208 Small Places, Large Issues them to communicate on a very large scale, at least in principle. They pay taxes to the government, are committed to following the written laws of the country and may make certain demands of the state. Abstract time, which measures the value of their labour power, is another anonymous structure; it is not only valid for you and me but for everybody who follows it, and it serves to synchronise a very large number of persons in an anonymous way. A brief comparison with the mechanisms of social integration prevalent among, say, the Dogon, the Yanomamö or the Trobriand islanders would indicate that capitalism and wagework entail not merely 'economic' changes, but also profound social and cultural changes. There is no simple determinism or a one-to-one relationship, but the capitalist system of production and exchange, once it has become an integral part of local society, inevitably creates new kinds of social relations as it contributes to defining premisses for social relations far beyond the domain of production. SUGGESTIONS FOR FURTHER READING Philippe Descoia and Gíslí Pálsson, eds: Nature and Society: Anthropological Perspectives, London: Routledge 1996. Kay Milton, ed.: Environmentalism. London: Routledge 1993. Eric Wolf: Peasants. Englewood Cliffs, NJ: Prentice-Hall 1966. 14 RELIGION AND RITUAL Rituals always have a desperate and manic aspect. — Claude Lévi-Strauss In a study of the Basseri pastoralists of southern Iran, Fredrik Barth (1961) expresses some surprise regarding their lack of religious interest. His surprise is caused by the fact that religion seems to loom large in the lives of most of the peoples described in classic anthropological studies. This may be a major reason why religion has always been a central field of inquiry in anthropology, even if, as Evans-Pritchard (1962) has pointed out, social scientists have themselves often been indifferent or hostile to religion. In attempting to give non-ethnocentric, comparatively useful definitions of politics, economy, nature, gender and other core concepts, it has repeatedly been shown that we run into problems usually related to the fact that these notions are in use, and have a specific meaning, in our own society and in the anthropological vocabulary, but not necessarily in other societies. This makes them problematic as 'etic' concepts. This problem is certainly valid where religion is concerned, and few concepts of social science have been defined, revised and criticised more often than this one. Only a little more than a century ago, it was commonplace in the professional literature to distinguish between religion and paganism on the one hand, and religion and superstition on the other. The concept of paganism was associated with non-Christian religions and, in particular, their practices of public rituals which expressed aspects of the content of the religion. The concept of superstition was largely reserved for descriptions of invisible interrelationships in the world which neither science, authorised religion nor 'common sense' could account for. From this kind of perspective, Islam and African ancestral cults would be located in the domain of paganism, while, say, the Trobriand islanders' belief that they die because of witchcraft and the common notion, in the Mediterranean region and elsewhere, that some persons are possessed by the evil eye, would be expressions of superstition. In contemporary anthropology, this corresponds to a frequently invoked distinction between religion and knowledge. Religion may thereby be said to include forms of social belief in supernatural powers which are public and which are given public expression through rituals. Knowledge can be defined 209 210 Small Places, Large Issues so as to include 'facts' which people are reasonably certain of and act upon, and which also have a social origin. Of course, knowledge can have a religious character, and the distinction is not absolute. In this chapter, we nevertheless concentrate on the social notions associated with contact with the sacred and the hereafter, and its expressions through ritual; leaving other forms of knowledge to the next chapter. Have I still not defined religion properly? If so, I join a large group of anthropologists in struggling with this concept. Ever since anthropologists began to study forms of belief in alien communities seriously, there has been disagreement as to what religion is. One of the oldest definitions, supported by Tylor, defines religion simply as beliefs in supernatural beings. The question of what is supernatural immediately poses itself here, for does that not vary just like every other form of knowledge - is it not the case that what is natural for us is supernatural to others and vice versa? Is the garden magic of the Trobrianders, which is as necessary to them as manure is to us, part of their religion or part of their production technology? Are ancestral spirits supernatural? If so, who says so? Another possibility in the struggle for a definition lies in following Durkheim, who assumed that a distinction is made between the profane and Is Understanding Religion Compatible with Believing? This provocative question is the title of an article by the philosopher Alasdair Macintyre (1970). His answer is, briefly, no. According to Macintyre, religion must be understood in sociological and logical terms, as stabilising and legitimising ideologies and as systems of signification and action which provide a certain shape and meaning to the world and to human existence; which explains why we are here and what happens when we die. If one is to believe in religion, he continues, one has to move to a completely different mode of thought, which easily accepts contradiction and lack of coherence, appealing to concepts such as 'the absurd' (Kierkegaard), 'paradox' (Karl Barth) or 'mystery' (Marcel). The sceptic and the believer, he claims, have no shared conceptual world - the sceptic, who 'understands' religion, cannot conceptualise the reality of the believer. Macintyre further seems to hold that the social context of modernity, which has created modern social science, is incompatible with religious faith since it is based on a wholly secular form of thought. Many anthropologists would be inclined to disagree with Macintyre (see Evans-Pritchard 1962). It seems likely, for example, that Catholics such as Victor Turner, Mary Douglas and E.E. Evans-Prichard have contributed somewhat to our understanding of religion. Religion and Ritual 211 the sacred in everysociety, and who confined religion to the sacred domain. Durkheim also wanted to show how the function of religion in 'primitive societies' consisted of creating solidarity and integration through rituals and 'collective representations'. In a famous statement, he claimed that religion at its most profound level means society's worship of itself. This view has its problems (notably the problems of functionalist explanation), for example in not explaining why the inhabitants of one society believe in ancestral spirits whereas their neighbours believe in