PER CHU AE Lait des 7 Et Ü ~, el 4 * RAR is PEU HTC Srey oie » i + hts HE A LNH ae + 5 ee oo 5 Bs i eet ARTS we “ ate FF Ses i @ ANNE HA A er AAU Ver J ae, ae ole LR : NV (aie AE AS tye ve DAS; > ie HOW NATIVES THINK aye RL EX fi , Lue 0M ‘f 0 MAT i, EME a Wb ie © he : AU ae RO Bid RNA Max: \ IRL MUR ARE vy ah 2 FA 14! VAT (fey var, AS k Li HOW NATIVES THINK (LES FONCTIONS MENTALES DANS LES SOCIÉTÉS INFERIEURES) BY PROF. LUCIEN LEVY-BRUHL OF THE SORBONNE Authorized Translation BY LILIAN, A. CLARE NEW YORK ALFRED : A - KNOPF (All vights reserved) Printed in Great Britain by Unwin Brothers, Lid., Woking AUTHOR’S FOREWORD La Menrautré Primitive, which appeared in English in 1923, and Les Fonctions Mentales dans les Sociétés Inférieures, published in the same language to-day, really make one and the same work in two volumes. It was due to the accident of circumstances alone that the second was translated into English before the first. The volumes can certainly be read separately, but, as the critics have remarked, the essential theories I maintain and the principles upon which I rely are all contained in the earlier one, now available to the English reader, and undoubtedly the two volumes will be best under- stood if read in the order in which they were written. The fact that at this date, after having been submitted to the tests of time and experience, the theories maintained in Les Fonctions Mentales still seem to me to hold good, would possibly not be sufficient guarantee of their truth—nor even the fact that my present researches in the same subject appear to confirm them. One is always fearful of being the dupe of flattering illusions concerning the cause he is pleading. Happily in this case I can appeal to other witnesses, less likely to be partial. Among those whose official functions, or vocations, bring them into constant relations with primitives, among the administrators and missionaries who share their everyday life, there are many who, hitherto strangers to me, have been good enough to write and tell me that these volumes have been of service tothem. They say that the reading of Les Fonctions Mentales and La Mentalité Primitive helps them to compre- hend much that had appeared both unintelligible and ridiculous in the way natives reason, and also in their customs, and that their relations with them are accordingly facilitated and improved. 6 HOW NATIVES THINK In this I find the most valuable reward and the best en- couragement that I can possibly receive, and this, too, emboldens me to present to-day to the English-reading public the translation of Les Fonctions Mentales, in the hope that it may be as well received as Primitive Mentality has been. I also desire to express my thanks to Mrs. Clare for her excellent translation and for the care with which she has compiled the index. L. LEVY-BRUHL. PARIS, October 1925. CONTENTS ERS OP a AL SENG a aOR OA LI RG A INTRODUCTION . ; : : : ‘ ; À i : : I. A brief definition of collective representations. The object of this book. Its bearing upon the work of sociologists and upon present-day psychology. II. Earlier theories. Comte and his teaching with regard to the higher mental functions. The mind of the primitive, from the point of view of ethnography, anthropology, and that of the English school in particular. ‘ BIT) A postulate granted by all: the human mind is at all times and everywhere true to type. The “ animism ”’ of Tylor and Frazer and their school implies acceptance of this postulate. IV. A critical examination of the methods of this school. Examples drawn from Frazer’s work. 1. It leads to probabilities merely. 2. It disregards the social nature of the phenomena to be ex- plained. The influence of this school upon associationistic psy- chology, and Herbert Spencer’s philosophy of evolution. V. Types of mentality differ among themselves as do social types. The paucity of documentary evidence, either contemporaneous or earlier, in determining these differences. To what extent, and by what methods, can this deficiency be made good ? PART I CHAPTER 1 COLLECTIVE REPRESENTATIONS IN PRIMITIVES’ PERCEPTIONS AND THE MYSTICAL CHARACTER OF SUCH . . . . I, Emotional and motor elements inherent in the collective representations of primitives. Mystic properties attributed to animals, plants, part of the human body, inanimate objects, the soil, the shape of manufactured articles. The persistence of this form and the danger of making any change in it whatever. Primitives do not perceive things as we do. Our traditional problems have to be reversed. PAGE 35 HOW NATIVES THINK PAGE Predominance of mystic elements in primitives’ perceptions; their ideas of images and portraits, names, shadows, dreams. q Perceptions peculiar to certain privileged persons. The primitive’s impermeability to experience. The reality he perceives at once natural and supernatural. The omnipresence of spirits. CHAPTER II THE LAW OF PARTICIPATION . ; . ‘ r ; f i OG a III. IV. PET à The difficulty of reconstructing the connections uniting the collective representations of primitives. Examples of connec- tions which are foreign to our thought, and cannot be explained either by a simple association of ideas, or by a childish appli- cation of the principle of causality. The law of participation. An approximate statement of this law. Primitive thinking is both mystic and prelogical, as proved by the collective representations respecting souls. Tylor’s animism ; his theory examined. The concept of “ soul ” a comparatively recent one. The law of participation determines the social group’s idea of itself and the human and animal groups around it. It is implied in the intichiuma ceremonies of the Aruntas, in their idea of mythical beings in the shape of animals, and in general in their group idea of the relations between human beings and animals. This law is implied in the primitive conception of the influences which persons and things exert upon each other (by contact, transference, contamination, sympathy, possession, etc.). The representation of personified spirits does not appear to be a primitive one. CHAPTER III THE FUNCTIONING OF PRELOGICAL MENTALITY . . . 110% Nu II. Logical and prelogical elements co-existent in prelogical men- tality. Such a mentality is essentially synthetic. How memory functions in prelogical mentality; its develop- ment. The sense of locality, and of direction. Abstraction, and the concepts peculiar to prelogical mentality. Generalization peculiar to prelogical mentality. Primitive classification. The concepts of mana, wakan, orenda, etc., and other collective representations of a similar nature involve the law of participation. » CONTENTS PAs Le CHAPTER IV THE MENTALITY OF PRIMITIVES IN RELATION TO THE LANGUAGES THEY SPEAK . . . . . ° . . . . I. Number in the language of primitive peoples: the dual, trial, and plural. II. These languages strive to express the detailed form, position, and movements of persons and things. III. An example taken from the language of Klamath Indians. The immense number of suffixes and prefixes, and their uses, IV. The custom of talking by gesture common in many undeveloped races. The similarity of sign language and vocal language. si The Lautbilder. V. The wealth or poverty of the vocabulary of primitive languages corresponds with the primitives” methods of abstraction and .generalization. VI. The mystic power of words. Languages peculiar to special circumstances or certain classes of people. Sacred languages. CHAPTER V PRELOGICAL MENTALITY IN RELATION TO NUMERATION . . I. Methods by which prelogical mentality supplements any defic- iency in numerals when it does not count beyond two or three. II. Number not at first distinct from the objects enumerated. Sometimes the number-series varies with the class of objects to be counted. ‘“‘ Classifiers.’” The same word may indicate several numbers in succession. III. There is no basis upon which primitives establish their numerical system, nor is there a natural one. The numerical system depends upon the collective representations of the social group, and the participations which these representations involve. IV. The mystic power inherent in numbers. Critical examination of Usener’s theory. The mystic value of the numbers four, five, six, etc. Mystic numbers in the Vedic texts. Reply to an objection, PAGE 139 181 IO HOW NATIVES THINK PARDEE CHAPTER VI INSTITUTIONS IN WHICH COLLECTIVE REPRESENTATIONS GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (1) I. Hunting. Mystic influences exercised upon the quarry (such as dances, fasts, incantations, etc.) in order to summon it, to paralyse or blind it. Mystic influences exerted upon the hunter, and prohibitions imposed upon him and his. Ceremonies calcu- lated to appease the spirit of the slaughtered game. II. Fishing. Mystic influences, similar to the above, exercised to ensure the presence of fish and to cause it to enter the nets. Mystic arts practised upon the fisherman, and prohibitions upon him and his. Rites of expiation and propitiation after fishing. III. Similar rites applied to warfare. IV. Rites the object of which is to secure regularity in the natural order of phenomena. The Aruntas’ intichiuma ceremonies. The mystic relation of the totemistic group to its totem. V. The couvade. Participation between child and parents. Customs pertaining to pregnancy, parturition, and early infancy. Persistence of participation, even at the time of initiation. CHAPTER VII INSTITUTIONS IN WHICH COLLECTIVE REPRESENTATIONS GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (11) I. Illness always produced by the influence of spirits in various forms. Diagnosis concerned with the discovery of the spirit involved. Treatment essentially mystic, spirit acting on spirit. Cherokee prescriptions. Classification of diseases. II. Death never “ natural.’’ The double significance of the term. Practices of divination to discover the power responsible for the death, and where it is to be sought for. Juxta hoc, ergo propter hoc. III. Divination, a means of discovering latent or hidden partici- pation. The divinatory significance of games. Sympathetic magic. PAGE CONTENTS CHAPTER VIII INSTITUTIONS IN WHICH COLLECTIVE REPRESENTATIONS Il. IV. VI. Vil. GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (III) . The dead continue to live. Contradiction of ideas which this continued existence involves. The several stages of dying. Practices immediately following a death. Premature burials. The condition of the dead between his decease and the funeral obsequies. The sentiments a dead man inspires. The ceremony which ends the period of mourning completes the death. The obligations which cease when this ceremony has taken place. The dead whose bodies do not decay are peculiarly maleficent ghosts. The destruction of the deceased’s personal effects. The sense in which they continue to belong to him. Property a mystic participation. The condition of the widow. Birth a reincarnation. Like death, it is going on at different times. Mystic idea of conception. White men are reincarna- tions of natives. Infanticide, and its significance to prelogical mentality. The naming of the child. The child previous to initiation not yet participating in the life of the social group. The mystic meaning of the rites of initia- tion. Apparent death and re-birth. Initiation of medicine-men, wizards, shamans, etc., and members of secret societies. Mystic meaning of the rites to which they are subjected. PART IV CHAPTER IX THE TRANSITION TO THE HIGHER MENTAL TYPES I. II. III. In communities of the lowest type, participations are felt rather than perceived. The paucity of myths in most of such com- munities. Among more advanced peoples, participations tend towards representation. Development of myths and symbols. The individualization of spirits. Myths and their mystic significance. The participations they express. How the interpretation is to be sought for. II PAGE 301 361 12 HOW NATIVES THINK General conditions of the retreat from prelogical mentality and the progress towards logical thought. How the imper- meability to experience is lessened as soon as its logical absurdity comes to be felt The development of conceptual thought. Logical thought cannot pretend to supplant prelogical mentality entirely. They are co-existent in the apparent unity of the thinking subject. Postulates and prejudices which till now have hindered the clear perception of their relations, and the a: comprehension of their conflicts. “SV se, INDEX e . . . . e L - LI LL . e e PAGE 387 INTRODUCTION I THE representations which are termed collective, defined as a whole without entering into detail, may be recognized by the following signs. They are common to the members of a given social group; they are transmitted from one generation to another; they impress themselves upon its individual members, and awaken in them sentiments of respect, fear, adoration, and so on, according to the circumstances of the case. Their existence does not depend upon the individual ; not that they imply a collective unity distinct from the indi- viduals composing the social group, but because they present themselves in aspects which cannot be accounted for by considering individuals merely as such. Thus it is that a language, although, properly speaking, it exists only in the minds of the individuals who speak it, is none the less an incontestable social reality, founded upon an ensemble of collective representations, for it imposes its claims on each one of these individuals ; it is in existence before his day, and it survives him. This fact leads at once to a very important result, one on which sociologists have rightly insisted, but which had escaped the notice of anthropologists. To be able to under- stand the processes by which institutions have been established (especially among undeveloped peoples), we must first rid our minds of the prejudice which consists in believing that collec- tive representations in general, and those of inferior races) in particular, obey the laws of a psychology based upon the analysis of the individual subject. Collective representations have their own laws, and these (at any rate in dealing with primitives 1) cannot be discovered by studying the “ adult, t By this term, an incorrect one, yet rendered almost indispensable through common usage, we simply mean members of the most elementary social aggregates with which we are acquainted. Le J . { L " Hu § L 4 14 HOW NATIVES THINK civilized, white man.’’ On the contrary, it is undoubtedly the study of the collective representations and their connections in uncivilized peoples that can throw some light upon the genesis of our categories and our logical principles. Durkheim and his collaborators have already given examples of what may be obtained by following this course, and it will doubtless lead to a theory of knowledge, both new and positive, founded upon the comparative method. , This great task cannot be accomplished save by a series of successive attempts. Perhaps their inception will be rendered easier if we determine which are the most general laws govern- ing collective representations in undeveloped peoples. To find out exactly what are the guiding principles of primitive mentality, and how these make themselves felt in the primi- tive institutions and customs, is the preliminary problem with which we have to deal in this volume. Without the labours of © those who have preceded me—the anthropologists and eth- nologists of various countries—and more particularly without the information afforded me by the French sociological school I have just mentioned, I could never have hoped to solve this problem, or even to present it in practical terms. The analysis by that school of numerous collective representations, and of the most important among them, such as the group ideas of what is sacred, of ‘“‘mana,” totem, magic and religious symbols, etc., has alone rendered possible this attempt to co-ordinate and systematize these representations among primitives. Basing my conclusions on those labours, I have been able to show that the mental processes of “primitives” do not coincide with those which we are accustomed to describe in men of our own type; I believe I have even been able to discover wherein the difference between them lies, and to establish the most general laws peculiar to the mentality of primitives. I have received practical help, too, from sie fairly large number of psychologists who, following Ribot, aim at showing the importance of the emotional and the motor elements of mental life in general and extending to the intellectual life, properly so called. To quote but two works, both Ribot’s Logique des Sentiments and Heinrich Maier’s Psychologie des emotionalen Denkens show how narrow were the limits within INTRODUCTION 15 which traditional psychology, under the influence of formal logic, sought to confine the life of thought. Mental processes are infinitely more elastic, complex, and subtle, and they com- prise more elements of the psychic life than a too “‘ simplist ”’ intellectualism would allow. Ribot’s observations on psycho- logy, therefore, have been very valuable to me. Nevertheless, the research I have undertaken differs widely from his. His analysis bears mainly upon subjects which are interesting from the emotional, passionate, or even pathological standpoint of our social aggregates, and hardly gives a glance at the collec- tive phenomena to be found among other peoples. I, on the other hand, propose to determine which are the most general laws governing collective representations (including their affective and motor elements) in the most undeveloped peoples known to us. I] That the higher mental functions should be studied by the comparative, that is, the sociological, method, is no new idea. Auguste Comte had already distinctly advocated it in his Cours de Philosoplie positive. He would divide the study of these functions between biology and sociology. His well- known dictum: ‘“‘ Humanity is not to be defined through man, but on the contrary, man through humanity’’ is designed to show that the highest mental functions remain unintelligible as long as they are studied from the individual alone. If we are to understand them we must take the development of the race into consideration.: In a man’s mental life everything which is not merely the reaction of the organism to the stimuli it receives is necessarily of a social character. The idea was a fertile one, but it did not bear fruit imme- diately, either in Comte or in his more or less direct followers. In Comte it found the way barred, so to speak, by a sociology that he believed he had constructed in its entirety, but which actually was a philosophy of history. He thought he had shown that the law of the three stages exactly expresses the intellectual evolution of humanity considered as a whole, and also that of any particular community. In establishing the 16 HOW NATIVES THINK science of the higher mental functions, therefore, he does not consider it necessary to begin with a comparative study of these functions in different types of human societies. In the same way, when constructing his ‘‘ cerebral chart,” he is not guided by physiology, so sure is he that the work of the physio- logists will confirm a priori his classification and location of the faculties. So too, when formulating the essential features of his theory of the higher mental functions, the law of the three stages suffices him, since the more special laws cannot fail to come into line with that one. Similarly, he constructed his theory according to the development of Mediterranean civilization, but without suspecting a priori that the laws thus discovered may not hold good of all the races of humanity. In one sense, then, Comte is the initiator of a positive science of the mental functions, and to a large extent the merit of having conceived it and of having shown it to be a sociological science ought to be accorded him. But he did not undertake the investigation of phenomena which such a science demands. He did not even attempt it, and at the time he wrote his Politique positive, he would doubtless have condemned it as ““useless.”’ Nevertheless, the patient and meticulous study of mental phenomena in differing types of human societies, the necessity of which Comte had not perceived, had been begun by others, and they pursued it perseveringly, not as philosophers but as experts, with the straightforward aim of discerning and classi- fying facts. I mean the anthropologists and ethnologists and the English school of anthropology in particular. Primitive Culture, the important work of its head, E. B. Tylor, which appeared in 1871, and marks an epoch in the history of anthro- pological science, led the way for a large number of zealous and well-trained collaborators, whose work is not unworthy of their leader. Thanks to their zeal, an immense mass of documents dealing with the institutions, customs, and lan- guages of so-called savage or primitive races, and with the group ideas governing these, has been accumulated. Work of a similar kind has been carried on in Germany and in France. In the United States, the Ethnological Bureau of the Smith- sonian Institute published some excellent monographs on the Indian tribes of North America. INTRODUCTION 17 Now as the documents increased in number, a certain uniformity in the phenomena of which they treated became continually more evident. As fast as undeveloped peoples were discovered, or more thoroughly studied, in the most distant quarters of the globe, sometimes at the very antipodes of each other, extraordinary likenesses, sometimes even exact resemblances, down to the smallest details, were found to exist. There were the same institutions ; the same religious or magical ceremonies ; the same beliefs and customs relating to birth and death; the same myths, and so on. Thus the comparative method established itself, as it were. Tylor makes continual and very felicitous use of it in his Primitive Culture; so, too, does Frazer in The Golden Bough, and so also do other representatives of the school, such as Sydney Hartland and Andrew Lang. By so doing, these writers have been the necessary pre- cursors of the positive science of the higher mental functions, but they did not originate it, any more than Comte, though the reasons in their case were not the same. Why did not the use of the comparative method lead them to it? Was it for want of having regarded the problem as universal, and, having once compared primitive races with each other, comparing them with our own? Not at all. The English school of anthropology, on the contrary, following the example of its head, is perpetually trying to show the relation between “savage’’ and “ civilized’ mentality, and to explain it. And it is just this explanation which has prevented their going any further. They had it ready-made. They did not look for it in the facts themselves, but imposed it on’ them. While testifying to the existence of primitive institutions and beliefs so entirely different from our own, they did not ask themselves whether, to account for them, they ought not to examine various hypotheses. They took it for granted that the facts could be explained in one way only. Do the collective repre- sentations of the communities in question arise out of higher mental functions identical with our own, or must they be referred to a mentality which differs from ours to an extent yet to be determined? Such an alternative as the latter did not occur to their minds. 18 HOW NATIVES THINK Ee Without entering upon a critical discussion of the method employed and the results obtained by these experts '~a discussion to which I could not devote the space it demands-— I should merely like to demonstrate in a few words the con- sequences to their theory of this belief in the identity of a “human mind” which, from the logical point of view, is always exactly the same at all times and in all places. Such an identity is admitted by the school as a postulate, or rather, anaxiom. There is no need to demonstrate it, or even formally enunciate it: it is an understood principle, and too evident for any consideration of it to be necessary. Accordingly, the collective representations of primitives, so often foreign to our ideas, and the connections we find between them, which are no less strange, do not raise problems the solutions of which may. either amplify or modify our conception of the. “‘ human mind.” We know already that this mind of theirs is not different from ourown. All we have to find out is the way in which mental functions exactly like ours have been able to produce these representations and their connections, and here comes in the general hypothesis which the English school of anthropology favours : it is animism. Frazer’s Golden Bough, for instance, clearly shows how animism accounts for many of the beliefs and practices found almost everywhere in primitive social aggregates, numerous traces of which survive in our own. We shall notice that the hypothesis comprehends two successive stages. In the first place, the primitive, surprised and moved by apparitions which present themselves in his dreams—where he sees once more the dead and the absent, talks and fights with them, touches them and hears them speak—believes in the objective reality of these representations. To him, therefore, his own existence is a double one, like that of the dead and absent who appear to him. He admits his actual existence as a living and conscious personality, and at the same time as a soul able to separate itself from him, to become external to him and 1 On this point see Durkheim’s articles in the Revue Philosophique of January and February 1909, entitled ‘ Examen critique des systèmes classiques sur l’origine de la pensée religieuse.” INTRODUCTION 10 manifest itself in a ‘“ phantom ” state. This belief would be universal among primitives, because all would be subject to the psychological self-delusion in which it originated. In the second place, when they want to account for the natural phenomena which make an impression on their senses, and assign a cause to them, they immediately apply and generalize the explanation they have accepted of their dreams and hallucinations. In all forms of being, and behind all natural phenomena, they imagine “‘souls,’”’ “‘ spirits,’ ‘‘ intentions,”’ similar to those they believe they have experienced in them- selves and their companions, and in animals. It is a simple and artless logical process, but not less spontaneous nor less inevitable to the “‘ primitive’’ mind, than the psychological illusion which preceded it and upon which it is based. Thus without any attempt at reflection, by the mere influence of mental processes which are the same to all, the primitive develops a “‘ philosophy ”’ of his own, childish and clumsy, no doubt, but yet perfectly consistent with itself. It propounds no problem that it cannot immediately solve to its complete satisfaction. If it were anyhow possible for all the experience acquired and transmitted through the centuries to be suddenly obliterated—if we were brought face to face with nature like the real “‘ primitives,’’ we should be certain to construct. for ourselves a “natural philosophy,” which would * also be primitive, and that philosophy would be a universal animism, quite correct from the logical point of view, consider- ing the paucity of positive data we should have at command. The hypothesis of animism in this sense, then, is a direct consequence of the axiom which dominates the researches of the English school of anthropology, and, according to our view, it is this that has prevented it from attaining to a positive science of the higher mental functions, to which the compara- tive method would seem to be leading. For while the resem- blance in institutions, beliefs and practices in undeveloped peoples of the most varied kind is explained by such a hypo- thesis, it takes no trouble to demonstrate that their superior mental functions are identical with our own. The axiom serves them as demonstration. The fact that myths and collective representations like those on which totemism is based, such as the belief in spirits, in separate and external € 20 HOW NATIVES THINK souls, and in sympathetic magic, are to be found in all human aggregates, is a necessary result of the structure of the “ human ' mind.” The laws governing the association of ideas, and the natural and irresistible application of the law of causality, combined with animism, would be certain to engender these collective representations and their connections. It is only the spontaneous working of an unvarying logical and psycho- logical process. Nothing is easier to explain if we once admit, as the English anthropological school implicitly does, that this process is the same in undeveloped peoples as it is with us. Must we admit this? That is what I have to find out. But even at this point it is clear that if there be any doubt about this axiom, animism, based upon it, would also be matter of doubt and could in no case serve as a proof. We should find it impossible, unless we argued in a vicious circle, to account for the spontaneity of animism in primitives by a specific mental construction, and affirm the existence of the latter in them by relying upon this same spontaneity of animism for support. An axiom and its deduction cannot be proved by each other. IV It now remains to be seen whether the animist theory is borne out by the facts, and whether it is sufficient to account for the institutions and beliefs of primitive peoples. This is the point upon which Tylor, Frazer, Andrew Lang and many other representatives of this school have expended both knowledge and skill. The tremendous amount of documentary evidence they bring forward in support of their position is almost incredible to those who have not read it. In dealing with this vast collection, however, two things must be borne in mind. The first is that we may consider the presence of the same institutions, beliefs, practices in a great number of widely separated but typically analogous peoples, as established. From this we may legitimately conclude that the mental processes which produce these similar ideas are the same, for it is clear that resemblances of such a kind, which occur so frequently and are so exact, cannot be merely fortuitous. But the accumulated facts which are conclusive upon this point INTRODUCTION 1 are not equally valid when it is a case of proving that these oe ideas have a common origin in the belief in animism, in a spontaneous ‘natural philosophy” which would be the earliest reaction of the human mind to the appeal of experience. Undoubtedly the explanation of each belief or practice thus obtained is usually plausible, and we can always imagine the working of the mental process which led to it in the primitive : but it is only plausible. And is it not the first rule of any prudent method never to take for granted that which is merely probable? Savants have learnt in so many cases that the apparently true is rarely the truth. There is a like reserve, in this respect, among philologists and natural philosophers. Must not the sociologist be equally cautious? The very language of the anthropologists and the form adopted by their demonstrations show that they do not go beyond probability, and the number of the facts brought forward adds nothing to the conclusive force of the argument. Uncivilized races have an almost universal custom of destroying a dead man’s weapons, his clothes, the things he used, even his house, and occasionally they sacrifice his wives and his slaves. How are we to account for this? “The custom,” says Frazer, ‘‘ may : have sprung from the idea that the dead were angry with the living for dispossessing them. . . . The idea that the souls of the things thus destroyed are dis- patched to the spirit-land . . . is less simple and therefore probably * later.” 2 Undoubtedly this custom may have arisen thus; but it may also be of different origin. Frazer’s theory does not preclude any other, and his wording even acknowledges this. As to the general principle upon which he relies, and which he expressly formulates a little further on: “‘ In the evolution of thought, as of matter, the simplest is the earliest,” it un- doubtedly proceeds from Herbert Spencer’s philosophy, but that does not make it any the more certain. I do not think it can be proved in the material world, and in what we know of the world of ‘ thought,” the facts would seem to contradict it. Frazer seems to be confusing ‘simple ’’ and “ undiffer- 1 The italics are mine (L. L.-B.). 2 “ Certain Burial Customs as Illustrative of the Primitive Theory of the Soul,” Journal of the Anthropological Institute (henceforth J.A.I,), xv. p. 75 (note 1), (1885). 22 HOW NATIVES THINK entiated’’ here. Yet we find that the languages spoken by peoples who are the least developed of any we know— Australian aborigines, Abipones, Andaman Islanders, Fuegians, etc.—exhibit a good deal of complexity. They are far less ‘simple ”’ than English, though much more “ primitive.” Here is another instance, taken from the same article by Frazer. It is a common practice in varying regions, and persistent from all time, to place in the mouth of a dead man either some grains of corn or a gold coin. Frazer quotes from a considerable number of documents in proof of this. Then he explains it thus. ‘The original custom may have been that of placing food in the mouth, for which in after times valuables (money or otherwise) were substituted, that the dead might buy his own food.”’! The explanation is a probable one, but in one case in which we can check it, it is incorrect. A similar custom has existed in China from time immemorial, and De Groot, following old Chinese documents, gives us the true reason. Gold and jade are substances which endure indefi- nitely. They are symbols of the celestial sphere, “‘ which is unchangeable, indestructible, beyond the influences of decay. . . . Hence jade and gold (pearls too) naturally endow with vitality all persons who swallow them ; in other words, they intensify their souls or shen, which are, like the heavens, composed of Yang matter; and they hold at a distance from the dead corruption and decay, thus furthering their return to life.’ 2 We must even go further. ‘“‘ Taoist and medical authors assert that, whoever eats jade, gold or pearls does not only prolong his life, but ensures also the existence of his body after death, saving it from putrefaction. This doctrine, by its mere existence, intimates that szen, who acquired immortality by eating such or other substances, were conceived to continue using their body after their earthly career, or removed to the region of the immortals, also corporally. A new light is thus shed on the custom of ancients and moderns to keep away corruption from the dead by placing the three precious things in their mouths or other apertures: it was an attempt to make sten of them.’’3 Elsewhere the dead are given money with which to make their purchases in the next world, but it is 1 J.A.I., xv. pp. 77-9 (note). > J. J. M. de Groot, The Religious System of China, i. p. 271. 3 Ibid., il. pp. 331, 332. INTRODUCTION 23 not put into their mouths. Here we have a belief similar to that which induces the Chinese to choose for their coffins woods distinguished for their durability, such as pine and cypress ‘‘ both of which are possessed of great vitality, so that the wood, when placed around the dead, might facilitate their return to life.”’t These are cases, similar to so many others we read of, of participation through contact. These two examples will doubtless suffice, although we could cite many similar ones. The ‘‘explanations’’ of the English school of anthropology, being never anything more than probable, are always affected by a co-efficient of doubt, which varies according to the circumstances of the case. They take it for granted that the ways which, to our minds, seem to lead naturally to certain beliefs and practices, are precisely those trodden by the members of the communities in which they are to be found. Nothing can be more risky than such a postulate, and it would possibly not be confirmed five times out of a hundred. In the second place, the phenomena to be accounted for— institutions, beliefs, practices—are all pre-eminently social phenomena. Should not the representations and the con- nections between the representations be of the same nature ? Are they not necessarily ‘ collective representations ” ? But - in such a case the animist theory becomes suspect, and with it the postulate upon which it was based, for both theory and postulate deal with the mental processes of the individual human mind only. Collective representations are social phenomena, like the institutions for which they account ; and if there is any one point which contemporary sociology has thoroughly established, it is that social phenomena have their own laws, and laws which the analysis of the individual qua individual could never reveal. Consequently any attempt to “explain ” collective representations solely by the functioning of mental operations observed in the individual (the association. of ideas, the naive application of the theory of causality, and so on), is foredoomed to failure. Since some of the data which are essential to the problem have been omitted, its defeat is certain. We might just as well hope to make scientific use of the idea of a human individual mind imagined to be devoid of 1 J. J. M. de Groot, The Religious System of China, i. p. 294. 24 HOW NATIVES THINK all experience whatever. Would it be worth while to try and reconstruct the method in which such a mind would represent the natural phenomena which occurred within and around him? As a matter of fact, we have no means of knowing what such a mind would be like. As far back as we can go, however primitive the races we may study, we shall never find any minds which are not socialized, if'we may put it thus, not already concerned with an infinite number of collective repre- sentations which have been transmitted by tradition, the origin of which is lost in obscurity. | The idea of an individual human mind absolutely free from all experience is, then, as fanciful as that of man prior to social life. It does not correspond with anything that we can grasp and verify, and the hypotheses implying it could but be arbitrary. If on the other hand, we start with collective — representations as our data, as a reality which will bear scientific examination, we shall certainly not have probable and seductive ‘‘explanations’’ to oppose to those of the English school. It will all be much less simple. We shall find ourselves faced by complex problems, and very frequently we shall not have sufficient data to solve them; very fre- quently, too, the solutions we shall propose will be hypo- thetical. But at least we may hope that by a positive study of collective representations we may arrive by degrees at a knowledge of the laws which govern them and thus obtain a more correct interpretation of the mentality of primitive peoples, and even of our own. One example, perhaps, will suffice to show how complete is the contradiction between the standpoint of the English school of anthropology and the one we hope to see adopted. Tylor writes : ‘‘ Conformably with the early child-like philosophy in which human life seems the practical key to the understanding of nature at large, the savage theory of the universe refers its phenomena in general to the wilful action of pervading personal spirits. It was no spontaneous fancy, but the reasonable inference that effects are due to causes, which led the rude men of old days to people with such ethereal phantoms their own homes and haunts, and the vast earth and sky beyond. Spirits are simply personified causes.’’! There is nothing simpler or 1 Primitive Culture (4th edit.), ii. pp. 108, 109. INTRODUCTION 25 more acceptable than this ‘explanation ” of an immense number of beliefs, provided we admit, as Tylor does, that they are the result of a ‘‘ reasonable inference.’’ But it is very difficult to grant this. When we consider the collective representations which, in undeveloped peoples, imply a belief in spirits pervading the whole of nature, which inspires the practices relating to these spirits, it does not seem as if these practices are the result of a mental curiosity seeking for causes. Myths, funeral rites, agrarian practices and the exercise of magic do not appear to originate in the desire for a rational explanation: they are the primitives’ response to collective needs and sentiments which are profound and mighty and of compulsive force. I do not maintain that this desire for an explanation does not exist. Like so many other potentialities which will be realized later when the social group develops, this curiosity is latent, and it may possibly be already manifest to a slight extent in the mental functioning of such peoples. But it is assuredly contrary to the facts to see in it one of the main directing controls of that functioning, and the origin of the collective representations relating to most of the natural phenomena. If Tylor and his followers are satisfied with such an ‘‘ explanation,” it is because they assume these beliefs to exist in individual minds similar to their own. As soon, however, as we take into consideration the collective nature of these ideas, the inadequacy of this explanation is apparent. Being collective, they force themselves upon the individual ; that is, they are to him an article of faith, not the product of his reason. And since the collective representations, as a rule, predominate most where the races are least advanced, the mind of the “‘ primitive ’’ has hardly any room for such ques- tions as “how?” or ‘“ why ? ’’ The ensemble of collective representations which master him and excite in him an in- tensity of feeling which we cannot even imagine, is hardly com- patible with that disinterested contemplation of a matter which a purely intellectual desire to probe into its cause would demand. Without entering upon a detailed discussion of the theory of animism here—for we shall find room for it later—we are allowed to think that Tylor’s dictum that “ spirits are per- sonified causes ”” does not suffice to account for the place held 26 HOW NATIVES THINK by spirits in the collective representations of primitives. To us, however, interested first of all in analysing these repre- sentations without any preconceived ideas about the mental processes upon which they depend, it may possibly be the “ spirits,” on the contrary, which will help us to understand what certain ‘causes ” are. Perhaps we shall find that the effect of the efficient cause—vexata questio to the philosophers —is a sort of abstract precipitate of the mystic power attri- buted to spirits. This is a theory which we intend to examine, and in any case we shall be suspicious of categorical and comprehensive pronouncements. The English school of anthropology, with its lofty theory of animism and its pre- conceived ideas, always has at least a probable explanation of the phenomena it has collected. If new phenomena manifest themselves, its theory is plastic and general enough to permit of its explaining these too; it only requires the exercise of a © little ingenuity. It sees in these a confirmation of its theory. But this confirmation has precisely the same validity as the probable “‘ explanations ”’ of which it is but a fresh example. We are sure to be asked how a savant such as Tylor, whose perspicacity is so marvellous and whose criticism is so acute in dealing with special phenomena, should have shown himself less exacting when a general theory was at stake, and how it is that his followers could have imitated his example in this matter. Perhaps we must take into account here the influence of contemporary English philosophy, and the theory of evolu- tion in particular. At the time Primitive Culture appeared, and for some years after, the philosophy of associationism seemed to have a definite sway. The evolutionism of Herbert Spencer, then greatly in repute, exercised considerable fascina- tion over many minds. They saw in it the expression of the most comprehensive philosophical synthesis: an expression which could at the same time be adapted to any class of natural phenomena whatever, and thus serve as a guiding line in scientific research. It could be applied to the history of the solar system as easily as to the genesis of organic matter, or that of the intellectual life. They might therefore anticipate that it would be extended to social phenomena, and Spencer did not disappoint them in this. He too, as we know, took as his govern- INTRODUCTION 27 ing hypothesis in explaining the mentality of primitives, a theory of animism based upon an associationistic philosophy. At the present day, Herbert Spencer’s evolutionism is somewhat severely judged. His generalizations appear hasty, presumptuous, and not well founded. Thirty years ago, how- ever, they were considered both substantial and potent. Tylor and his school believed they saw in them a guarantee for the continuity in man’s mental development which they formulate. Such a doctrine allowed them to present this development as an uninterrupted evolution, the stages of which could be noted, from the animistic beliefs of ‘‘ primi- tives ’’ to the Newtonian conception of the universe. At the same time, a little throughout his Primitive Culture, and especially in its concluding parts, Tylor takes the trouble to refute a theory according to which “‘savage”’ races are in reality degenerate ones—their representation of nature, their institutions and beliefs being the almost obliterated but yet recognizable remains of original revelation. But to oppose to this theory of a theological order, can Tylor find anything better than the theory of evolution which is, according to him, of a scientific order? The latter provides him with a rational interpretation of the facts. That which has been presented as the vestiges of a more perfect antecedent state, Tylor readily explains, from the evolutionist point of view, to be the rudimentary germ of a succeeding state which shall be more differentiated. Finally, when we remember how much apparent clearness and intelligibility the general theory of animism gives to the mass of evidence, we shall not be surprised at the success it has shared with the theory of evolution, nor wonder that the large majority of the English school of anthropology should have remained faithful to it until now. V The series of social phenomena are solidary with respect to each other, and they are placed in mutual relationship. A definite type of society, with its own institutions and customs, will therefore necessarily have its own mentality. Different mentalities will correspond with different social types, and all 28 HOW NATIVES THINK the more because institutions and customs themselves are at bottom only a certain aspect of collective representations, only these representations considered objectively, as it were. Thus we are led to perceive that the comparative study of the different types of human societies is linked up with the com- parative study of the collective representations and the connections between these, which dominate such societies. Similar considerations must have prevailed with the naturalists when, while retaining the idea of the identity of the essential functions in all living beings, or at any rate, of all animal nature, they decided to admit types differing funda- mentally from each other. There is no doubt that the pro- cesses of nutrition, respiration, secretion, reproduction, do not vary in essentials, whatever the organism which manifests them may be. But they may be manifested in an ensemble of conditions which are totally different from the histological or anatomical or physiological point of view. General biology made a great advance when it recognized that it could not (as Auguste Comte believed) find in the analysis of the human organism the means of attaining a clearer understanding of the organism of a sponge. From that time forward, the study of biology proper was not complicated by preconceived ideas of the subordination of one entity to another, notwithstanding the possibility of forms originally of a common type, prior to divergence. In the same way, there are features which are common to all aggregates of human beings, by which they may be distin- guished from the rest of the animal world. Language is spoken, traditions are transmitted, and institutions main- tained. The higher mental operations, therefore, have every- where a basis of homogeneity. This being once admitted, however, human aggregations, like organisms, may differ profoundly in their construction and as a consequence their higher mental operations will also present corresponding differences. We must then reject beforehand any idea of reducing mental operations to a single type, whatever the peoples we are considering, and accounting for all collective representations by a psychological and mental functioning which is always the same. If it be true that there are aggre- gates of human beings who differ from each other in construc- INTRODUCTION 29 tion as invertebrate animals differ from vertebrates, a com- parative study of the various types of collective mentality is just as indispensable to anthropology as comparative anatomy and physiology are to biology. Is there any need to say that this comparative study, conceived thus as universal, presents difficulties which at the moment are insurmountable ? In the present state of sociology we could not dream of undertaking it. The differentiation of types of mentality is as difficult as that of types of society, and for similar reasons. What I am attempting here, by way of introduction, is a preliminary study of the most general laws to which collective representations in inferior races are sub- jected, and more especially those of the most primitive peoples of which we know anything. I shall endeavour to construct, if not a type, at any rate an ensemble of characteristics which are common to a group of neighbouring types, and in this way to define the essential features of the mentality peculiar to undeveloped peoples. In order to bring out these features as clearly as possible, I shall compare that mentality with our own, i.e. with that of races which are the product of ‘‘ Mediterranean ”’ civilization, in which a rationalistic philosophy and positive science have been developed. For the first rough attempt at comparative study there is an evident advantage in making choice of the two mental types available, between which the difference is greatest. The essential dissimilarities between such will be most marked, and as a consequence they will be least likely to escape attention. Moreover, in starting: with these we shall be most easily able to approach the study of the inter- mediary or transitional types. Even when confined to such limits, the attempt will no doubt appear only too bold, and very uncertain as to result. It remains incomplete ; it opens up many more questions than it can answer; and leaves unsolved more than one vast problem which it touches in a superficial way. I am not forgetting all this, but, in examining a mentality which is so obscure, I have thought it better to confine myself to that which seemed to stand out most distinctly. Besides, as far as the mentality peculiar to our society is concerned, since it is only to serve me as a state for comparison, I shall regard it as 30 HOW NATIVES THINK sufficiently well defined in the works of philosophers, logicians and psychologists, both ancient and modern, without conjec- turing what sociological analysis of the future may modify in the results obtained by them up to the present. The real object of my researches, therefore, is the study, by means of the collective representations of primitives, of the mental processes which regulate them. But we do not know these representations and their con- nections except by the institutions, beliefs, myths, customs of undeveloped peoples, and how do we obtain these? It is nearly always from the accounts of travellers, sailors, mission- aries—in short, from documents to be found in the ethno- graphical records of the two hemispheres. Every sociologist ought to realize the value of these documents. It is a problem of capital importance, to which the ordinary rules of criticism apply, and one which I cannot broach here. I must, however, draw attention to the fact that the anxious care to observe undeveloped peoples scientifically, by means of a system as objective, precise and detailed as that which experts employ in determining natural phenomena, and as similar as it can be to this, is but of recent growth. And now that it does exist, a kind of irony ordains that it has scarcely any object. The last century has witnessed irreparable losses to the comparative study of human aggregates. The peoples whose institutions would have been of the greatest interest to this scientific study are rapidly becoming extinct in the most widely separated districts. The undeveloped peoples still remaining are doomed to disappear speedily, and it behoves careful observers to make haste. The enormous accumulation of the older records by no means makes up for what we are losing in this way. With rare exceptions, the facts collected in passing by travellers who only go through a country are of very little value. Major Powell remarks with good reason that these people can no more make known to us the institutions of a tribal community than they can give an exact description of the flora of a country, the fauna of a district, or the geological formation of a continent.! 