The category of the person Anthropology, philosophy, history Edited by Michael Carrithers Steven Collins Steven Lukes •::: CAMBRIDGE UNIVERSITY PRESS VI Contents 7. Self: public, private. Some African representations Godfrey Lienhardt 141 8. Between the earth and heaven: conceptions of the self in China Mark Elvin 156 Preface 9. Purity and power among the Brahmans of Kashmir Alexis Sanderson 1 90 10. Of masks and men Martin Hollis 217 11. An alternative social history of the self Michael Carrithers 234 12. The person Charles Taylor 257 Conclusion Steven Lukes 282 'A comprehensive knowledge of the facts is only possible through the collaboration of numerous specialists . . . . Only mutual supervision and pitiless criticism can yield firm results.' Behind these dry words lies the passionate communal spirit with which Marcel Mauss and his colleagues of the Annee sociologique school sought to forge a new understanding of Bibliography Index 302 304 human life. Of all their creations one of the most remarkable was Mauss's last essay, published in 1938, on the notion of self or person. The basic lines of argument had already been sketched by Durkheim forty years earlier. Mauss proposes that our seemingly natural and self-evident concep­ tions of our selves, our persons, are in truth artefacts of a long and varied social history stretching back, at least in principle, to the earliest human communities. Other societies have held very .different notions of the self, and each society's notion is intimately connected with its form of social organization. The notion least like ours, that of the 'character' or 'role' (personnage), Mauss finds in ethnographic materials from North America and Australia. In such societies each role was in daily life the locus of different rights, duties, titles and kinship names within the clan, and was on ceremonial occasions vividly exemplified by different masks or body paint. No general rules applied to 'roles' as such apart from the clan, nor were they thought to bear an inner conscience. A revolution then occurred in ancient Rome, when the 'role' - the 'mask' or persona - was made the locus of general rights and duties as a legal 'person' and a citizen of the state. To this more abstract 'person' was later added the notion of an inner conscience and inner life, chiefly viii Preface through Christianity. And this notion of person, now bearing both a conscience and a civic identity, became the foundation of modern political, social and legal institutions. This sketch does little justice to Mauss's rich argument, but will arm the reader to face its complexity. Perhaps because of this complexity subsequent scholars have conducted little 'mutual supervision' of it. We have attempted in this volume to re-create, albeit under very different circumstances, Mauss's communal enterprise. Each of the authors was asked to address himself to Mauss's essay, which is translated here. Most of them, whether implicitly or explicitly, have addressed each other as well. And almost all of them attended and gave a first version of their paper at a series of seminars held in May and June 1 980, in Wolfson College, Ox­ Contributors ford, to whose Fellows we are deeply grateful for hospitality and financial assistance at that time. Mauss's essay was given in French as the Huxley Memorial Lecture for 1938, and appeared under the title 'Une Categorie de l'Esprit Humain: La Notion de Personne, Celle de "Moi" , in the Journal of the Royal Anthropological Institute 68 ( 1 938). (It was reprinted in Mauss's Sociologie et anthropologie [Paris, 1 950] - with some printing errors.) A translation by Ben Brewster was published in Marcel Mauss: Sociology and Psychology (London, 1 979). The translation by W.D. Halls was commissioned for this volume with the permission of Routledge and Kegan Paul PLe. In all important passages French terms are given in parentheses in the text. The following are usual equivalents: moi - (the) self soi - (one's) self personne - person personnalite - personality personnage - role, character. The quotation from Mauss that prefaces this volume is taken from the autobiographical sketch presented as part of his application for member­ ship of the College de France in 1 930. It appeared in Revue Frant;aise de sociologie 20 ( 1 ) : 1 979. G. Lienhardt's paper originally appeared in the Journal of the Anthropological Society of Oxford, 1 980. L. Dumont's paper appeared in Religion 12: 1 982, 1-27, and is reproduced here with the consent of the author, the editor of Religion, and the publishers, © Academic Press, Inc. (London) Ltd. N.J. Allen is Lecturer in the Social Anthropology of South Asia at Oxford University. Michael Carrithers is Lecturer in Social Anthropology at Durham Uni­ versity. Steven Collins is Lecturer in the Study of Religions at Bristol University. Louis Dumont is Directeur d'Etudes at the Ecole des Hautes Etudes en Sciences Sociales, Paris. Mark Elvin is Lecturer in Chinese History at Oxford University. Martin Hollis is Professor of Philosophy at the University of East Anglia. J.S. La Fontaine is Professor of Social Anthropology at the London School of Economics. Godfrey Lienhardt is Reader in Social Anthropology at Oxford Univer­ sity. Steven Lukes is Fellow and Tutor in Politics and Sociology at Balliol College, Oxford University. A. Momigliano is Alexander White Professor in the Humanities at Chi­ cago University, Professor Emeritus of Ancient History at London University. Alexis Sanderson is Lecturer in Sanskrit at Oxford University. Charles Taylor is Professor of Political Science at McGill University. 