath Dhar (Marwah Publications, New Delhi, 1981), pp. ix-x.
Page 554
well as that of the Maury as with the Arthaśāstra. For, if the founder of the Guptas could be Sandrocottus, the broad agreement of the book's state-system with the account of Megasthenes and with the inscriptions of Aśoka would show that the Mauryan state-system is not exclusively indicated by the Arthaśāstra; for then Megasthenes would bring in the state-system of the Gupta dynasty for comparison.
As for Aśoka and the Guptas, it will not do to drive a wedge everywhere between them as if in several matters the latter stood on one side and the former on the other along with Kautilya and Megasthenes. Although, as we shall soon see, a very significant wedge can divide the last two from Aśoka, no ubiquitous division sets the Guptas apart from the Mauryas. U. N. Ghoshal1 has some pertinent remarks: "The Imperial Guptas continued the traditional machinery of bureaucratic administration with nomenclature mostly borrowed or adapted from earlier times. The mantri (High Minister), whose office is known to Kautilya's Arthaśāstra [V.3] evidently stood at the head of the civil administration." Aśoka's chief Mahāmatra at once comes to mind. Ghoshal2 also notes: "A link between the central and the provincial administration of the Imperial Guptas is furnished by the class of officers called Kumāra-mātyas and āyuktas. The amātyas stand for the general body of officials in the Arthaśāstra and the Jātakas, while the āyuktas (or āyuktakas) may be traced back to the yutas of the Aśokan inscriptions and the yuktas of the Arthaśāstra... The provinces called bhuktis were usually governed by officers called uparikas and sometimes by princes of the royal blood with the title Mahārāja-putra devabhattāraka. The uparikas represented the pradeiśikas of Aśoka's Empire and the amātyas at the provincial headquarters of the Sātavāhana administration, while the Mahārājaputra devabhattāraka had his prototype in the Kumāra viceroys of Aśoka's times." We may comment that an analogue to the Aśokan Kumāra viceroys may be discerned in the Gupta-created "order (or rank) called kumārāmātya to which belonged not only high imperial officers, but also officers on the personal staff of the Emperor, the Crown Prince and others as well as those in charge of districts".3
1."Political Theory and Administrative Organisation", The Classical Age, p. 349.
2.Ibid., p. 350.
3.Ibid.
Page 555
Here what we have termed "a very significant wedge" becomes relevant to our discussion. An important note by Raychaudhuri may be brought forward.1 It is in relation to the Arthaśāstra's officials Samādhartri and Sannidhātri, names epigraphically attested only in the post-Khāravela epoch. Raychaudhuri2 explains these designations as, respectively, "Chancellor of the Exchequer and Minister of the Interior" and "High Treasurer and Keeper of Stores". Then, discussing Revenue and Expenditure in Mauryan times, he3 writes: "The distinction between taxes Lévied in rural and in fortified areas respectively is known to the Arthaśāstra which refers to certain high revenue functionaries styled the Samādhartri and the Sannidhātri. No such officials are, however, mentioned in Maurya inscriptions. Greek writers, on the other hand, refer to 'treasurers of the state' or 'superintendents of the treasury'." What is remarkable in this note is that the Arthaśāstra, in the very act of evincing one of its few affinities with the age of Megasthenes and Sandrocottus, stands in contrast to the age of the Maury as, as though the two ages could never be the same.
This paradoxical result at once provokes the query: "How about the age of the Guptas? Do the officers listed by the Greek writers have an echo in it?"
We have no relevant records of Chandragupta I himself to scrutinize. The gold coins ascribed to him can give no information about any department of administration. Neither can the Meherauli Iron Pillar epigraph, which can be attributed to him instead of his namesake grandson, yield what we want: it recounts only his manifold conquests. We have to come to a later time. But this is nothing peculiar. Chandragupta Maurya has left us no inscriptions. His son Bindusāra too is devoid of them. Aśoka alone supplies us with pertinent material. His grandson Daśaratha is of little help. Gupta history has more members of the line with such material though none has so abundant a record to offer as Aśoka with his numerous edicts. We have to take advantage of those members and from one or another of them seek an answer to our query.
The best answer comes from around the year 188 of the Gupta
1.We have already commented on it in other words and in a more limited context in Supplement Four to Parts One and Two.
2.The Political History..., p. 283.
3.Ibid., p. 294.
Page 556
Era - c. 127 B.C. by our chronology and c. 508 by the current one. Vainyagupta, about whom "there is hardly any doubt that he belonged to the imperial family",1 shows himself associated in his Gunaigarh Inscription2 with East Bengal and a Dūtaka, a high officer, named Vijayasena. After the king's death, Vijayasena passes to one Gopachandra who is evidently in the Gupta tradition and calls himself not only by the general Gupta title Mahārājādhirāja but also by that particular Guptaism Bhattaraka.3 Vijayasena, in his Mallasaral Copper-Plate Inscription under Gopachandra, lists a number of officers among whom we find (1) Audrangika, Collector of the Udrahga tax, (2) Tāddāyuktaka, Treasury Officer, (3) HIrānyasāmudāyika, Currency Officer.4 In some other Gupta inscriptions we have a functionary in charge of land revenue, Dhruvādhikaranika - a Treasurer, Bhāndāgārādhikrita - a Collector of Taxes, Utkhatayita.5 Putting together all the designations we have an overwhelming correspondence to a special feature in the age of Megasthenes. We look in vain for anything equally unmistakable in the Edicts of Aśoka.
Raychaudhuri's negative verdict about them is borne out by the guess-work to which Mookerji1 is reduced when expounding the Aśokan administration from the names provided by the Edicts: "The Yutas, also mentioned in Kautilya's Arthaśāstra as Yuktas, along with their assistants, the Upayuktas, were probably treasury officers, whose main function was to manage the king's property, receive and spend the revenue and keep accounts." Mark that sign of vagueness covering a non-particularity, the adverb "probably".
Not that treasurers and revenue-collectors must be absent under Aśoka. The point at issue is that, in contrast to the account of Megasthenes and to the Gupta epigraphs, they are not specifically disclosed, not given distinct name and form.
Glancing back at all the facts surveyed, we may conclude: Sandrocottus's identification with the first Gupta is not at all put in jeopardy by the Arthaśāstra, which at least in one item draws a blank for the first Maurya, and the major part of the treatise can
1.A New History..., p. 190.
2.Ibid., p. 210.
3.The Gupta Empire, p. 130.
4.Ibid., pp. 130-31, 158.
5.Ibid., p. 158.
6."Aśoka the Great", The Age of Imperial Unity, p. 80.
Page 557
best be fitted into an epoch closely antecedent to the Junāgadh Inscription of Rudradāman I in 449 B.C.
POSTSCRIPT: S. R. GOYAL ON KAUTILYA
1
An impressive book with several features relevant to this Supplement has recently come out: Kautilya and Megasthenes by S. R. Goyal.1 We do not need to tackle any item in it touching on the Greek writer - for instance, the argument that he unmistakably reported the Indians of his time to have no "written letters" - nor is it necessary for us to discuss the argument that the Brāhmī script was invented only around Aśoka's day. Intriguing though these viewpoints are, the call upon us is confined to the book's contentions about Kautilya and his time.
Its most fundamental contention for our purpose is that the very idea of Chānakya's having a second name "Vishnugupta Kautilya" which goes with the treatise Arthaśāstra is invalid. If this highly unorthodox thesis is tenable, it renders useless all attempt to compare the contents of the Arthaśāstra to those of Megasthenes's Indica with the aim of proving Chandragupta Maurya, through his association with the alleged author Chānakya, a contemporary of the Greek writer.
Following in the footsteps of H. Jacobi who first opposed the two-name theory and of E. J. Johnstone and T. Burrow who built up the opposition in some detail, Goyal2 cogently lays out its credentials. As he has extended his case beyond the page or two where the central thesis is sketched, we shall use a few square brackets to integrate briefly the additional matter into the initial statement. We are told:
"...almost in all the early versions of the story of Chānakya, only this name (not Kautilya or Vishnugupta) occurs. The earliest reference in Sanskrit to the legend of Chānakya is found in the Mrchchhakatika of Śūdraka (probably fourth century A.D.) where this name appears in the Prakrit form Chanakka (1.39 and VIII. 34 and 35).... The Nandīsūtā of the Jainas [not later than the
1.Published by Kusumanjali Prakashan, Meerut, 1985.
2.Op. cit., pp. 22-23.
Page 558
5th century1] mentions Chanakka among a list of persons famous for their intellect; here the reference is no doubt to the political skill displayed by Chānakya in uprooting the Nandas. Elsewhere the same text refers to the Kautilīya Arthaśāstra without giving any hint of any connection between Chānakya and Kautilya. The [later secular] Jaina literature [the germs of whose traditions are traced to the early Nijjuttis of the first century A.D., traditions of which the fossils are found embedded in the Painnas whose lower limit might be fixed at about 100 B.C.2]... also refers to Chanakka or Chānakya, the minister of Chandragupta, without suggesting that he was also known as Kautilya. In the Kāshmirian Sanskrit versions (of Somadeva and Kshemendra) of the Brhatkathā of Gunādhya the story of Chānakya is found: it is quite likely, therefore, that it was included in the original Brhatkatha. The significant fact, however, is that in the Sanskrit versions of this work also only the name Chānakya appears, not Kautilya or Vishnugupta.
"In the Buddhist literature, the story of Chānakya is found briefly in the Mahāvamsa and in detail in its Tikā. According to Burrow,3 no trace of the name Kautilya is found in the Pali sources in connection with this story though it is found mentioned in separate contexts in some later Buddhist works."
Here an explanation is due in favour of Goyal. He does not give the epoch of the Mahāvamsa and it looks as if he has deliberately omitted the date because it is generally held to be the "sixth century A.D."4 - that is, a time which can hardly be considered early in the field of discussion involved. What we should not forget is a situation somewhat similar to that of the Brihatkathās of Somadeva and Kshemendra. The Sanskrit translations project beliefs not of the two translators' periods but of the earlier period of Gunadhya who wrote most probably in Paiśāchi Prākrit and about whose date R. C. Majumdar5 sums up in the prevailing chronological framework that it "is not later than A.D. 500, though some place it much earlier, - even in the first century A.D.",
1.Ibid., p. 7.
2.Ibid., pp. 25-26.
3.Annals of Bhandarkar Oriental Research Institute, Poona, XLVIII-XLIX, Golden Jubilee Vol., pp. 17-31.
4.R. C. Majumdar, "Ceylon", The Age of Imperial Unity, p. 234.
5."Literature", The Classical Age, p. 314.
Page 559
as does the Canadian Sanskritist A. K. Warder.1 The Mahāvamsa of the 6th century A.D. is a more extensive treatment, than the 4th-century Dīpavamsa, of a still earlier "common source now lost, viz. the Atthakathā-Mahāvamsa of the Mahāvihāra monastery, composed in old Sinhalese prose mingled with verse in the Pali language".2 Basing himself mostly on a work antedating even the 4th-century, Mahānama, the author of the Mahāvamsa, presents a picture substantially earlier than does the Nandīsūtra which knows of the Kautilīya Arthaśāstra.
Goyal3 is led to aver: "Thus we find that all the early versions of the story of Chānakya use only this name, and never Kautilya or Vishnugupta. Secondly, ...in all these versions nowhere is there any mention of Chānakya having been the author of a work on the science of government. These facts should be quite sufficient to make it clear that Chānakya, the Chancellor of Chandragupta Maurya, and Kautilya, the author of the Arthaśāstra, were originally two different persons..." Goyal4 tells us also that "the Kautilīya Arthaśāstra is not mentioned by any ancient work of the pre-Gupta age". In accord with modern historians this age means for him the time before 320 A.D., the presumed accession-date of the first of the Imperial Guptas. So, for him, the Arthaśāstra comes broadly in proximity to the Gupta age - a little more near it than any pre-Gupta "ancient work". He5 dates it "around 300 A.D." or "towards the close of the third century A.D." Hence, with the start of the Mauryas put by him where modern historians do, Chānakya and Kautilya in his eyes "were separated from each other by more than five hundred years".6
As an aside, we may wonder whether the usual description of Chānakya as the "minister" or "Chancellor" of Chandragupta Maurya has partly prompted a number of modern scholars to accept the tradition that he penned a political treatise for the guidance of his sovereign. Very few researchers realise that there is hardly any ground to ascribe this post to him. I myself became
1."Classical Literature", A Cultural History of India, edited by A. L. Basham (Clarendon Press, Oxford 1975), p. 176.
2.Majumdar, "Ceylon", op. cit., p. 234.
3.Op. cit., p. 23.
4.Ibid., p. 7.
5.Ibid., pp. 8, 21.
6.Ibid., p. 23.
Page 560
aware very lately of the misconception. It was my friend the veteran Indologist K. C. Varma who, though inclined to the popular Indian view about the Arthaśāstra, set me right with his letter of June 27,1986: "What is the proof that Chānakya/Kautilya was the Chancellor of the Mauryan empire? Nothing but a mistake of Jacobi, who compared him to Bismarck, has led to this error. The drama Mudrārākshasa vividly portrays the manner in which Chānakya contrived to persuade Rākshasa, Prime Minister of the Nandas, to become the P. M. of the Mauryas."
