Purity and Danger
When something is firmly classed as anomalous the outline of the set in which it is not a member is clarified. To illustrate this I quote from Sartre's essay on stickiness. Viscosity, he says, repels in its own right, as a primary experience. An infant, plunging its hands into a jar of honey, is instantly involved in contemplating the formal properties of solids and liquids and the essential relation between the subjective experiencing self and the experiencd world (1943, p. 696 seq.). The viscous is a state half-way between solid and liquid. It is like a cross-section in a process of change. It is unstable, but it does not flow. It is soft, yielding and compressible. There is no gliding on its surface. Its stickiness is a trap, it clings like a leech; it attacks the boundary between myself and it. Long columns falling off my fingers suggest my own substance flowing into the pool of stickiness. Plunging into water gives a different impression. I remain a solid, but to touch stickiness is to risk diluting myself into viscosity. Stickiness is clinging, like a too-possessive dog or mistress. In this way the first contact with stickiness enriches a child's experience. He has learnt something about himself and the properties of matter and the interrelation between self and other things.
I cannot do justice, in shortening the passage, to the marvellous reflections to which Sartre is provoked by the idea of stickiness as an aberrant fluid or a melting solid. But it makes the point that we can and do reflect with profit on our main classifications and on experiences which do not exactly fit them. In general these reflections confirm our confidence in the main classifications. Sartre argues that melting, clinging viscosity is judged an ignoble form of existence in its very first manifestations. So from these earliest tactile adventures we have always known that life does not conform to our most simple categories.
There are several ways of treating anomalies. Negatively, we can ignore, just not perceive them, or perceiving we can condemn. Positively we can deliberately confront the anomaly and try to create a new pattern of reality in which it has a place. It is not impossible for an individual to revise his own persona] scheme of classifications. But no individual lives in isolation and his scheme will have been partly received from others.
Culture, in the sense of the public, standardised values of a community, mediates the experience of individuals. It provides in advance some basic categories, a positive pattern in which
38
Secular Defilement
ideas and values are tidily ordered. And above all, it has authority, since each is induced to assent because of the assent
i        of others. But its public character makes its categories more frigid. A private person may revise his pattern of assumptions or not. It is a private matter. But cultural categories are public matters. They cannot so easily be subject to revision. Yet they
, cannot neglect the challenge of aberrant forms. Any given system of classification must give rise to anomalies, and any given culture must confront events which seem to defy its
I assumptions. It cannot ignore the anomalies which its scheme produces, except at risk of forfeiting confidence. This is why, I suggest, we find in any culture worthy of the name various provisions for dealing with ambiguous or anomalous events. L First, by settling for one or other interpretation, ambiguity is often reduced. For example, when a monstrous birth occurs, the denning lines between humans and animals may be threatened. If a monstrous birth can be labelled an event of a peculiar kind the categories can be restored. So the Nuer treat monstrous births as baby hippopotamuses, accidentally born to humans
k and, with this labelling, the appropriate action is clear. They gently lay them in the river where they belong (Evans-Pritchard, 1956, p. 84).
Second, the existence of anomaly can be physically controlled. ^ Thus in some West African tribes the rule that twins should be killed at birth eliminates a social anomaly, if it is held that two humans could not be born from the same womb at the same time. Or take night-crowing cocks. If their necks are promptly wrung, they do not live to contradict the definition of a cock as a bird that crows at dawn.
Third, a rule of avoiding anomalous things affirms and ^.ttrcngthens the definitions to which they do not conform. So where Leviticus abhors crawling things, we should see the abomination as the negative side of the pattern of things approved.
ji_Fourth, anomalous events may be labelled dangerous. Admittedly individuals sometimes feel anxiety confronted with anomaly. But it would be a mistake to treat institutions as if they evolved in the same way as a person's spontaneous reactions. Such public beliefs are more likely to be produced in the course of reducing dissonance between individual and general interpretations. Following the work of Festinger it is obvious that
39
Purity and Danger
a person when he finds his own convictions at variance with those of friends, either wavers or tries to convince the friends of their error. Attributing danger is one way of putting a subject above dispute. It also helps to enforce conformity, as we shall show below in a chapter on morals (Chapter 8).
Fifth, ambiguous symbols can be used in ritual for the same ends as they are used in poetry and mythology, to enrich meaning or to call attention to other levels of existence. We shall see in the last chapter how ritual, by using symbols of anomaly, can incorporate evil and death along with life and goodness, into a single, grand, unifying pattern.
To conclude, if uncleanness is matter out of place, we must approach it through order. Uncleanness or dirt is that which must not be included if a pattern is to be maintained. To recognise this is the first step towards insight into pollution. It involves us in no clear-cut distinction between sacred and secular. The same principle applies throughout. Furthermore, it involves no special distinction between primitives and moderns: we are all subject to the same rules. But in the primitive culture the rule of patterning works with greater force and more total comprehensiveness. With the moderns it applies to disjointed, separate areas of existence.
3
The Abominations of Leviticus
Defilement is never an isolated event. It cannot occur except in view of a systematic ordering of ideas. Hence any piecemeal interpretation of the pollution rules of another culture is hound to fail. For the only way in which pollution ideas make sense is in reference to a total structure of thought whose key-stone, boundaries, margins and internal lines are held in relation by rituals of separation.
To illustrate this I take a hoary old puzzle from biblical scholarship, the abominations of Leviticus, and particularly the dietary rules. Why should the camel, the hare and the rock badger be unclean? Why should some locusts, but not all, be unclean? Why should the frog be clean and the mouse and the hippopotamus unclean? What have chameleons, moles and crocodiles got in common that they should be listed together (Levi, xi, 27)?
To help follow the argument I first quote the relevant versions of Leviticus and Deuteronomy using the text of the New Revised Standard Translation.
Dcut. xiv 3. You shall not eat any abominable things. 4. These are the animals you may eat: the ox, the sheep, the goat, 5. the hart, the gazelle, the roe-buck, the wild goat, the ibex, the ante-\ lope and the mountain-sheep. 6. Every animal that parts the hoof and has the hoof cloven
in two, and chews the cud, among the animals you may eat. 7. Yet of those that chew the cud or have the hoof cloven you shall not eat these: The camel, the hare and the rock badger, because they chew the cud but do not part the hoof, are unclean for you. 8. And the swine, because it parts the hoof
Purity and Danger
but does not chew the cud, is chews the cud but does not
unclean for you. Their flesh part the hoof, is unclean to you.
you shall not eat, and their car- 5. And the rock badger, be-
casses you shall not touch. 9. cause it chews the cud but does
Of all that are in the waters not part the hoof, is unclean to
you may eat these: whatever you. 6. And the hare, because
has fins and scales you may eat. it chews the cud but does not
10. And whatever does not part the hoof, is unclean to have fins and scales you shall you. 7. And the swine, because not eat; it is unclean for you. it parts the hoof and is cloven-
11. You may eat all clean birds, footed but does not chew the
12. But these are the ones cud, is unclean to you. 8. Of which you shall not eat: the their flesh you shall not eat, eagle, the vulture, the osprey. and their carcasses you shall
13. the buzzard, the kite, after not touch; they are unclean to their kinds; 14. every raven you. 9. These you may eat of all after its kind; 15. the ostrich, that are in the waters. Every-the night hawk, the sea gull, thing in the waters that has the hawk, after their kinds; 16. fins and scales, whether in the the little owl and the great owl, seas or in the rivers, you may the water hen 17. and the peli- eat. 10. But anything in the can, the carrion vulture and seas or the rivers that has not the cormorant, 18. the stork, fins and scales, of the swarming the heron, after their kinds; the creatures in the waters and of hoopoe and the bat. 19. And all the living creatures that are in winged insects are unclean for the waters, is an abomination you; they shall not be eaten, to you. 11. They shall remain 20. All clean winged things you an abomination to you; of their may eat. flesh you shall not eat, and
their carcasses you shall have in
Lev. x% abomination. 12. Everything in
2. These are the living things the waters that has not fins and which you may eat among all scales is an abomination to you. the beasts that are on the earth. 13. And these you shall have
3. Whatever parts the hoof and in abomination among the is cloven-footed and chews the birds, they shall not be eaten, cud, among the animals you they are an abomination: the may eat. 4. Nevertheless among eagle, the ossifrage, the osprey, those that chew the cud or pan 14. the kite, the falcon accord-the hoof, you shall not eat ing to its kind, 15. every raven these: The camel, because it according to its kind, 16. the
4*
The Abominations of Leviticus
ostrich and the night hawk, the unclean. 27. And all that go on
sea gull, the hawk according to their paws, among the animals
its kind, 17: the owl, the cor- that go on all fours, are un-
morant, the ibis, 18. the water clean to you; whoever touches
hen, the pelican, the vulture, their carcass shall be unclean
19. the stork, the heron accord- until the evening, 28. and he
ing to its kind, the hoopoe and who carries their carcass shall
the bat. 20. All winged insects wash his clothes and be un-
that go upon all fours are an clean until the evening; they
abomination to you. 21. Yet are unclean to you. 29. And
among the winged insects that these are unclean to you among
go on all fours you may eat the   swarming   things that
those which have legs above swarm upon the earth; the
their feet, with which to leap weasel, the mouse, the great
upon the earth. 22. Of them lizard according to its kind, 30.
you may eat: the locust accord- the gecko, the land crocodile,
ing to its kind, the bald locust the lizard, the sand lizard and
according  to  its  kind,  the the chameleon. 31. These are
cricket according to its kind, unclean to you among all that
and the grasshopper according swarm; whoever touches them
to its kind. 23. But all other when they are dead shall be un-
winged insects which have four clean until the evening. 32.
feet are an abomination to you. And anything upon which any
24. And by these you shall be- of them falls when they are
come unclean; whoever touches dead shall be unclean,
their carcass shall be unclean 41. Every swarming thing that
until the evening, 25. and who- swarms upon the earth is an
ever carries any part of their abomination; it shall not be
carcass shall wash his clothes eaten. 42. Whatever goes on its
and be unclean until the even- belly, and whatever goes on all
ing. 26. Every animal which fours, or whatever has many
pans the hoof but is not cloven- feet, all the swarming things
footed or does not chew the that swarm upon the earth, you
cud is unclean to you: every- shall not eat; for they are an
one who touches them shall be abomination.
All the interpretations given so far fall into one of two groups: either the rules are meaningless, arbitrary because their intent is disciplinary and not doctrinal, or they are allegories of virtues and vices. Adopting the view that religious prescriptions are largely devoid of symbolism, Maimonides said:
43
Purity and Danger
'The Law that sacrifices should be brought is evidently of great use . . . but we cannot say why one offering should be a lamb whilst another is a ram, and why a fixed number of these should be brought. Those who trouble themselves to find a cause for any of these detailed rules arc in my eyes devoid of sense
As a mediaeval doctor of medicine, Maimonides was also disposed to believe that the dietary rules had a sound physiological basis, but we have already dismissed in the second chapter the medical approach to symbolism. Foe a modern version of the view that the dietary rules are not symbolic, but ethical, disciplinary, see Epstein's English notes to the Babylonian Talmud and also his popular history of Judaism (1959, p. 24):
'Both sets of laws have one common aim . . . Holiness^ While the positive precepts have been ordained for the cultivation of virtue, and for the promotion of those finer qualities which distinguish .the truly_rcligious and ethjcaLJtcing. the negative precepts arc defined to combat vice and suppress other evil tendencies and instincts which stand athwart roan's^striving towards holiness. . . . The negative religious laws are likewise assigned educational aims and purposes. Foremost among these is the prohibition of eating the flesh of certain animals classed as 'unclean'. This law has nothing totemic about it. It is expressly associated in Scripture with trie ideal of holiness. Its real object is to train the Israelite in self-controJ_ as thejndis-pensable first step for the attainment of holiness^
According to Professor Stein's The Dietary Laws in Rabbinic and Patristic Literature, the ethical interpretation goes back to the time of Alexander the Great and the Hellenic influence on Jewish culture. The first century a.d. letters of Aristeas teaches that not only are the Mosaic rules a valuabje^ji^ipline which 'prevents the Jews from thoughtless action and injustice', but they also coincide with what natural reason would dictate for achieving the jjpod life. So the Hellenic influence allows the medical and ethical interpretations to run together. Philo held that Moses* principle of selection was precisely to choose the most delicious meats:
The lawgiver sternly forbade all animals of land; sea or air whose flesh is the finest and fattest, like that of pigs and scale-less fish, knowing that they set a trap for the most slavish of senses, the taste, and that they produced gluttony',
44
The Abominations of Leviticus (and here we are led straight into the medical interpretation)
'an evil dangerous to both soul and body, for gluttony begets ' indigestion, which is the source of all illnesses and infirmities'.
In another stream of interpretation, following the tradition of Robertson Smith and Frazer, the Anglo-Saxon Old Testament I     scholars have tended to say simply that the rules are arbitrary I     because they are irrational. For example, Nathaniel Micklem says :
'Commentators used to give much space to a discussion of the question why such and such creatures, and such or such states and symptoms were unclean. Have we, for instance, primitive rules of hygiene? Or were certain creatures and states unclean because they represented or typified certain sins? It may be taken as certain that neither hygiene, nor any kind of typology, is the basis of uncleanness. These regulations are not by any means to be rationalised. Their origins may be diverse, and go j back beyond history . . .'
i      Compare also R. S. Driver (1895}:
' The principle, however, determining the line of demarcation
between clean animals and unclean, is not stated; and what it is lias been much debated. No single principle, embracing all the cases, seems yet to have been found, and not improbably
( more principles than one co-operated. Some animals may have
been prohibited on account of their repulsive appearance or uncleanly habits, others upon sinitary grounds; in other cases, again, the motive of the prohibition may very probably have been a religious one, particularly animals may have been supposed, like the serpent in Arabia, to be animated by superhuman or demoniac beings, or they may have had a sacramental significance in the heathen rites of other nations; and ^ the prohibition may have been intended as a protest against 'V'      these beliefs. . , .'
Z-    P. P- Saydon takes the same line in the Catholic Commentary ^    4m Holy Scripture (1953), acknowledging his debt to Driver and to Robertson Smith. It would seem that when Robertson Smith 1      applied the ideas of primitive, irrational and unexplainable to % ■   tome parts of Hebrew religion they remained thus labelled v   y and unexamined to this day.
Needless to say such interpretations are not interpretations at all, since they deny any significance to the rules. They express ' #1 45
Purity and Danger
bafflement in a learned way. Micklem says it more frankly when he says of Leviticus:
'Chapters xi to xv are perhaps the least attractive in the whole Bible. To the modern reader there is much in them that is meaningless or repulsive. They—are. concerned ^vith ritual 'u.D-cjearmgssl in respect of animals (ti) of childbirth (12), skin diseases and stained garments (13), of the rites for the purgation of skin diseases {14) of leprosy and of various issues or secretions of the human body (15). Of what interest can such subjects be except to the anthropologist? What can all this have to do with religion?' ---------
Pfciffer's general position is to be critical of the priestly and legal elements in the life of Israel. So he too lends his authority to the view that the rules in the Priestly Code are largely arbitrary:
'Only priests who were lawyers could have conceived of religion as a theocracy regulated by a divine law fixing exactly, and therefore arbitrarily, the sacred obligations of the people to their God. They thus sanctified the external, obliterated from religion both the ethical ideals of Amos and the tender emotions of Hosea, and reduced the Universal Creator to the stature of an inflexible despot. . . . From immemorial custom P derived the two fundamental notions which characterise its legislation: physical holiness and arbitrary enactment—archaic conceptions which the reforming prophets had discarded in favour of spiritual holiness and moral law.' (p. 91)
It may be true that lawyers tend to think in precise and codified forms. But is it plausible to argue that they tend to codify sheer nonsense—arbitrary enactments? Pfeiffer tries to have it both ways, insisting on the legalistic rigidity of the priestly authors and pointing to the lack of order in the netting out of the chapter to justify his view that the rules are arbitrary. Arbitrariness is a decidedly unexpected quality to find in Leviticus, as the Rev. Prof. H. J. Richards has pointed out to me. For source criticism attributes Leviticus to the Priestly source, the dominant concern of whose authors was for order. So the weight of source criticism supports -us in looking for another interpretation.
As for the idea that the rules are allegories of virtues and vices, Professor Stein derives this vigorous tradition from the
46
T
The Abominations of Leviticus
same early Alexandrian influence on Jewish thought (p. 145 seq.). Quoting the letter of Aristeas, he says that the High Priest, Eleazar:
'admits that most people find the biblical food restrictions not understandable. If God is the Creator of everything, why should His law be so severe as to exclude some animals even from touch (128 f)? His first answer still links the dietary restrictions with the danger of idolatry. . . . The second answer attempts to refute specific charges by means of allegorical exegesis. Each law about forbidden foods has its deep reason. Moses did not enumerate the mouse or the weasel out of a special consideration for them (143 f). On the contrary, mice are particularly obnoxious because of their destructiveness, and weasels, the very symbol of malicious tale-bearing, conceive through the ear and give birth through the mouth (164 f). Rather have these holy laws been given for the sake of justice to awaken in us devout thoughts and to form our character (161-168). The birds, for instance, the Jews are allowed to eat are all tame and clean, as they live by corn only. Not so the wild and carnivorous birds who fall upon lambs and goats, and even human beings. Moses, by calling the latter unclean, admonished the faithful not to do violence to the weak and not to trust their own power (145-148). Cloven-hoofed animals which part their hooves symbolise that all our actions must betray proper ethical distinction and be directed towards righteousness. . . . Chewing the cud, on the other hand stands for memory.'
Professor Stein goes on to quote Philo's use of allegory to interpret the dietary rules:
'Fish with fins and scales, admitted by the law, symbolise endurance and self-control, whilst the forbidden ones are swept away by the current, unable to resist the force of the stream. Reptiles, wriggling along by trailing their belly, signify persons who devote themselves to their ever greedy desires and passions. Creeping things, however, which have legs above their feet, so that they can leap, are clean because they symbolise the success of moral efforts/
Christian teaching has readily followed the allegorising tradition. The first century epistle of Barnabus, written to convince the Jews that their law had found its fulfilment, rook the clean and unclean animals to refer to various types of men, leprosy to mean sin, etc. A more recent example of this tradition is in
47
Purity and Danger
Bishop Challoner's notes on the Westminster Bible in the beginning of this century:
'Hoof divided and cheweth the cud. The dividing of the hoof and chewing of the cud signify discretion between good and evil, and meditating on the law of God; and where either of these is wanting, man is unclean. In like manner fishes were reputed unclean that had not fins and scales: that is souls that did not raise themselves up by prayer and cover themselves with the scales of virtue.' Footnote verse 3.
These are not so much interpretations as pious commentaries. They fail as interpretations because they are neither consistent nor comprehensive. A different explanation has to be developed for each animal and there is no end to the number of possible explanations.
