Rain, God, and Unity among the Kotas RicHARD KENT WoLF .. i 10 Father god who, with the black cow, set the first pillar Where are you? ... Lamp lamp where are you? come god! Are you in torf,ba·l? come god! Are you under the ne· rl tree? come god! ... Are you in ponic? come god! 'vefke· vefke·', a Kota song from Kolme·l In the story behind the epigraphical song and others like it, a community finds that divine presence depends on communal harmony, unity, and righteous living. Among the Kotas it is a formulation found in various guises: the folklore of devotional music, stories of divine locales, and the ritual structure of worship. In this essay I will analyse all of these forms in a rain-making ceremony the Kotas perform every summer, just before the monsoon. 1 The rain ceremony and the eclectic approach I have employed to discuss it will, I hope, allow us to explore in depth what unity means in Kota society, how Kotas have interpreted its relative absence as they have assessed their past and present, and how these questions may have been affected by notions of 'tribe,' Hinduism, and morality projected upon the Kotas from without. I adopted the tactic, now well established in anthropology, of gleaning central cultural issues through the analysis of one event in critical detail. What I hope to accomplish, on a broader methodological level, is a successful narrative synthesis of diverse elements: historical-anthropological, ethnomusicological, personal narrative (with several voices), descriptive, and symbolic-analytical. 232 Richard Kent Wolf Rain, God, and Unity among the Kotas 233 The first part of this essay provides a broad context for the analysis of the ceremony. There are three subsections. The first considers what it means for the Kotas to be a 'tribe' in south India. The second provides historical background on two anthropologists studying the Kotas in the thirties: Murray B. Emeneau and.the late David G. Mandelbaum. The third section introduces K. Sulh, a Kota reformer and a salient voice in the anthropological record of the tribe. The second part of this paper is a discussion of how the music of the rain ceremony derives meaning from rituals and mythohistorical events extrinsic to it. The third part is a narrative analysis of the ceremony as it unfolds, alternating between a macroview of the whole ceremony in performance and microviews examining the significance of each part in some detail. Finally, I o~fer a bro~d analytic view of the entire ceremony, including suggestiOns as to Its origins. To begin with, I will provide a rather rarified survey of the arguments and issues that will emerge, fleshed out and hopefully lucid, from the analysis. The ceremony is one among several. Kota ceremonies which institutionalize an association between VIllage unity, righteousness and the efficacy of the Kota gods. The t.rope Gf unity is evidenced at several analytical levels. In analysmg the overall form, we note first that the rain ceremony consists of a procession to all the 'god place~' that bor.der the vill~ge and th; sequential performance of a speCial repertOire of twelve god tunes . I would suggest that, by virtue of belonging to one cerem?ny, ~he constituent places and tunes belong to one category-albeit a .ntually constituted one-and thus establish a 'unity' as re.gards ntual function.2 Why this should be the case, and how this bear~ on behaviour and religious efficacy, will become clear as we consider what these places and musical pieces mean to the Kotas: . Unity in the physical space Kotas occupy is an~ther vanati?n on this trope. Spatial unity is affirmed as the processwn t.r~c~s vil~a~e boundaries and thereby secures the presence of diVImty Withm them. Phenomenologically, the ceremony would seem to create an 3 experience of temporal unity, by 'binding' the past with the present, and psychic unity, by attempting to insure that all the Kotas are of 'one mind'. It can thus be considered an argument that the Kotas perform for themselves, asserting fundamental social-cultural harmony.4 The ceremony exemplifies what James Fernan?ez has ~e­ scribed as 'returning to the whole' (1986: 188-213) m that, hke revitalization movements and 'perhaps all religions', it is 'fundamentally interested in restoring the relatedness of things' ( 1986: 191). When the integrity of the village landscape is recreated in procession, the power of and belief in the gods affirmed through prayer, and the association between land, place and person further established through the power of musical performance and the divine stories this music recalls, rain will come, as they say, 'without fail'." Stated more generally, when unity and righteousness have been 'achieved' (ritually, at least) the Kota gods behave as Kotas would have them behave and the community thrives: rain falls, crops grow abundantly, cattle multiply, children are born and survive, diviners divine and ritual leaders stay alive. The theme of unity in the ceremony as a whole is mirrored in constituent domains such as verbal discourse, in the ritual phrase, 'Is everybody joined?' and gesture, in the formal sharing of tobacco. In addition, while the enactment of the ceremony establishes a series of analogies and correspondences between its components, the landscape and music involved in the ceremony are, not surprisingly, symbolically loaded in and of themselves. 'Gods' places' (devrd erm) and 'god tunes' (devr ko!) generally encode perspectives on multiple pasts which seek to preserve the equation of divine efficacy with righteous living as a 'Kota'. 6 Encoding and re-encoding occur in the recounting of stories connected with these places and tunes: each story recalls an event in which divine beings appear to have 'come to the rescue' in moments of stress. The need to restore wholeness is felt 'particularly in times of stress-where literal routines break down and where we are constrained by false or moribund categories' (Fernandez 1986: 205-6). The tension between the collective creation of the whole, as embodied in the rain ceremony (and, in other ways, in other rituals as well), and the tendency toward community disintegration, embodied in disagreement, the negotiation of modernity, and physical disasters (disease, drought, or inability to complete a ritual), constitutes one of the enduring problematics of Kota society. This tension, I believe, provides an overall clue to understanding the rain ceremony. The problems of disunity, breaking of rules, changing of norms, and so forth, which are in a sense ritually countered by the rain ceremony, are at the same time presupposed by the very need for 234 Richard Kent Wolf Rain, God, and Unity among the Kotas 235 its yearly performance. During my investigation of the history of the ritual, I came across materials that may allow us to speculate about the creation of the rain ceremony. At the conclusion of this paper I will consider evidence suggesting that the ceremony may have been invented (or modified into its present form) in 1933, in response to a particular set of modern circumstances which threatened the community. As a strategy designed in a sense to mirror and in a sense to exemplify and evoke the disunity that I describe as a raison d' etrefor the ceremony, my account includes conflicting statements the Kotas have provided us about the ceremony. I also include conflicting accounts in order to avoid erasing the problems inherent in attempting to abstract, from situations 'on-the ground', coherent cultural systems. One is constantly reminded that individuals are involved in that peculiar process of culture-making which unfolds during ethnographic research. 7 It seems important, therefore, to consider situations in which people argue over the details of what occurs in a ritual as well as what it means. On the other hand, not all individuals can be represented, and not all voices are invested with equal authority to speak for the community. Thus, even though I feel compelled to counter implications that the Kotas, who number only 1,500, are an undifferentiated whole, I will not discredit the value of understanding 'the Kotas' through the thoughtful and creative products of unusually talented or otherwise unconventional individuals. Striking a meaningful balance (or synthesis) of synoptic, holistic views (the Kotas, the ceremony, the structure) with analyses of how meaning is actually created in specific times and places becomes, then, a central task. Understanding a cultural unit like the rain ceremony in structural terms, for example, must be broadened through a situational perspective-a perspective which recognizes that individual agents (the men who perform the ceremony, decide how the ceremony is to be performed, or argue about what a particular gesture means) create an event while at the same time subtly registering broad macroscopic cultural schema within ithere, the grammar of 'unity' and the discourse of 'tradition'. 8 This situational perspective involves not only actors of the present but also figures from the past, particularly because some accounts and interpretations of the rain ceremony are historical, appearing in the fieldnotes of the late David G. Mandelbaum. To use Mandelbaum's material critically it will be necessary to examine the dynamics of fieldwork during his first trip to India ( 1937-8); and, more generally, the problems associated with using other people's fieldnotes (cf. Smith 1990; Lutkehaus 1990; Sanjek 1990, passim), and the multiple histories which are produced when ethnographies of the past are introduced into frameworks based on fieldwork of the presentY A practical problem follows from attempting to combine these diverse sources into a single account. The dimensions of the problem include: (1) My observations of the rain ceremony in 1991 and 1992; (2) Kotas reporting on how the ceremony is 'supposed' to be done; (3) Kotas reporting on how the ceremony has actually been done, based on their memory; (4) Kotas reporting on how the ceremony has actually been done, based on what they have heard from other Kotas; (5) Kotas, no longer alive (like Sulli), whose reports are preserved in David Mandelbaum's fieldnotes or publications; (6) Kotas commenting upon what other Kotas have reported or what I have observed. The differences between these six perspectives are not merely problems of narrative, of course, but interesting problems in and of themselves. By sifting through, juxtaposing and comparing the histories, descriptions and interpretations I encountered in the process of understanding the rain ceremony, I hope to lead the reader through an epistemological journey that is in some ways like fieldwork (confusing, open-ended, difficult to control). But multiple perspectives, disagreement, and the contestation of 'history' itself are not only in the nature of fieldwork, they are embedded in the ways the Kotas view themselves as a people. Views of the Kotas: Inside and Outside, Then and Now THE KOTAS AS A 'TRIBE' The Kotas number about 1,500 (by their own estimates in 1990) and live in seven villages scattered widely across the Nilgiri Hills. The government of India lists the Kotas as a 'Scheduled Tribe'-a 236 Richard Kent Wolf Rain, God, and Unity among the Kotas 237 designation which purports to safeguard the community against 'exploitation' and to make it eligible for government welfare programmes. The Constitution however does not specify what constitutes a 'tribe', 10 and as a result the label does little to characterize the Kotas as a people. I do not wish to enter into a lengthy discussion about what a 'tribe' in India is, or whether such a term is applicable11-clearly the term is a polythetic one ( cf. Needham 1975). Instead, I want to explore the popular images inspired by the word 'tribe,' and its vernacular equivalents. The Hindi word iidiviisi, meaning 'original inhabitant', and the Tamil term pafankufi makka!, 'ancient race people', are of fairly recent origin. To my knowledge there are no terms in ancient Tamil or Sanskrit which indicate that modern notions of tribe were ever operative in ancient Indian principles of classifying people (George Hart 1993, pers. com.). Even though individual tribe-like (by modern standards) communities were named in many early texts, there were no terms for identifying 'tribe' as a unit of classification. What, then, does that term 'tribe' conjure up in the minds of Indians today? In a popular construction, the tribe is hypostatized and somehow equated with the primeval lndian. 12 'The Scheduled Tribes are the real sons of the Indian soil. They are the original people of India who have variously been known as backward tribes, primitive tribes, criminal tribes, Adimjatiand everything else' (Mehta 1991: 9) The image ramifies: as 'sons of the soil', the tribals are, if! eft to their own devices, at one with nature, self-sufficient. They hunt and gather, excel in crafts, are experts in the use of traditional medicines and are capable of sorcery. Tribal gods too are somehow seen to possess an abundance of divine power. Although, or perhaps because, these tribals are more imagined than real, the myth prevails. Many people assume that the tribes of today live in an eroded state, corrupted by modern society from some pre-modern natural state. In modern India, these populist notions of tribe cannot be considered in isolation from the way any community-a 'tribe', so calledconstructs itself as a tribe. And among the Kotas many of these notions do indeed dominate discussions about tribalness. Some brief examples may clarify what I mean. R. Mathi and S. Cindamani, two women in their late forties with whom I had developed a relatively familial relationship, were concerned with attempts by another community to claim tribal status. 13 Their rhetorical questions expressed exasperation, 'How can they be tribal? Do they know blacksmithing? Do they know how to make pots? Are they musicians?' Many Nilgiri tribals think of themselves in similar terms. R. Mathi's son, Duryodana 14 (my research assistant, close friend and companion), used to enjoy meeting other tribals and enthusiastically accompanied me whenever I visited other tribal villages. In such situations I observed him engaging in lively conversation with other tribal men about honey-collecting, hunting and trapping, and the characteristics of the local flora and fauna. He felt, and believed that other tribals felt, that the Nilgiri tribes share authority in domains of knowledge connected with the forest, craftsmanship and music. These feelings can be said to separate the ways tribal communities view each other from the ways they view non-tribals. The image of the tribal as the 'primeval Indian' is refashioned and reaffirmed in public displays such as Republic Day, when in the town ofOotacamund the Todas, Kotas, and lrulas are called to sing, dance, and play in full costume. The image tribals themselves construct of intertribal communities is affirmed in similar performances, both participatory and for display, in modern revitalization movements. 1" MA·MU·L Al':D 0CMU·L Moving now to a discussion of how Kotas construct themselves as a community, it is possible, and indeed desirable, to deal with concepts growing directly out of Kota discourse. Once such concept is associated with the word, ma· mu·l, a term (ironically) of Arabic origin, meaning for the Kotas 'the old way', or 'tradition, custom' . 16 Its opposite, ocmu·l, has connotations of foreignness and inauthenticity on one hand, and progressiveness, modernity and vitality on the other. 17 The Kota word ma · mu. l frequently refers to practices of the past-both the 'everyday' variety and those formalized ritually: ~ames, rituals, songs, dances, language, dress and ways of threshmg can all be discussed in terms of their ma· mu·l-ness. The term is also applied to people who support the old ways or who are themselves very old. 18 A related term is kaf (DEDR 1147), which means 'knot', and by extension 'custom'-that which has been 'tied,' or established firmly. 238 Richard Kent Wolf Rain, God, and Unity among the Kotas 239 Ocmu· l practices are regarded ambivalently: at times enthusiastically by those who regard new practices as somehow inauthentic, or threatening to the integrity of the Kotas; and at times sceptically by those who regard new practices as somehow inauthentic, or threatening the integrity of the Kotas. Those who introduce new practices are sometimes regarded with suspicion as well (see, e.g., Mandelbaum 1941; 1960), especially if these people are seen to benefit personally from such new practices. It seems reasonable to speculate that as the primevalist view of the tribal became increasingly disseminated in the popular media, and as it seems to have grown into one of the informal philosophical bases of public policy, tribes like the Kotas became increasingly self-reflexive about themselves not only as a distinctive ethnic group, but also as a tribal ethnic group. This reflexivity, at least currently, is expressed in the relationship between and the values assigned to ma · mu· l and ocmu· 1. 