Religion in Tibetan Society: A New Approach

RELIGION IN TIBETAN SOCIETY A NEW APPROACH PART TWO A STRUCTURAL MODEL Geoffrey Samuel Newcastle, N.S.W. In the second of this series of two papers I shall look at a particularly well studied Tibetan population, the Sherpas of Nepal. I hope to show that previous accounts of Sherpa religion have been incomplete in certain crucial particulars, and that the approach I have outlined gives a more comprehensible picture of the totality of Sherpa religious experience and activity. The majority of the Sherpas live in high valleys in the East of Nepal, on the south side of the main Himalayan range. Some have settled in the Kathmandu valley, and there is also a large population in the area of Darjeeling in India, but the centre of Sherpa society remains the two districts of Shorung (Solu in Nepali) and Khumbu in East Nepal, about 70 to 100 miles east of Kathmandu, and it is to these two districts that the anthropological accounts refer. The population of Sherpas in this area in 1965 was about 15,000 (Oppitz 1974: 235) The Sherpas trace their origin to a group of clans from East Tibet (Kham) who settled in Shorung and Khumbu, probably at the end of the 15 th and beginning of the 16 th century. While the distinctiveness of the Sherpas in relation to other Tibetan speaking populations in Nepal has been stressed by Christoph von Furer-Haimendorf (1964) there is no doubt that culturally the Sherpas are Tibetan. The Sherpas speak a Tibetan dialect, their way of life is in most respects characteristically Tibetan, they absorb Tibetan immigrants into their society easily and rapidly, and as I shall show their religion is quite within the mainstream of Tibetan religious practice. This point is worth making because one of the authors I shall discuss, Friedrich Funke, regards Sherpa religion as not typical of Tibetan religion today. I shall consider the grounds for his assertion below. The Sherpas have been probably the most accessible population of Tibetans in modern times. At any rate they have been the most intensively studied by anthropologists, as well as being the Tibetan population which has had the most extensive contact with Westerners in general over the last twenty-five years through their participation in mountaineering expeditions. Here I shall be mainly concerned with the three anthropological works dealing at length with religion; Furer--Haimendorf's general ethnography of 1964, and the studies of Sherpa religion by Friedrich Funke (1969) and Robert Paul (1970). Some other works will be referred to however, in the course of my argument. The Nyingmapa in Tibetan Religion To begin with I shall discuss some general features of Sherpa religion. While the practice of Buddhism by the Sherpas is typically Tibetan, it is typical of one particular

100 | Kailash Tibetan monastic order, the Nyingmapa (cf Snellgrove 1957, Tucci 1970). The Nyingmapa are the oldest of the four main orders. By origin they were not so much a homogeneous group deriving from a particular lama and his monastery, as were the other orders, but a kind of residual category of monasteries which remained outside the new monastic orders of the 12 th to 15 th centuries, the Kagyupa, Sakyapa and Gelukpa. Their most distinctive feature, referred to in their name, is their adherence to the 'old' (T rnying ma)¹ tantras associated with Padmasambhava and his missionary activity in Tibet in the 8 th century, and to the rituals discovered subsequently in the form of 'hidden' texts claimed to date from the time of Padmasambhava. Some of the characteristically Nyingmapa tantric ritual cycles, in particular the 'Union of the Precious Ones' widely used by the Sherpas, are centred on the deified figure of Padmasambhava. 2 Another important cycle of texts is that of the zhi 'khro (T; Sh shetu) the mandala of the tranquil and fierce divinities, best known in the West through the Tibetan Book of the Dead' which forms part of this cycle (cf Tucci 1949, Fremantle and Trungpa 1975). However while these cycles are to some degree specially connected with the Nyingmapa, they are used also by lamas and monks of the other monastic orders. More importantly, they are no different in basic nature from the rituals of the 'new' tantras characteristic of the other orders. There are some philosophical differences between the Nyingmapa and the others and interpretations of Buddhist philosophy given by the ethnographers I am discussing perhaps reflect views characteristic of the Nyingmapas on some occasions. These differences are of little or no significance to the average layman. 3 There are however some differences between the Nyingmapa and their polar opposites, the Gelukpa, the 'reformed' order founded by Tsongkhapa and which was dominant in the Dalai Lama's realm, which are significant in this discussion. In particular it is not necessary for Nyingmapa lamas to be celibate, and many are not. These non-celibate lamas range from relatively low status 'village lamas', that is laymen with some tantric training, to the heads of the most important Nyingmapa monasteries. Also, and consequently, the office of lama at a local temple, or of abbot at a monastery, is sometimes hereditary among the Nyingmapa. Historically it was the Gelukpa who reasserted the Indian Buddhist ideal of monastic celibacy and consequently eliminated hereditary abbacy in their monasT followed by italics indicates Tibetan terms in transliteration (following Wylie 1959); S indicates Sanskrit, and Sh the Sherpa dialect of Tibetan in phonetic transcription (following R.A. Paul 1970 or Furer-Haimendorf 1964; in some cases where the forms differ significantly I have given both). Tibetan and Sherpa proper names (Nyingmapa, Shorung) are used in the text in phonetic transcription without special indication. 2 Texts from this cycle are translated by Snellgrove (1957: 228-9, 245-61). 3 For philosophical peculiarities of the Nyingmapa, cf. Tucci 1970: 94-106; Guenther 1963, 1966, 1972. Specifically Nyingmapa concepts are possibly reflected in R.A. Paul's account, e.g. 1970: 342-3, 376-7. .

