6

TEST CASES

At a stage in her life when she had cut herself free from social ties, Simone Weil (1951) reproved the ancient Israelites for their adherence to rule, for the legalism of the priests and their rejection of the Dionysian mystery cults. But it was all very well for her to make the judgement. It would be impossible for the leaders of an occupied but still resisting nation to adopt an effervescent form of religion. To expect them to stop preaching a stern sexual morality, vigilant control of bodily boundaries, and a corresponding religious cult would be asking them to give up the political struggle. So long as they were set upon that, the choice was no more open to them to worship God with wine, song and dance than it was open to Simone Weil herself to make a strict religious commitment when she had relaxed her own commitment to society.

The principle of symbolic replication of a social state leads us to a different view of revivalist movements among the poor and oppressed. Compensation theory too glibly explains by reference to physical suffering and deprivation of civil and economic rights. Then it finds this explanation does not cover the throng of well-to-do women who so often predominate in these movements. So it turns to sexual frustration. In the following passage Norman Cohn adopts a simple Freudian approach to explain the female element in millenarian movements:

. . . emotional frustration in women of means and leisure but without social function or prestige. Throughout the history of Christianity this circumstance has contributed to the rise of revivalist movements and it still does so today. What ideal such a movement sets itself seems to depend chiefly on personal factors – in the first place on the particular personality of the prophet, which will appeal only to certain types of women. The antinomian and erotic millenarism of the Brethren of the Free Spirit does however indicate one recurrent possibility.... Do comparable movements occur in societies where sexual life is less guilt-ridden than it has usually been in Christendom?

(Cohn, 1962: 41)

The answer is yes: the tendency to celebrate sexual promiscuity is not a response to repression; it is more likely to be found where repression is least in evidence. Cohn also mentions as alternative causes, political oppression and sudden catastrophe (see Chapter 9 below).

My explanation of effervescent religious revivalism is tidier since it uses only one hypothesis to predict its occurrence both among the poor and deprived and among the females of the rich and privileged classes, and also its likely development following catastrophe. In all cases, it is the lack of strong social articulation, the slackening of group and grid which leads people to seek, in the slackening of bodily control, appropriate forms of expression. This is how the fringes of society express their marginality. It is enough to say that the experience of a certain kind of structuring of society gets expressed in a certain way, without invoking the emotionally distracting principle of deprivation. In relation to established authority, his area of the social structure is the wilderness from which prophets and new cults are observed to arise. The collection of essays on Spirit Mediumship and Society in Africa (ed. Beattie and Middleton, 1969) is a rich mine of examples. Robin Horton makes this very point in his contribution on the main varieties of spirit possession in Kalabari religion. He contrasts possession by big spirits and minor spirits.

Possession by the big spirits is subject to a public scrutiny and control which discourages any departure from a traditionally-prescribed content. Its principal significance is that it reminds people, sometimes at recurrent intervals and sometimes at times of communal crisis, of the presence and attributes of the spirits, and of the values they support. Its general setting leaves little room for individual innovation on the part of the medium. Possession by the minor water-spirits, on the other hand, is virtually free from public scrutiny and control. Although it too serves to remind people of the presence and attributes of the spirits, it has a much wider range of significance. Thus it provides a means of personal adjustment for those whose ascribed position in society is excessively irksome to them. It also provides the occasion for both narrative and dramatic art. Finally, it provides a means for the propagation of new ideas about the world. In all these contexts, the impress of individual innovation is very evident.

These differences are, I think, fairly readily explained. First of all, doctrines relating to the founding heroes and the big local water-spirits occupy a crucial place in the community's world view. They both interpret, validate and indicate means for the perpetuation of the established order of society and ecology. Any change in such doctrines is potentially a grave threat to the established order of things. Hence there is continual public scrutiny to insure that no such change takes place. Since possession by the big spirits gives dramatic reminders and illustrations of these doctrines, it must be included in the scrutiny.

Doctrines relating to the minor water-spirits are, by contrast, rather marginal to the community's world-view. Collectively, it is true, these minor spirits have an important part to play in the explanation of the peculiarities of individual life-courses. But no one spirit is involved with more than a few individuals. Again, these minor spirits are by definition the owners of distant creeks with which the community is not practically concerned. Particular spirits of this class, then, are important neither to the community as a whole nor to any considerable section of it. Hence doctrines concerning them are free from the scrutiny and control applied to doctrines concerning the big spirits. And this freedom extends to possession by them.

One consequence is that possession by the minor water-spirits comes readily under the influence of a variety of desires and needs other than those concerned with the explanation, prediction and control of the world. Thus it becomes caught up in the struggle to find a way round uncongenial ascriptions of status, and again in the struggle to elaborate forms of both narrative and dramatic art. Another consequence is that the individual ‘carrier’ is free to make his personal contribution to the content of possession: a freedom which has been amply exploited in both of these contexts.