Another traditional approach, also dating back to the letter of Arjsteas, is the view that what is forbidden to the Israelites is forbidden solely to protect them from foreign influence. For instance, Maimonides held that they were forbidden to seethe the kid in the milk of its dam because this was a cultic act in the religion of the Canaanites. This argument cannot be comprehensive, for it is not held that the Israelites consistently rejected all the elements of foreign religions and invented something entirely original for themselves. Maimonides accepted the view that some of the more mysterious commands of the law had as their object to make a sharp break with heathen practices. Thus the Israelites were forbidden to wear garments woven of linen and wool, to plant different trees together, to have sexual intercourse with animals, to cook meat with milk, simply because these acts figured in the rites of their heathen neighbours. So far, so good: the laws were enacted as barriers to the spread of heathen styles of ritual. But in that case why were some heathen practices allowed? And not only allowed—if sacrifice be taken as a practice common to heathens and Israelites—but given an absolutely central place in the religion. Maimonides' answer, at any rate in The Guide to the Perplexed, was to justify sacrifice as a transitional stage, regrettably heathen, but necessarily allowed because it would be impractical to wean the Israelites abruptly from their heathen past. This is an extraordinary statement to come from the pen of a rabbinical scholar, and indeed in his serious rabbinical writings Maimonides did not
48
The Abominations of Leviticus
attempt to maintain the argument: on the contrary, he there counted sacrifice as the most important act of the Jewish religion.
At least Maimonides saw the inconsistency and was led by it into contradiction. But later scholars seem content to use the foreign influence argument one way or the other, according to the mood of the moment. Professor Hooke and his colleagues have clearly established that the Israelites took over some Canaanite styles of worship, and the Canaanites obviously had much in common with Mesopotamian culture (1933). But it is no explanation to represent Israel as a sponge at one moment and as a repellent the next, without explaining why it soaked up this foreign element but repelled that one. What is the value of saying that seething kids in milk and copulating with cows are forbidden in Leviticus because they are the fertility rites of foreign neighbours (1935), since Israelites took over other foreign rites? We are still perplexed to know when the sponge is the right or the wrong metaphor. The same argument is equally puzzling in Eichrodt (pp. 230-1). Of course no culture is created out of nothing. The Israelites absorbed freely from their neighbours, but not quite freely. Some elements of foreign culture were incompatible with the principles of patterning on which they were constructing their universe; others were compatible. For instance, Zaehner suggests that the Jewish abomination of creeping things may have been taken over from Zoroastrianism (p. 162). Whatever the historical evidence for this adoption of a foreign element into Judaism, we shall see that there was in the patterning of their culture a pre-formed compatibility between this particular abomination and the general principles on which their universe was constructed.
Any interpretations will fail which take the Do-nots of the Old Testament in piecemeal fashion. The only sound approach is to forget hygiene, aesthetics, morals and instinctive revulsion, even to forget the Canaanites and the Zoroastrian Magi, and start with the texts. Since each of the injunctions is prefaced by the command to be holy, so they must be explained by that command. There must be contrariness between holiness and abomination which will make over-all sense of all the particular restrictions.
Holiness is the attribute of Godhead. Its root means 'set apart'. What else does it mean? We should start any cosmologica! enquiry by seeking the principles of power and danger. In the Old
49
Purity and Danger
Testament we find blessing as the source of alt good things, and the withdrawal of blessing as the source of all dangers. The blessing of God makes the land possible for men to live in.
Cod's work through the blessing is essentially to create order, through which men's affairs prosper. Fertility of women, livestock and fields is promised as a result of the blessing and this is to be obtained by keeping covenant with God and observing all His precepts and ceremonies (Deut. xxvm, 1-14). Where the blessing is withdrawn and the power of the curse unleashed, there is barrenness, pestilence, confusion. For Moses said:
'But if you will not obey the voice of the Lord your God or be careful to do all his commandments and his statutes which I command you to this day, then all these curses shall come upon you and overtake you. Cursed shall you be in the city, and cursed shall you be in the field. Cursed shall be your basket and your kneading trough. Cursed shall be the fruit of your body, and the fruit of your ground, the increase of your cattle, and the young of your flock. Cursed shall you be when you come in and cursed shall you be when you go out. The Lord will send upon you curses, confusion, and frustration in all that you undertake to do, until you arc destroyed and perish quickly on account of the evil of your doings, because you have forsaken me . . . The Lord will smite you with consumption, and with fever, inflammation, and fiery heat, and with drought, and with blasting and with mildew; they shall pursue you till you perish. And the heavens over your head shall be brass and the earth under vou shall be iron. The Lord will make the rain of your land powder and dust; from heaven it shall come down upon you until you are destroyed.' (Deut. xxvm, 15-24)
From this it is clear that the positive and negative precepts are held to be efficacious and not merely expressive: observing them draws down prosperity, infringing them brings danger. We are thus entitled to treat them in the same way as jve treat primitive ritual avoidances whose breach unleashes danger to men. The precepts and ceremonies alike are focussed on the idea of the holiness of God which men must create in their own lives. So this is a universe in which men prosper by conforming to holiness and perish when they deviate from it. If there were no other clues we should be able to find out the Hebrew idea of the holy by examining the precepts by which men conform to
50
The Abominations of Leviticus
(it. It is evidently not goodness in the sense of an all-embracing humane kindness. Justice and moral goodness may well illustrate holiness and form part of it, but holiness embraces other ideas as well.
Granted that its root means separateness, the next idea that emerges is of the Holy as wholeness and completeness. Much of f:     Leviticus is taken up with stating the physical perfection that is
(required of things presented in the temple and of persons approaching it. The animals offered in sacrifice must be without blemish, women must be purified after childbirth, lepers should be separated and ritually cleansed before being allowed to approach it once they are cured. All bodily discharges are defiling and disqualify from approach to the temple. Priests may only come into contact with death when their own close kin die. But the high priest must never have contact with death.
Levit. xxi
117. 'Say to Aaron, None of your descendants throughout their generations who has a blemish may approach to offer the bread of his God. 18. For no one who has a blemish shall draw near, i a man blind or lame, or one who has a mutilated face or a limb
too long. 19. or a man who has an injured foot or an injured 1 hand, to. or a hunch-back, or a dwarf, or a man with a defect
I in his sight or an itching disease or scabs, or crushed testicles;
■ ti. no man of the descendants of Aaron the priest who has a
blemish shall come near to offer the Lord's offerings by fire; ...'
j In other words, he must be perfect as a man, if he is to be a I priest.
f This much reiterated idea of physical completeness is also worked out in the social sphere and particularly in the warriors' camp. The culture of the Israelites was brought to the pitch of greatest intensity when they prayed and when they fought- The army could not win without the blessing and to keep the blessing in the camp they had to be specially holy. So the camp was to be preserved from defilement like the Temple. Here again all bodily discharges disqualified a man from entering the camp as they would disqualify a worshipper from approaching the altar. A warrior who had had an issue of the body in the night
;       should keep outside the camp all day and only return after
j sunset, having washed. Natural functions producing bodily waste were to be performed outside the camp (Deut. xxm,
%     10-1 j). In short the idea of holiness was given an external,
Purity and Danger physical expression in the wholeness of the body seen as a perfect container.
Wholeness is also extended to signify completeness in a social context. An important enterprise, once begun, must not be left incomplete. This way of lacking wholeness also disqualifies a man from fighting. Before a battle the captains shall proclaim:
Deut. xx
5. 'What man is there that has built a new house and has not dedicated it? Let him go back to his house, lest he die in the battle and another man dedicate it. 6. What man is there that has planted a vineyard and has not enjoyed its fruit? Let him go back to his house, lest he die in the battle and another man enjoy its fruit. 7. And what man is there that hath betrothed a wife and has not taken her? Let him go back to his house, lest he die in the battle and another man take her.'
Admittedly there is no suggestion that this rule implies defilement. It is not said that a man with a half-finished project on his hands is defiled in the same way that a leper is defiled. The next verse in fact goes on to say that fearful and faint-hearted men should go home lest they spread their fears. But there is a strong suggestion in other passages that a man should not put his hand to the plough and then turn back. Pedersen goes so far as to say that:
'in all these cases a man has started a new important undertaking without having finished it yet ... a new totality has come into existence. To make a breach in this prematurely, i.e. before it has attained maturity or has been finished, involves a serious risk of sin'. (Vol. Ill, p. 9)
If we follow Pedersen, then blessing and success in war required a man to be whole in body, whole-hearted and trailing no uncompleted schemes. There is an echo of this actual passage in the New Testament parable of the man who gave a great feast and whose invited guests incurred his anger by making excuses (Luke xiv, 16-24; Matt. xxn. See Black & Rowley, 1962, p. 836). One of the guests had bought a new farm, one had bought ten oxen and had not yet tried them, ahd one had married a wife. If according to the old Law each could have validly justified his refusal by reference to Deut. xx, the parable supports Pedersen's view that interruption of new projects was held to be bad in civil as well as military contexts.
52
j The Abominations of Leviticus
Other precepts develop the idea of wholeness in another direction. The metaphors of the physical body and of the new under-''       taking relate to the perfection and completeness of the individual 1       and his work. Other precepts extend holiness to species and \,      categories. Hybrids and other confusions are abominated.
Lev. xviii
I '23. And you shall not lie with any beast and defile yourself
with it, neither shall any woman give herself to a beast to lie with it: it is perversion.'
The word 'perversion' is a significant mistranslation of the rare 'jF      Hebrew word tebhel, which has as its meaning mixing or confusion. The same theme is taken up in Leviticus xix, 19.
j? 'You shall keep my statutes. You shall not let your cattle breed
if with a different kind; you shall not sow your field with two
kinds of seed; nor shall there come upon you a garment of cloth made of two kinds of stuff.'
All these injunctions are prefaced by the general command:
U 'Be holy, for I am holy.'
We can conclude that holiness is exemplified by completeness. Holiness requires that individuals shall conform to the class to l        which they belong. And holiness requires that different classes of things shall not be confused.
Another set of precepts refines on this last point. Holiness means keeping distinct the categories of creation. It therefore involves correct definition, discrimination and order. Under this head all the rules of sexual morality exemplify the holy. Incest and adultery (Lev. xvm, 6-20) are against holiness, in the simple sense of right order. Morality does not conflict with holiness, but holiness is more a matter of separating that which should be separated than of protecting the rights of husbands and brothers. Then follows in chapter xix another list of actions which are f      contrary to holiness. Developing the idea of holiness as order, not confusion, this list upholds rectitude and straight-dealing as y        holy, and contradiction and double-dealing as against holiness. lc      Theft, lying, false witness, cheating in weights and measures, all I      kinds of dissembling such as speaking ill of the deaf (and pre-L     sumably smiling to their face), hating your brother in your I|    heart (while presumably speaking kindly to him), these are
Purity and Danger clearly contradictions between what seems and what is. This chapter also says much about generosity and love, but these are positive commands, while I am concerned with negative rules.
We have now laid a good basis for approaching the laws about clean and unclean meats. To be holy is to be whole, to be one; holiness is unity, integrity, perfection of the individual and of the kind. The dietary rules merely develop the metaphor of holiness on the same lines.
First we should start with livestock, the herds of cattle, camels, sheep and goats which were the livelihood of the Israelites. These animals were clean inasmuch as contact with them did not require purification before approaching the Temple. Livestock, like the inhabited land, received the blessing of God. Both land and livestock were fertile by the blessing, both were drawn into the divine order. The farmer's duty was to preserve the blessing. For one thing, he had to preserve the order of creation. So no hybrids, as we have seen, either in the fields or in the herds or in the clothes made from wool or flax. To some extent men covenanted with their land and cattle in the same way as God covenanted with them. Men respected the first born of their cattle, obliged them to keep the Sabbath. Cattle were literally domesticated as slaves. They had to be brought into the social order in order to enjoy the blessing. The difference between cattle and the wild beasts is that the wild beasts have no covenant to protect them. It is possible that the Israelites were like other pastoralists who do not relish wild game. The Nuer of the South Sudan, for instance, apply a sanction of disapproval of a man who lives by hunting. To be driven to eating wild meat is the sign of a poor herdsman. So it would be probably wrong to think of the Israelites as longing for forbidden meats and finding the restrictions irksome. Driver is surely right in taking the rules as an a posteriori generalisation of their habits. Cloven-hoofed, cud-chewing ungulates are the model of the proper kind of food for a pastoral ist. If they must eat wild game, they can eat wild game that shares these distinctive characters and is therefore of the same general species. This is a kind of casuistry which permits scope for hunting antelope and wild goats and wild sheep. Everything would be quite straightforward were it not that the legal mind has seen fit to give ruling on some borderline cases. Some animals seem to be ruminant, such as the hare, and the hyrax (or rock badger), whose constant grinding of their
54
The Abominations of Leviticus
teeth was held to be cud-chewing. But they are definitely not cloven-hoofed and so are excluded by name. Similarly for animals which are cloven-hoofed but are not ruminant, the pig and the camel. Note that this failure to conform to the two necessary criteria for defining cattle is the only reason given in the Old Testament for avoiding the pig; nothing whatever is said about its dirty scavenging habits. As the pig does not yield milk, hide nor wool, there is no other reason for keeping it except for its flesh. And if the Israelites did not keep pig they would not be familiar with its habits. I suggest that originally the sole reason for its being counted as unclean is its failure as a wild boar to get into the antelope class, and that in this it is on the same footing as the camel and the hyrax, exactly as is stated in the book.
After these borderline cases have been dismissed, the law goes on to deal with creatures according to how they live in the three elements, the water, the air and the earth. The principles here applied are rather different from those covering the camel, the pig, the hare and the hyrax. For the latter are excepted from clean food in having one but not both of the defining characters of livestock. Birds I can say nothing about, because, as I have said, they are named and not described and the translation of the name is open to doubt. But in general the underlying principle of cleanness in animals is that they shall conform fully to their class. Thoss species are unclean which are imperfect members of their class, or whose class itself confounds the general scheme of the world.
To grasp this scheme we need to go back to Genesis and the creation. Here a three-fold classification unfolds, divided between the earth, the waters and the firmament. Leviticus takes up this scheme and allots to each element its proper kind of animal life. In the firmament two-legged fowls íly with wings. In the water scaly fish swim with fins. On the earth four-legged animals hop, jump or walk. Any class of creatures which is not equipped for the right kind of locomotion in its element is contrary to holiness. Contact with it disqualifies a person from approaching the Temple. Thus anything in the water which has not fins and scales is unclean (xi, 10-12). Nothing is said about predatory habits or of scavenging. The only sure test for cleanness in a fish is its scales and its propulsion by means of fins.
Four-footed creatures which fly (xi, 20-26) are unclean. Any
55
Purity and Danger
creature which has two legs and two hands and which goes on all fours like a quadruped is unclean (xi, 27). Then follows (v. 29) a much disputed list. On some translations, it would appear to consist precisely of creatures endowed with hands instead of front feet, which perversely use their hands for walking: the weasel, the mouse, the crocodile, the shrew, various kinds of lizards, the chameleon and mole (Danby, 1933), whose forefeet are uncannily hand-like. This feature of this list is lost in the New Revised Standard Translation which uses the word 'paws' instead of hands.
The last kind of unclean animal is that which creeps, crawls or swarms upon the earth. This form of movement is explicitly contrary to holiness (Levit. xi, 41-44). Driver and White use 'swarming' to translate the Hebrew sherec, which is applied to both those which teem in the waters and those which swarm on the ground. Whether we call it teeming, trailing, creeping, crawling or swarming, it is an indeterminate form of movement. Since the main animal categories are defined by their typical movement, 'swarming' which is not a mode of propulsion proper to any particular element, cuts across the basic classification. Swarming things are neither fish, flesh nor fowl. Eels and worms inhabit water, though not as fish; reptiles go on dry land, though not as quadrupeds; some insects fly, though not as birds. There is no order in them. Recall what the Prophecy of Habakkuk says about this form of life:
'For thou makest men like the fish of the sea, like crawling things that have no ruler.' (1, v. 14)
The prototype and model of the swarming things is the worm. As fish belong in the sea so worms belong in the realm of the grave, with death and chaos.
The case of the locusts is interesting and consistent. The test of whether it is a clean and therefore edible kind is how it moves on the earth. If it crawls it is unclean. If it hops it is clean (xi, v. 21). In the Mishnah it is noted that a frog is not listed with creeping things and conveys no uncleanness (Danby, p. 722). I suggest that the frog's hop accounts for it not being listed. If penguins lived in the Near East I would expect them to be ruled unclean as wingless birds. If the list of unclean birds could be retranslated from this point of view, it might well turn out that they are anomalous because they swim and dive as
56
The Abominations of Leviticus
well as they fly, or in some other way they are not fully birdlike.
Surely now it would be difficult to maintain that 'Be ye Holy' means no more than 'Be ye separate'. Moses wanted the children of Israel to keep the commands of God constantly before then-minds :
Deut. xi
'18. You shall therefore lay up these words of mine in your heart and in your soul; and you shall bind them as a sign upon your hand, and they shall be as frontlets between your eyes.
19. And you shall teach them to your children, talking of them when you are sitting in your house, and when you are walking by the way, and when you lie down and when you rise.
20. And you shall write them upon the doorposts of your house and upon your gates.'
If the proposed interpretation of the forbidden animals is correct, the dietary laws would have been like signs which at every turn inspired meditation on the oneness, purity and completeness of God. By rules of avoidance holiness was given a physical expression in every encounter with the animal kingdom and at every meal. Observance of the dietary rules would thus have been a meaningful part of the great liturgical act of recognition and worship which culminated in the sacrifice in the Temple.
57
4
Magic and Miracle
Once when a band of !Kung Bushmen had performed their rain rituals, a small cloud appeared on the horizon, grew and darkened. Then rain fell. But the anthropologists who asked if the Bushmen reckoned the rite had produced the rain, were laughed out of court (Marshall, 1957). How naive can we get about the beliefs of others? Old anthropological sources are full of the notion that primitive people expect rites to produce an immediate intervention in their affairs, and they poke kindly fun at those who supplement their rituals of healing with European medicine, as if this testified to lack of faith. The Dinka perform an annual ceremony to cure malaria. The ceremony is timed for the month in which it is to be expected that malaria will soon abate. A European observer who witnessed it remarked dryly that the officiant ended by urging everyone to attend the clinic regularly if they hoped to get well (Lienhardt 196.).
It is *iot difficult to trace the idea that primitives expect their rites to have external efficacy. There is a comfortable assumption in the roots of our culture that foreigners know no true spiritual religion. On this assumption Frazer's grandiose description of primitive magic took root and flourished. Magic was carefully separated from other ceremonial, as if primitive tribes were populations of Ali Babas and Aladdins, uttering their magic words and rubbing their magic lamps. The European belief in primitive magic has led to a false distinction between primitive and modern cultures, and sadly inhibited comparative religion. I do not propose to show how the term magic has been used by various scholars hitherto. Too much erudition has been
J8
Magic and Miracle
expended already on defining and naming symbolic actions which are held to be efficacious for altering the course of events (Goody, Gluckman).