1 A category of the human mind: the notion of person; the notion of self Marcel Mauss (translated by W. D. Halls) I: The subject1: the 'person' (personne) My audience and readers will have to show great indulgence, for my subject is really enormous, and in these fifty-five minutes I shall be able only to give you some idea of how to treat it. It deals with nothing less than how to explain to you the way in which one of the categories of the human mind - one of those ideas we believe to be innate - originated and slowly developed over many centuries and through numerous vicis­ situdes, so that even today it is still imprecise, delicate and fragile, one requiring further elaboration. This is the idea of 'person' (personne), the idea of 'self' (moi). Each one of us finds it natural, clearly determined in the depths of his consciousness, completely furnished with the fundaments of the morality which flows from it. For this simplistic view of its history and present value we must substitute a more precise view. A note on the principle underlying these kinds of research In so doing you will see an example - one that is perhaps not up to what you expected - of the work of the French school of sociology. We have concentrated most especially on the social history of the categories of the human mind. We attempt to explain them one by one, using very simply, and as a temporary expedient, the list of Aristotelian categories 2 as our point of departure. We describe particular forms of them in certain civilisations and, by means of this comparison, try to discover in what consists their unstable nature, and their reasons for being as they are. It was in this way that, by developing the notion of mana, Hubert and I believed we had found not only the archaic basis for magic, but also the very 2 M. Mauss general, and probably very primitive, form of the notion of cause. It was in this way that Hubert described certain features of the notion of time. Likewise our much regretted colleague, friend and pupil, Czarnowki, began - but, alas, never finished - his theory of the 'parcelling out of extension', in other words, of one of the features and certain aspects of the notion of space. Likewise also, my uncle and teacher, Durkheim, has dealt with the notion of the whole, after we had examined together the notion of genus. I have been preparing for many years studies on the notion of substance. Of these I have published only a very recondite extract which is not worth reading in its present form. I will mention to you also the numerous times that Lucien Levy-Bruhl has touched upon these questions in those works of his which deal with the primitive mentality, especially, as regards our subject, what he has termed 'the primitive mind' (l 'arne prirnitive). He, however, does not concentrate on the study of each special category, not even on the one we are going to study. But rather, in reviewing all of them, including the category of 'self', does he seek particularly to ascertain what element of the 'pre-logical' is contained in this study of the mentality of peoples, in relation to anthropology and ethnology rather than history. If you will permit me, let us proceed more methodically and restrict ourselves to the study of one single category, that of the 'self' (rnoi). This will be amply sufficient. In the present short space of time, I shall conduct you, with some daring and at inordinate speed, across the world and through time, guiding you from Australia to our European societies, from extremely ancient history to that of our own times. More extensive research studies could be undertaken, each one of which could be gone into much more deeply, but I can only claim to show you how such research might be organised. What I intend to do is to provide you with a summary catalogue of the forms that the notion has assumed at various times and in various places, and to show you how it has ended up by taking on flesh and blood, substance and form, an anatomical structure, right up to modern times, when at last it has become clear and precise in our civilisations (in our European ones, almost in our lifetime), but not yet in all of them. I can only rough out the beginnings of the sketch or the clay model. I am still far from having finished the whole block or carved the finished portrait. Thus I shall not discuss the