Religion in Tibetan Society | 101 teries. In both these respects the other two Buddhist monastic orders (Sakyapa and Kagyupa) represent an intermediate situation. 4 The fifth, quasi-Buddhist Tibetan monastic order, Bon, has no monasteries in Sherpa country, though it is of considerable importance among Tibetan populations in North-West Nepal (Snellgrove 1961, 1967, Furer-Haimendorf 1975). It nevertheless requires further mention because of the important part it plays in Funke's account. Funke, following Hoffmann and thus, indirectly, the polemical writings of Tibetan historians of the Gelukpa order, uses the term Bon to refer to both the pre-Buddhist religion of Tibet and the contemporary adherents of the monastic order of Bon. It has become apparent over the last decade or so that this usage is highly misleading, as well as historically inaccurate. The inaccuracy lies in labelling pre-Buddhist Tibetan religion as a whole as 'Bon' when the bon po were only one of several classes of ritual practitioners in Tibet during the royal period, and were probably largely indebted for their ritual practice to non-Tibetan sources. What is more important is that the Bon of today is a monastic order similar in all its basic features to the four strictly Buddhist monastic orders. It shares many of the same rituals in slightly modified form (e.g., the zhi 'khro cycle), and it stands in essentially the same relationship as the Buddhist orders to what Tucci (1970) calls the Volksreligion of Tibet and Stein (1972: 191-229) the 'nameless religion' of Tibetan tradition, that is the cult of local deities, the concern with malevolent spirits, the practice of exorcising ritual etc. These matters are in no respect Bon specialities, although it seems that the village magicians (T sngags pa) mentioned above tend to be associated with Bon or Nyingmapa rather than with Gelukpa, as might be expected from the Gelukpa emphasis on monastic celibacy for its students of tantra. At the same time both Nyingmapa and Bon do have monasteries like those of the Gelukpa, containing celibate monks observing their vows and studying Buddhist philosophy (cf Snellgrove 1961, Kvaerne 1976 b). At most one could say that there is something of a continuum in terms of involvement with 'folk religion' from the Gelukpa with their emphasis on the ideal of the celibate monk, academic study and right motivation, to the less exclusively academic and monastic Nyingmapa and Bon. Since we do not have adequate studies of the involvement of lamas in 'folk religion' in Gelukpa-dominated areas, it is not possible to evaluate the degree of variation between the two poles of the continuum. But even the Gelukpa carry out initiations to strengthen the life-force in laymen, perform monastic dances and other rituals against malevolent spirits, and perform rituals to guide the consciousness of the dead to a happy rebirth. And if my analysis of Tibetan religion is right, it is hard to see how the Gelukpa could have maintained their lay support if they were not involved with 'folk religion'; because these rituals are essential components of the reciprocity between monasteries and laymen. 4 On this question cf. Stein 1972: 191, 22-47; Kvaerne 1967 a and 1976 b; Macdonald 1971.

102 | Kailash Outline of Sherpa religion All of the three authors whom I am considering are in agreement on the general outlines of Sherpa religion, and their account is confirmed by other descriptions (e.g. Snellgrove 1957, S.O. Paul 1970) and by my own observations during a brief visit to Shorung. Some minor differences arise from the fact that Furer-Haimendorfs account refers to the region of Khumbu, while Funke and Paul both worked primarily in the more southerly area of Shorung. Minor differences in social structure between the two areas (cf.R.A.Paul 1970: 256-275) appear to have little effect on the basic forms of religious behaviour. Funke regards Shorung as less 'Tibetanized' in religious matters than Khumbu, and one might expect a stronger Nepali influence in Shorung, which is further south and nearer the Nepal Valley. However specific examples are not apparent of this influence in the field of religion. S The Sherpa's pantheon corresponds to that described in my first paper for Tibetans in general. The tantric gods and protective gods are mostly the concern of the monastic cult, with the exception of those few deities who are familiar to the layman: Avalokitesvara (Sh Pawa Cherenzi), Padmasambhava (Sh Curu Rimpoche ), Tara (Sh Drolmd). Lay ritual towards these gods consists of the recitation of mantras or short prayers. The local gods are of considerable importance. The most powerful are the yul lha (T) or regional deities, which are associated with major mountains in Shorung and Khumbu, and are known simply as Shorung yul lha and Khumbu yul lha. Several lesser gods are linked with smaller local mountains and hills, and these are of concern mostly to particular descent groups with a hereditary duty to perform an offering ritual of some kind to them, usually once or twice yearly. Two of the main Himalayan peaks, Everest and Gauri Shankar, are associated with two of a set of mountain goddesses well known in Tibet too, the Five Sisters of Long Life (T tshe ring mched Inga), but they have no popular cult among the Sherpas. These local deities are in general benevolent, but can be offended and cause illness and other misfortune. Overlapping with these mountain gods in power and general nature are another class of spirits, the water-spirits or lu (Sh; T klu, equated in literary texts with S naga). Like the mountain gods they can be helpful to man but are also capable of ill-will. They range from lu associated with a lake high up on the slopes of Shorung yul lha mountain, Uomi Tso or Milk Lake, (a place to which many Sherpas make an annual pilgrimage) to lu associated with local streams and lakes, and lu living in the houses of men (within which a small shrine 5 Except perhaps for the use of non-Sherpa shamans in Shorung, and the importance of the cult of the lu in both Shorung and Khumbu, both of which are discussed below. As will be seen, these are not very significant exceptions. 6 These correspond to the Tibetan names, 'phags pa spyan ras gzigs, gu ru rin po che, sgrol ma. $;

