A Reader in the Anthropology of Religion
Edited by Michael Lambek
jfk Blackwell *C/ Publishing
c /,     1.1.   I /•
10
The Problem of Symbols
E. E. Evans-Pritchard
From E. E. Evans-Pritchard, "The Problem of Symbols," in Nuer Religion (Oxford: Clarendon Press, 1956), pp. 123-43. Reprinted by permission of Oxford University Press. Abridged.
E. E. Evans-Pritchard (1902-73), or "E.-P.," as he was known, was arguably the most important anthropologist of his generation, certainly in the United Kingdom (he was professor at the University of Oxford). He was a superb fieldworker and theoretician; as in the chapter from Nuer Religion reproduced here, his theory is never far separated from his ethnographic exposition. Trained by Malinowski and influenced by Radcliffe-Brown, he was an original and independent thinker who shifted from a primarily structure-functionalist to a more historical and humanistic position.
While many anthropologists have taken religious symbols as complex variants of ordinary symbols as defined by Langer, thus as "packets" of meaning that could be unpacked (see chapters 11 and 12, respectively, as well as Turner 1967, for notable unpackings), Evans-Pritchard is unusual in following more clearly the point made by Langer that the relevant question is one of predication. In the selection from Nuer Religion reprinted here, Evans-Pritchard looks less at the content of the symbol (for which see the profound analysis in his preceding chapter on the refractional nature of Nuer deity, Kwoth) than on the meaning of the verb that establishes it as a predicate. He is concerned, then, with what it means for Nuer to say that twins are birds or what it means when Nuer replace cattle with cucumbers in their sacrifices. The question is thus not the meaning of twins, birds, oxen, or cucumbers qua symbols but what the Nuer mean when they say and do certain things about and with twins, birds, oxen, and cucumbers. In this very subtle essay Evans-Pritchard thus takes us to the complexities of religious meaning and the intricate gradations between literal and metaphorical predication.
It is sometimes remarked that Evans-Pritchard portrays Nuer religion in a manner not entirely dissimilar to Roman Catholicism, or rather that the arguments he makes to explain Nuer refractions of spirit or the material symbolism of spirit are not altogether different from the way in which a Catholic theologian might defend the concept of the
146
E. E. EVANS-PRITCHARD
Trinity or the Communion. This is not necessarily a criticism; some people have gone on to make the point that anthropologists who practice religion can have a better appreciation of the religion of their subjects than the agnostic majority. While this may be true in the case of Evans-Pritchard, I doubt the general validity of the argument. Moreover, while a religious sensibility is certainly evident in this essay, more to the point is Evans-Pritchard's philosophical acuity and ethnographic precision. I would, however, make the broader hermeneutic point that any anthropological interpretation will be informed by the position of the anthropologist, not only by the ethnographer's own religious formation but by the context of the debate over religious, philosophical, or political issues in which the anthropologist is immersed within his or her own time and tradition. Evans-Pritchard's professional debate, as we glimpse toward the end of his essay, was with Tylor, Levy-Bruhl, and others who made arguments that denigrated eitherthe conclusions or the very rationality of "primitive thought." As discussed above, Evans-Pritchard's oeuvre on religion is a masterly refutation of these arguments.
In the last chapter I discussed how the Nuer conception of Spirit is figured in different ways to different persons and categories and groups. In this chapter I consider the material forms in which Spirit manifests itself or is represented. God is, properly speaking, not figured in any material representations, nor are almost all the spirits of the above, though both God and his supra-terrestrial refractions may reveal themselves in signs. But the spirits of the below are represented in creatures and things. Our problem chiefly concerns these spirits of the below. It can be simply stated by the question: What meaning are we to attach to Nuer statements that such-and-such a thing is kwoth, spirit? The answer is not so simple.
There are several ways in which what we would render as "is" is indicated in the Nuer language. The one which concerns us here is the particle e. It is used to tell the listener that something belongs to a certain class or category and hence about some character or quality it has, as "e dit", "it is a bird", "'gat nath e car", "the Nuer is black", and "Duob e ram me goagh", "Duob is a good man." The question we are asking is what meaning or meanings it has for Nuer when they say of something "e kwoth", "it is Spirit" (in the sense either of God or of a divine refraction).
Nuer do not claim to see God, nor do they think that anyone can know what he is like in himself. When they speak about his nature they do so by adjectives which refer to attributes, such as "great" and "good", or in metaphors taken from the world around them, likening his invisibility and ubiquity to wind and air, his greatness to the universe he has created, and his grandeur to an ox with widespread horns. They are no more than metaphors for Nuer, who do not say that any of these things is God, but only that he is like (cere) them. They express in these poetic images as best they can what they think must be some of his attributes.
Nevertheless, certain things are said, or may be said, "to be" God - rain, lightning, and various other natural - in the Nuer way of speech, created - things which are of common interest. There is here an ambiguity, or an obscurity, to be elucidated, for Nuer are not now saying that God or Spirit is like this or that, but that this or that "is" God or Spirit. Elucidation here does not, however, present great difficulties.
God being conceived of as in the sky, those celestial phenomena which are of particular significance for Nuer, rain and lightning, are said, in a sense we have to
THE PROBLEM OF SYMBOLS
147
determine, to be him. There is no noun denoting either phenomenon and they can only be spoken of by verbs indicating a function of the sky, as "ce nhial deam", "the sky rained", and "ce nhial mar", "the sky thundered". Also pestilences, murrains, death, and indeed almost any natural phenomenon significant for men are commonly regarded by Nuer as manifestations from above, activities of divine being. Even the earthly totems are conceived of as a relationship deriving from some singular intervention of Spirit from above in human affairs. It is chiefly by these signs that Nuer have knowledge of God. It might be held, therefore, that the Nuer conception of God is a conceptualization of events which, on account of their strangeness or variability as well as on account of their potentiality for fortune or misfortune, are said to be his activities or his activities in one or other of his hypostases or refractions. Support for such a view might be found in the way Nuer sometimes speak of one or other of these effects. They may say of rain or lightning or pestilence "e kwoth", "it is God", and in storms they pray to God to come to earth gently and not in fury - to come gently, it will be noted, not to make the rain come gently.
I do not discuss this ontological question here beyond saying that were we to suppose that such phenomena are in themselves regarded as God we would misunderstand and misrepresent Nuer religious thought, which is pre-eminently dualistic. It is true that for them there is no abstract duality of natural and supernatural, but there is such a duality between kwoth, Spirit, which is immaterial rather than supernatural, and cak, creation, the material world known to the senses. Rain and lightning and pestilences and murrains belong to this created world and are referred to by Nuer as nyin kwoth, instruments of God.
Nevertheless, they and other effects of significance for men are Stoarj^iLa, signs or manifestations of divine activity; and since Nuer apprehend divine activity in these signs, in God's revelation of himself to them in material forms, the signs are, in a lower medium, what they signify, so that Nuer may say of them "e kwoth", "it is God". Rain and pestilence come from God and are therefore manifestations of him, and in this sense rain and pestilence are God, in the sense that he reveals himself in their falling. But though one can say of rain or pestilence that it is God one cannot say of God that he is rain or pestilence. This would make no sense for a number of reasons. In the first place, the situation could scarcely arise, God not being an observable object, in which Nuer would require or desire to say about him that he is anything. In the second place, the word kwoth does not here refer to a particular refraction of Spirit, a spirit, but to Spirit in its oneness, God, and he could not be in any way identified with any one of his manifestations to the exclusion of all the others. A third, and the most cogent, reason is that rain is water which falls from the sky and pestilence is a bodily condition and they are therefore in their nature material things and not Spirit. Indeed, as a rule, rain is only thought of in connexion with Spirit, and is therefore only said to be Spirit, when it does not fall in due season or falls too much or too violently with storm and lightning - when, that is, the rain has some special significance for human affairs. This gives us a clue to what is meant when Nuer say of something that it is God or that it is a spirit of the air, as thunder may be said to be the spirit wiu or a prophet of the spirit deng may be said to be deng - especially as Nuer readily expand such statements by adding that thunder, rain, and pestilence are all instruments (nyin) of God or that they are sent by (jak) God, and that the spirit deng has filled (gwang) the prophet through whom it speaks. In
148
E. E. EVANS-PRITCHARD
the statement here that something is Spirit or a spirit the particle e, which we translate "is", cannot therefore have the meaning of identity in a substantial sense. Indeed, it is because Spirit is conceived of in itself, as the creator and the one, and quite apart from any of its material manifestations, that phenomena can be said to be sent by it or to be its instruments. When Nuer say of rain or lightning that it is God they are making an elliptical statement. What is understood is not that the thing in itself is Spirit but that it is what we would call a medium or manifestation or sign of divine activity in relation to men and of significance for them. What precisely is posited by the hearer of any such elliptical statement depends on the nature of the situation by reference to which it is made. A vulture is not thought of as being in itself Spirit; it is a bird. But if it perches on the crown of a byre or hut Nuer may say "e kwoth", "it is Spirit", meaning that its doing so is a spiritual signal presaging disaster. A lion is not thought of as being in itself Spirit; it is a beast. But it may, on account of some event which brings it into a peculiar relation to man, such as being born, as Nuer think sometimes happens, as twin to a human child, be regarded as a revelation of Spirit for a particular family and lineage. Likewise, diseases, or rather their symptoms, are not thought of as being in themselves Spirit, but their appearance in individuals may be regarded as manifestations of Spirit for those individuals. Spirit acts, and thereby reveals itself, through these creatures. This distinction between the nature of a thing and what it may signify in certain situations or for certain persons is very evident in totemic relationships. A crocodile is Spirit for certain persons, but it is not thought to be in its nature Spirit, for others kill and eat it. It is because Nuer separate, and quite explicitly when questioned about the matter, spiritual conceptions from such material things as may nevertheless be said "to be" the conceptions, that they are able to maintain the unity and autonomy of Spirit in spite of a great diversity of accidents and are able to speak of Spirit without reference to any of its material manifestations.
So far I have been mostly speaking of the conception of God and of those of his refractions which belong to the category of the sky or of the above. With two possible exceptions,1 we cannot say that the things said "to be" these spirits are material symbols or representations of them; at any rate not in the same sense as we can speak of things being symbols of those lesser refractions of Spirit Nuer call spirits of the earth or of the below, in which God stands in a special relationship to lineages and individuals - such diverse things as beasts, birds, reptiles, trees, phosphorescent objects, and pieces of wood. These lesser refractions of Spirit, regarded as distinct spirits in relation to each other, cannot, unlike the spirits of the air, easily be thought of except in relation to the things by reference to which they derive their individuality, and which are said "to be" them.
When, therefore, Nuer say that the pied crow is the spirit buk or that a snake is Spirit, the word "is" has a different sense from what it has in the statement that rain is Spirit. The difference does not merely lie in the fact that kwoth has here a more restricted connotation, being spoken of in reference to a particular and exclusive refraction - a spirit - rather than comprehensively as God or Spirit in its oneness. It lies also in the relation understood in the statement between its subject (snake or crow) and its predicate (Spirit or a spirit). The snake in itself is not divine activity whereas rain and lightning are. The story accounting for a totemic relationship may present it as arising from a revelation of divine activity, but once it has become an established relationship between a lineage and a natural species, the species is a representation or
THE PROBLEM OF SYMBOLS
149
symbol of Spirit to the lineage. What then is here meant when it is said that the pied crow "is" buk or that a snake "is" Spirit: that the symbol "is" what it symbolizes? Clearly Nuer do not mean that the crow is the same as buk, for buk is also conceived of as being in the sky and also in rivers, which the pied crow certainly is not; nor that a snake is the same as some spiritual refraction, for they say that the snake just crawls on the earth while the spirit it is said to be is in the sky. What then is being predicated about the crow or snake in the statement that either is Spirit or a spirit?
It will be simpler to discuss this question in the first place in relation to a totemic relationship. When a Nuer says of a creature "e nyang", "it is a crocodile", he is saying that it is a crocodile and not some other creature, but when he says, to explain why a person behaves in an unusual manner towards crocodiles "e kwothdien", "it (the crocodile) is their spirit", he is obviously making a different sort of statement. He is not saying what kind of creature it is (for it is understood that he is referring to the crocodile) but that what he refers to is Spirit for certain people. But he is also not saying that the crocodile is Spirit - it is not so for him - but that certain people so regard it. Therefore a Nuer would not make a general statement that "nyang e kwotb", "crocodile is Spirit", but would only say, in referring to the crocodile, "e kwoth", "it is Spirit", the distinction between the two statements being that the first would mean that the crocodile is Spirit for everyone whereas the second, being made in a special context of situation, means that it is Spirit for certain persons who are being discussed, or are understood, in that context. Likewise, whilst it can be said of the crocodile that it is Spirit, it cannot be said of Spirit that it is the crocodile, or rather, if a statement is framed in this form it can only be made when the word kwoth has a pronominal suffix which gives it the meaning of "his spirit", "their spirit", and so forth; in other words, where the statement makes it clear that what is being spoken of is Spirit conceived of in relation to particular persons only. We still have to ask, however, in what sense the crocodile is Spirit for these persons.
Since it is difficult to discuss a statement that something which can be observed, crocodile, is something more than what is appears to be when this something more, Spirit, cannot be observed, it is helpful first to consider two examples of Nuer statements that things are something more than they appear to be when both the subject term and the predicate term refer to observable phenomena.
When a cucumber is used as a sacrificial victim Nuer speak of it as an ox. In doing so they are asserting something rather more than that it takes the place of an ox. They do not, of course, say that cucumbers are oxen, and in speaking of a particular cucumber as an ox in a sacrificial situation they are only indicating that it may be thought of as an ox in that particular situation; and they act accordingly by performing the sacrificial rites as closely as possible to what happens when the victim is an ox. The resemblance is conceptual, not perceptual. The "is" rests on qualitative analogy. And the expression is asymmetrical, a cucumber is an ox, but an ox is not a cucumber.
A rather different example of this way of speaking is the Nuer assertion that twins are one person and that they are birds.2 When they say "twins are not two persons, they are one person" they are not saying that they are one individual but that they have a single personality. It is significant that in speaking of the unity of twins they only use the word ran, which, like our word "person", leaves sex, age, and other distinguishing qualities of individuals undefined. They would not say that twins of the same sex were one dhol, boy, or one nyal, girl, but they do say, whether they are
150
E. E. EVANS-PRITCHARD
of the same sex or not, that they are one ran, person. Their single social personality is something over and above their physical duality, a duality which is evident to the senses and is indicated by the plural form used when speaking of twins and by their treatment in all respects in ordinary social life as two quite distinct individuals. It is only in certain ritual situations, and symbolically, that the unity of twins is expressed, particularly in ceremonies connected with marriage and death, in which the personality undergoes a change. Thus, when the senior of male twins marries, the junior acts with him in the ritual acts he has to perform; female twins ought to be married on the same day; and no mortuary ceremonies are held for twins because, for one reason, one of them cannot be cut off from the living without the other. A woman whose twin brother had died some time before said to Miss Soule, to whom I am indebted for the information, "Is not his soul still living? I am alive, and we are really children of God."
There is no mortuary ceremony even when the second twin dies, and I was told that twins do not attend the mortuary ceremonies held for their dead kinsfolk, nor mourn them, because a twin is a ran nhial, a person of the sky or of the above. He is also spoken of as gat kwoth, a child of God. These dioscuric descriptions of twins are common to many peoples, but the Nuer are peculiar in holding also that they are birds. They say "a twin is not a person (ran), he is a bird (dit)", although, as we have just seen, they assert, in another sense, that twins are one person (ran). Here they are using the word ran in the sense of a human being as distinct from any other creature. The dogma is expressed in various ways. Very often a twin is given the proper name Dit, bird, Gwong, guineafowl, or Ngec, francolin.3 All Nuer consider it shameful, at any rate for adults, to eat any sort of bird or its eggs, but were a twin to do this it would be much more than shameful. It would be nueer, a grave sin, for twins respect (thek) birds, because, Nuer say, birds are also twins, and they avoid any sort of contact with them. The equivalence of twins and birds is expressed particularly in connexion with death. When an infant twin dies people say "ce par", "he has flown away", using the word denoting the flight of birds. Infant twins who die, as so often happens, are not buried, as other infants are, but are covered in a reed basket or winnowing-tray and placed in the fork of a tree, because birds rest in trees. I was told that birds which feed on carrion would not molest the bodies but would look at their dead kinsmen - twins and birds are also said to be kin, though the usage may be regarded as metaphorical - and fly away again. When I asked a Nuer whether adult twins would be buried like other people he replied "no, of course not, they are birds and their souls go up into the air". A platform, not used in the normal mode of burial, is erected in the grave and a hide placed over it. The body is laid on this hide and covered with a second hide. Earth is then carefully patted over the upper hide instead of being shovelled in quickly, as in the burial of an ordinary person. I was told that the corpse is covered with earth lest a hyena eat it and afterwards drink at a pool, for men might drink at the same pool and die from contamination (nueer).
It is understandable that Nuer draw an analogy between the multiple hatching of eggs and the dual birth of twins. The analogy is explicit, and, through an extension of it, the flesh of crocodiles and turtles is also forbidden to twins on the ground that these creatures too, like birds, lay eggs. Miss Soule once had a girl twin in her household who refused fish for the same reason - the only case of its kind known to either of us. But the analogy between multiple births in birds and men does not adequately explain why it is with birds that human twins are equated when there are
THE PROBLEM OF SYMBOLS
151
many other creatures which habitually bear several young at the same time and in a manner more closely resembling human parturition. It cannot be just multiple birth which leads Nuer to say that twins are birds, for these other creatures are not respected by twins on that account. The prohibition on eating eggs is clearly secondary, and it is extended to include crocodiles and turtles - and by Miss Soule's girl fish also - not because they lay eggs but because their laying eggs makes them like birds. Moreover, it is difficult to understand why a resemblance of the kind should in any case be made so much of. The multiple hatching of chicks is doubtless a resemblance which greatly strengthens the idea of twins being birds, but it is only part of a more complex analogical representation which requires to be explained in more general terms of Nuer religious thought. A twin, on account of his peculiar manner of conception is, though not Spirit himself, a special creation, and, therefore, manifestation of Spirit; and when he dies his soul goes into the air, to which things associated with Spirit belong. He is a ran nbial, a person of the above, whereas an ordinary person is a ran piny, a person of the below. A bird, though also not in itself Spirit, belongs by nature to the above and is also what Nuer call, using "person" metaphorically, a ran nbial, a person of the above, and being such is therefore also associated with Spirit. It cannot, of course, be determined for certain whether a twin is said to be a person of the above because he is a bird or whether he is said to be a bird because he is a person of the above, but the connexion in thought between twins and birds is certainly not simply derived from the multiple birth similitude but also, and in my view primarily, from both birds and twins being classed by Nuer as gaat kwoth, children of God. Birds are children of God on account of their being in the air, and twins belong to the air on account of their being children of God by the manner of their conception and birth.
It seems odd, if not absurd, to a European when he is told that a twin is a bird as though it were an obvious fact, for Nuer are not saying that a twin is like a bird but that he is a bird. There seems to be a complete contradiction in the statement; and it was precisely on statements of this kind recorded by observers of primitive peoples that Levy-Bruhl based his theory of the prelogical mentality of these peoples, its chief characteristic being, in his view, that it permits such evident contradictions - that a thing can be what it is and at the same time something altogether different. But, in fact, no contradiction is involved in the statement, which, on the contrary, appears quite sensible, and even true, to one who presents the idea to himself in the Nuer language and within their system of religious thought. He does not then take their statements about twins any more literally than they make and understand them themselves. They are not saying that a twin has a beak, feathers, and so forth. Nor in their everyday relations with twins do Nuer speak of them as birds or act towards them as though they were birds. They treat them as what they are, men and women. But in addition to being men and women they are of a twin-birth, and a twin-birth is a special revelation of Spirit; and Nuer express this special character of twins in the "twins are birds" formula because twins and birds, though for different reasons, are both associated with Spirit and this makes twins, like birds, "people of the above" and "children of God", and hence a bird a suitable symbol in which to express the special relationship in which a twin stands to God. When, therefore, Nuer say that a twin is a bird they are not speaking of either as it appears in the flesh. They are speaking of the anima of the twin, what they call his tie, a concept which includes both what we call the personality and the soul; and they are speaking of the
152
E. E. EVANS-PRITCHARD
association birds have with Spirit through their ability to enter the realm to which Spirit is likened in metaphor and where Nuer think it chiefly is, or may be. The formula does not express a dyadic relationship between twins and birds but a triadic relationship between twins, birds, and God. In respect to God twins and birds have a similar character.
It is because Nuer do not make, or take, the statement that twins are birds in any ordinary sense that they are fully aware that in ritual relating to twins the actions are a kind of miming. This is shown in their treatment of the corpse of a twin, for, according to what they themselves say, what is a bird, the tie or anima, has gone up into the air and what is left and treated - in the case of adults platform burial being a convenient alternative to disposal in trees - as though it might be a bird is only the ring, the flesh. It is shown also in the convention that should one of a pair of twins die, the child who comes after them takes his place, counting as one of them in the various ceremonies twins have to perform and respecting birds as rigorously as if he were himself a twin, which he is not. The ceremonies have to be performed for the benefit of the living twin and their structure and purpose are such that there have to be two persons to perform them, so a brother or sister acts in the place of the dead.
This discussion of what is meant by the statement that a twin is a bird is not so far away from the subject of totemism as it might seem to be, for the stock explanation among the Nuer of a totemic relationship is that the ancestor of a lineage and a member of a natural species were born twins. The relationship of lineage to species is thereby made to derive not only from the closest of all possible relationships but also from a special act of divine revelation; and since the link between a lineage and its totem is the tutelary spirit of the lineage associated with the totem it is appropriate that the relationship should be thought of as having come about by an event which is a direct manifestation of Spirit.
However, an examination of the Nuer dogma that twins are birds was made not on account of totemic relationships commonly being explained in terms of twinship but because it was hoped that it would be easier to understand, in the light of any conclusions reached about what is meant by the statement that a twin is a bird, what Nuer mean when they say that some totemic creature, such as the crocodile, is Spirit. Certainly there is here neither the sort of metaphor nor the sort of ellipsis we found in earlier statements. Nor can Nuer be understood to mean that the creature is identical with Spirit, or even with a spirit, Spirit conceived of in a particular totemic refraction. They say quite definitely themselves that it is not; and it is also evident, for Nuer as well as for us, that a material symbol of Spirit cannot, by its very nature, be that which it symbolizes. Nevertheless, though crocodile and Spirit are quite different and unconnected ideas, when the crocodile is for a certain lineage a symbol of their special relationship to God, then in the context of that relationship symbol and what it symbolizes are fused. As in the case of the "twins are birds" formula, the relation is a triadic one, between a lineage and a natural species and God.
There are obvious and significant differences between the creature-Spirit expression and the cucumber-ox and bird-twin expressions. Cucumber, ox, man, and bird are all things which can be known by the senses; but where Spirit is experienced other than in thought it is only in its effects or through material representations of it. We can, therefore, easily see how Nuer regard it as being in, or behind, the crocodile. The subject and predicate terms of the statement that something is Spirit are here no longer held apart by two sets of visible properties. Consequently, while Nuer say that
THE PROBLEM OF SYMBOLS
153
totemic spirits and totems are not the same they sometimes not only speak of, but act towards, a totem as if the spirit were in it. Thus they give some meat of a sacrifice to the lion-spirit to lions, and when they sacrifice to the durra-bird-spirit they address also the birds themselves and tell them that the victim is for them. Nevertheless, they make it clear in talking about their totems that what respect they show for them is on account of their representing the spirits associated with them and not for their own sake.
Another difference is that whereas in the cases of the cucumber-ox and twin-bird expressions the equivalence rests on analogies which are quite obvious even to us once they are pointed out - the cucumber being treated in the ritual of sacrifice as an ox is, and twins and birds both being "children of God" and also multiple births -analogy is lacking in the creature-Spirit expression. There is no resemblance between the idea of Spirit and that of crocodile. There is nothing in the nature of crocodiles which evokes the idea of Spirit for Nuer, and even for those who respect crocodiles the idea of Spirit is evoked by these creatures because the crocodiles is a representation of Spirit in relation to their lineage and not because there is anything crocodile-like about Spirit or Spirit-like about crocodiles. We have passed from observation of resemblances to thought by means of symbols in the sort of way that the crocodile is used as a symbol for Spirit.
We are here faced with the same problem we have been considering earlier, but in what, in the absence of analogical guidance to help us, is a more difficult form. The difficulty is increased by Nuer symbols being taken from an environment unfamiliar to us and one which, even when we familiarize ourselves with it, we experience and evaluate differently. We find it hard to think in terms of crocodiles, snakes, and fig-trees. But reflection shows us that this problem is common to all religious thought, including our own; that a religious symbol has always an intimate association with what it represents, that which brings to the mind with what it brings to the mind. Nuer know that what they see is a crocodile, but since it represents Spirit to some of them it is for those people, when thought of in that way, also what it stands for. The relationship of members of a Nuer lineage to Spirit is represented by a material symbol by which it can be thought of concretely, and therefore as a relationship distinct from the relationships of other lineages to Spirit. What the symbols stand for is the same thing. It is they, and not what they stand for, which differentiate the relationships. There results, when what acts as a symbol is regarded in this way, a fusion between Spirit, as so represented, and its material representation. I would say that then Nuer regard Spirit as being in some way in, or behind, the creature in which in a sense it is beholden.
The problem is even more difficult and complex than I have stated it, because we might say that what are fused are not so much the idea of Spirit and its material representation as the idea of Spirit and the idea of its material representation. It is rather the idea of crocodile than the saurian creatures themselves which stands for Spirit to a lineage. If a Nuer cannot see Spirit he likewise in some cases seldom, if ever, sees his totem; so that it is no longer a question of a material object symbolizing an idea but of one idea symbolizing another. I doubt whether those who respect monorchid bulls or waterbuck often see a member of the class or species, and children in these and other cases must often be told about their totemic attachments before they have seen their totems. There must also be Nuer who respect dom palms who live in parts of Nuerland to the east of the Nile where this tree does not grow.4
154
E. E. EVANS-PRITCHARD
Indeed, I feel confident that one totem, the lou serpent, a kind of Loch Ness monster, does not exist, and if this is so, a totem can be purely imaginary. As this point has some theoretical importance for a study of totemism I draw attention to a further significant fact. Nuer do not speak of the spirit of crocodiles, lions, tamarind-trees, and so on, but always of the spirit of crocodile, lion, and tamarind-tree, and they would never say that crocodiles, lions, and tamarind-trees were somebody's spirit but always that crocodile, lion, and tamarind-tree, was his spirit. The difference in meaning between the plural and singular usage is not, perhaps, very obvious in English but it is both clear and vital in Nuer. It is the difference between crocodiles thought of as they are seen in rivers and crocodiles thought of as crocodile or as the crocodile, as a type of creature, crocodile as a conception. The point I am making is exemplified by the story already recorded (p. 65) of a man who gave up respecting lions because they killed his cattle. He still regarded lion-spirit, Spirit in the representation of lion, as a spirit connected with his family. But if a totemic relationship may be an ideal one, and has always something of the ideal in it, I would still say that Nuer regard Spirit as being in some way in, or behind, totemic creatures when they think of them as representations of Spirit. [...]
... The more Spirit is thought to be bound to visible forms the less it is thought of as Spirit and the more it is thought of in terms of what it is bound to. In other words, there are gradations of the conception of Spirit from pure unattached Spirit to Spirit associated with human, animal, and lifeless objects and more and more closely bound to what it is associated with the farther down the scale one goes. This scale of Spirit, as I have explained earlier, is related to segmentation of the social order and is represented by Nuer by levels of space as well as by levels and degrees of immanence. So when Nuer say of something that it is Spirit we have to consider not only what "is" means but also what "Spirit" means. Nevertheless, though the sense of "kwoth" varies with the context, the word refers always to something of the same essence; and what is being said, directly or indirectly, in the statements is always the same, that something is that essence.
We can make some contribution towards a solution of the problem in the light of this discussion. When Nuer say of something "e kwoth", "it is Spirit", or give it a name of which it can be further said "that is Spirit", the "is" does not in all instances have the same connotation. It may be an elliptical statement, signifying that the thing referred to is a manifestation of Spirit in the sense of God revealing himself in instruments or effects. Or it may be a symbolical statement, signifying that what in itself is not Spirit but represents Spirit to certain persons is for these persons Spirit in such contexts as direct attention to the symbolic character of an object to the exclusion of whatever other qualities it may possess. Or it may be a statement signifying something closer to identity of the thing spoken of with what it is said to be, Spirit. The statements never, however, signify complete identity of anything with Spirit, because Nuer think of Spirit as something more than any of its modes, signs, effects, representations, and so forth, and also as something of a different nature from the created things which they are. They are not able to define what it is, but when it acts within the phenomenal world they say it has come from above, where it is conceived to be and whence it is thought to descend. Consequently Spirit in any form can be detached in the mind from the things said to be it, even if they cannot always be so easily detached from the idea of Spirit.
THE PROBLEM OF SYMBOLS
155
I can take the analysis no farther; but if it is inconclusive it at least shows, if it is correct, how wide of the mark have been anthropological attempts to explain the kind of statements we have been considering. Anthropological explanations display two main errors. The first, best exemplified in the writings of Levy-Bruhl, is that when a people say that something is something else which is different they are contravening the Law of Contradiction and substituting for it a law of their own prelogical way of thinking, that of mystical participation.5 I hope at least to have shown that Nuer do not assert identity between the two things. They may say that one is the other and in certain situations act towards it as though it were that other, or something like it, but they are aware, no doubt with varying degrees of awareness, and readily say, though with varying degrees of clarity and emphasis, that the two things are different. Moreover, it will have been noted that in the seemingly equivocal statements we have considered, with perhaps one exception, the terms cannot be reversed. The exception is the statement that twins are birds, because it can also be said that birds are twins. That a hatch of birds are twins is a statement, to which we also can give assent, which does not derive logically from the statement that twins are birds but from a perception independent of that proposition; so it does not concern our problem. Rain may be said to be God but God cannot be said to be rain; a cucumber may be called an ox but an ox cannot be called a cucumber; and the crocodile may be said to be Spirit but Spirit cannot be said to be the crocodile. Consequently these are not statements of identity. They are statements not that something is other than it is but that in a certain sense and in particular contexts something has some extra quality which does not belong to it in its own nature; and this quality is not contrary to, or incompatible with, its nature but something added to it which does not alter what it was but makes it something more, in respect to this quality, than it was. Consequently, no contradiction, it seems to me, is involved in the statements.
Whether the predicate refers to a conception or to a visible object the addition makes the subject equivalent to it in respect to the quality which both now have in common in such contexts as focus the attention on that quality alone. The things referred to are not the same as each other but they are the same in that one respect, and the equivalence, denoted by the copula, is not one of substance but of quality. Consequently we cannot speak here, as Levy-Bruhl does, of mystical participation, or at any rate not in his sense of the words, because the two things are not thought to be linked by a mystical bond but simply by a symbolic nexus. Therefore, what is done to birds is not thought to affect twins, and if a totem is harmed the spirit of that totem may be offended but it is not harmed by the harm done to the totemic creature.
That the relation between the thing said to be something else and that something else it is said to be is an ideal one is indeed obvious, but anthropological explanations of modes of primitive though as wide apart as those of Tylor, Max Miiller, and Levy-Bruhl, are based on the assumption that though for us the relation is an ideal one primitive peoples mistake it for a real one; and those anthropologists who sponsor psychological explanations often make the same assumption. This is the second error. If my interpretation is correct, Nuer know very well when they say that a crocodile is Spirit that it is only Spirit in the sense that Spirit is represented to some people by that symbol just as they know very well that a cucumber is only an ox in the sense that they treat it as one in sacrifice. That they do not mistake ideal relations
156
E. E. EVANS-PRITCHARD
for real ones is shown by many examples in this book: the identification of a sacrificial spear with that of the ancestor..., the identification of man with ox in sacrifice..., the identification of a man's herd with that of the ancestor of his clan..., the identification of sickness and sin in a sacrificial context..., and the identification of the left hand with death and evil.... It is shown also in the symbolism of many of their rites, where their purpose is expressed in mimicry____
I think that one reason why it was not readily perceived that statements that something is something else should not be taken as matter-of-fact statements is that it was not recognized that they are made in relation to a third term not mentioned in them but understood. They are statements, as far as the Nuer are concerned, not that A is B, but that A and B have something in common in relation to C. This is evident when we give some thought to the matter. A cucumber is equivalent to an ox in respect to God who accepts it in the place of an ox. A crocodile is equivalent to Spirit only when conceived of as a representation of God to a lineage. Consequently, though Nuer do not mistake ideal relations for real ones, an ideal equivalence is none the less true for them, because within their system of religious thought things are not just what they appear to be but as they are conceived of in relation to God.