It would seem that mediaeval India identified Chānakya with Kautilya because he was in legend "the archetype of political cleverness" and Kautilya was "the greatest authority on the science of polity".1 Another cause may be the misunderstanding of "Kautilya" as a nickname signifying "crookedness" and thus being apt to that master-schemer Chānakya, whereas actually it must be the gotra designation of Vishnugupta. In any case, we gather from Goyal2 that the earliest work to "give Chānakya as another name of Vishnugupta or Kautilya" and at the same time to mention the Kautilīya Arthaśāstra was "the Daśakumāracharita of Dandin (6th century A.D.)". But from some remarks of Goyal's we may surmise that a still earlier assumption of Chānakya's having authored the Kautilīya Arthaśāstra had occurred.
After saying that we do not definitely know when the identification took place, he3 opines: "it must have taken place very shortly after the composition of the Arthaśāstra, for the author of the Mudrārākshasa is aware of the supposed identity of the two. Similarly, the Purānas refer to the destroyer of the Nandas by the name of Kautilya and the Pahchatantra (composed in the Gupta age) mentions Chānakya as an author on the science of polity." However, Goyal4 also notes: "...most of the Purānas (composed in the present form in the Gupta or post-Gupta period) and the Mudrārākshasa of Viśākhadatta (probably 4th century A.D.) refers to Kautilya as the destroyer of the Nandas without mentioning him as the author of the Arthaśāstra." Evidently this non-mention leads Goyal to be sure only of Dandin for the full statement of the
1.Ibid., p. 23.
2.Ibid., pp. 4, 7.
3.Ibid., pp. 23-24.
4.Ibid., p. 4.
Page 561
Chānakya-Kautilya equation.1 But reservation on his part about Viśākhadatta and the Purānas is really unnecessary. The Mudrārākshasa and the Purānas' dynastic sections omit to mention the literary composition because there was no dramatic call in the former nor any historical demand in the latter to bring it in when reporting a violent coup engineered by a double-named personage. To ascribe to Chānakya no other second name than Vishnugupta Kautilya argues knowledge of this particular name's bearer having been the author of the Arthaśāstra. B. P. Kangle2 even points out that the Mudrārākshasa makes use of the various strate-gems, diplomatic moves, etc. recommended in the Arthaśāstra and in some places refers to its teachings. He asks us to note that a certain passage in Act IV is the same as Kautilya's 5.4.1 and also how another in this Act compares with Kautilya's 9.1.42-43.
If Goyal's date - 400 A.D. - is preferable to Burrow's which he3 quotes when informing us that "according to Burrow Viśākhadatta (whom he places in the 6th century) was possibly the person responsible for this identification", then we may well hold that soon after c. 400 A.D. the Arthaśāstra was tampered with: "Once this identification became current, it was not unnatural for some scribe to add a verse at the end of the final Chapter of the Arthaśāstra stating that its author was responsible for the destruction of the Nandas. In the words of Keith,"4 It is the only passage which refers clearly to the defeat of the Nandas and there is no reason to believe that it belongs to the original work. There is already a metrical conclusion.' "5
Goyal may be declared successful in breaking Chānakya and Kautilya apart. His success suits our ends excellently, but his acceptance of the current chronological framework is quite dissonant with them. We have to inquire whether this framework can be set aside while utilizing his happy findings in their essence.
1.More correctly we should say: "the Chānakya-Vishnugupta equation". For, though the Kautilīya Arthaśāstra is meant by Dandin, Goyal is mistaken in using the name "Kautilya": this name does not figure in Dandin's account. See p. 60 of the book cited in the next fn.
2.The Kautilīya Arthaśāstra: A Study (the University of Bombay, 1965), Part III, p. 61 and fn. 7.
3.Op. cit., p. 24.
4.B. C. Law Volume, I, p. 494.
5.Goyal, op. cit., p. 24.
Page 562
2
A small initial hurdle is in connection with Dandin. After marking him as the first full and explicit instance of the orthodox view, Goyal1 adds that the Daśakumāracharita "significantly enough" refers to the Arthaśāstra "as a 'recent' work". As somewhat later he2 uses Dandin's mention of Kautilya "as a 'recent' author" to fault "the tradition contained in the last but one verse of the Arthaśāstra"', he does not appear to realize that he cannot lightly ignore Dandin's commitment to the very tradition of that verse. The assumed allusion to recentness is a casual one, whereas the reference to the Arthaśāstra\ author Vishnugupta being the mentor of Chandragupta Maurya and the exterminator of the Nandas is a clear-cut basic setting for the treatise in Dandin's dealings with it.
Besides, what should we take to be "recent"? Goyal has dated Dandin to the sixth century A.D.: can a book which he dates to the close of the third century - that is, about 300 years earlier - be ranked as a recent work? Can anybody in the twentieth century speak legitimately of Dryden and Johnson and Pope as recent writers? On the strength of Dandin's alleged word, more logical is J. Jolly's attempt to prove the contemporaneity of the Daśakumāracharita and the Arthaśāstra. But Kangle3 reminds us that Jacobi has shown Jolly's understanding of idānīm as "now" and "Maurya" as "a king" in Dandin's idānīm Mauryārthe... sam-ksiptā to be based on a faulty rendering of the text. At any rate, nothing in Dandin can stand against the implications of the other statement as expressing his chronological outlook, which is the orthodox one, entirely at variance with Goyal's.
No doubt, Goyal has several strings to his bow, but we must see exactly how far any of them can sound convincing in support of his late chronology. Thus, material akin to that of a particular age -say, the age of the Guptas - does not of necessity demonstrate that all the text was composed solely in proximity to it. There could be material gathered not merely from but even in different ages if it is something else than political information floating on from the past to be co-ordinated with current ideas. (We shall dilate on this
2.Ibid., p. 21.
3.Op. cit., III, p. 96.
Page 563
subject later.) Again, the simple non-mention of the Arthaśāstra anywhere does not by itself prove the book's existence only subsequent to the non-mentioning text. If the Nandīsūtra is the first to name it, we are not obliged to believe that in the time of the earlier compositions which are silent about it the Arthaśāstra was still unwritten: the silence may be there because no occasion has arisen to break it. As we shall soon explain, a group of factors are required to fix down the right occasion. In the meanwhile we may advert to the uncertainty of several dates taken for granted by Goyal.
Śūdraka, in whose Mrichchhakatika he discerns the earliest reference in Sanskrit to the Chānakya-legend, he traces to "probably fourth century A.D."; but M. A. Mehendale1 frankly confesses: "Nothing definite can be said either about the author or the play. It is even difficult to say whether the play was written before or after Kālidāsa, but the former view is generally accepted." The allusion to Kālidāsa, as we shall shortly note, throws a lot of issues into the melting-pot and can carry Sūdraka into the B.C. age. Majumdar2 warns us: "scholars are even now strongly divided in their opinion as to the age of Kālidāsa, and no definite opinion can be hazarded until more positive evidence is forthcoming."
Goyal has three glances at the Pahchatantra. We have already adverted to one. Earlier he3 specifies the Kathāmukha of this book as attributing to Chānakya a work on polity; in between he4 finds in the Kathāmukha a mention of the Kautilīya Arthaśāstra side by side with the Manusmriti and Vātsyāyana's Kāmasūtra as authoritative works on the trivargas. But from Majumdar5 we can infer that what Goyal refers to is a translation of an older work about which we learn: "The original of this work, now lost, goes back to the early centuries of the Christian era." In fact, there are several translations and according to Majumdar they belong to a period later than that dealt with in the volume in which he was writing -the period from A.D. 320 to A.D. 740.6 Two points draw our attention. Even at so late a time and in spite of Chānakya being
1."Language and Literature", The Age of Imperial Unity, p. 264.
2.Preface to The Age of Imperial Unity, p. li.
3.Op. cit., p. 4.
4.Ibid., p. 8.
6.K. M. Munshi, Foreword to The Classical Age, p. vii.
Page 564
made to author a work on polity no attempt is visible to identify this work with the Kautilīya Arthaśāstra. Secondly, if the original Pahchatantra goes back to the first few centuries of the Christian era we may not be able to affirm that the Arthaśāstra of Kautilya was not known before the period Goyal gives it.
Goyal1 has also remarked about the treatise: "It is...highly likely that Āryasūra (434 A.D.) the author of the Jātakamāla...knew it." But the date here is misleading. Mehendale2 informs us: "The Jātakamāla was translated into Chinese in A.D. 434, and Ārya Sūra therefore probably lived in the third or fourth century A.D." Knowledge of Kautilya's book in the third century - 200-300 A.D. - would render it as good as impossible for this book to have been written "around 300 A.D." What is more, Mehendale's inference is itself arbitrary. The time of the original need not have been so close to that of the translation. Between the translations of the Panchatantra and the original, more than four centuries could elapse. Between the translation by Somadeva (c. 1070 A.D.) and the original Brihatkathā of Gunādhya more than five centuries at the least and over a thousand years at the most could intervene. There is no reason why we should not credit a phrase like Warder's:3 "Ārya Sūra, possibly a contemporary of Mātrceta -probably soon after A.D. 176." Then we cannot place, à la Goyal, the Arthaśāstra "towards the close of the third century A.D.."
The chronology of this treatise can really be ascertained and the book declared unwritten prior to a particular period if five factors are involved. First, an author must bring in the relevant theme of polity. Secondly, there must be more than one author with such a theme so that a general situation is evidenced. Thirdly, they must be writing in broadly the same span of time. Fourthly, they must all bypass Kautilya. Fifthly, each one's allotted approximate range of date within the common spectrum of time must be beyond controversy. Concretely put, the crucial issue is: "Are there a number of texts refering to Arthasastric literature and each general date securely fixed to the time more or less preceding Goyal's 'around 300 A.D.' and all of them devoid of reference to the Kautilīya Arthaśāstra?"
Goyal has essayed to meet such a test. Let us look at his very
1.Op. cit., p. 7.
2."Language and Literature", The Age..., p. 267.
3.Op. cit., p. 178.
Page 565
words. Taking the Guptas to have commenced where modern historians put Chandragupta I - namely, 320 A.D. - he1 notes; "...while referring to the authorities on the science of polity, pre-Gupta literature usually mentions those scholars who are described by Kautilya as his own predecessors. Kautilya begins his work with salutation to Śukra and Brhaspati evidently ranking them as the founders of the two greatest schools of Arthaśāstra. In the body of his work, again, he quotes several times the views of the schools of Manu, Brhaspati and Usanas (Śukra) as well as Parāsara. Among individual teachers the most frequently quoted names are those of Bhāradvāja, Viśālāksha, Piśuna, Vātavyādhi, Bāhudantīputra and Kaunapadanta. Now, while referring to the authorities on the science of polity the pre-Gupta literature refers to those very predecessors of Kautilya with conspicuous omission of Kautilya himself. For example, in the Mahābhārata (which received its final form in the beginning of the Gupta age) it is said (XII.59) that the archetypal work of Brahmā on dandanīti was successively summarized by the gods Śiva (Viśālāksha) and Indra (Bahudantaka) as well as the sages Brhaspati and Kāvya (Śukra). In his Buddhacharita (1.46) Aśvaghosa (c. 100 A.D.) states that Śukra and Brhaspati created that Rājasastra which their fathers, Bhrgu and Ahgiras respectively, had not done. Similarly, in his Kāmasūtra (1.5.7) Vātsyāyana (c. 3rd century A.D.) has stated that out of the archetypal work of Brahma Manu prepared his treatise on Dharma, Brhaspati on Artha and Nandin on Kama. In his Pratima Nātaka at one place (Act V) Bhāsa makes Ravana enumerate the most important works on the various sciences including the Mānava Dharmaśāstra or the Manusmrti which the king of demons had studied. There, on the science of polity reference is made to the Arthaśāstra of Brhaspati, and not to that of Kautilya. These references prove that the pre-Gupta literature was not only ignorant of the existence of Kautilya, it positively referred to his predecessors as authorities on the science of polity."
Goyal's argument for a late Arthaśāstra could be binding if his dating of the various authors were really final. But is it? Mehen-dale2 speaks of "the question of the date of Bhāsa" as "a matter of prolonged controversy" since "Kālidāsa's reverential reference
1.Op. cit., pp. 7-8.
2."Language and Literature", The Age of Imperial Unity, p. 261.
Page 566
to him as a poet of established repute makes Bhāsa's date hinge upon that of Kālidāsa". There are two theories about Kālidāsa, one placing him in the 1st century B.C., the other in the 4th century A.D. To Mehendale the latter "seems more plausible", but Majumdar,1 weighing the arguments for it yet without supporting the other theory, says: "these are all mere conjectures which do not carry conviction.... we must admit that the evidence adduced in support of it is neither definite nor direct nor decisive." The great Sanskrit dramatist's chronology still hangs in the balance within a range of nearly 5 centuries. So it is not at all improbable that Bhāsa flourished earlier than the 2nd or 3rd century A.D. Indeed, a footnote to the provisional choice of the late dating tells us: "For an attempt to place Bhāsa in the 5th or 4th century B.C. see Pusalker, Bhāsa - A Study."