19 This all becomes relevant to the rain ceremony because the form and content of the ceremony itself point to a conscious awareness of tradition, and the past, that calls for some explanation. Ma · mu· l and ocmu· l are of course very close to what English speakers call tradition and modernity and indeed the issues involved with the tensions between them tend toward cultural universals. My intention here is to bring into focus how the Kotas deal with these issues within their cultural framework. But to do this it is necessary, of course, to recognize the complex historical matrix which gave rise to Kota culture in modern India in the first place. Al\'THROPOLOGISTS AMO~G THE KoTAs: DAVID MuRRAY G. MANDELBAUM AND B. EMENEAU It is often difficult to evaluate in an)thing but general terms the dynamics of influence between communities in contact (let alone broad entities like Hindu and tribal, Indian and English), the specific ways in which particular communities have responded to colonial and other rulership, or the impact of individuals (writers, administrators, scholars, or cultural leaders) on the history of a community, without considering suggestive historical moments, periods in which persons, places and events intersect in ways that impress upon the cultural history of a community a configuration that appears unique and lasting. Here I wish to consider one such consequential period, both because this consideration will facilitate a critical assessment of data from this period and because it will help situate the history of the rain ceremony itself. In the period beginning some ten years after a disease decimated Kolme ·1 village, two scholars arrived to study the Kotas: the Canadian linguist Murray B. Emeneau (b. 1904), who had by then (early fall of 1935) already earned a Ph.D. in Latin, Greek and Sanskrit ( 1931) and studied ethnological linguistics with Sapir;20 and the late David Mandelbaum (1911-87), who had written a doctoral thesis (1936) on the Plains Cree, under the direction of Edward Sapir. Mandelbaum arrived in the Nilgiris in April 1937 after a brief stint in Kerala among the tribes of Travancore and the Jews of Cochin (Mandelbaum 1939a; 1939b). Emeneau was the first philologically trained North American scholar to take an anthropological interest in Indian culture, and certainly the first to apply philological methods to unwritten Indian languages. Mandelbaum was the first American cultural anthropologist to conduct fieldwork in India. By virtue of these circumstances along, both scholars were staking out new ground for American anthropology. But significantly new events were taking place among the Kotas as well. The thirties was a time of stress for the Kotas of Kolme -1 because after the lice-borne 'relapsing fever' struck, the village was left without ritual specialists ( ca· tmga· m) to conduct ceremonies in the traditional fashion. Partly in response to this rievastating event, which was believed to be evidence that the Kotas were committing some terrible wrong, and partly to improve their social status in relation to their Hindu and other tribal neighbours, a movement arose to advocate worshiping a new set of deities and another movement arose attempting to modernize Kota ways-particularly as regards the slaughtering of bovine, the seclusion of women during menstruation, and the male style of wearing the hair long. K. SULLI IN TIME PERSPEGfiVE Leading in some of the reforms was a Kota school teacher, K. Sulli, 21 a man schooled by missionaries and who, as a child, even wished to convert to Christianity; although prevented from doing so by his father, Sulli maintained a life-long interest in Christianity and belief in Jesus Christ. 22 Emeneau's contacts in Ooty summoned Sulli, the only English-speaking Kota at the time, to facilitate his study of the Kotas' language. Emeneau had established a routine 240 Richard Kent Wolf Rain, God, and Unity among the Kotas 241 and style of working with Sulli before Mandelbaum arrived. For a rupee a day, Sulli would come to Emeneau's room in Ooty and dictate stories. Sulli took leave of his teaching job throughout the period of employment with the Americans. When Mandelbaum arrived in the Nilgiris after three months of exasperating fieldwork with informants (in Travancore) whom he felt to be singularly lacking in volubleness, Sulli seemed to harbour a change in luck. But soon, judging from comments in Mandelbaum's field journals (Mandelbaum n.d.), Mandelbaum became frustrated with Sulli's unfocused loquacity: Sulli was a tap that could not be turned off, wandering from topic to topic, tangent to tangent, and embellishing stories and events with elaborate detail. Neither Emeneau nor Mandelbaum knew or learned Tamil while they were in India, 23 and neither of them actually learned to speak the Kota language with any fluency. Although Sulli was the only Kota with whom either scholar could communicate directly, 24 Mandelbaum did work with other Kota informants with the help of Christian Badaga interpreters. 25 And he was careful to collect alternative interpretations and points of view from non-Kotas. 26 If Mandelbaum was forced to rely heavily on Sulli, it was not uncritically, however, as this account of Sulli's traits as an 'ethnological respondent' illustrates: One is that his recollection tends to be neater and more integrated than was the historical actuality. His narrative artistry is apt to gloss over inconsistencies or irregularities and to make one episode follow another in logical, abstracted sequences that may have more aesthetic symmetry than historical exactness. Sulli has the kind of integrating, abstracting mind which one may consider to be more properly the prerogative of the ethnological theorist than of the ethnologist's informant. Secondly, he is like any gifted narrator of events in which he took part and of which he finds reason to be proud. He tends to figure much larger in his account than he may have in the event. But when he gives an impersonal account of, say, ceremonies, these traits do not prevaiL Sulli, in tum, was influenced by his work with the linguist and the anthropologist. In the first instance, the association with two whom he called 'our Europeans' added to his prestige. It is not unlikely that this association gave him the final impetus, in 1937, to take the decisive step of cutting his hair (Mandelbaum 1960: 307; emphasis mine). I emphasized the sentence concerning Sulli's accounts of ceremonies to make the point that informants, or 'native collaborators' (Clifford 1988: 49), may position themselves (or may be viewed by others) in different ways as regards what they describe, teach, or interpret (a ceremony, an incident, a song). Sulli figures himself prominently in village events but not in ritual. But the annotations to Mandelbaum's fieldnotes of the thirties suggest Sulli' s narration of ritual suffered from a different problem. Sulli used the English verb 'to say' both for actual utterance and for what Sulli presumed actors to be thinking. Dravidian languages contain a verb which is used both to mark the quotations of a speaker and to indicate what a speaker is thinking: in Kota the verb is in(DEDR 868); I suspect Sulli's idiosyncratic tendency to confuse what was said from what was thought may have occurred as Sulli translated Kota thoughts into English sentencesYBut in some ways the bigger problem was that Sulli tended to volunteer explanations of and intentions behind rituals in the first place. Sulli, like other informants described in Casagrande's In the Company ofMen ( 1960), 'understood, often with real subtlety, what an ethnographic attitude toward culture entailed' (Clifford 1988: 49). Yet his role in controlling the kind of information Mandelbaum ultimately collected cannot be understimated. For fourteen months out of my two-year fieldwork in the Nilgiris I lived in Sulli' s home village, Kolme ·L The accounts I collected (1990-2) differed from those Mandelbaum and Emeneau reported in the thirties and forced me to think about my own informants and their memories, motives, and reliability in ways that presumably would have been different if these writings did not exist. Because the Kota memory (actual and/ or legendary) of Mandelbaum was a positive one, Mr K. Pucan (an eighty-year-old man who was highly regarded in the village, both as an elder and as a musician) and other Kotas who disputed what I had learned from Mandelbaum's notes, would tell me that Sulli was lying about one matter, ignorant concerning another, or inventing data concerning yet another-rather than blame Mandelbaum for what they believed was inaccurate. Some of my elderly friends in Kolme ·l had personal reasons for discrediting Sulli. For example, one lady in her eighties, Pa. Mathi, claimed that Sulli had kept a Kurumba man (member of a tribe feared for its sorcery) in his attic and through him killed her ten children and her brother. She further claimed that Sulli's abundant Wealth was a result of sorcery. Sulli is not remembered, other than by his family, for positive contributions to the Kota community. In 242 Richard Kent Wolf &in, God, and Unity among the Kotas 243 fact, by some he is remembered as a self-aggrandizing troublemaker. His status outside the community in some ways weakened his authority inside the village. 28 Despite the fact that Sulli was trying to 'modernize' Kota culture, despite an apparent interest in Christianity, and despite his familiarity with European ways,29 it appears that Sulli still commanded a detailed knowledge of Kota history, stories and ritual. He may not have been considered any more 'right' in his day than he is now (if one were to conduct a poll), but neither can all the differences between accounts by Sulli and by others be understood as only ideological differences. The twelve 'god tunes,' performed in the Kolme ·1 rain ceremony are found also in the yearly god ceremony, devr. 33 Pragmatic considerations have rendered obsolete ritual practices originally associated with some of these tunes, and in other cases, stories and other contextual material concerning these tunes have been forgotten. 34 The effect of these contextual changes has been, in one sense, an objectification of these tunes as independent musical objects-objects which can be employed in new contexts related, in principle at least, with the contexts for which they are named. The Music of the Rain Ceremony The following cultural analysis of music in the rain ceremony will consider first the overall structure of the musical repertoire and how it functions within a ritual framework. Then I will consider in turn the symbolism of each constituent 'god tune,' numbering each set of paragraphs for convenient cross-reference later in the text. Instrumental tunes, or kof. 30 ofthe Kotas differentiate, mark, and partially constitute ritual occasions.:lJ There is a repertoire for dancing, a repertoire for funerals and a repertoire for 'god'. Each is characterized by broad stylistic features, but the criteria for distinction between repertoires are to a greater extent contextual and singular (piece by piece) than they are musical. Within each repertoire, a particular tune may be associated with a particular action (in a funeral, for example, one tune is associated with lifting and carrying the bier to the cremation ground). In general, the ways in which the structure of instrumental melodies coarticulates with ritual structures can be analysed as a system of indigenous classification in and of itself. Melodies are accompanied by one of two basic rhythmic patterns, each of which may be elaborated in different ways. In Kolme -1 the rhythms are called ca · da da · k (ordinary variety') and tirugana · ! da· k ('turning dance variety'). Unlike the complex rhythms performed for various occasions by Paraiyaror Cakkaliyarensembles on the plains in Tamil Nadu, rhythms performed by Nilgiri tribes are not used to differentiate ritual occasions-although they do differentiate dances. 32 § 1. The longest and decidedly most important tune among these is called the o-la · gicd koJ35 or gury terdd kof ('temple opening' tune). It is used during the annual ceremony of opening the three Kota temples and during other rituals associated with this act. Although the opening of the temple itself is not among the rituals that have been discontinued over the years-it is one of the central features of the god ceremony-many accompanying activities have been abbreviated. For example, the mundka· no· ns (ritual leaders) throw grass onto the roof of the temple to suggest symbolically the ancient practice of rebuilding the temple each year. The 'temple opening tune', which is performed to call god regardless of whether or not the temple is directly involved, is in a sense the hypostasis of divine music. As such, this tune also constitutes a necessary beginning of the rain ceremony. The other eleven kofs are named as well. These names are one of the means through which each kof comes to recall a particular context, story or bit of Kota history or a deity, when the original practice which gave rise to these associations has been lost. Knowledge of these kofs and their stories is rather esoteric and for this reason it was difficult to cross-check some details. Because Mr K. Pu~an is one of the most widely respected musicians, his conception of mstrumental music should be of interest to us. But we must not accept Pucan's authority unconditionally: Mr S. Raman, another respected kof player, insists that some kofs do not have names (for example the difference between the 'goddess' kofand 'father god' ko! ?e believes, is factitious). 36 Nevertheless, the devr kof 'lore' was not Invented for my benefit 37 and provides useful insights into Kota cosmology and the rain ceremony. The general point is that the god tunes are 'meaningful' to Kotas 244 Richard Kent Wolf Rain, God, and Unity among the Kotas 245 in different degrees, not so much in different ways. A few Kota men and probably even fewer, if any, Kota women recognize each tune, know one or more alternate names for it, and know whatever story may be connected with it. More Kotas know that the tunes have names, and know some of the names, but do not know which tune is which; they may or may not know any extramusical associations. Finally some people may or may not recognize that tunes are god tunes by listening, but know that such a repertoire exists and know, because of the performance context, when the tunes are being performed. To them, the tunes are simply associated in some way with the gods. THE STORIEs BEHIND THE DEVR Ko~ §2. Some of the devr ko!s, or 'god tunes,' can be characterized as such because they are literally named after deities. The origin of the shortest such tune, the ni-lgiri co· ym ko! or 'Nilgiri god tune', is obscure, as is knowledge concerning the deity. According to Mr Valmand Kamatn of Kolme ·1, a former mundka · no· n, 38 the 'Nilgiri god' is a primordial Nilgiri divinity who resides in a place called Talko. r~9 at the top of a hill where buffaloes were tended (mala·'! impay!) . After some unspecified conflict or ritual misconduct the deity moved to Rangasami Peak. The ayr;,o· r mundka· no· n (the ritual leader for worship of the 'father god') of Kina· r village recalled a legend in which the Nilgiri co· ym revealed himself to a Kota and an Irula who were walking together. Both Kotas and Irulas began to worship the Nilgiri god at that same spot until some quarrel separated them: the Irulas began to worship on Rangasami Peak and the Kotas near their villageboth at the same time of the year (May). The lrulas are priests for 'Nilgiri Ranga,' the deity of Rangasami Peak. 40 It appears there is some identity between 'Rangasami' worshipped by the lrulas and 'Nilgiri God' of the Kotas. 41 Though Kotas from Kina. r village and Parga· r village worship at and attend Rangasami festivals it is doubtful that the deity is popularly identified with their own Nilgiri co· ym. The location of the 'original Nilgiri God' (that is, of the origin stories) was probably the environs of Kina. r village (as opposed to other Kota villages) because this village is relatively near Rangasami Peak. Today a small shrine for Nilgiri God stands just at the border of Kina· r village, where a festival honouring it is held each March. 42 Ko! players say they play the Nilgiri God tune while walking to and from places of worship because it is short and easy. In Kolme -1, since no shrine exists there for the Nilgiri God, the performance of the tune is not associated with any particular place. But inclusion of the Nilgiri God tune in the divine repertoire also marks the deity itself as a Kota one. Incorporating the Nilgiri God in the Kota pantheon is a gesture tying the Kotas of all seven villages to the land they inhabit: that is, the Nilgiri God is somehow seen to be, as its name suggests, a primordial divinity connected with the physical place of the Nilgiris, a numen. To the extent that the tune itself recalls stories of the god-in this case fairly unlikely because detailed knowledge of this deity is rare-the tune recalls a former relationship between Kotas and Irulas that was broken, resulting in worship of the same deity in different places. In a sense, then, the Nilgiri God is associated with both a pan-Nilgiri tribal identity and the division into different tribes, each in its locality, and each with its styles of worship. §3. Another incident of divisiveness, this time disunity within· a village rather than disagreement between two men, is associated with a god tune that resembles the Nilgiri God tune, called the Ki· rputn meyr;, gub!k oygd ko! ('tune for going to the herd of Kirputn's son'). Its story is rather suggestive: A long time ago many cows were kept in Kolme ·I. A terrible disease, affecting the throat and causing diarrhoea, began killing off the cows. The village diviner ( te· rha · m) was consulted and became possessed. The 'diarrhoea goddess' (be· ydamn), the diviner related, was affecting the cows because the men of Kolme -1 were not of one mind--each was out for himself. Ay~o· r (the Kota 'father god'), promising to send the goddess away from the village, began to sing this tune. The song mentions the names of the places along a route to the southwest on the way to the house of a Badaga-the son of one Ki · rpu~n. This man was apparently branding as his own any cow or buffalo that strayed into his herd-he also had a reputation for cruelty. Within 'eight to fifteen days' the disease left Kolme -1 and killed off the Badaga's cattle. 