1 Report of the Bureau of Ethnology of the Smithsonian Institute, Washington (henceforth cited as Æ. B. Rept.), iii. p. 62 INTRODUCTION 31 _ In most cases, too, those who first saw these peoples from within, even if they lived among them for some time, were too much engaged otherwise to be able to give accurate and detailed accounts, as complete as possible, of the institutions and customs which came to their notice. They observed what seemed to them most noteworthy and singular, the things that piqued their curiosity ; they described these more or less happily. But the observations thus collected were always merely side-issues to them, never the main reason for their sojourn among these peoples. Moreover, they did not hesitate to interpret phenomena at the time they described them: the very idea of hesitation would have seemed quite unnecessary. How could they suspect that most of their interpretations were simply misapprehensions, and that “ primitives’’ and “savages ’’ nearly always conceal with jealous care all that is most important and most sacred in their institutions and beliefs ? Nevertheless, as Tylor has clearly shown, many of these old observations are illuminated and corrected by the light of the knowledge we possess to-day. Some of them, indeed, are very valuable ; for instance, those made by certain mission- aries who resided for a long time in the community they describe, who seem to have assimilated its spirit somewhat, and in whose accounts we can easily separate observation properly so-called from the preconceived ideas which have mingled with it. Such, among others, are the Jesuit fathers who were the first to come in contact with the Indian tribes of North America,—Dobrizhoffer with the Abipones, in the eighteenth century,—and more recently, missionaries like Turner in Samoa, Codrington in Melanesia, etc. The earliest of this class of observers had the advantage of knowing nothing of sociological theories, and it often happens that their accounts are of all the more importance to us precisely because they do not understand what they are relating. As a set-off to this, however, they are often irritatingly incomplete, and silent upon the most essential points. Compared with these sketches, the correctness of which is never certain, which their designers have sometimes retouched or refashioned according to the taste of the period, the observa- tions made to-day by professional ethnographers appear like 32 HOW NATIVES THINK good, clear photographs. In fact, the collaborators of the Ethnological Bureau of the Smithsonian Institute of Washing- ton use both photographic and phonographic apparatus as part of their necessary equipment. It is from investigators aware of the difficulties of their task, and expert in the methods which admit of their attacking it with the best chances of success, that we look for documents. But even in such cases we must not overlook such precautions as a considered criticism demands. Many among them are missionaries, Catholic or Protestant, and they, like their predecessors of past centuries, are convinced that savages derive some kind of natural religion from God, and owe to the agency of the devil their most reprehensible customs. Many of them, both cleric and lay, have read the works of Tylor and Frazer, and have become their adherents. Since they have the self- imposed task of procuring fresh evidence in support of their - masters’ theories, they observe with prejudiced eyes. This becomes a serious disadvantage when they set out provided with a detailed set of questions drawn up in the spirit of a particular school. Their eyes seem to be screened in a way which prevents them from perceiving any fact not provided for in their catechism, and in relating what they do see, their preconceived interpretation can no longer be distinguished from the facts themselves. PART I CHAPTER I COLLECTIVE REPRESENTATIONS IN PRIMITIVES’ PERCEPTIONS AND THE MYSTICAL CHARACTER OF SUCH I BEFORE undertaking an investigation of the most general laws governing collective representations among undeveloped peoples, it may be as well to determine what the essential characteristics of these representations are, and thus avoid an ambiguity which is otherwise almost inevitable. The ter- minology used in the analysis of mental functions is suited to functions such as the philosophers, psychologists, and logicians of our civilization have formulated and defined. If we admit these functions to be identical in all human aggregates, there is no difficulty in the matter; the same terminology can be employed throughout, with the mental reservation that “savages ’’ have minds more like those of children than of adults. But if we abandon this position—and we have the strongest reasons for considering it untenable—then the terms, divisions, classifications, we make use of in analysing our own mental functions are not suitable for those which differ from them; on the contrary, they prove a source of confusion and error. In studying primitive mentality, which is a new subject, we shall probably require a fresh terminology. At any rate it will be necessary to specify the new meaning which some expressions already in use should assume when applied to an object differing from that they have hitherto betokened. This is the case, for instance, with the er “ collective representations.”’ a | In the current eine of psychology which classifies phenomena as emotional, motor, or intellectual, ‘ representa- 36 HOW NATIVES THINK tion” is placed in the last category. We understand by it a matter of cognizance, inasmuch as the mind simply has the image or idea of an object. We do not deny that in the actual mental life every representation affects the inclinations more or less, and tends to produce or inhibit some movement. But, by an abstraction which in a great many cases is nothing out of the ordinary, we disregard these elements of the repre- sentation, retaining only its essential relation to the object which it makes known to us. The representation is, par excellence, an intellectual or cognitive phenomenon. It is not in this way, however, that we must understand the collective representations of primitives. Their mental activity is too little differentiated for it to be possible to consider ideas or images of objects by themselves apart from the emotions and passions which evoke these ideas or are © evoked by them. Just because our mental activity is more | differentiated, and we are more accustomed to analysing its functions, it is difficult for us to realize by any effort of imagi- nation, more complex states in which emotional or motor elements are integral parts of the representation. It seems to us that these are not really representations, and in fact if we are to retain the term we must modify its meaning in some way. By this state of mental activity in primitives we must understand something which is not a purely or almost purely intellectual or cognitive phenomenon, but a more complex one, in which what is really “‘representation’’ to us is found blended with other elements of an emotional or motor character, coloured and imbued by them, and therefore imply- ing a different attitude with regard to the objects represented. Moreover, these collective representations are very often acquired by the individual in circumstances likely to make the most profound impression upon his sensibility. This is particularly true of those transmitted at the moment when he becomes a man, a conscious member of the social group, the moment when the initiation ceremonies cause him to undergo new birth,: when the secrets upon which the very life of the group depends are revealed to him, sometimes amid tortures which subject his nerves to the most severe tests. It would be difficult to exaggerate the intense emotional force of such t Vide Chap. VIII. pp. 352-3. PRIMITIVES’ PERCEPTIONS 37 representations. The object is not merely discerned by the mind in the form of an idea or image; according to the cir- cumstances of the case, fear, hope, religious awe, the need and the ardent desire to be merged in one common essence, the passionate appeal to a protecting power—these are the soul of these representations, and make them at once cherished, formidable, and really sacred to the initiated. We must add, too, that the ceremonies in which these representations are translated into action, so to speak, take place periodically ; consider the contagious effect of the emotional excitement of witnessing the movements which express them, the nervous exaltation engendered by excessive fatigue, the dances, the phenomena of ecstasy and of possession,—in fact everything which tends to revive and enhance the emotional nature of these collective representations. At any time during the intervals between the occurrences of these ceremonies, when- ever the object of one of these representations once more arises in the consciousness of the ‘‘ primitive,’ even should he be alone and in a calm frame of mind at the moment, it can never appear to him as a colourless and indifferent image. A wave of emotion will immediately surge over him, un- doubtedly less intense than it was during the ceremonies, but yet strong enough for its cognitive aspect to be almost lost sight of in the emotions which surround it. Though in a lesser degree, the same character pertains to other collective representations—such, for instance, as those transmitted from generation to generation by means of myths and legends, and those which govern manners and customs which apparently are quite unimportant; for if these customs are respected and enforced, it is because the collective representations relating to them are imperative and something quite different from purely intellectual phenomena. The collective representations of primitives, therefore, differ very profoundly from our ideas or concepts, nor are they their equivalent either. On the one hand, as we shall presently discover, they have not their logical character. On the other hand, not being genuine representations, in the strict sense of the term, they express, or rather imply, not only that the primitive actually has an image of the object in his mind, and thinks it real, but also that he has some hope or fear connected 38 HOW NATIVES THINK with it, that some definite influence emanates from it, or is exercised upon it. This influence is a virtue, an occult power which varies with objects and circumstances, but is always real to the primitive and forms an integral part of his repre- sentation. If I were to express in one word the general peculiarity of the collective representations which play so important a part in the mental activity of undeveloped peoples, I should say that this mental activity was a mystic … one. In default of a better, I shall make use of this term— not referring thereby to the religious mysticism of our com- munities, which is something entirely different, but employing the word in the strictly defined sense in which “ mystic ” implies belief in forces and influences and actions which, though imperceptible to sense, are nevertheless real. | In other words, the reality surrounding the primitives is . itself mystical. Not a single being or object or natural phenomenon in their collective representations is what it appears to be to our minds. Almost everything that we perceive therein either escapes their attention or is a matter of indifference to them. On the other hand, they see many things there of which we are unconscious. For instance, to the primitive who belongs to a totemic community, every animal, every plant, indeed every object, such as the sun, moon, and stars, forms part of a totem, and has its own class and sub-class. Consequently, each individual has his special affinities, and possesses powers over the members of his totem, class, and sub-class ; he has obligations towards them, mystic relations with other totems, and so forth. Even in communi- ties where this form does not exist, the group idea of certain animals (possibly of all, if our records were complete) is mystic in character. Thus, among the Huichols, “ the birds that soar highest . . . are thought to see and hear everything, and to possess mystic powers, which are inherent in their wing and tail feathers.”” These feathers, carried by the shaman, ‘“‘ enable him to see and hear everything both above and below the earth . . . to cure the sick, transform the dead, call down the sun, etc.’’! The Cherokees believe that fishes live in companies like human beings, that they have their villages, their regular paths through the waters, and that they conduct 1 C, Lumholtz, Unknown Mexico, ii. pp. 7-8. PRIMITIVES’ PERCEPTIONS 39 themselves like beings endowed with reason. They think, too, that illnesses—rheumatic affections in particular—pro- ceed from a mystic influence exercised by animals which are angry with the hunters, and their medical practices testify to this belief. In Malaya and in South Africa the crocodile, and in other places the tiger, leopard, elephant, snake, are the object of similar beliefs and practices, and if we recall the myths of which animals are the heroes, in both hemispheres, there is no mammal or bird or fish or even insect to which the most extraordinary mystic properties have not been attributed. Moreover, the magic practices and ceremonies which, among nearly all primitive peoples, are the necessary accompaniment of hunting and fishing, and the sacrificial rites to be observed when the quarry has been killed, are sufficiently clear testimony to the mystic properties and powers which enter into the collective representations relating to the animal world. It is the same with plant life. It will doubtless suffice to mention the snfichiuma ceremonies described by Spencer and Gillen, designed to secure, in mystic fashion, the normal reproduction of plants,—the development of agrarian rites, corresponding with the hunting and fishing ceremonial, in all places where primitive peoples depend wholly or partly on the cultivation of the soil for their subsistence—and lastly, the highly unusual mystic properties ascribed to sacred plants, as, for instance, the soma in Vedic India, and the hikuli among the Huichols. Again, if we consider the human body, we shall find that each organ of it has its own mystic significance, as the wide- spread practice of cannibalism and the rites connected with human sacrifices (in Mexico, for instance) prove. The heart, liver, kidney, the eyes, the fat, marrow, and so on, are reputed to procure such and such an attribute for those who feed on them. The orifices of the body, the excreta of all kinds, the hair and nail-parings, the placenta and umbilical cord, the blood, and the various fluids of the body, can all exercise magic influences.2 Collective representations attribute mystic power t J. Mooney, ‘“ The Sacred Formulas of the Cherokee,” E. B. Rept., vii. Pp. 375: + K. Th. Preuss, ‘ Der Ursprung der Religion und Kunst,’ Globus, Ixxxvi. Pp. 20; Ixxxvii. p. 19. | 40 HOW NATIVES THINK to all these things, and many wide-spread beliefs and practices relate to this power. So, too, certain parts of plants and animals possess peculiar virtues. ‘‘ Badtz is the name given to the evil principle which . . . attends (like an evil angel) everything in his life. . . . Von de Wall describes it as the ‘enchanting or destroying influence which issues from any- thing; for example, from a tiger which one sees, from a poisonous tree which one passes under, from the saliva of a mad dog, from an action which one has performed.’ ”’ 1 Since everything that exists possesses mystic properties, and these properties, from their very nature, are much more important than the attributes of which our senses inform us, the difference between animate and inanimate things is not of ‘the same interest to primitive mentality as it is to our own. As a matter of fact, the primitive’s mind frequently disregards it altogether. Thus rocks, the form or position of which - strike the primitive’s imagination, readily assume a sacred character in virtue of their supposed mystic power. Similar power is ascribed to the rivers, clouds, winds. Districts in space, direction (the points of the compass), have mystic significance. When the Australian aborigines assemble in large numbers, each tribe, and each totem of a tribe, has its own place, a place assigned to it by virtue of its mystic affinity with a particular spatial region. Facts of a similar nature have been noted in North America. I shall not lay any stress _on the rain, lightning, or thunder, the symbols of which play so important a part in the religious ceremonies of the Zuñi, the Australian aborigines, and all aggregates where a pro- longed drought is a serious menace to the very existence of the group. Finally, in Loango, the soil “is something more to the Bafioti than the scene upon which their lives are played out. There is in the ground, and there issues from it, a vital influ- ence which permeates everything, which unites the present and the past. . . . All things that live owe their powers to the soil. . . . The people regard their land as a fief from their god ... the ground is sacred.’’2 The same belief obtains among the North American Indians, who consider it sacrilege to till the ground, for by so doing they would run a risk of 1 W. W. Skeat, Malay Magic, p. 427. 2 Dr. Pechuël-Loesche, Die Loango-Expedition, iii. 2, pp. 194 et. seq. (1907). PRIMITIVES’ PERCEPTIONS 41 offending the mystic power and drawing down dire calamities upon themselves. Even things made, and constantly used, by man have their mystic properties and can become beneficent or terrifying according to circumstances. Cushing, who had lived among the Zuñis, had made them adopt him, and whose unusual versatility of mind led him finally to think like them, says that they, “no less than primitive peoples generally, conceive of everything made . . . whether structure or utensil or weapon, .. as living .. . a still sort of life, but as potent and aware nevertheless and as capable of functioning not only obdurately and resistingly, but also actively and powerfully in occult ways, either for good or for evil. As for living things they observe every animal is formed, and acts or functions accord- ing to its form—the feathered and winged bird flying, because of its feathered form, the furry four-footed animal running and leaping, and the scaly and finny fish swimming—. . . So the things made or born in their special forms by the hands of man also have life and function variously according to their various forms.” Even the differences in the claws of beasts, for example, are supposed to make the difference between the hugging of the bear and the clutching of the panther. ‘‘ The forms of these things not only give their power, but also restrict their power, so that if properly made, that is made and shaped strictly as other things of their kind have been made and shaped, they will perform only such safe uses as their prototypes have been found to serve.” It is therefore of the utmost importance that they shall be faithfully reproduced, so that one may not have to fear the unknown “ powers ”’ which a fresh form might possess.! In this way, according to Cushing, we can account for the extraordinary persistence of the same forms among primitive peoples, including even the most minute details of the orna- mentation with which they decorate the products of their industries and arts. The Indians of British Guiana, for instance, “ show extraordinary skill in many of the things they manufacture but they never improve upon them. They make them exactly as their fathers did before them.’’2 This is not, 1 F. H. Cushing, ‘‘ Zuñi Creation Myths,” E. B. Rept., xiii. pp. 361-3. 2 Bernau, Missionary Labours in British Guiana, p. 46 (1847). 42 HOW NATIVES THINK as we have been told, merely the result of habit, and of a spirit of conservatism peculiar to these peoples. It is the direct result of active belief in the mystic properties of the things, properties connected with their shape, and which can be controlled through this, but which would be beyond the power of man to regulate, if there were the slightest change of form. The most apparently trifling innovation may lead to danger, liberate hostile forces, and finally bring about the ruin of its instigator and all dependent upon him. In the same way, any change effected by manual labour in the state of the soil, building, digging, mining, the making of a pavement or the demolition of a building, or even a slight modification in its shape by the addition of a wing, may be the cause of the greatest misfortunes. | ‘ Should anyone fall suddenly ill and die,” says De Groot, . 1 ‘“his kindred are immediately ready to impute the cause to somebody who has ventured to make a change in the estab- lished order of things, or who has made an improvement in his own property. . . . Instances are by no means rare of their having stormed his house, demolished his furniture, assailed his person. . .. No wonder Chinamen do not repair their houses until they are ready to fall and become uninhabitable.” 1 The steeple to be placed on the Catholic church in Pekin raised such a storm of protestation that the erection of it had to be abandoned. This mystic belief is intimately associated with that which the Chinese call the fungshw. But we find similar instances in other places. Thus, in the Nicobar Isles, ‘some of the chief men of Mus, Lapati, and Kenmai came and requested me to postpone fixing the beacon until the arrival of their people from Chowra, for they said that in consequence of this new work, and of a tree that had been felled down by Mr. Dobie in their graveyard, near the object, the sea was annoyed and had caused high wind and big surf, until they supposed that their friends would be drowned at sea.” 2 In Loango, “the stranger who goes away must not demolish his buildings or lay waste his plantations, but leave them just as they are. That is the reason why the natives protest when Europeans take down whole houses which they 1 The Religious System of China, i. p. 10 2 Solomon, “ Diaries Kept in Car Nicobar” J .A.I., XXXii, p. 230. VC VIT ail PRIMITIVES’ PERCEPTIONS 43 had built, to transport them elsewhere. The corner-stones and pillars at least should not be taken out of the ground. . . It is even forbidden to carry away the trunks of trees, to make excavations for mines, and so forth. A contractor exposes himself to serious trouble if, consulting his own wishes, he is so presuming as to make a new path, even if much shorter and more convenient than the one in use.’”! This is not mere misoneism, the dislike of any change which breaks established custom. With the old road, they know how matters stand, but they are ignorant of the unforeseen consequences, possibly calamitous, which might ensue upon the abandonment of it and the opening up of a fresh one. A road, like everything else, has its own peculiar mystic properties. The natives of. Loango say of an abandoned path that it is ‘ dead.” To them, as to us, such an expression is metaphorical, but in their case it is fraught with meaning. For the path, “‘in active existence ’’ has its secret powers, like houses, weapons, stones, clouds, plants, animals, and men—in short, like everything of which the primitive has a group idea. ‘ All things have an invisible existence as well as a visible one,” say the Igorots of the Philippine Islands.2 From these facts and many similar ones which we might quote, we can draw one conclusion: primitives perceive nothing in the same way as we do. The social milieu which surrounds them differs from ours, and precisely because it is different, the external world they perceive differs from that which we apprehend. Undoubtedly they have the same senses as ours—rather more acute than ours in a general way, in spite of our persuasion to the contrary—and their cerebral : structure is like our own. But we have to bear in mind that which their collective representations instil into all their perceptions. Whatever the object presented to their minds, it implies mystic properties which are inextricably bound up with it, and the primitive, in perceiving it, never AGN these from it. To him there is no phenomenon which is, strictly speaking, a physical one, in the sense in which we use the term. “The 1 Dr. Pechuél-Loesche, Die L ri a PEE, lii. 2, pp. 209-12. + Jenks, The Bontoc Igorot, p. 196 (Manila, 1905). 44 HOW NATIVES THINK rippling water, the whistling wind, the falling rain, any natural phenomenon whatever, a sound, a colour,—these things are never perceived by him as they are by us, that is, as more or less compound movements bearing a definite relation to preceding and to subsequent movements. His perceptive organs have indeed grasped the displacement of a mass of material as ours do; familiar objects are readily recognized according to previous experience; in short, all the physio- logical and psychological processes of perception have actually taken place in him as in ourselves. Its result, however, is immediately enveloped in a state of complex consciousness, dominated by collective representations. Primitives see with eyes like ours, but they do not perceive with the same minds. We might almost say that their perceptions are made up of a nucleus surrounded by a layer of varying density of repre- __ sentations which are social in their origin. And yet such a simile seems somewhat clumsy and inexact, for the primitive has not the least feeling of such a nucleus and surrounding layer ; it is we who separate them; we, who by virtue of our mental habits cannot help distinguishing them. To the primitive the complex representation is still undifferentiated. The profound difference which exists between primitive mentality and our own is shown even in the ordinary percep- tion or mere apprehension of the very simplest things. Primi- tive perception is fundamentally mystic on account of the mystic nature of the collective representations which form an integral part of every perception. Ours has ceased to be so, at any rate with regard to most of the objects which surround us. Nothing appears alike to them and to us. For people like ourselves, speaking the language familiar to us, there is insur- mountable difficulty in entering into their way of thinking. The longer we live among them, the more we approximate to their mental attitude, the more do we realize how impossible it is to yield to it entirely. It is not correct to maintain, as is frequently done, that primitives associate occult powers, magic properties, a kind of soul or vital principle with all the objects which affect their senses or strike their imagination, and that their perceptions are surcharged with animistic beliefs. It is not a question of association. The mystic properties with which things and PRIMITIVES’ PERCEPTIONS 45 beings are imbued form an integral part of the idea to the primitive, who views it as a synthetic whole. It is at a later stage of social evolution that what we call a natural pheno- menon tends to become the sole content of perception to the exclusion of the other elements, which then assume the aspect of beliefs, and finally appear superstitions. But as long as this “ dissociation ’’ does not take place, perception remains an undifferentiated whole. We might call it “ polysynthetic,” like words in the languages spoken by certain primitive peoples. In the same way, we shall find ourselves in a blind alley, whenever we propound a question in such terms as: How would the primitive’s mind explain this or that natural phenomenon? The very enunciation of the problem implies a false hypothesis. We are supposing that his mind appre- hends these phenomena like our own. We imagine that he simply perceives such facts as sleep, dreaming, illness, death, the rise and decline of the heavenly bodies, rain, thunder, etc., and then, stimulated by the principle of causality, tries to account for them. But to the mentality of undeveloped peoples, there are no natural phenomena such as we under- stand by the term. Their mentality has no need to seek an explanation of them; for the explanation is implied in the mystic elements of the collective representations of them. Therefore problems of this nature must be inverted. What we must seek is not the logical process which might have resulted in the interpretation of phenomena, for this mentality never perceives the phenomenon as distinct from the interpretation ; we must find out how the phenomenon became by degrees detached from the complex in which it first found itself enveloped, so that it might be apprehended separately, and how what originally was an integral part of it should later on have become an “ explanation.” I] The very considerable part played by collective represen- tations in the primitives’ perceptions does not result alone in impressing a mystic character upon them. The same cause leads to another consequence, and these perceptions are 46 HOW NATIVES THINK accordingly oriented differently from our own. In that which our perceptions retain, as well as in that which is disregarded, the chief determining factor is the amount of reliance that we can place upon the unvarying reappearance of phenomena in the same given conditions. They conduce to effect the maxi- mum ‘objective ”” validity, and, as a result, to eliminate everything prejudicial or merely unnecessary to this objec- tivity. From this standpoint, too, primitives do not perceive as we do. In certain cases where direct practical interests are at stake, we undoubtedly find that they pay great attention to, and are often very skilful in detecting differences in, impres- sions which are very similar, and in recognizing external signs of objects or phenomena, upon which their subsistence, and possibly even their lives, depend. (The shrewdness of the Australian aborigines in finding and profiting by the dew 1 which has fallen during the night,’ and other similar facts, are an example of this.) But, even setting aside that which these fine perceptions owe to training and memory, we still find that in most cases primitives’ perceptive powers, instead of tending to reject whatever would lessen objectivity, lay special stress upon the mystic properties, the occult forces of beings and phenomena, and are thus oriented upon factors which, to us, appear subjective, although to primitives they are at least as real as the others. This characteristic of their perceptions enables them to account for certain phenomena, the “.explana- tion’’ of which, when based solely upon mental or logical processes in the individual, does not appear adequate. It is a well-known fact that primitives, even members of communities which are already somewhat advanced, regard artificial likenesses, whether painted, carved, or sculptured, as real, as well as the individual they depict. “ To the Chinese,” says De Groot, ‘‘ association of images with beings actually becomes identification, both materially and psychically. An image, especially if pictorial or sculptured, and thus approach- ing close to the reality, is an alter ego of the living reality, an abode of the soul, nay it is that reality itself. . . . Such intense association is, in fact, the very backbone of China’s inveterate idolatry and fetish-worship.” 2? In support of his 1 Eyre, Journals of Expeditions of Discovery into Central A ustralia, ii. P. 247. 2 J. J. M. de Groot, The Religious System of China, ii. pp. 340-55. PRIMITIVES’ PERCEPTIONS 47 Statement, De Groot gives a long series of tales which seem wholly incredible, but which Chinese authors find perfectly natural. A young widow has a child by a clay statue of her husband; portraits are endued with life: a wooden dog starts running; paper animals, horses, for instance, act exactly like living animals; an artist, meeting a horse of a certain colour in the street, recognizes it as a work of his. . From these the transition to customs which are very general in China is an easy one.—Such customs as placing upon the tombs of the dead miniature figures of animals, burning paper money there, for instance. In North America, the Mandans believe that the portraits taken by Catlin are alive like their subjects, and that they rob these of part of their vitality. It is true that Catlin is inclined to draw a long bow, and his stories must be taken with a grain of salt. In this respect, however, the beliefs and sentiments he attributes to the Mandans are exactly what we find noted elsewhere in similar circumstances. ‘I know,” Says one man, ‘that this man put many of our buffaloes in his book, for I was with him, and we have had no buffaloes since to eat, it is true,”’! They pronounced me the greatest medicine-man in the world,” writes Catlin, ‘ for they said I had made living beings —they said they could see their chiefs alive in two places— those that I had made were a little alive—they could see their eyes move—could see them smile and laugh, and that if they could laugh, they could certainly speak, if they should try, and they must therefore have some life in them.” : Therefore, most Indians refused him permission to take their likenesses. It would be parting with a portion of their own substance, and placing them at the mercy of anyone who might wish to possess the picture. They are afraid, too, of finding them- selves faced by a portrait which, as a living thing, may exercise a harmful influence. | ” We had placed,” say the Jesuit missionaries, “ images of St. Ignatius and St. Xavier upon our altar. They regarded them with amazement; they believed them to be living persons, and asked whether they were ondaqui”’ (plural form of wakan, supernatural beings): “in short, that which * Catlin, The North American Indians, i. pp. 122-3 (Edinburgh, 1903). 48 HOW NATIVES THINK they recognize as superior to humanity. They inquired whether the tabernacle were their dwelling, and whether these ondagui used the adornments which they saw around the altar.” ! In Central Africa, too, ‘‘I have known natives refuse to enter a room where portraits were hanging on the walls, because of the masoka souls which were in them.’’? The same author tells the story of a chief who allowed himself to be photographed, and who, several months later, fell ill. In accordance with his request, the negative had been sent to England, and ‘‘his illness was attributed to some accident having befallen the photographic plate.” Thus the similitude can take the place of the model, and possess the same properties. In Loango, the followers of a ve ey EE Oe certain eminent wonder-worker used to make a wooden image ~— of their master, imbued it with “‘ power,” and gave it the name of the original. Possibly even they would ask their master to make his own substitute, so that after his death, as well as during his life, they could use it in performing their miracles.3 On the Slave Coast, if one of twins happens to die, the mother “|. . to give the spirit of the deceased child something to enter without disturbing the survivor, carries about, with the latter, a little wooden figure, about seven or eight inches long, roughly fashioned in human shape, and of the sex of the dead child. Such figures are nude, as an infant would be, with beads around the waist.’’4 With reference to the Bororo of Brazil we read ‘‘ they begged Wilhelm most earnestly not to let the women see the drawings he had made of the bull- roarers ; for the sight of the drawings would kill them as the real things would.” 5 Many similar instances had already been collected by Tylor.é Are these to be explained from a purely psychological point of view, as is so frequently the case, by the association of ideas ? Must we say, with De Groot, that it is impossible for 1 Ed. Thwaites, Relations des Jésuites, v. p. 256 (1633). 2 Hetherwick, ‘‘Some Animistic Beliefs of the Yaos,” /J.A.I., xxxil. pp. 89-90. 3 Dr. Pechuél-Loesche, Die Loango-Expedition, ili. 2, pp. 378-9 (1907). 4 A. B. Ellis, The Yoruba-speaking Peoples, p. 80. 5 K. von den Steinen, Unter den Naturvôlhern Zentralbräsiliens, p. 386. 6 Primitive Culture, il. pp. 169 et seq. ag ae eae ae ee TEE PRIMITIVES’ PERCEPTIONS 49 them to distinguish a mere resemblance from identity, and admit that primitives suffer from the same illusion as the child who believes her doll to be alive? First of all, however, it is difficult to decide whether the child herself is quite sure of it. Perhaps her belief is part of the game and at the same time sincere, like the emotions of grown-up people at the theatre, shedding real tears about misfortunes which they nevertheless know to be but feigned. On the contrary, it is impossible to doubt that the primitives’ beliefs which I have just mentioned are serious ; their actions prove it. How then can a portrait be “ materially and psychically ” identified with its original ? To my mind, it is not on account of a childish trust in analogy, nor from mental weakness and confusion ; it is not due to a naive generalization of the animist theory, either. It is because, in perceiving the similitude, as in looking at the original, the traditional collective representations imbue it with the same mystic elements. If primitives view the pictured resemblance differently from ourselves, it is because they view the original otherwise also. In the latter we note its objective and actual charac- teristics, and those only: the shape, size, and proportions of the body; the colour of the eyes; the facial“expression, and so forth; we find these reproduced in the picture, and there, too, we find these alone. But to the primitive, with his perceptions differently oriented, these objective features, if he apprehends them as we do, are neither the only ones nor the most important ; most frequently, they are but the symbols or instruments of occult forces and mystic powers such as every being, especially a living being, can display. As a natural consequence, therefore, the image of such a being would also present the mingling of characteristics which we term objective and of mystic powers. It will live and prove berteficial or malevolent like the being it reproduces : it will be its surrogate. Accordingly we find that the image of an unknown—and consequently dreaded—object often inspires extraordinary dread. ‘I had,” says Father Hennepin, “a pot about three feet high shaped like a lion, which we used for cooking our food in during the voyage. . . . The savages never ventured to touch it with their hands unless they had previously covered them with beaver skins. They imparted 4 50 HOW NATIVES THINK such terror of it to their wives that the latter had it fastened to the branches of a tree, for otherwise they would not have dared to sleep or even enter the hut if it were inside. We wished to make a present of it to some of the chiefs, but they would neither accept it nor make use of it, because. they feared that it concealed some evil spirit which might have killed them.” We know that these Indians in the valley of the Mississippi had never before seen a white man, or a lion, or a cooking utensil. The likeness of an animal they did not know awakened in them the same mystic fears that its appearance among them would have done. | This identification which appears so strange to us must therefore occur naturally. It does not arise out of gross mental hallucination or childish confusion of ideas. As soon as we realize how primitives view entities, we see that they © view reproductions of them in exactly the same way. If their perceptions of the originals ceased to be mystic, their images would also lose their mystic properties. They would no longer appear to be alive, but would be what they are to our minds—-merely material reproductions. In the second place, primitives regard their names as something concrete and real, and frequently sacred. Here are a few of the many proofs of it. “The Indian regards his name, not as a mere label, but as _ a distinct part of his personality, just as much as are his eyes or his teeth, and believes that injury will result as surely from the malicious handling of his name as from a wound inflicted on any part of his physical organism. This belief was found among the various tribes from the Atlantic to the Pacific.” 2 On the East African coast, ‘‘ there is a real and material connection between a man and his name, and . . . by means of the name injury may be done to the man. . . . In con- sequence of this belief the name of the king . . . is always kept secret. . . . It appears strange that the birth-name only, and not an alias, should be believed capable of carrying some of the personality of the bearer elsewhere . . . but the native . ..1 L. Hennepin, Nouveau Voyage de l'Amérique Septentrionale, pp. 366-7. ; _? J. Mooney, “The Sacred Formulas of the Cherokees,” £. B. Rept., vil. P. 343. ; À f N PRIMITIVES’ PERCEPTIONS 51 view seems to be that the alias does not really belong to the man.””"1 | Accordingly all kinds of precautions become necessary. A man will avoid uttering his own name? and the names of others, while the names of the dead, above all, will never be pronounced ; very frequently, too, even ordinary words in which the name of a dead person is implied will fall into desue- tude. Alluding to a name is the same thing as laying hands on the very person or being that bears thename. It is making an attack upon him, outraging his individuality, or again, it is invoking his presence and forcing him to appear, a proceeding which may be fraught with very great danger. There are excellent reasons, therefore, for avoiding such a practice. “When they (the Santals) are hunting and see a leopard or a tiger they will always call the attention of their companions to the fact by calling out ‘a cat,’ or some similar name.” 3 With the Cherokees, too, “it is never said that a person has been bitten by a snake, but that he has been ‘ scratched by a brier.” In the same way, when an eagle has been shot for a ceremonial dance, it is announced that ‘ a snow-bird has been killed,’ the purpose being to deceive the rattlesnake or eagle spirits which might be listening,’ 4 The Warramunga, instead of mentioning the snake Wollunqua by its name when speaking of it, call it Urkulu nappaurima, “‘ because,” say they, “if they were to call it too often by its right name, they would lose their control over it, and it would come out and eat them all up.” 5 At the beginning of a fresh epoch in his life,—at his initia- tion, for instance—an individual receives a new name, and it is the same when he is admitted to a secret society. A town changes its name to indicate that it is commencing a new era ; Yedo becomes Tokyo.6 A name is never a matter of indiffer- ence; it implies a whole series of relationships between the man who bears it and the source whence it is derived. ‘ A name implies relationship, and consequently protection ; * A. B. Ellis, The Ewe-speaking Peoples, pp. 98-9. 2 Rivers, The Todas, D: 627: | 3 Bodding, ‘ On Taboo Customs amongst the Santals,” Journal of the Asiatic Society of Bengal, iii. p. 20 (1898). _-4 J. Mooney, The Sacred Formulas of the Cherokees, p. 352. 5 Spencer and Gillen, The Northern Tribes of Central Australia, DP. 227. 6 Chamberlain, Things Japanese, p. 344 (1902). . | 52 HOW NATIVES THINK favour and influence are claimed from the source of the name, whether this be the gens or the vision. A name, therefore, shows the affiliation of the individual ; it grades him, so to speak.’’! In British Columbia, ‘ names, apart from the staz or nickname, are never used as mere appellations to distin- guish one person from another, as among ourselves, nor do they seem to have been used ordinarily as terms of address. They are primarily terms of relation or affiliation, with his- toric and mystic reference. They were reserved for special — and ceremonial occasions. The ordinary terms of address ; among the Salish tribes, as among other primitive peoples, were those expressive of age.’2 With the Kwakiutl, ‘each clan has a certain limited number of names. Each individual « has only one name at a time. The bearers of these names form the nobility of the tribe. When a man receives the totem of. his father-in-law, he at the same time receives his name, while. the father-in-law gives up the name, and takes what is called ‘an old man’s name,’ which does not belong to the names constituting the nobility of the tribe.”3 Finally, De Groot « notes that ‘‘ the Chinese have a tendency to identify names w with the persons who bear them, a tendency which may be. classed on a level with their inability, already illustrated by \ numerous instances, of clearly discriminating between ( semblances or symbols and the realities which these call to w mind.” 4 | This last comparison seems perfectly correct, to my mind, « and I think as De Groot does, that the same cause may account M for both tendencies. This cause is not to be found in a childish ~ association of ideas, however. It is in the collective represen- — tations which, forming an integral part of their perception of entities, form an integral part also of their perception of the likeness and the name which betokens them. The reality of the similitude is of the same kind as that of the original—that M is, essentially mystic, and it is the same with the reality of the name. The two cases are alike except in one point—that t Dorsey, ‘‘ Siouan Cults,”’ E. B. Rept., xi. p. 368. i 2 Hill Tout, “ Ethnology of the StatlumH of British Columbia,” J.AJ., © XXXV. P. 152. | 3 F. Boas, “ The North-western Tribes of Canada,’’ Reports of the British « Association, p. 675 (1898). 4 The Religious System of China, i. p. 212. PRIMITIVES’ PERCEPTIONS 53 which appeals to the sight in the first case, appeals to the hearing in the second, but otherwise the process is identical. The mystic properties in the name are not separated from those in the beings they connote. To us the name of a person, an animal, a family, a town, has the purely external signifi- cance of a label which allows us to discern without any possi- bility of confusion who the person is, to what species the animal belongs, which family and which town it is. To the primitive, however, the designation of the being or object, which seems to us the sole function of the name, appears a mere accessory and of secondary importance : many observers expressly state that that is not the real function of the name. To make up for this, there are very important functions of which our names are deprived. The name expresses and makes real the relationship of the individual with his totemic group; with the ancestor of whom he is frequently a rein- carnation ; with the particular totem or guardian angel who has been revealed to him in a dream ; with the invisible powers who protect the secret societies to which he belongs, etc. How does this arise? Evidently because beings and objects do not present themselves to the primitive’s mind apart from the mystic properties which these relations involve.\ As a natural consequence, names derive their characteristics from the characteristics of these same beings and objects. The name is mystic, as the reproduction is mystic, because the perception of things, oriented differently from our own, through the _ collective representations, is mystic. We can therefore extend also to names Cushing’s acute reflections, already quoted, with regard to the forms of objects. Names condition and define the occult powers of | the beings who participate in them. Hence are derived the feelings and fears they awaken, and the precautions to which these fears lead. The problem is not to discover how the simple term ‘‘ is associated ’’ with mystic elements which are never separable from it in the minds of primitives. What is given is the ensemble of collective representations of a mystic nature expressed by the name. The actual problem is to ascertain how these collective representations become gradually impaired and dissociated, how they have assumed the form of “ beliefs ”” less and less closely ‘‘ attached ’’ to the name, until 54 HOW NATIVES THINK the moment arrives when, as with us, it serves but as a distinctive designation. The primitive is, as we know, no less careful about his shadow than he is about his name or his counterfeit present- ment. If he were to lose it he would consider himself hope- lessly endangered. Should it come into the power of another, : he has everything to dread. Folklore of all countries has made us familiar with facts of this kind; we shall cite but a © few of them only. In the Fiji Islands, as in many places inhabited by people of a similar stage of development, it is a | mortal insult to walk upon anybody else’s shadow. In East q Africa, murders are sometimes committed by means of a knife . or nail thrust through the shadow of a man; if the guilty person is caught in the act he is executed forthwith. Miss Kingsley in reporting this fact, shows clearly to what extent © the West African negroes dread the loss of their shadow. ‘It strikes one as strange,’’ she writes, ‘‘ to see men who have been © walking, say, through forest or grass land, on a blazing hot morning quite happily, on arrival at a piece of clear ground or _a village square, most carefully go round it, not across, and you © will soon notice that they only do this at noontime, and learn that they fear losing their shadow. I asked some Bakwiri I once came across who were particularly careful in this matter, w why they were not anxious about losing their shadows when ~ night came down and they disappeared in the surrounding darkness, and was told that was all right, because at night all © shadows lay down in the shadow of the Great God, and so got « stronger. Had I not seen how strong and how long a shadow, ~ be it of man or tree or of the great mountain itself, was in the u early morning time ?’”’ ! De Groot notes similar precautions in China. ‘‘ When the | lid is about to be placed on the coffin, most of the bystanders | not belonging to the nearest kindred retire a few steps, or even « make off for the side apartments, as it is dangerous to health and detrimental to good luck to have one’s shadow enclosed in ~ a coffin.’”’2 What then zs the shadow? It is not the exact | | À equivalent of what we call the soul; but it is of the nature of « the soul, and where the soul is represented as multiple, the 1 Mary Kingsley, West African Studies, p. 176. 2 J. J. M. de Groot, The Religious System of China, i. pp. 94, 240. PRIMITIVES’ PERCEPTIONS shadow (according to Miss Kingsley) is sometimes one of the souls. On his side, De Groot says: “ We find nothing in the books of China which points positively to an identification of shadows and souls.’’