Aśvaghosha too has to be dated in relation to Kālidāsa. Currently he is set in the 1st century A.D., but a debate goes on whether he preceded or succeeded Kālidāsa, for, as Mahājan2 informs us, "there are certain admitted resemblances between some verses of Aśvaghosha and Kālidāsa". Mehendale3 remarks: "The poems of Aśvaghosha, though not widely read these days, had strength enough to influence the diction and incidents in the works of Kālidāsa." The view which puts Kālidāsa in the 4th century A.D. finds nothing derogatory in Kālidāsa borrowing from a lesser poet and improving upon the original. So if Kālidāsa existed in the 1st century B.C., as he very well might have, Aśvaghosha can retreat into the B.C. period and may easily be a number of centuries earlier than Kālidāsa. If Bhāsa could be, à la Pusalker, in the 5th or 4th century B.C., Aśvaghosha could be still earlier since, as Mehendale4 observes, "Bhāsa's Prakrit is later than that of Aśvaghosha".
As for Vātsyāyana, the issue is again complex. His non-mention of Kautilya's Arthaśāstra may simply mean that he chose Brihas-pati's work as a better model in the sphere of Artha just as he chose Manu's treatise on Dharma and Nandin's on Kāma out of whatever else might have been available. From Goyal5 himself we
1."Literature", The Classical Age, p. 303.
2.Op. cit., p. 453.
3."Language and Literature", The Age of Imperial Unity, p. 278.
4.Ibid., p. 261.
5.Op. cit., p. 10, fn. 4.
Page 567
learn that Winternitz was not deterred from placing Vātsyāyana in the 4th century A.D. and Kautilya in the 3rd. But surely Goyal is right in interpreting Vātsyāyana's omission in the same light as he does that of the rest and making him precede Kautilya? Actually, this author's date can be calculated with some historical force from only one item recorded by Goyal:1 "the fact that he has referred to Kuntala Sātakarni." At the moment we do not need to come to a precise perspective. We shall do so before long. Meanwhile a general picture can be projected. If the uncertainty of Kālidāsa's date allows us to shift both Bhāsa and Asvagosha substantially back in time and if Vātsyāyana and these two omit Kautilya in similar lists, the first named of the three can go into a fairly pre-Christian period along with the latter pair. Then there will be sufficient antiquity for the initial Arthaśāstra as a successor of the Kāmasūtra.
The chronology of the Mahābhārata is also nothing pinned-down. Led by "the huge conglomeration of matter of the most diverse type" and by "considerations of style, language and metre", Mehendale,2 though opting, in the framework of the current time-scheme, for a particular time-bracket for the poem's origin and development, grants: "It is... extremely difficult to speak of the age of the epic except in a general way. Scholars hold widely divergent views on the subject, but we may accept for all practical purposes the one expressed by Dr. Winternitz in the following words: 'The Mahābhārata cannot have received its present form earlier than the 4th century B.C. and later than the 4th century A.D.' " Obviously, this does not mean that we could be certain, as Goyal implies, of the epic having "received its final form in the 4th century A.D." where "the beginning of the Gupta age" is usually seen. We have only a working lower limit: the present form might have been taken fairly before it. Even within the usual chronology it has been possible to think like Mahājan:3 "The Mahābhārata in the present form seems to have been well-known in the time of Patanjali in the second century B.C." Mehendale" also observes that "Patanjali definitely knew a Pāndu epic" and adds that "Pānini explains the formation of the names of the
1.Ibid., p. 10.
2."Language and Literature", The Age of Imperial Unity, p. 251.
3.Op. cit., p. 115.
4.Op. cit., p. 251.
Page 568
epic personages and the word 'Mahābhārata'." It is perfectly possible that a version of good length took shape before Pānini's time, especially as he was not alone in alluding to the term "Mahābhārata" no less than to "Bhārata" to indicate a narrative of the Bhāratas' battle. In Mehehdale's words,1 "the Aśvalāyana Griha-sūtra knows both Bhārata and Mahābhārata as sacred books." As to Pānini's date Mehendale2 opines: "On the whole, we may place the great grammarian about the fifth century B.C." But he3 honestly prefaces this estimate by saying: "The date of Pānini is not definitely known and has been variously estimated between the seventh and fourth centuries B.C." A footnote by Mehendale4 cites Belvalkar, Systems of Sanskrit Grammar, p. 15, for 700-600 B.C. ascribed to Pānini. It should be reasonable to hold that except for certain patent interpolations or additions of a minor kind and barring some major parts like the Bhagawad Gītā which bespeaks a highly developed cult of Krishna the Avatar which Pānini is far from evidencing, the epic in substantial bulk is very old. Including those major parts and excluding only some minor items, its final form could easily have been available in advance of Patahjali's Mahābhāshya. So, with regard to this poem, as to all the other texts brought forward by Goyal, there is no positively established chronology and the marked flux of opinion easily permits us to argue a much greater antiquity for the Arthaśāstra than Goyal proposes.
3
To get an idea where approximately in the trans-Goyal antiquity we should situate the initial form of the Arthaśāstra, we cannot do better than carry a suggestion of Goyal himself from his chronological framework over to ours. He finds Vātsyāyana's sex- compendium the most helpful text, for at the same time that it omits Kautilya's work it provides by "its plan, language, style and basic attitude towards life" the closest resemblance to Kautilya's polity-compendium. Goyal,5 after this generalisation, goes on to
2.Ibid., p. 269.
3.Ibid., p. 568.
4.Ibid., p. 569, fn, 1.
5.Ibid., p. 9.
Page 569
particulars about the Arthaśāstra and the Kāmasūtra:
"Like the latter the former is written in the Sūtra style. In both the works verses from ancient texts have been quoted. Both the treatises are divided into adhikaranas which are subdivided into prakaranas. Each of them is, on the one hand, based on the floating mass of traditional material on its subject and, on the other, bears a distinct stamp of the original thinking of its author. Further, each of them cites the opinion of its author in the third person (iti Kautilyah and iti Vātsyāyanah), a style which is only rarely found in ancient Indian literature. In the Kāmasūtra there is a short adhikarana named aupanishadika which deals with artificial means of increasing youth and beauty, recipes for fascinating and making the desired man or woman submissive as well as for increasing sexual vigour, etc. Similarly, Kautilya has given an adhikarana of the same name in which he has described various mantras and recipes for producing illusive appearances, spreading disease and killing people on a mass scale, remaining without food for days together, making others sleep, etc. The attitude of both these masters is completely amoral; both of them proceed on the assumption that everything is fair in love and war. The amoral attitude of Kautilya in his inculcating the benefits of defeating an opponent by guile, or in his recommending unscrupulous methods for getting rid of inconvenient ministers and princes or in his formulating ingenious means of extorting taxes to fill the treasury is comparable to the indifference of Vātsyāyana to uprightness, as we see, for instance, in his complacent instruction regarding the ways of deceiving maidens, or in making shameless use of other people's wives for profit as well as pleasure or in his teaching'of calculated and sordid tricks to the harlot for winning love and lucre. These facts suggest that Kautilya and Vātsyāyana were the products of the same cultural milieu. In the words of Jolly 'no long interval can have passed between the composition of twosuch cognate productions'1.... Actually, according to a tradition recorded by Hemachandra, Vātsyāyana and Kautilya were the names of the same person.2 No corroborative evidence for this is so far available but in view of the evidence discussed above it can hardly be denied that Kautilya and Vātsyāyana must have been
1.Proceedings of the Oriental Conference, Allāhābād, 1926.
2.Introductory Essay in N. N. Law's Studies in Ancient Hindu Polity, pp. xiii-xiv.
Page 570
contemporaries or near contemporaries. And as in his KamaSūtra Vātsyāyana mentions the Arthaśāstra of Brhaspati and not of Kautilya, it may be presumed that the Kautilīya Arthaśāstra came into existence after the composition of the work of Vātsyāyana."
Now, Vātsyāyana is to be dated in relation to Kuntala Sātakarni. So we have to take stock properly of the point of his reference to this king. It is not as if he were this king's contemporary. What he gives is only a curious anecdote suiting his theme. Mahājan,1 while enumerating the Sātavāhanas (Āndhras) of the Purānic series, writes apropos of Kuntala: "It is stated in the Kāmasūtra of Vātsyāyana that Kuntala Sātakarni struck Malayavati, his chief queen, with fingers held like a pair of scissors and as a result of it that queen died." Hence all we can say is that Vātsyāyana is later in time than Kuntala. To date him in general we have to allot the latter his right historical place.
D. C. Sircar2 considers him as well as two others - Āpilaka and Bāla, about whom too we read in literature even outside the Purānas - to belong not to the main line but to collateral ones ruling at the same time in different parts of the Deccan. Kuntala Sātakarni he assigns to a branch ruling in the Kuntala country comprising the North Kanara District of the Bombay State and parts of Mysore, Belgaum and Dharwar." Sircar3 continues: "The Purānic lists make him a predecessor of Gautamīputra Sātakarni, and a commentator of the Kāmasūtra explains the name as being due to the king's birth in the Kuntala country." Then, following the idea of collateral lines, Sircar supposes that in the wake of the southern expeditions of Gautamīputra's son Pulumāvi the lieutenants of this king established themselves in the Kannada region, but before offering the supposition he is frank enough to say: "When exactly Kuntala came under Sātavāhana influence cannot be ascertained." Gautamīputra himself is credited by Sircar4 with rule extending to "the Krishna in the south". And he appears to have reconquered what the closely preceding members of his family had lost to the Śaka Nahapāna.5 Still earlier the king Sātakarni I is said to have exercised sway over wide regions and
1.Op. cit., p. 309.
2."The Sātavāhanas and the Chedis", The Age of Imperial Unity, p. 200.
3.Ibid., p. 207.
4.Ibid., p. 201.
5.Ibid., p. 200.
Page 571
probably his immediate predecessors to have done the same.1 If, as the Purānas aver, Kuntala Sātakarni, though later than Sātakarni I, preceded Gautamīputra, his very name should indicate a southern extension of the Sātavāhana kingdom long before Pulumāvi.
There is nothing conclusive to bar the Purānic view of Kuntala Sātakarni. Although Sircar argues for collateral lines and for a period of merely 300 years for the main one, he2 admits: "There is difference of opinion amongst scholars as regards the problem of Sātavāhana chronology". Both Vincent Smith in an older generation and Dr. Gopalachari recently have accepted a fairly longer duration and a larger number of kings in the main line.3 About Āpīlaka who in the Purānic list precedes Kuntala Sātakarni, Sircar4 himself, referring to "the discovery of a copper coin of king Śivaśri Āpīlaka in Chhattisgarh", informs us: "On numismatic grounds some writers ascribe Āpīlaka's coin to a date earlier than Gautamīputra, while others prefer to connect it with the later Sātavāhanas." With a greater antiquity than Gautamīputra open to Āpīlaka, Kuntala may well be one of Gautamīputra's predecessors. Then since Vātsyāyana together with Bhāsa and Aśvaghosha can be pushed into an appreciably ancient time as a result of the undeniable possibility that Kālidāsa flourished in the 1st century B.C., the Āndhra-Sātavāhanas may quite credibly have begun, as we have calculated from Purānic chronological material, in 802 B.C. and Gautamīputra mounted the throne in the year we have found for him: 510 B.C. In that case Kuntala Sātakarni goes still beyond this year and Vātsyāyana who comes after him should fit plausibly enough in the earlier part of the period in which we have set the initial form of the Arthaśāstra - a little prior to the Junāgadh Inscription of Rudradāman 1 which we have dated to 479 B.C. Around 500.B.C. could be the time of that form.
The crucial conjunction Goyal has presented of the Mahābhārata, Aśvaghosha, Vātsyāyana and Bhāsa to date Kautilya has failed to stand in the way of our chronology. Consequently, room is found also for the post-Sātavāhana Guptas just where we have situated them in time - their era commencing in 315 B.C.
1.Ibid., pp. 198-99.
2.Ibid., p. 195.
3.Mahājan, op. cit., pp. 307-08, in which after enumerating all the reasons pleaded by historians like Sircar the author favours the opinion against them.
4.Op. cit., p. 210.
Page 572
The lower end of our time-bracket for the Arthaśāstra has been near the Gupta age, in conspicuous difference from the Maurya period. Here Goyal brings a trio of pointers worth quoting. There is Kautilya's use of "the word pratyanta in the sense of a region which is included in a kingdom but is on its border (cf 1.17.42, V.2.3, etc.)." Goyal1 comments: "The word carried the same meaning in the Gupta age. For instance, in the Allāhābād pillar inscription of Samudragupta it is used for those states which were on the frontier of his empire but were included in it. However in the Aśokan inscriptions the term prachamta or pachchanta (Prakrit form of pratyanta) is used for the neighbouring states outside the empire of Aśoka. This fact should be regarded as a strong indication of the posteriority of Kautilya to Aśoka, and of his nearness to the Gupta age."
The two other terms bear on particular states or tribes. Although Mādra as a state was known in ancient times, "the significance of the inclusion of the Mādrakas in the list of the rāja-śabdopajīvin samghas [republics living by the designation of rāja] has... so far remained unnoticed. The Classical writers who have given a detailed description of the Punjāb states on the eve of the establishment of the Maurya empire do not mention them at all. But they certainly existed as a republic in the pre-Samudragupta period as they are included in the list of the republican tribes which submitted to the Gupta emperor.... Here it may also be noticed that the Arthaśāstra (III.18.8) refers to the Prājjunikas who have also been mentioned [as Prārjunās] in the Allāhābād pillar inscription of Samudragupta. They are not noticed anywhere else in the entire literature and epigraphy of ancient India."2
4
We pass over several points well made by Goyal which have already been observed by other writers who have tried to separate Kautilya's period from Megasthenes's. As a grand finale we shall draw on Goyal for a point few have emphasised as he has. Irrespective of whether or not Chānakya and Kautilya are the same person, what basically cuts off the author of the Arthaśāstra
1.Op. cit., p. 12.
