Study Of Mahabharata

Author(s): J W De Jong
Publisher: J W De Jong
Catalog link: https://jainqq.org/explore/269713/1

Page #1
THE STUDY OF THE MAHABHARATA A brief survey (Part II) by J. W. de Jong As mentioned before, the mythical interpretation of the Mahabharata proposed by Alfred Ludwig in the last two decades of the nineteenth century did not meet with much approval. The theory of naturalistic mythology which had been applied by Max Muller and other scholars to the Veda did not seem to have any relation to the story told in the epic. Moreover, at the end of the nineteenth century and the beginning of the twentieth, the development of ethnological studies brought about the decline of the entire school of naturalistic mythology. In 1947 an entirely new mythological interpretation of the epic was suggested by a Swedish scholar, Stig Wikander, in an article which the following year was made more accessible in a French translation by Georges Dumezil. In the following years both Wikander and Dumezil published a number of articles in which they further developed their interpretation of the epic. In 1968 Georges Dumezil published a comprehensive study of the Mahabharata. Wikander pointed out that in the past no satisfactory explanation had been given for two fundamental features of the epic: the divine birth of the five Pandavas and their joint marriage with Draupadi. Scholars such as Winternitz considered the divine birth as a most fantastic story which could scarcely have belonged to the old poem (History of Indian Literature, vol. 1, pp. 329330). Even as recently as 1973, van Buitenen wrote: "There is no reason at all why Bhisma should be the son of the river goddess, why Karna should be the offspring of the Sun, why the Pandavas should have been begotten by various deities" (The Mahabharata, I, p. xix). As to the marriage of Draupadi with the five Pandava brothers, Lassen explained it allegorically. Ludwig accepted this explanation, while Dahlmann suggested a mythological explanation. Other scholars accepted it as a true fact, some of them explaining it as a non-Aryan custom to be found among hilltribes. However, Wikander has no difficulty in showing the unsatisfactoriness of the solutions proposed. According to him, the only interpretation possible is to be found in the theory of the tripartite ideology which Dumezil had developed since 1938 in a series of publications. This ideology which is peculiar to the ancient Indo-European peoples, is a system of three functions: sacred power, physical force and the pro

Page #2
Fa Hua Wen Hua Yan Jiu (Di 11Hao ) In India these three functions are exercised by the duction of material prosperity. three varna: brahman, ksatra and vis. However, Dumezil stresses the fact that this tripartite theology is not necessarily, as in India, reflected in a threefold division of the society, but can be only an ideal, and at the same time an instrument for the analysis and interpretation of the forces which rule the world and the life of mankind. Dumezil traces this ideology in the theology, the mythology, the rituals, the institutions and the literature of the Indo-European peoples. Wikander points out that the Pandavas fit in completely with this system. Yudhisthira represents the legitimate sovereign, Bhima and Arjuna the warriors (ksatriya), and Nakula and Sahadeva the mass of the people engaged in raising cattle and horses. Their divine fathers represent the three functions: Dharma the first, Vayu and Indra the second and the Asvins the third. Wikander explains that Bhima corresponds not to the Vedic Vayu but to the Indo-Iranian Vayu. This proves that the divine birth of Bhima and the other Pandavas is not the result of a secondary "mythisation" of heroes as maintained by historicists, as in that case we would expect to find their divine fathers to represent the epic pantheon in which Visnu and Siva are the most important gods. In the Rgveda the first function, the sacred power, is exercised by Mitra and Varuna who represent two different aspects of it. According to Wikander the god Dharma incarnates the Mitra aspect of the first function." This is clearly evident in the disguise adopted by Yudhisthira at the court of King Virata: at the second game of dice played by Yudhisthira it had been stipulated that the losers would live in the forest for twelve years and live in disguise in society for a further, thirteenth year. Yudhisthira chose to become a brahman experienced in the game of dice. Wikander sees in this disguise the manifestation of the passive attitude of the sacred King.. The marriage of the Pandavas with Draupadi reflects the association of a female deity with the male gods of the three functions in the tradition of the Indo-Iranians. This female deity is a goddess of fecundity who is associated with the gods of the third function, but this type of goddess can be combined in different ways with one or more of the male representatives of the three main functions. In his conclusion Wikander remarks that the fact that the Pandavas are not mentioned prior to the Mahabharata - a problem which has occupied many scholars - is not surprising. In the ritual texts the mythical system composed by the juxtaposition of Mitra-Varuna, Indra-Vayu and the Asvins is still valid, but in the Mahabharata we find an entirely different mythology. The Pandavas are epic survivals of the Vedic gods and as such could only emerge when the latter had lost their central importance in the religion. Wikander and Dumezil have further developed the ideas first outlined by Wikander 2

Page #3
THE STUDY OF THE MAHABHARATA in 1947. Both scholars underline the difference between the heroes representing the first two functions and the twins (Nakula and Sahadeva) representing the third function. Yudhisthira, Bhima and Arjuna are the sons of Kunti, who married Pandu in a svayamvara marriage. The twins are sons of Madri, whose brother Salya is an incarnation of an Asura (Samhrada or Samhlada). Pandu's alliance with Madri is a marriage by purchase, an asuric (asura) marriage. In the Veda and the ritual texts there is practically no difference between the two Asvins. However, the epic clearly differentiates between Nakula and Sahadeva. In this connection Dumezil remarks that some very ancient theological and mythological doctrines not found in the Vedic hymns must have been known by oral tradition to the authors of the Mahabharata. In other Indo-European mythologies we find twins with different specialities and different destinies. Dumezil concludes that the Vedic theology of the Asvins is the result of a reform aiming at identifying closely two divine beings between which social classes other than the priestly normally made a distinction. We see here an example of the method applied by Dumezil on several occasions. If a theological or mythical item which can be deduced from the epic material cannot be explained by a Vedic or Indo-Iranian precedent, he posits that it is legitimate to suppose that it had been purposively transformed by the authors of the Vedic hymns. This supposition is all the more probable when something corresponding can be found among other Indo-European peoples. 3 Wikander believed to have discovered, in the Iranian goddess Anahita, the divine prototype of Draupadi. Dumezil qualifies the former as a trivalent goddess who in some way assures the synthesis of the three functions. Her Indian equivalent is Sarasvati, who according to Dumezil is clearly trivalent. The authors of the Mahabharata have translated this conceptual relationship between a trivalent goddess and gods of the three functions into a marriage between Draupadi and the five Pandavas. We find different stories in the Mahabharata to justify this scandalous marriage. However, in the Mahabharata Draupadi is said to be an incarnation of the goddess Sri, the wife of Visnu. Dumezil discovers in Sri also a goddess who is the synthesis of the advantages which correspond to the three functions. It is obvious that Dumezil is not entirely satisfied with his own explanations with regard to the trivalent Indo-Iranian goddess (Anahita, Armaiti, Sarasvati) and the nature of Sri. He writes that the trivalent Indo-Iranian goddess seems to have had her foundation in the third function, but that the heroine who represents her as a result of the transposition, has a special affinity for the second function just as does the goddess (Indrani, Sri), her immediate prototype, in the "intermediary stage" of the operation.' Dumezil has not been able to solve this problem. When he remarks that the trans