linguistic problem which, for. the sake of completeness, should indeed be tackled. In no way do I maintain that there has ever been a tribe, a language, in which the term 'I', 'me' (je, A category of the hurnan rnind 3 rnoi) (you will note that we still decline it with two words) has never existed, or that it has not expressed something clearly represented. This is far from the case: as well as possessing the pronoun, a very large num­ ber of languages are conspicuous for their use of many 'positional' suffixes, which deal for the most part with the relationships existing in time and space between the speaker (the subject) and the object about which he is speaking. Here the 'self' (rnoi) is everywhere present, but is not expressed by 'me' (rnoi) or 'I' (je). However, in this vast domain of lan­ guages my scholarship is only mediocre. My investigation will concern solely law and morality. Nor shall I speak to you of psychology, any more than I shall of linguistics. I shall leave aside everything which relates to the 'self' (rnoi), the conscious personality as such. Let me merely say that it is plain, particularly to us, that there has never existed a human being who has not been aware, not only of his body, but also at the same time of his individuality, both spiritual and physical. The psychology of this awareness has made immense strides over the last century, for almost a hundred years. All / neurologists, French, English and German, among them my teacher Ribot, our esteemed colleague Head, and others, have amassed a great deal of knowledge about this subject and the way this particular awareness is formed, functions, deteriorates, deviates and dissolves, and about the considerable part it plays. My subject is entirely different, and independent of this. It is one relating to social history. Over the centuries, in numerous societies, how has it slowly evolved - not the sense of 'self' (rnoi) - but the notion or con­ cept that men in different ages have formed of it? What I wish to show you is the succession of forms that this concept has taken on in the life of men in different societies, according to their systems of law, religion, customs, social structures and mentality. One thing may alert you to the drift of my exposition: I shall show you how recent is the word 'self' (rnoi), used philosophically; how recent "the category of 'self' " (rnoi), "the cult of the 'self' " (rnoi) (its aberration) ; and how recent even "the respect of 'self' " (rnoi), in particular the respect of others (its normal state). 4 M. Mauss of logic, we will make an excursion into that kind of museum of facts (I dislike the word 'survivals', when it is used for institutions still active and proliferating) which ethnography affords us. II: The 'role' (personnage), and the place of the 'person' (personne) The Pueblos Let us start with the fact that has been the point of departure for all this research. I borrow it from the Pueblo Indians, the Zuni - or more accurately from those of the Pueblo of Zuni, so admirably studied by Frank Hamilton Cushing (who was fully initiated into the Pueblo), and by Mathilda Cox Stevenson and her husband for a great number of years. Their work has been criticised, but I believe it to be reliable and, in any case, unique. It is true that there is nothing 'very primitive' about things. The 'Cities of Cibola' were once converted to Christianity and have preserved their baptismal registers. Yet, at the same time they have practised their ancient laws and religions - almost in the 'aboriginal state', if one may say so: this was roughly that of their predecessors, the cliff dwellers and the inhabitants of the 'mesa' as far as Mexico. In their material civilisation and social constitution they were, and have remained, very comparable to the Mexicans and to the most civilised Indians of the two Americas. 'Mexico, that Pueblo', writes admirably the great L. H. Morgan, who was so unfairly treated, and yet the founder of our disciplines.3 The document below is by Frank Hamilton Cushing,4 an author much criticised, even by his colleagues at the Bureau of American Ethnology. Yet, knowing his published work and having considered very carefully what has appeared on the Zuni and the Pueblo in general, strengthened also by what I believe I know about a large number of American societies, I persist in considering him one of the best portrayers of societies of all time. If you will allow me, I will pass over everything concerning the orientation and distribution of the characters (personnages) in the ritual, al­ though this has very great importance, to which we have already drawn attention elsewhere. But I cannot omit two points: The existence of a limited number of forenames in each clan; and the definition of the exact role played by each one in the 'cast-list' of the clan, and expressed by that name. In each clan is to be found a set of names called the names of childhood. These names are more of titles than of cognomens. They