This freedom also makes possession by the minor water-spirits a promising channel for innovations in belief and doctrine which may eventually come to assume importance in the community at large. Since the utterances of these spirits on the heads of their carriers are virtually free from public scrutiny and control, they can serve as vehicles of new ideas which would be scotched at birth if they came from one of the big spirits. Remember how, during the time of the first Christian conversions, it was a minor water-spirit that went about telling people to join the churches since the day of the oru was over.

Through the same freedom, it is even possible that these marginal spirits may provide the material for renewing and readapting the very core of the community's world view. We don't have much evidence on this; but one case is suggestive. This is the case of the spirit carried by one of the two men in our sample of people possessed by ‘women's oru’. This spirit was first announced as the owner of a distant creek, far away from the community's own sphere of interest and operations. Later, it came to announce itself as a controller of local waters who acted together with the established water-spirit Duminea. During a visit by a Shell prospecting party, it assumed responsibility for the oil resources of the neighbouring creeks; and when oil was found, the community gave it the credit for the discovery. For some time now, the community has been on the brink of treating it as an object of public cult.

With this case in mind, we may look again at some of the myths which tell how village heroes originally came out of the world of the water-people to live with men. I have already offered an intellectualist interpretation of such myths. But it would seem possible to supplement this with a more historical (though highly speculative) interpretation. It is that the heroes, and other big spirits, were originally introduced to the community as minor water-spirits on the heads of oru kuro people; and that they stayed incubating on the sidelines until, at some time of social upheaval and change requiring new interpretative concepts, they came out to make grander claims for themselves. Elsewhere, I have described how in former times Kalabari got rid of spirits who seemed to have no further usefulness to the community. We may have a clue here as to how they got themselves new spirits to meet new challenges to their way of life.

(Horton, 1969: 45–7)

In his Malinowski lecture on Spirit Possession (1966) Ioan Lewis applies the useful distinction between the main morality cult and peripheral cults: he finds that people who are peripheral to the central focus of power and authority tend to be possessed by spirits who are peripheral to the main pantheon and whose morality is dubious. So the allocation of spiritual powers reflects the location of people along a dimension from the centre to the margins. So women subject to their husbands, serfs subject to their masters, indeed any in a state of subjection, constitute his category of the peripheral. For lack of a hypothesis about why these people should incline towards cults of bodily dissociation, the argument insidiously slides towards deprivation as the explanation and means of recognizing peripheral possession cults. But what about the Bog Irish of London? The argument is unable to deal with the many cases of people who are obviously and consciously deprived, and yet do not react in the predicted way.

It is no accident that women so often form the main membership of possession cults. The social division of labour involves women less deeply than their menfolk in the central institutions – political, legal, administrative, etc. – of their society. They are indeed subject to control. But the range of controls they experience is simpler, less varied. Mediated through fewer human contacts, their social responsibilities are more confined to the domestic range. The decisions they take do not have repercussions on a very wide range of institutions. The web of their social life, though it may tie them down effectively enough, is of a looser texture. Their social relations certainly carry less weighty pressure than those which are also institutional in range. This is the social condition they share with slaves and serfs. Their place in the public structure of roles is clearly defined in relation to one or two points of reference, say in relation to husbands and fathers. As for the rest of their social life, it takes place at the relatively unstructured, interpersonal level, with other women in the case of women, with other slaves and serfs in the case of slaves and serfs. Of course I would be wrong to say that the network of relations a woman has with others of her sex is unstructured. A delicate patterning certainly prevails. But its significance for society at large is less than the significance of men's relations with one another in the public role system. A quarrel between women has not anything like the same repercussions as a quarrel between their husbands. If they want to give their social relations with one another a more central structuring, they can only do so by embroiling their menfolk. Their links with one another are only as strong as the links between the menfolk to whom they are attached. Women, serfs and slaves (especially released slaves), are inevitably pinned only weakly into the central structure of their society. A small setback can harm them more irrevocably than those whose more complex links give a better chance of recovery. Their options are few. They experience strong grid. Therefore they are susceptible to religious movements which celebrate this experience. Unlike those who have internalized the classifications of society and who accept its pressures as aids to realizing the meanings they afford, these classes are peripheral. They express their spiritual independence in the predicted way, by shaggier, more bizarre appearance, and more ready abandonment of control.

I do not wish to embark on the difference between male and female dress in expressing this difference. For it is complicated by sexual functions. Instead, consider the distinctive appearance of prophets. They tend to arise in peripheral areas of society, and prophets tend to be shaggy, unkempt individuals. They express in their bodies the independence of social norms which their peripheral origins inspire in them. It is no accident that St John the Baptist lived in the desert and wore skins, or that Nuer prophets wear beards and long hair in a fashion that ordinary Nuer find displeasing. Everywhere, social peripherality has the same physical forms of expression, bizarre and untrimmed.

