Pramāṇa · Search · About · API

Age of the Nandas and Mauryas c11.s20: Excursus on the Arthasastra

Age of the Nandas and Mauryas c11.s20

Language
English

Excursus on the Arthasastra

c11.s20.1

There is not, and probably there never will be, a general agreement about the date and authorship of the *Arthaśāstra* of Kauṭilya; but doubts regarding the age and genuineness of the work have not been allowed to hinder the free use of the book in studies on Mauryan administration and society.

c11.s20.2

The volume of polemical literature that has grown round the work is too great to be reviewed here in its entirety. Prominent among those who have stood up for the genuineness of the work and a Mauryan date for it are: Shama Sastri who discovered the book and edited and translated it (1909—15) for the first time, Jacobi, V. A. Smith, Jayaswal, Ganapati Sastri—who issued a fresh edition of the text with an excellent commentary and J. J. Meyer who translated the work into German, and more recently Breloer. On the other side we have Jolly, Keith, Winternitz, O. Stein, F. W. Thomas, E. H. Johnston. Other scholars like Hillebrandt hold that the present text contains a genuine core much overlaid with additions and emendations by later hands.

c11.s20.3

In the introductions to his edition of the Arthaśāstra and his English translation, Dr. Shama Sastri briefly set forth the evidence, internal and external, in favour of the work being accepted as a genuine production of Kauṭilya, the celebrated Chancellor of Chandragupta Maurya. That evidence is strong and remains unshaken in spite of everything that has been said to the contrary.

c11.s20.4

Some of the objections taken are trivial and due to misunderstanding of Indian literary practice or of the Sanskrit idiom. That Kauṭilya (Crooked) is not a name that a great minister of State would have borne, that if he was the author he would not cite himself as ‘iti Kauṭilyaḥ’, much less refute himself, and that Daṇḍin refers to the work of ācārya Viṣṇugupta as recent, are examples. Other objections, vague and inconclusive, are of value only as indices to the prejudices of those that propose them : such are, for instance, the view that the Chancellor of the first Mauryan emperor had enough to do otherwise and could not have found time to write so systematic a treatise on politics and administration ; that the Arthaśāstra is so full of pedantry and schematic classifications that it could only have been written by a Pandit and no practical administrator or statesman ; that the polity of the Arthaśāstra is a small state and not an all-India empire like that of the Mauryas. The last statement appears a little more plausible than the rest only if we forget that the Arthaśāstra contains a specific reference to the whole of India being the Cakravartikṣetram (IX i), that Indian imperialism seldom resulted in the destruction of the polity of subject states, and that in the whole range of Indian political literature there is no other work with a better claim than the Arthaśāstra to be considered a manual of Imperial polity.

c11.s20.5

The KA., it has been argued, is too cyclopaedic a work to be considered the production of a single author, and in several technical sciences, civil and military, in architecture, metallurgy and so on, it represents a state of advancement which we can hardly ascribe to India of the fourth century B. C. This ignores Kauṭilya’s express statement that he consulted all the *arthaśāstras* of his predecessors in the field (*yāvanti arthaśātrāṇi pūrvācāryaiḥ prastāvitāni*), and *arthaśāstra* is a wide term which, as Winternitz has recognised, embraces technology, science, and all knowledge of practical arts besides politics. For the knowledge of agriculture, forestry, methods of elephant-lore, horse-training, mining and so on, the author may well have drawn upon such pre-existing works. And what right has any one to judge *a priori* the level of attainment in the practical arts that could legitimately be ascribed to India of the Mauryan epoch ? Let us not forget that the Aśoka pillars still exhibit a fine polish which time and neglect have not effaced, and of which the secret is yet to be discovered by the technicians of our own age. J. J. Meyer discusses these questions more fully in his introduction to his German translation of KA.