On the continent magic has remained a vague literary term, described but never rigorously defined. It is clear that in the tradition of Mams' Theorie de la Magie, the word does not connote a particular class of rituals, but rather the whole corpus of ritual and belief of primitive peoples. No special focus is centred on efficacy. We owe to Frazer the isolating and hardening of the idea of magic as the efficacious symbol (see Chapter 1). Malinowski further developed the idea uncritically and gave its currency renewed life. For Malinowski magic takes its origin in the expression of an individual's emotions. Passion, as it contorted his face, and caused the magician to stamp his foot or shake his fist, also led him to enact his strong desire for gain or revenge. This physical enactment, almost involuntary at the start, a deluded wish-fulfilment, was for him the basis of the magic rite (see Nadel, p. 194). Malinowski had such original insights into the creative effect of ordinary speech that he profoundly influenced contemporary linguistics. How could he have barrenly isolated magic rite from other rites and discussed magic as a kind of poor man's whisky, used for gaining conviviality and courage against daunting odds? This is another aberration which we can lay to the door of Frazer, whose disciple he claimed to be.
Since Robertson Smith drew a parallel between Roman Catholic ritual and primitive magic, let us gratefully take the hint. For magic let us read miracle and reflect on the relation between ritual and miracle in the minds of the mass of believers in the miracle-believing ages of Christianity. There we find that the possibility of miracle was always present; it did not necessarily depend on rite, it could be expected to erupt anywhere at any time in response to virtuous need or the demands of justice. It inhered more potently in some material objects, places and persons. It could not be laid under automatic control; the saying of the right words or sprinkling of holy water could not guarantee a cure. The power of miraculous intervention was believed to exist, but there was no certain way of harnessing it. It was as different and as like Islamic Baraka or Teutonic Luck or Polynesian Mono as each is different from the other. Each primitive universe hopes to harness some such marvellous power to the
59
Purity and Danger
needs of men, and each supposes that a different set of links has to be reckoned with, as we shall see in the next chapter. In the miraculous period of our Christian heritage miracle did not only occur through enacted rites, nor were rites always performed in the expectation of miracle. It is realistic to suppose an equally loose relation holds between rite and magic effect in primitive religion. We should recognise that the possibility of magic intervention is always present in the mind of believers, that it is human and natural to hope for material benefits from the enactment of cosmic symbols. But it is wrong to treat primitive ritual as primarily concerned with producing magical effects. The priest in a primitive culture is not necessarily a magic wonder worker. This idea has barred our understanding of alien religions, but it is only a recent by-product of a more deep-rooted prejudice.
A contrast between interior will and exterior enactment goes deep into the history of Judaism and Christianity. Of its very nature any religion must swing between these two poles. There must be a move from internal to external religious life, if a new religion endure even a decade after its first revolutionary fervour. And finally, the hardening of the external crust becomes a scandal and provokes new revolutions.
So the rage of the Old Testament prophets was continually renewed against empty external forms paraded instead of humble and contrite hearts. From the time of the first Council of Jerusalem, the Apostles tried to take their stand on a spiritual interpretation of sanctity. The Sermon on the Mount was seen as the deliberate Messianic counterpart of the Mosaic law. St. Paul's frequent references to the law as part of the old dispensation, a bondage and a yoke, are too familiar to need quotation. From this time on the physiological condition of a person, whether leprous, bleeding, or crippled, should have become irrelevant to their capacity to approach the altar. The foods they ate, the things they touched, the days on which they did things, such accidental conditions should have no effect on their spiritual status. Sin was to be regarded as a matter of the will and not of external circumstance. But continually the spiritual intentions of the early Church were frustrated by spontaneous resistance to the idea that bodily states were irrelevant to ritual. The idea of pollution by blood, for example, seems to have been a long time dying, if we judge by some early Penitcntials.
60
Magic and Miracle
See the Penitential of Archbishop Theodore of Canterbury, a.d. 668-690:
'If without knowing it one eats what is polluted by blood or any unclean thing, it is nothing; but if he knows, he shall do penance according to the degree of pollution. . . .'
He also requires from women 40 days of purgation after the birth of a child, and enjoins penance of three weeks' fast on any woman, lay or religious, who enters a church, or communicates during menstruation (McNeill & Gamer).
Needless to say, these rules were not adopted as part of the Corpus of Canon Law, and now it is difficult to find instances of ritual uncleanness in Christian practice. Injunctions, which in their origin may have been concerned with removing pollution of blood, are interpreted as carrying only a symbolic spiritual significance. For example, it is usual to reconsecrate a church if blood has been shed in its precincts, but St. Thomas Aquinas explains that 'bloodshed' refers to voluntary injury leading to bloodshed, which implies sin, and that it is sin in a holy place which desecrates it, not defilement by bloodshed. Similarly, the rite for purification of a mother probably does derive ultimately from Judaic practice, but the modern Roman Ritual, which dates back to Pope Paul V (1605-21), presents the churching of women simply as an act of thanksgiving.
The long history of protestantism witnesses to the need for continual watch on the tendency of ritual form to harden and replace religious feeling. In wave upon wave the Reformation has continued to thunder against the empty encrustation of ritual. So long as Christianity has any life, it will never be time to stop echoing the parable of the Pharisee and the Publican, to stop saying that external forms can become empty and mock the truths they stand for. With every new century we become heirs to a longer and more vigorous anti-ritualist tradition.
This is right and good as far as our own religious life is concerned, but let us beware of importing uncritically a dread of dead formality in ourselves into our judgments of other religions. The Evangelical movement has left us with a tendency to suppose that any ritual is empty form, that any codifying of conduct -is alien to natural movements of sympathy, and that any external religion betrays true interior religion. From this it is a short step to assuming something about primitive religions. If they
61
Purity and Danger
are formal enough to be reported at all, they are too formal, and without interior religion. For example, Pfeifler's Books of the Old Testament has this anti-ritualist basis which leads him to contrast 'the old religion of cult' with the prophets' 'new one of conduct'. He writes as if there could be no spiritual content in the old cult (pp. 5j scq.). The religious history of Israel he presents as if the stern, insensitive lawgivers were in conflict with the prophets, and never allows that both could have been engaged in the same service, or that ritual and codification could have something to do with spirituality. According to Pfeiffcr the lawyer priests:
'sanctified the external, obliterated from religion the ethical ideals of Amos and the tender emotions of Hosea, and reduced the universal creator to the status of an inflexible despot. . , . From immemorial custom P derived the two fundamental notions which characterised its legislation: physical holiness and arbitrary enactment—archaic conceptions which the reforming prophets had discarded in favour of spiritual holiness and moral law.' (p. 91)
This is not history, but sheer anti-ritualist prejudice. For it is a mistake to suppose that there can be religion which is all interior, with no rules, no liturgy, no external signs of inward states. As with society, so with religion, external form is the condition of its existence. As the heirs of the Evangelical tradition we have been brought up to suspect formality and to look for spontaneous expressions like the Minister's sister whom Mary Webb made to say, 'Home-made cakes and home-made prayers are always best'. As a social animal, man is a ritual animal. If ritual is suppressed in one form it crops up in others, more strongly the more intense the social interaction. Without the letters of condolence, telegrams of congratulations and even occasional postcards, the friendship of a separated friend is not a social reality. It has no existence without the rites of friendship. Social rituals create a reality which would be nothing"without them. It is not too much to say that ritual is more to society than words are to thought. For jt is very possible to know something and then find words for it. But it is impossible to have social relations without symbolic acts.
We shall understand more about primitive ritual if we clarify further our ideas about secular rites. For us, individually, every-
62
Magic and Miracle
day symbolic enactment does several things. It provides a focussing mechanism, a method of mnemonics and a control for experience. To deal with focussing first, a ritual provides a frame. The marked off time or place alerts a special kind of expectancy, just as the oft-repeated 'Once upon a time' creates a mood receptive to fantastic tales. We can reflect on this framing function in small personal instances, for the least action is capable of carrying significance. Framing and boxing limit experience, shut in desired themes or shut out intruding ones. How many times is it necessary to fill a weekend case to find out how to exclude successfully all tokens of unwanted office life? One official file, packed in a weak moment, can spoil the whole effect of the holiday. I quote here Marion Milner on framing:
'. . . the frame marks off the different kind of reality that is within it from that which is outside it; but a temporal—spatial frame marks off the special kind of reality of a psycho-analytic session . . . makes possible the creative illusion called transference . . .' (1955)
She is discussing the technique of child analysis and mentions the locker in which the child patient keeps his play objects. It creates a kind of spatio-temporal frame which gives him continuity from one session to the next.
Not only does ritual aid us in selecting experiences for concentrated attention. It is also creative at the level of performance. For an external symbol can mysteriously help the co-ordination of brain and body. Actors' memoirs frequently recount cases in which a material symbol conveys effective power: the actor knows his part, he knows exactly how he wants to interpret it. But an intellectual knowing of what is to be done is not enough to produce the action. He tries continually and fails. One day some prop is passed to him, a hat or green umbrella, and with this symbol suddenly knowledge and intention are realised in the flawless performance.
The Dinka herdsman hurrying home to supper, knots a bundle of grass at the wayside, a symbol of delay. Thus he expresses outwardly his wish that the cooking may be delayed for his return. The rite holds no magic promise that he will ■now be in time for supper. He does not then dawdle home thinking that the action will itself be effective. He redoubles his haste. His action has not wasted time, for it has sharpened
63
Purity and Danger
the focus of his attention on his wish to be in time (Lienhardt). The mnemonic action of rites is very familiar. When we tie knots in handkerchiefs we are not magicking our memory, but bringing it under the control of an external sign.
So ritual focusses attention by framing; it enlivens the memory and links the present with the relevant past. In all this it aids perception. Or rather, it changes perception because it changes the selective principles. So it is not enough to say that ritual helps us to experience more vividly what we would have experienced anyway. It is not merely like the visual aid which illustrates the verbal instructions for opening cans and cases. If it were just a kind of dramatic map or diagram of what is known it would always follow experience. But in fact ritual does not play this secondary role. It can come first in formulating experience. It can permit knowledge of what would otherwise not be known at all. It does not merely externalise experience, bringing it out into the light of day, but it modifies experience in so expressing it. This is true of language. There can be thoughts which have never been put into words. Once words have been framed the thought is changed and limited by the very words selected. So the speech has created something, a thought which might not have been the same.
There are some things we cannot experience without ritual. Events which come in regular sequences acquire a meaning from relation with others in the sequence. Without the full sequence individual elements become lost, imperceivable. For example, the days of the week, with their regular succession, names and distinctiveness: apart from their practical value in identifying the divisions of time, they each have meaning as part of a pattern. Each day has its own significance and if there are habits which establish the identity of a particular day, those regular observances have the effect of ritual. Sunday is not just a rest day. It is the day before Monday, and equally for Monday in relation to Tuesday. In a true sense we cannot experience Tuesday if for some reason we have not formally noticed that we have been through Monday. Going through one part of the pattern is a necessary procedure for being aware of the next part. Air travellers find that this applies to hours of the day and the sequence of meals. These are examples of symbols which are received and interpreted without having been intended. If we admit that they condition experience, so we must admit also
64
Magic and Miracle
that intended rituals in regular sequence can have this as one of their important functions.
Now we can turn to religious rites again. Dürkheim was well aware that their effect is to create and control experience. It was his main preoccupation to study how religious ritual makes manifest to men their social selves and thus creates their society.
I But his thought was channelled into the English stream of
anthropology by Radcliffe-Brown, who modified it. Thanks to Dürkheim the primitive ritualist was no longer seen as a pantomime magician. That was a notable advance on Frazer. Furthermore, Radcliffe-Brown refused to separate religious ritual from secular ritual—another advance. Malinowski's magician became no different from any flag-waving patriot or superstitious salt-thrower, and these were treated alongside the Roman Catholic abstaining from meat or the Chinese putting rice on a grave.
1 Ritual was no more mysterious or exotic.
I In dropping both the words Sacred and Magic, Rad cliff e-
I Brown seemed to restore the thread of continuity between secular
j and religious ritual. But unfortunately this failed to broaden
) the field of enquiry. For he wanted to use 'ritual' in a very
j narrow and special sense. It was to substitute for Durkheim's
cult of the sacred and so be restricted to the enactment of socially significant values (1939). Such-like constraints on the use 1 of words are intended to help understanding. But so often they
distort and confuse. Now we have got to the position in which Ritual replaces Religion in anthropologists' writings. It is used carefully and consistently to refer to symbolic action concerning the sacred. As a result the other, commoner kind of non-sacred ritual without religious efficacy has to be given another name if it is to be studied at all. So Radcliffe-Brown removed with one hand the barrier between sacred and secular, but put it back with the other. He also failed to follow up Durkheim's idea that .ritual belongs within a social theory of knowledge, but treated it as part of a theory of action, taking on uncritically some assumptions about 'sentiments' current in the psychology of his day. Where there are common values, he said, rituals express and focus attention on them. By rituals the necessary sentiments are generated to hold men to their roles. Childbirth taboos express to the Andaman Islanders the value of marriage and maternity and the danger to life in child labour. In war dances before a truce, the Andamans work off their sentiments f 65
Purity and Danger
of aggression. Food taboos instil sentiments of respect for seniority, and so on.
'I'llis approach is stultifying. Its main value is in requiring us to take taboos seriously because they express concern. But why the food taboos or visual or touch taboos should single out these particular foods or sights or contacts for avoidance is left unanswered. Radcliffe-Brown, somewhat in the spirit of Maimon-ides, implies that the question is silly, or that its answer is arbitrary. Even more unsatisfactory, we are left with little clue about people's concerns. It is obvious that death and childbirth should be a matter of concern. Thus Srinivas writing under the influence of Radcliffe-Brown says of Coorg avoidances and purifications:
'The pollution resulting from birth is milder than the pollution consequent on death. But in both cases pollution affects only the concerned kindred, and it is the means by which concern is defined and made known to everyone.' (1952, p. 102)
But he cannot apply the same reasoning to all pollutions. What sort of concern about bodily emissions, such as faeces or spittle, has to be denned and made known to everyone?
In the end the English received Durkheim's teaching when better field-work had raised understanding to the level of Durkheim's armchair insight. Lienhardt's whole discussion of Dinka religion is largely devoted to showing how rituals create and control experience. Writing of Dinka rain ceremonies, performed in the droughts of spring, he says:
'The Dinka themselves know, of course, when the rainy season is approaching . , . the point is of some importance for the correct appreciation of the spirit in which Dinka perform their regular ceremonies. In these their hujnansymbolic action .moves with the rhythm of the natural world around them," recreating that rhythm in moral terms and not merely attempting to coerce it into conformity with human desires.'
Lienhardt moves on in the same vein to sacrifices for health, for peace and to cancel the effects of incest. Finally he reaches the burial alive of Masters of the Fishing Spear, the rite by which the Dinka face and triumph over death itself. Throughout he insists on the rituals' function in modifying experience. Often it works retroactively. Officiants may solemnly deny the quarrels and misconduct which are the actual occasion of a
66
Magic and Miracle
sacrifice. This is not a cynical perjury at the altar itself. The object of the ritual is not to deceive God but to re-formulate past experience. By ritual and speech what has passed is restated so that what ought to have been prevails over what was, permanent good intention prevails over temporary aberration. When an act of incest has been committed, a sacrifice can alter the common descent of the pair and so expunge their guilt. The victim is cut in half alive, longitudinally through the sexual organs. So the common origin of the incestuous pair is symbolically negated. Similarly in peace-making ceremonies there are actions of blessing and purification as well as mimic battles:
'It seems that gesture without speech was enough to confirm in the external physical universe, an intention conceived interiorly in the moral. . . . The symbolic action in fact, mimes the total situation in which the parties in the feud know themselves to be including both their hostility and their disposition towards peace without which the ceremony could not be held. In this symbolic representation of their situation they control it, according to their will to peace, by transcending in symbolic action the only type of practical action (that is, continued hostilities) which for the Dinka follows from the situation of homicide.'
Later again (p. 291) he continues to hammer the point that ritual has as one of its objectives to control situations and to modify experience.
Only by establishing this point can he interpret the burial alive of the Dinka Spear Masters. Hence the fundamental principle is that certain men, closely in contact with Divinity, should not be seen to enter upon physical death.
'Their deaths are to be, or are to appear deliberate, and they are to be the occasion of a form of public celebration . . . the ceremonies in no way prevent the ultimate recognition of the ageing and physical death of those for whom they are performed. This death is recognised; but it is the public experience of it, for the survivors, which is deliberately modified by the performance of these ceremonies . . . the deliberately contrived death, though recognised as death, enables them to avoid admitting in this case the involuntary death which is the lot of ordinary men and beasts.'
The Master of the Fishing Spear does not kill himself. He requests a special form of death which is given by his people,
67
Purity and Danger
for their own sake, not for his. If he were to die an ordinary death, the life of his people which is in his keeping, goes with him. His ritually contracted death separates his personal life from this public life. Everyone should rejoice, because on this occasion there is a social triumph over death.
Reading this account of Dinka attitudes to their rituals one gets the impression that the author is like a swimmer heading against a heavy tide. All the time he has to push aside the flow of arguments from simple-minded observers who have taken the ritual at its Aladdin-and^hej.lamp_face value. Of course Dinka hope that their rites will suspend the natural course of events. Of course they hope that rain rituals will cause rain, healing rituals avert death, harvest rituals produce crops. But instrumental efficacy is not the only kind of efficacy to be derived from their symbolic action. The other kind is achieved in the action itself, in the assertions it makes and the experience which bears its imprinting.
Once this has been forcefully spelled out for Dinka religious experience we cannot escape its truth. We can apply it even more fully to our own selves. First we should allow for the fact that very little of our ritual behaviour is enacted in the context of religion. Dinka culture is unified. Since all their major contexts of experience overlap and interpenetrate, nearly all their experience is religious, and so therefore is all their most important ritual. But our experiences take place in separate compartments and our rituals too. So we must treat the spring millinery and spring cleaning in our towns as renewal rites which focus and control experience as much as Swazi first fruit rituals.
When we honestly reflect on our busy scrubbings and cleanings in this light we know that we are not mainly trying to avoid disease. _We are separating, placing boundaries, making visible statements about the home that we are- intending to create out of the material house. If we keep the bathroom cleaning materials away from the kitchen cleaning materials and send the men to the downstairs lavatory and the women upstairs, we are essentially doing the same thing as the Bushman wife when she arrives at a new camp (Marshall Thomas, p. 41). She chooses where she will place her fire and then sticks a rod in the ground. This orientates the fire and gives it a right and left side. Thus the home is divided between male and female quarters.
We moderns operate in many different fields of symbolic
68
Magic and Miracle
action. For the Bushman, Dinka and many primitive cultures, the field of symbolic action is one. The unity which they create
; by their separating and tidying is not just a little home, but a total universe in which all experience is ordered. Both we and the Bushmen justify our pollution avoidances by fear of danger. They believe that if a man sits on the female side his male
, virility will be weakened. We fear pathogenicity transmitted through micro-organisms. Often our justification of our own avoidances through hygiene is sheer fantasy. The difference between us is not that our behaviour is grounded on science and theirs on symbolism. Our behaviour also carries symbolic
1 meaning. The real difference is that we do not bring forward from one context to the next the same set of ever more power-
! ful symbols: our experience is fragmented. Our rituals create a lot of little sub-worlds, unrelated. Their rituals create one single,
) symbolically consistent universe. In the next two chapters we shall show what kinds of universes are produced when ritual
I        and political needs work freely together.