It is necessary all the time to remind ourselves that we are only dealing with distant ages and remote places in order to understand ourselves. The ceteris paribus rule allows me to use a more local example. In noticing that Lloyd George wore his hair long and loose it is relevant only in that it was longer and looser than other members of his cabinet, not that it was longer than Roundheads or shorter than Royalists wore it two centuries earlier. It becomes extremely interesting that the verdict of two contemporaries on this long-haired premier ranges him clearly with other peripheral prophets. It is argued that he would never have come to power but for the national chaos of mid-war in 1916 and that only recurring crises of great magnitude kept him in power until 1922. Even within the political scene, he owed his promotion to a ‘revolt of the cabin-boys’ (Taylor, 1970: 189), a phrase which suggests a desperate abdication of reason and control to someone from the margins. He was the only politician who retired far richer than he began. The irregularity of his personal life is no secret. Not for him the main morality cult. When he made a speech he trembled and poured with sweat as if the divine afflatus was on him. Keynes, who saw him at the Peace Conference of 1919, wrote:

How can I convey to the reader, who does not know him, any just impression of this extraordinary figure of our time, this syren, this goat-footed bard, this half-human visitor to our age from the hag-ridden magic and enchanted woods of Celtic antiquity? One catches in his company that flavour of final purposelessness, inner responsibility, existence outside or away from our Saxon good and evil, mixed with cunning, remorselessness, love of power, that lend fascination, enthralment and terror to the fair-seeming magicians of North European folklore.

(Keynes, 1933: 36–7)

Here is a stirring description of a peripheral prophet-leader which corresponds closely to the ethnographic examples in respect of appearance, morality and social provenance. Having illustrated it I shall now attempt to test my hypothesis more rigorously. It requires that peoples sharing a common culture who differ radically in social organization along the specified dimensions shall show the predicted variations in religious behaviour. If I concentrate on variations in terms of the original diagram, I will try to contrast different possible patterns across the top right-hand quadrant. I will seek to contrast weaker and stronger control by grid and group. As an illustration I will compare the Nuer and Dinka, two neighbouring Nilotic tribes who are radically different in the crucial aspects of their religious behaviour. For the Western Dinka, the state of trance is treated positively as a central cult, the source of blessing and strength. For Nuer, trance is dangerous. I shall look for objectively definable social variables to account for the difference.

Those neighbouring tribes of pastoralists who raid one another, capture and enslave one another and speak related Nilotic languages are both organized on the basis of agnatic lineages. The Nuer have been observed in a number of short visits by Evans-Pritchard and their published record is now very full. Before he wrote a large volume on Nuer religion, he produced monographs describing their ecology, political institutions and kinship and marriage, and while this work was in production, he also published numerous short articles on diverse aspects of their life. One supposes that the problem his pupil faced as ethnographer of the Dinka was delicate and complex. First, it could be tedious to recite at length the parallel chronicle of the Dinka, noting both where their practice conformed to and diverged from Nuer practice. Second, such a programme would not allow for the different perspective of a younger man, seeing the culture from the shoulders, as it were, of his teacher. This perspective might light upon different aspects, it might be more profound, and still be perfectly compatible with the earlier work on the Nuer. On the other hand, discoveries about how Dinka institutions worked might produce an embarrassing situation in which one or other or both of the two friends might find that his work discredited the findings of the other. To such a complex of delicacies we can attribute the various emphases and omissions in Godfrey Lienhardt's Dinka studies.

As far as ecological pressures and political institutions are concerned, he has adopted the alternative of summarizing very comprehensively the differences between the two peoples (1958). As far as family structure, incest, exogamy, marriage are concerned, he adopted the solution of omission. As far as religion is concerned, he adopted his own original perspective, taking the interpretation of symbolic behaviour to profound, new levels. My puzzle about whether Nuer and Dinka social experience is different in ways which would account for the difference in their symbolic order is therefore complicated by lack of detailed information about marriage regulations. From the pattern of the rest of their social and symbolic behaviour I will hazard a guess that Dinka observe less stringent, less far-reaching rules of incest and exogamy and are less consistent in using the pattern of their cattle transfers to define their categories of permitted and prohibited sexual relations.

But first let me fill in the contrast at the symbolic level, in what concerns their attitudes to spirit possession. The Nuer regard it as dangerous. Evans-Pritchard says, ‘Seizure of a man by a spirit may be temporary or permanent. When it is permanent the possessed person becomes a prophet . . .’ (1940a). Sickness is often attributed to temporary spirit possession: sacrifice is made to the spirit to appease it, and the victim, once cured, must continue to make sacrifices ‘to let it know that he has not forgotten it. Otherwise it may cause trouble again.’ The process of cure, apart from sacrifice, includes a séance of singing, drumming, rattling and clapping until someone close to the victim becomes possessed. In the case he describes it is the sick man's father who becomes a medium to make known the demands of the spirit. I quote:

As the singing, rattling and clapping continued, Rainen began to twitch and shake from head to foot and then suddenly leaped into the air and fell back on the floor of the hut where he lay stiff as though in a paroxysm. After lying tense and prostrate for a while he sat up, but shortly afterwards collapsed again. Then for about a quarter of an hour he threw himself wildly about the hut, writhing and twitching as though in agony. He reminded me of a hen which has had its throat cut in the Muslim way, and is thrown on the ground to die. If the people around had not broken most of his falls he might have injured himself. As it was, he complained to me on the following day of the soreness of his arms and legs. Every now and again he barked like a dog. In describing these spasms Nuer say that the spirit wrestles with the man it possesses.