Religion in Tibetan Society | 103 is often erected for them). They are particularly offended by pollution of their lakes and streams, or by bad smells in the case of household lu (cf. Funke 1969: 27-33, S.O. Paul 1970: 241-250). The lu appear to be the most important representatives for the Sherpas of the class of deities which includes also the minor gods of earth and fields (T sa bdag 'earth-owner' etc.). The Sherpas know these beings too - the sa bdag (Sh sabtak) are regarded as the husbands of lu (who are generally female) according to Ortner (S.O. Paul 1970: 242-248) - but they are given relatively little emphasis. The Sherpa lamas know the full range of Tibetan malevolent spirits, though for the laymen they are generally all lumped together in a general category of shrindi or hrendi (Sh: T srin bdud?). As in the general model discussed in the first paper, these beings are unambiguously malevolent and protection against them is one of the major purposes of all collective ritual. Witches (Sh pem) and malignant ghosts (SH norpa) are also objects of some concern and are protected against in similar ways. There are a number of different kinds of religious practitioners among the Sherpas, conforming to the general classes suggested in the previous paper. In the first place (though historically they are of recent origin) there are several monasteries - in the sense of communities of celibate monks - such as those of Tengboche in Khumbu, discussed by Furer-Haimendorf (1964) and Jiwong in Shorung, visited by Snellgrove (1957). Most of these monasteries have incarnate lamas as their abbots or (where in their minority) titular heads. Buddhist tantric rituals are also performed by lay tantric priests (Sh lama or banzen) belonging mostly to a hereditary lama clan. Some of the larger village temples have a number of these 'village lamas' (to use Furer-Haimendorf's term) attached to them, who perform rituals regularly in the temple, and thus constitute communities of married lamas similar to those of Tibet. Two groups of specialists deal with divination and minor rituals to placate local gods and spirits; the lhawa (Sh; T tha pa) or spirit-mediums who divine while in a state of possession by a local god, and the mindung or minung (Sh; T mig mthong) who use various material aids to clairvoyance (the distinction between these types is not rigid and a range of different techniques is used, cf.R.A.Paul 1976 a: 144). From Furer-Haimendorf's account (1964:254-263) it is clear that both Ihawa and mindung work in close connection with village and monastic lamas, and are in a generally subordinate position to the lamas. Lamas and lhawa may recommend clients to each other, but only lamas can see the higher tantric deities, such as the gods of Dewachen (Sh; T bde ba can, the Western Paradise of Amitabha), let alone incarnate them (through the processes of tantric ritual). Lhawa may be trained by lamas (Furer-Haimendorf 1964: 255-56); the reverse would be inconceivable. Lhawa and mindung usually recommend the making of offerings to propitiate the shrindi, witch, malignant ghost or local deity who is responsible for the illness or misfortune in question.

} 104 | Kailash In Shorung many of the minung/Thawa practitioners are not Sherpa. They belong to non-Sherpa castes resident in the area such as the Newar and Chetri (R.A. Paul 1976 a: 144). While the style of their performances may be closer to that of Nepali shamans in other parts of Nepal (cf. Hitchcock and Jones 1976) than to that of the Tibetan practitioners furter north, the role that they perform from the point of view of the Sherpas is inter-changeable with that of Sherpa or Tibetan mediums and soothsayers. The collective rituals of the villagers are concerned with the protection of the village, its crops and herds from malevolent spirits, and the general welfare of the village community. These rituals are all of standard Tibetan types, and are taken from tantric liturgical cycles such as that of the 'Union of the Precious Ones', referred to above (cf. also Funke 1969: 255-279). The life-initiation (Sh tseong = T tshe dbang) to strengthen the life-force of the villagers, and the symbolic destruction of evil spirits in the form of an effigy, are both incorporated in the major annual village festival, dumje ( Sh), which includes masked dances in the larger centres (cf. Furer-Haimendorf 1964 : 185-205, Funke 1969: 116-138). The annual manirimdu ceremonies at the monasteries include the same two basic elements, though the dance-sequences are much more elaborate and are performed by the monks (cf. Furer-Haimendorf 1964: 210-224, Jerstad 1969, R.A. Paul 1970: 430). One annual collective ritual, the niungne fast (Sh; T smyung gnas) is concerned with 'other-worldly' ends, namely the accumulation of merit by those who participate (cf. Haimendorf 1964: 180-185, Funke 1969: 116, S.O. Paul 1970: 180 ). The acquisition of 'merit' (good karma) at an individual level is however a constant theme of Sherpa life, as with other Tibetans and among Theravada Buddhists. Merit can be acquired by recitation of scriptures and other ritual acts, and by the construction and repair of religious shrines such as mani-walls and chorten (Sh; T mchod rten = S stupa), circular edifices containing religious relics. One also gains merit through such socially useful acts as building bridges and rest-houses, giving charity and acting as a peace-maker (Furer-Haimendorf 1964: 272- 275). Bad karmic acts ('sin' in Furer-Haimendorf's terminology, though the word is perand haps misleading) include theft, cheating, adultery, killing or hurting any living creature, the like. As Furer-Haimendorf observes, Sherpas are aware that their way of life necessarily involves the commission of some of these 'sins' - yak-breeders have to castrate their bulls, although it inflicts pain upon them, trees must be cut down and living beings occasionally killed. Generally the so-called 'sins' are not so much seen as socially reprehensible, but rather as actions carrying with them their own punishment, through the operation of karma. What is expected is that the necessary minimum of sin involved in daily life should be made up for by the performance of meritorious works. Social disapproval is more concerned with breaches of the rules of clan exogamy, or disobedience to orders of the village assembly. Merit can also be gained by the sponsoring of ritual performances by others, usually monks or village lamas, either at their monastery or at one's own home. Such ceremonies are performed particularly in cases of illness, and appear to make up much of the work pe 19 pe T D g S 1 1