forest spirits, granted that both would be socially integrating beliefs. A rather different approach to religion is represented in an influential essay by Geertz, 'Religion as a Cultural System'. He defines religion like this: (1) a system of symbols which acts to (2) establish powerful, pervasive, and long-lasting moods and motivations in men by (3) formulating conceptions of a general order of existence and (4) clothing these conceptions with such an aura of factuality that (5) the moods and motivations seem uniquely realistic. (Geertz 19 73, p. 90) In other words, he argues that instead of looking at the social functions of religion, we ought to explore what religion means to people, how it helps to make sense of the world and how it gives a meaning and direction to human existence. We should study religion itself, not its social causes, and the ultimate aim of such an investigation ought to be to understand how the world and human existence appear meaningful to the believer. This hermeneutical, or interpretive, procedure is evident in Evans-Pritchard's earlier work on Nuer religion (1956), which is a detailed ethnographic exposition aimed at translating Nuer beliefs into European (and Christian theological) concepts, thereby rendering them comparable to other religious phenomena, as well as relating religion to social organisation. The approach advocated by Geertz and foreshadowed by Evans-Pritchard has been very influential in anthropological research on religion in recent decades. This shift in perspective is a part of the general change in anthropological thinking mentioned earlier, whereby the main trends have swung from an interest in functions, structure and social integration, which was dominant until the 1960s, to a concern with the interpretation of meaning, symbols and social process. Several examples of this change have been noted in earlier chapters; in the study of religion, it has brought, among other things, a growing interest in relating meaning to experience and in understanding cosmologies. This chapter shares such a concern, but in the sections dealing with rituals - religion as practice - we return to some of the problems first articulated by Durkheim, as well as presenting approaches to understanding the ideological aspects of religion and ritual. ORAL AND WRITTEN RELIGIONS The distinction between oral and written religions is important and has a bearing on other aspects of culture and society too. Written religions, or 212 Small Places, Large Issues 'religions of the book' (Goody 1986), are linked to a sacred text (like the Koran) or a collection of sacred texts (like the Bible), and the believers are expected to have at least a minimal knowledge of the contents of the works. Such religions, including Judaism, Islam and Christianity, regard their content as tied to the text and not to a particular cultural context. Since they are text-bound, these religions can be disseminated throughout the world to peoples who in other regards live very different kinds of lives. Islam, for example, is the main religion in countries as different as Java, Niger, Egypt and Iran; whereas Christianity dominates in countries like the Philippines, El Salvador and Germany. The written religions, and particularly the monotheistic ones with their origins in the Middle East, can also be described as religions of conversion - systems of belief to which one can be converted and in which one has to affirm one's faith. Unlike other religions, they tend to be exclusive and not to accept 'syncretism'. Christian missionaries in Africa have therefore despaired at the sight of Africans cheerfully worshipping their God as well as water spirits and ancestral spirits. Some written religions fulfil this pattern only partly; notably the Asian ones (Hinduism and Buddhism are the largest) which have a less fixed doctrine, more flexible practices, and insist less on obedience to texts than the monotheistic script religions do. However, the kinds of religion characteristically studied by anthropologists are markedly different from religions based on scriptures. First of all, they are locally confined. No Nuer or Kaguru in his right mind would expect the whole world's population to become 'disciples' of their revered spirits or even of their highest god, thoth (Nuer) or mulungu (Kaguru). The gods are frequently physically associated with revered places in the tribal area. For this reason, missionaries and others misguidedly held tribal peoples to be 'animists': holding the belief that trees, springs and rocks are imbued with divine powers {'anima' is Latin for spirit). Second, oral religions tend to be embedded in the social practices of society, whereas written ones are often more detached from other social institutions. This distinction, which is not absolute, corresponds to the previously discussed institutional differentiation in modern societies, which is largely lacking in small-scale traditional societies. However, we should note that one of the first specialised (non-food producing) occupations that develops as societies become more differentiated is the priestly one. Shamans, that is people who through the medium of the trance enter into communion with the spiritual realm, exist as a specialised profession even in acephalous and otherwise undifferentiated communities, such as traditional Inuit society. A somewhat related distinction, which was introduced by Robert Redfield (1955), concerns little and great traditions. Redfield argues that different strains and logics of religion and knowledge exist side by side in many societies; they may be radically different, but are often interrelated. The Mediterranean belief in the evil eye, for example, clearly belongs to a 'little' tradition (neither Christianity nor Islam - the 'great' traditions of the region Religion and Ritual 213 - supports the notion), as does the worship of saints in Muslim societies. In some coastal Portuguese communities, villagers believe in a wide range of healing rituals, sorcery, magic supposed to secure fishing luck and 'supernatural' explanations of misfortune (Brogger 1990). These beliefs exist alongside the official doctrines of the Catholic Church, although they contradict the teachings of Christianity and seem to have done so for centuries. The same individuals believe simultaneously in the 'great' and 'little' traditions. In other societies, such as Indian ones, there may be a more clear-cut social distinction between the religious traditions. Brahmanic Hinduism is the official 'great' tradition, its beliefs and rituals sanctioned in the ancient Veda texts and centuries of monitored ritual practices. Low-caste Hindus nevertheless have their own rituals and beliefs, often more reminiscent of oral than written religions, which coexist with the 'high' or 'great' tradition but are socially segregated from it. Oral religions are characterised by their local relevance, relative lack of dogma and tight integration with the 'non-religious' domains. An ethnographic example may illustrate these points. AN ORAL RELIGION IN AFRICA According to the Kaguru, God (mulungu) created the world, but they are uncertain as to when it happened (Beidelman 1971, p. 32). This God appears quite rarely in the lives of the Kaguru, however; usually they consult ancestral spirits instead of bothering the great mulungu when they are in trouble. The ancestors (and perhaps particularly ancestresses - the Kaguru are matrilineal) arrived from the north and the east in a mythical past, founded the present-day clans and divided the land between them. Each clan is assumed to have a 'mystical' connection (Beidelman's term) with its land. For the harvest to be good, the clan members must carry out annual rites so that the ancestral spirits will bless the land and secure its fertility. Notions and practices of this kind clearly go a long way to explain why land cannot be sold or bought in many traditional African societies. When the Kaguru wish to consult an ancestor or a different spirit, they leave the village and enter a hillside or go into the bush. Usually, the consultations concern practical issues such as rites of passage and festivals intended to ensure the fertility of soil and women. The ancestral spirits are believed to wield real power over the living. A Kaguru woman who had lost several children thus blamed her deceased father, claiming that he was feeling lonely and had called her children to come and keep him company. Since the spirits are this powerful, it is important to pay them respect continuously and sacrifice to them regularly. They enter into every realm of life; birth, rites of passage, fertility and politics. 214 Small Places, Large Issues THE AFTERLIFE The Kaguru, like virtually every people in the world, are concerned with the afterlife. All existent religions deal with death and try to reconcile life and death. Most peoples have notions about an afterlife which in a sense represents an idealised version of life here and now, devoid of the trivialities, problems and frustrations of this life. The Kaguru envision the afterlife as a mirror-image of life in Kaguruland, but marked by abundance and lack of conflict. The Norse Vikings, for their part, assumed that people (or at least men), after an honourable death, came to Valhalla, where the fire burned day and night, there was plenty of fighting and an abundance of roasted meat. During fieldwork in Trinidad, moreover, I once found a pamphlet from a North American missionary organisation in my mailbox. It painted, in vivid colours, the Christian paradise as a kind of amusement park, where people could fly, where there were video shows presenting highlights from biblical times, and so on. The more intellectually oriented, and orthodox, versions of Christianity, like other written religions, rather depict the afterlife in more abstract terms. Oral religions tend to be more concrete on this point too. According to the traditional religion of the mountain Sami of northern Scandinavia, people are allowed to keep everything, including their reindeer herds, in the afterlife, the main difference being that pastures are abundant there. This kind of notion explains why the Sami (and many other peoples) were buried with their favourite clothes on, with their tools and, in the case of some hierarchical societies, their favourite slaves. Notions of the afterlife, be they abstract or concrete, obviously give an impression of continuity and serve to demystify death. It should be added, though, that there are also peoples who do not believe in an afterlife; this unusual view seems to be particularly widespread in two of the most individualistic types of society we know of, namely hunters and gatherers (Woodburn 1982) and modern industrial and post-industrial societies. THE LOGIC OF ANCESTRAL CULTS The great attention given to ancestors and ancestral spirits, which is found in most non-literate societies, also clearly deals with the problem of continuity - both in society and in the individual lifespan - when a life is suddenly stopped. In an article on ancestral cults in Africa, Kopytoff (19 71) has observed that there is not necessarily a sharp distinction between living humans and ancestral spirits. Living people become wiser, 'drier' and less mobile the older they become; the ancestors are thus perceived, he argues, merely as extremely wise, dry and immobile persons. There is no rigid boundary between life and death in this scheme, rather a gradual transition to another phase, which begins long before death. Religion and Ritual 2, The respect paid to ancestors also has a politically legitimating and socially stabilising effect. When age is a criterion of wisdom and a qualification for political office, which is nearly inevitable in kinship-based societies, politics becomes conservative. The ancestors showed the way, the elders are their intermediaries and their younger descendants have to listen and obey. As Robin Horton (1970) and others have pointed out, a great many rituals in African tribal societies are dramatic re-creations of the past intended to please the ancestral spirits by showing that the living are faithful to the values and practices taught by their elders. The political aspects of ancestral cults are significant in practice, even if they cannot explain why people hold beliefs in ancestral spirits. In a general sense, one may perhaps state that any religion, like other kinds of ideology, must simultaneously legitimate a political order and provide a meaningful world view for its adherents, such as a reconciliation with one's own inevitable death. The death of a ruler, which signifies discontinuity, is always associated with crisis, and the belief in ancestral spirits may mitigate the effects of the rupture. It should be remarked at this point that we have added a political dimension to Geertz's cultural definition of religion, which may lend some support to Marx's famous statement to the effect that religion is the opiate of the people. By this he meant that it functioned as a drug and diverted interest from the real political issues to silly fantasies about a happy afterlife for the pious and obedient. (Of course, a religious non-Marxist might retort that Marxism is the opiate of Marxists.) We now turn to a closer look at the interrelationship between the cultural (ideational) aspects of religion and its social and political dimensions. RITUAL: RELIGION IN PRACTICE Most people in the world are faced with various practical problems of an economic and social character. Some of them can only be resolved with the help of specialists, and in many societies such specialists are people with a priviliged access to higher powers. Other kinds of problems are existential ones; they may deal with the mystery of birth or the fear of death, or simply the ultimate meaning of life. Rituals are largely directed towards problems of the latter kind, dramatising them and giving them articulation - if not necessarily resolving them. Ritual has been defined as the social aspect of religion. If we may define religion as systems of notions about the supernatural and the sacred, about life after death and so on (with its obvious political implications), then rituals are the social processes which give a concrete expression to these notions. Very generally speaking, we may suggest that rituals are rule-bound public events which in some way or other thematise the relationship between the earthly and the spiritual realms. 