(Evans-Pritchard, 1956: 36)

Note that the occasion of a spirit possessing a man is first known by the illness of the latter: in the next stage it possesses someone else with whom it wrestles violently, before it discloses its name and demands gifts. Then after some hard bargaining, the spirit is persuaded to go. The séance is in the charge of a prophet, one whose state of possession is permanent. And note the strange picture of a Nuer prophet (ibid., facing p. 306) which shows the unkempt hair and beard ‘both of which are objectionable to ordinary Nuer’. Their prophets stand outside the structuring of normal Nuer society. A prophet who is inspired by a spirit must give it a name

which distinguishes it as his particular spirit from the spirits of other prophets of his neighbourhood who are his rivals for renown and influence; for the attachment here is to individuals who built it up through a personal following and not, at least primarily, to social groups.

(ibid.: 117)

To this extent they sustain my thesis that bodily control tends to be relaxed where social control is weak. The Nuer prophets look strange and unkempt; they operate outside the normal social structure, competing with one another for influence in a social dimension peculiar to themselves and distinct from the balanced opposition of lineage segments. By normal Nuer values, prophets have different moral standards. They are greedy, grasping and eccentric. They cure sickness caused by spirits and state their price in terms of cattle to be dedicated to spirits. Spirit-dedicated cattle represent transfers which restrict the flow of wealth through the secular channels of marriage payments and compensations. In fact, though individual prophets do well enough by their vocation, spirits are distinctly a nuisance to the general run of Nuer who would like their help but would also like them to stay away.

To sum up, using Ioan Lewis's terms, spirit possession is not part of the main morality cult of the Nuer but a peripheral cult. Whereas, among the Dinka, a benign form of spirit-possession lies at the centre of their religion and has none of these attributes.

Dinka spear-master clans and warrior clans are linked in descent and politics, the former specialized in ritual and the latter in political leadership, producing a balanced duality of power (Lienhardt, 1958: 118–19).

Leadership, in any Dinka political segment, necessarily involves the presence of two different classes or categories of clan, the warriors and the spear-masters, which are of equal and complementary status.... The Dinka masters of the fishing-spear are not merely ritual agents for composing feuds when the parties to them both wish it; they do not, as is usual for the Nuer leopard-skin chiefs, stand outside the structure of agnatic groups with which the political groups are identified in their relations with one another...

(ibid.: 130–1)

Needless to say there is no question of spear-masters being shaggy, immoral, greedy, strange or grasping. They do not operate outside the social structure, but as a normal part of it.

As to their idea of spirit, Dinka in their rituals are much readier to expect good to come from abandoning themselves to it completely. Each clan has its own divinities, but the spear-master clans collectively recognize the divinity Flesh, the word which comes from the twitching of the flesh after a sacrificial beast is skinned which makes it appear to have a life of its own (Lienhardt, 1961: 136–7). The divinity Flesh is specially manifest in the quivering of legs and thighs at the onset of possession. The divinity Flesh ensures that the man who speaks by virtue of it speaks absolutely truthfully. Flesh means righteousness and justice. For my general comparison, it is worth noting that Dinka who are possessed by other, lesser divinities, become ‘hysterically’ possessed and are in a state of danger (ibid.: 57ff. and 137), whereas the trembling of the spear-masters at sacrificial rites is always more controlled and safe. Thus the Dinka have two degrees of bodily uncontrol, associated appropriately with the centre and periphery of their religion, the centre and periphery of their social categories.

The Nuer attitude to possession is that it is dangerous in the first phase, and produces an abnormal, specialized role in the second phase; a role whose specialized task is to counteract the dangers of first phase possession. (The distinction between these phases has been developed by Ioan Lewis in a paper on witchcraft and spirit possession.) The Dinka attitude is that trance is the primary manifestation of unspecialized benign power. It is not restricted to a specialized role in the sense of calling for special initiation, by affliction, asceticism or training, but is open to all the adult males of a clan, and normally experienced by them all.