Page #4
** (119) position effected by the ancient Indian scholars has its limits, and that it is a joy for the mind to see how an imaginative liberty as to the details is combined with care in preserving the main lines, one cannot suppress a certain feeling of unease that such an important and central theme of the Mahabharata as the polyandric marriage of Draupadi remains far from being explained, even to the satisfaction of Dumezil himself. Before her marriage with Pandu, Kunti evoked the god Surya with the help of a magical formula. Karna was born from her union with Surya. Kunti entrusted Karna in a basket to the waters and he was adopted by Radha and the charioteer Atiratha. Dumezil compares the two mothers of Karna to the two mothers of the sun in the Veda, the goddesses of the Night and the Dawn. According to a Vedic myth, Indra and Surya were enemies and Indra stole or destroyed one of the wheels of Surya's chariot. This same myth has been transposed by the authors of the Mahabharata. The wheel of Karna's chariot is first broken by Bhima, and in a later fight with Arjuna one wheel of his chariot sinks into the ground. Karna was born dressed in a coat of mail (sahajam kavacam, kavacam svanisargajam) and with earrings. Dumezil points out that the epic of the Ossetes speaks of a coat of mail of the sun. He believes that the Ossetes have retained a facet of the myth of the Sun which is not found in the Rgveda. Much more problematical is the relation of Pandu and his brothers Dhrtarastra and Vidura to their divine prototypes. Neither Pandu nor Dhstarastra are said to be sons or incarnations of a god, and Vidura, the son of Vyasa and a sudra woman, is the incarnation of Dharma, the father of Yudhisthira. Dumezil remarks that it is strange that no hero is the son or incarnation of the two great sovereign Vedic gods, Mitra and Varuna, or of the two minor gods Aryaman and Bhaga. He points out that in the pantheon of the Mahabharata Aryaman and Bhaga do not play any role, Mitra hardly any, and Varuna is nothing more than the god of the waters of the ocean. According to Dumezil, Pandu ought to have been the incarnation of the Vedic and pre-Vedic Varuna, and Dhstarastra and Vidura of Bhaga and Aryaman respectively. Their character and behaviour derive from these gods. In the case of Yudhisthira, Mitra was replaced by Dharma because by the time of the composition of the epic the word dharma expressed the substance of the Vedic and pre-Vedic Mitra. Dumezil makes every effort to prove his thesis but doubt remains. One of the most venerable persons in the epic is Bhisma, the great-uncle of the Pandavas. He is the incarnation of the god of the sky, Dyu, one of the eight Vasus. The Veda does not give any information on this god, to whom no hymn is addressed. However, Dumezil points out that the Scandinavian god Heimdallr explains many of

Page #5
THE STUDY OF THE MAHABHARATA 5 the strange aspects of Bhisma. He concludes that here again the authors of the Rgveda had omitted great parts of the traditional mythology which existed since the time of the Indo-Europeans. The fact that the archaic mythology of Dyu, which had been eliminated from the Rgveda, never reappears in any form in any period of the Indian mythology, proves that the transposition of gods into heroes, which forms the basis of the epic, must have taken place at a very early period when this mythology was still well known in certain circles. The battle between the Kurus and the Pandavas led to the almost complete destruction of the two warring parties. The tenth book describes how RudraMahadeva manifests himself to Asvatthaman (the son of Drona and incarnation of Siva-Rudra, Death, Anger and Desire), and then enters into his body. Asvatthaman penetrates into the camp of the Pandavas and massacres all the warriors with the exception of the five Pandavas themselves and Krsna, who are absent. Krsna assures the Pandavas that their lineage will be continued by the future son of Abhimanyu, the son of Arjuna and Krsna's sister. Asvatthaman kills the embryo, but Krsna promises the Pandavas that the embryo will be restored to life. Dumezil remarks that the massacre of the Pandava warriors by Asvatthaman is the epic transposition of an eschatological crisis, the end of a yuga (age of the world), in which Rudra (incarnated in Asvatthaman) and Visnu (incarnated in Krsna) confront each other. However, the doctrine of the ages of the world is not to be found in the Veda. * Dumezil draws attention to Scandinavian eschatology instead, and explains that the Scandinavian god Vidarr corresponds to Visnu. Dumezil concludes that the functional identity of these two saviour gods of the Indians and the Scandinavians seems to confirm the existence of a common eschatology which perhaps goes back to the IndoEuropeans. In his concluding chapter Dumezil recapitulates the characteristics of the epic mythology which are not found in the Vedic mythology and which testify to its antiquity. He supposes that the conflict between Kesna and Asvatthaman, i.e. between Visnu and Siva, is perhaps due to a recent Sivaite influence. In the past scholars tried to discover a main hero in the epic, for instance Arjuna. Dumezil points out that the solidarity of the five Pandavas excludes such an interpretation. As to the existence of a historical basis, he points out that the heroes of the epic find their explanation in the functional mythology he has demonstrated. In a note, he remarks that, if one wishes, it is possible to assume a historical event the memory of which would have been completely overlaid by non-historical matters. He adds that such an assumption would be undemonstrable and useless. Finally Dumezil calls attention to several problems which he has not been able to solve. For instance,