2.Ibid., p. 14.
Page 573
from Megasthenes in Goyal's eyes are the utterly different aims of the two. He1 forcefully tells us:
"The attitude of a large number of scholars who believe in the traditional date of the Arthaśāstra is dichotomous. K.A.N. Shastri and P.C. Bagchi, for example, concede that 'the Arthaśāstra is, as its name implies, a general normative manual of polity laying down arrangements suitable for any independent kingdom at any time' and that 'Megasthenes recorded the impressions he derived by observing institutions in their actual working round about 300 B.C.'2 But after recording this opinion they forget it and use the Arthaśāstra for the study of the Maury a state and society and explicitly maintain that 'The machinery of government described in the Arthaśāstra may well be accepted as a representation of what obtained in Chandragupta's reign and that of Bindusāra.'3 But once it is conceded that the Arthaśāstra is a normative work, it becomes ridiculous to simultaneously believe that the condition of the state and society depicted in [Kautilya's] work necessarily reflects the actual conditions of his age, and specifically of the early Maurya period the known facts about which do not always correspond with norms laid down by him."
A few pages earlier we hear from Goyal:4
"Even if one concedes the theoretical possibility that the Arthaśāstra as a whole or in its kernel belongs to the early Maurya period, the problem remains whether or not the text may be used to reconstruct the picture of the state and society of the period. For nobody can deny that the Arthaśāstra does not purport to be the factual description of the state and society of the age in which it was composed; it is a normative work in which the author discusses the ideas of his predecessors and also of his own on polity, social organisation and economic activities. The assumption that his ideas may be used to know the Indian conditions of the early Maurya period will be as illogical as the assumption that one can know the facts about the condition of the society of independent India by studying the Gandhian Constitution for Free India. On no matter discussed in the Arthaśāstra one can be certain that Kautilya has stated a 'fact'. But historians usually forget this and
1.Ibid., p. 42.
2.In A Comprehensive History of India, II, p. 52.
3.Ibid., p. 58.
4.Op. cit., p. 29.
Page 574
after assigning it to the age of Chandragupta Maurya (which is itself a highly questionable belief) use it for the reconstruction of the socio-political and economic condition of that period. As a result of this arbitrary assumption and illogical procedure the differences between the Arthaśāstric norms on the one hand and the definitely known facts from the Indica of Megasthenes, the Aśokan edicts and other sources on the other, are neglected, minimized and sought to be explained away on highly subjective grounds. Instead, emphasis is given on their superficial similarities."
Remembering the normative nature of Kautilya's book we should think rather of this or that emperor founding his kingdom's organisation on the norms laid down in it than of its reflecting Chandragupta Maurya's or any other emperor's organisation of his kingdom. On this principle the organisation depicted by Megasthenes can hardly be said to show that his contemporary Sandrocottus imitated Kautilya's picture. Keith's judgment, as quoted by Goyal,1 is sound: "the similarities which are visible between the two authorities depend on-matters of a general character which are equally valid today... On the other hand, the differences...often touch on essential facts and point essentially to a distinction in date between the two authorities."2
Hence an analysis of agreements and disagreements in the politico-administrative sphere is "an unprofitable exercise so far as the date of the Arthaśāstra is concerned".3 As "the authors of both the Arthaśāstra and the Dharmaśāstra literature freely utilized earlier works on their respective subjects...it is quite natural...that early and late materials appear side by side in their treatises".4 That being so, Goyal regards it "as axiomatic that the dates of the composition of such works should be determined by their material of the later period".5 A study of the internal and external evidence pointing to the latest period discernible is, therefore, to Goyal the most definite and conclusive procedure.6 Although, in deference to several scholars' choice of the Mauryan period and belief in correspondence with Megasthenes, he has attended to their arguments,
1.Ibid., p. 5.
2.B. C. Law Volume, I, p. 483.
3.Goyal, op. cit., p. 7.
4.Ibid., p. 6.
5.Ibid., pp. 6-7.
6.Ibid.,
Page 575
his main line of research has been, as we have shown in brief, in the direction of internal and external signposts. They have led him to the close of the third century A.D. They lead us to almost the start of the third century B.C. for the final form - but under a few special conditions.
Not the Mauryas but the Guptas are then on the Magadhan throne. Also, the final form is not the original work with incorporation of material from other times: the original work begins about two centuries earlier when Kuntala Sātakarni is already in the past and either Gautamīputra Sātakarni or whoever was his predecessor is reigning. This view contradicts Goyal's assertion:1 "The treatise gives every impression of being the work of a single individual." Our grounds are unlike Thomas R. Trautmann's, whose recent computerised study "has virtually proved," as Basham2 acknowledges in a statement of which Goyal is unaware, the Arthaśāstra to be "a conflation of at least three earlier texts, composed by different hands." Working within the same system of chronology as Goyal, Trautmann suggests that the earliest layer belongs to 150 A.D. and the latest to c. 250 A.D.3 The second date nearly agrees with Goyal's own, but taken out of the conventional chronology Trautmann would be in step with our picture of a book developing through a number of periods, particularly as his dating suggestions are provisional and he confesses, as quoted by Goyal:4 "the conclusions I have reached contain no implications for the dating of the Arthaśāstra more specific than the one that there are several dates, and that the long books need not have been composed simultaneously."
Here is a denial, akin to ours, of the claim that only one person, taking into account varied floating material of diverse epochs, has penned the treatise. But we would go by the route we indicated when registering the possibility that material might be seen as hailing from certain ages in such a shape that we should have to take it to fix the collector of it to the very time it held good. This material, as we made clear, would have to be other than the politico-administrative kind: the latter kind could very legitimately be such as a writer around Goyal's "third century A.D." could
1.Ibid., p. 21.
2.Foreword to op. cit. in the preceding Supplement.
3.Goyal, op. cit., p. 2, fn. 3.
4.Ibid., p. 3, fn. 3 of p. 2 continued.
Page 576
gather while reviewing older accounts of Artha and weaving them into his own discourse.
5
One example of material of a non-assimilable nature would be reference to the currency of the time in which the part of the Arthaśāstra dealing with it was written. R. Shamasastry1 who was the first to discover and translate and comment on Kautilya's book drew attention to the subject years ago. He pointed out in some detail how the grammarian Patanjali, whom most historians date to the 2nd century B.C. but whom we have assigned to c. 75 B.C., mentions in his Mahābhāshya (1,2,3) an ancient system of currency obsolete and non-existent in his time, consisting of a pana of four pādas and of sixteen māshas. This is precisely the system described in item 103, chapter XIX, Book Two of the Arthaśāstra. The coins involved are also listed in chapter XII of the same Book. So Kautilya is obviously anterior to Patanjali. Patanjali does not say what prevailed in his day, but this "something else", according to Shamasastry, "was probably one-twentieth of a kārshapana, as stated in the Kātyāyana-Smriti".
Thus Kautilya goes past even Katyāyān who is commonly placed in the third century B.C. How far back he can go on the strength of his pana is difficult to say. Jolly and J. Meyer have declared that only Baudhāyana and Kautilya have a pana of sixteen māshas and that like Baudhāyana, who was a southerner, Kautilya must have been from the South.2 Kangle3 rebuts: "It is not true that all authorities except Baudhāyana and Kautilya know a pana of 20 Māsas. Manu, whom no one regards as a southerner, states in 5.134-136 that a kārsāpana (which may be presumed to be equal in weight to the pana) is one suvarna in weight, i.e. 16 māsas, not 20." V. S. Agarwal4 is in agreement with Kangle: "Māsha was both a silver and a copper coin... A silver māsha was one-sixteenth part of a karshāpāna and weighed 2 rattis (3.6 grs.),
1.Kautilya's Arthaśāstra, Seventh Edition (Mysore Printing and Publishing House, Mysore, 1961), pp. xxviii-xxx.
2.Goyal, op. cit., p. 28.
3.Op. cit., Ill, p. 114.
4.India as Known to Pānini, p. 207.
Page 577
as stated by Manu (VIII.135)." Goyal1 tells us that the Manu-smriti "is usually assigned to the second century B.C. or later." If not only in Patanjali's time but in Kātyāyana's as well the pana of sixteen māshas was replaced by one of twenty, Manu in the original and not the later expanded version must be older than Kātyāyana. Comparison to Baudhayana would carry Kautilya to c. 5th century B.C. according to the conventional chronology's dating of the earliest Dharmasūtras. Pānini, says Agarwal2 who dates him to c. 450 B.C., "uses both names, kārshāpana (V.1.29) and pana (V.1.34)" and mentions, among several multiples of this coin" 1/16 as māsha "(V. 1.34)." So Kautilya belongs to a very ancient tradition in his coinage. But he differs from Pānini in that the latter (V.127) "knows of a heavier kārshāpana called vimśatika equivalent to 20 māshas as against the standard karshapana of 16 māshas"3 and even a "trimśatka of thirty parts, i.e. māshas"4 (V.1.24) - "a name which is found only in the Ashtādhyāyī and not elsewhere".5
The data from Pānini would seem to show that while the trirhsatka died out the vimśatika which is found in the time of both Kātyāyana and Patanjali continued and that Kautilya lived in a locality where only the pana of 16 māshas was current but in a time after which even locally the pana he mentions ceased being used, as would appear to be the case since Kātyāyana and Patanjali who belonged to different localities are aware merely of 20 māshas to a pana. When precisely the Kautilyan currency stopped cannot be estimated, but it must be between Pānini's epoch and Kātyāyana's. Therefore the whole Arthaśāstra cannot be as late as Goyal makes out and all of it is not such floating politico-administrative items from various ages as can have been put together by a writer at the close of the third century A.D.
Another piece of information may also be placed beside the facts of currency in the Arthaśāstra. After detailing the group of deities in the book, Goyal tells us that the reference to Śiva along with Senāpati who is the same as Skanda-Kārttikeya, the God of Victory, and with Sankarshana suggests that this material used by Kautilya belongs to the first century of the Christian era and that
1.Op. cit., p. 9.
2.Op. cit., pp. 465, 464.
3.Ibid., p. 268.
4.Ibid., p. 269. 5. Ibid., p. 271.
Page 578
the same indication is given by the numerous references to temples and images of the deities.1 We can understand Kautilya using materials of various times in connection with the theme of polity and administration, but he is not concerned with collecting religious data of several ages. Where these data occur, the implication must be a recording of practices contemporary with those parts of the book. What would be the sense of giving past practices instead of present ones? And if present practices are there, then by Goyal's own testimony to their time - "the first century of the Christian era" - the Arthaśāstra cannot have been composed in c. 300 A.D. It is even challengeable whether in the terms of the current system of chronology his spotlighting of the first century A.D. is accurate for the religious associations reported.
T. M. P. Mahādevan2 writes: "The earliest historical record to mention the worship of Śiva is that of Megasthenes, the Greek envoy at Pātaliputra about 300 B.C. He describes two Indian deities under the names of Dionysus and Herakles, generally identified with Śiva and Krishna respectively. Patanjali in the second century B.C. refers in his Mahābhāshya to Śivabhāgavatas as also to images of Śiva and Skanda which were sold by the Mauryas to raise money." Beyond Megasthenes's c. 300 B.C. we are pointed by Mahādevan's information:3 "The fact that Śiva is classed among minor gods both iri the Āpastamba Griya Sūtra and Kautilya's Arthaśāstra shows... that his position of unquestioned supremacy is not established at a very early period." Sahkarshana (alias Baladeva, Balarāma, Rāma) too comes into prominence not before "the first century B.C."4 when the Ghosundi Inscription calls him "Bhagavat and Sarvesvara jointly with Vāsudeva".5 "The Buddhist Niddesa works" - "c. first century B.C."6 - "also mention the votaries of Baladeva side by side with those of Vāsudeva".7 The manner of his mention in the Arthaśāstra, like that of Śiva, that is, among minor deities, indicates a sufficiently earlier stage.
1.Op. cit., p. 32. Actually Goyal has the expression - "the last century of the Christian era" - which makes no meaning and must be a printer's mistake.
2."Religion and Philosophy", The Age of Imperial Unity, p. 456.
3.Ibid., p. 451.
4.V. M. Apte, "Religion and Philosophy", ibid., p. 449.
6.Sircar, "Religion and Philosophy", ibid., p. 437.
7.Apte, loc. cit.
Page 579
Another indication of antiquity is underlined by B. P. Sinha:1 "Kautilya refers to Samkarshana but not with Krishna; moreover the reference to the worshippers of the god Samkarshana as ascetics with shaven head or braided hair [13.3.54] points to a time earlier than when Samkarshana was a full-fledged member of the Vaishnava Vyūha, inalienable with Krishna or one of the Vīras."