216 Small Places, Large Issues The issue is somewhat more complex than this suggests, although it may be a fruitful beginning towards a useful definition. In fact, several of the greatest anthropologists of the twentieth century have devoted many years to trying to understand ritual. Since the early structural-functionalist accounts of rituals as manifestations of society's worship of itself, where the integrating functions of ritual were stressed, anthropology has eventually developed quite complex theories about what ritual is and how it works. One principal perspective emphasises that rituals simultaneously legitimate power, and are thus important vehicles of ideology, and give the participants strong emotional experiences; another perspective focuses on the ability of rituals to give people an opportunity to reflect on their society and their own role in it. Victor Turner stresses the multivocality or ambiguity of ritual symbols (1969). One of the most famous analyses of rituals does not even deal with a religious ritual but a cockfight (Geertz 1973), while in the mid-1990s a team of Norwegian anthropologists carried out research on the 1994 Winter Olympics, which they see as an enormous ritual celebrating and legitimating modernity. This is, in other words, a very complex field, and it is important because the ritual can be seen as a synthesis of several important levels of social reality: the symbolic and the social, the individual and the collective; and it usually brings out, and tries to resolve - at a symbolic level - contradictions in society. RITUALS AND INTEGRATION Max Gluckman (1982 [1956]) has described a number of curious rituals from Southern Africa, whereby customary rules, conventions and hierarchies are turned upside down. One of them took place among the Swazi at the coronation of a new king. When this was about to happen, every citizen was expected to mock and criticise the king in public, making a grand spectacle of his inadequacy. Similarly, open social criticism was allowed at the medieval carnivals in parts of Europe, but not during the rest of the year. Actual social conflicts are allowed to play themselves out as theatrical performances. Gluckman describes several other 'rituals of rebellion' and concludes that 'by allowing people to behave in normally prohibited ways, [they] gave expression, in a reversed form, to the normal Tightness of a particular kind of social order' (1982, p. 116; but see de Heusch 2000 for another interpretation). Gluckman thus sees these rituals as functional in that they transform conflicts in a harmless direction, but he is also aware that strong experiences on the part of the participants are necessary for the rituals to be possible at all. In other words, he sees an interaction between individual motivations and societal 'functions'. In a famous study of ritual among the Tsembaga Religion and Ritual 217 Maring in highland New Guinea, Roy Rappaport (1968) looked almost exclusively at the functional aspect. The Tsembaga Maring are horticulturalists and pig-raisers organised in local groups of 200 to 300 persons. Their political system is egalitarian, with the 'big man' and shaman the only formally recognised authorities. They are considered warlike and are frequently feud with neighbouring groups. Rappaport argues that there is an intrinsic functional link between war activities and the ritual cycle of the Tsembaga. Every twelfth to fifteenth year, the Tsembaga organise the kaiko festival, which lasts for a full year and culminates in a declaration of war on the enemies of the local group. At the kaiko parties prior to this, large numbers of pigs are sacrificed to the ancestors and lavish ceremonial gift exchange . takes place within the group. This year-long religious ritual begins when there are 'enough pigs', say the Tsembaga. Rappaport, however, holds that the kaiko festival begins when there are so many pigs in the village that they destroy more values (crops) than they produce (meat). The kaiko can thus be seen as a regulating response to the competition from and parasitism of the pigs. Moreover, when there are many pigs the population is less concentrated - the women, who are swineherds, have to move further and further away from the village during the day - and are more vulnerable to military attacks. The violent activity following the large-scale pig slaughter also serves to disperse the population, since the losers of the war have to move and raise new settlements, thus decreasing the pressure on the ecosystem. The ecosystem, including the people in it, is analysed as a self-regulating system. Rappaport shows that the kaiko begins, and violence breaks out, when the number of pigs and humans in a given area has reached a critical level; then the ecosystem is near the limits of its sustainability. After the completion of the ritual cycle, the critical values decrease and the system is stabilised. Rappaport's monograph was debated for years after its publication. The critics pointed out that a system cannot be 'rational' in this way - that an ecosystem cannot conceivably know the limits of its sustainability and trigger rituals and war when the critical values appear. It was also stressed that humans are themselves the causes of their actions, that they take decisions within a cultural universe and that it is highly unlikely ecological fluctuations can create' rituals. What Rappaport succeeds in showing is, in the end, a statistical correlation between ecological pressure and ritual activity. The problems inherent in Rappaport's analysis, which he addressed himself in highly sophisticated ways (Rappaport 1984), are the classic problems of functionalist explanation. Although the rituals are 'functional' in the sense that they contribute to the long-term survival of Tsembaga society, this does not explain why they exist. Other institutions would also have taken care of that; besides, other societies change too. Functionalist accounts of rituals were challenged in Edmund Leach's monograph on the Kachin (1954). In contrast to the conventional anthropological wisdom of the day, Leach 218 Small Places, Large Issues discovered that the religious-ritual complex among the Kachin in no way functioned smoothly; instead, it spurred dissension and imbalance. IDEOLOGICAL AND SOCIAL AMBIGUITIES Unlike their neighbours the Shan, who are Buddhists, the Kachin worship their own gods and spirits (nats). The world of nats is conceived of as an extension of the earthly hierarchy, since the nats are ancestral spirits. They belong to lineages, and in a manner analogous to the distinctions of society, the Kachin distinguish between aristocratic nats and commoner nats. Kachin society, Leach argues, is not stable socially or ideologically, and there are two rival views of how society should be run, called 