Now to return to the question of efficacy. Mauss wrote of
I primitive society repaying itself with the false coin of magic. The metaphor of money admirably sums up what we want to assert of ritual. Money provides a fixed, external, recognisable sign for what would be confused, contradictable operations; ritual makes visible external signs of internal states. Money
' mediates transactions; ritual mediates experience, including ^•pcial experience.__Money provides a standard for measuring worth; ritual standardises situations, and so helps to evaluate them. Money makes a link between the present and the future, so does ritual. The more we reflect on the richness of the metaphor, the more it becomes clear that this is no metaphor. Money is only an extreme and specialised type of ritual.
In comparing magic with false currency Mauss was wrong. Money can only perform its role of intensifying economic interaction if the public has faith in it. If faith in it is shaken, the currency is useless. So too with ritual; its symbols can only have effect so long as they command confidence. In this sense all money, false or true, depends on a confidence trick. The test of
* money is whether it is acceptable or not. There is no false money except by contrast with another currency which has more total acceptability. So primitive ritual is like good money, not false money, so long as it commands assent.
- 69
Purity and Danger
Note that money can only generate economic activity by virtue of the feed-back from public confidence in it. What about ritual? What kind of effectiveness is generated by confidence in the power of its symbols? Using the analogy with coinage we can revive the question of magical efficacy. There are two possible views: either the power of magic is sheer illusion, or it is not. If it is not illusion, then symbols have power to work changes. Miracles apart, such a power could only work at two levels, that of individual psychology and that of social life. We know very well that symbols have power in social life; the analogy with currency provides an illustration. But has the Bank Rate anything to do with Shamanistic cures? Psychoanalysts claim to work cures by manipulating symbols. Has the confrontation with the subconscious anything to do with primitive spell-binding and loosing? I now cite two marvellous studies which must render scepticism out of date.
One is Turner's analysis of a Shamanistic cure, 'An Ndembu Doctor in Practice' (1964), which I summarise briefly. The technique of the cure was the famous one of cupping blood and seeming to extract a tooth from the body of the patient. The symptoms were palpitations, severe pains in the back and disabling weakness. The patient was also convinced that the other villagers were against him and withdrew completely from social life. Thus there was a mixture of physical and psychological disturbance. The doctor proceeded by finding out everything about the past history of the village, conducting seances in which everyone was encouraged to discuss their grudges against the patient, while he aired his grievances against them. Finally the blood-cupping treatment dramatically involved the whole village in a crisis of expectation that burst in the excitement of the extraction of the tooth from the bleeding, fainting patient. Joyfully they congratulated him on his recovery and. themselves on their part in it. They had reason for joy since the long treatment had uncovered the main sources of tension in the village. In future the patient could play an acceptable role in their affairs. Dissident elements had been recognised and shortly left the village for good. The social structure was analysed and rearranged so that friction was, for the time, reduced.
In this absorbing study we are shown a case of skilful group therapy. The back-biting and envy of the villagers, symbolised by the tooth in the sick man's body, was dissolved in a wave of
70
Magic and Miracle
enthusiasm and solidarity. As he was cured of his physical symptoms they were all cured of social malaise. These symbols worked
i at the psycho-somatic level for the central figure, the sick man, and at the general psychological level for the villagers, in changing their attitudes, and at the sociological level in so far as the pattern of social statuses in the village was formally altered
,       and in so far as some people moved in and others moved away as a result of the treatment. In conclusion Turner says:
'Stripped of its supernatural guise, Ndembu therapy may well ^ offer lessons for Western clinical practice. For relief might be
given to many sufferers from neurotic illness if all those included in their social networks could meet together and publicly confess their ill will towards the patient and endure in their turn the recital of his grudges against them. But it is likely that nothing less than ritual sanctions for such behaviour and belief in the doctor's mystical powers could bring about such humility and compel people to display charity towards their suffering neighbour,'
t This account of a Shamanistic cure points to the mani-
pulation of the social situation as the source of its efficacy. The other enlightening study says nothing whatever about the social situation but concentrates on the direct power of the symbols to work upon the mind of the sufferer. Levi-Strauss
1 («949 & 1958), has analysed a Cuna Shaman's song which is chanted to relieve a difficult delivery in child birth. The doctor does not tquch the patient. The incantation is to have its effect merely by recital. The song starts by describing the difficulties of the midwife and her appeal to the Shaman. Then the Shaman at the head of a band of protective spirits, sets out (in song) for the house of Mum, a power responsible for the foetus, which has captured the soul of the patient. The song describes the quest, the obstacles and dangers and victories of the Shaman's party until they finally give battle to Muu and her confederates. Once Muu is conquered and frees the captive soul, the labouring mother is delivered of her child and the song ends. The interest of the song is that the landmarks on the Shaman's journey to Muu are literally the vagina and womb of the pregnant woman, in the depths of which he finally fights for her victoriously. By repetition and minute detail, the song forces the patient to attend to an elaborate account of what has gone
Purity and Danger
wrong in her labour. In one sense the body and internal organs of the patient are the theatre of action in the story, but by the transformation of the problem into a dangerous journey and battle with cosmic powers, by shuttling back and forth between the arena of the body and the arena of the universe, the Shaman is able to impose his view of the case. The patient's terror is focussed on the strength of mythic adversaries and her hopes of recovery fixed on the powers and ruses of the Shaman and his troups.
The cure would consist then in making an emotional situation thinkable; and in making the mind accept pains which the body refuses to bear. It is of no importance that the mythology of the Shaman docs not correspond to objective reality: the patient believes in it. The protective powers and the malevolent ones, the supernatural monsters and magic animals form part of a coherent system which underlies the native conception of the universe. The patient accepts them, or rather she has never doubted them. What she does not accept is this incoherent and arbitrary pain which is an intrusive element in her system. By appeal to the myth, the Shaman places it in a unified scheme where everything belongs. But the patient, having understood, does not resign herself: she gets better.'
Like Turner, Levi-Strauss also concludes his study with very pertinent suggestions for psychoanalysis.
These examples should be enough to shake a too complacent contempt of primitive religious beliefs. Not the absurd Ali Baba, but the magisterial figure of Freud is the model for appreciating the primitive ritualist. The ritual is creative indeed. More wonderful than the exotic caves and palaces of fairy tales, the magic of primitive ritual creates harmonious worlds with ranked and ordered populations playing their appointed parts. So far from being meaningless, it is primitive magic which gives meaning to existence. This applies as much to the negative as to the positive rites. The prohibitions trace the cosmic outlines and the ideal social order.
72
5
Primitive Worlds
'Now what are the characteristic marks of the sea-anemone', George Eliot muses, 'which entitle it to be removed from the hands of the botanist and placed in those of the zoologist?'
For us ambiguous species merely provoke essayists to elegant reflections. For Leviticus the rock badger or Syrian hyrax is unclean and abominable. Certainly it is an anomaly all right. It looks like an earless rabbit, has teeth like a rhino and the small hoofs on its toes seem to relate it to the elephant. But its existence does not threaten to bring the structure of our culture tumbling round our ears. Now that we have recognised and assimilated our common descent with apes nothing can happen in the field of animal taxonomy to rouse our concern. This is one reason why cosmic pollution is more difficult for us to understand than social pollutions of which we have some personal experience.
~ Another difficulty is our long tradition of playing down the difference between our own point of vantage and that of primitive cultures. The very real differences between 'us' and 'them' arc made little of, and even the word 'primitive' is rarely used. Yet it is impossible to make any headway with a study of ritual pollution if we cannot face the question of why primitive culture is pollution-prone and ours is not. With us pollution is a matter of aesthetics, hygiene or etiquette, which only becomes grave in so far as it. may create social embarrassment. The sanctions We social sanctions, contempt, ostracism, gossip, perhaps even police action. But in another large group of human societies the effects of pollution are much more wide ranging. A grave pollution is a religious offence. What is the basis of this differ-
73
Purity and Danger
cnce? We cannot avoid the question and must attempt to phrase an objective, verifiable distinction between two types of culture, jjrimitive and modern. Perhaps we Anglo-Saxons are more concerned to emphasise our sense of common humanity. We feel there is something discourteous in the term 'primitive' and so we avoid it and the whole subject too. Why else should Professor Herskovits have renamed the second edition of 'Primitive Economics* to 'Economic Anthropology' if his sophisticated West African friends had not expressed dislike of being lumped together with naked Fuegians and Aborigines under this general sign? Perhaps it is partly also in healthy reaction to early anthropology : 'Perhaps nothing so sharply differentiates the savage from the civilised man as the circumstance that "the former observes tabus, the latter does not' (Rose, 1926, p. 111). No one can be blamed for wincing at a passage such as the following, though I do not know who would take it seriously:
'We know that the primitive man of today has mental equipment very different from that of the civilised man. It is much more fragmented, much more discontinuous, more "gestalt-frcc". Professor Jung once told me how, in his travels in the African bush, he had noticed the quivering eye-balls of his native guides: not the steady gaze of the European, but a darting restlessness of vision, due perhaps to the constant expectation of danger. Such eye movements must be co-ordinated with a mental alertness and a swiftly changing imagery that allows little opportunity for discursive reasoning, for contemplation and comparison.' (H. Read, 1955)
If this were written by a Professor of Psychology it might be significant, but it is not. I suspect that our professional delicacy in avoiding the term 'Primitive' is the product of secret convictions of superiority. The physical anthropologists have a similar problem. While they attempt to substitute 'ethnic group' for the word 'race', (see Current Anthropology, 1964) their terminological problems do not inhibit them from their task of distinguishing and classifying forms of human variation. But social anthropologists, to the extent that they avoid reflecting on the grand distinctions between human cultures, seriously impede their own work. So it is worth asking why the term 'primitive' should imply any denigration.
Part, of our difficulty in England is that Levy-Bruhl, who first posed all the important questions about primitive cultures and
74
Primitive Worlds
their distinctiveness as a class, wrote in deliberate criticism of the English of his day, particularly of Frazer. Furthermore, Levy-Bruhl laid himself open to powerful counter-attack. Most text-books on comparative religion are emphatic about the mistakes he made, and say nothing about the value of the questions he asked. (For example, F. Bartlett, 1923, pp. 283-4, and P. Radin, 1956, pp. 230-1.) In my view he has not deserved such neglect, t Levy-Bruhl was concerned to document and to explain a pecu-
liar mode of thought. He started (1922) with the problem set by an apparent paradox. On the one hand there were convincing reports of the high level of intelligence of Eskimo or Bushmen f        (or of other such hunters and gatherers, or primitive cultivators or herdsmen), and on the other hand reports of peculiar leaps made in their reasoning and interpretation of events which suggested that their thought followed very different paths from our own. He insisted that their alleged dislike of discursive J*1        reasoning is not due to intellectual incapacity but to highly i       selective standards of relevance which produce in them an 'in-I       superable indifference to matters bearing no apparent relation to those which interest them'. The problem then was to discover the principles of selection and of association which made the primitive culture favour explanation in terms of remote, invisible mystic agencies and to lack curiosity about the intermediate links in a chain of events. Sometimes Levy-Bruhl seems to be putting his problem in terms of individual psychology, but it is clear that he saw it as a problem of the comparison of cultures first and as a psychological one only in so far as individual psychology is affected by cultural environment. He was interested in analysing 'collective representations', that is standardised assumptions and categories, rather than in individual I     aptitudes. It is precisely on this score that he criticised Tylor 1     and Frazer, who tried to explain primitive beliefs in terms of H    individual psychology, whereas he followed Dürkheim in seeing ^     collective representations as social phenomena, as common pat-%    terns of thought which are related to social institutions. In this %    he was undoubtedly right, but as his strength lay more in )■■     massive documentation than in analysis he was unable to apply his own precepts.
■ What Levy-Bruhl should have done, Evans-Pritchard has said, was to examine the variations in social structure and relate them ~      to concomitant variations in the patterns of thought. Instead
I. "
Purity and Danger
he contented himself with saying that all primitive people present uniform patterns of thought when contrasted with ourselves, and laid himself open to further criticism by seeming to make primitive cultures more mystical than they are and making civilised thought more rational than it is (Evans-Pritchard, Levy-Bruhl's Theory of Primitive Mentality). It seems that Evans-Pritchard himself was the first person to listen sympathetically to Levy-Bruhl and to direct his research to carrying Levy-Bruhl's problems into the more fruitful field which Levy-Bruhl himself missed. For his analysis of Azande witchcraft beliefs was exactly an exercise of this sort. It was the first study to describe a particular set of collective representations and to relate them intelligibly to social institutions (1937). Many studies have now ploughed lines parallel to this first furrow, so that from England and America a large body of sociological analysis of religions has vindicated Durkheim's insight. I say Durkheim's insight and not Levy-Bruhl's advisedly, for in so far as he contributed his own original slant to the matter, so Levy-Bruhl earned the just criticism of his reviewers. It was his idea to contrast primitive mentality with rational thought instead of sticking to the problem adumbrated by the master. If he had stayed with Durkheim's view of the problem he would not have been led into the confusing contrast of mystical with scientific thought, but would have compared primitive social organisation with complex modern social organisation and perhaps have done something useful towards elucidating the difference between organic and mechanical solidarity, between two types of social organisation which Dürkheim saw to underlie differences in beliefs.
Since Levy-Bruhl the general tendency in England has been to treat each culture studied as wholly sui generis, a unique and more or less successful adaptation to a particular environment (see Beattie, i960, p. 83, 1964, p. 272). Evans-Pritchard's criticism that Levy-Bruhl treated primitive cultures as if they were more uniform than they really are has stuck. But it is vital now to take up this matter again. We cannot understand sacred contagion unless we can distinguish a class of cultures in which pollution ideas flourish from another class of cultures, including our own, in which they do not. Old Testament scholars do not hesitate to enliven their interpretations of Israelite culture by comparison with primitive cultures. Psychoanalysts since Freud,
76
1;
Primitive Worlds
and metaphysicians since Cassirer are not backward in drawing general comparisons between our present civilisation and others very different. Nor can anthropologists do without such general distinctions.
The right basis for comparison is to insist on the unity of human experience and at the same time to insist on its variety, on the differences which make comparison worth while. The only way to do this is to recognise the nature of historical progress and the nature of primitive and of modern society. Progress means differentiation. Thus primitive means undifferentiated; modern means differentiated. Advance in technology involves differentiation in every sphere, in techniques and materials, in productive and political roles.
We could, theoretically, construct a rough gradient along which different economic systems would lie according to the degree to which they have developed specialised economic institutions. In the most undifferentiated economies roles in the productive system are not allocated by market considerations and there are few specialised labourers or craftsmen. A man does what work he does as part of performing his role as, say, son or brother or head of family. The same goes for the processes of distribution. As there is no labour exchange, so there is no supermarket. Individuals get their share of the community's product in virtue of their membership; their age, sex, seniority, their relationship to others. The patterns of status are etched by grooves of obligatory gift-making, along which rights to wealth are channelled.
Unfortunately for economic comparison there are many societies, small in scale, based on primitive techniques, which are not organised in this way, but rather on principles of market competition (see Pospisil). However, development in the political sphere lends itself very satisfactorily to the pattern I wish to introduce. There are not, in the most small-scale type of society, any specialised political institutions. Historical progress is marked by the development of diverse judicial, military, police, parliamentary and bureaucratic institutions. So it is easy enough to trace what internal differentiation would mean for social institutions.
On the face of it the same process should be traceable in the intellectual sphere. It seems unlikely that institutions should diversify and proliferate without a comparable movement in
77
Purity and Danger
the realm of ideas. Indeed we know that it does not happen. Great steps separate the historical development of the Hadza in Tanganyikan forests, who still never have occasion to count beyond four, from that of West Africans who for centuries have reckoned fines and taxes in thousands of cowries. Those of us who have not mastered modern techniques of communication such as the language of mathematics or of computers can put ourselves in the Hadza class compared with the ones who have become articulate in these media. We know only too well the educational burden our own civilisation carries in the form of specialised compartments of learning. Obviously the demand for special expertise and the education for providing it create cultural environments in which certain kinds of thinking can flourish and others cannot. Differentiation in thought patterns goes along with differentiated social conditions.
From this, basis it ought to be straightforward to say that in the realm of ideas there are differentiated thought systems which contrast with undifferentiated ones, and leave it at that. But the trap is just here. What could be more complex, diversified and elaborate than the Dogon cosmology? Or the Australian Murin-bata cosmology, or the cosmology of Samoa, or of Western Pueblo Hopi for that matter? The criterion" we are looking for is not in elaborateness and sheer complication of ideas.
There is only one kind of differentiation in thought that is relevant, and that provides a criterion that we can apply equally to different cultures and to the history of our own scientific ideas. That criterion is based on the Kantian principle that thought can only advance by freeing itself from the shacules of its own subjective conditions. The first Copernican revolution, the discovery that only man's subjective viewpoint made the sun seem to revolve round the earth, is continually renewed. In our own culture mathematics first and later logic, now history, now language and now thought processes themselves and even knowledge of the self and of society, are fields of knowledge progressively freed from the subjective limitations of the mind. To the extent to which sociology, anthropology and psychology are possible in it, our own type of culture needs to be distinguished from others which lack this self-awareness and conscious reaching for objectivity.
Radin interprets the Trickster myth of the Winnebago Indians on lines which serve to illustrate this point. Here is a primitive
78
'I
Primitive Worlds
parallel to Teilhard de Chardin's theme that the movement of evolution has been towards ever-increasing complextfication and self-awareness.
These Indians lived technically, economically and politically in the most simple undifferentiated conditions. Their myth contains their profound reflections on the whole subject of differentiation. The trickster starts as an unselfconscious, amorphous being. As the story unfolds he gradually discovers his own identity, gradually recognises and controls his own anatomical parts; he oscillates between female and male, but eventually fixes his own male sexual role; and finally learns to assess his environment for what it is. Radin says in his preface:
'He wills nothing consciously. At all times he is constrained to behave as he docs from impulses over which he has no control ... he is at the mercy of his passions and appetites . . . possesses no defined and well-fixed form . . . primarily an inchoate being of indeterminate proportions, a figure foreshadowing the shape of man. In this version he possesses intestines wrapped around his body and an equally long penis, likewise wrapped round his body with his scrotum on top of it.'
Two examples of his strange adventures will illustrate this theme. Trickster kills a buffalo and is butchering it with a knife in his right hand:
'In the midst of all these operations suddenly his left arm grabbed the buffalo. "Give that back to me, it is mine! Stop that or I will use my knife on you!" So spoke the right arm. "I will cut you to pieces, that is what I will do to you," continued the right arm. Thereupon the left arm released its hold. But shortly after, the left arm again grabbed hold of the right arm . . . again and again this was repeated. In this manner did Trickster make both his arms quarrel. That quarrel soon turned into a vicious fight and the left arm was badly cut up. . . .*
In another story Trickster treats his own anus as if it could act as an independent agent and ally. He had killed some ducks and before going to sleep he tells his anus to keep guard over the meat. While he is asleep some foxes draw near:
'When they came close, much to their surprise however, gas was expelled from somewhere. "Pooh" was the sound made. "Be careful! He must be awake", so they ran back. After a while one of them said "Well, I guess he is asleep now. That was
79
Purity and Danger
only a bluff. He is always up to some tricks. So again they approached the fire. Again gas was expelled and again they ran back. Three times this happened . . . Then louder, still louder, was the sound of g,-s expelled. "Pooh! Pooh! Pooh!" Yet they did not run away. On the contrary they now began to eat the roasted pieces of duck. . . .'