! But, on the other hand, ghosts have no shadows. And De Groot concludes by saying: that “the shadow is a part of the personality which has an im- mense influence on his destiny,” a characteristic which applies equally, as we have seen, to a person’s picture or his name. I shall therefore refer it to the same theory. If we ask ourselves: how has the primitive come to associate with the idea of his shadow beliefs which we find to be almost universal ? we might reply by an ingenious explanation, and one which would be psychologically probable, but it would be unsound, because the problem cannot be propounded in such terms as these. To enunciate it thus would be to imply that the idea of his shadow to the primitive is the same as to us, and the rest is superimposed: Now it really is nothing like that. The perception of the shadow, as of the body itself, like that of the image or the name, is a mystic perception, in which that which we properly call the shadow,—the design upon the ground of a figure which recalls the form of a being or object lighted from the opposite side—is only one element among many. We have not to discover how the perception of the shadow has been placed in juxtaposition or united with such and such a representation: these indeed form an integral part of the perception, so far as we can trace it in past obser- vations. For this reason I should be prepared to take up a counter-position to that of De Groot. ‘The Chinese,” he says, ‘‘ are even to these days without ideas of the physical causation of shadows. . . . They must needs see in a shadow something more than a negation of light.’”’2 I, on the con- trary, should say: the Chinese, having a mystic perception of the shadow, as participating in the life and all the properties of the tangible body, cannot represent it as a mere “‘ negation of light.” To be able to see a purely physical phenomenon in the production of the shadow, it would be necessary to have an idea of such a phenomenon, and we know that such an idea is lacking to the primitive. In undeveloped communities, there 1 J. J. M. de Groot, The Religious System of China, ii. p. 83. 2 Tbid. 56 HOW NATIVES THINK is no perception unaccompanied by mystic qualities and occult properties, and why should the shadow be any exception ? Finally, the same considerations apply equally to another class of phenomena—dreams—which occupy an important place in the primitive mind. To primitives the dream is not, as it is to us, simply a manifestation of mental activity which occurs during sleep, a more or less orderly series of repre- sentations to which, when awake, the dreamer would give no credence, because they lack the conditions essential to objective validity. This last characteristic, though it does not escape the primitives, seems to interest them but slightly. On the other hand, the dream, to them, is of far greater significance than to us. It is first a percept as real as those of the waking state, but above all it is a provision of the future, a communi- : cation and intercourse with spirits, souls, divinities, a means of: establishing a relation with their own special guardian angel, and even of discovering who this may be. Their confidence in the reality of that which the dreams makes known to them is very profound. Tylor, Frazer, and the representatives of the English school of anthropology have brought together a vast number of facts which bear witness to this, collected by investigators of primitive peoples of the most diverse types. Shall I, too, quote some? In Australia ‘ Sometimes a man dreams that someone has got some of his hair or a piece of his food, or of his ’possum rug, or indeed anything almost that he has used. If he dreams this several times he feels sure of it and calls his friends together, and tells them of his dreaming too much about ‘ that man,’ who must have something belong- ing to him. . . . Sometimes natives only know about having their fat taken out by remembering something of it as in a dream.”’ : To the North American Indians, dreams, natural or induced, have an importance which it would be difficult to overestimate. ‘‘ Sometimes it is the rational mind which is wandering, whilst the mind which feels continues to animate the body. Sometimes the familiar spirit gives wholesome advice upon what is about to happen; and sometimes it is a visit from the soul of the object of which one dreams. But in 5 Howitt, ‘ On Australian Medicine-men,” /.4.1., xvi. I, pp. 29-30. PRIMITIVES’ PERCEPTIONS 57 whatsoever fashion the dream may be conceived, it is always regarded as a sacred thing, and as the most usual method employed by the gods of making their will known to men. . .. Frequently it is an order from the spirits.” In Lejeune’s Relations de la Nouvelle France, it is stated that the dream is “the god of the heathens’’; and an observer of our own times says: “‘ Dreams are to savages what the Bible is to us, the source of Divine revelation—with this important difference that they can produce this revelation at will by the medium of dreams.’’? Consequently the Indian will at once carry out . what has been commanded or simply indicated to him in a dream. Mooney tells us that among the Cherokees, when a man dreams that he has been bitten by a snake, he must follow the same treatment as if he had really been bitten, for it is a witch-snake that has done the injury, and if he did not, swelling and ulceration would ensue, possibly even many years later.3 In the Relations de la Nouvelle France we read that “a warrior, having dreamed that he had been taken prisoner in battle, anxious to avert the fatal consequences of such a dream, called all his friends together and implored them to help him in this misfortune. He begged them to prove themselves true friends by treating him as if he were an enemy. They therefore rushed upon him, stripped him naked, fettered him, and dragged him through the streets with the usual shouts and insults, and even made him mount the scaffold. . . . He thanked them warmly, believing that this imaginary captivity would ensure him against being made prisoner in reality. . . . In another case, a man who had dreamed he saw his hut on fire, did not rest till he had witnessed its burning in reality. . . . A third, not believing that his dream would be sufficiently realistic if he were burned in effigy insisted on having fire applied to his legs, as is done with captives about to undergo capital punishment. . . . It was fully six months before he recovered from the burns.”’ 4 The Malays of Sarawak never doubt their blood-relation- 1 Charlevoix, Journal d'un Voyage dans l'Amérique Septentrionale, iii. PP- 353-5. 2 A. Gatschet, The Klamath Language, p. 77 (Contributions to North American Ethnology, ii. 1). 3 ** Myths of the Cherokee,”’ E. B. Rept., xix. p. 295. 4 Années, pp. 46-8 (1661-2), 58 HOW NATIVES THINK ship with a certain animal, if they have dreamed about it. ‘‘Wan’s great-great-grandfather became blood-brother to a crocodile. ... Wan had several times met this crocodile in dreams. Thus in one dream he fell into the river where there were many crocodiles about. He climbed on to the head of one which said to him, ‘ Don’t be afraid,’ and carried him to the bank. Wan’s father had charms given him by a crocodile and would not on any account kill one, and Wan clearly regards himself as being intimately related to crocodiles in general.” 1 In short, to conclude with a peculiarly happy dictum of Spencer and Gillen: ‘‘ What a savage experiences during a dream is just as real to him as what he sees when he is awake.’’ 2 To explain these phenomena, are we to rely upon the current theory which refers them to a psychological illusion : constantly obtaining among primitives? They would be unable to distinguish an actual perception from one which, though powerful, is merely imaginary. In all cases of life-like representation, the belief in the objectivity of this representa- tion would appear. Accordingly, the apparition of a dead person would induce the belief that he was actually present. The representation of one’s own self, in a dream, acting, travelling, conversing with persons who are at a distance or who have disappeared, would convince one that the soul does indeed leave the body during sleep and travels whither it feels conscious of going. ‘‘ The supreme confusion in the thought of non-civilized individuals,” says Powell, ‘‘ is the confusion of the objective with the subjective.” Without disputing the general accuracy of the psycho- logical law which is invoked here, I should like to point out that it does not wholly account for the way in which primitives represent their dreams, and the use to which they put them. In the first place, they distinguish very clearly between per- ceptions which come to them through dreams from those they receive in the waking state, however similar they may other- wise be. They even recognize different categories of dreams, t Hose and Macdougall, “ Relations between Men and Animals in Sara- walk? FAL xxi. DA HEOT. 2 Spencer and Gillen, The Northern Tribes of Central Australia, p. 451. ee PRIMITIVES’ PERCEPTIONS 59 and attribute a varying degree of validity to them. ‘The Ojibbeways have divided dreams into various classes, and given each a special name. The excellent Bishop Baraga, in his lexicon of that language, has collected the Indian names for a bad dream, an impure dream, an ominous dream, as well as for a good or happy dream.”’! ‘The Hidatsa have much faith in dreams, but usually regard as oracular only those which come after prayer, sacrifice, and fasting.’’2 It is therefore with full knowledge of the circumstances and due reflection that primitives accord the one kind of perception as complete a credence as the other. Instead of saying, as people do, that primitives believe in what they perceive in the dream, although it is but a dream, I should say that they believe in it because it is a dream. The “illusion ” theory does not suffice. How do we account for the fact that, know- ing well that the dream 7s a dream, they should nevertheless rely upon it? This cannot be explained as a mere psycho- logical process effected in the individual. Here again, we are obliged to take into account the collective representations which make both the perception and the dream something entirely different for the primitive from what they would be for us. Our perception is directed towards the apprehension of an objective reality, and this reality alone. It eliminates all that might be of merely subjective importance, and in this it is in contrast with the dream. We do not understand how any- thing that is seen in a dream can be placed on a par with that which we see in the waking state: and if such a thing does occur we are forced to believe that it is the result of a very strong psychological illusion. But. with the primitives there is no such violent contrast as this. Their perception is oriented in another fashion, and in it that which we call objec- tive reality is united and mingled with, and often regulated by, mystic, imperceptible elements which we nowadays characterize as subjective. In this way, in fact, it is closely related to the dream. Or, if we prefer to put it thus, we may say that their dream is a perception like the others. It isa compound into which the same elements enter, which awakens t Kohl, Kitcht Gama: Wanderings round Lake Superior, p. 236. 3 Dorsey, ‘‘ Siouan Cults,” E. B. Rept., xi. p. 516. 60 HOW NATIVES THINK the same sentiments, and which even impels to action. Thus the Indian who has a dream and risks his existence upon its truth, is not ignorant of the difference between this dream and a similar perception which he might have in his waking state. But, since his perception in the waking state and in the dream are alike mystic, this difference means very little to him. In our eyes, the objective reality of the perception is the measure of its validity ; in his, such a consideration is only secondary, or rather, is of no importance at all. That which to us is perception is to him mainly the com- munication with spirits, souls, invisible and intangible mysterious powers encompassing him on all sides, upon which his fate depends, and which loom larger in his consciousness than the fixed and tangible and visible elements of his repre- sentations. He has therefore no reason to depreciate the dream, and consider it as a subjective and dubious representa- ~ tion, in which he must place no trust. The dream is not a form of inferior and illusory perception. On the contrary, it is a highly favoured form, one in which, since its material and tangible elements are at a minimum, the communication with invisible spirits and forces is most direct and most complete. This accounts for the confidence which the primitive has in his dreams, a confidence which is at least as great as that he accords his ordinary perceptions. It accounts also for his seeking after means of procuring dreams which shall be revela- tory and, among the North American Indians, for instance, for the whole technique of securing the sincerity and validity of dreams. Thus the young man, arrived at the age of initia- tion, who is going to try and see in a dream the animal which will be his guardian angel, his personal totem, has to prepare himself for this purpose by carrying out a series of observances. “ He first purifies himself by the mpi or steam bath, and by fasting for a term of three days. During the whole of this time, he avoids women and society, is secluded in his habits and endeavours in every way to be pure enough to receive a revelation from the deity whom he invokes” . . . then he subjects himself to various tortures “until the deities have vouchsafed him a vision or revelation.” 1 t Dorsey, “ Siouan Cults,” E. B. Rept., xi. pp. 436-7. OE ES ON PE PRIMITIVES’ PERCEPTIONS 61 This, too, accounts for the deference and respect shown to dreamers, seers, prophets, sometimes even to lunatics. A special power of communicating with invisible reality, that is, a peculiarly privileged perception, is attributed to them. All these well-known facts naturally result from the orientation of the collective representations which obtains in primitive peoples, and which endows with mysticism both the real world in which the “ savage’ dwells, and his perception of it. III Further differences between the primitives’ perception and our own arise out of this mystic character. To us one of the essential signs by which we recognize the objective validity of a perception is that the being or phenomenon perceived appears to all alike under the same conditions. If, forinstance, one person alone among a number present hears a certain sound repeatedly, or sees an object close by, we say that he or she is subject to delusions, or has a hallucination. Leibniz, Taine, and many others have insisted upon the agreement between the subjects who are perceiving as a means of distin- guishing between real ‘‘ and imaginary phenomena.”’ Current opinion on this point, too, is wholly on the side of the philo- sophers. With the primitives, on the contrary, it constantly happens that beings or things manifest themselves to certain persons to the exclusion of others who may be present. No one is astonished at this, for all regard it as perfectly natural. Howitt writes, for instance: ‘‘ Of course, the Ngarang was invisible to all but the wivarap (medicine-man).” 1 A young medicine-man in training, who is telling of his initiation, remarks: ‘‘ After that I used to see things that my mother could not see. When out with her I would say ‘ Mother, what is that out there yonder?’ She used to say ‘ Child, there is nothing.” These were the jir (or ghosts) which I began to see.’’? The aborigines observed by Spencer and Gillen think that during the night the sun visits the place where it arises in the morning, ‘and that it might actually be seen at night times by . . . clever medicine-men, and the fact that it can t Howitt, ‘‘ On Some Australian Medicine-men,” /.A.1., xvi. i. p. 42. 2 Ibid., p. 50. 62 HOW NATIVES THINK not be seen by ordinary persons only means that they are not gifted with sufficient power, and not that it is not there.” In their case, as with many other aggregates of the same stage of development, the medicine-man extracts from the body of the sufferer a small object only visible to the operator. “‘ After much mysterious searching he finds and cuts the string which is invisible to everyone except himself. There is not a doubt amongst the onlookers as to its having been there.’’? In the form of witchcraft which the Australian aborigines called ‘““ pointing the death bone,’’ a complicated series of operations would be carried on without anyone’s perceiving them. ‘ The blood of the victim, in some fashion which is unperceived, flows from him to the medicine-man, and thence to the receptacle where it is collected; at the same time, by a corresponding movement a bone, a magic stone proceeds from the body of the sorcerer to the body of his victim—-still invisibly—-and, entering there, induces a fatal malady.”’ 3 We find the same beliefs in Eastern Siberia. ‘In the Alarsk department of the Government of Irkutsk... if . anyone’s child becomes dangerously ill, the Buryats.. . believe that the crown of his head is being sucked by Onok- hoi, a small beast in the form of a mole or cat. . . . No one except the shaman can see this beast.”’ 4 In North America, among the Klamaths of Oregon, the kiuks (medicine-man) who is called to treat a case of disease must consult the spirits of certain animals. ‘‘ Such persons only as have been trained during five years for the profession of conjurers can see these spirits, but by them they are seen as clearly as we see the objects around us.’’5 ‘‘ Dwarfs can be seen only by those initiated into the mysteries of witchcraft.’’ 6 Among the Tarahumares “large serpents, which only the shaman can see, are thought to live in the rivers. They have horns and very big eyes.” 7 ‘The great Hikuli’’ (a sacred plant personified) “‘ eats with the shaman, who alone is able to t The Native Tribes of Central Australia, pp. 561-2. 2 Ibid., p. 532. 3 W. E. Roth, Ethnological Studies among the N.W. Central Queensland Aborigines, No. 264. _ 4 V. Mikhailovski, Shamanism in Siberia and European Russia, analysed 100 I RIVE D. 00 UCI Fp 14%: 5 À. Gatschet, The Klamath Language, p. xcviii. «Ibid; Pickem 7 C. Lumholtz, Unknown Mexico, i. p. 340. PRIMITIVES’ PERCEPTIONS 63 see him and his companions.’’! In one of the Huichol cere- monies, the heads of the does are placed with the heads of the bucks, because they, too, have horns, “though only the shaman sees them.”’ 2 All such phenomena are to be expected if it be true that the perception of primitives is oriented differently from our own, and not pre-eminently concerned, as ours is, with the characteristics of the beings and manifestations which we call objective. To them the most important properties of the beings and objects they perceive, are their occult powers, their mystic qualities. Now one of these powers is that of appearing or not appearing in given circumstances. Either the power is inherent in the subject who perceives, who has been prepared for it by initiation, or else holds it by virtue of his participa- tion in some superior being, and so on. In short, mystic relations may be established between certain persons and certain beings, on account of which these persons are exclu- sively privileged to perceive these beings. Such cases are analogous to the dream. The primitive, far from regarding the mystic perception in which he has no part, as suspect, sees init, as in the dream, a more precious, and consequently more significant communication with invisible spirits and forces. IV Conversely, when collective representations imply the presence of certain qualities in objects, nothing will persuade the primitives that they do not exist. To us, the fact that we do not perceive them there is decisive. It does not prove to them that they are not there, for possibly it is their nature not to reveal themselves to perception, or to manifest themselves in certain conditions only. Consequently, that which we call experience, and which decides, as far as we are concerned, what may be admitted or not admitted as real, has no effect upon collective representations. Primitives have no need of this experience to vouch for the mystic properties of beings and objects: and for the same reason they are quite indifferent to the disappointments it may afford. Since experience is 1 C, Lumholtz, Unknown Mexico, p. 372. 2 Id., Symbolism of the Huichol Indians, p. 68. 64 HOW NATIVES THINK limited to what is stable, tangible, visible, and approachable in physical reality, it allows the most important of all, the occult powers, to escape. Hence we can find no example of the non-success of a magic practice discouraging those who believe in it. Livingstone gives an account of a prolonged discussion which he had with the rain-makers, and ends by saying: ‘I have never been able to convince a single one of them that their arguments are unsound. Their belief in these ‘charms’ of theirs is unbounded.’’! In the Nicobar Islands, “the people in all the villages have now performed the cere- mony called tanangla, signifying either ‘support’ or ‘ pre- vention.’ Its object is to prevent illness caused by the north- east monsoon. Poor Nicobarese! They do the same thing year after year, but to no effect.” 2 Experience is peculiarly unavailing against the belief in the virtues of “‘ fetishes ’’ which secure invulnerability : a method of interpreting what happens in a sense which favours the belief is never lacking. In one case an Ashanti, having pro- cured a fetish of this kind, hastened to put it to the proof, and received a gunshot wound which broke his arm. The “ fetish man ’’ explained the matter to the satisfaction of all, saying that the incensed fetish had that moment revealed the reason to him. It was because the young man had had sexual rela- tions with his wife on a forbidden day. The wounded man confessed that this was true, and the Ashantis retained their convictions.3 Du Chaillu tells us that when a native wears an iron chain round his neck he is proof against bullets. If the charm is not effectual, his faith in it remains unshaken, for then he believes that some maleficent wonder-worker has produced a powerful ‘‘ counter-spell,”’ to which he falls a victim.4 Elsewhere he says: ‘‘ As I came from seeing the king, I shot at a bird sitting upon a tree, and missed it. I had been taking quinine, and was nervous. But the negroes standing around at once proclaimed that this was a fetish-bird, and therefore I could not shoot it. I fired again, and missed again. Hereupon they grew triumphant in their declarations, while I. . . loaded again, took careful aim, and to my own t Missionary Travels, pp. 24-5 (1857). 2 Solomon, ‘‘ Diaries kept in Cap Nicobar,” J.A.I., xxxii. p. 213. 3 Bowditch, Mission to Ashanti, p. 439. + Explorations and Adventures in Equatorial Africa, p. 338. PRIMITIVES’ PERCEPTIONS 65 satisfaction and their dismay, brought my bird down. Immediately they explained that I was a white man, and not entirely amenable to fetish laws ; so that I do not suppose my shot proved anything to them after all.” ! It is the same in Loango. ‘I had been presented,” writes Pechuél-Loesche, ‘‘ with a very fine collar, made of hair from the tail of an elephant . . . and adorned with teeth from a sea-fish and a crocodile. These teeth were to preserve me from any danger connected with water. . . . It frequently hap- pened that my boat was upset when I was crossing the bar, and one day I had great difficulty in reaching the shore. I was told quite seriously that it was the teeth alone that had saved me, for without them my swimming powers would not have sufficed to help me clear the heavy breakers. JI was not wearing the collar, but its efficacy was in no manner of doubt from that fact’? The fetish and the medicine-man always have the last word. | Jo Primitive man, therefore, lives and acts in an environment of beings and objects, all of which, in addition to the properties that we recognize them to possess, are endued with mystic attributes. He perceives their objective reality mingled with another reality. He feels himself surrounded by an infinity of imperceptible entities, nearly always invisible to sight, and always redoubtable : ofttimes the souls of the dead are about him, and always he is encompassed by myriads of spirits of more or less defined personality. It is thus at least that the matter is explained by a large number of observers and anthro- pologists, and they make use of animistic terms to express this. Frazer has collected many instances which tend to show that this phenomenon obtains everywhere among undeveloped peoples.3 Is it necessary to quote some of them? “ The Orâon’s imagination tremblingly wanders in a world of ghosts. Every rock, road, river, and grove is haunted.” .. . Some- times, too, there are ‘‘ malignant spirits.’ 4 Like the Santals, Mundas, and the Ordons of Chota-Nagpur, ‘the Kadars believe themselves to be compassed about by a host of invisible powers, some of whom are thought to be the spirits t Explorations and Adventures in Equatorial Africa, p. 179. 2 Die Loango-Expedition, ili. 2, Pp. 352. 3 The Golden Bough (2nd edit.), ili. pp. 41 et seq. 4 Risley, Tribes and Castes of Bengal, 11, pp. 143-5. 5 66 - HOW NATIVES THINK of departed ancestors, while others seem to embody nothing more definite than the vague sense of the mysterious and uncanny with which hills, streams, and lonely forests inspire the savage imagination. . . . Their names are legion, and their attributes barely known.’”’! In Korea, “ spirits occupy every quarter of heaven and every foot of earth. They lie in wait for a man along the wayside, in the trees, on the rocks, in the mountains, valleys, and streams. They keep him under a constant espionage day and night. . . . They are all about him, they dance in front of him, follow behind him, fly over his head and cry out against him from the earth. He has no refuge from them even in his own house, for there they are plastered into or pinned on the walls or tied to the beams. . . . Their ubiquity is an ugly travesty of the omnipresence of God.’ In China, according to the ancient doctrine, “ the universe is filled up in all its parts with legions of shen and kwei. . . . Every being and every thing that exists is animated either by a shen, or bya kwet, or by ashen and a kweï together.” 3 With the Fang of East Africa, ‘spirits are everywhere; in - rocks, trees, forests, and streams; in fact, for the Fang, this life is one continual fight against spirits corporal and spiritual.’ 4 ‘In every action of his daily life,” writes Miss Kingsley, ‘‘ the African negro shows you how he lives with a great, powerful spirit world around him. You will see him before starting out to hunt or fight rubbing medicine into his weapons to strengthen the spirits within them, talking to them the while ; telling them what care he has taken of them, reminding them of the gifts he has given them, though these gifts were hard for him to give, and begging them in the hour of his dire necessity not to fail him. You will see him bending over the face of a river talking to its spirit with proper incan- tations, asking it when it meets a man who is an enemy of his to upset his canoe, or drown him, or asking it to carry down with 7 some curse to the village below which has angered him.” Miss Kingsley lays great stress upon the homogeneity of the t Risley, Tribes and Castes of Bengal, i. p. 369. 2 G. H. Jones, ‘‘ The Spirit Worship in Korea,” Transactions of the Korea Byanch of the Royal Asiatic Society, ii. i. p. 58 Py M de Groot, The Religious System of China, iv. p. 51. 4 Bennett, “‘ Ethnographical Notes on the Fang,” /.A.1., xxix. p. 87. 5 West African Studies, p. 110. PRIMITIVES’ PERCEPTIONS 67 African native’s representations of everything. ‘ The African mind naturally approaches all things from a spiritual point of view . . . things happen because of the action of spirit upon spirit.” When the doctor applies a remedy “the spirit of the medicine works upon the spirit of the disease.’’ The purely physical effect is beyond the power of conception unless it be allied with the mystic influence. Or rather, we may say that there is no really physical influence, there are only mystic ones. Accordingly it is almost impossible to get these primi- tives to differentiate, especially when it is a case of an accu- sation of murder through the practice of witchcraft, for instance. Here is a typical case. ‘I explain to my native questioner,’’ says Nassau, ‘‘ that if what the accused has done in fetich rite with intent to kill, had any efficiency in taking away life, I allow that he shall be put to death; if he made only fetiches, even if they were intended to kill, he is not guilty of this death, for a mere fetich cannot kill. But if he used poison, with or without fetich, he is guilty.” ‘ But even so,” adds Nassau, “‘ the distinction between a fetich and a poison is vague in the thought of many natives. What I call a ‘ poison’ is to them only another material form of a fetich power, both poison and fetich being supposed to be made efficient by the presence of an adjuvant spirit.””2. This means that to their minds the mere fetich kills as certainly as the poison does. More certainly even; for the poison kills only by virtue of a mystic power of which, in certain circum- stances, it may be deprived. The idea of its physical properties which is so clear to the European mind, does not exist for the African. We thus have good authority for saying that this mentality differs from our own to a far greater extent than the language used by those who are partisans of animism would lead us to think. When they are describing to us a world peopled by ghosts and spirits and phantoms for primitives, we at once realize that beliefs of this kind have not wholly disappeared even in civilized countries. Without referring to spiritualism, we recall the ghost-stories which are so numerous in our folk- lore, and we are tempted to think that the difference is one of degree only. Doubtless such beliefs may be regarded in our t West African Studies, p. 330. 2 Fetichism in West Africa, p. 263. 68 HOW NATIVES THINK communities as a survival which testifies to an older mental condition, formerly much more general. But we must be careful not to see in them a faithful, though faintly outlined, reflection of the mentality of primitives. Even the most uneducated members of our societies regard stories of ghosts and spirits as belonging to the realm of the supernatural : between such apparitions and magical influences and the data furnished by ordinary perception and the experience of the broad light of day, the line of demarcation is clearly defined. Such a line, however, does not exist for the primitive. The one kind of perception and influence is quite as natural as the other, or rather, we may say that to him there are not two kinds. The superstitious man, and frequently also the religious man, among us, believes in a twofold order of reality, the one visible, palpable, and subordinate to the essential laws of motion; the other invisible, intangible, “ spiritual,” form- ing a mystic sphere which encompasses the first. But the primitive’s mentality does not recognize two distinct worlds in contact with each other, and more or less interpenetrating. © To him there is but one. Every reality, like every influence, is mystic, and consequently every perception is also mystic. CHAPTER II THE LAW OF PARTICIPATION I Ir the primitives’ collective representations differ from ours through their essentially mystic character ; if their mentality as I have endeavoured to show, is oriented in another direc- tion than our own; we must admit, too, that their repre- sentations are not connected with each other as ours would be. Must we then infer that these representations obey some other system of logic than the one which governs our own understanding ? That would be going too far, for such a hypothesis would exceed that which facts warrant us in affirming. Nothing proves that the connections of collective representations must depend solely upon laws of a logical kind. Moreover, the idea of a logic different from our own could only provide us with a negative concept, devoid of mean- ing. Now as a matter of fact we can at least endeavour to comprehend how representations are connected in the minds of primitives. We understand their language, we make bargains with them, we succeed in interpreting their insti- tutions and their belief: all this shows that there is a possible transition, a practicable method of communication between their mentality and our own, With these exceptions, however, the mentalities are different, and the disparity becomes the more perceptible as the comparative study advances, and the documentary evidence admits of its being extended. The explorer who travels rapidly through communities of an uncivilized type has no time to probe into this problem; in fact, he scarcely ever thinks of it as such. First, he ascertains the curious persistence of certain traits of human nature, manifesting themselves in the most varied conditions, and he then expresses 70 HOW NATIVES THINK his surprise when brought face to face with methods of think- ing and acting of which both the origin and reason escape him. He leaves to his reader the task of discovering how these successive impressions may be reconciled, or else he con- fines himself to some general ‘‘ explanations ’’ derived from traditional psychology and logic, if he happens to possess a smattering of either. But when we listen to observers who have lived for a longer time among primitives, and especially those who have endeavoured to penetrate their method of thinking and feeling, we hear something quite different. Whether it be the North American Indians (of whom Cushing and Powell tell us), the negroes of the French Congo (studied by Miss Kingsley), the Maoris of New Zealand (known to Elsdon Best), or any other “primitive” people whatever, a “ civilized being ’’ can never expect to see his thought following exactly the same course as that of the primitive, nor to find again the path by which the latter has travelled. “The mentality of the Maori is of an intensely mystical nature. ... We hear of many singular theories about Maori beliefs and Maori thought, but the truth is that we do not understand either, and, what is more, we never shall. We shall never know the inwardness of the native mind. For that would mean retracing our steps for many centuries, back into the dim past, far back to the time when we also possessed the mind of primitive man. And the gates have long closed on that hidden road.” # Cushing had acquired a kind of mental “ naturalization ” among the Zuñis. Not content with living with, and like, them for many years, he had made their religious rulers adopt him and admit him to their secret societies; in their sacred ceremonies he, like their priests, had his own rôle, and carried it out. The unfortunately rare works of his which have been published give us, however, the feeling of a form of mental activity with which our own would never exactly correspond. Our habits of thought are too far removed from those of the Zuñis. Our language, without which we can conceive nothing, and which is essential to our reasoning, makes use of categories which do not coincide with theirs. Lastly and t Elsdon Best, “ Maori Medical Lore,’’ Journal of the Polynesian Society, xiii. P. 219 (1904). THE LAW OF PARTICIPATION TI chiefly, the ambient social reality, of which the collective representations and, to a certain extent, the language, are functions, is with them too far removed from our own case. Thus the mentality of inferior peoples, though not so impenetrable as it would be as if it were regulated by a logic | different from our own, is none the less not wholly com- prehensible to us. We are led to imagine that it does not exclusively obey the laws of ow logic, nor possibly of any laws which are wholly logical. An analysis of some characteristic phenomena may perhaps serve to throw light on this point. Observers have frequently collected the arguments, or rather, the connections between representations which have seemed strange and inexplicable to them. I shall quote but a few instances. ‘‘ A certain drought at Landana was attri- buted to the missionaries wearing a certain kind of cap during the services: the natives said that this stopped the rain, a great outcry that the missionaries must leave the country was raised. . . . The missionaries showed the native princes their own garden, that their cultivation was being ruined for want of water, and asked if it was probable that they would spoil their own crops. The natives remained unconvinced, and only when the rains at length fell plentifully did the excite- ment subside.” Pechuël-Loesche relates an exactly similar case, and adduces enough analogous facts to allow of our generalizing on the matter. ‘‘ After the Catholic missionaries had landed, there was a scarcity of rain, and the plantations were suffering from drought. The people at once took it into their heads that this was the fault of these clerics, and especi- ally due to the long robes they wore, for such had never been seen before. There was besides a white horse which had recently been landed, and it had prevented trading, and occa- sioned many troublesome discussions. A contractor had a great deal of trouble because he had replaced the pole of native wood which was badly warped, upon which his flag was erected, by an upright mast which had just been imported. A shiny mackintosh coat, an unusual hat, a rocking-chair, any instru- ment whatever, can give rise to disquieting suspicions. The entire coast population may be disturbed at the sight of a sailing-ship with unfamiliar rigging, or a steamer which has t Philips, ‘‘ The Lower Congo,” J.A.J., xvii. p. 220. 72 HOW NATIVES THINK one more funnel than the others. If anything vexatious should occur, it is at once attributed to something unusual that has taken place.” ! In New Guinea, “‘ at the time my wife and I took up our residence in Motumotu,” writes Edelfelt, “a kind of pleurisy epidemic prevailed along the coast. . . . Of course my wife and I were accused of bringing the messenger of destruction, and for this loud rumours were about that we—including the Polynesian teachers—should suffer death. . . . Somegne or something was still the cause of it, and they blamed a poor unfortunate sheep I had, that was killed to please them; but the epidemic raged as violently as ever. Now our two goats were blamed; these animals, however, lived it all out. Finally, they levelled their abuse and accusations against a large picture of Queen Victoria, which hung in our dining- room. Previous to the epidemic appearing people came in from long distances to see this picture, and . . . would . look at it several hours at a time. . . . The harmless image © of our gracious Queen became eventually . . . an imaginary destroyer of health and life, and they requested me to take the picture down ; to this I did not concede.”’ 2 At Tanna, in the New Hebrides, “‘it can hardly be said that there is any sequence of ideas, properly so called. Ifa person were passing along a path, and some creature (say a snake) fell on him out of a tree, and next day, or “the week after, he heard of the death of a son in Queensland, he would connect the two. A turtle came ashore one night and laid a nestful of eggs. It was captured in the act. * Such a thing had never taken place in the memory of the people. The conclusion was that Christianity was the cause of the turtle coming ashore to lay its eggs, and the right thing to do was to offer the turtle ss the missionary who had brought the worship of Jehovah.” Ideas are linked Fe in the same way in North America. “One evening, when talking about the animals of the country, I wanted to let them know that we had rabbits and leverets 1 Die Loango-Expedition, iii. 2, p. 83. a ‘ Customs and Superstitions of New Guinea Natives,’ Proceedings of Royal Geographical Society of Australasia, Queensland Branch, vii. p. 1 (1891-2). 3 Gray, “ Notes on the Natives of Tanna,”’ J/.AJ., xxviii. p. 131. THE LAW OF PARTICIPATION 73 in France, and I showed them to them by making the shadow on the wall in the firelight. It happened quite by chance that they caught more fish than usual the next morning ; they believed that the shadow pictures I had made for them were the cause of this, and they begged me to do this every evening and to teach them how, a thing which I refused to do, as I would not minister to this foolish superstition of theirs.’’ 2 . Lastly, in New Guinea. ‘‘ A man returning from hunting or fishing is disappointed at his empty game-bag, or canoe, and turns over in his mind how to discover who would be likely to have bewitched his nets. He perhaps raises his eyes and sees a member of a neighbouring friendly village on his way to pay a visit. It at once occurs to him that this man is the sorcerer, and watching his opportunity, he suddenly attacks him and kills him.” 2 | The current explanation of such phenomena consists of saying that primitives apply the principle of causality undis- cerningly, and that they confound cause and antecedent. | This would be the very common logical fallacy known as the} fallacy of post hoc, ergo propter hoc. The primitives, we are told, have not the least idea that this can be wrong. That the representations are consecutive in their minds is a sufficient guarantee that they are connected: or rather, they do not consider the connection needs any guarantee. The observers themselves frequently suggest this explanation. “To natives,’ says Pechuél-Loesche, “there is no such thing as chance. Oécurrences which are close together in point of time, even if widely removed in space, readily appear to them to be linked by a causal relation.” 3 It is true, and later on we shall discover the reason for. it, that there is no such thing as chance to the primitive. As for the rest, the explanation suggested, if not wholly incorrect, is assuredly incomplete. Primitives are undoubtedly prone, as much as, and possibly more than, civilized beings, to the fallacy of post hoc, ergo propter hoc. But in the facts I have quoted, some simple cases from a very large number 1 Father Sagard, Le Grand Voyage au Pays des Hurons, pp. 256-7 (1632). a Guise, ‘‘ Wangela River, New Guinea,” J.A.J., xxviii. p. 212. 3 Die Loango-Expedition, iii. 2, Pp. 333. 74 HOW NATIVES THINK of instances there is something more and something different ; it is not merely an artless and erroneous application of the principle of causality. It is not only direct anteriority in time which makes the connection between one fact and another. The sequence perceived or remarked may suggest the connection, but the connection itself is not in any way confused with this sequence. It consists in a mystic relation which the primitive represents to himself—and of which he is convinced as soon as he represents it to himself—between the antecedent and the consequent: the first having the power to produce the second and make it apparent. This is a result of the very facts related by Pechuél-Loesche, if we regard them in the light of what has already been established respecting the mystic properties in the forms taken by persons and things.t What effects may not the mystic virtues of a cassock, a steamer with three funnels, a mackintosh, a tent- pole, in short, any unusual object produce ? Who knows the consequences which may arise from their very presence ? _ ‘‘ Anything strange is uncanny to the native,” say Spencer and Gillen.2 In the case of Queen Victoria’s portrait, the fallacy post hoc, ergo propter hoc is obviously an inadequate explanation. The portrait was well known to the natives long before the epidemic broke out. They did not attach any blame to it except in the fourth place, after having suc- cessively imputed the outbreak to the missionary, his sheep, and his two goats. If, after that, they laid the trouble at its door, it was undoubtedly because of the magic power they believed to be attaching to this unusual object. The case of the Hurons, related by Sagard, can be similarly explained. To comprehend facts such as these, and refer them to one common principle, we must go back to the mystic nature of the collective representations in the mentality of undeveloped peoples and look for the same characteristic in the connec- tions between them. The sequence in time is one element of the connection, but it is not always a necessary element, and it is never all-sufficing. If it were otherwise, how could we explain the fact that the most unvarying and most evident 1 Vide Chap. I. pp. 41- 2 The Northern Tribes of otis Australia, pp. 31-2. THE LAW OF PARTICIPATION 75 sequences in phenomena so often escape the notice of primi- tives? For instance, “‘ the Ja-Luo believe that the sun goes back to the east overhead during the night, hidden behind the heavens. They do not associate the daylight sky with the light of the sun, but look upon it as something quite distinct, and asked what became of it at night.’’! Dobrizhoffer points out that the Abipones are often unable to comprehend direct and self-evident relations between cause and effect. ‘A wound inflicted with a spear often gapes so wide that it affords ample room for life to go out and death to come in; yet if the man dies of the wound, they really believe him killed, not by a weapon, but by the deadly arts of the jugglers. . . . They are persuaded that the juggler will be banished from amongst the living, and made to atone for their relation’s death, if the heart and tongue be pulled out of the dead man’s body immediately after his decease, roasted at the fire, and given to dogs to devour. Though so many hearts and tongues are devoured, and they never observe any of the jugglers die, yet they religiously adhere to the custom of their ancestors by cutting out the hearts and tongues of infants and adults of both sexes, as soon as they have expired.”’ 2 Thus not only does the time sequence in phenomena of the most impressive kind often remain unperceived by the primitive’s mind, but he frequently believes firmly in an order of succession which is not borne out by experience, for experi- ence can no more undeceive primitives than it can teach them. In an infinite number of cases, as we have seen, their minds are impervious to experience. Therefore when they make the cassocks of the missionaries responsible for the drought, or attribute an epidemic to a portrait, it is not merely the sequence in point of time which impresses their minds and becomes a causal relation to, them. The mental process is a different and rather more complex one. That which we call experience and the natural order of phenomena does not find in primitives, minds prepared to receive and be impressed by it. On the contrary, their minds are already preoccupied t Hobley, “‘ British East Africa, Kavirondo and Nandi,” /.A.J., XXXiiil. P. 358. ; 4 2 An Account of the Abtpones, il. p. 223. 76 HOW NATIVES THINK with a large number of collective representations, by virtue of which objects, whatever they may be, living beings, inani- mate objects, or articles manufactured by man, always present themselves charged with mystic properties. Consequently while these minds are very often unheedful of the objective relations, they pay great attention to the mystic connections, whether virtual or actual. These preformed connections are not derived from the experience of the present, and experience is powerless against them. IT Let us then no longer endeavour to account for these con- nections either by the mental weakness of primitives, or by the association of ideas, or by a naïve application of the principle of causality, or yet by the fallacy post hoc, ergo propter hoc ; in short, let us abandon the attempt to refer their mental activity to an inferior variety of our own. Rather let us consider these connections in themselves, and see whether they do not depend upon a general law, a common founda- tion for those mystic relations which primitive mentality so frequently senses in beings and objects. Now there is one element which is never lacking in : such relations. In varying forms and degrees they all involve a ‘ ‘participation ’’ between persons or objects which form part of a collective representation. For this reason I shall, in default of a better term, call the principle which is peculiar to “ primitive” mentality, which governs the connections and the preconnections of such representations, the law of participation. At the moment it would be difficult to formulate this law in abstract terms. It will be sufficiently defined in the course of this chapter, although that which we desire to define scarcely enters the ordinary framework of our thought. However, in default of a wholly satisfactory formula, we can make an attempt to approximate it. I should be inclined to say that in the collective representations of primitive mentality, objects, beings, phenomena_can be, though. ina way incomprehensible to_us, both themselves and. something other than themselves. In a fashion which is no less incom- prehensible, they give forth and they receive mystic powers, QC } à THE LAW OF PARTICIPATION 77 virtues, qualities, influences, which make themselvés. felt outside, without ceasing to remain where they are. | In other words, the opposition between the one and the à many, the same and another, and so forth, does not impose upon this mentality the necessity of affirming one of the terms if the other be denied, or vice versa. This opposition is of but secondary interest. Sometimes it is perceived, and . frequently, too, it is not. It often disappears entirely before the mystic community of substance in entities which, in our thought, could not be confused without absurdity. For instance, “the Trumai (a tribe of Northern Brazil) say that they are aquatic animals.—The Bororo (a neighbouring tribe) boast that they are red araras (parakeets). This does not merely signify that after their death they become araras, nor that araras are metamorphosed Bororos, and must be treated as such. It is something entirely different. “The Bororos,’”’ says Von den Steinen, who would not believe it, but finally had to give in to their explicit affirmations, ‘give one rigidly to understand that they are araras at the present time, just as if a caterpillar declared itself to be a butterfly.”’! It is not a name they give themselves, nor a relationship that they claim. What they desire to express by it is actual identity. That they can be both the human beings they are and the birds of scarlet plumage at the same time, Von den Steinen regards as inconceivable, but to the mentality that is governed by the law of participation there is no diffi- culty in the matter. All communities which are totemic in form admit of collective representations of this kind, implying similar identity of the individual members of a totemic group and their totem. From the dynamic standpoint also, the creation of entities and phenomena, the manifestation of such and such an occur- rence, are the result of a mystic influence which is communi- cated, under conditions themselves of mystic nature, from one being or object to another. They depend upon a participation which is represented in very varied forms: contact, trans- ference, sympathy, telekinesis, etc. In many aggregates of an undeveloped type the abundance of game, fish, or fruit, the regularity of the seasons, and the rainfall, are connected : K. von den Steinen, Unter den Naturvilkern Zentralbräsiliens, pp. 305-6. 