A general remark of Sinha's2 is worth pondering: "Reference to the worship of Indra, Yama, Brahmā and Prajāpati or Kasyapa [2.24.27], the famous Vedic deities, certainly suggests quite an early period for the Arthaśāstra." We may add H. D. Bhatta-charya's observation:3 "Indra and Prajāpati the two outstanding divine figures of the Vedic and the Brāhmanic age respectively." In the light of it we cannot help noting Kangle's rendering4 of the Arthaśāstra'?, sloka: "Salutation to Kaśyapa, the Lord of creation, and to the god (of rain) always. May the divine Sita prosper in my seeds and my grains." The last phrase justifies Kangle's explanatory "of rain" pointing to Indra. Bhattacharya's other observation5 may also be attended to: "During the period of the Brāhmanas, Prajāpati occupied the topmost position and was looked upon as the creator of gods, men and demons... When the post-Brāhmanical age of rationalism was ushered in and the cult of sacrifice fell into comparative disuse, the worship of Prajāpati declined. But the theists coined a new name for him and called him Brahma, first of the later Hindu Trinity." We have also F. W. Thomas6 wondering - as we have seen in an earlier part of our book - how the Buddhist scriptures which are commonly considered contemporaneous with Megasthenes and therefore with the Mauryas in the conventional chronology give us Brahmā and Indra as the main gods worshipped whereas the Greek ambassador reports Śiva and Krishna under the names of Dionysus and Heracles as receiving the greatest share of popular adoration. Instead of separating the Mauryas from Megasthenes and thinking Brahma and Indra natural to the religious denominations enumerated by Aśoka's edicts which betray no sign of the Krishna-cult, Thomas
1.Readings in Kautilya's Arthaśāstra (Agam Prakashan, Delhi, 1976), p. 172.
2.Ibid., p. 170.
3.Op. cit., p. 475.
4.Op. cit., II, p. 175.
5.Op. cit., p. 464.
6.The Cambridge History of India, (1923), I, p. 485.
Page 580
imagines those scriptures to be "archaising" - for no reason he can provide. The interplay of all these testimonies among themselves and with the Arthaśāstra's deliverances in the religious field creates clearly the broad impression of an antique period appreciably prior to .the epoch of Megasthenes and also different in religious climate from the Mauryan epoch which is conventionally believed to start with Megasthenes. Of course, this impression shifts us to a still greater distance from Goyal's chronology for the book.
Furthermore, what is of special significance in the antiquating direction is the book's allusion to some among the minor deities whose worship is rare: Goyal1 mentions "the temples of Śiva, Vaiśravana, Aśvins", and he adds: "The reference to the temples of Vaiśravana (Kubera) and Aśvins is interesting, for temples of these deities are not mentioned in any other source." Kangle's comment2 on the text concerned - in connection with Otto Sten's comparison of the Arthaśāstra with the Silpaśāstra - is: "It is to be noted that Śiva and Vaiśravana are mentioned in a devatādvandva compound of Patanjali on Pānini, 6.3.26, suggesting an early date for them. The Aśvins are well-known Vedic deities, whose worship in later times is not known. Clearly the tradition of the Arthaśāstra must be regarded as much earlier than that of the Silpaśāstra and not contemporary with it." Here again we have for the Arthaśāstra, on the one hand, a plausible date older than Goyal's "towards the close of the third century A.D." and even "the first century of the Christian era" and, on the other, a pre-Patanjali period for which we have no fixed pointer.
Vaiśravana (Kubera), however, is not unique to Kautilya. Bhattacharya3 tells us: "It appears that the worship of Kubera-Vaiśravana was... not uncommon, for not only does Kautilya refer to the installation of his image in a fort, but there is sculptural evidence to prove that Kubera with his two nidhis - the conch-shell and the cornupopia exuding coins under a banyan tree (kalpa-vriksha) on the top of a column - was a favourite cult-object at one time." Still, we realise that the time of Kautilya's Kubera differed from the days when this deity had a particular position related to the
1.Op. cit., p. 31.
2.Op. cit., p. 85.
3."Religion and Philosophy", The Age..., p. 464.
Page 581
four quarters. Goyal1 gives the information from the Arthaśāstra: "It is said that Brahmā, Indra, Yama and Senapati are the presiding deities of the city in the north, the east, the south and the west respectively." Kangle2 draws attention to the fact that the presiding deities over the north and the west "are not Kubera and Varuna respectively as later". Not that Kubera and Varuna would necessarily fail to be such guardians in early times, but by and large their vogue for those quarters would be absent. Varuna is worshipped in the Arthaśāstra (13.2.16) in a context having nothing to do with the quarters. The tradition to which Kautilya attests for the quarters is completely missing in later times. Hence it drives us further into the B.C. age.
To complete our chronological survey of the rare religious rites, we may glance at the information Agarwal gleans from Pānini. Touching on the worship of various Vedic deities with oblation and performance of appropriate rituals by different classes of priests, Agarwal3 picks out from Pānini the mention not only of Indra and Varuna (IV.1.49) but also of the Nāsatya (VI.3.75), the two Aśvins. Here we have an extraordinary correspondence to Kautilya's disclosures. Again, there is Pānini's reference" to the bhakti of Mahārāja or Kubera (IV.3.97) who is also classed as a devatā (IV.2.35) to whom oblations were offered.5 We learn too from Agarwal6 that Patahjali (1.436) refers to the temples dedicated to Kubera besides those to Keśava (Krishna) and Rama (Krishna's brother Balarāma)." Apropos of Pānini's allusion to "the worship of Mahārāja, which was but another name of Vessavana-Kubera, who headed the group of the Four Kings or Regents of the Four Quarters and was the king of the Yakkhas in the North", Agarwal7 adds: "Pānini also mentions the descendants of Dhritarājan (VI.4.135) who may be identified as the Lokapāla Dhatarattha ruling in the East, at the head of the Gandhabbas." From these pieces of information based evidently on later Prakrit literatufe, we cannot gather for sure who were Pānini's Lokāpalas in the North and the East, but we can see that Kautilya's Lokapāla
2.Op. cit., II, p. 80, fn. to 2.4.19.
3.Op. cit., p. 350.
4.Ibid., p. 359.
6.Ibid., p. 460.
7.Ibid., p. 363.
Page 582
in the East was Indra as contrasted to Dhritarājan who may or may not have been Pānini's.
Two unusual features shared with Pānini - the cult of the Aśvins and that of Kubera - and the second of them shared with Patahjali suggest strongly a chronological position for Kautilya of the religious data in the interval between the authors of the Ashtādhyāyi and the Mahābhāshya, but closer to the author of the former treatise by virtue of that rarest of rare religious post-Vedic practices: the worship of the Nasatya.
When we cast about for analogues to at least some of the gods in the Arthaśāstra''s collection - Yama, Indra, Vaiśravana (Kubera), Sankarshana and Varuna - the only comparable list that strikes us is the Nānāghāt Inscription "belonging to the queen of a Sātavāhana performer of numerous Vedic sacrifices" and beginning "with an adoration to the gods Dharma, Indra, Sahkarshana and Vāsudeva, the Moon and the Sun, and the four lokapālas, viz. Yama (differentiated from Dharma), Varuna, Kubera and Vasava (differentiated from Indra)."' We cannot help marking that except for the inclusion of the Moon and the Sun and for the presence of Vāsudeva as a member like the others of a cluster of equal deities we have the religious milieu of the Arthaśāstra. Hence we realise the appropriateness of shifting the commencement of the Sātavāhanas considerably back in time from their current chronology. But, persuaded by Sircar's conjecture2 that the epigraph, "much of which is damaged", should be attributed to the wife of Sātakarni I, an early Sātavāhana king, our book has dated it close to our commencement of this dynasty in 802 B.C. There is no internal compelling factor in Sircar's favour. Neither Sātakarni I nor his wife Nayanika is named. Only two Kumāras, VediŚri and Śaktiśri, are mentioned. Their mother comes in but remains anonymous. We have no knowledge of either of the Kumāras succeeding Sātakarni I. Sircar3 says: "Kumāra Śaktiśri has been identified with the prince Śakti-Kumāra, son of king Sālivāhana of Pratishthāna, mentioned in literature." This hypothetical identification is of little help. It suggests the identity of Sātakarni I with the legendary Sālivāhana, reputed to be the founder of the Era of 78 A.D. No Sātavāhana whose coins or epigraphs have been found used any
1.Sircar, "Religion and Philosophy", The Age..., p. 438.
2."The Sātavāhanas and the Chedis", ibid., p. 199.
Page 583
era: everyone reckoned only in his regnal years. Further, what about Vediśri, obviously by his precedence of Śaktiśri the elder son of whoever is relevant to the inscription, and thus the claimant to the crown? Most Purānas give Purnotsanga as the successor of Sātakarni I, while some place, as Sircar1 has noted, Sātakarni II immediately after him. All that we actually know of the family of Sātakarni I from the "label" inscriptions in the Nānāghāt pass is that his wife was Nāyanikā and that a Kumāra named Hakusiri may have been his son and that Sātakarni himself was either the son or the grandson of Simuka-Sātavāhana.2 As the Purānas name Krishna as the successor of Simuka and an epigraph in a cave in the Nāsik hills attests to the sovereignty of king Krishna of the Sātavāhana-family,3 one of the two labels which in the series of eight are totally lost may have named Krishna. No Kumāra other than Hakusiri is in sight. It is a sheer guess that, as Sircar4 puts it, "Hakusiri" is "probably a Dravidian corruption of Sanskrit Śaktiśri". As our survey of the development of Bhāgavatism shows inscriptionally that this cult was in an immature state during the whole Sātavāhana epoch, we would be no less served by a somewhat later dating which would bring the Nānāghāt epigraph more in the neighbourhood of c. 500 B.C., our chronology for the initial form of Kautilya's book.
However, none of the subsequent kings who had substantial achievements known to us - Gautamīputra Sātakarni, Vāsishthi-putra Pulumāvi and Yajnasri Sātakarni - had a successor identifiable with Vediśri or Śaktiśri. If we overlook this lacuna, any of these three can be linked to the Nānāghāt Inscription with as much _ likelihood as Sātakarni I. Gautamīputra, the most famous of them, is praised in several inscriptions.5 He is said to have totally uprooted the Kshaharāta dynasty and to have extirpated the Śakas together with the Yavanas and the Pahlavas and is also described as the lord of many countries, the unique Brāhmana, the one whose chargers drank the water of the three seas. Well might he have performed the one Rājasūya and the two Aśvamedha sacrifices listed in the Nānāghāt Inscription. His mother figures in two
1.Ibid., p. 198, fn. 2.
2.Ibid., pp. 197-98, 199.
3.Ibid., p. 197.
4.Ibid., p. 199.
5.Ibid., pp. 182, 200-02.
Page 584
epigraphs;1 it would be natural if his wife figured in one. He had "very probably" even two sons by the same wife - Vāsishthī-putra Pulumāvi and Vāsishthīputra Sātakarni2 - whose non-official names during their minority may have been Vediśri and Śaktiśri.
Pulumāvi too was a conqueror. Yajhaśri who was next to Pulumāvi's successor was a still greater conqueror. But if the lacuna we have spoken of is too glaring we may pass over the three notable Sātavāhanas and seek for our man only among comparatively obscure Sātavāhanas - obscure where our knowledge is concerned - whose successors cannot be determined. There is a promising group - Mātharīputra Śakasena, Śaka Sāta, Vāsishthīputra Chatarapana Sātakarni - whose coins or inscriptions have been found but whose relations among themselves and with the other Sātavāhanas are unknown.3 If related in any way among themselves, they are likely to "have ruled over wide dominions" since one of them, Śakasena, may be identified with Śaka Sātakarni who is said to have ruled thus.4 Significantly, the last-named in the group is connected with Nānāghāt. Sircar5 speaks of "Vāsishthīputra Chatarapana Sātakarni of a Nānāghāt record of the thirteenth year of his reign". His wife may have been the queen of the other Nānāghāt inscription, and his two sons the Kumāras named in it.
No doubt, the palaeographist will protest: "How can the Nānāghāt Inscriptions be taken so far back in time wheri their characters are more developed and therefore later than those of the Besnagar epigraph of Heliodorus which cannot be much earlier than the end of the second century B.C.?" On more than one occasion in our book we have pointed out the shortcomings of palaeography as a final determinant of dates and sequences. We may add here that palaeographical opinions change from time to time. Before the discovery of the Besnagar Inscription, epigraphists, as Sircar6 has recounted, "usually compared the script of records like the Nānāghāt and Hāthigumphā inscriptions with that of the Aśokan epigraphs and assigned them to the beginning of the
1.Ibid., p. 204, fn. 1.
2.Ibid., p. 203.
4.Ibid.
6.Ibid., p. 195, fn. 1.
Page 585
second century B.C." Another discovery could upset the present Sircarian position. In his dealings with the Sātavāhanas we find Sircar himself at one place willing to make palaeography take the back-seat. Looking at coins discovered in the Chanda District, Madhya Pradesh, bearing the names of the kings Sātakarni and Pulumāvi as well as of some other Sātavāhana kings, Sircar1 writes: "In the absence of metronymics, it is with some diffidence that we propose to identify Sātakarni and Pulumāvi of these coins with Gautamīputra Sātakarni and Vāsishthīputra Pulumāvi. The palaeography of the Chanda coins of Pulumāvi seems to be earlier than that of the ordinary issues of Vāsishthīputra Pulumāvi; but, as Rapson has rightly noticed, the palaeographical test is not quite reliable in these cases." We may suggest that where crucial reconstructions of history for various weighty reasons are concerned, the palaeographical theory has to find ways to assimilate important changes of perspective. Historical occasion and not the palaeographical vision ultimately decides chronology. The Besnagar Inscription is itself dated to the end of the second century B.C. by Sircar2 not because of palaeography but because the king Amtalikita (the Greek Antialcidas), whose ambassador Heliodorus was, has to be dated in relation to the Indo-Greek kings Lycias and Heliocles on the one hand and on the other the Indian king Kautsiputra Bhāgabhadra to whose court Heliodorus came.