When Trickster woke up and saw the duck gone:
'. . . "Oh, you loo, you despicable object, what about your behaviour? Did I not tell you to watch this fire? You shall remember tins! As a punishment for your remissness, I will burn your mouth so that you will not be able to use it!" So he took a piece of burning wood and burned the mouth of his anus . . . and cried out of pain he was inflicting on himself,'
Trickster begins, isolated, amoral and unselfconscious, clumsy, ineffectual, an animal-like buffoon. Various episodes prune down and place more correctly his bodily organs so that he ends by looking like a man. At the same time he begins to have a more consistent set of social relations and to learn hard lessons about his physical environment. In one important episode he mistakes a tree for a man and responds to it as he would to a person until eventually he discovers it is a mere inanimate thing. So gradually he learns the functions and limits of his being.
I take this myth as a fine poetic statement of the process that leads from the early stages of culture to contemporary civilisation, differentiated in so many ways. The first type of culture is not prc-logical, as Levy-Bruhl unfortunately dubbed it, but pre-Copernican. Its world revolves round the observer who is trying to interpret his experiences. Gradually he separates him-srlf from his environment and perceives his real limitations and powers. Above all this pre-Copernican world is personal. Trickster speaks to creatures, things and parts of things without discrimination as if they were animate, intelligent beings. This personal universe is the kind of universe that Levy-Bruhl describes. It is also the primitive culture of Tylor and the animist culture of Marctt, and the mythological thought of Cassirer.
In the next few pages I am going to press as hard as I can the analogy between primitive cultures and the early episodes of the Trickster myth. I will try to present the characteristic areas of non-differentiation which define the primitive world view. I shall develop the impression that the primitive world view is
80
Primitive Worlds
subjective and personal, that different modes of existence are confused, that the limitations of man's being are not known. This is the view of primitive culture which was accepted by iylor and Frazer and which posed the problems of primitive \ mentality. I shall then try to show how this approach distorts the truth.
First, this world view is man-centred in the sense that explanations of events are couched in notions of good and bad j      fortune, which are implicitly subjective notions ego-centred in \      reference. In such a universe the elemental forces are seen as linked so closely to individual human beings that we can hardly speak of an external, physical environment. Each individual carries within himself such close links with the universe that he is like the centre of a magnetic field of force. Events can be explained in terms of his being what he is and doing what he has done. In this world it makes good sense for Thurber's fairy ,      tale king to complain that falling meteors are being hurled at
Ihimself, and for Jonah to come forward and confess that he is the cause of a storm. The distinctive point here is not whether the working of the universe is thought to be governed by spiritual beings or by impersonal powers. That is hardly relevant. Even powers which are taken to be thoroughly impersonal are held to be reacting directly to the behaviour of individual humans.
I.        A good example of belief in anthropocentric powers is the I      IKung Bushmen belief in N.'ow, a force thought to be responsible |;     for meteorological conditions at least in the Nyae-Nyae area of Bechuanaland. N.'ow is an impersonal, amoral force, definitely >     a thing and not a person. It is released when a hunter who has
(one kind of physical make-up kills an animal which has the corresponding element in its own make-up. The actual weather at any time is theoretically accounted for by the complex interaction of different hunters with different animals (Marshall). This hypothesis is attractive and one feels it must be intellectually satisfying since it is a view which is theoretically capable of being verified and yet no serious testing would ever be f practical.
;.•        To illustrate further the man-centred universe I quote from I    -what Father Tempels says of Luba philosophy. He has been criticised for implying that what he says so authoritatively from his intimate knowledge of Luba thought applies to all the Bantu.
81
Purity and Danger
But I suspect that in its broad lines his view on Bantu ideas of vital force applies not merely to all the Bantu, but much more widely. It probably applies to the whole range of thought which I am seeking to contrast with modern differentiated thought in European and American cultures.
For the Luba, he says, the created universe is centred on man (pp. 43-5). The three laws of vital causality are:
(1) that a human (living or dead) can directly reinforce or diminish the being (or force) of another human
(2} that the vital force of a human can directly influence inferior force-beings (animal, vegetable or mineral)
(3) that a rational being (spirit, dead or living human) can act indirectly on another by communicating his vital influence to an intermediary inferior force.
Of course there are very many different forms which the idea of a man-centred universe may take. Inevitably ideas of how men affect other men must reflect political realities. So ultimately we shall find that these beliefs in man-centred control of the environment vary according to the prevailing tendencies in the political system (see Chapter 6). But in general we can distinguish beliefs which hold that all men are equally involved with the universe from beliefs in the special cosmic powers of selected individuals. There are beliefs about destiny which are thought to apply universally to all men. In the culture of Homeric literature it was not certain outstanding individuals whose destiny was the concern of the gods, but all and each whose personal fate was spun on the knees of the gods and woven for good or ill with the fates of others. Just to take one contemporary example, Hinduism today teaches, as it has for centuries, that for each individual the precise conjunction of the planets at the time he was born signifies much for his personal good or ill-fortune. Horoscopes are for everybody. In both these instances, though the individual can be warned by diviners about what is in store for him, he cannot change it radically, only soften a little the hard blows, defer or abandon hopeless desires, be alert to the opportunities that will lie in his path.
Other ideas about the way in which the individual's fortune is bound up with the cosmos may be more pliable. In many parts of West Africa today, the individual is held to have a complex personality whose component pans act like separate
82
Primitive Worlds
persons. One part of the personality speaks the life-course of the individual before he is born. After birth, if the individual strives for success in a sphere which has been spoken against, ;      his efforts will always be in vain. A diviner can diagnose this spoken destiny as cause of his failures and can then exorcise his prenatal choice. The nature of his predestined failure which a man has to take account of varies from one West African society to another. Among the Tallensi in the Ghana hinterland the conscious personality is thought to be amiable and uncompetitive. His unconscious element which spoke his destiny before birth is liable to be diagnosed as over-aggressive and rivalrous, and so makes him a misfit in a system of controlled statuses. By contrast the Ijo of the Niger Delta, whose social organisation is fluid and competitive, take the conscious component of the self to be full of aggression, desire to compete and to excel. In ■      this case it is the unconscious self which may be pre-destined to ':      failure because it chose obscurity and peace. Divination can discover the discrepancy of aims within the person, and ritual can put it right. (Fortes, 1959; Horton, 1961.) These examples point to another lack of differentiation in the i      personal world view. We saw above that the physical environ-I     ment is not clearly thought of in separate terms, but only with (      reference to the fortunes of human selves. Now we see that the I     self is not clearly separated as an agent. The extent and limits of its autonomy are not defined. So the universe is part of the I.     self in a complementary sense, seen from the angle of the in-I     dividual's idea, not this time of nature, but of himself. I'       The Tallensi and Ijo ideas about the multiple warring per-I     sonalities in the self seem to be more differentiated than the Homeric Greek idea. In these West African cultures the bind-■j    ing words of destiny are spoken by pan of the individual him-I    self. Once he knows what he has done he can repudiate his $   earlier choice. In Ancient Greece the self was seen as a passive £   victim of external agents:
^       'In Homer one is struck by the fact that his heroes with all
fi.       their magnificent vitality and activity feel themselves at every turn not free agents but passive instruments or victims of other powers ... a man felt that he could not help his own emotions. I An idea, an emotion, an impulse came to him; he acted and
f presently rejoiced or lamented. Some god had inspired him or ,': blinded him. He prospered, then was poor, perhaps enslaved;
Purity and Danger
he wasted away with disease, or died in battle. It was divinely ordained, his portion apportioned long before. The prophet or diviner might discover it in advance; the plain man knew a little about omens and merely seeing his shaft hit its mark or the enemy prevailing, concluded that Zeus had assigned defeat to himself and his comrades. He did not wait to fight further but fled.' (Onians, 1951, p. 302)
The pastoral Dinka living in the Sudan similarly are said not to distinguish the self as an independent source of action and of reaction. They do not reflect on the fact that they themselves react with feelings of guilt and anxiety and that these feelings initiate other states of mind. The self acted upon by emotions they portray by external powers, spiritual beings who cause misfortune of various kinds. So in an effort to do justice to the complex reality of the self's interaction within itself the Dinka universe is peopled with dangerous personal extensions to the self. This is almost exactly how Jung described the primitive world view when he said:
'An unlimited amount of what we now consider an integral part of our own psychic being disports itself merrily for the primitive in projections reaching far and wide.' (p. 74)
I give one more example of a world in which all individuals are seen as personally linked with the cosmos to show how varied these linkages can be. Chinese culture is dominated by the idea of harmony in the universe. If an individual can place himself to ensure the most harmonious relationship possible, he can hope for good fortune. Misfortune may be attributed to lack of just such a happy alignment. The influence of the waters and the airs, called Feng Shwe, will bring him good fortune if his house and his ancestors' graves arc well placed. Professional geomancers can divine the causes of his misfortune and he can then rearrange his home or his parental graves to better effect. Dr. Freedman in his forthcoming book holds that geomancy has an important place in Chinese beliefs alongside ancestor worship. The fortune which a man can manipulate thus by geomantic skills has no moral implications; but ultimately it must be brought to terms with the reward of merit which in the same set of beliefs is meted out by heaven. Finally then, the whole universe is interpreted as tied in its detailed workings to the lives of human persons. Some individuals arc more successful in
84
Primitive Worlds
dealing with Feng Shwe than others, just as some Greeks have a more splendid fate decreed for them and some West Africans a spoken destiny more committed to success.
Sometimes it is only marked individuals and not all humans who are significant. Such marked individuals draw lesser men in their wake, whether their endowment is for good or evil fortune. For the ordinary man in the street, not endowed himself, the practical problem is to study his fellow men and discover whom among them he ought to avoid or follow.
In all the cosmologies we have mentioned so far, the lot of individual humans is thought to be affected by power inhering in themselves or in other humans. The cosmos is turned in, as it were, on man. Its transforming energy is threaded on to the lives of individuals so that nothing happens in the way of storms, sickness, blights or droughts except in virtue of these personal links. So the universe is man-centred in the sense that it must be interpreted by reference to humans.
But there is a quite other sense in which .the primitive undifferentiated world view may be described as personal. Persons are essentially not things. They have wills and intelligence. With their wills they love, hate and respond emotionally. With their intelligence they interpret signs. But in the kind of universe I am contrasting with our own world view, things are not clearly distinct from persons. Certain kinds of behaviour characterise person to person relations. First, persons communicate with one another by symbols in speech, gesture, rite, gift and so on. Second, they react to moral situations. However impersonally the cosmic forces may be defined, if they seem to respond to a person-to-person style of address their quality of thing is not fully differentiated from their personality. They may not be persons hut nor are they entirely things.
Here there is a trap to avoid. Some ways of talking about things might seem to the naive observer to imply personality. Nothing can necessarily be inferred about beliefs from purely linguistic distinctions or confusions. For instance a Martian anthropologist might come to the wrong conclusion on overhearing an English plumber asking his mate for the male and female parts of plugs. To avoid falling into linguistic pitfalls, I confine my interests to the kind of behaviour which is supposed to produce a response from allegedly impersonal forces.
It may not be at all relevant here that the Nyae-Nyae Bush-
85
Purity and Danger
men attribute male and female character to clouds, any more than it is relevant that we use 'she' for cars and boats. But it may be relevant that the pygmies of the Ituri forest, when misfortune befalls, say that the forest is in a bad mood and go to the trouble of sjnging to it all night to cheer it up, and that they then expect their affairs to prosper (Turnbull). No European mechanic in his senses would hope to cure engine trouble by serenade or curse.
So here is another way in which the primitive, undifferentiated universe is personal. It is expected to behave as if it was intelligent, responsive to signs, symbols, gestures, gifts, and as if it could discern between social relationships.
The most obvious example of impersonal powers being thought responsive to symbolic communication is the belief in sorcery. The sorcerer is the magician who tries to transform the path of events by symbolic enactment. He may use gestures or plain words in spells or incantations. Now words are the proper mode of communication between persons. If there is an idea that words correctly said are essential to the efficacy of an action, then, although the thing spoken to cannot answer back, there is a belief in a limited kind of one-way verbal communication. And this belief obscures the clear thing-status of the thing being addressed. A good example is the poison used for the oracular detection of witches in Zandeland (Evans-Pritchard, 1937). The Azande themselves brew their poison from bark. It is not said to be a person but a thing. They do not suppose there is a little man inside which works the oracle. Yet for the oracle to work the poison must be addressed aloud, the address must convey the question unequivocally and, to eliminate error of interpretation, the same question must be put in reverse form in the second round of consultation. In this case not only does the poison hear and understand the words, but it has* limited powers of reply. Either it kills the chicken or it does not. It can only give yes and no answers. It cannot initiate a conversation or conduct an unstructured interview. Yet this limited response to questioning radically modifies its thing-status in the Azande universe. It is not an ordinary poison, but more like a captive interviewee filling in a survey questionnaire with crosses and ticks.
The Golden Bough is full of examples of belief in an impersonal universe which, nevertheless, listens to speech and responds
86
Primitive Worlds
to it one way or another. So are modern field-workers' reports. Stanner says: 'Most of the choir and furniture of heaven and earth are regarded by the Aborigines as a vast sign system. Anyone who understanding^ has moved in the Australian bush with aborigine associates becomes aware of the fact. He moves, not in a landscape but in a humanised realm saturated with significations.'
Finally there are the beliefs which imply that the impersonal universe has discernment. It may discern between fine nuances in social relations, such as whether the partners in sexual intercourse are related within prohibited degrees, or between less fine ones such as whether a murder has been committed on a fellow-tribesman or on a stranger, or whether a woman is married or not. Or it may discern secret emotions hidden in men's breasts. There are many examples of implied discernment of social status. The hunting Cheyenne thought that the buffaloes who provide their main livelihood were affected by the rotten smell of a man who had murdered a fellow-tribesman and they moved away, thus endangering the survival of the tribe. The buffalo were not supposed to react to the smell of murder of a foreigner. The Australian Aborigines of Arnhemland conclude their fertility and initiation ceremonies with ceremonial copulation, believing that the rite is more efficacious if sexual intercourse takes place between persons who are at other times strictly prohibited (Berndt, p. 49). The Lele believe that a diviner who has had sexual intercourse with the wife of his patient, or whose patient has had sexual intercourse with his wife, cannot heal him, because the medicine intended to heal would kill. This result is not dependent on any intention or knowledge on the part of the doctor. The medicine itself is thought to react in this discriminating way. Furthermore, the Lele believe that if a cure is effected and the patient omits to pay his healer promptly for his services, early relapse or even a more fatal complication of the illness will result. So Lele medicine, by implication, is credited with discerning debt as well as secret adultery. Even more intelligent is the vengeance magic bought by the Azande which detects unerringly the witch responsible for a given death, and does capital justice on him. So impersonal elements in the universe are credited with discrimination which enables them to intervene in human affairs and uphold the moral code.
In this sense the universe is apparently able to make judg-
87
Purity and Danger
ments on the moral value of human relations and to act accordingly. Malweza, among the Plateau Tonga in Northern Rhodesia, is a misfortune which afflicts those who commit certain specific offences against the moral code. Those offences are in general of a kind against which ordinary punitive sanctions cannot be applied. For example, homicide within the group of matrilineal kinsmen cannot be avenged because the group is organised to avenge the murder of its members by outsiders (Colson, p. 107). Malweza punishes offences which are inaccessible to ordinary sanctions.
To sum up, a primitive world view looks out on a universe which is personal in several different senses. Physical forces are thought of as interwoven with the lives of persons. Things are not completely distinguished from persons and persons are not completely distinguished from their external environment. The universe responds to speech and mime. It discerns the social order and intervenes to uphold it.
I have done my best to draw from accounts of primitive cultures a list of beliefs which imply lack of differentiation. The materials I have used are based on modem fieldwork. Yet the general picture closely accords with that accepted by Tylor or Marett in their discussions of primitive animism. They are the kind of beliefs from which Frazer inferred that the primitive mind confused its subjective and objective experiences. They are the same beliefs which provoked Levy-Bruhl to reflect on the way that collective representations impose a selective principle on interpretation. The whole discussion of these beliefs has been haunted by obscure psychological implications.
If these beliefs are presented as the result of so many failures to discriminate correctly they evoke to a startling degree the fumbling efforts of children to master their environment. Whether we follow Klein or Piaget, the theme is the same; confusion of internal and external, of thing and person, self and environment, sign and instrument, speech and action. Such confusions may be necessary and universal stages in the passage of the individual from the chaotic, undifferentiated experience of infancy to intellectual and moral maturity.
So it is important to point out again, as has often been said before, that these connections between persons and events which characterise the primitive culture do not derive from failure to differentiate. They do not even necessarily express the thoughts
88
Primitive Worlds 1 of individuals. It is quite possible that individual members of such cultures hold very divergent views on cosmology. Vansina recalls affectionately three very independent thinkers he encountered among the Bushong, who liked to expound their per-■ sonal philosophies to him. One old man had come to the conclusion that there was no reality, that all experience is a shifting illusion. The second had developed a numerological type of metaphysics, and the last had evolved a cosmological scheme of great complexity which no one understood but himself (1964). It is misleading to think of ideas such as destiny, witchcraft, mana, magic as part of philosophies, or as systematically thought out at all. They are not just linked to institutions, as Evans-Pritchard put it, but they are institutions—every bit as much as Habeas Corpus or Hallow-e'en. They are all compounded part of belief and part of practice. They would not have been recorded in the ethnography if there were no practices attached to them. Like other institutions they are both resistant to change and sensitive to strong pressure. Individuals can change them I       by neglect or by taking an interest.
If we remember that it is a practical interest in living and I        not an academic interest in metaphysics which has produced I        these beliefs, their whole significance alters. To ask an Azande whether the poison oracle is a person or a thing is to ask a kind of nonsensical question which he would never pause to ask himself. The fact that he addresses the poison oracle in words does not imply any confusion whatever in his mind between things and persons. It merely means that he is not striving for intellectual consistency and that in this field symbolic action seems i        appropriate. He can express the situation as he sees it by speech '        and mime, and these ritual elements have become incorporated into a technique which, to many intents and purposes is like 1        programming a problem through a computer. I think that this !       is something argued by Radin in 1927 and by Cellner (1962) when he points to the social function of incoherences in doc-,        trines and concepts.
Robertson Smith first tried to draw attention away from be-liefs considered as such, to the practices associated with them. ' f And much other testimony has piled up since on the strictly ||! practical limitation on the curiosity of individuals. This is not |i a peculiarity of primitive culture. It is true of 'us' as much as H of 'them', in so far as 'we' are not professional philosophers. As 1 89
Purity and Danger
business man, farmer, housewife no one of us has time or inclination to work out a systematic metaphysics. Our view of the world is arrived at piecemeal, in response to particular practical problems.