This implies experience on an imaginative level of thought where the mind moves in figures, symbols, metaphors, analogies, and many an elaboration of poetic fancy and language; and another reason why there has been misunderstanding is that the poetic sense of primitive peoples has not been sufficiently allowed for, so that it has not been appreciated that what they say is often to be understood in that sense and not in any ordinary sense. This is certainly the case with the Nuer, as we see in this chapter and in many places elsewhere in this book, for example, in their hymns. In all their poems and songs also they play on words and images to such an extent that no European can translate them without commentary from Nuer, and even Nuer themselves cannot always say what meaning they had for their authors. It is the same with their cattle-and dance-names, which are chosen both for euphony and to express analogies. How Nuer delight in playing with words is also seen in the fun they have in making up tongue-twisters, sentences which are difficult to pronounce without a mistake, and slips of the tongue, usually slips in the presence of mothers-in-law, which turn quite ordinary remarks into obscenities. Lacking plastic and visual arts, the imagination of this sensitive people finds its sole expression in ideas, images, and words.
In this and the last chapter I have attempted to lay bare some features of the Nuer conception of Spirit. We are not asking what Spirit is but what is the Nuer conception of kwoth, which we translate "Spirit". Since it is a conception that we are inquiring into, our inquiry is an exploration of ideas. In the course of it we have found that whilst Nuer conceive of Spirit as creator and father in the heavens they also think of it in many different representations (what I have called refractions of Spirit) in relation to social groups, categories, and persons. The conception of Spirit has, we found, a social dimension (we can also say, since the statement can be reversed, that the social structure has a spiritual dimension). We found also that Spirit, in the Nuer conception of it, is experienced in signs, media, and symbols through which it is manifested to the senses. Fundamentally, however, this is not a relation of Spirit to things but a relation of Spirit to persons through things, so that, here again, we are ultimately concerned with the relation of God and man, and we
THE PROBLEM OF SYMBOLS
157
have to consider not only what is the God-to-man side of the relationship, to which attention has so far mostly been given, but also the man-to-God side of it, to which I now turn.
NOTES
1 The spear win may be said to stand for the spirit wiu, and the pied crow may be said to stand for the spirit buk which is the most terrestrially conceived of among the greater spirits.
2 I have given a more detailed account in "Customs and Beliefs Relating to Twins among the Nilotic Nuer", Uganda Journal, 1936.
3 That the names, at least all those I have heard, are taken from birds lowest in the scale of Nuer reckoning requires comment, especially in view of the argument I later develop. It may be due to the Nuer habit of speaking of their relation to God - the birth of twins constitutes such a context - by comparing themselves with lowly things. On the other hand, it may be simply in keeping with the logic of the analogy. Twins belong to the class ' of the above but are below; just as guineafowl and francolin belong to the class of birds, which as a class is in the category of the above, but are almost earthbound.
4 Dr. Lienhardt tells me that a number of lineages in western Dinkaland respect creatures which no longer exist there. A Dinka who travelled with him to other parts of the Southern Sudan was astonished when he first saw his totem, an elephant. Nana Kobina Nketsia IV of Sekondi permits me to say that the first time he saw his totem, the buffalo, was last year in a film at Oxford. Professor I. Schapera tells me that the ruling family of the senior tribe in the Bechuanaland Protectorate, the Kwena, have been living for a hundred years in a region where their totem, the crocodile, is unknown (see also what he says in The Tswana (International African Institute), 1953, p. 35, and Hugh Ashton, The Basuto, 1952, p. 14). Other examples could be cited. It may help us to appreciate the point better if we consider the nearest parallels in our own country. When we think of the lion as our national symbol we do not think of the mangy creatures of the African bush or in zoos. Nor does it incommode us that there are no unicorns and never have been any.
5 I refer to his earlier writings, in particular Les Fonctions mentales dans les societes inferieures (1910) and La mentalite primitive (1922). The second part of his last book, L'Experience mystique et les symboles chez les primitifs (1938), which took account of modern research, is a brilliant discourse on the problems we have been discussing.
11
On Key Symbols
Sherry B. Or trier
From Sherry B. Ortner, "On Key Symbols," American Anthropologist 75 (1973): 1338-46. Reprinted by permission of the American Anthropological Association. Notes deleted.
Sherry Ortner is an American anthropologist currently teaching at Columbia University, New York. She is renowned for the clarity of her synthetic appraisals of theory, of which this is an early one (see also 1984, 1995), her contributions to gender theory and feminist anthropology (1974, 1996a; Ortner and Whitehead 1981), and her ethnography of Sherpa Buddhism (1978, 1989, 1999b). Ortner was a student of Geertz who, despite her interest in seeking broader and more "muscular" forms of social explanation than he provides (Dirks, Eley, and Ortner, eds. 1994), has remained one of the most articulate spokespersons for the significance of culture in social theory.
If Langer and White provide the ground for understanding the symbolic basis of culture, there was an emphasis in American anthropology to go further and seek key or core symbols that would sum up the particular focus, ethos, or worldview of a given cultural system. Whether or not such symbols can work this way in the ethnographic portraits of anthropologists, they often do appear to serve such ends for the members of the groups in which they are found. In this essay Ortner captures a particularly exciting moment in the discipline as symbolic anthropology emerged into prominence, and usefully articulates the various ways anthropologists have conceived, derived, or applied dominant symbols and how the symbols operate. Ortner suggests both the ways systems of meaning are organized and the ways different kinds of symbols work to condense, produce, or invite meaning. What she offers is a series of Weberian "ideal types," not species of symbols. This essay represents the state of the art in the early 1970s; Ortner has since moved well beyond it in an attempt to fully historicize cultural accounts. For a recent essay that links meaning with power see Ortner (1999a).
Victor Turner (1967) provides several great essays on religious symbols from a Dur-kheimian perspective. Douglas (1975) offers a classic analysis of a central symbol, the Lele pangolin, while De Boeck (1994) is a more recent intricate symbolic interpretation from the same Central African culture area as Turner and Douglas. A general account
ON KEY SYMBOLS
159
of symbols is to be found in Firth (1973), while Munn (1973; cf. 1986, 1990) is an excellent discussion of the transformative properties of ritual symbols. Schneider moves away from individual symbols to account for the symbolic constitution of persons and the conceptions underlying such cultural domains as kinship and religion (1977 [1969], 1980 [1968]). The notion of cultural scenarios is well developed in Schieffelin (1976). Both Pepper (1942) and White (1973) offer suggestive accounts of the power of underlying metaphors. In addition to the essay by Tambiah (chapter 24, below), the best anthropological work on metaphor is by Fernandez (1986b; ed. 1991) and Sapir and Crocker, eds. (1977). For more directly linguistic accounts, full of examples of the prevalence of metaphor in ordinary language, see Lakoff and Johnson (1980).
This paper reviews the use of the notion of "key symbol" in anthropological analysis. It analyzes phenomena which have been or might be accorded the status of key symbol in cultural analyses, categorizing them according to their primary modes of operating on thought and action.
It is by no means a novel idea that each culture has certain key elements which, in an ill-defined way, are crucial to its distinctive organization. Since the publication of Benedict's Patterns of Culture in 1934, the notion of such key elements has persisted in American anthropology under a variety of rubrics: "themes" (e.g., Opler 1945; Cohen 1948), "focal values" (Albert 1956), "dominant values" (DuBois 1955), "integrative concepts" (DuBois 1936), "dominant orientations" (F. Kluckhohn 1950), and so forth. We can also find this idea sneaking namelessly into British social anthropological writing; the best example of this is Lienhardt's (1961) discussion of cattle in Dinka culture (and I say culture rather than society advisedly). Even Evans-Pritchard has said, "as every experienced field-worker knows, the most difficult task in social anthropological field work is to determine the meanings of a few key words, upon an understanding of which the success of the whole investigation depends" (1962:80). Recently, as the focus in the study of meaning systems has shifted to the symbolic units which formulate meaning, the interest in these key elements of cultures has become specified as the interest in key symbols. Schneider (1968) calls them "core symbols" in his study of American kinship; Turner (1967) calls them "dominant symbols" in his study of Ndembu ritual; I called them "key symbols" in my study of Sherpa social relations (Ortner 1970).
The primary question of course is what do we mean by "key"? But I will postpone considering this problem until I have discussed the various usages of the notion of key symbols in the literature of symbolic analysis.
Two methodological approaches to establishing certain symbols as "core" or "key" to a cultural system have been employed. The first approach, less commonly used, involves analyzing the system (or domains thereof) for its underlying elements - cognitive distinctions, value orientations, etc. - then looking about in the culture for some figure or image which seems to formulate, in relatively pure form, the underlying orientations exposed in the analysis. The best example of this approach in the current literature is David Schneider's (1968) analysis of American kinship; Schneider first analyzes the kinship system for its basic components - nature and law
160
SHERRY B. ORTNER
- and then decides that conjugal sexual intercourse is the form which, given its meaning in the culture, expresses this opposition most succinctly and meaningfully. Schneider expresses his debt to Ruth Benedict, and this debt turns out to be quite specific, since the other major work which embodies this method is Benedict's The Chrysanthemum and the Sword (1967). The sword and the chrysanthemum were chosen by Benedict from the repertoire of Japanese symbols as most succinctly, or perhaps most poetically, representing the tension in the Japanese value system which she postulated. She did not arrive at this tension through an analysis of the meanings of chrysanthemums and swords in the culture; she first established the tension in Japanese culture through analysis of various symbolic systems, then chose these two items from the repertoire of Japanese symbols to sum up the opposition.
In the second, more commonly employed approach, the investigator observes something which seems to be an object of cultural interest, and analyzes it for its meanings. The observation that some symbol is a focus of cultural interest need not be very mysterious or intuitive. I offer here five reasonably reliable indicators of cultural interest, and there are probably more. Most key symbols, I venture to suggest, will be signaled by more than one of these indicators:
(1) The natives tell us that X is culturally important.
(2) The natives seem positively or negatively aroused about X, rather than indifferent.
(3) X comes up in many different contexts. These contexts may be behavioral or systemic: X comes up in many different kinds of action situation or conversation, or X comes up in many different symbolic domains (myth, ritual, art, formal rhetoric, etc.).
(4) There is greater cultural elaboration surrounding X, e.g., elaboration of vocabulary, or elaboration of details of X's nature, compared with similar phenomena in the culture.
(5) There are greater cultural restrictions surrounding X, either in sheer number of rules, or severity of sanctions regarding its misuse.
As I said, there may be more indicators even than these of the key status of a symbol in a culture, but any of these should be enough to point even the most insensitive fieldworker in the right direction. I should also add that I am not assuming that there is only one key symbol to every culture; cultures are of course a product of the interplay of many basic orientations, some quite conflicting. But all of them will be expressed somewhere in the public system, because the public symbol system is ultimately the only source from which the natives themselves discover, rediscover, and transform their own culture, generation after generation.
It remains for us now to sort out the bewildering array of phenomena to which various investigators have been led to assign implicitly or explicitly the status of key cultural symbol. Anything by definition can be a symbol, i.e., a vehicle for cultural meaning, and it seems from a survey of the literature that almost anything can be key. Omitting the symbols established by the first approach cited above, which have a different epistemological status, we can cite from the anthropological literature such things as cattle among the Dinka and Nuer, the Naven ritual of the Iatmul, the Australian churinga, the slametan of the Javanese, the potlatch of the northwest coast, the forked stick of Ndembu rituals, and from my own research, the wheel-
ON KEY SYMBOLS
161
image in Tibet and food among the Sherpas. We could also add such intuitive examples as the cross of Christianity, the American flag, the motorcycle for the Hell's Angels, "work" in the Protestant ethic, and so on.
The list is a jumble - things and abstractions, nouns and verbs, single items and whole events. I should like to propose a way of subdividing and ordering the set, in terms of the ways in which the symbols operate in relation to cultural thought and action.
The first major breakdown among the various types of symbols is along a continuum whose two ends I call "summarizing" vs. "elaborating." I stress that it is a continuum, but I work with the ideal types at the two ends.
Summarizing symbols, first, are those symbols which are seen as summing up, expressing, representing for the participants in an emotionally powerful and relatively undifferentiated way, what the system means to them. This category is essentially the category of sacred symbols in the broadest sense, and includes all those items which are objects of reverence and/or catalysts of emotion - the flag, the cross, the churinga, the forked stick, the motorcycle, etc. The American flag, for example, for certain Americans, stands for something called "the American way," a conglomerate of ideas and feelings including (theoretically) democracy, free enterprise, hard work, competition, progress, national superiority, freedom, etc. And it stands for them all at once. It does not encourage reflection on the logical relations among these ideas, nor on the logical consequences of them as they are played out in social actuality, over time and history. On the contrary, the flag encourages a sort of all-or-nothing allegiance to the whole package, best summed up on a billboard I saw recently: "Our flag, love it or leave." And this is the point about summarizing symbols in general - they operate to compound and synthesize a complex system of ideas, to "summarize" them under a unitary form which, in an old-fashioned way, "stands for" the system as a whole.
Elaborating symbols, on the other hand, work in the opposite direction, providing vehicles for sorting out complex and undifferentiated feelings and ideas, making them comprehensible to oneself, communicable to others, and translatable into orderly action. Elaborating symbols are accorded central status in the culture on the basis of their capacity to order experience; they are essentially analytic. Rarely are these symbols sacred in the conventional sense of being objects of respect or foci of emotion; their key status is indicated primarily by their recurrence in cultural behavior or cultural symbolic systems.
Symbols can be seen as having elaborating power in two modes. They may have primarily conceptual elaborating power, that is, they are valued as a source of categories for conceptualizing the order of the world. Or they may have primarily action elaborating power; that is, they are valued as implying mechanisms for successful social action. These two modes reflect what I see as the two basic and of course interrelated functions of culture in general: to provide for its members "orientations," i.e., cognitive and affective categories; and "strategies," i.e., programs for orderly social action in relation to culturally defined goals.
Symbols with great conceptual elaborating power are what Stephen Pepper (1942) has called "root metaphors," and indeed in this realm the basic mechanism is the metaphor. It is felt in the culture that many aspects of experience can be likened to, and illuminated by the comparison with, the symbol itself. In Pepper's terms, the symbol provides a set of categories for conceptualizing other aspects of experience,
162
SHERRY B. ORTNER
or, if this point is stated too uni-directionally for some tastes, we may say that the root metaphor formulates the unity of cultural orientation underlying many aspects of experience, by virtue of the fact that those many aspects of experience can be likened to it.
One of the best examples of a cultural root metaphor in the anthropological literature is found in Godfrey Lienhardt's discussion of the role of cattle in Dinka thought. Cows provide for the Dinka an almost endless set of categories for conceptualizing and responding to the subtleties of experience. For example: "The Dinkas' very perception of colour, light, and shade in the world around them is ... inextricably connected with their recognition of colour-configurations in their cattle. If their cattle-colour vocabulary were taken away, they would have scarcely any way of describing visual experience in terms of colour, light and darkness" (1961:13). More important for Lienhardt's thesis is the Dinka conceptualization of the structure of their own society on analogy with the physical structure of the bull. " 'The people are put together, as a bull is put together,' said a Dinka chief on one occasion" (ibid.: 23), and indeed the formally prescribed division of the meat of a sacrificed bull is a most graphic representation of the statuses, functions, and interrelationships of the major social categories of Dinka society, as the Dinka themselves represent the situation.
In fact, as Mary Douglas points out, the living organism in one form or another functions as a root metaphor in many cultures, as a source of categories for conceptualizing social phenomena (1966). In mechanized society, on the other hand, one root metaphor for the social process is the machine, and in recent times the computer represents a crucial modification upon this root metaphor. But the social is not the only aspect of experience which root-metaphor type symbols are used to illuminate; for example, much of greater Indo-Tibetan cosmology - the forms and processes of life, space, and time - is developed on analogy with the quite simple image of the wheel (Ortner 1966).
A root metaphor, then, is one type of key symbol in the elaborating mode, i.e., a symbol which operates to sort out experience, to place it in cultural categories, and to help us think about how it all hangs together. They are symbols which are "good to think," not exactly in the Levi-Straussian sense, but in that one can conceptualize the interrelationships among phenomena by analogy to the interrelations among the parts of the root metaphor.
The other major type of elaborating symbol is valued primarily because it implies clear-cut modes of action appropriate to correct and successful living in the culture. Every culture, of course, embodies some vision of success, or the good life, but the cultural variation occurs in how success is defined, and, given that, what are considered the best ways of achieving it. "Key scenarios," as I call the type of key symbol in this category, are culturally valued in that they formulate the culture's basic means-ends relationships in actable forms.
An example of a key scenario from American culture would be the Horatio Alger myth. The scenario runs: poor boy of low status, but with total faith in the American system, works very hard and ultimately becomes rich and powerful. The myth formulates both the American conception of success - wealth and power - and suggests that there is a simple (but not easy) way of achieving them - singleminded hard work. This scenario may be contrasted with ones from other cultures which present other actions as the most effective means of achieving wealth and power, or which formulate wealth and power as appropriate goals only for certain segments of
ON KEY SYMBOLS
163
[he society, or, of course, those which do not define cultural success in terms of wealth and power at all. In any case, the point is that every culture has a number of such key scenarios which both formulate appropriate goals and suggest effective action for achieving them; which formulate, in other words, key cultural strategies.
This category of key symbols may also include rituals; Singer seems to be making the point of rituals as scenarios when he writes of "cultural performances" (1958), in which both valued end states and effective means for achieving them are dramatized for all to see. Thus this category would include naven, the slametan, the potlatch, and others. The category could also include individual elements of rituals - objects, roles, action sequences - insofar as they refer to or epitomize the ritual as a whole, which is why one can have actions, objects, and whole events in the same category.
Further, scenarios as key symbols may include not only formal, usually named events, but also all those cultural sequences of action which we can observe enacted and reenacted according to unarticulated formulae in the normal course of daily life. An example of such a scenario from Sherpa culture would be the hospitality scenario, in which any individual in the role of host feeds a guest and thereby renders him voluntarily cooperative vis-a-vis oneself. The scenario formulates both the ideally valued (though infrequently attained) mode of social relations in the culture - voluntary cooperation - and, given certain cultural assumptions about the effects of food on people, the most effective way of establishing those kinds of relations. Once again then, the scenario is culturally valued - indicated in this case by the fact that it is played and replayed in the most diverse sorts of social contexts - because it suggests a clear-cut strategy for arriving at culturally defined success.
I have been discussing the category of key symbols which I called "elaborating" symbols, symbols valued for their contribution to the sorting out of experience. This class includes both root metaphors which provide categories for the ordering of conceptual experience, and key scenarios which provide strategies for organizing action experience. While for purposes of this discussion I have been led by the data to separate thought from action, I must hasten to put the pieces back together again. For my view is that ultimately both kinds of symbols have both types of referents. Root metaphors, by establishing a certain view of the world, implicitly suggest certain valid and effective ways of acting upon it; key scenarios, by prescribing certain cuturally effective courses of action, embody and rest upon certain assumptions about the nature of reality. Even summarizing symbols, while primarily functioning to compound rather than sort out experience, are seen as both formulating basic orientations and implying, though much less systematically than scenarios, certain modes of action.
One question which might be raised at this point is how we are to understand the logical relationships among the types of key symbols I have distinguished. As the scheme stands now, it has the following unbalanced structure:
summarizing   vs. elaborating
root metaphor
key scenario
164
SHERRY B. ORTNER
I would argue that this asymmetry follows from the content of the types: the meaning-content of summarizing or sacred symbols is by definition clustered, condensed, relatively undifferentiated, "thick," while the meaning-content of elaborating symbols is by definition relatively clear, orderly, differentiated, articulate. Thus it is possible to make distinctions among the different ordering functions of elaborating symbols, while the denseness of meaning of summarizing symbols renders them relatively resistant to subdivision and ordering by types. Nonetheless, in the interest of systematic analysis, we may raise the question of whether such subdivisions are possible, and in particular whether the thought/action distinction which subdivides elaborating symbols (into root metaphors and key scenarios) also crosscuts and subdivides summarizing symbols.
The important mode of operation of summarizing symbols, it will be recalled, is its focusing power, its drawing-together, intensifying, catalyzing impact upon the respondent. Thus we must ask whether some summarizing symbols primarily operate to catalyze thought or in any case internal states of the actor, while others primarily operate to catalyze overt action on the part of the actor. Now it does seem possible, for example, to see the cross or some other religious symbol as primarily focusing and intensifying inner attitude, with no particular implied public action, while the flag or some other political symbol is primarily geared to focusing and catalyzing overt action in the public world. Yet, intuitively at least, this distinction seems relatively weak and unconvincing compared to the easily formulated and grasped distinction between the two types of elaborating symbols: static formal images serving metaphor functions for thought (root metaphors), and dramatic, phased action sequences serving scenario functions for action (key scenarios). Of course, as I said, root metaphors may imply particular modes of, or at least a restricted set of possible modes of, action; and key scenarios presuppose certain orderly assumptions of thought. But the distinction - the former geared primarily to thought, the latter to action - remains sharp.
Summarizing symbols, on the other hand, speak primarily to attitudes, to a crystallization of commitment. And, in the mode of commitment, the thought/action distinction is not particularly relevant. There may certainly be consequences for thought and action as a result of a crystallized commitment, but commitment itself is neither thought nor action. The point perhaps illuminates the generally sacred status of summarizing symbols, for they are speaking to a more diffuse mode of orientation in the actor, a broader context of attitude within which particular modes of thinking and acting are formulated.
This is not to say that nothing analytic may be said about summarizing symbols beyond the fact that they catalyze feeling; there are a number of possible ways of subdividing the catalog of sacred symbols in the world, some no doubt more useful or illuminated than others. My point is merely that the particular factor which subdivides elaborating symbols - the thought/action distinction - does not serve very powerfully to subdivide the category of summarizing symbols, since the summarizing symbol is speaking to a different level of response, the level of attitude and commitment.
We are now in a position to return to the question of "key" or central status. Why are we justified in calling a particular symbol "key"? The indicators provided earlier for at least provisionally regarding certain symbols as key to a particular culture were all based on the assumption that keyness has public (though not necessarily
ON KEY SYMBOLS
165
conscious) manifestation in the culture itself, available to the observer in the field, or at least available when one reflects upon one's observations. But the fact of public cultural concern or focus of interest is not why a symbol is key; it is only a signal that the symbol is playing some key role in relation to other elements of the cultural system of thought. The issue of keyness, in short, has to do with the internal organization of the system of cultural meaning, as that system functions for actors leading their lives in the culture.
Broadly speaking, the two types of key symbols distinguished above, defined in terms of how they act upon or are manipulated by cultural actors, also indicate the two broad modes of "keyness" from a systemic point of view, defined in terms of the role such symbols are playing in the system; that is, a given summarizing symbol is "key" to the system insofar as the meanings which it formulates are logically or affectively prior to other meanings of the system. By "logically or affectively prior" I mean simply that many other cultural ideas and attitudes presuppose, and make sense only in the context of, those meanings formulated by the symbol. The key role of an elaborating symbol, by contrast, derives not so much from the status of its particular substantive meanings, but from its formal or organizational role in relation to the system; that is, we say such a symbol is "key" to the system insofar as it extensively and systematically formulates relationships - parallels, isomorphisms, complementarities, and so forth - between a wide range of diverse cultural elements.
This contrast between the two modes of "keyness" may be summed up in various ways, all of which oversimplify to some extent, but which nonetheless give perspective on the point. (1) "Content versus form": The keyness of a summarizing symbol derives from its particular substantive meanings (content) and their logical priority in relation to other meanings of the system. The keyness of an elaborating symbol derives from its formal properties, and their culturally postulated power to formulate widely applicable modes of organizing cultural phenomena. (2) "Quality versus quantity": The keyness of a summarizing symbol derives from the relative fundamentality (or ultimacy) of the meanings which it formulates, relative to other meanings of the system. The keyness of an elaborating symbol derives from the broadness of its scope, the extent to which it systematically draws relationships between a wide range of diverse cultural elements. (3) "Vertical versus lateral": The keyness of a summarizing symbol derives from its ability to relate lower-order meanings to higher-order assumptions, or to "ground" more surface-level meanings to their deeper bases. (The issue here is degree of generality of meaning. Whether more general meanings are termed "higher" or "deeper," "ultimate" or "fundamental," by a particular cultural analyst seems a matter of personal preference.) The keyness of an elaborating symbol by contrast derives from its ability to interconnect disparate elements at essentially the same level, by virtue of its ability to manifest (or bring into relief) their formal similarities.
All of these terminological contrasts - form/content, quantity/quality, lateral/ vertical - are really perspectives upon the same basic contrast, for which we have no more general term; that is, when we say a summarizing symbol is "key" to the system, we mean that its substantive meanings have certain kinds of priority relative to other meanings of the system. When we say an elaborating symbol is key to the system, we refer to the power of its formal or organizational role in relation to the system.
166
SHERRY B. ORTNER
But at this point we must stop short of reifying the distinctions, for, in practice, the contrast between the two broad types of key symbols and the two modes of "key-ness" may break down. It seems empirically to be the case that an elaborating symbol which is accorded wide-ranging applicability in the culture - played in many contexts, or applied to many different sorts of forms - is generally not only formally apt but also substantively referential to high-level values, ideas, cognitive assertions, and so forth. Indeed, insofar as such high level formulations are made, a key elaborating symbol of a culture may move into the sacred mode and operate in much the same way as does a summarizing symbol. And, on the other hand, some summarizing symbols may play important ordering functions, as when they relate the respondent not merely to a cluster of high-level assumptions and values, but to a particular scenario which may be replayed in ongoing life. (One may think, for example, of the Christian cross evoking, among other things, not only a general sense of God's purpose and support, but also the particular scenario of Christ's martyrdom.)
Thus we are brought to an important point, namely, that we are distinguishing not only types of symbols, but types of symbolic functions. These functions may be performed by any given symbol - at different times, or in different contexts, or even simultaneously by different "levels" of its meaning. While there are many examples of summarizing and elaborating symbols in their relatively pure forms, the kinds of functions or operations these symbols perform may also be seen as aspects of any given symbols.
To summarize the original scheme briefly, key symbols may be discovered by virtue of a number of reliable indicators which point to cultural focus of interest. They are of two broad types - summarizing and elaborating. Summarizing symbols are primarily objects of attention and cultural respect; they synthesize or "collapse" complex experience, and relate the respondent to the grounds of the system as a whole. They include most importantly sacred symbols in the traditional sense. Elaborating symbols, on the other hand, are symbols valued for their contribution to the ordering or "sorting out" of experience. Within this are symbols valued primarily for the ordering of conceptual experience, i.e., for providing cultural "orientations," and those valued primarily for the ordering of action, i.e., for providing cultural "strategies." The former includes what Pepper calls "root metaphors," the latter includes key scenarios, or elements of scenarios which are crucial to the means-end relationship postulated in the complete scenario.
This scheme also suggests, at least by the choices of terms, the modes of symbolic analysis relevant to the different types of key symbols. The first type (summarizing symbols) suggests a range of questions pertaining to the cultural conversion of complex ideas into various kinds of relatively undifferentiated commitment - patriotism, for example, or faith. The second type (root metaphors) suggests questions applicable to the analysis of metaphor in the broadest sense - questions of how thought proceeds and organizes itself through analogies, models, images, and so forth. And the third type (key scenarios) suggests dramatistic modes of analysis, in which one raises questions concerning the restructuring of attitudes and relationships as a result of enacting particular culturally provided sequences of stylized actions.
This article has been frankly programmatic; I am in the process of implementing some of its ideas in a monograph on Sherpa social and religious relations. Here I
ON KEY SYMBOLS
167
have simply been concerned to show that, although a method of cultural analysis via key symbols has been for the most part unarticulated, there is at least incipiently method in such analysis. It is worth our while to try to systematize this method, for it may be our most powerful entree to the distinctiveness and variability of human cultures.
REFERENCES CITED
Albert, Ethel. 1956. The Classification of Values: A Method and Illustration. American
Anthropologist 58: 221^8. Benedict, Ruth. 1934. Patterns of Culture Boston: Houghton-Mifflin.
-1967. The Chrysanthemum and the Sword. Cleveland and New York: World.
Cohen, A. K. 1948. On the Place of "Themes" and Kindred Concepts in Social Theory.
American Anthropologist 50: 436-43. Douglas, Mary. 1966. Purity and Danger. New York: Praeger.
DuBois, Cora. 1936. The Wealth Concept as an Integrative Factor in Tolowa-Tututni Culture.
In Essays in Anthropology Presented to A. L. Kroeber. Robert Lowie, Ed. Berkeley:
University of California Press. -1955. The Dominant Value Profile of American Culture. American Anthropologist 57:
1232-9.
Evans-Pritchard, E. E. 1962. Social Anthropology and Other Essays. New York: Free Press. Kluckhohn, Florence. 1950. Dominant and Substitute Profiles of Cultural Orientation. Social Forces 28: 376-93.
Lienhardt, Godfrey. 1961. Divinity and Experience. Oxford: Clarendon Press. Opler, Morris E. 1945. Themes as Dynamic Forces in Culture. American Journal of Sociology 51: 198-206.
Ortner, Sherry B. (Sherry O. Paul). 1966. Tibetan Circles. M.A. thesis, University of Chicago.
-1970. Food for Thought: A Key Symbol in Sherpa Culture. Ph.D. thesis, University of
Chicago.
Pepper, Stephen. 1942. 'World Hypotheses. Berkeley and Los Angeles: University of California Press.
Schneider, David M. 1968. American Kinship. Englewood Cliffs, N.J.: Prentice-Hall. Singer, Milton. 1958. The Great Tradition in a Metropolitan Center: Madras. In Traditional India: Structure and Change. Milton Singer, Ed. Philadelphia: American Folklore Society. Turner, Victor. 1967. The Forest of Symbols. Ithaca: Cornell University Press.
13
Myth in Primitive Psychology
Bronislaw Malinowski
From Bronislaw Malinowski, "Myth in Primitive Psychology," in Magic, Science and Religion and Other Essays (Garden City, NY: Doubleday & Co., Inc., 1954 [1926]), pp. 100-26, 145. Reprinted by permission of Macmillan Publishers, Inc. Abridged.
Bronislaw Malinowski (1884-1942) is one of the founders of modern anthropology. In a series of remarkable works written largely in the 1920s, he demonstrated the importance of rich ethnographic observation. Born in Poland, he spent most of his teaching career at the London School of Economics, where he influenced a large cohort of students, many of whom made their own important contributions to the ethnographic corpus on religion (notably Firth 1940, 1967, and Richards 1956, among many others).
From this entry readers may glimpse how Malinowski was able to capture the imagination of earlier generations of audiences, both professional and public. Malinowski invites us to share his romantic and pleasurable field odyssey in the Trobriand Islands of Melanesia while making no bones about the superiority of his approach over preceding ones. This essay is famous for developing the argument of myth as "charter" but it can be seen, I think, that Malinowski escapes the narrowly functionalist interpretations that are often placed on his ideas. Among other things, he emphasizes the immediacy and "living reality" of myth as well as its discursive and pragmatic (hence dialogical) qualities (that is, the recitation of myths as speech events). While it would be wrong to suggest either that all myths have political functions or that the interest or value of any given myth can be reduced to its instrumental political function, Malinowski was undoubtedly correct to look at the place of myth in legitimating particular forms of social organization and loci of power or interest and the contestation this inevitably brings. Myth becomes a language of legal argument.
Useful developments and exemplifications of Malinowski's approach include Leach's discussion of "myth as a justification for faction and social change" in highland Burma (1964: Chapter IX) and Andriolo's essay on genealogy in the Old Testament (1973). Leach himself later turned to structuralist (Levi-Straussian) analyses of the Old Testament (1969). Whereas Malinowski claims the meaning of myth is on the surface, many
MYTH IN PRIMITIVE PSYCHOLOGY
177
writers would disagree. Marx and Freud are both noted for arguing that meaning is concealed and needs to be recovered, while structuralists see the issue as one of revealing the codes or grammar by which meaning is produced.
