7—
Appropriating the Epic:
Gender, Caste, and Regional Identity in Middle India

Joyce Burkhalter Flueckiger

Indian epic includes not only the well-known Mahabharata and Ramayana— narratives dating from antiquity that have survived for centuries—but also a myriad of other traditional stories in verse that vary in their themes and social meanings from community to community. In the central Indian region of Chhattisgarh, the Candaini epic vividly draws on the local folklore repertoire; it is an oral tradition performed—like other contemporary Indian epic—in discrete episodes that figure in the larger narrative known to the audience. The narrative, a distinctly nonheroic tale that challenges conventional notions of epic, relates a love story and features a female heroine. Joyce Flueckiger argues that it derives much of its cultural and political meaning from its strong identification with the region and thus stands as perhaps the most resonant expression of what might be called Chhattisgarhi folklore, constantly viewed by the community as its "own" epic.^[1]

In the first essay of this volume,^[2] Gregory Nagy suggests that a particular genre can be identified as epic only by placing it in relationship to other genres performed within a particular folklore community. Accordingly, features of narrative, poetic composition, and heroic characters and themes that have typically characterized the analytic category of epic would not in and of themselves be enough to give definition to "epic."^[3] Such is true in India, where there are numerous folk narrative traditions that are long, sung, and heroic but that do not hold the significance of "epic" for the communities in which they are performed. What distinguishes "epic" from these narratives is the nature of the relationship epic narratives have with the communities in which they are performed: "Epics stand apart from other 'songs' and 'stories' in the extent and intensity of a folklore community's identification with them; . . . the oral epic is the most geographically wide-spread form that still preserves a community's identity."^[4]

Thus narratives that serve as "epic" in one region, in which performers

and audiences self-consciously identify the narrative as "theirs," may be performed in another region without the level of necessary self-identification to be categorized as "epic." For example, the Dhola-Maru epic tradition of northern and western India is also performed in the central Indian region of Chhattisgarh, the area of study for this essay, and yet it is known here specifically as a Rajasthani (western Indian province and cultural region) story, representing a somewhat exoticized "other," exemplified by the hero flying away on a desert camel not native to Chhattisgarh. The northern and central Indian martial epic of Alha is also performed in Chhattisgarh but is associated with specific historical kingdoms outside the region and is perceived to be someone else's history. Likewise, although the pan-Indian Ramayana epic tradition is arguably the most significant religious narrative in the plains of Chhattisgarh, its singers and audiences call it a Hindi, rather than Chhattisgarhi, story (katha ) The hero and heroine, Ram and Sita, are divine royalty and, in dramatic performances of the tradition, are dressed in generic north Indian royal costuming rather than Chhattisgarhi dress and jewelry that would identify them by region and caste. Placed in the context of these long, sung, heroic narratives available in the repertoire of Chhattisgarhi regional performance genres, the epic of Candaini stands apart in the extent to which it has been appropriated by various communities within the region as "their own." This essay examines the ways in which this process of appropriation has identified and given identity to the folklore region of Chhattisgarh.

Geographic and Social Boundaries of the Epic

Performance of the Candaini narrative is not limited to Chhattisgarh; its performance spreads across geographic and linguistic borders, from Chhattisgarh to the Gangetic plains of northern India, in the province of Uttar Pradesh (or U.P.). The tradition is called Candaini in Chhattisgarh and Canaini or Loriki (from the names of its hero and heroine) in U.P. Candaini differs from many Indian epic traditions in that it is not associated with a particular caste or regional historical events, nor is it associated with a religious cult. Thus it can and has been appropriated by a spectrum of communities as "theirs" in a way in which many other narrative traditions cannot be.

While folklorists may identify the narrative traditions in these two regions as "the same" because of their shared characters, constant plot elements, and shared motifs, it is important to point out that the wide geographic mapping of Candaini is a reality to those folklorists and not to the epic's performers and audiences. They know and understand the tradition as rooted in geographically circumscribed performance and social contexts, as being identified with—"belonging to"—specific communities, in this case, a cowherding caste in U.P. and the broader regional folklore community in Chhattisgarh. None of the singers whom I met on the plains of Chhattisgarh knew the

"same" story was sung in U.P. When I mentioned this to one of the epic singers, he exclaimed: "Do you mean they really sing our Chhattisgarhi Candaini way up there?"

When I went to Chhattisgarh to begin my dissertation fieldwork in 1980, one of my first "strategies" was to elicit from villagers a core repertoire of folklore genres that they considered to be "Chhattisgarhi," unique to or characterizing that linguistically and historically defined region. I would ask something like "What do you celebrate here in Chhattisgarh, what do you sing?" A core repertoire gradually emerged from the varied responses. Its genres did not exhaust their performance repertoire but included those traditions whose performance the inhabitants themselves perceived to be identified with and give identity to the region. The epic traditions of Candaini and Pandvani (a regionalized variant of the pan-Indian Mahabharata tradition) were always on this list.

The longer I was in Chhattisgarh and as I became more knowledgeable about its various performance traditions, and thus not perceived to be quite such an unknowing outsider, indigenous commentary began to break down the nature of the social communities with which performance traditions were identified, by caste, age, and gender. Candaini and Pandvani most always retained their regional identification, however; the community with which they are primarily identified is more inclusive, having a wider geographical and social spread, than that of any other genre from the core repertoire. Candaini was repeatedly identified as "a Chhattisgarhi story," "our story."