c11.s20.6

It has been said that no writer earlier than A.D. 300 is definitely known to allude to Kauṭilya ; but the Girnār inscription of Rudradāman (A.D. 150) knows of *praṇaya*, *viṣṭi*, and other technical terms in the sense in which they are employed by him, and the Tolkāppiyam the earliest extant Tamil grammar, borrows the whole list of *Tantrayuktis* given at the end of the *Arthaśāstra* and adopts them with very minor and insignificant changes.

c11.s20.7

Kauṭilya deprecates the use of wood in fortifications and defences ; but that Pāṭaliputra was surrounded by a wooden palisade is evident from Greek writings and the results of excavations. But to assume that Kauṭilya must belong to a much later age is not by any means the only way in which this discrepancy could be explained. Other equally inconclusive evidences that have been cited as proof of a much later date for the *Arthaśāstra* than the Mauryan epoch are : the use of Sanskrit as the language of royal rescripts contemplated in the *Śāsanādhikāra* which is in striking contrast with the employment of Prākrit in the inscriptions for several centuries from the days of Asoka ; and the mention of Pārasamudra and Cinabhūmi (II, ii) which remind us of Palasimundu of the Periplus and of the contact with the Chinese silk trade of a later epoch.

c11.s20.8

Several attempts have been made along different lines to arrive at a date for the *Arthaśāstra* later than the Mauryan epoch. Jolly has compared the *Arthaśāstra* with the *Dharmaśāstras* ; while he has succeeded in discovering many close parallels, it cannot be said that any clear conclusion can be drawn from them about the relative age of KA. Jolly has himself changed his mind ; in 1913¹ he was ready to concede that the text of Yājñavalkya as we know it did not exist when KA was written ; he said that while many neologisms are common between KA and the youngest smṛtis, there is little clue to decide which of them is earlier ; he was struck by many differences between KA and the smṛtis (torture, ordeals, divorce etc.) and accounted for them by assuming that they were due to deep-lying differences in standpoint known from the beginning between a code of custom and a text-book of politics. Finally he said that the kernel of KA belonged to about 300 B.C. and much of its contents was seen to be genuine by numerous early citations as shown by Zachariae, Hillebrandt, Hertel and Jacobi. The resemblances with the later smṛtis puzzled him and he left it an open question if the smṛtis revised the old law in the light of the KA or later ideas entered into KA and fused themselves inseparably with the text. Ten years later, in 1923 Jolly wrote : ‘One cannot help arriving at the conclusion that Kauṭilya must have been acquainted with the whole body of Dharmaśāstra literature much as we now have it, from the earliest Dharmaśāstras down to the most recent metrical smṛtis and smṛti fragments’. Even the great authority of Jolly cannot gain assent for so extravagant a conclusion in preference to the hesitant tone that marked his expressions a decade earlier, especially when we find him adding as an afterthought : ‘It is true that some facts seem to point the other way, so that Yājñavalkya instead of Kauṭilya would have to be regarded as the borrower, either directly or indirectly through the medium of a common source.’

c11.s20.9

¹ ZDMG. 1913, pp. 49-96.

c11.s20.10

² Introd. pp. 17-18.

c11.s20.11

Jolly employs a more general argument. ‘Generally speaking’, he says, ‘the Dharmaśāstra or science of duty and religion has far better claims to a high antiquity than the Arthaśāstra or science of gain which in its turn is older than the Kāmaśāstra or science of love, the three sciences based on Trivarga having followed each other in point of time as well as in rank and value’.¹ One may well doubt this; for even the earliest dharmasūtras known contain a core of rājanīti, the subject-matter of Arthaśāstra. Even granting the correctness of this sequence in the evolution of the sciences, we can get no decision from it about the age of any single work in view of the long tradition each of the sciences lays claim to. A plausible *priori* case can be made out for the view that early Indo-Aryan life was gayer and more materialist in its outlook, that the sciences of gain and love had a better chance of coming up in that period than in a later time when Indians became more and more intent on the other world and began to lay greater stress on dharma and conceived of mokṣa as the *summum bonum* of life. In truth, we know as yet too little of the growth of the concept of *puruṣārthas* to be in a position to affirm or deny the truth of Jolly’s proposition regarding the sequence of the sciences. But the Indian writers are seen to have recognised, in their works the interdependence of the different ends of human endeavour, and no work is purely a work of dharma or artha to the exclusion of the other. And one of the best summaries of general dharma is found in the pages of the *Carakasamhitā*, a work of medicine. It is not much use laying stress on the open recognition in KA of ‘reprehensible practices such as the murder of distinguished officers, the levying of highly oppressive taxes, the corrupt system of espionage’, and seek to draw any inference from it either about the age of the work or on the character of administration in the age which produced the work. The author of the *Kāmasūtra* gives the clue which has not received the attention that it merits. He says :