43 This story exemplifies how divine favour depends on community solidarity. The story teaches, by example, that villagers should value the interests of the community over the interests of the individual: disunity had caused the cows to become ill. It further establishes the Badagas as an 'out group' and reinforces certain negative stereotypes of Badagas that some Kotas hold. The tune, ostensibly 246 Rirhard Kent Wolf Rain, God, and Unity among the Kotas 24 7 composed by AyJ:lo· r ('father god') himself, is significant both as a divine 'product,' and as the aural reminder of a story and the cultural values that story encodes. §4. A fourth tune, called the padnet devr a· td ko! or 'eighteen god calling tune', is a bit puzzling. Thus far I have found reference to 'gods' only in Emeneau's Kota Texts (1944, III: 17), but the eighteen gods were not named. Mr Pucan suggested that eighteen refers not to gods but to feast days ( u · tm) of the god festival. Nowadays the Kotas of Kolme ·l alternate one-day and three-day feasts each yearalthough it is widely known that the feasts have been shortened over the years. 44 Whether the tune is named after eighteen gods, or as many feasts, or both, the tune seems to have previously had further significance or extrinsic association. Nowadays, to the extent that Kotas reflect at all upon the 'meaning' of the tune, the eighteen god tune is tied to vague notions of divinity, bygone practice, and tradition ( ma· mu·l), but not to a particular story. The lengthy feasting may have been connected, in a rather instrumental way, with the arduous process oftemple reconstruction. Cane and bamboo collected from the forest were fashioned into a thatched enclosure. Supporting beams for the temples were collected from a 'milk tree' (pa·l marm, unidentified botanically, see DFJJR 41 00) in the 'shola' forest 45 called kuy te·l (lit.: 'hole forest'), about four kilometres southwest ofKolme ·1 on the route to Me. na. r village. without the cooperation of all Kotas: if one Kota violates a taboo, it constitutes an obstacle for the whole village. In subsequent years the musicians played the post-cutting tune at some time during the actual process of gathering posts. In modern times when a permanent structure was erected the post-cutting ritual was unnecessary. But the post-cutting tune remains a nod to the past; whether understood as a reference to building the temple (a reference perceived by all because the name itself gives a clue), or whether understood in terms of the story (a less accessible reference because not all Kotas know the story). §6. The stories associated with the Ki· rputn meyt~ tune and the postcutting tune suggest that the Kota gods respond favourably to Kota pleas for help. At one time, music was itself used for supplication. Each of three tunes, called moyr patd ko4s (complaint experiencing tunes), were used to petition a deity, 'big father god' (dorf,ayJ:lo·r), 'little father god' (kunayJ:lo· r) ,47 'mother god[dess]' (amno· r). Among these, the kunayJ:lO· r moyr patd ko! bears a strong melodic resemblance to the 'eighteen god calling tune' but is supported by a different rhythmic pattern. 48 Although these tunes are no longer used individually to petition particular deities they retain a sacral potency as part of the devr ko! repertoire. They are, like several of the devr kofs, sonic components of bygone rituals which have come to be preserved in a new context. §5. The kab ercd ko!, or 'post-cutting tune', originated in an event connected with collecting this 'milk tree'. Because somebody committed a ritual fault during the god ceremony one year, the Kotas could not locate a milk tree from which to cut a pillar for the temple. The diviner, consulted on this matter, put his wrists together behind his back (the characteristic posture for possession among the Kotas), shook and hopped backwards to the shola and revealed a milk tree. As he hopped, he hummed a tune which the ko( players memorized and subsequently adopted as the 'post-cutting tune'. 46 Like the Ki· rputn rneyJ:l ko{, ( §3) this tune was ostensibly composed by god himself; and also like the Ki· rputn meyJ:l ko!, this tune is a metonym for Kota religious values. Proper worship cannot be conducted §7. Kotas 'make god' when they enact the god ceremony and part of this creative act used to include rebuilding the temples. But a more literal 'making' includes the fashioning of the 'faces' of the gods by pasting gold coins and silver ornaments (which were offerings to the god) on patties of cowdung. One man from each of three 'families' (kuytss) is responsible for arranging these coins in the manner of a face on the right entrance-pillar of each of the three temples (one man per temple). Some Kotas think this practice is of recent origin and thus consider it inauthentic ( ocmu ·l, see above). This sentiment is of some consequence since some Kotas define themselves differently from Hindus (though not all do) by emphasizing the idea that Kotas do not represent their deities anthropomorphically. To diffuse the significance of the 'face' as something 248 Richard Kent Wolf Rain, God, and Unity among the Kotas 249 comparable to a Hindu idol, they explain that coins and ornaments had long been pasted on the temples as offerings to the gods, but over the years people had begun arranging the many ornaments into a decorative pattern, a face. For many of the Kotas who consider themselves Hindu, the face is an anthropomorphic representation of the god and, in this sense, is like a Hindu idol. For these Kotas the age of the practice is not the issue-new or old, it is a practice culturally valid-and the responsibility of fashioning the face is also an honour. The two perspectives on making the god's face are but a simple illustration of how politically charged and open to interpretation the concepts of ma· mu·l ('tradition') and ocmu·l ('new rule') themselves are. Leaving aside the question of how the ornaments are arranged, and what that means, we should recognize that the practice of offering money or precious metals to a deity is itself a widespread and ancient religious practice in India. Among the Kotas, the coins used to be heated in a smithy within the temple premises and fashioned into small bows and arrows and other tools associated with the Kota gods. 49 Since these quintessential Kota skills, blacksmithing and hunting, were said to have been taught to the Kotas by their gods, icons of these practices were also considered holy. Forging these icons was thus another way of 'making god'. 50 Before the faces are fashioned during the god ceremony, the first three coins must be placed on each of the three temples by the two ritual leaders (mundka· no· ns). It is this initializing process which is elaborated ritually, rather than the subsequent arranging of the coins. While the ritual leaders paste the coins, the brass trumpets (kob, literally 'horn', always played in pairs) are sounded and the patm pacd ko/, 51 or coin placing tune, is played. 52 Before discussing the next devr ko!, let us further consider the significance of blacksmithing and hunting. Recall for a moment that tribals in India are 'sons of the soil' in the popular imagination and that Indians attribute to the tribal a pre-modern self-sufficiencyall part of the more general idea that tribes are 'at one with nature.' The Kotas quite consciously think of their gods as having something to do with nature (writ large, in the Western sense): the gods are embodied in fire, water, stones, special bows and arrows, and in the blacksmith's shop. During the god ceremony diet is restricted to what used to grow in the Nilgiris (even if it was cultivated) and no intoxicating substances are supposed to be ingested. But the idea of 'nature' is itself culturally constructed (cf. Schneider 1968) and to my knowledge there is no Kota word for 'nature' as an abstract concept. Kotas use the English word as well as the Tamil equivalent iyarkai. 53 I strongly suspect that the Kota tendency to associate Kotaness and Kota gods with nature is tied up in a modem self-reflexivity that comprehends the general way tribals are viewed in India. I am not claiming a modern origin for the practices, but rather for the interpretation. It is in part this interpretation, I believe, that allows the Kotas to think of diverse cultural and religious practices as belonging to a whole. §8. Hunting is associated not only in a general way with the Kota 'father god', 54 but also in a specific way with a deity named ve · fka · r co· ym ('hunting god'), worshipped just outside Gudalur town (near the Kota village of Kala· c) by several Nilgiri and Wynaad tribes and castes. The origin stories of this god vary depending on the community telling the story. Kotas of Kurgo -j and Kala· c village still attend the yearly festival for ve· tka· r co· ym (sometime around October each year), and the men play instruments and dance. 55 Several Kota villages have god tunes named after the hunting god. In Kolme ·1, the hunting god tune was once used in a god ceremony ritual (it has since been abandoned). At the end of the god ceremony the mundka. no· ns used to lead a procession to the sacred place called totf,ba· land shoot an arrow, symbolically to kill a bison believed to be connected with their gods. This bison had repeatedly disturbed the Kotas during worship. 56 It is believed that, for many years, a bison would come to that place at the end of the god ceremony. More recently, after a bison no longer appeared every year, the practice was formalized into a ritual of renewal called devr kaytd ( 'god washing'). Except during the god ceremony, the deities are believed to reside in the back rooms (kakui) of the two ritual leaders' houses (dotf,vay, 'big house') in the form of a bow and arrow. For the 'god washing' ritual, the bow and arrow were brought to to¢ba·l under tight security: if women or non-Kotas were to see the deity, the consequences, it is said, would be deadly. Along the way an arrow was shot toward Doddabetta Peak (north) and to the west. When they reached totf,ba · l, the silver points of arrows were washed, new bamboo was collected for the bow, tavt ('hill guava,' DEDR 3112) wood for the shaft of the arrow and new twine 250 Richard Kent Wolf Rain, God, and Unity among the Kotas 251 (pobit na· r[) were extracted from tree bark for the bow string, 57 The hunting god tune, which was part of this (now abandoned) god washing ritual, is now performed more generically as part of the god tune repertvire. My impression is that the 'hunting god' is known to most Kotas as a deity in some way connected with their hunting past. As one of the god tunes, it evokes a history and self-identity congruent with the lore of the other god tunes-a history that valourizes a tribal way of life (huntiug) in an environment that is substantially their own (the Nilgiris) where divinity is associated with and evident in nature (the bison, the forest materials which the bow and arrow are created as well as in the capacity to control the environment (success in hunting, the use of fire to forge metal arrowheads). Description and Analysis of the Rain Ceremony The description that follows will alternate between a macroview of the whole ceremony in performance, and an extended macroview discussing the significance of each part. The micro sections will be marked with smaller type to facilitate reading. I will begin with a personal narrative that harks back to the epigraph of this paper. LEARNING ABOUT THE RAil~: CEREMONY §9. The last two god tunes are area· y! ko{s, tunes played while men sit under the rough-hewn hut (area· y!) erected in the temple area during the god festival. All the god tunes are known generically as area· y! ko{s by those who do not know the individual names or stories. THE SIGNIFICANCE OF THE Goo TUNES IN THE RAIN CEREMONY Each of the god tunes is named after a particular practice or story whose theme or use is part of 'making god' during the god festival. The fact that the god tune repertoire is also used in the rainmaking ceremony suggests a strong association between the two. All gods of the god tunes are called to participate in a community attempt to bring rain. All these tunes, some from diverse origins, are combined in a single, pointed effort to invoke god. This act of combining suggests that all these tunes are of one category. 58 Bringing these diverse god-related (whether in ritual, origin story, or merely in name) tunes together is a gesture that establishes 'the relatedness of things' (cf. Fernandez [1986: 191), in regard to 'returning to the whole'): the relatedness of different divinities, the unity of the Nilgiris as a region, the importance of right conduct and social harmony, and so fmth. Collecting different tunes together into a single repertoire is a musical equivalent of what happens in the procession, which is like a miniature pilgrimage. 59 By incorporating sites of divine presence into a single ritual, the Kotas constitute a spatial unity within the village. One day in May 1991 I went to Indu Nagar-6° to meet my Kota friend L.Gunasekaran, an English literature student who was helping me learn the Kota language. He planned to help me interview Mr K. Puc an, a senior instrumentalist and village elder ( dorf,a. !) . Fortunately for me, Mr Pucan and his wife were next door visiting their son, Dr Varadharajan, and even more fortunately, Mr Pucan had agreed to sing for me-a significant breakthrough because he had until now insisted that only women could sing. The halting, slightly out-of-tune quality of Pucan's voice, affected as it was by an age of eighty years, was more than compensated for by his way of making performance into a story. Pucan chose to render two songs which Kolme · I women sing during important festivals. Women sing these songs while dancing in a circle and clapping hands. The songs share much stylistically with women's kummi songs of Tamil Nadu and Kerala. Women's singing and dancing are necessary and auspicious endings to entire festivals and to structural units such as specific rituals within festivals; they are considered ea · trm, or rituals. One of these ritual songs, from which the epigraph is extracted, has its Kolme · 1 origin preserved in a well-known story. In the beginning, a black cow called Basavan led the Kota people to the place called na{kal and indicated that a village should be founded. The first post for the first house in the village was erected there. The Kotas milked the cow and built a temple. Now there are three temples. Big father god temple, little father god temple, and mother goddess temple. The mundka· no· n (ritual leader) does puja for these temples. One time the mundka· no· n made a mistake and the temple oil lamp disappeared, cloaking the temple in darkness. He could not do puja and, singing this song, went in search of the lamp. The song mentions the names of holy places and trees."! 252 Richard Kent Wolf Rain, God, and Unity among the Kotas 253 While telling about each of these holy sites Pucan informed us that on the coming Monday, ka·rJky (monetary offerings to the deity) were to be given at each site as part of a ceremony to bring rain. Going from site to site the twelve devr ko{s, or 'god tunes', are performed, and for these tunes, he explained, 'there is meaning' (artm, DBIA 20). The process of learning and what he meant by 'meaning' was one of the most fruitful and rewarding as well as frustrating activities in my ethnomusicological fieldwork. Ultimately, what I learned about the 'god tunes' has been summarized above. The sites themselves, also of some significance, led me to consider the rain ceremony in a broader Kota context. THE RAIN CEREMONY IN OTHER VILLAGES AND IN RELATION TO OTHER CEREMONIES The Rain Ceremony At about 9:45 in the morning on 20 May I991, men and boys of Kolme ·1 village assembled at the set of two upright stones ( na(kal) northwest of the temple area (guryva·l). The natkal, a location for village council meetings (ku· tm) and place of assembly before ritual, is significant as the site on which a divine 'black cow' (ka· ra· v) was said to have indicated to the Kotas that they should found a village-as told in the story associated with the epigraph. The rain ceremony (may ca· trm) of four of the seven Kota villages, Kolme ·1 (eight kilometres south of Ootacamund), Ticga· r (fifteen kilometres northeast of Ootacamund), Kurgo j (twenty-five kilometres northwest of Ootacamund), and Me· na · r (thirty kilometres south-southwest of Ootacamund), is a procession to trees and stones associated with divinity. Through prayer, monetary offerings, 'jumping' (chanting sacred syllables while stepping rhythmically around a circle), and instrumental music, Kotas beseech their gods to remain in the village and look after their well-being. 62 The rain ceremonies in each village are composed not only of musical elements but also of other ritual elements found in devr (lit. 'god'), the annual god festival. Celebrating the god festival is called devr gicd, that is 'doing' or 'making' god; suggesting, I think, that the constitution of divinity is a creative act: it lies in performance. The god ceremony is also sometimes called paca·l devr. The placename paca·l refers to the green grassy area around the three central temples. This reference to place is diacritically important: it distinguishes the original god ceremony from another which is devoted to the god Rangrayi:t-the Kota version of the Hindu god Ranganathan, worshipped at Karamadai (a town that lies between Coimbatore and Mettupalaiyam). 