are determined A category of the human mind 5 upon by sociologic and divinistic modes, and are bestowed in childhood as the 'verity names' or titles of the children to whom given. But this body of names relating to any one totem - for instance, to one of the beast totems - will not be the name of the totem beast itself, but will be names both of the totem in its various conditions and of various parts of the totem, or of its functions, or of its attributes, actual or mythical. Now these parts of functions, or attributes of the parts or functions, are subdivided also in a six-fold manner, so that the name relating to one member of the totem j for example, like the right arm or leg of the animal thereof - would correspond to the north, and would be the first in honor in a clan (not itself of the northern group) ; then the name relating to another member - say to the left leg or arm and its powers, etc. - would pertain to the west and would be second in honor; and another member - say the right foot - to the south and would be third in honor; and of another member - say the left foot - to the east and would be fourth in honor; to another - say the head - to the upper regions and would be fifth in honor; and another - say the tail - to the lower region and would be sixth in honor; while the heart or the navel and center of the being would be first as well as last in honor. The studies of Major Powell among the Maskoki and other tribes have made it very clear that kinship terms, so called, among other Indian tribes (and the rule will apply no less or perhaps even more strictly to the Zunis) are rather devices for determining relative rank or authority as signified by relative age, as elder or younger, of the person addressed or spoken of by the term of relationship. So that it is quite impossible for a Zuni speaking to another to say simply brother; it is always necessary to say elder brother or younger brother, by which the speaker himself affirms his relative age or rank; also it is customary for one clansman to address another clansman by the same kinship name of brother-elder or brotheryounger, uncle or nephew, etc.; but according as the clan of the one addressed ranks higher or lower than the clan of the one using the term of address, the word-symbol for elder or younger relationship must be used. With such a system of arrangement as all this may be seen to be, with such a facile device for symbolizing the arrangement (not only according to number of the regions and their subdivisions in their relative succession and the succession of their elements and seasons, but also in colours attributed to them, etc.) and, finally, with such an arrangement of names correspondingly classified and of terms of relationship significant of rank rather than of consanguinal connection, mistake in the order of a ceremonial, a procession or a council is simply impossible, and the people employing such devices may be said to have written and to be writing their statutes and laws in all their daily relationships and utterances. Thus, on the one hand, the clan is conceived of as being made up of a certain number of persons, in reality of 'characters' (personnages). On the other hand, the role of all of them is really to act out, each insofar as it concerns him, the prefigured totality of the life of the clan. 6 M. Mauss So much for persons and the clan. The 'fraternities' are even more complicated. Among the Pueblo of Zuni, and clearly among the others too - the Pueblos of Sia and Tusayan, in the Hopi tribe, those of Walpi and Mishongnovi - the names do not merely correspond to the organisation of the clan, its processions and ceremonies, whether private or public. They correspond principally to ranks in the fraternities, in what the original terminology of Powell and the Bureau of American Ethnology designated 'Fraternities', viz., 'Secret Societies', which we might very exactly compare to the Colleges of the Roman Religion. There were preparations in secret, and numerous solemn rituals reserved for the Society of the Men (Kaka or Koko, Koyemshi, etc.), but also public demonstrations - almost theatrical performances - and, especially at Zuni, and above all among the Hopi, mask dances, particularly those of the Katchina. These were visits of spirits, represented by their delegates upon earth, who bore their titles. All this, which has now become a spectacle for tourists, was still very much alive less than fifty years ago, and is so even today. Miss B. Freire Marecco (now Mrs Aitken) and Mrs E. Clew Parsons continue to add to our knowledge and to corroborate it. Moreover, let us add that these lives of individuals, the driving force of clans and of the societies superimposed upon them, not only sustain the life of things and of the gods, but the 'propriety' of things. They not only sustain the life of men, both here and in the after-life, but also the rebirth of individuals (men), sole heirs of those that bear their forenames (the