c11.s20.12

¹ *Ib.* p. 20.

c11.s20.13

*na śāstram-astityetena prayogo hi samīkṣyate |*
*śāstrārthān vyāpino vidyāt prayogāmstu-ekadeśikān ||*

c11.s20.14

Science is all-embracing thought; practice is quite another affair. The relentless logic with which the implications of state policy are worked out in the KA is an example of the perfection of scientific thought such as it was, and is no index to everyday practice.

c11.s20.15

The *Kāmasūtra* of Vātsyāyana borrows from KA its plan and scope, and many peculiar terms as well as entire paragraphs. Hence, says Jolly, ‘no long interval of time can have passed between the composition of two such cognate productions.’¹ Jacobi, as Jolly knows, did not think so, and in truth there can be no rule governing the interval necessary between an original and its imitation. The same consideration applies to the similarity of textual structure and of *tantrayuktis* between KA and the *Suśruta* to which attention has been drawn.² J. J. Meyer has also studied the relation between KA and the *smṛtis* with a view to determine the place of KA ; though his conclusion on the age of KA puts it in the Mauryan age, his views regarding the relative ages of the other *smṛtis* have not gained general acceptance and it seems unlikely that they ever will.³

c11.s20.16

¹ Introd. pp. 21-24.

c11.s20.17

² IC. IV. pp. 439-40.

c11.s20.18

³ *Uber das Wesen* and IHQ. IV. (1928) pp. 570-92.

c11.s20.19

The comparison of the KA with the epics has also led to no better results. Following up a line of thought suggested by Jacobi in the first instance, Charpentier compared the legendary examples cited by KA with the corresponding episodes in the *Mahābhārata*, and he reached the conclusion that the current text of the epic must have come into existence sometime between the date of the KA and that of the *Kāmandakīya Nītisāra*.⁴ He also pointed out that KA (1, 5) defines *itihāsa* in a manner that shows that K. was not thinking of our epic. On the other hand Hillebrandt and Meyer point out that the *Mahābhārata* knows all the predecessors of K. mentioned by him, but not K. himself ; and that the *kaccit* chapters in the *Rāmāyaṇa* (II 100) and the *Mahābhārata* (II 5) which agree with one another closely contain phrases which recall whole chapters of KA to the mind.⁵ Hillebrandt shows further that the *Rāmāyaṇa* is fully acquainted with the detailed terminology of the *Arthaśāstra* and apparently draws several verses from older systematic treatises on the science.

c11.s20.20

⁴ WZKM. 28. (1914). pp. 211-40.

c11.s20.21

⁵ Meyer : *Das Arthasastra* Intr. pp. xxxvii ff. Also Hillebrandt : *Altindische-Politik*. pp. 6-16.

c11.s20.22

Clearly there is no hope of reaching along this line any precise chronological indications regarding the date of KA.