63 The name for this latter ceremony, ko·jka·l devr, 'Koj area god', 64 is also qualified by a placename, ko-jka·l, the former location of a ko·jtree. 65 Placenames are not merely convenient reference points in these contexts. They reflect the fact that Kota gods, like many 'village gods' of South Asia, are numens: they inhabit particular places. As people arrived, the ko!var (band of musicians) 66 prepared their instruments. The drummers tightened their drum heads by heating them by a fire and the ko! (double-reed instrument with conical bore) players adjusted their instruments, dampening the reeds and testing the sound. Elders and ko! players sat with their backs against the stones and faced east, while the other musicians sat to their left in a semicircle. Other men sat on the right. 67 All were dressed in the white clothing appropriate for worship: a cloak called vara· rand a waistcloth called murJ¢. Pucan led as the musicians played three of the god tunes ( devr ko{), starting with the tune usually associated with the ritual of opening the temples during the god ceremony. THE PROCESSION Briefly, the procession begins at the na(kal (stones marking a meeting place); proceeds southeast to a place called ponic where three god tunes are played and coins are offered to the deity; returns through the village and proceeds northwest to toif,ba·lwhere god is worshipped in similar fashion; returns toward the village via the arckal ('king stones'), where three more tunes are played; and finally, all meet at the na(kal again and share tobacco. Mter Pucan and Raman played the o·la· gicd ko! and two other tunes, 68 two men sounded the brass hom, called kob, another beat a large half-coconut-shaped drum (roughly 2.5 feet in diameter) called the e· rtaba[k. From this point onwards, the narration of the rain ceremony proper is indicated in regular type; in smaller type are extended explanations of what individual actions, places, or other elements mean. 254 Richard Kent Wolf Rain, God, and Unity among the Kotas 255 Both instruments sonically highlight moments when the procession leaves and arrives in a place. Calling attention to important moments is the general function of these instruments, whether those moments are distressing, as in the announcement of a death, or joyful, as in the opening of the temples."9 to lead in this worship, in 1991 only one diviner was alive, and in 1992 there were none. Sulli narrated to Mandelbaum a prayer which is no longer recited at ka. dman ka1J4y,1 2 but adds to our understanding of the ceremony (full text and translation in appendix). 73 Gods kama{raya, am no· r, ay1Jo · r, ra · mco · mi, rahgco. mi and be{damn! We are your sons, children of god. Having finished sowing, we are experiencing hardship without any rain. Letting tears fall at god's feet, we come to bring you along. Since 74 kama{raya lives in venve{m and rahgray1J lives in Karamadai how are we to prosper? For that we come, elders and youngsters, with our musical instruments and horns, saying 'we will catch hold of you and go'. So that you don't forget us, we come in the manner of supplicants and grab (i.e. bow at) your feet. The idea contained in the phrase 'since kama{raya lives in venve{m and railgray'IJ lives in Karamadai how are we to prosper?' is that since Kota gods are originally from areas outside of Kolme ·1, they might return to these places of origin because the Kotas of Kolme ·1 are not living righteously.75 We have seen similar themes in the stories behind the god tunes and the song vefke· vefke· (in the epigraph) as well: a goddess caused cattle to become ill because the village was i;>ecoming factionalized ( §3); the 'milk tree' was rendered invisible because the people were not living righteously (§5); and the lamp disappeared because the ritual leader had committed some fault. kama{raya, ay1Jo· rand amno· rare names for the three Kota gods associated with three temples in the centre of the village. 76 But the other three gods, rahgco· mi, m· mco· mi and be{damn are the new trinity of gods (shared by Hindus and other tribals) introduced from Karamadai town in 1925 (Mandelbaum 1960: 226). Since there are no longer any diviners alive for the three gods (for whom the Kolme·l temple called ko·jk·l was constructed), these gods are not currently represented in the rain ceremony. 77 The processional order was loosely structured: a man carrying the e· rtabatk on his back led the procession while a boy struck the drum from behind. The ritual specialists lagged behind the rest of the group. Pucan and Raman played the first three ko!s, but as Pucan was too old and infirm to keep up with the group, Raman and his son Duryodana played for the rest of the ceremony. The men walked from the natkal about half a kilometre southeast to a place about fifty metres above the village and paused to worship at a place called ka· dmankar:u!y, one of several 'raised level grounds of worship' ( kubit km:u!y) surrounding Kolme ·1 village. These 'worship places' are significant in that they are the last places from which the village can be seen when leaving and the first places from which the village can be seen when returning (Emeneau 1944, II:245). Just as Hindus value darsan, Kotas value the visual interaction between deity and person (Babb 1981; Eck 1981)-although other parallels between Hindu puja and Kota worship are few. In this case, the view of the village is the view of the deity, since Kota deities are seen to be especially effective, active, and present within the village boundaries. The idea that divinity is accessible through vision is expressed in a number of everyday activities, especially in coming and going from the village. Kotas pray facing the temple before commencing any excursion beyond the village. They may even pray before someone else begins an excursion, as did R. Mathi, whenever her son and I would leave town. Upon returning to the village Kotas pray first at the nearest kubi{ka1Jrf,y, then at holy sites along the way, and finally at the temple. These are not special activities for special days; rather they are ingrained and habitual. Praying at a particular spot includes clasping the hands together in an anjali (kay mu·vd) and turning to one direction or another while uttering 'god' (co· ym). These directions are frequently believed to be the locations of other important deities or shrines-although a given person may not have visited all the locations. It is enough to gives respect by facing the directions in which the deities reside. 70 Mter worshipping by clasping the hands together, turning to the right or to the left, 78 and bowing down ( acf,murtd), the men walked east-southeast for just over a kilometre to the first of a series of stone circles that are part of the place called panic 79 (one of the sacred sites mentioned in the song ve!ke· ve!ke· ) or ertva1J ( 'place fruit'). 80 This first place is called the parykm vecd erm or the 'offering keeping place'. .:4lthuryodana was forced to lead the procession. Such events underscore the community value ascribed to music. !his concludes the detailed description of the rain ceremony as ~ 'Wltnessed it. The remaining discussion concerns the context of the rain ceremony as a whole. 262 Richard Kent Wolf Rain, God, and Unity among the Kotas 263 Reflections upon the VVhole THE RAIN CEREMONY AND THE HARVEST On the Tuesday following Monday's rain ceremony, the village diviner was once obliged to climb the temple tak102 tree and announce 'ind ma. p i¢uge·' 'today take leave'.) 103 The following two Tuesdays were and are also supposed to be such formal days of rest. Until the recently deceased Va. Kamaten called an end to the practice, the Kotas would also refrain from working for three Fridays after the three Tuesdays and then for three Sundays after the three Fridays. When this final day of ritually calculated leave had passed, it would be time for the harvest to begin. According to my calculations, based on these days ofleave, the rain ceremony would have occurred nearly six weeks before the harvest. Va. Kamaten changed the rules to suit the tea harvest, which must continue throughout the rainy season. Just as Saturdays are religious holidays to Jews, and Sundays t? Christians, Mondays, and to a lesser extent, Saturdays are traditional days of ritual observance for the Kotas. 104 The days of leave associated with the rain ceremony, taken to show respect for the 105 gods, do not fall on these traditional weekly holidays. I wo~ld suggest that, since the ceremony marks howeverth~ formal be?mning of the monsoon season, these special days of ntual leave m a sense 'stand for' the long monsoon periods which people would have at one time spent in their homes anyway. In the old days, they say, the sun would not shine for months on end. And, as I interpret them, the three sets of three one-day-a-week vacations, which interlock and yet exclude other holidays, provided continuity and a means of practically reckoning time between the sowing season and the harvest season. SuMMARY: VILLAGE UNITY CoNSTITUTED IN THE RAIN CEREMONY the Kota gods are contained. By offering coins at these places, invoking god, praying, performing music and so forth, the Kotas affirm the geographical integrity of the village landscape in specifically deity-oriented and community-oriented terms. Like the twelve tunes, which together constitute the musically divine, the four places, ponic, to¢ba-l, arckal, and natkal, enshrine and contain within their borders the spatially divine. Both the places and the tunes are meaningful, in part, because they remind Kotas of stories about the power of the gods. In this sense, they encode a kind of positively valued past, a past in which the gods worked to the Kotas' benefit. The rain ceremony combines these elements in a uniquely Kota bricolage, drawing upon the 'remains and debris of events . . . fossilized evidence of the history of an individual or a society' (Levi-Strauss 1966: 21-2; de Certeau 1984: 29-42). Rather than relegating the rain ceremony to that mystical, universal phenomenon which Levi-Strauss calls 'mythical thought', I prefer to think of it as a 'strategic' (de Certeau 1984: 35-6) use of the past, a performance aimed, in part, at delimiting a 'place as its own' .105 The temporal remove of 'a past' is only relevant in relation to a later period, let us call it a 'modern' period, from which a 'disjuncture' from the past can be experienced and reflected upon. 106 Put in another way, modernity, in my usage, is relational rather than temporally fixed: it marks a community's self-awareness of the gap between the present and an 'authentic' past (for the Kotas, that constructed authenticity called ma· mu-l). For the Kotas, it is a modern period in which one might expect the integrity of the landscape, the memory or faith in the gods, and the unity of the people to be a trenchant issue; trenchant enough to urge a 'return to the whole'. It is possible to locate one such moment of disjuncture (it happens to be in the twentieth century)-there may be more-but to what extent is it reasonable to search for origins of the rain ceremony in this period? What would constitute evidence of such origins? KOTA VIEWS ON THE HISTORY OF THE RAIN CEREMONY I suggested that the performance of god tunes is a mu~ical parallel of the procession to holy places. Ponic and to¢ba · l are m some ways eastern and western boundaries of the village: not of the village interior, where the houses are built, but the village boundaries as ritually constituted in prayer to village deities; borders of ritual games; demarcation ofland; and incorporation in 'god' stories and songs. They are the borders within which all sorts of ceremonies for Significant new features of rain propitiation may have come about during a period of drought: In 1933 there was no rain at all in the district. They asked the god and the diviner said, 'Go to ponic and call the god with your flutes [ko(s), then come 264 Richard Kent Wolf Rain, God, and Unity among the Kotas 265 to to(iba-l (there they bow down and pay a rupee to the god), then come and sit in front of the ra· mayt~o· r temple, at the arckal . . .' When they reached there, a big storm and heavy rain came. It fell only in Kolme·l not across the boundary. [Sulli's account in Mandelbaum n.d., 14 May 1937, 3-4; my emphasis added). The first passage is puzzling. Did Sulli mean that the first rain ceremony of this form (the procession and coin offerings), in Kolme ·1, occurred in 1933? If the ceremony was already a yearly occurence, as it is now and as it was in 1937, 107 would the diviner have needed to tell them to enact a ceremony they already knew? Were there gaps in Mandelbaum's account (did he ask, 'give me an example of a time when the rain ceremony needed to be performed' or 'when was the first rain ceremony?') 108 Even if the first ceremony was in 1933, where did the form come from? Can we assign Kota gods (the purported inventors of all ceremonies) the same agency as human bricoleurs? Parallel ceremonies exist in other villages now and some did in 1937 as well, according to Mandelbaum's Kurgo-j Kota informants and a Badaga interpreter. 109 What does it mean for a ceremony to 'originate' in a place, by divine decree in a sense, if it exists in similar form elsewhere? Between my two years of fieldwork in the Nilgiris I came back to the United States, began studying Mandelbaum's fieldnotes, and discovered Sulli's (14 May 1937) account. When I returned to the Nilgiris I asked Pucan and Raman when the rain ceremony began. Both insisted that it originated long ago ( ma· mu·l ca· trm, 'traditional ritual'). I mentioned Sulli' s account and they both scoffed. Sulli could not be trusted, they told me. Pucan told me on several occasions that he felt Sulli was misinforming Mandelbaum. But since Pucan did not know English, he could not intervene. How Pucan perceived Sulli to be giving misinformation is a different questionhis English was insufficient to understand discussions between Mandelbaum and Sulli and he would seldom have been present during their work sessions. Rather, I think, Pucan's overall distrust of Sulli prompted him to think Mandelbaum was being misled. Pucan indicated the ceremony originated very long ago, during a year when there was no rain in the district. In quite another context I asked the late Va. Kamaten about the significance of ponic as a place and he replied with this interesting answer, 110 'Once drought hit Kolme -1 and people went [there] and built a stone [structure] and were praying [to] it. When they were praying it started raining, and from then onwards it became a custom. This happened more than a Century ago'; [interview, Va. Kamaten, 9 January, 1992; my emphasis]. Sulli, Pucan and Va. Kamaten agree that the ponic ceremony originated one year when there was no rain. It seems unlikely that Sulli would have mistaken the year of drought since his I 937 discussion with Mandelbaum took place only four years after the remembered drought. But what might cast suspicion on Sulli' s dating of the first ceremony is his support for the new gods and the possible desire to show their role in bringing rain. Pucan and Va. Kamaten, as two of the oldest members of the village, either may have forgotten that the ceremony began in their lifetimes or may have wished to make the ceremony appear more 'traditional' by asserting its origin to be earlier than it actually was. Sulli and Mandelbaum discussed the rain ceremony on other occasions, and Mandelbaum recorded another Kolme-1 man's description of the rain ceremony, but no mention was made of the ceremony's origins: this also seems suspicious to me, if the ceremony was indeed four years old. The matter of when the ceremony originated will have to remain unresolved. Can we approach the problem from the other side? With due respect for the limits of such speculation, let us ask: if the ceremony, in its present form, did originate in 1933 or thereabouts, why that form at that time? What was special about the 1930s that might have called for historical reflection, reaffirmation of boundaries and reassertion of social unity? I alluded already to the social and religious upheaval of the period, beginning first with the epidemic of 1924. Let us examine that period in more detail. One of Mandelbaum's major articles on the Kotas, 'Social Trends and Personal Pressures', is part of the 'culture and personality' discourse of his generation. Here he describes the 1924 'epidemic of relapsing fever' which decimated the Kotas. In Kolme ·1, all the ritual leaders and diviners for the three ayr:to· rand amno· r temples died, as did many others. 111 In 1925 the Kotas began their yearly god ceremony without any ritual specialists; and when the villagers prayed to the gods to indicate successors for these men, 'as had not happened before in the memory of the people, no man was super:naturally propelled to the temple pillar to be diviner; none was ~Seized as priest' (Mandelbaum 1960: 226). Mandelbaum goes on to explain, as I confirmed recently with other Kotas, that a rather irresponsible and untrustworthy man by 266 Richard Kent Wolf Rain, God, and Unity among the Kotas 267 the name ofKucvayn became possessed by what he claimed was the god Ranganathan of Karamadai. Kotas have had long-standing involvement with worship of this deity, at least according to Kotas today. 