The social boundaries of the performance communities (and note I have shifted to plural here) associated with the Candaini epic tradition in Chhattisgarh have shifted rather dramatically in the last twenty to twenty-five years; so it is important to look carefully at what it means for an epic to be "ours," asking who is the "we" that is being represented. Further, at what level is identification being made, textually, performatively, or both? I suggest that increasing mass media and literacy in Chhattisgarh in recent years have affected both the performances that identify and the identity of the "we."

The Epic Story

Epic narratives exist both as oral and performance traditions, a distinction Laurie Sears and I made in Boundaries of the Text between a general knowledge of the "whole story" (a summary) that many in the folklore community would be able to relate and the epic as it is performed in a marked, artistic enactment of that oral tradition.^[5] The performed epic in India is sung in episodes,^[6] with the assumption that audience members frame the performance both within the larger epic story (oral tradition) as well as within the folklore repertoire of which it is a part. Thus while scholars have spent considerable energy recording epic stories "from beginning to end," counting

the number of hours and pages required to do so, this is not how the epic is received by indigenous audiences. Further, there are certain episodes of the epic that are more frequently performed than others; and there may be episodes that exist only in the oral tradition, and not in performance at all.

What follows is a narrative summary of primarily the Chhattisgarhi epic variant, drawn from the oral tradition (summaries that were told to me) and performances I attended. I have noted some of the major differences between this and the U.P. variant of the epic, and more of the substantive differences between the two variants will become apparent in the analyses that follow. In Chhattisgarh, Candaini is the story of the hero Lorik and heroine Candaini, both from the Raut cowherding caste. The hero and heroine are each married to other partners, but Candaini leaves her husband when she learns he has been cursed by the goddess to be impotent for twelve years. On her way back to her maternal village, Candaini is accosted in the jungle by the untouchable Bathua. She cleverly escapes his evil intentions, but he chases after her and terrorizes the inhabitants and cattle of the village. In desperation, the villagers ask the hero Lorik to rescue them; ultimately he defeats Bathua through nonmartial (and, I might add, rather unheroic) means. During this contest, Candaini first lays eyes on him, falls in love, and proceeds to seduce him. After some delays, primarily due to Lorik's hesitancy and cowardice in decision making, the hero leaves his wife Manjari, and he and Candaini elope together to Hardi Garh.

In Chhattisgarh, Candaini performances center upon and elaborate various adventures from this elopement journey (urhar; literally, "flight"). In fact, when I asked villagers what the story was about, most responses began with some variant of "It is the story of the elopement of Lorik and Candaini." Eventually, Lorik receives word that his brothers have all died in battle, and their wealth and cattle have been dissipated throughout the Chhattisgarhi countryside, leaving his mother and wife destitute. Lorik returns home with Candaini to avenge his family's honor. He succeeds in reclaiming his cattle, through battle in the U.P. variant and by wandering the countryside as a mendicant, collecting his cattle, in Chhattisgarh. When the task is completed, he takes up the position of head of the surviving extended family, including his first wife. But, it is said, Lorik did not take pride in his success. In U.P. versions, he finds that his former physical prowess and strength have dissipated, and he kills himself. In Chhattisgarh, sad and dissatisfied after his return, Lorik one day mysteriously wanders off into the countryside, never to be seen again.

In the Chhattisgarhi village of Garh Rivan (home of Lorik in the epic and a present-day village near the cattle bazaar town of Arang in Raipur District), one performer sang the epic's closing episode as that of a lovers' argument. As the couple was sitting in a boat in the middle of the village tank (or pond), the argument got so vehement that the boat overturned. Candaini swam to the bank and took refuge in a goddess temple. The goddess was so angered

at Candaini's sudden and inauspicious intrusion that she beheaded our heroine, only to regret her action later and restore the head. In a village goddess temple on the banks of the tank of Garh Rivan, there are today two images (one beheaded and the other whole) of the heroine Candaini, which keep the goddess company. The heroine is not called a goddess, but simply honored as "our rautin," or cowherdress. Lorik, it is said, was never seen after this episode and is presumed to be still wandering in the Chhattisgarhi countryside.

The narrative as performed in both Chhattisgarh and northern India is not a religious epic, nor are its performances an integral part of any particular ritual or festival, although it is often performed at two festivals that have themselves been "imported" into the Chhattisgarhi ritual calendar, ganes caturthi and durga puja, perhaps as a way of localizing them. Villagers say the epic is sung primarily for "entertainment" (manoranjan ) nonprofessional performers may sing for small groups of friends and neighbors, and professionals may perform at annual village fairs or provide entertainment during long winter evenings. These nonritual performance contexts do not, however, diminish the significance of the epic for the communities in which it is performed. In U.P., while the characters are not deified, they are they held up as models to be emulated, of "who we would like to be." In Chhattisgarh, by contrast, they are "who we are," in larger-than-life proportions.

The U.P. Variant as Caste Epic

To understand the differences in the performatively identified communities of the Gangetic plains of U. P. and Chhattisgarh, it will be useful for us now to take a closer look at both narrative and performative variation in these two areas. My analysis of the U.P. epic variant is based on two published versions of the epic collected and transcribed by S. M. Pandey in the 1970s, one in the dialect of Awadhi and the other in Bhojpuri, as well as upon personal communication with Pandey in the early 1980s.^[7] I will call this U.P. variant the Loriki/Canaini tradition, taken from Pandey's titles. The Chhattisgarhi data is drawn from my own fieldwork (1980 through 1993, intermittently) and Verrier Elwin's translation of a partial version.^[8]