I believe I have summarized correctly a different religious bias in the two tribal cultures. By contrasting Nuer prophets with Dinka spear-masters I am able to emphasize a different value placed on bodily dissociation. The Nuer prophet has a special function in war, to compose hymns and rouse the fighting men. His role is not parallel to that of the Dinka spear-masters, whose clans are specialized for religion in juxtaposition with the warrior clans. The true parallel of the Dinka spear-master among the Nuer is the leopard-skin priest. In certain Nuer lineages a ‘priestly virtue’ is transmitted to their members (Evans-Pritchard, 1956: 292–3), which is an effective power to curse and bless and to perform sacrifice for certain occasions. Exactly the same word is used in each language to describe the source of this power, ring, which means flesh. It would seem that, at some stage in their development, the Nuer priestly lineages were also juxtaposed symmetrically with secular lineages in a similar pattern to the Dinka one (ibid.: 293). Among the Dinka, Flesh is the chief divinity, whose cult we have described. It is a cult of physical dissociation and the attributes of flesh are etherialized into intellectual and moral qualities. Nuer priests invoke ‘the spirit of our flesh . . . which refers to the spiritual source of sacerdotal power’ (ibid.: 109). Even if it were to transpire that the Nuer concept of flesh closely parallels the Dinka one, and this in spite of the many connotations of ring, flesh, with the physical as compared with other aspects of life (ibid.: 55, 154, 159), the much greater prominence and centrality of priestly possession as the channel of divine power for the Dinka would still distinguish theirs from Nuer religion. However great the difficulties of assessment, the question still remains that two neighbouring peoples, with two related languages, related histories and distinct political institutions, frequently at war with one another, have made different emphases in their use of the bodily mode.

The Nuer cosmos seems to be more rational and regulative. The connection they make between sickness and sin is so close that, though they believe in evil eye and in the work of fetishes and ghosts, they ‘generally appear to feel that suffering is due to some fault of theirs’ (ibid.: 21, 22 and 176). Moral faults, inherent in man's nature, tend to accumulate and predispose to disaster (ibid.: 193). Though they believe in luck, it does not intervene as an explanation of misfortune (ibid.: 195). The caprice element ranks low in their cosmological ideas. There are no inexplicable shocks or surprises in their universe. At death they mourn quietly, but not ostentatiously, since this might give the impression that God had not a perfect right to do whatever he wishes with them. Their general outlook on humankind and its fate is pessimistic. Though he writes about sacrifices to avert natural disasters, Evans-Pritchard says the Nuer are not very interested or hopeful about the efficacy of rites to change their fortune in hunting, agriculture or the seasons. ‘Nuer rather turn their eyes inwards, to the little closed social world in which they live, they and their cattle. Their sacrifices are concerned with moral and spiritual, not natural crises’ (ibid.: 200). Nuer are much more concerned than Dinka with automatic pollution. They recognize classes of offences which entail automatic misfortune. Incest is one such class; adultery, disrespect to in-laws and homicide are others. To each class of offence is imputed a particular class of illness. Incest produces skin disease; adultery afflicts the injured husband with pains in the lumbar region; and so on. A wider range of misfortunes may befall the man who uncovers his genitals in the presence of his in-laws or who drinks water after shedding human blood before purification. Most of the Nuer sacrifices are made to atone for one or other such offence, assumed to be the cause of illness. Here is an area of taboo-mindedness in Nuer religion which is at variance with their general disinclination for fetishes, charms and spells. But in these two areas of their life social constraints are strongly felt, marriage rules and local loyalty in fighting. Adultery, marriage and homicide are the main occasions for the transfer of cattle and the Nuer have less cattle than the Dinka. These transgressions held to be automatically dangerous express social relations in which Nuer most strongly feel the constraints of living in society.

By contrast, Dinka seem much less pollution-prone or taboo-minded. There is less emphasis in Lienhardt's book on Dinka religion on the piacular element in sacrifices. This could be a difference of focus in the observer. But I think not. Certainly the Dinka seem to have a more optimistic world view. They do not expect their universe to be rational: ‘The Dinka are in a universe which is largely beyond their control and where events may contradict the most reasonable human expectation’ (Lienhardt, 1961: 54). They seem much less sin-conscious. An element of caprice is clearly linked with misfortune in their concept of the intriguing divinity, Macardit. He represents

the final explanation of sufferings and misfortunes which cannot be traced to other causes more consonant with Dinka notions of Divinity as just.... Macardit presides over the ending of good things, the inevitable and sometimes brutal curtailment of human life and fertility . . . a malign divinity specially associated with women

(ibid.: 81–3)

So misfortune is not regularly traced to human faults. Is their notion of death less philosophical and passive than that of the Nuer? What does it signify that they cannot bear to speak of it, and bury their dead without looking at the grave as they shovel the earth in backwards? (Lienhardt, 1962). And what different quality of their beliefs leads them to bury their most famous spear-master alive so as to enact a kind of social triumph of life over death (Lienhardt 1961: 318)? Sexual promiscuity is said to follow such a ceremony. Do the stricter Nuer have no such moments of licensed orgy in which marriage ties and incest rules are overriden? Or did their ethnographer just not happen to record them?

God, for the Nuer, is dangerous (Evans-Pritchard, 1956: 177, 195– 6, 198). They are torn by a wish to keep him at a distance and have him near at hand to help them. The Dinka god, as we have seen, comes right down to possess intimately the bodies of his worshippers. He is not distant, evidently. Is he dangerous? I maintain that we are dealing here not with the different bias of two reporters who saw similar things differently. On the contrary, their close association gave them the same bias, as near as can be achieved by conscientious observation. The differences are precisely those we would predict from small differences in ecology and social structure summed up by Lienhardt in Tribes without Rulers (1958).

