Chapter 1 - Introduction

I INTRODUCTION The present thesis is an attempt at Mahabharata criticism from the viewpoint of an archaeologist, and hence the title "Archaeology and the Mahabharata". The looseness and the vagueness of the term, however, may give rise to doubts and misgivings about the contents and the approach of the work. It will, therefore, be our endeavour here, to enunciate at some length, the choice of this particular subject, the need for such a critical study based on archaeology, the guiding principles that have shaped the work, and the actual method adopted. The Mahabharata along with the Ramayana is one of the two great national epics of India. It is a colossal 1. Throughout this work 'Mahabharata' is used as an abbreviation of the full term 'Mahabharata'.

work totalling to a stupendous figure of a hundred- -thousand verses, presenting within its vast fabric the most diverse literary material of great cultural and historical value. The central theme is a bloody fratricidal war around which is woven an intricate web of the most fascinating stories, myths, legends, fables and parables, which have become an inseparable part of India's national consciousness, their spellbinding magic as alive today as it was over two-thousand years ago. The Mahabharata is, in fact, not simply an epic narrative, but a whole literature, representative of the wisdom, culture and traditions of an ancient race. The great value of this thesaurus for the historian can never be overestimated. Whatever his special field of investi- -gation, he will always have occasion to turn, again and again, to this "inexhaustible mine of culture"1. 1. These and other quotes are from Dr. V. S. Sukt hankar's Prolegomena (1933) to the Adiparva. The entire discussion on the composition of the Critical Edition that occurs here, is also primarily based on the same prolegomena. 2

= 3 But there is always another side to a coin, and so it happened, that it was this very encyclopaedic nature of the Epic which militiated against its historical value. It is common knowledge and an easy inference that the unweildly bulk of the Mahabharata could not have come from the imagination of any one person. It is clearly a product of generations of literary, ne intellectual and cultural activity. There was much in it that was genuid, but much also that was spurious, repetative and worthless. Its text had been in a fluid state for centuries, and was tampered with, on and again, not only by posts of genius, but also by very ordinary scribes. The memory of this fluidity of the original is preserved in the Epic to this day. We are told that Vyasa taught his 'Bharata' to his five pupils - Sumantu, Jaimini, Paila, Suka and Vaisampayana and these five rhapsodists, direct pupils of the author, published in their turn, five separate versions of the Epic (1.57.74). Vaisampayana's version was recited by him in the form of a free recitation before Arjuna's great-grandson Janamejaya and the assembled guests at an elaborate sacrificial session. It was recited again, a second time, before Saunaka and the other inhabitants of the Naimisa forest by Sauti, who had herd it only at the first recitation and somehow committed the whole poem to memory, certainly not 'verbatim et literatim'

- 4 as Dr. Sukthankar points out. It is thus obvious that the process of transmission in the beginning was by word of mouth and went somewhat like this first memorising the poem and then reciting it freely, but as faithfully as possible. Slowly in the course of time, as the Epic was being handed down from generation to generation, from place to place, and from bard to bard, its wordings and its contents started swelling and varying, till such a time in its chequered history, when it came to be put down in writing. But unfortunatly the process of additions and alterations did not end, as one would rightly expect it to, despite this "scriptal fixation". It continued unabated and unchecked, and there is no better proof of this than the large number of varying Mss which represent a very late phase in the evolution of the text. The inflated bulk of the Epic as preserved in these Mss was. thus far from belonging to any one age or culture but became a confused conglomerate of several ages and cultures. The original nucleus, whatever it be, had been gradually swollen over the centuries, by continuous additions, to such an extent, that no historian could approach this rich treasure- -house of culture without being assailed by doubts about the dubious nature of the material and its loose chronological set-up. The utter desparation of a scientifically inclined mind, when confronted with this