Uttar Pradesh is in the heartland of orthodox Brahminic Hinduism, while Chhattisgarh lies on its periphery. Chhattisgarh's cultural and religious traditions are influenced by the high percentage of tribal groups that have now been integrated into the Hindu caste system. These include folk performance and festival genres, social and marriage patterns, and women's dress, tattoos, and jewelry. Of particular interest to us in our examination of the epic is the relatively higher status of women in Chhattisgarh compared to women in U.P. This may be partially explained by tribal influences, but also influential is the fact that the rice-growing economy of central India requires a higher

proportion of female labor participation than does the wheat-growing economy of the north. Thus women in Chhattisgarh are not considered to be quite the economic liability that they are in U.P.^[9]

In both performance areas, U.P. and Chhattisgarh, the epic tradition seems to have originated with the local cowherding castes—Ahirs in U.P. and Rauts in Chhattisgarh. In U.P., where Ahir males continue to be both primary performers and audience members, however, the tradition has remained more closely identified with that caste. Pandey cites two Awadhi proverbs in U.P. that clearly identify Canaini with the Ahir caste:

However clever an Ahir be 
Nothing but Canaini singeth he.

However many times an Ahir may read the Puranas 
He will not sing anything but Canaini. ^[10]

Certain clans of Ahirs in U.P. identify with the epic more than just performatively; they look to the epic as the history of their caste. Gwal Ahir singers of the contemporary folk-song genre called virha believe the Loriki-Canaini to be the oldest extant record of their caste group. Although most of them admit to not knowing the epic well, they claim that many of their songs and narratives are based upon it, and many social and religious traditions unique to the caste derived from it.^[11]

The differences between caste-epic identification in U.P. and Chhattisgarh can be partially attributed to the differences in each caste's self-perception, status, organization, and ideology. The Ahirs of U.P. have traditionally viewed themselves as a local warrior caste and continue to promote that image of themselves. As certain Ahirs gained in political and economic power in the late nineteenth century, they joined forces in an effort to raise their caste status by appropriating customs (such as donning the sacred thread) and ideologies of the ksatriya varna caste category (a process the Indian anthropologist Srinivas has called "sanskritization"). ^[12] Another way to confirm their warrior status was to try to associate themselves with the Yadav cowherding caste of the divine cowherder Krishna, calling themselves Yadavs instead of Ahirs. Ahir intelligentsia "rewrote" certain historical documents to prove this connection^[13] and formed a national Yadav organization that continues to coordinate and promote the mobility drive of the caste. Integral to this movement are retellings of caste history that reflect its martial character; the epic is an important channel for some of these retellings. Hence the cowherder Lorik is portrayed as a warrior first, whose primary role is to defend the honor of the caste, often through a defense of the honor of its women. Consequently, epic battles rather than the elopement become the central episodes of the narrative, and the elopement is consciously underplayed. Elopement and the freedom of individual choice it implies threaten caste

endogamy and the strict maintenance of caste boundaries necessary in the effort to raise status. Further, the implicit freedom of elopement contradicts the social control of women articulated elsewhere in the U.P. variant of the epic.^[14]

Further, the male hero Lorik is the central character of the U.P. variants, rather than the heroine Candaini, as is the case in Chhattisgarh. The northern tradition is, in sum, a male, martial epic tradition that has been appropriated to promote a particular image of the Ahir caste.^[15] A common saying in eastern U.P. asserts: "If Loriki is recited for one month, somewhere there will be a battle."^[16] The martial ethos of the epic is perhaps most dramatically visualized in a bazaar pamphlet titled (in Hindi) Lorikayan: The Battle of Hardigarh (interestingly, this episode is the only one that has been published in this popular format).^[17] Its cover pictures Lorik as the classical indian warrior, standing on a battlefield, holding up a broken chariot wheel, with bodies and weapons strewn across the field and arrows flying through the air.

The Chhattisgarhi Variant as Regional Epic

Older Chhattisgarhi informants told me in 1980 that in Chhattisgarh, too, Candaini singers used to be primarily from the cowherding Raut caste. Its multicaste audiences and the seemingly easy adaptation of the epic to innovative performance styles available to performers from a wide spectrum of castes suggest, however, that it was never "caste-owned" in the sense that it is in U.P. A possible explanation for differences in the caste-epic relationship are the respective castes' self-image.

One fifty-year-old Raut male gave the following account of the dispersion of the caste. In "former days" all the Rauts of the area used to go to Garh Rivan (the home of Lorik in the epic and the present-day village mentioned above) to celebrate the Raut festival of matar.^[18] Then one year, King Kadra, of a basket-weaving caste, battled against the Rauts. Many Rauts were killed, and the survivors scattered from Garh Rivan and settled "here and there." Since that time, according to the informant, Rauts have no longer gathered at Garh Rivan to celebrate matar, but rather celebrate it in their own villages. We cannot know from such an account whether or not the caste was, in fact, ever a cohesive martial or administrative power. Their perception, however, is that they were once stronger and more unified than they are now.