reincarnation of women is a completely different matter). Thus, in short, you will understand that with the Pueblo we already see a notion of the 'person' (personne) or individual, absorbed in his clan, but already detached from it in the ceremonial by the mask, his title, his rank, his role, his survival and his reappearance on earth in one of his descendants endowed with the same status, forenames, titles, rights and functions. The American North-West If I had time, another group of American tribes would well deserve in this study a detailed analysis of the same facts. These are the tribes of the American North-West - and it is to the great credit of your Royal Anthropological Institute and the British Association to have instigated a complete analysis of their institutions. This was begun by Dawson, the great geologist, and so magnificently continued, if not completed, by the great works of Boas and his Indian assistants, Hunt and Tate, and by those of Sapir, Swanton and Barbeau, etc. A category of the human mind 7 Here also is posed, in different terms but ones identical in nature and function, the same problem - that of the name, the social position and the legal and religious 'birthright' of every free man, and even more so, of every noble and prince. I shall take as a starting-point the best known of these important societies, the Kwakiutl, and confine myself only to some broad facts. One word of caution: just as with the Pueblos, so also with the Indians of the North-West, we must not think of anything in any way primitive. Firstly, one section of these Indians, in fact those in the North, the Tlingit and Haida, speak languages which according to Sapir are tonal languages related to those derived from a root which it has been agreed to call proto-Sino-Tibetan-Burman. And even, if I may tell you of one of my impressions as an ethnographer - if not an 'armchair' one, at least a 'museum' one - I have a very strong recollection of a display exhibit concerning the Kwakiutl, the work of the esteemed Putnam, one of the founders of the ethnological section of the American Museum of Natural History. It was a very large ceremonial boat, with figures life-size, with all their religious and legal paraphernalia, which represented the Hamatse, the cannibal princes, arriving from the sea to carry out a ritual - doubtless a marriage. With their very rich robes, their crowns of red cedar bark, their crewmen less sumptuously attired but nonetheless magnificent, they gave me an exact impression of what, for example, Northern China in the very remote past might have looked like. I believe that this boat, with its somewhat romanticised representations, is no longer exhibited; it is no longer the fashion in our ethnographic museums. No matter, for at least this one had had its effect upon me. Even the Indian faces vividly recalled to me the faces of the 'Paleo-Asiatics' (so called because we do not know under what to classify their languages) . And, from this point in civilisation and of settlement, we have still to reckon with many long and varied developments, revolutions and new formations that our esteemed colleague, Franz Boas, perhaps with undue haste, is attempting to trace back. The fact remains that all these Indians, and in particular the Kwakiutl, installed in their settlements a whole social and religious system where, in a vast exchange of rights, goods and services, property, dances, ceremonies, privileges and ranks, persons as well as groups give satisfaction to one another. We see very clearly how, from classes and clans, 'human persons' adjust to one another and how, from these, the gestures of the actors in a drama fit together. Here all the actors are theoretically the sum total of all free men. But this time the drama is more than an aes- 8 M. Mauss thetic performance. It is religious, and at the same time it is cosmic, mythological, social and personal. Firstly, as with the Zuni, every individual in each clan has a name, even two names, for each season, one profane (summer) (WiXsa), and one sacred (winter) (LaXsa). These names are distributed between the various families, the 'Secret Societies' and the clans cooperating in the rituals, occasions when chiefs and families confront each other in innumerable and interminable potlatch, 5 about which I have attempted else­ where to give some idea. Each clan has two complete sets of its proper names, or rather its forenames, the one commonly known, the other secret, but which itself is not simple. This is because the forename, actually of the noble, changes with his age and the functions he fulfils as a consequence of that age.6 As is said in an oration, made, it is true, about the clan of the Eagles, i.e. about a kind of privileged group among privileged clans: For that they do not change their names starts from (the time) when long ago II 0' maxt!iilaLe', the ancestor of the numaym G ig i1gam of the I Q !omoya'ye, made the seats of the Eagles; and those went