Kautilya's Maxims
Argument for the ReconstructionThis is a consecutive overview of the considerations which lead us to conclude that the Kautilya maxims within Arthashâstra (ArS) may be earlier than the admittedly late date of most of that text, and, more specifically, may come from a source close to the historical Kautilya. In effect, we argue for a several-layered ArS, in which the Kautilya maxims occur only in the earliest layer. The argument is long. It proceeds backwards, from the latest to the earliest material. Some points are further investigated on associated pages, which are linked below as they occur in the argument. Here are the subdivisions of the main argument:
We should note at the outset that the ArS is divided into 15 "books," each in one or more chapters, and those in turn into 180 sections (150 sections, if continuations are counted as one). The chief fact on which the present argument is based is that the Kautilya maxims do not occur evenly distributed within ArS, but are confined to certain books and chapters. It can then be verified that the linguistic features which have often been pointed out as signs of late date for the ArS as a whole are clustered in the ArS books and chapters which do not contain Kautilya maxims. The implication is that the Kautilya maxims comprise a linguistically early stratum within ArS, and that the rest of ArS represents an expansion (perhaps in several installments) beyond that original stratum.
It will be easiest to follow the process in reverse, starting with certain elements which have been widely recognized as either late in content or extraneous in form.
Kautilya. The next to last line of the text reads "Seeing the manifold errors of the writers of commentaries on scientific treatises, Vishnugupta himself composed the sutra and the bhasya." Kautilya's personal name Vishnugupta is mentioned nowhere else in the work; indeed, the supposed author is called Kautilya in the Book 15 ending formula directly preceding this line, and in the whole-text ending formula directly following that. Kangle 2/516 calls the Vishnugupta attribution "clearly a later addition." It would seem to have been added after the work was otherwise formally complete. Its intent may have been to temper the then-negative implications of the name or epithet Kautilya (etymologically "The Devious One"), for a later age which held deviousness in less esteem. The variant name Kautalya, preferred by some scholars, but attested only later, may be a similar attempt by an interested posterity to mitigate a previous reputation,. Kangle (3/109-113) argues that the Kautilya form is earlier attested, in the texts, in the commentaries, and in the echoes in other works. We find that argument entirely sufficient.
At the outer edges of the text are two framing elements. The first section of Book 1, exceptionally, is unnumbered. It summarizes the 180 following sections of ArS. It can only have been written when the work had reached that total size, and when its sections had been numbered from 1 to 180. The framing statement respects this number 180 by not renumbering itself as section 1.
The last of those 180 sections (a single section, which makes up all of Book 15) is the other framing item. It lists the rhetorical devices used in the preceding 179 sections. This too is outside the work as such, though within its numbering system.
These two lists could have been added at any time after the completion of the ArS proper. The unnumbered introductory overview is obviously the later of the two, and section 180 itself need be only slightly earlier. In what follows, we will ignore both, and consider that ArS, for our purposes, consists of 14 books divided into a total of 179 sections.
ArS is largely in prose, but includes about 380 slokas or verse passages (Kangle 2/5 n18). These are not randomly placed. They occur at the end of every book and every chapter within each book; some also interrupt the prose sections. The chapter-final verses are sometimes summary in character, but sometimes they simply continue the preceding exposition. The interruptive verses typically elaborate the preceding lines. Some of them begin with transition words like "Therefore" (1/15:17). Literarily, they cannot stand without the prose text which they interrupt or conclude, whereas the prose text continues to make sense if these non-final verses are removed. The verses are thus clearly later than the prose text. Olivelle Dharma xxv notes that slokas began to replace the earlier sutra style in the administrative writings "around the beginning of the common era." The presumption is that, to update ArS stylistically, the ends of all its chapters were at some point rewritten as slokas, and some explanatory slokas were also added within the text. This might have happened as early as c100. In visualizing the text as it would have appeared at an earlier period, we should then consider the interruptive verses as not being present, and regard the chapter-final verses as having been originally written in prose. On this assumption, the content of the chapter-final slokas remains part of the text, whereas the content of the interruptive slokas is eliminated from further consideration. This affects our inventory of presumptively original Kautilya maxims.
As may be seen in the Inventory, the Kautilya maxims are not distributed evenly over the fifteen books (or as we have concluded above, actually fourteen books), as we would expect them to be if they were a mere authenticating gesture used by the author of a late work to connect it with a respected earlier name. In that case, we would expect the sayings to be as well distributed as the chapter-final attribution statements mentioned above. Instead, the maxims cluster. Most of them occur in only five books (ArS 1, 3, 7, 8, 9). The few which occur elsewhere can easily be related to the themes of those five groups. For example, 5/6:23 and 5/6:32, on the subject of the transition to the heir apparent, might be thought to belong thematically with the comments on preparing the heir apparent in 1/17:22 and 1/17:30. Book 5 contains no other Kautilya maxims than these two. In fact, if we mentally relocate these two maxims to Book 1, thus placing them with the other ArS maxims on the same subject, Books 4-6 include no Kautilya citations at all. Further, the nine books with few or no Kautilîya maxims (ArS 2, 4-6, and 10-14, and of course 15) do not cite any other authorities either (a reference to the Puranas in 3:7 is part of a passage which has other difficulties as well; Kangle 2/215 calls it "a late marginal comment that has got into the text"). They read like contemporary descriptions, not like authority statements. This is a different strategy altogether.