Page #6
Fa Hua Wen Hua Yan Jiu (Di 11Hao ) many names of heroes, such as Yudhisthira, Nakula, Sahadeva Bhisma, Drona, etc., cannot be explained. Another problem is the fact that many Vedic gods-the Rbhus, Tvastar, Parjanya, Pusan, Yama, Savitar, Soma, etc. are not represented. The same is true of many goddesses, above all Usas, the goddess of the Dawn, and Prthivi, the goddess of the Earth. The myth of Surya and his two mothers has been transposed to Karna whose two mothers, Kunti and Radha, however, do not at all resemble the goddesses of the Night and the Dawn. A third problem is that of the incarnated demons. It is surprising to see that Salya, the brother of Madri, who is the mother of the twins Nakula and Sahadeva, is the incarnation of a demon, Asura Samhrada ; and that Rukmin, the brother of Krsna's wife Rukmini, is the incarnation of a raksasa. The fourth problem is the surprising relationship between some heroes. Why is Asvatthaman (the incarnation of Siva) a son of Drona, who is an incarnation of Brhaspati ? This leads to the fifth and most important problem. Why in the battle between Good and Evil, i.e. between Kali-Duryodhana and Dharma-Yudhisthira, do so many good heroes take the side of Evil? In this respect Dumezil quotes the names of Bhisma, Krpa and Drona (cf. p. 204). We see here that this question, which was first mooted by the elder Holtzmann in 1846 and which led to the so-called inversion-theory, still remains unanswered. Needless to say, it is greatly to the credit. of Dumezil to have concluded his study of the Mahabharata not on a note of triumph but with a list of problems which remain to be solved. The publication in 1968 of Dumezil's Mythe et epopee I stimulated Madeleine Biardeau to study the Mahabharata in her classes at the Ecole Pratique des Hautes Etudes, Ve section: Sciences religieuses. One can find the results of her research in the first place in her annual reports published in the Annuaire. The most important publications to date, however, are two long articles published in the Bulletin de l'Ecole Francaise d'Extreme Orient in 1976 and 1978." These two articles were preceded by three articles in the same journal under the title: Etudes de mythologie hindoue. I. Cosmogonies puraniques." The articles published in 1976 and 1978 carry the title Etudes de mythologie hindoue. II. Bhakti et avatara. It is important to note that Biardeau's study of the epic follows upon her study of the puranic mythologies. Whereas Dumezil's study of the epic was the continuation of his studies in comparative or functional Indo-European mythology, Biardeau examines the epic in the light of classical Hinduism as found in the epics, the purana-s, other religious texts, dharma codes, and philosophical systems, and as observed today in the cults of the great Hindu temples. Van Buitenen wrote: "While they share fundamental underlying views and attitudes, Dumezil and Biardeau stand on the two sides of the Great Divide of the traditional periodization of Indian civilisation into Vedic and post

Page #7
THE STUDY OF THE MAHABHARATA Vedic : Biardeau to the south of it, as she takes the Mahabharata to be part and parcel of a Hinduism stretching indefinitely futureward; Dumezil to the north of it, as he mines the epic for a treasure of reminiscences of not only a Vedic but a pre-Vedic, Indo-Iranian, even Indo-European trifunctional inheritance." Most probably both scholars would agree with this characterisation of their work but they would hardly be willing to agree with the following remark of van Buitenen: "Neither in so doing shows great respect for, or even much interest in, the possible integrity of the Mahabharata as unique product in the growth of Indian civilization; the text seems fated to be relevant to anything but itself."10 Already in 1969 Biardeau remarked that the cosmic crisis which forms the plot of the epic does not find its model in the Vedic literature but in the puranic stories of reabsorption or dissolution (pralaya) and recreation (pratisarga)." In the article published in 1976 she is at pains to show that when the epic speaks of the end of a period of the world (yuganta), it is the entire symbolism of the pralaya which is brought into play. The first phase of the pralaya is the cosmic fire fanned by the breath of Rudra. In the second phase, rains extinguish the fire and a flood submerges the three worlds with an all-covering ocean. On it sleeps Visnu-Narayana who is represented as lying on the waters or upon the cosmic snake Sesa. A lotus grows from his navel and in this lotus is born the god Brahma who recreates the three worlds. The battle of the Mahabharata is said to take place in the twilight or junction (sandhya) between two yuga-s, the third and the fourth. The four yuga-s, Keta, Treta, Dvapara and Kali, are of diminishing duration : 4000, 3000, 2000 and 1000 divine years. Each is preceded and followed by a twilight. The Kotayuga has two twilights of 400 divine years, and the twilights of the following yuga-s have a duration of 300, 200 and 100 divine years respectively. In each successive yuga the dharma loses a quarter of its value. In the Kaliyuga the earth is oppressed by the weight of the adharma and a new mahayuga must begin. The four yuga-s constitute a mahayuga, and a kalpa contains one thousand mahayuga-s. The pralaya takes place at the end of a kalpa. Although the events described in the epic take place between two yuga-s, the symbolism is that of the end of a kalpa. Biardeau points to the theme of the fire at the end of a yuga, to which many allusions are made in the epic. She also compares the reabsorption of all beings into Visnu-Narayana, the great Yogin, when he is plunged in his yogic sleep, with the ascetic life of the Pandavas in the forest, an attenuated form of pralaya in so far as it is in opposition to civilized life regulated by dharma in the cities and the villages. The exile of the Pandavas which follows the fateful game of dice can be seen in two different perspectives. In the opinion of the Kauravas, the game of dice was the test

Page #8
Fa Hua Wen Hua Yan Jiu (Di 11Hao ) the diksa preceding the final sacrifice of the five brothers. However, the truth is that the exile is the period of consecration before the sacrifice of the coming battle. The idea of the sacrifice occupies a central place in Biardeau's interpretation of classical Hinduism for which one has to consult her study of the sacrifice in Hinduism.18 Held explained the battle of the epic as a cosmic ritual in which two parties representing the two halves of the cosmos oppose each other (op. cit., p. 332). According to Biardeau, the battle is a sacrifice comparable to the cosmic conflagration at the end of a kalpa, a sacrifice carried out by Rudra-Siva for the sake of the deliverance of all beings. In the epic the battle is at the same time identified with a sacrifice of cosmic dimensions and compared to the end of a yuga (yuganta). The war is a cosmic crisis which recalls to mind the junction between the Kaliyuga and the Krtayuga. The epic story relates the degradation of the dharma due to bad kings. This leads to a catastrophe from which the dharma rises again intact through the actions of the dharmic princes, Yudhisthira and his four brothers. In the epic the supreme god is Visnu-Narayana, who descends as an avatara on earth in the form of Krsna. Biardeau points out the contradictions in the concept of the avatara. The god is the supreme Purusa, the great yogin who is plunged in meditation, but with a part (amsa) of himself he appears on earth in order to assure the victory of the dharma over the adharma. The avatara must destroy that which is bad and restore the right order. He is both destroyer and creator. At the end of the kalpa, during the two phases of the pralaya, Rudra-Siva and Visnu-Narayana are active. However, in the period between two yuga-s Visnu alone must fulfil the functions of both Rudra-Siva and Visnu-Narayana. The avatara unites in himself two aspects which are opposed and complementary. He is Visnu but provided with a Rudraic component. In the third place the avatara can be at the same time a brahman and experienced in the art of war. An example of such an avatara is Parasurama, Rama with the axe, who kills twenty-one times all the adharmic ksatriya-s. Orthodox brahmanism makes the obtainment of salvation (heaven or deliverance) depend on one's belonging to the highest castes. The god of the bhakti assures salvation to all, women and sudra-s included, and even to wicked people. This universality of salvation does not arise from the love of god for his creatures so much as from his yogic impartiality. The god, Mahayogin, is a projection of the human yogin who has broken with the world and the rules of the castes. When Visnu-Narayana is plunged in his yogic sleep, he carries in himself all creatures without distinction between good and evil. In the epic, the relation between the king and the avatara is of special importance. Biardeau points out that the duties of the king were the main concern of the authors of the epic, and the avatara is the form