In the more recent past, Chhattisgarhi Rauts have traditionally seen themselves as "village servants," who herd and milk the village cattle, rather than as warriors who protect caste honor and boundaries.^[19] Lorik, as a Chhattisgarhi Raut, is not portrayed as the U.P. martial hero brandishing a sword, riding on a horse, but primarily as a lover whose only weapon is his herding staff and who travels on foot. Further reflecting a Chhattisgarhi ethos in which women have more mobility and arguably higher status than their sis-

ters in the Gangetic plains, the heroine is the primary initiator of action in Chhattisgarhi performances; it is frequently she who protects and saves Lorik rather than the other way around. Thus, although the singing of the epic may have been first associated with the cowherding caste of its singers, the tradition as it has been documented in the last fifteen to twenty years does not suggest a strong caste identity.^[20]

Part of what gives the epic tradition its regional identification in Chhattisgarh is its performance contexts and the broad social base of its audiences and performers today. Two basic performance styles of Candaini have developed in Chhattisgarh. Both styles are most commonly called simply Candaini, but when the styles are distinguished, the first is called Candaini git or song, and the second naca, or dance-drama. As mentioned earlier, traditionally, Candaini git singers were male members of the Raut caste who sang the epic both professionally and nonprofessionally to primarily male audiences, but with women sitting on the sidelines. Rauts sang without musical accompaniment; but essential to their performance was a companion (ragi or sangvari ) who joined in the last words of every line and served as a respondent. Today, it is difficult to find Rauts who still sing in the git style without instrumental accompaniment. The only such singer I knew died in 1988, and none of his sons were interested in learning or continuing his father's tradition. As one informant observed, "How can this [that is, style with no musical accompaniment] compete with video halls?" The repetitious response by the ragi, however, is still one of the primary characterizing features of both instrumentally accompanied git and naca Candaini performance styles.

The dates and circumstances in which members of the Satnami caste took up the git style of Candaini performance are undocumented and vague in caste and regional memory. Yet when I was looking for epic performances in the 1980s, I was frequently told that I would find Candaini only in those areas with large numbers of Satnamis. The Satnamis are a sect that converted in the 1800s from the outcaste Camar, a leather-working caste, but whose conversion did not raise their status from that of the lowest caste groups. It is probable that when they began to sing Candaini professionally, it began to attract more diverse audiences and to take on its current regional identification. The Satnamis added musical accompaniment to the git performance style, including, minimally, harmonium and tabla; but, as mentioned above, they have retained the combination of lead singer and one or more "companions," whose response lines end with mor or tor.^[21]

Because I have little comparative data to use from "purely" Raut performances, it is difficult to know exactly how the narrative may have shifted when the Satnamis began to sing the epic professionally, particularly in its portrayal of the "villain" character, the Camar Bathua who tries to accost Candaini in the jungle. In one Satnami performance, however, Lorik meets Bathua again

after their initial confrontation in the heroine's maternal village. Bathua reappears as the bodyguard of a foreign king whom Lorik has offended (by chopping off the nose of one of his subjects), so the king sends Bathua to punish him. This time their confrontation is martial, and Lorik is unable to defeat the untouchable physically. He is pinned to the ground, and Candaini has to beg Bathua for mercy. The Camar gives in but says Lorik must tie him up so that the king will think he has been defeated, not compassionate. Lorik eventually wins the kingdom through both battle and trickery and names it after the untouchable Bathua. When I later discussed this episode with several non-Satnami villagers, they told me that Satnamis have tended to glorify the character of Bathua and that a Raut singer would never have included such an episode, glorifying the heroism of the Camar.

The second Candaini performance style, called naca (literally, "dance"), includes song and dance, spoken conversations between characters, and narration in the git, responsive style.^[22] According to naca performers, the naca is said to have developed in the early 1970s in response to the strong influence of the increasingly popular Hindi cinema, an essential element of which is also song and dance. A naca troupe consists of up to eight or ten performers, some of whom are actors and others musicians. An important feature of the naca is the inclusion of costuming and minimal props. The hero Lorik carries a herding staff and wears traditional Raut festival dress, decorated with peacock feathers and cowrie shells; male performers put on saris and typical Chhattisgarhi jewelry to act out the female roles. The musicians sit at the side of the stage and accompany the songs of the actors or provide their own sung narration in the Candaini git style. Candaini is only one of many narratives performed in the naca style, but naca troupes that specialize in Candaini do so to the exclusion of other narratives. Although this style has grown in popularity, it is expensive to patronize. When sufficient funds for the naca cannot be raised, or if troupe members are singing nonprofessionally, the git style, without dance, can still be heard.^[23]

The performance context of the naca is important in establishing the epic's regional character. Troupes are usually multicaste, heavily represented by Satnamis, but also by other middle-level castes, including Rauts. One performance troupe I met consisted of ten members from six different castes. Troupes are hired by village/neighborhood councils for annual village fairs or festivals or as independent entertainment events. Occasionally, a family will sponsor a performance to celebrate the birth of a son or a wedding. Naca audiences, too, represent the caste spectrum of a particular village or urban neighborhood, male and female. Nacas are performed in public space such as a village or town square or main street, accessible to everyone. Persons from surrounding villages frequently walk several miles to attend nacas in neighboring villages, and audiences may reach as many as 200 participants.

The enthusiastic and responsive participation of women in the primary

audience of the Candaini naca stands in sharp contrast to the all-male audiences and performance contexts of the U.P. variants of the epic. In 1980 when I asked female audience members if women ever sang Candaini in Chhattisgarh, they all answered negatively. I did hear segments of the epic narrative and reference to its characters in other female performance genres, which they did not, however, identify as "Candaini" because of the performance context and singing style. "To sing Candaini" means to sing in a public context and, more specifically, to incorporate at some level the responsive singing style of the Candaini ragi, with his end-line words of tor or mor. What these women were singing was identified by context as a harvest-dance song, not by content as Candaini.