Myth as it exists in a savage community, that is, in its living primitive form, is not merely a story told but a reality lived. It is not of the nature of fiction, such as we read today in a novel, but it is a living reality, believed to have once happened in primeval times, and continuing ever since to influence the world and human destinies. This myth is to the savage what, to a fully believing Christian, is the Biblical story of Creation, of the Fall, of the Redemption by Christ's Sacrifice on the Cross. As our sacred story lives in our ritual, in our morality, as it governs our faith and controls our conduct, even so does his myth for the savage.
The limitation of the study of myth to the mere examination of texts has been fatal to a proper understanding of its nature. The forms of myth which come to us from classical antiquity and from the ancient sacred books of the East and other similar sources have come down to us without the context of living faith, without the possibility of obtaining comments from true believers, without the concomitant knowledge of their social organization, their practiced morals, and their popular customs - at least without the full information which the modern fieldworker can easily obtain. Moreover, there is no doubt that in their present literary form these tales have suffered a very considerable transformation at the hands of scribes, commentators, learned priests, and theologians. It is necessary to go back to primitive mythology in order to learn the secret of its life in the study of a myth which is still alive - before, mummified in priestly wisdom, it has been enshrined in the indestructible but lifeless repository of dead religions.
Studied alive, myth, as we shall see, is not symbolic, but a direct expression of its subject matter; it is not an explanation in satisfaction of a scientific interest, but a narrative resurrection of a primeval reality, told in satisfaction of deep religious wants, moral cravings, social submissions, assertions, even practical requirements. Myth fulfills in primitive culture an indispensable function: it expresses, enhances, and codifies belief; it safeguards and enforces morality; it vouches for the efficiency of ritual and contains practical rules for the guidance of man. Myth is thus a vital ingredient of human civilization; it is not an idle tale, but a hard-worked active force; it is not an intellectual explanation or an artistic imagery, but a pragmatic charter of primitive faith and moral wisdom. [...]
In the subsequent chapters of this book we will examine a number of myths in detail, but for the moment let us glance at the subjects of some typical myths [from the Trobriand Islands]. Take, for instance, the annual feast of the return of the dead. Elaborate arrangements are made for it, especially an enormous display of food. When this feast approaches, tales are told of how death began to chastise man, and how the power of eternal rejuvenation was lost. It is told why the spirits have to leave the village and do not remain at the fireside, finally why they return once in a year. Again, at certain seasons in preparation for an overseas expedition, canoes are overhauled and new ones built to the accompaniment of a special magic. In this
178
BRONISLAW MALINOWSKI
there are mythological allusions in the spells, and even the sacred acts contain elements which are only comprehensible when the story of the flying canoe, its ritual, and its magic are told. In connection with ceremonial trading, the rules, the magic, even the geographical routes are associated with corresponding mythology. There is no important magic, no ceremony, no ritual without belief; and the belief is spun out into accounts of concrete precedent. The union is very intimate, for myth is not only looked upon as a commentary of additional information, but it is a warrant, a charter, and often even a practical guide to the activities with which it is connected. On the other hand the rituals, ceremonies, customs, and social organization contain at times direct references to myth, and they are regarded as the results of mythical event. The cultural fact is a monument in which the myth is embodied; while the myth is believed to be the real cause which has brought about the moral rule, the social grouping, the rite, or the custom. Thus these stories form an integral part of culture. Their existence and influence not merely transcend the act of telling the narrative, not only do they draw their substance from life and its interests - they govern and control many cultural features, they form the dogmatic backbone of primitive civilization.
This is perhaps the most important point of the thesis which I am urging: I maintain that there exists a special class of stories, regarded as sacred, embodied in ritual, morals, and social organization, and which form an integral and active part of primitive culture. These stories live not by idle interest, not as fictitious or even as true narratives; but are to the natives a statement of a primeval, greater, and more relevant reality, by which the present life, fates, and activities of mankind are determined, the knowledge of which supplies man with the motive for ritual and moral actions, as well as with indications as to how to perform them.
In order to make the point at issue quite clear, let us once more compare our conclusions with the current views of modern anthropology, not in order idly to criticize other opinions, but so that we may link our results to the present state of knowledge, give due acknowledgment for what we have received, and state where we have to differ clearly and precisely.
It will be best to quote a condensed and authoritative statement, and I shall choose for this purpose of definition an analysis given in Notes and Queries on Anthropology, by the late Miss C. S. Burne and Professor J. L. Myres. Under the heading "Stories, Sayings, and Songs," we are informed that "this section includes many intellectual efforts of peoples" which "represent the earliest attempt to exercise reason, imagination, and memory." With some apprehension we ask where is left the emotion, the interest, and ambition, the social role of all the stories, and the deep connection with cultural values of the more serious ones? After a brief classification of stories in the usual manner we read about the sacred tales: "Myths are stories which, however marvelous and improbable to us, are nevertheless related in all good faith, because they are intended, or believed by the teller, to explain by means of something concrete and intelligible an abstract idea or such vague and difficult conceptions as Creation, Death, distinctions of race or animal species, the different occupations of men and women; the origins of rites and customs, or striking natural objects or prehistoric monuments; the meaning of the names of persons or places. Such stories are sometimes described as etiological, because their purpose is to explain why something exists or happens."1
MYTH IN PRIMITIVE PSYCHOLOGY
179
Here we have in a nutshell all that modern science at its best has to say upon the subject. Would our Melanesians agree, however, with this opinion? Certainly not. They do not want to "explain," to make "intelligible" anything which happens in their myths - above all not an abstract idea. Of that there can be found to my knowledge no instance either in Melanesia or in any other savage community. The few abstract ideas which the natives possess carry their concrete commentary in the very word which expresses them. When being is described by verbs to lie, to sit, to stand, when cause and effect are expressed by words signifying foundation and the past standing upon it, when various concrete nouns tend towards the meaning of space, the word and the relation to concrete reality make the abstract idea sufficiently "intelligible." Nor would a Trobriander or any other native agree with the view that "Creation, Death, distinctions of race or animal species, the different occupations of men and women" are "vague and difficult conceptions." Nothing is more familiar to the native than the different occupations of the male and female sex; there is nothing to be explained about it. But though familiar, such differences are at times irksome, unpleasant, or at least limiting, and there is the need to justify them, to vouch for their antiquity and reality, in short to buttress their validity. Death, alas, is not vague, or abstract, or difficult to grasp for any human being. It is only too hauntingly real, too concrete, too easy to comprehend for anyone who has had an experience affecting his near relatives or a personal foreboding. If it were vague or unreal, man would have no desire so much as to mention it; but the idea of death is fraught with horror, with a desire to remove its threat, with the vague hope that it may be, not explained, but rather explained away, made unreal, and actually denied. Myth, warranting the belief in immortality, in eternal youth, in a life beyond the grave, is not an intellectual reaction upon a puzzle, but an explicit act of faith born from the innermost instinctive and emotional reaction to the most formidable and haunting idea. Nor are the stories about "the origins of rites and customs" told in mere explanation of them. They never explain in any sense of the word; they always state a precedent which constitutes an ideal and a warrant for its continuance, and sometimes practical directions for the procedure.
We have, therefore, to disagree on every point with this excellent though concise statement of present-day mythological opinion. This definition would create an imaginary, non-existent class of narrative, the etiological myth, corresponding to a non-existent desire to explain, leading a futile existence as an "intellectual effort," and remaining outside native culture and social organization with their pragmatic interests. The whole treatment appears to us faulty, because myths are treated as mere stories, because they are regarded as a primitive intellectual armchair occupation, because they are torn out of their life context, and studied from what they look like on paper, and not from what they do in life. Such a definition would make it impossible either to see clearly the nature of myth or to reach a satisfactory classification of folk tales. In fact we would also have to disagree with the definition of legend and of fairy tale given subsequently by the writers in Notes and Queries on Anthropology.
But above all, this point of view would be fatal to efficient field work, for it would make the observer satisfied with the mere writing down of narratives. The intellectual nature of a story is exhausted with its text, but the functional, cultural, and pragmatic aspect of any native tale is manifested as much in its enactment, embodiment, and contextual relations as in the text. It is easier to write down the story than to observe the diffuse, complex ways in which it enters into life, or to study its
180
BRONISLAW MALINOWSKI
function by the observation of the vast social and cultural realities into which it enters. And this is the reason why we have so many texts and why we know so little about the very nature of myth.
We may, therefore, learn an important lesson from the Trobrianders, and to them let us now return. We will survey some of their myths in detail, so that we can confirm our conclusions inductively, yet precisely.
Myths of Origin
We may best start with the beginning of things, and examine some of the myths of origin. The world, say the natives, was originally peopled from underground. Humanity had there led an existence similar in all respects to the present life on earth. Underground, men were organized in villages, clans, districts; they had distinctions of rank, they knew privileges and had claims, they owned property, and were versed in magic lore. Endowed with all this, they emerged, establishing by this very act certain rights in land and citizenship, in economic prerogative and magical pursuit. They brought with them all their culture to continue it upon this earth.
There are a number of special spots - grottoes, clumps of trees, stone heaps, coral outcrops, springs, heads of creeks - called "holes" or "houses" by the natives. From such "holes" the first couples (a sister as the head of the family and the brother as her guardian) came and took possession of the lands, and gave the totemic, industrial, magical, and sociological character to the communities thus begun.
The problem of rank which plays a great role in their sociology was settled by the emergence from one special hole, called Obukula, near the village of Laba'i. This event was notable in that, contrary to the usual course (which is: one original "hole," one lineage), from this hole of Laba'i there emerged representatives of the four main clans one after the other. Their arrival, moreover, was followed by an apparently trivial but, in mythical reality, a most important event. First there came the Kayla-vasi (iguana), the animal of the Lukulabuta clan, which scratched its way through the earth as iguanas do, then climbed a tree, and remained there as a mere onlooker, following subsequent events. Soon there came out the Dog, totem of the Lukuba clan, who originally had the highest rank. As a third came the Pig, representative of the Malasi clan, which now holds the highest rank. Last came the Lukwasisiga totem, represented in some versions by the Crocodile, in others by the Snake, in others by the Opossum, and sometimes completely ignored. The Dog and Pig ran round, and the Dog, seeing the fruit of the noku plant, nosed it, then ate it. Said the Pig: "Thou eatest noku, thou eatest dirt; thou art a low-bred, a commoner; the chief, the guya'u, shall be I." And ever since, the highest subclan of the Malasi clan, the Tabalu, have been the real chiefs.
In order to understand this myth, it is not enough to follow the dialogue between the Dog and the Pig which might appear pointless or even trivial. Once you know the native sociology, the extreme importance of rank, the fact that food and its limitations (the taboos of rank and clan) are the main index of man's social nature, and finally the psychology of totemic identification - you begin to understand how this incident, happening as it did when humanity was in statu nascendi, settled once for all the relation between the two rival clans. To understand this myth you must have a good knowledge of their sociology, religion, customs, and outlook. Then, and
MYTH IN PRIMITIVE PSYCHOLOGY
181
only then, can you appreciate what this story means to the natives and how it can live in their life. If you stayed among them and learned the language you would constantly find it active in discussion and squabbles in reference to the relative superiority of the various clans, and in the discussions about the various food taboos which frequently raise fine questions of casuistry. Above all, if you were brought into contact with communities where the historical process of the spread of influence of the Malasi clan is still in evolution, you would be brought face to face with this myth as an active force.
Remarkably enough the first and last animals to come out, the iguana and the Lukwasisiga totem, have been from the beginning left in the cold: thus the numerical principle and the logic of events is not very strictly observed in the reasoning of the myth.
If the main myth of Laba'i about the relative superiority of the four clans is very often alluded to throughout the tribe, the minor local myths are not less alive and active, each in its own community. When a party arrives at some distant village they will be told not only the legendary historical tales, but above all the mythological charter of that community, its magical proficiencies, its occupational character, its rank and place in totemic organization. Should there arise land quarrels, encroachment in magical matters, fishing rights, or other privileges the testimony of myth would be referred to.
Let me show concretely the way in which a typical myth of local origins would be retailed in the normal run of native life. Let us watch a party of visitors arriving in one or the other of the Trobriand villages. They would seat themselves in front of the headman's house, in the central place of the locality. As likely as not the spot of origins is nearby, marked by a coral outcrop or a heap of stones. This spot would be pointed out, the names of the brother and sister ancestors mentioned, and perhaps it would be said that the man built his house on the spot of the present headman's dwelling. The native listeners would know, of course, that the sister lived in a different house nearby, for she could never reside within the same walls as her brother.
As additional information, the visitors might be told that the ancestors had brought with them the substances and paraphernalia and methods of local industry. In the village of Yalaka, for instance, it would be the processes for burning lime from shells. In Okobobo, Obweria, and Obowada the ancestors brought the knowledge and the implements for polishing hard stone. In Bwoytalu the carver's tool, the hafted shark tooth, and the knowledge of the art came out from underground with the original ancestors. In most places the economic monopolies are thus traced to the autochthonous emergence. In villages of higher rank the insignia of hereditary dignity were brought; in others some animal associated with the local subclan came out. Some communities started on their political career of standing hostility to one another from the very beginning. The most important gift to this world carried from the one below is always magic; but this will have to be treated later on and more fully.
If a European bystander were there and heard nothing but the information given from one native to the other, it would mean very little to him. In fact, it might lead him into serious misunderstandings. Thus the simultaneous emergence of brother and sister might make him suspicious either of a mythological allusion to incest, or else would make him look for the original matrimonial pair and inquire about the
182
BRONISLAW MALINOWSKI
sister's husband. The first suspicion would be entirely erroneous, and would shed a false light over the specific relation between brother and sister, in which the former is the indispensable guardian, and the second, equally indispensable, is responsible for the transmission of the line. Only a full knowledge of the matrilineal ideas and institutions gives body and meaning to the bare mention of the two ancestral names, so significant to a native listener. If the European were to inquire who was the sister's husband and how she came to have children, he would soon find himself once more confronted by an entirely foreign set of ideas - the sociological irrelevance of the father, the absence of any ideas about physiological procreation, and the strange and complicated system of marriage, matrilineal and patrilocal at the same time.
The sociological relevance of these accounts of origins would become clear only to a European inquirer who had grasped the native legal ideas about local citizenship and the hereditary rights to territory, fishing grounds, and local pursuits. For according to the legal principles of the tribe all such rights are the monopolies of the local community, and only people descendent in the female line from the original ancestress are entitled to them. If the European were told further that, besides the first place of emergence, there are several other "holes" in the same village, he would become still more baffled until, by a careful study of concrete details and the principles of native sociology, he became acquainted with the idea of compound village communities, i.e., communities in which several subclans have merged.
It is clear, then, that the myth conveys much more to the native than is contained in the mere story; that the story gives only the really relevant concrete local differences; that the real meaning, in fact the full account, is contained in the traditional foundations of social organization; and that this the native learns, not by listening to the fragmentary mythical stories, but by living within the social texture of his tribe. In other words, it is the context of social life, it is the gradual realization by the native of how everything which he is told to do has its precedent and pattern in bygone times, which brings home to him the full account and the full meaning of his myths of origin.
For an observer, therefore, it is necessary to become fully acquainted with the social organization of the natives if he wants really to grasp its traditional aspect. The short accounts, such as those which are given about local origins, will then become perfectly plain to him. He will also clearly see that each of them is only a part, and a rather insignificant one, of a much bigger story, which cannot be read except from native life. What really matters about such a story is its social function. It conveys, expresses, and strengthens the fundamental fact of the local unity and of the kinship unity of the group of people descendent from a common ancestress. Combined with the conviction that only common descent and emergence from the soil give full rights to it, the story of origin literally contains the legal charter of the community. Thus, even when the people of a vanquished community were driven from their grounds by a hostile neighbor their territory always remained intact for them; and they were always, after a lapse of time and when their peace ceremony had been concluded, allowed to return to the original site, rebuild their village, and cultivate their gardens once more. The traditional feeling of a real and intimate connection with the land; the concrete reality of seeing the actual spot of emergence in the middle of the scenes of daily life; the historical continuity of privileges, occupations, and distinctive characters running back into the mythological first beginnings - all this obviously makes for cohesion, for local patriotism, for a feeling
MYTH IN PRIMITIVE PSYCHOLOGY
183
of union and kinship in the community. But although the narrative of original emergence integrates and welds together the historical tradition, the legal principles, and the various customs, it must also be clearly kept in mind that the original myth is but a small part of the whole complex of traditional ideas. Thus on the one hand the reality of myth lies in its social function; on the other hand, once we begin to study the social function of myth, and so to reconstruct its full meaning, we are gradually led to build up the full theory of native social organization.
One of the most interesting phenomena connected with traditional precedent and charter is the adjustment of myth and mythological principle to cases in which the very foundation of such mythology is flagrantly violated. This violation always takes place when the local claims of an autochthonous clan, i.e., a clan which has emerged on the spot, are overridden by an immigrant clan. Then a conflict of principles is created, for obviously the principle that land and authority belong to those who are literally born out of it does not leave room for any newcomers. On the other hand, members of a subclan of high rank who choose to settle down in a new locality cannot very well be resisted by the autochthons - using this word again in the literal native mythological sense. The result is that there come into existence a special class of mythological stories which justify and account for the anomalous state of affairs. The strength of the various mythological and legal principles is manifested in that the myths of justification still contain the antagonistic and logically irreconcilable facts and points of view, and only try to cover them by facile reconciliatory incident, obviously manufactured ad hoc. The study of such stories is extremely interesting, both because it gives us a deep insight into the native psychology of tradition, and because it tempts us to reconstruct the past history of the tribe, though we must yield to the temptation with due caution and scepticism. [...]
As far as the sociological theory of these legends goes the historical reconstruction is irrelevant. Whatever the hidden reality of their unrecorded past may be, myths serve to cover certain inconsistencies created by historical events, rather than to record these events exactly. The myths associated with the spread of the powerful subclans show on certain points a fidelity to life in that they record facts inconsistent with one another. The incidents by which this inconsistency is obliterated, if not hidden, are most likely fictitious; we have seen certain myths vary according to the locality in which they are told. In other cases the incidents bolster up non-existent claims and rights.
The historical consideration of myth is interesting, therefore, in that it shows that myth, taken as a whole, cannot be sober dispassionate history, since it is always made ad hoc to fulfill a certain sociological function, to glorify a certain group, or to justify an anomalous status. These considerations show us also that to the native mind immediate history, semi-historic legend, and unmixed myth flow into one another, form a continuous sequence, and fulfill really the same sociological function.
And this brings us once more to our original contention that the really important thing about the myth is its character of a retrospective, ever-present, live actuality. It is to a native neither a fictitious story, nor an account of a dead past; it is a statement of a bigger reality still partially alive. It is alive in that its precedent, its law, its moral, still rule the social life of the natives. It is clear that myth functions especially where
184
BRONISLAW MALINOWSKI
there is a sociological strain, such as in matters of great difference in rank and power, matters of precedence and subordination, and unquestionably where profound historical changes have taken place. So much can be asserted as a fact, though it must always remain doubtful how far we can carry out historical reconstruction from the myth.
We can certainly discard all explanatory as well as all symbolic interpretations of these myths of origin. The personages and beings which we find in them are what they appear to be on the surface, and not symbols of hidden realities. As to any explanatory function of these myths, there is no problem which they cover, no curiosity which they satisfy, no theory which they contain. [•••]
The science of myth in living higher cultures, such as the present civilization of India, Japan, China, and last but not least, our own, might well be inspired by the comparative study of primitive folklore; and in its turn civilized culture could furnish important additions and explanations to savage mythology. This subject is very much beyond the scope of the present study. I do, however, want to emphasize the fact that anthropology should be not only the study of savage custom in the light of our mentality and our culture, but also the study of our own mentality in the distant perspective borrowed from Stone Age man. By dwelling mentally for some time among people of a much simpler culture than our own, we may be able to see ourselves from a distance, we may be able to gain a new sense of proportion with regard to our own institutions, beliefs, and customs. If anthropology could thus inspire us with some sense of proportion, and supply us with a finer sense of humor, it might justly claim to be a very great science. [...]
NOTE
1   Quoted from Notes and Queries on Anthropology, pp. 210 and 211.
15
Land Animals, Pure and Impure
Mary Douglas
From Mary Douglas, "Preface" and "Land Animals, Pure and Impure," in Leviticus as Literature (Oxford: Oxford University Press, 1999), pp. v-viii, 134-51. Reprinted by permission of Oxford University Press. Abridged.
Mary Douglas is a distinguished British anthropologist who trained at the University of Oxford and taught at University College London and elsewhere. She is the author of a large number of works, of which the key ones with respect to religion include Purity and Danger (1966), Natural Symbols (1970), and the essays in Implicit Meanings (1975, cf. 1996). She has worked among the Lele of Central Africa, on contemporary British and American society, and on the Old Testament (1993, 1999). Her work has been the subject of an extensive study by Fardon (1999).
The present essay is taken from a recent, comprehensive interpretation of the biblical book of Leviticus. As Douglas explains, its origins lie in her much earlier analysis of "The Abominations of Leviticus" in Purity and Danger. That work, in turn, owed its immediate inspiration to Steiner's essay on taboo (1956), which in some respects paralled Dürkheim on the sacred and Levi-Strauss on totemism in showing its subject as relationally rather than substantively constituted. In her enormously influential analysis Douglas argued that systems of classification and the anomalies they inevitably produce are central objects of religious concern in any society. "Dirt," she famously stated, is simply "matter out of place" and hence always relative to a particular system of classification. Rather than explaining systems of purity and taboos in terms of biological germ theory (as though, somehow, the ancient Israelites had known about the dangers of trichinosis in uncooked pork), Douglas reverses the prevalent argument and sees modern concerns with hygiene as simply another instance of symbolic ordering. Purity and Danger thus helped develop a radical shift in the way that symbolic systems of small- and large-scale societies could be compared to one another, and was also inspirational for establishing the significance of the metaphoric properties of the body and food within cultural systems. Douglas provides a forthright critique of the ail-too prevalent perspective she called "medical materialism," which is not entirely distinct from the ideology that Sahlins critiques in the preceding essay. This form of
LAND ANIMALS, PURE AND IMPURE
195
argument which reduces symbol to function is evident in Harris's popular explanations for the refusal to eat pigs in the Middle East and cows in South Asia (1985), and subject to lively refutation in Sahlins (1976). However, Douglas differs from Sahlins in that while she has embraced new directions in structuralism and phenomenology, she has always remained true to her Durkheimian roots in seeing symbolic practice as rooted in, and addressing, social boundaries and relations.
In Leviticus as Literature Douglas revisits the question of symbolic classification in the Bible. Her account is no longer merely a very clever piece of analysis but an interpretation that is grounded in a deep and emergent understanding of the religion of ancient Judaism and the styles of reasoning and writing used by the authors of the different books of theTorah (Pentateuch). Remaining rigorously opposed to piecemeal moralist and materialist explanations of individual food taboos, in this chapter she essays an explanation of the rules of impurity in terms of the logic of the whole. The argument now attends to the positive meaning of ritual purity and the rules of sacrifice rather than ostensible disgust or puzzlement at anomalies. It also draws on the model developed elsewhere in the book of a system of analogies between Mount Sinai, the tabernacle, animal offerings, and the bodies of the people of Israel. Somewhat to Douglas's own puzzlement, and much to her credit, the interpretation moves beyond her inclination to seek direct correspondences between the social and symbolic orders. For other work on symbolic classification see Needham, ed. (1973); for an additional anthropological foray into the Bible, Leach (1969).
... To study the book of Leviticus as an anthropologist has been a project very dear to my heart. It seemed far beyond my reach. Yet not to do it would be to leave dangling a number of threads from early work. Let me explain some things about my training which have influenced my attitude to the Bible. Young anthropologists in Oxford in the late 1940s and 1950s were heirs to an old debate about human rationality, a debate provoked by the experience of science and biased by the experience of empire. Nineteenth-century rationalists centred on what they thought of as the natives' intellectual problems. Gross superstitions, naive magic, and immoral gods, were explained by reference to moral evolutionism. The mind of the primitive in aeons past had been hampered by illogical mental habits and proneness to letting emotions govern reason, and the same handicaps were thought to afflict present-day backward peoples. However, in reaction, for the students of my generation the main text was Evans-Pritchard's Witchcraft, Oracles and Magic (1937). From this we learnt that people from alien traditions, trusting in their gods and ancestors and fearing their witches, were every bit as logical as we (or just as illogical). It is actually no more "logical" to believe in a divinely created moral universe than to believe in an amoral self-generating universe. Foundational beliefs stand beyond the operations of logic. Our researches were framed by an interest in the moral construction of the universe and the nature of belief.
In those days it was axiomatic for anthropologists that, however peculiar they might seem to us, the strange beliefs of a foreign tradition make sense. Explanations of other minds based on mystery, mystique, native credulity or mysticism, were out. Moral evolution was replaced by a down-to-earth approach to alternative ways of living and dying. We took on a hardy scepticism and a nuts-and-bolts demand for evidence. The point about doing fieldwork was to learn how a world-view was
196
MARY DOUGLAS
adapted to what the people were trying to achieve, especially to what they were doing towards living together in society. Hence our attention to ritual and symbolism. Rain rites, for example, would be a collective act of affirmation. The rite did not attempt to prove the priests' control of meteorology, it was done to affirm publicly the moral aspect of the natural order. Spectacular ceremonials to appease the gods were also performed for the sake of influencing each other's minds.
I would never have felt impelled to attempt an anthropological reading of Leviticus if during African fieldwork I had not been confronted by local dietary rules, and so thought of looking up the passage in chapter 11 on the forbidden animals. I actually cited Leviticus and the parallel passage in Deuteronomy in my "Animals in Lele religious symbolism" (1957). What I wrote ten years later about uncleanness and pollution in Purity and Danger (1966) was driven by fieldwork experience, stiffened by training in Oxford anthropology and enriched with some reading about the psychology of perception. But before looking up those baffling chapters, I had never read the Bible, either at school or at university or subsequently. When I came eventually to read the scholarly commentaries on the Mosaic dietary laws I was surprised to find so much disagreement on such an important subject. Though with some minor variations scholars almost unanimously associated the forbidden animals with unpleasant characteristics, there was no agreement and no satisfactory explanation either in the book or outside it about why each particular species should have been selected and not others which might equally be abominated.
Reflecting on these animals I was drawn to focus on the class of unclassifiable things. The forbidden land animals were certainly described as such a class, and I extended it with some confidence to water creatures and speculatively to those in the air that could not be identified. I proposed a theory of anomaly, a universal feeling of disquiet (even of disgust) on confrontation with unclassifiables. Taking the Levitical classification system as it revealed itself, the said abominable species failed to show the taxonomic requirements of inhabitants of the three environmental classes, land, air, water, and the abominability of species that "go upon the belly" in all environments went by the same rule: the forbidden animals were species that escaped being classified. Consistently with the main thrust of social anthropology of my period, the argument explained abominability, but denied magicality and favoured the rationality of the Mosaic dietary code. It was gratifying to find that some Bible scholars accepted the idea that the puzzles of the abominable animals in Leviticus and Deuteronomy could be laid to rest, the prohibitions being part of the process of tidying up the classifications of the environment (see Levine's FPS Commentary, Leviticus (1989), 243). But a puzzle remained.
The central argument of Purity and Danger was that classifications are not otiose. They do something, they are necessary in organization. The pollution theory that I have seen develop over the last thirty years shows that where lines of abominability are drawn heavy stakes are at issue. The classification of the universe is part and parcel of social organization, and the categories are useful in defining who can be admitted where, and who comes first and who comes second or nowhere at all. This works so effectively elsewhere that I was implicitly waiting for it to be found true of biblical pollution (see my "Sacred contagion" (1996)). It applies well enough, in fairly obvious ways, for the cult of the tabernacle and the dignity of the priesthood, but for the organization of society the doctrine of pollution did nothing except draw a boundary round the people of Israel against outsiders. Nothing happens at the
LAND ANIMALS, PURE AND IMPURE
197
level of action to explain the selection of forbidden animals. Against everything I believe, the cognitive scheme which left these creatures unclassified hung in the air uselessly. If chapter 11 of Leviticus was a case for pollution theory the classifying of the animals should correspond to some important classifying for the internal organization of society. But the more that pollution theory developed, and the more that pollution was seen as the vehicle of accusations and downgradings, the more I was bound to acknowledge that it does not apply to the most famous instance of the Western tradition, the Pentateuch. All of this volume is an attempt to explain why. General pollution theory still stands, but its application to the Bible is limited. The forbidden animals turn out to have a much more interesting role than ever I imagined. [...]
*
Behold, I establish my covenant with you and your descendants after you, and with every living creature that is with you, the birds, the cattle, and every beast of the earth with you, as many as came out of the ark. (Gen 9: 9-10)
For thou lovest all things that exist, and hast loathing for none of the things which thou hast made, for thou wouldst not have made anything if thou hadst hated it. (Wisdom 11:24)
God met his people on Mount Sinai and continues to meet them in the tabernacle. In sacrifice the body of the sacrificial animal becomes another microcosm in its own right, corresponding to the tabernacle and the holy mountain. Then the sequence of cultic laws is interrupted by the narrative in chapters 8-10. When the law-giving is resumed it develops a different bodily microcosm. This time the body of the worshipper is made analogous to the sanctuary and the altar. Whatever will render the altar impure will do the same for the Israelite's body. The laws of impurity sketch out the parallel in meticulous detail over chapters 11-15. The animal that is taken into the body by eating corresponds to that which is offered on the altar by fire; what is disallowed for the one is disallowed for the other; what harms the one harms the other. One thing that the book never says is that it is bad for the health of the body to eat any of the forbidden animals.
Land Animals under the Covenant
Chapter 11 is probably the best known in Leviticus because it deals with the Mosaic dietary laws. It has been taken to imply that the forbidden animal meats are abominable, detestable, or unedifying in one way or another. Taking account of the full context, which is the rest of the Pentateuch, it would be difficult to overlook one biblical principle: God is compassionate for all living things; not only to the humans, he is good to all his creatures (Psalm 145: 8-9). So if he himself does not detest them, why should he tell humans to detest any of his animals? This is a serious and central doctrinal problem.
Two kinds of covenant are the basis of chapter 11. First the covenant with Noah and his descendants in which God said he would never again punish the land and the
198
MARY DOUGLAS
living things on it for the evil things done by mankind, and made the rainbow its sign. It is emphatically also a covenant with the animals: "This is the sign of the covenant which I make between me and you and every living creature that is with you, for all future generations" (Gen 9: 12).. ."I will remember my covenant which is between me and you and every living creature of all flesh..." (Gen 9: 15, repeated in vv. 16 and 17). A few verses earlier in the same chapter God has required a reckoning for the life-blood of humans. "Only you shall not eat flesh with its life, that is, its blood. For your life-blood I will surely require a reckoning; of every beast I will require it and of man" (Gen 9: 4,5). In Genesis God gave man dominion over animals. Robert Murray1 has argued persuasively that dominion for Genesis always entails responsibility. Leviticus presents the further implications of human dominion over animals. A one-sided pledge from God cannot quite be called a covenant: the animals are not bound by any counter-obligations, unless by a stretch of the imagination the command to them at the creation to go forth and multiply counts as such.
Later, the covenant with Abraham is a promise of fertility to his descendants. It does not mention the animals, but extravagantly it says that his descendants will be as innumerable as the dust (Gen 13: 16). It echoes the blessing of Genesis to Noah and his sons, "Be fruitful and multiply, and fill the earth" (Gen 9: 1), and "Be fruitful and multiply, bring forth abundantly on the earth and multiply in it" (Gen 9: 7), the very words used for his blessing on the creatures of the water and the air after they had been created (Gen 1: 22). There is no doubt that this God is concerned with fertility, and that his promise is linked with their obedience.
The covenant with Moses on Sinai is the explicit assertion of God's overlordship over the people of Israel and their livestock. It specifically includes the servants and the cattle in the sabbath observance (Exod 20: 8). From householder to children, to servants, to cattle, the animals come under the lines of authority drawn by the Sinai covenant. Sabbath observance only affects the work animals, the ox that treads the husks off the grain, that draws the cart, that turns the water-wheel.2 Exodus also makes the point strongly by requiring that the male first-born of the domestic animals be offered to the Lord just as the first-born of humans. "Consecrate to me all the first-born; whatever is the first to open the womb among the people of Israel, both of man and of beast, is mine" (Exod 13: 2). "You shall set apart to the Lord all that first opens the womb. All the firstlings of your cattle that are males shall be the Lord's" (Exod 13: 12). The rule for land animals which always sounds so complicated is quite simple when the covenant is seen to be its guiding principle. God is the feudal Lord. From this it follows that no one is allowed to harm God's people or use God's things, nor must his followers harm each other, or harm the other living beings on his territory without his express permission. This he gives for the killing of herd animals in sacrifice, and use of their carcasses.
