Notes on contributors
James l. watson was born in 1943 in the USA. He received his PhD from the University of California, Berkeley. He is currently Lecturer in Asian Anthropology at the School of Oriental and African Studies, University of London. He is the author of Emigration and the Chinese lineage (1975), and the editor of Between two' cultures: migrants and minorities in Britain (1977) and Asian and African systems of slavery (1980).
james woodburn was born in 1934 in the UK. He took his MA and PhD at the University of Cambridge. He is currently Senior Lecturer in Social Anthropology at the London School of Economics. He is the author of various articles on the Hadza of Tanzania.
1    Introduction: death and the regeneration of life
MAURICE BLOCH and JONATHAN PARRY
Introduction
This volume focuses on the significance of symbols of fertility and rebirth in funeral rituals, though all the contributors have found it impossible to exclude consideration of many other aspects of the treatment of death which are related to this central theme.
While it would take us too far from our central concerns to embark on a systematic historical review of the various ways in which our problem has been approached in the literature of anthropology and related disciplines, a few preliminary remarks may help to place the collection in relation to some of its direct predecessors.
The observation that notions of fertility and sexuality often have a considerable prominence in funeral practices excited the attention of anthropologists and their public from the very beginning of the discipline. The Swiss anthropologist Bachofen was one of the first to pay any systematic attention to the topic in his Versuch über Grabersymbolik der Alten which was published in 1859 and parts of which have been translated into English under the title 'An essay on ancient mortuary symbolism' (in Myth, religion and mother right, Bachofen, 1967). His study was principally concerned with Greek and Roman symbolism, particularly as manifested in the Dionysian and Orphic mystery cults, and its starting point was the significance of eggs as symbols of fertility and femininity in some Roman tombs and in funerary games. The eggs were painted half-black and half-white, representing the passage of night and day and the rebirth of life after death. "The funeral rite', Bachofen concludes, 'glorifies nature as a whole, with its twofold life and death giving principle ... That is why the symbols of life are so frequent in the tomb ...' (p.39)
The theme was picked up by a number of subsequent writers. It became, for example, a central preoccupation of Frazer's The golden bough (1890) which more ponderously reviews the material on the ancient mystery cults considered by Bachofen. The key question here
Maurice Bloch and Jonathan Parry
Introduction
is how killing can be a rite of fertility and renewal, and in particular how the killing of divine kings regenerates the fertility of the community. Although Frazer ranges widely, the extent to which his central ideas derive from classical examples is striking. What appears to be the fruit of cross-cultural comparison was in fact little more than an excursus on the ideas which inspired the Dionysian cults.
A comparable use of similar sources is to be found in Jane Harrison's influential Themis (1912), where the logic by which the mystery cults transformed death into birth is pursued much further, and where she goes beyond Frazer in discussing the significance of such symbolism as a way of linking the social order with the natural order. The combined impact of the works of Frazer and Harrison on literary circles in the first part of this century was considerable, as is well illustrated by the use made of the death and rebirth theme by Robert Graves in The white goddess and T. S. Eliot in The waste land. The irony is that - quite contrary to the spirit of Frazer and Harrison - Eliot used their work as justification for an antirationalist mystical point of view.
A discipline anxious to establish its academic respectability may well have been disposed to distance itself from the over-enthusiastic way in which its findings were sometimes used. But it was no doubt more directly because, by the 1940s, the central preoccupations of most anthropologists had moved away from a concern with systems of belief towards an emphasis on social morphology, that they subsequently seem to have shied away from any systematic consideration of the place of fertility in funerary symbolism. An exception here - more in tune with the spirit of an earlier generation - was Lord Raglan, on whom the influence of Hocart was formative and who was still preoccupied with the theme of the life-giving death of the divine king in a work published in 1945. Of course, specific ethnographic studies such as Evans-Pritchard's (1948) discussion of Shilluk kingship and G. Wilson's (1939) article on 'Nyakyusa conventions of burial' have a direct bearing on the issue, but it is no longer handled in the wide comparative manner characteristic of the earlier authors.
A quite different tradition concerned with the symbolism of death stems from Durkheim's pupil, Robert Hertz, whose 'Contribution to the study of the collective representation of death' was published in the 1905-6 volume of the Annie Sociologique (English translation, 1960). Hertz knew of Frazer's work, and Harrison knew Hertz's essay. But neither seems to have been particularly influenced by the theories of their predecessor, to whose work they merely appeal for confirmation of the striking parallels between funerary and initiation rituals. The difference between Hertz's study and those of Bachofen and Frazer is
both theoretical and ethnographic. Unlike these earlier writers. Hertz does not turn to Greece and Rome for his sources, but primarily to funerary cults of Malayo-Polynesian-speaking peoples. Although beliefs concerning the soul provide a key element in Hertz's argument, at the time it was written the informed reader - familiar with the works of other anthropologists like Tylor and Frazer and with those of folklorists and theologians - would probably have been struck by his comparative lack of attention to them. Nor was the link-up between death and sexuality given the same prominence as this was largely absent from his sources. The major symbolic themes on which Hertz elaborates are rather the characteristic South-East Asian contrast between the bones and the flesh, the pattern of double obsequies, and the parallels he discovers between the state of the corpse, the fate of the soul and the ritual condition of the mourners. It might in fact be argued that much the same criticism as we have made of Frazer would also apply to Hertz: that is, his general model was somewhat over narrowly related to the particular ethnographic material with which he started.
The central theoretical purpose of Hertz's essay is clear enough if we put it into the intellectual context in which it was written. The argument of the essay parallels his teacher Durkheim's famous study of Suicide (Dürkheim, 1952; first published in 1897). Durkheim's main point was that although we think of suicide as a supremely individual and personal act, it also has a social and non-individual aspect; as is shown by the fact that different types of society produce different rates of suicide. This social aspect, argues Dürkheim, can be studied in its own terms and suicide cannot be seen as a purely individual phenomenon. Hertz similarly chooses a topic which in the thought of his time was seen as peculiarly private and individual - the emotions aroused at the time of death. But 'death has not always been represented and felt as it is in our society' (Hertz, 1960:28); and following Durkheim's example, Hertz set out to show that these emotions - as well as the conception of death (for us occurring in an instant but for others a lengthy process) and the practices surrounding it - are in fact social and can be studied as sociological facts. Thus the detailed attention to the sequence of mortuary rites is intended to show how these rituals organise and orchestrate private emotions, a point which is illustrated with the example of weeping which, Hertz argues, is both institutionally governed and the manifestation of an emotion which appears falsely internal. As in the case of suicide, what had at first appeared as supremely individual, turns out to be the product of socially-constructed emotions and beliefs.
More than this, Hertz was concerned to emphasise that the problem
2
3
Maurice Bloch and Jonathan Parry
which has to be met at death stems from the fact that the deceased was not only a biological individual but a 'social being grafted upon the physical individual' whose 'destruction is tantamount to a sacrilege' against the social order (1960:77). 'Society' had to meet this threat by recuperating from the deceased what it had given of itself and regrafting it on to another host. There are thus two phases to the mortuary rituals: a phase of disaggregation (represented by the temporary disposal of the corpse) followed by a phase of reinstallation (represented by the secondary burial) from which the collectivity emerges triumphant over death. This dual process is mirrored in beliefs about the fate of the soul and the ritual condition of the mourners. It takes time for the collectivity to readjust to the death of one of its members, and this finds expression in the idea of a dangerous period when the departed soul is potentially malevolent and socially uncontrolled, and in the separation of the mourners from everyday life. The final ceremony, however, involves the reassertion of society manifested by the end of mourning and by the belief that the soul has been incorporated into the society of the dead and has settled down -in the same way as the collective consciousness of the living has been resettled by the funerary rituals. It is not, then, a matter of the fate of the soul determining the treatment of the corpse, but rather of the nature of society and the state of the collective conscience determining both the treatment of the corpse and the supposed condition of the soul.
The transfer of the soul from one social order to another (albeit imaginary) order is, however, invoked to explain the parallels between the symbolism of mortuary ceremonies, initiation rites and marriages; each of these involves a transfer in which a new social identity is grafted onto the individual. It is for this reason, Hertz argues, that funerals are double, not only overtly in the Indonesian and Malagasy examples considered, but also covertly in other cases. There are two jobs to be done: on the one hand a disaggregation of the individual from the collectivity, and on the other the re-establishment of society requiring a reallocation of the roles the deceased once occupied. Consistent with such an analysis, 'the death of a stranger, a slave, or a child will go almost unnoticed; it will arouse no emotion, occasion no ritual' (Hertz, 1960:76). Such individuals have not been fully incorporated into the social order, which therefore remains largely unmoved by their deaths for it is 'not as the extinction of animal life that death
occasions social beliefs, sentiments and rites____Since society has not
yet given anything of itself to the (new-born) child, it is not affected by its disappearance and remains indifferent' (1960:76, 84) - thus illus-
Introduction
trating once more the socially-determined nature of the emotional and ritual reactions to death,
Hertz's emphasis on the problem of reallocation necessitated by death led to much important work, such as Goody's (1962) analysis of the way in which the roles and property of a deceased LoDagaa are redistributed. But this focus has tended to take the discussion away from the relation between death and fertility symbolism, although Goody's own ethnographic evidence contains some intriguing information on the topic which might, we think, be worth a closer analysis.
The parallels which Hertz noted between death and other rites of passage were, of course, to be emphatically restated by Van Gennep (1909), though his analysis of funeral rituals was far less interesting. It is in relation to these parallels that Hertz's concerns briefly converge with those of Bachofen and Frazer. Every life-cycle ritual 'implies the passage from one group to another: an exclusion, i.e. a death, and a new integration, i.e. a rebirth' (Hertz, 1960:81). The rebirth which occurs at death is not only a denial of individual extinction but also a reassertion of society and a renewal of life and creative power - a position which is easily reconcilable with Frazer's.
The interests of both Frazer and Hertz are taken up in a recent book by Huntington and Metcalf (1979), though no new analytical framework is proposed and their intention is seemingly rather to stress what is valuable in a number of earlier contributions. Since they make no real attempt at a synthesis between the quite disparate theoretical approaches they draw upon, and do not consider the extent to which they are compatible with each other, the result is somewhat eclectic.
Both authors have worked in societies which display the classic Hertzian theme of the double funeral (Huntington in South Madagascar and Metcalf in Borneo), and it is for Hertz's analysis that they reserve their most fulsome praise. What is striking, however, is that their discussion of his essay almost totally ignores his central preoccupations with the social construction of emotion and with the relationship between the biological individual and the social collectivity. What they approvingly stress is rather the point of method he makes in emphasising the need to pay close attention to the treatment of the corpse, and the parallels he discovers between the state of the corpse and the fate of the soul. The particular example with which they choose to illustrate this last argument - the example of the Berawan of Borneo - is however shown to be rather equivocal; and we suspect that this might also be the case for other material on which Hertz relies. But while the matter was certainly important for Hertz, it is only a part of
4
5
Maurice Bloch and Jonathan Parry
a much wider and more general thesis concerned with the non-individual nature of reactions to death.
More in line with the Frazer-Bachofen tradition, another central theme of Huntington and Metcalf's book is the way in which 'the life values of sexuality and fertility (often) dominate the symbolism of funerals'. The centre-piece of their discussion of this issue is provided by a fascinating and elegant analysis of the funerals of the Bara of southern Madagascar, and we shall return to their interpretation of this case later on. Their more general views on the connection between death and fertility are, however, less fully elaborated and seem to amount to little more than the observation - reminiscent of Frazer -that such symbolism is a reassertion of life in the face of death.
Sociological, symbolical and psychoanalytical interests all come together in Morin, 1970 (first edition 1951), and more recently in Thomas (1975); but both of these studies are intended as a critique of western ways of dealing with death and as a result are of a very different character to this book. Nonetheless several of the points which Thomas makes are re-echoed in our introduction, although for different ends.