Page #9
THE STUDY OF THE MAHABHARATA of the divine which ultimately justifies these duties. The king is obliged to wage war for the sake of the dharma which the brahmans teach him. He is not allowed to retire to the forest and to leave the earth without a ruler. He must protect his subjects by the administration of punishment (dandaniti). His activities are impure but nevertheless necessary for the benefit of the world. Biardeau remarks that the universalism of the bhakti is concerned in particular with the salvation of the king. If it is possible to "save" his martial activity, to impregnate it with values so that his own dharma is as pure as that of the brahmans, then his entire activity will have been brought into the sphere of bhakti. This is probably the most important theme in Biardeau's study of the epic. This theme shows itself in concrete form in the relation between the avatara and the ideal king. In the epic the ideal king is Arjuna, and in her second article Biardeau studies in detail the figure of Arjuna (BEFEO, 't. LXV, pp. 87-203). The second part of this article is devoted to a study of the two Krsnas, the ksatriya Kssna of the Mahabharata and the cowherd Krsna of the Harivamsa (pp. 203-237). The dharma of the warrior finds its full justification in the Bhagavadgita. Biardeau remarks that she is not concerned with the problem of whether the Gita is an original part of the epic or a later addition. The essential point is that this text can only be understood in its epic context. 14 According to the orthodox brahmanic tradition, one can choose between living in the world following the prescripts of the dharma and engaging in sacrifices, or the life of a renouncer who refrains from sacrifices together with their "fruits". The bhakti religion tries to overcome this dichotomy and to make salvation accessible to everybody. It integrates renunciation in the form of yoga. In the Bhagavadgita, Krsna teaches that acts must be performed without concern for the fruits thereof. One must do one's duty as a yogin with the mind fixed on Krsna. All unselfish acts become sacrifices and are therefore free from violence. In the battle of the epic the ksatriya-s are offerers (yajamana) who perform the sacrifice of the war. It is the duty of Arjuna to wage war in order to assure the victory of dharma. It is not possible to study here in more detail Biardeau's interpretation of the mythological meaning of the epic. Her work is still in progress and she does not hesitate to correct her own interpretations. As she points out, the entire meaning of the epic cannot be discovered in a single structure because its unity consists in an integrated system of structures. As the exploration proceeds by "trial and error", new perspectives continually emerge.15 The differences between Dumezil's functional Indo-European mythology and Biardeau's structuralist Hinduist mythology become clearly visible when one compares their

Page #10
10. Fa Hua Wen Hua Yan Jiu (Di 11Hao ) treatment of the same topic. In studying Asvatthaman's massacre of the Pandava warriors, Dumezil drew attention to the Scandinavian eschatology and postulated the existence of an eschatology common to both the Indians and the Scandinavians which perhaps derived from an Indo-European eschatology. Biardeau recalls the action of Rudra-Siva during the first phase of the pralaya, when the three worlds are engulfed in a cosmic conflagration. When Asvatthaman kills the embryo in the womb of the wife of Abhimanyu, son of Arjuna and Krsna's sister Subhadra, Krsna promises to resuscitate the child which will be born dead. Here, Biardeau recalls the flood and the recreation. The waters of the flood in which the rest of the conflagration disappears are like a fecund womb which gives birth to the three worlds. It may be interesting to compare also their treatment of an apparently very minor episode of the epic. The young Kauravas and Pandavas are educated by the brahman Drona. Once, the ball (vita) they are playing with falls down a well (kupa) and they are unable to recover it. Drona mocks them for the uselessness of their warrior skill. He throws a seal-ring in the well and first recovers the ball with the help of blades of grass consecrated by a mantra. Then he pierces an arrow through the ring, takes the arrow out of the well and gives the ring to the young princes. Dumezil explains this story by comparing it with that of Trita, who is saved from a well by Brhaspati in the Rgveda (I, 105, 17). The story of Trita is told in the ninth book of the Mahabharata. While travelling, the three brothers Ekata, Dvita and Trita encounter a wolf. Trita runs away, follows the river Sarasvati and falls down a well. His brothers hear his cries for help but abandon him. He performs a mental sacrifice and is heard by Brhaspati, the priest of the gods. The gods go to the well. The river goddess Sarasvati descends into the well and Trita is lifted out of it. According to Dumezil, the authors of the story of the ball and the ring have not used the epic version of the story as found in book nine of the Mahabharata, but the Vedic version, since Drona acts alone whereas Brhaspati gives the alert to the gods who come and assist Sarasvati,16 Biardeau is not interested in the origin of the story but in its symbolism. Drona recovers the ball through the power of mantras, which is the proper brahmanic power. He shows his ksatriya power by his use of a bow and arrow for retrieving the ring. This episode presages the whole epic drama, where the ruling ksatriya-s are by and by reduced to a position in which they have to rely on a brahman in order to save the Earth from complete destruction and regain their royal power. The ball like any round object calls to mind a sacrificial offering like, for example, the pinda, the rice-ball offered to the Pitys. In the epic sacrifice of war, Earth is the offering par excellence, the victim to be sacrificed and reborn anew, an idea which