From Goyal's own book we can elicit an example of palaeography's untrustworthiness in particular situations. Though indebted to Sircar the palaeographist for assigning to the post-Aśokan era a number of inscriptions thought by some scholars to be pre-Aśokan, Goyal3 demolishes a palaeographical argument against his theory about the Brāhmī script: "One of the most important arguments cited in favour of the existence of Brāhmī in the pre-Aśokan period is the supposed existence of regional or local variations in the script of the Aśokan edicts which, if true, will tend to prove a pre-Aśokan beginning for Brāhmī. Biihler believed in the existence of such regional variations. G. H. Ojha accepted the regional influence only partially, while R. B. Pandey has concluded, without giving any argument in support, that 'regional subvarieties are also traceable in the Aśokan inscriptions'.
1.Ibid., p. 209.
2."The Yavanas", The Age..., pp. 115-16.
3.Op. cit., pp. 96-97.
Page 586
But a closer study of the provenance of the Brāhmī letters found in the Aśokan inscriptions shows that the theory of the regional variations is a myth because, as has been shown in recent studies of the Brāhmī script,1 all the different forms of the same letters may be found in the records of the same region, sometimes even in the same inscription. Therefore the differences in the forms of letters which we notice are not indicative of regional peculiarities; they were mainly due to the individual stylistic characteristics either in engraving, or in writing the draft that was supplied to and copied by the engraver." May we not let "individual stylistic characteristics" have a large say in all our assessments of the chronology of epigraphs instead of going rather mechanically by a theory of letter-development?
Apropos of the Nānāghāt Inscription and palaeography we may cast a look at the Ghosundi Inscription which Sircar2 regards as being of the same age: the first century B.C. This means that its characters are also more developed than those of the Besnagar Inscription which he puts anterior to the Nānāghāt. So they are far removed from the characters of the Aśokan edicts dated usually around the middle of the 3rd century B.C. But astonishingly D. R. Bhandarkar3 refers thus to the Ghosundi Inscription: "This record has been assigned by Buhler to the period between B.C. 350 and 250. The inscription cannot therefore be of any time later than that of Aśoka." Palaeographically, Buhler's judgment implies that Ghosundi's Brāhmī could be even earlier than Aśoka's. If so, Nānāghāt's Brāhmī need not be pinned down as posterior to Besnagar's but may go as far into the past as historical perspectives may demand.
We may dare to hold that everything seems to hang together in our scheme justifying the antiquity which we make the starting-point of the Arthaśāstra. And this drive deeper into the B.C. age receives a very surprising confirmation from a curious phrase in that treatise. The phrase (14.3.44) is part of a passage where strange-named devilish spirits are worshipped. Kangle's rendering4
1.C. S. Upasak, The History and Palaeography of Mauryan Brāhmī Script, (Nālandā, 1960), p. 14 ff; T. P. Verma, The Palaeography of the Maurya Brāhmī Script in North India (Varanasi, 1971), p. 2, and "Fresh Light on the Origin of Brāhmī", Journal of the Oriental Institute, Baroda, XIII, 1964, pp. 360-71.
2."Religion and Philosophy", The Age..., p. 438.
3.Aśoka, p. 208.
4.Op. cit., II, p. 587.
Page 587
runs: "(I bow) to Armalāva, to Pramīla, to Mandolūka, to Ghato-bala and to the service of Krsna and Kamsa, and to Paulomī, the successful." Kangle's footnote1 says: "krsna-karhsopacāram 'the service of Krsna and Karhsa, i.e. those who wait upon these two spirits' (Meyer). That Krsna and Karhsa here are unrelated to the heroes of the Harivarmśa is likely, though not certain. So Paulomī may or may not refer to Saci, Indra's wife." The predominant impression is of a list of demons and a demoness. We may recall Sinha's remarking that Sankarshana in the Arthaśāstra is uncom-panioned by Krishna and is himself worshipped by queer ascetics who are quite other than the devotees connected with the well-known combined cults of Krishna-Vāsudeva, Sankarshana and other god-heroes. We should not be taken aback on seeing the name "Krishna" in unusual company and particularly linked to Kamsa. Kamsa in later tradition is notorious as an evil tyrant whom Krishna destroyed. Already "the Mahābhāshya, quoting passages from a Kāvya on the Kamsa-vadha episode, points to the pre-Christian origin" of this aspect of the Krishna saga.2 But in the Arthaśāstra the destroyer and the destroyed are comrades-in-arms. Obviously the treatise preserves the remnant of a legend from pre-Vaishnava times.
Benjamin Walker2 sums up the old information: "The name often occurs in the Vedas and other early literature without reference to any deity. Krishna, son of Devakī, is mentioned in the Chhāndogya Upanishad as a scholar who composed a hymn and as a pupil of the sage Ghora Āngirasa. Another Krishna is a rishi, the son of Viśvaka, while yet another Krishna was a 'loud-yelling' non-Aryan asura chieftain of the Jamnā region who led a 'godless legion' of ten thousand followers and committed great havoc until he was defeated and skinned by Indra. One Krishna was also a Dravidian god of youth. A Vedic passage speaks of a leader of fifty thousand Krishnas, who was captured and slain together with all his pregnant wives so that he might leave no issue. There is evidence to suggest that he was 'a hater of the Brāhmīnic faith' who declared T will surely cause the worship of cows, through force if need be'." Very plausibly an Asura Krishna like the "loud-yelling" chieftain is the associate of Kamsa in the Arthaśāstra.
1. Ibid., p. 586. 2. Apte, loc. cit., p. 451.
3. Hindu World: An Enclyclopedic Survey of Hinduism (George Allen & Unwin, London, 1968), Vol. I,
Page 588
A hint of this pejorative nomenclature from the Veda is in the paragraph in Kangle just preceding our context: "I bow to Bali, the son of Virochana, to Śambara of the hundred guiles, to Nikumbha, to Naraka, to Kumbha, to Tantukaccha, the great Asura." Śambara, who at once rivets our gaze with the power of trickery and magic attributed to him, gets from Walker1 the note: "Śambara in the Vedas is a demon, and appears like Vritra to be a personification of drought. He opposed Divodāsa, and was defeated by Indra. In the Purānas he carried off Pradyumna and was killed by him." Along with the growing legend of the divine son of Vāsudeva, there seems to have been a lingering legend of a devilish spirit of the same name who could team with the likes of Śambara such as Karhsa. So we should be in an age prior to the epoch when Krishna was a recognisable divine being with a substantial following, even if he was not yet a supreme Avatar r The epoch of Megasthenes would appear to be posterior to the age of the Arthaśāstra's Krishna - indeed a far cry from the time Goyal allots to this treatise. Our dating of the earliest parts of it to c. 500 B.C. on various grounds apart from its 14.3.44 as well as other contexts looks fairly reasonable.
All in all, Goyal's fascinating, forceful and detailed study does not compel us to alter our outlook. In fact it provides us at many points with pathways along which, with a different vision from his, we can attain with greater precision our own goal.
1. Ibid., p. 92.
Page 589
THE NEW CHRONOLOGY AND A GENERAL SUMMARY
This chronology is counted backward and forward with certainty or probability, according to various considerations, from 315 B.C., the accession-date fixed for Chandragupta I, founder of the Imperial Guptas, identified with Sandrocottus of the Greek accounts on the basis of logical analysis of the modern chronological stand, the traditional-Purānic time-scheme and the evidence, both chronological and historical, gleaned from the Indica of Megasthenes, the Greek ambassador at the court of Sandrocottus.
C. 1482 or 1452 B.C. The Bhārata War, fought 9 or 8 generations of spiritual teachers - each generation said to have been of 30 years - before the 36th year of Buddha's life when his ministry started.
C. 1446 or 1415 B.C. The death of Krishna, marking the Kaliyuga which started 36 years after the Bhārata War, and synchronising with the accession of Parīkshit, the grand-nephew of Yudhish-thira, to the throne of Hastīnapura.
C. 1171 B.C. The death of Mahāvira, at the age of 72, 3 years before that of Buddha.
C. 1168 B.C. The Parinirvāna (death) of Buddha at the age of 80, 218 years before the accession of Aśoka Maurya.
C. 950 B.C. The accession of Aśoka whose personal name is now known from an inscription of his to have been Priyadarśin, either 49 years (by the Purānic calculation) or a little longer period (according to the calculation of the Ceylonese Chronicles) after the accession of his grandfather Chandragupta Maurya to the Magadhan throne.
During the major part of his reign Aśoka set up Edicts solely on stone pillars and slabs: even his Pillar Edict VII engraved in his 27th year mentions only these two materials and not rocks at all. His numerous Rock Edicts came afterwards.
Along with his famous 14 R.E.s all over India in several dialects of Prākrit, he set up on a rock-face at Kandahār an Edict in an Aramaic studded with extreme archaisms and very ancient Avestan words, differing most markedly from the Aramaic prevalent in the time modern historians give to Aśoka: c. 269-232 B.C. It has not only all the signs of the usual Aśokan
Page 590
mode of sentiment and style but also an unmistakable pointer to the people addressed in it. It has no reference to those two religious denominations which figure again and again in Aso-ka's Edicts: the Brāhmanas and Śramanas. This lack at once reminds us of Aśoka's statement in one version of R.E. XIII that the Brāhmanas and Śramanas went everywhere in his empire except in the province of the people he called Yonas (=Sanskrit Yavanas).
In such antiquity as the Aramaic here suggests, the Yonas could not be what modern historians suppose them to have been from the resemblance of "Yona" to the Persian "Yauna" which equates to "Ionian": the Greeks. According to all literary Indian tradition the Yonas are a tribe of "degraded kshatriyas", closely associated with another such tribe, the Kambojas. The grammarian Pānini who first mentions the Sanskrit form "Yavana" has the same suggestion in his Gana-pātha. In Buddhist literature, a border-state of Yonas is said to have existed when Buddha was born, and Buddha is also made to refer to the Yonas and Kambojas as his contemporaries. Aśoka also links the Yonas with the Kambojas. Everything pulls us away from the Greeks and joins the Aramaic version to the ancient non-Greek Yonas.
From certain practices attributed by a Buddhist Jataka to several Kambojas we understand that some of them must have followed the Mazdean religion. The Mazdean religion, in a specially orthodox form, intolerant of "daiva-worship", was just the one that would strictly exclude Bramanas and Śramanas. The Greeks never banished non-Hellenic religions. Aśoka's Yonas to whom no Brāhmanas and Śramanas ministered, were obviously Irānianized Indians steeped in Mazdeanism.
It is in this light that we have to look at the Greek version engraved in the blank space above the Aramaic, in preference to the one below. Linguistically, that version is datable to the period 275-225 B.C. Its contents are not so close as those of the Aramaic to the Indian incriptions of Aśoka. It appears to be a generalised translation of the Aramaic adapted to Greek needs. It must have been engraved as a result of Greek interest in Indian cultural matters such as is amply evinced from the time of Megasthenes's Indica down to the duration of the reign of Sandrocottus's son Amitrachates to whose court also a Greek
Page 591
ambassador had come - Daimachus - in the wake of Megasthenes and his successor Dionysius. Ever since the sojourn of Alexander the Great in India a Greek colony had existed in "Arachosia", the region around Kandahār. Precisely in the period 275-225 B.C., to which the Greek version of the Kandahār bilingual is dated, the head of the Alexandria Library wanted, as Epiphanius reports, Hindu books to be translated into Greek.
To the same period belongs a Greek translation of parts of R.E.s XII and XIII, engraved on a portion of a stone house in Kandahār or nearby. None of the large number of Aśokan Edicts in either Prākrit or Aramaic was ever on any house. Such an engraving cannot be thought of as by his orders, it must be the work of the Arachosian Greek community.
In Aśoka's own age, more than 600 years earlier, the Yona tribe lived in the very location of this later community. Contiguous to it was the domain of the outsider "Yona rāja Arhtiyoka". The name has been equated with that of one of Alexander's successors, a king of Syria and Western Asia: "Antiochus". According to us, it denotes not a Greek but an Irānianized Indian, such as must have been the "Yavana rāja Tusāspha" mentioned in the Junāgarh Inscription of Rudradāman I, as the viceroy of Aśoka in Saurashtra.
Beyond this Arhtiyoka were four rājas who are not labelled as "Yona": Turamāya, Arhtekini, Magā (or Makā), Alikasudara. These too have been equated with the post-Alexandrine Ptolemy of Egypt, Antigonus of Macedon, Magas of Cyrene and Alexander of Epirus or Alexander of Corinth. But it is forgotten that the 5 Greek kings concerned were not the only ones in the post-Alexandrine age. There were some others of equal if not greater importance whose omission from Aśoka's field of Dftawia-propagation is inexplicable e.g. Eumenes of Pergamon (262-240 B.C.) and, nearer home, Diodatus of Bactria (256-245 B.C.). One may also ask why one of the two neighbouring Alexanders - either that of Epirus or that of Corinth - was dropped.