In discussing Azande ideas about witchcraft Evans-Pritchard insists on this concentration of curiosity on the singularity of an individual event. If an old and rotten granary falls down and kills someone sitting in its shadow, the event is ascribed to witchcraft. Azande freely admit that it is in the nature of old and rotten granaries to collapse, and they admit that if a person sits for several hours under its shadow, day after day, he may be crushed when it falls. The general rule is obvious and not an interesting field for speculation. The question that interests them is the emergence of a unique event out of the meeting point of two separate sequences. There were many hours when no one was sitting under that granary and when it might have collapsed harmlessly, killing no one. There were many hours when other people were seated by it, who might have been victims when it fell, but who happened not to be there. The fascinating problem is why it should have fallen just when it did, just when so-and-so and no one else was sitting there. The general regularities of nature are observed accurately and finely enough for the technical requirements of Azande culture. But when technical information has been exhausted, curiosity turns instead to focus on the involvement of a particular person with the universe. Why did it have to happen to him? What can he do to prevent misfortune? Is it anyone's fault? This applies, of course, to a theistic world view. As with witchcraft only certain questions are answered by reference to spirits. The regular procession of the seasons, the relation of cloud to rain and rain to harvest, of drought to epidemic and so on, is recognised. They are taken for granted as the back-drop against which more personal and pressing problems can be solved. The vital questions in any theistic world-view are the same as for the Azande: why did this farmer's crops fail and not his neighbour's? Why did this man get gored by a wild buffalo and not another of his hunting party? Why did this man's children or cows die? Why me? Why today? What can be done about it? These insistent demands for explanation are focussed on an individual's concern for himself and his community. We now know what Durkheim knew, and what Frazcr, Tylor and Marett did not. These quet-
90
Primitive Worlds
tions are not phrased primarily to satisfy man's curiosity about the seasons and the rest of the natural environment. They are phrased to satisfy a dominant social concern, the problem of how to organise together in society. They can only be answered, it is true, in terms of man's place in nature. But the metaphysic is a by-product, as it were, of the urgent practical concern. The anthropologist who draws out the whole scheme of the cosmos which is implied in these practices does the primitive culture great violence if he seems to present the cosmology as a systematic philosophy subscribed to consciously by individuals. We can study our own cosmology—in a specialised department of astronomy. But primitive cosmologies cannot rightly be pinned out for display like exotic iepidoptera, without distortion to the nature of a primitive culture. In a primitive culture the technical problems have been more or less settled for generations past. The live issue is how to organise other people and oneself in relation to them; how to control turbulent youth, how to soothe disgruntled neighbours, how to gain one's rights, how to prevent usurpation of authority, or how to justify it. To serve these practical social ends all kinds of beliefs in the omniscience and omnipotence of the environment are called into play. If social life in a particular community has settled down into any sort of constant form, social problems tend to crop up in the same areas of tension or strife. And so as part of the machinery for resolving them, these beliefs about automatic punishment, destiny, ghostly vengeance and witchcraft crystallise in the institutions. So the primitive world view which I have defined above is rarely itself an object of contemplation and speculation in the primitive culture. It has evolved as the appanage of other social institutions. To this extent it is produced indirectly, and to this extent the primitive culture must be taken to be unaware of itself, unconscious of its own conditions.
In the course of social evolution institutions proliferate and specialise. The movement is a double one in which increased social control makes possible greater technical developments and the latter opens the way to increased social control again. Finally we find ourselves in the modern world where economic interdependence is carried to the highest pitch reached by mankind so far. One inevitable by-product of social differentiation is social awareness, self-consciousness about the processes of communal life. And with differentiation go special forms of social coercion,
9'
Purity and Danger
special monetary incentives to conform, special types of punitive sanctions, specialised police and overseers and progress men scanning our performance, and so on, a whole paraphernalia of social control which would never be conceivable in small-scale undifferentiated economic conditions. This is the experience of organic solidarity which makes it so hard for us to interpret the efforts of men in primitive society to overcome the weakness of their social organisation. Without forms filled in triplicate, without licences and passports and radio-police cars they must somehow create a society and commit men and women to its norms. I hope I have now shown why Levy-Bruhl was mistaken in comparing one type of thought with another instead of comparing social institutions.
Wc can also see why Christian believers, Moslems and Jews are not to be classed as primitive on account of their beliefs. Nor necessarily Hindus, Buddhists or Mormons, for that matter. It is true that their beliefs are developed to answer the questions 'Why did it happen to me: Why now?' and the rest. It is true that their universe is man-centred and personal. Perhaps in entertaining metaphysical questions at all these religions may be counted anomalous institutions in the modern world. For unbelievers may leave such problems aside. But this in itself does not make of believers promontories of primitive culture sticking out strangely in a modern world. For their beliefs have been phrased and rephrased with each century and their inter-meshing with social life cut loose. The European history of ecclesiastical withdrawal from secular politics and from secular intellectual problems to specialised religious spheres is the history of this whole movement from primitive to modem.
Finally we should revive the question of whether the word 'primitive' should be abandoned. I hope not. It has a defined and respected sense in art. It can be given a valid meaning for technology and possibly for economics. What is the objection to saying that a personal, anthropocentric, undifferentiated world-view characterises a primitive culture? The only source of objection could be from the notion that it has a pejorative sense in relation to religious beliefs which it does not carry in technology and art. There may be something in this for a certain section of the English-speaking world.
The idea of a primitive economy is slightly romantic. It is true that we are materially and technically incomparably better
9a
Primitive Worlds
equipped, but no one would frankly base a cultural distinction on purely materialist grounds. The facts of relative poverty and wealth are not in question. But the idea of the primitive economy is one which handles goods and services without the intervention of money. So the primitives have the advantage over us in that they encounter economic reality direct, while we are always being deflected from our course by the complicated, unpredictable and independent behaviour of money. But on this basis, when it comes to the spiritual economy, we seem to have the advantage. For their relation to their external environment is mediated by demons and ghosts whose behaviour is complicated and unpredictable, while we encounter our environment more simply and directly. This latter advantage we owe to our wealth and material progress which has enabled other developments to take place. So, on this reckoning, the primitive is ultimately at a disadvantage both in the economic and spiritual field. Those who feel this double superiority are naturally inhibited from flaunting it and this is presumably why they prefer not to distinguish primitive culture at all.
Continentals seem to have no such squeamishness. 'Le primitif enjoys honour in the pages of Leenhard, Levi-Strauss, Ricoeur and EHade. The only conclusion that I can draw is that they are not secretly convinced of superiority, and are intensely appreciative of forms of culture other than their own.
93
6
Powers and Dangers
Granted that disorder spoils pattern; it also provides the materials of pattern. Order implies restriction; from all possible materials, a limited selection has been made and from all possible relations a limited set has been used. So disorder by implication is unlimited, no pattern has been realised in it, but its potential for patterning is indefinite. This is why, though we seek to create order, we do not simply condemn disorder. We recognise that it is destructive to existing patterns; also that it has potentiality. ^symbolises both danger_an4_^ower. ■
Ritual recognises the potency of disorder. In the disorder of the mind, in dreams, faints and frenzies, ritual expects to find powers and truths which cannot lie reached by conscious effort. Energy to command and special powers of healing come to those who can abandon rational control for a time. Sometimes an Andaman Islander leaves his band and wanders in the forest like a madman. When he returns to his senses and to human society he has gained occult power of healing (Radcliffe Brown, •933« P- '39)- This is a very common notion, widely attested. Webster in his chapter on the Making of a Magician {The Sociological Study of Magic), gives many examples. I also quote the Ehanzu, a tribe in the central region of Tanzania, where one of the recognised ways of acquiring a diviner's skill is by going mad in the bush. Virginia Adam, who worked among this tribe, tells me that their ritual cycle culminates in annual rain rituals. If at the expected time rain fails, people suspect sorcery. To undo the effects of sorcery they take a simpleton and send him wandering into the bush. In the course of his wanderings he unknowingly destroys the sorcerer's work.
94
Powers and Dangers
In these beliefs there is a double play on inarticulateness. First there is a venture into the disordered regions of the mind. Second there is the venture beyond the confines of society. The man who comes back from these inaccessible regions brings with him a power not available to those who have stayed in the control of themselves and of society.
This ritual play on articulate and inarticulate forms is crucial to understanding pollution, Inritualform it is treated as ifit were quick with power to maintain itself in being, yet always liable to attack. Formlessness is also credited with powers, some dangerous, some good. We have seen how the abominations of Leviticus are the obscure unclassifiable elements which do not fit the pattern of the cosmos. They are incompatible with holiness and blessing. The play on form and formlessness is even more clear in the rituals of society. : First, consider beliefs about persons in a marginal state. These
are people who are somehow left out in the patterning of society, who are placeless. They may be doing nothing morally wrong, but their status is indefinable. Take, for example, the unborn I        child. Its present position is ambiguous, its future equally. For I        no one can say what sex it will have or whether it will survive the hazards of infancy. It is often treated as both vulnerable and dangerous. The Lele regard the unborn child and its mother as in constant danger, but they also credit the unborn child with capricious ill-will which makes it a danger to others. When pregnant, a Lele woman tries to be considerate about not approaching sick persons lest the proximity of the child in her womb causes coughing or fever to increase. Among the Nyakyusa a similar belief is recorded. A pregnant |        woman is thought to reduce the quantity of grain she approaches, I       because the foetus in her is voracious and snatches it. She must I        not speak to people who are reaping or brewing without first mak-!        ing a ritual gesture of goodwill to cancel the danger. They speak }        of the foetus 'with jaws agape* snatching food, and explain it by I        the inevitability of. the 'seed within* fighting the 'seed without*.
I The child in the belly ... is like a witch; it will damage food
j like witchcraft; beer is spoiled and tastes nasty, food docs not
) grow, the smith's iron is not easily worked, the milk is not good.'
Even the father is endangered at war or in the hunt by his wife's i       pregnancy (Wilson, pp. 138-9). V 95
Purity and Danger
Levy-Bruhl noted that menstrual blood and miscarriage sometimes attract the same kind of belief. The Maoris regard menstrual blood as a sort of human being manque. If the blood had not flowed it would have become a person, so it has the impossible status of a dead person that has never lived. He quoted a common belief that a foetus born prematurely has a malevolent spirit, dangerous to the living (pp. 390-6). Levy-Bruhl did not generalise that danger lies in marginal states, but Van Gennep had more sociological insight. He saw society as a house with rooms and corridors in which passage from one to another is dangerous. Danger lies in transitional states, simply because transition is neither one state nor the next, it is undefinable. The person who must pass from one to another is himself in danger and emanates danger to others. The danger is controlled by ritual which precisely separates him from his old status, segregates him for a time and then publicly declares his entry to his new status. Not only is transition itself dangerous, but also the rituals of segregation are the most dangerous phase of the rites. So often do we read that boys die in initiation ceremonies, or that their sisters and mothers are told to fear for their safety, or that they used in the old days to die from hardship or fright, or by supernatural punishment for their misdeeds. Then somewhat tamely come the accounts of the actual ceremonies which are so safe that the threats of danger sound like a hoax (Vansina, 1955). But we can be sure that the trumped up dangers express something important about marginality. To say that the boys risk their lives says precisely that to go out of the formal structure and to enter the margins is to be exposed to power that is enough to kill them or make their manhood. The theme of death and rebirth, of course, has other symbolic functions: the initiates die to their old life and are reborn to the new. The whole repertoire of ideas concerning pollution arid purification are used to mark the gravity of the event and the power of ritual to remake a man—this is straightforward.
During the marginal period which separates ritual dying and ritual rebirth, the novices in initiation are temporarily outcast. For the duration of the rite they have no place in society. Sometimes they actually go to live far away outside it. Sometimes they live near enough for unplanned contacts to take place between full social beings and the outcasts. Then we find them behaving like dangerous criminal characters. They are licensed to waylay,
96
Powers and Dangers
^ steal, rape. This behaviour is even enjoined on them. To behave anti-socially is the proper expression of their marginal condition (Webster, 1908, chapter III). To have been in the margins is to have
■f been in contact with danger, to have been at a source of power. It is consistent with the ideas about form and formlessness to treat initiands coming out of seclusion as if they were themselves charged with power, hot, dangerous, requiring insulation and a time for cooling down. Dirt, obscenity and lawlessness are as relevant symbolically to the rites of seclusion as other ritual expressions of their condition. They are not to be blamed for misconduct any more than the foetus in the womb for its spite and greed.
It seems that if a person has no place in the social system and is therefore a marginal being, all precaution against danger must 1        come from others. He cannot help his abnormal situation. This I       is roughly how we ourselves regard marginal people in a secular, not a ritual context. Social workers in our society, concerned with the after-care of ex-prisoners, report a difficulty of resettling ,       them in steady jobs, a difficulty which comes from the attitude I       of society at large. A man who has spent any time 'inside' is put f       permanently 'outside' the ordinary social system. With no rite I       of aggregation which can definitively assign him to a new posi-tion he remains in the margins, with other people who are sirni-;       lar credited with unreliability, unteachability, and all the wrong social attitudes. The same goes for persons who have entered institutions for the treatment of mental disease. So long as they stay at home their peculiar behaviour is accepted. Once they have been formally classified as abnormal, the very same behaviour is counted intolerable. A report on a Canadian project I       in 1951 to change the attitude to mental ill-health suggests that I       there is a threshold of tolerance marked by entry to a mental hospital. If a person has never moved out of society into this marginal state, any of his eccentricities are comfortably tolerated by his neighbours. Behaviour which a psychologist would class at once as pathological is commonly dismissed as 'Just a quirk', I       or 'He'll get over it*, or 'It takes all sorts to make a world'. But I       once a patient is admitted to a mental hospital, tolerance is !       withdrawn. Behaviour which was formerly judged to be so ]       normal that the psychologist's suggestions raised strong hostility, I      was now judged to be abnormal (quoted in Cumrning). So men-I;-'     tal health workers find exactly the same problems in rehabilitat-
Purity and Danger ing their discharged patients as do the prisoners' aid societies. The fact that these common assumptions about ex-prisoners and lunatics are self-validating is not relevant here. It is more interesting to know that marginal status produces the same reactions the world over, and that these are deliberately represented in marginal rites.
To plot a map of the powers and dangers in a primitive universe, we need to underline the interplay of ideas of form and formlessness. So many ideas about power are based on an idea of society as a series of forms contrasted with surrounding non-form. There is power in the forms and other power in the inarticulate area, margins, confused lines, and beyond the external boundaries. If pollution is a particular class of danger, to see where it belongs in the universe of dangers we need an inventory of all the possible sources of power. In a primitive culture the physical agency of misfortune is not so significant as the personal intervention to which it can be traced. The effects are the same the world over: drought is drought, hunger is hunger; epidemic, child labour, infirmity—most of the experiences are held in common. But each culture knows a distinctive set of laws governing the way these disasters fall. The main links between persons and misfortunes are personal links. So our inventory of powers must proceed by classifying all kinds of personal intervention in the fortunes of others.
The spiritual powers which human action can unleash can roughly be divided into two classes—internal and external. The first reside within the psyche of the agent—such as evil eye. witchcraft, gifts of vision or prophecy. The second are external symbols on which the agent must consciously work: spells, blessings, curses, charms and formulas and invocations. These powers require actions by which spiritual power is discharged.
This distinction between internal and external sources of power is often correlated with another distinction, between uncontrolled and controlled power. According to widespread beliefs, the internal psychic powers are not necessarily triggered off by the intention of the agent. He may be quite unaware that he possesses them or that they are active. These beliefs vary from place to place. For example, Joan of Arc did not know when her voices would speak to her, could not summon them at will, was often startled by what they said and by the train of events which her obedience to them started. Trie Azande
9«
Powers and Dangers
believe that a witch does not necessarily know that his witchcraft is at work, yet if he is warned, he can exert some control to check its action.
By contrast, the magician cannot utter a spell by mistake; specific intention is a condition of the result. A father's curse usually needs to be pronounced to have effect.
Where does pollution come in the contrast between uncontrolled and controlled power, between psyche and symbol? As I see it, pollution is a source of danger altogether in a different class: the distinctions of voluntary, involuntary, internal, external, are not relevant. It must be identified in a different way.
First to continue with the inventory of spiritual powers, there is another classification according to the social position of those endangering and endangered. Some powers are exerted on be-' i half of the social structure; they protect society from malefactors i| against whom their danger is directed. Their use must be approved by all good men. Other powers are supposed to be a danger to society and their use is disapproved; those who use them are malefactors, their victims are innocent and all good men would try to hound them down—these are witches and sorcerers. This is the old distinction between white and black magic
Are these two classifications completely unconnected? Here I tentatively suggest a correlation: where the social system explicitly recognises positions of authority, those holding such positions are endowed with explicit spiritual power, controlled, conscious, external and approved—powers to bless or curse. Where the social system requires people to hold dangerously ambiguous roles, these persons are credited with uncontrolled, unconscious, dangerous, disapproved powers—such as witchcraft and evil eye.
In other words, where the social system is well-articulated, I look for articulate powers vested in the points of authority; where the social system is ill-articulated, I look for inarticulate powers vested in those who are a source of disorder. I am suggesting that the contrast between form and surrounding non-form accounts for the distribution of symbolic and psychic powers: external symbolism upholds the explicit social structure and internal, unformed psychic powers threaten it from the non-structure.
This correlation is admittedly difficult to establish. For one
99
Purity and Danger
thing it is difficult to be precise about the explicit social structure. Certainly people carry round with them a consciousness of social structure. They curb their actions in accordance with the symmetries and hierarchies they see therein, and strive continually to impress their view of the relevant bit of structure on other actors in their scene. This social consciousness has been so well demonstrated by Goffman that there should be no need to labour the point further here. There are no items of clothing or of food or of other practical use which we do not seize upon as theatrical props to dramatise the way we want to present our roles and the scene we are playing in. Everything we do is significant, nothing is without its conscious symbolic load. Moreover, nothing is lost on the audience. Goffman uses dramatic structure, with its division of players and audience, stage and back-stage, to provide a frame for his analysis of everyday situations. Another merit of the analogy with theatre is that a dramatic structure exists within temporal divisions. It has a beginning, climax and end. For this reason Turner found it useful to introduce the idea of social drama to describe clusters of behaviour which everyone recognises as forming discrete temporal units (1957). I am sure that sociologists have not finished with the idea of drama as an image of social structure but for my purpose it may be enough to say that by social structure I am not usually referring to a total structure which embraces the whole of society continually and comprehensively. I refer to particular situations in which individual actors are aware of a greater or smaller range of indusiveness. In these situations they behave as if moving in patterned positions in relation to others, and as if choosing between possible patterns of relations. Their sense of form makes demands on their behaviour, governs their assessment of their desires, permits some and over-rides others.