'gumlao' and 'gumsa'. Gumlao refers to an egalitarian ideology with no ranking of lineages and conscious attempts to avoid the tendency towards hierarchy inherent in the mayu-dama relationships (see Chapter 8), while gumsa refers to a more hierarchical form. To some extent, Leach shows that Kachin societies oscillate between gumlao and gumsa. Now each local community has a patron saint, a mung nat, who is worshipped and sacrificed to during a ritual lasting for several days at a public place in the village. During hierarchical gumsa periods, the mung nat is regarded as a spiritual member of the chiefs lineage; within the egalitarian gumlao system, he is considered the ancestor of all the lineages. The sky nat Madai, the ruler of the spiritual realm, is only recognised within the gumsa system. Edmund R. Leach (1910-89) was initially trained as an engineer, but was converted to social anthropology at the age of 2 5, when he joined Malinowski's select group of students. His stationing in Burma during the Second World War gave him the ethnographic raw material for a book which has become one of the most important monographs of the subject, Political Systems of Highland Burma (1954). The book was an explicit attack on the then dominant view that societies were generally well integrated and stable, and that myths and ideology first and foremost strengthened such a stability. Leach described a society which continuously carried the germ of its own transformation, whose political system changed in a cyclical manner. His next monograph, Pul Eliya (1960) from Sri Lanka, was much more ethnographic in character, but around the same time, he launched his attack on what he saw as the exaggerated interest in ethnographic 'butterfly collecting' at the expense of theoretical development (Rethinking Anthropology, 1961). Leach was a sharp polemicist, and had a life-long dialogue with Lévi-Strauss and structuralism, which he regarded with a mixture of admiration and scepticism. Religion and Ritual 219 Leach's argument is as follows. The spiritual world is construed as a mirror-image of society. Rituals - which largely consist of private and public sacrifices - are chiefly indirect and oblique ways of talking about society. Therefore it is understandable that one does not sacrifice to the 'king' of the spirits during the egalitarian gumlao phases. Further, there is an intrinsic relationship between the myths (religion, or the cognitive aspect of religion) and the rituals, since the rituals dramatise the myths. However, Kachin myths are ambiguous and can be told in different ways. Some of these confirm the gumsa system; other, only slightly modified versions, confirm gumlao ideals. The different slants can be presented at the same time by different persons wishing to make different points. Leach shows that this ambiguity in myth and ritual practice in no way creates social stability. The inconsistencies in the Kachin ritual system are, in his view, fundamental and are therefore an eternal source of tension in society. Quite unlike what Durkheim, Malinowski and others had argued, Leach shows that the myths and rituals positively encourage a lack of stability, since they offer themselves to conflicting interpretations. Kachin beliefs and rituals nevertheless always function ideologically in the sense of legitimating a particular power structure, but the ideology, reflecting instability in Kachin social organisation, is ambiguous. Much later research on ritual has concentrated on one of the dimensions Leach deals with, namely what he calls 'symbolic statements about the social order'. One interesting study in this vein is Bruce Kapferer's analysis of demon exorcism among the Sinhalese in Sri Lanka (19 8 4). I shall not go into the details of the ritual; suffice it to say that these exorcisms are large, well-attended and heavily dramatised events which usually take place in the front yard of the home of the patient (the possessed person). In his analysis, Kapferer stresses that the rituals enable the participants to see the world more clearly than usual, and to reflect on their own position in it. For this to be possible, however, they must be able to move to and fro between the ritual, spiritual context and the everyday context: otherwise the two realms would remain separate. Paradoxically therefore, Kapferer writes, the part of the audience which is furthest away from the central stage is best able to carry out this kind of reflection. The patient, the relatives and the first rows of spectators are too immersed in the event to reflect on it, while the people at the back, sipping their tea and chatting together in low voices, are able to see the ritual at a distance and thereby use it consciously in their self-reflection. Here Kapferer finds a quite different pattern from Geertz (1973) in his famous analysis of the Balinese cockfight, where he argues that the only participants who fully understand all the symbolic nuances of the fight (which is laden with cultural symbolism) are those who are central betters - who are placed at the inner circle near the light itself. Only they engage in what Geertz calls 'deep play', which may be an euphemism for religious communion. In Kapferer's example, the opposite proved to be the case: the 220 Small Places, La r ge Issues people of the inner circle understood little, because they were too deeply immersed in the drama. It is very likely that this difference is caused by differences between the respective empirical findings. Such dissimilarities, of course, seem to complicate even further the matter of building a general theory of ritual. Nevertheless, we should be aware that from the system ecology of Rappaport, to the complex multi-level analysis of Leach, to the hermeneutic approach represented by Kapferer and Geertz, there is a clear continuity in that they all agree that ritual is an oblique, indirect way of making complex statements with a strong ideological dimension about society. Through a number of studies of the Ndembu of present-day Zambia, Victor Turner (1967, 1969, 19 74) has developed a sophisticated model of ritual symbolism, which sums up many of the concerns taken up by others. THE MULTIVOCALITY OF SYMBOLS The use of symbols is central to rituals, and studies of ritual symbols must not merely investigate which symbols are being used, but must also look into their mutual relationship and their meaning (what they symbolise). In Christianity, white symbolises virtue and purity while black signifies evil and darkness; the number seven has sacred connotations, and the wafer consumed at communion has the paradoxical quality of being simultaneously an ordinary wafer and a part of Christ's body. The wafer can thus be seen as a liminal object forming a bridge between this world and the spiritual realm. In this way, it can truly be said that rituals both say something and do something. Moreover, many of the symbols of Christianity are ambiguous. In Turner's terminology, they are multivocal, which literally means that 'several voices can be heard'. Several meanings can be read into the number seven, and it is not universally agreed what the holy communion really means. Turner sees the milk tree (Diplorrhyncus mossambicensis) as a central symbol at initiation rituals (Turner 1967). The tree is notable in that it secretes a thick, white, milkish fluid when its bark is cut. The Ndembu explain that it is important in the initiation of girls because the milk tree stands for human breast milk and for the breasts themselves. They also say that the tree 'belongs to mother and child'; that it symbolises the mother-child tie. In other words, it seems to have two main meanings: a biological and a social one. The Ndembu also emphasise that the milk tree expresses the continuity of the matrilineage and the cohesion of the tribe. 