Page #11
11 THE STUDY OF THE MAHABHARATA goes back to Vedic literature, in which Varaha dives into the water to raise as much mud as he can hold on his snout and spreads it afterwards so as to make it prthivi ("the wide one"). In the epic and puranic cosmogonies, Varaha has become Narayana and is identified with Sacrifice : the Earth he raises out of the ocean on one of his tusks is his sacrificial offering. This cosmogonic symbolism pervades the whole Mahabharata war, since the end of a yuga is enacted and another yuga introduced. Drona's ring recalls the royal cakra, the power of regulating the world and dharma through uninterrupted time cycles : in other words, though a brahman, he aspires to kingly status. The deteriorated condition of the ksatra is the excuse for this aspiration and allows him to hold that status till his death, but from the very first day of the war he wishes for the Pandavas's victory, that is the victory of dharma, which would be the end of his royal power."? Dumezil, and Biardeau study the epic from different points of view but their approaches are not mutually exclusive. However, there is no doubt that for Biardeau the proper mythological meaning of the epic is to be discovered not by tracing the Vedic, pre-Vedic and Indo-European origin of its themes, but by considering the epic as the principal document on the duties of warriors according to the classical Hindu religion based on bhakti. An entirely different approach to the study of the Indian epic is to be found in P. A. Grintser's book entitled The ancient Indian epic. Genesis and typology (Moscow, 1974). Grintser examines the Indian epic in relation to other epics and applies the results obtained from the study of folktales and of oral literature to the study of the Indian epic. The first part of his book, "Oral and written tradition in the ancient Indian epic", is devoted to the genesis of the epic. Of the seven chapters, three deal entirely with the characteristics of the oral tradition as found in the epic. Grintser follows the methods developed by an American scholar, Milman Parry (1902-1935), in the study of the Homeric and Serbian epics, and by A. B. Lord in his book The Singer of Tales (Cambridge, Mass. 1960). Parry proved that the style of the Iliad and the Odyssey is typical of oral poetry because it is characterized by the use of fixed formulas. Parry defined a formula as "a group of words which is regularly employed under the same metrical condition to express a given essential idea". He showed that in the Homeric language only one formula with a certain meaning is admissible in a given metrical position. Grintser remarks that this principle is only applicable to the Indian epic to a limited degree as there is a greater metrical variety in the sloka than in the Homeric hexameter. Moreover, the abundance of synonyms in Sanskrit allows for the existence of metrically and semantically equivalent formulas. Grintser divides the formulas into six groups: I. Attributive formulas, e.g. kunti

Page #12
Fa Hua Wen Hua Yan Jiu (Di 11Hao ) putro yudhisthirah 'Yudhisthira, son of Kunti'; bhimo bhimaparakramah 'Bhima of terrible prowess'; purusavyaghrah tiger among men', etc. II. Narrative formulas, e.g., svabahubalam asritah / asritya 'relying on the force of his arms'; krtva pradaksinam 'having made a circumambulation', etc. III. Auxiliary formulas, e.g., etasminn eva kale tu at that time'; tasya tad vacanam srutva 'having heard those words of him'; etc. IV. Formulas of direct speech, e.g., nesyami / nayami yamasadanam 'I will bring (you) to the abode of Yama'; tistha tistheti cabravit 'he said "Stop, stop "'; srotum icchami 'I wish to hear'. V. Maxims: esa dharmah sanatanah 'such is the eternal law'; yato dharmas tato jayah where dharma is, there victory is'. VI. Similes purnacandranibhanana 'faces resembling a full moon'; dandapanir ivantakah 'like the god of death with a stick in his hand'. These are only a few examples. In an appendix, Grintser lists 594 formulas, indicating whether they occur in even or uneven pada-s. In the Sanskrit sloka, the metre of 95 per cent of all verses in the Mahabharata, there is less variation in the second half of the even pada-s, and consequently formulas are mostly found in that position. As it is necessary for the singer to express the same idea sometimes in the first half of a half-sloka and sometimes in its second half, there are two sets of formulas with identical or similar meaning for the two positions, for instance, in uneven pada-s: yudhisthiro raja 'king Yudhisthira', and in even pada-s: dharmaputro yudhisthirah 'the son of Dharma, Yudhisthira'. 12 The basic element of the formula is a supporting word in the second half of the pada, for instance rsabha in the following formulas: purusarsabha, nararsabha, manujarsabha, etc. In the Indian epic more than one formula of similar meaning can occur in one and the same metrical position, for instance purusarsabha and manujarsabha, rathinam varah and jayatam varah. Also formulas of different meaning can be construed by means of the same supporting word, for example, dvijasattamah and rajasattamah, sarvalaksanasampannah and rupayauvanasampannah. Finally, Grintser points out that the poet can create so-called formulaic expressions on the basis of the supporting word of a fixed formula. For instance, the supporting word samkruddhah in the formula abhyadhavata samkruddhah forms the basis for such formulaic expressions as pidayamasa samkruddhah, tatah parthas ca samkruddhas, tam simham iva samkruddham, etc. The use of formulas sometimes results in their being applied in cases where their meaning does not fit the context. For instance, when Kicaka pursues Draupadi with amorous proposals, she says to him: paradarasmi bhadram te 'I am the wife of another, good fortune to you' (IV. 13. 14a). When Uttara follows Arjuna, her future father-in-law, she is compared to an elephant-cow: anvagacchad visalaksi sisum gajavadhur iva 'the large-eyed one followed him as an elephant-cow

Page #13
THE STUDY OF THE MAHABHARATA her baby' (IV. 35. 9).18 According to Grintser it is only rarely that formulas are used inappropriately in the Indian epic. In Chapter Three of his book, Grintser analyses Chapter 49 of Book 6 which describes a duel between Drona and Dhrstadyumna. He arrives at the conclusion that only 23 out of the 164 pada-s are entirely free from formulas, and that 77% of the entire chapter consists of formulas or formulaic expressions. About the same percentage obtains in other battle chapters, whereas it is lower in narrative chapters. The didactic portions of the epic are also poorer in formulas and formulaic expressions, but many of the maxims they contain are found also in the same or in slightly different form in other didactic works. According to Grintser, the presence of such a great number of formulas and formulaic expressions in the Indian epic proves its origin and nature. He remarks that P. G. Bogatyrev and other scholars have pointed out that the basic law of oral literature is the combination of tradition and improvisation. Different aspects of the technique of oral literature (themes, repetitions, catalogues and inserted episodes) are studied by Grintser in Chapter Four. A theme, according to Lord, is an element of narration or of description in traditional oral poetry which is repeated. Among permanent themes in the old Indian epic are councils of gods and kings, meetings and receptions of guests, descriptions of the departure of heroes to the forest and of their forest adventures, descriptions of duels, ascetic exploits, armaments, military expeditions, but also of lamentations, prophetic dreams, evil omens and nature scenes. A theme is constructed on the basis of a determined sequence of traditional elements and contains a standard choice of formulas, but, although there is a great degree of uniformity, variations are far from being excluded. Grintser quotes different examples of themes in the epic, e.g. duel scenes, the lamentation of women or other relatives for slain warriors, etc. The use of traditional themes leads to repetitions which are often literal. Grintser remarks that these repetitions are conditioned by the requirements of the composition of the epic and by the formulatory nature of its language. It is not correct to maintain that a repeated passage is unauthentic, as has been done by Jacobi. Repetitions of themes and episodes result from the special nature of the oral performance of the epic. In the first place, the epic singer continues his recital during many days, It is therefore very probable that at the beginning of a new session and before a partly new audience he repeats an episode in order to clarify the sequence of events. In the second place, the same traditional epic topics are related by many different singers and often in substantially different versions. The singer who knows these versions sometimes wishes to incorporate two or three versions of the same epic event. Another characteristic of oral literature is the presence of contradictions and inconsistencies.