There is a further point. In general, the topics of the ten ArS books which contain few or no Kautilya citations imply a later stage of political evolution than the other five. Among other things, they contain minute specifications for the operation of a large state bureaucracy (ArS 2), they describe elaborate police mechanisms (ArS 4), they expound the highly ramified system of secret agents for which the ArS is notorious among modern readers (ArS 5), and they expound the mandala or "circle of kings" theory (ArS 6), another favorite of later readers, but on its face a suspiciously schematic view of internal relations in a multi-state system. These books, then, may well be a later stratum than the books which include groups of Kautilya citations.
Additionally, Hartmut Scharfe's vocabulary arguments for a c0150 date for the entire ArS turn out to apply not to the whole work, but essentially to material in this apparently later stratum. Scharfe's c0150 date may be accepted as the earliest at which these late ArS extensions could have been compiled. Whether they were added at one time or over an extended period does not affect the Kautilya question, and can be left to be clarified by future research. We do confidently identify them as late material. We put them aside, and will not discuss them further.
Even in the ArS books where the Kautilya maxims cluster, they cluster unevenly within those books. In ArS 8, for instance, which as a whole abounds with Kautilya maxims, all such maxims are found in the first four chapters, with none in the fifth. ArS 8 in general deals with a series of problems and emergencies encountered by the state. Only its fifth and final chapter discusses military problems. The Kautilya maxims themselves consider political and diplomatic matters, but not specifically military ones. In the late Book 10, a more extended consideration of military matters, it is conspicuous that two ancient authorities are cited (in 10:6) on the makeup of the basic battle group, but without the usual final citation from Kautilya. It then seems likely that the last chapter in ArS 8 and other such chapters in the generally early books have been added to extend the range of those books into areas beyond the specific "Kautilya" prototype. Since the addition of chapters to existing books is a different and less drastic structural device than the creation of whole new books, the chapters without citations (within books which do contain citations) may be assumed to define a middle layer. That layer would consist of the last chapter of Book 8 and similar material: the first thematic ventures beyond the limits of the early ArS, and of Kautilya himself. The addition of whole books on the same new topics would be a second and later strategy, and the military Book 10 would belong to that later layer.
Removing from consideration, as irrelevant to our present purpose, both the middle and later material as defined above, it follows that the early layer of ArS must then comprise approximately the chapters containing significant numbers of Kautilya citations. But even these chapters do not wholly consist of those citations; they normally continue with statements further developing the position attributed to Kautilya, or else they precede the Kautilya citations by other material on the same general topic. These more developed statements are presumably the original voice of the Arthashâstra compiler, as distinct from the quoted voice of Kautilya. It would then be a relevant and interesting task to define exactly how far the original Arthashâstra theorist (apart from the later writer who expanded the ArS far beyond the five chapters which were probably its original compass) had progressed beyond the position defined by the Kautilya citations.
Law. This early but non-Kautilyan material should thus probably be thought of as added to the core of Kautilya sayings on a given subject. In some chapters, that expansion process can be partly tracked, and a relative dating can be suggested. See for example the separate Law page, which discusses in some detail the expansion process as it can be seen exemplified in the early part of ArS 3.
We come at last to the Kautilya citations themselves. These are almost always given in response to one or more opinions by other and presumably earlier authorities. They often supply a corrective or revisionist "last word" to them. Some of the non-Kautilya names cited are identified with known schools of interpretation, though there are difficulties in matching their content to that of the works which supposedly preserve the traditions of some of those schools. Is it possible that the Early Arthashâstra author has quoted the opinions of those schools in order to refute them by his own position, which he speciously presents under the respected name of Kautilya? In that case, the motive for the ArS would be controversialist, and its position, and that of the cited "Kautilya" which validates it, might relate to a later century, and thus be irrelevant to Kautilya in the late 04c.
Manu. The easiest test of this possibility is with the ArS citations of "Followers of Manu." There was a later school of legal interpretation identified with Manu; its text is the Laws of Manu (Manu Smrti). This is a wholly versified text (in sloka rather than sutra form), and thus, by Olivelle's canon, cannot be from the BC centuries. Do the Manu citations in ArS define a position which is recognizably that of the Manu Smrti? There are points of similarity, and thus possible continuity, but on the whole they do not appear to define the same position. Kangle notes (2/6, 3/80) that the "Manu" citations in ArS largely diverge from the views of the Manu Smrti. The same seems to be generally true of other ArS quotes that can be compared with later school doctrines associated with those names. Then the evident controversial intent of the earliest ArS cannot as a whole be referred to the substantially later period from which the Manu Smrti and other extant representatives of those schools must (for other reasons) be dated. It is more plausible to conclude that the Kautilya quotes are arguing with predecessors of the known schools, and of course also with some other viewpoints of that time which happen not to have later successors.