The present collection follows Huntington and Metcalf in trying to combine the concerns of the two rather disparate traditions we have outlined. Like Frazer and Bachofen we are primarily interested in the way in which the symbolism of sexuality and fertility is used in the mortuary rituals; but with Hertz we share a concern with the social implications of mortuary practices, though not his view of society as an entity acting for itself. If we can speak of a reassertion of the social order at the time of death, this social order is a product of rituals of the kind we consider rather than their cause. In other words, it is not so much a question of Hertz's reified 'society' responding to the 'sacrilege' of death, as of the mortuary rituals themselves being an occasion for creating that 'society' as an apparently external force. It is therefore particularly important for us to consider cases, like the four hunter-gatherer societies discussed by Woodburn, where at best this ritual recreation of the social order occurs only in the most attenuated form.
We have tried to combine the two strands deriving from Frazer and Hertz in that to a greater or lesser extent each of our authors is interested not only in the cultural logic of the kind of symbolism which preoccupied Frazer, but also (and here we are more in step with the sociological orientation of Hertz) in seeing this symbolism in relation to the organisational aspects of the society in which it occurs. For us, sociological analysis and symbolical analysis are not alternatives but
Introduction
need to be combined - and this we attempt to do in the present essay.
While all the contributors to this volume have attempted to ground their analysis firmly in a specific cultural context, several of the papers are explicitly comparative - though in rather different ways. Thus we have comparisons between the death-related practices of different categories of people within the same society (Parry; chapter 3), between different societies of the same economic type (Woodburn; chapter 7) or of the same cultural region (Strathern; chapter 4), as well as a discussion of the logic behind the variability between mortuary symbolism in different types of social system (Bloch; chapter 8). But further than this, we would claim that our papers are sufficiently closely related as to be mutually illuminating and to invite parallels and a continual cross-referencing. In however cursory and inadequate a way, we shall endeavour in the rest of this introduction to direct the reader's attention to at least some of the general considerations which might emerge from such an exercise.
'Fertility' and the vision of life as a 'limited good'
At the outset we should make it clear that we do not use the term 'fertility' in any restricted or technical way, but in the dictionary sense of 'fecundity' or 'productiveness'. If death is often associated with a renewal of fertility, that which is renewed may either be the fecundity of people, or of animals and crops, or of all three. In most cases what would seem to be revitalised in funerary practices is that resource which is culturally conceived to be most essential to the reproduction of the social order.
The mortuary rituals of the four hunter-gatherer societies considered by Woodburn display little concern with ensuring the continuity of the human group itself, or the replacement of its personnel. The emphasis is rather on the group's ability to appropriate nature - an ability which is put in jeopardy by the birth, sexuality and death of humans and which is restored by such rituals as the molimo of the Mbuti and the epeme dances of the Hadza. Harris stresses that in Laymi ideology the value of land is paramount, while large families are disapproved of because they upset the balance between people and land. Consistent with this, it is agricultural rather than human fertility which is the primary value and which is elaborated upon in the mortuary rituals. The Merina world, by contrast, is premised on a total identification between specific groups of people and specific areas of land, and the fertility which is ensured by the proper combination of ancestral corpses and ancestral land is the generalised fertility of both
6
7
Maurice Bloch and Jonathan Parry
Introduction
the group and its material means. In Strathern's Gimi example it is more a matter of reproducing the clan (which requires the mediation of the forest and the cannibalistic necrophagy of women). In the Hindu case discussed in this book, by contrast, we seem to be dealing in part with a more general notion which symbolically equates the funerary rites with the mystical regeneration, not of specific groups, nor merely of the deceased himself, but ultimately of the entire cosmos - a regeneration brought about by the 'sacrifice' that occurs on the cremation pyre.
The logic of Hindu sacrifice rests on the implicit assumption that a life must be relinquished if life is to be attained, and this in turn suggests that - from one point of view at least - life is seen as a 'limited good'. The papers by Bloch and Parry draw explicit attention to such a world view, which is also clearly implicit in several of Strathern's examples. Another obvious illustration is provided by Malinowski's (1948) discussion of Trobriand beliefs - according to which there would appear to be a given stock of souls in each sub-clan which is absolutely constant. On death the soul of a sub-clan member goes to Tuma, the island of the dead, where it settles down amongst its kin for another lifetime as a ghost. When it returns again to the land of the living it will find its way into the womb of a woman of its own sub-clan. Each sub-clan thus has a given quantum of soul-substance, some of which is contained in the living on Kiriwina while the rest is with the dead on Tuma.
It is not difficult to see - as Bloch points out - that this basic theme of life as a limited resource lends itself to various permutations. A more belligerent variant is to attempt to deny your enemies of their corpses in order to prevent them from recuperating the life-essence they contain.(Bloch's 'negative predation'); while a further escalation would be a system of 'positive predation' in which you endeavour to appropriate to yourself the life-essence of others by killing them. The purpose of this may either be to enhance the vitality of the killers themselves (as in the Jivaro case) or the vitality of the non-human resources on which they depend, as is suggested by Barth's report (1975:151) that the killing of a Baktaman enemy promotes the growth of the taro.
It is clear that such conceptions imply that death is a source of life. Every death makes available a new potentiality for life, and one creature's loss is another's gain. The corollary, that the regeneration of life is a cause of death, is illustrated by our Hindu and New Guinea examples, where sexual relations (especially for the male) are seen as entailing a depletion of life-essence. But in the Daribi case (Strathern's
paper) both sexes suffer. The man loses his 'juice' in ejaculation while a woman loses hers by breast-feeding; and this loss must be made good by eating meat, including the consumption of human flesh which is a 'way of supplementing one's vital juices' - (an example of 'positive predation' requiring an actual ingestion of the victim). The Etero provide a transformation on the same general theme: since sexual relations imply a transfer of life-essence, and since wives are disruptive outsiders, heterosexual intercourse is viewed as a somewhat prodigal activity. Male homosexual relations within the lineage on the other hand are approved, for they keep substance within the group and help young boys to grow. At this stage, however, the more general point we wish to stress is that there is a logical connection between the conception of life as a limited good and the idea that death and reproduction are inextricably related.
Given a world view of this kind it is therefore obvious why the rituals surrounding death should be so thoroughly permeated by the symbolism of rebirth. But such symbolism is, of course, by no means always associated with such a view of life, and at the most general level may be related to the fact that almost everywhere religious thought consistently denies the irreversible and terminal nature of death by proclaiming it a new beginning. Conception and birth are the most striking and obvious symbols available for asserting such a dogma. What complicates the matter, however, is that while the overwhelming majority of cultures deny that death is an individual extinction, the extent to which they use the symbolism of procreation to do so seems highly variable, and this variation needs to be accounted for. Moreover, biological reproduction - as we argue below - is a highly ambiguous symbol, and is often dramatically acted out in the mortuary rituals more as representative of something to be overcome than as an affirmation of regeneration.
Neither that which is regenerated nor the symbolic means by which the regeneration occurs can therefore be taken as self-evident. This must be examined in each case, and the answers must be seen in relation to the wider social and cultural context. It is only then that we can begin to account for the variation.
Death and the denial of duration
Leach (1961) has outlined what is essentially a sophistication of the argument about the way in which religious ideology uses the promise of rebirth to negate the finality of death. He suggests that our inherently ambiguous concept of time facilitates the assertion that
8
9
Maurice Bloch and Jonathan Parry
Introduction
birth comes after death as day comes after night. The category 'time' covers two quite different kinds of experience: time as repetitive and time as irreversible duration. By merging both kinds of experience within the same category we manage, Leach argues, to muddle them up and to avoid recognising irreversibility by assimilating it to repetition. As a result birth appears to follow death, just as death follows birth. '... If it were not for religion we should not attempt to embrace the two aspects of time under one category at all. Repetitive and non-repetitive acts are not, after all, logically the same.' In the paired essay, Leach discusses the way in which time is created by festivals which act as the boundary-markers by which duration is divided. 'We talk of measuring time, as if time were a concrete thing waiting to be measured; but in fact we create time by creating intervals in social life' (1961:135). The relationship between Leach's theory of taboo (Leach, 1966fl) and the idea that festivals serve to carve up the naturally continuous world into discontinuous chunks is obviously close.
In one way or another this interest in the relationship between concepts of time and death recurs throughout this volume. In the Laymi case it is the festivals of the dead which mark out the agricultural cycle and divide the year between a period of household production and a period of communal consumption. While people toil the ancestors are on holiday in the world of the living. But after the First Fruits ceremony, when their descendants are liberated from their labours for a period of leisure and conviviality, the dead must return to the cultivation of red chillis in the inverted world of Tacna over the sea. Death itself is unpredictable (its unpredictability being symbolised by the games of chance played during the final preparations for the interment) and this aspect is stressed in order to represent the irreversibility of time. The spirits of the recently dead are similarly unpredictable. But these capricious spirits are tamed by a series of rituals, so that at the festival of All Saints, which initiates the agricultural year, they can - as it were - be socialised into a source of potential recurrent fertility. While death as an event may defy all regularity, the dead are eventually incorporated into the predictable cycle of the year and are harnessed (however imperfectly) to the reproduction of social life. In this way - as Leach's argument would imply and as Harris suggests - the discontinuous is ultimately merged with the cyclic; and death is consequently transformed into a process which is essential for the continuation of life.
For Leach the problem with duration is that it implies the irreversibility of individual death, and is therefore psychologically unpalatable.
By contrast, the emphasis in Bloch's contribution to this volume is rather that duration and the contingency of events (as manifested pre-eminently in death) present a problem - not of individual psychology - but of an essentially political nature to do with the legitimation of traditional authority. In such systems positions of authority are conceptualised as belonging to an eternal and unchanging order, and their inviolability is therefore premised on a denial of history. It is thus that things have always been and must always remain. But individuality and the flow of events pose a problem for this theoretically static world and a threat to its continuity in that different role holders are patently different and the social order is not eternal. Both must consequently be negated. As one might expect, the funeral is one of the principal means by which this negation occurs (Thomas, 1975:178). The mechanism involved consists in a radical devaluation of the deceased's individuality by identifying this with putrescence and pollution 'which are elaborately constructed because once constructed they can be expelled'. But the mortuary rituals do not leave it at that. Having as far as possible erased individuality, they reassert continuity by equating death with birth into the depersonalised collectivity of ancestors which is the source of the continuing fertility of the living. The denial of ideologically threatening duration is thus accomplished by a denial 'of the main discontinuous processes in the social group, i.e. death'. Where we find the classic Hertzian pattern of double obsequies, the first disposal is associated with the time-bound individual and the polluting aspects of death, and the second with the regenerative aspects which re-create the permanent order on which traditional authority is based.
Middleton's paper also relates the themes of death, time and the legitimation of the social order. A central opposition in Lugbara thought is that between the 'home' (the domain of controlled fertility, presided over by male authority represented by the symbol of ordered speech) and the wild 'bushland' (the domain of uncontrolled sexuality and power, identified with barren female eroticism). Death comes in from the wilderness. Associated with this dichotomy are two different kinds of time. In Middleton's terminology, 'duration' governs the world of the home. 'Order, hierarchy and authority are associated with and sanctioned by differences in genealogical generation and by age', and thus-depend on the passing of time. But the wild exists in a timeless vacuum in which there is no duration, change or growth, and which lacks order, fertility and hierarchical authority (as opposed to uncontrolled power). With the passing of generations the dead will eventually move into the timelessness of the wild and cease to exercise
10
11
Maurice Bloch and Jonathan Parry
authority. But in the meantime the main concern of funerary rituals - of important men at least - is to hang on to the 'soul' which is the durational authoritative aspect of the person and which is installed in a shrine at the very centre of the home, while getting rid of the 'spirit' which is the 'timeless' wild aspect.