'The milk tree is our flag', said an educated Ndembu, invoking an apt analogy to an important multivocal symbol in nation-states. Turner also notes that the milk tree can symbolise contradiction and fission. Especially at the girls' initiation rites, the tree forms the focus for the female spirit of community and their opposition to male dominance; the Religion and Ritual 221 women dance around it, sing libellous songs against the men and so on. Further, says Turner, the tree represents the individual novice, as a young milk tree is being blessed at the same time that she enters the liminal phase. Thus, the tree represents the tension between individual and society. It can also represent a conflict between the mother of the novice and the other women: she loses her child who is becoming an adult, and is not allowed to join in the dance around the tree. Finally, the milk tree may represent the matrilineage of the novice and so serve as a reminder of the contradiction between the unity of the tribe and the separation of the lineages. The milk tree is a dominant symbol, and Turner argues that all rituals are focused on similar symbols. Dominant symbols have the following characteristics. First, they are condensed, that is to say many different phenomena are given a common expression. Second, a dominant symbol amounts to a fusion of divergent meanings. In this way, otherwise different people can sense likeness and express solidarity through these symbols - such as flags in nation-states, which mean different things to different people and so are able to give different people the impression that they are the same. Third, dominant ritual symbols entail a polarisation of meaning. At one pole (the ideological), there is a set of meanings to do with the social and political order of society. At the other pole (the sensory), physiological and biological meanings are expressed. (To this, we would probably add emotional meanings today.) The milk tree thus represents, at one pole, matrilineality and the unity of Ndembu society (among other meanings); at the other pole it expresses breast milk and the mother-child relationship. A major insight in Turner's work is that symbols have to be multivocal, or ambiguous, to create solidarity: since persons are different, the symbols must be capable of meaning different things to different people. This could be said of rituals in general too, and Leach's Kachin study is a clear case in point. Another important insight from Turner, who belonged to a generation of British anthropologists concerned with bridging the gap between interpretation of meaning and accounting for social structure, is the idea that ritual symbols must speak both about politics (social structure, legitimation) and about existential or emotional cravings: they must be capable of fusing personal experiences with political legitimation if they are to be effective. THE INHERENT COMPLEXITY OF RITUAL In an important study of the changing significance of circumcision rituals in Madagascar, Maurice Bloch (1986) has developed further the points made here concerning ritual, social integration, ideology and power. The Merina (formerly known as Hovas), who live in the Malagasy highlands, are the most powerful ethnic group in the island, having ruled most of its area for centuries, subjecting neighbouring peoples to Merina rule. Strongly hierarchically organised, the Merina succeeded in retaining some 222 Small Places, Large Issues of their traditional power during French colonialism, even after surrendering to France in 189 5. The Merina have a bilateral kinship organisation, and the fundamental unit of local organisation is the deme (Bloch 1971), consisting of a largely endogamous local group associated with a particular territory. The Merina are famous for their imposing tombs and their elaborate ancestral worship. In his principal study of Merina ritual, Bloch concentrates on male circumcision, which is the single most important public ritual. Merina boys are circumcised very young (between the ages of one and two years old), and this ritual involves practically the whole local group, as well as deme members living elsewhere. They give contributions to the child's family which are proportional to their closeness as kin. During the ritual they sing and dance throughout the night, thereby dramatising the unity of the kin group. The circumcision usually takes place in the child's parents' house, but other deme members are appointed as the child's 'father' and 'mothers'. The child's 'father' is the circumciser, while the 'mothers' are young women who have a special responsibility for the child during the ritual, which can last for days. Other people also have special assignments; adolescent boys are expected to act mischievously and make practical jokes, while the men cook the food, which is usually a female task. In a largely structuralist analysis of the symbolism of the ritual, Bloch describes how symbolic meanings are contrasted and inverted, and how both the social and the cosmic order are dramatised in suggestive, non-verbal ways. The unity represented in the ancestors' tombs and the division represented in the houses (inhabited by people from different lineages) are juxtaposed, while the male-female opposition is also expressed at a variety of levels. For example, the ritual always takes place during the cold season; in Bloch's view, this establishes a continuity between the life-giving ceremony of circumcision and the stone tombs of death and deme unity. The fact that the circumciser and 'mothers' are not close kin to the child, which negates biological kinship, also lends support to this view. During the very complex proceedings, objects and acts represent different forces and social relationships which interact in ambiguous ways; these include strong vital elements (which are 'wild' and include the vazimba, the mythical enemies of the Merina), intermediaries (which mediate and domesticate those vital forces, making them useful to the Merina), the tomb (unity and undifferentiated descent) and devalued entities (such as 'women on their own' and division; Bloch 1986, p. 99). The central contradiction in the ritual is, in Bloch's multi-layered analysis, the symbolic association between blessing and its opposite, namely violence. Violence is enacted both symbolically and literally; the latter