- 5 literary giant, is best summed up in the words of the noted Indological Oldenberg when he described the Mahabharata as "a monstrous chaos" The problem faced by the historian was how to evolve an order out of this chaos. Under these circumstances the need for a critical assessment of the Mahabharata material was naturally acutely felt. However, any such attempt at throwing out the spurious and the redundant had to be first made in the realm of literary criticism. Efforts were made in edition the direction by many a renowned Western scholars, but the gigantic enterprise was finally taken up and complet- -ed, after half-a-century's hard work, by the Bhandarkar Oriental Research Institute in Pune, during the years 1919 to 1970. The Critical Edition, as the BORI came to be known, was very breifly "a corrective edition" of the Mahabharata, based on the oldest Mss tradition, after serutiniz- -ing all the available Mss at that given time, from public as well as private collections, within and outside the country. The scientific method followed by its brilliant editors left nothing to chance, not even a single word or phrase, when they carefully and painstakingly checked and re-checked mud the numerous Mss. In the words of the General Editor, Dr. Sukthankar, the aim was to produce "a critical edition of the Mahabharata in the preparation of which all important versions of the Great Spic shall have been taken into consideration, and all important Mss collated,

estimated and turned to account". Towards this immense but worthly task were collected, as a first step, all the Mss that could be traced in catalogues of public and private libraries, as well as in private collections accessible to the editor at the time. They were then classified broadly into recensions by the script in which they were written. Corresponding to the two main types of Indian scripts, Northen and Southern, there were two main recensions of the Epic, each of which was again divided into a number of sub-recensions called "versions" by the editor, corresponding to different provincial scripts in which these texts were written. The final classification that emerged was as follows Vyasa's original 'Bharata' 1 Ur-Mahabharata' "that ideal but impossible desideratum" which at present looks like being lost forever. N B S T KN V BD ? M . 6

N: is the ultimate source from which all versions of the Northen recension are, directly or indirectly derived. u: is the lo is the lost archetype of the North-Western group, appreciably shorter than any of the other known versions (Texsus simplicior). $: is the Sarada or the Kashmiri version. K: is the specific Devanagari version allied to the Sarada or the Kashmiri version, but sharply distinguished from other Devanagari versions. however, largely contaminated. - 7 It is r: is the intermediate source between the North-Western and the Southern groups, from which all versions of the Central sub-recension are derived. E: is the archetype of the Eastern group which is free from additions and alterations made later in certain Devanagari Mss. N: is the Nepali version of the eastern group. V: is the Maithili version of the same group, from Bihar. B: is the Bengali version of the same group. D: is represented by Devanagari versions other than those of the Kashmir group. There are sub-version like :Da: with the commentary of Arjunamisra Dr: with the commentary of Ratnagarbha Dn: with the commentary of Nilakantha, which has POOHA 5 LIBRARY

been accepted as the 'vulgate' text, the most popular but also the most contaminated and inflated version. S: is the ultimate source from which all versions of the Southern recension are, directly or indirectly derived, and which is appreciably larger than the Northen, and far more elaborate and ornate (texus ornatior). A: is the lost archetype of the two Southern versions like the :T: the Telugu version from the Andhra country . 8 and G: the Grantha version from the Tamil country. M: It contains a large number of curruptions and secondary additions from which the Malyalam version from Kerala is remarkably free. be This was considered, by the editor, to the best classification possible of the critical apparatus, neither totally arbitrary nor absolutely perfect! Underlying the superficial differences of scripts were deep textual differences, since scribes as a rule were familiar only with their own provincial seript, unable to copy any other, while the cheif medium of contamination between different recensions and versions was the Devanagari script "which was sort of a vulgar script, widely used and understood CHAR POORA & LIBRARY