112 But still, the community was slow to accept a new addition to the village pantheon. But in 1926, when new priests were still not chosen, the community heeded the words of Kucvayn when Ranganathan spoke through him demanding a temple be built. Only then, he claimed, would diviners for the old temples be chosen. Two other men were possessed by Rama and Betdamn respectively and Kucvayn chose a priest for the three new gods (Mandelbaum 1941). Indeed the next year all five vacant posts were filled as prophesied by Kucvayn. But ambivalence over the new temples was still not erased. Two factions in the village remained bitterly opposed until around 1960 ( S. Raman, pers. com.). A well-known mourning song (a· I!) captures, with emotive richness, the atmosphere of hostility of the period; since it also illustrates the way in which value can be negatively assigned to village spaces, it is worth including here. The song was composed byTuj, S. Raman's father, in mourning the loss of a young girl named Rangumathi (a name, incidently, which combines the prototypical female Kota name Mathi with the name of the new deity, Ranga). Rangumathi was the daughter of one K. Payrn (older brother of the K. Puc an of this paper). My mother 115 Rangumathi, what will I say Rangumathi? Kolme ·1 village, where umbrellas hardly fit 114 'the middle-street people's cremation site we want,' they say, Rangumathi! 115 My mother Rangumathi, what will I say Rangumathi? 'that-street people's cremation site we don't want,' they say, Rangumathi! My mother Rangumathi, what will I say Rangumathi? Kolme ·1 village, village where umbrellas hardly fit My mother Rangumathi, what will I say Rangumathi? Payrn angered the ma· mu·l people ('traditionalist' faction who did not support the new gods) of his own exogamous division, which occupied one segment of the village (a·ke·r or 'that street'), by agreeing to fill a ritual position for another exogamous division (occupying naryke· r, 'middle street'). The vacancy was left by a man who became a diviner for one ofthe new deities. When Payrn's daughter Rangumathi died, a quarrel arose in the village as to where the daughter should be cremated, whether in the site reserved for the 'middle-street' people, or the site used for the rest of the people. This story illustrates how village factionalism is tied not only to different deities but also to the space on which Kota culture is inscribed. It is ultimately irrelevant whether this particular event occurred before or after the rain ceremony was instituted (almost definitely afterward, if it originated in 1933 or before), rather it characterizes a relatively extended state of disunity among the Kotas of Kolme ·1, beginning with the arguments over new deities in the middle 1920s, and continuing for several decades. Now, although a few families still tacitly avoid worshipping at the ko·jka· l temple (the single temple erected for the three new gods), the new deities have generally been embraced . The new gods were only one of a number of changes occurring at the time-sartorial changes and changes in hairstyle, a movement to do away with the menstrual seclusion hut-all modern in outlook but also contrary to Kota cultural conventions. These conventions were brought into clear focus when they were challenged. Kotas were compelled to reconsider practices they had never previously needed to question. Believing that the gods had or would leave the village and return to their places of origin, the Kotas became acutely aware of their ineffectiveness in coping with a deadly epidemic; and they associated this ineffectiveness with the exigencies of modernity: they needed new gods for the new times. But new gods created new problems of village divisiveness. In this sense, for several decades after the plague of 1924, the 'literal routines' of the Kota community broke down and the community was made to believe, by new factions in the village, that the traditionalists (ma· mu·l kaci) were causing the community to be 'considered by false or moribund categories.' 116 Then, in 1933, there was no rain. Kota stories like those associated with the cattle of Kirputn's son (§3) and the milk tree (§5) demonstrate that divination in Kota society locates a physical problem (sick cows or inability to find a particular tree) in a societal one (violation of a taboo or social dissonance). A Kota deity responded, through a diviner, to this 'time of stress' and outlined a course of action to restore rain whose ritual vocabulary and grammar constituted a discourse of holism. It is therefore not unreasonable to s.uggest that the rain ceremony may have served a revitalizing ftmction in the 1930s. 268 Richard Kent Wolf Rain, God, and Unity among the Kotas 269 Conclusion: Performed Identities in Space and Time By juxtaposing tunes and places of religious significance, the rain ceremony establishes a common context through which the interrelatedness of music, people, places, the gods and community memory is performed. The rain ceremony is in this way an argument-not a verbal argument, but an expressive one-which links the notion of unity or wholeness to the efficacy of the gods and the continuity of the Kota community. Here rain is transformed into a key symbol: a literal nourisher of the Kota people and an indicator that the Kota community is still vital-spiritually, socially, and physically. By ritually reinscribing the landscape, the rain ceremony persuades the gods to remain in the village and defines the 'Kota' realm against that of the 'outside'. We may regard as a case in point the recent introduction of the kuncrtvart (the new shrine, 'little place fruit') to prevent encroachment by the bo· yars. Based on this principle of inclusion and exclusion, Sulli's assertion that rain fell 'only in Kolme -1, not across the boundary' is logical, even if not literally true. This analysis of the rain ceremony suggests that Kota constructions of what we might call 'spatialized subjectivities' meaningfully parallel those found on the plains. Margaret Trawick has suggested in a recent paper, for example, that Tamil Paraiyars identify so strongly with the places they inhabit that 'place' and 'people' become virtually equated (Trawick 1991: 224). 'Place' emerges as a rhetorical substitution for 'people' in the verbal style that Cevi, a Tamil agricultural labourer, employs in her performance of so~g (1991: 236-7). The parameters ofTamil identity with the land, m this example, are manipulated through language use. Through a series of carefully framed questions, E. Valentine Daniel explored shades of meaning and semantic penumbrae of Tamil terms of place. He found that iiru nii!{u Ve!fii!ars conceptualized territories described by Sanskrit-derived terms (ticam, kiriimam) in a manner systematically different from those described by Tamil ones ( ur, niitu). He argued that the Tamil terms provided a 'person-centric definition of space . . . in keeping with the personcentric orientation of Hindu culture'. m The Sanskrit-derived terms were 'relatively context free, universal, and fixed' (Daniel1984: 70) · Tamils located ur boundaries in landmarks: first the shrine for a sentinel deity, Karuppu, and then other shrines, intersecting roads, and other points of significance. We will recognize that the iiru nii!!iilarconception of 'the village' is virtually identical to the Kota conception of the village as constituted in ritual. 118 Perhaps Kotas are affectively attached to features of the landscape, in part, because they strongly value the latter close to their gods. On a spatio-sociallevel, the integrity of Kolme ·1 village thus comes to depend upon keeping the gods within its borders. These borders, like those of the Tamil ur, are divine locales. The constituent rituals of the rain ceremony designed to please the gods residing there thus achieve, in effect, a 'binding' of the village. 119 Divine localities in South Asia are almost always associated with origin stories that explain or illustrate their power. To state that 'place-histories' are 'a unique Tamil expression of Hinduism' (Gonda 1962, 2: 40; cited in Harman 1989: 25) would be stretching a point because 'many of the histories of particular Tamil temples appear, either originally or exclusively, in the Sanskrit language' (Harman 1989: 25), and the phenomenon is certainly not confined to India. 120 But in the Nilgiris many places are not just locations of divinity but are themselves divinities-among the Kotas and even more so among the Todas, 121 mountains are a principal example of this phenomenon. But once again, as Zvelebil warns us, One must . . . be cautious in ascribing an 'areal status' to this cultural trait. Hinduism in general, particularly in the South, knows of mountains which are gods and are worshipped as such: the best-known example is probably Arunacala-Tiruvannamalai, which is worshipped as the 'solidified' form of the fiery Linga (Zvelebill979, III:l49n.). Thus we find that the Kotas assign significance to places and organized spaces in manners akin to those found in other parts of South Asia, and particularly South India. It may be that what makes Kota space 'Kota' is indeed the particular claims on the landscape rather than the forms these claims take. 122 The claims of morality and spiritual power placed upon the landscape and the musical repertoire are qualitative claims about the past. In fact, rather than referring to 'the' past, these claims refer to 'a' particular past. The past is selected: it is a favourable past, a strategically significant past. The particular past is emphasized in the present, under sacralized conditions. To support and sustain selection and symbolic manipulation of the past is not a phenomenon confined to the Kotas: it is characteristic human 270 Richard Kent Wolf Rain, God, and Unity among the Kotas 271 behaviour, particularly in the context of social and political movements. But as Arjun Appadurai has argued, there are important culturally imposed limits on the past as a symbolic resource (Appadurai 1981). The many 'pasts of an Indian village' ( cf. Cohn 1987 [1961]), or any other cultural unit, must be understood in culturally specific and historically situated terms. The Kota method of linking place, past and identity is as much South Asian in character as it is Kota. We have seen that the god tunes function in much the same way as place in this place-past-identity equation. The type of musical sign in operation here is what Christian Kaden has termed a 'detached or loosened symptom' (1984: 126; cited and translated in Stockman 1991: 329)-that is, when the context and function of music are 'loosened from their contextual conditions (a lament from actual death and its rituals, a march or dance from a situation of real marching or dancing), they refer to the original reality more in the sense of a symbol' (Stockman 1991: 329). Thus the templeopening tune and other god tunes come to have more richly nuanced meanings when they are incorporated into contexts other than those for which they were named. They also accrue meaning when conceptualized as a unit rather than as units. Instead of merely being copresent during isolated rituals, they are inscribed in a musical genre called the 'god tunes'. 123 Their sequential performances during the god ceremony and the rain ceremony are musical rituals unto themselves. But the meanings of the music reach deeply into other ritual and historical values. POSTSCRIPT degree. For a time, there was a possibility of fission in Kota society, through a process which has often occurred in Indian village life when the devotees of a new dispensation form themselves into a new endogamous group. But the reforms now seem quite right to most Kotas who have grown up in the past twenty years; and the integrative forces-including their apprehensions about the Badagas-are strong enough to avert any split in the Kota community . . . They can happily worship two sets of gods because, like most peoples of India, they find no necessary contradiction in doing so, as long as their social practices and identity do not suffer thereby. And as has commonly happened in India, the new deities are being amalgamated with the older gods. Many of Sulli's fellows agree with him that the new gods are just the old ones who have decided to take new names, new forms, and to prescribe new ritual practices (Mandelbaum 1941: 254). This phenomenon of inclusion is so widespread in India that for some people it, along with caste, has been erroneously thought to be somehow a defining feature of Hinduism. 125 But this does not mean that the Kotas do not strive to establish and maintain a certain 'Kotaness.' Within this modern Indian context, the rain ceremony strives for a rather special kinds of Kotaness. Unlike the unfortunate ritual leader who lamented as he searched all the holy places ofthe village 'god, god, where are you?' the Kotas celebrate on these holy places and beseech the gods for rain, asserting by their actions, 'god, god, here you are' and, by extension, 'so are we.' NOTES 1. The Kotas have managed in course of time to maintain a remarkably strong sense of uniqueness within India while at the same time adjusting to and incorporating both modern and Hindu features of Indian society around them. Most Kotas find no contradiction in stating they are Kota and they are Hindu; that they belong to a tribe and have their own deities and that these deities are also 'Siva' and so forth. What Mandelbaum wrote in 1941 is perhaps even more true today. 124 The new ways are so much with the tide of the times that even the most obdurate of the surviving conservatives have yielded to them in some The data for this paper derive from ceremonies I witnessed in I 991 and 1992 in Kolme -1 and Ticga · r villages, and from historical and contemporary descriptions of the ceremony in these and other villages. Kota village names, rather than equivalents in English, Tamil or Badaga languages, have been employed throughout this paper. Common English spellings for the seven Kota villages are given below: Kota English Kolme -1 Ticga · r Me-na·r Porga · r Kina· r Kurgo -j Kala· c Kollimalai Trichygady Kundah Kotagiri Kotagiri Kil Kotagiri Sholur Kokal Gudalur Rain, God, and Unity arrwng the &tas 273 272 Richard Kent Wolf 10. I have. ~etained t~e ~ng~ish .word 'Kota' instead of the Kota word fto.vhoWeVtr• to faohtate Enghsh md1catwn of gender and number, i.e. 'Kota Woman' rather than the linguistically correct ko· ty, and 'Kotas' rather than ko· v ·ZA11 spellings accord with Emeneau's phonology in DEDR, etc. ~· Nancy Munn suggests that iconicity between symbol vehicles and meaning is a generalizable phenomenon: 'the form of the vehicles is intrinsic to the message carried, and certain qualities of the referents are put into sucial circulation as part of this form' (1974: 580). That is, the overall organization of the ceremony, its formal features (the way in which the places and tunes are used or presented), should structurally replicate the meaning of its constituent elements (the places and tunes themselves). 'Time-binding' is Fernandez's skillful turn of phrase for 'one of the chief contributions of culture', its ability to knit together, to unify, and to bridge the apparent dichotomy between past and present (Fernandez 1966:69). Central concepts in Victor Turner's work on ritual are particularly revelant here: the capacity of ritual to unify apparently disparate significata, and to condense many ideas and relationships into a symbol vehicle that would require considerable exegesis if expressed verbally. David Lan's subtle analysis of the Zimbabwean revolution notes an analogous relationship among people, land and rain among the Shona of Zimbabwe, where proper regard for the spirits of dead chiefs was necessary to ensure rain, 'peace and plenty'. Furthermore, 'amongst the Shona the right ofoWJlership in land is demonstrated and proved by the ability of a partiallar set of ancestors to control its fertility. The people whose ancestors bring the rain awn the land' (Lan 1985:98). In this regard, Kota gods are to Kotas what Shona ancestors are to the Shona. Lan's book served as an inspiration in my awn analysis, as in my initial field investigation. . Although the meanings behind sacred sites and musical tunes could conce~ ably be manipulated in new and different ways, they generally are not. That 1 ~ to say, all ceremonies directed towards the gods involve a similar style 0 'selectivity', the process of' constructing around the dominant symbol a conte~ 0 of symbolic objects, activities, gestures, social relationships between a~or~ ritual roles, and verbal behaviour (prayers, formulas, chants, songs, recltatl~n of sacred narratives, and so on) that both bracket and underline those of Its referents deemed pertinent in the given situation' (Turner 1977:187). th' See Daniel (1984:13) on a use of the term 'culture' which comprehends IS creative process. . ·. hip For more on this 'representational hypothesis' concerning the relauo~s between macro- and micro-sociological perspectives see Knorr-Cetina's In:: ductory essay in Advances in Social Theory and Methodology: Tmuard an I~conof Micro- and Macrrrsociologies ( 1981: 30-40 and passim). See also Marcus cise summary (1986:169 n.6). th essaY For a critical analysis of the use of history in ethnography see e ·ng in 'Ethnography and the Historical Imagination', and other essays appe~ide the book of the same name (Comaroff and Comaroff, 1992). My pre.s:ich the does not, however, aspire to be the new 'historical ethnography gross Comaroffs propose-particularly because my subject matter is, on a level, easily circumscribed. Rather, it empowers the President oflndia, in cooperation with a State Governor, to designate which 'tribes or tribal communities' are to be scheduled (Mehta 1991:29). For a broad spectrum of arguments see Mehta (1991), Thapar (1977). Singh (1972), and for a review, Babiracki (1990). For a discussion and critique of a related idea, 'primordialism', in the discourse of ethnic identity and nationalism, see Brass (1991:16 and passim). If a community is officially classified as a 'tribe' it is eligible for various forms of government sponsorship. Duryodana is a Tamil name (actually a Sanskrit name drawn from the Mahiibhiirata, pronounced as if it were Tamil). At birth, all men and women are given one of several Kota names. First born males in Kolme .J, for example, are named Kama~n. Nicknames are used to distinguish among those with the same name. These include 'flat-nose', 'eyebrow', 'shit' and other which relate either to some distinctive physical features or to some funny childhood story. Nowadays it is common to give a child Tamil or other Indian name in addition to a Kota name. The tribes in the Gudalur area ofTamil Nadu and in the Wynaad ofKerala are currently working together under the auspices of ACCORD-Action for Community Organization, Rehabilitation and Development. They hold an intertribal gathering every year which features traditional sports, games, music, dance and drama as well as newly created ceremonies expressing intertribal solidarity. The word probably entered Kota through Tamil: see the Tamil Lexicon. The term results from a 'false' etymology: ma· mu·l is interpreted as a combination of the prefix ma: from the Tamilization of the Sanskrit mahii, great, with the Kota word mu· ~ 'direction' (metaphorically extended to 'way or manner'). ocis a Kota prefix meaning 'new'-but the connotation, often negative, is foreign or forbidden, 'oca·! eel' for example, refers to the stench of a 'new' (i.e. non-Kota) man. See Mandelbaum (1960; 1941) on village factionalism based on attitudes towards ma·mu·l. What may seem even more obvious, at least on the surface, is that since both terms are not Dravidian, the concepts themselves may not be indigenous. The assignment of such social-cultural meanings to linguistic phenomena is very tricky, however, and would warrant consideration of a kind I am not equipped to carry out. Although over his three years in the Nilgiris Emeneau worked primarily on the Toda language, he also made the first significant contributions to the study of the Kota, Kodagu and Kolami languages. Sulli's life is discussed at length in Mandelbaum's article, 'A Reformer of His People' (1960). Very few Kotas converted to Christianity and those who have are effectively separated from the community. Reverend Metz' s disparaging account of Kota attitudes toward. missionaries in 1864 captures the spirit, if not behaviour, of Kotas today: 'when I endeavour to address them they drown my voice with their dreadful music, or compel me to retire by abusing me in the most obscene and offensive language or barking at me in the style of their own half-wild dogs. II. 12. 