In recent years, there have been a handful of individual female performers who have performed professionally the git style of Candaini, accompanied by male ragas and musicians. They are usually self-taught and have gained meteoric popularity because of their unusual position as professional, public female performers. Several audience members told me: "Who wouldn't go to hear a woman? There's more entertainment in that!" One such female performer is Suraj Bai, who, in 1987, was hailed in a local English-language newspaper as "the melody queen." She had represented Chhattisgarh at national and state folk festivals and had performed on nationwide television and radio; yet, the newspaper article bemoaned, she still worked as a daily-wage laborer. Over the last five years in Chhattisgarh, the epic tradition of Pandvani is experiencing a similar rise in popularity, primarily attributable to the fact that the tradition is being performed by two professional female singers, Tijan Bai and Ritu Varma, who have gained notoriety through their performances on television and radio. Both women have traveled extensively around India and even as far as Paris and New York for festivals of India.

Although Candaini female performers are still unusual, the worldview expressed by both female and male performers of the Chhattisgarhi epic is a female-centered one.^[24] The heroine Candaini is the dominant character in the pair of lovers and the initiator of most of the epic action. She and other women are not portrayed as property to be exchanged and protected (as they are frequently depicted in the U.P. variants); rather, they are resourceful and take initiative, relying not on the ritual power of their chastity, as women frequently do in dominant-discourse narratives, but upon their own intuitive common sense.

Candaini's dominant role in the Chhattisgarhi epic first becomes evident when she makes the decision to leave her husband when their relationship is not fulfilling to her. Then it is she, rather than Lorik, who initiates their relationship; she sees him in the competition with her assailant Bathua and sets about to seduce him. In one version, she asks her brother to build a swing for her next to the path that Lorik uses every day to get to his wrestling grounds. As Lorik passes by, Candaini asks him to swing her. When he de-

clines, she curses him. This so angers him that he violently swings her, causing her to fall off the swing and giving him the opportunity to catch her.^[25]

The next time they meet, Candaini suggests a joking sexual relationship with Lorik by calling him her devar (younger brother-in-law), with whom such a relationship is permissible. Having grown up in the same village, they would normally call each other brother and sister, precluding a sexual relationship; changing the terms of address is often one of the first indications of a change in the nature of a relationship in Chhattisgarhi rural life and oral traditions. Finally, Candaini openly invites Lorik to visit her during the night, telling him how to get past various guards that stand at the entrance to her palace. As their relationship develops, it is also she who suggests and pushes for the elopement to Hardi Garh.

Candaini's resourcefulness and courage are illustrated by numerous examples from Chhattisgarhi episodes of the epic. In one performance, when the couple is eloping and their way is blocked by a flooded river, Candaini, not Lorik, figures out how to cross. She first procures a small boat from the ferryman stationed at the crossing. Lorik accuses her of negotiation of more than transportation with the ferryman, however, and in jealousy splits the boat and its owner in two with his sword. He then goes into the jungle and cuts down some green wood to build a raft, which, of course, immediately sinks. It is Candaini who knows it must be built with dry bamboo, tied together with lengths of a forest vine. When the ferryman's wife comes to bring him his breakfast and sees him dead, she immediately suspects the eloping couple of murder and creates a magical mouse to hide in their raft.

Halfway across the river, the stowaway mouse bites through the ropes holding together the raft. Candaini manages to reach the far shore, but Lorik does not know how to swim and starts to drown. The heroine unties her braid, jumps in, and saves him, presumably by pulling him ashore with her hair.^[26] Candaini's ingenuity and physical strength in this episode stands in sharp contrast to a similar scene in the U.P. variant in which Lorik's wife calls upon the power of her chastity (her faithfulness to her husband, sat ) to cause the waters of a river to part.

A female worldview is again reflected in a wonderful episode of the eloping couple's journey through a kingdom of all women. Candaini sends Lorik into the town to buy them some betel leaf (pan ) He is tricked by the pan- seller to follow her home, where she "keeps her best pan" (to feed pan to a member of the opposite sex in Chhattisgarhi folklore is often to initiate a sexual relationship, or may be used as a metaphor for intercourse itself). Once the pan-seller has trapped Lorik in her house, she threatens to beat him with a bamboo pole and stuff his skin with straw, poke his eyes out with a needle, and, finally, brand him with a hot crowbar unless he promises to marry her. After each threat, he gives in, only to recant a few minutes later. Finally, Candaini comes looking for her partner and meets the pan -seller in

the bazaar. The pan -seller begs the epic heroine to help her with a man who refuses to marry her. Candaini discovers a sari-clad Lorik in the woman's courtyard, having been so disguised so as to hide his male identity in the all-female kingdom. Once his identity is made known, the two women agree to play a round of dice to determine who will win him as husband. Note that although this is a reversal of the gender roles in Sanskritic, male dicing games, which are played to win a woman in marriage or as a sexual partner, the motif of women dicing over the fate of men is found in other Chhattisgarhi folk narratives. Candaini triumphs in her dice game with the pan -seller and frees Lorik from his captivity. One can hardly imagine the martial hero of the U.P. variants of the epic permitting the pan -seller's physical humiliations to be forced upon him or to be dependent upon rescue by a woman in a women's world.

Even in several episodes in which Lorik takes the primary role in a confrontation, it is still a woman who tells him how he can win, and the means are rarely traditional "heroic" ones. The first such confrontation is between Lorik and the Camar Bathua. Candaini's mother says the only man who can successfully confront Bathua is the "sporting hero Lorik."^[27] Lorik's wife, Manjari, however, warns him that he will not be able to defeat the Camar in a normal wrestling competition. She suggests that the confrontation be one in which the men are buried up to their waists in separate pits by the other man's wife. The man who can first get out of his pit and beat the other man will be the winner. Lorik agrees to this. When the women are burying each other's husbands, Manjari begins to throw gold coins on the ground. This so distracts the Camar's wife that she only loosely packs the dirt around Lorik and then runs to pick up the coins. Meanwhile, Manjari has time to bury Bathua firmly. When the time comes for the men to try to get out of their pits, Bathua is stuck. Lorik jumps right out and soundly defeats the Camar.