We conclude that the later schools identified with some of the Kautilya opponents had diverged, at some points substantially, from the ideas of their founders, and that the early ArS is reflecting them at an earlier stage in their development. It remains also possible that some of these schools were not really in lineal descent from their claimed founder, but were borrowing an early name to authenticate later doctrines. Into these complications we have not ventured. We have left them for future research by those properly equipped.
What, then, is the nature of the authority citations as a whole? We may dismiss the possibility that they are themselves a text, accumulated in a line leading from the earliest cited authority to Kautilya (the latest), and representing doctrinal evolution in that series of thinkers within one tradition which was simply codified or continued by the earliest ArS writer. For one thing, there is no one succession of such thinkers in the ArS citations (Bharadvaja is the earliest name in one sequence, and the followers of Manu stand at the head of another), and many Kautilya positions are given in opposition to a more vaguely attributed opinion of "the teachers." Thus, several different traditions, some of them not very well defined except as traditional wisdom, would seem to be represented. For another, some of those citations are more than a little implausible on their face, and may have been distorted or even concocted for the purposes of the ArS core compiler. The non-Kautilya positions as a whole have the appearance of being assembled for Kautilya to refute, or, more rarely, to homogenize or approve.
Unanswered Sayings. But we cannot dismiss all of them as invented, since some non-Kautilya maxims are quoted without a concluding Kautilya wrap-up, apparently for their own value as early opinions. In the absence of a Kautilya opinion, we must assume that they were themselves authoritative for the Early ArS compiler. Two such maxims are the statements of Usanas (10/6:1) and Brhaspati (10/6:2) on the standard battle array of the chariot army of the time. These disagree with each other, but they are not followed by a capping or harmonizing comment from Kautilya. Kautilya in general has nothing to say about battle (as distinct from war), and ArS 10, which is concerned with battle, apparently respects this fact by citing other authorities instead. Here, at least, the other authorities are not straw men set up for Kautilya to improve on.
Dialogue. A third point is that the successive authority positions, as expressed in ArS, convey an effect of dialogue among the authorities, which is unlikely if those positions were quoted from the separately preserved maxims of their respective schools. It is more likely that they were rephrased by whoever assembled them. Here again, but on different grounds, we reach the suspicion that the group of citations did not exist as a text prior to its use by the early ArS. It then seems likely that the early ArS compiler himself did the assembling and rephrasing of these pre-Kautilya maxims, and structured them so as to maximize the force of the Kautilya maxims themselves.
There seems to be no ground in the ArS material, or in traditions reported elsewhere, to assume that Kautilya at any point wrote down his own maxims, or founded his own school of law or statecraft. Even in the ArS 15 "Vishnugupta" addendum, he is remembered as a doer rather than a teacher. At the same time, there is no doubt that he is the focal point of the ArS assemblage of citations, and that the core ArS text continues past Kautilya's contribution in what seems to be the same spirit. The simplest conjecture that will cover this situation is that the ArS core author was a political theorist of the early Maurya period, acquainted with the principles of Kautilya, and concerned to formulate them in text form, and that he may have done so shortly after the death of Kautilîya, in order to fill the gap left by the fact that Kautilya did not write his own text, or found his own school. He set Kautilya's positions in the context of previous opinions, gathered from the whole range then available, and connected and expounded them. The core ArS, on this understanding, is not exactly a text of the Kautilya school, there being no Kautilya school in the strict sense, but a functional substitute for it: a systematic and eponymous attempt to preserve Kautilya's statecraft as subsuming and replacing earlier contributions to that science.
If the core ArS, though not by Kautilya, is a Kautilyan text, as this suggestion supposes, one detail of the final ArS becomes perhaps more intelligible. This is the "mandala" theory developed in ArS 6 (for a diagram, see Spellman 157), of the pattern of enmities and allegiances among the successively more distant kingdoms surrounding the one with which the text is concerned. Such a multi-state system would have been eliminated by the Maurya unification, and the mandala theory would thus have been obsolete, or at any rate less compelling, for any writer after Kautilya (Spellman, noting this difficulty, dates the mandala theory to c0500). That a late book of the ArS develops the diplomacy of Kautilya in that direction shows, in our opinion, that Kautilya's own world continued to be "classical" for the author of that book, just as the Han-dynasty (02c) chapters of the Gwandz still speak in terms of the multi-state system which was the context of that work's early layers, but which the Chin unification of 0221 had eliminated as a political fact. The late ArS is then not exclusively based on the world of its own time; it is also concerned to elaborate in theory the world still embodied in the perceptions of its eponymous figure, Kautilya. Keeping Kautilya's heritage alive will then have ranked with keeping it current, as a motive for extending the text. This suggests that the Arthashâstra, over the period during which it grew as a text, regarded itself as the definitive repository of Kautilyan statecraft.
Pending further results from study of the text and archaeological evidence, we feel that it is plausible on present evidence to conclude that the Kautilya maxims of the ArS represent him and his predecessors, not indeed in their own words, but in a formulation that may be close to them in substance.
Kautilya's Maxims is Copyright © 2001- by E Bruce and A Taeko Brooks
Contact The Project / Exit to Kautilya Page