All the names can be regarded as unusual Indian or Indo-Irānian ones. The alleged Greek equivalents can be criticized and the components of the names shown to be explicable without resort to them. All in all, it is possible to prove
Page 592
"Antiochus" indecisive, "Ptolemy" impossible, "Magas" unnecessary, "Antigonus" inaccurate and "Alexander" unlikely as well as gratuitous.
Aśoka can further be proved to have had no connection with Ceylon: his "Tāmbapamnī" and "Tāmbapamnīya" refer only to the South-Indian region of the Tāmraparnī river. The supposed relationship with King Devanariipiyatissa of Ceylon is a misapplication to Aśoka of another Indian monarch's dealings with Ceylon.
914 B.C. The end of Aśoka's reign, leading to a dismemberment of his empire - his sons and his grandsons (e.g. Dasaratha and Samprati) ruling independently in different parts of the country. Even the supposed last Maurya Brihadratha - who met his death at the hands of his commander Pushyamitra Śunga - was one of Aśoka's immediate or at least proximate successors.
Pushyamitra started the Śunga dynasty in the eastern provinces, where Daśaratha had been ruling. It lasted for 112 years but in its last 45 its power was superseded for all practical purposes by the Kānva dynasty, whose first member was a minister of the Śungas.
802 B.C. The approximate epoch of the grammarian Pānini's Ashtādhyāyi. This book speaks of its author's native province Gandhāra as an independent kingdom. The current dating of him - c. 450 B.C. - overlooks the fact that this date is part of the period during which Gandhāra was in the possession of the Achaemenid emperors of Persia, beginning with Cyrus (558-530 B.C.) and ending with Darius III who was defeated by Alexander in 330 B.C. In the immediate post-Aśokan period from 914 B.C. onwards when the Mauryan empire got fragmented, tradition posits Virasena, one of Aśoka's successors, to have set up as an independent ruler in Gandhāra. Under his descendants this province is likely to have continued as an independent kingdom to 802 B.C. and beyond. There are also other chronological clues in the Ashtādhyāyi and its companion Ganapātha. The former's word Yavanāni ("Yavana script") finds its gloss in the latter where the Yavanas, along with the Kambojas, are spoken of as "shaven-headed", a description rendering impossible the usual identification of them with the Greeks. Yavanāni indicates not the Greek but the Aramaic script. Nothing in Pānini shows Greek influence, whereas there are traces of vocables
Page 593
with Semitic derivations. A significant pointer to the age of the Ashtādhyāyī are the side-by-side words Vāsudevaka ("one who has bhakti for Vāsudeva") and Arjunaka ("one who has bhakti for ArJunā"). Their juxtaposition proves that Vāsudeva was not a unique object of devotion , and it directs us to the early stage of the cult of Vāsudeva-Krishna, which gets no mention in either the early Buddhist books or Aśoka's inscriptions although both these writings list a variety of contemporary religious sects . Moreover, bhakti in Pānini is a term applied also to cakes, suggesting "fondness" as the basic sense, so that the Vāsudeva-cult could hardly have been even a fully religious movement just initiated.
Pānini's time coincided with the start of the Āndhra Sātavāhana dynasty which put an end to the Kanvas as well as to whatever Śungas had survived as rois fainéants with the Kānvas. The Sātavāhanas, who ruled for 412 years, had their principal seat in the Eastern and Central Deccan but with a fluctuating overlordship of Magadha. During their whole period the Vāsudeva-Krishna cult was still in its early stage, as may be inferred from two inscriptions, one (Nānāghāt) in the first part of the period and the otherin the last.
711 B.C. The Mālava or Krita Era,founded perhaps by a Mālava hero named Krita. It is an error to think of either "Mālava" or "Krita" as an alternative to "Vikrama" and thus pointing to the last-named term 's era of 57 B.C. In no inscription is "Vikrama" accompanied by "Malava" or "Krita" . Only the latter two interchange.
551 B.C. The Śaka Era brought into Sind from Irān by the two Śaka families , the Kshaharātas and the Kārdarnakas, soon after the conquests to the west of the Indus by the first Achaemenid emperor Cyrus (558-530 B.C.). They call themselves "Satraps" (=Provincial Governors) without mentioning any overlord. It can be presumed that the overlord at first was Cyrus and afterwards Darius (522-486 B.C. ) who extended most the Achaemenid Empire to the vicinity of the Indus.
510 B.C. The accession of the 23rd Sātvāhana king, Gautamīputra Sātakarni, who in his 18th regnal year defeated Rishabhadatta, the viceroy of the Kshaharāta ruler Nahapāna, and caused the end of the Kshaharāta dynasty.
C. 500 B.C. The epoch of the Nānāghāt Inscription and the major
Page 594
part of the political treatise Kautilīya Arthaśāstra whose religious data those of this epigraph resemble. The turns of speech in the Arthaśāstra's original form anticipate closely some expressions in the Junāgarh Inscription of the Kārdamaka ruler Rudradāman I as well as in some Gupta epigraphs. A few crucial points of administration also tally with the Gupta regime rather than the Maurya. Its mention of Sanskrit as the State language indicates too a proximity to the conditions under Rudradaman I and the Guptas.
479 B.C. The Junāgarh Inscription of Rudradāman I of the Kardamaka dynasty, dated [Śaka] 72 from 551 B.C.
390 B.C. The beginning of an uncertain period of 75 years in Magadha after the end of the Sātavāhanas, during the later part of which the clan of the Lichchhavis appear to have controlled that province.
326 B.C. The crossing of the Indus by Alexander on April 13, early morning.
The Nāga king Chandrarhsa (known as Xandrames to the Greeks) reigned over the Indian interior about the Ganges and - at the head of the Gangetic peoples termed the Gangaridai by the Greeks - waited beyond the Ganges to give battle to Alexander if he should advance deeper into India.
315 B.C. The accession of Chandragupta I (known to the Greeks as Sandrocottus), founding the dynasty of the Imperial Guptas at Pātaliputra (Greek Palibothra) and initiating the Gupta Era in the year of his accession which seems to have tallied with his marriage to the Lichchhavi princess Kumāradevī. This was a few years after his limited and local kingship in the north-west, heading an army of rebels to expel Alexander's prefects from that region.
C. 305 B.C. Seleucus Nicator, successor of Alexander in the East, crossed the Indus at the river's 7 mouths to Sind but was met by Chandragupta I, pushed back, chased up the right bank of the river and defeated. The enemy and his troops were designated "Bāhlīkas" by Chandragupta I in the inscription he set up in the wake of his victory - the famous Meherauli Pillar Inscription of King "Chandra", topped by the emblem of Garuda (Eagle) consecrated to the God Vishnu. From this epigraph we may conclude that the Greeks, who in all their early advances towards and into India came via Bactria, old Bāhlīka, modern
Page 595
Balkh, were known in the 4th century B.C., as Bāhlīkas and not as Yonas or Yavanas. The latter terms came into vogue in a subsequent period - and then too it was not exclusively applied to the Greeks. Seleucus established friendly relations with the Indian monarch.
C. 302 B.C. The arrival of Megasthenes, the ambassador of Seleucus, at the court of Chandragupta I, for a long stay, during which he gathered material for his Indica. This book evinces knowledge of the traditional-Purānic chronology. Besides providing a picture, partly precise partly ambiguous, of the military, political and social state of the country, it sketches the various religious practices. It has no noticeable pointer to Buddhism but records, among other things, that the worship of Heracles (the Greek for "Hari-Krishna") which is the cult of Vāsudeva-Krishna (Vaishnavism, Bhagavatism) was in full swing at Mathurā among the Śūrasenas as well as elsewhere and that Vāsudeva-Krishna was connected also with the South-Indian Pandyas. All this information makes a total contrast to what we gather from Aśoka's Edicts and reveals for the first time the maturity of Vaishnavism-Bhāgavatism which thenceforth grew more and more, as can be seen from the Besnagar Inscription of the last quarter of the 2nd century B.C., and whose presence is amply attested by the Guptas who were predominantly Vaishnavites and Bhāgavatas.
C. 285 B.C. The accession of Samudragupta (known to the Greeks as Amitrachates, Sanskrit Amitrachchhettā, meaning "Mower of enemies", akin to the title given to Samudragupta in later Gupta inscriptions, Sarvarājochchhettā, "Mower of all Kings). Samudragupta, though a Vaishnavite, was a great patron of Art, Literature and Philosophy in general and encouraged Buddhism, so that it revived from the slump into which it seems to have fallen before and during the time of his father Chandragupta I.
280 B.C. The probable time of the final recension of the Bhagavad Gītā.
275-225 B.C. The Greek version of the Kandahār inscription was set up by the Greek colony in Arachosia most probably in the time of Samudragupta, in whose honour a famous epigraph was engraved at Allāhābād on the same stone-pillar as an Aiokan edict. The period of Samudragupta's reign was part of a post-
Page 596
Alexandrine epoch showing, as we have indicated, the intense interest of the Greeks in matters Indian.
During Samudragupta's reign King Devānampiyatissa who, by Ceylon's traditional chronology, ruled from 308 to 268 B.C. in that island figured under one of the categories mentioned in the Allāhābād Pillar inscription which speaks of Samudragupta's suzerainty over "the peoples of Simhala and other islands". This category is of countries "soliciting imperial charters confirming them in the enjoyment of their territories". The Ceylonese Chronicles strongly suggest Devānampiyatissa to have had a status feudatory to the contemporary King of North India. These Chronicles, however, have linked Devānampiyatissa to Aśoka. Even if Aśoka reigned, as modern historians hold, in 269-232 B.C., his inscriptions cannot be understood to connect him with Ceylon, nor does Devānampiyatissa's reign-period - 308-268 B.C. - give grounds for the belief that he was formally consecrated on his throne by Aśoka, whose reign started one year before Devānampiyatissa's ended and whose conversion to Buddhism and possible despatch of a Buddhist mission to Ceylon could not have preceded the conscience-searing Kalihga war in the 8th year of his reign.
The Allāhābād Pillar Inscription's expression Daivaputra-Shāhi-ShāhānuShāhi refers to Antiochus I Soter, the son of Seleucus Nicator by a Persian wife, and his successor who inherited from Seleucus not only Syria but all Western Asia, including the old Persian Empire. He ruled from 281 to 262 B.C., practically the same period as Samudragupta. Greek historians tell us of his friendly relationship with Amitrachates, the son of Sandrocottus. As master of the old Persian Empire and himself half-Persian, he would inherit the title of the Persian monarchs - not only Kshāyathiya, Indian Shāhi, meaning "King" but also Kshāyathiyānam Kshāyathiya, whose Indian equivalent is Shāhānushāhi, "King of kings". The Indian Daivaputra would correspond to the divine honours the generals of Alexander assumed after their own chief had been introduced to the Indian ascetics at Taxila as "Son of God". The spelling Daiva rather than the usual Deva is suggestive of a Persian connection, since in the Achaemenid inscriptions this spelling is always found.
The Śaka-Murunda of the Allāhābād Pillar are the Northern
Page 597
and Western Śakas following the Śaka Era of 551 B.C., and the former's associate tribe of the Murundas of Lahghan who later migrated to the area about the Upper Ganges delta. The Murundas and the Northern Śakas were subsumed under the name "Assakenoi" (=Aśvakayana) whose capital was "Massaga" (=Maśakavati) and who therefore were a shoot of the Scythian (Śaka) tribe called Massagetae by Greco-Roman historians and known as Maśakas, a warrior tribe of the Śakas, in Sanskrit literature.
259 B.C. Death of Samudragupta and, as calculated by one tradition of the Purānas, the end of the hypothesised Kaliyuga of 3102 B.C. The Greek name Amitrachates could stand not only for Sanskrit Amitrachchhettā ("Mower of enemies) but also for Sanskrit Amritakhāda ("Eater of Ambrosia") and this significance would be apt for one whom the Allāhābād Pillar designates "a god dwelling on earth" and achintya purusha in human form. Samudragupta in such an aspect corresponds to the Purānic Viśvasphūrti (also called Viśvasphani) who emerges as an Avataric figure, "Vishnu's peer", in a context describing the chaotic condition at the Kaliyuga's end.
Chandragupta II, Samudragupta's son, styled Vikramāditya, comes to the throne. He is known as the destroyer of the Western Śakas. His coins replacing those of these Śakas were struck some time between the 90th year of the Gupta Era of 315 B.C. and his last year on the throne, the 94th. These years are 225-221 B.C. The last coins of the Śakas are in the year 310 of their era. According to us, this era was of 551 B.C. So the Satraps' last coins date to (551-310) 241 B.C. They fall suitably within the reign of Chandragupta II (259-221 B.C.).
221-181 B.C. The reign of Kumāragupta I, son of Chandragupta II, into which the famous Mandasor (Daśapura) Inscription No. 52 mentioning a Kumāragupta as "the ruler of the earth" fits perfectly with both its chronological points - Mālava years 493 and 527 - falling, as they should, within the same king's rule if we calculate from our Mālava Era of 711 B.C.
181-169 B.C. The reign of Skandagupta, son of Kumāragupta I, also styled Vikramāditya. During this reign, there was again the engraving of a new inscription on a rock which already carried 14 edicts of Aśoka, as well as an epigraph of the Śaka Rudradāman I, which actually mentions Aśoka's governor of
Page 598
Śaurāshtra, "Yavana rāja Tusāspha". Here is another instance of an Aśokan text accompanied by additional matter during the Gupta period.