c11.s20.23

Another inconclusive effort to place KA about A. D. 250 is that made by E. H. Johnston¹. He argues that Kauṭilya's work must have come into existence after the time of Aśvaghoṣa, but not very long after; Aśvaghoṣa does not use the term *vijigīṣu* though he knows the forms *jigiṣat* and *jigiṣu*, and he remains well within the limits of Dharma in his references to politics; hence he must have preceded Kauṭilya. Yet, the common use of neologisms by the two writers (of which examples are cited) shows that no long interval separated the two writers. And the fact that Āryaśūra (A. D. 434) unlike Aśvaghoṣa parades his knowledge of *Arthaśāstra* in the *Jātakamālā*, and refers to Kauṭilya, shows clearly that he came later than Kauṭilya. But the only certainty that emerges from Johnston's arguments is that Kauṭilya wrote earlier than Āryaśūra's time. For Aśvaghoṣa was under no compulsion to accept Kauṭilya's outlook and vocabulary if he had the work before him; many later writers are known to have declined to do so like Daṇḍin and Bāṇa, and to have condemned his doctrines and methods.

c11.s20.24

¹ *JRAS.* 1929, pp. 77-89

c11.s20.25

O. Stein's attempt to demonstrate that Megasthenes and Kauṭilya could not have belonged to the same period cannot be held a success. The detailed comparison of the fragments of Megasthenes with corresponding passages from the *Arthaśāstra* is valuable so far as it goes; but as Breloer has pointed out the method followed is too superficial and mechanical. On important subjects like the ownership of the soil, slavery, social organisation, legal procedure, and administrative arrangements it is quite possible, as we have seen, to explain the apparent differences and discover much closer similarity than Stein has found between the statements of the Hellenistic ambassador and of the Brahmin Chancellor of the first Mauryan emperor. And Stein apparently fails to note that some of his results e.g. the difference between the two writers on milestones point logically to the conclusion that Megasthenes must have written after Kauṭilya. But we may not follow Breloer the whole length². For we see little reason to accept his view that the theory of State-landlordism was introduced into India for the first time in the Mauryan epoch and borrowed from Ptolemaic Egypt ; as a matter of fact, as Breloer admits, there is no clear statement of this theory in the whole of Kauṭilya's work. According to Egyptian notions, the State was the ‘house’ of the king and its territory his estate¹. Such an idea was not accepted in India even by extreme advocates of State-landlordism who only made the king adhipati suzerain or major partner, whose rights were strictly limited by law and custom. Much less are we persuaded that the Kauṭilyan polity was a completely planned economy after the Nazi model as Breloer seeks to make out in the latest and most comprehensive of his Kauṭilya-studies. The principle of local and sectional autonomy was too deeply rooted in Indian cultural tradition for even the all-powerful bureaucratic control and regulation of the Mauryan empire to extinguish it or even to curtail its operation to any very considerable extent ; witness the multiplicity of metayage arrangements in ch. II 14 on sitādhyakṣa which finds parallels in several other sections of the work. In the war years German scholarship turned, possibly had to turn, to Nazi propaganda which no one should take seriously².

c11.s20.26

² Cf. *IHQ.* XI (1935), pp. 328-50.

c11.s20.27

¹ Rostovtzeff, Soc. and Ec. His. of the Hellenistic world (1941) p. 269

c11.s20.28

² Cf. Hauer, Glaubengeschite der Indo-Germanen i where Hitler is compared to Śrī Krishna.