- 9 in India". The classification over, the next task was to find a unifying thread through this "luxuriant growth and indiscriminate fusion of versions", each version consisting of a large number of Mss. The study of the Mss revealed that theirs was an extremely complex inter-relationship, with a long, intricate and haphazard process of synthesis lying behind each version of the extant Epic text. But notwithstanding this, it was as clear as daylight, to anyone who wished to see, that behind this bewildering profusion was a family heritage, shared and nurtured in common, however distant and nebulous it might seem to us today. To trace this common heritage backwards, was ultimately the problem of unifying the Hss tradition, as far as it was possible, by grasping whatever loose threads that were available in this otherOne such loose but sure - wise enormous entangled mass. end was provided by the Sarada (Kashmir) version. It is the shortest known version, but not an abridged or truncated one, for despite its smaller size there is relatively little in it which is not found, at the same time, in all other versions of both the Northen and the Southern recensions. It is thus, not just the shortest and the most compact of the versions, but also the least contaminated one, the main reason why it was chosen as the norm for the Gritical Edition. At the other end of

- 10 the Sarade version was the 'vulgate' of Nilakantha, probably the most popular, but also the most contaminated edition in Devanagari. The text used or prepared by Nilakantha is smoothed and eclectic, but certainly inferior, for he followed a typical open-door policy of include anything and everything and exclude nothing. The golden means between these two extremities lay through a maize of versions in Nepali, Maithili, Bengali and Devanagari, from all over Eastern and Korthen India, and in Telugu, Grantha and Malayalam script from all over the South. Of these the Bengali version, though closely allied to the Vulgate, is considered far superior to it, being free from a large number of later accretions. The Nepali and the Maithili versions are, more or less, on the Bengali pattern, while a large number of the Devanagari versions are "simply misch- -codices of small trustworthiness and no special value for critical purpose". The Southern versions are not only very much longer than the Northen, but much more richer in details, leaving little or nothing to the imagination of the reader or hearer. Their precise, schematized and practicle approach is in sharp contrast to the vague, unsystematic and somewhat naive approach at story-telling by the Northeners. Their exuberant and ornate details appealed insinctively to the Indian mind and hence Nilakantha's vulgate borrows heavily

- 11 from the Southern elements. These diverse trends were finally united in the form of the Critical Edition by following a rigorous and comparative method which involved the examination evaluation of as many MSS as possible, their groups, families and traditions. In the words of Dr. Sukthankar "originality and authenticity are, unfortunately not the prerogative of any single recension or version or Mss. They must be established laboriously, chapter by chapter, line by line, word by word, syllable by syllable". The three guiding principles that were followed were (1) originality of agreement between independant versions, (11) concurrence of the largest number of independant versions, (111) and, in case of controversy, the acceptance of that reading which best explains how the other readings may have arisen. The new edition of the BORI was thus, undoubtedly, a great step forward towards a purer and a more cohensive edition of the bh, leaving far behind the confusing and frighetening mess of innumerable versions and MSS, their contradictions, spurious additions and copying errors. The value of this edition is further enhanced by the systematic and scholarly presentation of the entire "apparatus criticus", the variants by way of foot-notes to the text, and the longer passages as appendices at the end of each Parva. The reader, thus, has in front of him, as never before, the critical evidence of all possible MSS sources, prepared with PGORA 6. LIBRARY

- 12 meticulous care and conveniently presented. He can form his own opinions with the help of this readymade data in case he happens to disagree with the editor. The composition of this edition was a landmark The in Mahabharata criticism and Epic studies in general. historian was now treading on much surer grounds than before. At his disposal was an Epic text which was the nearest possible to the original based on sound scientific principles of literary criticism. The importance of the Critical Edition for any historical research is invaluable. Therefore, in choosing the Critical Edition as the basic text for our present study in preference to the numerous editions of the Nbh that exist in print, we are acknowledging our great debt to the editors of this masterly piece of work. We have deliberately dwelt at some length on the preparation of the Critical Edition, not simply to justify our choice of this particular edition, but more so, to give the correct background to the present attempt at Mahabharata criticism. A question may naturally arise as to the purpose, necessity, properity and usefullness of further criticism in the same field after so masterly an attempt. In answer, one can, in all humbleness assert, that the Critical Edition can not be and is not the last word in Mahabharata criticism. It is but the first concrete step which has paved the way for further critical studies. No less a person than Dr. Sukthankar himself has taken some pains * POCHA 6 LIBRARY