Candaini's beauty and a male's desire for her are the source of several major conflicts in the Chhattisgarhi variant, and in these situations she is physically threatened and needs physical protection like the women in the U.P. versions. As we have seen above, however, if Lorik is left to his own strength and resources, he may or may not be able to provide Candaini with the necessary protection. Judging by her resourcefulness in other situations, one senses that if she had no male to protect her physically, Candaini would come up with alternative solutions. Furthermore, when her chastity is protected by Lorik, only her personal honor is at stake. The personal honor of a Chhattisgarhi Raut woman does not necessarily extend to the honor of her family and caste. One of the main episodes in the U.P. variant making this connection between the three levels of honor—the story of Lorik saving Manjari from having to marry a king outside the Ahir caste is not present at all in the reported and performed versions I have seen in Chhattisgarh. The other U.P. episode making this association explicit is Lorik's defeat of Bathua,

which saves the honor of Candaini and the Ahir caste. In Chhattisgarhi versions, Candaini's mother, in asking Lorik for help, is not as concerned with honor as with physical safety: Bathua is terrorizing the entire village, so that everyone is afraid to go out of their homes, and the cattle are dying from lack of fodder and water.^[28]

As the role of women increases in importance in the Chhattisgarh variant, we have seen that the character of the hero also shifts. He is no longer the ideal protector and warrior. When he does engage in battle, he usually employs nonmartial, often unheroic, means to win; when the battle is honest, he battles without the aid of large armies, elephants, or other military paraphernalia, which support him in U.P. versions. He is a simple cowherder whose weapons are his own physical strength and herding staff. In this epic variant that centers around elopement love, the hero's status as warrior is less important than that as lover.

An important way Lorik's lover role is highlighted in Chhattisgarh is through the elaboration of the character of Bawan Bid, Candaini's impotent first husband. His impotence and passivity give emphasis to Lorik's sexual prowess and virility. One naca performance portrayed Bawan as a buffoon who is always wiping his nose with his fingers and licking the snot off of them. During the twelve years of his impotence, he wanders the forest as a sadhu (religious ascetic) but is easily frightened by any strange noise and welcomes Canda's company when she comes to the forest to try to persuade him to give up his asceticism. Both Satnami and Raut versions agree that Bawan Bir's impotence is the result of a curse cast upon him by the goddess Parvati. A Satnami version of the curse incident recounts that Bawan used to tease the Raut girls who picked up cow dung in the jungle every day. One day, Parvati took the form of one of these girls, and Bawan began to tease her. She revealed her true form to him and cursed him with impotence for his audacity. The Raut version says that one day Bawan Bir left a leaf cup of milk sitting on the ground, from which he had drunk. Shiva, in the form of a snake, came up to the cup and drank out of it. Subsequently, he began to acquire the rather obnoxious personality of Bawan Bir, quarreling with and scolding his wife, Parvati. When Parvati realized why this personality transformation had occurred, she cursed Bawan with impotence.

Bawan Bir is also impotent in the U.P. epic variant; the fact, however, is given little elaboration in the performances reported by Pandey. In the Awadhi version, we learn of the impotence in a single line. The performer tells his audience that Bawan is a eunuch with no hair on his body, but he gives no reason for the condition, although the audience knows the reason is a curse from Durga. Another story circulates in Ballia, U.P., that Bawan encircled his large penis around a Shiva liniga, a phallic representation of Shiva, and that the god cursed him with impotence for trying to compete with him.^[29] Whatever the reason, Bawan's impotence is overshadowed in the

U.P. versions by his martial nature. He, too, is a powerful warrior when he battles and defeats Lorik's older brother and confiscates all of their family wealth and cattle, and again in the battle in which Lorik regains this wealth at the end of the epic.

Appropriating the Performative "Exterior" of the Tradition

As varying social groups have appropriated the epic within its traditional performance range from U.P. to Chhattisgarh, both the textual content, "interior," and performative "exterior" of the tradition have responded to and reinforced the identities of these groups. While the epic in U.P. has served to represent the caste both to itself and to other castes in the region, in the Ahirs' effort to consolidate and raise their caste status, in Chhattisgarh, traditionally, it has been self-reflexive, mirroring the region to itself, contributing to a Chhattisgarhi self-awareness of difference, particularly, for example, regarding the status of women and marriage customs.

To say that the region has "appropriated" the epic in the Chhattisgarhi contexts described above is, perhaps, to give unwarranted self-conscious agency to a relatively loose social body.^[30] In the last ten to fifteen years, however, "appropriation" is the word to describe the emergence of "new" performance contexts and audiences for Candaini, both within and outside of Chhattisgarh. The tradition has been self-consciously crafted and packaged for both Indian and international audiences as representative of the region (not a particular caste, class, or gender). This appropriation coincides with increased availability of mass media technologies and communications (television and radio), as well as the academic and popularized interest in "ethnicity" that has developed in India over the last decade (as evidenced, for example, in international festivals of India and modified "ethnic dress" as high fashion among the upper middle class of urban India).