Religion in Tibetan Society / 105 performed by the village lamas and the monks for the villagers (cf. Furer-Haimendorf 1964: 164-166, R.A. Paul 1970: 484 ff.). The other main class of occasions on which they perform rituals for the lay population are the rites accompanying and following death. These are of standard Tibetan type, involving the recitation of the 'Tibetan Book of the Dead' (Evans-Wentz 1960, Fremantle and Trungpa 1975) by lamas and monks, and the guiding of the consciousness of the deceased to a happy rebirth by the presiding lama. The Sherpas however usually dispose of their dead by cremation, not by feeding the bodies to birds and wild animals in the Tibetan manner. (In Tibet cremation is generally reserved for high lamas.) Furer-Haimendorf and the anatomy of Sherpa values Furer-Haimendorf's book is valuable, perhaps, mainly at a descriptive and ethnographic level. It provided in fact the first, and so far, the only adequate ethnographic account of a Tibetan community, and to suggest that it is theoretically weak is perhaps to accuse the author of failing to do what he did not set out to do. However since this paper is concerned with theory, this aspect of Furer-Haimendorf's book deserves some consideration. To the extent that The Sherpas of Nepal does have a theoretical orientation it lies in Furer-Haimendorf's concern with values and moral concepts. Furer-Haimendorf's fieldwork has been with Hindu, Buddhist and tribal societies, many of them along the HinduBuddhist interface of the Himalayas. For him religion is of interest primarily in terms of its role in shaping morality and values in these contrasted societies, and this dimension of religion permeates the Sherpa book and much of Furer-Haimendorf's other work. A paper published two years before the monograph on the Sherpas is an explicit comparison of the moral concepts and values of the Sherpas, the Hindu Chetris of Nepal, and the tribal Dafla of Assam (Furer-Haimendorf 1962). Furer-Haimendorf says of the 1964 work on the Sherpas: What I have set out to do is to describe and analyse the type of society in which the Sherpas have developed their spirit of independence, their ability to cooperate smoothly for the common good, their courtesy and gentleness of manner and their values which are productive of an admirable balance between this-worldly and otherworldly aims (1964:xix). It is in this context that he describes the openness of the Sherpas to outside immigrants, the relative (but only relative) lack of status distinctions and emphasis on equality in family life and in village organization, and it is in this context too that he discusses reli gion. Furer-Haimendorf's material on religion takes up the last four chapters - more than half of his book. These chapters deal in turn with monasteries and village lamas, with the practice of religion (meaning the seasonal rituals and the death rituals), with the 'control of invisible forces' (shrindi, witches, ghosts and local deities), and with values and moral con-

106 | Kailash . . . . cepts. This last chapter is evidently meant to sum up the book's argument and to answer the question implicit in the quotation given above: what is it in the religion of the Sherpas which is conducive to the Sherpa's 'spirit of independence, their ability to co-operate their courtesy and gentleness of manner' etc? Furer-Haimendorf finds his answer in the ideology of sin and merit, in which 'sinful' actions are not so much socially despised as simply tied up inevitably with their own punishment. For the Sherpas 'the individual[is] a free moral agent, responsible for his actions and capable of moulding his fate in the next life' (Furer-Haimendorf 1964: 288). Furer-Haimendorf sees this attitude as being at the basis of the tolerance and humanity of the Sherpa way of life. I would not personally argue with the point Furer-Haimendorf is making here. Most anthropologists, myself included, who have worked with Sherpa or Tibetan populations would concur with Furer-Haimendorf's sympathetic picture of their character and way of life. It seems natural to link this to their Buddhist religion, especially in view of the marked contrast to both the Hindu and tribal peoples in neighbouring regions (cf FurerHaimendorf 1962). At the same time it is noticeable that the 'moral concepts and values' framework, such as it is, fails to encompass Furer-Haimendorf's material on religion. He relates Sherpa morality to the Buddhist concepts of karma, merit and sin; but most of the material on religion given in the previous three chapters has little to do with any of these concepts. In some ways it is to the advantage of Furer-Haimendorf's work that description is not subordinated here to a theoretical scheme. By contrast Funke and Robert Paul, whose books are shaped by their authors' theoretical frameworks, provide in some respects much more one-sided accounts. Funke and the diffusionist approach to Sherpa religion Friedrich Funke's book (1969) reads strangely to an English-speaking anthropologist, since its approach is within a tradition which is completely out of fashion within the EngIt should be noted that Furer-Haimendorf describes the Sherpa ethos as differing somewhat from the of other Tibetan groups: Although the Tibetan-speaking people of the western border regions also practice Buddhism and live in the seclusion of remote and sparsely populated Himalayan valleys, one does not encounter there a basic outlook comparable in humanity and breadth with that prevailing in Khumbu. Thus we would err if we ascribed the general ethos of Sherpa society solely to the effect of Buddhist ideology. (1964: xix). As I mentioned in Part I of this article, Furer-Haimendorf suggests that another relevant factor is the relative freedom of the Sherpa communities from the control of external authorities.