- 13 It to point out the limitations inherent in a literary criticism of such magnitude and complexity. He has made it clear in no uncertain terms that the constituted text of the Critical Edition is no autograph copy of Vyasa, nor a reconstruction of the Ur-bh, nor even a replica of the poem recited by either Vaisampayana or Sauti. is, in fact, doubtful if it can come anywhere near even to the Satasahasri Samhita mentioned in the land-grands of the 5-6 th century A.D., the earliest Mss utilized being a 14 th century A.D. Nepali Mss. All other HSS, notwithstanding the Sarada, are from between the 15 th end the 18 th century A.D. Apart from this obvious late- -ness of the Mss tradition on which the edition is entirely based, it also suffers from some of those very drawbacks which are a hallmark of the other editions. It can not be dated and labelled as belonging to a particular place, personality or age, and it is very much "a mosaic of old and new as any other edition of the Epic". Its shaky position is brought out by Dr. Sukthankar without mincing any words. He says: average adhyaya of this edition (as of any other edition) we may read a stanza of the 2 nd century B.C., followed by one written in the 2 nd century A.D. Sometimes the gap will occur in the middle of a line, precisely as in every other edition. This uneveness and these inequalities are inevitable, conditioned as they are by the very nature of the text and the tradition". There are thus "In an

= 14 many an element in the constituted text of the Critical Edition which a scientific historian or even a lay reader might find difficult to digest as genuine or original, on the evidence of cultural, social or religious history, or simply their own intrinsic value. But they have to be left untouched on purely literary grounds since they are found in all versions and Mss. Nobody need doubt their late interpolation in the Epic text, but this unfortunately took place at a date to which the Mss tradition does not hark back. Therefore as pointed out by Winternitz "the elimination of such passages is not the business of an editor, but must be left to that critical study of the Epic, of which the Critical Edition is only the beginning and the only safe basis" (1934). The present thesis makes another attempt, a a a very tame one compared to the first, at a critical appraisal of the Mahabharata material, to extricate it from an overgrowth of additions and interpolations and sift it into meaningful chronological brackets. By taking the first bold step the Critical Edition has opened up vast new avenues, so far unexplored, to scholars from other branches of learning. This is where archeology can step in. The archeological evidence excavated from various sites all over the country will be of immense value to put the cultural matter in the Epic in its proper chronological perspective. Now, for the first time, traditional literary evidence can be checked and reschecked, corroborated and contradicted,

with a more objective standard of measure, with well- -dated and well-stratified archeological finds. a As a first step, therefore, the evidence from all important excavated sites was sought and collected. In dealing with a loosely dated, multi-faceted culture complex like the Mahabharata, this evidence had to be utilised in its strict time and space context, so as to make the epic data a little more meaningful historically. The archeological evidence utilized falls into three or a four cultural phases, following one after the other in chronological sequence, but not necessarily with direct evolutionary links. I. The Indus Valley Civilization (2500-1700 B.C.) with important settlements in Sind, Punjab, Rajasthan, Kutch, Saurashtra. C II. Chaleolithic settlements (1600-1200 B.C.), mostly of Central India and the Deccan. III. Early Iron Age settlements, particularly those of the Gangetic region, marked by the finds of the PG sherds (1100-600 B.C.) Some of the key sites are mentioned in the Mahabharata IV. Early Historical sites, particularly of Northen and Central India, characterized by NBP sherds and Mauryan finds (600-200 B.C.). Many of these sites too are mentioned in the Mahabharata - 15 B. C R POORA 6 LIBRARY