In Skandagupta's reign and even a little earlier - in the last part of Kumāragupta's - tribals from beyond the Indus irrupted into the Gupta empire. In the inscriptions on the Junāgarh rock they are called by the general name "Mlechchha". In the Bhitari Pillar Inscription of Skandgupta the name "Hūna" is said to have been found, but actually it is a reconstruction by Fleet from a mutilated word.
140 B.C. A powerful chief named Toramāna who had made Kāshmir his seat invaded Mālava state which was then under the Gupta king Bhānugupta. Toramāna is usually reckoned as a Hūna, but we have no evidence to that effect. He might have been a Kushāna or a Śaka. Called Shāhi in one inscription of his, he seems to have been a Śaka: Śakas no less than Kushānas are called Shāhis.
137 B.C. Toramāna died after a short rule over Mālava. His son Mihirakula ruled it for 15 years.
124 B.C. Bhānugupta engaged him in a battle but unsuccessfully. The battle is recorded in the Eran Inscription of Gupta year 191 in honour of Bhānugupta's feudatory Goparāja who lost his life in it.
The composition by the astronomer Varāhamihira of his earliest book, Panchasiddhāntikā, which dates itself in year 427 of the Śaka Era (of 551 B.C.). 122 B.C. Two Mandasor inscriptions of the year 589 of the Mālava Era of 711 B.C. record the triumphs of a Mālava chief named Yaśodharman, the greatest triumph being his victory over Mihirakula who is said to have been humbled for the first time. Yaśodharman differentiates Mihirakula from "the chiefs of the Hūnas," and, while not mentioning any fight with the Hūnas, concentrates on the defeat of Mihirakula.
Yaśodharman defeating the Śaka Mihirakula and liberating Mālava in 122 B.C. suggests an equation between him and the traditional Vikramāditya, the Mālava king who is designated "Śakari" ("Śaka-slayer") and, on a minor scale, "Hūnari", and who is associated with the era of 57 B.C. Vikramāditya is credited with 60 years of kingship. If Yaśodharman is taken to have ruled from 122 to 57 B.C. he would have a reign of 65
Page 599
years. Although the era of 57 B.C. is commonly regarded as having been founded by Vikramāditya, a persistent Jain tradition as well as a number of inscriptions considers it as founded on Vikramāditya's death by a grateful Mālava. Thus its establishment may be related to the end of Yaśodharman's reign. Even the account Yaśodharman gives of his empire's extent at the beginning of his reign in 122 B.C. agrees remarkably with the traditional picture of Vikramāditya's far-flung kingdom.
100 B.C. Mihirakula raised his head again but was vanquished by Narasirhhagupta Bālāditya, a Gupta emperor.
Around the same date come two Besnagar inscriptions expressive of Bhāgavatism. By this time the term "Yavana" for the Greeks has been brought into use in the midst of some other applications of it and it will soon be applied in general to all foreigners.
C. 75 B.C. The epoch of the grammarian Patahjali's Mahābhāshya in which we have evidence of Bhāgavatism in full flower with the mention of both Vāsudeva-Krishna and Sankarashana and with reference to the former's Vyūhas (forms or phases of conditioned Spirit) but with no allusion to the aspect of him as Nārāyana, an aspect found in a slightly later time.
Around this date the poet and dramatist Kālidāsa may have flourished.
57 B.C. The era traditionally associated with the King Vikramāditya of Ujjayinī, in whose court flourished, according to the Indian tradition, the "nine jewels" among whom Varāhamihira is put. Though several of them would seem really to belong to different epochs, the tradition is right about Varāhamihira.
C. 50-10 B.C. The Ghosundi Inscription which records the erection of a stone enclosure of worship for Bhagawat Sankarashana and Vāsudeva, within the enclosure of Nārāyana. Here Sahkarshana seems to have precedence of Vāsudeva. A slight deviation from the Vāsudeva-cult persisting from the 4th century B.C. is noticed.
42 B.C. The death of Varāhamihira in Śaka 509 as reported by Āmarāja.
C. 70-80 A.D. The Periplus of the Erythraean Sea, written by an Egyptian Greek.
78 A.D. The initial year of the Śaka Era commonly believed to be the era of the Śakas coming into power but actually one
Page 600
marking their destruction, as shown by several inscriptions and a number of literary documents. Everything considered, the destroyer seems best figured by the various Indian legends about King Sālivāhana of Pratishthāna as the founder of the Śaka-kala. The name "Sālivāhana" derives from "Sālavāhana" which comes from "Sātavāhana". By the new chronology, he is a late scion of the famous Āndhra-Sātavāhana dynasty which ruled from Pratishthāna. The historical circumstances disclosed by a critical study of the Periplus and of the later Geography of Ptolemy fit him into just the period required for destroying the Śakas who had returned victoriously to Ujjayinī some time after the death of Vikramāditya, their earlier conqueror, according to Indian traditions. These traditions bring both Pratishthāna and the Sātavāhanas on the scene, telling us as they do that Kālakāchārya, who had led the Śakas to Ujjayinī to avenge the ravishment of his sister by Vikramāditya's father Gardabhilla, retired after Gardabhilla's defeat to the Sātavāhana court at Pratishthāna.
At some point in the last 20 years or so of the 1st century A.D., Kahishka, the greatest of the Kushānas, may have started his reign in Northern India, though not all historians agree to this period. Mostly, he is taken to have established the era of 78 A.D. with his accession-year. But even if this accession-year be accepted, how can the era of his accession be termed Śaka-kāla? He was a Kushāna, not a Śaka, and Indian literature has always distinguished his tribe from that of the Śakas by designating it Turushka, Tukhāra or Tushāra. With the removal of the Western Satraps to a past more than 600 years earlier -the Satraps who are currently considered provincial governors under the Kushānas - Kanishka's supposed connection with these Śakas who are said to have followed the era of his accession and made Śaka-kāla possible as its name becomes out of the question.
Kanishka and his successors issued gold coins - the dīnāras -resembling on the one side the Roman denarius aureus and on the other the earlier Gupta coins, also called dīnāras, which had been issued from the time of Chandragupta II (259-221 B.C.) and had derived their name from an old Persian term for a weight or else an old Indian term for an ornament.
Here an allusion may be made to the comparative chronology
Page 601
of the Kushānas and the Guptas. The Gupta coins at the start are said to be imitations of those of the Kushānas. But the very first Gupta issues - the gold marriage-coins of Chandragupta I -are so markedly different that just where the Kushāna-model should be most operative it is found ineffective in a striking degree. The later Gupta coins may be considered as having influenced the Kushāna issues. Then there is the question of the goddess Ardoksho appearing in the latter as well as in some early Gupta coins. Ardoksho is often called "Roman" and therefore taken to date the early Guptas to a time when the influence of Rome on India had commenced - the Kushāna age in particular. Thus this age is thought to have preceded the Guptas. But the figure the Kushānas term Ardoksho is really a Persian deity, the same as the Avestan goddesses Asi Vanuhl of "good fortune" and Pārendī of "abundance". The appearance of the cornucopia, a Greek motif, along with her on Gupta coinage shows that in the post-Alexandrine epoch the Greek symbols came to be mingled with the Persian. The presence of Persian customs in the time of Sandrocottus whom we have identified with Chandragupta I is evident in the king's hair-washing ceremony reported by Megasthenes, a ceremony Herodotus ascribes to the Achaemenid kings. The incense-fed fire-altar on the Gupta coins could be another sign of Persia's cultural invasion. And their later repeating Archer-type might be a reflection of the most important of Persian coins, Darius I's famous "Darics" where the reverse has a kneeling Archer. The Gupta dīnāra varied in weight and the increase under Kumāragupta I to 132 grains compares with the Daric's 130, while the weight of the Kushāna issues remained the same from beginning to end: 120 grains on the average, close to the Roman denarius's 124. The major Guptas can easily be put before the Kushānas. As the Gupta dynasty, according to us, extended from 315 B.C. to 320 A.D., some of its members were naturally preceded by the families of Kadphises and Kanishka.
C. 130-140 A.D. The Greek geographer Ptolemy mentions a "Siroptolemeios" and a "Tiastenes" ruling at "Baithana" and "Ozéné" respectively. The reference is to two members of the families known as (Śri) "Pulomās" (also called "Āndhras") of Pratishthāna and as "Chashtanas" (from the Śaka Chashtana of the Kārdamaka line) of Ujjayinī. But the Greek geographer's
Page 602
information is often faulty with regard both to place and to time and it is very likely that the two rulers who are indicated by their dynastic and not personal names belonged to an earlier period.
C. 175 A.D. A late Gupta emperor named Śrīgupta, mentioned by the Chinese traveller I-tsing (last quarter of the 7th century A.D.) as having flourished 500 years before him.
C. 304-320 A.D. A minor Samudragupta most probably preceded by a minor Chandragupta, was contemporaneous - towards the beginning of his reign - with a Gadahara chief who, between 230 and 340 A.D., issued coins with the name "Samudra" on them, just as his probable predecessor issued them with the stamp "Chandra". The period concerned cannot hold the great Samudragupta even on the reckoning of the currently favoured chronology. For, in the Allāhābād Pillar Inscription Chandragupta I is stated to have handed over his throne to Samudragupta: the latter, to correspond to the terms in which he is named, must have been at least 25 to 30 years old at the time. But, with his father coming to the throne in 320 A.D. on marrying Kumāradevī, his reign-period could not have begun before 345 A.D. at the earliest and most probably began in c. 350 - both of which years are too late for the Gadahara chief.
The minor Samudragupta was also the San-meou-to-lo-kiu-to, whom the Chinese writer Wang-hiuen-t'se travelling in India in 647-48 A.D. mentions as having done a favour to the Ceylonese king Chi-mi-kia-po-mo (Sirimeghavanna). Sirimeghavanna, according to the Ceylonese tradition which reckoned from a hypothetical Nirvāna-era of 543 B.C., ruled from 304 to 332 A.D.
320 A.D. The Gupta dynasty reaches the end of its 645 years -after clear initial prominence, obvious middle-stage decline and evident closing revival of a rather unpleasant character on the whole. According to the Arab traveller and scientist Albērūnī (1031 A.D.), the end of the Guptas came to be celebrated as the Gupta Era starting 241 years after the era of 78 A.D. popularly called the Śaka Era. His actual words are: "As regards the Gupta-kāla, people say that the Guptas were wicked powerful people, and that when they ceased to exist this date was used as the epoch of an era." The use of this Gupta Era has nothing to do with the era employed by the Imperial Guptas
Page 603
who began their career in 315 B.C. with the king whom Megasthenes and other Greek writers mention as Sandrocottus.
399-414 A.D. The Chinese pilgrim Fa-hien's visit to India, supposed to be during the reign of Chandragupta II, but Fa-hien does not report the name of any reigning king and some of his information is at stark variance with what we definitely know about the life in Gupta times.
C. 425-575 A.D. A line of kings, with ups and downs of possession of Magadha, known to historians as the Later Guptas but having no ascertainable connection with the Imperial Guptas, started with Krishna-gupta and practically ended with Aditya-sena. Their very presence in this period would seem to exclude the existence of the Imperial Guptas as masters of Magadha and to discredit the current chronology and support the traditional-Purānic for Chandragupta I and his successors.
530 or 1165 A.D. The Gōkāk Plates of Dejja Mahārāja with their year 845 of an era called "Āguptāyikānām rājnām" which, if we attend to the force of the Sanskrit "Ā", means "of the kings who go up to the Guptas". "Up to" can take us either to the start of the Guptas or - more pointedly - to their end where their total number has been completed. In the second instance we would have a reference to Albērūnī's Gupta-Kala marking the end of the Gupta dynasty and then the date of the Gōkāk Plates would be 845 years added to 320 A.D. In the first instance the reference would be to 315 B.C. and the date 845 years after it. There is no third possibility and either alternative serves to support the Purānas' chronology of the Imperial Guptas, one negatively, the other positively.
630-643 A.D. The Aihole Inscription - in 634 A.D. - of King Pulakesin II, the successful southern opponent of King Harsha of Kanauj who is mentioned by the Chinese scholar Hiuen-Tsang as his patron during his travels in India in 630-643 A.D. Pulakeśin dates his inscription in a double way. He specifies the Śaka year 556 which, counted from 78 A.D., brings us to 634 A.D. - within the period of Hiuen-Tsang's association with Harsha. Then Pulakesin specifies 3736 years after the Kaliyuga, pointing from 634 A.D. to 3102 B.C., this yuga's traditional date. He also connects the Bhārata War intimately with the Kaliyuga.
A word on Hiuen-tsang's chronology of the Guptas will be in
Page 604
place here. The defeat of Mihirakula by the Gupta king Bālāditya, after Yaśodharman's victory over the former, is put by Hiuen-Tsang "some centuries ago", and not in c. 530 A.D. as modern historians believe - that is, fairly close to 633 A.D. when the Chinese pilgrim visited Śakala which was historically connected with Mihirakula. Other Chinese authorities are at one with Hiuen-Tsang on this point. Again, the latter's sense -though exaggerative - of the Guptas having flourished soon after Buddha's Nirvāna shows in general his sense of the antiquity of these kings instead of their proximity to his own time as the current chronology of them (320-569 A.D.) has it. He is one more support to our new time-scheme.
Page 605
Home
Disciples
Amal Kiran
Books
Share your feedback. Help us improve. Or ask a question.