Any local, personal view of the whole social system will not necessarily coincide with that of the sociologist. Sometimes in what follows, when I speak of social structure, I will be referring to the main outlines, lineages and the hierarchy of descent groups, or chiefdoms and the ranking of districts, relations between royalty and commoners. Sometimes I will be talking about little sub-structures, themselves chinese-box-like, containing others which fill in the bare bones of the main structure. It seems that individuals are aware in appropriate contexts of all these structures and aware of their relative importance. They do
100
Powers and Dangers
; not all have the same idea of what particular level of structure is relevant at a given moment; they know there is a problem
1 of communication to be overcome if there can be society at I all. By ceremony, speech and gesture they make a constant j effort to express and to agree on a view of what the relevant ;      social structure is like. And all the attribution of dangers and
I powers is part of this effort to communicate and thus to create :      social forms,
I The idea that there may be a correlation between explicit authority and controlled spiritual power was first suggested to me by Leach's article in Rethinking Anthropology. In develop-
' ing the idea I have taken a somewhat different direction. Controlled power to harm, he suggests, is often vested in explicit key points in the authority system, and contrasted with the unin-i      tentional power to harm supposed to lurk in the less explicit,
)       weakly articulated areas of the same society. He was mainly
I * concerned with the contrast of two kinds of spiritual power used I in parallel contrasting social situations. He presented some societies as sets of internally structured systems interacting with one
> another. Living within one such system people are explicitly conscious of its structure. Its key points are supported by beliefs in controlled forms of power attached to controlling positions. For instance, Chiefs among the Nyakusa can attack their foes by a kind of sorcery which sends invisible pythons after them. { Among the patrilineal Tallensi, a man's father has a correspondingly controlled right of access to ancestral power against him, and among the matrilineal Trobrianders the maternal uncle is thought to support his authority with consciously controlled spells and charms. It is as if the positions of authority were wired up with switches which can be operated by those who reach the right places in order to provide power for the system as a whole.
This can be argued along familiar Durkheimian lines. Religious beliefs express society's awareness of itself; the social structure is credited with punitive powers which maintain it in being. This is quite straightforward. But I would like to suggest that
! those holding office in the explicit part of the structure tend to be credited with consciously controlled powers, in contrast with those whose role is less explicit and who tend to be credited with unconscious, uncontrollable powers, menacing those in better defined positions. Leach's first example is the
'  i. 101
Purity and Danger
Kachin wife. Linking two power groups, her husband's and her brother's, she holds an interstructural role and she is thought of as the unconscious, involuntary agent of witchcraft. Similarly, the father in the matriiineal Trobrianders and Ashanti, and the mother's brother in patrilineal Tikopia and Taleland, is credited with being an involuntary source of danger. These people are none of them without a proper niche in the total society. But from the perspective of one internal sub-system to which they do not belong, but in which they must operate, they are intruders. They are not suspect in their own system and may be wielding the intentional kind of powers on its behalf. It is possible that their involuntary power to do harm may never be activated. It may lie dormant as they live their life peacefully in the corner of the sub-system which is their proper place, and yet in which they are intruders. But this role is in practice difficult to play coolly. If anything goes wrong, if they feel resentment or grief, then their double loyalties and their ambiguous status in the structure where they are concerned makes them appear as a danger to those belonging fully in it. It is the existence of an angry person in an interstitial position which is dangerous, and this has nothing to do with the particular intentions of the person.
In these cases the articulate, conscious points in the social structure are armed with articulate, conscious powers to protect the system; the inarticulate, unstructured areas emanate unconscious powers which provoke others to demand that ambiguity be reduced. When such unhappy or angry interstitial persons are accused of witchcraft it is like a warning to bring their rebellious feelings into line with their correct situation. If this were found to hold good more generally, then witchcraft, defined as an alleged psychic force, could also be defined structurally. It would be the anti-social psychic power with which persons in relatively unstructured areas of society are credited, the accusation being a means of exerting control where practical forms of control are difficult. Witchcraft, then, is found in the non-structure. Witches are social equivalents of beetles and spiders who live in the cracks of the walls and wainscoting. They attract the fears and dislikes which other ambiguities and contradictions attract in other thought structures, and the kind of powers attributed to them symbolise their ambiguous, inarticulate status.
toa
Powers and Dangers
Pondering on this line of thought, we can distinguish different types of social inarticulateness. So far we have only considered witches who have a well-defined position in one sub-system and an ambiguous one in another, in which they none the less have duties. They are legitimate intruders. Of these Joan of Arc can be taken as a splendid prototype: a peasant at court, a woman in armour, an outsider in the councils of war; the accusation that she was a witch puts her fully in this category.
But witchcraft is often supposed to operate in another kind of ambiguous social relation. The best example comes from the witchcraft beliefs of the Azande. The formal structure of their society was pivoted on princes, their courts, tribunals and armies, in a clear cut hierarchy down to princes' deputies, through local governors, to heads of families. The political system afforded \ an organised set of fields for competition, so that commoners did not find themselves in competition with nobles, nor poor against rich, nor sons against fathers, nor women against men. Only in those areas of society which were left unstructured by the political system did men accuse each other of witchcraft. A man who had defeated a close rival in competition for office might accuse the other of bewitching him in jealousy, and co-wives might accuse one another of witchcraft. Azande witches were thought to be dangerous without knowing it; their witchcraft was made active simply by their feelings of resentment or grudge. The accusation attempted to regulate the situation by vindicating one and condemning the other rival. Princes were supposed not to be witches, but they accused one another of sorcery, thus conforming to the pattern I am seeking to estalv ltsh.
Another type of unconscious power to harm emanating from inarticulate areas of the social system is illustrated by the Man-dari, whose land-owning clans build up their strength by adopting clients. These unfortunates have, for one reason or another, lost their claim to their own territory and have come to a foreign territory to ask for protection and security. They are landless, inferior, dependent on their patron who is a member t of a land-owning group. But they are not completely dependent. To some real extent the patron's influence and status depend on his loyal following of clients. Clients who become too numerous and bold can threaten their patron's lineage. The explicit struc-\    ture of society is based on land-holding clans. By these people
Purity and Danger
clients are held likely to be witches. Their witchcraft emanates from jealousy of their patrons and works involuntarily. A witch cannot control himself, it is his nature to be angry and harm emanates from him. Not all clients are witches, but hereditary lines of witches are recognised and feared. Here are people living in the interstices of the power structure, felt to be a threat to those with better defined status. Since they are credited with dangerous, uncontrollable powers, an excuse is given for suprcss-ing them. They can be charged with witchcraft and violently despatched without formality or delay. In one case the patron's family merely made ready a big fire, called in the suspect witch to share a meal of roast pig, and forthwith bound him and put him on the fire. Thus the formal structure of land-holding lineages was asserted against the relatively fluid field in which landless clients touted for patronage.
Jews in English society are something like Mandari clients. Belief in their sinister but undefinable advantages in commerce justifies discrimination against them—whereas their real offence is always to have been outside the formal structure of Christendom.
There are probably many more variant types of socially ambiguous or weakly defined statuses to which involuntary witchcraft is attributed. It would be easy to go on piling up examples. Needless to say, I am not concerned with beliefs of a secondary kind or with short-lived ideas which flourish briefly and die away. If the correlation were generally to hold good for the distribution of dominant, persistent forms of spiritual power, it would clarify the nature of pollution. For, as I see it, ritual pollution also arises from the interplay of form and surrounding formlessness. Pollution dangers strike when form has been attacked. Thus we would have a triad of powers controlling fortune and misfortune: first, formal powers wielded by persons representing the formal structure and exercised on behalf of the formal structure: second, formless powers wielded by interstitial persons: third, powers not wielded by any person, but inhering in the structure, which strike against any infraction of form. This three-fold scheme for investigating primitive cosmologies unfortunately comes to grief over exceptions which are too important to brush aside. One big difficulty is that sorcery, which is a form of controlled spiritual power, is in many parts of the world credited to persons who ought, according to
104
Powers and Dangers
my hypothesis, be charged with involuntary witchcraft. Malevolent persons in interstitial positions, anti-social, disapproved, working to harm the innocent, they should not be using conscious, controlled, symbolic power. Furthermore, there are royal chiefs who emanate unconscious, involuntary power to detect disaffection and destroy their enemies—chiefs who according to my hypothesis should be content with explicit, controlled forms of power. So the correlation I have tried to draw does not hold. However, I will not throw it aside until I have looked more closely at the negative cases.
One reason why it is difficult to correlate social structure with type of mystic power is that both elements in the comparison are very complex. It is not always easy to recognise explicit authority. For example, authority among the Lele is very weak, their social system makes a criss-cross of little authorities, none very effective in secular terms. Many of their formal statuses are supported by the spiritual power to curse or bless, which consists in uttering a form of words and spitting. Cursing and blessing are attributes of authority; a father, mother, mother's brother, aunt, pawn owner, village head and so on, can curse. Not any 'i     one can reach out for a curse and apply it arbitrarily. A son \     carmot curse his father, it would not work if he tried. So this pattern conforms to the general rule I am seeking to establish. f     But, if a person who has a right to curse refrains from formulat-
iing his curse, the unspit saliva in his mouth is held to have power to cause harm. Better than harbour a secret grudge, anyone with a just grievance should speak up and demand redress, lest the saliva of his ill-will do harm secretly. In this belief we have both the controlled and uncontrolled spiritual power attributed to the same person in the same circumstances. But as their pattern of authority is so weakly articulated, this is hardly a negative case. On the contrary, it serves to warn us that authority can be a very vulnerable power, easily reduced to nothing. We should be prepared to elaborate the hypothesis to take more account of the varieties of authority.
There are several likenesses between the unspoken curse of the Lele and the witchcraft beliefs of the Mandari. Both are tied to a particular status, both are psychic, internal, involuntary. But the unspoken curse is an approved form of spiritual power, while ( the witch is disapproved. Where the unspoken curse is revealed as the cause of harm restitution is made to the agent, when
Purity and Danger witchcraft is reveaJed the agent is brutally attacked. So the unspoken curse is on the side of authority; its link with cursing makes this clear. But authority is weak in the case of the Lele, strong in the case of the Mandari. This suggests that to test the hypothesis fairly we should display the whole gamut from no formal authority at one end of the scale to strong effective secular authority at the other end. At either extreme I am not prepared to predict the distribution of spiritual powers, because where there is no formal authority the hypothesis does not apply, and where authority is firmly established by secular means it less requires spiritual and symbolic support. Under primitive conditions authority is always likely to be precarious. For this reason we should be ready to take into account the failure of those in office.
First consider the case of the man in a position of authority who abuses the secular powers of his office. If it is clear that he is acting wrongly, out of role, he is not entitled to the spiritual power which is vested in the role. Then there should be scope for some shift in the pattern of beliefs to accommodate his defection. He ought to enter the class of witches, exerting involuntary, unjust powers instead of intentionally controlled powers against wrongdoers. For the official who abuses his office is as illegitimate as an usurper, an incubus, a spanner in the works, a dead weight on the social system. Often we find this predicted shift in the kind of dangerous power he is supposed to wield.
In the Book of Samuel, Saul is presented as a leader whose divinely given powers are abused. When he fails to fill his assigned role and leads his men into disobedience, his charisma leaves him and terrible rages, depression and madness afflict him. So when Saul abuses his office he loses conscious control and becomes a menace even to his friends. With reason no longer in control, the leader becomes an unconscious danger. The image of Saul fits the idea that conscious spiritual power is vested in the explicit structure and uncontrolled unconscious danger vested in the enemies of the structure.
The Lugbara have another and similar way of adjusting their beliefs to abuse of power. They credit their lineage elders with special powers to invoke the ancestors against juniors who do not act in the widest interests of the lineage. Here again we have conscious controlled powers upholding the explicit structure.
106
Powers and Dangers
But if an elder is thought to be motivated by his own personal, ' selfish interests, the ancestors neither listen to him nor put their power at his disposal. So here is a man in a position of authority, \ jr. improperly wielding the powers of office. His legitimacy being W in doubt, he must be removed, and to remove him his antagonists m accuse him of having become corrupt and emanating witchcraft, H a mysterious, perverted power which operates at night (Middle-It ton). The accusation is itself a weapon for clarifying and H strengthening the structure. It enables guilt to be pinned on if the source of confusion and ambiguity. So these two examples 'f symmetrically develop the idea that conscious power is exerted f;      from the key positions in the structure and a different danger
from its dark, obscure areas. I I        Sorcery is another matter. As a form of harmful power which !      makes use of spells, words, actions and physical materials, it can ♦       only be used consciously and deliberately. On the argument we have been following, sorcery ought to be used by those in control of key positions in the social structure as it is a deliberate, controlled form of spiritual power. But it is not. Sorcery is found |>      in the structural interstices where we have located witchcraft, as 'S      well as in the seats of authority. At first glance it seems to cut ! I      across the correlation of articulate structure with consciousness. I      But on closer inspection this distribution of sorcery is consistent with the pattern of authority that goes with sorcery beliefs.
I* In some societies positions of authority are open to competi-i tion. Legitimacy is hard to establish, hard to maintain and always liable to reversal. In such very fluid political systems we I would expect a particular type of beliefs in spiritual power. I Sorcery is unlike cursing and invocation of ancestors in that |: it has no built-in device to safeguard against abuse. Lugbara I cosmology, for example, is dominated by the idea of the an-1 cestors upholding lineage values; the Israelite cosmology was I; dominated by the idea of the justice of Jehovah. Both these 1 sources of power contain an assumption that they cannot be I deceived or abused. If an incumbent of office misuses his power, | spiritual support is withdrawn. By contrast, sorcery is essentially f     a form of controlled and conscious power that is open to abuse.
In the Central African cultures, where sorcery beliefs flourish, I      this form of spiritual power is developed within the idiom of medicine. It is freely available. Anyone who takes the trouble to acquire sorcery power may use it. In itself it is morally and
•°7
Purity and Danger
socially neutral and it contains no principle for safeguarding against abuse. It works ex opere operato, equally well whether the intentions of the agent are pure or corrupt. If the idea of spiritual power in the culture is dominated by this medical idiom, the man who abuses his office and the person in the unstructured crevices have the same access to the same kind of spiritual powers as the lineage or village head. It follows that if sorcery is available to anyone who wants to acquire it, then we should suppose that positions of political control are also available, open to competition, and that in such societies there are not very clear distinctions between legitimate authority, abuse of authority and illegitimate rebellion.
The sorcery beliefs of Central Africa, west to east from the Congo to Lake Nyasa, assume that malign spiritual powers of sorcery are generally available. In principle these powers are vested in the heads of matrilineal descent groups and are expected to be used by these men in authority against enemy outsiders. There is a general expectation that the old man may turn his powers against his own followers and kin. and if he is disagreeable or mean, their deaths are likely to be attributed to him. He always risks being dragged down from his little elevation of senior status, degraded, exiled or put to the poison ordeal (Van Wing, p. 359-60, Kopytoff, p. 90). Then another contender will take his official role and try to exercise it more warily. Such beliefs, as I have tried to show in my study of the Lele, correspond to a social system in which authority is weakly denned and has little real sway (1963). Marwick has claimed for similar beliefs among the Cewa that they have a liberating effect, since any young man can plausibly accuse of sorcery a reactionary old incumbent of an office which he himself is qualified to occupy when the senior obstacle has been removed (1952). If sorcery beliefs really serve as instruments for self-promotion they also ensure that the ladder of promotion is short and shaky.
The fact that anyone may lay hands on sorcery power and that it is available for use against, or on behalf of society suggests another cross-classification of spiritual powers. For in Central Africa sorcery is often a necessary adjunct to roles of authority. The mother's brother must be acquainted with sorcery to be able to combat enemy sorcerers and to protect his descendants. It is a double-edged attribute, for if he uses it unwisely he can be ruined. Thus there is always the possibility,
toS
Powers and Dangers
even the expectation, that the man in an official position will fail to fill it creditably. The belief acts as a check on the use of secular power. If a leader among the Cewa or Lele becomes unpopular the sorcery beliefs contain an escape clause enabling his dependents to get rid of him. This is how I read the Tsav beliefs of the Tiv, checking as much as validating the eminent lineage elder's authority (Bohannan). So freely available sorcery is a form of spiritual power biased towards failure. This is a cross-classification which puts witchcraft and sorcery in the same bracket. Witchcraft beliefs are also tilted to expect role failure and to deal with it punitively, as we have seen. But witchcraft beliefs expect failure in interstitial roles, while sorcery beliefs expect failure in official roles. The whole scheme in which spiritual powers are correlated with structure becomes more consistent if we contrast those powers which are biased towards failure with powers which are biased towards success.
Teutonic notions of Luck, and some forms of baraka and mana are success-biased beliefs which parallel sorcery as a failure-biased belief. Mana and Islamic baraka exude from official positions, regardless of the intention of the incumbent. They are either dangerous powers to strike or benign powers for good. There are chiefs and princes exerting mana or baraka whose merest contact is worth a blessing and a guarantee of success, and whose personal presence makes the difference between victory and defeat in battle. But these powers are not always so well anchored to the outlines of the social system. Sometimes baraka can be a free-floating benign power, working independently of the formal distribution of power and allegiance in society.
If we find such free-lance benign contagion playin:* an important role in people's beliefs, we can expect either that formal authority is weak or ill-defined or that, for one reason or another, the political structure has been neutralised so that the powers of blessing cannot emanate from its key points.
Dr. Lewis has described an example of an un-sacralised social structure. In Somalitand there is a general division in thought between secular and spiritual power (1963). In secular relations power derives from fighting strength and the Somali are militant and competitive. The political structure is a warrior system where might is right. But in the religious sphere the Somalia are Muslims and hold that fighting within the Muslim com-
109
Purity and Danger
munity is wrong. These deeply held beliefs de-ritualisc the social structure so that Somali do not claim that divine blessings or dangers emanate from its representatives. Religion is represented not by warriors but by men of God. These holy men, religious and legal experts, mediate between men as they mediate between men and God. They are only reluctantly involved in the warrior structure of society. As men of God they are credited with spiritual power. It follows that their blessing (baraka) is great in proportion as they withdraw from the secular world and are humble, poor and weak.
If this argument is correct it should apply to other Islamicised peoples whose social organisation is based on violent internal conflict. However the Moroccan Berbers exhibit a similar distribution of spiritual power without the theological justification. Professor Gellner tells me that Berbers have no notion that fighting within the Moslem community is wrong. Moreover it is a common feature of competitive segmentary political systems that the leaders of the aligned forces enjoy less credit for spiritual power than certain persons in the interstices of political alignment. The Somali holy man should be seen as the counterpart of the Tallensi Earth shrine priest and the Nuer Man of the Earth. The paradox of spiritual power vested in the physically weak is explained by social structure rather than by the local doctrine which justifies it. (Fortes and Evans-Pritchard, 1940, p.22).
Baraka in this form is something like witchcraft in reverse. It is a power which does not belong to the formal political structure, but which floats between its segments. As witchcraft accusations are used to reinforce the structure, so do people in the structure try to make use of baraka. Like witchcraft and sorcery its existence and strength is proved empirically, post hoc. A witch or sorcerer is identified when a misfortune occurs to someone against whom he has a grudge. The misfortune indicates there is witchcraft at work. The known grudge indicates the possible witch. It is his reputation for quarrels which essentially focusses the charge against him. Baraka is also identified empirically, post hoc. A piece of marvellous good fortune indicates its presence, often quite unexpectedly (Westermarck, I, chapter II). The reputation of a holy man for piety and learning focusses interest on him. Just as the witch's bad name will get worse with every disaster that befalls her neighbours, so the saint's
110
Powers and Dangers
good name will improve with every stroke of good fortune. The snow-ball effect is the same.