At first sight this situation might appear to be quite at odds with the line of argument which Bloch develops. While Bloch stresses the theoretically 'timeless' character of the Merina descent group and sees 'duration' as a threat to legitimate authority, Middleton seems to be describing a situation in which the 'durational' time of the home and the shrine is at the heart of authority, while the 'timelessness' of the wild is associated with the lack of it. The difference, however, is more apparent than real. What underlies the Lugbara concept of 'duration' is the idea of seniority, which legitimates the authority of the elders and thus maintains the continuity of the lineage. By contrast, the 'timelessness' which the Lugbara associate with the wild implies a lack of continuity and hence the absence of properly constituted authority -which is precisely what lies behind the Merinas' determined denial of discontinuity. The two cases are therefore more analogous than they might appear, the illusion of dissimilarity being largely the product of the ethnographers' discrepant use of English words. The essential point is that in both instances legitimate authority is founded on the orderly and faithful replication of the pattern ordained by the ancestors. What lies outside this orderly world, but threatens to engulf it, is unrestrained and insubordinate individuality - which the Merina identify with biological birth and death, and which the Lugbara identify with the non-procreative sexuality of women as opposed to their controlled fertility under the proper supervision of responsible men,
In one way or another the funerary rituals of each of the three societies we have considered so far in this section attempt to negate the unpredictability of death, for - as we would see it - an uncontrolled event of such centrality puts in question the extent to which the social order can really govern the lives of its members. The most threatening quality of death commonly appears to be its aleatory character (a sentiment from which we ourselves are to some extent distanced by the fact that we live in an environment where - for the first time in human history - survival to old age has become the norm). The symbolic negation of the apparent arbitrariness of death is, however, often accomplished by a rhetorical emphasis on what is being denied -as, for example in the case of Laymi gambling.
This endeavour to control the contingency of death is highlighted by
Introduction
the commonly encountered discrepancy between the event of physical death and the social recognition of it. After the Lugbara has said his last words to his heir, the latter emerges from the hut in which they have been closeted and calls out the cere - the personal chant - of the dying man, an appropriation which would be unthinkably evil at other times. This marks the moment of succession; and even if the patient lingers on after it, he is socially dead and his mortuary rites are performed as if he were dead. In the case of the rain-maker, the discontinuity between physical and socially-recognised death is likely to be very much more pronounced. He undergoes death - including a symbolic burial - at the time of his initiation, and when he is physically dead his corpse is interred at night and in silence, in a manner which is quite different from normal burial and which seems to approximate to the mere disposal of a carcass. An even more extreme example is provided by the Dogon (Paulme, 1940) where in some cases funerary rites are performed for people who are presumed to be, but in fact are not, dead. When this occurs, and the 'dead' man returns, not even his closest kin will recognise him and he is forced to remain a nameless beggar until his physiological death.
All this is strikingly paralleled by the Indian case. The Hindu ascetic, who performs his own funerary rites at the time of his initiation, henceforth exists in the world as a wandering ghost, and his corpse is not cremated but simply immersed in the Ganges. The effigy of a missing person who is presumed dead will be cremated, and his subsequent mortuary rituals performed. If he then reappears, he does so as an intrusive ghost who has no place in the world of the living, and (in theory) nobody at all will eat with him. In all these examples the social recognition of death precedes the physical event. But in the case of the Hindu householder this order is (with the exception just noted) reversed. The message encoded in the cremation rituals of one who has died 'properly' is that death 'really' occurs mid-way through the cremation when the chief mourner cracks open the deceased's skull with a bamboo stave in order to release the 'vital breath' from the body, and it is at this point that death pollution is commonly said to begin. The same sequence recurs in the case of those who have died a 'bad' or 'untimely' death. Here too an effigy of the deceased is constructed; a piece of lighted camphor is placed in its navel, and it is only when the flame burns itself out that the deceased is regarded as truly dead. Again this ritual performance discounts the actual physiological death and re-runs the event so that it conforms to the ideal of a controlled release of life.
The conquest of time is - on Parry's analysis - a central concern of
12
13
Maurice Bloch and Jonathan Parry
both the mortuary rituals of the Hindu householder and the practices of the Aghori ascetic, though the two cases deny duration in rather different ways. The ideal Aghori lives on the cremation ground, consumes his food from a human skull, eats excrement and the putrid flesh of corpses, consorts with menstruating prostitutes, and in other respects too inverts the proper order. Parry argues that all this represents a systematic attempt to escape from time, and hence from death. By systematically combining opposites the Aghori recaptures a primordial and static condition of non-duality, and identifies himself with Lord Siva, who transcends time. Like the Lugbara prophet, he does not die; he attains samadhi, a perpetual state of deep meditation or suspended animation in which he is immune from the normal consequences of death.
In the case of the Aghori ascetic, time is - as it were - halted. The ideal death of the householder in the sacred city of Benares suggests, however, a different kind of victory over events. Firstly, the person who has gone to Benares before his death has chosen to die in a particular place at a particular time, unlike those who are caught unaware by untimely death. Secondly, the symbolism of cremation aligns death with a perpetual cyclic renewal of time; for every cremation in Benares may be seen as an act of self-sacrifice which re-enacts the original cosmogonic sacrifice, and rekindles the fires of creation at the very spot where creation began. Consistent with this, Benares is seen as immune to the ravages of time and exists perpetually in the Golden Age of origins, while the rest of the world has progressively run down into the Black Age of the Kali Yuga (Parry, 1981).
To the non-Indianist such concepts may at first perhaps appear arcane and culturally specific. But our own cultural background provides us with a parallel which is not after all so remote. The cremation ground at Manikarnika ghat, which is the navel of the cosmos and the scene of its original creation, may be likened to Golgotha, which has often been represented not only as the site of Christ's crucifixion, but also as the centre of the world where Adam was created and buried. The blood of the Saviour falls on the skull of Adam interred at the foot of the Cross, and redeems him and mankind (Eliade, 1965:14).
We thinke that Paradise and Calvarie, Christs Crosse, and Adams tree, stood in one place;
Looke Lord, and finde both Adams met in me; As the first Adams sweat surrounds my face,
May the last Adams blood my soule embrace.
(John Donne, Hyme to God my God, in my sickness)
Introduction
Here again, then, death becomes an act of universal regeneration, which renews time and which is performed at the centre of the world, the place of original cosmogony (Parry, 1981) - a death that is regularly re-enacted in the ritual of the Mass.
It is evident, then, that individuality and unrepeatable time are problems which must be overcome if the social order is to be represented as eternal. Both are characteristically denied by the mortuary rituals which, by representing death as part of a cyclical process of renewal, become one of the most important occasions for asserting this eternal order. But in contemporary western cultures the individual is given a transcendental value, the ideological stress is on his unique and unrepeatable biography, and he is conceived of in opposition to society and his death is therefore not a challenge to its continuity. Moreover, while man's nature may be seen as immutable, the existing social order is not. It is therefore not surprising that in this context the symbolic connection between death and fertility should be far more weakly stressed than it is in the cases we have considered here. Other aspects of the ideology preclude any such elaboration, or render it superfluous.
'Good' and 'bad' death
Both the impulse to determine the time and place of death, and the dissociation of social death from the termination of bodily function, clearly represent an attempt to control the unpredictable nature of biological death and hence dramatise the victory of order over biology. The specificity and contingency of the event itself is suppressed so that death can be represented as part of a repetitive cyclical order. The 'good' death is thus the one which suggests some degree of mastery over the arbitrariness of the biological occurrence by replicating a prototype to which all such deaths conform, and which can therefore be seen as an instance of a general pattern necessary for the reproduction of life. By contrast, in nearly all of our examples, those deaths which most clearly demonstrate the absence of control are those which are represented as 'bad' deaths and which do not result in regeneration (Thomas, 1975:192).
For the Merina there is no worse nightmare than that one's body will be lost so that it cannot enter into the communal tomb, since the secondary burial of the corpse not only recharges the fertility of the descent group and its land, but also rescues the deceased himself from complete obliteration. Without this reburial not only is a potential source of regeneration lost to the group, but the death of the individual is truly terminal.
14
15
Maurice Bloch and Jonathan Parry
Introduction
In the Lugbara case the 'good' death is that which occurs in the home, the place of the shrines of the ancestors and of legitimate authority represented by the symbol of speech. The dying man must speak clearly to his heir, who then marks his orderly succession by taking over the personal chant of the moribund; In this way the proper order of the lineage is maintained in the locality where the lineage is anchored and continuity is guaranteed by the smooth transfer of authority. 'Bad' death on the other hand occurs at the wrong place, away from the ancestral shrines to which the deceased's soul cannot therefore easily return; and at the wrong time so that the orderly succession of speech cannot occur. The regenerating element of the dead man is lost in its antithesis, the disorganised wild.
Again, for the pious Hindu the 'good' death is that of the man who, having fulfilled his duties on this earth, renounces his body (as the ascetic has earlier renounced his) by dying at the right place and the right time, and by making of it a sacrifice to the gods. 'Bad' death, by contrast, is the death of the. person who is caught short, his body still full of excrement, and his duties unfinished. It is the death of one whose youthfulness belies the likelihood of a conscious and voluntary renunciation of life, or of one whose body is contaminated by a disease which makes it unfit as a life-creating sacrifice.
In all these examples the 'good' death not only promises a rebirth for the individual but also a renewal of the world of the living; while the 'bad' death represents the loss of regenerative potential. But in some cases a 'bad' death is not merely a lost potential for, but is an actual threat to, fertility. The Laymi can harness an ordinary death to the agricultural cycle, but the death of an unbaptised infant is positively harmful for the crops. The point is particularly well illustrated by the famous case of the Dinka Spearmaster (Lienhardt, 1961). The 'proper' death of the Spearmaster - at which he presides over his own burial alive - robs death of its contingency, and is an occasion for joy at which there must be no mourning. Such a death enhances the fertility and prosperity of the community of which he is the embodiment. On the other hand famine results if he is taken by death, rather than his taking it.
This last example brings us to a consideration of what in a great many cultures is the supreme example of 'bad' death - the suicide, whose self-destruction is regarded with such incomparable horror that the soul may forever be excluded from the society of the dead and must wander the earth as a lonely and malignant ghost, while the corpse may not be accorded the normal rites of disposal (as in Christian cultures where it could not traditionally be buried in consecrated
ground). This total rejection of the self-inflicted death might at first sight appear to represent a marked contrast to those cases - like that of the Dinka Spearmaster or the pious Hindu - where death must appear to have been chosen. However, the apparent difference is superficial and it is quite possible for both conceptions to coexist. In the Hindu case, for example, suicide is also the bad death par excellence, and is conceptualised as something quite different from the voluntary renunciation of life which is the 'good' death. While the one is a surrender to the disappointed desires of life and thus evinces an acute involvement with the material world, the other stems from absence of desire for life and a calm indifference to mundane existence. The 'good' death - whether we consider the Hindu case or that of the Merina or Lugbara - is a kind of handing over of a vitality which can then be recycled. In this way it recalls the supreme altruistic gift of the Christian martyr, or even of Christ himself, by whose death life is supposedly renewed. By contrast, the suicide acts for himself alone, and loses for others his regenerative power.
In spite of this clear ideological difference between suicide and self-sacrifice, the categorisation of any particular instance is not always unambiguous. While from one point of view the hunger-striker sacrifices himself for the cause, the authorities present his death as suicide. Indeed we would suggest that what feeds the horror with which suicide is so often regarded is that it is an act which, by its apparent similarity, almost parodies the death which is the ultimate manifestation of altruistic self-abnegation.
A further ambiguity may arise over whether any given death was 'good' or 'bad'. Even though in principle the distinction between them may be sharply drawn, in practice it is not always possible to be entirely confident about the case in hand. Despite the most fastidious endeavours, a mistake of ritual detail may have nullified the efficacy of the rites by which the rebirth of the deceased is guaranteed; while the mourners' actual experience is likely to tell them that the death they mourn fell some way short of the perfect case represented in the ideology. In almost every instance there remains a place for the suspicion that the victory over discontinuity which is acted out in the mortuary rites is an illusory one, and that death has not been successfully harnessed to the cycle of regeneration.