Page #14
14 ** * (115) Grintser remarks that many contradictions can be explained by the requirements of the story: the singers each time propose an explanation which corresponds to a certain element in the story and bring into the foreground the detail which is most appropriate in a given context. Typical also for oral poetry is the occurrence of enumerations of peoples, phenomena and objects. The abundance of this kind of enumeration is apparently to be explained by the fact that knowledge of them constitutes an indispensable part of the training of the epic poet. One of the most important characteristics of the Indian epic is the plentitude of all kinds of inserted episodes, differing in content and extent, which are not connected with the basic narrative. Grintser mentions with approval Vittore Pisani's theory that inserted episodes do not appear in a chaotic way but are meant to fill the temporal hiatuses which occur when the tempo of the main narrative is considerably slowed down.20 Often, also, these inserted stories do not simply duplicate or illustrate an epic event or supplement the characterisation of a person, but they reflect some essential facet of the main idea of the epic. For a more detailed discussion of Grintser's study of the oral technique in the Indian epic see 'Recent Russian publications on the Indian epic', Adyar Library Bulletin vol. 39 (1975), pp. 1-42. The same article also analyses two articles by Ja. V. Vasilkov on oral epic poetry. In 1979 Vasilkov published an article entitled 'Epic and pilgrimage' (On the meaning of the "pilgrimage" theme in the Mahabharata, Literatury Indii (Moscow, 1979), pp. 3-14. Grintser subsequently published in 1975 an article on 'Epical formulas in the Mahabharata and the Ramayana', Tipologiceskie issledovanija po folklory (Moscow, 1975), pp. 156-181. An even more recent article deals with the "Stylistical development of the theme in the Sanskrit epic', Pamjatniki kniznogo eposa (Moscow, 1978), pp. 16-47. In the last three chapters of the first part of his book, Grintser examines a number of topics which have been studied by many scholars in the past, such as the different versions and recensions of the epic, and the absolute and relative chronology of the two epics and their historical basis. Grintser draws attention to the fact that several versions are known to exist in the case of other ancient epics also. In the Mahabharata there are, in his opinion, traces of combinations of several versions. Grintser supposes that the text of the Mahabharata which has come down to us is an expansion of a much shorter narrative, which concentrated around the battle on the Kuru field, by stories and teachings told by the Bhargava brahmans. When the Mahabharata was reworked in the process of being written down, the didactic element was reinforced. As to the historical foundation of the epic, brief mention has already been made of Grintser's opinion which is in agreement with the views expressed by A. N.

Page #15
15 THE STUDY OF THE MAHABHARATA Veselovsky in his lectures on the comparative history of the epic (1884-6).21 Grintser firmly believes that the Mahabharata has a real historical basis and surmises that the battle on the Kuru field took place about 1000 B.C. Grintser draws attention to the fact that Troy really existed and that historical data are found in Beowulf, the Edda and the Chanson de Roland. He also believes in the existence of an Indian heroic age which lasted from the fourteenth to the tenth century B. C. This period witnessed the first centuries of the occupation of India by the invading Aryans and continued until the internecine war of the Bharatas. According to Grintser, there is usually a gap of several centuries, sometimes up to five or six, between the heroic age itself and its reflection in the epic. In the first part of his book, Grintser studied the genesis of the Indian epic and showed the importance of the oral tradition in the creation and development of both * Indian epics. In the second part, "Typology of the ancient Indian Epic", he com pares the Indian epic with other epics and with folktales which are also the products of oral tradition. Grintser begins by drawing attention to the fact that there is a remarkable uniformity in the motifs and plots of heroic epics. Viktor Zhirmunsky (1891-1971), who wrote extensively on the medieval European epics and on the epics of the peoples of central Asia, pointed out that the similarities between the heroic epics of different peoples have almost always a typological character and are based upon the existence of a similar social reality and of an identical level of development of the social conscience. Grintser adds that the typological correspondence between epics of different peoples has produced also a similarity in the organisation of the epic material as a whole. This similarity can be explained by the fact that their plots are constructed in similar ways and that they share the same principles and methods of composition. G.S. Levy divides the ancient epics into three groups: the epic of creation, the epic of the quest and the epic of the heroic war.22 The Ramayana, which describes the search of Rama for his abducted wife Sita, belongs to the second group; the Mahabharata to the third group. However, according to Grintser, there is undoubtedly a similarity between both Indian epics with regard to many constituent elements. The same motif of the abduction of the wife of the hero also has an important compositional function in the Mahabharata. When Yudhisthira lost the game of dice, Duhsasana dragged Draupadi by her hair into the hall. This outrage inflicted on Draupadi plays a primal role among the causes of the war between the Kauravas and the Pandavas. The outrage is a mitigated variant of the abduction and is characteristic for epics in which the non-human adversary is replaced by a human adversary. The motif of the abduction is duplicated many times in the epic. For instance, in the Fourth Book

Page #16
Fa Hua Wen Hua Yan Jiu (Di 11Hao ) Kicaka importunes Draupadi, and in the Third Book Draupadi is abducted by Jayadratha. A. Lord considers duplications to be one of the most important principles of composition of the epic. In the Mahabharata, the legend of the struggle of the hero for his wife is removed to the pre-history of the Pandavas. Draupadi is an incarnation of the goddess Sri, the wife of Indra. The struggle of the gods and the asuras for the possession of Sri is, as it were, transposed into the discord between the Pandavas and the Kauravas. In the Ramayana, the abductor is the ten-headed demon Ravana, who can be compared typologically to the serpent or dragon in fairy-tales. The theme of the struggle with the serpent-aggressor is specific to archaic epic forms. With the transition from archaic forms of the epic to more developed forms, the fantastic forms of the hero's rivals (above all of the serpent) are supplanted by the figure of the real, human opponent. The process of transformation of the form of the opponent receives its logical conclusion in epics of the type of the Iliad or of the Mahabharata. In the Mahabharata, the enemies of the heroes resemble almost not at all the archaic monsters and dragons of the subterranean world. Traces of their archetypal meaning can be found in the indications given in the epic that the Kauravas are incarnations of the Asura demons. In the Mahabharata, folk-tale motifs, although they are present, are usually so transformed that they are no more than vague memories. However, if one takes into account the nature of the transformation, it appears that the folk-tale compositional scheme which was valid for the Ramayana is applicable also to the Mahabharata. Analysing the contents of the Mahabharata, Grintser shows that, like the Ramayana, the Mahabharata can be divided into two mutually-connected but formally independent parts: the obtaining of the wife and the quest for the abducted wife. In his conclusion Grintser remarks that the hero of the epic, in contrast to the passive hero of the fairy-tale, is an active hero. In his morphological analysis of the folk-tale, Vladimir Propp (1895-1970) distinguished as two important elements the obtainment of gifts and the liquidation of a misfortune or fulfilment of a lack.23 In the Mahabharata they are as a rule brought about by military exploits or a mortal encounter. Moreover, the hero is above all active himself and relies on his own strength. Another difference between the composition of the ancient Indian epic and that of the fairy-tale is that the concatenation of the compositional links is much weaker in the epic than in the fairy-tale. 16 Grintser concluded this chapter on the Mahabharata by asking if it is not possible to show that both the fairy-tale and the epic, in principle independent one from the other, are ultimately based on some common compositional model. This problem is studied by him in Chapter Three, "The mythological motives of the epic". Grintser examines successively the following mythological motifs in heroic epics and illustrates