The failure-biased powers have a negative feed-back. If anyone potentially possessing them tries to get above himself, the accusation cuts him down to size. The fear of accusation works like a thermostat on everyone in advance of actual quarrels. It is a control device. But the success-biased powers have the possibility of a positive feed-back. They could build up and up indefinitely to an explosion. As witchcraft has been called institutionalised jealousy, so baraka can work as institutionalised admiration. For this reason it is self-validating when it works in a freely competitive system. It is on the side of the big battalions. Empirically tested by success, it attracts adherents and so earns more success. 'People in fact become possessors of baraka by being treated as possessors of it.' (Gellner 1962).
I should make it clear that I do not believe that baraka is always available to competing elements in tribal social systems. It is an idea about power which varies in different political conditions. In an authoritative system it can emanate from the holders of authority and validate their established status, to the discomfiture of their foes. But it also has the potentiality of disrupting ideas about authority and about right and wrong, since its only proof lies in its success. The possessor of baraka is not subject to the same moral restraints as other persons (Westermarck, I, p. 198). The same applies to Mana and Luck. They can be on the side of established authority or on the side of opportunism. Raymond Firth came to the conclusion that at least in Tikopia, Mana means success (1940). Tikopian Mana expresses the authority of hereditary chiefs. Firth reflected on whether the dynasty would be endangered if the chief's reign were not a fortunate one, and concluded (correctly as it happened) that the chiefship would be strong enough to ride such a storm. One of the great advantages of doing sociology in a teacup is to be able to discern calmly what would be confusing in a larger scene. But it is a drawback not to be able to observe any real storms and upheavals. In a sense all colonial anthropology takes place in a teacup. If mana means success it is an apt concept for political opportunism. The artificial conditions of colonial peace may have disguised this potential for conflict and rebellion which the success-biased powers imply. Anthropology has often been weak in political analysis. The equivalent of a
111
Purity and Danger
paper constitution without any dust or conflict or serious estimate of the balance of forces is sometimes offered in lieu of an analysis of a political system. This must necessarily obscure interpretation. It may be helpful to turn to a pre-colonial example.
Luck, for our Teutonic ancestors, like the opportunist or freelance forms of mana and baraka, also seems to have operated freely in a competitive political structure, fluid, with little in the way of hereditary power. Such beliefs can follow swift changes in the lines of allegiance, and change judgments of right and wrong.
I have tried to push as far as possible the parallel between these success-biased powers and witchcraft and sorcery, both failure-biased and both capable of operating independently of the distribution of authority. Another parallel with witchcraft is in the involuntary nature of these success forces. A man discovers he has baraka because of its effects. Many men may be pious and live outside the warrior system, but not many have great baraka. Mana too may be exerted quite unconsciously, even by the anthropologist, as Raymond Firth wryly recounts when a magnificent haul of fish was attributed to his mana. The Sagas of the Norsemen are full of crises resolved when a man suddenly discovers his Luck or finds that his Luck has deserted him (Gronbech, Vol. I, ch. 4).
Another characteristic of success power is that it is often contagious. It is transmitted materially. Anything which has been in contact with baraka may get baraka. Luck was also transmitted partly in heirlooms and treasures. If these changed hands, Luck changed hands too. In this respect these powers are like pollution, which transmits danger by contact. However, the potentially haphazard and disruptive effects of these success powers contrasts with pollution, austerely committed to support the outlines of the existing social system.
To sum up, beliefs which attribute spiritual power to individuals are never neutral or free of the dominant patterns of social structure. If some beliefs seem to attribute free-floating spiritual powers in a haphazard manner, closer inspection shows consistency. The only circumstances in which spiritual powers seem to flourish independently of the formal social system are when the system itself is exceptionally devoid of formal structure, when legitimate authority is always under challenge or when the rival segments of an acephalous political system resort to media-
Powers and Dangers
tion. Then the main contenders for political power have to court for their side the holders of free-floating spiritual power. Thus it is beyond doubt that the social system is thought of as quick with creative and sustaining powers.
Now is the time to identify pollution. Granted that all spiritual powers are part of the social system. They express it and provide institutions for manipulating it. This means that the power in the universe is ultimately hitched to society, since so many changes of fortune are set off by persons in one kind of social position or another. But there are other dangers to be reckoned with, which persons may set off knowingly or unknowingly, which are not part of the psyche and which are not to be bought or learned by initiation and training. These are pollution powers which inhere in the structure of ideas itself and which punish a symbolic breaking of that which should be joined or joining of that which should be separate. It follows from this that pollution is a type of danger which is not likely to occur except where the lines of structure, cosmic or social, are clearly denned.
A polluting person is always in the wrong. He has developed some wrong condition or simply crossed some line which should not have been crossed and this displacement unleashes danger for someone. Bringing pollution, unlike sorcery and witchcraft, is a capacity which men share with animals, for pollution is not always set off by humans. Pollution can be committed intentionally, but intention is irrelevant to its effect—it is more likely to happen inadvertently.
This is as near as I can get to defining a particular class of dangers which are not powers vested in humans, but which can be released by human action. The power which presents a danger for careless humans is very evidently a power inhering in the structure of ideas, a power by which the structure is expected to protect itself.
External Boundaries
The idea of society is a powerful image. It is potent in its own right to control or to stir men to action. This image has form; it has external boundaries, margins, internal structure. Its outlines contain power to reward conformity and repulse attack. There is energy in its margins and unstructured areas. For symbols of society any human experience of structures, margins or boundaries is ready to hand.
Van Gennep shows how thresholds symbolise beginnings of new statuses. Why does the bridegroom carry his bride over the lintel? Because the step, the beam and the door posts make a frame which is the necessary everyday condition of entering a house. The homely experience of going through a door is able to express so many kinds of entrance. So also are cross roads and arches, new seasons, new clothes and the rest. No experience is too lowly to be taken up in ritual and given a lofty meaning. The more personal and intimate the source of ritual symbolism, the more telling its message. The more the symbol is drawn from the common fund of human experience, the more wide and certain its reception.
The structure of living organisms is better able to reflect complex social forms than door posts and lintels. So we find that the rituals of sacrifice specify what kind of animal shall be used, young or old, male, female or neutered, and that these rules signify various aspects of the situation which calls for sacrifice. The way the animal is to be slaughtered is also laid down. The Dinka cut the beast longitudinally through the sexual organs if the sacrifice is intended to undo an incest; in half across
"4
External Boundaries
the middle for celebrating a truce; they suffocate it for some occasions and trample it to death for others. Even more direct is the symbolism worked upon the human body. The body is a model which can stand for any bounded system. Its boundaries can represent any boundaries which are threatened or precarious. The body is a complex structure. The functions of its different parts and their relation afford a source of symbols for other complex structures. We cannot possibly interpret rituals concerning excreta, breast milk, saliva and the rest unless we are prepared to see in the body a symbol of society, and to see the powers and dangers credited to social structure reproduced in small on the human body.
It is easy to see that the body of a sacrificial ox is being used as a diagram of a social situation. But when we try to interpret rituals of the human body in the same way the psychological tradition turns its face away from society, back towards the individual. Public rituals may express public concerns when they use inanimate door posts or animal sacrifices: but public rituals enacted on the human body are taken to express personal and private concerns. There is no possible justification for this shift of interpretation just because the rituals work upon human flesh. As far as I know the case has never been methodically stated. Its protagonists merely proceed from unchallenged assumptions, which arise from the strong similarity between certain ritual forms and the behaviour of psychopathic individuals. The assumption is that in some sense primitive cultures correspond to infantile stages in the development of the human psyche. Consequently such rites are interpreted as if they express the same preoccupations which fill the mind of psychopaths or infants.
Let me take two modern attempts to use primitive cultures to buttress psychological insights. Both stem from a long line of similar discussions, and both are misleading because the relation between culture and individual psyche are not made clear.
Bettelheim's Symbolic Wounds is mainly an interpretation of circumcision and initiation rites. The author tries to use the set rituals of Australians and Africans to throw light on psychological phenomena. He is particularly concerned to show that psychoanalysts have over-emphasised girls' envy of the male sex and overlooked the importance of boys' envy of the female •ex. The idea came to him originally in studying groups of
tic
Purity and Danger
schizophrenic children approaching adolescence. It seems very likely that the idea is sound and important. I am not at all claiming to criticise his insight into schizophrenia. But when he argues that rituals which are explicitly designed to produce genital bleeding in males are intended to express male envy of female reproductive processes, the anthropologist should protest that this is an inadequate interpretation of a public rite. It is inadequate because it is merely descriptive. What is being carved in human flesh is an image of society. And in the moiety- and section-divided tribes he cites, the Murngin and Arunta, it seems more likely that the public rites are concerned to create a symbol of the symmetry of the two halves of society.
The other book is Life against Death, in which Brown outlines an explicit comparison between the culture of 'archaic man' and our own culture, in terms of the infantile and neurotic fantasies which they seem to express. Their common assumptions about primitive culture derive from Roheim (1925): primitive culture is autoplastic, ours is alloplastic. The primitive seeks to achieve his desires by self-manipulation, performing surgical rites upon his own body to produce fertility in nature, subordination in women or hunting success. In modern culture we seek to achieve our desires by operating directly on the external environment, with the impressive technical results that are the most obvious distinction between the two types of cultures. Bettelheim adopts this summing up of the difference between the ritual and the technical bias in civilisation. But he supposes that the primitive culture is produced by inadequate, immature personalities, and even that the psychological shortcomings of the savage accounts for his feeble technical achievements :
'If preliterate peoples had personality structures as complex as those of modern man, if their defences were as elaborate and their consciences as refined and demanding; if the dynamic interplay between ego, superego and id were as intricate and if their ego's were as well adapted to meet and change external reality —they would have developed societies equally complex, though probably different. Their societies have, however, remained small and relatively ineffective in coping with the external environment. It may be that one of the reasons for this is their tendency to try to solve problems by autoplastic rather than alloplastic manipulation.' {p. 87)
Let us assert again, as many anthropologists have before, that
116
External Boundaries
there are no grounds for supposing that primitive culture as such is the product of a primitive type of individual whose personality resembles that of infants or neurotics. And let us challenge the psychologists to express the syllogisms on which such a hypothesis might rest. Underlying the whole argument is the assumption that the problems which rituals are intended to solve are personal psychological problems. Bettelheim actually goes on to compare the primitive ritualist with the child who hits his own head when frustrated. This assumption underlies his whole book.
Brown makes the same assumption, but his reasoning is more subtle. He does not suppose that the culture's primitive condition is caused by individual personal traits: he allows very properly for the effect of cultural conditioning on the individual personality. But he proceeds-to consider the total culture as if it, in its totality, could be compared to an infant or a retarded adult. The primitive culture resorts to bodily magic to achieve its desires. It is in a stage of cultural evolution comparable to that of infantile anal eroticism. Starting from the maxim:
'Infantile sexuality is autoplastic compensation for the loss of the Other; sublimation is alloplastic compensation for loss of Self.' (p. 170)
he goes on to argue that 'archaic' culture is directed to the same ends as infantile sexuality, that is escape from the hard realities of loss, separation and death. Epigrams are, by their nature, obscure. This is another approach to primitive culture which I would like to see fully spelt out. Brown develops the theme only briefly, as follows:
'Archaic man is preoccupied with the castration complex, the incest taboo and the desexualisation of the penis, that is, the transference of the genital impulses into that aim-inhibited libido which sustains the kinship systems in which archaic life is embedded. The low degree of sublimation, corresponding to the low level of technology, means by our previous definirions, a weaker ego, an ego which has not yet come to terms (by negation) with its own pregenital impulses. The result is that all the fantastic wishes of infantile narcissism express themselves in unsublimated form so that archaic man retains the magic body of infancy.' (p. 298-9)
These fantasies suppose that the body itself could fulfil the
"7
Purity and Danger infant's wish for unending, self-replenishing enjoyment. They are a flight from reality, a refusal to face loss, separation and death. The ego develops by sublimating these fantasies. It mortifies the body, denies the magic of excrement and to that extent faces reality. But sublimation substitutes another set of unreal aims and ends by providing the self with another kind of false escape from loss, separation and death. This is how I understand the argument to run. The more material that an elaborate technology imposes between ourselves and the satisfaction of our infantile desires, the more busily has sublimation been at work. But the converse seems questionable. Can we argue that the less the material basis of civilisation is developed, the less sublimation has been at work? What precise analogy with infantile fantasy can be valid for a primitive culture based on a primitive technology? How does a low level of technology imply *an ego which has not yet come to terms (by negation) with its own pregenital impulses*? In what sense is one culture more sublimated than another?
These are obviously technical questions in which the anthropologist cannot engage. But on two points the anthropologist has something to say. One is the question of whether primitive cultures really can be said to revel in cxcremcntal magic. The answer to this is surely No. The other is whether primitive cultures seem to be seeking an escape from reality. Do they really use their magic, excremental or other, to compensate for loss of success in external fields of endeavour? Again the answer is No.
To take the matter of excremental magic first. The information is distorted, first as to the relative emphasis on bodily as distinct from other symbolic themes, and second as to the positive or negative attitudes to bodily refuse seen in primitive ritual.
To take up the latter point first: the use of excrement and other bodily exuviae in primitive cultures is usually inconsistent with the themes of infantile erotic fantasy. So far from excrement, etc., being treated as a source of gratification, its use tends to be condemned. So far from being thought of as an instrument of desire, the power residing in the margins of the body is more often to be avoided. There are two main reasons why casual reading in ethnography gives the wrong impression. The first is an informant's bias and the second an observer's bias.
Sorcerers are supposed to use bodily refuse in pursuing (heir nefarious desires. Certainly in this sense excremental magic
118
External Boundaries
ministers to its user's desires, but information about sorcery is usually given from the alleged victim's point of view. Vivid accounts of the materia medica of sorcery can always be had from supposed victims. But recipe books of charms dictated by confessed sorcerers are rarer. It is one thing to suspect that others are using bodily refuse unlawfully against oneself, but this does not mean that informants think of these materials as available for their own use. So a kind of optical delusion makes what often belongs on the negative side of the balance sheet appear on the positive side.
The observer's bias also exaggerates the extent to which primitive cultures make positive magical use of bodily relicts. For various reasons best known to psychologists, any reference to excremental magic seems to leap to the reader's eye and absorb attention. Thus a second distortion is introduced. The full richness and range of symbolism tends to be overlooked, or assimilated to a few scatologic principles. Take as an illustration of this bias Brown's own discussion of the Trickster Myth of the Winnebago Indians which we mentioned in Chapter 3. Anal topics occur only two or three times in the course of the long series of Trickster's adventures. I quoted one of these occasions, where Trickster tried to treat his anus as a separate person. Brown's impression of the myth is so different that at first I mistakenly thought he had gone back in erudite fashion to a more primary source than Radin's when he said that:
The Trickster of primitive mythologies is surrounded by unsublimated and undisguised anality.'
According to Brown the Winnebago Trickster, who is also a great culture hero, 'can create the world by a filthy trick out of excrement, mud, clay'. He cites as example an episode in the myth in which Trickster defies a warning not to eat some bulb which fills his belly with wind, each eruption of which lifts him higher and higher. He calls the humans to hold him down, but in thanks for their attempt to help him in a last final eruption he scatters them all far and wide. Search the story as told by Radin in vain for any sign that Trickster's defaecation is creative in any way. It is rather destructive. Search Radin's glossary and introduction and learn that Trickster did not create the world and is not in any sense a culture hero. Radin considers the quoted episode to have an altogether negative moral, and one consistent
119
Purity and Danger
with the theme of Trickster's gradual development as a social being. So much for the bias which reads too much excremental magic into primitive cultures.
The next point concerning cultural parallels with anal eroticism is to ask in what sense any primitive cultures are in flight from the realities of separation and loss. Do they try to ignore the unity of death and life? On the contrary, my impression is that those rituals which most explicitly credit corrupt matter with power are making the greatest effort to affirm the physical fullness of reality. So far from using bodily magic as an escape, cultures which frankly develop bodily symbolism may be seen to use it to confront experience with its inevitable pains and losses. By such themes they face the great paradoxes of existence, as I shall show in the last chapter. Here I only touch on the subject briefly because it bears on the parallel with infantile psychology as follows: insofar as ethnography supports the idea that primitive cultures treat dirt as a creative power it contradicts the idea that these cultural themes can be compared with the fantasies of infantile sexuality.
To correct the two distortions of evidence to which this subject is prone we should classify carefully the contexts in which body dirt is thought of as powerful. It may be used ritually for good, in the hands of those vested with power to bless. Blood, in Hebrew religion, was regarded as the source of life, and not to be touched except in the sacred conditions of sacrifice. Sometimes the spittle of persons in key positions is thought effective to bless. Sometimes the cadaver of the last incumbent yields up material for anointing his royal successor. For example, the decayed corpse of the last Lovedu queen in the Drakensberg mountains is used to concoct unguents which enable the current queen to control the weather (Krige, pp. 273-4). These examples can be multiplied. They repeat the analysis in the previous chapter of the powers attributed to the social or religious structure for its own defence. The same goes for body dirt as ritual instrument of harm. It may be credited to the incumbents of key positions for defending the structure, or to sorcerers abusing their positions in the structure, or to outsiders hurling bits of bone and other stuff at weak points in the structure.
But now we are ready to broach the central question. Why should bodily refuse be a symbol of danger and of power? Why should sorcerers be thought to qualify for initiation by shedding
120
External Boundaries
blood or committing incest or anthropophagy? Why, when initiated, should their art consist largely of manipulating powers thought to inhere in the margins of the human body? Why should bodily margins be thought to be specially invested with power and danger?
First, we can rule out the idea that public rituals express common infantile fantasies. These erotic desires which it is said to be the infant's dream to satisfy within the body's bounds are presumably common to the human race. Consequently body symbolism is part of the common stock of symbols, deeply emotive because of the individual's experience. But rituals draw on this common stock of symbols selectively. Some develop here, others there. Psychological explanations cannot of their nature account for what is culturally distinctive.
Second, all margins are dangerous. If they are pulled this way or that the shape of fundamental experience is altered. Any structure of ideas is vulnerable at its margins. We should expect the orifices of the body to symbolise its specially vulnerable points. Matter issuing from them is marginal stuff of the most obvious kind. Spittle, blood, milk, urine, faeces or tears by simply issuing forth have traversed the boundary of the body. So also have bodily parings, skin, nail, hair clippings and sweat. The mistake is to treat bodily margins in isolation from all other margins. There is no reason to assume any primacy for the individual's attitude to his own bodily and emotional experience, any more than for his cultural and social experience. This is the clue which explains the unevenness with which different aspects of the body are treated in the rituals of the world. In some, menstrual pollution is feared as a lethal danger; in others not at all (see Chapter 9). In some, death pollution is a daily preoccupation; in others not at alt. In some, excreta is dangerous, in others it is only a joke. In India cooked food and saliva are pollution-prone, but Bushmen collect melon seeds from their mouths for later roasting and eating (Marshall Thomas, p. 44).
Each culture has its own special risks and problems. To which particular bodily margins its beliefs attribute power depends on what situation the body is mirroring. It seems that our deepest fears and desires take expression with a kind of witty aptness. To understand body pollution we should try to argue back from the known dangers of society to the known selection of bodily themes and try to recognise what appositeness is there.
121