-Religion in Tibetan Society (107 lish-speaking world today, that of diffusionism. Much of Funke's book is concerned with explaining Sherpa religion through a procedure of dividing it into various supposed components and then advancing hypotheses to account for the presence of those components, mostly in terms of diffusion from other culture areas. Since I am concerned primarily with understanding Sherpa religion as a totality, as a functioning aspect of Sherpa society, much of Funke's argument is irrelevant to my approach. However I think it can be shown that not only does his approach lead to a somewhat one-sided view of Sherpa religion, but that he also fails to make a very good job of what he himself sets out to do. As with the classical diffusionists, Funke places much emphasis on material culture. Much of the book is taken up with extensive descriptions of village temples ( particularly that of Junbesi in Shorung) and of religious paintings, and with a number of translations of ritual texts. These translations are perhaps the most useful part of the book, but Funke does little with them, mainly because he wants Sherpa religion to be 'non-Lamaist' - he regularly used 'Lamaism' for Tibetan Buddhism, in the style of Waddell - and the texts . obviously don't fit this picture. By contrast with the extensive description of material objects, there is relatively little in the book on how the Sherpas themselves view their religion. Funke's view of Tibetan religion derives from Waddell (1967) and particularly from Hoffmann (1961). As I mentioned above, Hoffmann's views on Bon are no longer generally accepted; unfortunately Funke incorporated them in his own work in a somewhat simplistic form. Thus for Funke Tibetan Buddhism ('Lamaism') is among the Sherpas a thin veneer over pre-Aryan hill cults, old Persian and in particular the 'pre-Buddhist Bon religion': When the protoclans of the Sherpa emigrated from Kham and moved southwards to their present-day habitat in Nepal, they brought with them age-old forms of beliefs and rites and these have survived there up to the present day, whereas in Tibet proper these components of archaic religious views gradually lost their separate identity during five centuries of political and religious power struggles. It is true that in modern times the Tibetan Nyingmapa sect from its border monastery of Rongphu on the north side of Mt. Everest has constantly sought to missionise those Sherpa living across the border and thus win them back to orthodox Lamaism, but their efforts have met with little or no success. (1969: 289; Funke's English Summary ). 8 Incidentally Funke's approach is relatively mild by comparison with that of Siegbert Hummel, another diffusionist who took a short look at Sherpa religion in a paper based on some of Furer-Haimendorf's data (1967). After an exhilarating tour through the myths and rites of most of Europe and Asia, Hummel succeeds in demonstrating that the local god of the Sherpas of Khumbu is-a variant of Santa Claus. He has published much material on Tibet proper in a similar vein.

108/Kailash This is a very odd statement, considering that Snellgrove clearly regarded the Sherpas as typical Nyingmapa followers, and that the accounts of Furer-Haimendorf, Jerstad, Robert Paul and Sherry Ortner, and for that matter that of Funke himself, bear constant witness to the very great importance of Buddhism to the Sherpas. It is all too easy to pick holes in Funke's argument. His basic mistake lies in thinking that the concern with local deities and malevolent spirits among the Sherpas is anomalous for a Tibetan society. In fact it is equally characteristic, on all available evidence, of Tibetans in Tibet proper, and as I showed earlier it is central to Tibetan religion in all its forms. Funke's texts, which are standard Tibetan Buddhist liturgies for dealing with local gods and spirits, indicate as much. If the Sherpas are not really Buddhists, then neither was anyone in Tibet proper, with the possible exception of a few hermits and yogis. I shall not criticise Funke's arguments any further in detail, but it is I think worth having a closer look at the effect of his theoretical outlook on his description of Sherpa religion. For Funke, explanation consists, in general, in tracing origins of individual items, and occasionally in relating them to standard categories such as the cult of vegetation powers. Thus while Funke describes a number of the same rituals and institutions as FurerHaimendorf the dumje festival, the minung or soothsayers - the emphasis in his account is on how each ritual or institution is 'non-Lamaist', is derived from Bon, etc. Certainly there are local influences on Sherpa religion. The functioning of the lu as house-spirits, and their greatly increased importance in comparison to Tibet proper, may well be related, as Funke suggests, to the non-Tibetan climate of Sherpa country, with its monsoonal rains, and also perhaps to the importance of the corresponding naga spirits in the whole area, for example among the Newars for whom they also serve as house-spirits (cf. Nepali 1965: 323-326). Yet what I find striking is that the lu simply form part of the standard Tibetan category of local deities, and are treated with exactly the same attitude and methodology as the local deities of Tibet. The basic structure of Tibetan religion persists among the Sherpas in an unchanged form. This is something that Funke is unable to see. In addition he has little to say about the monks, the lamas and the monasteries, or about the 'ideology of merit' (to use Tambiah's convenient phrase) among the Sherpas. Indeed it is hard to see why there should be all this Buddhist activity, and all the emphasis on merit-making, if Buddhist concepts were as meaningless to the Sherpas as Funke would have them. Paul: Freud and the Sherpa Buddhist Psyche Funke and Paul both worked in Shorung, and within a couple of years of each other. Yet it is often hard to believe they are talking about the same people. The Sherpas of 9 Compare Furer-Haimendorf's description of the installation of a lu shrine (1964: 267-8) with Norbu's account for the local deities of East Tibet (1966: 187-8, 190-2).