does not just occur in the act of removing the child's foreskin, but also in the killing of a bull to be consumed during the ritual. Each act of violence, however, is associated with a tsodrano or blessing ritual, where important men and Religion and Ritual 223 women call on,God and ancestors to give them their blessings. These elders then 'blow water' on to the child and the spectators, thereby mediating the blessings given by God and the ancestors. This contrast - between blessing and violence, life and death - is in Bloch's view a central contradiction in Merina ideology. Drawing on sources enabling him to trace descriptions of the ritual back to the eighteenth century, Bloch shows how its central symbolic features have remained remarkably uniform despite important social and political upheavals - the growth of the centralised Merina state, colonialism and independence - and despite great variations in its size and social importance. During the period of the centralised state, the royal circumcision was the main state ritual, and was used to legitimate royalty, tax collection and centralised hierarchy at the expense of weakening the demes. For this purpose, the symbolic content of the ritual was altered slightly, although its key elements remained unchanged. During French colonialism the ritual was a small-scale family undertaking, whereas since independence in 19 60 it has increased in importance and taken on anti-elite connotations. Bloch's detailed analysis of both the symbolic and social elements of the ritual process and its changing historical significance shows that ritual is not determined by an easily intelligible set of factors. Bloch expresses dissatisfaction with conventional anthropological approaches to ritual. On the one hand, various functionalist explanations (Marxist as well as non-Marxist variants) are inadequate for reasons discussed earlier in this chapter. On the other, what he calls intellectualist and symbolist views, including those of Geertz and Evans-Pritchard (in his 1956 book), which 'see religion as a speculation on nature and an intellectual accommodation of the beyond' (Bloch 1986, p. 8), fail to place the beliefs and rituals in a proper social context. In concluding his own historical analysis of the Merina circumcision ritual, Bloch concludes: 'Rituals are events that combine the properties of statements and actions. It is because of this combination that their analysis has proved endlessly elusive' (1986, p. 181). A fact which cannot be elaborated here, but which should be kept in mind, is that it is not always easy to distinguish clearly between theatrical and ritual performances. Schechner (1994, p. 622) proposes a continuum where the theatre represents entertainment and the ritual efficacy; the theatre stands for fun and appreciation, the ritual for results and beliefs. However, not least in contemporary Western dramatism, the object of a theatre performance may well be to make the audience reflect on the conditions of existence (consider, for example, the relationship between the Sinhalese exorcism ritual and Beckett's plays) or to act politically (as with Brecht). The close connection between dance and ritual, furthermore, is demonstrated in recent etnographies of dancers, such as Helena Wulffs study of transnational ballet dancers (Wulff 1998). 224 Small Places, Large Issues An Actor-centred Perspective on Ritual In Chapters 11 and 12, actor-based definitions of politics and economy are presented. Rather than stating that politics, for example, is the social distribution of power, authority and rights, or that economy is society's routines for production, distribution and consumption, one may thus define politics and economy as aspects of action. Leach has proposed a definition of the same kind regarding ritual (Leach 1968). Common anthropological definitinons of ritual would locate them to ritual institutions - churches, mosques, sacrificial grounds, etc. - and focus on the systemic level. This would ultimately be misleading, Leach argues, and instead he calls attention to the ritual act, seeing rituals as an aspect of culturally standardised actions. The expressive, symbolic aspect of a conventional act - everything which is not obviously goal-directed - is ritual, says Leach, and points out that ritual acts do not necessarily take place in what we think of as 'ritual contexts'. This kind of definition, we should note, does not exclude the more 'substantivist' definition focusing on the ritual institutions, but complements it by focusing on the acts themselves and not merely the social framework. POLITICAL RITUAL IN STATE SOCIETIES Careful to avoid a simplistic reductionist explanation, Bloch analysed the ideological dimension of ritual perhaps more carefully than any earlier anthropologist. One of his main points, which he shares with many other anthropologists, is that rituals and ritual symbolism have to be ambiguous because they are representations of a social world that is contradiction-ridden. So 'the message of ideology cannot be maintained simply as a statement... because it is by its very nature in contradiction with human experience in the world' (Bloch 1986, p. 195). This is important. Ideology always simplifies and imposes hierarchy and a particular social order. In the case of the Merina, ideology as mediated by the circumcision ritual also served to justify state violence. In modern state societies, the oblique ideological dimension of ritual is no less evident than in non-modern societies (see Handelman 1990). National flags, for example, are sufficiently ambiguous (or multivocal) to be able to create a symbolic bond and a sense of community between persons who are very different and who represent contradicting interests. Insofar as they are able to interpret the flag in different ways, and thus identify with it on different grounds by relating it to different kinds of personal experiences, citizens can actualise themselves as a nation through such simple national symbols. In this way, state rituals may indeed function as charters for collective action. Religion and Ritual 225 In situation&jof social transformation, rituals belonging to the former social order may be reproduced, although their meaning may change, in order to give an impression of legitimacy. As Kertzer notes (1988, p. 46), New Year celebrations have been a constant feature of Russian society since pagan times. When the Russians became Christian, the Church merged these festivities with Christmas, and after the 1917 Bolshevik revolution the new Soviet leaders actively sponsored the festival after a brief interlude of attempting to abolish it, but tried to remove as much of its Christian content as possible (see also Mach 1993, pp. 130 ff.). During periods perceived as turbulent, where, say, a new political power structure tries to replace the old one, the new leaders may try to appropriate ritual symbols associated with the old, familiar order in order to create an impression of continuity and legitimacy. Such a use of familiar symbols in order to render an unfamiliar situation familiar, whether