The uncertainty which surrounds death is also manifest in the existence of a number of beliefs which are different and even contradictory to the central ones - (for example, the belief in ghosts in Christian Europe). Such beliefs are usually less elaborated and less emphatic, but their occurrence is extremely'common and is illustrated
16
17
Maurice Bloch and Jonathan Parry
Introduction
by the Hindu, Chinese, Laymi, Merina and IKung examples in the book. This inconsistency stems in part, we suggest, from the way in which death is transformed into regeneration by acting out a victory over (and thus giving recognition to) the finality and uncontrollability of death. The 'good' regenerative death can only be constructed in antithesis to an image of 'bad' death, which it therefore implies. It requires and must even emphasise what it denies, and cannot obliterate that on which it feeds. We have already alluded to this antithetical process in our discussion of time, and it will emerge much more clearly in the sections which follow. All we wish to propose here is that since the dominant ideological representation is created out of its contrary, the negative aspects of death are accorded a prominence which it is hard to entirely erase. As a result, that which is asserted by the central ideology is unlikely to appear as a complete cosmology, and thus allows space for the elaboration of all sorts of subsidiary beliefs which are inconsistent with it.
Fertility and female sexuality
We have noted above that sexuality in general - and female sexuality in particular - is often seen as the cause of death; and also that the fertility which is regenerated by the mortuary rites may be either human or natural (or both). In order to push the analysis further we need to examine some of the ways in which these various elements - death, female sexuality, human reproduction and natural fertility - may be combined. More specifically, we must examine the fact that female sexuality is often associated with death only to be opposed to the 'real' fount of human and natural creativity; and that sexuality may be seen as the source of death and human procreation, which stands in opposition to non-human fertility.
The Judaeo-Christian tradition provides an example of this second possibility. In this tradition, it is the temptation of Eve which brings death into the world and results in the expulsion from Paradise, where the bounty of God's creation had provisioned the primeval couple without any significant effort on their part. But when they sinned against God's ordinance, God cursed Eve: T will greatly multiply thy sorrow and thy conception; in sorrow thou shalt bring forth children ...' (Genesis 3:16). Adam's punishment, however, was that the ground was cursed so that it brought forth thorns and thistles, and 'in the sweat of the face shalt thou eat bread'. He was then banished 'to till the ground from whence he was taken' (Genesis 3:17-23). Although
sexuality (initiated by woman) results in human fertility, this contrasts with the divine asexual fertility of Eden, is intrinsically flawed and is accompanied by death and a decline of natural fertility. Human and natural fertility are thus opposed to each other.
Much the same idea seems to be present in two of Woodburn's examples. For the Hadza, female reproduction is mystically incompatible with hunting big game: if the hunter's wife is menstruating the poison of his arrows loses its efficacy, and if she is pregnant a wounded animal will recover from its wounds. Again, for the Baka a pregnancy or death amongst the clanspeople of the same camp will ruin the hunt, until - intriguingly - the identity of the pregnant woman or dead person is known. Here again, then, human fertility is in some way antithetical to man's appropriation of nature.
In other instances the biological fact that human reproduction is the consequence of human sexuality is suppressed, or even denied, by the way in which creativity is ritually represented. So while Huntington and Metcalf - like Frazer before them - take commonsense as their guide and assume that the sexuality which is enacted in the mortuary rites must be a symbolic regeneration of life, we would perversely suggest that sexuality may be ritually elaborated as something to be overcome as antithetical to fertility. By this we do not intend to resurrect the old controversy over whether all peoples are aware of the connection between copulation and conception. Our point is rather the one suggested by Leach's (1966&) discussion of 'Virgin birth'; symbolic representations may totally transform what is perceived in other contexts. That is, the dissociation of fertility from sexuality is dramatically established by ritual in a way which denies what we and they know of biology. This disjunction is often effected by using gender symbolism to concretise the hierarchical contrast. Sexuality is set in opposition to fertility as women are opposed to men.
Again the Lugbara case provides us with an illustration. We have already noted that here the untamed sexuality of women is identified with the dangerous, socially-unproductive wild. This contrasts with the fertility associated with the compound, the ancestral shrine, and the authority of elders - a sacred ancestral fertility controlled by men in which women would appear to be but passive partners. What is more, it is not only female sexuality which belongs in the bushland but also parturition, for during her labour a woman is surrounded by taboos which symbolically remove her hut from the settlement and relocate it in the wild. It is not until the biological process has been disposed of that the new-born child can be introduced into the social world as a gift
18
19
Maurice Bloch and Jonathan Parry
made by the elders by virtue of their control over the source of legitimate fertility. Sexuality, biological birth and death are made to represent an intrusion of the wild, the natural sphere of women - an intrusion acted out in the unrestrained and orgiastic sexual pairing which occurs outside the homestead during the death dances and which cannot result in conception. Consistent with all this, the Lugbara myth of the origin of death (and also incidentally of the origin of social division) places the blame squarely on a woman, who cut the rope (or tree) by which men could return to converse with Divinity.
In such a context it would be wholly misleading to interpret the symbolism of sexuality in terms of a regeneration of fertility. But this, we suggest, is precisely the flaw in Huntington's striking and suggestive discussion of the funerary rituals of the Bara of southern Madagascar (Huntington, 1973; Huntington and Metcalf, 1979:98-118). Life for the Bara, Huntington argues, is a precarious balance between the sterile forces of 'order' associated with bone derived from the father, and the chaotic forces of 'vitality' associated with flesh derived from the mother. In death the balance is upset: the corpse is reduced to bone, order and sterility; and the purpose of the mortuary rituals is to restore the equilibrium 'through a symbolic increase in vitality' induced by unrestrained sexuality. So on this analysis the roles of women, flesh and orgiastic sexuality in the mortuary rituals are as agents of a regeneration brought about by the restoration of the female components of the person.
In the light of this interpretation it is perhaps surprising to find that the first and temporary burial of the corpse is clearly intended to effect a complete separation of the bones from the flesh, and that once exhumed the bones must be cleansed of any residue. If, in other words, the object is to restore the female element, it seems somewhat paradoxical to insist on the complete elimination of the flesh (the quintessentially female component of the person) - the more so since in other Malagasy societies (like the Merina) we find that people go to some lengths to retain both the bones and the flesh (in the form of the dust of the corpse).
It is, incidentally, tempting to see this variation as associated with the fact that the Bara stress exogamy while the Merina stress endogamy. More generally, it would seem that those systems which make a distinction between kin and affines are the ones which are likely to pick up on the common contrast between male bones and female flesh, and to be concerned to separate them at death (e.g. the Bara and the Chinese example discussed here); while those systems which allow no such distinction and which stress endogamy as an ideal
Introduction
are much more likely to be concerned with the corpse as a whole (e.g. the Merina and the Laymi of pre-Catholic times).
Another interpretation of the Bara data, more consistent with this aspect of their mortuary practice and more in line with the case we are arguing, is however possible. Immediately after death two huts are cleared: one for the men and one for the women (from which males are rigorously excluded). The corpse is secluded in the latter for a period of three days and two nights. During the daytime the people mourn, while the nights are given over to promiscuous sexual pairing initiated by the erotic dancing and provocative singing of the girls - behaviour which would be completely unacceptable to the Bara at other times. This period is terminated by what Huntington describes as a 'burial by capture'. The men go to the women's hut and forcibly remove the corpse over the anguished protests of the female mourners. Relays of young men run with the coffin towards the mountain of the ancestors, pursued by a group of young girls - hair dishevelled and clothes in disarray - who try to hinder their progress.
Often the girls intervene physically to stop the journey to the tomb and there ensues a tug-of-war over the coffin as the girls try to pull it back to the village. When this fails, the girls may run ahead and line up across the boys' path. The boys charge, using the coffin as a battering ram to penetrate this female barrier and continue toward the tomb. (Huntington and Metcalf, 1979:115-6)
Having reached the tomb, the deceased is reborn (head first like a foetus) into the world of the ancestors.
All this suggests a ritual drama in which women are given the role of an unacceptable obscene sexuality, in which they deliberately endeavour to implicate men, which takes place at night and which must be broken through during the day - as the obstructive cordon of girls must be broken through - in order to attain a proper rebirth into the world of the ancestors. In other words, it is the necessary defeat of women, sexuality and biology which is enacted, rather than their indispensable part in the re-creation of life. In this case then - as with the Lugbara - sexuality is, we suggest, opposed to fertility. It is associated with flesh, decomposition and women, while true ancestral fertility is a mystical process symbolised by the tomb and the (male) bones. Consistent with this ultimate repudiation of sexuality is the fact that a Bara woman is buried in her father's tomb and never in her husband's (Huntington and Metcalf, 1979:107). As with the Dobuan case, which we will consider later on, the final triumph over death is also a triumph over the necessity for affines and over the world of sexual reproduction which they represent.
20
21
Maurice Bloch and Jonathan Parry
Introduction
Women and putrescence
On our analysis, then, the symbolism of the mortuary rites of the Bara and Lugbara identifies women with sexuality, and sexuality with death. Victory over death - its conversion into rebirth - is symbolically achieved by a victory over female sexuality and the world of women, who are made to bear the ultimate responsibility for the negative aspects of death. In line with this, the sexuality of women is often closely associated with the putrescence of the corpse as, for example, Pina-Cabral's (1980) fascinating discussion of the cults of the dead in contemporary North-western Portugal shows.
Here the flesh (as metonym for the body) is what binds the soul to the mundane world, and its corruption is a necessary step towards spiritual purification. Three or four years after the burial the grave is opened and the bones are cleaned in order to rid them of the flesh, thus marking the final separation of the deceased from the living. But sometimes (surprisingly often) the disinterred corpse has not decomposed. There are two possible interpretations of this. The first, favoured by the priests, is that the deceased must have died with a large burden of unexpiated sin. This calls for a ritual 'lifting of excommunication' which consists of whipping the corpse while reciting prayers so as to unify him (or her) with the body of the church -(the incorrupt flesh being the material manifestation of his exclusion from it) - and to allow the soul to go to heaven. The laity, however, generally interpret an incorrupt corpse as an indication of sainthood, and if the body is reburied after the ritual whipping and still does not decompose, then the popular view is held to have been vindicated and the priests will be forced to take it seriously.
The significance of this material for our argument is that the female corpses which are found to be incorrupt - or at least those amongst them which eventually come to be venerated - are the corpses of women who are held to have been paragons of sexual purity. (The incorruption of male corpses, by contrast, is attributed to exemplary social and economic behaviour.) The parallel here is with the Blessed Virgin whose Assumption was in corporeal form, and whose ascent to heaven was not conditional on the decomposition of her immaculately conceived and virginally intact body. It could hardly be clearer that it is women's sexuality which causes the corruption of their flesh. Putrescence is a consequence of concupiscence, and an eternal preservation of the body is the reward of those who renounce its pleasures.
This association between (female) sexuality and decomposition is implicit in several of the case studies provided in this book. One of the
most remarkable features of Watson's Chinese material is the extent to which the corpse itself is an object of terror and its putrescence is the source of severe pollution. It is even said that marauding brigands have been kept at bay by the false report of a death in the village. The first burial allows the flesh to rot away, and when the body is later disinterred the bones are meticulously cleaned of the last vestiges of flesh before being reburied. Once this has happened the bones emanate a power which, if properly tapped, promotes the fertility of the descendants.1 It is, as one might by now expect, the men who handle this aspect of matters, while the married women (as opposed to unmarried girls) are crucial in the ritual sequence dealing with the putrescent flesh - from which the men absent themselves if they possibly can. True they have some excuse, for a man's male essence (yang) is depleted every time he handles a corpse, and by the seventh occasion he is permanently polluted, while a women is not affected in the same way. The hair is absorbent, like blotting-paper, and the women rub their hair on the coffin 'out of respect' - thus symbolically soaking up the pollution of the decaying flesh. In other terms, what seems to be happening is that at death the women reabsorb the flesh that is their contribution to thechild at birth, and Watson demonstrates that this in turn renews their powers as biological reproducers.