The question of whether they do or do not come under the covenant is paramount. Leviticus divides land animals into two categories, first, the herds and flocks which share the lives of their owners, travel with them, and provide their sustenance, and second, all the rest. The pure animals come under the terms of the covenant of their masters, and their treatment is strictly regulated. The feudal relationship extends from God to his people and to their livestock.
The teaching about the sanctity of blood derives from this feudal relationship. God protects the people of Israel, his rites give them covering, sacrifice is the means he has given to them for expiation. Sacrifice protects them from the consequences of their
LAND ANIMALS, PURE AND IMPURE
199
own behaviour, even from his just anger. They are never, ever, allowed to eat blood, but he has given them the right to consecrate the lives of their herd animals, to use their blood to make atonement to him for their sins, and to eat the blood-free flesh for their own nourishment (Lev 17: 11). This solemn injunction teaches the sanctity of life (the life is in the blood). In religious terms, the mosaic dietary code is an invitation to Israel to join in the divine work of creation by living a life that honours the way God made the world and the covenants God has made with his people.
The Two Texts
The two texts in Leviticus and Deuteronomy start by running in close parallel. They give a perfectly logical classification which echoes the opening chapters 1—7 on sacrifice with a description of the domesticated ruminants of their herds, cattle, sheep, goats, which may be consecrated for offering on the altar. Then follows in both texts a careful set of rules to discriminate near-misses, candidates for entry into the class of domestic ruminants which fail because they show one but not both the required criteria.
Marching in step, the two texts say that the unclean animals are only "unclean for you". Because the dietary rules about land animals derive from the covenant, they only apply to the people of Israel. Deuteronomy seems to say by its prefatory remark that abominable things are the things forbidden as unclean in 14: 7-8. The equation of unclean with abominable in Deuteronomy is the source of the idea that the
Table 1   Land animals, pure and impure, Deuteronomy 14: 3-9 and Leviticus 11: 2-8
Deuteronomy
3. You shall not eat any abominable thing.
4. These are the animals you may eat: the ox, the sheep, the goat,
5. the hart, the gazelle, the roebuck, the wild goat, the ibex, the antelope, 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 foot 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 but does not chew the cud,
is unclean for you. Their flesh you shall not eat and their carcasses you shall not touch.
Leviticus
2. These are the living things which you may eat among all the beasts that are on the earth.
3. Whatever parts the hoof and is cloven-footed and chews the cud, among the animals, you may eat.
4. Nevertheless among those that chew the cud or part the hoof, you shall not eat these: The camel, because it chews the cud but does not part the hoof, is unclean to you.
5. And the rock badger, because it chews the cud but does not part the hoof, is unclean to you.
6. And the hare, because it chews the cud but does not part the hoof, is unclean to you.
7. And the swine, because it parts the hoof and is cloven-footed but does not chew the cud, is unclean to you.
8. Of their flesh you shall not eat and their carcasses you shall not touch.
200
MARY DOUGLAS
forbidden animals have some detestable characteristic, the focus of so much scholarly ingenuity. But in Leviticus the unclean animals are not abominable.
The microcosm is based on the body of anyone of the congregation of Israel about to take nourishment; the body is equivalent to the altar and so is his hospitable board round which he gathers his family and friends. Way back in the time of the Leviticus writer the body was already the analogue of the altar. Not a secular analogue, for in a total religious system (such as that of Leviticus) the word secular does not have much meaning. The table, and all who eat at it, and everything that has been cooked for them to eat, are under the same law of holiness.J Body for altar, altar for body, the rules which protect the purity of the tabernacle are paralleled by rules which protect the worshipper. What he can eat without contracting impurity and what can be offered to God in sacrifice are the same.
An interesting difference between the texts of Deuteronomy 14 and Leviticus 11 is that the Leviticus opening, with its reference to living beings on the earth (Lev 11:2), recalls the account of the creation in Genesis, "Let the earth bring forth living creatures according to their kinds: cattle and creeping things and beasts of the earth, according to their kinds" (Gen 1:24). The opening verses of Leviticus II are only the beginning of a larger survey of land animals (the beasts of the earth) which is not featured in Deuteronomy.
When Leviticus has listed the different types of land animals that are impure and must not be eaten, look round and see what is left - nothing! All these land animals are either clean or unclean, pure or impure. Now that verse 26 is taken to refer to animals with solid hoofs, we can use the following comprehensive typology of the land animals:
(i) animals of the flocks and herds, ruminants, split hooves;
(ii) ruminants without split hoofs, e.g. camel, rock badger, hare;
(iii) non-ruminant with split hoofs, pig;
(iv) solid hooves, e.g. asses, horses;
(v) paws, e.g. lion, civet cat, dog, hyena;
(vi) list of eight land animals that go on their belly: the mole, the mouse, the great lizards, the gecko, the land crocodile, the lizard, the sand lizard, and the chameleon.
Table 2   Living beings on the earth
humans (bipeds)
animals (quadrupeds)
ruminant
non-ruminant
(i) split hoofs (1-4)
(ii) no split hoofs (4, 5, 6)
(iii) split hoofs (7)
(iv) solid hoofs (26)    (v) paws (27)   (vi) land-swarmers (29-38)
LAND ANIMALS, PURE AND IMPURE
201
The zoological criteria are good enough to make an exhaustive list of land animals. By the end of the chapter everything living on the land has been included: going on hoofs, going on paws, gliding on the belly - what else is there? Everything has been accounted for....
The first set are under the covenant, and clean. All other land animals, excluding the first set, are unclean or impure, and their dead bodies not to be touched or eaten. Deuteronomy makes a point of allowing the killing and eating of wild counterparts of domestic herds (Deut 13: 5), of which Leviticus does not speak. Were Leviticus explicitly to permit secular slaughter, it would undermine the covenant basis of the Levitical rules constraining humans from eating animals that they have not reared. A similar silence covers what is to be done with blemished animals from their flocks and herds which are not allowed to be sacrificed. Presumably they are quietly eaten.
Many four-footed animals are not specifically mentioned in the catalogue; for instance, rodents such as squirrel, rabbit, rat, cat, do not need to be listed since they are covered by the rules for animals that go on their paws. Finally, not on the chart above, there are other land-dwellers indicated as "swarming things that go on the belly, go on all fours, or with many feet" (Lev 11: 42), including land-dwelling insects, snakes, worms, spiders, centipedes. Leviticus calls these by a word that is translated as abominable, but Leviticus uses a different word from that used by Deuteronomy and also translated the same way. Leviticus also confuses the issues by applying a general defilement term for land-swarmers: the people must not "defile" themselves by contact with their carcasses (Lev 11: 43-4). The sanction for infringing the rule is not severe. Though uncleanness is very contaminating, the sanction is just to wait till sundown (Lev 11: 31, 39). If someone has gone further than just touching an unclean corpse, the sanction is still mild, he must wash his clothes and remain unclean until evening; and anyone who carries its carcass shall wash his hands and remain unclean until evening (Lev 11: 40). We should not exaggerate the penalties or the severity of the rules.
The rule of uncleanness only affects contact with the dead carcass. In this respect contact with the carcasses of land animals are accorded a similar, though lesser, impurity as contact with human corpses. The people of Israel are enjoined to have towards their livestock some of the responsibilities of a feudal lord to his followers. Both humans and livestock are called to be separate and pure in the interior circle of a world of unclean nations or unclean animals. Milgrom says that the effect of the criteria for edible quadrupeds (Lev 11: 3) is to limit Israel to three domestic species: sheep, goats, and cattle.4 But there is more to it than just not eating. Notice some of the tacit restrictions that follow from these rules. All land animals have been classified within the system. They can all be touched alive, but the only ones that can be touched after death are the classified ruminants. This means that only the latter can be killed for sacrifice. While they are alive camels and asses can be harnessed, loaded, ridden, dogs can be beaten, cats can be kicked, mice can be trapped, without incurring impurity, but once they are dead they convey uncleanness.
In effect the rule against touching a dead animal protects it in its lifetime. Since its carcass cannot be skinned or dismembered, most of the ways in which it could be exploited are ruled out, so it is not worth breeding, hunting, or trapping. These unclean animals are safe from the secular as also from the sacred kitchen. The rule is a comprehensive command to respect the dead body of every land animal. If anyone were to take it seriously it would be very restrictive. The verb to touch has also the
202
MARY DOUGLAS
idea of harming, damaging, laying hands upon as if to steal or strike. An example is in Genesis when Abimelech commands his followers not to harm Isaac: "Whoever touches this man or his wife shall be put to death" (Gen 26: 11), and says to Isaac: "We have not touched you and have done to you nothing but good" (Gen 26: 29). The rule of not touching the corpse makes the skins useless for fur coats or fur blankets, no leather waistcoats or bags, no shoe leather or wine-skins. Their bones and teeth cannot be carved for combs, buttons, containers, dice, jewellery, utensils. Their gut cannot be used for stringed instruments, or their stomachs or bladders for bags, or their sinews for sewing. In practice the penalty is so light that the rule would hardly prevent a taxidermist or tanner from pursuing his trade, so long as he purified his clothes and himself before approaching the tabernacle. Nonetheless it is still unequivocally forbidden to touch these creatures when dead. The tremendous domestic complications entailed by the high degree of contagiousness might deter the furrier, and it would be awkward for the wearer of a mink coat to have to keep washing, and no one else could so much as shake hands with the wearer without afterwards performing the same ablutions. To be classified unclean ought to be an advantage for the survival of the species.
Interpretations of Uncleanness/lmpurity
Though Moses admonished Aaron and his sons to distinguish between the holy and the common and between the clean and the unclean (Lev 10), he did not explain what is meant by unclean, or holy. The sages did not make much sense of it at all. Why pig is counted unclean in the Bible has been the subject of much speculation. Changing the word to "impure" does not really help. Why ever should a non-ruminant with cloven hoofs be counted impure?
Some scholars favour the idea that pig has to do with cults of the dead in Egypt or Canaan, but this does not explain the uncleanness of the other three animals named with it. When the classification is so comprehensive it cannot be sound to take the animals as if they were separate items in a catalogue without headings or subheadings. If some of the rules have the effect of banning predators, or blood, or carrion-eaters, it is not a comprehensive explanation. It is also too Hellenistic, too oriented to feelings, for this book. The same for the idea of Philo, the first-century c E Jewish philosopher,5 that the forbidden species each signifies a vice or virtue. In a long, rambling homily he takes each forbidden animal separately and explains its prohibition in terms of symbols. His fanciful allegories are not rooted anywhere in the Bible text, only in the imagination of the philosopher, and so inevitably his moral preoccupations dominate the reading. He derives from Leviticus 11 a lesson to control gluttony, passions, and desire. The animals that go on the belly are forbidden so as to teach the people not to pay attention to their bellies. It is not a naturalist explanation, he does not consider the forbidden animals to be bad in themselves, rather the contrary:
All the animals of land, sea or air whose flesh is the finest and fattest, thus titillating and exciting the malignant for pleasure, he sternly forbade them to eat, knowing that they set a trap for the most slavish of the senses, the taste, and produce gluttony, an evil very dangerous both to soul and body.
LAND ANIMALS, PURE AND IMPURE
203
Philo was not working from any tradition that was close to the writing of the text, for he does not draw on the rest of Leviticus or Deuteronomy to construct his sermons.
Some moralizing interpreters have regarded the laws as oblique commands that will restrain human "omnivorousness and ferocity";6 others again as arbitrary commands to test obedience.7 Others give up on interpreting at all, treating the rules as inexplicable, though deriving from an ancient time when they presumably once made some kind of sense. Some treat the Levitical scheme as a relic of a pastoral way of life.8 For this to be serious there would need to be a theory of why some relics remain strong when their supporting context has passed away, while others are forgotten. In default of such a theory it can still be argued that the Leviticus writer could not betray all the pastoral tradition. His commitment to the idea of the ancient covenant would preserve rules about herd animals. However, the general disarray among rival interpretations testifies to the lost tradition. Not surprisingly the general public is ready to believe that there is something abhorrent about the creatures which the book tells them to abhor. To this day it is common to hear distinguished scholars explain ritual purity by natural reactions:
Many people wince at having to pick up a dead animal; most people (except two-year-olds) try to avoid touching defecation; corpses inspire a natural feeling of awe, and we hesitate to touch them; washing off semen and blood is almost natural, and certainly not hard to remember. Even gnat-impurity, which sounds picky, is not hard to understand. Who wants a fly in one's soup?9
Most of the discussion is based on the Deuteronomy formula, not on Leviticus. The first mistake in this quotation is to have used lines from Deuteronomy as if they came from Leviticus and as if they all meant the same thing. Deuteronomy says that winged insects are unclean, but not Leviticus. Humans are constantly under their attack, flies feast on babies' eyes, they breed their maggots in the larder, walk contemptuously over food, they suck blood and sting to frenzy. It is plausible that invasive insects and creepy-crawlies might be disliked universally. Another mistake is to use supposedly natural or "almost natural" reactions to justify all the unclean-ness rules. The nuisance value of insects makes this explanation plausible, but the rules are not mostly about insects. Why revile shy animals like hares and useful animals like camels by classifying them with the naturally dislikeable? The naturalist explanation must be wrong for a book so sophisticated as Leviticus, and for anthropologists it is always wrong to take natural as a universal category, forgetting that nature is culturally defined.
Again, the text itself specifically says that these are rules made for the people of Israel; what is designated as unclean for them is not unclean for the whole of humanity. Thus the rules of impurity are not a way of promoting a universal hygienic principle or pronouncing a general health warning. The only explanation will be in the rest of the rule system. Many civilizations have been built on camel meat, or pork, and though hyrax is hard to get, there is nothing bad for you in adding hyrax or hare to your diet; some people habitually eat blood, and the dietary value of suet fat can be underestimated. One popular explanation for the banning of the water-swarmers (which in Leviticus are not unclean) is that they are scavengers:
204
MARY DOUGLAS
pigs will eat carrion; shrimps and crabs feed upon dead fish; so dirty feeders are forbidden. This explanation is weak because a lot of animals would opportunistically consume carrion if they found it, and anyway the text says nothing about carrion-feeding animals.
The concept of dirtiness has contaminated the conceptual field; the idea of disgust at eating unclean things dominates interpretation. The kitchen, medical, and bathroom senses intrude. Leviticus certainly plays upon disgust at bodily exudations in its long disquisition on uncleanness of bleeding and leprosy in chapters 12-15. If impure was not originally a term of vilification it certainly has become one. Appeals to medical and aesthetic principles are not the way to interpret an enigmatic law in Leviticus: the only safe path is to trace the contrast sets and parallels the book itself develops. The impurity of an animal kind is part of the technical meaning of ritual purity.
In itself the idea of impurity is not difficult to translate. The word is well chosen from secular contexts where unclean, defiled, impure, dirty correspond to a situation which calls for an act of cancellation. But washing, polishing, burnishing, are too superficial to carry all the meanings of purification. In Christianity impure is used for the defilement of sin, with frequent reference to the parallel with a soiled garment, it is taught that the repentant soul requires a cleansing rite. But the listed uncleannesses of Leviticus are not sin in general, they are a separate set of sins, they depend on physical contact only, and the central principle is that the contaminated body has contagious power, which entails that all its future physical contacts convey contamination. The rules prescribe how the object spreading defilement must be washed, destroyed, or somehow stopped, according to the gravity of the defilement.
The word for impure, tame, is worked very heavily in Leviticus and used sparsely elsewhere in the Bible.10 We may ask why it became such a favourite word for the priestly writer, but the first question is how it relates to holiness. Once again, the most illuminating passage to explain what it is about is the warning that God gave to Moses about the sanctity of Mount Sinai. In Exodus 19: 10-24 he tells Moses to make a fence round the mountain and to prevent the people from approaching the mountain, not even to touch the edge of it. He tells Moses to tell them to purify themselves, to wash their clothes and be ready for the day when he will appear to them, but to wait until they are summoned by trumpet: "Go down and warn the people, lest they break through to the Lord to gaze, and many of them perish. And also let the priests who come near to the Lord consecrate themselves, lest the Lord break out upon them" (Exod 19: 21-2). The danger is two-edged: the people might break through or the Lord might break out, and in either case, people will die. This is the effect of holiness. The holy thing that is not correctly guarded and fenced will break out and kill, and the impure person not correctly prepared for contact with the holy will be killed. Furthermore, a person who has had the misfortune to "contract" holiness, to use Milgrom's term, may inadvertently contaminate other unprotected things or persons, merely by contact.11
The nearest usage in European languages for the idea of contagion is in the discourse of honour, especially with reference to the virtue of women or the honour of a knight. The taint of dishonour gives a fair idea of impurity and violation. In Mediterranean cultures a woman's honour must be protected at all costs; if she is defiled her violator must be killed; if her father or brothers fail to cancel the offence
LAND ANIMALS, PURE AND IMPURE
205
they will be dishonoured too, and the whole family. It is not a metaphor, it is a concept about behaviour that has practical consequences: none of her sisters will be able to marry, no respectable person will do business on equal terms with the menfolk, they will not be able to hold up their heads at a meeting, the contaminated family is ruined.12
Israel had a patronal society in which the patron-client relation is expressed by the client's respect for the honour of the patron. At meetings between lord and vassals the latter bring specified gifts of food to be ceremoniously shared. Leviticus says that the cereal offering must expressly be given with the "salt of the covenant" (Lev 2: 13), which suggests that the terminology and values of covenant would have been current and easily interpretable for the people for whom the book was written. Defilement as a violation of holiness is a particularly apt expression for an attack on the honour of God perceived as a feudal lord. The word for holy has the sense of "consecrated", "pledged", "betrothed", as "sacrosanct" in modern English, something forbidden for others,13 not to be encroached upon, diluted, or attacked. A key text for understanding impurity in Leviticus would be: "For you are a people holy to the Lord your God; the Lord your God has chosen you to be a people for his own possession, out of all the peoples that are on the face of the earth" (Deut 7: 6). This is followed by a reference to the principle of requital on which the covenant rests (Deut 7: 9-10). A few verses later the same text goes on to say what being holy or reserved to the Lord entails in terms of behaviour. It corresponds to the requirements of chastity and fidelity in the discourse of honour and betrothal, which is similar to, or rather, modelled upon the discourse of alliance and covenant. "You shall therefore be careful to do the commandment, and the statutes, and the ordinances, which I command you this day" (Deut 7: 11).
In addition to obedience the covenant with the overlord requires protecting his honour, or abstaining from insult. His power protects his people or his things and places, and to insult any of them is an insult to his honour. The parallel with the discourse of honour explains why sins cause uncleanness to adhere to the sanctuary and to the altar. Jacob Milgrom is curious to know why the altar should need atonement when the altar has not sinned. He develops a convincing theory of contagion from sin clustering on and around the holy places until it is washed off by the rite of atonement.14 His analysis of contagious impurity is impeccable, but one can notice that the language of dirt and ablution is unnecessarily materialist. In the courts of chivalry a warrior would recognize that his armour is dishonoured if he himself is impeached: as well as his children, and father and mother, his helmet, his coat of arms, his house, all are tainted and made worthless by the contagious dishonour. Blood washes off the major taint, a noble gift cancels a minor fault. In the same way, bringing uncleanness into the Lord God's sanctuary makes it impure since the place shares in the insult to God.
Leviticus has first described the pure animals as ruminant hoof-cleavers, and then has gone on to exclude "ruminants" which do not cleave the hoof and the one non-ruminant species which does (the pig). This order of listing gives the impression of excluded animals trying to get into the privileged enclosure so that they too could be consecrated and share in the Lord's cult. There would have been pressure from enterprising cooks seeking to alleviate the monotony of the menu. The sense of pressure to be included adds to the meaning of the animals excluded for having only one but not both defining features.
206
MARY DOUGLAS
In the midrash the image of a reclining pig stretching out its cloven hoofs and saying: "Look, I'm pure," while concealing the fact that it does not chew the cud, is used to characterize the hypocrisy of the Roman empire, which posed as being dedicated to law and justice, while actually oppressing the peoples it ruled.15
Frivolously one can ask why pig or any other animal would seek to be accounted pure when the pure animals are destined for early death and the fire of the altar. On a secular view, having one but not both the criteria for purity would be a saving blessing, but the context is religious.
The meaning of purity depends on the sense of God's awful majesty, manifest in his creation. Exodus describes it in a narrative of volcanic explosion, thunder, fire. Deuteronomy describes it with words about God's power, and with verbal warnings of disaster. Leviticus conveys it by double, triple, multiple microcosms. The people, with their children and their servants and their domestic animals too, benefit from his covenant. As vassals of God their unworthiness is immeasurable, but yet they are invited to eat at his table, and may eat the food that is offered to him. Sacrifice is a communal feast. Theoretically the people of Israel never eat meat except in God's company, in his house and with his blessing. They have been singled out for the honour of being consecrated to God, to be his people. The height and the depth of this honour is inexpressible. At another level it is a parallel honour for their flocks and herds, the cloven-hoofed ruminants, to be singled out of all animal kinds to be consecrated to God. This paradigm turns the covenant animals into vassals in relation to the people of Israel, as are the people of Israel the vassals of God.
Sacred Contagion
We still can ask what interest Leviticus could have had in elaborating the concept of holiness and impurity in these ways. The full answer must relate to the fact that belief in the maleficent power of demons has been demolished. The theodicy has to be changed: his friends will no longer be able to tell a sick man that he has been seized by a leprosy demon or a woman that her child has died because a female demon took it. Suffering and sorrow still remain, and death. The priests are expected to explain, give comfort, and help. This is what the doctrine of purity does. If you fall sick, it could be that God has broken out on you because you unknowingly incurred holiness or impurity. This is a close parallel to the superseded idea that a demon might have caught you. A sacrifice will put it right, or a wash and waiting till evening, according to the gravity of the transgression. The word "unclean" is particularly apt for relating the field of demonological medicine to the new regime, it affords a theory of pain and suffering free of demons and affords an alternative explanation for bodily afflictions.
So why should touching unclean animals provoke a dangerous breaking-out of this kind? The insult to God is to have come into his sacred place after profane and contagious contact with the corpse of one of his creatures. Taken together the food purity rules and the touch purity rules are part of a unified doctrine in which corpse pollution, bloodshed, and unsanctified death are classed as breaches of covenant.
LAND ANIMALS, PURE AND IMPURE
207
It has been a puzzlement to Christian readers that Leviticus puts unclean contact into the same bracket as breaches of the moral code. However, there is nothing puzzling about both kinds of disobedience to the Lord's command being treated together. To touch an unclean thing and then to approach the tabernacle puts the person in need of atonement. Leviticus in chapter 5 begins the topic of uncleanness:
Or if anyone touches an unclean thing, whether the carcass of an unclean beast or a carcass of unclean cattle, or a carcass of unclean swarming things, and it is hidden from him, and he has become unclean, he shall become guilty (Lev 5:2). When a man is guilty in any of these, he shall confess the sin he has committed and he shall bring his guilt offering to the Lord for the sin he has committed... and the priest shall make atonement for him for his sin. (Lev 5: 5-6)
And again in the summing up, chapter 7 emphasizes the contagions principle:
Flesh that touches any unclean thing shall not be eaten; it shall be burned with fire. All who are clean may eat flesh, but the person who eats of the flesh of the sacrifice of the Lord's peace offerings while an uncleanness is upon him, that person shall be cut off from his people. And if anyone touches an unclean thing, whether the uncleanness of man or an unclean beast or any unclean abomination, and then eats of the flesh of the sacrifice of the Lord's peace offerings, that person shall be cut off from his people.
(Lev 7: 19-21)
This is very emphatic language, repetitive, classificatory, and redolent of mythopoetic analogy. Such a statement from an archaic thought style cannot be decoded into modern terms. The interpreter must not read emotional quality into language which is primarily cast in a spatio-temporal mode. The contact has been forbidden, and the person who has become contagious shall not carry the contagion sacrilegiously to defile the holy place or to eat the flesh of the Lord's peace offerings; he will be punished. A domestic ruminant is the designated medium of atonement and the priest following the instructions for a sin offering in chapters 4 and 5 will make atonement for a sinner and he will be forgiven. He can live his ordinary life in this contagious state, but because of the contagion he and other persons he may contact will commit sacrilege if they take part in the cult of the tabernacle. He can expect to be criticized by his believing fellows and be made to take the blame for a community-wide disaster, and possibly expelled, like Jonah by the sailors.
Unclean is not a term of psychological horror and disgust, it is a technical term for the cult, as commentators have often pointed out. To import feelings into the translation falsifies, and creates more puzzles. The technique of delayed completion postpones the meanings until chapter 17. At that point Leviticus commands the people not to eat blood, not to eat an animal that has died an unconsecrated death, that is, an animal that has died of itself, or an animal torn by beasts, presumably with its blood still in it (Lev 17: 8—16; see also Deut 14: 21). The dietary laws thus support the law against unconsecrated killing. The Leviticus writer's reverential attitude to life, animal and human, explains the animal corpse pollution rules. "Thou shalt not stand upon [profit from] another's blood" (Lev 19:16). The case
208
MARY DOUGLAS
of the animal's blood and the case of the human's blood are parallel. Ritual impurity imposes God's order on his creation.
NOTES
1 Murray 1992.
2 The ass that carries loads and persons is not mentioned here. A half-way category, it is given a half-way treatment when it comes to offering the first-born to the tabernacle: "Every firstling of an ass you shall redeem with a lamb ..." (Exod 13: 13).
3 After the destruction of the temple when the Mishnah substituted the cleanness of the worshipper's body and food for that of the altar and sacrifice, they already had a strongly developed precedent in Leviticus, chapter 11. Neusner 1977.
4 Milgrom 1989.
5 Philo 1939: 99-102.
6 Kass 1994: 12.
7 Maimonides 1881.
8 Houston 1993.
9 Sanders 1990: 145.
10 tame, impure, occurs 89 times in the Bible; 47 times in Leviticus; 8 times in Deuteronomy, and not at all in Exodus.
11 Milgrom 1991: 443-56 argues that in Leviticus holy things do not transmit holiness to persons, but contact kills them, whereas Ezekiel following an older tradition taught that holy things could transmit holiness to persons as well as to objects.
12 Campbell 1964.
13 Tigay 1996: 86.
14 Milgrom 1983.
15 Tigay 1996: 139.
REFERENCES
Campbell, John, Honour, Family and Patronage (Oxford University Press, 1964). Douglas, M., "Animals in Lele religous symbolism", Africa 27 (1957).
-"Sacred Contagion", in John Sawyer (ed.), Leviticus (Sheffield Academic Press, 1996),
86-106.
-Purity and Danger: An Analysis of the Concepts of Pollution and Taboo (London,
Routledge & Kegan Paul, 1966). Evans-Pritchard, E. E., Witchcraft, Oracles, and Magic among the Azande (Clarendon Press,
1937).
Houston, Walter, Purity and Monotheism: Clean and Unclean Animals in Biblical Law
(JSOTS 140, Sheffield Academic Press, 1993). Kass, L. R., The Hungry Soul, Eating and the Perfection of Our Nature (Free Press, 1994). Levine, Baruch, The FPS Commentary, Leviticus (Jewish Publication Society, 1989). Maimonides, Moses, Guide for the Perplexed (London, 1881).
Milgrom, Jacob, "Israel's sanctuary: the priestly 'Picture of Dorian Gray'", in Studies in Cultic Theology and Terminology, Studies in Judaism in Late Antiquity, ed. Jacob Neusner, 36 (Brill, 1983), 390^00.
-"Rationale for cultic law: the case of impurity", Semeia 45 (1989), 103-9.
-Leviticus 1-16, The Anchor Bible (Doubleday, 1991).
LAND ANIMALS, PURE AND IMPURE
209
Murray, Robert, The Cosmic Covenant, Heythrop Monographs, 7 (Sheed and Ward, 1992).
Neusner, Jacob, "History and Structure, the Case of the Mishnah" FAAR 45/2 (1977), 161-92.
Philo, with English translation, ed. E H. Colson, viii: The Special Laws (Harvard University Press, 1939).
Sanders, E. P., Jewish Law from Jesus to the Mishnah, Five Studies (SCM Press, 1990). Tigay, J., Deuteronomy, the People's Torah (Jewish Publication Society, 1996).
18
The Winnebago Trickster Figure
Paul Radin
From Paul Radin, "The Nature and the Meaning of the Myth," in The Trickster: A Study in American Indian Mythology (reproduced by kind permission of the Philosophical Library, New York, 1972 [1956]), pp. 132-54.
Paul Radin (1883-1959) was an outstanding American anthropologist of the Boasian school noted for his close work with the Winnebago. As the present essay shows, Radin had tremendous appreciation for the insight conveyed in Amerindian narrative.
This excerpt from Radin's interpretive essay on the trickster myth of the Winnebago Indians is best read in conjunction with his record of the entire cycle set out earlier in the book. The trickster is an amoral yet essentially good-natured creature whose adventures as both duper and dupe express a kind of ontogeny, a coming into being of consciousness and differentiation that may be understood at both psychological and cosmological levels. The Trickster cycle exemplifies the role of chaos and ambiguity in creation myth. Radin also points to the centrality of the comic. As Diamond (1972) indicates, the trickster personifies a frank recognition and acceptance of human ambivalence.
Trickster figures are known from many parts of the world. Ind ividua I trickster gods are found within the pantheon in polytheistic religions like those in ancient Greece or in West Africa. Tricksters are central to the storytelling of many Native American groups; I recommend Howard Norman's exquisite renderings of Swampy Cree versions (1976). A recent work on Southern African Bushmen emphasizes the role of trickster figures there and makes a number of useful connections with the literature on shamanism (Guenther 1999). Comedy and irony are also central to spirit possession and widespread in religious traditions, albeit rather absent from the sustained seriousness characteristic of most public or "official" rituals and representations of Christianity and Islam.
As a Boasian, Radin worked closely with gifted raconteurs in order to record cultural texts, and paid careful attention to what participants themselves said about the material and to the individual variants. In his classic accounts of Primitive Man as Philosopher (1957a [1927]) and Primitive Religion (1957b [1937]), Radin argued powerfully against the Levy-Bruhlian view of a prelogical mentality and, indeed, against the
THE WINNEBAGO TRICKSTER FIGURE
245
entire evolutionist approach, the "fundamentally misleading doctrine that primitive peoples represent an early stage in the history of the evolution of culture" (1957a [1927]: x). For Radin there are always individuals holding more or less religious perspectives or who are more or less inclined to engage in philosophical speculation. Radin's portrait of intellectual and reflective speculation is far broader and richer than that portrayed by Tylor. Just as he demonstrates, in the chapter immediately following the ones presented here, the satirical dimension of the Trickster tales, so he locates "skepticism and critique" among certain thinkers in any "primitive" society (1957a: Chapter XIX). Other significant members of the Boasian school include Sapir (1956 [1928]), Lowie (1935, 1948), and Benedict (1934). They frequently emphasized the emotional dimension in religious expression and its relation to personality. For a lively rendition of the Americanist language-based tradition by contemporary practitioners see Valentine and Darnell (1999).
The Winnebago word for trickster is wakdjunkaga, which means the tricky one. The corresponding term for him in Ponca is ishtinike, in the kindred Osage, itsike and in Dakota-Sioux, ikto-mi. The meaning of the Ponca and Osage words is unknown, that of the Dakota is spider. Since all these three stems are clearly related etymologic-ally, the question arises as to whether the Winnebago rendering, the tricky one, does not really mean simply one-who-acts-like-Wakdjunkaga, and is thus secondary. In no other Siouan language is the stem for tricky remotely like wakdjunkaga. It seems best, then, to regard the real etymology of wakdjunkaga as unknown.
The similarity of the exploits attributed to Wakdjunkaga and all other trickster-heroes in North America is quite astounding. The only possible inference to be drawn is that this myth-cycle is an old cultural possession of all the American Indians, which has remained, as far as the general plot is concerned, relatively unchanged. Just because of this fact the specific differences between the Winnebago myth-cycle and the others assume special importance and demand explanation. To do this adequately it will first be necessary to summarize the plot of the Wakdjunkaga myth-cycle in considerable detail.1
The cycle begins with an incident found in no other version, namely Wakdjunkaga pictured as the chief of the tribe, giving a warbundle feast on four different days. He, although host and consequently obligated to stay to the very end, is described as leaving the ceremony in order to cohabit with a woman, an act which is absolutely forbidden for those participating in a warbundle feast. On the fourth day he stays to the end and invites all the participants in the feast to accompany him by boat. Hardly has he left the shore when he returns and destroys his boat as useless. At this piece of stupidity some of his companions leave him. He then starts on foot, but after a short time destroys both his warbundle and his arrowbundle and finds himself eventually deserted by everyone and alone; alone, that is, as far as human beings and society are concerned. With the world of nature he is still in close contact. He calls all objects, so our text tells us, younger brothers. He understands them; they understand him.