or not this is intentional, is characteristic of the legitimation of contemporary ethnic movements (Abner Cohen 1974) and nationalism; this is dealt with in Chapter 18. RITUALS OF MODERNITY: SPORTS Although ritual is frequently seen as 'enacted religion', it must be kept in mind that the most famous analysis of a single ritual, namely Geertz's interpretation of the Balinese cockfight, concentrates on an entirely secular ritual; and as noted above, dance may well be understood in a framework partly shared with the study of ritual. Other non-religious rituals certainly also merit attention. The affinities between rock concerts and religious rituals are obvious, but the most important rituals in the contemporary world are arguably those to do with sports. Estimates suggest that between a quarter and a third of the world's population followed the finals of the 1998 football World Cup on television. Until the 1990s, anthropological studies of sports were marginal, but today, several important studies -particularly of football - exist (including MacClancy 1994; Armstrong and Giulianotti 1997; Archetti 1999). Already in 1982, however, Marc Augé argued that the focus of football studies should be shifted from social history to religious anthropology. Common to most studies of football is a concern with the forms of social identity. Roberto DaMatta (1991) uses Turner's notion of the social drama to understand the dynamics of football fan-dom; Eduardo Archetti (1999) has - among other things - analysed the game as a celebration of masculinity and the star pla)'ers as religious icons; others have called attention to its class dimension (evident in many European countries), its pivotal role (at least at the operational level) in expressing national identity, and even its potential for bridging generation gaps (boys support the same teams as their fathers). As a multivocal symbolic realm, football can also be a vehicle for the expression of political views. Its unparalleled global character (international 226 Small Places, Large Issues tournaments, certainly; but also the fact that Manchester United is virtually the home team of Singapore) needs further exploration. The unpredictability of the outcome of any given game , further, contributes to blurring the boundary between football, religion and witchcraft. A research question which cannot be answered straightforwardly and conclusively, concerns what exactly is the 'object of worship' in spectator sports such as professional football. It is all of the above and more. It is perhaps nowhere more evident than in the study of religion, rituals and practical/cognitive systems of knowledge that anthropological research generates insights which would not have been available without fleldwork. For instance, contrary to much theoretical philosophy, anthropological research has shown how it is fully possible, in practice, to hold notions which are contradictory in theory. Different kinds of knowledge are used in different kinds of situations, and as long as they are not confronted in the same situation they may easily coexist in the mind of one person. In a study of medical systems in polyethnic Mauritius, Linda Sussman (1983) shows that Mauritians may well consult three or four different kinds of doctors - who in a sense work within totally different realities and have irreconcilable views on illness and healing - to be on the safe side. If they have a backache, they may see a Chinese herbal doctor, an Indian ayurvedic doctor, a European physiotherapist and an African traditional healer. The general point here is that meaning is use: that religious as well as other knowledge becomes important to people only when it can be used for something, only when it is connected to their experience. Rituals, in this regard, dramatise the rather abstract tenets of religion, render the content of religion concrete and recognisable, link it to experience and legitimate the social and political order. Moreover, different kinds of knowledge are made relevant in different situations. Therefore it does not necessarily lead to a practical contradiction to believe in both the Bible and the scientific theory of evolution, as long as the two bodies of thought are kept in separate realms. Similarly, a Kachin may be favourable to both gumlao and gumsa values, but not simultaneously; and a West Indian may be (indeed, most are) favourable to values of both respectability and reputation. SUGGESTIONS FOR FURTHER READING E.E. Evans-Pritchard: Theories of Primitive Religion. Oxford: Clarendon 1965. Clifford Geertz: Religion as a Cultural System, in Clifford Geertz, The Interpretation of Cultures. New York: Basic Books 1973. Jeremy MacClancy, ed.: Sport, Identity andEthnicity. Oxford: Berg 1994. Victor W. Turner: The Forest of Symbols: Aspects ofNdembu Ritual. Ithaca, NY: Cornell University Press 1967. 15 MODES OF THOUGHT Animals are divided into (a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (1) et cetera, (m) having just broken the water pitcher, (n) that from a long way off look like flies. — Jorge Luis Borges (quoting from 'a certain Chinese encyclopedia') WHORF'S HYPOTHESIS AND THE PROBLEM OF TRANSLATION Benjamin Lee Whorf was an insurance salesman in the US in the 1920s. A recurrent problem in his job concerned the interpretation of words; their precise meaning was often extremely significant with regard to indemnity payments. What did it mean, for example, that a fire was 'self-inflicted'? And what did it mean that a drum of petrol was 'empty'? In some cases, it could be empty of petrol, but full of petrol gas and highly explosive. A fire which was caused by an empty petrol drum exploding could, however, not be defined as self-inflicted. Whorf s company lost some money on such cases. Some years later, Whorf developed an hypothesis on the relationship between language and the non-linguistic world which has enjoyed great influence in anthropology. Whorf s teacher in linguistic anthropology, Edward Sapir, played a part in the development of the idea, and the hypothesis is sometimes named the Sapir-Whorf hypothesis, but I shall speak of it as Whorf's hypothesis (Whorf 1956). It postulates that there is an intimate connection between the categories and structure of a language and the ways in which humans are able to experience the world. Whorf paid special attention to the language of the Hopis, which was almost without nouns as we know them and which also lacked the standard verb conjugations common to Indo-European languages. Since the language of the Hopis had these peculiar characteristics, Whorf argued, they would experience the world in a fundamentally different way from the descendants of European settlers in North America, who had brought their languages and grammars to the continent. The language of the Hopis was process-oriented and focused on movement, whereas English and other European languages were oriented towards things and nouns in general. Whorf argued that there was an intrinsic connection between the life-world of a people and its language; that every people will develop the 227