c11.s20.29

The great value of the Kauṭilya studies of Breloer cannot be gainsaid. They offer several convincing interpretations by which the apparent conflicts between Megasthenes and Kauṭilya are resolved ; and they have done well to stress the fact that after Alexander's astonishing career, the world was no longer the same as before³. The great economic and political revolutions initiated by the establishment and the early break up of Alexander's empire, the increase in trade due to the multiplication of armed camps in the wars of ‘the successors’ and the divisions of the empire, the accumulation of large fortunes by some and the coming up of a proletariat, the dispersion of the large gold treasures of Persia and the quickening transition from a rural to a money economy, and the rise of several large territorial states under absolute monarchs were the chief features of the new epoch. India was drawn more and more into this welter, and Chandragupta and his teacher grew up in this atmosphere of rapid change and new formations. War, trade, diplomacy and travel, opened up numerous channels of increasing contacts with the outer world, and it should be no wonder if the unique character of the *Arthaśāstra* is partly due to the stress of foreign ideas and influences pressing their way into the political and administrative system of the newly founded empire of the Mauryas. There is much force in Rostovtzeff's observation that 'if one believes in the historical character and the early date of the kernel of the *Arthaśāstra* of Kauṭilya and in the radical centralization of Indian government effected by Chandragupta on Hellenistic lines, one may say that Chandragupta did more to Hellenize India than Demetrius and Menander.'¹ It is, however, not merely a question of Hellenic influence; for we know with certainty that the administrative system of the Hellenistic monarchies in Asia and Africa 'was practically a continuation of that of the Persian kings and it is equally certain that no such continuation would have been possible without the help of the documents and information assembled in the Persian archives'². In the natural reaction against Spooner's pompous announcement of a 'Zoroastrian period' in Indian history, there is a danger of Persian influences on Indian life being either altogether denied or considerably under-estimated. The mass and variety of detailed statistical information which the *Arthaśāstra* requires the officials of the State to collect and arrange for ready reference (e.g. in II 35 on *samāhartā* and 36 on *nāgaraka*) is something unique in all Indian political literature; we are tempted to suppose that the model for Kauṭilya and the Hellenistic states was furnished by the practice of Persian kings and satraps maintaining lists of inhabited centres of each satrapy, together with approximate estimates of their population and material resources, lists which they employed alike for purposes of taxation and preparation for war³. Kauṭilya's categorical statement in III (i) that a royal edict (*rājaśāsana*) overrides *dharma*, *vyavahāra* and *caritra* is somewhat exceptional in the political literature of India though Nāradasmṛti follows the Arthaśāstra on this point ; the more usual rule is to require the king to frame his edicts in conformity to Dharma, and some writers even go the length of implying that if this condition was not satisfied the royal order was invalid. Kauṭilya’s exaltation of the royal edict above canon and custom deserves to be compared with the growing dominance of royal legislation and jurisdiction, and the active exercise of royal authority in the sphere of civil law that were characteristic of the Persian empire and the Hellenistic monarchies¹.

c11.s20.30

³ KS. i. 108 ff.

c11.s20.31

¹ *Op. cit.* pp. 550-1.

c11.s20.32

² *Ib.* p. 1034.

c11.s20.33

³ *Ib.* p. 1033.

c11.s20.34

¹ *Ib.* pp. 1067-8.

c11.s20.35

Sylvain Lévi has argued that the mention of coral from Alexandria (*pravālam Alakandakam* II, 11, 41) shows that the Arthaśāstra must be later than the first century A.D. when the coral trade shifted to India according to Pliny and the Periplus². But there are numerous references to *pravāla* (coral) in the earliest strata of the Mahābhārata not to speak of the occurrence of the word in the gaṇapāṭha where we cannot be sure of the sense of the term. There is little reason to doubt that the coral was known in India much earlier than the first century A.D., and we know that coral was a valued article of trade in the Hellenistic world.

c11.s20.36

² *IHQ.* 12 (1936). pp. 120-33

c11.s20.37

It has been pointed out³, lastly, that while Kauṭilya prescribes at II 6 the recording of the year, month, *pakṣa* and the day in specifying dates, the inscriptions of Aśoka nowhere follow this system while an approach to it is seen in the Kuṣāṇa records which give the regnal year, season and day, and the exact adoption of the rule of Kauṭilya is found for the first time in the Girnar inscription of Rudradāman. Rudradāman’s inscription also employs the terms *praṇaya* and *viṣṭi* in the same senses as those known to KA. But this only shows that the composer of the Girnar praśasti may have known the KA, and means little for the date of that work. The Kuṣāṇa inscriptions do not follow the system of KA and may be left on one side ; Aśoka gives only the year from his coronation in his records and no other detail, evidently following the Persian models which he consciously followed. The Persian monarchs knew of a system of dating quite like that of the KA, but they did not employ it on all occasions, and the chronology of Darius's inscriptions is rather vague. Let us not forget also that the prescription in the KA regarding the manner of dating occurs in the chapter on revenue collection, and obviously bears a close relation to the form of accounts and is not connected with the issue of proclamations or edicts.