This is clearly an introduction and its purpose is manifest. Wakdjunkaga is to be desocialized, to be represented as breaking all his ties with man and society. Why our raconteur began the cycle in this particular fashion it is impossible to say, but it is
246
PAUL RADIN
best to assume that it is a literary device. Presumably he has decided that Wakdjun-kaga is to be depicted as completely unconnected with the world of man and as gradually evolving from an amorphous, instinctual and unintegrated being into one with the lineaments of man and one foreshadowing man's psychical traits. He has, in short, like many another epic writer, begun in medias res. What he seems to be saying is: "Here is Wakdjunkaga pretending to be thoroughly socialized and about to embark on a warparty. But let me tell you what he really is: an utter fool, a breaker of the most holy taboos, a destroyer of the most sacred objects!" And then he proceeds, in kaleidoscopic fashion, to reduce Wakdjunkaga to his primitive self.
The exploits that follow tell us precisely who Wakdjunkaga is. (See incidents 4-10 in text.) In the first he treacherously lures an old buffalo to destruction, kills him in most cruel fashion and butchers him. No ethical values exist for him. And how does he kill and butcher the buffalo? With only one hand, his right. The next incident shows why only one hand has been used. He is still living in his unconscious, mentally a child, and this is here symbolized by the struggle between his right and his left hands in which his left hand is badly cut up. He himself is hardly aware of why this has occurred. He can only ejaculate, "Why have I done this?" In contradistinction to Wakdjunkaga, the world of nature is represented as conscious, and the birds, in a language he cannot understand, exclaim, "Look, look! There is Wakdjunkaga. There he goes!"
In the next incident he is still Wakdjunkaga the undifferentiated and instinctual. He comes upon a being with four little children who must be fed in a certain manner and at a certain time lest they die. In short, the principle of order must be recognized. But he knows no such principle. The father warns him that if the children die because Wakdjunkaga has failed to follow his instructions, he will kill him. Yet Wakdjunkaga, because of his own hunger, disobeys the instructions given him and the children die. Immediately the father is upon him. Wakdjunkaga is pursued around the island world, that is the universe, and only by jumping into the ocean surrounding it does he escape death.
As he swims aimlessly in the water, not knowing where the shore is, if, indeed, one exists, completely without bearings, he asks fish after fish where he can find land. None of them knows. Finally he is told that he has been swimming along the shoreline all the time.
He has barely landed, that is, he has barely got his bearings, when he attempts to catch some fish. But all he can obtain is the water through which some fish have passed. Out of this he enthusiastically prepares a soup, and fills himself to his utmost capacity. As he lies there, practically incapacitated, his stomach shining from being distended, a dead fish drifts by. He seizes it, but he cannot eat any more and he buries it.
Here we find Wakdjunkaga completely unanchored. He is not only isolated from man and society but - temporarily at least - from the world of nature and from the universe as well. Small wonder, then, that he is described as thoroughly frightened and as saying to himself, "That such a thing should happen to Wakdjunkaga, the warrior! Why I almost came to grief." What the author intended here — the enraged father, the pursuit, the headlong flight and the submerging in the ocean - may well have been meant as a description of what can happen to anyone who leads the life instinctual.
THE WINNEBAGO TRICKSTER FIGURE
247
However, there is also another point involved here. Being frightened is, in Winnebago symbolism, generally the indication of an awakening consciousness and sense of reality, indeed, the beginning of a conscience. And that seems to be borne out by the next incident (11 in text), where Wakdjunkaga is represented as imitating what he takes to be a man pointing at him but which turns out to be a tree stump with a protruding branch. The important point here is his reaction to his blunder and stupidity. "Yes, indeed," so he says, "it is on this account that the people call me Wakdjunkaga, the foolish one! They are right." He has one of the necessary traits of an individualized being now, a name. In Winnebago society a child had no legal existence, no status, until he received a name.
The episode which follows (12 in text) is known throughout North America in a practically identical form. It describes how Wakdjunkaga persuades some ducks to dance for him with eyes closed and how he wrings their necks as they dance, though most of them succeed in escaping. He roasts the few he has killed and, exhausted from his encounter, goes to sleep after instructing his anus to keep watch. His anus does its best to awaken him when foxes appear, but to no avail, and Wakdjunkaga awakens to find the ducks have been eaten. In anger he punishes his anus by burning it, and when he can endure the pain no longer, he exclaims, "Ouch! This is too much!. ..Is it not for such things that I am called Wakdjunkaga? Indeed, they have talked me into doing this, just as if I had been doing something wrong!" Important for our purpose is this exclamation, and also the one in incident 14, where Wakdjunkaga discovers that he has been devouring parts of his own intestines and commenting upon how delicious they taste: ^Correctly indeed am I named Wakdjunkaga, the foolish one! By being called thus I have actually been turned into a wakdjunkaga, a foolish one!"
With these incidents (12, 13, 14 in text) we have reached a new stage in Wakd-junkaga's development. The emphasis is now upon defining him more precisely, psychically and physically. He is now to be shown emerging out of his complete isolation and lack of all identity, and as becoming aware of himself and the world around him. He has learned that both right and left hands belong to him, that both are to be used and that his anus is part of himself and cannot be treated as something independent of him. He realizes, too, that he is being singled out, even if only to be ridiculed, and he has begun to understand why he is called Wakdjunkaga. But he does not as yet accept responsibility for his actions. In fact, he holds other people, the world outside of himself, as compelling him to behave as he does.
It is only at this point that we are told anything specific about Wakdjunkaga's appearance. Every Winnebago, of course, knew what it was. Why then are we informed about it just here? The answer seems to be that his original appearance is now to be altered. He is now to be given the intestines and anus of the size and shape which man is to have.
That this episode has not been placed here just through accident is proved by the fact that in the episode which immediately follows we have the first mention of his penis, of its size and of his manner of carrying it in a box on his back. And for the first time are we made aware of his sexuality. In all other specifically trickster myths lust is his primary characteristic; in these all his adventures reek with sex. If in the Winnebago Wakdjunkaga cycle it is not mentioned until now, this is because the author or authors who gave this cycle its present shape wished to give us not a series of Trickster's adventures as such but the evolution of a Trickster from an undefined
248
PAUL RADIN
being to one with the physiognomy of man, from a being psychically undeveloped and a prey to his instincts, to an individual who is at least conscious of what he does and who attempts to become socialized. Sex is treated primarily in its relation to Wakdjunkaga's evolution. Sexual escapades do not really seem to interest our raconteurs as such.
It is not strange then that the first sexual episode related of Wakdjunkaga should consist of his waking from his sleep to find himself without a blanket. He sees it floating above him, and only gradually recognizes that it is resting on his huge penis erectus. Here we are brought back again to the Wakdjunkaga whose right hand fights with his left, who burns his anus and eats his own intestines, who endows the parts of his body with independent existence and who does not realize their proper functions, where everything takes place of its own accord, without his volition. "That is always happening to me," he tells his penis.
It is not an accident that this episode is placed just here. It belongs here for it is to serve as an introduction to giving Wakdjunkaga an understanding of what sex is. Quite properly, we first have the symbol for masculine sexuality and an example of how it is thought of socially, namely as analogous to the banner raised by the chief when the tribal feast is given, and that then there follows an example (incident 16 in text) of how it is used concretely and properly.
Incident 16, the sending of the penis across the water so that Wakdjunkaga can have intercourse with the chief's daughter, is as well known in North America as that of the hoodwinked ducks. In most of the trickster cycles it is immaterial where it is placed. Here, clearly, this is not true. It belongs here, for it is to be used to indicate how meaningless and undifferentiated Wakdjunkaga's sex drive still is inherently; indeed, to show how meaningless it is for all those involved. Penis, cohabitation are only symbols here; no sense of concrete reality is attached to them. That Wakdjunkaga has as yet developed no sense of true sex differentiation is made still clearer by the episode where he transforms himself into a woman (incident 20 in text).
Immediately following incident 16 we find the well-known theme of how he begs the turkey-buzzard to carry him on his back and fly with him. Whatever may be its larger psychological implications, this incident seems to play no role in the drama of Wakdjunkaga's development and must be regarded as an interlude. His rescue by women, after turkey-buzzard has treacherously dropped him into a hollow tree, is part of the secondary satire on man and society that permeates the whole cycle and about which we will have more to say in the next section.
We have now reached the crucial episode where Wakdjunkaga changes his sex and marries the chief's son. The overt reason given for his doing this is that he and his companions have been overtaken by winter and are starving and that the chief and his son have plenty. This episode like the preceding ones is well known; no trickster cycle omits it. The reason generally given is that Trickster does it to avenge some insult. The change of sex is a trick played on an oversexed individual in order to show to what lengths such a person will go, what sacred things he will give up and sacrifice to satisfy his desires. Such is its role in one of the most famous of all North American Indian trickster cycles, that of Wisaka of the Fox tribe.2 But here in the Winnebago cycle it is not to avenge an insult but ostensibly to obtain food that the transformation of sex has occurred.
Taken in conjunction with the sex episodes which have preceded and the two incidents that follow, its meaning becomes clear. It is part of Wakdjunkaga's sex
THE WINNEBAGO TRICKSTER FIGURE
249
education. This must begin by sharply differentiating the two sexes. It is as if Wakdjunkaga were being told: this is the male; this, the penis; this is cohabitation; this is the female organ; this is pregnancy; this is how women bring children into the world. Yet how can Wakdjunkaga, with his generalized sexual organs, arranged in the wrong order and still living distinct from him in a receptacle on top of his body, how can he be expected to understand such matters? For that reason Wakdjunkaga's sex life, indeed, his whole physical life, is for him still something of a wild phantasmagoria. This phantasmagoria reaches its culmination point in incidents 20 and 21 of the text. Satire, Rabelaisian humour and grotesqueness are combined in these passages with amazing effect. Thrice, within very short intervals before the visit to the chief's son, the man-woman, Wakdjunkaga, is made pregnant; she, a woman, does her own courting; the man-woman becomes pregnant again. Whose are the children he brings forth? We are purposely left in ignorance in order to stress the fact that it makes no difference. Parenthood is immaterial, for they are born of a man-woman.
We have here reached a point where ordinary words and terms are indeed-completely inadequate. Only symbols, only metaphors, can convey the meaning properly. As soon as the last child is born he begins to cry and nothing can stop him. A specialist at pacifying children, an old woman who has passed her climacteric, that is, one who is beyond sex, is called, but she is helpless. Finally the infant cries out, "If 1 could but play with a piece of white cloud." To translate this into meaning then becomes the task of a special shaman. So it is with the child's other requests. They seem all unreasonable and unseasonable. What else can we expect in this phantasmagoria? Yet these requests are, at the same time, reasonable and have concrete non-symbolic significance. Not for the child, however, but for Wakdjunkaga, who is waiting for spring to come and for the time when he can obtain his food himself. Be it remembered: at no time is Wakdjunkaga represented as becoming a victim of this phantasmagoria. He always remains his old primordial self. He has as yet not learned very much and has forgotten even less.
The denouement arrives when Wakdjunkaga is chased around the fireplace by his mother-in-law, when his vulva drops from him and he is revealed as his true self. Ordinarily on such an occasion in the Wakdjunkaga cycle he is represented as laughing at the discomfiture of those on whom he has played a trick. But here he runs away. The reason is clear; the situation is fraught with too many difficulties. Too many taboos have been broken, the sensibilities of too many people have been outraged, too many individuals have been humiliated. It is serious enough for a chief's son to be indulging in what turns out to be homosexual practices, but far more serious is the situation in which the chief's wife, Wakdjunkaga's "mother-in-law", finds herself. Among the Winnebago the mother-in-law taboo was very strict, yet here she is openly associating with one who could have married her daughter and become her son-in-law and thus a person with whom she is not allowed to speak and with whom no joking is possible. The right to joke with and to tease an individual implies a very special relationship. It can only take place between a very restricted number of blood-relatives and a less restricted number of relatives-by-marriage. Joking between a mother-in-law and son-in-law is simply unthinkable. Apparently it was even unthinkable in this Walpurgisnacht atmosphere, for the narrator does not use the term daughter-in-law when he speaks of the chief's wife teasing Wakdjunkaga, but the term hiciga, brother's son's wife. The fact that Wakdjunkaga when
250
PAUL RADIN
functioning as the daughter-in-law could not possibly be hiciga a Winnebago audience, of course, would know, but, under the circumstances, any term was better than to call him daughter-in-law.
The shock of all these revelations to those participating in this comic-tragic drama is clear, and our raconteur has expressed this shock by bringing his narrative to a full stop. He apparently feels that one must get out of this insane atmosphere quickly. I think he has done this very astutely. He has Wakdjunkaga not only run away but suddenly come to some realization of what he was doing. Suddenly, and for the first time in the cycle, he is pictured as a normal man with a wife to whom he is legally married and a son for whom it is still necessary to provide. In short, he is suddenly represented as a good citizen, as a thoroughly socialized individual. And so he returns to his home, is received there with joy and stays with his family until his child is well able to take care of himself. The only indication that it is Wakdjunkaga with whom we are here dealing is found in the last three sentences of this episode. "I will now go around and visit people for I am tired of staying here. I used to wander around the world in peace but here I am just giving myself a lot of trouble." In these words we have his protest against domestication and society with all its obligations. Doubtless this also voices the protest of all Winnebago against the same things.
The biological education of Wakdjunkaga is now to be resumed. The next adventure is a utilization of a strictly Rabelaisian theme found throughout aboriginal America, the talking laxative bulb (incidents 23 and 24 in text). Although he now possesses intestines of normal human size he knows nothing about them. He comes upon a bulb which tells him that whoever chews it will defecate. Nature has never taunted him in this fashion before. So he takes the bulb and chews it to find that he does not defecate but only breaks wind. This expulsion of gas increases in intensity progressively. He sits on a log, but is propelled into the air with the log on top of him; he pulls up trees to which he clings, by their roots. In his helplessness he has the inhabitants of a village pile all their possessions upon him, their lodges, their dogs, and then they themselves climb upon him, for he tells them that a large warparty is about to attack them. And so the whole world of man is now on Wakdjunkaga's back. With a terrific expulsion of gas he scatters the people and all their possessions to the four quarters of the earth. And there, we are told, he stood laughing until his sides ached.
Apart from the grotesque humour and the obvious satire, is there anything else involved here? Yes. Broadly speaking, a Winnebago would say this is an illustration of what happens when one defies nature even in a minor fashion, that this is what happens when man climbs on Wakdjunkaga's back.
But this world to which he has fled to escape from society, the world where he could wander around in peace, has not finished its test with him. He now begins to defecate. The earth is covered with excrement. To escape it he takes refuge in a tree, but to no avail, and he falls into mountains of his own excrement. Blinded by the filth clinging to him he gropes helplessly for a path to water. The trees whom he asks for information mock and mislead him. Finally he reaches the water and can cleanse himself.
However, despite this reminder of ignorance, knowledge concerning himself and the outside world comes to him slowly. No sooner has he cleansed himself completely than he mistakes the reflection in the water of plums growing on a tree on the shore for the plums themselves.3
THE WINNEBAGO TRICKSTER FIGURE
251
There now follows a series of incidents (27-46 in text) that have little bearing on the education of Wakdjunkaga. Apart from their manifest satiric implications they are more or less the typical adventures of all North American tricksters. They exemplify all the traits customarily attributed to him, the meaningless cruelty he inflicts upon others in order to obtain food, and how, at the last moment, he is always frustrated and cheated, cheated in fact, not only by others but by himself (see incidents 30, 31); how he comes to grief by trying to imitate others (incidents 32, 33, 41-4); and how occasionally he turns the tables on his tormentors (incidents 34, 45, 46). From a literary and psychological point of view our myth-cycle breaks down after incident 26, where Wakdjunkaga is knocked unconscious by diving after the reflection of plums in the water, although some of the threads are pulled together, albeit not too well, after incident 38.
What should have followed incident 26,1 feel, is the episode where Wakdjunkaga, through the instrumentality of chipmunk, is taught where his genitals should be placed on his body and the proper order of penis and testicles. (See incidents 38 and 39 in text.)
The words of Wakdjunkaga in the dialogue between him and chipmunk are worth noting. They are meant to point out that Wakdjunkaga is at last to become aware concretely of his sex. "Is it not your penis you are carrying on your back?" chipmunk shouts at him, and Wakdjunkaga answers, "What an evil person it is who mentions that! He seems to have full knowledge of what I am carrying on my back." Again chipmunk shouts at him, "Your testicles together!" and Wakdjunkaga answers, "Why, this being must have been watching me closely." Throughout Wakdjunkaga acts bewildered and embarrassed. At first he behaves purely passively, although he follows the instruction. He becomes angry only when chipmunk finally shouts at him his last injunction - "Put the head of the penis on top, put it on top!" It is then, when his genitals are in their right place and correctly arranged, when he has really become aware of his sex and his masculinity, it is only then, that he pursues his tormentor. He attacks chipmunk with his penis, not, ostensibly, in order to cohabit with him but to punish and destroy him for making him aware of his genitals and of his sex. It is his final protest at becoming a mature male. Be it remembered that his penis is still of tremendous length. The farther he penetrates the hole in which chipmunk has sought refuge, the more of his penis the latter bites off until it finally has been reduced to human size. In such fashion does Wakdjunkaga become a male and attain sex consciousness.
A very important addendum now follows. In contrast to the manner in which he disposes of the sloughed-off portions of his intestines, namely, by eating them himself, the parts of the penis which chipmunk has bitten off are thrown into the water and transformed into food plants for man.
Wakdjunkaga's resistance to attaining sexual maturity has innumerable larger psychological and psychoanalytical implications the explanation of which, however, I must leave to others. What I would like to stress here are two questions: first, the fact that he cannot himself reduce his large and amorphous genitals to their normal human size, arrange them in their proper order or place them properly. This must be accomplished through some outside agency. Yet, on the other hand, he himself is represented as responsible for reducing the size of his intestines. Second, it might be asked, whether there is involved in the final act, where chipmunk in his hole bites off large parts of Wakdjunkaga's penis, some form of emasculation or some form of
252
PAUL RADIN
cohabitation. My own belief is that neither is involved, but that we are still dealing with Wakdjunkaga's biological evolution and that what is being implied here symbolically is his transition from a generalized natural and procreative force to a concrete heroic human being. This, I feel, is expressly stated in his exclamation, "Of what a wonderful organ have I been deprived! But why should I say this? I can make useful objects of all these pieces of my penis for human beings!" Thus from being an unconscious benefactor he has now become a conscious benefactor not only of mankind but of nature as well.
Having attained biological maturity one would have imagined that the narrative would then indicate how he attains full psychical and social-ethical maturity. But the incidents that follow show this very inadequately and inconsistently, if at all. It was perhaps actually an impossible thing to do, considering Wakdjunkaga's traditional associations. One of the reasons for this failure, at least from a literary-psychological point of view, lay probably in the fact that one basic exploit or rather, series of exploits, connected with Wakdjunkaga and without which in the minds of the Winnebago the Wakdjunkaga cycle was unthinkable, had still to be included, namely his visits to various animals, the manner in which he was entertained by them and the manner in which he attempts, quite unsuccessfully, to reciprocate their hospitality. (See incidents 41-4.) But these episodes could only with the greatest of difficulty be used to illustrate any progressive development in Wakdjunkaga's character. An attempt, however, seems clearly to have been made, at least in one direction, namely, to show him as developing some sense of social and moral responsibility.
In the incident immediately following the transformation of the gnawed-off pieces of his penis we are told of Wakdjunkaga's meeting with coyote and his attempt to compete with him as a keen scenter. Its only significance in our cycle is to serve as an introduction to the theme of his visits to the muskrat, snipe, woodpecker and polecat (incidents 41-4), and to motivate his turning the tables on coyote (incident 46). What we have in the coyote episode is a very abbreviated form of a competition between Wakdjunkaga and coyote which plays a much greater role in trickster myths in other parts of North America.
The most that our raconteur can do with the episode of the visits to the various animals is to credit Wakdjunkaga with wishing to provide his family with food, to present him as a harmless, vainglorious blunderer and fool, and as one who succeeds in finally obtaining revenge on those who have humiliated him or desire to do so, like mink and coyote. (See incidents 40 and 46.) This is all part of his socialization. Thus, for example, after polecat visits him and kills innumerable deer for his family we have the following idyllic scene. It is really best to quote it:
"Well, wife, it is about time for us to go back to the village. Perhaps our relatives are lonesome for us especially for the children." "I was thinking of that myself," replied his
wife____Then they packed their possessions and began to carry them away____After a
while they got near their home and all the people in the village came out to greet him and help him with the packs. The people of the village were delighted. "Kunu, firstborn, is back," they shouted. The chief lived in the middle of the village and alongside of him they built a long lodge for Wakdjunkaga. There the young men would gather at night and he would entertain them for he was a good-natured fellow. The prodigal son has made good and returned!
THE WINNEBAGO TRICKSTER FIGURE
253
This reads almost like an account of the return of a successful warleader, or at least a great hunter. Yet something of his old unregenerate self still adheres to him, as is seen in the delight he takes in humiliating mink and coyote. However, a Winnebago audience would have sympathized with this humiliation of mink and coyote. They would have agreed that Wakdjunkaga was a very good-natured person, a blundering fool, it is true, but more sinned against than sinning, one who really meant well but whose good intentions always went amiss. It is in this light that we must interpret the two episodes (incidents 47 and 48) which follow, Wakdjunkaga's removal of natural obstacles in the Mississippi River that would interfere with the free movement of human beings.
But before proceeding to the discussion of these, a few words about the implications of one of the points in the fourth of his visits, that to the polecat, seem in point. In that delightful and Rabelaisian episode polecat kills deer by shooting them with wind he expels from his anus. He "loads" Wakdjunkaga with four such shots to take home with him. Wakdjunkaga is now faced with a new situation. In the case of his visits to muskrat, snipe and woodpecker (41-3), all he had to do to get himself into difficulties and inflict pain upon himself was to imitate them. But now, provided with the means for really accomplishing what his host, polecat, had done, how was he to fail, for fail he must? The problem is simply solved: he must waste these provisions. So, without any reason, he persuades himself that polecat has deceived him, and he shoots at four objects in succession blowing them to pieces - at a knoll, at a tree, at an enormous rock and at a rocky precipitous hill, the last the symbol of a sacred precinct. It is his last act of defiance against the world of nature with which he had, until recently, been on such intimate terms. It is Caliban protesting against the civilization which had been forced upon him.
It would be quite erroneous to think that the author-raconteurs of our cycle were trying in incidents 47 and 48 to exhibit to us a Wakdjunkaga who had now become a wholly beneficent being, a semi-deity in fact. What we have here is a purely secondary addition with no actual connection with what has preceded. It represents largely the influence of the most sacred of all Winnebago narratives, the Origin Myth of the Medicine Rite. There, after Earthmaker has created the universe and all its inhabitants, animal and human, he discovers that evil beings4 are about to exterminate man. In order to help them he sends Wakdjunkaga, the first being comparable to man he has created, down to earth. This is what is meant when we are told that Wakdjunkaga suddenly remembered the purpose for which he had been sent to the earth. In the Origin Myth of the Medicine Rite Wakdjunkaga is described as failing completely. Not even Earthmaker apparently could properly "rehabilitate" him. But on earth Wakdjunkaga could accomplish nothing. As the myth phrases it, "Every variety of small evil animals began to play pranks on him and plague him and he finally sat himself down and admitted to himself that he was incapable of doing anything."5 Yet in spite of all his trickster antecedents he has here, for a moment, been elevated to the rank of a true culture-hero, although the specific role he is being given belongs properly to an entirely different hero, or rather heroes, the Twins.
As I have indicated above, I think that a large part of this transformation of the character of Wakdjunkaga is due to the role he plays or was intended to play in the founding of the Medicine Rite. However, to judge from the fact that there seems to have been a difference of opinion among the Winnebago two generations ago, and one which was definitely not of recent origin, as to how he was to be evaluated, it
254
PAUL RADIN
may very well be that people always interpreted him and his activities in two ways. But to this we will return in the following section when we deal with the Winnebago attitude toward Wakdjunkaga in the first decade of this century.
In the last scene (incident 49) we get still another picture of him. We see him as a deity, an aspect of his nature completely neglected in our cycle, and as the elemental trickster, an ageing trickster, indeed almost a demiurg, taking his last meal on earth. He is pictured sitting on top of a rock with his stone kettle, eating. He perpetuates this last meal for all time, leaving in the rock the imprint of his kettle, of his buttocks and his testicles. He then departs and, since he is the symbol for the procreating power as such and the symbol for man in his relation to the whole universe, he first dives into the ocean and ascends to that island-world over which he presides, that lying immediately under the world of Earthmaker____
The above summary should give the reader some idea of the composite nature of the Winnebago trickster cycle and the degree to which the various episodes composing it have been welded together into a new whole. To obtain a better conception of the success the Winnebago achieved in this regard one must read the trickster cycles of other American Indian tribes. Then it will become clear to what an extent in the Winnebago Wakdjunkaga episodes, incidents, themes and motifs have been integrated, and then the consummate literary ability with which this has been done will stand out sharply. That this literary remodelling and reinterpretation is secondary there can be no question. It is apparently due to special circumstances in Winnebago history and to the existence of a special literary tradition there. To form some idea of what the Wakdjunkaga cycle was originally we must, however, divest our version of all those features which have made it an aboriginal literary masterpiece. This we shall attempt to do in the concluding section of this introduction when we deal with the North American Indian trickster-cycle in general.
The Attitude of the Winnebago toward Wakdjunkaga
Much of the analysis given in [the previous] section is the analysis of an outsider, of a white man, and it goes without saying that such an analysis has its dangers and pitfalls, no matter how well such an outsider thinks he knows an aboriginal culture. It is always best to let members of the culture themselves speak, and I shall, therefore, attempt to present now in a few words what were the ideas and evaluations of contemporary Winnebago - I am speaking of 1908-18 - in regard to Wakdjunkaga and how he was pictured in Winnebago literature. In those years when the new Peyote religion was spreading throughout the tribe and many Winnebago began to make evaluations and re-evaluations of their culture, Wakdjunkaga found both defenders and antagonists. Let me commence with the statement of an old conservative which he prefaced to a myth not included in the cycle being given here:
"The person we call Wakdjunkaga," so he said, "was created by Earthmaker, and he was a genial and good-natured person. Earthmaker created him in this manner. He was likewise a chief. He went on innumerable adventures. It is true that he committed many sins. Some people have, for that reason, insisted that he really was the devil.6 Yet, actually, when you come to think of it, he never committed any sin at all. Through
THE WINNEBAGO TRICKSTER FIGURE
255
him it was fulfilled that the earth was to retain for ever its present shape, to him is due the fact that nothing today interferes with its proper functioning. True it is that because of him men die, that because of him men steal, that because of him men abuse women, that they lie and are lazy and unreliable. Yes, he is responsible for all this. Yet one thing he never did: he never went on the warpath, he never waged war.
"Wakdjunkaga roamed about this world and loved all things. He called them all brothers and yet they all abused him. Never could he get the better of anyone. Everyone played tricks on him."
What this particular Winnebago is undoubtedly trying to say is that Wakdjunkaga represented the reality of things, that he was a positive force, a builder, not a destroyer. The reference to his not having gone on the warpath is very illuminating. It indicates that, for this particular individual, Wakdjunkaga's failure to help mankind by destroying those who were plaguing it was not a reprehensible thing because it would have meant violence, meant waging war. If Wakdjunkaga was thus useless after he had prepared the earth for man that is quite intelligible. That men do not understand him, that they misinterpret and laugh at his activities, this too is intelligible. He does not belong in the world of men but to a much older world.
In contrast to this sympathetic attitude we have that of the members of the Peyote rite. They used Wakdjunkaga and his cycle to point a moral. It would be quite erroneous to imagine that this was an entirely new attitude; it existed long before the Peyote rite came into existence. The attitude of the Peyote people is best illustrated in the following homily:
The older people often spoke to us of Wakdjunkaga. However, we never knew what they meant.7 They told us how, on one occasion, he wrapped a racoon-skin blanket around himself and went to a place where there were many people dancing. There he danced until evening and then he stopped and turned around. There was no one to be seen anywhere, and then he realized that he had mistaken for people dancing the noise made by the wind blowing through the reeds.
So do we Winnebago act. We dance and make a lot of noise but in the end we accomplish nothing.
Once as Wakdjunkaga was going toward a creek he saw a man standing on the other side, dressed in a black suit and pointing his finger at him. He spoke to the man but the latter would not answer. Then he spoke again and again but without receiving any reply. Finally he got angry and said: "See here! I can do that too." So he put on a black coat and pointed his finger across the creek. Thus both of them stood all day. Toward evening, when he looked around again, he noticed that the man across the creek who had been pointing his finger at him was really a tree stump.
"O my! What have I been doing all this time? Why did I not look before I began? No wonder the people call me the Foolish-One!"
Wakdjunkaga was walking around with a pack on his back. As he walked along someone called to him. "Say, we want to sing." "All right," said he. "I am carrying songs in my pack and if you wish to dance, build a large lodge for me with a small hole at the end for an entrance." When it was finished they all went in and Wakdjunkaga followed them. Those who had spoken to him were birds. He told them that, while they were dancing, they were not to open their eyes for if they did their eyes would become red. Whenever a fat bird passed Wakdjunkaga would choke it to death, and if the bird squeaked he would say, "That's it! That's it! Give a whoop!"
256
PAUL RADIN
After a while one of the birds got somewhat suspicious and opened its eyes just the least little bit. He saw that Wakdjunkaga was choking all the birds he caught to death, and he cried out, "Let all those who can run save themselves for he is killing us!" Then this bird flew out through the top of the house. Wakdjunkaga took the birds he had killed and roasted them. But he did not get a chance to eat them for they were taken away from him.
So are we Winnebago. We like all that is forbidden. We say that we like the Medicine Rite; we say that it is good and yet we keep it secret and forbid people to witness it. We tell members of the society not to speak about it until the world comes to an end. They are, in consequence, afraid to speak of it. We, the Winnebago, are the birds and Wakdjunkaga is Satan.
Once as Wakdjunkaga was going along the road someone spoke to him. He listened and he heard this person saying, "If anyone eats me, faeces will come out of him." Then Wakdjunkaga went up to the object that was talking and said, "What is your name?" "My name is Blows-himself-away." Wakdjunkaga would not believe it and so he ate this object. (It was a shrub.) After a while he blew himself away. He laughed. "O, pshaw! I suppose this is what it meant." As he went along it grew worse and worse, and it was really only after the greatest hardship that he succeeded in returning home.
So are we Winnebago. We travel on this earth all our lives and then, when one of us tastes something that makes him unconscious,8 we look upon this very thing with suspicion upon regaining consciousness...
Here we have Wakdjunkaga as both the glorified image of man and as the tempter. The Winnebago term used for Satan here is Hereshguina. The latter is the great evil spirit who is believed to have existed from the beginning of time, who is as old as Earthmaker and always negating what Earthmaker creates. According to one etymology his name means "he-of-whose-existence-one-is-doubtful". That Wakdjunkaga should be equated with him by the members of the semi-Christian Peyote rite is not strange.
This insistence on Wakdjunkaga's purely negative side is a very old attitude. We find it among the Dakota-Sioux and the Ponca. But equally old is the interpretation of his character and of his positive activities to which I have referred before and which finds its best expression in a very old myth, The Two Boys.9 In this myth he is represented as actively helping Hare in his endeavours to secure the powers that are eventually to help one of the great spirits to victory over his enemy. In this myth Wakdjunkaga is represented as addressing Earthmaker as follows:
Father, it is well. That which we desired, this you have given us precisely as we wished it and without any hesitation. It is my friend Hare who is to see that our purpose is attained. He is the only one who can accomplish it. All the spirits in the lodge from which we have come listen and obey what he says, for his are good thoughts. It is he who helped the human beings before, and he will do this for them too.