Radio, television, and the cassette industry have provided significant new contexts for folklore performance, including the epic. Akashvani (All India Radio) has both local (Chhattisgarhi) and national (Hindi) programming, with regularly scheduled folklore programs as a part of both. Such programming expands the social boundaries of groups to whom many performance genres are traditionally available; songs that women used to sing among themselves while transplanting rice or in the privacy of their courtyards are now blared over speakers from tea stalls and bus stands in urban neighborhoods and village main streets. Although prior to its appearance on media channels, the epic was spoken of as being "Chhattisgarhi," its performance on radio and television has solidified the epic's geographic regional identity, drawing its boundaries more literally than "live" epic performances, since such programming is limited to specified districts but also

has become uniformly available throughout those districts, even in those villages and neighborhoods in which the epic has never been performed.

In 1985 when I was trying to trace down various performance traditions in the burgeoning town of Dhamtari, I was frequently asked why I didn't simply turn on the radio on Wednesday afternoons for Akashvani's Chhattisgarhi folklore programming, from which I could simply tape the "best singers" directly from the radio, without all the complications of live performance. Both radio and television performances are taped in rather sterile recording rooms, with specific time frames (much abbreviated from any live performance), without a live audience with whom to interact and jointly craft the performance. Further, these performances are taped under the direction of radio-station personnel who often have certain aesthetic criteria that they feel "typify" the particular genre in question, although most of them are not "native" to the region. These criteria include less repetition, more instrumentation, and particular voice quality and stage presence of singers. When I articulated some of these differences between a half-hour radio performance of a Candaini episode and its elaboration during a four-hour, late-night epic performance in a village square, adding that there was little manoranjan (literally, "entertainment," but with the implication of emotional satisfaction) hearing it over the radio, these same informants generally wholeheartedly agreed, although they often felt somewhat differently about television performances. In the mass media, the epic is taken out of its traditional performance contexts and recontextualized in a setting in which it "represents" on an external performative level through style and instrumentation, but in which its interior is frozen, unresponsive, generic.

Radio and television programming has affected the careers of particular singers who are chosen and promoted by the staff. This has been the case especially for the female epic performers referred to above. Once heard repeatedly on local radio or television, they are then invited to statewide folklore singing competitions and folklore festivals in major urban centers, such as New Delhi, Bombay, Calcutta, and even London, to "represent" Chhattisgarh. As individual singers themselves become famous, the genres associated with them become more popular as well, both within and outside of the region.

Representative of the growing "academic" interest in Chhattisgarhi folklore by members of an urban, educated class, who have not traditionally participated in epic performance as singers or audience, is the playwright/director Habib Tanvir, a Muslim born in Chhattisgarh's heartland, now living in New Delhi. His troupe, Naya Theatre (New Theatre), consists of a majority of actors and actresses drawn from Chhattisgarh's villages, nonliterate "traditional" dancers and performers. Along with his interest in experimental theatrical forms, an overriding concern of Tanvir's is to promote the appreciation and preservation of Chhattisgarhi folk performance traditions.

To this end, he has held numerous folklore workshops in Chhattisgarh itself for performers of these traditions. The aim of these workshops is for performers to share with each other their repertoires and for Tanvir himself to document them, often then integrating their themes and forms into his "new theatre." In a 1985 interview, while in Calcutta staging his play Charan Das Chor, Tanvir explained this task as follows:

I had to work in two ways. I had to purify their forms and themes to make them more authentic and contemporary. I found that the folk form was getting spoiled and diluted by the combined influence of urbanization, mass media, and low-grade Hindi films. The first part of my job was to weed out the falsities and purify the form. Not for the sake of purity, but because the folk form is both beautiful and a powerful medium for a message.^[31]

For one of his Chhattisgarhi folklore workshops, held in the late 1970s, Tanvir called together the "best" Candaini singers he had met in his tours of the region. Singers from a range of castes shared their stylistic and thematic repertoires. One of these singers was the Satnami Devlal; he was also one of several workshop participants chosen to go to Delhi to work with Tanvir for several more weeks. According to Devlal, Tanvir stressed to the singers the importance of keeping their tradition alive, and that one of the ways to do this was to keep the entire narrative in performance, singing it "from the beginning," when the hero and heroine were children and so on, rather than focusing so exclusively on the elopement episode.

I attended (and was the primary patron of) one of Devlal's Candaini performances that resulted in what I have called a "failed performance," with most of the approximately 200-member audience walking away within the first hour of the performance. I have analyzed the reasons for this elsewhere,^[32] but one important reason cited by audience members was that he was singing "stories we don't know," from this reconstructed larger repertoire of epic episodes. What they expected and wanted to hear was some variation of the elopement narrative. Devlal was also experimenting with form. He framed the performance as if it would be a naca, a form influenced by the corrupting "low-grade Hindi films" to which Tanvir referred, but did not wear the expected costuming, did not perform the expected "song and dance"; so that another major complaint of the dissatisfied audience was that "he should have worn a sari."

Literate and nonliterate residents of Chhattisgarh alike have voiced, over the years during which I have returned to Chhattisgarh since 1980, a certain unease about Tanvir's appropriation of Chhattisgarhi folklore for display outside the region. Even as he is attempting to promote an appreciation of the region and its performance genres, many inhabitants feel that the process serves no benefit to Chhattisgarh itself. Several residents of the town in which Devlal performed, who have known him since his childhood and over the

years in which he developed his epic-singing skills, complained that when Tanvir chose particular singers such as him, they often forgot the Chhattisgarhi roots from which they had come, were no longer satisfied to sing in "traditional" contexts, demanded too much money, and were no longer responsive to their audiences.