Religion in Tibetan Society | 109 Shorung as described by Funke are animists under a thin veneer of Buddhism; as seen by Paul they are Buddhists whose admitted belief in local gods and spirits receives little attention. In Paul's case as in Funke's, it seems arguable that the author has allowed his theoretical perspective to dominate his presentation of the data to an extent where the overall picture is seriously distorted. In the preface to his thesis Paul cites Melford Spiro as his chief mentor; 'it was he who first interested me in the problems of psychological anthropology, religion and Buddhism. His influence may be discerned on every page of this thesis' (1970: ii-iii). Despite this avowal, Paul's argument is rather different in character from Spiro's, and the thesis is by no means merely a rewriting of Spiro's work on Burmese religion for the Sherpas. The most obvious difference lies in the attitude to psychoanalytic theory. Spiro makes considerable occasional use of Freud, and of Kardiner, in his two books on Burma (1967, 1971). and his other writings on religion (e.g. 1966), but for Spiro the Freudian hypotheses form part of, and are subordinate to, an explicitly functionalist framework. Paul, by contrast, is a very thorough-going Freudian indeed. Somewhat paradoxically, in view of Paul's adoption of such a typically modern Western world-view as that of Freud, he also gives the im pression of taking Buddhism as a philosophy much more seriously than Spiro does. In fact much of his description of the classes of Sherpa deities, and of the nature of monastic life, reads like a restatement of Buddhist philosophy in Freudian language. Paul more or less admits this at the end of the work: In answer to the question I asked myself when beginning this project, namely "What is Sherpa monasticism all about?" I am forced to reply, "It is about precisely what it claims to be about, that is, a solution to the problem of how to live in the face of death, and of how to be saved, which is after all the most important question I can imagine". (R. A. Paul, 1970: 629) Such a viewpoint has some very real advantages. Paul's approach goes a considerable way towards making Sherpa Buddhist thought intelligible to a Western reader, at an emotional as well as at an intellectual level. (I have no space to discuss his analysis in detail. but refer the reader to Paul's article on a Sherpa temple (1976 b) which presents some of the same material as the thesis.) I would also emphasize that Paul's ethnography is very thorough and impressive; Spiro's influence is certainly noticeable here. As far as I can tell, Paul alone of the three authors I am discussing here spoke Sherpa dialect, and his knowledge of Tibetan Buddhist philosophy, though mainly from Sherpa oral sources, is also much better than that of Furer-Haimendorf or Funke. However I would suggest that his perspective leads to an account of Sherpa religion that is strong on the personal meaning to the Sherpas of their religion, but weak on religion as part of the functioning of Sherpa society as a whole. I think it goes along with this that local gods, and particularly the lu, the water spirits, and the demons and ghosts, receive very summary treatment in Paul's account. They are not really relevant to his theme: 1

110|Kailash The worship of zhibtak [local gods] is quite elaborately developed in Sherpa religion, but since it is not really concerned with the problems of "high religion" (even though it is not distinguished from other internal parts of the religion), a discussion of these gods would carry us too far from the present line of enquiry. I will likewise avoid an extensive discussion of these fascinating creatures [the lu], since they are not really central to soteriological problems. (R.A. Paul, 1970: 373, 373-4) The ghosts and demons receive slightly over a page of over 600 pages: It is beyond the scope of this paper to discuss them in any detail, and since they too are largely unconcerned with cosmological, eschatological problems, but with those of the workaday world, I will simply point out that, like all malevolent spirits, they may symbolize a number of different psychological phenomena. (1970: 375). I will continue the above quotation to give some idea of Paul's approach: First, they may be objectifications of individual fears of external threats and rivals They may also be projections of repressed hostilities of all kinds. Second...they may represent threats posed by the id, or rather the libido, to the stable organization of the ego. . . www.imm and Paul's thesis is full of such The reader's response to this kind of analysis psychoanalytic interpretations - is doubtless a function of his attitude to Freud's theories. But whatever one thinks of demons and ghosts as objectifications of internal and external threats to the ego, it remains true that demons and ghosts as supposed agents of disease and misfortune which is what they primarily are for the Sherpas - fall outside the scope of Paul's analysis. Such matters are simply part of the 'workaday world". It is here that Paul departs markedly from Spiro's approach. In his well-known paper on religion Spiro isolates cognitive, substantive and expressive desires met by religion (1966: 109-117). Paul is concerned with the expressive desires - the desires to give expression to painful drives and feelings not allowed overt manifestation by the culture. To some degree he discusses too the cognitive desire, the wish to have an explanation for the universe and man's place within it. But he is scarcely concerned at all with the substantive desires - with religious practice, that is, as a technique to get things done, as a way to achieve material ends. Yet substantive desires - for protection from illness, for prosperity, for a good rebirth, even for the attainment of Buddhahood are the needs which are most real for the Sherpas. Gods; monkhood and other religious roles; religious practice: all are for Paul manifestations of internal psychological problems. The study of tantra 'is concerned with actually attaining the state of supreme power which is inherent in the id-dominated state of mind' (1970: 410). Maybe it is; but I have suggested already that it is the uses to which that power is put which are critical for the nature of Tibetan, and Sherpa, religion. -