Both the theme of a particular affinity between women and rotting corpses, and an ambivalence about biological reproduction similar to the one we have encountered elsewhere, are worked through with emphasis in Gimi culture as Strathern's paper shows. 'Out of compassion' the women short-circuit the normal process of putrescence by consuming the flesh of a deceased male, thus freeing his spirit to return to the pre-eminently male domain of the rain forest which is ideologically represented as the source of productivity in the Gimi universe, and which is the abode of the ancestral spirits who 'collectively constitute a forest reservoir from which emerges the animating life-force of new generations' (Gillison, 1980:154). The recycled spirit is redeposited in women by men so that they may nurture it, but special ritual measures must be taken to force them to relinquish it at the proper time, for 'the female tends permanently to retain and in that way destroy (i.e. reabsorb) what she nourishes' (Gillison, 1980:148).
In a number of ways we are dealing with a complex set of representations which endeavour to overcome the spectre of a tyrannous biology. Not only is the good death a controlled release of life and copulation an insufficient cause of procreation, but even birth itself is (as in the Gimi case) induced by male authority, or (as in our
22
23
Maurice Bloch and Jonathan Parry
Introduction
Lugbara example) represented as an act of patriarchal benevolence to which the mother's travail in the 'wild' of the confinement hut is seemingly little more than a regrettable prelude. In this attempt to master the world of biology, gender symbolism often provides - we have suggested - the crucial mechanism. Fertility is separated from and made superior to the biological processes of sex and birth by analogy with the taken-for-granted difference between the sexes.
Bloch outlines a very general model in terms of which these various elements are combined, though his discussion starts out from the specifics of the Merina case. Merina women are identified with the domestic sphere of the household, the individual interests of which are seen as a threat to the unity of the undivided deme. In the rituals women are thus given the dramatic task of representing the divisions which are to be overcome. Consistent with our earlier discussion of the deceased as representative of individuality and duration, it therefore makes perfect sense in terms of Merina cultural logic that women should be associated with the polluting world of the time-bound individual with which the first funeral is concerned, and which must be transcended by the second burial. This transcendence is demonstrated by the elaborate assault on the world of women which occurs during this ceremony, when the de-personalised ancestor is incorporated into the collective tomb which is the source of pure fertility. A similarly antagonistic dichotomy applies to birth: physical birth - which is represented as an exclusively female activity - is polluting, and is subsequently transcended by the circumcision ceremony at which the child is torn away from the divisive and impure world of women to be reborn into the pure and undivided world of the descent group. Merina women may thus be said to act as representatives of that aspect of people which must be removed (biology and individuality) and are therefore associated with the decomposition of the corpse and the pollution which this causes. They take upon themselves the negative aspects of death and act as the defeated protagonists in a mock battle from which rebirth and fertility emerge victorious.
In this case it is the undifferentiated category of women which is associated with the pollution of the corpse. Among the Cantonese, however, this role is given only to married women; while elsewhere a distinction is drawn between women as daughters and sisters, and women as wives and mothers. In Ngubane's (1976) discussion of the Zulu, for example, we find that the chief mourner is always a married woman, who is aided by other married women of the lineage. The ethnography again reveals the association between, and devaluation
of, biological birth and biological death, for it is almost as if she is made to give birth to the corpse.
When a widow delivers a corpse to the lineage men at the doorway, her action represents delivering a baby to the lineage at birth. The corpse is tied up in such a manner that it more or less represents a foetus in the womb - with its knees and arms bent up. The hut in which the corpse and the mourners are, symbolizes the confinement hut as well as the womb itself ... Having delivered the corpse to the men, in a dramatization of birth, soon afterwards the conception is dramatized, when the chief mourner enters the round hole (representing the womb), receives the corpse from the lineage men and places it in the niche - to be born into the other world. (Ngubane, 1976)
In terms of the present discussion, the really significant point is that among the Zulu it is as wives and mothers that women are associated with negative polluting mystical forces, while as daughters and sisters they deal as diviners with positive mystical forces. It is tempting to see the significant variables here as exogamy, versus the endogamy of the Merina who do not in the matter of mourning significantly discriminate between different categories of women.
While an intimate association between women and the pollution of death appears to be extremely widespread it is not universal. It is absent from the South American Laymi case where the ideology continually stresses, not the subordination of one sex to the other, but rather their parallelism and complementarity. Nor would it seem to be a particularly prominent feature of the Hindu mortuary rituals - which is perhaps partly to be explained by the fact that here there are Untouchables, low-grade Funeral Priests and other specialists to shoulder the burden that is elsewhere assumed by women (Parry, 1980). Even where pollution is as far as possible off-loaded on the women, they are not of course the only ones to be infected by it. In our Cantonese case, for example, it is also absorbed and removed by the funeral specialists (who are paid for their pains), and by the direct heirs of the deceased (whose recompense is explicitly conceptualised as a share in the inheritance proportionate to the amount of pollution they soak up).
But whether or not putrescence is associated with women, an apparently superfluous emphasis on its horrors is common to a wide range of different kinds of society. This highlighting of decomposition is particularly striking in the case of the Cantonese with their endless insistence on averting their gaze from the corpse at critical junctures of the ritual. During the second burial the Merina display an uncompromising determination to force the participants into the closest possible proximity to the terrifying decomposed corpses. Again, the
24
25
Maurice Bloch and Jonathan Parry
theme of decomposition seems to have held a particular fascination for the western European mind during the late Mediaeval period - witness those funerary monuments which not only represent the putrescent corpse but for good measure also remind us of the worms wriggling in and out of it. But no more dramatic instance of this luxuriance in putrescence is perhaps to be found than that of the Aghori ascetics who live on the cremation grounds and consume their food out of human skulls.
It is, of course, obvious that such symbolism provides a potent warning against the vanities of the flesh and the transience of the sensual world - a denigration of the world of the senses which is particularly clear in the Aghori case, but which also seems to underlie the fifteenth and sixteenth century European fascination with decomposition which Huizinga (1965:136) suggestively describes as 'a spasmodic reaction against excessive sensuality'.
It is, however, to a different aspect of this hyperbolical elaboration that we would draw particular attention here. In all these systems death is harnessed to the cycle of regeneration and converted into birth. One of the key ways, we suggest, by which this restitution of life is dramatised is by the elaborate construction, and subsequent negation, of its antithesis - decomposition and decay. An emphasis on biological processes is used to darken the background against which the ultimate triumph over biology (and hence over death) can shine forth all the more brightly. This is perhaps clearest in the two Malagasy cases we have cited, where that which is to be overcome is carefully set up in order to be the more emphatically knocked down. But it is also significant that above the late Mediaeval representation of the maggot-infested corpse we may sometimes discover the pure, radiant and incorrupt soul leaving behind its corruptible shell and arising into heaven. Symbolically it is the corruption of the corpse which creates the purity of the soul - a point which Catholic belief itself comes close to recognising in the notion that it is the flesh which binds the soul to the profane world, putrescence thus becoming a necessary prelude to spiritual purification (see above p.22). Again, in the Cantonese case, it is the decomposition of the flesh which will eventually permit the recovery of fertility from the bones. From the participants' point of view, then, putrescence is in all these instances seen as a prerequisite for the distillation of life out of death. From the outside analyst's point of view it would, however, be preferable to say that the symbolism of regeneration actually derives its force from its juxtaposition to the antithetical symbolism of decomposition. The vigil of the Aghori on the cremation ground reveals this process clearly, for his morbid
Introduction
revelry in putrescence only serves to underline his claim to have transcended the world of biology and pollution, and to have conquered death. By wallowing in decay and death the Aghori histrionically proclaims his victory over them. In the light of all this it becomes significant that where - as in the four hunter-gatherer societies discussed by Woodburn - the symbolism of regeneration is weakly elaborated, there is little or no symbolic preoccupation with the process of decomposition, for nothing here is created in antithesis to it.
We have suggested, then, that the negative aspects of death are commonly seen as inseparable from other biological phenomena (like copulation and parturition); that in common with other biological processes, decomposition and decay are often (though not always) pre-eminently associated with women; and that this world of biology is elaborately constructed as something to be got rid of so as to make way for the regeneration of the ideal order. It is to this reconstituted ideal order that we turn our attention in the two sections which follow. The first of these focuses primarily on the commentary which this ideological representation of the community makes on marriage and exchange. In this connection, the Dobuan data from Melanesia with which we start, provides us with a crucial case. In the second section we shall pick up on a prominent theme to emerge from our re-analysis - the role of the tomb as a crystallised embodiment of the ideal community.
Eternity and the end of affinity
Perhaps the most dramatic feature of Fortune's (1963) ethnography concerns the Dobuan view of affines - 'Those-resulting-from-marriage' - as a constant danger to the exogamous matrilineage (shsw). All deaths are caused by a human agency, and the first suspicion of 'treacherous secret murder' by sorcery falls on the village kin of the surviving spouse, for
One marries into a village of enemies, witches and sorcerers, some of whom are known to have killed or to be the children of those known to have killed members of one's own village ... In the dark spaces between villages (at night) the agents of death roam - and death dealing spirits of women and men of all other villages, witches and sorcerers all. (Fortune, 1963:23)
It is hardly surprising, then, that marriages are extremely tense affairs and that the conjugal bond is fragile. A man is only really safe with his village 'sisters', with whom marriage is strongly discountenanced because such a union would not set up economic exchanges between two villages (Fortune, 1963:69). 'Incestuous' relations between classi-
26
27
Maurice Bloch and Jonathan Parry
ficatory 'brothers' and 'sisters' are, however, common; and the constant suspicion that one's spouse is conducting an affair with a sibling may prompt a jealous husband to time his wife's absences when she goes to the bush to defaecate, or to recruit children as spies.
A kind of suppressed yearning for the safety of incestuous reproduction is clearly apparent in Dobuan notions about plant biology, and in particular about the yams on which their subsistence is based. The particular strain of seed yams which any gardener cultivates is exclusively inherited within the susw; and although husbands and wives pool their production, they cultivate separate gardens in which they each tend yams of their own matrilineal strain. This is no mere matter of an idiosyncratic culinary preference; no other variety would grow for them. Just as seed yams remain within the lineage, so the garden magic which promotes their growth can, in theory, only be transmitted within the descent group.
All this is more than an question of a particular strain of yams being associated with a particular lineage; it is rather that the yams are part of the lineage. For the Dobuans all creation is the result of the metamorphosis of one thing into another. Yams are metamorphosed people, and they still retain many of their human characteristics. They have ears and hear, are susceptible to magical charms, walk about at night and give birth to children - though significantly there is no mention of them propagating sexually. They are tomot - 'human beings' (as opposed to Europeans who are of a different species). But further than this, they are lineage kin descended from the founding ancestress of the shsu. Their flesh - the Dobuan idiom - is planted in the gardens as the corpses of the ancestors are planted in the village mound which is the focal point of the settlement. What reproduces the lineage in a material sense, then, is - in Dobuan ideology - the flesh of its own kind. The consumption of yams, or at least of the yams grown from one's own lineage's seed strain, thus amounts to an act of symbolic endo-cannibalism.
Eating on Dobu is an intensely private affair. Meals are generally only shared by the members of a single conjugal unit (Fortune, 1963:74) though the key element in the formal recognition of marriage is the feeding of the bride and groom by their respective mothers-in-law (p.26). Commensality is closely associated with affinity and more loosely eating with sexuality. Consuming one's own yams becomes a metaphor for consuming one's own sisters. What we seem to have here, then, is the symbolic equation between cannibalism and incest that is such a pronounced feature of Strathern's ethnography.