78 HOW NATIVES THINK with the performance of certain ceremonies by individuals destined thereto, or to the presence or to the well-being of a sacred personality who possesses a special mystic power. Qr yet again, the newborn child feels the effects of everything its father does, what he eats, etc. The Indian, out hunting or engaged in warfare, is fortunate or unfortunate according to whether his wife, left behind in the camp, eats, or abstains from eating, certain foods, or is doing or not doing certain things. The collective representations abound in relations of this nature. What we call the natural relation of cause and effest passes unnoticed, or is of but slight importance. It is the mystic participations which are in the front rank, and frequently occupy the whole field. On this account the mentality of primitives may be called prelogical with as good reason as it may be termed mystic. These are two aspects of the same fundamental quality, rather than two distinct characteristics. If we take the content of the representations more particularly into account, we shall call it mystic—and, if the connections are the chief considera- _ tion, we pronounce it prelogical. By prelogical we do not mean to assert that such a mentality constitutes a kind of antecedent stage, in point of time, to the birth of logical thought. Have there ever existed groups of human or pre-human beings whose collective representations have not yet been subject to the laws of logic? We do not know, and in any case, it seems to be very improbable. At any rate, the mentality of these undeveloped peoples which, for want of a better term, I call prelogical, does not partake of that nature. It is not antilogical; it is not alogical either. By designating it “ prelogical’’ I merely wish to state that it does not bind itself down, as our thought does, to avoiding contradiction. It obeys the law of participation first and foremost. Thus oriented, it does not expressly delight in what is contradictory (which would make it merely absurd in our eyes), but neither does it take pains to avoid it. It is often wholly indifferent to it, and that makes it so hard to follow. As has been said, these characteristics apply only to the collective representations and their connections. Considered as an individual, the primitive, in so far as he thinks and acts independently of these collective representations where THE LAW OF PARTICIPATION 79 possible, will usually feel, argue and act as we should expect him to do. The inferences he draws will be just those which would seem reasonable to us in like circumstances. If he has brought down two birds, for instance, and only picks up one, he will ask himself what has become of the other, and will look for it. If rain overtakes and inconveniences . him, he will seek shelter. If he encounters a wild beast, he will strive his utmost to escape, and so forth. But though on occasions of this sort primitives may reason as we do, though they follow a course similar to the one we should take (which in the more simple cases, the most intelligent among the animals would also do), it does not follow that their mental activity is always subject to the same laws as ours. In fact, as far as it is collective, it has laws which are peculiar to itself, and the first and most universal of these is the law of participation. The very material upon which this mental activity is exercised has already undergone the influence of the law of participation, for we must realize that their collective represen- tations are very different from our concepts. The latter, the material with which our logical thought works, are themselves the result, as we know, of previous operations of a similar kind. The mere expression of a general abstract term, such as man, animal, organism, virtually connotes a number of separate judgments which involve definite relations between many concepts. But the collective representations of primi- tives are not, like our concepts, the result of intellectual pro- cesses properly so called. They contain, as integral parts, affective and motor elements, and above all they imply, in the place of our conceptual inclusions or exclusions, parti- cipations which are more or less clearly defined, but, as a general rule, very vividly sensed. Why, for example, should a picture or portrait be to the primitive mind something quite different from what it is to ours? Whence comes that attributing of mystic properties to it, of which we have just had an instance? Evidently from the fact that every picture, every reproduction “ parti- cipates’’ in the nature, properties, life of that of which it is the image. This participation is not to be understood as a share—as if the portrait, for example, involved a fraction of the whole of the properties or the life which the model 80 HOW NATIVES THINK possesses. Primitive mentality sees no difficulty in the belief that such life and properties exist in the original and in its reproduction af one and the same time. By virtue of the mystic bond between them, a bond represented by the law of participation, the reproduction is the original, as the Bororo are the araras. Therefore one may obtain from the one what one gets from the other; one may influence the second by influencing the first. For this reason, if the Mandan chiefs let Catlin take their portrait, they will not sleep quiet in their graves. Why is this? Because, by virtue of an inevit- able participation, anything that happens to their pictures, delivered over to strange hands, will be felt by them after their death. And why is the tribe so uneasy at the idea that the repose of their chiefs should be thus disturbed ? Evidently —though Catlin does not tell us so—it is because the welfare of the tribe, its prosperity, its very existence depend, by virtue of this same participation, upon the condition of its chiefs, whether living or dead. The same argument applies to the other collective repre- sentations, the mystic character of which we have already demonstrated: those concerning a man’s name and his shadow, for instance. There is one which we should dwell on particularly, however, because it serves as a rallying- point for the entire theory of primitive mentality. It is the representation of the “soul,” the starting-point of the doctrine known as animism, the principles of which Tylor formulates thus: ‘‘It seems as though thinking men, as yet at a low level of culture, were deeply impressed by two groups of biological problems. In the first place, what is it that makes the difference between the living body and the dead one? What causes waking, sleep, trance, disease, death ? In the second place, what are those human shapes which appear in dreams and visions ? Looking at these two groups of phenomena, the ancient savage philosophers probably made their first step to the obvious inference that every man has two things belonging to him, namely, his life and his phantom. These two are evidently in close connection with the body, the life as enabling him to feel and think and act, the phantom as being his image or second self ; both, also, are perceived to be things separable from the body, the life as able to go THE LAW OF PARTICIPATION 81 away and leave it insensible or dead, the phantom as appear- ing to people at a distance from it, The second step... merely the combining the life and the phantom. As both belong to the body, why should they not also belong to one another, and be manifestations of one and the same soul ? . . . This, at any rate, corresponds with: the actual conception of the personal soul or spirit among the lower races, which may be defined as follows: a thin unsubstantial human image, in its nature a sort of vapour, film or shadow; the cause of life and thought in the individual it animates, in- dependently possessing the personal conscience and volition of its corporal owner, past or present; capable of leaving the body far behind, to flash swiftly from place to place; mostly impalpable and invisible, yet also manifesting physical power, and especially the appearing to men, waking or asleep, as a phantasm separate from the body to which it bears a likeness ; continuing to exist and appear to men after the death of that body ; able to enter into, possess, and act in the bodies of other men, animals, and even of things. ... These are doctrines answering in the most forcible way to the plain evidence of men’s senses, as interpreted by a fairly consistent and rational primitive philosophy.” 1 It is indeed a favourite idea of Tylor’s that animism is a doctrine which is all the more consistent and satisfactory from the logical standpoint, the nearer to its source that we can view it, that is, in its most primitive form. Later, when complicated by new elements and endeavouring to resolve more subtle problems and to generalize, it becomes obscure and entangled. At its source, its clarity is absolute because it imposes itself, so to speak, on the naive reflection of the ‘‘savage philosopher’? when faced by the facts. And the satisfaction which this philosopher would find in his hypo- thesis is savoured by the savant of to-day in his turn, when he affirms that this hypothesis is the spontaneous and universal product of an intellectual activity which is always identical at bottom, and which, like his own, is impelled by the rational need for supplying an answer to the problems which the facts propound to the intelligence. It is a very seductive theory, and it seems as if, were we 1 Primitive Culture, i. ee 428-9 (4th edit., 1903). | | \ } 82 HOW NATIVES THINK | in the place of the ‘‘ savage philosopher,” we should reason as he does—that is, as we make him reason. But have such “savage philosophers” ever existed? Do the collective representations of the soul in the lower races constitute a sort of doctrine born of the necessity for resolving biological problems ? Nothingseems more unlikely. Nothing seems more improbable even, if it be certain that the mentality of such races is oriented differently from our own, and that their ~ collective representations are, above all, mystical by nature, primitives being much more concerned about the mystic virtues inherent in things than about the logical coherence of their own thought. That is why, the more rational and consecutive this primitive “‘ philosophy ’’ of the soul becomes, the more reason will there be to fear that, in spite of the amount of evidence collected and the skill of those who inter- pret it, it will still be very far removed from that which it pretends to explain. As a matter of fact, in almost every case where there has been an adequately prolonged and careful investigation, we must abandon the happy simplicity of ‘one and the same soul ’’ manifesting itself as ‘‘life on the one side and phantom on the other.” The collective representations offer something much more complex and far less easy to ‘‘ explain.” To take a few examples from the West Coast of Africa, Ellis has found a number of collective representations which do not tally in any way (as he himself expressly states) with the idea of a soul such as Tylor has defined, quoted above, The natives, says Ellis, distinguish the Ara from the srahman. “The kra existed before the birth of the man, probably as the kra of a long series of men, and after his death it will equally continue its independent career, either by entering a newborn human body, or that of an animal, or by wandering about the world as a sisa or kra without a tenement. The general idea is that the sisa always seeks to return to a human body and again become a kra, even taking advantage of the temporary absence of a kra from its tenement to take its place. ... The kra can quit the body it inhabits at will, and return to it again. Usually it only quits it during sleep, and the occurrences dreamed of are believed to be the adventures of the kra during its absence. The srahman or ghost-man only THE LAW OF PARTICIPATION 83 commences its career when the corporeal man dies, and it simply continues in the Dead-land, the existence the cor- poral man formerly led in the world. Thus we must consider “separately (1) the man; (2) the indwelling spirit or kra; (3) the ghost or srahman, though in another form the last is only the continuation of the first in shadowy form.” This differentiation is applied in every case. ‘‘ When the bush is torn up and withers, the kra (so to speak) of the bush enters a seedling bush, or a seed, and the ghost-bush goes to Dead-land. Similarly, the kra (so to speak) of the sheep, when that sheep is killed, enters a newborn lamb, and the ghost-sheep goes to Dead-land for the use of ghost-men. .. . Dead-land itself, its mountains, forests, and rivers are, the Tshi-speaking negro holds, the ghosts of similar natural features which formerly existed in the world.” | The kra, therefore, is not the soul. ‘‘ The soul, in the accepted sense of the word, is the ‘animating, separable, surviving entity,’ the ‘ vehicle of individual personal exist- ence’ (Tylor), whereas every kra has been the indwelling spirit of many men, and probably will be of many more. The kra in some respects resembles a guardian spirit, but it is more than that. Its close connection with the man is indi- cated by the fact of its nocturnal adventures during its absence from the body being remembered by that man when he awakes. The latter even feels physically the effect of his kva’s actions, and when a negro awakes feeling stiff and unrefreshed, or with limbs aching from muscular rheumatism, he invariably attributes these symptoms to the fact of his kra having been engaged in some struggle with another, or in some severe toil. . . . It has, though doubtless in a shadowy form, the very shape and appearance of the man, and both the mind and the body of the latter are affected by, and register the results of, the kva’s actions.” “When the kra leaves the body of the man it inhabits, that man suffers no physical inconvenience; it goes out, when he is asleep, without his knowledge, and if it should leave him when he is awake, he is only made aware of its departure by a sneeze or a yawn. When, however, the soul, ‘ the vehicle of personal individual existence,’ leaves the body, that body falls into a condition of suspended. animation ; 84 HOW NATIVES THINK it is cold, pulseless, and apparently lifeless. Sometimes, though rarely, the soul returns after such an absence, and then the man has been in a swoon or trance; more generally it does not return ; then the man is dead.”’ ! How therefore are we to regard the relations of a man with his kra which certainly, as Ellis tells us, is not his soul? It is equally incorrect to say that his kra is himself and that he is not himself. It is not the individual, for it existed before him and survives him; it is nevertheless not himself, for upon awaking the individual remembers what the kra has done, endured, and suffered during the night. If we persist in submitting these representations to the claims of logical. thought, we shall not only not discover in them the “ consistent and rational philosophy ” which Tylor and his school expect, but they will remain unintelligible. On the other hand, we shall understand them to the extent to which they may be ‘ wnderstood,” if we refer them to the general law of partici- pation. The individual, while living, participates in the kra that inhabits him, that is, he to a certain extent is this ka, and at the same time he is not: for the actual contradiction does not disturb his prelogical mentality. At the moment of death, this participation comes to an end.? Representations which are just as involved are to be found in most undeveloped aggregates. Since they did not suggest any meaning which investigators could accept (judging them as they did by the rules of logical thought) these writers frequently tried to guard against absurdity by supposing that primitives: admitted of several souls. It thus became possible to distribute among these that which would have been irreconcilable and incompatible in a single one. Spencer and Gillen, for instance, refer frequently to multiple souls in dealing with the tribes of Central Australians.3 Among the natives of the Torres Straits, Haddon speaks of parts of 5 A. B. Ellis, The Ewe-speaking Peoples, pp. 15-21, 106. Cf. The Tshi- speaking Peoples, p. 149. 2 Traces of representations of this kind were found among the Greeks who, according to Rohde (Psyche, i. pp. 4, 6, 257; ii. pp. 141, 157, 183-4, 304-5), distinguished between a vital principle, a soul or shade in Hades after death, and another principle, which inhabited the body during life, but which neither disease nor death could affect. 3 The Native Tribes of Central Australia, p. 515; The Northern Tribes of Central Australia, p. 450. THE LAW OF PARTICIPATION 85 the soul. “There was a belief that part of the mar: left at death, and part remained until frightened away.” In North America, a plurality of souls seems to be the rule. ‘They distinguish four souls in the same body. An old man told us sometime ago that some natives had two or three souls, and that his own had left him more than two years before, to go away with his dead parents, and that he now only possessed the soul of his body, which would go down to the grave with him. By this we know that they imagine that the body has a soul of its own, which some call the soul of their nation (?) and that besides there are others which leave it sooner or later according to their fancy.” 2 ‘It is believed by some of the Hidatsa that every human being has four souls in one. They account for the phenomenon of gradual death where the extremities are apparently dead while consciousness remains, by supposing the four souls to depart, one after another, at different times. When dissolution is complete, they say that all the souls are gone, and have joined together again outside of the body.” 3—‘‘ The Mandan believe that each person has several spirits dwelling within him; one of which is black, another brown, and a third light-coloured, the last alone returning to the Lord of Life.”’s The Dacotans recognize four: ‘ (1) a spirit of the body, which dies with the body ; (2) a spirit which always remains with or near the body ; (3) the soul which accounts for the deeds of the body and is supposed by some to go south or by others west. The fourth always lingers with the small bundle of the hair of the deceased, kept by the relatives until they have a chance to throw it into the enemy’s country, when it becomes a roving spirit, bringing disease and death.” 5 Some Sioux even admit of five souls. In British Columbia, ‘‘ man is believed to have four souls. The main soul is said to have the shape of a mannikin, the others are the shadows of the first. In disease either the lesser souls or the main one leave the body. Shamans can easily return the shadows, but not the main soul. If the latter leaves the body the sick one must die. After death the main soul goes to the sunset, and here it remains. The 1 ** The Western Tribes of Torres Straits,” J.AJ., xix. p. 317. > Relations du Père Lejeune, p. 146 (1630). 3 Dorsey, ‘‘ Siouan Cults,” E. B. Rept., xi. p. 517. , 4 Ibid., p. 512. s Lynd, cited by Dorsey, ibid., p. 484. 86 HOW NATIVES THINK shadows become ghosts. They revisit the places which the deceased frequented during lifetime, and continue to do the same actions which he did when alive.” 1 These reports, to which we might add many others, are far from being unanimous regarding the functions of the various souls. Nevertheless they all attest a multiplicity of souls in the same individual, and diversity in the functions of these souls. They indicate, too, though less regularly and clearly, that the fate of these souls is not the same after death. Is it not permissible to imagine that this multiplicity is pre-emi- nently expressive of the impossibility which the investigators found in reconciling what the ‘‘ savages ”’ had told them, with their own preconceived ideas upon the soul ? Gross misunder- standing and misconstruction were inevitable. The mission- aries and explorers were making use of terms (soul, spirit, ghost, etc.) defined for them by prolonged evolution in religion, philosophy, and literature, and they found themselves dealing with collective representations which were essentially mystic and prelogical, not yet reduced to a conceptual form, and paying very little regard to logical laws. Consequently nearly all they report needs to be revised and corrected, As a general rule, an observer’s report is the more suspect, the more readily it agrees with the current conception of the soul. On the other hand, observations frequently manifest a charac- teristic obscurity, or even present an inextricable confusion, which is a faithful reflection of the perplexity felt by their authors. | I shall quote but two instances. “It is difficult to say precisely what the Fijians believe to be the essence of the immortal part of a man. The word ‘ yalo’ has the following meanings. Yalo (with pronoun-suffixed) means Mind, as Yalo-ngu. Yalo (with possessive pronoun separated) means Shade or Spirit. Yaloyalo means Shadow. From the posses- sive pronoun being suffixed we may gather that the mind was regarded as being as intimately connected with a man’s body as his arm, but that the spirit could be detached from it.” 2 Among the Yakuts, “the elementary soul of the object in t F. Boas, ‘‘ The North-western Tribes of Canada,” Reports of the British Association, p. 461 (1894). 2 B. A. Thompson, ‘ The Ancestor Gods of the Fijians,” /J.A.J., xxiv. p. 354 (note). THE LAW OF PARTICIPATION 87 general (ichchi) which, it appears, merely expresses the fact of existence, differs from the soul of living things (sur). Life begins when respiration begins (fy). Living objects have a double soul therefore—ichchi and sur ; animals when dead or even ill lose their sur and preserve their ichchi alone, and in the case of death this disappears also. Men, and among animals, horses alone, have a triple soul: ichchi, sur, and the ‘but. The human kut is small, no larger than a little piece of coal. Sometimes the shaman summons from beneath the earth, in the left or female part of the house, the kut of those who are sick. . . . The kut sometimes abandons man during his sleep, and wanders afar. If some misfortune befalls it during its travels, its owner falls ill. The kui is like a faint picture, like the shadow. As the shadow has three parts, one large and faint, one small and darker, and the centre quite black, so man has three souls. When he loses one, he suffers from a feeling of uneasiness; when he loses two, he becomes ill: and when he loses all three, he dies.”’! The confusion in this report and the evident impossibility of recon- ciling the different parts with the definitions given are very significant: they may however help us to understand what this pretended plurality of souls really amounts to. It is noteworthy that without thinking of a prelogical and mystic mentality, without even having made any research into the problem which occupies my mind, Pechuél-Loesche should have arrived at the same conclusion, touching the plurality of souls, as I do. ‘If we were generalizing hastily,” he says in conclusion, ‘‘ we might speak of a belief in two souls, or even in three or four. The first would be power (Potenz) the creative principle (the essence of the ancestors passing to their descendants), possibly also part of the uni- versal soul. Then there would be the personal or indi- vidual soul; finally the dream-soul, and the wandering or desert-soul (Wildnisseele). But such a conception would be incorrect.”2 In my opinion, these various souls express ‘ participations,” irreducible to logical intelligibility, but the most natural thing in the world to prelogical mentality. We 1 Sieroshewiski, Douze Ans chez les Yakoutes, cited -J.A.I., xxxi. p. 108 (note). 2 Die Loango-Expedition, iii. 2, pp. 296-300. 88 HOW NATIVES THINK can find sufficient evidence upon the subject of the desert-soul, which is what Miss Kingsley calls the ‘‘ bush-soul.” The negroes of Calabar, she says, suppose there to be four souls: ‘‘ (a) the soul that survives death, (6) the shadow on the path, (c) the dream-soul, and (d) the bush-soul. This bush-soul is always in the form of an animal in the forest— never of a plant. Sometimes when a man sickens it is because his bush-soul is angry at being neglected, and a witch-doctor is called in, who, having diagnosed this as being cause of the complaint, advises the administration of some kind of offering to the offended one. . . . The bush-souls of a family are usually the same for a man and his sons, for a mother and her daughters... . Sometimes all the children take the mother’s, sometimes all take the father’s... .” 1 “This bush-soul may be in only an earth pig, or it may be in a leopard, and ... no layman can see his own soul. It is not as if it were even connected with all earth pigs or all leopards, as the case may be, but it is in one particular earth pig or leopard or other animal. . . . If his bush-soul ~ dies, the man it is connected with dies. Therefore if the hunter who has killed it can be found out—a thing a witch- doctor cannot do unless he happens by chance to have had his professional eye on that bush-soul at the time of the catas- trophe; . . . that hunter has to pay compensation to the family of the deceased. On the other hand, if the man belong- ing to the bush-soul dies, the bush-soul animal has to die too.””2 Miss Kingsley has collected very detailed information about the diseases of this bush-soul, as well as those of the dream-soul, and of the treatment that such diseases require. There was an exactly similar idea existing in Central America. Of the Guatemaltecs Gage tells the following quaint story: ‘Many are deluded by the Devil to believe that their life dependeth upon the life of such and such a beast (which they take unto them as their familiar spirit) and think that when that beast dieth they must die; when he is chased, their hearts pant; when he is faint, they are faint ...’’3 This evidently refers to a forest-soul, 1 Travels in West Africa, pp. 459-60. 2 West African Studies, pp. 209-10. 3 Bancroft, The Native Races of the Pacific States of North America, iii. P. 120. THE LAW OF PARTICIPATION 89 We accordingly find nothing among primitives which exactly corresponds with the single soul which, according to Tylor, can manifest itself in the double form of vital principle and of phantom. Undoubtedly they everywhere believe in the objective reality of what they see in dreams, and everywhere, too, they are convinced that the ghosts of the dead return, at least for some time, to haunt the places in which they dwelt while living. But what we have just learnt about this proves indeed that their collective representa- tions on this matter do not proceed from any need to account for these apparitions by a uniform concept of the “soul.” On the contrary, I should say that originally (to the extent that such a term is permissible) the idea of a soul is not found among primitives. That which takes its place is the repre- sentation, usually a very emotional one, of one or more co- existent and intertwined participations, as yet not merged into the distinct consciousness of an individuality which is really one. The member of a tribe, totem, or clan, feels himself mystically united with the animal or vegetable species which is his totem, mystically united with his dream-soul, his forest- soul, and so on. These communions, the intensity of which is renewed and increased at certain times (the sacred ceremonies, rites of ini- tiation, and so forth), by no means exclude each other. They have no need to be expressed in definite terms to be profoundly felt, and felt by every member of the group. Later, when these rites and ceremonies have gradually ceased to be under- stood, and then to be practised, such participations retained in customs and myths will fall, so to speak, into the form of “‘ multiple souls,” as has happened in the case of the Calabar negroes, so carefully studied by Miss Kingsley. And later still, nearer our own times, as the example of the Greeks show, these multiple souls will in their turn be crystallized into a single soul, not without the distinction between the vital principle and the spiritual inhabitant of the body still remaining apparent. In short, the “ soul ” properly so called, which serves as the starting-point, of Tylor’s theory, and which is, according to him, the object of the savage’s primitive philosophy, does not appear, according to my view of the matter, except in peoples of a relatively advanced type. If — 90 HOW NATIVES THINK he has referred it so far back it is not because he did not know the facts, for Tylor himself quotes a number of cases in which the multiplicity of “souls’’ is expressly mentioned. But his interpretation of these facts was, as it were, imposed upon him by his postulate, according to which the mentality of lower races obeys the same logical laws as our own thought. Let us abandon this postulate, and at once the mystical and pre- logical nature of this mentality appears, and with it the law of participation which governs its collective representations. From this it follows too that the concept of the soul can no longer be considered save as the product of thought which is already somewhat advanced, and not yet known to primitive peoples. III It scarcely ever happens that the primitive mind acquires its collective representations in an isolated state, apart from the connections with which they are usually bound up. The mystic character which attaches to them necessarily involves relations, which are also mystic, between the various objects of their thought. We may then admit a priori, as it were, that the same law of participation which | governs the forma- existing between them. To establish this position, it will be necessary to study the method in which the main inter- relations between persons and things is actually occur in the prelogical mind. r In the first place, the very existence of the social groups, in its relations with that of the individuals which compose them, is most frequently represented (and felt at the same time) as a participation, a communion, or rather, a number of participations and communions. This characteristic is apparent in all the primitive societies about which we have fairly detailed and reliable information. It has been fully demonstrated in the two volumes on the tribes of Central Australia by Spencer and Gillen, Among the Aruntas ‘each individual is the direct incarnation of an Alcheringa ancestor or of the spirit part of some ae animal. The totem of any man is regarded . . . as the same thing : as him- THE LAW OF PARTICIPATION OI self. . . . Each totemic group is supposed to have a direct control over the numbers of the animal or plant the name of which he bears. . . .”’1 Lastly, each totem is mystically bound to a strictly defined locality or portion in space which is always occupied by the spirits of the totemic ancestors, and this is called ‘‘ local relationship.” 2 The collective representation in this caseis exactly like that which so astonished Von den Steinen when the Bororos “ rigidly ’”? maintained that they were araras, and the Trumai that they were aquatic animals. Every individual is both such and such a man or woman, alive at present, a certain ancestral individual, who may be human or semi-human, who lived in the fabulous age of the Alcheringa, and at the same time he is his totem, that is, he partakes in mystic fashion of the essence of the animal or vegetable species whose name he bears. The verb ‘‘ to be”’ (which moreover is non-existent in most of the languages of undeveloped peoples) has not here the ordinary copulative sense it bears in our languages. It signifies something different, and something more. It en- compasses both the collective representation and the collective consciousness in a participation that is actually lived, in a kind of symbiosis effected by identity of essence. This is the reason that the members of a definite totemic group alone have the power to carry out the infichiuma ceremonies, the aim of which is to secure a regular reproduction of a certain animal or plant species.3 This explains, too, the performances, ceremonies, dances (with or without masks, painted decora- tions, costumes, tattooings) which are to be met with in many primitive peoples, and which are designed to attain the same end—the bison dances of the North American Indians, stag dances of the Huichol Mexicans, the serpent dances of the Zuñis and in other pueblos, for instance. With the Australian tribes, we may say that Spencer and Gillen have put a finger on the very spot, not only revealing the mystic and at the same time utilitarian significance of the intichiuma ceremonies, but the intimate relation which exists between the individual, his totemic group and his totemic species, a relation which cannot be expressed by a concept, t The Native Tribes of Central Australia, pp. 202-4. + Ibid., pp. 303, 544. 3 Ibid., pp. 169-70 02 HOW NATIVES THINK just because the wholly mystic relation consists of a participa- tion which cannot be comprised within the limits of logical thought. ‘At the first glance it looks much as if all that they intended to represent was the behaviour of certain animals, but the ceremonies have a much deeper meaning, for each performer represents an ancestral individual who lived in the Alcheringa. . . . It is as a reincarnation of the never-dying spirit part of one of their semi-animal ancestors that every member of the tribe is born, and therefore, when born he, or she, bears of necessity the name of the animal or plant to which the Alcheringa ancestor was a transformation or a descendant.’’! The ceremonies and dances, therefore, are intended to revive and maintain, by means of the nervous exaltation and ecstasy of movement not wholly unlike that seen in more advanced societies, the community of essence in which the actual individual, the ancestral being living again in him, and the animal or plant species that forms his totem, are all mingled. To our minds, there are necessarily three distinct realities here, however close the relationship may be. To the primitive minds, the three make but one, yet at the same time are three. In this way the influence which the ceremonies exercise over the totemic species is more than direct : it is 2mmanent. How can the primitive doubt their effective power? The most invincible logical certitude gives way before the sym- biotic feeling which is an accompaniment of collective repre- sentations thus lived and translated into action. Another aspect of this participation or rather communion is revealed by the part played in the life of the individual and of the community among the Aruntas, by the sacred objects ts known as churinga. _ These things (pieces of wood or stone of an oblong shape, and generally decorated with mystic designs) are most carefully preserved and deposited in a sacred place which women and children dare not approach. Every totemic group has its own, and from the standpoint of logical thought it would be very difficult to define exactly what churinga are, or are not. The external souls of individuals ; the vehicles of ancestral spirits and possibly the bodies of these ancestors themselves; extracts of totemic essence; 1 The Native Tribes of Central Australia, p. 228. THE LAW OF PARTICIPATION 93 reservoirs of vitality—-they are all of these in turn and simul- taneously. The recognition of their mystic power attains its highest intensity at the moment of the initiation cere- monies, which we shall examine in detail later.t At this stage, however, I may note, on Spencer and Gillen’s authority, the deep religious respect which surrounds the churinga, the care taken to preserve them, and the veneration and precaution with which they are handled. ‘During the whole time the presence of the churinga’’ (which had been lent to a neigh- bouring tribe, and were being examined on their return) “seems to produce a reverent silence as if the natives really believed that the spirits of the dead men to whom they have belonged in times past were present, and no one, while they were being examined, ever spoke in tones louder than a whisper.’ 2? Everywhere, the very expressions used by investi- gators suggest the idea of participation. “ A man who possesses such a churinga as the churinga snake one will constantly rub it with his hand, singing as he does so about the Alcheringa history of the snake, and gradually comes to feel that there is some special association between him and the sacred object —that a virtue of some kind passes from it to him and also from him to it.’’3 Can we be surprised, therefore, that the churinga is represented, or rather, felt, to be a living being ? It is something very different from a piece of wood or stone. It is intimately connected with the ancestor ; it has emotions as we have, and these can be calmed when it is stroked with the hand, in the same way in which those of living people may be soothed. From a participation which is directly represented and actually felt, such as described by Von den Steinen and by Spencer and Gillen, it is an easy transition to the belief, which is so common in undeveloped races, in a relationship between men and animals, or rather, between certain groups of men and certain predetermined animals. Such beliefs are often expressed in myths. Among the Aruntas, Spencer and Gillen collected a number of legends relating to semi-human or semi- animal beings, which establish a living transition between t Vide Chap. VIII. pp. 354-8. 2 The Native Tribes of Central Australia, p. 303. 3 The Northern Tribes of Central Australia, pp. 277-8. 4 Ibid., p. 265 (note). 94 HOW NATIVES THINK the two classes. Very frequently, too, the expressions used by the observers are significant. Thus, in describing a totemic ceremony, it is said: “‘ The particular rat-man or man-rat— for, as already said, the identity of the human individual is sunk in that of the object with which he is associated, and from which he is supposed to have originated—to whom this ceremony referred, is supposed to have travelled to Walyirra, where he died, and where his spirit remained associated as usual with the churinga.’’* Spencer and Gillen find in these mythical ideas “a crude attempt to describe the origin of human beings out of non-human creatures who were of various forms ; some of them were representations of animals, others of plants, but in all cases they were to be regarded as inter- mediate stages in the transition of an animal or plant ancestor into the human individual who bears its name as that of his or her totem.” 2 Among the peoples who are more advanced in develop- ment, the idea of these mythical animals is slightly different. The ancestors of the totemic group are not exactly like animals that exist to-day, but they have a mystic share of both the animal and the human nature. The primitives project upon them, as it were, the participation which forms the union between the social group and its totemic animal. In British Columbia, for instance, “‘I sought to learn from him (i.e. the usual informant) whether his people were known as the otter-people, and whether they looked upon the otter as their relatives, and paid regard to these animals by not killing or hunting them. He smiled at the question and shook his head, and later explained that although they believed their remote ancestor to have been an otter, they did not think it was the same kind of otter as lived now. The otters from which they were descended were otter-people, not animals, who had the power to change from the forms of men and women to those of the otter. All the animals in the old-time were like that, they were not just common animals and nothing else ; they were people as well and could take the human or the animal form at will, by putting on or taking off the skin or other natural clothing of the animal. . .. Among the Thompsons they have a distinct term in their t The Native Tribes of Central Australia, p. 231. 2 Ibid., p. 392. THE LAW OF PARTICIPATION 95 language by which these mystic beings are distinguished from ordinary animals.” 1 Here again mystic participations account for relationships that the lower races consider as natural and evident, however absurd and unimaginable they appear to European investi- gators. Du Chaillu tells us that the African “‘ king Quengueza refused to eat the meat offered him. ‘It is roondah for me.’ He explained that the meat of the Bos brachicheros was for- bidden to his family, and was an abomination to them, for the reason that many generations ago one of their women gave birth to a calf instead of a child. I laughed, but the king replied very soberly that he could show me a woman of another family whose grandmother had given birth to a crocodile, for which reason the crocodile was roondah to that family. . . . They are religiously scrupulous on this matter . scarce a man is to be found to whom some article of food is not roondah,”’2 There is no need to lay stress on such beliefs, for they are excessively general, and experience, in the rare cases in which it can be appealed to, proves power- less against them Rajah Brooke tells of a man who had had his leg mutilated by an alligator, despite his mystic re- lationship with that animal. ‘I asked him if he had since retaliated on the alligator tribe. He replied, ‘No; I never wish to kill an alligator, as the dreams of my forefathers have always forbidden such acts; and I can’t tell why an alligator should have attacked me, unless he mistook me for a stranger, and that was the reason the spirits saved my life.’ ’’ 3 When a social group or an individual thus considers itself solidary with, or related to, a totemic animal, when he or it objectifies this participation in the actual relations with the animal, does it signify the animal species considered in its entirety and, so to speak, in an abstract fashion, or all the members of the species considered collectively, or is it a certain animal in particular ? This question presents logical thought with hypotheses which are distinct and mutually exclusive, and we have to choose between them. Prelogical mentality, however, hardly ever separates them (except in the case of the 3 Hill Tout, ‘‘ The Halkomelem of British Columbia,” J.A.I., xxiv. p. 325. 2 Equatorial Africa, pp. 308-9; J.A.J., xxiv. p. 325. 3 Ten Years in Sarawak, i. p. 235. 96 HOW NATIVES THINK “ forest-soul ”” given above) just because the law of participa- tion, which mainly governs them, allows of his thinking of the individual in the collective and the collective in the indi- vidual without any difficulty. Between the bear and bears, . and the bison and bisons, and one fish and fish in general, . such a mentality pictures a mystic participation, and neither the species as a whole nor the individual existence of its repre- sentatives means the same to him as it would do to us. Are the honours so frequently paid in ceremonious style to an animal killed in the chase, meant for that animal in particular, or for the spirit of the genus whose goodwill must be secured? There is no alternative in the case: such honours are paid to both, as one and indivisible. ‘ A French- man having one day thrown away a mouse which he had just caught, a little girl picked it up, intending to eat it. The child’s father snatched it from her and began to caress the dead animal affectionately. . . . ‘I am doing this,’ he said, ‘so that the spirit of the mouse may not trouble my child when she has eaten it.’’?! This guardian spirit, interpreted : as a concept for rational thought, is a relation which in reality can find no expression in its categories: it is the interpretation of a participating relation between the individual animal and the collective. And this participation does not find its reason and its proof, as it seems to us, in the identity of anatomical structure or physiological function, of external characteristics which are apparent and can be verified by experience: it is conceived and felt “in terms of spirit” like every reality which the prelogical mind perceives. To such a mind that which is of supreme interest in the animal (setting aside the need of obtaining food from it; and even eating the flesh of an animal is above all participating mys- tically in its nature), is not its visible form or qualities ; it is the spirit of which it is the manifestation, upon which the mystic relations of this animal (whether regarded individually or collectively, matters little), with the human groups in question depend. Invisible and intangible, this “ spirit ” is at once in each and all. Such “ multipresence ” proves no stumbling-block to prelogical mentality. t Charlevoix, Journal d'un Voyage dans l’Amérique Septentrionale, iii. Pp. 299-300. THE LAW OF PARTICIPATION 97 Bancroft relates a Californian belief, which seems incredible to him, and which throws strong light upon the mystic par- ticipation between the individual and the species. “They called this bird (the buzzard) the panes, and once every year they had a festival of the same name, in which the principal ceremony was the killing of a buzzard without losing a drop of its blood. It was next skinned, all possible care being taken to preserve the feathers entire. . . . Last of all the body was buried within the sacred enclosure amid great apparent grief from the old women, they mourning as over the loss of relative or friend. Tradition explained this: the panes had indeed been once a woman, whom, wandering in the mountain ways, the great god Chinigchinich had come suddenly upon and changed into a bird. How this was con- nected with the killing of her every year anew by the people and with certain extraordinary ideas held relative to that killing is, however, by no means clear; for it was believed that as often as the bird was killed it was made alive again, and more, and—faith to move mountains—that the birds killed in one same yearly feast in many separate villages were one and the same bird !’’! IV So far we have been considering especially, as far as the collective representations of primitives are concerned, what we might call the participating relations from the static point of view, that is, those which govern the existence of objects, natural phenomena, individuals or species. Let us now pass to the dynamic aspect, and consider the influence and sway which beings and objects mutually exercise. To tell the truth, it is one of the characteristic features of primitive mentality that in very many cases the difference between the two points of view tends to disappear. It is often impossible to decide whether an influence is transitive or immanent, and in spite of the difficulty which we feel in conceiving that which we consider contradictory, it is both at the same time. This is the case with the influence exercised by the totemic group upon the animal or plant which is its totem, by means of the t The Native Races of the Pacific States of North America, iii, p. 108. ff 98 HOW NATIVES THINK intichiuma ceremonies. In North America, again, the members of the wind totem are supposed to have a special influence on blizzards, and they are entreated to send a breeze when the mosquitoes become too troublesome.! In the Torres Straits, ‘an Umai man (having the dog as his totem) was credited with understanding the habits of dogs, and with ability to exercise special control over them.” 2 Among the tribes of Central Australia ‘an euro man gives to a plum-tree man a ‘sung-over ’ churinga for the purpose of assisting him to hunt the animal.’’3 In the Kaitish tribe, the headman of the water totem must religiously refrain from using a pointing stick or bone against an enemy, for if he did so, the water would become foul and infectious.4 These facts and many others which re be cited serve to show how insensibly prelogical mentality establishes a connection between the influence exercised_ upon. oneself and the influence exercised on something else, When a certain act of the water totem chief makes the water undrinkable, it is impossible to decide whether the influence exercised is imagined as transient or as immanent: prelogical mentality does not differentiate. But that which we perceive clearly in the relations of the totemic groups with the being or object or species which is their totem, fa profound and informed analysis of prelogical mentality would discover among an infinitude of other relations, which this same mentality would also picture as subject to the law of participation, Thus there exists a mystic participation between each totemic group and a certain spot which pertains to it, that is, between this totemic group and a certain point of the compass. The cardinal points, in their turn, are bound up (also by mystic participation) with colours, winds, mythical animals, and these with rivers, or sacred forests, and so on, almost to infinity. The natural environment of a certain group, tribe or family of tribes, for instance, thus appears in their collective representations, not as an object or a system of objects and phenomena governed by fixed laws, according to the laws of rational thought—but as an unstable ensemble of mystic 1 Dorsey, “ _Siouan Cults,’”’ E. B. Rept., xi. p. 410. 2 Haddon, ‘ The West Tribes of Torres Straits,’ J.4.1., xix. p. 325. 3 Spencer and Gillen, The Native Tribes of Central Australia, PP. 202-4. 4 Id., The Northern Tribes of Central Australia, p. 463. THE LAW OF PARTICIPATION 99 actions and reactions, of which persons, things, phenomena, are but the vehicles and manifestations, an ensemble which depends upon the group, as the group depends upon it. Oriented differently from our own, and pre-eminently engaged with mystic qualities and connections, with the law of participation as its supreme guide and control, primitive mentality perforce interprets in a fashion other than ours what we call nature and experience. Everywhere it perceives the communication of qualities (through transference, contact, projection, contamination, defilement, possession, in short, through a number of varied operations) which, either instan- taneously or in the course of time, bring a person or a thing into participation with a given faculty; and these qualities have the power of consecrating him or it, or the contrary, at the beginning or end of aceremony.t Later on, from a formal point of view, and to demonstrate the actual functioning of prelogical mentality, I shall pass in review a certain number of magical or religious practices derived from these ideas, and they will appear to be inspired and maintained by this belief in a participation. The beliefs relating to the different kinds of taboo fall under this head. When an Australian aboriginal or a Maori, terrified at the idea that he has uncon- sciously partaken of a food which is forbidden, dies from having infringed his taboo, it is because he feels himself hopelessly impregnated by a death-dealing influence which has entered his body with the food in question. This same influence was imparted to the food through participation : for example, it may have been the remains of a chief’s repast which an unfortunate man of the common people has eaten inadvertently. | The same ideas are at the bottom of the universal belief which affirms that certain men become animals—tigers, wolves, bears, etc.—whenever they put on the skins of such. To the primitives, such an idea is wholly mystic. They are not concerned with knowing whether the man, in becoming a tiger, ceases to be a man, and later, when he becomes a man again, isnolongeratiger. That whichis of paramount import- ance to them is the mystic virtue which makes these individuals ‘“ participable,” to use Malebranche’s term, of both tiger t Cf. Hubert and Mauss, Mélanges ad’ Histoire des Religions, pp. 22-32, 66-7 100 HOW NATIVES THINK and man in certain conditions, and consequently more for- midable than men who are never anything but men, and tigers which are always tigers only. “ How is this?” said the worthy Dobrizhoffer to the Abi- pones, ‘you daily kill tigers in the plain without dread, why then should you weakly fear a false imaginary tiger in the town?” ‘ You Fathers don’t understand these matters,” they reply, with a smile. ‘‘ We never fear, but kill, tigers in the plain, because we can see them. Artificial tigers we do fear, because they can neither be seen or killed by us.” : So, too, the Huichol who adorns himself with the éagle's feathers does not do so solely, or even mainly, for decoration. He believes that he can, by means of these feathers, transfer to himself some of the acute visual power, strength, and shrewdness of the bird. Again it is participation which is the basis of the collective representation that dictates this action. As a general rule, the processes employed by primitives to obtain the results they desire afford enlightenment as to their ideas of natural forces, and the manifestation of the living beings and phenomena which surround them; for we may say that they either imitate the manifestation, such as they suppose it to be, or else they imagine it as a symbol of what they do themselves. Now these processes, as we shall see in detail later on, are essentially mystic, and almost always involve the elements of participation. Their idea of the forces of ambient nature, then, presents the same character- istics. This is a fresh reason for abandoning the plausible and seductive, yet incorrect theory according to which, by spontaneous and inevitable application of anthropomorphic analogy, primitives see everywhere in nature wills, spirits, souls like their own. Far from sanctioning our attributing to them previous reflection upon their own activity, or a generali- zation which is founded on the result of such reflection, the facts forbid of our endowing them with that logical and con- sistent natural ‘‘ philosophy ”’ which, at any rate in its origin, animism would be. We must undoubtedly take into account the stupendous t Dobrizhoffer, An Account of the Abipones, ii. pp. 77-8. THE LAW OF PARTICIPATION IOI number of facts collected and classified by Tylor, Frazer, and their collaborators and disciples, and grant that, according to these facts, we never find in the collective representations of primitives anything that is dead, motionless, without life. There is abundance of evidence that all entities and all objects even those which are inanimate and inorganic, even things manufactured by men’s hands, are imagined capable of exer- cising and undergoing the most diverse influences. The Malay) miner believes that tin ore may have special affinities for certain people and cannot be discovered by others ; and we have noted how very important, according to Cushing’s account, is the exact shape to be given to familiar objects of manufacture. But it does not therefore follow that the tin ore or the domestic utensils have a soul conceived by analogy with the human soul. It is only permissible to conclude that the primi- tive’s ideas of entities and objects and all that relates to them, are mystic, and governed by the law of participation. | (It may be that at a certain stage in the development of the mentality, | the members of a given social group tend both to acquire a | more distinct consciousness of their own personality, and to. imagine similar personalities outside themselves in animals, | trees, rocks, etc., or in gods and spirits. But neither this, representation, nor this generalized analogy, is a natural and | original product of this mentality. Pechuél-Loesche has made a detailed study of this matter, as far as the Bafioti of the east coast of Africa are concerned.? We cannot here reproduce or even summarize his arguments, which are based upon a very meticulous observance of beliefs and customs. His conclusion is that such words as “ will ” or “soul” or “spirit”? should be altogether eliminated. There is indeed something in the beings and the phenomena, but it is neither soul nor spirit nor will. If it be absolutely necessary to give it expression,.it would be best to use “ dy- nanism”’ instead of ‘‘animism.’”’ Pechuél-Loesche quotes a traveller of the seventeenth century, according to whom “ these people know neither God nor devil, for they cannot give either any proper name; but they confine themselves to applying the: term mokisie ’ to everything in which they perceive t Skeat, Malay Magic, p. 259. 2 Die Loango-Expedition, iil, 2, pp. 356-7. - 102 HOW NATIVES THINK hidden force.” He notes too that the wonder-workers have just as much fear of the souls of the dead as the laymen. “If one should ask a celebrated nganga by the help of whose soul or spirit he is performing, he will look at you without answer- ing, with a glance full of fear. He has never had any idea of that sort of thing; it is much too dangerous. . . . In short, the Bafioti do not know elementals. According to them there are forces of power and life existing everywhere (and to-day they regard them as proceeding from one supreme god)—then there are themselves, and between the two, the souls of the dead. There is nothing more. It is with these forces, and not with souls or spirits, that ih black magic and its opponent, the white magic, operate.” It is the same with the most TT aevelopeal tribes of South America. ‘The most primitive animist idea consists in regarding nature as animate everywhere (Allbescelung) ; an idea which is by no means derived as a result of a know- ledge of the human soul, but which is formed at the same time as that by means of simple analogy.” 2 Junod the missionary very aptly expresses the character of this idea of nature. ‘The Ba-Ronga,’’ he says, “‘ like their allied race, the Bantus, are animists. In their eyes, the world is full of spiritual influences, sometimes favourable but more often formidable, which must be warded off. Do they imagine these clearly ? No: their animist ideas are very vague. . . . On the other hand, there are two or three conceptions which are very familiar to them, and these figure more clearly on the vague background of their beliefs. These are their conceptions of khombo (misfortune), sila (defilement), and ya (that which is forbidden).’’ 3 Even investigators of the Tylor and Frazer school, in describing what they see, use expressions which tend to modify their masters’ theory in the way I have indicated. Thus ‘the root idea,” says Skeat, ‘‘seems to be an all-pervading animism, involving a certain common vital principle (Semangat) in man and Nature, which, for want of a more suitable word, has been here called the Soul.’’4 In the island of Borneo, t Die Loango-Expedition, pp. 276-7, 313 2 eer Ehrenreich, Die Mythen und Legenden der Stid-A mericanischen Urvolker, eat 3 TR AO Les Ba-Ronga, p. 471. 4 Malay Magic, p. 579. THE LAW OF PARTICIPATION 103 the natives “ may be said to attribute a soul or spirit to almost every natural agent and to all living things.” But how are we to understand this animist doctrine? ‘‘ They feel them- selves to be surrounded on every hand by spiritual powers, which appear to them to be concentrated in those objects to which their attention is directed by practical needs. Adapt- ing a mode of expression familiar to psychologists, we may say that they have differentiated from a continuum of spiritual powers a number of spiritual agents with very various degrees of definiteness. Of these the less important are extremely vaguely conceived, but are regarded as being able to bring harm to men.” This continuum of spiritual powers, antecedent to the definite individualities which are the result of differentiation, we find described in North America in almost exactly the same terms by Miss Alice Fletcher. ‘‘ The Indians,” she says, “regarded all animate and inanimate forms, all phenomena as pervaded by a common life, which was continuous and similar to the will power they were conscious of in themselves. This mysterious power in all things they called Wakonda, and through it all things were related to man, and to each other. In the idea of the continuity of life, a relation was maintained between the seen and the unseen, the dead and the living, and also between the fragment of anything and its entirety.” 2 Would it be possible to bring more thoroughly home, in terms of animism, the mystic representations governed by the law of participation, which form the basis of prelogical mentality ? Finally, in his recent book,3 A. C. Kruijt also admits that, instead of the traditional animism, the mentality of primitives first of all imagines a continuum of mystic forces, a principle of continuous life, an Allbeseelung, and that individualities or personalities, souls, spirits, only appear in the second place. 3 It is therefore permissible to believe that, the more carefully one collects data, the more one detaches them from the animist interpretation which observers have only too frequently 1 Hose and Macdougall, ‘‘ Men and Animals in Sarawak,” J.A.J., xxxi. Pp. 174. 2 “ The Signification of the Scalp-lock (Omaha Ritual),’” J.A.J., xxvi. P- 437: : 3 Het Animisme in den Indischen Archipel, pp. 1-2 (1906). 104 HOW NATIVES THINK incorporated with them, even unconsciously, the more evident it will appear that the mentality of primitives, being mystic, is necessarily prelogical also : which means that, preoccupied above all with the mystic powers and properties of persons and things, it conceives of their relations under the law of participation without troubling about contradictions which rational thought cannot possibly tolerate. CHAPTER III THE FUNCTIONING OF PRELOGICAL MENTALITY I Ir would be idle to institute any comparison between the discursive processes of prelogical mentality and those of our thought, or to look for any correspondence between the two, for we should have no grounds on which to base a hypothesis. We have no a priori reason for admitting that the same pro- cess is used by both. The discursive operations of our rational thought—the analysis of which has been made familiar to us through psychology and logic—require the existence and the employment of much that is intricate, in the form of cate- gories, concepts, and abstract terms. They also assume an intellectual functioning, properly so called, that is already well-differentiated. In short, they imply an ensemble of con- ditions which we do not find existing anywhere in social aggregates of a primitive type. On the other hand, as we have seen, prelogical mentality has its own laws, to which its discursive operations must necessarily submit. In order to determine what these operations are, and how they are accomplished, our only resource is to describe and analyse them according to the direct connections we have observed between the collective representations. This is a most difficult task, both on account of the character of these same operations, and of the incompleteness of the documents at our disposal. The attempt I am about to make, therefore, will undoubtedly yield an imperfect and very unfinished outline only, but it will not have been useless if it shows that ‘The operations of prelogical mentality depend upon the law of participation, and cannot be explained apart from it. Before we begin to analyse these operations, we must first of all say something about the co-existence of the laws ‘of 106 HOW NATIVES THINK contradiction and participation. Are we to suppose that certain operations are governed entirely by the first, and others just as exclusively by the second, of these laws? Do we imagine, for instance, that every individual representation is the result of thought that is already logical, whilst collective representations alone submit to a law peculiar to prelogical mentality ? Water-tight compartments of this kind are incon- ceivable, if only because it is very difficult—indeed almost impossible—to trace a distinct line of demarcation between individual and collective representations. What can be more individual, to all appearances, than sense-perceptions ? Never- theless we have noted the extent to which the primitive’s sense-perceptions are enveloped in mystic elements which cannot be separated from them and which undoubtedly are collective in their nature. The same may be said of most of the emotions experienced and of most of the move- ments which take place almost instinctively at the sight of a certain object, even quite an ordinary one. In these communities as much as in our own, perhaps even more : so, the whole mental life of the individual is profoundly socialized, We must therefore expect to see the influence of the law of participation exercised, not only pre-eminently in what we have called collective representations, but also making itself felt more or less emphatically in all mental operations. Con- versely, the effect of the law of contradiction is already more or less strong and constant, first of all in operations which would be impossible without it (such as numeration, inference, etc.) and then also in those which are governed by the law of participation. There is nothing but what is changing and unstable, and this is one of the greatest difficulties with which we have to contend. In the mentality of primitive peoples, the logical and prelogical are not arranged in layers and separated from each other like oil and water in a glass. They permeate each other, and the result is a mixture which is a very difficult matter to differentiate. Since the laws of logic abso- lutely exclude, in our own thought, everything that is directly contrary to itself, we find it hard to get accustomed to a men- tality in which the logical and prelogical can be co-existent and make themselves equally perceptible in mental processes. FUNCTIONING OF PRELOGICAL MENTALITY 107 The prelogical element which our collective representations still contain is too small to enable us to reconstruct a mental state in which the prelogical, when dominant, does not exclude what is logical. What strikes us first of all is that prelogical mentality is little given to analysis. Undoubtedly in a certain sense every act of thought is synthetic, but when it is a question of logical thought this synthesis implies, in nearly every case, a previous analysis. Relations are expressed by judgments only after the food for thought has first been well digested, and subjected to elaboration, differentiation, and classification. Judgment deals with ideas which have been rigidly defined, and these are themselves the proof and product of previous logical processes. This previous work, in which a large number of successive analyses and syntheses occur and are recorded, is received ready-made by every individual in our communities when he first learns to talk, by means of the education inseparably bound up with his natural development ; so much so indeed that certain philosophers have believed in the supernatural origin of language. In this way the claims of logical thought are urged, established, and then confirmed in each individual mind by the uninterrupted constraining force of his social environment, by means of language itself and of what is transmitted by language. This is a heritage of which no member of our community is deprived, and which none would ever dream of refusing. Logical discipline is thus imposed upon his mental operations with irresistible force. The fresh syntheses which it effects must submit to the definitions of the concepts employed, definitions which the previous logical operations have legitimatized. In short, his mental activity, in whatever form it may be exercised, must submit to the law of contradiction. The conditions under which prelogical mentality operates are altogether different. There is no doubt that it, too, is transmitted socially by means of language and concepts without which it could not be exercised. It also implies work which has been previously accomplished, an inheritance handed down from one generation to another. But these concepts differ from ours,t and consequently the mental t Vide Chap, III. pp. 126-7. 108 HOW NATIVES THINK operations are also different. Prelogical mentality is essen- tially synthetic. By this I mean that the syntheses which compose it do not imply previous analyses of which the result has been registered in definite concepts, as is the case with those in which logical thought operates. In other words, the connecting-links of the representations are given, as a rule, with the representations themselves. In it, too, the syn- theses appear to be primitive and, as we have seen in our study of perception, they are nearly always both undecomposed and undecomposable. This, too, explains why primitive mentality seems both impervious to experience and insensible to contradiction in so many instances. Collective repre- sentations do not present themselves separately to it, nor are they analysed and then arranged in logical sequence by it. They are always bound up with preperceptions, pre- conceptions, preconnections, and we might almost say with prejudgments ; and thus it is that primitive mentality, just because it is mystical, is also prelogical. But, someone may object, if the mental operations of un- — civilized peoples differ from logical thinking in their mode of functioning, if their paramount law is the law of participa- tion, which a priori allows of these preconnections and partici- pations of participations which are so infinitely varied, if their mentality does finally escape the control of experience, will it not appear to us unbridled and unregulated, and just as purely arbitrary as it is impenetrable ? Now in nearly all inferior races we find, on the contrary, that the mentality is stable, . fixed and almost invariable, not only in its essential elements, but in the very content and even in the details of its representations. The reason is that this mentality, although not subordinate to logical processes, or rather, precisely because it does not submit to them, is not free. Its uniformity reflects the uniformity of the social structure with which it corresponds. Institutions fix beforehand, so to speak ne varietur, the combinations of collective representations which are actually possible. The number of the connecting-links between the representations and the methods by which they are connected are predetermined at the same time as the representations themselves. It is especially in the pre- connections thus established that the predominance of the FUNCTIONING OF PRELOGICAL MENTALITY 109 law of participation and the weakness of the strictly intellectual claims are made manifest. Moreover, collective representations as a rule form part of a mystical complex in which the emotional and passionate elements scarcely allow thought, as thought, to obtain any mastery. To primitive mentality the bare fact, the actual object, hardly exists. Nothing presents itself to it that is not wrapped about with the elements of mystery: every object it perceives, whether ordinary or not, moves it more or less, and moves it in a way which is itself predestined by tradition. For except for the emotions which are strictly individual and dependent upon immediate reaction of the organism, there is nothing more socialized among primitives than are their emotions. Thus the nature which is perceived, felt, and lived by the members of an undeveloped community, is necessarily predetermined and unvarying to a certain extent, as long as the organized institutions of the group remain un- altered. This mystical and prelogical mentality will evolve only when the primitive syntheses, the preconnections of collective representations are gradually dissolved and decom- posed; in other words, when experience and logical claims win their way against the law of participation. Then, in submitting to these claims, “‘ thought,” properly so called, will begin to be differentiated, independent, and free. Intellectual operations of a slightly complex kind will become possible, and the logical process to which thought will gradually attain, is both the necessary condition of its liberty and the indis- pensable instrument of its progress. IT In the first place, in prelogical mentality memory plays a much more important part than it does in our mental life, in which certain functions which it used to perform have been taken from it and transformed. Our wealth of social thought is transmitted, in condensed form, through a hierarchy of concepts which co-ordinate with, or are subordinate to, each other. In primitive peoples it consists of a frequently enormous number of involved and complex collective repre- sentations. It is almost entirely transmitted through the rm Py IIO | HOW NATIVES THINK memory. During the entire course of life, whether in sacred or profane matters, an appeal which without our active voli- tion induces ws to exercise the logical function, awakens in the primitive a complex and often mystic recollection which regulates action. And this recollection even has a special tone which distinguishes it from ours. The constant use of the logical process which abstract concepts involve, the, so to speak, natural use of languages relying upon this process, disposes our memory preferably to retain the relations which have preponderating importance from the objective and logical standpoint. In prelogical mentality both the aspect and tendencies of memory are quite different because its contents are of a different character. It is both very accurate and very emotional. It reconstructs the complex collective representations with a wealth of detail, and always in the order in which they are traditionally connected, according to relations which are essentially mystic. Since it thus, to a certain extent, supplements logical functions, it exercises the privileges of these to a corresponding degree. For instance, a representation inevitably evoked as the result of another frequently has the quality of a conclusion. Thus it is, as we shall see, that a sign is nearly always taken to be a cause. The preconnections, preperceptions, and preconclusions which play so great a part in the mentality of uncivilized peoples do not involve any logical activity ; they are simply committed to memory. We must therefore expect to find the memory extremely well developed in primitives, and this is, in fact, reported by observers. But since they un- reflectingly assume that memory with these primitives has just the same functions as with us, they show themselves both surprised and disconcerted by this. It seems to them that it is accomplishing marvellous feats, while it is merely being exercised in a normal way. ‘“‘ In many respects,’’ say Spencer and Gillen of the Australian aborigines, “their memory is phenomenal.’”’ ‘‘ Not only does a native know the track of every beast and bird, but after examining any burrow, he will at.once, from the direction in which the last track runs, tell you whether the animal is at home or not. . . . Strange as it may sound... the native will recognize the footprint FUNCTIONING OF PRELOGICAL MENTALITY xt of every individual of his acquaintance.’’! The earliest explorers of Australia had already referred to this marvellous power of memory. Thus Grey tells us that three thieves were discovered by their footprints. ‘‘I got hold of an in- telligent native of the name of Moyee-e-nan, and accompanied by him, visited the garden whence the potatoes had been stolen ; he found the tracks of three natives, and availing him- self of the faculty which they possess of telling who has passed from their footmarks, he informed me that the three thieves had been the two wives of a native . . . and a little boy named Dal-be-an.’”’2 Eyre is astonished at ‘the intimate knowledge they have of every nook and corner of the country they inhabit ; does a shower of rain fall, they know the very rock where a little water is most likely to be collected, the very hole where it is the longest retained. . . . Are there heavy dews at night, they know where the longest grass grows, from which they may collect . . . water in great abundance.”’ 3 W. E. Roth also draws attention to the exceptional memo- rizing powers of the North Queensland aborigines. He has heard them “reciting a song the delivery of which takes upwards of five nights for its completion (the Molonga set of corroborees). . . . The wonder is increased when it is remembered that the words are sung in a language of which the singers of both localities (ninety miles apart) are entirely ignorant. . . . A tribe will learn and sing by rote whole corroborees in a language absolutely remote from its own, and not one word of which the audience or performers can understand the meaning of. That the words are very care- fully committed to memory, I have obtained ample proof by taking down phonetically the same corroborees as performed by different-speaking people living at distances upwards of one hundred miles apart.” 4 | Von den Steinen has reported similar experiences in his explorations of the Xingu basin. “Every tribe knew the songs of the neighbouring tribes, without understanding their 1 Spencer and Gillen, The Native Tribes of Central Australia, p. 25. 2 Grey, Journals of Two Expeditions of Discovery in North-western and Western Australia, ii. p. 351. 3 Eyre, Expeditions into Central Australia, ii. p. 247.° 4 W.E. Roth, Ethnographical Studies among the N.W. Central Queensland Aborigines, Nos. 191, 199. 112 HOW NATIVES THINK exact meaning, a fact I was able to prove on numerous occa- sions.” In a large number of North American tribes we find incantations of a sacred character, faithfully transmitted from generation to generation, which are not understood, either by those who officiate or those who listen. In Africa, too, Livingstone likewise expressed surprise at the wonderful memory displayed by certain natives. “ These chief’s mes- sengers have most retentive memories ; they carry messages of considerable length great distances, and deliver them almost word for word. Two or three usually go together, and when on the way the message is rehearsed every night, in order that the exact words may be kept to. One of the native objections to learning to write is that these men answer the purpose of transmitting intelligence to a distance as well as a letter would.” 2 One specially noticeable form of the memory so highly developed in natives is that which preserves to the minutest detail the aspect of regions they have traversed, and this permits of their retracing their steps with a confidence which amazes Europeans. Among the North American Indians this topographical memory “is something marvellous ; it is quite enough for them to have been in a place once only for them to have an exact image of it in their minds, and one which will never be lost. However vast and untravelled a forest may be, they cross it without losing their way, once they have got their bearings. The people of Acadia and the Gulf of St. Lawrence would often embark in the frail canoes to go to Labrador. . . . They would sail for thirty or forty leagues without a compass, and disembark at the precise spot at which they had decided to land. . . . Even when the sky is overcast, they will follow the sun for several days without making a mistake.’ Charlevoix is inclined to attribute this to an innate faculty. ‘‘ This gift is inborn ; it is not the result of their observations, nor a matter of habit; children who have scarcely ever left their village go about with as much confidence as those who have travelled the country.” Like the Australian aborigines, ‘they have a marvellous gift for knowing whether anyone has passed a certain place. On 1 K. von den Steinen, Unter den Naturvélkern Zentralbräsiliens, p. 268. 2 Livingstone, Zambesi and its Tributaries, p. 267 (1865). FUNCTIONING OF PRELOGICAL MENTALITY 113 the shortest grass, the very hardest ground, and even on the stones they discover tracks, and by their direction and the outline of the foot and the way in which the person has stepped aside, they distinguish the footprints of different races, and the tracks of men from those of women.” 1 Pechuël-Loesche, who made a study of similar phenomena on the west coast of Africa, rightly distinguishes between what he calls the sense of locality (Ovtsinn) and the sense of direction (Richtsinn). What we call sense of locality is simply a memory for places; it is an acquired aptitude, founded on a very strong memorizing faculty, and on the recognition of an infini- tude of detail which allows one to return to the same region in space... . Beyond and above this sense of locality is the feeling of direction (Richtungsgefühl), or the sense of direction (Richisinn). It is not a special sense, it is the sense of locality carried to its highest degree of perfection, and therefore again a form of memory. He who has acquired it will never lose his way again. Undoubtedly, “ he will not be certain of arriving without fail at a given point, but at any rate he will always start in the direction which leads to it . . . under the open sky, in the fog, rain, snow, or in the depths of night. Nevertheless I have noticed that this sense may be entirely at fault in violent storms. . . . Persons gifted with a strong sense of locality appear to be exempt from vertigo and sea-sickness.” 2 This analysis help us to understand similar observations made by other explorers when speaking of certain individuals belonging to primitive peoples. Thus an Australian named Miago “‘ could indicate at once and correctly the exact direc- tion of our wished-for harbour, when neither sun nor stars were shining to assist him. He was tried frequently, and under very varying instances, but strange as it may seem, he was invariably right. This faculty—though somewhat analogous to one I have heard attributed to the natives of North America—had very much surprised me when exercised on shore, but at sea, out of the sight of land, it seemed beyond belief, as assuredly it is beyond explanation.” This same Miago ‘‘ remembered accurately the various places we had 1 Charlevoix, Journal d'un Voyage dans l’Amérique Septentrionale, ili. p. 239. 2 Die Loango-Expedition, iii. 2, pp. 28-9. 8 II4 . HOW NATIVES THINK visited during the voyage ; he seemed to have carried the ship’s track in his memory with the most careful accuracy.” 1 The same facultv has been noted in Fuegians. “ Niqueaccas was so well acquainted with all the coast between 47° and the Straits of Magellan that upon being taken to a high hill, immediately after landing from a cruise, in which they had been far out of sight of land, he pointed out the best harbours and places for seal then visible. . . . The boy Bob, when only ten years old, was on board the Adonea at sea. As the vessel approached land, Low asked him where a harbour could be found. As soon as he understood what was meant, which was an affair of some difficulty, for he could then speak but very little English, he got up on the vessel’s bulwark, and looked anxiously around. After some hesitation, he pointed to a place where the ship might go, and then went to the lead- line, and made signs to Mr. Low that he must sound as he approached the land... an extraordinary proof of the degree in which the perceptive and retentive faculties are enjoyed by these savages.”’ 2 ) It was evidently a case of thoroughly well-developed “sense of locality ’’ attaining the degree of superiority in which it becomes, as Pechuél-Loesche calls it, a feeling of direction, but there is nothing marvellous in it but a local memory that is out of the common. Von den Steinen has given us a good, though less surprising, description of a similar case. ‘‘ Antonio (a Bakairi) would see and hear everything, and commit to memory the most insig- nificant details, and by means of such signs of locality he exercised the faculty which educated people call the sense of direction. If I had not convinced myself by frequently questioning him I could scarcely have believed that anybody, without written notes, could have acquired in a single voyage on an ordinary river so exact a knowledge of the special features of its course. Antonio not only remembered every bend but he was able to tell me when I asked him, whether there were two or three bends before we arrived at such and such a place. He had the map of it in his head; or, to t Stokes, Discoveries in Australia, i. pp. 222-3 (1846). a Fitzroy, Narrative of the Surveying Voyages of the ‘ Adventure" and the “ Beagle,’ ü. pp. 192-3. FUNCTIONING OF PRELOGICAL MENTALITY 115 express it more accurately, he had retained in their right order a number of apparently unimportant facts such as a tree here, a gunshot there, a little further on some bees, and so on.’’ ! This extraordinary development of memory, and of a memory which faithfully reproduces the minutest details of sense-impressions in the correct order of their appearance, is shown moreover by the wealth of vocabulary and the gram- matical complexity of the languages. Now the very men who speak these languages and possess this power of memory are (in Australia or Northern Brazil, for instance) incapable of counting beyond two or three. The slightest mental effort involving abstract reasoning, however elementary it may be, is so distasteful to them that they immediately declare themselves tired and give it up. We must admit therefore, as we have already said, that with them memory takes the place (at very great cost, no doubt, but at any rate it does take the place) of operations which elsewhere depend upon a logical process. With us, in everything that relates to intellectual functions, memory is confined to the subordinate rôle of regis- tering the results which have been acquired by a logical elaboration of concepts. But to prelogical mentality, recol- lections are almost exclusively highly complex representations which succeed each other in unvarying order, and in which the most elementary of logical operations would be very difficult (since language does not lend itself thereto), supposing that tradition allowed of their being attempted, and granting that individuals possessed enough boldness to entertain the idea. Our thought, in so far as it is abstract, solves at one swoop a great number of problems implied in one single state- ment, provided that the terms employed are sufficiently general and definite. This is what prelogical mentality could. not even imagine, and this accordingly makes it so difficult for us to reconstruct such a mentality. The amanuensis of the eleventh century who laboriously reproduced page by page the manuscript which was the object of his pious endeavour, is no further removed from the rotary machine of the great newspaper offices which prints off hundreds of thousands of copies in a few hours, than is prelogical mentality, in which t K. von den Steinen, Unter den Naturvélkern Zentralbräsiliens, pp. 155-7. 110. HOW NATIVES THINK the connections between the representations are preformed, and which makes use of memory almost entirely, from logical thought, with its marvellous stock of abstract concepts. AS III AX VS y Are we to take it for granted, then, that this mentality, even in the very lowest social aggregates, makes no use of concepts whatever? Certainly not: the ofttimes compli- cated language that it speaks, the institutions transmitted from generation to generation, are sufficient to prove the contrary. Yet the concepts that are used in such aggregates for the most part differ from our own. The mind that forms and employs them is not merely prelogical. It is essentially mystic, and if its mystic character determines, as we have already seen, the way in which it perceives, it exercises no less influence upon its methods of abstraction and generaliza- tion, that is, the way in which it creates its concepts. Especi- * al M ally in that which concerns representations which are strictly collective, prelogical mentality most frequently arrives at its abstraction by the law of participation. One can imagine that it is extremely difficult to give proof of this, the testimony afforded by observers being necessarily interpreted by concepts with which they are familiar, and which come within the limits of our logical thought. Nevertheless Spencer and Gillen have reported a certain number of facts. which permit of our seeing fairly clearly how prelogical mentality practises abstraction. ‘“ When asked the meaning of certain drawings . . . the natives will constantly answer that they are only play work, and mean nothing .. . but . . . similar drawings, only drawn on some ceremonial object or in a particular spot, have a very definite meaning. . . . The same native will tell you that a special drawing in one spot has no meaning, and yet he will tell you exactly what it is supposed to signify when drawn in a different spot. The latter, it may be remarked, is always on what we may call sacred ground, near to which the women may not come.” !—‘‘ A nurtunja (a sacred pole), is symbolic of one, and only one, thing, though, as far as 1 The Native Tribes of Central Australia, p. 617. ie FUNCTIONING OF PRELOGICAL MENTALITY 117 its appearance and structure are concerned, it may be pre- cisely similar to a nurtunja which means something totally different. Suppose, for example, as on the last occasion, a large churinga or a nurtunja represented a gum-tree, then, in the mind of the native it becomes so closely associated with that object that it cannot possibly mean anything else ; and if a precisely similar churinga or nurtunja were wanted an hour afterwards to represent, say an emu, then a new one must be made.’’! Conversely, the same object, in different circumstances, may have very diverse meanings. “The various parts of the waninga’’ (a sacred symbol of a totemic animal or plant) ‘‘ have very different meanings, but it must be remembered that the same structure will mean one thing when it is used in connection with one totem, and quite a different thing when used in connection with another.” 2 Finally, with regard to designs which seem to have a geometric appearance, which have been collected from these same abori- gines, Spencer and Gillen say : ‘‘ The origin of these geometric drawings is quite unknown, and their meaning, if they have one, is a purely conventional one. Thus, for example, a spiral or series of concentric circles incised on the face of one particular churinga will signify a gum-tree, and a precisely similar design on another churinga will indicate a frog.” 3 Here we have very clear instances of what we shall call mystic abstraction which, different as it is from logical abstrac- tion, is none the less the process which primitive mentality would frequently make use of. In fact, if exclusive attention be one of the primary conditions of abstraction, and if exclusive attention be necessarily paid to the features which are most interesting and important to the subject, we know which these features will be to a mentality which is mystic and prelogical. Beyond and above aught else, they are those which establish relations between the visible, tangible, concrete objects and the invisible and mysterious forces which compass them about, the spirits, phantoms, souls, etc., which secure to persons and things their mystic properties and powers. The attention of primitives, like their perception, is oriented differently from our own Their abstraction accordingly t The Native Tribes of Central Australia, p. 346. 2 Ibid., p. 308. 3 The Northern Tribes of Central Australia, p. 697. 118 HOW NATIVES THINK is carried out differently, and dominated by the law of participation. Such abstraction is very difficult for us to reconstruct effectively. How are we to understand the fact reported by Spencer and Gillen in the first of the observations just quoted : that two precisely similar drawings when found in different places may represent, in the one case, a certain definite object, and in the other, nothing at all? To our minds, the most essential relation which a drawing bears is that of resemblance. Undoubtedly such a drawing may possess a symbolic and religious significance and at the same time arouse mystic ideas accompanied by strong emotions: such drawings, for instance, as the frescoes of Fra Angelico in St. Mark’s, Florence. But those are features which are evoked by association of ideas and the resemblance remains the fundamental relation. On the contrary, that which interests prelogical mentality above all is the relation of the semblance, as of the object represented, to the mystic force within it. Without such participation, the form of the object or the design is a negli- gible factor. This is the reason that the design, when traced or engraved upon a sacred object, is more than a semblance ; it participates in the sacred nature of the object and is imbued with its power. 1 Thus the European observer, when attempting to interpret the designs made by primitives, is nearly sure to go wrong. Von den Steinen proved this in Brazil. On his side Parkinson says: ‘‘ We find ourselves faced with a difficult problem. The Mitteilungen sees in these drawings serpents, and in fact there is something that recalls the head and body of such; but the Baining affirm that they represent pigs. . . . The figure that follows might if necessary pass for a face, but according to the natives, it represents a club, though it has not the remotest resemblance to such a thing. Certainly nobody in the wildest flights of imagination would have conceived of such an explanation. . . . I was inclined to regard the three circular forms which follow, as eyes, but the natives robbed me of this illusion, assuring me that eyes could not be reproduced. The explanations of the decorations have been given me by the Baining themselves; there can be no doubt on this point therefore, since those who execute them associate a definite idea with their drawings, although in nearly every case we are unable to see the con- nection, for the design does not in any way resemble the object in question. We see how incorrect it is to interpret the ornamental decorations of a primitive people according to their resemblance to any object known to us. The Baining see in these conventional designs a shell, a leaf, a human form, etc. The idea is so firmly fixed in their minds that one can see the stupefied wonder on their faces when they are asked the meaning of these designs : they cannot conceive that anybody should fail to recognize at once the meaning of the decorations ”” (Parkinson, Dreissig Jahre in der Südsee, pp. 621-7 ; cf. pp. 234, 235). FUNCTIONING OF PRELOGICAL MENTALITY xrr9 When this same design is found elsewhere, upon an object which possesses no sanctity, it is less than asemblance. Having no mystic signification, it has none of any kind. The details which Catlin gives us in his tales of the portraits which he made of the Mandan chiefs confirms this view. Catlin talks incessantly of the surprise and fear which the sight of these portraits excited in the Mandans. Nevertheless these same Indians, from time immemorial, were accustomed to represent the most striking events in their history, and even rude traces of the chiefs’ features, on their banners. How then are we to explain the terror which these portraits of Catlin’s caused? By their greater fidelity to the originals ? No. The truth was that the Mandans found themselves confronted by unwonted semblances, implying a mystic participation hitherto unknown to them, and consequently, like everything unknown, highly dangerous. Their own drawings also expressed participation, but it was a definite one, hence their security. Those made by Catlin expressed a different one, because the methods he used were strange to them, and his likenesses were “speaking.” Thus in this case, as in the preceding ones, prelogical mentality makes its abstraction from the mystic standpoint. If there is no mystic participation felt, the form of the likeness is unperceived, or at any rate, does not arrest the attention. This is what the European observer interprets by saying that the design has then ‘‘no meaning whatever.’ It does not mean that the primitive fails to recognize the design, but that unless he abstracts mystically, he makes no abstraction at all. The note which relates to the nurtunja is equally clear. The Aruntas cannot imagine that the same nurtunja figures first as a tree, and then as an emu: rather than that, they take the trouble of making a second nurtunja, otherwise exactly like the first, when they want to represent the emu. In this we might see a ritual observance which does not allow of the same object being used, with a religious significance, more than once. But Spencer and Gillen sweep away such an explanation. They explicitly state that the Aruntas attribute a different signification to two objects which are similar. It is an admirable instance of mystic abstraction. One of the two nurtunja participates mystically of the nature 120 HOW NATIVES THINK of the tree, the other of the emu’s nature, and this suffices to render them totally different, so that one cannot be sub- stituted for the other. Their identity in form does not interest the Aruntas any more than we should be interested in the identical sounds of “‘ weigh’”’ and “‘ way,” for instance. Just as we constantly use these words without paying any attention to that identity, so prelogical mentality remains indifferent to the resemblance in form between the two objects. Its attention is fixed on the mystic participation which gives each its sacred character. Similarly, on a certain churinga, one design represents a gum-tree ; on another an absolutely similar design stands for a frog, and the observers conclude accordingly that to the Australian aborigines these designs bear a purely “ conven- tional’? meaning. They ought not to say “conventional,” but ‘‘ mystic’’ however. The design possesses no interest but in so far as it realizes a mystic participation, and this, in its turn, depends entirely upon the mystic nature of the churinga upon the surface of which the design has been traced. Their resemblance is no more noticed by the primitive than would the relative places of tonic and fifth in different scales be regarded by the musician when he looks at his score. Spencer and Gillen themselves say that an arrangement that signifies one thing when used in connection with a certain totem is something quite different when it relates to another. But the churinga have the same mystic character as the totems, and thus the same participations become possible. From the first observation recorded above it follows that the place occupied by a person, an object, an image is of para- mount importance, at any rate in some cases, to the mystic properties of such a person, object, or image. There is a corresponding participation between a definite place, as a place, and the objects and entities which are found there, and it thus possesses certain mystic properties peculiar to it. To the primitive mind space does not appear as a homogeneous unity, irrespective of that which occupies it, destitute of properties and alike everywhere. On the contrary, each social group among the tribes of Central Australia, for instance, feels itself mystically bound up with the portion of ground it occupies or travels over; it has no conception that it might FUNCTIONING OF PRELOGICAL MENTALITY xr2r occupy another, or that some other group might inhabit the region it fills. Between the soil and the group participa- tions exist, amounting to a kind of mystic property which cannot be transferred, stolen, or acquired by conquest. More- over, in the region thus defined, each locality with its charac- teristic aspect and form, its own peculiar rocks and trees, springs and sand-heaps, etc., is in mystic union with the visible or invisible beings who have revealed their presence there, or dwell there, and with the individual spirits who there await their reincarnation. Between themselves and their locality there is reciprocal participation: without them it could not be what it is, and it is equally necessary to make them what they are. This it is which Spencer and Gillen designate ‘local relationship,” ! and it accounts for those “totemic pilgrimages’’ of which they have furnished so interesting a description. But if it be thus, we have fresh reason for believing that the prelogical mind does not, in general, practise abstraction at all as ours are accustomed to do. The condition of our abstraction is the logical homogeneity of the concepts which permits of their combination. Now this homogeneity is closely bound up with the homogeneous representation of space.\ If the prelogical mind, on the contrary, imagines the various regions in space as differing in quality, as determined by their mystic participations with such and such groups of persons or objects, abstraction as we usually conceive of it becomes very difficult to such a mind, and we shall find that its place is taken by the mystic abstraction which is the result of the law of participation. IV The principles and processes peculiar to prelogical mentality appear to stand out more distinctly when it generalizes even than when it is a case of abstracting. I am not speaking of concepts which more or less resemble our own, the existence of which is testified by the linguistic vocabulary, and which t The Native Tribes of Central Australia, pp. 14, 303, 544; The Northern Tribes of Central Australia, p. 29. 122 | HOW NATIVES THINK fairly represent what have been termed generic ideas, such as man, woman, dog, tree, and so on. In the following chapter we shall find that the element of generality in these concepts is usually restricted and counterbalanced by the very special determination of the classes of beings or objects they designate. With this exception such concepts correspond very fairly with certain of our general ideas. But in the collective repre- sentations, properly so called, of primitives, particularly in those relating to their institutions and religious beliefs, we find generalizations of quite a different kind, which it is ex- tremely difficult for us to reconstruct, the analysis of which would probably allow us to seize the mystic and prelogical mind in the very act, as it were. We might endeavour to trace back such generalizations, starting, for instance, from certain myths and certain totemic beliefs which are attested by rites and ceremonies. But if it be possible it would be better still to apprehend them directly, and in the very combination of the elements which form them. In Lumholtz’ excellent works upon Unknown Mexico, we find — observations with respect to the Huichol Indians which throw strong light upon the way in which the prelogical mind generalizes. “Corn, deer, and hikuli ” (a sacred plant) “are, in a way, one and the same thing to the Huichol.’’! At first this identi- fication seems absolutely inexplicable. To make it intelligible, Lumholtz explains it on utilitarian grounds: ‘‘ Corn is deer (food substance) and hikuli is deer (food substance) and corn is hikuli . . . all being considered identical in so far as they are food substances.’”’! This explanation is a probable one, and it undoubtedly becomes the one held by the Huichols themselves as the forms of their ancient faiths gradually lose their primitive meaning in their minds. But according to Lumholtz’ own explanation, the Huichols who express them- selves thus view the matter from another aspect: it is the mystic properties in these things, so differently regarded by us, that unites them in one and the same idea. The hikuli is a sacred plant which men (destined and prepared for this end by a series of very complicated rites) gather every year with great ceremonial, in a remote district, and at the cost | tC, Lumholtz, Symbolism of the Huichol Indians, p. 22. FUNCTIONING OF PRELOGICAL MENTALITY 123 of much fatigue and personal privation: the existence and the well-being of the Huichols are mystically connected with the harvesting of this plant. Their corn harvest is absolutely dependent upon it. If the hikuli were to fail, or were not gathered according to the obligatory rites, the cornfields would not yield their usual crops. But deer, in their relations with the tribe, present the same mystic characteristics. Deer- hunting, which takes place at a certain definite time of year, is an essentially religious function. The welfare and preserva- tion of the Huichols depends on the number of deer killed at this time, just as they depend upon the quantity of hikuli which is gathered; and the chase is accompanied by the Same ceremonial practices and evokes the same collective emotions as the search for the sacred plant. Hence results the identification of the hikuli, deer, and corn, which we find asserted several times over. “A layer of straw had been spread outside of the temple at the right side of the entrance, and on this the deer was carefully deposited. It was thus received in the same way as the corn-rolls, because in the Indian conception, corn is deer. According to the Huichol myth, corn was once a deer.’”’ —‘‘ To the Huichol so closely are corn, deer, and hikuli associated that by consuming the broth of the deer meat and the hikuli they think the same effect is produced, namely, making the corn grow. Therefore, when clearing the fields they eat the hikuli before starting the day’s work.”’ 2 It seems then that in these collective representations of the Huichols (representations which are inseparable, as we know, from intense religious emotions, which are also collec- tive), the hikuli, deer, and corn participate in mystic qualities of the highest importance to the tribe, and, for this reason, are considered as “‘the same thing.” This participation, which rs felt by them, does not present the confusion which we, despite all our efforts, see in it. Just because their collective representations are bound together by the law of participation, nothing seems simpler or more natural to them and, we may add, more necessary. The prelogical and mystic mentality is exercised in this way without constraint or effort, and without being yet controlled by the claims of logical thought. 1 C. Lumholtz, Unknown Mexico, ii. p. 45. a Ibid, ii. p. 268. 124 HOW NATIVES THINK But this is not all, and Lumholtz will show us how the participations just instanced are compatible with others of a like nature. ‘It has been pointed out,” he writes, “ that the deer is considered identical with hikuli and hikuli identical with corn, and certain insects identical with corn. The same tendency to consider heterogeneous objects as identical may be observed in the fact that a great variety of objects are considered as plumes. Clouds, cotton wool, the white tail of a deer, the deer’s antlers, and even the deer itself are con- sidered as plumes, and all serpents are believed to have plumes.” 