Page #17
THE STUDY OF THE MAHABHARATA 17 them with many examples of which we will adduce only those which are taken from the Mahabharata: 1. The divine birth of the heroes. In the Mahabharata all five Pandava brothers have divine fathers. Grintser remarks that this motif is very archaic, belongs to the oldest stratum of epic poetry and has mythological sources. However, taken as a whole, the motif of the divine birth serves in the epic above all as an indication of the eminent qualities of the epic hero. 2. The first exploit(s) of the hero. In many epics the hero fights duels with a monster. Already at a very early age he shows an indomitable warrior spirit. In the Mahabharata, Bhima falls from the lap of his mother soon after his birth and shatters a mountain with his body (I. 114. 11-13). Later he kills the raksasa-s Hidimba and Baka (I. 142, 153). 3. Death or near-death of the hero. The motif of the death of the hero, or more correctly his near-death, plays an important role in the epic. Often somebody else is killed instead of the hero, normally his closest friend, a comrade or brother, who functions as his deputy or as a substitute. This theme is found in the Mahabharata also, in the episode of the fire in the lacquer house in which a Nisada woman with her five sons is burned as substitute for the Pandavas and their mother (I. 136). 4. Journey of the hero to the subterranean world. In Book 18 (Chapter 2) of the Mahabharata, Yudhisthira, accompanied by the messenger of the gods, descends to hell. 5. Exile. The Pandavas live in exile for thirteen years. 6. Battle between relatives. In the Mahabharata, the Kauravas and Pandavas are first cousins. 7. Rejection of the love of a goddess by the hero. In heaven Arjuna rejects the love of the apsaras Urvasi (Crit. Ed. III, Appendix I, No. 6). In the following chapter, which deals with the abduction of the wife and the search for her, Grintser remarks that the Ramayana repeats to a certain degree the composition of the calendar-myth in which the goddess of fertility is in the power of the forces of the subterranean world. The god-husband sets out to search for her, defeats the abductor-demons and brings ("revives") the victim back to the earth. In the Mahabharata, the five Pandavas are sons of the gods Dharma, Vayu, Indra and the Asvin-s, but at the same time they are all considered to be incarnations of Indra. Draupadi is an incarnation of his wife, the goddess Sri. Grintser refers to the legend in the Markandeya Purana where the story of the Pandavas is connected with the myth of Indra's killing of Vrtra.24 According to Grintser this legend explains an aspect which is very essential and fundamental in the Indian mythological tradition. This tradition considers the struggle of the Kauravas and the Pandavas to be an earthly reflection of the struggle of Indra as king of the gods with the Asura demons and above all with Vrtra. Grintser remarks that the main function of the motif of the struggle with the dragon is cosmogonical. However, according to him, Indra's

Page #18
18 #**ILWER 11) battle with Vrtra also has a calendar, agricultural meaning. In this legend there is not only the archaic layer of the agricultural myth (the theft of cows, the liberation of the waters and the return of fertility to the plants and the cattle) and the central motif of the struggle with the dragon, but also the vague outlines of the idea of the forced imprisonment of the god, his death and his resurrection to life. Dumezil wrote that the myth which is transposed in the whole of the Mahabharata is the preparation for, and the bringing about of, the "end of the world" (or of an age of the world) followed by a rebirth (or the beginning of a new age of the world).25 Grintser remarks that one must speak rather of a calendar myth, although neither in structure nor in the general idea is it possible to draw a strict dividing line between the calendar myth and the eschatological myth as interpreted by Dumezil. Examining the relations between myth and folk-tale, Grintser remarks that they possess the same common structure but differ in their semantics. Both are constructed as chains of losses and gains, but in the myth these gains have a cosmic or ritual character (the heavenly luminaries, fire, sacred potion, etc.), whereas in the folk-tale they are more individual treasures (miraculous objects, a bride, a bird, etc.). The myth takes place in an ahistoric time which precedes historic time, while the "once" of the folk-tale relates to historic time. The opposition between myth and folk-tale is that of sacral and non-sacral. The compositional scheme of the archaic myth and folk-tale forms the basis for the plot of both the fairy-tale and the epic. Of central importance for the agricultural civilisations of antiquity is the calendar or agricultural myth. Serving as semantic models for these myths are the daily solar cycle, the cycle of the seasons of the year and the organic life-cycle. As a rule the agricultural myth is told in the male variant (the god, not the goddess, disappears to the world beyond the grave and his wife or sister goes to search for him). On the contrary, in the overwhelming majority of cases the classical epic tells of the search for the wife by the husband. This actualisation in the epic of the feminine variant of the myth is apparently connected with the influence of the social institution of "the exchange of women " or of the exogamous marriage. The epic is not a heroised story and not a "spoilt " or re-interpreted myth, but it is a representative of a compositional scheme which is common to both folk-tale and myth. This traditional compositional scheme was the means by which the heterogeneous epic material was organised, and not the source for the contents of the epic. The myth is related to ahistoric time, but in the epic one finds concrete and determined facts. In the epic, history is not superimposed on the myth but it is only told according to the mythical canon; the true meaning of the epic poetry reveals itself clearly as what remains of this canon is the compositional scheme. It is only in this sense that one can speak