This metaphorical association between yams and women, and
Introduction
i
! between sexual and alimentary consumption, is further revealed by
: the elaborate magical procedures which are directed as the seduction
; of other people's yams. 'Just as it is considered good form to try to
v seduce other men's wives whenever possible, so in gardening ...
every man should try to entice the yams, greatly desired personal • beings in metamorphosised form, from other persons' gardens'
'; (Fortune, 1963:134). With regard to both the objective is the same: to
; appropriate what belongs to others while hanging on to your own. But
( what needs to be kept firmly in mind is that such predation has a
f limited pay-off not only in that it invites mystical retribution, but more
importantly in that stolen yams are barren. It is only your own yams t which will reproduce, just as it is only your own sisters that can
reproduce the lineage. The symbolic assimilation of women to endogenous yams thus denies to affines any real role in the propagation of the descent group. Dobu is not after all so remote from the Trobriand Islands with their famous denial of physiological paternity.
Not surprisingly, this desire of the lineage to turn in on itself and abrogate all relations with outsiders emerges as a central feature of the symbolism of Dobuan mortuary practices. In life the pattern of alternating residence means that individuals must spend half of their time in the dangerous and hostile world of their spouse's village. Even at home one can hardly be safe, for the settlement will include the spouses of one's own brothers and sisters, who are serving time in their [ affinal village and who are probably witches and sorcerers.
All this however is only a problem of life, for at death one attains a permanent haven in the village mound where one is at last free from untrustworthy outsiders. This liberation of the deceased (as well as the liberation of the survivors from the affinal relationship which he or she contracted and which may well have been the cause of the death) is dramatised in various aspects of the funerary sequence. The house -which is the key symbol of the conjugal unit and from which all others are excluded - remains deserted until the end of mourning, when it is razed to the ground. The corpse itself is claimed by the susu, while 'Those-resulting-from-marriage' kneel outside and can neither look on it, nor participate in its subsequent display, decoration and interment.3 » After this the surviving spouse remains a kind of prisoner in the
j deceased's village until the end of an arduous year of mourning,
i Initially the spouse is incarcerated, with blackened body and a black
' rope around the neck, in a small walled enclosure of plaited coconut
j fronds which is built underneath the now abandoned house, where he
j or she remains 'sitting on a mat all day, walled off, speaking to no one
f and seeing no one' (Fortune, 1963:11). After a couple of months the
28
29
Maurice Bloch and Jonathan Parry
survivor emerges to do toilsome work for the deceased's lineage kin, but must hide when the skull of the deceased is brought out. At the end of this arduous regime the widow, or widower, is led out of the village and may never return.
All this has to be placed in the context of a situation in which the rules of residence ensure that the empirical manifestation of the descent group is highly problematical in that its living representatives have no territorial integrity. Yet the smsm is geographically anchored -by the large communal burial mound which stands at the centre of its notional settlement, and around the periphery of which are grouped the individual houses of the various conjugal families who are the part-time members of the local community. The physical marginality of these houses, and their impermanence provide a fitting symbol of affinity itself. The burial mound, by contrast, is a place of permanence and stability. Here the ideal unity of the susu is finally realised after the aggravating flux of life. It is not too much to say, then, that burial constructs a Dobuan vision of Utopia in which the boundary between one descent group and another is hermetically sealed. What is achieved when 'Those-resulting-from-marriage' are sent packing after death is the creation of the ideal community represented by the mound where the generations succeed each other without the unpleasant necessity of exchange - in much the same way as the lineage yams reproduce themselves asexually,
A similar ambivalence about exchange is, as we read the evidence, implicit in much of Strathern's ethnography. According to the Melpa theory of genetics, bone comes from the father, while from one point of view flesh is the contribution of the mother. In another way however, this is denied, for the flesh of the living is also seen as a product of the fertility brought about by the putrescence of the dead, whose bodies regenerate the soil and feed the plants on which the living subsist and which creates their substance. In a further respect, too, Melpa symbolism seems to suggest an attempt to deny - or at least replace -women, if not as the source of birth then as the source of the nurturing towards adulthood. Shortly after birth, the child's navel-string and its mother's placenta are buried together, and a cordyline or a banana tree is planted over the spot. As the plant grows, so does the child. 'The navel-string', says Strathern, 'which once connected the child to the mother in her womb, now connects it to the earth, and the link is represented by the cordyline ...'. What this seems to suggest is that as soon as they possibly can, the Melpa replace women by the clan territory. They would seem to be working out the same phantasy as is revealed by their proclaimed abhorrence of cannibal-
Introduction
istic consumption - the phantasy of a world without exchange. But whereas - on Strathern's analysis - cannibalism shows the disadvantages of such a world, this ideological displacement of women would seem to display a surreptitious yearning after it. Our own interpretation of Melpa cannibalism, however, would be rather different; and would be consistent with - rather than contradict - what we have said about the replacement of women. That is, we would argue that Strathern's interpretation should be turned on its head: since the typical cannibal is an in-marrying wife, the problem which cannibalism highlights is not so much the dangers of non-exchange, but rather the dangers attendant on the necessity of exchange with others exemplified by a reliance on outsider women. Seen like this, the difference between the Melpa and the Etero is less extreme than Strathern suggests. It is not so much that the beliefs surrounding the imaginary cannibalism of the Melpa are (as Strathern's analysis would suggest) an emphatic repudiation of what the Etero practice, but rather that Etero practice corresponds to the Melpa phantasy of an enviable order without exchange - an order which they create in their rituals but which their society, based on the foundation of exchange, makes impossible.
While Melanesian peoples in general are commonly associated with a maximum ideological elaboration of exchange, the two Melanesian societies we have looked at here simultaneously entertain the countervailing vision of an ideal order without exchange. This observation does little more, however, than echo what Levi-Strauss expressed in far more general terms and in far more vivid language in his concluding paragraph to The elementary structures of kinship and marriage:
To this very day, mankind has always dreamed of seizing and fixing that fleeting moment when it was permissible to believe that the law of exchange could be evaded, that one could gain without losing, enjoy without sharing, At either end of the earth and at both extremes of time, the Sumerian myth of the golden age and the Andaman myth of the future life correspond, the former placing the end of primitive happiness at a time when the confusion of languages made words into common property, the latter describing the bliss of the hereafter as a heaven where women will no longer be exchanged, i.e., removing to an equally unattainable past or future the joys, eternally denied to social man, of a world in which one might keep to oneself. (Levi-Strauss, 1969:496-7).
To these observations we would however add that in the converse case - where the exchange of women between groups is ideologically discountenanced - certain aspects of ritual life may express an equally ambivalent attitude towards the consciously articulated values. This is
30
31
Maurice Bloch and Jonathan Parry
illustrated by the striking contrast between the (imagined) cannibalism of the Melpa and the necrophagy of the Aghori ascetics. That the former is associated with greed and consumption while the latter stands for an ascetic denial of normal consumption should be obvious enough. But what is more germane to the point at issue is Parry's observation (chapter 3) that 'with the destruction of boundaries implied by the consumption of flesh, excrement and so on, goes an affirmation of the irrelevance of caste boundaries'. In this respect also the symbolic load carried by Aghori necrophagy would seem to be precisely the reverse of the message encoded in the myth of Melpa cannibalism. In both instances cannibalism represents an antithesis to the prevailing order. But while for the Melpa it stands, we have argued, for the dangers of exchange in a world premised on the principle of exchange, in the Aghori case it stands for the suppressed potential of the exchange of substance to dissolve the social barriers between groups in a world premised on their exclusiveness and closure (the orthodox Indian ideal being much more like that of the Etero, for whom bodily substance is to be conserved rather than exchanged).
At another level, however, this contrast with the Melanesian material disappears. In the Dobuan case, we have argued, the eternal order is created by the repudiation of exchange. But there is a sense in which this is also true of the Aghori. Exchange not only presumes an alter, but also creates or maintains differentiation and - where it takes an asymmetrical form - the hierarchy of castes. But all this is precisely what the Aghori denies. For him everything in the universe has the same essence, and the distinction between ego and alter, or between Brahman and Untouchable, belongs merely to the worid of illusory appearances. By realising this state of non-differentiation through his ascetic discipline, he attains an eternity in which there is neither death nor birth. As with the Dobuans, then, permanence is only achieved by overcoming the differentiation on which exchange is based. There are no others in eternity. In the Dobuan case they are eliminated by the suppression of affinity, while in the Aghori case they are eliminated with the boundary-maintaining pollution practices that separate alter from ego, and both from perpetuity.
Tombs and the social order
If the avoidance of exchange constructs the image of permanence, we have seen how, in the Dobuan case, this construction takes a material form in the burial mound.
Introduction
A further example of the graveyard as a symbolic representation of the social order is given in Firth's (1936) famous ethnography of Tikopia. The focus of traditional Tikopian social organisation is the paito, a term which Firth has variously translated as 'house' or 'lineage', but which might also have been glossed as 'tomb'. The paito is given its identity by the house-site on which it stands and from which its members take their name. This name identifies its bearers as the heirs to a long line of previous residents on the site, and therefore merges descent with locality. But that is not all. One half of the traditional Tikopian house is not actually lived in because underneath the mats which cover the floor are buried the former occupants of the house; the previous bearers of the name. Tikopians live, therefore, on the cramped borders of their tombs and take their identity from what is essentially a necropolis. As in Dobu, tombs in Tikopia are the enduring units of society and provide the material symbol of their continuity.
W.A. Douglass' ethnography of a Spanish Basque village, Death in Murelaga (1969), describes a situation reminiscent of the Tikopian case. Here the primary unit of social organisation is the rural farmstead, or basseria, which is managed by one domestic group (or household) consisting of a couple, their children, and perhaps also an aged parent (or parents) and any unmarried siblings. Only one of the couple's children will be appointed as the heir to the farmstead, while the other children will be given a share of the moveable property and expected to set up on their own. A basseria cannot be dismembered by sale or inheritance and, like the Tikopia paito, theoretically exists in perpetuity: 'in rural Basque society social continuity is provided not by descent groups but rather by the immutability of households' (Douglass, 1969:88). People derive their names from the farmsteads on which they reside; their social identity is completely bound up with a specific household and they interact with others as members of it (p.115).
Each basseria is associated with a sepulturie on the floor of the local church, which is not only a family stall in which the women sit on Sundays, but also a symbolic burial plot and the focus of mourning rituals and rites to the dead. This sepulturie is the responsibility of the heir of the farm. Up until the late eighteenth century it was the real burial site, and is still seen by people as a family tomb. Today, however, corpses are buried in the graveyard outside the church in plots allocated by household. But the real grave is of little importance.
Whenever there is a death, the gravediggers exhume a previously buried corpse, deposit the bones in the ossuary, and then bury the newly deceased. Little attention is paid to where the individual is buried. In a few months the
32
33
Maurice Bloch and Jonathan Parry ''
weeds claim the grave, and it is often impossible to distinguish the outlines of the grave plot. (Douglass, 1969:72)
Such neglect contrasts with the attention devoted to the sepulterie. On ■! Sundays the women surround it with candles and during mourning it is the focus of Masses said for the collective dead. If the basseria is sold, ; the new owners assume full control of the sepulturie and the vendors relinquish all rights. Masses said for the collective dead are devoted to ! all former owners of the farmstead traced through any line of descent or none, but excluding kin who have moved out. In this way the sepulturie represents the uninterrupted integrity of 'the basic unit of rural society and economy' (Douglass, 1969:6).
The role of the tomb as a symbol of the continuity of the property-holding kinship group is also nicely illustrated by the scene in J Galsworthy's novel To Let, where Soames sits by the family vault in Highgate cemetery morosely pondering the end of the old order and the dispersal of the Forsytes, augured by the cremation of Susan Hayman and the uncertain future of the family mausoleum. This ; volume offers several other examples of groups creating themselves and sealing their association with a particular locality by the construction of sepulchres. Harris' discussion of the Laymi illustrates well how the location and inclusiveness of the cemetery reflects changes in the character and autonomy of the local community. Consistent with their | objective of undermining the relative independence of the pre-colonial " local groups, the Spanish authorities centralised the cemeteries. Now ; the trend is again towards the political detachment of the local t community from the wider society, and is in turn reflected in a renewed dispersal of the graveyards.