First, the Dinka are about four times more numerous than Nuer; they live more densely. Then, most Dinka settlements straggle across savannah forest in continuous settlement; their sense of local bounded unit should be weaker than that of Nuer who live in discrete, wet season villages and concentrate in large cattle camps in the dry season. The pattern of Dinka transhumance has two phases of congregation, one in the dry season, their permanent settlements, the other in the height of the wet season. When

flooding reduces the area available for grazing, such small groups (of cattle camps) are drawn together and converge on the few best sites in their neighbourhood. Toward the end of the rainy season, herdsmen of each tribe are concentrated in several sub-tribal camps.

(Lienhardt, 1958: 100)

Most significantly, Dinka speak of their political communities as cattle camps, and these

are more fluid in composition and less fixed in their spatial relations to each other, than are permanent settlements.... Dinka country sets less rigid limits to movements in the wet season, and to the expansion of settlements, than are set by much of Nuer land for its people. These different ecological conditions are consistent with some differences between Nuer and Dinka political segmentation.

(ibid.: 101)

Lienhardt goes on to describe Dinka political theory which rests upon the assumption that groups expand, segment and break away from each other. ‘Fission and fusion do not take place within a single genealogical framework’ (ibid.: 104). Their political theory is one of expansion. In spite of the much higher density of population, it is a mobile social system in a sense in which the Nuer is not. ‘They see their history as a spreading out and separation of peoples on the ground . . . a notion which includes the notion of a measure of personal leadership’ (ibid.: 118). The political framework is more confused and more fluid, genealogies are more muddled (ibid.: 106) than Nuer; lineages have fewer orders of formal segmentation. From this I conclude that it would be surprising if their categories for incest and bridewealth were as strict as those of the Nuer. Dinka recognize that they are capable of less wide-scale co-operation against common enemies than Nuer (ibid.: 108). These facts suggest that it is reasonable to place the Dinka further than Nuer along the diminishing lines of grid and group towards zero. If their idea of formal sin were alleged to be more highly developed than among the Nuer, it would be surprising, since these people are related in less inclusive and more easily evaded categories. I would not expect techniques of reconciliation and religious techniques of coercion to be so well developed. I would expect Nuer to be more magical and taboo-minded than the Dinka. The differences in their social structures, in the light of my hypothesis about symbolic behaviour, seem amply to justify the differences in the reporting of their religions. If it had been the other way, and the Dinka were reported to be more conscious of sin, more interested in purification, more afraid of the dangers of spirit possession, then the regular patterns apparent elsewhere would have suggested a subjective bias. But the more positive and more central use of bodily dissociation in Dinka religion turns out to be correlated with the predicted social variables.

To complete the demonstration, I would like to include another Nilotic tribe, the Mandari. On my reading of the ethnography they would appear on a diagram showing strength of grid and group as in diagram 5.

For the Mandari, grid and group are strongest; for the Nuer, they are weaker; for the Dinka, grid and group are weaker than for the Nuer.

Diagram 5 Three Nilotes

Group, of course, refers to several possible levels of allegiance. The Mandari are acutely conscious of the widest one:

It is most important to bear in mind . . . that the small Mandari population in its tiny country is surrounded by powerful and numerous neighbours in vast territories and also that the Mandari population is itself made up of levels of immigrants dispersed around separated cores of original landowners, again of various kinds.

(Buxton, 1963b: 49)

‘Mandari country was composed of a very large number of small chiefdoms between whom relations were friendly, competitive or openly hostile’ (Buxton, 1958: 71). So, at the level of small local chief-dom, group allegiance was important for Mandari. Within each chief-dom relations were stratified in a hierarchy which attributed ritual ties with the land to early groups of immigrants and validated their claims as landowners. The rights to political priority became matters of very tense competition, as their attitude to clients as potential witches shows (Buxton, 1963b). Restraints of grid and group seem to be highly charged emotional experiences. It is gratifying to find that their attitude to sin and to purification is very formalist. To the Mandari, sins, or pollutions, are specific acts; they are not made better or worse by the intentions of the sinner. Jean Buxton died suddenly in 1971. I had earlier spoken to her about this subject and she generally con-firmed my impression of the greater magicality in the bias of Mandari culture. I recall a lively account she gave in a seminar in University College London of the complex colour and thermal categories into which Mandari class kinds of sins and kinds of illnesses. They prescribe sacrificial beasts with appropriate sex and markings for the purification of each kind of offence. But the pernickertiness of their rules drives them to desperate ritual shifts for transforming incorrectly ‘hot’ beasts, red ones or male ones, into the prescribed ‘cool’, white, black or female forms. Their highly magical, formal approach to sin is in conformity with their attitude to spirit possession which seems closely to correspond to that which I have outlined for the Nuer. Possession is dangerous, causes sickness in the first phase and produces specialists in the second phase who are adept at countering the dangers they have themselves survived (Buxton, 1968: 40). She describes an ‘elaborate specialist treatment’ for a person whose sickness has been caused by a nyok, the vengeful spirit of a dead dog.