To this speech of Wakdjunkaga Earthmaker replies in the following fashion:
Firstborn, you are the oldest of all those I have created. I created you good natured: I made you a sacred person. I sent you to the earth to remain there so that human beings would listen to you, honour you and obey you and that you might teach them by what means they could secure a happy life. This was the purpose for which you were
THE WINNEBAGO TRICKSTER FIGURE
257
created. What happened to you after you reached the earth that you brought upon yourself alone. It is because of your own actions and activities that you became the butt of everyone's jest, that everyone took advantage of you, even the smallest of insects. How is it then that now you are presenting as a model to be followed that very individual, Hare, who did do what I told him to? You, although you were given the greatest of powers, made light of my creation. It was not anything I told you to do. It is therefore your own fault if people call you the Foolish-One. I created you to do what your friend Hare actually did. I did not create you to injure my creation.
This apparent bewilderment of the Winnebago supreme deity concerning the reasons for Wakdjunkaga's actions and this disavowal of responsibility for them, it will be important to keep in mind. [...]
NOTES
1 The story is presented as "The Winnebago Trickster Cycle" in the volume from which this extract comes, pp. 3-60.
2 W. Jones, Fox Texts, Publications of the American Ethnological Society, Leyden, 1907, pp. 315 ff.
3 This incident is probably of European origin.
4 These are not represented as having been created by him.
5 Another version of the same myth states, "He was like a small child crawling about____All
one saw of him was his anus. He accomplished no good and in fact injured Earthmaker's creation."
6 He is referring to the followers of the Peyote rite.
7 That is, they did not understand the significance of Wakdjunkaga's actions. The following episode is not found in our version of the myth.
8 He is referring to the eating of the peyote.
9 Cf. Special Publications of Bollingen Foundation, No. 3, Basel, 1954.
21
Closure and Multiplication: An Essay on Polynesian Cosmology and Ritual
Alfred Gell
From Alfred Gell, "Closure and Multiplication: An Essay on Polynesian Cosmology and Ritual," in Daniel de Coppet and André Iteanu, eds., Cosmos and Society in Oceania (Oxford: Berg Publishers, 1995), pp. 21-56. Abridged.
Alfred Gell (1945-97) was an outstanding iconoclastic thinker who taught at the London School of Economics and wrote on time and art, among many other subjects. Gell's essay is a powerful account of Polynesian cosmology that begins with the startling assertion that the Polynesian world is one of immanence rather than transcendence and that such immanence brings problems in its wake. Rather than the carefree life that Polynesians are often portrayed as having, Gell paints a picture of anxiety in which you had to "watch your back," not so much from personal enemies as from the pervasive effects of the sacred, Gell's aesthetic sensibility is evident in the acute analysis of Polynesian carving and tattooing, supplemented by his own drawings and those of earlier western visitors to Polynesia (only some of which are reproduced here). See also his longer study Wrapping in Images (1993). The anthropology of art is a rich field overlapping that of religion.
Gell's analysis is based entirely on secondary sources; his own fieldwork was carried out in Melanesia (1975) and tribal jndia. He is the author of a particularly fine essay on taboo (1979). Other essays on Oceanian cosmologies are to be found in de Coppet and Iteanu (1995); see also Strathern (1988) and Bamford (1998), as well as many excellent ethnographies.
The main features of Polynesian cosmological beliefs show a high degree of consistency and have become reasonably well known, through the efforts of a host of nineteenth-century writers, mostly missionaries or government officials, and subse-
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
291
quent re-working by Polynesia specialists. The Central-Eastern Polynesian culture area, comprising the Society Islands (Tahiti) the Marquesas, New Zealand, and Hawaii to name only the larger sub-regional systems, all maintained broadly similar systems of ritual and belief, which were founded on the cosmological scheme in which the creation of the cosmos came about through conflict and separation (Hanson 1982). The cosmos was originally one but this afforded no scope of action to the creator god {Ta'aroa, Tane, etc.) whose initial act was to bring about the separation of sky and earth, night and day, thus creating the "bi-cameral world" of the Polynesians which was divided into the po, the other-world, the world of night {po), darkness, the original gods, the dead, etc. and the ao, the world of light, day (ao) life, human activity, and so forth. This original separation was subsequently elaborated differently in different places, and might become formidably complicated, as is suggested by Figure 1, which shows a diagram of the cosmos created by a Tuamotuan informant for an inquisitive missionary in the late nineteenth century. The cosmos as a whole takes the form of a nested arrangement of "shells" {'apu), each associated with a particular cosmogonic episode of separation; and pathways or ladders are indicated by means of which intercourse is possible between different cosmological levels.
Polynesian thought about the universe differed from Judaeo-Christian "creationist" thought in that it was predicated, not on the creation of the universe ex nibilo, by God, but on the initial existence of everything in an all-embracing plenum or
x.f
i * -'J -:Kj : . :i3 '
«Kl V
■S2E
r-y.
ö '<---*
: C i
•; o, : qI
: > /$ : C -
■   ■ C
Figure 1   The Tuamotuan conception of the Cosmos by Paiore, c. 1820
292
ALFRED GELL
tightly-bound continuum. The creative epoch occurred as a process of "differentiation" within this pre-existing plenum, undertaken by a God who made "cuts". The Tahitian creation chant, part of which I am about to cite, is really wrongly so called, because nothing whatsoever is "created" in it. Instead, what the God does is to articulate, or differentiate, the world into its distinct components and qualities, but the substance of the newly-articulated cosmos remains what it always was, nothing other than the God himself. The name borne by Ta'aroa, the Tahitian creator god, actually means "the sever-er". Ta'aroa is called thus because he made the initial severance between the po and the ao, and he created the various natural phenomena by rearranging his own body:
So he overturned his shell and raised it up to form the sky.. .and he slipped out of another shell which covered him which he took for rock and sand. But his anger was not yet appeased, so he took his spine for a mountain range, his ribs for mountain slopes, his vitals for the broad fleeting clouds, his flesh for the fatness of the earth, his arms and legs for the strength of the earth, his finger-nails and toe-nails for scales and shells for the fishes, his feathers for trees, shrubs and creepers, to clothe the earth, his intestines for lobsters, shrimps and eels for the rivers and seas, and the blood of Ta'aroa got heated and drifted away for redness for the sky and the rainbows ... Ta'aroa had crusts, that is, shells, so everything has a shell.
The sky is a shell, that is, endless space in which the God placed the sun, the moon, the sporades [the scattered or individual stars], and the constellations of the gods. The earth is a shell to the stones, the water, the plants which spring from it. Man's shell is woman, because it is by her that he comes into the world, and woman's shell is woman, because she is born of woman. One cannot enumerate the shells of all the things the world produces. (Henry 1928: 339-40)
This projection of the body of the God into the phenomenal world, though developed in an exceedingly poetic way in the texts assembled by Teuira Henry, was not in itself an unorthodox conception of cosmogenesis. But it had profound consequences for the subsequent development and elaboration of indigenous religious thought. Because the po and the ao were originally emergent parts of a continuum, their separation had, for the Polynesians, only a provisional character, and they continually threatened to merge together again. Indeed the very instruments that kept them apart, kept them together and in communication. This was brought out most perspicuously in relation to the four props (to'o) which Ta'aroa positioned in order to keep the sky up and open a space (atea) for the ao, the world of light and of humanity. The props that separated the po and the ao also, by definition, joined them together, and thus became the exemplary prototypes of sacred objects, endowed with the power to communicate the divine presence. Society Islands politics revolved around rival claims to control certain ultra-sacred wooden billets, called to'o, which were counterparts of the "props" of the cosmos as a whole. But, being so sacred, these wooden billets had themselves to be separated from the world, whose continued existence/differentiation they guaranteed; so they were confined to an elaborate ark, and were massively bound around with sennit cordage, and other wrappings, and were only uncovered on the most sacred and propitious occasions, requiring many human sacrifices to allay the danger they presented to mortals, however sanctified by birth and ritual preparations (Henry 1928; Oliver 1974).
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
293
The cosmogonic scheme of differentiation rather than creation ex nihilo and of immanent father" fh¥frtranscehcTent divinity was,TfhTnk^the source of certain onto-logical anxieties that played an enormous part in Polynesian life. For us, the immanent deus sive natura of Spinoza represents an optimistic rather than a pessimistic deism, a blessed relief from the angry and punishing Almighty God of traditional Christianity, set apart from His creation and judging it harshly. But that is because the idea of an immanent God was never really naturalized in Christian Europe, however much eighteenth-century intellectuals may have hankered after one. In Polynesia the situation was the precise opposite; the immanence of theGods was the source of continuous anxiety (the proverbial hedomsmljTthTSouth Sea Islanders was founded on a sense_of acute and aj^ingjiysteria) and the rapidity and enthusiasm with which the ) ', Polynesians accomplished their conversion to Christianity stemmed from their untold j j relief upon discovering that God was, after all, transcendent, not part of this world, j j /
In this essay, the theme I want to address is Polynesian anxiety about immanence, '/ about differentiation, and some of the ways in which this fundamental anxiety shaped the patterns of their cultural life. That is to say, I want to explore some of the ways in which the underlying conception of the creation of the world through a process of differentiation (splitting, separation) gave rise to countervailing pressure towards fusion, absorption, collapse, etc., which had to be continually held in check through ritual action. In particular I want to identify the symbolic strategies through which social individuals sought to keep immanence at bay, while, at the same time, acknowledging the fundamental fact 'tTiat~immanence pervaded the world and constituted it. The essential strategy of counter-immanence (averting the danger of being absorbed, de-differentiated, etc.) was the~recapitulation, in one form or another, of the process of separation on which both-the cosmos in general, and the being of the individual, were founded. In other words, the cosmogonic activities of Ta'aroa the severerjiad to be recapitulated at the level of the individual subject; not out of a vainglorious desire to emulate the God, but certainly for the same reasons that the creative gods engaged in the original splitting, i.e. in order to open a space, to establish a difference. And just as the original cosmogonic act was the opening up of a domain, within the plenum, which was relatively non-sacred (the ao, the world of light) so the means of ritually securing the subject was the opening of a space between the subject and the sacred domain whence he originated.
Most important Polynesian ritual operated in precisely the inverse sense to Christian communion, i.e. the intention was to cause the divinity to leave (some part of) the world, rather than to induce the divinity to enter (some part of) it. There were no rituals through which persons (e.g. kings) were consecrated, because they were already all-too-sacred, but action in the world was impossible for persons of high intrinsic sanctity, because they did not truly belong to it, belonging, instead, to the po. Thus, for instance, the non-executive sacred ruler of Tonga, the Tu'i Tonga, was not mourned when he died, and his funeral mound was covered with excrement rather than offerings, because he had never lived (Bott 1982). Instead of being sanctified, Polynesians of high intrinsic (i.e. genealogically-based) rank had to be elaborately de-sanctified- Thus, the main passage-rites of the Society Islands were the amo'a or "head-releasing" rituals through which children, who were considered to be "little gods" (and highly tapu) were made "secular" by being offered blood drawn from the congregation, which replaced their sacred substance and freed them for ordinary worldly social interactions.
294
ALFRED GELL
We can begin to consider the consequences of the cosmology of immanence for personal ontology by dwelling for a moment on the general Polynesian word for "god", atua. This word is based on the morpheme tua, which means "back", or the far, invisible, side of any object. The particularly sacred parts of a human being were the back (the spine) and the head - i.e. precisely those parts of the body that cannot normally be seen by their possessor. The atua (spiritual element) of the person was the tua (back) of the person. Tua also refers to elders and ancestors, the "back" portions of time, and these were also sacred. Immanent divinity was not a contingent presence, but a categorical, dimensional, feature of objects, spaces, times. Ordinary human intercourse was front to front, expressed in western Polynesia (and Fiji) in the fundamental social morality of "facing one another" (fa'agaga, cf. Shore 1982), while it was always strictly forbidden to move about behind some important person, invisibly to them. The back was the "individual" as opposed to the cosmological po, the inaccessible, threatening, but ever-present unseen. But it was possible also to intervene so as to protect this aspect of the person.
I think one can identify two basic strategies through which the integrity of the ('person could be maintainecLagainst cosmological collapse, to which I have attached the labels of "closure">and "multiplication". "Closure" is the provision of extra reinforcement, hardening the target of spiritual danger, while "multiplication" is the strategy of reduplicating the person in myriad forms. We can see both strategies at work in the cosmological myths referred to above: thus Ta'aroa multiplies himself by opening up the ao, assuming his myriad immanent forms; but also he closes it off with the "shells" (secured by the "props") so that it cannot fall in again. In fact, the entire cosmos consisted of such "shells", as Figure 1.1 shows. But, as the chant also specifies, people were also "shells", and so were their kin-groups. (In the eighteenth century, 'apu, "shell" was the standard Tahitian expression for the bilateral kindred, the basic unit of social organization, cf. Maori: hapu.) But rather than continue with these general observations, let me turn to consider some more specific imagery of closure and multiplication in the context of Polynesian art traditions.
Polynesian carving is particularly rich in Janiform images (Figure 2), which, though they occur elsewhere (particularly in India), nowhere assume the importance they have in this part of the globe. Figure 2, which we may take as a typical instance, is one of a small number of images of Siamese twin goddesses carved from whales' teeth, all of which were actually collected in Fiji, where they were regarded as the consorts of Ndeng'ei, the main god; but they are known to have been carved in Ha'apai (part of the Tongan group of islands), and they probably originally represented the Tongan Siamese twin culture heroines Nafanua and Tokupulu. The Tongan Siamese twin goddesses (Colocott 1921; Reiter 1907), who play an important role in the Tongan creation myth (and were responsible for assuring plentiful catches for Tongan fishermen), have Samoan counterparts in the form of the twins Taema and Tilafaega, also known as Nafanua the goddesses of tattooing and warfare respectively (Kraemer 1902). I do not want to expand on the mythological detail, but I would like to offer some suggestions as to why Janiform images of this type had so much ritual significance in the area.
The Siamese twin archetype combines closure and multiplication in their most elementary forms. The most outstanding feature of a Siamese twin is symmetry about the vertical axis, not only in one plane (left/right) but also about the other vertical plane (front/back). In other words, a Siamese twin has no "back", only a
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
295
IP
Figure 2  Tongan twin deities carved in whale ivory. Collected in Fiji by Gordon
(Haddon Museum)
middle and a periphery. Such a being is protected, or encompassing, in a way in which no ordinary mortal can be. For merely one-dimensionally symmetric beings
the back always remains the most vulnerable area____      Siamese twins are immune
from such dangers, and have therefore a kind of completeness which justifies their being considered among the immortals. Indeed, they seem to adumbrate creation (as splitting) itself, since their image seems to capture the moment at which the god created the world by splitting apart the plenum, which was himself. Siamese twins obviously qualify as "multiplied" beings, but at the same time they are closed in that their twofold symmetry closes off that part of the ordinary being which is open to assault from the back.
It would therefore be natural to expect that one of the strategies through which ordinary mortals might seek to recapitulate, in themselves original cosmogonic • splitting, and thereby assure the integrity of their persons, might be to imitate the twins, and attach a secondary person to their backs. We are able to know that this
296
ALFRED GELL
possibility occurred to the minds of the inhabitants of Western Polynesia, from where the statuettes of the Siamese twin goddesses come, not because they practised artificial twinning, but because they were specifically disallowed from doing so. George Turner, a nineteenth-century missionary ethnographer, recorded that in Samoa it was forbidden for members of the family to sit learning against one another back-to-back, on pain of immediate fatal, divine retribution (Turner 1884). I believe that in contemporary Christian Samoa it would still be thought very improper to adopt this posture. Turner also noted that all fruit or vegetables that grew together to form a fused pair (as happens sometimes with bananas, taro, yams, etc.) were immediately to be offered to the gods and might on no account be eaten. He makes both these observations in the same paragraph, and leaves one in no doubt that they were connected; but he offers no more specific explanation. In the light of what has been said so far, I think one can reasonably infer that the principle of the "sacredness of doubles" and the prohibition on back-to-back sitting relate to the image of divinities as protected, encompassing back-less beings who escape from the fundamental ontological deficit of ordinary mortals, i.e. an asymmetrical relationship with tua, the back. To become such a double being was to usurp a privilege afforded to the gods (or goddesses) alone. It was sacrilege, in other words.
But although overt doubling-up was not a feasible symbolic strategy, since it infringed too much on divine privilege (like eating "double" fruit or vegetables) there were other, more indirect means of achieving a similar result. At this point, multiplication and closure begin to diverge. Let me take multiplication first. Returning to the Siamese twifTImage, one^possible reading of it is that it represents a mutual and symmetrical parturition (note that the Western Polynesian mythic twins are all female). There is a direct connection between the idea of doubling and birth. Momentarily, at least, any woman giving birth is a Siamese twin, and indeed in the Marquesas (if not anywhere else, so far as I know) the word for "birth" (fanau) also referred to fu^e^iiojib]e„p„bjects, such as double bananas, etc. (Dordillon 1934-5). It goes without saying that there is a strong relationship between cosmogony and birth, where cosmogony is specifically not conceptualized as creation ex nihilo: the God of the Old Testament does not noticeably "give birth" to the cosmos whereas the corresponding Polynesian texts, as has frequently been noticed, are permeated with procreative symbolism. Birth among mortals was also a process of cosmological significance recapitulating creation, in that children (in some places only the children of elite kindreds) were believed to come from the po and to be, in fact gods; exceptionally sacred, ranking higher than their parents (Oliver 1974). Because women were \ capable of recapitulating divine cosmogonic acts and were the means through which I divinities entered the ao, it follows that women were not just more sacred, but also i moreprotected against "jffirfJ^l^iYdHCh--l ag-Hp-Hiffprpnfj^ggin) TKaif
imen. Women were "naturally" noa (me'ie, etc.: various synonyms), i.e. "clear", noh-\sacrecl in the sense of being not endangered by other persons, though very capable of harming others, especially males, by infringing their tapu (Hanson 1982).
Tapu (Tahitian ra'a) was the general quality oFpefsonal, sacred distinctiveness, which was continually subject to dispersion. It is usually glossed as "sacredness", and so much has been written on the subject, that I hesitate to embark on a discussion of it here. But, as briefly as possible, here are the salient points. Objects and persons were tapu insofar as they were held to be invested with po-derived, ato-derived attributes and qualities. Because these sacred properties were imma-
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
297
nent, they were liable to dispersion: any object, or person invested with tapu quality remained so only insofar as that object, or person did not lapse back or become merged, with the generality of things. Immanent sacredness can be localized and preserved only via isolation from non-sacred things: it is quite different from our kind of sacredness, which is transcendental in which the sacred object is secular until it has been consecrated by ritual action. With immanent sacredness it is the object, or the person which is sacred simply by virtue of being that object or person because, in some way, it is itself an atua. Broadly speaking, things and persons contained tapu in proportion to the extent to which their existence recapitulated cosmogonic acts. Persons (especially of the elite) were sacred because their coming-into-being recapitulated the birth of gods. Creative activities, such as carving, canoe-building or tattooing were sacred because the activity of craftsmen is godlike, they cut things, make lines, hollow things out, and so on. Another way of putting this is to say that tapu is difference" in the sense that any thing or person that is non-substitutable, whose being there and then makes a difference, is characteristically tapu, and must be protected from indiscriminate contacts that diminish this difference. The effect of the philosophical attitude of immanence, just as is the case in India, is to place the ' whole weight of the religious system on the preservation of distinctions and boundaries (Dumont 1970). Where there is a god in everything, the need to keep things apart is overwhelming, because only this separation preserves essences, and essence precedes existence. Where boundaries are transgressed, annihilation follows.
To return to the question of the differential "sacredness" of males and females. Male tapu qualities were intrinsically vulnerable, though they also had their means of defence, wrucHT will consider shortly. But women, though no less imbued with tapu than males (sometimes more) were not vulnerable to quite the same extent, j because a woman is, so to speak, a complete system of differences in herself, a | microcosm, whereas males were irreducibly incomplete beings, dependent on , women for their presence in the world (and their departure from it). Women were containers and conductors of tapu quality, whereas men simply possessed tapu like a static electrical charge, and could be much more easily deprived of it. As the Tahitian i chant says, "woman is the shell of woman, because she is born of woman". On the surface this indicates a transitive relation
woman (1) —» woman (2) —> (3) -> (n)
in which one woman gives birth to another. But the phraseology suggests at the same time a recursive, self-referential relationship
woman (1) —> woman (1) —> (1) —>...
i.e. a woman giving birth recapitulates her own birth, ad infinitum. Whereas the schemes, also given in the chant,
woman —> man
admits of no such interpretation. Hence, I would argue, the procreative powers of women make them temporally "symmetrical" beings (beings with no past and no future) in a way precisely analogous to the spatial symmetry of Siamese twins along
298
ALFRED GELL
the spatial axis front/back. They are thus, like the Siamese twins, both multiple and closed, their tapu not lessened by the fact that it is in no danger from without.
For males this solution does not exist: they faced an acute problem in becoming sufficiently noa/me'ie in order to carry on a normal existence, while not dispersing their tapu quality in such a fashion as to lose their distinctiveness. They had essentially two possibilities: they could find other means of imitating the natural characteristics of Siamese twins, women, etc., which enabled them to contain their sacredness so that it was not in danger from without, or they could disperse their sacredness under "controlled" conditions. I will consider instances of both of these.
One means that men could and did employ in order to convert themselves from mortals, with the fundamental ontological deficit that mortality implies, into "complete" beings, recapitulating cosmogony, was', tattooing, which was widely practised in Polynesia, especially by men (almost exclusively_,by men in Western Polynesia). Tattooing was an obligatory passage rite of late adolescence for males, particularly in Samoa, which produced tattoo artists for the Tongan elite as well. As was noted above, tattooing was thought to have originated from the Siamese twin goddesses, Tilafaenga and Teama (who brought it from Fiji). This mythical association of tattooing with Siamese twins underlines the latent equivalence between the strategy of multiplication-twinning and the strategy of tattooing which was predicated mainly on "closure". One is able to know this by making a study of tattooing ritual and the nomenclature of tattoo designs. Figures 3, 4, and 5 show tattoo designs, two from Western Polynesia (Samoa), the other from Eastern Polynesia (the Marquesas). Of the three, the Samoan examples are the simpler, so I will discuss it first.
Samoan tattooing was a necessary preparation.for war and for sexual life; untat-tooed males might not engage in either. Both of these were dangerous activities: it has already been made clear that women, by virtue of their position as mediators between the po and the ao, were a potent source of spiritual danger. Warfare was as well, indirectly, in that death in battle was never merely--a matter of military incompetence, but would always be traced to spiritual debility. Tattooing strengthened young males and prepared them for warlike and amatory exploits. But how was tattooing conceptualized? To begin with, let us note that the area that was first tattooed was the lower part of the back, i.e. tua, and that over this vulnerable part were placed designs of a specifically protective nature (Buck 1930; Kraemer 1902; Stair 1897). The back-design as a whole was called pe'a, "flying fox", alluding, among other things, to the motto of the king-making descent-group of Sava'i, the Tonumaipe'a, descendants of Nafanua, which was "salvation comes from the flying fox". More generally, the flying fox was a well-protected animal; anyone who has seen a flying fox securely wrapped in its enfolding wings will understand why. The design, which vaguely recalls the wing-shape of a flying fox, wrapped round the body. The back-design also has another name, tapulu, which , means approximately "tattooed wrapping" (ta+pulu). Other motifs are "bindings" (fusi) "beams" (aso), etc., suggesting that the metaphor of tattooing is the construction of a housing or defensive screen covering the body. Finally, the last and ritually perhaps the most significant tattoo was applied to the navel (pute). Horatio Hale (1846) enquired about this, and was told the navel had to be tattooed out of "shame", because that was the part of the male body that was connected to the mother. One can certainly discount the idea that it was modesty as we would understand it that required the tattooing of the navel; everything suggests that it
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
299
Figure 3  Samoan male tattooing (rear), by Kramer®
was the need to seal off the body, definitively, from the po and from the danger of dispersion via female "conductors" of tapu quality. The mark over the pute was called a fusi, a binding, making clear that this was indeed the idea.
The metaphoric basis of Samoan male tattoo was therefore biased towards closure rather than multiplication. But the implicit identity between the two (Milner 1969) is nonetheless suggested in the famous proverb
Tupu le tane, ta le tatau Tupu fafine, fanafanau.
"The man grows up, he is tattooed, The woman grows up, she gives birth."
But the imagery of multiplication could also be incorporated into the general anti-dispersive prophylaxis of tattooing. In order to see this one has to turn from the
300
ALFRED GELL
Figure 4  Samoan female tattooing
elegant restraint of Samoan tattoo, to the more florid style of the Marquesas. In the Marquesas, tattooing was called "pabu tiki, literally, "wrapping in images" (tiki). It was obligatory for the same reasons as in Samoa, i.e. warfare and sexuality, and was applied in collective ceremonies which bear a strong resemblance to their Western Polynesian prototypes (W. Handy 1922).
The Marquesan tattooing style follows design principles which can be traced back 2,000 years to Lapitan pottery decoration, in that it consists of delimited "zones with decorative 'infills' (Green 1979). Between the zones there are narrow undecor-ated strips, so that the form as a whole articulates into distinct segments (in Marquesan, paka, "crusts"). The effect of the zone-infill principle is to interfere with the perception of outlines, so that the bodyseems to dissolve into^myriad fragments. This fragmentation, in turn corresponds to trie "non-unitary way in which the Marquejans^treaterTtheir bodies conceptually, in that it was customary (for chiefs, especially) to have separate proper names for individual body-parts in addition to a personal name (Linton 1923). Thus a man (Roger, say) would have a head called Peter, a back called William, an arm called Charles and genitals called Henry, etc. Each of these separately-named members of the body would have its own life, be subject to its own tapu restrictions, and so on. The same principle applied to important artefacts, such as canoes, which also had separate proper names for all their different parts. In this, one perceives the passion for distinctiveness which, I have argued, is the corollary of the religious premises of immanence; difference, non-substitutability, is at a premium because it is only provisional, and can be annihilated at a stroke____
The effects of tattooing was to de-totalize the body_into_djst^tfragments, exaggerating differences and conserving tapu"equality. But the fragments themselves, the
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
301
HANAUP
1 hope-vehine
2 tt'i pepedipw «"t rt malahoata ß ntho peala
3 li'i pepehipu mil ji &
poep
lepoe hno
Figure 5   Marquesan tattooing (side), by Von den Steinen*
"crusts", were a means of giving integrity to the body as a whole, by "wrapping" it. Often this is brought out in the nomenclature of individual motifs, which metaphorically evoke certain well-protected, enclosed creatures, objects and spaces (W. Handy 1922; Von den Steinen 1925). The most important animal motif was honu or kea, the generic name for testudinates, turtles and tortoises. These were particularly sacred creatures, which could only be eaten by chiefs, as a result of prestige conferred on them by their excellent armour, which corresponded to the ideal of "closure". Other "closure" motifs were shellfish motifs [poi'i), crab motifs (karu), enclosed garden motifs (papua), and a large class of motifs based on ipu (calabash, bowl, dish, etc.). But inspection of Figure 5 also reveals that, besides "enclosing" the body, Marquesan tattoo also protected it from immanence by multiplying it. The person was multiplied via an entourage of tattooed supernumeraries thronging his armoured integument. There are two types of such motifs: homunculi (etua, "gods") and tattooed "secondary faces", as Von den Steinen calls them. There is a clear structural relationship
302
ALFRED GELL
between the tattooing style of the Marquesas, in which the self was furnished with a collection of subsidiary selves to surround and protect it, and the type of Siamese twin image found in Figure 2,... The effect of the tattooing of "etua" motifs and secondary faces was to convert the body into such an n-tuplet. This is revealed particularly clearly in one Langsdorff's engravings of a young Marquesan Warrior sketched in 1801 by a Russian expedition (Figure 6). While mid-nineteenth-century Marquesan tattooing had become somewhat abstract, the more readable eighteenth-century style featured an important motif consisting of a death's-head positioned directly over the mid-line of the back. This design {mata Komo'e) had specifically military and protective functions, representing a dead (divinized) chief who was to be avenged. The Marquesan warrior, with his ghostly companion at his back, warding off harm, provides a striking male counterpart to the female Siamese twin images we considered earlier (cf. Figure 2). [•••]
Figure 6  Marquesan tattooing (incomplete), by Von den Steinen1
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
303
My general argument, up to now, has been that Polynesian cosmology is correlated with the religious attitude of immanence. The world is a space within the deity (the plenum of diety) that is kept apart through the preservation of difference. But everything in the ao has only a relative - one could say, "an embattled" - existence, and must eventually be consumed by the night, as the Maori culture-hero Maui was crushed within the grinding genitals of Hine-nui-te-po, in his vain attempt to confer immortality on himself, and on mankind (to have achieved this he would have had to pass right through her, exiting through her mouth into the unimaginable country beyond the night). The Polynesians' refusal of transcendence explains both the extraordinary passion of their mourning ceremonies, and the unbounded relief, which I mentioned before, afforded them by Christian eschatology. What I want to turn to now is certain Polynesian mortuary practices that reflected the underlying religious attitudes to which I have referred.
The outcome of ordinary deaths, in most Polynesian belief systems, was that the dead person would return to the po, a journey which was usually imagined as going under the sea, towards the west and the setting sun. On arrival in the pitch-black underworld, the dead person would be eaten by the atua there, who would gradually scrape and gnaw the flesh away until only the bones were left. Only then would the deceased (on the most optimistic assumptions) become an atua himself or herself. The images of darkness, submersion, excoriation, cannibalization and extinction -not as punishment, but as a consequence of the mere fact of death - relate to the fundamental nemesis that must await the person in a cosmos pervaded by immanence. Death assumes a particularly terrifying shape in this type of intellectual milieu, and it is surely true that the Polynesians went further than most in increasing its terrors in the here and now in the treatment they meted out to the weak and captive whom they made into cannibal victims and human sacrifices.
The typical pattern of Polynesian mortuary practices (in Central and East Polynesia, anyway) was, first of all, exaggerated mourning, usually involving self-scarification by the bereaved, especially women, and the dispatching of funerary victims, in the case of great chiefs. There followed the exposure of the corpse and "mummification" (so called, though it was not really comparable to the ancient Egyptian variety), followed eventually by the final disposal of the dry remains, usually only the bones, in secret places, most frequently caves. Some bones were kept in family temples, and some were turned into ornaments, weapons and tools; but it would not be true to say that they ever became significant cult objects.
The only aspect of these mortuary practices that I propose to discuss here is the middle phase of the sequence, i.e. the desiccation or mummification of the corpse. And the point that I want to make about this phase of the mortuary sequence is that, once again, it reflects the basic cosmological scheme of immanence. Here I have one final illustration... of a scene witnessed, iii 1844, Tjy the missionary... Pere Aimable Petithomme... (Williamson 1933). In this year died Iotete, in his hey-day called "le roi de Tahuata", a Marquesan chief of great stature, cunning, and renown, who sought to attain the status of the Pomare "Kings" of Tahiti, but who was eventually humbled (Thomas 1986). Petithomme, visiting the widow of this chief shortly after his death, was surprised to discover her stripping the skin, bit by bit, off his body.
The missionary asked the queen why she was obliged to remove her late husband's skin in this way. She replied that the goddess Oupu, who ruled over the afterworld, had decreed that none who had tattooed skins might enter the "Havaiki" inhabited
304
ALFRED GELL
by the gods themselves, "a land of delights planted with all sorts of excellent fruits and adorned by the waters of a blue and calm lake", but would be consigned for ever to a swamp where the sun never shone. This obligation to excoriate the corpse (together with the obligation to feed the gods of the po with numerous mortuary sacrifices) had to be fulfilled in order to assure a "desirable" afterlife.
One can easily see that excoriation of the corpse simply pre-empted the "scraping" that the dead were expected to undergo at the hands of the atua in the po according to the standard conception of the afterlife in Polynesian religion. [...]
But why did the queen specifically state that it was necessary to efface the tattooing on the skin of her dead consort in order to assure his place in paradise? Here the arguments of this paper comes full circle. We saw earlier that tattooing was conceptualized as "wrapping" {pahu) - wrapping that preserved the integrity of the self and prevented the diffusion of difference and tapu. Tattooing was a necessary part of worldly existence, part of a battery of essentially "defensive" structures that constituted the Polynesian ego - "character armour", to borrow a very apposite phrase from the ego-psychology of W. Reich (1950). But tattooing was incompatible with unmediated sacredness. For all the effort and inventiveness that the Marque-sans lavished on the art of tattoo, they explicitly believed that their gods were not tattooed.