13. 14. 2. 3. 4. 15. 5. 16. 17. 6. 18. 19. 7. 8. 20. 21. 22. 9. 27 4 Richard Kent Wolf Rain, God, and Unity among the Kotas 275 Thus I have often had to leave their villages, with a heavy heart at their apparently hopeless condition' (Metz 1864: 134). 23. Emeneau feels that his lack of familiarity with Tamil allowed him to learn the structures of Kota and Toda without a Tamil bias (1993, pers. com.). 24. Intensive learning from a few knowledgable informants is a common, pragmatic, and perhaps necessary component of fieldwork. The probleiLS engendered by relying on a small number of informants are not so much those of 'getting it wrong' or creating lacunae, since completeness is a fiction anyway, as they are the silencing of the dialogic process which ultimately informs an ethnography. Some recent ethnographies still explore the possibilities of understanding culture through the individual (e.g. Crapanzano 1980; Shostak 1981). But the significance of these ethnographies, Clifford notes, 'is the transformation of 'cultural' text (ritual, an institution, a life history, or any unit of typical behaviour to be described and interpreted) into a speaking subject, who sees as well as is seen, who evades, argues, probes back'. This is effected by 'staging dialogues or narrating interpersonal confrontations' (Clifford and Marcus 1986: 14). Mandelbaum represented Sulli in such interactive settings-although seldom with the anthropologist in broad view. 25. The Badagas (lit. 'northerners') comprise a number of castes who migrated to the Nilgiris from Mysore, probably in the sixteenth century. 1bey participated in a close system of reciprocal obligations with the Nilgiri tribes until the middle of this century when imbalances in population, the rise of a cash economy, and political fractionalism Jed to the gradual breakdown of this system. See Hockings' Ancient Hindu Refugees and other references in Blue Mountains 11989). The role of a Badaga, Christian or otherwise, as an interpreter or an informant in research about the Kotas is a precarious one, given the distrust some Kotas have and do harbour towards them. 26. Particularly from the Badaga M.N. Thesingh, a man who served as an informant to subsequent scholars, including Paul Hockings. 27. On Sulli's ability to speak English and Emeneau's role in improving it, see Emeneau (1940, I: v). 28. I have not been able to ascertain what Sulli's position was vis-li-visothers in the community before he became active in pushing for reforms. To the extent that present-day observation can illuminate the problem, Kotas tend to hold m~m­ bers of their community who have succeeded in the modern world (education, salaried, urban jobs, etc.) in high esteem provided they do not try to use this success to bolster their power within the village. 29. This familiarity was undoubtedly gained through his missionary education. But he also gained experience as the lone Kota, who, by virtue of knowing the English language, could mediate between Kotas and the English (Sulli n.d.). 30. An identification of the idea of 'tune' with the instrument it is played on IS indicated by the use of one term, koh for both. The word kof could also be translated as 'melody,' but I prefer 'tune' because 'tune' seems to convey the notion of a musical unit, a song without words; ('Do you know that tune?' means 'Do you know that composition?' 'Do you know that melody?' could refer to a particular part of a composition). 'Composition' might also be suitable,.but the term carries with it Western notions of invention that are not appropnate. 31. This practice is common to most ofthe Nilgiri and Wynaad tribes who maintain traditions of instrumental music and dance: these include Iru!as, Bena 32. 33. 34. 35. 36. 37. 38. 39. Kurumbas,Alu Ku[Umbas (and probably Palu Ku[llmbas as well), Kanu Nayakas and PaQiyas. Although two ko!s may in rare cases be distinguished through their accompanying rhythm (da·k), the melody alone is usually diacritical. The origin of the number twelve is unclear. Pucan claims to have learned only twelve of what were once sixteen devr kofs. Each time we (myself and other interested Kotas) have attempted to count the ko!s as they are performed sequentially (whether in ritual context or isolated for performance), we have arrived at a different number. The number twelve has no particular significance in Kota culture. Sulli, in Mandelbaum's notes on the rain ceremonies, mentions that eighteen 'songs' are played, some sung and some on instruments. In other contexts, Sulli and other informants have mentioned nine god tunes and twelve god tunes. Because there seems to be some question as to what is counted within the special repertoire and what is excluded from it, I hesitate to conclude that the repertoire has shrunk, although it does appear that fewer men are competent to play these tunes now than were able to in the 1930s. Most of the special twelve devr kofs are generically called area· y! ko!s as they are played together as a set underneath an enclosure called arca·y{. These processes have been discussed systematically, e.g., Tambiah (1981) on ossification in ritual and Kaden (1984) and Stockman (1991) on semiotic aspects of decontextualization in music. Manuel (1993: 131-52) devotes a chapter to a related process through which melodies become 'resignified' in Indian popular music. Although I have not been able to figure out the morphology exactly (and it was not readily apparent to Emeneau either, when I asked him), the term o· fa· gicd apparently refers to the making (gicd) of the sound of the holy syllable 'o·!y' as is commonly the practice during peak moments of worship. Raman, the husband of R. Mathi and father of Duryodana, is also the classificatory younger brother ofPucan. Deep respect mixed with competitive feelings and moral and ideational differences characterize their relationship. Pucan and Raman, as my closest musical and cultural teachers and friends, encapsulate for me the absolute contestability of cultural meaning and content. This lore also appears in passing throughout manyofMandelbaum's fieldnotesfrom several informants, from several villages, and through different interpreters. Toward the end of his 1930s fieldwork period, Mandelbaum began eliciting stories connected with these tunes. He obtained information similar to what I collected. I had not seen these stories before my fieldwork ended, although I had read a version of the rain ceremony as Sulli dictated it. Va. Kamatn, who died in Februaryofl992, had lost his position as mundka· no· n when his wife died. Among the Kotas, ritual leaders and diviners must be married and their first wives (who have important ritual roles) must be living. Talko· r (also talkavrand tafku. r) is a place near Me ·na·rvillage that lies along the route to the Kota land of the dead (located near the Toda land of the dead, near MukurtiPeak; see Emeneau 1944, II: 195). It is the site at which the culture hero Kote · rveyki · Q was said to have shot an arrow into a stone and released a spring. Aversion of this story, recorded by Emeneau in the 1930s (1944, I: 137) is also told today, and is preserved in a well-known mourning song. Mr Raman, in the context of discussing the spirits of the dead (a· na· (o· r, literally 'those 276 Richard Kent Wolf of that land'), mentioned talko· r as the name of the place that the dead go to. He may have been confused: he also thought the place may be called dodde· r vetm, which means 'big god peak'. Doqde· rvetm is also a holy place near Me·na·r. Neither place, however, is considered to be the land of the dead. Zvelebil provides an account in which two Irula brothers quarrelled and parted. The next day when they met, the older brother asked what the younger had eaten. Both claimed to have eaten fruit and milk, but in fact the older had eaten meat. They quarrelled again and vomited to prove what they had eaten. Through magic of the older brother, the younger vomited meat. He was sent to settle in Karamadai on the plains and the older brother went to settle on Rangasami Peak. The account continues to relate how the divinity of each brother was discovered. What is significant here is the association of the Karamadai deity and Rangasami as brothers (Zvelebil 1988: 137-8) The Karamadai temple is now considered Tamil Vaisnava (i.e. Hindu), but an Irula priest also plays a role in the puja, the details of which I do not know. Zvelebil reports, misinterpreting Emeneau somewhat, 'Kotas worship the god at Karamadai in the Vaisnava temple there including him in a prayer which is said by the Kota priest and the diviner in the field before they begin to plow . . . ' At the time of Emeneau's research, the prayers including the names of older Kota gods were apparently extended to include the newly arrived Karamadai gods. The Kotas used to visit Karamadai and participate in the festival there; after bringing the god(s) to Kolme ·1 they no longer actually went to Karamadai. Zvelebil identifies Betdamn ('peak goddess'), the goddess included in the 'new' Kolme·l trinity, with Railgamma be!f ('Ranga mother peak'). Although there was reluctance to admit the Karamadai gods into Kolme·l village, there was and is a belief that the deity in Karamadai was originally a tribal one. A Kota origin story for the temple (provided by K. Pucan) is similar to the one reported by Zvelebil (1979 III: 137), butthe protagonists are a Kota and a Kurumba. If the Irula story of two brothers is merely a transformation of the Kota story of a Kota and an Irula (the Kina· r mundka. no. n 's story), the Kota Nilgiri co· ym may actually be the deity now called Rangasami (or Rangasvami, Ranganathan, etc.). This conjecture is also supported by Emeneau' s story of kole· rveky male· rveky, culture heroes and best friends who have been memorialized (and apotheosized) in stones on hills near Kina· r village and Me· na · r village (talko· r, see previous footnote) respectively. A variation on this story, found today, maintains that both sites are parts of the same parts of the same deified man: the head near Me. na. r (thus justifYing the tal morpheme of talko· r, meaning head) and the rest of the body near Kina· r. See previous note for an extended discussion of this identity. Although the people from Kinar village also have (ar had) a Nilgiri god tune, I was unable to hear or record an example to compare it with that of Kolme .1. This is a version told by K. Pucan. I found two other versions of the same story in Mandelbaum's 1937-8 fieldnotes, one by a musician and one by Sulli. Sulli's version did not mention that a tune was associated with the story. The 'eighteen songs' to which Sulli refers in Mandelbaum's notes might also have some connection with this. See also Emeneau's Kota Texts (1944, II: 337, n.5), where eighteen days of the god festival are mentioned, and eighteen gods are also mentioned (1944, III: 17). 45. Rain, God, and Unity among the Kotas 277 46. 40. 47. 48. 49. 50. 51. 52. 53. 41. 42. 43. 44. 'A "shola" is . . . an evergreen forest of the elevated Nilgiri Plateau, located along stream banks or in hollows and surrounded by large tracts of savanna' (Lengerke and Blasco 1989: 54). Mandelbaum collected a version of the same story from Kurgoj village. I recorded a version of the kab ercd ko! in Kurgo j, played on the bugi· r (a fiveholed bamboo trumpet, not a 'flute' as in DEDR 4239) by K. Mundan (brother of S. Cindamani). His version of the tune is quite different from that of S. Raman and K. Pucan. It may be an entirely different tune with the same name and story attached. Comparison is difficult in part because I did not ask the musicians themselves to compare the tunes and because versions of the 'same' tune on the bugi· rand on the ko! will sometimes differ because the latter has six holes and the former only five. Since kunayrzo· r, according to one story, is said to have come to Kolme ·1 from Porga · r village out of attraction for the music, the last day of the Kolme ·1 god festival, 'dance day,' is celebrated in honour of him. This is one of the few instances in which the accompanying rhythmic pattern, or da· k, is a diagnostic feature of a kof. Before the lamp was introduced, the only objects in the temple were a few large stones within which the fire was built. Ornaments are no longer forged within the temple. A stock of ornaments and coins is kept with the ritual leader or another trusted member of the community. Here the question of how Kota 'icons' relate to Hindu 'icons' is relevant: modern Hindu gods often look like people. Kota icons like the bow and arrow are different in that they embody divine practice. In a way this is comparable to the way the musical instrument, the vi?Jii, or any book, is the goddess Sarasvati. Also called patm kacd kof, or 'coin offering tune'. The tune is rhythmically different from but melodically similar to the amno· r moyr pa!d kof. These observations of similarity and difference are Pucan's; he thinks about the tunes in this way to keep track of which ones he has played and which ones he has yet to play at a given time. The period during which Dr Varadhar~an and the other two men arrange the coins into faces is one of the loosely structured periods during which young boys can practice playing the musical instruments without fear of reprimand. The instruments are loud and should not be played in the village at other times unless a particular occasion calls for it. It is difficult to determine when the Tamil word iyarkai began to take on Rousseauistic meanings. The word is derived from the Tamil verb iyalmeaning 'to be possible'; iyal as a noun means 'nature' in the sense of 'property' or 'quality.' The abstract noun iyarkai has meant not only 'nature, disposition, inherent quality' but also 'that which is natural', as opposed to that which is made (ceyarkai). This latter meaning has been traced to the purananura (written circa first century BC-AD first century; cf. Tamil Lexicon). But even in this opposition we should not read 'nature' versus 'culture'. The absence of a Kota word for 'nature' is not conclusive evidence that the concept is foreign; but neither is the presence of a word for nature in modern Tamil usage evidence that the concept is ancient. It seems likely that the modern iyarkai, when it is used in the sense of Nature, with all the romantic connotations, is the result of 278 Richard Kent Wolf syncretism-but how, when and if this occurred really cannot be traced. I thank Dr James Lindholm for providing insight into some of these points. Ayno· r is also known in Kolme ·1 and some other villages as Kamap-aya, a name for which M. B. Emeneau has not yet been able to discover a satistactory etymology. A god of the same name is important to the Todas as well, as I have found this deity mentioned in one Toda song I recorded. Vasaumalli, a Toda woman who translated the song for me, commented predictably 'they [the Kotas] copied it [worshipping the hunting deity) from us'. Several other villages have a site of divinity called tfH/ba·l and an associated origin story similar to that told in Kolme ·I. This is the descriptionS. Raman (age fifty-seven in 1992) recalls hearing when he was a thirteen year-old boy. A village diviner described the ritual to him. Even at the turn of the century when Thurston was collecting information on the Kotas the shooting of the bison was a bygone practice, but 'what takes place at the present day is said to be unknown to the villagers, who are forbidden to leave their houses during the absence of the hunting party' (1909, IV: 16). I made this point in a paper entitled 'The apotheosis of a Musical Repertoire: Identity and History in the 'god tunes' of the Kota Tribe (Nilgiri Hills, south India)' (Wolf 1993). Schechner constructs a similar argument about the Ramlila processions of Ramnagar, which 'condenses much of the Indian subcontinent into a comprehensible single sacred space with nine main stations . . . ' (1983:257). Indu Nagar, the headquarters of Hindustan Photo Films, lies just outside Ootacamund. This version of the story is modern and fashioned for my understanding in several ways. For one thing, Kota worship is not called puja: puja is a specifically Hindu ritual which usually involves worship of an anthropomorphic deity with fire, incense, flowers and coconut. Kota worship is much simpler in form. Mr Pucan did not disclose the 'mistake' committed by the mundka· no· nand, as I feel other Kotas may not want this information revealed, I leave it out of my account. The modernity of this version of the story also lies in its location. The introduction of a lamp into the temple is a phenomenon of the last half-century or so, and some villages still reject this Hindu influence. A version of the story related by Mrs Cindamani of Kolme ·1 identifies the mundka· no·!. wife of the mundka· no· n, as the singer of the song. In this version she attempts to coax the god back after her husband committed a ritual error. In this version it is the lamp in the house that disappears: this makes a bit more sense since the lamp is lit in the temple only once a year. She finally finds the lamp at ponic, one of the numinous sites involved in the rain ceremony. The rain ceremonies of Kina·r village (fifty-five kilometres north-northeast from Ootacamund) and Porga·r village (on the edge of Kotagiri town, thirty kilometres east of Ootacamund) and an additional ceremony at Ticga·r are different. They involve young unmarried girls circling the temples and the village while sprinkling water. Kala·c village (on the edge Gudalur town) practices no ceremony for bringing rain, in part, I gather, because rainfall is timely and abundant. See Mandelbaum (1941) for the story of how this Karamadai deity came to be worshipped in Kolme ·1. 