- 16 a V. The archeological evidence is at times, further stretched, to Sunga-kugana period, to Indo-Greek, Saka and Parthian settlements (200 B.G. - 300 A.D.), if it is found relevant. Theoretically, however, the 4 th-5 th century A.D. is accepted as the lowest chronological limit in order to accomodate the latest additions and interpolations, which according to literary and inscriptional evidence, do not cross the mark of 600 A.D., by which time the Mahabharata had, more or less, come to acquire its present prestigious form (Winternitz 1972:463). If any further additions were made after 600 A.D., and they probably were, they were minor and inconsequential. However, the evidence at our disposal hardly needs to be stretched beyond 300 A.D. as will be demonstra- -ted in the final concluding chapter. a The reliance on archeological evidence clamps an automatic, but not an altogether unexpected restriction on the scope of the work. Since excavated evidence pertains mainly to material aspects of life, to concrete objects and finds, the present study too, restricts itself, more or less, to only those aspects of material and cultural life of the Mahabharata which can count upon a archeological data for comparison. This approach necessarily leads to the exclusion of many an interesting aspects of social, religious and political history, with the consequent danger that a lop-sided appraisal of the POONA 6 LIBRARY

- 17 Epic data might be the result. But at the same time, it is hoped that their deliberate exclusion will make the work more compact, more cohesive and very likely more readable. a Archeological evidence, however, has its own obvious limitations which can be ignored only at the cost of imperilling the authenticity of the study. First and foremost, the sites taken into consideration are not always fully excavated. Oftener than not it is only a small area which is actually dug, either due to inadequate finances or lack of time or sometimes simply due to an ancient site being superimposed by a highly-populated modern settlement. An excavated area, therefore, can never claim to be fully representative of all the sections of an ancient inhabitation. The second hurdle is often the non-availability of detailed reports. One has to mostly depend on sketchy or at best little more than sketchy reviews, not yet worked out systematically, which appear in journals and periodicals. Detailed and systematic reports are available only in a few cases. often results in inadequate and imprecise descriptions of finds. The greatest cause for disillusionment, however, is the realization that not everything is preserved in these ancient remains. The decaying agencies of Nature and the perishable nature of many materials have conspired to remove all traces of many an objects, while the existence of many others can only be infered and conjectured. This

- 18 a Archeological discoveries being mainly a matter of chance, there is all likelihood that somethings have been over- -looked while others still lie buried inside the earth, waiting for the spade to bring them to light at some later date. Thus while working within the framework of a archeology, one has to constantly remind oneself that the picture inside the frame is not yet complete, lacking in details, waiting for the final touch-ups, still a bit to 'fresh', if we may put it so. Whenever such contingencies arise it is a dengerous to take archeological evidence at face-value and hold it up against literary descriptions, for it will amount to gross distortion of facts, a misinterpretation of history, a concious attempt at misleading. To avoid these pitfalls in the present work, aid has been freely sought from allied fields like epigraphy, numismatics and sculpture, where too, the evidence furnished is as much precisely dated as actual excavated finds. In fact, sculptural depictions and those on the coins often provide the graphic touch to the somewhat imaginary literary descriptions. However, very often it so happens that literary descriptions are borne out the best only by other literary descriptions. Some of these literary texts are dated, even if somewhat loosely, and therefore, we have not hesitated to use their evidence too, in the same spirit as that of inscriptions, coins and sculptured

- 19 monuments. In so doing the overriding desire was to avoid drawing foot-loose conclusions, although on the other hand, there was also the danger of tuning the work into a mere literary criticism. The literary sources whose evidence is constantly utilized are the Pali Cononical texts, the Astadhyayi of Panini and the ArthaSastra of Kautilya. Notwithstanding the fact that these literary records have diverses geographical background and diverse fields of emphasis, their evidence dating mostly from the pre-Christian era, but not generally going beyond the 5 th century B.C., has proved to be of great importance. The Vedic literature too is often utilized as a backdrop to these developments. d Thus an all-round effort is being made to capture the Epic evidence in a well-set chronological framework, with the aid of all available evidence, but relying mainly of archeological finds wherever they are present. The tighter and more compact this famework becomes, it will be able to set forth the Epic material, clearly and precisely, as never before, in its proper historical context. Once this is done, with even a small measure of success, a a historian or an archeologist approaching the libh, need not be dismayed by the prospect of groping his way through dark corridors, for he will be treading on much firmer grounds than before. The need for striking these fresher and firmer grounds will be readily acknowledged when one realises that inspite of the extensive digging work of the