Finally, when the fragile defensive structures on which they depended caved in, and life was at an end, it was necessary to dismantle the defences bit by bit, in order fully to achieve the fusion and de-differentiation of absolute death. Not to attain a subsequent state of immortality, but an antecedent one: to make it as if the living individual had never been at all. Hence the mortuary sequence is essentially a playing out of the cosmogonic process in reverse, enacting gestation and birth in reverse (or "de-conception": cf. Mosko 1983) at the hands of the female mourners/ cannibals. The incremental logic of the passage through life's stages was replaced by a subtractive logic of remorseless exposure, desiccation and dissipation. This was not a "new birth" as Dante imagines it, in his famous lines on the flaying of Marsyas, who finds everlasting life "Traeste dalla vagina delle membre sue" - "dragged from the sheath (vagina) of his own limbs - so much as birth undone, life stripped away, and a return to the status quo ante, in which the creation of the universe and all its consequent effects had been cancelled and annulled.
NOTE
Further discussion of the topics raised in this essay are continued in a book written by myself and published by Clarendon Press entitled Wrapping in Images: Tattooing in Polynesia.
REFERENCES
Bott, E. (1982). Tongan Society at the Time of Captain Cook's Visit. Memoirs of the
Polynesian Society, 44. Wellington. Buck, Sir P. (1930). Samoan Material Culture. Bishop Museum Bulletin, No. 75. Honolulu.
AN ESSAY ON POLYNESIAN COSMOLOGY AND RITUAL
305
Colocott, V. (1921). Notes on Tongan Religion. Journal of the Polynesian Society, 30, 152-63, 227^0.
Dordillon (1934-5). Dictionnaire de I'homme des lies Marquises. Societe des Oceanistes, Paris.
Dumont, L. (1970). Homo Hierachicus. University of Chicago Press, Chicago.
Gell, A. (1993). Wrapping in Images: Tattooing in Polynesia. Clarendon Press, Oxford.
Green, R. (1979). Early Lapita Art. In: S. Mead (ed.), Exploring the Visual Art of Oceania.
University of Hawaii Press, Honolulu. Hale, H. (1846). United States Exploring Expedition 1838-42 under the Command of
Captain Wilkes... Ethnology and Philology. Lea and Blanchard, Philadelphia. Handy, E. (1923). Native Culture of the Marquesas. Bishop Museum Bulletin, 9. Honolulu. Handy, W. (1922). Tattooing in the Marquesas. Bishop Museum Bulletin, 1. Honolulu. Hanson, F. (1982). Female Pollution in Polynesia. Journal of the Polynesian Society, 91,
335-81.
Henry, T. (1928). Ancient Tahiti. Bishop Museum Bulletin, 48. Honolulu. Kraemer, A. (1902). Die Samoa-Inseln. E. Schweizerbart, Stuttgart.
Linton, R. (1923). The Material Culture of the Marquesas. Bishop Museum Memoirs, No. 8, Part 5. Honolulu.
Milner, G. (1969). Siamese Twins, Birds and the Double Helix. Man, NS 4, 5-24.
Mosko, M. (1983). Conception, De-conception and Social Structure in Bush Mekeo culture.
Mankind, 14, 24-32. Oliver, D. (1974). Ancient Tahitian Society. University of Hawaii Press, Honolulu. Reich, W. (1950). Selected Writings. Vision Press, London. Reiter, F. (1907). Traditions Tonguiennes. Anthropos, 2, 743-54.
Shore, B. (1982). Sala'ilua: A Samoan Mystery. Columbia University Press, New York. Stair, J. (1897). Old Samoa. Religious Tract Society, London. Steinen, K. Von den (1925). Die Marquesaner und ihre Kunst. D. Reimer, Berlin. Thomas, N. (1986). Le Roi de Tahuata. Journal of Pacific History, 21, 3-20.
-(1990). Marquesan Societies. Oxford University Press, Oxford.
Turner, G. (1884). Samoa 100 Years Ago and Long Before. Macmillan, London. Williamson, R. (1933). The Religious and Cosmic Beliefs of Central Polynesia. Cambridge University Press, Cambridge.
22
Cosmological Deixis and Amerindian Perspectivism
Eduardo Viveiros de Castro
Eduardo Viveiros de Castro, "Cosmological Deixis and Amerindian Perspectivism," Journal of the Royal Anthropological Institute, n.s. 4 (3) (Oxford: Blackwell Publishers, 1998): 469-88.
Eduardo Viveiros de Castro is a Brazilian anthropologist who has taught in Rio de Janeiro, Paris, and Cambridge. Here he provides a brilliant and original account of Amerindian cosmology, situating it with respect to the literature on totemism, animism, and myth. He shows the lively fieldwork-inspired legacy of Levi-Strauss in Brazilian anthropology and is also able to link his account of lowland (Amazonian) South American groups to people as far away as the Cree of northern Canada. He describes what he calls "perspectivism" and '^joau^f^iu^isn^^e ideas that humans, animals, and spirits see both themselves and one another differently from different kinds of bodies.
Viveiros de Castro's essay is also an intervention in the literature on shamanism, providing a particularly sharp account of what it is the shaman does and linking shamanism in a fascinating but nonreductionjstj^yjyj^iiujiting,. The discussion of bpdyTderoratfWcanTDe instructively compared to Gell's account of Polynesian tattooing. Note also that whereas the thrust of Stallybrass and White's arguments might be to see universals in a practice like masking, Viveiros de Castro distinguishes very sharply between masking in Amerindian society and in the West. For a more complete picture, turn to his monograph (1992).
Further references to Amazonian and Northern North American cosmologies can be found in the essay. Other excellent Americanist accounts of cosmology include Hallowell (1960,1967 [1955]) on Ojibwa, Witherspoon (1977) on Navajo, and Ortiz (1969) on Tewa.
This study discusses the meaning of Amerindian "perspectivism": the ideas in Amazonian cosmologies concerning the way in which humans, animals and spirits see both
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
307
themselves and one another. Such ideas suggest the possibility of a redefinition of the classical categories of "nature", "culture" and "supernature" based on the concept of perspective or point of view. The study argues in particular that the antinomy between two characterizations of indigenous thought - on the one hand "ethnocentrism", which would deny the attributes of humanity to humans from other groups, and on the other hand "animism", which would extend such qualities to beings of other species - can be resolved if one considers the difference between the spiritual and corporal aspects of beings.
... la reciprocity de perspectives ou j'ai vu le caractere propre de la pensee mythique...
Levi-Strauss 1985: 268
Introduction
This article deals with that aspect of Amerindian thought which has been called its "perspectival quality" (Arhem 1993): the conception, common to many peoples of the continent, according to which the world is inhabited by different sorts of subjects or persons, human and non-human, which apprehend reality from distinct points of view. This idea cannot be reduced to our current concept of relativism (Lima 1995; 1996), which at first it seems to call to mind. In fact, it is at right angles, so to speak, to the opposition between relativism and universalism. Such resistance by Amerindian perspectivism to the terms of our epistemological debates casts suspicion on the robustness and transportability of the ontological partitions which they presuppose. In particular, as many anthropologists have already concluded (albeit for other reasons), the classic distinction between Nature and Culture cannot be used to describe domains internal to non-Western cosmologies without first undergoing a rigorous ethnographic critique.
Such a critique, in the present case, implies a redistribution of the predicates subsumed within the two paradigmatic sets that traditionally oppose one another under the headings of "Nature" and "Culture": universal and particular, objective and subjective, physical and social, fact and value, the given and the instituted, necessity and spontaneity, immanence and transcendence, body and mind, animality and humanity, among many more. Such an ethnographically-based reshuffling of our conceptual schemes leads me to suggest the expression, ''rmdrinatj^lism", to designate one of the contrastive features of Amerindian thought in relation to Western "multiculturalist" cosmologies. Where the latter are founded on the mutual implication of the unity of nature and the plurality of cultures - the first guaranteed by the objective universality of body and substance, the second generated by the subjective particularity of spirit and meaning - the Amerindian conception would suppose a spiritual unity and a corporeal diversity. Here, culture or the subject would be the form of the unlversaTTwTiil^^ object would be the form of the particular.
This inversion, perhaps too symmetrical to be more than speculative, must be developed by means of a plausible phenomenological interpretation of Amerindian cosmological categories, which determine the constitutive conditions of the relational contexts we can call "nature" and "culture". Clearly, then, I think that the distinction between Nature and Culture must be subjected to critique, but not in order to reach the conclusion that such a thing does not exist (there are already toe
308
EDUARDO VIVEIROS DE CASTRO
many things which do not exist). The flourishing industry of criticisms of the Westernizing character of all dualisms has called for the abandonment of our conceptually dichotomous heritage, but to date the alternatives have not gone beyond the stage of wishful unthinking. I would prefer to gain a perspective on our own contrasts, contrasting them with the distinctions actually operating in Amerindian perspectivist cosmologies.
Perspectivism
The initial stimulus for the present reflections were the numerous references in Amazonian ethnography to an indigenous theory according to which the way humans perceive animals and other subjectivities that inhabit the world - gods, spirits, the dead, inhabitants of other cosmic levels, meteorological phenomena, plants, occasionally even objects and artefacts - differs profoundly from the way in which these beings see humans and see themselves, r* Typically, in normal conditions, humans see humans as humans, animals as animals and spirits (if they see them) as spirits; however animals (predators) and spirits see humans as animals (as prey) to the same extent that animals (as prey) see humans as spirits or as animals (predators). By the same token, animals and spirits see themselves as humans: they perceive themselves as (or become) anthropomorphic beings when they are in their own houses or villages and they experience their own habits and characteristics in the form of culture - they see their food as human food (jaguars see blood as manioc beer, vultures see the maggots in rotting meat as grilled fish, etc.), they see their bodily attributes (fur, feathers, claws, beaks etc.) as body decorations or cultural instruments, they see their social system as organized in the same way as human institutions are (with chiefs, shamans, ceremonies, exogamous moieties, etc.). This "to see as" refers literally to percepts and not analogically to concepts, although in some cases the emphasis is placed more on \the_categorical rather than on the sensory aspect of the phenomenon.
In sum, animals are people, or see themselves as persons. Such a notion is virtually always associated with the idea that the manifest form of each species is a mere envelope (a "clothing") which conceals an internal human form, usually only visible to the eyes of the particular species or to certain trans-specific beings such as shamans. This internal form is the "soul" or "spirit" of the animal: an intentionality or subjectivity formally identical to human consciousness, materializable, let us say, (T in a human bodily schema concealed behind an animal mask. At first sight then, we '   would have a distinction between an anthropomorphic essence of a spiritual type, common to animate beings, and a variable bodily appearance, characteristic of each { individual species but which rather than being a fixed attribute is instead a change-\\ able and removable clothing. This notion of "clothing" is one of the privileged expressions of metamorphosis - spirits, the dead and shamans who assume animal form, beasts that turn into other beasts, humans that are inadvertently turned into animals - an omnipresent process in the "highly transformational world" (Riviere 1994: 256) proposed by Amazonian ontologies.1
This perspectivism and cosmological transformism can be seen in various South American ethnographies, but in general it is only the object of short commentaries and seems to be quite unevenly elaborated.2 It can also be found, and maybe with
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
309
even greater generative value, in the far north of North America and Asia, as well as amongst hunter-gatherer populations of other parts of the world.3 In South America, the cosmologies of the Vaupes area are in this respect highly developed (see Arhem 1993; 1996; Hugh-Jones 1996; Reichel-Dolmatoff 1985), but other Amazonian societies, such as the Wari' of Rondonia (Vilaca 1992) and the Juruna of the Middle Xingu (Lima 1995; 1996), also give equal emphasis to the theme.
Some general observations are necessary. Perspectivism does not usually involve all animal species (besides covering other beings); the emphasis seems to be on those species which perform a key symbolic and practical role such as the great predators and the principal species of prey for humans - one of the central dimensions, possibly even the fundamental dimension, of perspectival inversions refers to the relative and relational statuses of predator and prey (Arhem 1993: 11-12; Vilaca 1992: 49-51). On the other hand, however, it is not always clear whether spirits or subjectivities are being attributed to each individual animal, and there are examples of cosmologies which deny consciousness to post-mythical animals (Overing 1985: 249 sqq.; 1986: 245-6) or some other spiritual distinctiveness (Baer 1994: 89; Viveiros de Castro 1992a: 73-4). Nonetheless, as is well known, the notion of animal spirit "masters" ("mothers of the game animals", "masters of the white-lipped peccaries", etc.) is widespread throughout the continent. These spirit masters, clearly endowed with intentionality analogous to that of humans, function as hypostases of the animal species with which they are associated, thereby creating an intersubjective field for human-animal relations even where empirical animals are not spiritualized.
We must remember, above all, that if there is a virtually universal Amerindian notion, it is that of an original state of undifferentiation between humans and Smmals, described in mythology. Myths are filled with beings whose form, name and behaviour inextricably mix human and animal attributes in a common context of intercommunicability, identical to that which defines the present-day intra-human «world. The differentiation between "culture" and "nature", which Levi-Strauss i showed to be the central theme of Amerindian mythology, is not a process of differentiating the human from the animal, as in our own evolutionist mythology. The original common condition of both humans and animals is not animality but rather humanity. The great mythical separation reveals not so much culture distinguishing itself from nature but rather nature distancing itself from culture: the myths tell how animals lost the qualities inherited or retained by humans (Brightman 1993: 40, 160; Levi-Strauss 1985: 14, 190; Weiss 1972: 169-70). Humans are those who continue as they have always been: animals are ex-humans, not humans ex-animals. In sum, "the common point of reference for all beings of nature is not humans as a species but rather humanity as a condition" (Descola 1986: 120).
This is a distinction - between the human species and the human condition -which should be retained. It has an evident connexion with the idea of animal .clothing hiding a common spiritual "essence" and with the issue of the general if meaning of perspectivism. For the moment, we may simply note one of its main corollaries: the past humanity of animals is added to their present-day spirituality hidden by their visible form in order to produce that extended set of food restrictions or precautions which either declare inedible certain animals that were mythically co-substantial with humans, or demand their desubjectivization by shamanistic means before they can be consumed (neutralizing the spirit, transubstantiating the meat into plant food, semantically reducing it to other animals less proximate to humans),
310
EDUARDO VIVEIROS DE CASTRO
under the threat of illness, conceived of as a cannibal counter-predation undertaken by the spirit of the prey turned predator, in a lethal inversion of perspectives which transforms the human into animal.4
It is worth pointing out that Amerindian perspectivism has an essential relation with shamanism and with the valorization of the hunt. The association between shamanism and this "venatic ideology" is a classic question (for Amazonia, see Chaumeil 1983: 231-2; Crocker 1985: 17-25). I stress that this is a matter of symbolic importance, not ecological necessity: horticulturists such as the Tukano or the Juruna (who in any case fish more than they hunt) do not differ much from circumpolar hunters in respect of the cosmological weight conferred on animal predation, spiritual subjectivation of animals and the theory according to which the universe is populated by extra-human intentionalities endowed with their own perspectives. In this sense, the spiritualizationo^_rjlarjls_^meteorological phenomena or artefacts seems to me to be^e£on33xy^o£ derivative iricorm2arisori_wit^ spiritualization of animals: the animal is the extra^louman r^ojtotype of the-Qther, maintaining privileged relalfohs^wTtfT other prototypical figures of alterity, such as affines (Arhem 1996; Descola 1986: 317-30; Erikson 1984: 110-12). This hunting ideology is also and above all an ideology of shamans, insofar as it is shamans who administer the relations between humans and the spiritual component of the extra-humans, since they alone are capable _of assuming the point of view of such beings aad,Jii.particular, are cajjable of returning to tell the tale. If Western muTticultural-ism is relativism as public policy, then Amerindian perspectivist shamanism is multi-naturalism as cosmic politics.
Animism
The reader will have noticed that my "perspectivism" is reminiscent of the notion of "animism" recently recuperated by Descola (1992; 1996 [sic]). Stating that all conceptualizations of non-humans always refer to the social domain, Descola distinguishes three modes of objectifying nature: totemism^ where the differences between natural species are used as a model for social distinctions; that is, where the relationship between nature and culture is metaphorical in character and marked by discontinuity (both within and between series); animism; where the "elementary categories structuring social life" organize the relations between humans and natural species, thus defining a social continuity between nature and culture, founded on the attribution of human dispositions and social characteristics to "natural beings" (Descola 1996 [sic]: 87-8); and naturalism, typical of Western cosmologies, which supposes an ontologicaj_dj^lity^etweenjiajjir.e, the domain of necessity, and cuh-ture, the domain of spontaneity, areas separated by metonymic discontinuity. The "aiwmic mode" is characteristic of societies in which animals are the "strategic focus of the objectification of nature and of its socialization" (1992: 115), as is the case amongst indigenous ^peoples of America, reigning supreme over those social morphologies lacking in elaborate internal segmentations. But this mode can also be found co-existing or combined with totemism, wherein such segmentations exist, the Bororo and their aroe/bope dualism being such a case.s
These ideas form part of a theory which I cannot discuss here as fully as it would merit. I merely comment on the contrast between animism and naturalism but from
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
311
a somewhat different angle from the original one. (Totcmism, as defined by Descola, seems to me to be a heterogeneous phenomenon, primarily classificatory rather than cosmological: it is not a system of relations between nature and culture as is the case in the other two modes, but rather of purely logical and differential correlations.)
Animism could be defined as an ontology which postulates the social character ol relations between humans and non-humans: the space between nature and society is itself social. Naturalism is founded on the inverted axiom: relations between society and nature are themselves natural. Indeed, if in the animic mode the distinction "nature/culture" is internal to the social world, humans and animals being immersed in the same socio-cosmic medium (and in this sense "nature" is a part of an encompassing sociality), then in naturalist ontology, the distinction "nature/culture" h internal to nature (and in this sense, human society is one natural phenomenor amongst others). Animism has "society" as the unmarked pole, naturalism has "nature": these poles function, respectively and contrastively, as the universal dimension of each mode. Thus animism and naturalism are hierarchical and meto-nymical structures (this distinguishes them from totemism, which is based on i metaphoric correlation between equipollent opposites).
In Western naturalist ontology, the nature/society interface is natural: humans art organisms like the rest, body-objects in "ecological" interaction with other bodies and forces, all of them ruled by the necessary laws of biology and physics; "product ive forces" harness, and thereby express, natural forces. Social relations, that is contractual or instituted relations between subjects, can only exist internal to humar society. But how alien to nature - this would be the problem of naturalism - ar< these relations? Given the universality of nature, the status of the human and socia world is unstable and, as the history of Western thought shows, it perpetualh oscillates between a naturalistic monism ("sociobiology" being one of its curren avatars) and an ontological dualism of nature/culture ("culturalism" being its con temporary expression). The assertion of this latter dualism, for all that, only re inforces the final referential character of the notion of nature, by revealing itself t( be the direct descendant of the opposition between Nature and Superstore. Cultun is the modern name of Spirit - let us recall the distinction between Naturwis senschaften and Geisteswissenschaften - or at the least it is the name of the com promise between Nature and Grace. Of animism, we would be tempted to say tha the instability is located in the opposite pole: there the problem is how to administe the mixture of humanity and animality constituting animals, and not, as is the cas amongst ourselves, the combination of culture and nature which characteriz humans; the point is to differentiate a "nature" out of the universal sociality.
However, can animism be defined as a projection of differences and qualitie internal to the human world onto non-human worlds, as a "socio-centric" mode in which categories and social relations are used to map the universe? This intei pretation by analogy is explicit in some glosses on the theory: "if totemic system model society after nature, then animic systems model nature after society" (Arher 1996:185). The problem here, obviously, is to avoid any undesirable proximity wit the traditional sense of "animism", or with the reduction of "primitive classifies tions" to emanations of social morphology; but equally the problem is to go beyon other classical characterizations of the relation between society and nature such a Radcliffe-Brown's.6
312
EDUARDO VIVEIROS DE CASTRO
Ingold (1991; 1996) showed how schemes of analogical projection or social modelling of nature escape naturalist reductionism only to fall into a nature/culture dualism which by distinguishing "really natural" nature from "culturally constructed" nature reveals itself to be a typical cosmological antinomy faced with infinite regression. The notion of model or metaphor supposes a previous distinction between a domain wherein social relations are constitutive and literal and another where they are representational and metaphorical. Animism, interpreted as human sociality projected onto the non-human world, would be nothing but the metaphor of a metonymy.
Amongst the questions remaining to be resolved, therefore, is that of knowing whether animism can be described as a figurative use of categories pertaining to the human-social domain to conceptualize the domain of non-humans and their relations with the former. Another question: if animism depends on the attribution of human cognitive and sensory faculties to animals, and the same form of subjectivity, then what in the end is the difference between humans and animals? If animals are pe^lejjhefi^hT^tkrthe^^ Why, to be precise, the perspectivism?
Finally, if animism is a way of objectifying nature in which the dualism of nature/ culture does not hold, then what is to be done with the abundant indications regarding the centrality of this opposition to South American cosmologies? Are we dealing with just another "totemic illusion", if not with an ingenuous projection of our Western dualism?
Ethnocentrism
In a well-known essay, Levi-Strauss observed that for "savages" humanity ceases at the boundary of the group, a notion which is exemplified by the widespread auto-ethnonym meaning "real humans", which, in turn, implies a definition of strangers as somehow pertaining to the domain of the extra-human. Therefore, ethnocentrism would not be the privilege of the West but a natural ideological attitude, inherent to human collective life. Levi-Strauss illustrates the universal reciprocity of this attitude with an anecdote:
In the Greater Antilles, some years after the discovery of America, which the Spanish were dispatching inquisitional commissions to investigate whether the natives had a soul or not, these very natives were busy drowning the white people they had captured in order to find out, after lengthy observation, whether or not the corpses were subject to putrefaction. 1 (1973: 384)
The general point of this parable (from which Levi-Strauss derived the famous moral: "The barbarian is first and foremost the man who believes in barbarism") is quite simple: the Indians, like the European invaders, considered that only the group to which they belong incarnates humanity; strangers are on the other side of the border which separates humans from animals and spirits, culture from nature and supernature. As matrix and condition for the existence of ethnocentrism, the nature/culture opposition appears to be a universal of social apperception.
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
313
At the time when Levi-Strauss was writing these lines, the strategy of vindicating the full humanity of savages was to demonstrate that they made the same distinctions as we do: the proof that they were true humans is that they considered that they alone were the true humans. Like us, they distinguished culture from nature and they too believed that Naturvölker are always the others. The universality of the cultural distinction between Nature and Culture bore witness to the universality of culture as human nature. In sum, the answer to the question of the Spanish investigators (which can be read as a sixteenth-century version of the "problem of other minds") was positive: savages do have souls.
Now, everything has changed. The savages are no longer ethnocentric but rather cosmocentric; instead of having to prove that they are humans because they distinguish themselves from animals, we now have to recognize how mhuman we are for opposing humans to animals in a way they never did: for them nature and culture are part of the same sociocosmic field. Not only would Amerindians put a wide berth between themselves and the Great Cartesian Divide which separated humanity from animality, but their views anticipate the fundamental lessons of ecology which we are only now in a position to assimilate (Reichel-Dolmatoff 1976). Before, the Indians' refusal to concede predicates of humanity to other men was of note; now we stress that they extend such predicates far beyond the frontiers of their own species in a demonstration of "ecosophic" knowledge (Arhem 1993) which we should emulate in as far as the limits of our objectivism permit. Formerly, it had been necessary to combat the assimilation of the savage mind to narcissistic animism, the infantile stage of naturalism, showing that totemism affirmed the cognitive distinction between culture and nature; now, neo-animism reveals itself as the recognition of the universal admixture of subjects and objects, humans and non-humans against modern hubris, the primitive and post-modern "hybrids", to borrow a term from Latour (1991).
Two antinomies then, which are, in fact, only one: either Amerindians are ethno-centrically "stingy" in the extension of their concept of humanity and they "totemi-cally" oppose nature and culture; or they are cosmocentric and "animic" and do not profess to such a distinction, being models of relativist tolerance, postulating a multiplicity of points of view on the world.
I believe that the solution to these antinomies7 lies not in favouring one branch over the other, sustaining, for example, the argument that the most recent characterization of American attitudes is the correct one and relegating the other to the outer darkness of pre-post-modernity. Rather, the point is to show that the "thesis" as well as the "antithesis" are true (both correspond to solid ethnographic intuitions), but that they apprehend the same phenomena from different angles; and also it is to show that both are false in that they refer to a substantivist conceptualization of the categories of Nature and Culture (whether it be to affirm or negate them) .^which is not applicable to Amerindian cosmologies.
I' The first point to be considered is that the Amerindian words which are usually (^translated as "human being" and which figure in those supposedly ethnocentric self-i^signations do not denote humanity as a natural species. They refer rather to the sTjcTaTxQruiitir>n of personhood, and they function (pragmatically when not syntactically) less as nouns than as pronouns. They indicate the position of the subject; they are enunciative4nlrkers, not names. Far from manifesting a semantic shrinking of a common name to a proper name (taking "people" to be the name of the tribe), these
314
EDUARDO VIVEIROS DE CASTRO
words move in the opposite direction, going from substantive to perspective (using "people" as a collective pronoun "we people/us"). For this very reason, indigenous categories of identity have that enormous contextual variability of scope that characterizes pronouns, marking contrastively Ego's immediate kin, his/her local group, all humans, or even all beings endowed with subjectivity: their coagulation as "ethnonyms" seems largely to be an artefact of interactions with ethnographers. Nor is it by chance that the majority of Amerindian ethnonyms which enter the literature are not self-designations, but rather names (frequently pejorative) conferred by other groups: ethnonymic objectivation is primordially applied to others, not to the ones in the position of subject. Ethnonyms are names of third parties; they belong to the category of "they" not to the category of "we". This, by the way, is consistent with a widespread avoidance of self-reference on the level of personal onomastics: names are not spoken by the bearers nor in their presence; to name is to externalize, to separate (from) the subject.
Thus self-references such as ^pef^ple" mean "person", not "member of the human species", and they are personal pronouns registering the point of view of the subject talking, not proper names. To say, then, that animals and spirits are people is to say that they are persons, and to attribute to non-humans the capacities of conscious intentionality and agency which define the position of the subject. Such capacities are objectified as the soul or spirit with which these non-humans are endowed. Whatever possesses a soul is a subject, and whatever has a soul is capable of having a point of view. Amerindian souls, be they humans or animal, are thus indexical categories, cosmological deictics whose analysis calls not so much for an animist psychology or substantialist ontology as for a theory of the sign or a perspectival pragmatics (Taylor 1993«; 1993b; Viveiros de Castro 1992b).
Thus, every being to whom a point of view is attributed would be a subject; or better, wherever there is a point of view there is a subject position. Whilst our constructionist epistemology can be summed up in the Saussurean formula: the point of view creates the object - the subject being the original, fixed condition whence the point of view emanates - Amerindian ontological perspectivism proceeds along the lines that the point of view creates the subject; whatever is activated or "agented" by the point of view will be a subject.8 This is why terms such as wari' (Vilaca 1992), dene (McDonnell 1984) or masa (Arhem 1993) mean "people", but they can be used for - and therefore used by - very different classes of beings: used by humans they denote human beings; but used by peccaries, howler monkeys or beavers they self-refer to peccaries, howler monkeys or beavers.
As it happens, however, these non-humans placed in the subject perspective do not merely "call" themselves "people"; they see themselves anatomically and culturally as humans. The symbolic spiritualization of animals would imply their imaginary hominization and culturalization; thus the anthropomorphic-anthropocentric character of indigenous thought would seem to be unquestionable. However, I believe that something totally different is at issue. Any being which vicariously occupies the point of view of reference, being in the position of subject, sees itself as a member of the human species. The human bodily form and human culture - the schemata of perception and action "embodied" in specific dispositions - are deictics of the same type as the self-designations discussed above. They are reflexive or apperceptive schematisms by which all subjects apprehend themselves, and not literal and constitutive human predicates projected metaphorically (i.e. improperly) onto non-
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
315
humans. Such deictic "attributes" are immanent in the viewpoint, and move with it (Brightman 1993: 47). Human beings - naturally - enjoy the same prerogative and therefore see themselves as such.9 It is not that animals are subjects because they are
( humans in disguise, but rather that they are human because they are potential subjects. This is to say Culture is the Subject's nature; it is the form in which every subject experiences its own nature. Animism is not a projection of substantive
>. human qualities cast onto animals, but rather expresses the logical equivalence of the reflexive relations that humans and animals each have to themselves: salmon are to (see) salmon as humans are to (see) humans, namely, (as) human.10 If, as we have observed, the common condition of humans and animals is humanity not animality, this is because "humanity" is the name for the general form taken by the Subject.
Multinaturalism
With this we may have discarded analogical anthropocentrism, but only apparently to adopt relativism.11 For would this cosmology of multiple viewpoints not imply that "every perspective is equally valid and true" and that "a correct and true representation of the world does not exist" (Arhem 1993: 124)?
But this is exactly the question: is the Amerindian perspectivist theory in fact / asserting a multiplicity of representations of the same world? It is sufficient to consider ethnographic evidence to perceive that the opposite applies: all beings see ("represent") the world in the same way - what changes is the world that they see. Animals impose the same categories and values^rTreality aTiTflmans ~5o: their v worlds, like ours, revolve arourTcTnunting and fishing, cooking and fermented drinks, cross-cousins and war, initiation rituals, shamans, chiefs, spirits. "Everybody is involved in fishing and hunting; everybody is involved in feasts, social hierarchy, chiefs, war, and disease, all the way up and down" (Guedon 1984:142). If the moon, snakes and jaguars see humans as tapirs or white-lipped peccaries (Baer 1994: 224), it is because they, like us, eat tapirs and peccaries, people's food. It could only be this way, since, being people in their own sphere, non-humans see things as "people" do. But the things that they see are different: what to us is blood, is maize beer to the jaguar; what tVtFe souls of the dead is a rotting corpse, to us is soaking manioc; what we see as a muddy waterhole, the tapirs see as a great ceremonial house.
(Multi)cultural relativism supposes a diversity of subjective and partial representations, each strivfng to grasp an external and unified nature, which remains perfectly indifferent to those representations. Amerindian thought proposes the opposite: a representational or phenomenological unity which is purely pronominal or deictic, indifferently applied to a radically objective diversity. One single "culture", multiple "natures" - perspectivism is multinaturalist, for a perspective is not a representation.
A perspective is not a representation because representations are a property of the mind or spirit, whereas the point of view is located in the body.12 The ability to adopt a point of view is undoubtedly a power of the soul, and non-humans are subjects in so far as they have (or are) spirit; but the differences between viewpoints (and a viewpoint is nothing if not a difference) lies not in the soul. Since the soul is formally identical in all species, it can only see the same things everywhere - the difference is given in the specificity of bodies. This permits answers to be found for
316
EDUARDO VIVEIROS DE CASTRO
our questions: if non-humans are persons and have souls, then what distinguishes them from humans? And why, being people, do they not see us as people?
Animals see in the same way as we do different things because their bodies are different from ours. I am not referring to physiological differences - as far as that is concerned, Amerindians recognize a basic uniformity of bodies - but rather to affects, dispositions or capacities which render the body of every species unique: what it eats, how it communicates, where it lives, whether it is gregarious or solitary, and so forth. The visible shape of the body is a powerful sign of these differences in affect, although it can be deceptive since a human appearance could, for example, be concealing a jaguar-affect. Thus, what I call "body" is not a synonym for distinctive substance or fixed shape; it is an assemblage of affects or ways of being that constitute a habitus. Between the formal subjectivity of souls and the substantial materiality of organisms there is an intermediate plane which is occupied by the body as a bundle of affects and capacities and which is the origin of perspectives.
The difference between bodies, however, is only apprehendable from an exterior viewpoint, by an other, since, for itself, every type of being has the same form (the generic form of a human being): bodies are the way in which alterity is apprehended as such. In normal conditions we do not see animals as people, and vice-versa, because our respective bodies (and the perspectives which they allow) are different. Thus, if "culture" is a reflexive perspective of the subject, objectified through the concept of soul, it can be said that "nature" is the viewpoint which the subject takes of other body-affects; if Culture is the Subject's nature, then Nature is the form of the Other as body, that is, as the object for a subject. Culture takes the self-referential form of the pronoun "I"; nature is the form of the non-person or the object, indicated by the impersonal pronoun "it" (Benveniste 1966a: 256).