Perhaps clearest of all is the Merina case. The fundamental unit of Merina social organisation is the deme, an ideally endogamous kindred associated with a specific territory by a number of tombs. The :■ tombs not only contain the remains of members of the deme but may actually be said to create the deme, which - like the Dobuan swsw - has no territorial integrity this side of eternity since most of its members J may actually reside elsewhere. After death, however, the expatriate , Merina returns 'home'. This regrouping of the dead, which is a central symbol of the culture and which underlies the joy of the second funeral, is achieved by the entry of the new corpses into the collective mausoleum. In this way the tomb and the reunited dead within it represent the undivided and enduring descent group, and as a result is the source of blessings and the fertility of the future.
The force of this symbol of the tomb as the representation of the f eternal undivided group can only be sustained by down-playing the |
Introduction
individuality of the corpses which enter it - which the Merina do by breaking up their individual dead in order to 'group' them. The same process is also found in a less material form among the Laymi. At the festival of All Saints 'the spirit is first welcomed and mourned individually and then there is a move towards collectivity in the graveyard'. The collective tomb or cemetery in which the group finds continuing life brooks no individuals, or is at least antithetical to the long-term maintenance of individual memory.
On the face of it the generality of this anti-individualism in funerary rituals would appear to be completely undermined by those cases in which the tomb of a single individual - for example, a king - becomes the focus of the continuity and fertility of the community. Such a situation has been discussed by Bloch (1981) for western Madagascar, though probably the most obvious example is that of the Egyptian pyramids (Frankfort, 1948). In passing we may observe that in such cases the growth in the size and significance of royal tombs seems to be accompanied by a diminution, or even a total eradication, of the tombs of the subjects as permanent objects. The central problem, however, is that the renewal of life implies the denial of individual death; yet the identification of the royal tomb as the source of regeneration comes close to emphasising an individual lifespan. One solution to this dilemma is to stress - as do the Merina - the royal line rather than the individual monarch. In the ritual of the royal bath the living Merina king goes to the tomb and washes himself with water associated with the collectivity of his forebears. He thereby proclaims his complete mystical identity with his predecessors and successors. Another and more radical possibility is to deny altogether the transient nature of the ruler's life by transforming his body into a permanent death-transcending mummy. In the one case it is the individuality of the king that is denied, in the other it is also his mortality.
In this section we have focused on the way in which tombs are used to construct an idealised material map of the permanent social order. It is worth pointing out, however, that the general process we are dealing with does not necessarily require the corpse in its entirety, or even any of the physical remains of the deceased. The same result can also be achieved by utilising only a part of the corpse (as appears to have been the case with the skull houses of the Melpa), or some immaterial aspect of the deceased. This possibility is realised by the Lugbara (as well as by many other African peoples). Here what is retained of the deceased is his 'soul', which is established in a shrine as a 'ghost' after the threatening individual aspect of the deceased's person has been cast away into the 'wild' thereby allowing his moral
34
35
Maurice Bloch and Jonathan Parry
lineage aspect to be created anew in the 'home'. If we bear in mind that there is a sense in which ancestor worship creates the lineage in such societies, it is clear that we are once again dealing with the fabrication of an ideal social order out of the transformed remains of the dead.
These various systems - the Dobuan, Tikopian, Laymi, Merina and Lugbara - are all cases where the community in its enduring aspect is constructed by reference to the dead. Whether this is actually accomplished by means of skulls, corpses, tombs or shrines is perhaps of little significance in terms of the overall logic of the ideology - which only goes to show how misleading it may be to extrapolate collective representations about death directly from the evidence of material culture. In all these instances what is created by the mortuary symbolism is a particular group or division of society - a lineage, for example, or a local community. But this is not - as Parry's discussion of the Indian data indicates - invariably the case.
At first sight Hindu funerary practices could not be more different than the cases we have just considered. In Hinduism nothing of the individual is preserved which could provide a focal symbol of group continuity. The physical remains of the deceased are obliterated as completely as possible: first the corpse is cremated and then the ashes are immersed in the Ganges and are seen as finally flowing into the ocean. The ultimate objective seems to be as complete a dissolution of the body as possible.
This last statement needs however to be qualified. The Hindu concept of the 'body' (sharira) does not entirely correspond with our own, and the real aim of the mortuary rites is more specifically the radical destruction of the 'gross' physical body {sthula sharira). The limb by limb creation of a new body for the deceased is the central purpose of the ritual sequence of the first ten days after cremation. But this new container for the soul is a 'subtle' or 'ghostly' body (suksham or preta sharira)) of a more refined and less elemental form. What is more, the preta sharira appears to be a transitory form which will also soon be dissolved, or transformed, with the assimilation of the deceased into the category of the ancestors. A ball of rice representing the newly created body of the deceased is laid alongside three other rice balls representing his father, father's father and father's father's father, It is then cut up into three pieces, the first of which is merged with the father's rice ball, the second with the grandfather's and so on. In Benares the three ancestral balls (which now contain the three fragments of the deceased) are subsequently rolled into one and immersed in the Ganges. In other words, the process of generation is
Introduction
reversed so that the generating absorb the generated (until - as the ultimate immersion of the consolidated rice ball perhaps suggests -total annihilation). It is as if the genealogy of the dead had been reversed. Whether, when all this is done, the deceased can be said to exist in bodily form is not altogether clear from informants' statements. The theologically unsophisticated tend to talk as though he now exists in a purely incorporeal state in the 'form of air' (vayu rup), while those who are better versed in the texts present it as a matter of acquiring an even more ethereal bodily envelope. But whichever view one takes, the underlying progression is the same: a gross material body is replaced by more and more refined forms.
All this bears testimony to the influence of the ascetic ideal and to a preoccupation with abandoning the sensual, material order. The Brahmanic theory of the four stages of life supposes that at the end of his life a man will eventually renounce the world and become a wandering and homeless ascetic. In fact, of course, this ideal is hardly ever realised in practice. But it is here, we suggest, that cremation fits in as a kind of surrogate for the ascetic's abrogation of the body. It is a kind of catching up on the renunciation of carnal existence which should ideally be the conclusion to every (male) life. Cremation must therefore be represented as a voluntary act of self-sacrifice consciously undertaken by a living individual. The conventional observation that renunciation is a kind of death may thus be reversed: death takes the form of a kind of last-ditch renunciation. Here we have one explanation of why the ascetic himself is not cremated. He has already accomplished what cremation belatedly achieves for the householder. To burn his corpse would not merely be redundant; it would also be to insinuate that his renunciation had been inadequate. Cremation and renunciation are both directed towards the same end: a repudiation of the 'gross' body. This repudiation is somewhat melodramatically proclaimed by the Aghori ascetics whose practices demonstrate that they can with impunity reverse what remains essential for those who are still bound to the mundane world and are still carnal beings. Having reached beyond carnality, they bombastically declare themselves impervious to the most polluting substances and actions.
The Hindu case, with its insistence on a total destruction of the physical body, is therefore quite different from the Malagasy, Dobuan, Laymi and Lugbara cases where an attempt is made to retain a pure and regenerating aspect of the deceased in order to construct and reproduce an ordered group. Beyond this difference there is however a more fundamental similarity. The overwhelming evidence is that the
36
37
Maurice Bloch and Jonathan Parry
theme of life out of death is as central to Hindu thought as it is to any of the other systems we have considered. But while elsewhere the mortuary rituals guarantee the continuation of specific groups, in Benares they re-enact the creation of the whole universe and regenerate the cosmic order. The real difference between this case and the others is the difference between a system in which the mortuary rituals dissolve the fundamental units of society into an undifferentiated universe, and those systems where they shore up these units and give them a permanent and transcendental value. In the eyes of the Hindu ascetic (traces of whose ideology we have discovered in the mortuary practices of the householder) this world of appearances - where caste is divided from caste and man from god - is merely, after all, the product of illusion (maya).
We are back, then, with the contrast with which we ended the last section. On the one hand we have those systems where the eternal order is achieved by the abolition of exchange relations between groups, which produces a static fossilisation of these groups in their sepulchres. On the other hand we have the Hindu system where - in order to attain a timeless eternity - exchange is abolished by the destruction of the internal divisions of society, whose existence implies the necessity of exchange.
The limits of the ideological representation
In the previous sections we have discussed how 'fertility' is created out of death. In order that this eternal, stable, life-giving element can be constructed, it is antithetically contrasted with another order built up by reference to such notions as 'biology', 'individuality', 'flesh', the 'gross body', and 'exchange'. What all these things have in common is that they refer to life and people as they are known in the everyday world, though seen for the purpose of the ideology in a particularly hostile light.
The elimination of this element creates the transcendent but this process is, as we have already noted, inevitably problematic. The problem lies in the very use of the antithesis. It recognises what it devalues. We saw, for example, that funerary rituals which deny death in order to construct an eternal source of fertility where life and death are merged, have first to revel in decomposition, biology and the dangers of exchange.
There is clearly, however, an even more fundamental reason why ideology cannot completely eliminate the natural world of biological
Introduction
process or the social world of exchange. To refuse the first would be to jeopardise the physical continuity of the group, while to refuse the second would be to deny society itself. That there is a real problem here is shown by the Aghoris who come close to repudiating all social relationships, and even more clearly by the Dinka Spearmaster who does indeed defeat the contingency of biological events, but only at the cost of abolishing himself. As illustrated by both these instances, the ultimate irony of death is that its final conquest is only achieved by embracing it oneself.
Such a solution is clearly not available to any on-going social system, and this fact most poignantly reveals the limits of ideology. What is more, the ideology has to be put to work in that very world which it denies, and it must therefore be compromised. This compromise is manifested symbolically and this explains certain aspects of the funerary rituals we have not so far examined, and which take the form of the reintroduction, in certain ritual contexts, of what had enthusiastically been denied in others.
Lugbara funerary practices, for example, expel that part of the person - the 'spirit' - which is symbolically and linguistically associated with the 'wild', and project an image of uncontrolled sexuality as sterile. But it is then as if the amorphous and destructive power of the wild must be, with the greatest caution, reintroduced so that it can augment the fertility of the ancestors. This power is brought into the 'home' by women; and more particularly by their sexuality, which by itself is dangerous and unproductive, but which - like fire - is at the centre of social life once it has been tamed. Control of the wild always remains problematic, though divination is one means to this end, A more successful attempt to channel its power seems to be that of the rain-maker who is able to contain within himself the force of divinity and to control it for the benefit of man.
The same dialectic is found in a different form in the Cantonese example. There the key symbols of legitimate authority and the enduring lineage order are the ancestral halls and the ancestral tablets, which must at all costs be protected from the contamination of the corpse. This terrifying object is to be disposed of as quickly as may be. Yet the elements of which it is composed must not be irretrievably lost. The flesh is, as we have seen, recouped by women - through their hair and by way of the absorbent green cloths which they wear at funerals and which reappear in the harnesses they use to carry their infants. However the greatest power resides in the bones, and this can only be recovered under the strictest precautions.' In order to confer fertility
38
39
Maurice Bloch and Jonathan Parry
and good fortune on some or all of the deceased's descendants (who are merely a segment of the wider descent group) his bones must be located according to the finest geomantic calculations, which if inaccurate may - like an uncontrolled intrusion of the Lugbara 'wild' -also bring disaster. The Chinese symbolism of the bones and flesh would seem therefore to be significantly different from the apparently similar bone/flesh dichotomy we have encountered among the Bara and the Melpa. In these latter cases the bones stand as symbols for the permanent units of society. But in China this symbolic function pre-eminently belongs to the ancestral hall and the tablets it contains; while both bones and flesh represent a dangerous vitality which can only be recovered under strict control. As Freedman (1966:143; 1967) made clear, the bones stand for the amoral advantage of some as against the moral authority wielded over the collectivity by the ancestral spirits in the lineage hall.