down to the I numayms. And the name-keeper Wiltse'stala says, I 'Now our chiefs have been given everything, and I will go right down (according to the order of rank).' I Thus he says, when he gives out the property; for I will just name the names II of one of the head chiefs of the numayms of the I Kwakiutl tribes. They never change their names from the beginning, I when the first human beings existed in the world; for names can not go out I of the family of the head chiefs of the numayms, only to the eldest one I of the children of the head chief. IF What is at stake in all this is thus more than the prestige and the authority of the chief and the clan. It is the very existence of both of these and of the ancestors reincarnated in their rightful successors, who live again in the bodies of those who bear their names, whose perpetuation is assured by the ritual in each of its phases. The perpetuation of things and spirits is only guaranteed by the perpetuating of the names of individuals, of persons. These last only act in their titular capacity and, conversely, are responsible for their whole clan, their families and their tribes. For instance, from conquest in war are acquired: a rank, a power, a religious and aesthetic function, dancing and demoniacal possession, paraphernalia, and copper objects in the form of buckler shields - real crown shapes in copper, important currency for present and future potlatch: it suffices to kill the one possessing them, or to seize from him one of the trappings of ritual, robes or masks, so as to inherit his names, his goods, his obi i- A category of the human mind 9 gations, his ancestors, his 'person' (personne), in the fullest sense of the word.8 In this way ranks, goods, personal rights, and things, as well as their particular spirit, are acquired. This huge masquerade in its entirety, this whole drama, this compli­ cated ballet of ecstatic states, concerns as much the past as the future, becomes a test for its performer, and proof of the presence within him of the naualaku, an element of an impersonal force, or of the ancestor, or of the personal god, in any case of the superhuman power, spiritual and ultimate. The potlatch of victory, of the copper won by conquest, corre­ spond to the impeccable dance, to a successful state of possession. There is no time left to develop all these subjects. Almost from an anecdotal viewpoint, I would like to draw your attention to an institution, an object commonly found from the Nootka right up to the Tlingit of North Alaska. This is the use of those remarkable shutter masks, which are double and even triple, which open up to reveal the two or three creatures (totems placed one upon the other) personified by the wearer of the mask.9 You can see some very fine examples of them in the British Museum. And all those celebrated totem poles, those soapstone pipes, etc., all those objects which have become rubbishy goods designed for the tourists brought there by train or on cruises - all these may be analysed in the same way. A pipe I believe to be Haida in origin, one to which I have hardly given any attention, in point of fact represents a young initiate in his pointed headdress, presented by his spirit father, likewise behatted, bearing the grampus. Beneath the one initiated, to whom they are subordinate in descending order: a frog - doubtless his mother - and a crow, doubtless his maternal grandfather. We shall not deal with the very important case of change of name during a lifetime - particularly that of a noble. It would entail expounding a whole succession of curious facts regarding substitution: the son, a minor, is temporarily represented by his father, who assumes provisionally the spirit of the deceased grandfather. Here also we would need to set out a complete proof of the presence among the Kwakiutl of dual uterine and male descent, and of the system of alternate and displaced generations. Moreover, it is very remarkable that among the Kwakiutl (and their nearest kin, the Heiltsuk, the Bellacoola, etc.) every stage of life is named, personified by a fresh name, a fresh title, whether as a child, an adolescent or an adult, both male and female. Thus one may possess a name as a warrior (naturally this does not apply to women), as a prince or princess, as a chief or a female chieftain. There is a name for the feast that 10 M. Mauss men and women give, and for the particular ceremonial that belongs to them, for their age of retirement, their name in the society of seals (those retired: no states of ecstacy or possession, no responsibilities, no gains, save those arising from past memories) . Finally is named their 'secret society', in which they are protagonists (a bear - frequent among women, who are represented in it by their menfolk or their sons - wolves, Hamatse [cannibals], etc.). Names are also given to: the chief's house, with its roofs, posts, doors, ornamentation, beams, openings, double-headed and double-faced snake, the