1 Here, then, we have the deer, which was already corn, and hikuli, which is also plumes. Lumbholtz frequently lays stress upon this. ‘‘ Hairs from the tail of a deer are tied round outside of the feathers ’’ (on a ceremonial arrow). “It will be remembered that not only are deer-antlers plumes, but also the deer himself, and here is a striking illustration of the conception of the animal, his hair being employed in the place of feathers.’’ 2 Now we know from other passages that to the Huichol mind feathers possess mystic properties of a very special kind. ‘“‘ Birds, especially eagles and hawks . . . hear everything ; and the same is the case with their plumes ; they also hear, the Indians say, and have mystic powers. Plumes are to the Huichols health, life, and luck-giving symbols. By their help, the shamans are capable of hearing everything that is said to them from below the earth and from all the points of the world, and perform magic feats. The feathers of the vulture and of the raven are not considered as plumes. All plumes are desirable as attachments to ceremonial objects ; therefore a Huichol has never too many of them. There is, however, one plume of special merit, and that is, strange to say, the deer. Everyone who kills a deer comes into possession of a precious plume, that insures him health and luck. . . . Not only the antler, but the whole body of the deer is, in the Huichol mind, a plume, just as a bird is called a plume; and I have met with instances where the hair from the tail of a deer actually served as plume attachments on ceremonial arrows.”’ 3 It is, then, the presence of mystic qualities both in birds 1 C. Lumholtz, Symbolism of the Huichol Indians, p. 212. 2 Ibid., p. 96. 3 Ibid., p. 21. ap FUNCTIONING OF PRELOGICAL MENTALITY 125 (and their feathers) and the deer (and the hair of its tail) which makes the Huichol saying: ‘‘ The deer is a plume ” intelli- gible. Lumholtz accounts for this as due to “a strong tendency to see analogies ; what to us are called heterogeneous phenomena are by them considered as identical.” 1 But what really is this tendency? And how can the Huichols perceive any analogy between an eagle’s feather, a grain of corn, the body of a deer, the hikuli plant, if it be not mystic analogies, and all the more so, because it is not merely a question of analogy or association, but of identity ? Lum- holtz is very emphatic on this point: to the Huichols, the deer is hikuli, the hikuli is corn, the corn 7s a plume.—Else- where we learn that most gods and goddesses ave serpents, and so are the pools of water and the springs in which the deities live: so too are the staffs used by the gods. From the standpoint of logical thought, such “identities ” are, and remain, unintelligible. One entity is the symbol of another, but not that other. But to the prelogical mind these identi- ties are comprehensible, for they are identities of participa- tion. The deer ts hikuli, or corn, or plume, just as the Bororo is an arara, and as, in general, the member of a totemic group is his totem. The facts Lumholtz has related are of pro- found significance. It is by virtue of participation that the eagle’s feather possesses the same mystic properties as the eagle itself, and the whole body of the deer the same as those in its tail; and it is by virtue of participation, too, that the deer becomes identified with the eagle’s plume or the hikuli plant. Without insisting further on this point, we have here the principle of a generalization which proves disconcerting to logical thought, but is quite natural to prelogical mentality. It is presented to our minds in the form of that which, for want of a better term, we have called the preconnections of the collective representations, since ‘identities ’’ of the kind we have been considering always occur to each individual mind at the same time as the representations themselves, and this accounts for the profound difference between these “ repre- sentations ’’ and our own, even when it is a case of fairly general concepts which are somewhat similar. When a 1 C. Lumholtz, Unknown Mexico, ii. p. 233. 126 HOW NATIVES THINK primitive, an Australian or Indian, for instance, imagines ‘* deer ’’ or “‘ plume ”’ or “‘ cloud,” the generic image presented to his mind implies something quite different from the some- what analogous image which the European mind would conjure up in the same circumstances. Our concepts are surrounded by an atmosphere of logical potentiality. This is what Aristotle meant when he said that we never think of the particular as such. When I imagine Socrates as an individual, I think of the man Socrates at the same moment. When I see my horse or my dog, I certainly perceive their special characteristics, but these also as belong- ing to the species horse or dog. Strictly speaking, their image may be imprinted on my retina and appear to my consciousness as quite distinctive, as long as I am not paying attention to it. But directly I apprehend it, it is inseparable from everything connoted by the terms “horse” and “ dog ’”’— that is, not only from an infinite number of other potential images like the first, but also from the sustained consciousness which I have both of myself and of a possible, logically ordered, — conceivable world of experience. And since each of my concepts can be broken up into others which in their turn can be analysed, I know that I can pass from these to others by definite stages which are the same for all minds resembling | my own. I know that logical processes, if they be correct, and their elements drawn from experience as they should be, will lead me to definite results which experience will confirm, however far I may pursue them. In short, logical thought implies, more or less consciously, a systematic unity which is best realizable in science and philosophy. And the fact that it can lead to this is partially due to the peculiar nature of its concepts, to their homogeneity and ordered regularity. This is material which it has gradually created for itself, and without which it would not have been able to develop. Now this material is not at the command of the primitive mind. Primitive mentality does indeed possess a language, but its structure, as a rule, differs from that of our languages. It actually does comprise abstract representations and general ideas; but neither this abstraction nor this generalization resembles that of our concepts. Instead of being surrounded by an atmosphere of logical potentiality, these representations FUNCTIONING OF PRELOGICAL MENTALITY 127 welter, as it were, in an atmosphere of mystic possibilities. There is no homogeneity in the field of representation, and for this reason logical generalization, properly so called, and logical transactions with its concepts are impracticable. The element of generality consists in the possibility—already predetermined —of mystic action and reaction by entities upon each other, or of common mystic reaction in entities which differ from each other. Logical thought finds itself dealing with a scale of general concepts varying in degree, which it can analyse or synthesize at will. Prelogical thought busies itself with col- lective representations so interwoven as to give the impression of a community in which members would continually act and react upon each other by virtue of their mystic qualities, participating in, or excluding, each other. V Since abstraction and generalization mean this for pre- logical mentality, and its preconnections of collective repre- sentations are such, it is not difficult to account for its classi- fication of persons and things, strange as it frequently appears to us. Logical thought classifies by means of the very opera- tions which form its concepts. These sum up the work of analysis and synthesis which establishes species and genera, and thus arranges entities according to the increasing generality of the characters observed in them. In this sense classifica- tion is not a process which differs from those which «have preceded or will follow it. It takes place at the same time as abstraction and generalization: it registers their results, as it were, and its value is precisely what theirs has been. It is the expression of an order of interdependence, of hierarchy among the concepts, of reciprocal connection between persons and things, which endeavours to correspond as precisely as possible with the objective order in such a way that concepts thus arranged are equally valid for real objects and real persons. It was the governing idea which directed Greek philosophical thought, and which inevitably appears as soon as the logical mind reflects upon itself and begins consciously to pursue the end to which it at first tended spontaneously. But to the primitive mind this predominating concern for 128 HOW NATIVES THINK objective validity which can be verified is unknown. Charac- teristics which can be discerned by experience, in the sense in which we understand it, characteristics which we call objec- tive, are of secondary importance in its eyes, or are important only as signs and vehicles of mystic qualities. Moreover, the primitive mind does not arrange its concepts in a regular order. It perceives preconnections, which it would never dream of changing, between the collective representations; and these are nearly always of greater complexity than concepts, properly so called. Therefore what can its classifications be? Per- force determined at the same time as the preconnections, they too are governed by the law of participation, and will present the same prelogical and mystic character. They will betoken the orientation peculiar to such a mind. The facts already quoted are sufficient proof of this. When the Huichols, influenced by the law of participation, affirm the identity of corn, deer, hikuli and plumes, a kind of classification has been established between their represen- tations, a classification the governing principle of which is a common presence in these entities, or rather the circulation among these entities, of a mystic power which is of supreme importance to the tribe. The only thing is that this classifica- tion does not, as it should do in conformity with our mental processes, become compacted in a concept which is more comprehensive than that of the objects it embraces. For them it suffices for the objects to be united, and felt as such, in a complexity of collective representations whose emotional force fully compensates, and even goes beyond, the authority which will be given to general concepts by their logical validity at a later stage. In this way the classifications to which Durkheim and Mauss have called our attention, noting their very different characteristics from those which distinguish our logical classi- fications, may again be explained. In many undeveloped peoples—in Australia, in West Africa, according to Dennett’s recent book,! among the North American Indians, in China and elsewhere—we find that all natural objects—animals, plants, stars, cardinal points, colours, inanimate nature in general—are arranged, or have been originally arranged, in t At the Back of the Black Man’s Mind (London, 1906). FUNCTIONING OF PRELOGICAL MENTALITY 129 the same classes as the members of the social group, and if the latter are divided into so many totems, so, too, are the trees, rivers, stars, etc. A certain tree will belong to such and such a class, and will be used exclusively to manufacture the weapons, coffins, etc., of men who are members of it. The sun, according to the Aruntas, is a Panunga woman, that is, she forms part of the sub-group which can only intermarry with members of the Purula sub-group. Here we have some- thing analogous with that which we have already noticed about associated totems and local relationship, a mental habit quite different from our own, which consists in bringing together or uniting entities preferably by their mystic parti- cipations. This participation, which is very strongly felt between members of the same totem or the same group, between the ensemble of these members and the animal or plant species which is their totem, is also felt, though un- doubtedly to a lesser degree, between the totemic group and those who have the same location in space. We have proofs of this in the Australian aborigines and in the North American Indians, where the place of each group in a common camping- ground is very precisely determined according to whether it comes from north or south or from some other direction. Thus it is felt once more between this totemic group and one of the cardinal points, and consequently between this group and all that participates in it, on the one hand, and this cardinal point and all that participates in it (its stars, rivers, trees, and so forth), on the other. In this way is established a complexity of participations, the full explanation of which would demand exhaustive acquaintance with the beliefs and the collective representations of the group in all their details. They are the equivalent of, or at least they correspond with, what we know as classi- fications: the social participations being the most intensely felt by each individual consciousness and serving as a nucleus, as it were, around which other participations cluster. But in this there is nothing at all resembling, save in appearance, our logical classifications. These involve a series of concepts whose extent and connotation are definite, and they constitute an ascending scale the degrees of which reflection has tested. The prelogical mind does not objectify nature thus. It lives 9 130 HOW NATIVES THINK it rather, by feeling itself participate in it, and feeling these participations everywhere; and it interprets this complexity of participations by social forms. If the element of generality exists, it can only be sought for in the participation extending to, and the mystic qualities circulating among, certain entities, uniting them and identifying them in the collective representation. In default of really general concepts, therefore, primitive mentality is conversant with collective representations which to a certain extent take their place. Although concrete, such representations are extremely comprehensive ih this respect, that they are constantly employed, that they readily apply to an infinite number of cases, and that from this point of view they correspond, as we have said, with what categories are for logical thought. But their mystic and concrete nature has often puzzled investigators. These did indeed note its importance and could not fail to draw attention to it, though at the same time they realized that they were face to face with a method of thinking which was opposed to their own — mental habits. Some examples in addition to those already quoted will help to make us realize these representations, which are general without however being at the same time abstract. In the Yaos,t Hetherwick notes beliefs which appear in- comprehensible to him. He cannot understand how it is that the lisoka (the soul, shade or spirit) can be at once both per- sonal and impersonal. In fact, after death the asoka becomes mulungu. This word has two meanings: one, the soul of the dead, the other, “‘ the spirit world in general, or more properly speaking the aggregate of the spirits of all the dead.” This would be conceivable if mulungu meant a collective unity formed by the union of all the individual spirits ; but this explanation is not permissible, for at the same time mulungu signifies “a state or property inhering in something, as life or health inheres in the body, and it is also regarded as the agent in anything mysterious. ‘It is mulungu’ is the Yao exclamation on being shown anything that is beyond the range of his under- standing.” This is a characteristic trait which we shall find 1 Hetherwick, ‘‘ Some Animistic Beliefs among the Yaos of Central Africa,” J'A.I., xxxii. pp. 89-95. FUNCTIONING OF PRELOGICAL MENTALITY 131 in all collective representations of this nature: they are used indifferently to indicate a person or persons, or a quality or property of a thing. To get out of the difficulty, Hetherwick distinguishes between what he calls ‘three stages of animistic belief: (x) the human /isoka or shade, the agent in dreams, delirium, etc.; (2) this lisoka regarded as mulungu, and an object of worship and reverence, the controller of the affairs of this life, the active agent in the fortunes of the human race; (3) mu- lungu as expressing the great spirit agency, the creator of the world and all life, the source of all things animate or inani- mate.’ Itseems as if Hetherwick, like the French missionaries of old in New France, tends to interpret what he observes by the light of his own religious beliefs, but he adds, in good faith: ‘And yet between these three conceptions of the spirit nature no definite boundary line can be drawn. The distinction in the native mind is ever of the haziest. No one will give you a dogmatic statement of his belief on such points.” If Hetherwick did not get from the Yaos the answers he wanted, it may possibly have been because the Yaos did not understand his questions, but it was largely because he did not grasp their ideas. To the Yaos the transit from the personal soul, before or after death, to the impersonal soul or to the mystic quality which pervades every object in which there is something divine, sacred and mystic (not super- natural, for on the contrarv nothing is more natural to primi- tive mentality than this kind of mystic power) is not felt. To tell the truth, there is not even such transit: there is “identity governed by the law of participation ” such as we found in the case of the Huichols, entirely different from logical identity. And through the perpetual working of the law of participation, the mystic principle thus circulating and spreading among entities may be represented indiffer- ently as a person or subject, or a property or power of the objects which share it, and consequently an attribute. Prelogical mentality does not consider there is any difficulty about this. It is the same with the North American Indians, about whom we have abundant and definite information. Miss 132 HOW NATIVES THINK Alice Fletcher,! in describing the mysterious power called wakanda, writes of their idea of the continuity of life, by which ‘‘ a relation was maintained between the seen and the unseen, the dead and the living, and also between the fragment of anything and its entirety.’”’ Here continuity means what we call participation, since this continuity obtains between the living and the dead; between a man’s nail-parings, saliva, or hair and the man himself; between a certain bear or buffalo and the mystic ensemble of the bear or buffalo species. Moreover, like the mulungu just spoken of, wakanda or wakan may signify not only a mystic reality, like that which Miss Fletcher calls “life,” but a characteristic, a quality belonging to persons and things. Thus there are wakan men, who have gone through many previous existences. ‘““They arise to conscious existence in the form of winged seeds, such as the thistle, . . . and pass through a series of inspiration, with different classes of divinities, till they are fully wakamzed and prepared for human incarnation. They — are invested with the invisible wakan powers of the gods. . . .’’ 2 Similarly, day and night are wakan. The term is explained thus by an Indian: ‘‘ While the day lasts a man is able to do many wonderful things, kill animals, men, etc. . . . But he does not fully understand why the day is, nor does he know who makes or causes the light. Therefore he believes that it was not made by hand, i.e. that no human being makes the day give light. Therefore the Indians say that the day is wakan...5o 1s, the “sun: iy) eee)’, It 4s a propertysen mystic quality inherent in things that is meant. And the Indian adds: ‘‘ When it is night, there are ghosts and many fearful objects, so they regard the night as wakan. . . 3 A yet earlier investigator, quoted by Dorsey, had already remarked: ‘‘ No one term can express the full meaning of the Dacota’s wakan. It comprehends all mystery, secret power and divinity. . . . All life is wakan. So also is every- thing which exhibits power, whether in action, as the winds and drifting clouds, or in passive endurance, as the boulder by the wayside. . . . It covers the whole field of fear and 1 “ The Signification of the Scalp-lock,” J.A.I., xxvii. p. 437. * Dorsey, “ Siouan Cults,” £. B. Rept., xi. p. 494. 3 Ibid., p. 467. FUNCTIONING OF PRELOGICAL MENTALITY 133 worship; but many things that are neither feared nor wor- shipped, but are simply wonderful, come under this designation.”’ ! We may be inclined to ask, what, then, is not wakan? Such a question would in fact be urged by logical thought which demands the strict definition of its concepts, and a rigorously determined connotation and extent. But pre- logical reasoning does not feel the need of this, especially when dealing with collective representations which are both con- crete and very general. Wakan is something of a mystic nature in which any object whatever may or may not partici- pate, according to circumstances. ‘‘ Man himself may become mysterious by fasting, prayer and vision.’’2 A human being is not necessarily wakan or not wakan, therefore, and one of the duties of the medicine-man in this matter is to avoid errors which might have fatal results. Wakan might be compared with a fluid which courses through all existing things, and is the mystic principle of the life and virtue of all beings. “A young man’s weapons are wakan: they must not be touched by a woman. They contain divine power. . . . A man prays to his weapons on the day of battle.” If the observer recording these facts interprets them at the same time (as usually happens), and if he has not noted the difference between prelogical reasoning and logical thinking, he will be led direct to anthromorphic animism. Here, for instance, is what Charlevoix tells us about the same North American Indians: ‘‘ If one is to believe the savages, there is nothing in nature which has not a corresponding spirit : but there are varying orders of spirits, and all have not the same power. When they fail to understand a thing, they attribute supreme virtue to it, and they then account for it by saying ‘it is a spirit.’’’3 That means that this thing is “‘ wakan”’ ; just as the Yaos say “it is mulungu !” Although Spencer and Gillen uphold the animistic theory, they are too keen observers not to have themselves noticed how very puzzling these collective representations are to our logical thinking. They remarked that certain words are some- t Dorsey, ‘‘ Siouan Cults,” E. B. Rept., xi. pp. 432-3. : 2 [bid., p. 365. 3 Charlevoix, Journal d'un Voyage dans l Amérique Septentrionale, iii. p. 346. 134 HOW NATIVES THINK times used as substantives, and then again as adjectives. For instance, arungquiltha to the Aruntas is “a supernatural evil power.” ‘A thin ostrich or emu is either avungquiltha or is endowed with arungquiltha. The name is applied indiscrimin- ately either to the evil influence or to the object in which it is, for the time being, or permanently, resident.” 1 Else- where, Spencer and Gillen state that avungquiltha is sometimes personal and sometimes impersonal. ‘They believe that eclipses are caused by the periodic visits of the arungquiltha, who would like to take up his abode in the sun, permanently obliterating its light, and that the evil spirit is only dragged out by the medicine-men.’”’? Even the churinga, which these aborigines regard as a sacred, living being and, according to some observations made, as the body of a personal ancestor, is on other occasions considered to be a mystic property inherent in things. ‘‘ Churinga,’ say Spencer and Gillen explicitly, “is used either as a substantive, when it implies a sacred emblem, or as a qualifying term, when it implies | sacred or secret.” 3 In the Torres Straits, also, ‘‘ when anything behaved in a remarkable or mysterious manner it could be regarded as a zogo . . . rain, wind, a concrete object or a shrine can be a zogo; a zogo can be impersonal or personal ; it belonged in a general way to particular groups of natives, but it was a particular property of certain individuals, the zogo le, who alone knew all the ceremonies connected with it, because the rites were confined to them. . . . I do not know how the term can be better translated than by the word ‘sacred.’ The term zogo is usually employed as a noun, even when it might be expected to be an adjective.”’ 4 Hubert and Mauss, in their acute analysis of the idea of the mana of the Melanesians, described by Codrington, and also that of the Huron orenda, have clearly brought out their relation to the idea of wakan.5 What we have just said about the latter applies equally to these and to other similar con- t The Native Tribes of Central Australia, p. 548 (note). 2 Ibid., p. 566. Cf. The Northern Tribes of Central Australia, p. 629. 3 Ibid., p. 139 (note). 4 The Cambridge Anthropological Expedition to Torres Straits, vi. pp. 244-5. 5 Esquisse d’une théorie générale de la magie, Année Sociologique, vii. pp. 108 et. seq. (1904). FUNCTIONING OF PRELOGICAL MENTALITY 135 ceptions of which it would be an easy matter to find examples elsewhere, also interpreted as animistic. Such an idea is that of wong, which we find in West Africa. “The Guinea Coast negro’s generic name for a fetish-spirit is wong; these aerial beings dwell in temple-huts and consume sacrifices, enter into and inspire their priests, cause health and sickness among men, and execute the behests of the mighty Heaven-god. But part or all of them are connected with material objects and a native can say ‘ In this river, or tree or amulet, there is a wong.’ ... Thus among the wongs of the land are rivers, lakes and springs ; districts of land, termite-hills, trees, croco- diles, apes, snakes, elephants, and birds.’’? It is from a mis- sionary’s report that Tylor has borrowed this account, and it is by no means difficult to find in it, not only “ the three stages of animistic belief’? which Hetherwick noticed in the Yaos, but also a collective representation entirely similar to wakan, mana, orenda, and many others. Collective representations of such a nature are to be found, more or less clearly indicated, in nearly all the primitive peoples who have been studied at all closely. They dominate, as Hubert and Mauss have well demonstrated, their religious beliefs and magic practices. It is possibly through them that the difference between prelogical mentality and logical thought can be best defined. When face to face with such repre- sentations the latter is always dubious. Are they realities which exist per se, or merely very general predicates? Are we dealing with one single and universal subject, with a kind of world-soul or spirit, or with a multiplicity of souls, spirits, divinities? Or again, do these representations imply, as many missionaries have believed, both a supreme divinity and an infinite number of lesser powers? It is the nature of logical thought to demand a reply to questions such as these. It cannot admit at one and the same time of alternatives which seem to be mutually exclusive. The nature of prelogical mentality, on the contrary, is to ignore the necessity. Essen- tially mystic as it is, it finds no difficulty in imagining, as well as feeling, the identity of the one and the many, the individual and the species, of entities however unlike they be, by means of t Tylor, Primitive Culture (4th edit), i. p. 205. 136 HOW NATIVES THINK participation. In this liesits guiding principle ; this it is which accounts for the kind of abstraction and generalization peculiar to such a mentality, and to this, again, we must mainly refer the characteristic forms of activity we find in primitive peoples. —- test BA RSS + Wi CHAPTER IV THE MENTALITY OF PRIMITIVES IN RELATION TO THE LANGUAGES THEY SPEAK THE essential characteristics of the mentality of a given social group should, it seems to me, be reflected to some extent in the language its members speak. In the long run the mental habits of the group cannot fail to leave some trace upon their modes of expression, since these are also social phenomena, upon which the individual has little, if any, in- fluence. With differing types of mentality, therefore, there should be languages which differ in their construction. We could not venture very far upon the strength of so general a principle, however. In the first place, we do not know whether even in primitive peoples there is a single one who speaks his own language—that is, a language which exactly corresponds, according to the hypothesis suggested above, with the type of mind which his group ideas express. On the contrary it is probable that by reason of migration, inter- mingling and absorption of groups we shall nowhere encounter the conditions which such a hypothesis implies. Even in the period known to history a social group very often adopts the language of another group which has conquered it, or been conquered by it. We can therefore safely establish nothing more than a very general correspondence between the characteristics of the languages and those of the mentality of the social groups, confining ourselves exclusively to such characteristics as are to be found in the language and the mentality of all the groups of a certain kind. In the second place, the languages of primitive peoples are still very little known. Of very many of them we possess no more than very imperfect vocabularies. They may perhaps allow of our placing them, provisionally, in a certain linguistic 140 HOW NATIVES THINK family, but they are wholly insufficient for the purpose of comparative study, and in the opinion of those best qualified to judge, a comparative grammar of the different families of spoken languages would be an impossible achievement. Lastly, the construction of the languages spoken by primitive peoples conveys both that which is peculiar to their own mental habits and that which they have in common with ourselves. We have already found that prelogical does not mean antilogical. We cannot lay down a principle that these languages must have special grammars differing specifically from our own. We are compelled, therefore, to leave these too vast problems untouched, and discover by some more modest method how far an examination of their languages may confirm what I have already said about the mentality of primitives. Leaving grammar, properly so called, out of the question, I shall primarily search for what may be revealed of the mind of such peoples in the construction of their sentences and in their vocabulary, and I shall choose "my examples preferably from the languages of the North American Indians, which have been specially studied by those who collaborate in the Washington Bureau of Ethnography. This does not, however, preclude me from quoting, for the purpose of comparison, from other tongues which may belong to quite different language groups. I Perhaps the most salient characteristic of most of the languages of the North American Indians is the care they take to express concrete details which our languages leave understood or unexpressed. ‘A Ponka Indian in saying that a man killed a rabbit, would have to say: the man, he, one, animate, standing (in the nominative case), purposely killed by shooting an arrow the rabbit, he, the one, animal, sitting (in the objective case) ; for the form of a verb to kill would have to be selected, and the verb changes its form by inflection or incorporated particles to denote person, number, and gender (as animate or inanimate) and gender again as sitting or lying, and case. The form of the verb would also express whether the killing was done accidentally or purposely, and whether PRIMITIVES’ LANGUAGE 141 it was by shooting . . . and if by shooting, whether by bow and arrow, or by gun.”! So too, in the Cherokee tongue, “‘instead of the vague expression we, there are distinct modifications meaning I and thou, I and ye, I and ye two, I and he, I and they, etc., and in the plural I, thou, and he or they, I, ye, and he and they, etc., etc. In the simple conjuga- tion of the present of the indicative, including the pronouns in the nominative and oblique cases, there are not less than seventy distinct forms. ... Other nice distinctions; the various forms of the verb denote whether the object be animate or inanimate, whether or not the person spoken of, either as agent or object, is expected to hear what is said, and in regard to the dual and plural numbers, whether the action terminates upon the several objects collectively, as if it were one object, or upon each object considered separately, etc.” 2 These languages, therefore, like our own, recognize a number category, but they do not express it in the same way. We oppose the singular to the plural: a subject or object is either singular or plural, and this mental habit involves a rapid and familiar use of abstraction, that is, of logical thought and the matter it deals with. Prelogical mentality does not proceed thus, however. ‘To the observing mind of the Klamath Indian,” says Gatschet in his excellent Klamath grammar, “the fact that things were done re- peatedly, at different times, or that the same thing was done severally by distinct persons, appeared much more important than the pure idea of plurality, as we have it in our language.’’3 Klamath has no plural form, but it makes use of a distributive reduplication, and every time that this form indicates plurality, it is merely because this idea of distributive reduplication happens to coincide with the idea of plurality. “ Thus nep means hands as well as hand, the hand, a hand, but its distributive form nénap means each of the two hands or the hands of each person when considered as a separate individual. Kitcho’l means star, the stars, constellation or constellations ; kichdéktchd] means each star or every star t Powell, ‘‘ The Evolution of Language,” E. B. Rept., i. p. 16. 2 Gallatin, Transactions of the American Ethnological Society, ii. pp. 130-1. 3 A. Gatschet, The Klamath Language, p. 419. 142 HOW NATIVES THINK or constellation considered separately. Pddshat means you became blind of one eye; papddsha i, you are totally blind, you lost the use of each of your eyes.”’ ! Does this mean that the Klamath language cannot express a plural? By no means; but it does so in varied forms. “The plural number of the subject of the sentence may be indicated in different ways: (1) analytically by adding to the noun a numeral or an indefinite pronoun (a few, some, all, many, etc.); (2) by the noun being collective, or one of the substantives designating persons, which possess a form for the real plural; (3) the large majority of substantives having no real plural, their plurality is indicated in the intran- sitive verbs by the distributive form, and in a few transitive verbs by a special form which has also a distributive function ; (4) the dual form serves for two, three, or four subjects of certain intransitive verbs.” 2 To judge from these examples, which are by no means exceptional, if prelogical mentality primarily makes no use of the plural form it is because such a form is not sufficiently — explicit, and does not specify the particular modality of the plural. The primitive’s mentality needs to differentiate between two, three, a few, or many subjects or objects, to indicate whether they are together or separate. As we shall see later, it has no general terms for ‘“ tree ”” or “‘ fish’’ but special terms for every variety of tree or of fish. It will therefore have methods of rendering, not the pure and simple plural, but the varying kinds of plural. As a general rule we shall find this peculiarity the more marked when we are considering languages spoken by the social groups in which prelogical mentality is still dominant. In fact, in the Australian dialects, in those of the New Hebrides and Melanesia, and in those of New Guinea, we find used, sometimes as well as the plural form properly so called, sometimes without it, the dual, trial, and even what we might term the quadrial forms. Thus in the Papuan language of the island of Kiwai, ‘“‘ nouns are often used with- out any mark of number; but when the noun is the subject of a verb, it is usual to distinguish number by means of a suffix, The singular is shown by the suffix vo, the dual by A. Gatschet, The Klamath Language, pp. 262-3. 2 Ibid., pp. 578-9. PRIMITIVES’ LANGUAGE 143 the word toribo, the trial by the word potoro. The plural is shown by the word sivio preceding, or by the word sirioro following. The singular suffix vo is very commonly omitted. Potoro is used also for four, and its real meaning is therefore probably ‘a few.” The vo suffixed in potoro and sirioro is probably the same as the singular 70, and suggests that potorois a set of three, a triad, and sirioro a lot, a number.” ! In this same language we find a multiplicity of verbal suffixes, simple and compound, the function of which is to specify how many subjects act upon how many objects at a given moment. Here are examples of suffixes : yudo means the action of two on many in past time. yumo means the action of many on many in the past. durudo, the action of two on many in present time. duyumo, the action of many on many in the present. amadurodo, the action of two on two in present time. amarudo, similar action in past time. amarumo, the action of many on two in past time. ibidurudo, the action of many on three in present time. ibidurumo, similar action in past time. amabidurumo, action of three on two in present time, etc.? To my mind the desire for concrete specification could scarcely be expressed more clearly, as far as number is con- cerned. Therefore we may say that these languages possess a whole system of plurals. “The dual number, and what is called the trial, are in Melanesian languages, with the exception of a very few words, really no distinct number, but the plural with a numeral attached.”3 This remark of Codrington’s applies exactly to the languages of British New Guinea. It amounts to saying that these languages express as fully as possible a plural which is deter- minate in number, and not simply a plural. The same phenomenon occurs frequently in the Australian languages. Thus, ‘‘in all the dialects having the Tyattyalla structure, there are four numbers, the singular, dual, trial, and plural. The trial has also forms in the first person (inclusive 1 The Cambridge Expedition to the Torres Straits, iii. p. 306. 2 Codrington, Melanesian Languages, p. 111, quoted in The Cambridge pose to Torres Straits, iii. p. 428. 3 Ibid. ' 144 HOW NATIVES THINK and exclusive). The trial has also been found by me in the Thaguwurru and Woiwurru tongues. . . . The existence of trial was reported years ago in Aneityum and some other islands in the Pacific Ocean, and was observed to some extent in the pronouns of the Woddouro tribe in Victoria by Mr. Tuckfield.”’ 1 ‘‘ Although the dual is generally used, a trial is often met with in the Bureba language (Murray River).”’ 2 ‘The trial number, as existing in the native languages of Victoria, is different in character from that observed in some other countries. For example, in the New Hebrides, the case endings of the dual, trial and plural are independent and differ from each other in form. ... But among the Victorian tribes, the trial number is formed by adding another case-ending to that of the plural.’’3 In the Motu language of New Guinea, W. G. Lawes, the missionary, reports that the dual and the trial of pronouns is formed by additions to the plural. This is the fact which Codrington had noted. In New Mecklenburg, in the Bismark Archipelago, forms of quadrial (Vierzahl) over and above the trial, had been ~ encountered. These quadrial forms are to be found also in Nggao (Solomon Isles) and at Araga and Tanna, in the New Hebrides. They are the counterpart of the Polynesian ‘plurals,’’ which in reality are trials.4 The diversity of these forms does not prevent our recognizing a common tendency in all. Sometimes we have the dual and trial as independent forms, co-existing with the plural, properly so called, in the New Hebrides ; sometimes there are plurals completed by a suffix which specifies a number, as in the languages of Melanesia and New Guinea, and some Australian tongues. Occasionally the distributive reduplication is prior to the plural proper and supplies its place; or again, the plural seems to be wanting, and there are various ways of providing for it. For example, “ In Déné-dindjié there is no plural, and the idea of it is expressed by the addition of the word ‘many’ to the singular... . The Peaux de liévre and the Loucheux make use of the dual 1 Mathews, ‘“‘ The Aboriginal Languages of Victoria,’’ Journal of the nine aeaied ed South Wales, pp. 72-3 (1903). 3 Id., ‘ Languages of the Kamitaroi,” /.4.1., xxxiii. pp. 282-3. 4 P. W. Schmidt, Anthropos, ii. p. 905 (1907). PRIMITIVES’ LANGUAGE 145 element and this form indiscriminately.” 1 Occasionally, too, there is variety in the formation of the plural. Thus, in the Abipone language, ‘the formation of the plural number of nouns is very difficult to beginners, for it is so various that hardly any rule can be set down. . . . Moreover, the Abipones have {wo plurals: more than one, and many. Joalez, some men; Joalivipi, many men.”? This differentiation is a familiar one in Semitic languages also. We see in them the various methods (the list of which we have not exhausted) by which languages express the various modalities of numera- tion. \Instead of indicating plurality in general, they specify what sort of plurality is intended: of two things together, or of three. Beyond three, a good many languages say “many.” This is doubtless why we find no special forms for the plural, beyond the dual and the trial, in the tongues of the most primitive peoples we know of. Little by little, as the mental functioning is modified and representations necessarily become less concrete, there is a tendency to reduce plural forms to the simple plural. We lose the trial form first of all, and then the dual. Junod notes an isolated survival of the dual form in the Ronga language.3 The history of Greek demonstrates a continuous decay of the dual form which is significant.4 II It is not in the number category alone that the need for concrete expression manifests itself in the languages of primitives. There is at least the same wealth of forms | which endeavour to render the varied modality of action | denoted by a verb. For instance, in the language of the Ngeumba tribe of the Darling River, New South Wales, in the past and future tenses, terminations vary to indicate whether the action described occurred in the immediate, recent, or remote past, or will take place at once or in the near or remote future; whether there has been, or will be, repetition or continuity of action; and yet other modifications of verbal suffixes. These terminations are the same for all the persons in 1 Petitot, Dictionnaire de la Langue Dènè-dindjié, p. li. 2 Dobrizhoffer, An Account of the Abipones, il. p. 163. 3 Junod, Grammaire Ronga, P. 135. 4 Cuny, Le Duel en Grec, pp. 506-8. IO 146 HOW NATIVES THINK the singular, dual, and plural. Therefore there are different forms to express— I shall beat (indefinite). I shall beat in the morning. I shall beat all day. I shall beat in the evening. I shall beat in the night. I shall beat again, etc. In the Kafir tongue, with the help of auxiliary verbs, we can obtain six or seven imperative forms, all with differing meanings: 1. Ma unyuke e ntabeni—Stand up to go up the hill. 2. Ka unyuke e ntabeni—Make one move to go up the hill. 3. Suka u nyuke ntabeni—Wake up to go up the hill. 4. Hamb’o kunyuka—Walk up to go up. 5. Uz’ unyuke e niabeni—Come to go up the hill, etc., etc. Though all the above expressions may be rendered by ‘go up the hill,” yet properly form (1) supposes a change of occupation, (2) may be used only of a momentary action ; (3) will best be said to one who is too slow to perform an order, - (4) to one who has to go some way before beginning to go up the hill, and (5) conveys an order or prayer which allows delay in the execution, etc., etc.? The extraordinary prolixity of the verbal forms in the languages of the North American Indians is well known, and in those known as Indo-European it seems to have been no less. In the Abipone language it creates, as Dobrizhoffer says, ‘‘a labyrinth most formidable.’’3 In Northern Asia, “the Aléutian verb, according to Venianimof, can take more than four hundred endings, to indicate mood, tense, and person, without reckoning the tenses which may be formed with the help of auxiliaries. It is clear that originally each of these many forms corresponded with a definite shade of meaning, and that the Aléutian of former days, like the Turkish language of our times, was marvellously versatile in responding to the very minutest verbal modality.” 4 t Mathews, “‘ Aboriginal Tribes of New South Wales and Victoria,’”’ Journal of the Royal Society of New South Wales, pp. 220-4 (1905). Cf. ibid., pp. 142, 151, 166 (1903). 2 Torrend, Comparative Grammar of the South African Bantu Languages, 23% 3 Dobrizhoffer, op. cit., ii. pp. 172-80. « V. Henry, Esquisse d'une Grammaire Raisonnée de la Langue Aléoute,” PP. 34-5- PRIMITIVES’ LANGUAGE 147 If the need for concrete expression and the accumulation of forms capable of expressing any peculiarities of action, or subject and object, are indeed features common to very many of the languages spoken by primitive peoples ; if these features tend to grow weaker or to disappear as communities advance in development, it is permissible to inquire what it is with which they correspond in that which we have called the mentality peculiar to these peoples. It is a mentality which makes little use of abstraction, and even that in a different method from a mind under the sway of logical thought ; it has not the same concepts at command. Will it be possible to go yet a little further and find, in examining the matter at its disposal—that is, the vocabulary of its languages—any positive indications of the manner of its functioning ? The Klamath language, which may be taken to represent a very numerous family of languages in North America, obeys a well-marked tendency which Gatschet calls “ pictorial,” a tendency to delineate that which one desires to express. ‘A motion performed in a straight line is referred to difi- erently from a motion performed sidewise or obliquely or at a distance from the one speaking, circumstances which it would seldon occur to us to express in European languages.’’t It is above all in its primitive form that the Klamath language displays this characteristic. At that time “it seems to have left unnoticed the expression of number in verbs, as well as in nouns and found no more necessity to define it than to define sex. Only a little more attention was paid to the categories of mode and tense, for what was done in all these belongs to later periods of linguistic development. Concrete categories alone were then accounted of importance ; for all relations bearing upon locality, distance, and indivi- duality or severalty are distinguished with superior accuracy, and even tense is marked by means of particles which were originally locomotive.” 2 In short, it is especially spatial relations, all that can be retained and reproduced by visual and muscular memory, that the Klamath language aims at expressing, and this the more exclusively in the most remote period of its history. 1 A, Gatschet, The Klamath Language, p. 460. 2 Ibid., pp. 433-4. 148 HOW NATIVES THINK Like nearly all the languages of primitive peoples, it has no verb ‘to be.” ‘“ The verb gz which takes the place of it is, in fact, the demonstrative pronoun ge, ke (this one, this here), in a verbified shape, and having assumed the verbal form, it came to signify to be here, to be at this or that place, to be at this time or at such a time.’’! In a general way, that which relates to time is expressed by words which were first of all applied to spatial relations. “In Klamath, as in many other languages, there are only two tense-forms, one for the completed and the other for the incompleted act or state ...and... both forms, whether appearing in the verb or in some substantives, originally had a locative character, now pointing to distance in time only.” 2 The spatial element predominates in the same way where case is concerned. Setting aside the “‘three purely gram- matical cases (subjective, direct objective, and possessive), all the other cases, as instrumental, inessive, adessive, etc., are either locative, or take their origin in some locative relation of the noun to the verb.’’3 Even the possessive was origin- ally locative, and the partitive also, which “is but another form of the prefix ¢a, and originally both referred to objects standing erect, as men, animals, trees, etc., the suffixed 7 pointing to location on, upon something.’’4 It is the same with the inessive. ‘‘ As the first of the five post-position cases, I have placed the one formed of the pronominal element i, hi. . . . It occurs in nominal inflection as a case-terminal by itself, and also enters the composition of several others, as ti, xént, émi, kshi, ksaki . . . From its primary signification upon the ground, have developed those of within, at home, in the lodge, for one’s or another’s benefit or disadvantage, and the temporal one when, at the time when.” 5 Finally as to the directive case: ‘‘ this case post-position is a combination of the two pronominal elements #4 and la, which we find to be the components of a large number of affixes. It is most generally connected with verbs of motion, and corresponds with our to, toward. . . . It is connected with the names of the cardinal points of the horizon, and . . . the original 1 A. Gatschet, The Klamath Language, pp. 430-1. 2 Ibid., p. 402. 3 Ibid., p. 467. 4 Ibid., p. 476; 5 Ibid., p. 485. PRIMITIVES’ LANGUAGE 149 use made of this particle seems to have been that of pointing to objects visible at long distances.” * We must refer to Gatschet’s work for ‘‘ a long series of locative case- endings.”’ 2 If we pass on to the demonstrative pronouns we shall find that there, too, we have a great many spatial pecularities most minutely expressed. Klamath is not content with distinguishing #his from that ; it distinguishes, both in animate and inanimate kind : this (so near as to be touched) this (close by, right here) this (standing, being before you) this (present, visible, within sight) that visible (though distant) that absent that absent (departed) that (beyond sight) All these forms are in use both for the subjective and the objective case,3 and this is not, as we know, a peculiarity of the Klamath tongue. In most primitive languages, personal and demonstrative pronouns exhibit a large number of different forms, in order to express the relations of distance, relative position, visibility, presence or absence, between subject and object, etc. To quote but one or two examples taken from the languages of wholly undeveloped peoples, in the Wongai- bon tongue demonstrative pronouns are both numerous and varied, and represent divers grades of meaning which depend upon the position of the object with regard to the speaker as well as with regard to cardinal points. It is the same, too, with the Dyirrigan and Yota-yota languages.4 In the case of the Yahgans, pronouns are very numerous... . they have, the three numbers, and are declined like nouns. The Yahgans, when making use of pronouns, always indicate the position of the person spoken of. ... For instance, they speak of him or her in relation to an object at the upper end of the : A. Gatsch et, The Klamath Language, p. 489. 2 Ibid., pp. 479 et seq. 3 Ibid., pp. 538 et. seq. _ 4 Mathews, “ Languages of . . . Queensland, New South Wales and Victoria,” Journal of the Royal Society of New South Wales, pp. 151, 163, 170 (1903). 