The nyok screams ‘Ahrrr! Ahrrrr!’ and the doctor wrestles with it over the hole. He sways to and fro hanging on to the spear, while the spectators hold on to him; he foams at the mouth or blood pours from his nostrils. The nyok is said eventually to weaken and then to cease its struggles.... Treatment is in line with other exorcisms, and with convulsion therapy, where possession is induced by rhythmic rattle shaking.

(ibid.: 60)

Thus I have compared three Nilotic peoples on the basis of grid and group. The material is very suggestive. I am tempted to find that their religious behaviour upholds the hypothesis. The weaker the social constraints, the more bodily dissociation is approved and treated as a central ritual adjunct for channelling benign power to the community. The stronger the social pressures, the more magicality in ritual and in the definition of sin.

It is important to notice two aspects of the Nuer–Dinka comparison which will be relevant to any wider discussion. First, the control of grid and group is not a function of population density. The Dinka in question live at a density of up to 60 to the square mile, the Nuer at roughly seven. Grid and group are a function of order and constraint in social relations and these can be as easily absent in dense as in sparse populations. The Dinka seem to think that they can break off social relations and start afresh quite easily. Their sense of living in an expanding economy may possibly be derived from their successful cattle husbandry. It makes a great difference to the quality of social life if people are sharing resources which seem to be expanding, dwindling or static. Paul Spencer (1965) has drawn attention to this variable in his comparison of Rendille and Samburu pastoralists in Kenya. He relates the much stricter social controls operating in Rendille society to the fact that Rendille ‘believe that their camel herds, if they are growing at all, are growing at a slower pace than the human population’ while ‘the Samburu believe that their own cattle herds increase at a faster rate than their human population’ and that the poorest of men may build up substantial herds (p. 293). If the Nuer believe themselves to be husbanding a static livestock population and the Dinka an expanding one, this would be another explanation for why the latter take social control more lightly. Economic expansion and restriction turn out to be much more significant variables affecting cosmology than absolute population density as such.

I would not wish to brush off as unimportant the difficulties of interpretation which beset this argument. Any kind of illustrative material is intensely difficult to find for exactly the same reasons: no reports are exhaustive; none can avoid bias; there is an enormous subjective element of selection in any ethnographic observation. My particular kind of hypothesis depends on very close, objective assessment for its validation. Nothing will do but research which has been specially designed, not to prove it, but to test it.

Reviewing the shift from formal rite to positive approval of trance, I have referred here and there to co-varying ideas about sin. These should now be made more explicit. Along the series from maximum formality and control in symbolic behaviour to maximum informality and uncontrol there is a corresponding series in attitudes to wrong-doing. At the pole of maximum formality, the idea of wrong-doing takes no account of internal motive, or of the state of mind of the actor. Wrong-doing is bad in itself, its dangers are automatically unleashed, blame falls automatically, and the wrong is known ex opere operato. It exactly parallels the attitude to ritual in the case of extreme magicality. At the pole of maximum informality the idea of wrong-doing is entirely concerned with internal states of mind. The actual consequences of the act are of less concern than the wishes and intentions of the actor. Responsibility ends with securing right motives. To take homicide as an example, at one end of the range we have automatic pollution of blood, at the other unintended manslaughter distinguished from homicide. I would expect these variations to coincide smoothly with variations in formality and informality and both in accordance with the hold of grid and group on individuals in relevant contexts. Thus I would expect to find whole cultures where ideas of sin are more internal, less taboo-ridden than the ideas of their close neighbours, who experience more effective and all-embracing social constraints.

To show how this can be examined, let me break my methodological strictures and dare to compare widely separated peoples. The comparison of the pygmies with Hadza hunters in Tanzania is so illuminating in this matter of sin that I must discuss them together. I have earlier described the fluidity of the Mbuti pygmy camps. Their groupings are so undefined, so unimportant in their lives that no tribe in Africa seems to emphasize group membership less than they, with the exception of the Hadza. The latter move even more freely in and out of camp, forming new ones and moving away. The description of camp groupings poses a serious problem of method for their ethnographer (J. C. Wood-burn, 1964). We would expect then that they would have internalized the idea of sin even more completely than the pygmies. But this is not so. The Mbuti pygmies have no conception of pollution, neither pollution of death, nor of birth, nor of menstruation. But the Hadza fear pollution of menstrual blood. To interpret this taboo, I need to leave the question of group and return to grid. The pygmies are as free of social categories as they are of bounded groups. Neither sex, age, nor kinship order their behaviour in strictly ordained categories. Turnbull writes:

It would, of course, be ridiculous to deny that there is any system of kinship, but it is certain that the kinship system does not have the same importance as a focal point of social control as it may have in other African societies. To my mind this is undeniably linked to the ad hoc nature of the society, with its almost complete lack of concern for the past, as for the future.... The effective kinship terminology at once reflects the situation, which only becomes confused when any attempt is made to relate the terms to their usage in village society. It distinguishes generations rather than kin and cuts indiscriminately across actual kinship boundaries...