Conclusion Religion in Tibetan Society | 111 | _ _. I would argue then that the material given by Furer-Haimendorf, Funke and Paul all fits well into the model which I have suggested in the first of these papers - and that if it is looked at in terms of my model, it makes much more sense, and allows for a much clearer understanding of the role of religion in Sherpa or Tibetan society, than if one uses the fo Furer-Haimendorf does not give much styles of analysis advanced by Funke or Paul analysis, and he misses the essential links between the 'power' of Buddhist tantric deities and defence against the local gods, ghosts and spirits, mainly I think because of his lack of knowledge of what the rituals and the philosophy mean to the Sherpas and Tibetans. At the same time, Furer-Haimendorf does not claim to give an analysis of the Sherpa religious system, and the book has value in its own right. As an ethnographic account of the Sherpas it gave us the first adequate description of any Tibetan society, and provided the basis for much that came later. Funke and Paul on the other hand both set out specifically to study Sherpa religion, and both, I would argue, give one-sided accounts. Funke emphasizes the ghosts, spirits and local gods, and explains them as historically derivative from early (pre-Buddhist) Tibetan religion. Paul emphasizes the Buddhist pantheon and the monastic role, and explains them by translating them into Freudian terminology. Neither are able to see local gods and demons, tantric gods, Buddhist monks, and lamas, as all parts of a single system, in which each element can only be fully understood by reference to that total system. To restate my point in somewhat different language: both the beliefs in local gods and malevolent spirits, and the lamas and monasteries with their practice of the literary tradition of tantric Buddhism, are essential parts of religion in Tibetan societies. Funke's derivation of the Sherpa animistic beliefs from pre-Buddhist sources doubtless has historical truth in it, though I see no reason to agree with his contention that the Sherpas are more involved with these spirit-cults than any other Tibetan group. However his suggestion that the spirit-cults are the dominant feature of the religious system is at most a half-truth, and a misleading one, since it treats Buddhism as a thin veneer over basically animistic beliefs. This scarcely fits with the importance of the 'ideology of merit' for the Sherpas. Also, and in my view more significantly, it is precisely because the Sherpa believe in these spirits, who are dealt with for the most part through Buddhist ritual technology and by 10 A recent analysis of Sherpa ritual symbolism by Sherry Ortner (1975), while approaching the problem from quite a different direction, is I think compatible with the model suggested in the present paper. Ortner correctly sees Sherpa ritual as being concerned with channelling the power which the Sherpas recognize within Buddhism for their own everyday concerns (ibid.: 165-6).

112/Kailash Buddhist ritual practitioners whose position of supremacy is openly acknowledged by all that the lamas and their monasteries' 11 have such a central place in Sherpa and in Tibetan society. www. On the other hand Paul treats monasticism, on the whole, in isolation from the everyday concerns of Sherpa life. His Freudian arguments could explain why Sherpas become monks, and also perhaps why Sherpas regard monks and lamas as repositories of power. He is not concerned however with what that power is used for, which is, in general, to meet the Sherpas' substantive desires for long life and deliverance from illness and misfortune. My analysis, as presented in the first of these papers, would therefore describe the Sherpas neither as primarily Buddhist nor as primarily animist. All Sherpas are some of both, and the emphasis they place on each varies from layman to monk, and from individual to individual. At the same time animism and Buddhism are not two separate religions (as Spiro (1967) argues, wrongly in my view, for Burma; they are inextricably bound up with each other, and indeed confirm each other. A Sherpa could scarcely be an animist without being a Buddhist, because only Buddhism provides the power to keep the spirits under control; and most Sherpas would hardly be Buddhists if they were not also animists, for the same reason. It this has not been sufficiently obvious in previous work on Tibetan religion, the fault lies perhaps in an unwillingness to recognize the degree of religious specialization which is present in Tibetan culture. One might suggest that Funke and Paul, each in their different ways, minimize the difference between lamas and laymen. Lamas and laymen may well share the same basic frame of reference, but their roles are different, and it is only when this rather simple point has been explicitly recognized that the complementarity of these roles, and therefore of 'Buddhism' and 'animism', can be discussed. 1 believe that much the same is true for 'Buddhism' and 'animism' in the Theravada countries (as Tambiah's study (1970) of N.E. Thailand suggests) and that it is perhaps also true for the 'folk' and 'literary' elements in Indian religion. This subject however will be reserved for discussion elsewhere. 11 The monasteries (in the sense of communities of celibate monks) are of course a relatively new feature in Shorung and Khumbu, though communities of married lamas go back probably to the initial settlement of Tibetans in this area (cf. Furer-Haimendorf 1964: 130; R.A. Paul 1970: 50). These were doubtless similar to communities of married lamas still found among other Tibetan populations in Nepal (Snellgrove 1961). As was argued in my first paper, celibate monasticism is of secondary importance in Tibetan religion, though it has never been entirely absent since the monastery of Samye was founded back in the eighth century. The magical power of the lama is the essential factor from the point of view of the lay Tibetan, and this does not depend on whether the lama is a celibate monk.