ceremonial boat, the dogs. To the lists set out in the Ethnology of the KwakiutllO it must be added that the dishes, the forks, the copper objects, everything is emblazoned, endowed with life, forming part of the persona of the owner and of the familia, of the res of his clan. We have singled out the Kwakiutl, and in general the peoples of the North-West, because they really do represent the extremes, an excessiveness which allows us better to perceive the facts than in those places where, although no less essential, they still remain small-scale and involuted. Yet we must understand that a large part of the Americans of the prairies, in particular the Sioux, possess institutions of this kind. Thus the Winnebago, who have been studied by our colleague Radin, have in point of fact these successions of forenames, which are determined by clans and families, who distribute them according to a certain order, but always following precisely a kind of logical distribution of attributes or powers and natures,l 1 founded upon the myth of the origin of the clan, and legitimating the right of some person or another to assume the role. Below is an example of this origin of the names of individuals which Radin gives in detail in his model autobiography of Crashing Thunder: Now in our clan whenever a child was to be named it was my father who did it. That right he now transmitted to my brother. Earthmaker, in the beginning, sent four men from above and when they came to this earth everything that happened to them was utilized in making proper names. This is what our father told us. As they had come from above so from that fact has originated a name Comes-from-above; and since they came like spirits we have a name Spirit-man. When they came, there was a drizzling rain and hence the names Walking-in-mist, Comesin-mist, Drizzling-rain. It is said that when they came to Within-lake they alighted upon a small shrub and hence the name Bends-the-shrub; and since they alighted on an oak tree, the name Oak-tree. Since our ancestors came with the thunderbirds we have a name Thunderbird and since these are the animals who cause thunder, we have the name He-who-thunders. Similarly we have Walks-with-a-mighty-tread, Shakes-the-earth-down-withhis-face, Comes-with-wind-and-hail, Flashes-in-every-direction, Only-a-flash- A category of the human mind 1 1 of-lightning, Streak-of-lightning, Walks-in-the-clouds, He-who-has-longwings, Strikes-the-tree. Now the thunderbirds come with terrible thunder-crashes. Everything on earth, animals, plants everything, is deluged with rain. Terrible thundercrashes resound everywhere. From all this a name is derived and that is my name - Crashing-ThunderY Each one of the names of the thunder birds which divide up the different elements of the thunder totem is that of ancestors who are perpetually reincarnated. (We even have a story of two reincarnations.) 1 3 The men who reincarnate them are intermediaries between the totemic animal and the protecting spirit, and the things emblazoned and the rites of the clan or of the great 'medicines'. All these names and bequeathals of 'roles' (personna lites) are determined by revelations whose limits, indicated by his grandmother or the elders, are known to the beneficiary beforehand. We discover, if not the same facts, at least the same kind of facts, almost everywhere in America. We could continue this exposition for the world of the Iroquois and the Algonquin, etc. Australia It is preferable to revert for a moment to more summary and more primitive facts. Two or three items concern Australia. Here also the clan is in no way conceived to be entirely reduced to an impersonal, collective being, the totem, represented by the animal species and not by individuals - on the one hand men, on the other, animals. 1 4 Under its human aspect it is the fruit of the reincarnation of spirits that have migrated and are perpetually being reborn in the clan. (This is true for the Arunta, the Loritja and the Kakadu, etc.) Even among the Arunta and the Loritja, these spirits are reincarnated with very great precision at the third generation (grandfather-grandson) and at the fifth, where grandfather and great-great-grandson are homonyms. Here again it is the fruit of uterine descent crossed with male descent. We can, for example, study in the distribution of names by individuals, clans and exact matrimonial category (eight Arunta categories) the relationship of these names to the eternal ancestors, to the ratapa, in the form they take at the mo­ ment of conception, in the foetus and in the children that they bring to the light of day, and between the names of these ratapa and those of adults (which are, in particular, those of the functions fulfilled at clan and tribal ceremonies). 1 5 The art underlying all these kinds of distribution is not only to arrive at religion, but also to define the position of the individual in the rights he enjoys and his place in the tribe, as in its rites.