lest 25 years and more, crucial gaps still exist in our J - 20 knowledge of the archeological map of India. Literary sources still remain out best guides and our major sources of information as regards ancient Indian life. No doubt, invaluable light has been thrown on the so called "dark ages", from the end of the Indus civilization to the beginnings of the Historical Period, by the disco- -veries of numerous Chalcolithic and Iron Age settlements all over India. Yet anyone can hardly claim that the picture is absolutely clear. The horizons are still shrouded in mystery. This is the very same period which is believed to have witnessed the triumphant entry of wave after wave of immigrating Aryans, into the Punjab, later spilling over into the Gangetic doab - Aryans (whoever they be), whose foot-prints have as yet not been clearly traced on the sifting sands of Indian history. The primary sources of our knowledge for this period are still the Vedic literature and the Epic songs. Fortunately, the Vedas have been preserved in their pristine purity. Not so the Epic texts. Therefore, it is all the more important that every attempt at a critical appraisal of the Epic material, from whatever viewpoint, archeological or otherwise, should be made till the 'lost horizons' have been cleared once and for all. a But let us also make it explicit, here and now, that this 'clearing' process does not in any way mean a 'cleansing' process. No attempt is made whatsoever, in

- 21 the present thesis, to throw out any part of the constituted text of the Critical Edition on the grounds of its not being genuine, or spurious and later. There has been no preconcieved notion at all about any fixed date for the Mahabharata All that is sought to be done is to arrange the rich cultural data into meaningful chronological brakets, thereby indicating the likely period of time when these elements might have been incorporated into the Epic. This flexible bracketing may also help, to some extent, in understanding how and when the epic text was being expanded, and the probable motivating factors and trends which led to such expansionist activities. Beyond this the scope of the work does not extend. We are content to function within this limitation. The thesis is no textual criticism, and hence the text remains intact. For the same reason we have chosen to ignore almost completely the variants and appendices, and instead put our faith in the Critical Edition as it is, as authentic and pure enough for our purpose. It is a measure of Dr. Sukthankar's "scrupulous consciousness" as Winternitz (1934) has put it, that he expects the reader to improve upon his text with whatever variants he chooses to or to question his choice in relegating certain passages to the appendices. But with the high standard of workman- -ship set by him, it is doubtful if anyone will be so eager, after all, to come forward and question. We, for our part, have chosen not to. It would have been an uncalled for FOONA 6 LIBRARY

- 22 widening of the limited scope of our study. Ours is finally a problem of synthesis, of putting together the evidence from various sources to a a or the case, For see, if it will, at any stage, coincide in time. this purpose a simplistic method of presenting the facts is adopted. Each chapter opens with an exhaustive review of the Mahabharata material, followed by a brief outline of other close literary parallels and finally by a detailed statement of actual remains, if any, their location and find-spots, chronological and cultural strata. However, it is not just a simple statement of facts as is likely to appear. Definite inferences are drawn wherever the literary and the archeological evidence seems either to be mutually complimentary/to totally contradictory. But since this is not always pasale, so whenever the literary annals or the archeological records fall short of expectations, it is thought best only to record the facts faithfully. If the facts permit, then the chapter is concluded with a brief summary which tries to loosely arrange the material in some kind of homogeneous chronological groups. These various conclusions, appearing at the end of each individual chapter, are synchronized in the concluding chapter, which attempts to give a broad pattern of how and when the various elements came to be incorporated into the Epic text. The effort might seem a trifle too ambitious for our limited capacity, so that only the comming pages will demonstrate how far we have succeeded, if at all, in our endeavour.