Page #19
19 THE STUDY OF THE MAHABHARATA of a "substitution" of the mythological models by historical ones, because, in essence, in the epic the heroes are treated historically from the very outset. Therefore the divine origin of the persons - a basic characteristic of the myth - serves in the epic as the most obvious sign of the exclusive nature of the hero. The hero of the myth is as it were endowed in advance with greatness and superiority, but the hero of the epic must acquire these qualities and therefore the quest for glory dominates him continually. The characters of the folk-tale act automatically according to the demands of the plot, but the epic hero acts consciously, following the epic idea of honour (Yudhisthira agrees to play the game of dice, though he knows about its consequences). In the centre of the epic is the figure of the warrior-hero who relies above all on his own strength. This figure is not proper to the myth or the folk-tale. According to Zhirmunsky, the similarity in the plots of different epics is not to be explained by borrowing, but "reflects regularities and relations of the objective reality and at the same time is conditioned by the historic characteristics of the human conscience, which reflects this reality." Zhirmunsky and other scholars such as V. Ja. Propp, E.M. Meletinskij (1918- ) and B. N. Putilov (1919- ), consequently give preference to the historical-typological approach to the study of the epic. A. Lord and the followers of the mythological school see in the myth not only a compositional model for the epic but even more a semantic model. However, Grintser disagrees and considers it wrong to look for the meaning of the epic in the myth and to see in it a transformation of the myth. In the process of oral creation the epic singers made use of a compositional scheme which is also proper to the myth. However, this scheme is only the form, the matrix, in which the purely epic material is cast. In studying the epic, the meaning of the myth and comparison with the myth are essential for the understanding of the syntax of the epics, but not for the understanding of their semantics, specific character and problems. The epic subject matter, by virtue of the conservative nature of oral creation, possesses an amazing continuity, but using this subject matter as a basis, epics are created in different countries and periods, each possessing their own concrete meaning. NOTES The Mahabharata, Part II 1. Pandavasagan och Mahabharatas mytiska forutsattningar', Religion och Bibel, Nathan Soderblom-sallskapets Arsbok VI (1947), pp. 27-39. Translated with a few omissions by Georges Dumezil as La legende des Pandava et la substructure mythique du Mahabharata', Jupiter Mars Quirinus IV. Paris, 1948, pp. 37-53. Commented upon by Dumezil, pp. 55-85. 2. Stig Wikander, 'Sur le fonds commun indo-iranien des epopees de la Perse et de l'Inde',

Page #20
# ILUX (115) La Nouvelle Clio 7 (1950), pp. 310-329; Fran Bravalla till Kurukshetra ', Arkiv for Nordisk Filologi LXXV (1960), pp. 183-193 ; 'Germanische und indo-iranische Eschatologie', Kairos II (1960), pp. 83-88 Tansar-nameh och Mahabharata', Acta Or. 30 (1966), pp. 213-217; 'Nakula et Sahadeva', Orientalia Suecana 6 (1971), pp. 66-96; Georges Dumezil, Karna et les Pandava', Orientalia Suecana 3 (1954), pp. 60-66; Les pas de Krsna et l'exploit d'Arjuna', Orientalia Suecana 5 (1956), pp. 183-188; Les Dieux des Germains, chap. III" Le drame du monde". Paris, 1959; 'La transposition des dieux souverains mineurs en heros dans le Mahabharata', IIJ 3 (1959), pp. 1-16. 3. Mythe et epopee. I. Paris, 1968, pp. 31-257: "La terre soulagee". 4. A. Ludwig, Das Mahabharata als Epos und als Rechtsbuch, Prag. 1896, p. 76; "So muszen wir auch die vermahlung der Draupadi mit den funf Pandava als den auszdruck eines bundnisses der Pancala mit denselben auffaszen. Disz ist freilich wol als eine spatere historisirung anzusehn." J. Dahlmann, Das Mahabharata als Epos und Rechtsbuch. Berlin, 1895, p. 97: "Die funf Indra bilden eine Einheit, welche in der engen Verbruderung und ungetheilten Gemeinschaft der funf Pandava sich verkorpert. Die eine Gottin Cri wird sichtbar in Draupadi und vertheilt sich in ihren funf Eigenschaften auf die funf Indra. So tritt hier die engste Verbindung, die vollkommenste Einheit zwischen beiden gottlichen Gestalten zu Tage. In dem gemeinsamen Besitz der Cri symbolisiert sich der gemeinsame Besitz (sahabhojanam) des einen ungetheilten Vermogens (ratnasya)." 5. Cf. E. W. Hopkins, JAOS 13 (1888), p. 354; The Great Epic of India. New York, 1901, p. 376; H. Jacobi, GGA (1896), p. 72; M. Winternitz, JRAS (1897), p. 758; G. Dumezil, Jupiter Mars Quirinus IV. Paris, 1948, p. 38, notes 1 and 2. 6. On Mitra and Varuna see Georges Dumezil, Les dieux souverains des Indo-Europeens. Paris, 1977, chapitre premier "Mitra-Varupa ". 7. Mythe et epopee, I. Paris, 1968, p. 122. 8. Tome LXIII (1976), pp. 111-263 ; Tome LXV (1978), pp. 87-238. 9. Tome LIV (1968), pp. 19-45; Tome LV (1969), pp. 59-105; Tome LVIII (1971), pp. 17-89. Published in one volume in 1981 as Volume CXXVIII of the Publications de l'Ecole Fran caise d'Extreme-Orient. 10. The Mahabharata. Volume 3. Chicago, 1978, p. 163. 11. Ecole pratique des Hautes Etudes. Ve section--Sciences religieuses. Annuaire 1969-1970. Tome LXXVII. Paris, 1969, p. 169. 12. BEFEO LXIII, p. 135. 13. Madeleine Biardeau-Charles Malamoud, Le sacrifice dans l'Inde ancienne (Paris, 1976), pp. 7-154. 14. Cf. Etudes de mythologie hindoue, t. 1. Paris, 1981, pp. 144-149; Le sacrifice dans l'Inde ancienne (Paris, 1976), pp. 119-138. 15. Cf. Annuaire, t. 85. Paris, 1977, p. 167. 16. Mythe et epopee. I, pp. 198-201. 17. The ball-game; myth and rite', Rtam, Ludwik Sternbach Felicitation Volume (Lucknow, 1981), pp. 875-881. 18. Cf. Annuaire, t. 77, pp. 168-169; BEFEO LXIII, p. 218, n. 1. 19. The critical text has sisur and van Buitenen translates : "the large-eyed girl followed him as a baby elephant follows its mother." Probably he changed gajavadhar to gajava dhum. However, this reading is not found in the critical apparatus. 20. 'The Rise of the Mahabharata', A Volume of Eastern and Indian Studies presented to Pro

Page #21
21 THE STUDY OF THE MAHABHARATA fessor F.W. Thomas. Bombay, 1939, pp. 166-176; Mahabharata, Episodi selecti. Torino, 1954, pp. 41-59. . 21. Istoriceskaja poetika [Historical poetics]. Leningrad, 1940, p. 471. 22. G.S. Levy, The Sword from the Rock. An Investigation into the Origins of Epic Literature and the Development of the Hero. London, 1953, pp. 13 ff. 23. Morphology of the folktale. Indiana University, 1958. First published in Russian in 1928: Morfologija skazki. 24. Cf. G. Dumezil, Mythe et epopee, I. Paris, 1968, pp. 113-116. 25. Op. cit., p. 210.