By contrast with these cases, the main concern of the Laymi would seem to be to expel the dead altogether, without retaining any aspect of their personality. The community, cannot, however, exist without the fertility which they control. However regrettable the necessity, the dead must therefore be brought back at the festival of All Saints so that they can preside over the growth of the crops in the months which follow. But these ancestors are half devils, who belong to the evil sphere and whose power is potentially malevolent. The period during which they remain in the world is a period of sadness, toil and restraint; and the festival which welcomes them emphasises social differentiation and is a time for fighting and battles, for the blood spilt is necessary for the earth to bring forth its harvest. The dead confer fertility, but the price which has to be paid is high and they are inseparably associated with the world of particularistic interests. At Carnival, however, this world is abolished, fights are prohibited, individuals are made anonymous, individual rights are transcended (as, for example, in the legitimised stealing of the standing crops), and the ancestors (whose work is now done) can at last be called devils to their face and unceremoniously driven out of the village to make way for a season of sociability and communal solidarity.
As with the Lugbara and the Chinese, we are once again confronted with a recognition of the necessity of coming to terms with what ideology would prefer to eliminate altogether, and with the dangers of doing so. This danger is again revealed by the special vulnerability of pregnant women to funerals which we find among the Laymi, the Merina and the Chinese. It is as if their condition makes them peculiarly susceptible to the untamed power which is unleashed, as a
Introduction
consequence of which they would - as the Merina believe - bear
anti-social monsters.
Death and the legitimation of authority
The unavoidable compromise which the ideology has to make with the practical world of nature and exchange is however mitigated by its careful reintroduction under the strict control of authority. This brings us back to the political significance of the representations we have been considering. At several points we have stressed the relationship between mortuary beliefs and practices and the legitimation of the social order and its authority structure. This relationship is perhaps clearest in those instances where that order is built up by transforming the dead into a transcendent and eternal force - as, for example, in the Lugbara, Cantonese and Merina cases. In these instances the social group is anchored, not just by political power, but by some of the deepest emotions, beliefs and fears of people everywhere. Society is made both emotionally and intellectually unassailable by means of that alchemy which transforms death into fertility. This fertility is represented as a gift made by those in authority which they bestow by their blessings.
If the political implications of funerary practices which emphasise the permanence of distinct social groups are clear enough, it is by no means obvious that any such significance can be attached to the apparently radically different Hindu case, where the mortuary rites do not eternalise distinct social groups but actually seem to do the exact opposite. However, even in this case, the pattern is recognisable. The transcendent authority of the Brahmans, who sacralise the social order, is reinforced by the theory of death as a cosmogonic sacrifice, for this theory locates the ultimate source of regenerative power in the ritual sphere, and places its control in the hands of those who operate the sacrifice. As for the ascetic who is striving after liberation from the cycle of rebirth, his whole endeavour is founded on an attempt to escape the inexorability of karma. He thereby acknowledges its reality for the man-in-the-world, and hence the principal ideological justification for the inequalities of the world.
The contrasting position of the Brahman and the ascetic illustrates well the complexity of the relation between the ideological construction and authority. The difficulty is that a position of real authority cannot be entirely rooted in a pristine ideological order, since as we saw in the previous section, this removes the actor from the world where his authority is to be exercised. He must at once be part of the
40
41
Maurice Bloch and Jonathan Parry
ideal world where death is replaced by eternal fertility and part of this world where death and time remain. As a result he has to keep a foot in both camps. Viewed in another light, the intellectual problem of the limits of ideology, discussed above, is also a political problem of authority. The challenge is one of achieving a workable balance between the ideological construction and the reality of death, duration, exchange and power.
It is not surprising, then, that this legitimating function is necessarily entirely hidden from the actors themselves. At certain points in history the political significance of such practices may become transparent -as, for example, during the Maoist period in China, when on the other side of the border a systematic propaganda campaign was launched against the kind of mortuary system which Watson describes for the Cantonese of Hong Kong. Similarly, in the Andean example it is evident that the Spanish conquerors well appreciated the significance of the Laymi graveyard in the ideological reproduction of the autonomous local community and therefore attempted to destroy it.
There are cultures, however, in which the handling of death is not put to work as a device for the creation of ideology and political domination. The four hunter-gatherer societies discussed by Wood-burn are cases in point. In none of these instances is there any systematic attempt to transform death into a rebirth or a regeneration of either the group or the cosmos. 'When you're dead that's an end of you', say the Baka. Indeed, as Woodburn points out, the force of the analogy between death and rebirth is missing when you not only enter and leave the world naked, but remain naked while in it; where there is no transcendental authority to be created the dead can be left alone.
NOTES
We would like to thank the following for comments on an earlier draft: C.J. Fuller, R.L. Stirrat and R. Thomas.
1 A fascinating detail is that the power of the bones is transmitted through the medium of the flesh of a roasted pig. Though it is going beyond the evidence available to us, it is tempting to see in this an element of the symbolic necrophagy which is suggested by the comparison with the symbolism of Melpa pig prestations discussed by Strathern.
2 The widow's (or widower's) family must pay an arm-shell to the shsk representatives who dig the grave. The arm-shell is an essential brideprice payment, and a case is cited in which a nubile daughter was given in lieu (Fortune, 1963:194). From this we may infer that the payment is a kind of replacement for the deceased. It is almost as if the affines were acknowledging their complicity In the death.
42
Introduction
; REFERENCES
Bachofen, J.J. 1967. Myth, religion and mother right, (trans. E. Mannheim) London: Routledge and Kegan Paul. ' Barth, F. 1975. Ritual and knowledge among the Baktaman of New Guinea. Yale: Yale
University Press.
i Bloch, M. 1981. 'Tombs and states', in Mortality and immortality: the anthropology
and archaeology of death, ed. S.C. Humphreys and H. King. London: Academic Press.
: Douglass, W.A. 1969. Death in Murelaga: funerary rituals in a Spanish Basque
village. Seattle: University of Washington Press. ' Dürkheim, E. 1952. Suicide. London: Routledge and Kegan Paul.
Eliade, M. 1965. The myth of the eternal return. Princeton: University Press
(Bollingen Series XLVI). Evans-Pritchard, E.E. 1948. The divine kingship of the Shilluk ofthe Nilotic Sudan. i The Frazer Memorial Lecture of 1948. Cambridge: Cambridge Unniversity
Press.
Firth, R. 1936. We, the Tikopia: a sociological study of kinship in primitive Polynesia. London: George Allen & Unwin Ltd. i Fortune, R. 1963. Sorcerers of Dobu: the social anthropology of the Dobu Islandersof
the Western Pacific. London: Routledge and Kegan Paul. ; Frankfort, H. 1948. Kingshipand thegods. Chicago: University of Chicago Press.
Frazer, J.G. 1890. The golden bough. London: MacMillan. Freedman, M. 1966. Chinese lineage and society. London: The Athlone Press. 1967. 'Ancestor worship: two facets of the Chinese case', in Social organiza-* Hon: essays presented to Raymond Firth. London: Frank Cass & Co. Ltd.
I Gillison, G. 1980. 'Images of nature in Gimi thought', in Nature, culture and
! gender, eds C. MacCormack & M. Strathern. Cambridge: Cambridge
University Press.
•"; Goody, J. 1962. Death, property and the ancestors. London: Tavistock.
Harrison, J. 1912. Themis. Cambridge: Cambridge University Press, \, Hertz, R. 1960. 'A contribution to the study of the collective representation of
death', in Death and the right hand, (trans. R. & C. Needham) London:
Cohen & West.
t Huizinga, J. 1965. The waning of the Middle Ages. Harmondsworth: Penguin
j Books.
i Huntington, R. 1973. 'Death and the social order: Bara funeral customs
i (Madagascar)'. African Studies, 32(2):65-84.
J Huntington, R. & Metcalf, P. 1979. Celebrations of death: the anthropology of
i mortuary ritual. Cambridge: Cambridge University Press,
j Leach,E.R. 1961. 'Two essays concerning the symbolic representation of time',
I in Rethinking anthropology. London: Athlone Press,
i 1966a. 'Anthropological aspects of language: animal categories and verbal
abuse', in New directions in the study of language, ed. E.H. Lenneberg. ! Cambridge, Massachusetts: The M.I.T. Press.
[ 1966b. 'Virgin birth', in Proceedings of the Royal Anthropological Institute of
Great Britain and Ireland. f Levi-Strauss, C. 1969. The elementary structures of kinship and marriage, (trans,
j JH. Bell, J.R. von Sturmer & R. Needham) London: Eyre and Spottis-
I woode.
Maurice Bloch and Jonathan Parry
Lienhardt, G. 1961. Divinity and experience: the religion of the Dinka. Oxford:
Oxford University Press. Malinowski, 13,1948. 'Baloma; the spirits of dead', in Magic, science and religion.
London: Faber and West. Morin, E. 1970. VHommc et la mort. Paris: Edition du Seuil. Ngubane, H. 1976. 'Some notions of "purity" and "impurity" among the
Zulu', Africa, 46:274-83. Parry, J.P. 1980. 'Ghosts, greed and sin: the occupational identity of the
Benares funeral priests', Man (n.s.) 15(1):88-111. 1981. 'Death and cosmogony in Kashi', Contributions to Indian Sociology,
15:337-65.
Paulme, D. 1940. Organisation sociale des Dogons, (institut de droit compare:
Etudes de sociologie et d'ethnologie juridique, 32). Pina-Cabral, J. de 1980. 'Cults of death in Northwestern Portugal', jouurnal of
the Anthropological Society of Oxford, 9(1):1-14. Raglan, Lord 1945. Death and rebirth. London: Watts & Co. Thomas, L-V. 1975. Anthropologic de la inort. Paris: Payot. Van Gennep, A. 1909. Ics rites de passage. Paris: Emile Nourry. Wilson, G. 1939. 'Nyakyusa conventions of burial', Bantu Studies 13:1-31.
44
2    The dead and the devils among the Bolivian Laymi
OLIVIA HARRIS
Introduction
... the dead go directly to Puquinapampa and Corapona. There they meet together and it is said that there they enjoy much feasting and conversation between the dead men and the dead women; and that whenn they leave there they go to another place where they endure much work, hunger, thirst and cold, and when it is hot the heat is too great; and thus they bury them with their food and drink. And they always take care to send them provisions to eat and drink; and after six months they make another similar feast for the dead, and after a year another; but they do not take out the said deceased in a procession as they do in Chinehaysuyu, they leave him inside his cave and underground chamber and they call the town of the dead amayan marcapa (town of the ghosts). (Waman Puma, 1613 (1936:294)
Waman Puma here relates the burial customs of Qullasuyu, which included what is today highland Bolivia, and formed the Southeastern quarter of the fourfold Inka state. His description, though separated by 350 years from the present day, offers illumination for the account that follows of mortuary rites in the Laymi ethnic group of northern Potosi, Bolivia. The Andean writer expresses a poignant contrast between the time of feasting and 'conversation' and the subsequent hard labour and suffering experienced by the dead. While he does not connect this polarity directly with calendrical rites, he does note that ceremonies for the dead are performed twice a year; the divided experience of the deceased is echoed in a temporal arrangement that divides the year into two halves. Today too the Laymi divide their year into two contrasted halves, each marked by a feast of the dead; one half is a time for sorrow and hard work, while the other is dedicated to feasting, pleasure and rest from their labours.
In other respects what we know of the mortuary practices of pristine Andean cultures stands in stark opposition to the present; the break was effected historically through the 'extirpation of idolatrous practices', the veritable military campaign waged against autochthonous Andean religion by the Christian priests and their heirs in the first century of European imperial rule.1 Many of the sixteenth-century
45