If, in the eyes of Amerindians, the body makes the difference, then it is easily understood why, in the anecdote told by Levi-Strauss, the methods of investigation into the humanity of the other, employed by the Spanish and the inhabitants of the Antilles, showed such asymmetry. For the Europeans, the issue was to decide whether the others possessed a soul; for the Indians, the aim was to find out what kind of body the others had. For the Europeans the great diacritic, the marker of difference in perspective, is the soul (are Indians humans or animals?); for the Indians it is the body (are Europeans humans or spirits?). The Europeans never doubted that the Indians had bodies; the Indians never doubted that the Europeans had souls (animals and spirits have them too). What the Indians wanted to know was whether the bodies of those "souls" were capable of the same affects as their own - whether they had the bodies of humans or the bodies of spirits, non-putrescible and protean. In sum: European ethnocentrism consisted in doubting whether other bodies have the same souls as they themselves; Amerindian ethnocentrism in doubting whether other souls had the same bodies.
As Ingold has stressed (1994; 1996), the status of humans in Western thought is essentially ambiguous: on the one hand, humankind is an animal species amongst others, and animality is a domain that includes humans; on the other hand, humanity is a moral condition which excludes animals. These two statuses co-exist in the problematic and disjunctive notion of "human nature". In other words, our cosmology postulates a physical continuity and a metaphysical discontinuity between humans and animals, the former making of man an object for the natural sciences, the latter an object for the "humanities". Spirit or mind is our great differentiator: it
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
317
raises us above animals and matter in general, it distinguishes cultures, it makes each person unique before his or her fellow beings. The body, in contrast, is the major integrator: it connects us to the rest of the living, united by a universal substrate (DNA, carbon chemistry) which, in turn, links up with the ultimate nature of all material bodies.13 In contrast to this, Amerindians postulate a metaphysical continuity and a physical discontinuity between the beings of the cosmos, the former resulting in animism, the latter in perspectivism: the spirit or soul (here not an immaterial substance but rather a reflexive form) integrates, while the body (not a material organism but a system of active affects) differentiates.
The Spirit's Many Bodies
The idea that the body appears to be the great differentiator in Amazonian cosmologies - that is, as that which unites beings of the same type, to the extent that it differentiates them from others - allows us to reconsider some of the classic questions of the ethnology of the region in a new light.
Thus, the now old theme of the importance of corporeality in Amazonian societies (a theme that much predates the current "embodiment" craze - see Seeger et al. 1979 [sic]) acquires firmer foundations. For example, it becomes possible to gain a better understanding of why the categories of identity - be they personal, social or cosmo-logical - are so frequently expressed through bodily idioms, particularly through food practices and body decoration. The universal symbolic importance of food and cooking regimes in Amazonia - from the mythological "raw and the cooked" of Levi-Strauss, to the Piro idea that what literally (i.e. naturally) makes them different from white people is "real food" (Gow 1991); from the food avoidances which define "groups of substance" in Central Brazil (Seeger 1980) to the basic classification of beings according to their eating habits (Baer 1994: 88); from the ontological productivity of commensality, similarity of diet and relative condition of prey-object and predator-subject (Vilaca 1992) to the ominipresence of cannibalism as the "predicative" horizon of all relations with the other, be they matrimonial, alimentary or bellicose (Viveiros de Castro 1993) - this universality demonstrates that the set of habits and processes that constitute bodies is precisely the location from which identity and difference emerge.
The same can be said of the intense semiotic use of the body in the definition of personal identities and in the circulation of social values (Mentore 1993; Turner 1995). The connexion between this overdetermination of the body (particularly of /its visible surface) and the restricted recourse in the Amazonian socius to objects ^-capable of supporting relations - that is, a situation wherein social exchange is not mediated by material objectifications such as those characteristic of gift and commodity economies - has been shrewdly pinpointed by Turner, who has shown ^how the human body therefore must appear as the prototypical social object. ^However, the Amerindian emphasis on the social construction of the body cannot be taken as the culturalization of a natural substract but ratherasjheproductionat a distinctlxJluman body, meaning~^tom7^Kumaii7Su^ a process seems to be expressing not soinlich~a~wIsn"~to "de-animalize" the body through its cultural marking, but rather to particularize a body still toojgeneric, differentiating it from the bodies of other human collectivities as well as from those of other species. The
318
EDUARDO VIVEIROS DE CASTRO
body, as the site of differentiating perspective, must be differentiated to the highest degree in order completely to express it.
The human body can be seen as the locus of the confrontation between humanity and animality, but not because it is essentially animal by nature and needs to be veiled and controlled by culture (Riviere 1994). The body is the subject's fundamental expressive instrument and at the same time the object par excellence, that which is presented to the sight of the other. It is no coincidence, then, that the maximum social objectification of bodies, their maximal particularization expressed in decoration and ritual exhibition is at the same time the moment of maximum animaliza-tion (Goldman 1975:178; Turner 1991; 1995), when bodies are covered by feathers, colours, designs, masks and other animal prostheses. Man ritually clothed as an animal is the counterpart to the animal supernaturally naked. The former, transformed into an animal, reveals to himself the "natural" distinctiveness of his body; the latter, free of its exterior form and revealing itself as human, shows the "supernatural" similarity of spirit. The model of spirit is the human spirit, but the model of body is the bodies of animals; and if from the point of view of the subject culture takes the generic form of "I" and nature of "it/they", then the objectification of the subject to itself demands a singularization of bodies - which naturalizes culture, i.e. embodies it - whilst the subjectification of the object implies communication at the level of spirit - which culturalizes nature, i.e. supernaturalizes it. Put in these terms, the Amerindian distinction of Nature/Culture, before it is dissolved in the name of a common animic human-animal sociality, must be re-read in the light of somatic perspectivism.
It is important to note that these Amerindian bodies are not thought of as given but rather as made. Therefore, an emphasis on the methods for the continuous fabrication of the body (Viveiros de Castro 1979); a notion of kinship as a process of active assimilation of individuals (Gow 1989; 1991) through the sharing of bodily substances, sexual and alimentary - and not as passive inheritance of some substantial essence; the theory of memory which inscribes it in the flesh (Viveiros de Castro 1992a: 201-7), and more generally the theory which situates knowledge in the body (Kensinger 1995: ch. 22; McCallum 1996). The Amerindian Bildung happens in the body more than in the spirit: there is no "spiritual" change which is not a bodily transformation, a redefinition of its affects and capacities. Furthermore, while the distinction between body and soul is obviously pertinent to these cosmologies, it cannot be interpreted as an ontological discontinuity (Townsley 1993: 454-5). As bundles of affects and sites of perspective, rather than material organisms, bodies "are" souls, just, incidentally, as souls and spirits "are" bodies. The dual (or plural) conception of the human soul, widespread in indigenous Amazonia, distinguishes between the soul (or souls) of the body, reified register of an individual's history, site of memory and affect, and a "true soul", pure, formal subjective singularity, the abstract mark of a person (e.g. McCallum 1996; Viveiros de Castro 1992fl: 201-14). On the other hand, the souls of the dead and the spirits which inhabit the universe are not immaterial entities, but equally types of bodies, endowed with properties -affects - sui generis. Indeed, body and soul, just like nature and culture, do not correspond to substantives, self-subsistent entities or ontological provinces, but rather to pronouns or phenomenological perspectives.
The performative rather than given character of the body, a conception that requires it to differentiate itself "culturally" in order for it to be "naturally" different,
COSMOLOGICAL DEIXIS AND AMERINDIAN PER5PECTIVISM
319
has an obvious connexion with interspecific metamorphosis, a possibility suggested by Amerindian cosmologies. We need not be surprised by a way of thinking which posits bodies as the great differentiators yet at the same time states their transform-ability. Our cosmology supposes a singular distinctiveness of minds, but not even for this reason does it declare communication (albeit solipsism is a constant problem) to be impossible, or deny the mental/spiritual transformations induced by processes such as education and religious conversion; in truth, it is precisely because the spiritual is the locus of difference that conversion becomes necessary (the Europeans wanted to know whether Indians had souls in order to modify them). Bodily metamorphosis is the Amerindian counterpart to the European theme of spiritual conversion.14 In the same way, if solipsism is the phantom that continuously threatens our cosmology - raising the fear of not recognizing ourselves in our "own kind" because they are not like us, given the potentially absolute singularity of minds - then the possibility of metamorphosis expresses the opposite fear, of no longer being able to differentiate between the human and the animal, and, in particular, the fear of seeing the human who lurks within the body of the animal one eats15 - hence the importance of food prohibitions and precautions linked to the spiritual potency of animals, mentioned above. The phantom of cannibalism is the Amerindian equivalent to the problem of solipsism: if the latter derives from the uncertainty as to whether the natural similarity of bodies guarantees a real community of spirit, then the former suspects that the similarity of souls might prevail over the real differences of body and that all animals that are eaten might, despite the shamanistic efforts to de-subjectivize them, remain human. This, of course, does not prevent us having amongst ourselves more or less radical solipsists, such as the relativists, nor that various Amerindian societies be purposefully and more or less literally cannibalistic.16
The notion of metamorphosis is directly linked to the doctrine of animal "clothing", to which I have referred. How are we to reconcile the idea that the body is the site of differentiating perspectives with the theme of the "appearance" and "essence" which is always evoked to interpret animism and perspectivism (Arhem 1993: 122; Descola 1986: 120; Hugh-Jones 1996; Riviere 1994)? Here seems to me to lie an important mistake, which is that of taking bodily "appearance" to be inert and false, whereas spiritual "essence" is active and real (see the definitive observations of Goldman 1975: 63). I argue that nothing could be further from the Indians' minds when they speak of bodies in terms of "clothing". It is not so much that the body is a clothing but rather that clothing is a body. We are dealing with societies which inscribe efficacious meanings onto the skin, and which use animal masks (or at least know their principle) endowed with the power metaphysically to transform the identities of those who wear them, if used in the appropriate ritual context. To put on mask-clothing is not so much to conceal a human essence beneath an animal appearance, but rather to activate the powers of a different body.17 The animal clothes that shamans use to travel the cosmos are not fantasies but instruments: they are akin to diving equipment, or space suits, and not to carnival masks. The intention when donning a wet suit is to be able to function like a fish, to breathe underwater, not to conceal oneself under a strange covering. In the same way, the "clothing" which, amongst animals, covers an internal "essence" of a human type, is not a mere disguise but their distinctive equipment, endowed with the affects and capacities which define each animal.18 It is true that appearances can be deceptive
320
EDUARDO VIVEIROS DE CASTRO
(Hallowell 1960; Riviere 1994); but my impression is that in Amerindian narratives which take as a theme animal "clothing" the interest lies more in what these clothes do rather than what they hide. Besides this, between a being and its appearance is its body, which is more than just that - and the very same narratives relate how appearances are always "unmasked" by bodily behaviour which is inconsistent with them. In short: there is no doubt that bodies are discardable and exchangeable and that "behind" them lie subjectivities which are formally identical to humans. But the idea is not similar to our opposition between appearance and essence; it merely manifests the objective permutability of bodies which is based in the subjective equivalence of souls.
Another classic theme in South American ethnology which could be interpreted within this framework is that of the sociological discontinuity between the living and the dead (Carneiro da Cunha 1978). The fundamental distinction between the living and the dead is made by the body and precisely not by the spirit; death is a bodily catastrophe which prevails as differentiator over the common "animation" of the living and the dead. Amerindian cosmologies dedicate equal or greater interest to the way in which the dead see reality as they do to the vision of animals, and as is the case for the latter, they underline the radical differences vis-ä-vis the world of the living. To be precise, being definitively separated from their bodies, the dead are not human. As spirits defined by their disjunction from a human body, the dead are logically attracted to the bodies of animals; this is why to die is to transform into an animal (Pollock 1985: 95; Schwartzman 1988: 268; Turner 1995:152; Vilaca 1992: 247-55), as it is to transform into other figures of bodily alterity, such as affines and enemies. In this manner, if animism affirms a subjective and social continuity between humans and animals, its somatic complement, perspectivism, establishes an objective discontinuity, equally social, between live humans and dead humans.19
Having examined the differentiating component of Amerindian perspectivism, it remains for me to attribute a cosmological "function" to the transspecific unity of the spirit. This is the point at which, I believe, a relational definition could be given for a category, Supernature, which nowadays has fallen into disrepute (actually, ever since Dürkheim), but whose pertinence seems to me to be unquestionable. Apart from its use in labelling cosmographic domains of a "hyper-uranian" type, or in defining a third type of intentional beings occurring in indigenous cosmologies, which are neither human nor animal (I refer to "spirits"), the notion of supernature may serve to designate a specific relational context and particular phenomenological quality, which is as distinct from the intersubjective relations that define the social world as from the "inter-objective" relations with the bodies of animals.
Following the analogy with the pronominal set (Benveniste 1966a; 1966b) we can see that between the reflexive "I" of culture (the generator of the concepts of soul or spirit) and the impersonal "it" of nature (definer of the relation with somatic alterity), there is a position missing, the "you", the second person, or the other taken as other subject, whose point of view is the latent echo of that of the "I". I believe that this concept can aid in determining the supernatural context. An abnormal context wherein a subject is captured by another cosmologically dominant point of view, wherein he is the "you" of a non-human perspective, Supernature is the form of the Other as Subject, implying an objectification of the human I as a "you" for this Other. The typical "supernatural" situation in an Amerindian world is the meeting in the forest between a man - always on his own - and a being which is
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
321
seen at first merely as an animal or a person, then reveals itself as a spirit or a dead person and speaks to the man (the dynamics of this communication are well analysed by Taylor 1993a).20 These encounters can be lethal for the interlocutor who, overpowered by the non-human subjectivity, passes over to its side, transforming himself into a being of the same species as the speaker: dead, spirit or animal. He who responds to a "you" spoken by a non-human accepts the condition of being its "second person", and when assuming in his turn the position of "I" does so already as a non-human. The canonical form of these supernatural encounters, then, consists in suddenly finding out that the other is "human", that is, that it is the human, which automatically dehumanizes and alienates the interlocutor and transforms him into a prey object, that is, an animal. Only shamans, multinatural beings by definition and office, are always capable of transiting the various perspectives, calling and being called "you" by the animal subjectivities and spirits without losing their condition as human subjects.21
I would conclude by observing that Amerindian perspectivism has a vanishing point, as it were, where the differences between points of view are at the same time annulled and exacerbated: myth, which thus takes on the character of an absolute discourse. In myth, every species of being appears to others as it appears to itself (as human), while acting as if already showing its distinctive and definitive nature (as animal, plant or spirit). In a certain sense, all the beings which people mythology are shamans, which indeed is explicitly affirmed by some Amazonian cultures (Guss 1989: 52). Myth speaks of a state of being where bodies and names, souls and affects, the I and the Other interpenetrate, submerged in the same pre-subjective and pre-objective milieu - a milieu whose end is precisely what the mythology sets out to tell.
NOTES
1 This notion of the body as a "clothing" can be found amongst the Makuna (Arhem 1993), the Yagua (Chaumeil 1983: 125-7), the Piro (Gow, pers. comm.), the Trio (Riviere 1994) and the Upper Xingu societies (Gregor 1977: 322). The notion is very likely pan-American, having considerable symbolic yield, for example, in North-west Coast cosmologies (see Goldman 1975 and Boelscher 1989), if not of much wider distribution, a question I cannot consider here.
2 For some examples see amongst many others: Weiss 1969: 158; 1972 (Campa); Baer 1994: 102, 119, 224; Renard-Casevitz 1991: 24-31 (Matsiguenga); Grenand 1980: 42 (Wayapi); Viveiros de Castro 1992a: 68 (Arawete); Osborn 1990: 151 (U'wa); Jara 1996: 68-73 (Akuriyo).
3 See for example, Saladin d'Anglure 1990; Fienup-Riordan 1994 (Eskimo); Nelson 1983; McDonnell 1984 (Koyukon, Kaska); Tanner 1979; Scott 1989; Brightman 1993 (Cree); Hallowell 1960 (Ojibwa); Goldman 1975 (Kwakiutl); Guedon 1984 (Tsimshian); Boelscher 1989 (Haida). See also Howell 1984; 1996; and Karim 1981, for the Chewong and Ma'Betisek of Malaysia; for Siberia, Hamayon 1990.
4 See Arhem 1993; Crocker 1985; Hugh-Jones 1996; Overing 1985; 1986; Vilaca 1992.
5 Or, as we may add, the case of the Ojibwa, where the co-existence of the systems of totem and manido (Levi-Strauss 1962a: 25-33) served as a matrix for the general opposition between totemism and sacrifice (Levi-Strauss 1962b: 295-302) and can be directly interpreted within the framework of a distinction between totemism and animism.
322
EDUARDO VIVEIROS DE CASTRO
6 See Radcliffe-Brown 1952: 130-1, who, amongst other interesting arguments, distinguishes processes of personification of species and natural phenomena (which "permits nature to be thought of as if it were a society of persons, and so makes of it a social or moral order"), like those found amongst the Eskimos and Andaman Islanders, from systems of classification of natural species, like those found in Australia and which compose a "system of social solidarities" between man and nature - this obviously calls to mind Descola's distinction of animism/totemism as well as the contrast of manidoltotem explored by Levi-Strauss.
7 The uncomfortable tension inherent in such antinomies can be gauged in Howell's article (1996) on Chewong cosmology, where the Chewong are described as being both "relativist" and "anthropocentric" - a double mischaracterization, I believe.
8 "Such is the foundation of perspectivism. It does not express a dependency on a predefined subject; on the contrary, whatever accedes to the point of view will be subject..." (Deleuze 1988: 27).
9 "Human beings see themselves as such; the Moon, the snakes, the jaguars and the Mother of Smallpox, however, see them as tapirs or peccaries, which they kill" (Baer 1994: 224).
10 If salmon look to salmon as humans to humans - and this is "animism" - salmon do not look human to humans (they look like salmon), and neither do humans to salmon (they look like spirits, or maybe bears; see Guedon 1984: 141) - and this is "perspectivism". Ultimately, then, animism and perspectivism may have a deeper relationship to totemism than Descola's model allows for.
11 The attribution of human-like consciousness and intentionality (to say nothing of human bodily form and cultural habits) to non-human beings has been indifferently denominated "anthropocentrism" or "anthropomorphism". However, these two labels can be taken to denote radically opposed cosmological outlooks. Western popular evolutionism is very anthropocentric, but not particularly anthropomorphic. On the other hand, "primitive animism" may be characterized as anthropomorphic, but it is definitely not anthropocentric: if sundry other beings besides humans are "human", then we humans are not a special lot.
12 "The point of view is located in the body, says Leibniz" (Deleuze 1988: 16).
13 The counterproof of the singularity of the spirit in our cosmologies lies in the fact that when we try to universalize it, we are obliged - now that supernature is out of bounds -to identify it with the structure and function of the brain. The spirit can only be universal (natural) if it is (in) the body.
14 The rarity of unequivocal examples of spirit possession in the complex of Amerindian shamanism may derive from the prevalence of the theme of bodily metamorphosis. The classical problem of the religious conversion of Amerindians could also be further illuminated from this angle; indigenous conceptions of "acculturation" seem to focus more on the incorporation and embodiment of Western bodily practices (food, clothing, interethnic sex) rather than on spiritual assimilation (language, religion etc.).
15 The traditional problem of Western mainstream epistemology is how to connect and universalize (individual substances are given, relations have to be made); the problem in Amazonia is how to separate and particularize (relations are given, substances must be defined). See Brightman (1993: 177-85) and Fienup-Riordan (1994: 46-50) - both inspired by Wagner's (1977) ideas about the "innate" and the "constructed" - on this contrast.
16 In Amazonian cannibalism, what is intended is precisely the incorporation of the subject-aspect of the enemy (who is accordingly hyper-subjectivized, in very much the same way as that described by Harrison [1993: 121] for Melanesian warfare), not its desubjectiv-ization as is the case with game animals. See Viveiros de Castro 1992a: 290-3; 1996: 98-102; Fausto 1997.
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
323
17 Peter Gow (pers. comm.) tells me that the Piro conceive of the act of putting on clothes as an animating of clothes. See also Goldman (1975: 183) on Kwakiutl masks: "Masks get 'excited' during Winter dances".
18 " 'Clothing' in this sense does not mean merely a body covering but also refers to the skill and ability to carry out certain tasks" (Riviere in Koelewijn 1987: 306).
19 Religions based on the cult of the ancestors seem to postulate the inverse: spiritual identity goes beyond the bodily barrier of death, the living and the dead are similar in so far as they manifest the same spirit. We would accordingly have superhuman ancestrally and spiritual possession on one side, animalization of the dead and bodily metamorphosis on the other.
20 This would be the true significance of the "deceptiveness of appearances" theme: appearances deceive because one is never certain whose point of view is dominant, that is, which world is in force when one interacts with other beings. The similarity of this idea to the familiar injunction not to "trust your senses" of Western epistemologies is, I fear, just another deceitful appearance.
21 As we have remarked, a good part of shamanistic work consists in de-subjectivizing animals, that is in transforming them into pure, natural bodies capable of being consumed without danger. In contrast, what defines spirits is precisely the fact that they are inedible; this transforms them into eaters par excellence, i.e. into anthropophagous beings. In this way, it is common for the great predators to be the preferred forms in which spirits manifest themselves, and it is understandable that game animals should see humans as spirits, that spirits and predator animals should see us as game animals and that animals taken to be inedible should be assimilated to spirits (Viveiros de Castro 1978). The scales of edibility of indigenous Amazonia (Hugh-Jones 1996) should therefore include spirits at their negative pole.
REFERENCES
Ärhem, K. 1993. Ecosofia makuna. In La selva humanizada: ecologia alternativa en el trópico
húmedo colombiano (ed.) F. Correa. Bogota: Instituto Colombiano de Antropológia, Fondo
FEN Colombia, Fondo Editorial CEREC. -1996. The cosmic food web: human-nature relatedness in the northwest Amazon. In
Nature and society: anthropological perspectives (eds) P. Descola &C G. Pálsson. London:
Routledge.
Baer, G. 1994. Cosmologťa y shamanismo de los Matsiguenga. Quito: Abya-Yala. Benveniste, E. 1966a. La nature des pronoms. In Problémes de linguistique generale. Paris: Gallimard.
-1996b. De la subjectivité dans le langage. In Problémes de linguistique generale. Paris:
Gallimard.
Boelscher, M. 1989. The curtain within: Haida social and mythical discourse. Vancouver:
Univ. of British Columbia Press. Brightman, R. 1993. Grateful prey: Rock Cree human-animal relationships. Berkeley: Univ. of
California Press.
Carneiro da Cunha, M.M. 1978. Os mortos e os outros. Säo Paulo: Hucitec.
Chaumeil, J.-P. 1983. Voir, savoir, pouvoir: le chamanisme chez les Yagua du nord-est
péruvien. Paris: Ecole des Hautes Etudes en Sciences Sociales. Crocker, J. C. 1985. Vital souls: Bororo cosmology, natural symbolism, and shamanism.
Tucson: Univ. of Arizona Press. Deleuze, G. 1988. Le pli: Leibniz et le baroque. Paris: Minuit.
Descola, P. 1986. La nature domestique: symbolisme et praxis dans ľécologie des Achuar. Paris: Maison des Sciences de l'Homme.
324
EDUARDO VIVEIROS DE CASTRO
Descola, P. 1992. Societies of nature and the nature of society. In Conceptualizing society (ed.) A. Kuper. London: Routledge.
Erikson, P. 1984. De l'apprivoisement ä l'approvisionnement: chasse, alliance et familiarisation en Amazonie amerindienne. Techn. Cult. 9, 105-40.
Fausto, C. 1997. A dialetica da predacäo e familiarizagäo entre os Parakanä da Amazonia oriental. Thesis, Museu Nacional, Univ. of Rio de Janeiro.
Fienup-Riordan, A. 1994. Boundaries and passages: rule and ritual in Yup'ik Eskimo oral tradition. Norman: Univ. of Oklahoma Press.
Goldman, I. 1975. The mouth of heaven: an introduction to Kwakiutl religious thought. New York: Wiley-Interscience.
Gow, P. 1989. The perverse child: desire in a native Amazonian subsistence economy. Man (N.S.) 24, 567-82.
-1991. Of mixed blood: kinship and history in Peruvian Amazonia. Oxford: Clarendon
Press.
Gregor, T. 1977. Mebinaku: the drama of daily life in a Brazilian Indian village. Chicago: Univ. of Chicago Press.
Grenand, P. 1980. Introduction a l'etude de l'univers wayüpi: ethno-ecologie des Indiens du
Haut-Oyapock (Guyane Frangaise). Paris: SELAF/CNRS. Guedon, M.-F. 1984. An introduction to the Tsimshian world view and its practitioners. In
The Tsimshian: images of the past, views for the present (ed.) M. Seguin. Vancouver: Univ.
of British Columbia Press. Guss, D. 1989. To weave and to sing: art, symbol and narrative in the South American rain
forest. Berkeley: Univ. of California Press. Hallowell, A. I. 1960. Ojibwa ontology, behavior, and world view. In Culture in history:
essays in honor of Faul Radin (ed.) S. Diamond. New York: Columbia Univ. Press. Hamayon, R. 1990. La chasse a l'äme: esquisse d'une theorie du chamanisme Sibirien.
Nanterre: Societe d'Ethnologie. Harrison, S. 1993. The mask of war: violence, ritual and the self in Melanesia. Manchester:
Univ. Press.
Howell, S. 1984. Society and cosmos: Chewong of peninsular Malaysia. Oxford: Univ. Press. -1996. Nature in culture or culture in nature? Chewong ideas of "humans" and other
species. In Nature and society: anthropological perspectives (eds) P. Descola & G. Pälsson.
London: Routledge.
Hugh-Jones, S. 1996. Bonnes raisons ou mauvaise conscience? De l'ambivalence de certains Amazoniens envers la consommation de viande. Terrain 26, 123—48.
Ingold, T. 1991. Becoming persons: consciousness and sociality in human evolution. Cult. Dyn. 4, 355-78.
-1994. Humanity and animality. In Companion encyclopedia of anthropology: humanity,
culture and social life (ed.) T. Ingold. London: Routledge. -1996. Hunting and gathering as ways of perceiving the environment. In Redefining
nature: ecology, culture and domestication (eds) R.F. Ellen & K. Fukui. London: Berg. Jara, F. 1996. El camino del Kumu: ecologia y ritual entre los Akuriyö de Surinam. Quito:
Abya-Yala.
Karim, W.-J. 1981. Ma'betisek concepts of living things. London: Athlone Press. Kensinger, K. 1995. How real people ought to live: the Cashinahua of eastern Peru. Prospect
Heights: Waveland Press. Koelewijn, C. with P. Riviere 1987. Oral literature of the Trio Indians of Surinam. Dordrecht:
Foris.
Latour, B. 1991. Nous n'avons jamais ete modernes. Paris: Editions La Dicouverte. Levi-Strauss, C. 1962a. he totemisme aujourd'hui. Paris: Presses Universitaires de France. -1962b. La pensee sauvage. Paris: Plön.
-1973 [1952]. Race et histoire. In his Anthropologie structurale deux. Paris: Plön.
COSMOLOGICAL DEIXIS AND AMERINDIAN PERSPECTIVISM
325
-1985. La potiere jalouse. Paris: Plon.
Lima, T. S. 1995. A parte do cauim: etnografia juruna. Thesis, Museu Nacional, Univ. of Rio de Janeiro.
-1996. O dois e seu multiplo: reflexoes sobre o perspectivismo em uma cosmologia tupi.
Mana 2:2, 21-47.
McCallum, C. 1996. The body that knows: from Cashinahua epistemology to a medical anthropology of lowland South America. Med. Antbrop. Q. 10:3, 1-26.
McDonnell, R. 1984. Symbolic orientations and systematic turmoil: centering on the Kaska symbol of dene. Canad. J. Antbrop. 4, 39-56.
Mentore, G. 1993. Tempering the social self: body adornment, vital substance, and knowledge among the Waiwai. /. Archaeol. Antbrop. 9, 22-34.
Nelson, R. 1983. Make prayers to the Raven. Chicago: Univ. of Chicago Press.
Osborn, A. 1990. Eat and be eaten: animals in U'wa (Tunebo) oral tradition. In Signifying animals: human meaning in the natural world (ed.) R. Willis. London: Unwin Hyman.
Overing, J. 1985. There is no end of evil: the guilty innocents and their fallible god. In The anthropology of evil (ed.) D. Parkin. London: Basil Blackwell.
-1986. Images of cannibalism, death and domination in a "non-violent" society. /. Soc. -
Amer. 72, 133-56.
Pollock, D. 1985. Personhood and illness among the Culina of western Brazil. Thesis, Univ. of Rochester.
Radcliffe-Brown, A. R. 1952 [1929]. The sociological theory of totemism. In Structure and
function in primitive society. London: Routledge & Kegan Paul. Reichel-Dolmatoff, G. 1976. Cosmology as ecological analysis: a view from the rain forest.
Man (N.S.) 12, 307-18. -1985. Tapir avoidance in the Colombian northwest Amazon. In Animal myths and
metaphors in South America (ed.) G. Urton. Salt Lake City: Univ. of Utah Press. Renard-Casevitz, F.-M. 1991. he banquet masque: une mythologie de I'etranger. Paris: Lierre
&C Coudrier.
Riviere, P. 1994. WYSINWYG in Amazonia. JASO 25, 255-62.
Saladin d'Anglure, B. 1990. Nanook, super-male; the polar bear in the imaginary space and
social time of the Inuit of the Canadian Arctic. In Signifying animals: human meaning in the
natural world (ed.) R. Willis. London: Unwin Hyman. Schwartzman, S. 1988. The Panara of the Xingu National park. Thesis, Univ. of Chicago. Scott, C. 1989. Knowledge construction among the Cree hunters: metaphors and literal
understanding./. Soc. Amer. 75, 193-208. Seeger, A. 1980. Corporacao e corporalidade: ideologia de concepcao e descendencia. In his
Os indios e nos. Rio de Janeiro: Campus. Tanner, A. 1979. Bringing home animals: religious ideology and mode of production of the
Mistassini Cree hunters. St John's Memorial Univ. of Newfoundland. Taylor, A.-C. 1993a. Des fantomes stupefiants: language et croyance dans la pensee achuar.
L'Homme 126-8, 33/2-4, 429-47. -1993b. Remembering to forget: identity, mourning and memory among the Jivaro. Man
(N.S.) 28, 653-78.
Townsley, G. 1993. Song paths: the ways and means of Yaminahua shamanic knowledge.
L'Homme 126-8, 33:2-4, 449-68. Turner, T. 1991. "We are parrots, twins are birds": play of tropes as operational structure. In
Beyond metaphor: the theory of tropes in anthropology (ed.) J. Fernandez. Stanford:
Stanford Univ. Press.
-1995. Social body and embodied subject: bodiliness, subjectivity, and sociality among the
Kayapo. Cult. Anthrop. 10, 143-70. Vilaca, A. 1992. Comendo como gente: formas do canibalismo Wari' (Pakaa-Nova). Rio de
Janeiro: Editora da UFRJ.
326
EDUARDO VIVEIROS DE CASTRO
Viveiros de Castro, E. 1978. Alguns aspectos do pensamento yawalpiti (Alto Xingu): classi-
ficagöes e transŕbrmacôes. Bol. Mus. nac. 26, 1-41.
--1979. A fabricacäo do corpo na sociedade xinguana. Bol. Mus. nac. 32, 2-19.
-1992a. Trom the enemy's point of view: humanity and divinity in an Amazonian society.
Chicago: Univ. of Chicago Press. -1992b. Apresentacäo to A. Vilaca. In Comendo como gente: formas do canibalismo
Wari'. Rio de Janeiro: Editoral da UFRJ. -1993. Alguns aspectos da afinidade no dravidianato amazônico. In Amazonia: etnológia
e história indígena (eds) E. Viveiros de Castro & M. Carneiro da Cunha. Säo Paulo: Núcleo
de História Indígena e do Indigenismo (USP/FAPESP). -1996. Le meurtrier et son double chez les Araweté: un exemple de fusion rituelle. Syst.
Pens. Afr. Noire 14, 77-104. Wagner, R. 1977. Scientific and indigenous Papuan conceptualizations of the innate: a
semiotic critique of the ecological perspective. In Subsistence and survival: rural ecology
in the Pacific (eds) T. Bayliss-Smith & R.G. Feachem. London: Academic Press. Weiss, G. 1969. The cosmology of the Campa Indians of eastern Peru. Thesis, Univ. of
Michigan.
-1972. Campa cosmology. Ethnology 11, 157-72.