64. Rain, God, and Unity among the Kotas 279 54. 55. 65. 66. 67. 56. 57. 68. 69. 70. 58. During the time of Mandelbaum's and Emeneau's research, the new gods were referred to in relation to an adjacent area called ke· rva· y--literally 'street mouth', presumably an entryway into the village. Why the point of reference has changed I have not been able to determine. For the word ko-j see Emeneau in this volume, par. 17 and footnote 15. This term is a contraction of ko!partahatk, or kof, the double reed instrument, par, the barrel drum, and taba(k, the frame drum. This arrangement was undoubtedly more orderly than it would have been had I not been present and videotaping the event. But in general, at gatherings at the natkal, elders are seated as has been described and others in a group to the east, either circled around or loosely facing the rest of the group. In this case they played the 'eighteen god calling tune' (§4) and 'Nilgiri god tune' (§2). The order is not fixed, but Pucan keeps a general sequence in mind in order to remember which tunes have been played and which have not. See Needham (1967) on the role of percussion in association with 'the formal passage from one status or condition to another'. I did not note the exact direction while the ceremony was being conducted. While mapping the area later, my assistant Duryodana pointed to the direction and I noted it with my compass. On another occasion I asked Pucan's son, Dr P. Varadhar~an, and he said the direction of prayer was east, 'facing the sun'. 59. 60. 61. 62. 63. 71. The ka · dman ka7J4Y may be significant as a marker of the past in two respects. It is near a tree where Todas used to offer clarified butter at Kota god ceremonies and would sit to watch the proceedings. Todas were not permitted (by their own 'pollution' rules as well as by Kota orthodoxy) to enter Kota villages. The general region of the ka'fJr!Y also encompasses an abandoned buffalo or cow pen (to· y) which may have had a significance which is now lost. 72. It may never have been a formal prayer; see earlier part of this essay on Sulli 's tendency to verbalize thoughts. 73. Mandelbaum (n.d., field notes from 14 May 1937 [misdated 13 March], p.7). I have revised the transliteration system and standardized the spelling to conform with my system. I have also translated directly from the Kota, rather than reproducing Sulli's English which, in this case, overlooks important subtleties. 74. I used 'since' to translate a conditional tense in Kota that could also be translated as 'while'. There is a slight possibility that I misinterpreted what Mandelbaum had written as 'itme· l', since vowel length was not indicated. The 'e' should be long for 'itme·l', to mean a conditional form of the verb 'to exist in place' (DEDR480). If the 'e' is short, the words should be separated and the 'el' is short for 'elm'-'all'. In this case translation would read '0 Kamatraya who lives in . . . and Rangrayn who lives in ... and all (in the sense of etcetera)'. This is close to the rough translation that appears in Mandelbaum's notes, but I think it is in this case an unlikely one based on my experience with the language. 75. S. Raman has repeatedly told me that the gods left Kolme ·llong ago and that is why no new diviner has become possessed in the last several decades. See also Mandelbaum (1941). 76. Kama(raya is another name for ayno· r (see p. 270, nt. 54). There are two ayno· r temples, one for 'big' t!Y'IJQ· rand one for 'little' flY'IJQ· !· 280 77. Richard Kent Wolf It is not clear the extent to which the new gods ever played a role in the rain ceremony. In other, less elaborate descriptions by Sulli and in those Mandelbaum collected from other Kolme ·1 informants, the new gods appear less prominently than they do throughout the May 1937 account. I think 'left' is normative but I know no reason for it. The same place as ponac in Kota Texts (Emeneau 1944, III: 131-2). Here a man/ demon would wait and attack men by swallowing them through his anus, until conquered in te · lo · yna · r by a man named Cing ra -jn. Although I was not able to learn any more about ponic during fieldwork, Mandelbaum noted that stones used to be taken from ponic to the village during the god ceremony to be used in the construction of the shed (area· y!) (Mandelbaum n. d: 27 October 1937). This first stone circle is about a metre high and two metres in diameter. One must remove footwear if entering within an area of about seven metres in radius from the circle. The stones of the first two circles in ponic are supposed to be of ancient origin, but the structures themselves are periodically rebuilt and painted white. Pucan said three coins, mu· nd pafl-m, are offered to the deity but I only noticed one rupee coin and in Sulli's account only one rupee was offered. During the most important days of the god festival, for example, attendance is taken and those not present are punished with additional or unpleasant tasks. Women are usually involved in other activities during ceremonies. If the presence of all women is required in any ceremony connected with Kota gods, the women must not be menstruating. During the god ceremony, the potential problem this creates is solved through a ritual in which all women are symbolically secluded in the menstrual hut for one day (they all go to one area below the village: they cannot actually fit in one house). By doing this they are considered free from the effects of menstruation ( ti·!), which, as in Hinduism, must not come in contact with the gods. But this temporary fiction is only partially accepted: the women who are actually menstruating cannot dance on the paca! (grassy area between the temples). The rain ceremony is conducted on a Monday, a day on which working in the fields is prohibited. Women are busy on this day cleaning the house, puritying the house and the area around it with cowdung, cooking and other domestic activities. They play no direct part in the procession, offering of coins, or music making. Unified action by the ritual leaders is, in an extended sense, a performed version of what Ortner (1979: 94) calls a summarizing symbol; although it would be more accurately termed a summarizing icon, in the Peircian sense of icon (Peirce 1955: 102): a sign which resembles its object. More precisely, it is what Peirce terms an iconic legisign, 'any general law or type, in so far as it requires each instance of it to embody a definite quality which renders it fit to call up in the mind the idea of a like object'. Unified action by the ritual leaders is a 'type', even a Kota 'law' of behaviour, which suggests similar unified action be followed by the rest of the Kotas. Sulli and several other of Mandelbaum's informants described versions of the same ritual and a version was also documented by Thurston (1909, IV: 23). I am told the ritual is still practiced in Kolme ·1 today when a woman gives birth to her first child, but I have not had the opportunity to witness it. Rain, God, and Unity among the Kotas 281 78. 79. 80. 81. 82. 83. 84. 85. The tOij,ba · lis a site of a ne· r! tree in Kolme ·1, and to my recollection, in each Kota village which has a place of similar name. 86. In this ceremony the power of the Kota gods is affirmed in a spontaneous overflowing of milk. This milk is milked in secret by the ritual leader and kept in the kakui (back room) of his house, the same room in which 'god' in the form of a bow and arrow resides. 'May milk overflow', or 'pa·l pongum', is a redolent image in many Kota prayers which means to Kotas, 'let our population increase'. It is a metaphor for community vitality. For an account of the milk ceremony, see Emeneau's Kota Texts (1944, IV: 300-9). 87. In Tamil the word titfu also refers to the lasting and defiling effects of death. In Kota the word ke· r is reserved for this 'death pollution'. I have in general avoided the English word 'pollution' because it tends to lump different concepts into one category. The effects of death and menstruation are dealt with in very different ways. The verbal root of ti ·! (DEDR 3274) in many Dravidian languages means 'to touch' or to defile by touching; ke· ris related to the Dravidian verbal root ker-, 'to perish or be spoiled'. The noun form ke· rand related forms in other Dravidian languages mean 'ruin, death' or 'evil'. Although menstruation is defiling to men and to the gods, it is not 'evil' or the same thing as death. 88. The divine value accorded fruit helps explain why the other name of ponic, ertvafl., or 'place fruit', may be significant as a place name. Nowadays people think of ertvan as a name, not as a word with a significant etymology. 89. 1.48 metres in height, 2 metres in diameter, with a 30 em. high rectangular block on top. 90. As was typical in many ritual situations, in 1992 there was considerable disagreement about which direction was correct. One village elder, Mr Motakambaten, claimed prayer must be in the direction of the village temples, i.e. West. The previous year's headman, Mr Bullan, argued that worship should be directed to the north, towards DoQ.Q.abe\\a. Pucan, who was considered the ultimate authority, said that prayer should be directed northwest. Pucan's son, Dr Varadharajan, who recounted this dispute, suggested, with some doubt, that northwest might be the direction of ve· !ka · r UYfl- (see previous discussion of the ve· !ka· r co· ym ko! or 'hunting god tune' [§8]). The notion that the direction be connected with some sort of significant place or deity is important. It is also interesting that questions of what should be done, and why, are open for debate. Northwest may have been ultimately arrived at as a compromise but the 'reason' may have been invented after the fact. 91. Other Nilgiri tribes and the Badagas chant similar syllables in a similar style. The Toda 'dance songs' documented by Emeneau are executed in this style. The Kota chants have no semantic content. 92. On other ritual occasions possession is desirable and its failure to occur is a bad sign. In Ticga·r, possession of the diviner was (in 1991) a formal part of the rain ceremony. 93. Kotas of today sometimes describe the offerings in similar terms. Dr. Varadhar~an-not without a note of humour-described offerings to the god as 'bribes'. 94. Since the text is not given in the vernacular, it is unclear here, as it is at times elsewhere, whether Sulli was explaining his interpretation or whether a prayer is recited during the ritual. 282 95. Richard Kent Wolf In puja flowers have at least two major classes of meanings. As a fragrant offering to the deity, flowers are a form of honour, welcome and beautification. In this way, they support other activities like the offering of incense, fruit and music and the dressing of the deity [see, e.g., extended description of puja in Babb (1975: 3I-67)]. At times the colour of the flowers is important, red flowers being associated with certain deities and yellow flowers with others. Hanchett (1988: I2l) reports how villagers' interpretations ('the goddess likes it') differ from an educated Brahmin's interpretations of the tala flower: the tight buds are a model for chastity and the yellow color is fitting for the goddess of wealth. Daniel (1984: 190--223) describes a divination ritual in which the meaning of flowers hinges on the correlation between colours and personal 'qualities' (ku>:~am), hot vs. cold, past, present and future, and so forth. Gold (1988: I24, 190-261 and passim) shows how the word 'flower' is used by Rajasthani pilgrims to denote (and suggest identities between) 'icons embodying spirits of dead children, bone remains of cremated adults, womb, and unborn children'. These are but a few examples which suggest flowers are 'elaborating symbols' (Ortner I979: 96) in Hinduism. Setting off an area as 'sacred' is a common strategy in India for protecting land. Further details concerning this legend can be found in Mandelbaum's fieldnotes, e.g. his II February 1938 annotations to notes of 24 April I937. There is considerable variation in accounts of this process. Many Kotas talk of Tipu Sultan without really knowing when and where he actually ruled. The numinous quality of gods and their association with particular spots is, of course, a widespread religious theme in India. Such spots often become pilgrimage sites, representing in broad geographical terms what the sites in the rain ceremony trace, as a processional route, in and around the village. During pabm ('festival'), for example, the entire village eats off the same plate in every house of the village (Emeneau 1944, IV: 294-30I). Tobacco sharing as a strong symbolic statement of unity is also indicated by Sulli' s description of occasions in which tobacco was not accepted by all male members of the village: these men were permitted to voice their grievances, and tobacco was shared only after the disputes were settled. The diviner did not climb the tak tree in 199I because he could hardly walk. 'Leave' does not mean refraining from work, but refraining from certain kinds of work: in particular, agricultural work. Kotas told me that their weekly days of leave were signs of respect for their gods. Sulli provided an alternate explanation for the observance of a holiday on Monday (Mandelbaum n.d.). Apparently the Kotas used to work for their Badaga 'partners' (those with whom they had established reciprocal obligations) on Mondays. After the Kotas and Badagas discontinued their relationship, the Kotas continued to recognize these days as out of the ordinary by refraining from working in their own fields. I draw upon de Certeau's special use of the word 'strategy', 'the calculation (or manipulation) of power relationships that becomes possible as soon as a subject with will and power (a business, an army, a city, a scientific institution) can be isolated. It postulates a place that can be delimited as its own and serve as the base from which relations with an exteriority composed of targets or Rain, God, and Unity among the Kotas 283 I06. 107. 108. 96. 97. 98. 99. 109. 100. 101. llO. 102. 103. 104. Ill. 112. 113. ll4. 105. 115. 116. threats (customers or competitors, enemies, the country surrounding the city, objectives and objects of research etc.) can be managed'. 