150 HOW NATIVES THINK wigwam, or facing the door, or to a person at the bottom of a creek or a valley—at the left or right of the wigwam or in its interior—in the wigwam, near the threshold—outside the house. All these pronouns are of three classes, according to whether they refer to the position of the person who is speaking, the person spoken to, or the one spoken of. . . . It is the same with the demonstrative pronouns.! The post-positions in Klamath are exceptionally numerous, and nearly all express spatial relations. ‘‘ Those of our prepositions which are of an abstract nature, as about, on behalf of, for, concerning, etc., are expressed by inflectional suffixes, appended to the verb or noun, and all the post- positions we meet are of a concrete, locative signification. Even the few temporal post-positions are locative at the same time.’’2 In Gatschet’s book the reader will find a list of the “‘ principal ”” of these, forty-three in number. “Temporal adverbs have all evolved from locative adverbs, and hence often retain both significations. ... Adverbs of space are very numerous and multiform, almost all the pronominal radices having contributed to the list.’ 3 Gatschet enumerates fifty-four of these which, he says, are the most frequently met with. There are special ways of expressing ‘close by,’ “in front,” ‘‘here at the side,” etc. Without unduly prolonging the list of proofs, which it would be quite easy to multiply, we may therefore regard as established the conclusion formulated by Gatschet in the following terms. ‘‘ The concrete categories of position, loca- tion and distance are of such paramount importance to the conception of rude nations as are to us those of time and causality.’’4 Every sentence in which concrete beings or objects are in question (and in such languages there are scarcely any others) must accordingly express their spatial relation. This essential point may be compared with the necessity of giving every noun in our language a gender. “The student” (of language) says Powell, ‘‘ must entirely free his mind from the idea that gender is simply a distinction of sex. In Indian tongues ” (possibly, too, in Bantu and 1 T. Bridges, ‘ A few Notes on the Structure of the Yahgan,” J.4.1., XXlii. pp. 53-80. 2 A. Gatschet, The Klamath Language, pp. 554 et seq. 3 Ibid., pp. 562 et seq., 583. 4 Ibid., p. 306. | PRIMITIVES’ LANGUAGE 151 in Indo-European languages), “‘ genders are usually methods of classification primarily into animate and inanimate. The animate may again be divided into male and female, but this is rarely the case. Often by these genders all objects are classified by characteristics found in their attitude or supposed constitution. Thus we may have the animate and inanimate, one or both, divided into the standing, sitting, lying ... or... into the watery, mushy, earthy, stony, woody, fleshy.”’ * In Klamath, for instance, ‘‘ whenever an animate or inani- mate subject or object is referred to as being somewhere, either indoor or outdoor, around, below, between, or above somebody or something, in the water or on the ground, the verb gz, to be, is not employed, but the adverbial idea becomes verbified in the form of some intransitive verb, so that below, e.g. becomes 7-utila, to be or lie below, underneath. The mode of existence has also to be distinctly qualified in that verbi- fied term; it has to be stated whether the object or subject was standing, sitting or lying, staying, living, sleeping. Usually the idea of staying and living coincides with that of sitting, and sleeping with that of lying on a certain spot.”’ 2 In other languages, modifications of the pronouns satisfy this demand. With the Abipones, for example, if the object of discourse is— MASCULINE. FEMININE; present, it is designated by eneha anaha if it be seated it is designated by hiniha haniha if it be lying it is designated by hiriha hariha if it be standing it is designated by haraha haraha if walking and visible it is designated by ehaha ahaha if walking and invisible it is designated by ekaha akaha If the object alone is— seated itis designated by ynitara lying it is designated by iritara walking it is designated by ekatara absent it is designated by okatara standing it is designated by eratara 3 t Powell, ‘ The Evolution of Language,” E. B. Rept., i. pp. 9-10. 2 A. Gatschet, The Klamath Language, pp. 674-5. 3 Dobrizhoffer, An Account of the Abipones, ii. p. 166. 152 HOW NATIVES THINK Since facts like these are ascertainable in nearly all the languages of primitive peoples actually known to us, we may regard the necessity to which they bear witness as an essential element of primitive mentality. Ill But primitive mentality does not demand alone that the relative positions of things and persons in space, as well as their distance from each other, be expressed. It is not satisfied unless the language expressly specifies besides, the details regarding the form of objects, their dimensions ; way of moving about in the various circumstances in which they may be placed; and to accomplish this, the most divers forms are employed. The Klamath language, which will serve as a type once more, mainly has recourse to affixes, of which it possesses a surprising number. A few examples only will suffice to show the extent to which this meticulousness is prosecuted. I shall consider prefixes first. (1) Prefixes indicating form and dimensions. a, a verb and noun prefix, denoting long, high objects, such as poles, sticks, and also human beings when their stature is being considered. It differs from ég, tk, which are no longer met with save as part of a root syllable denoting the immobility of a subject placed upright, in that it denotes long objects which are not necessarily in an upright position. For instance: aggédsha—to describe a circle (the hand of a watch). ahkätchga—to break (poles and sticks). alahia—to show (a tree). The prefix a is seen also in the initial syllable az or ei, when referring to a movement made with the head, as aka, to move the head forward. (2) Prefixes denoting a special method of dealing with definite objects. dy, y, prefixes of transitive verbs and their derivatives, indicating an action performed with or upon a number of elongated persons or objects, or upon objects considered collectively, when not standing in an invariable upright PRIMITIVES’ LANGUAGE 153 position. If there be but one object, the prefixes are 4, e, ksh, u. . . . For example: idsah—to cause to move, or carry away (a single object, éna). itpa—to fetch away, to remove (a single object, dtpa). (3) Prefixes denoting movement in a certain direction, hi, ke, ge, k, g, prefixes of transitive and intransitive verbs and their derivatives, to indicate an action accomplished obliquely, from the side, or a lateral movement towards an object. kidpka—to lie down across (ipka, to be lying). kimdédsh—an ant (anything which walks or moves sideways). Km is a prefix formed from the combination of the prefix 4 (shortened form of ki) and ma (abbreviated to m), the latter indicating a curvilinear movement or object, km accordingly denotes a lateral or curvilinear movement, or the turning movement of an object, like a cord, thread, or wrinkle. kmukdligi—to wrinkle (as the effect of moisture). (4) Prefixes denoting form and movement. 1, prefixed to verbs and nouns which describe or indicate the outside of a round or spheroidal, cylindrical, discoid or bulbed object, or a ring; also voluminous; or again, an act accomplished with an object which bears such a form ; or a circular or semi-circular or waving movement of the body, arms, hands, or other parts. Therefore this prefix is to be found connected with clouds, celestial bodies, rounded slopes on the earth’s surface, fruits rounded or bulbed in shape, stones and dwellings (these last being usually circular in form). It is employed, too, for a crowd of animals, for enclosures, social gatherings (since an assembly usually FOOD the form of a circle), and so forth. (5) Prefix denoting a movement in a definite medium. ich, ts, a prefix which appears in terms exclusively denoting the movements observed in water and other liquids, the floating of objects on or in water, the flow or movement of liquids themselves. tchéwa—to float (from éwa, used of water-birds). ichla’lya—to sink (from é/ya). Hen HOW NATIVES THINK (6) A compound prefix indicating a certain movement or form. shl, a prefix compounded of the prefixes sk and J, and indicating, in nouns, as in verbs (almost invariably transitive) objects of a slender, flexible shape, of the nature of leaves, such as linen, cloaks, hats, other articles of clothing or things in which one may be wrapped, and also other objects which may have folded surfaces; even baskets, because they are flexible. Shlaniya—to stretch (a skin, for instance). Shla-ish—a mat. Shläpa—to open, be in flower. Shlapsh—bud. In summing up, Gatschet gives a recapitulation of the Klamath language-prefixes, which the question of space forbids me to quote. I shall indicate the most important at least, in order that a glance at the various functions of the prefixes may enable us to see the predominance of the function which serves to specify spatial relations, forms, and methods of moving and acting. A. Prefixes relating to verbs, auxiliary, reflective, causative, transitive, and intransitive, etc. B. Prefixes relating to number : singular, plural. C. Prefixes relating to the form and contour of subject or object: (1) forms which are round or spherical or large; (2) flat, smooth, flexible, like threads ; (3) forms like leaves, and like coverings for the body ; (4) long, elongated, and tall forms. D. Prefixes relating to the attitude and position ; as upright, straight, rigid. E. Prefixes relating to movement, (x) in the air; (2) below; (3) outside, in or on water and other liquids; (4) performed by an oblique movement; (5) in a zigzag on the ground ; (6) in the form of a wave; (7) with the head; (8) with the hands or arms ; (9) with the back or feet. F. Prefixes referring to relations expressed by adverbs ; i.e. locative prefixes.! The number of suffixes, and the variety of their functions, is far greater even than those which our study of prefixes t A. Gatschet, The Klamath Language, PP. 302-3. > PRIMITIVES’ LANGUAGE rs has made manifest. I shall not enter into detail about the relations they express. I shall merely note that they serve to reproduce, among others, the following ideas: to begin, continue, cease, return to, to be accustomed to do, either frequently or at the beginning, to pass to, to move to a longer or shorter distance, to move in a zigzag or in a straight line, to go up, along the ground, or below, to describe circles in the air, to come towards or go away from (the subject or object being either visible or invisible), to change one’s place in the hut or outside it, on the water or below its surface, and finally an infinity of other details, many of which would be neither observed nor expressed by us, but which strike the Indian mind more forcibly than they would our own.? Gatschet notes that prefixes refer rather to the category of form, while suffixes preferably relate to those of the way of acting, movement, and repose. But it is not always easy to maintain this difference, as we shall see from the following list of suffixes, a list which is much abridged, only the headings of which I reproduce here. À. Suffixes describing movement: (1) in a straight line, or for a short distance; (2) towards the ground; (3) towards some other object, or towards the subject of the verb; (4) far away, to separate; (5) above or below something ; (6) on a horizontal plane; (7) circular (whether inside or outside the house) ; (8) around an object; (9) turning or winding ; (10) vibratory or oscillating; (11) down; (12) in the water. B. Suffixes to denote staying, or remaining at rest: (x) in the interior of the hut or some other enclosed space; (2) out- side, beyond certain limits; (3) on, above, or on the surface of; (4) around, encompassing something ; (5) below, beneath ; (6) between; (7) at a distance from; (8) in the woods or marshes or on the cliffs; (9) in the water; (10) around and near the water. C. Suffixes descriptive of acts accomplished by living beings or by parts of their bodies: (1) frequentative; (2) iterative ; (3) habitual; (4) in movement; (5) outside; (6) above, on the surface of; (7) below, beneath; (8) with a weapon or instrument ; (9) with or on the body; (10) with the mouth; t A, Gatschet, The Klamath Language, p. 305. 2 Ibid., p. 396 156 HOW NATIVES THINK (11) with the back; (12) near or in the fire; (13) in taking away; (14) in making a gesture; (15) in somebody’s interest ; (16) in calling by name; (17) with verbs expressing desire ; (18) as regarding the degree of accomplishment attained (inceptive, continuative, executed in part only, completely, lastingly).1 This method of specifying the details of the action expressed or the object denoted may be pursued almost indefinitely by the help of affixes. To take an example from Klamath, the verb gdlepka means to raise oneself, to mount to. By adding an k it indicates mounting upon something by using one’s hands. Then ge’hldptcha signifies doing this en route, while walking or travelling, and finally ge’ hldpichapka expresses the fact that one does this at a distance from, and unseen by, others. In the passage quoted, the last of these expresses the act of a prisoner who escapes on horseback during the night. “To carry a child’’ may be expressed in a variety of ways, the main differences being whether the infant is carried on its plank-cradle or without it; on the arm or on the back; whether borne to the hut, or outside it, etc.3 Details which would be absolutely insignificant to us become the ground of fine distinctions between verbs which we should call synonymous, but which are not so to the Indian. Gatschet tells us that occasionally their reason for expressing the same act or the same condition by different verbs is not due to a difference in the act or condition, but to divergences in the subjects and objects of the verb as to shape, quality, and number. . . . They have eight terms to express seizing, twelve for separating, fourteen for washing. Many other instances illustrative of the niceties of perception and the wealth of descriptive terms in the language might be given. This quality, however, as we know, is not the sole pre- rogative of the Klamath Indians. Such a characteristic is found, no less marked, among their neighbours, and it is common to most of the languages spoken by the North American tribes. In the Hurons’ tongue, ‘‘in describing a journey, the expressions used differ according to whether it was accomplished by land orsea. The active verbs increase 1 A. Gatschet, The Klamath Language, pp. 397-8. 2 Ibid., p. 68. 3 Ibid., pp. 698-9. PRIMITIVES’ LANGUAGE 157 in proportion with the things which may be done ; for instance, the verb expressing eating varies with the number of comestibles in the case. Action is described differently for animate or inanimate objects; to see a man, and to see a stone, requires two verbs. To use something which belongs to the user must be expressed differently from the verb which indicates the use of some other person’s property.t With the Nez-percés, verbs assume different forms according to whether the subject or object is advancing or retreating.? In the language of the Yahgans, there are ten thousand verbs, the number of which is considerably increased by the use of prefixes and suffixes, to indicate whence one comes or whither one goes, to north, south, east, or west, above, below, outside, inside, and we are told that these differences are almost inexhaustible, even without reckoning the locative adverbs.3 The Abipones, we are told, have an incredible number of synonyms, for they have different words to indicate wound- ing by the teeth of man or animal, by a knife, a sword, or an arrow; to express fighting with a spear, arrows, the fists, or indulging in wordy warfare ; to indicate that the two wives of a man are fighting about him, etc. . . . Different particles are affixed to indicate exactly the place and varying positions of the subject of discourse; above, below, around, in the water, in the air, on the surface, etc. There are many diverse forms of the verb “to follow,’’ for example, and a person coming, going, following with his hand something below or above him, following with his eyes, or with his mind, or following other people, may all be expressed.4 In South Africa, Livingstone found that verbs possessed the same power of expressing delicate shades of meaning. “It is not the want, but the superabundance of names that misleads travellers, and the terms used are so multifarious that good scholars will at times scarcely know more than the subject of conversation. . . . We have heard about a score of words to indicate different varieties of gait—one walks leaning t Charlevoix, Journal d'un Voyage dans l’Amérique Septentrionale, iii. pp. 196-7. 2 Bancroft, The Native Races of the Pacific States of North America, iii. p. 622. | 3 T. Bridges, ‘‘ Notes on the Structure of the Yahgan,” /.4.1., xxiii. pp. 53-80. 4 Dobrizhoffer, op. cit., ii. pp. 186-090. 158 HOW NATIVES THINK | forward, or backward; swaying from side to side; loungingly or smartly ; swaggeringly; swinging the arms, or only one arm ; head down or up, or otherwise : each of these modes of walking was expressed by a particular verb. . . .”’1 IV From these and many similar facts which might be quoted, we see that the languages of primitive peoples “always express their ideas of things and actions in the precise fashion in which these are presented to the eye or ear.”’ 2 They have a common tendency to describe, not the impression which the subject receives, but the shape and contour, position, move- ment, way of acting, of objects in space—in a word, all that can be perceived and delineated. They try to unite the graphic and the plastic elements of that which they desire to express. We may perhaps understand this need of theirs if we note that the same peoples, as a rule, speak another language as well, a language whose characteristics necessarily react upon the minds of those who use it, influencing their way of thought and, as a consequence, their speech. These peoples, in fact, make use of sign-language, at least in certain circumstances, and where it has fallen into disuse, there are still traces which show that it assuredly has existed. Very frequently, moreover, it is used without the explorers becoming aware of it, either because the natives do not employ it in their presence, or because the fact has escaped attention. One of them, according to Roth, took these gestures for masonic signs ! 3 Nevertheless, in cases where the most undeveloped peoples are concerned, we have some explicit testimony. Spencer and Gillen have observed this in Australia. ‘‘ Amongst the Warramunga, widows are not allowed to speak sometimes for as long a period as twelve months, during the whole of which time they communicate only by means of gesture language, in which latter they are so proficient that they t Livingstone, The Zambesi and its Tributaries, p. 537. _ 4 Schoolcraft, Information, ii. p. 341 3 W. E. Roth, Ethnological Studies among the N.W. Central Queensland Aborigines, No. 72. = order à PRIMITIVES’ LANGUAGE 159 prefer, even where there is no obligation upon them to do so, to use it in preference to speaking. Not seldom, when a party of women are in camp, there will be almost perfect silence and yet a brisk conversation is all the while being conducted on their fingers, or rather with their hands and arms, as many of the signs are made by putting the hands, or perhaps the elbows, in varying positions.”: ‘In the case : of the widows, mothers, and mothers-in-law ”’ (of Northern tribes) ‘this ban’’ (of silence) ‘extends over the whole period of mourning, and even at the expiration of this the women will sometimes voluntarily remain silent. . . . There is a very old woman in the camp at Tennant Creek who has not spoken for more than twenty-five years.”2 In South Australia, ‘after a death . . . the old women may refuse to speak for two or three months, expressing what they want to say by gestures with the hands—a sort of deaf and dumb language which the men are as adept in as the women.” 3 Like the Cooper’s Creek natives, those of the Port Lincoln district make use of many signs which are very necessary to the chase, not uttering a word the while. By using their hands they can inform their companions what animals they have found, and exactly where these are... . They have, too, signs for all varieties of game.4 Howitt collected a certain number of the signs used by the Cooper’s Creek natives in their gesture-language.s Roth has given us a fairly detailed dictionary of it, and he was able to prove that the | language he had thus formulated was understood and practised throughout the North of Queensland. In the Dieyerie tribe, it was found that an extensive sign-language existed side by side with the oral one, and that all animals, natives, both men and women, the sky, the ground, walking, riding, jumping, flying, swimming, eating and drinking, and a vast number of other things or acts all possessed their 1 The Native Tribes of Central Australia, Pp. 500—1. * The Northern Tribes of Central Australia, PP. 525, 527. 3 “‘ On the Habits of the Aborigines in the District of Powell Creek,’’ etc., J.AJ., xxiv. p. 178. ¢ Wilhelmi, Manners and Customs of the Natives of the Port Lincoln District, quoted by Brough Smyth, The Aborigines of Victoria, i. p. 186. 5 Ibid., ii. p. 308. : 6 Ethnological Studies among the N.W. Central Queensland Aborigines, chap. iv. 160 HOW NATIVES THINK own particular signs which enabled a conversation to be carried on without a single spoken word." In the Torres Straits, gesture-language was noted both in the eastern and western islands, and Haddon regrets that he did not collect its signs.2 It has been met with in (German) New Guinea also.3 To give but one instance in Africa, “ the Masai have a sign-language which is well developed, as Fischer reports.” 4 Dobrizhoffer noticed an Abipone medicine-man who could communicate with others secretly, so that nobody should hear a sound, and this he did by means of gestures in which hands, arms, and head all played their part. His colleagues replied, and thus they were able to keep in touch with each other.s Language of this kind appears to be very general throughout the whole of South America. The Indians of the various tribes do not understand each other’s speech, but they can communicate with each other by signs.6 Finally, it seems to be clear that in North America sign- language has been used everywhere: we have only to recall Mallery’s monograph on the subject which appeared in the first volume of the Reports of the Bureau of Ethnography. It is a real language, possessing its own vocabulary, forms, and syntax. ‘‘ We might,” says one explorer, “ formulate a complete grammar of this language by gesture. ¢ . . We may judge of its prolixity from the fact that Indians of different tribes, who do not know a word of each other’s oral language, can gossip together for half a day, and tell each other all manner of things merely by the movements of their fingers or heads and feet.” 7 Boas relates that a language of this kind was still fairly prevalent in the interior of British Columbia in 1890.8 1 S, Gason, The Dieyerie Tribe, in Woods, The Native Tribes of South Australia, p. 290. 2 The Cambridge Expedition to Torres Straits, iii. pp. 255-62. Cf. J.A-L., xix. p. 380. 3 Hagen, Unter den Papuas, pp. 211-12. 4 Dr. G. A. Fischer, quoted by Widenmann, “ Die Kilimandjaro-Bevôl- kerung,’’ Petermann’s Mitteilungen, No. 129 (1889). 5 Dobrizhoffer, op. cit., 11. p. 327. 6 Spix and Martius, Travels in Brazil, ii. p. 252. 7 Kohl, Kitcht Gami: Wanderings round Lake Superior, pp. 140-1. 8 F, Boas, ‘ The North-western Tribes of Canada,” Report of British Association, pp. 291 et seq. (1890). 3 CO NE Pie a eae PRIMITIVES’ LANGUAGE 161 In most primitive societies, therefore, two languages, the one oral, the other by means of gesture, exist side by side. Are we to assume that they do so without exerting any mutual influence, or must we believe, on the other hand, that the same mentality expresses itself by both, and conversely is modelled upon them? The latter view appears the more acceptable, and it is indeed confirmed by the facts. In a very important work upon “‘ manual concepts,” F. H. Cushing ! lays stress upon the relations between language expressed by manual movements and the spoken language. He demon- strated how the Zuñi order of the cardinal points and the formation of nouns of number originated in definite move- ments of the hands. At the same time he demonstrated in his own case, the resourcefulness of a method which belongs to him, and which his personal genius (the expression is not too strong) as well as the circumstances of his life, enabled him to apply very happily. To understand the mentality of ‘ primitives,” we must endeavour to reconstitute in ourselves conditions which re- semble theirs as closely as possible. On this point we are all agreed. Cushing lived among the Zuñis ; he lived with them, and like them ; he was initiated into their rites and ceremonies ; he became a member of their secret societies, and really was as one among the rest. But he did more than this, and herein lies the originality of his method. With infinite patience he revived the primitive functions of his own hands, living over again with them their experiences of prehistoric days, with the same material and under the same conditions as at that period, when the hands were so at one with the mind that they veally formed a part of it. The progress of civilization was brought about by reciprocal influence of mind upon hand and vice versa. To reconstitute the primitives’ mentality, he had to rediscover the movements of their hands, movements in which their language and their thought were inseparably united. Hence the daring yet significant expression ‘‘ manual concepts.” The primitive who did not speak without his hands did not think without them either. The difficulties which the application of the method suggested and employed by Cushing entail are considerable. He alone, probably, * “ Manual Concepts,” American Anthropologist, v. pp. 291 et seq. II , 162 HOW NATIVES THINK or men endowed with the same unusual tendencies and the same patience as he, would be able to put it into practice profitably, but it certainly led him to valuable results. For instance, Cushing shows how the extreme specializing of verbs, which we have noted everywhere in the languages of primitives, is a natural consequence of the part which the manual movements play in their mental activity. He declares this to have been a grammatical necessity, and says that in the primitive mind thought-expressions, expression- concepts, complex yet mechanically systematic, were effected more quickly than, or as quickly as, the equivalent verbal expression came into being.? | Speaking with the hands is literally thinking with the hands, to a certain extent; therefore the features of these ‘manual concepts” will necessarily be reproduced in the verbal expression of thought. The general processes in expression will be similar: the two languages, the signs of which differ so widely as gestures and articulate sounds, will be affiliated by their structure and their method of inter- preting objects, actions, conditions. If verbal language, therefore, describes and delineates in detail positions, motions distances, forms, and contours, it is because sign-language uses exactly the same means of expression. In this respect there is nothing more instructive than the sign-language of N.W. Queensland, of which Roth has given us a detailed description. In this language, as in the other, the real vital unit is not primarily the isolated sign or gesture any more than the word, but the sentence or the complex ensemble, of varying length, which expresses in inseparable fashion a complete meaning. The significance of a gesture lies in the “context.” Thus the gesture of a boomerang may express not only the object itself, but also, according to the context, the idea of reaching or killing something with it, or of making, or stealing, it. An interrogative gesture awakens the idea of a question, but the nature of the demand depends upon that which has preceded or is to follow.? Moreover, the ‘‘ideograms’”’ which serve to denote persons, things, and actions, are nearly always descriptive t ** Manual Concepts,” American Anthropologist, pp. 310-11. 3 W. E. Roth, op. cit., No. 72. À PRIMITIVES’ LANGUAGE 163 of movement. They reproduce either the attitudes or familiar movements of living beings (quadrupeds, birds, fishes, etc.) or the movements used in capturing them, or in creating or employing some object, etc. For instance, to denote the porcupine, manual movements exactly describe its quaint way of burrowing into the earth and throwing it aside, its quills, its manner of raising its little ears. To express water, the ideogram reproduces the way in which the native laps up the liquid he has taken in his hand. For collar, the two hands are put in the position of encircling the neck, with a gesture of closing them behind, and so forth. Weapons are minutely described to the eyes by the gestures employed in making use of them. In short, the man who speaks this language has at his disposal a great number of fully-formed visual motor associations and the idea of persons or things, when it presents itself to his mind, immediately sets these associa- tions going. We may say that he imagines them at the moment he describes them. His verbal language, therefore, can but be descriptive also. Hence the importance given to contour, form, situation, position, method of movement, visual characteristics of persons and things in general; hence the classification of objects according to whether they are standing, lying, seated, etc. Mallery tells us that the words of an Indian language which are synthetic and undifferentiated parts of speech are strictly analogous in this respect with the gestures which are the elements of sign-language, and that the study of the latter is valuable for the purpose of comparison with the words. The one language explains the other, and neither can be studied to advantage if the other be unknown.! Mallery’s study of the sign-language of the North American Indians was very searching, and he endeavoured to formulate a syntax of it. Of this we have but to retain that which throws light upon the mental habits of those using it, and at the same time illustrates their verbal language. The latter is necessarily descriptive. It may even happen that it is accompanied by gestures which are not only a spontaneous expression of emotion, but an indispensable element of the language itself. With the Halkomelem of British Columbia, t Mallery, “‘ Sign-language,” E. B. Rept., i. p. 351. 164 HOW NATIVES THINK for instance, ‘it may boldly be affirmed that at least a third part of the meaning of their words and sentences is expressed _ in those aids to primitive language, gestures and tonal diff- erences... ' The Coroados of Brazil complete and perfect the meaning of their sentences by their accent, the speed or slowness of the pronunciation, and certain signs made with hand or mouth, or other gestures. If the Indian wishes to indicate that he is going to the wood, he says ‘wood go,” and a movement of his mouth shows the direction that he intends to take.? Even among the Bantu peoples who, as a rule, belong to a type of community which is fairly advanced, the oral language, itself very descriptive, is constantly accompanied by movements of the hand joined with the demonstrative pronouns. It is true that such movements are no longer actual signs, like those which constitute a language by gesture ; but they are aids to the exact description which is given by means of words. For instance, a native will scarcely ever be heard to use a vague expression such as “ he has lost an eye’; but since he has noticed which eye it was, he will say, pointing to one of his own, “‘ this is the eye he has lost.” In the same way, he will not say that two places lie at a distance of three hours’ journey, but rather, ‘If you start when the sun is there, you will arrive when it is there,” at the same time indicating different parts of the sky. So, too, first, second, and third are not indicated by words, but by the pronoun ¢his, with the first, second or third finger extended.3 It is not even essential that these ‘‘ aids’’ to description should be gestures and movements exclusively. The need for description may seek its fulfilment by means of Laut- bilder, as the German explorers call them, i.e. delineations or reproductions of that which they wish to express, obtained by means of the voice. Westermann tells us that the language of the Ewe tribes is richly endowed with the means of in- terpreting an impression received by direct sounds. Such prolixity proceeds from the almost irresistible tendency to 1 Hill Tout, “ Ethnographical Reports ... Halkomelem British Columbia,’ J.A.I., xxxiv. p. 367. * Spix and Martius, Travels in Brazil, ii. pp. 254- 3 Torrend, Comparative Grammar of the South African Bantu Languages, p. 218. PRIMITIVES’ LANGUAGE 165 imitate all one hears or sees, and in general, all one perceives, and to describe it by means of a sound or sounds, chiefly, of movements. But there are also imitations or vocal repro- ductions of these Lautbilder for sounds, odours, tastes, tactile impressions. There are some used in connection with the expression of colour, fulness, degree, grief, well-being, and so on. It does not admit of doubt that many words, properly so called (nouns, verbs, and adjectives), have their origin in these Lautbilder. Properly speaking, they are not onoma- topceic words; they are descriptive vocal gestures rather. An example will best explain them. “In the Ewe language,” says Westermann, ‘as in many related languages, we find a very special kind of adverb... . which as a rule describes a single action or state, or a single property of objects, which therefore are applicable to a single verb only, and are never found in connection with any other. Many verbs, especially those descriptive of a transmission through the organs of sense, have a whole series of such adverbs to give more precision to the action, state or property they express... . These adverbs are actually Lautbilder, vocal imitations of sense-impressions. . . . For instance, the word zo, to walk, may be joined with the following adverbs, which are used only with it, and which describe various kinds of walking, or different gaits : 1 Zo bafo bafo—the gait of a little man whose limbs shake very much while he is walking. Zo béhe behe—to walk with a dragging step, like a feeble person. Zo bia bia—the gait of a long-legged man, who throws his legs forward. Zo boho boho—of a corpulent man, who walks heavily. Zo bula bula—to walk in a dazed fashion, without looking ahead. Zo dzé dze—an energetic and firm step. Zo dabo dabo—a hesitating, feeble step, shaking. Zo goe gde—to walk swinging the head and the buttocks. Zo gowu gowu—to walk with a slight limp, the head bent forward. Zo hloyt hloyi—to walk with many things, or with clothes floating around. Zo ka ka—to walk proudly, upright, without moving the body unnecessarily. Zo kédzo kodzo—the gait of a tall man or animal, with the head slightly bent. Zo kondobre hondobre—like the last, but with a feeble and lifeless step. 1 Cf. Livingstone’s observation, quoted on pp. 157-8. 166 HOW NATIVES THINK Zo kondzva kondzva—to walk with long steps, drawing in the abdomen. Zo kpadi kpadi—to walk with the elbows close to the sides. Zo kpo kpo—to walk quietly and easily. Zo kpudu kpudu—the short hasty step of a little man. Zo kundo kundo—like kondobre kondobre—but not in any unfavourable sense. Zo ltimo l#mo—the quick gait of small animals like rats and mice. Zo moe moe—like gôe gôe. Zo pla pla—to walk with small steps. Zo st si—the light step of small people who sway. Zo taka taka—to walk carelessly and heedlessly. Zo tyatyra tyatyra—a quick but rigid step. Zo tyende tyende—to walk shaking the abdomen. Zo tya tya—to walk quickly. Zo tyadi tyadi—to walk with a limping or dragging step. Zo ty6 tyé—the well-poised and firm step of a very tall person. Zo wido wudo—the quiet step (in a favourable sense), specially women’s. Zo wla wla—a quick, light, unencumbered gait. | Zo wui wui—quick, rapid. Zo wé wé—the walk of a fat man who advanced with a rigid step. Zo wiata wiata—to advance with a firm and energetic step; said specially of people with long legs. These thirty-three adverbs do not exhaust the list of those used to describe the manner of walking. Moreover, most of them may be met with in two forms: an ordinary and a diminutive one, according to the stature of the subject of discourse.t Naturally, there are similar adverbs or Lautbilder for all the other movements, such as running, climbing, swim- ming, riding, driving, for instance.? Finally, these descriptive adverbs are not joined to the verb as if the idea occurred in two points of time: firstly a conception of walking in general, and then the particular method being specified by means of the Lautbild. On the contrary, to the minds concerned, the conception of walking in general never presents itself alone : it is always a certain way of walking that they thus delineate vocally. Westermann even notes that as the delineation by degrees gives place to a real concept, the special adverbs tend to disappear, and other more general ones, such as very, much, to a great extent, etc., etc., are substituted.3 The same descriptive auxiliaries are noted in the Bantu languages. In Loango, for instance, ‘‘each man uses the * Dr. Westermann, Grammatik dev Ewesprache, pp. 83-4. a TDIGS. Ded 30, 3 Ibid., p. 82. PRIMITIVES’ LANGUAGE 167 language in his own way, or . . . to speak more correctly, the language issues from the mouth of each according to the circumstances and his own mood at the time. This use of language is unrestrained and natural as the sounds uttered by the birds, and I cannot think of any more apt comparison.”’ 1 To put it in another way, words are not something fixed and immutable once for all, but the vocal gesture describes, delineates, expresses graphically, in the same way as the gesture of the hands, the action or thing in question. In the Ronga tongue there are “ certain words which the Bantu grammarians regard generally as interjections or onomatopceics. They are usually vocables of one syllable only, by means of which natives express the sudden and direct impression which a sight, sound, or idea, makes upon them, or describe a move- ment, an apparition, a noise. It is quite enough to have listened to some of the perfectly free and unrestrained conver- sations of negroes to note the immense number of expressions of this kind which they have at command. We may be inclined to say perhaps: ‘It is merely a childish way of speaking, not worth the trouble of listening to.’ The truth is quite the contrary, however. The naturally versatile and ready-witted mind of the race is reflected in this picturesque language. Through such words it succeeds in expressing shades of meaning which a more restrained language could not render. Again, these little words have been the origin of numerous verbs, and would deserve recognition on that account. . . . Nevertheless it must be owned that these descriptive adverbs vary very much with individuals. Some among them embellish their speech to an extent which makes it incomprehensible for the uninitiated, and even invent new expressions. Many of these words, however, are actually incorporated into the language understood by all.’ 2 V The plastic and essentially descriptive character of the languages, both verbal and sign-languages, confirms what we have already said with regard to the special form of 1 Dr. Pechuël-Loesche, Die Loango-Expedition, iii. 2, pp. 91-5. 2 Junod, Grammaire Ronga, pp. 196-7. 168 HOW NATIVES THINK abstraction and generalization proper to primitive mentality. The primitive mind is well acquainted with concepts, but these are not at all like ours: it forms them in another way, and uses them in a method which differs from that of logical thought. “It is our aim,” says Gatschet, “ to speak clearly and precisely ; the Indian’s is to speak descriptively ; while we classify, he individualizes.”” The following example shows the difference clearly. The Delaware word madholineen is composed of nad, a derivative of the verb naten, to seek: of hol, from amochol, a boat, and ineen, Which is the verbal termination for the first person plural. It means “ Find the boat for us.” It is the imperative of a verb expressing : I am finding the boat for you, him, etc., which is conjugated like any other verb . . . but is always used in a special sense. It always signifies: find the boat, and expresses a particular act, having no general meaning; it does not mean: “ find any boat.” This is otherwise in classical languages. The Latin aedifico, belligero, mdifico, do not mean build a special edifice, make war on a particular nation, construct a certain specified nest. . . . So too, doypauuarée, diroypadéw, prrodoged, giroSeomorevouat, dilavOpwréw, do not express a preference for a certain book, picture, etc. They express a general love of literature, painting, and so forth. Had they a special meaning at some remote period of their history ? There is nothing to tell us so, and we know nothing about it. But what we do know is, that in the development of primitive American languages, verbs taken in a special sense appeared first of all, and that if one wanted to give them a general application, it was done by inserting an adverbial particle which means “‘ habitually.” : Again, while it cannot be denied that those who speak these languages have a concept of hand, foot, ear, etc. ; their concepts do not resemble ours. They have what I should call an “image-concept,” which is necessarily specialized. The hand or foot they imagine is always the hand or foot of a particular person, delineated at the same time. Powell tells us that in many Indian tongues of North America there is no distinct word for eye, hand, arm, or the other parts * Gallatin, in Transactions of the American Ethnological Society, ii. pp. 136-8, PRIMITIVES’ LANGUAGE 169 or organs of the body; but such are always found incor- porated with or attached to a pronoun which signifies the possessor. If an Indian were to find an arm that had fallen from an operating-table, he would say “‘ I have found his arm ”” (i.e. someone’s), and such linguistic peculiarity, though not universal, is met with frequently.1 It is to be found in many other languages, too. For instance, the Bakairi of Brazil do not say “‘ tongue,” but always add a pronominal adjective, my tongue, your tongue, etc.; and this rule applies to the other parts of the body.2 The same holds good for terms denoting relationship, father, mother, etc., which are very rarely used alone. In the Marshall Archipelago, “there is no generic term for ‘father,’ the word never being used save in conjunction, and applied to a particular person. It is the same for ‘ mother, brother, sister,’ ”” etc.3 The language spoken by the natives of the Gazelle peninsular of the Bismark Archipelago, ‘like most of the Melanesian languages, and some of the Micronesian (of the Gilbert Islands) and of Papua, make use of the possessive pronoun as a suffix when expressing substantives denoting relationship, parts of the body, and some prepositions.” 4 Grierson had noted that in the north-eastern provinces of India, the word “‘ father,” as a general idea, not connected with any special person, and therefore requiring a certain amount of abstractive thought, was never used alone, but always in conjunction with a possessive pronoun. . . . A hand, also, could not be imagined save as belonging to somebody, and even when the possessive form of the sentence rendered the pronoun unnecessary, the tendency to specialization was so strong that it was still added, as “‘ of my mother her hand.” 5 In the Angami tongue, nouns denoting parts of the body, or expressing relationship, had to be preceded by a possessive pronoun.6 The same held good for the Sema language.7 This is a very common feature, and it helps us to understand how it is that in primitive societies we ate those complicated 1 “ The Evolution of Language,’’ E. B. Rept., i. p. a K. von den Steinen, Unter den Naturvôlkern Zentralbrasiliens, D.,92. ‘Die Ebon- -Gruppe im Marshall’s Archipel,” Journal des Muséum our. i. pp. 39-40 4 Parkinson, Dreissig Jahre in der Südsee, p. 730. . 5 Grierson, Linguistic Survey of India, iii. 3, pp. 16-17. A Sbid:. Urge pico: 1:1Did} 12, Pp, 223) 170 HOW NATIVES THINK degrees of rueunehiD which prove so confusing to the European, and which he masters only with difficulty. He tries to conceive of them 77 abstracto, but the native never envisages them thus. In his childhood he learned that certain persons stood in such and such a relation to certain others, and the learning required no more trouble or thought than the rules of his (frequently just as complicated) mother-tongue. The nearer the mentality of a given social group approaches the prelogical, the more do these image-concepts predominate. The language bears witness ‘to this, for there is an almost total absence of generic terms to correspond with general ideas, and at the same time an extraordinary abundance of specific terms, those denoting persons and things of whom or which a clear and precise image occurs to the mind as soon as they are mentioned. Eyre had already remarked upon this with the Australian aborigines. He states that generic terms such as tree, fish, bird, etc., were lacking, although specific terms were applied to every variety of tree, fish or bird.t We are told that the natives of the Tyers Lake district, Gippsland, have no words for these either, but all the species such as bream, perch, mullet, are distinguished in each class? The Tasmanians had no words to represent abstract ideas, and though they could denote every variety of gum-tree or bush, by name, they had no word for tree. They could not express qualities, such as hard, soft, hot, cold, round, tall, short, etc. To signify “‘hard’’ they would say: like a stone; for tall, big legs; round, like a ball; like the moon; and so on, always combining their words with gestures, designed to bring what they were describing before the eyes of the person addressed.3 In the Bismark Archipelago ‘there are no names for colours. Colour is always indicated in the following way. The object in question is always compared with another, the colour of which has been accepted as a kind of standard. For instance, they will say : This looks like, or has the colour of a crow. In the course of time, the substantive alone has been used in adjectival sense. . . . Black is named after 1 Eyre, Journals of Expeditions of Discovery into Central Australia, ii. PP. 392-3. 2 Bulmer, quoted by Brough Smyth, The Grete of Victoria, ii. p. 27. 8 Ibid., 112} Di 413. PRIMITIVES’ LANGUAGE 171 the various things from which this colour is obtained, or else a black object is named. Thus the word “ kotkot” (crow) is used to denote black. Everything that is black, especially things that are glossy black, is called thus. Lzkutan or lukutan also means black, but rather in the sense of dark; toworo is the black colour derived from burnt candle-nuts ; Juluba, the black mud in the mangrove swamps ; dep, the black colour obtained from burning canary-wood gum; wtur, the colour of burnt betel-nut leaves mixed with oil. All these words are used for black, according to the circumstances of the case: there are just as many for other colours, white, green, red, blue, and so forth,’’ } It is the same with the Coroados of Brazil. ‘ Their languages extend only to the denomination of the objects immediately surrounding them, and often express the pre- dominant quality of things by imitative sounds. They distinguish with great precision the internal and external parts of the body, the various animals and plants, and the relation of such natural objects to each other is frequently indicated in a very expressive manner by the words them- selves; thus the Indian names of monkeys and palms were guides ‘to us in examining the genera and species, for almost every species has its particular Indian name. But it would be in vain to seek among them words for the abstract ideas of plant, animal, and the abstract notions colour, tone, sex, species, etc. Such a generalization of ideas is found only in the frequently used infinitive of the verbs to walk, eat, drink, dance, see, hear, etc.”’2 In California, ‘‘ there are no genera, no species: every oak, pine, or grass has its separate name.” 3 | Everything being represented in these ‘‘ image-concepts,” i.e. delineations in which the slightest peculiarity is shown —and this not only for the natural species of all animate nature, but for objects of all kinds, whatever they be, all movements or actions, all states or qualities which language expresses—it follows that these “ primitive ” languages have € t Parkinson, Dreissig Jahre in der Siidsee, pp. 143-5. Cf. The Cambridge Expedition to Torres Strants, ii. pp. 55-68. 2 Spix and Martius, Travels in Brazil, ii. pp. 252-3. Cf. The Cambridge Expedition to Torres Straits, ii. 1, pp. 44, 64. 3 Powers, Tribes of California, p. 419. 172 HOW NATIVES THINK a wealth of vocabulary unknown to ours. This extensive vocabulary has been the source of wonder to many explorers. ‘They ”’ (the Australian aborigines) ‘have names for almost every minute portion of the human frame; thus, in asking the name for the arm, one stranger would get the name for the upper arm, another for the lower arm, another for the right arm, another for the left arm, etc.”’1 “The Maoris have a most complete system of nomenclature for the flora of New Zealand. They are acquainted with the sex of trees, etc., and have distinct names for the male and female of some trees. Also they have different names for trees which change the form of their leaves, at the different stages of growth. In many cases they have a special name for the flowers of tree or shrub... different names for young unexpanded leaves and for the berries. . . . The koko or tut bird has four names ; two for the male and two for the female according to the seasons of the year. There are different names for the tail of a bird, of an animal, of a fish; three names for the cry of the kaka parrot (in anger, fear, or in ordinary circumstances).”’ 2 Speaking of the Bawenda tribe of South Africa, Gottschling says: “ For every kind of rain there is a special name in their language. . . . There is not a single geographical fact of their country but they have given it a name of their own. Even geological features have not escaped their notice, for they have specific names for every kind of soil and also for every sort of stone or rock. . . . There is not a tree, shrub or plant that has not a name in their language. They dis- tinguish even every kind of grass by a different name.” 3 Livingstone found the Bechuana vocabulary a source of wonder. “He” (Dr. Moffat) ‘“‘ was the first to reduce their speech to a written form, and has had his attention directed to the study for at least thirty years, so he may be supposed to be better adapted for the task” (of trans- lating the Bible) ‘‘than any man living. Some idea of the copiousness of the language may be formed from the fact that even he never spends a week at his work without dis- 1 Grey, Journals, etc., ii. p. 209 (1841). 2 Elsdon Best, ‘‘ Maori Nomenclature,” J/.A./., xxxii. pp. 197-8. 3 E, Gottschling, ‘‘ The Bawenda,” J.A.J., xxxv. p. 383. PRIMITIVES’ LANGUAGE 173 covering new words.’’!—With regard to India, Grierson speaks of “ the great number of terms for closely related ideas in the Kuki-Chin language,” making the comparison of the vocabularies of different dialects a matter of some difficulty. ‘Then in Lushei,” he says, “there are ten terms for ants, all probably denoting various kinds of ants; twenty terms for basket ; different words for different kinds of deer, but no general word for deer.” 2—The North American Indians ‘‘ have even many expressions, which may be almost called scientific, for frequently recurring forms of the clouds, and the characteristic features of the sky physiognomy which are quite untranslatable, and for which it is hopeless to seek an equivalent in Euro- pean languages. Thus the Ojibbeways, for instance, have a peculiar fixed name for the appearance of the sunshine between two clouds. In the same way they have a distinct appellation for the small blue oases which at times are seen in the sky between dark clouds.’’3—The Klamath Indians have no generic term for fox, squirrel, butterfly or frog; but each species has its own name. There is an almost countless number of substantives in the language.4—The Lapps have a great many terms to denote various kinds of reindeer, according to their age. . . . There are twenty words for ice, eleven for cold, forty-one for snow in all its forms, twenty-six verbs to express freezing and thawing, and so on. Hence they resist any attempt to exchange their language for Norwegian, which is much more limited in this respect.s— Finally, the Semitic languages, and even those we ourselves use, have known such wealth of vocabulary. ‘‘ We must imagine every Indo-European language as resembling the modern Lithuanian speech, poor in general terms, yet well supplied with specific ones indicating all particular actions and the details of familiar objects.’’ 