(Turnbull, 1965: 109–10)

So also with sex, they place little emphasis on separate male and female spheres. Men and women share in tasks of erecting huts and even in hunting. Social categories are markedly weak. It is the young men who operate the system of social control on behalf of the camp.

This is perfectly compatible with their general lack of concern for sin. In this kind of culture people would readily believe Mary Kingsley's missionaries, who taught that ‘a little talk with Jesus makes it all alright’. Informality is the key-note of their religious practice. I have not counted how often the words ‘intimacy’ and ‘joy’ occur in Turn-bull's account, but they are very frequent.

Even more leaderless and free in movement from camp to camp and spouse to spouse, the Hadza are divided by a social category so dominating and all-inclusive that Woodburn is tempted to describe it as a quasi-group. Wherever Hadza are, and whatever they do, they are always controlled by the division between the sexes. This division is between two hostile classes, each of which is capable of organizing itself for defence or virulent attack against the other. This extraordinarily intense consciousness of sexual difference is the only permanent level of organization the Hadza ever achieve. It is the background of male competition for wives, of female collusion between mothers and daughters to exact the maximum of trade goods from husbands in return for grudgingly given sexual satisfaction. The very low level of the division of labour between the sexes is itself an added difficulty in the way of building up a set of long-term conjugal relations. Keep in mind the insecurity of a man's hold on his wife, and their belief in menstrual pollution appears to have a practical value. When a Hadza woman menstruates, she must avoid certain activities which would be polluted by her contact. But not only must she rest. Her husband of the moment, whoever he may be, must himself abstain from manly activities lest he endanger the rest of the camp's chance of success in hunting. So his menstrual couvade is a kind of claim he affirms regularly by asserting the physiological connection between himself and his wife and the wide-spread dangers of disregarding it (Woodburn, 1964: 204–78; Douglas, 1968b).

Here we find, amid a general lack of concern for purity and danger, a strong regard for certain specific, symbolic boundaries. A symbolic expression of the tie between husband and wife (sanctioned by the threat of danger to the whole camp) reflects the one relationship which is highly valued. It expresses the one social category which is an active regulator of behaviour, the distinction of men and women, and it draws the boundary between the sexes in such a way as to incorporate the husband, in the restricted context of his conjugal claims, within the same line which encompasses his wife. So the rule, which in its general form sets all women apart from all men and treats them as dangerous, in its particular incidence sets each woman apart from other women, but not apart from her husband; he, in his turn, is, in virtue of his married state, set apart from other men. Thus the pollution rule draws very precise lines of incorporation and exclusion. No jump of the imagination is required to see this formal taboo, in an otherwise taboo-free system, as expressing the pressure of social relations. Thus I take it as a starting point for demonstrating the hypothesis that when social relations are not finely ascribed, when they are easily broken off and carry little in the way of obligation or privilege, the formal aspect of wrong-doing is disregarded. The more fluid and formless are social relations, the more internalized the idea of wrong-doing. The full demonstration requires, as I insisted earlier, the social and symbolic behaviour of the Hadza to be compared from this angle with their close neighbours and similarly for the pygmies. But until the interest of such an examination has been suggested, the work of analysis will not be carried out.

These examples throw considerable light on present theological attitudes to the subject of sin. What is taken to be a more advanced, enlightened doctrine appears merely as the usual expression of a less differentiated experience of social relations. We have here a glimpse of the sociological matrix in which ideas about sin and the self are generated. No simple evolutionary pattern emerges. It is not a history of the victory of liberal tolerance over bigoted intolerance. The relation of self to society varies with the constraints of grid and group: the stronger these are, the more developed the idea of formal transgression and its dangerous consequences, and the less regard is felt for the right of the inner self to be freely expressed. The more that social relations are differentiated by grid and group, the more the private individual is exhorted to pour his passions into prescribed channels or to control them altogether. In the small-scale primitive social system (whether we identify with high classification or with small group) a continual feedback process modifies the public pattern of roles so that no great discrepancy can arise. How the private individual sees his interests and how society at large expects him to respond will more or less coincide. All the pressures upon him are personal pressures. He is as capable of modifying them by his action as they are of controlling him. The great difference between us, in modern industrial society, and them, in small-scale primitive society, is that the feedback is lost. The pressures exerted on the individual are not modified by his reaction. There is no arguing with or explaining to the industrial system any more than there is arguing with the weather. The strongest social controls are not exerted in the personal mode.

We find ourselves generally with the other primitive societies I have identified with strong grid. They share with us the paradox that the sense of sin is weakened as social control is strengthened. Just as society demands more and more urgently that our passions flow in the channels it prescribes, we are more and more deaf to its inducements. Because of the disjunction between its classifications and our aims we hear the more insistent demand of the inner self to be given full expression.