Drawing on a workshop held for Candaini performers, Tanvir later wrote a script based on the epic to be performed by his Naya Theatre troupe, called Son Sagar, the name of one of Lorik's beloved cattle. I was able to sit in on one of the rehearsals of this play in 1985. The actors and actresses of the troupe are Chhattisgarhi, as is the language of the play; it opens with a traditional vandana invocation to the goddess Sarasvati) and is framed and interspersed with lines sung in the traditional git style. But, performed on a modern stage, outside of traditional performance contexts, it is not Chhattisgarhi Candaini, at least as it is understood by the folklore regional community. Although, according to Tanvir, there is room for improvisation, the lines are relatively fixed, memorized, and unable to be responsive to particular contexts and audiences—and if even if they were, the performance contexts would not be Chhattisgarhi.

In newly emerging performance contexts such as radio, television, and the modern stage, the epic has become decontextualized, so that it can be performed anywhere. In a sense, the audiences are not "live"; they are dispersed, unknown, and unseen. Further, the Chhattisgarhi dialect of the sung "text" is itself often not understood fully, if at all, by newly emerging Hindi or English-speaking audiences. What characterizes the epic for these audiences is its performative exterior, the unique singing and instrumental styles of epic performance, which themselves become relatively frozen, or at least enough so that they are recognizable as "Chhattisgarhi." In these contexts, the epic tradition has become an artifact, frozen in time and space, held up for admiration and nostalgia; thus though perhaps unresponsive to what may be perceived to be more traditional shifting performative and social contexts "on the ground," so to speak, it is responsive in a very different way to newly emerging middle-class audiences.

The Candaini living epic tradition has shown a tenacious ability to adapt to shifting and emergent performance contexts: to take up the cause of a caste trying to raise its status in U.P., and in Chhattisgarh to integrate non-Raut singers into the circle of its performers and instrumentation and the naca song and dance into its performance style as it competes with Hindi cinema and video halls. Over the last decade, however, while performers continue to be drawn from low-caste groups, the performance contexts of the Chhattisgarhi epic have bifurcated. The first are those live performances in traditional, late-night, open-air village squares in which primarily lower-class/caste audiences continue to interact with and help to shape the interior "text" of the tradition. It remains to be seen how flexible this interior

can be in its interaction with a rapidly changing social world, how long or in what ways its performances can compete with video halls, movie theaters, and television, and who the singers and performers will be in the next generation. The second context is physically distanced from its audiences, on stage or over the airwaves, audiences that now include an increasingly educated middle class. For these audiences, the epic's narrative interior no longer reflects "who we are," but its performative exterior may remind them nostalgically of "who we were."

Works Cited

Babb, Lawrence.

1975.
The Divine Hierarchy: Popular Hinduism in Central India. New York: Columbia University Press.

Bauman, Richard.

1977.
Verbal Art as Performance. Rowley, Mass.: Newbury House.

Blackburn, Stuart, and Joyce Burkhalter Flueckiger.

1989.
Introduction to Oral Epics in India, edited by S. Blackburn, P. Claus, J. Flueckiger, and S. Wadley, 1-11. Berkeley: University of California Press.

Blackburn, S., P. Claus, J. Flueckiger, and S. Wadley, eds.

1989.
Oral Epics in India. Berkeley: University of California Press.

Bose, Nandini, and Pradipta Bhattacharjee.

1984.
"Folk Theatre: Neither Exclusive nor Esoteric." The Sunday Statesman, April 22.

Chadwick, Nora Kershaw, and Victor Zhirmunsky.

1969.
The Oral Epics of Central Asia. Cambridge: Cambridge University Press.

Coccari, Diane.

1984.
"The Bir Babas of Banaras: An Analysis of a Guardian Deity in North Indian Folk Hinduism." Ph.D. dissertation, University of Wisconsin.

Elwin, Verrier.

1946
Folksongs of Chhattisgarh. Madras: Oxford University Press.

Flueckiger, Joyce Burkhalter.

1996.
Gender and Genre in the Folklore of Middle India. Ithaca, N.Y.: Cornell University Press.

1989.
"Caste and Regional Variants of an Epic Tradition." In Oral Epics in India, edited by S. Blackburn, P. Claus, J. Flueckiger, and S. Wadley, 33-54. Berkeley: University of California Press.

1988.
"'He Should Have Worn a Sari': A Failed Performance of a Chhattisgarhi Oral Epic." The Drama Review T 117: 159-169.

Flueckiger, Joyce Burkhalter, and Laurie J. Sears, eds.

1991.
Introduction to Boundaries of the Text: Epic Performances in South and Southeast Asia, 1-16. Ann Arbor: Center for South and Southeast Asian Studies, University of Michigan.

Johnson, John William.

1980
"Yes Virginia, There is an Epic in Africa." Research in African Literatures 11: 308-326.

Khedkar, V. K.

1959.
The Divine Heritage of the Yadavas. India: n.p.

Mandelbaum, David G.

1972.
"Cultural Adaptations and Models for Mobility." In Society in India:

Change and Continuity, 2: 442-467. Berkeley: University of California Press.

Miller, Barbara D.

1981.
The Endangered Sex: Neglect of Female Children in Rural North India. Ithaca, N.Y: Cornell University Press.

Narayana Rao, V.

Forthcoming
"What is Folklore in India? "In South Asian Folklore: An Encyclopedia, edited by Peter J. Claus and Margaret Mills. New York: Garland Publishing.

Oinas, Felix.

1972.
"Folk Epic." In Folklore and Folklife, edited by Richard Dorson, 99 -115. Chicago: University of Chicago Press.

Pandey, S. M.

1982.
The Oral Epic Canaini. Allahabad: Sahitya Bhawan Pvt.

1979.
The Oral Epic Loriki. Allahabad: Sahitya Bhawan Pvt.