Religion in Tibetan Society | 113 REFERENCES Evans-Wentz, W.Y. 1960. The Tibetan book of the dead. (3 rd edn, paperback reprint.) London: Oxford Univ. Press. Fremantle, F. and Trungpa, C. 1975. The Tibetan book of the dead. Berkeley and London: Shambhala. Funke, F.W. 1969. Religioses Leben der Sherpa. (Khumbu Himal, 9.) Innsbruck and Munchen: Univ. Verlag Wagner. Furer-Haimendorf, C. von. 1962. Moral concepts in three Himalayan societies. In Indian Anthropology. Essays in memory of D.N. Majumdar. (eds.) T.N. Madan and Gopala Sarana. London: Asia Publishing House. 1964. The Sherpas of Nepal. Buddhist Highlanders. London: John Murray. 1975. Himalayan Traders. Life in Highland Nepal. London: John Murray. Guenther, H.V. 1963. Indian Buddhist thought in Tibetan perspective: infinite transcendence versus finiteness. History of Religions 3, 83-105. 1966. Some aspects of Tibetan religious thought. History of Religions 6, 70-87. 1972. Buddhist philosophy in theory and practice. Harmondsworth: Penguin Books. Hitchcock, J.T. and Jones, R.L. (eds.). 1976. Spirit possession in the Nepal Himalayas. Warminster: Aris and Phillips. Hoffmann, H. 1961. The religions of Tibet. London: Allen and Unwin. Hummel, S. 1967. Khum-bu-yul-lha der weisse Gott der Sherpa. Acta Orientalia (Academiae Scientiarium Hungaricae). 20, 353-361. Jerstad, L.G. 1969. Mani-Rimdu. Sherpa dance-drama. Seattle and London: Univ. of Washington Press. Kvaerne, P. 1976 a. The Bonpo-Buddhist controversy. Tibetan Review vol. 11 no. 3. (March 1976), 9-15. 1976 b. Who are the Bonpos? Tibetan Review vol. 11 no. 9 (September 1976), 30-33. Macdonald, A.D. 1971. Une lecture des Pellit Tibetain 1286, 1287, 1038, 1047 et 1290. Essai sur la formation et l'emploi des mythes politiques dans la religion royale de Sron-bean-sgam-po. In Etudes tibetaines dediees a la memoire de Marcelle Lalou. (Librairie d'Amerique et d'Orient.) Paris: Adrien Maisonneuve. Nepali, G.C. 1965. The Newars. An ethno-sociological study of a Himalayan community. Bombay: United Asia Publications.

114 |Kailash Norbu, N. 1966. Appendix II. Pho lha, Phug lha, rLun rta. (English summary.) In Tibetan folk songs. (ed.) G. Tucci (Artibus Asiae Supplementum 22). Ascona: Artibus Asiae. Oppitz, M. 1974. Myths and facts: reconsidering some data concerning the clan history of the Sherpa. Kailash, Vol. II, no. 1, 121 - 131. Ortner, S.B. 1975. Gods' bodies, gods' food. A symbolic analysis of a Sherpa ritual. In The Interpretation of Symbolism (ed.) R. Willis. London: Malaby Press. Paul, R.A. 1970. Sherpas and their religion. Unpublished Ph. D. dissertation, University of Chicago. 1976 a. Some observations on Sherpa shamanism. In Hitchcock and Jones 1976. 1976 b. The Sherpa temple as a model of the psyche. American Ethnologist 3, 131-146. Paul, S.O. (= S.B. Ortner). 1970. Food for thought; a key symbol in Sherpa culture. Unpublished Ph. D. dissertation, University of Chicago. Snellgrove, D.L. 1957. Buddhist Himalaya. Oxford: Bruno Cassirer. 1961. Himalayan pilgrimage. Oxford: Bruno Cassirer. 1967. Four lamas of Dolpo. (Two volumes.) Oxford: Bruno Cassirer. Spiro, M.E. 1966. Religion; problems of definition and explanation. In Anthropological approaches to the study of religion (ed.) M. Banton (A.S.A. Monographs 3). London: Tavistock. 1967. Burmese supernaturalism. A study in the explanation and reduction of suffering. Englewood Cliffs, N.J.: Prentice-Hall. 1971. Buddhism and society. A great tradition and its Burmese vicissitudes. London: Allen and Unwin. Stein, R.A. 1972. Tibetan civilization. London: Faber. Tambiah, S.J. 1970. Buddhism and the spirit cults in North-East Thailand (Cambridge Studies in Social Anthropology, 2). Cambridge: University Press. Tucci, G. 1949. Il libro tibetano dei morti. Il libro della salvazione dall'esistenza intermedia. Milano. 1970. Die Religionen Tibets. In Die heligionen Tibets und der Mongolei, G. Tucci and W. Heissig. (Die Religionen der Menschheit, 20.) Stuttgart, Berlin, Koln, Mainz: Verlag W. Kohlhammer. Waddell, L.A. 1967. The Buddhism of Tibet, or Lamaism. (2 nd edn, reprinted.) Cambridge: W. Heffer and Sons. Wylie, T.V. 1959. A standard system of Tibetan transcription. Harvard Journal of Asiatic Studies 22,261-267.