c11.s20.38

³ *IC.* IV. p. 442.

c11.s20.39

The author of the *Arthaśāstra* has been hailed on the one hand as the Indian Bismarck and a Realpolitiker, and dismissed on the other as a Pandit, a schematic theoretician whose logical categories had no relation to realities. An open-minded study of the entire work will reveal that there is some support from the work for either view; while the author does not shrink from pursuing traditional theories relentlessly to their logical conclusion as e.g. in the *maṇḍala* theory, still there are other sections of the work particularly the *adhyakṣa pracāra* which, like modern administrative manuals, deal with details of day to day work in public offices. And we should not omit to notice that in its concern for details of practical administration the *Arthaśāstra* is unique in the whole range of the Artha literature of ancient India. And some of its administrative terms like *puruṣa*, *yukta*, *mahāmātra* and so on re-appear in the edicts of Aśoka.

c11.s20.40

Considering its age and its technical character, we have reason to think that the text tradition of the KA has been quite good. Its length is indicated as 6,000 *ślokas* in the text itself and by Daṇḍin; and our text, according to Dr. Shama Sastri is of about the same extent.¹ But scribal errors, particularly in the transcription of unfamiliar geographical names about which Bühler has uttered a clear warning, and occasional interpolations or even rehandlings of parts of the text are not altogether ruled out of the range of probability. In a very useful and penetrating analysis of the *Śāsanādhikāra* (II 10)², Stein has sought to establish that this chapter as it stands bears a composite appearance and shows signs of having been remodelled in the light of Roman Imperial letters of a later time. But in the opinion of the present writer the *Arthaśāstra* has stood the test of very vigorous criticism for so long that its genuineness must now be recognised to have been placed beyond doubt and that, with minor reservations, the work may properly be looked upon as the authentic production of the scholar and statesman who took a hand in the foundation of the Mauryan empire.

c11.s20.41

¹ p. VII of the English translation.

c11.s20.42

² ZII. VI (1928) pp. 45-71.

c11.s20.43

CHAPTER VI

c11.s20.44

ASOKA AND HIS SUCCESSORS

c11.s20.45

The reign of Asoka forms the brightest page in the history of India. The ruler himself takes rank easily among the masterminds of the world, and under his leadership India came to occupy the foremost place among the civilized nations of the time. Inheritor of an extensive and highly organised empire, Asoka proved fully worthy of his heritage; he was a man of unbounded energy and he gave himself without stint to the tasks of perfecting the administration of his empire and ensuring the happiness of his subjects. The range of his sympathies was wide, and he was by no means unwilling to adapt foreign models of administration and art to the growing needs and tastes of his country.

c11.s20.46

His own inscriptions clearly reveal to us the chief stages in the history of his reign and the motives underlying his activities. For more than a century these famous records have been the subject of patient and critical study by generations of scholars, and as a result of this study there has emerged a general agreement about the meaning of all inscriptions, though a few expressions still remain obscure. But the inscriptions are by no means evenly spread over the reign; most of them fall into two large groups, one about the thirteenth and fourteenth years after the king's coronation, and another in the twenty-seventh and twenty-eighth years. Though they do mention a few events with the time of their occurrence, they are far from being a full record of the reign. In this respect they constitute a complete contrast to the difficult Hathigumpha inscription of Kharavela, and the much later prasastis of the various mediaeval dynasties.

Linked texts (13)