'Disjuncture' is a key word in a current anthropological discourse about modernity. In a recent, rather influential article, Arjun Appadurai argues that displaced communities and intensive cultural transactions constitute an interactive system which is 'strikingly new' (Appadurai 1990). I would argue that such disjunctures are apparent even among communities like the Kotas, whose recent geographical ('ethnoscapes') displacements have been minimal but who are inextricably entangled in national and global 'finanscapes,' 'technoscapes,' 'mediascapes' and 'ideoscapes'. I would locate the primary disjuncture, among the Kotas, as temporal rather than territorial; but as we have seen in the rain ceremony, a self-awareness of inhabited space is an important (but not necessarily new) cultural value. What has occurred musically, the collection of tunes from diverse sources into a single repertoire (broadly encoding a positively valued past), is an example of 'reevaluation' (Slobin 1992: 70) which has occurred within the community and not for outsiders. That the rain ceremony was a yearly event in I937 is indicated in Mandelbaum's handwritten annotations to his typed notes from I4 May 1937. From the way the notes tend to include questions and answers, and from how they were annotated from inquiries on later days, I tend to think Sulli volunteered this information without prompting. Timothy, one of Mandelbaum's Christian Badaga interpreters, said, 'These fellows know how to make rain at the end of April. They are lazy and do not fork their fields in March and April as do the Badagas, or they hire someone to do it for them, so at the end of April their ground is as hard as stone. They go around their boundaries one day with music and by three o'clock that day an inch of rain must fall' (11 July 1937). The English gloss for the Kota is provided by L. Gunasekaran. The actual interview was conducted by Gunasekaran's father, Lakshmanan, because the family wanted a record of his speech, stories, and general knowledge of the old days. Va. Kamaten was literally on his deathbed, his conversation was very difficult to understand, and his narratives were interspersed with frequent gaps in memory. The recently deceased diviner of Kolme ·1, who became diviner after the plague, said forty of the eighty people living in the village at the time succumbed to the disease. See more on this deity in the discussion of the Nilgiri God tune (§2) and above, note 45. Term of affection for a girl. This phrase refers to a golden age in which the village was more populated. 'The houses at Kolme ·1 were so many and so closely spaced that the old-style leaf umbrella, which cannot be folded, could not be carried through the village' (Emeneau 1944,1:81). The streets or ke· rs correspond to exogamous units in the village. The 'middle street' people use a cremation site separate from that used jointly by the other two exogamous divisions. Here I return to these quotations to reiterate the relevance of Fernandez's arguments, 'Particularly in times of stress-where literal routines break down 284 Richard Kent Wolf and where we are constrained by false or moribund categories--do we turn to figurative language and the argument of images for a wider and more transcendent view of things. These are the times with which ritual and revitalization are most associated' (1986: 205-6). Ramanujan's lnteriur Landscape provides a very different, ancient Tamil conception of peace which coarticulates with emotions, time periods, flora and fauna, and music. Incidentally, u.r is also a Kota word, referring generically to non-Kota villages. A Kota village is termed ko· ka· ~or 'place of Kotas'. Kotas use specific words for the places Badagas, Todas and Kurumbas inhabit as well, i.e. for the communities with whom they have had extended contact. In the adjoining Coimbatore district, Beck (1976) provides an analysis of an entirely different sort that demonstrates a relationship between spatial orientation, person and the cosmos itself. Stories of Australian aboriginal places (this 'singing into existence' is captured nicely in Chatwin's 1987 novel, The Songlines) and Christian European pilgrimage sites are prominent examples. Keith Bassoa has written suggestively on the use of placenames and stories in everyday language among the Apache (Basso 1990). Writings on South Asian place-histories are extensive and need not be enumerated here. For further reference consult Harman (1989: 25-7). Although as Walker (1986: 156) notes, the 'gods of the sacred [dairy] places' are considered much more important than the mountain deities. Pilgrimage is a common structure through which diverse identities may be simultaneously instantiated. Gold (1989), for example, shows how individual life aspirations are projected into pilgrimage, and Fuller (1992: 211), the symbolic role of pilgrimage in Maharashtrian cultural unity. For a more general discussion of Indian regionalism, see Bernard Cohn's (1986) essay 'Regions Subjective and Objective: Their Relation to the Study of Modern Indian History and Society,' and the comments that follow in the volume. Actually, the tunes that belong to this 'detached' category are the 'twelve god tunes' (or arca·yf kof.s, tunes played under the rough-hewn hut [§l-§9]). They form a sub-genre of a more general genre of 'god tunes' which are still tied to the contexts for which they are named. I am told the factions remained until about 1960. Sulli may well have exaggerated a picture of social harmony to bolster his own image as an effective reformer. Inden's recent (1990) critique on Indology and histories of India puts such essentialisms in their place. .APPENDIX Rain, God, and Unity among the Kotas 285 1. Text to ve/ke· ve/ke· as sung by K. Pucan, 16 May 1991 ka· r a· vo· r kabitvd ay7Jo· r e· yo· !i· ro co· ym e· yo· !i· ro ve{ke· ve/ke· e· yo· !i· ro co· ym! (these first two lines were not lined up well with the tune and are not strictly characteristic of the verse-form of this song) ve!ke· ve/ke· e· yo· {i· ro va· co· ym totf,ba-le o· {i· ro va· co· ym ve/ke· ve{ke· e· yo· {i· ro va· co· ym ne· rl marka·l o· {i· ro va· co· ym vile marka·l o· !i· ro va· co· ym ve/ke· ve{ke· e· yo· !i· ro va· co· ym ne· rl marka·l o· fi· ro va· co· ym ponic lie o· fi· ro va· co· ym ve{ke· ve{ke· e· y o-!i· ro va· co· ym ve/ke· ve/ke· e· yo· {i· ro va· CO· ym viky marka·l o· {i· ro va· co· ym 117. 118. 119. 120. 121. 122. 123. 124. 125. 286 Richard Kent Wolf Rain, God, and Unity among the Kotas 287 2. Prayer at ponic as Sulli dictated to Mandelbaum (I have standardized the spelling) co· mi kamatraya amno· r ayJJo· r ra· mco· mi rangco· mi betdamn (calling gods, three old Kota deities and three new ones, Ramaswamy, Rangaswamy and Bet damn). a· m nin mog kunj vilaJJ el vitl may we're your sons 1 sowing and all having completed rain i·la· d kactm patr co· ym ka·lk being without difficulty experiencing god's leg at copni · r viglk a· m elm ni · mn pacrvaglk vade· m to fall in prayer we all you to grab and come came kama(r ayJJ venvetd itlme· rangrayJJ Kamatr father (god) living in Venvetm and Rangr father (god) ka· rmo· r!ume· itme·l a· m ennm varkvd2 in Karamadai also since you are living we how prospering adnk· yr a· m doqa· fo· r kuna· fo· r emd par for that reason we with big men and small men our barrel drum tabatk kofo· r kobo· r ni· m paco· ko· m i!lr frame drum with shown with horn with you we grab and go thus vade· m emn marva· d a· m a· to· n da· kl we come us without forgetting (?) one who calls like va· ve· rm ad nimd pa· dtn a· m pace· m ? (we come?) that your feet we grab 3. Song composed byTu-j (father ofS. Raman) for Payrn's daughter Rangumathi, as sung by Kembi of Kurgo-j village, 7 June 1992. talala . . . en av ranguma· dy enke· ngo ranguma· dy kolme·l ko· ka·l ko· r a· rad ko· ka·l talala . . . naryke· TO· r du· vk ve· ku· Jla· re ranguma· dy talala . . . en av ranguma· dy enke· ngo ranguma· dy talala . . . a· ke· TO· T dU· Vk Ve· ra· Jlre ranguma· dy en av ranguma· dy enke· ngo ranguma· dy kolme·l ko· ka·l ko· r a· rad ko· ka·l en av ranguma· dy enke· ngo ranguma· dy talala . . . I. Mogmeans 'soni' and kunjmeans 'men as children of god' (DEDR 1646). 2. Probably variant conjugation of DEDR 5372 (vadkj')-to prosper-is perhaps pronounced more like Tamil val? 288 Richard Kent Wolf Rain, God, and Unity among the Kotas 289 REFERENCES Appadurai, Arjun 1981. 'The Past as a Scarce Resource'. Man n.s. 16: 201-19. ----1990. 'Disjuncture and Difference in the Global Economy'. Public Culture 2, no. 2: 1-23. Babb, Lawrence A. 1975. The Divine Hierarchy: Popular Hinduism in Central India. New York: Columbia University Press. ----1981. 'Glancing: Visual Interaction in Hinduism'.joumal of Anthropological Research 37: 387-401. Babiracki, Carol 1990. Musical and Cultural Interaction in Tribal India: The Karam Repertory of the Mundas of Chotanagpur. Unpublished Ph.D. dissertation; University of Illinois, UrbanaChampaign. Basso, Keith H. 1990 Western Apache Language and Culture: Essays in Linguistic Anthropology. Tucson: University of Arizona Press. Beck, Brenda 1976. The symbolic merger of body, space and cosmos in Hindu Tamilnadu. Contributions to Indian Sociology n.s. 10: 213-43. Brass, Paul R. 1991. Ethnicity and Nationalism: Theory and Comparison. New Delhi: Sage publications. Burrow, T., and M. B. Emeneau 1984. A Dravidian Etymological Dictionary [DEDRJ. 2nd edition: Oxford: Clarendon Press. Chatwin, Bruce 1987. The Songlines. New York: Penguin Books. Clifford, James 1988. The Predicament of Culture: Twentieth Century Ethnography, Literature, and Art. Cambridge, Mass.: Harvard University Press. ---1990. 'Notes on (Field) Notes' In Roger Sanjek, Fieldnotes, 47-70. Ithaca: Cornell University Press. Clifford,James, and George E. Marcus. (ed.) 1986. Writing Culture: The Poetics and Politics of Ethnography. Berkeley: University of California Press. Cohn, Bernard S. 1966. 'Regions subjective and objective: Their relation to the study of modern history and society'. In Robert Crane (ed.). Regions and Regionalism in South Asian Studies. An Exploratory Study. Durham: Duke University. ----1987 (1961]. 'The pasts of an Indian village'. In Bernard S. Cohn. An Anthropologist Among the Historians and Other Essays, 88-99. Delhi: Oxford University Press. Comaroff, John, and Jean Comaroff. 1992. Ethnography and the Historical Imagination. Boulder: Westview Press. Crapanzano, Vincent 1980. Tuhami: Portrait of a Moroccan. Chicago: University of Chicago Press. Daniel, Valentine 1984. Fluid Signs. Being a Person the Tamil Way. Berkeley: University of California Press. de Certeau, Michel. 1984. The Practice ofEveryday Life. Steven Randall trans. Berkeley: University of California Press. Eck, Diana L. 1981. Darsan: Seeing the Divine in India. Chambersburd, PA: Anima Books. Emeneau, Murray B. 1937-8. 'Ritual games of the Kotas'. Bulletin of the Sri Rama Varma Research Institute 5: 114-22; 6: 1-6. - - - - 1944. Kota Texts. 4 parts. Berkeley: University of California Press. Emeneau, M.B., and T. Burrow 1962. Dravidian Borrowings from IndoAryan. [DBIAJ. Berkeley: University of California Press. Fernandez, James W. 1966. 'Unbelievably subtle words: Representation and Integration in the Sermons of an Mrican Reformative Cult'. History ofReligions: An Internationaljournal for Comparative Historical Studies 6, no. 1: 43-69. - - - - 1982. Bwiti: An Ethnography of the Religious Imagination in Africa. Princeton: Princeton University Press. - - - - 1986. Persuasions and Performances: The Play of Tropes in Culture. Fuller, C. J. 1992. The Camphor Flame: Popular Hinduism and Society in India. Princeton: Princeton University Press; New Delhi: Penguin Books India. Gold, Ann Grodzins 1989. Fruiiful journeys. The Ways of Rajasthani Pilgrims. Delhi: Oxford University Press. Gonda,Jan. 1962. Les Religions del'Inde. 3 vols. tr., L.Jospin. Paris: Payot. (Cited in Harman 1989). Hanchett, Suzanne 1988. Coloured Rice: Symbolic Structures in Hindu Family Festivals. Delhi: Hindustan Publishing Corporation. Harman, William P. 1989. The Sacred Marriage of a Hindu Goddess. Delhi: Motilal Banarsida:ss. Hill Fox (Pseud. for A. K Ranganathan) 1960. Nilakiri Carittiram ( munram piikam). (Nilgiri History (third part)]. Ootacamund: Ti kiraQ.t Indian Publications. . Hockings, Paul 1980. Ancient Hindu Refugees. Badaga Social History 1550-1975. The Hague: Mouton Publishers. ( ed.) 1989. Blue Mountains. The Ethnography and Biogeography of a South Indian Region. Delhi: Oxford University Press. 290 Richard Kent Wolf Rain, God, and Unity among the Kotas 291 Hockings, Paul, and Christiane Pilot-Raichoor 1992. A Badaga-English Dictionary. [BED]. Berlin: Mouton de Gruyter. Inden, Ronald 1990. Imagining India. Cambridge, :MA: Blackwell. Kaden, Christian 1984. Musiksoziologie. Berlin: Verlag Neue Musik. (Cited in Stockman 1991). Knorr-Cetina, Karin, and A. V. Cicourel 1981. Advances in Social Theory and Methodology: Toward an Integration ofMicr(}o and Macr(}o sociologies. Boston: Routledge and Kegan Paul. Lan, David 1985. Guns and Rain: Guerillas and Spirit Mediums in Zimbabwe. Berkeley: University of California Press. Lengerke, Hans J. von, and Franois Blasco 1989. 'The Nilgiri environment'. In Paul Hockings (ed.) Blue Mountains, 20-78. Delhi: Oxford University Press. Levi-Strauss, Claude 1966. The Savage Mind. Chicago: University of Chicago Press. Lutkehaus, Nancy 1990. 'Refractions of Reality: On the Use of Other Ethnographers' Fieldnotes. In Roger Sartiek (ed.) Fieldnotes, 303-23. Ithaca: Cornell University Press. Mandelbaum, David G. n. d. Field Notes. Bancroft Library collection no. 89/129cz. Berkeley: University of California. - - - - 1939a. 'Agricultural Ceremonies among Three Tribes of Travancore'. Ethnos 4: 114-128. - - - - 1939b. 'The Jewish Way of Life in Cochin'. Jewish Social Studies 1: 423-60. - - - - 1940. The Plains Cree. Anthropological papers of the American Museum of Natural History (New York), 37 part 2. - - - - 1941. 'S~cial Trends and Personal Pressures'. In Spier, Hallowell and Newman (eds.) Language, culture and personality: Essays in Memory ofEdward Sapir, ed. Menasha Wisconsin: Sapir Memorial Fund. Repr. in Charles Leslie (ed.). Anthropology of Folk Religion, 221-55. New York: Vintage Books (1960). 1960. A reformer of his people. In Joseph Casagrande (ed.). In The Company of Men. New York: Harper. 1966. 'Transcendental and Pragmatic Aspects of Religion'. American Anthropologist, 68: 1174-91. - - - - 1989. 'The Kotas in Their Social Setting'. In Paul Hockings, (ed.) Blue Mountains, 144-85. Delhi: Oxford University Press. Manuel, Peter 1993. Cassette Culture: Popular Music and Technology in North India. Chicago: University of Chicago Press. Marcus, George E. 1986. 'Contemporary Problems of Ethnography in the Modern World System'. In Clifford and James Marcus, (eds.), Writing Culture George E. The Poetics and Politics ofEthnographY! 165-93. Berkeley: University of California Press. Mehta, P. L. 1991. Constitutional Protection to Scheduled Tribes in India: In Retrospects and Prospects. Delhi: H. K. Publishers. Metz,Johann Friedrich 1864. The Tribes Inhabiting the Neilgherry Hills: Their Social Customs and Religious Rites. 2nd edition. Mangalore: Basel Mission Press. Munn, Nancy D. 1974. 'Symbolism in a Ritual Context: Aspects of Symbolic Action'. In J. J. Honigmann (ed.) Handbook of social and cultural anthropology, 579-612. New York: Rand McNally. Needham, Rodney 1967. 'Percussion and Transition'. Man (n.s.) 2: 606-14. - - - - 1975. 'Polythetic Classification: Convergence and Consequences'. Man, (n.s.) 10: 349-69. Ortner, Sherry. 1979. 'On key symbols'. In William A. Lessa and Evon Z. Vogt. (eds.) Reader in Comparative Religion: an Anthropological Approach, 4th edn.; 92-8. New York: Harper and Row. Peirce, Charles S. 1955. Philosophical Writings of Peirce. Justus Buchler, (ed.). New York: Dover. Ramanujan, A. K. n.d. The Interior Landscape: Love poems from a Classical Tamil Anthology. Delhi: Clarion Books. Sanjek, Roger, (ed.) 1990. Fieldnotes: The Makings of Anthropology. Ithaca: Cornell University Press. Schechner, Richard 1983. Performative Circumstances: From the Avant Garde to Ramlila. Calcutta: Seagull Books. Schneider, David 1968. American Kinship: A Cultural Account. Englewood Cliffs, N.J.: Prentice Hall. Shostak, Marjorie 1981. Nisa: The Life and Words of a ! Kung Woman. Cambridge, Mass.: Harvard University Press. Singh, K. Suresh 1972. The Tribal Situation in India. Patna: Indian Institute of Advanced Study. Slobin, Mark 1992. 'Micromusics of the West: A Comparative Approach'. Ethnomusicology' 36, no.1: 1-87. Smith, Robert]. 1990. Hearing Voices, Joining the Chorus: Appropriating someone else's fieldnotes. In Roger Sanjek. (ed.) Fieldnotes, (ed.) 356-70. Ithaca: Cornell University Press. Stockman, Doris 1991. Interdisciplinary Approaches to the Study of Musical Communication Structures. In Bruno Nettl and Philip V. Bohlman. (eds.) Comparative Musicology and Anthropology of l 292 Richard Kent Wolf Music: Essays on the History ofEthnomusicology, 318-41. Chicago: University of Chicago Press. Sulli, K. n. d. 'Culliyin varalaru' (Sulli's life history). Unpublished handwritten autobiography (in Tamil). Tambiah, Stanley 1981. 'A Performative Approach to Ritual'. Proceedings of the British Academy 65: 113-69. Tamil Lexicon. 1982. Madras: University of Madras. Thapar, Romesh (ed.) 1977. Tribe, Caste and Religion in India. Meerut: G. Wasani for Macmillan India, Ltd. Thurston, Edgar, and K. Rangachari (eds.) 1909. 'The Kotas'. Castes and Tribes of Southern India, 4: 3-31 Madras: Superintendent, Government Press. Trawick, Margaret 1991. 'Wandering Lost: A Landless Labour's Sense of Place and Self'. In Aijun Appadurai, Frank J. Korom and Margaret Mills (eds.) Gender, Genre and Power in South Asian Expressive Traditions, 224-66. Philadelphia: University of Pennsylvania Press. Turner, Victor 1977. 'Symbols in Mrican Ritual'. In Janet L. Dolgin, David S. Kemnitzer, and David M. Schneider (eds.) Symbolic Anthropology: A Reader in the Study of Symbols and Meanings, 18394. New York: Columbia University Press [Reprinted in Science (1973), 179: 1100-5. Walker, Anthony 1986. The Toda of South India: A New Look. Delhi: Hindustan Publishing Corporation. Wolf, Richard K. 1993. 'The Apotheosis of a Musical Repertoire: Identity and History in the 'god tunes' ofthe Kota tribe (Nilgiri Hills, South India)'. Unpublished paper read at the Midwestern Meeting of the Society for Ethnomusicology, Ann Arbor, MI. Zvelebil, Kamil1979. The Irula (Erla) Lanr;uage. 3 parts. Wiesbaden: Otto Harrassowitz. - - - - 1988. The Irulas of the Blue Mountains. Syracuse: Maxwell School of Citizenship and Public Mfairs, Syracuse University.