6 | The same tendency accounts for the vast number of special names given to single objects, and particularly to the least peculiarity in the soil. In New Zealand, with the t Livingstone, Missionary Travels, pp. 113-14. * Grierson, Linguistic Survey of India, iii. 3, p. 16. 3 Kohl, Kitcht Gami: Wanderings round Lake Superior, p. 220. 4 A. Gatschet, The Klamath Language, pp. 464, 500. 5 Keane, “ The Lapps: their Origin, etc.” J.A.I., xv. p. 235. 6 A. Meillet, Introduction à l’Etude Comparative des Langues Indo-Euro- péennes (2nd edit.), p. 347. ; 174 HOW NATIVES THINK Maoris, ‘‘ everything has its name: their houses, canoes, weapons, and even garments have distinctive appellations given them. . . . Their lands and roads are all named; so also the sea beaches round the islands, their horses, cows, and pigs, even their trees . . . rocks and fountains. Go where you will, in the midst of an apparently untrodden wilderness ; ask, has this spot a name ? and any native belonging to that district will immediately give one.’’! In Southern Australia ‘every range has its name; likewise every mountain has its particular name; so that blacks can state the precise mountain or hill in an extensive range they will meet. I have upwards of two hundred names of mountains in the Australian Alps... even every bend in the river Murray has a name.’’2 In Western Australia, the natives ‘‘ have names for all the conspicuous stars, for every natural feature of the ground, every hill, swamp, bend of a river, etc., but not for the river itself.” 3—Lastly, not to prolong this list unduly, in the Zambesi district, ‘every knoll, hill, mountain, and every peak on a range has a name, and so has every watercourse, dell, and plain. In fact, every feature or portion of the country is so distinguished by appropriate names that it would take a lifetime to decipher their meaning.” 4 VI On the whole, therefore, the characteristics of the languages spoken by primitives correspond with those we have noted in their mentality. The image-concepts, which are a kind of delineation, allowing of a limited generality and elementary abstraction only, yet involve remarkable development of memory, and thus give rise to the wealth of form and vocabulary. Where logical thought has obtained the upper hand the social treasure of acquired knowledge is transmitted and preserved by means of concepts. Each generation in instructing that which succeeds teaches it to analyse these concepts, to draw out what is included in them, to recognize and make use of the resources of abstract reasoning. In the 1 R. Taylor, Te Ika a Maui, pp. 328-9. 2 Quoted by Brough Smyth, in The Aborigines of Victoria, ii. p. 122 (note). 3 Ibid., p. 266. 4 Livingstone, The Zambesi and its Tributaries, pp. 537-8. PRIMITIVES’ LANGUAGE 175 peoples whom we are considering, on the contrary, this treasure is still entirely, or almost entirely, explicit in the language itself. It cannot fail to be transmitted, because the children try to imitate their parents’ speech, without any teaching, properly so called, without any intellectual effort, simply by memory. It is accordingly not susceptible to progress. Supposing that the mieu and the institutions of a social group do not change, its general mentality not changing either, its wealth of image-concepts would be transmitted from generation to generation without any great variation. When it does change, other changes are at work also, and usually it becomes impoverished. Advance in conceptual and abstract thought is accom- panied by a diminution in the descriptive material which served to express the thought when it was more concrete. The Indo-European languages have undoubtedly evolved in this sense. In British Columbia, ‘on the coast, when a masculine or a feminine article is used, the same terms serve for male and female relations. Here’ (among the Salish) ‘ where there is no grammatical distinction between the sexes, separate terms are used. It is worth remarking that the Bilqula, who have grammatical distinction of sex, distinguish between but afew of these terms This may indicate that the separate forms have been lost by the tribes who use gram- matical sex.’’! The increasing generality of the concepts causes them by degrees to lose the exactness which characterized them when they were at the same time, and primarily, images, delineations, and vocal gestures. “Little by little,’ says Victor Henry, ‘the idea of these finer shades of meaning becomes obscured, so that the present-day Aleutians make use of one single verbal form with many different meanings, or several forms with one acceptation impartially, and the native who is questioned about the reason which makes him prefer one form to another will usa, be unable to account for his preference.’’ 2 This gradual impoverishment, iS is the general rule, shows ety that the specializing terminology, and the 1 F. Boas, “ The North-western Tribes of Canada,” British Association eo) pp. 690-1 (1890). ? V. Henry, Esquisse d’une Grammaire Raisonnée de la Langue Aléoute, P: 54. à 176 HOW NATIVES THINK meticulous attention to detail, were not the result of desired and conscious effort, but merely of the necessity which the mode of expression demanded. Image-concepts could only be rendered and communicated by a kind of delineation, either by means of actual gestures, or oral expressions which are a species of vocal gesture, of which the “ auxiliary descriptive adverbs ’’ have provided a very clear example. As soon as the development of general ideas and abstract concepts permitted men to express themselves more easily, they did so, without troubling about the loss of the graphic precision which resulted. In fact, the ingenuity, extent and delicacy of the distinctions perceived and expressed, between the varieties of the same species of plants or animals, for example, must not lead us to believe that in them we have a mentality oriented like our own, towards the recognition of objective reality. We know that their mentality is otherwise oriented. In the reality of persons and things as their collective repre- sentations suggest them, the mystic, invisible factors, the occult powers, the secret participations, hold an incomparably higher place than the elements we consider objective. We need no other proof of this than the part played by notions like mana, wakan, orenda, taboo, contamination, and so forth. It is even sufficient to consider the classifications of entities that have been established. With primitive peoples, the principle of classification, disdaining the most striking objec- tive traits, is founded preferably upon a mystic participation. The sum total of all entities is divided up as are the individuals of the social group; trees, animals, stars, belong to this or that totem or clan or phratry. In spite of appearances, then, these minds, which evidently have no idea of genera, have none of species, families, or varieties either, although they are able to delineate them in their language. Their classification 1s something purely pragmatic, born of the necessity for action and expression, in which reflection has — no part. So little of knowledge is therein that, for real knowledge to be formed, this material of thought and expres- sion must first give place to another, and the image-concepts, which are at once both general and particular, must be super- seded by concepts which are really general and abstract. The language, too, will have been forced to lose the mystic PRIMITIVES’ LANGUAGE 177 character which it necessarily assumes among primitive peoples. To their minds, as we already know, there is no perception unaccompanied by a mystic complex, no pheno- menon which is simply a phenomenon, no sign that is not more than a sign: how then could a word be merely a word ? Every form that an object assumes, every plastic image and every delineation has mystic virtues: verbal expression, which is an oral delineation, must perforce possess them also. And this power does not pertain to proper nouns alone, but to all terms, whatever they may be. Moreover, names which express very specialized image-concepts do not differ from proper nouns nearly as much as our common nouns do. Hence it follows that the use of words can never be a matter of indifference: the mere fact of uttering them, like the tracing of a drawing or the making of a gesture, may establish or destroy important and formidable participations. There is magic influence in the word, and therefore precaution is necessary. Special languages for certain occasions, languages reserved for certain classes of persons, begin to take shape. Thus, in a great many aggregates, we meet with a different language for men and women. Frazer collected many examples of this. ‘It is a feature common to all the American nations,” says Gallatin, ‘‘ that women use different words from men for those purposes ”’ (to express relationship) ; “and that the difference of language between men and women seems in the Indian languages to be almost altogether confined to that species of words . . . and to the use of interjections.’”’* At the time when the young men are initiated and become full members of the tribe, it often happens that the seniors teach them a secret language unknown by and unintelligible to the uninitiated. ‘I have on several occasions reported the existence of a secret or cabalistic language used only by the men at the initiation ceremonies of several native tribes in New South Wales. While the novitiates are away in the bush with the elders of the tribe, they are taught a mystic name for surrounding natural objects, animals, parts of the body, and short phrases of 1 J. G. Frazer, “ Men’s Language and Women’s Language,” Fortnightly Review, January 1900. Cf. Man, No. 129 (1901). + Gallatin, Transactions of the American Ethnological Society, ii. pp. 131-2. 12 178 HOW NATIVES THINK general utility.” Frequently, too, the members of the secret societies which are so common a feature in social groups of a primitive type, are initiated into a language spoken and understood by themselves alone; their introduction to the society, or their promotion to a sufficiently exalted rank therein, gives them the privilege of using this mystic language. Among the Abipones, “‘ persons promoted to the rank of nobles are called Hecheri and Nelereycait, and are distinguished from the common people even by their language. They generally use the same words, but so transformed by the interposition or addition of other letters, that they appear to belong to a different language. : . . Moreover, they have some words peculiar to themselves, by which they supersede those in general use.” 2 In hunting it is essential to avoid uttering the name of the animals hunted, and in fishing, that of the fish that one desires to capture. Accordingly silence is enjoined, or the use of sign-language, and this accounts for the appearance of a special language when they are looking for camphor, or going fishing, or starting on a warlike expedition. A great many words are taboo, when the person of the king is in question: to eat or sleep or sit may not be expressed in the ordinary Malay words; special terms are essential. More- over, when the king is dead, his name must be uttered no more.3 We know that this was a very common custom in Madagascar. ‘‘ There are many words which are used in a certain sense to the king (or queen) and these words cannot be used in this special sense with the common people ; especially those which have reference to the state or health of the living king. . . . Other words are common to kings and chiefs only. . . . The king has power to make certain words fady, viz. to prohibit their use, it may be for a time or entirely.”’ 4 In many primitive societies, a woman and her son-in-law must avoid each other and not enter into con- 1 Mathews, ‘‘ Languages of Some Native Tribes,’’ Journal of the Royal Society of New South Wales, pp. 157-8 (1903). Cf. Webster, Primitive Secret Societies, pp. 42-3. ’ 2 Dobrizhoffer, An Account of the Abipones, ii. pp. 204-5. 3 Skeat, Malay Magic, pp. 35, 212, 315, 523. Cf. Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. pp. 414-31. 4 Last, ‘‘ Notes on the Languages Spoken in Madagascar,” J.A.JI., xxv p. 68. PRIMITIVES’ LANGUAGE 179 versation. Nevertheless, ‘ throughout the central and south- west districts of Victoria and in the south-east corner of South Australia there is a hybrid tongue or jargon in use, com- prising a short code of words, by means of which a mother-in- law can carry on a limited conversation in the presence of her son-in-law respecting some events of daily life.” 1 That which finally proves the mystic worth and power in words as words, is the widespread custom, in magic cere- monies and even in ritual and religious ceremonies, of using songs and formulas which are unintelligible to those who hear them, and sometimes even to those who utter them. For these songs and these formulas to be effective, it is enough that they have been transmitted by tradition in a sacred language. For instance, with the tribes of Central Australia, Spencer and Gillen say: ‘‘ As usual, in the case of sacred ceremonies, the words have no meaning known to the natives, and have been handed down from the Alcheringa.”2 In myths it appears that a change of language is frequently mentioned ; for instance, ‘at this spot the Achilpa changed their language to that of the Arunta people.’ 3 Another part of the tribe ‘‘camped apart and then moved on to Ariltha, where they changed their language to the Ilpirra language.” 4 ‘Somewhere out west of the river Say the women (Unthippa) changed their language to Arunta.’’s So, too, in Fiji, Banks Islands, Tanna, New Guinea, the songs used in the sacred ceremonies are unintelligible to those who are singing.6 Throughout North America similar facts come to light. Jewitt noted, though without understanding, them in the Indians of Nootka Sound. ‘ They have,” he says, ‘‘ a number of songs which they sing on various occasions: war, whaling, fishing, marriages and feasts, etc. The language of most of these appears to be very different, in many respects, from that used in their common conversation, which leads me to believe, either that they have a different mode of expressing 1 Mathews, “ Aboriginal Tribes of New South Wales and Victoria,’”’ Journal of the Royal Society of New South Wales, p. 305 (1905). a The Northern Tribes of Central Australia, PP. 286, 462, 460, 606. 3 The Native Tribes of Central Australia, p. 410. 4 Ibid., p. 416. §Thid.; p. 442. 4: 6 Sidney H. Ray, ‘‘ Melanesian and New Guinea Songs,” J.A.I., xxvi. PP. 430-45. 180 HOW NATIVES THINK themselves in poetry, or that they borrow their songs from their neighbours.” ! Catlin fully realized the mystic meaning of this. ‘‘ Every dance has its peculiar step, and every step has its meaning; every dance also has its peculiar song, and that is so intricate and mysterious oftentimes, that not one in ten of the young men who are dancing and singing it know the meaning of the song which they are chanting over. None but the medicine-men are allowed to understand them; and even they are generally only initiated into these secret arcana on the payment of a liberal stipend for their tuition, which requires much application and study.” “A great portion of the phraseology of the Ojibwa ritual is in an archaic form of language, and is thus unintelligible to the ordinary Indian, and frequently to many members of the society. This archaic phraseology naturally appears impres- sive and important to the general populace, and the shamans delight to dwell on such phrases, not only to impress their hearers, but to elevate themselves as well.” 3 Of the Klamath Indians, ‘“ many ...do not understand all these songs, which contain many archaic forms and words, and the con- jurors themselves are generally loth to give their meaning, even if they should understand them.”’4 What we call the meaning of the words or the form matters little. The people remain indefinably attracted by them, because their mystic virtue and magic efficacy have been known from time imme- morial. The most accurate and intelligible translation could not take the place of these incomprehensible songs, for they could not fulfil the same office. t Jewitt, Adventures and Sufferings, p. 97. 2 Catlin, The North American Indians, i. p. 142; ii. p. 181. 3 Hoffman, ‘‘ The Memomini Indians,’”’ E. B. Rept., xiv. p. 61. 4 À. Gatschet, The Klamath Language, p. 160. CHAPTER V PRELOGICAL MENTALITY IN RELATION TO NUMERATION It is quite possible future works dealing with linguistics may confirm the theory I have suggested in the preceding chapter. Nevertheless, in what follows I shall confine myself to proving it in one particular point, upon which our documentary evidence is fairly complete and accessible: that is, the way in which different peoples, especially those of the most primitive type we know, practise numeration. The various methods of counting and calculating, of forming and using the names for numbers, will possibly enable us to see, actually at work, the mentality of primitives where it differs specifically from logical thought. This will serve as a specimen of proofs which I cannot enter upon in detail. I In a great many primitive peoples—those in Australia, South America, etc.—the only names for numbers are one and two, and occasionally three. Beyond these, the natives , say “ many, a crowd, a multitude.” Or else, for three, they say two, one; for four: two, two; for five: two, two, one. Hence the opinion has frequently been formed that mental inaptitude or extreme indolence prevents them from dis- criminating any number higher than three. This is a hasty conclusion, however. It is true that these “ primitives ” form no abstract concept of four, five, six, etc.; but we cannot legitimately infer from this that they do not count beyond two or three. Their minds do not readily lend them- selves to operations familiar to us, but by the processes which are peculiar to them they can obtain the same results to a 182 HOW NATIVES THINK certain extent. Where synthetic representations are not analysed, there is more demand upon memory. Instead of the generalizing abstraction which provides us with our con- cepts, properly so called, and especially with those of number, their minds make use of an abstraction which preserves the specific characters of the given ensembles. In short, they count and even calculate in a way which, compared with our own, might properly be termed concrete. Since we count. by means of numbers and hardly ever count in any other way, we admit that in primitive societies which have no names for any number beyond three, it would be impossible to count further. But are we obliged to take it for granted that the apprehension of a definite plurality of objects can take place in one way only? Is it impossible for the mentality of primitive peoples to have its own peculiar operations and processes to attain the end we reach by numeration ? As a matter of fact, if a well-defined and fairly restricted group of persons or things interests the primitive ever so little, he will retain it with all its characteristics. In the representation he has of it the exact number of these persons or things is implied: it is, as it were, a quality in which this group differs from one which contained one more, or several more, and also from a group containing any lesser number. Consequently, at the moment this group is again presented to his sight, the primitive knows whether it is complete, or whether it is greater or less than before. A capability of this nature has already been noted, in very simple cases, among certain animals. It does happen that a domestic animal, a dog, ape, or elephant, perceives the disappearance of an object in a restricted ensemble with which it is familiar. In many species, the mother shows by unmistakable signs that she knows that one of her little ones has been taken from her. If we remember that according to most observers the primitives’ memory is ‘‘ phenomenal,” as Spencer and Gillen express it, or ‘‘ miraculous,” as Charlevoix pronounces it, we have stronger reason for believing that they can easily do without numerals. With the help of custom, each sum-total which matters to them is retained in their memory with the same exactness as that which makes them 1 Ch. Leroy, Lettres sur les Animaux, p. 123. PRIMITIVES’ NUMERATION 183 recognize unerringly the track of such and such an animal or person. If anything is missing from the sum-total, they instantly perceive it. In the representation so faithfully preserved, the number of persons or things is not differentiated : nothing allows of its being expressed separately. It is none the less perceived qualitatively, or, if you prefer it, felt. Dobrizhoffer has testified to this fact with the Abipones. They refuse to count as we do, i.e. by means of numerals. They are not only ignorant of arithmetic, but they dislike it. Their memory generally fails them (because they are required to make use of processes with which they are not familiar). “They cannot endure the tedious process of counting : hence to rid themselves of questions on the subject, they show as many fingers as they like, sometimes deceived themselves, sometimes deceiving others. Often,’’ says Dobriz- hoffer, ‘‘ if the number about which you ask is more than three, an Abipon, to save himself the trouble of showing his fingers, will cry ‘ Pop,’ which means ‘ many,’ or Chic leyekalipi, ‘ in- numerable.’ ”’ 1 ( Yet they have their own way of accounting for numbers. “When they return from an excursion to hunt wild horses, or to shoot tame ones, none of the Abipones will ask them ,, _ “how many horses have you brought home?’ but ‘how , much space will the troop of horse which you have brought _ home occupy ?’’’2 And when they are about to start on an excursion, “‘as soon as they are mounted, they all look round, and if one dog be missing out of the many which they keep, begin to call him. . . . I often wondered how, without being able to count, they could so instantly tell if one were missing out of so large a pack.’’3 This last is a very charac- teristic reflection of Dobrizhoffer’s. It explains why it is that the Abipones and peoples like them, who do not make use of numerals, are at a loss to deal with them when they are first taught them. So, too, the Guaranis have no numerals above four (al- though they have terms corresponding with the Latin: singult, bint, trint, quaterni).4 They, ‘like the Abipones, t Dobrizhoffer, An Account of the Abipones, ii. p. 170. 2 Ibid 3 Ibid., ii. pp. 115-16. 4 So, too, the Australian aborigines, who have no numerals above three. are yet able to conjugate in the singular, dual, trial, and plural numbers. 184 HOW NATIVES THINK when questioned respecting things exceeding four, immediately reply: ‘innumerable.’ . . . Generally speaking we found the art of music, painting, and sculpture easier learnt than numbers. They can all pronounce the numbers in Spanish, but are so easily and frequently confused in counting that you must be very cautious how you credit what they say in this matter.” 1 Numeration is an instrument, the need for which they do not | \recognize, and the use of which is unfamiliar to them. They _do not want numbers, apart from the totals which they can \'count so easily in their own way. But, we may ask, if this be so, is not the only thing possible for primitives to represent these totals and preserve them in their memory ? Are not the very simplest additions or sub- tractions beyond their powers? Not at all; they can perform such operations, for the prelogical mind in this respect (as in its language in general) proceeds in a concrete fashion. It has recourse to the representation of the movements which add units to the original whole or else subtract them from it. In this it has an instrument much less effective though more complicated than abstract numbers, but which permits it to perform the simple operations. It associates a regular series of movements and of the parts of the body connected with such movements with successive totals in such a way as to recall any of these at need by repeating the series from the beginning. If it be a case of fixing the day upon which a number of tribes are to meet for the common celebration of certain ceremonies, it will have to be several months ahead, because so much time is needed to tell all who are interested, and allow them to reach the spot agreed upon. How do the Australian aborigines start about it? “To indicate the precise time upon which the people should assemble . . . could be done by counting the different stages or camps to be made on the journey, or the number of ‘moons.’ If the number to be counted was large, recourse was had to the various parts of the body, each of which had a recognized name, and an understood position in this method of enumera- tion. So many parts thus enumerated, counting from the little finger of one hand, meant so many stages, or days, or months, as the case might require.” (One side of the body * Dobrizhoffer, op. cit., ii. pp. 171-2. PRIMITIVES’ NUMERATION 185 would be gone over, and then the other if necessary.) Howitt rightly observes that “ this method of counting fully disposes of any belief that the paucity of numerals in the language of the Australian tribes arises from any inability to conceive of more numbers than two, three, four.’ : Whence does this paucity arise, indeed, if not in the habits peculiar to the prelogical mind? As a matter of fact, in nearly every case in which we note this scarcity of numerals— a scarcity which we should consider dependent upon the number not being differentiated from that which is being enumerated—we find also this concrete method of numbering. In the Murray Islands, Torres Straits, ‘‘ the only natives’ numerals are metat (1) and meis (2). Any higher numbers would be described either by reduplication; e.g. neis-netat =2,I1=3; nes neis = 2, 2 = 4, etc., or by reference to some_part of the body...By the latter method a total of 31 could be counted. The counting commenced at the little finger of the left hand, thence counting the digits, wrist, elbow, arm- pit, shoulder, hollow above the clavicle, thorax, and thence in reverse order down the right arm, ending with the little finger of the right hand. This gives 21. The toes are then resorted to, and these give 10 more.’”’2 “ Dr. Wyatt Gill says: ‘ Any- thing above ten the Torres Straits Islanders count visibly,3 thus: touch each finger, then the wrist, elbow, and shoulder joint on the right side of the body; next touch the sternum and proceed to the joints of the /e/t, not forgetting the fingers of the left hand. This will give 17. If this suffice not, count the toes, the ankle, knee, and hip joints (right and left). This will give 16 more, the entire process yielding 33. Anything beyond can be enumerated only by help of a bundle of sticks.’ ’’ 4 Haddon clearly recognized that there were no numerals, nor even numbers properly so called, to be seen, but simply a method, an “ aid to memory ” to recall a given total at need. “ There was,” he says, ‘‘ another system of counting by com- * Howitt, ‘ Australian Message Sticks and Messengers,” J.A.I., xviii. PP. 317-19. * Hunt, “ Murray Islands, Torres Straits,” J.A.I., xxviii. p. 13. 3 This is a striking expression, which reminds us of the language of primitives, in which verbal utterance seems like a “ tracing’ of visual and motor images. 4 A. Haddon, “ The West Tribes of Torres Straits,” J.4.I., xix. pp. 305-6. 186 HOW NATIVES THINK © mencing at the little finger of the left hand, then following on with the fourth finger, middle finger, index, thumb, wrist, elbow joint, shoulder, left nipple, and ending with the little finger of the right hand (19 in all). The names are simply those of the parts of the body themselves and are not numerals. In my opinion, this system could only have been used as an aid in counting, like using a knotted string, and not as a series of actual numbers. The elbow joint (kudu) might be either 7 or 13, and I could not discover that kudu really stood for either of those numbers, but in a question of trade a man would remember how far along his person a former number of articles extended, and by beginning again on the left little finger he could recover the actual number.” ? So, too, in British New Guinea, we find the following system used in reckoning : I = monou—little finger of the left hand; 2 = veeve—next finger ; = haupu—middle finger ; 4 = moreere—index ; 5 = aiva—thumb; 6 = ankova—wrist ; | 7 = mirika mako—between wrist and elbow; 8 = na—elbow; 9 = ava—shoulder ; 10 = ano—neck; II = ame—left breast; 12 = unkari—chest ; 13 = amenekai—right breast ; 14 = ano—right side of the neck, etc.? We notice that the same word, ano, for the neck, either on the right or left side, does for 10 and for 14, which would be quite impossible if it were a question of numbers and numerals. Yet there is no ambiguity here, for it is the naming of the parts of the body in a fixed order that eliminates confusion. | A British scientific expedition to the Torres Straits brought to light a certain number of facts which fully confirm the preceding. I shall quote but a few of them. At Mabuiag, ‘counting is usually performed on the fingers, beginning with the little finger of the left hand. There was also a system of 1 “The West Tribes of Torres Straits,’ J.A.J., xix. p. 305. 2 James Chalmers, ‘‘ Maipua and Namau Numerals,” J.4.1., xxvii. p. 141. PRIMITIVES’ NUMERATION 187 counting on the body, by commencing at the little finger of the left hand: 1. kutadimur (end finger); 2. kutadimur gurunguzinga =a thing following the end finger (fourth finger) ; 3. a get= middle finger; 4. klak nitui get (index finger) = spear-throwing finger; 5. kabaget = paddle finger (thumb) ; 6. perta or tiap = wrist; 7. kudu = elbow joint; 8. zugu kwuick = shoulder ; 9. susu madu = breast flesh, ster- num; 10. kosa dadir = right nipple; 11. wadogam susu madu = other side breast flesh ; and so on, in reverse order, preceded by wadogam (other side) ; the series ending with the little finger of the right hand. . . . The names are simply those of parts of the body themselves, and are not numerals.” 1 Manus, a native of the Murray Islands, counted as follows: 1. kebi ke—little finger ; 2. kebi ke neis—little finger two; 3. etp ke—middle finger ; . 4. baur ke—spear finger (index) ; 5. au ke—big finger (thumb) ; 6. kebt kokne—little bone joint (wrist) ; 7. kebt kokne soy—little bone back (back of the wrist ) ; 8. au kokne—big bone joint (inner part of the elbow) ; 9. au kokne sor—bone joint back (elbow) ; 10. tugar—shoulder ; 11. kenani—armpit ; 12. gilid—pit above clavicle ; 13. nano—left nipple ; 14. koporv—navel ; 15. nerhep—top of chest ; 16. op nerpek—front of throat ; 17. nerut nano—other nipple ; 18. nerut gilid ; 19. nerut kenant, etc., up to 29. kebi ke nevute—little finger another.? In British New Guinea also, counting is accomplished by enumerating certain parts of the body, in a way that differs slightly from the preceding, but also going back on the right side after having begun on the left. ‘‘ This is done in the Elema district, I= haruapu, 2 = wrahoka, 3 — iroihu, 4 = han (index), 5 = hui (thumb), 6 = aukava (wrist), 7 = farae (forearm), 8= ari (elbow), 9 = kae (armlet), 10 = horu (shoulder), 11 = kavave (neck), 12 = avako (ear), 13 = ubuhae t The Cambridge Expedition to Torres Straits, iii. p. 47. 2 Ibid., ili, pp. 86-7. 188 HOW NATIVES THINK (eye), 14 = overa (nose), 15 = ubwauka (eye), 16 = avako kar (other eye), 17 = karave haukai (neck, other side), etc., etc., up to 27, ukai haruapu. . . . In the numbers from I5 onwards, kai, ukai, haukai, probably mean other or second.” Here is a final example, from a Papuan language spoken in the north-east of British New Guinea. ‘‘ According to Sir William MacGregor, the practice of counting on the body is found in all the lower villages on the Musa river. They begin with the little finger of the right hand, use the fingers of that side, then proceed by the wrist, elbow, shoulder, ear, and eye of that side, thence to the left eye and shoulder, and down the left arm and hand to the little finger. Many of them in count- ing become greatly confused on reaching the face.””? ‘ Here is an example of this method: 1 = anusi (little finger of the right hand), 2, 3, 4 = doro (are the ring, middle, and index fingers respectively of the right hand), 5 = ube: (thumb), 6 = tama (right wrist), 7 = unubo (elbow), 8 = visa (shoulder), 9 = denoro (right ear), 10 = diti (right eye), 11 = dit (left eye), 12 = medo (nose), 13 = bee (mouth), 14 = denoro (left ear), 15 = visa (left shoulder), 16 = unubo (left elbow), 17 = tama (left wrist), 18 = uber (thumb), 19, 20, 2I = doro (the index, middle, and ring fingers of the left hand), 22 = anust (little finger of the left hand).”’ 3 Here we see very clearly that the terms used are not numerals. How could the word doro stand for 2, 3, 4 and 19, 20, 21 alike, if it were not differentiated by the gesture which indicates the special finger of the right or left hand ? Such a process admits of the counting to fairly high numbers, when the parts of the body enumerated in a fixed order are themselves associated with other objects more easily handled. Here is an instance taken from the Dayaks of Borneo. It was a case of announcing to a certain number of insurgent villages which had been conquered, the amount of fine which they would have to pay. How would the native messenger accomplish his task? ‘‘ He brought a few dry leaves, which he tore into pieces; these I exchanged for paper, which served better. He arranged each piece separately on a table, and used his fingers in counting as well, until he reached 10, when he lifted his foot on the table, and took each toe to accord : OST PDO 2 Op. cit., p. 364. 3 Ibid. PRIMITIVES’ NUMERATION 189 with each bit of paper answering to the name of a village, name of chief, number of followers, and amount of fine : after having finished with his toes, he returned to his fingers again, and when my list was completed, I counted forty-five bits of paper arranged on the table. He then asked me to repeat them once more, which I did, when he went over the pieces, his fingers and toes, as before. ‘ Now,’ he said, ‘ this is our kind of letter; you white men read differently to us.’ Late in the evening he repeated them all correctly, placing his finger on each paper, and then said: ‘ Now, if I recollect them to-morrow morning it will be all right ; so leave these papers on the table,’ after which he mixed them all in a heap. The first thing in the morning he and I were at the table, and he proceeded to arrange the papers as on the evening before, and repeated the particulars with complete accuracy ; and for nearly a month after, in going round the villages, far in the interior, he never forgot the different amounts, etc’: The substitution of pieces of paper for the fingers and toes is particularly noteworthy, for it illustrates a clear case of abstraction, still really concrete, with which the prelogical mind is familiar. The inhabitants of Torres Straits, who have very few numerals, have ‘‘ a custom of purchasing canoes on the three- years-hire system,” and at the end of that period they are supposed to have paid for them. This method necessitates a fairly complicated method of counting and some elementary calculation. Even Australian natives, who have no numeral above two, find some way of adding. ‘The Pitta-Pitta aboriginal has words for the first two numerals only... . Beyond 4, the savage will generally speak of ‘a lot, a large number.’ He certainly has visible conceptions of higher numbers” (this expression recalls that used by Haddon, already quoted), ‘‘ and I have often had a practical demon- stration of the fact by asking him to count how many fingers and toes he has, and telling him to mark the number in the sand. He commences with the hand open, and turns his fingers down by two, and for every two he will make a double stroke on the sand. ... The strokes he makes... ere * Brooke, Ten Years in Sarawak, i. pp. 139-40. | 2 A. Haddon, ‘The West Tribes of Torres Straits,’ J.A.I. xix. PP. 316, 342. 190 HOW NATIVES THINK parallel one beside the other, and when the numeration is complete, he calls pakoola (2) for every two of them. This method of counting is common throughout the district, and often practised by the elders of the tribe to ascertain the number of individuals in camp.”’ } Without describing concrete numeration so precisely as in the cases just cited, it frequently happens that observers allow us to perceive it in that which they report. The missionary Chalmers, for instance, tells us that ‘‘among the Bugilai of British New Guinea, he has found the following numerals : I = tarangesa (small finger of the left hand), 2 = meta kina (next finger), 3 = guigimeta (middle finger), 4 = topea (next to middle), 5 = manda (thumb), 6 = gaben (wrist), 7 = trank- gimbe (elbow), 8 = podei (shoulder), 9 = ngama (left breast), 10 = dala (right breast).”’ 2 From the facts we have just quoted it is allowable to suppose that a more searching observation would have revealed that these are names for parts of the body used in concrete numera- tion rather than numerals. Moreover, such numeration may unconsciously become half-abstract and half-concrete, as the names (especially the first five) gradually bring before the mind a fainter representation of the parts of the body and a stronger idea of a certain number which tends to separate itself and become applicable to any object whatever. How- ever, nothing proves that numerals are formed thus. The contrary seems even to be the rule for the numbers I and 2. In the western tribes of Torres Straits, Haddon notes names for numbers up to six, and adds: “‘ Beyond that they usually say ras or ‘a lot.’ . . . I also obtained at Muralug nabiget = 5, nabiget nabiget = 10, nabtkoku = 15, nabikoku nabikoku = 20. Get means hand, and 2oko foot.” But he adds: ‘* Nabiget can hardly be said to be the name of the number five, but that there were as many objects referred to as there are fingers on one hand.’’3 In other words, the number is not yet an abstract one. In the Andaman Isles, in spite of the ‘“‘ wealth of formative particles, the numerals are limited to i and 2. Three really 1 W. E. Roth, Ethnological Studies Among the N.W. Central Queensland Aborigines, No. 36. (The italics are the author’s.) 2 James Chalmers, ‘‘ Maipua and Namau Numerals,” J.A.J., xxvii. p. 139. 3 “‘ The Western Tribes of Torres Straits,” J.4.1., xix. pp. 303-5. PRIMITIVES’ NUMERATION 191 means ‘one more,’ 4 = some more, 5 = all, and here their arithmetic may be said to stop altogether. In some groups, however, 6, 7 and perhaps even 10 may be reached by the aid of the nose and fingers. First the nose is tapped with the little finger of either hand to score I, then with the next finger for 2, and so on up to 5, each successive tap being accompanied with the word anka (and this). The process is then continued with the second hand, after which both hands are joined together to indicate 5 + 5, the score being clenched with the word ardura (all). But few get as far as this, and the process usually breaks down at 6 or 7.”’1 When it is possible to trace them back to their original meaning, numerals proper frequently reveal the existence of concrete numeration similar to, if not identical with, that of which we have given some instances. But instead of going over the different parts of the upper half of the body with an ascending movement to return on the other side in descending, this concrete enumeration is connected with the movements made by the fingers while counting. Thus are produced those concepts which Cushing has very aptly called “ manual ’— concepts of which he has made an original and exhaustive study, and one which we might term experimental, since an essential part of his system consisted in recalling the psycho- logical condition of the primitives by forcing himself to accom- plish exactly the same movements as theirs. Here are the “manual concepts ” which serve the Zuñis for counting the earlier numbers : 1 = opinte—taken as a starting-point ; 2 = kwilli—raised with the preceding one; 3 = ha’i—the finger which divides equally; = awiie—all the fingers raised except one; 5 = 6pte—the one cut off; © = topalik’ya—another added to what is already counted ; 7 = kwillik’ya—two raised with the rest; 8 = hailik’ya—three raised with the rest; 9 = lenalik’ya—all except one raised; 10 = dstem’thila—all the fingers; 11 = dstem’th la topayathl’ tona—all the fingers and one more raised, etc.? t Portland, “‘ The Languages of the South Andaman LEMDES OPA RUES PP: 393-5. * American Anthropologist, p. 289 (1892). 192 HOW NATIVES THINK In his book entitled The Number Concept, Conant quotes similar systems of “manual concepts.” Here is a final instance, reported of the Lengua Indians of Chaco in Paraguay : “ Thlama (x) and anit (2) are apparently rootwords ; the rest appear to depend upon them, and upon the hands. Anta- thlama (3) appears to be made by these two words joined ; 4 = ‘two sides alike,’ and 5 = ‘one hand’; 6 =" arrived at the other hand, one’; 7 = ‘arrived at the other hand, two,’ and so on. 10 =‘ finished the hands’; 11 = ‘arrived at the foot, one’; 16 = ‘arrived at the other foot, one’; 20 | — ‘finished the feet.’ Beyond that comes ‘ many, and if a very large number is required, the ‘hairs of the head’ are called into requisition.” ! But we must note that cases vary according to the state of development attained. The Zuñis count up to a thousand at least, and there is no doubt that they have real numerals, although the concrete enumeration of former times still appears under these. The Chaco Indians of Paraguay, on the other hand, like the Australian aborigines, do indeed seem to make use of a regular and constant series of concrete terms in which numbers are implied, though not yet differentiated. II It is usually admitted as a natural fact, requiring no examination, that numeration starts with the unit, and that the different numbers are formed by successive additions of units to each of the preceding numbers. This is, in fact, the most simple process, and the one which imposes itself upon logical thought when it becomes conscious of its functioning. Omnibus ex nihilo ducendis sufficit unum. Prelogical mentality, however, which has no abstract concepts at com- mand, does not proceed thus. It does not distinctly separate the number from the objects numbered. That which it expresses by speech is not really numbers, but “ number- totals,” the units of which it has not previously regarded singly. To be able to imagine the arithmetical series of whole numbers, in their regular order, starting from the unit, it must have separated the number from that which the number 1 Hawtrey, “ The Lengua Indians of the Paraguayan Chaco,” /.4 de XXXI. p. 290. PRIMITIVES’ NUMERATION 193 totals, and this 1s precisely what it does not do. It imagines, on the contrary, collections of entities or objects which are familiar to it both by nature and by number, the latter being felt and perceived, though not conceived in the abstract. Accordingly, Haddon says of the western tribes of Torres Straits, that ‘‘ they have a decided tendency to count by twos, or couples.” And Codrington says: ‘In counting by couples in the Duke of York Island they give the couple different names, according to the number of them there are. The Polynesian way was to use numerals with the understanding that so many pairs, not so many single things, was meant ; hokorua (20) meant 40 (20 pairs). In this instance, again, we might suppose that the natives start with the unit 2; agreeing to regard it as equal to 1. But Codrington adds : ‘In Fiji and the Solomon Islands there are collective nouns signifying tens of things very arbitrarily chosen, neither the number nor the name of the thing being expressed.’ (This is what we have just termed “ number-totals,” perfectly definite, but not differentiated.) ‘‘ Thus in Florida na hua is ten eggs; na banara, ten baskets of food. ... In Fiji bola is a hundred canoes, ovo a hundred coco-nuts, and salavo a thousand coco-nuts. . . . In Fiji four canoes in motion are 4 waga sagat va, from gai, to run. In Mota two canoes sailing together are called aka peperu (butterfly—two canoes), from the look of the two sails, etc.” | As these ‘‘ number-totals ”’ may be varied indefinitely, the prelogical mind will find itself possessed of a very small number of numerals, properly so called, and of a surprisingly | vast multiplicity of terms in which number is implied. Thus, ‘in the Melanesian tongues, “when persons or things under certain circumstances are reckoned, the numeral is not simply used, but is introduced by a word which more or less describes the circumstances. If ten men are spoken of regarded as in a company together, it would not be o tanun sanaval, but 0 tanun pul sanaval, pul meaning to be close together; ten men in a canoe are tanun sage sanaval ; etc.” ? In this respect we have a very noteworthy observation regarding the natives of New Pomerania. “ Counting above Io was far more trouble to them than our little ones find in * Melanesian Languages, pp. 241-2. * Ibid, pp. 304-5. 13 194 HOW NATIVES THINK ‘ once one is one, twice one is two.” They did not use their toes. After many attempts, it was found that they do not differentiate between 12 and 20; both are called sanaul lua, 10 plus 2 as well as 10 multiplied by 2. It is clear that they feel no need to distinguish them in speech, for they never count in abstract numbers, using numbers only in connection with substantives (number-totals): for instance, 12 coco-nuts, 20 taro roots, a heap of 10 serving as the ‘unit in the latter case. Then one can see whether it is a case of 10 coco-nuts and 2 more, or of 2 heaps of I0.’’! Very frequently different names are given to totals com- posed of different things, even though the number may be the same. Then the languages seem to possess multiple lists of numerals; but we must note that the number is wholly differentiated. In his very serviceable book, Conant has collected a good many facts of this kind, of which I shall quote a few only. In the Carrier tongue, one of the Déné dialects of Western Canada, the word tha means 3 things; thane, 3 persons; that, 3 times ; thatoen, in 3 places; thauh, in 3 ways; thatltoh, 3 things together; thahultoh, the 3 times considered as a whole.z In the Tsimshian language of British Columbia, we find seven distinct series of numbers used to count different classes of objects. The first serves for counting when there is no definite object referred to, the second for flat objects and animals, the third for round objects and the divisions of time, the fourth for men, the fifth for long objects, the numbers being combined with the word kan (tree) ; the sixth. for canoes, and the seventh for measures. This last seems to comprise the word anon (hand). Boas gives a table of the first ten numbers in the seven classes (see page 195). We shall note that the first class, that of the words which are used for counting in general, is almost identical with the second, with the exception of a slight difference in r and 8. It is therefore allowable to suppose that the first class is not formed at the same time as the others, or independently of them, but, on the contrary, that there were special numerals 1 Dr. Stephan, “ Beitrage zur Psychologie der Bewohner von Neu- Pommern,”’ Globus, Ixxxviii. p. 206 (1905). | 2 Morice, The Déné Languages, quoted by Conant in The Number Concept, p. 86. 195 PRIMITIVES’ NUMERATION juoodx . sde, 43 ueysjood ¥ redx 12dx de,48 de 48 OI JUOITUSP919% H9801979H | UPHSJOEWI9J9Y [89201999 9CUI939H 9PUI994 9979 6 juoçopre}4n À 4}e71n0 4 ueYSp9r[},49 [epeorxn Jen À 1e74n4 Jrepuens g quooprebds 3 x11ebds 3 ueysrebdo 3 reprebdse 3 J1ebde 3 Jebds 3 Jebdo,3 E PUOTOP[e ATE Ueys}yOe A -| repre, x We He I He 9 JUOTISUOJ0% HSU0079% ueHS3U0J9, [eosus0y 9039 9U079% 9U079X ¢ yuorebdyebz xbsdreb} uexsdeebz | repbdreb} bdreb} bdreb} bdyeb3 v jquoopns yueys}es ueysj[es [eqns apn8 quens quens € [eqns x31°2dre,s ueysdoes repebds pdno8 zebds 3ebds 3 z [2.4 JOeUIe FY UvYSINMe H [e,X [919,8 yes xe48 I | 90089 ‘syoofqo Suot . re uen} ‘spoofqo punoy | “siofgozenx | ar 196 HOW NATIVES THINK for certain categories of objects before there were any for This is confirmed by an examination of the neighbouring languages in British Columbia. The number of numeral series there is almost “unlimited.” in the Heiltsuk dialect. simple counting. OBJEcT COUNTED. I. 2e maalok Here are some 3e animated being menok yutuk round object menskam masem yutqsem long object ments’ak mats’ak yututs’ak flat object menaqsa matlqsa yutqsa day op’enequls matlp’enequls yutqp’enequls fathom op’enkh matlp’enkh yutqp’enkh united — matloutl yutoutl group nemtsmots’utl matltsmots’ult yutqtsmots’utl full cup mengqtlala matl’aqtlala yutqtlala empty cup menqtla matl’aqtla yutqtla full box menskamala masemala yutqsemala empty box menskam masem yutqsem loaded canoe mentsake mats’ake yututs’ake canoe and its crew ments’akis mats’akla yututs’akla all on the shore — maalis — all in the house — maalit] — etc. Of the Kwakiutl, Boas says: ‘‘ Besides the class-suffixes for animated beings, round, flat, long objects, days, fathoms, the numerals may take any of the noun suffixes. . . . The number of classes is unlimited. They are simply compounds of numerals and the noun-suffixes.””? This is unusual copious- ness, but it can be readily understood when we look back upon the general character of these languages, which are but slightly abstract, and pre-eminently “ pictorial.” It is hardly surprising that the numerals do not stand alone. This accounts, too, for a peculiarity of the Micmac tongue of North America, which Conant pronounces “ extremely ,noteworthy.” In it, he says, numerals are really verbs | instead of being adjectives or, as we occasionally find, nouns, They are conjugated in all the divers forms of mood, tense, person, and number. For instance, nazooktaich means “‘ there is one” (now); naiooktaichcus, ‘there was one” (imperfect) ; 1 F, Boas, ‘“ The North-western Tribes of Canada,” Report of the British Association for the Advancement of Sciences, p. 658 (1890). a Ibid., pp. 655-6. PRIMITIVES’ NUMERATION 197 and encoodaichdedou, ‘there will be one”’ (future). The various persons are shown by the following inflections : Present. Ist person iahboosee-ck we are two 2nd person tahboosee-yok you are two 3rd person tahkboo-sisik they are two Imperfect. Ist person tahboosee-egup we were two 2nd person tahboosee-yogup you were two 3rd person {ahboosee-sibunik they were two Future. 3rd person tahboosee-dak they will be two, etc. There is a negative conjugation also: fahboo-seekw, they are not two; mah tahboo-seekw, they will not be two; and so on, natooki meaning one, and tahboo two. Conant explains these forms by saying that the numbers are verbs here, and they are being conjugated, But he might just as reasonably have said that these verbs are numerals, numerative verbs. We who know that primitive languages are not divided into parts of speech corresponding exactly with our own, and that it is better to consider that they contain words “ functioning as verbs,’ although under other aspects they may be nouns, adjectives, etc., shall simply say that in the case under review, that which we call numerals in our languages are here ‘‘ functioning as verbs.” It is not only in North America that we find facts of this kind. In India, Grierson collected similar instances. Thus in the Kuki Chin group of the Tibeto-Burman family of languages, ‘‘ the numerals are, in this way, restricted in their sphere so as to apply to some special kind of objects.” And he reports that these languages show a “‘ tendency to specialize and individualize.” 2 For example, in the Rang Khôl tongue, the prefix dar is used when the numerals refer to money, and dong when they refer to houses.3 These prefixes vary, 1 Schoolcraft, Archives of Aboriginal Knowledge, v. p. 587, quoted by Conant, The Number Concept, p. 160. ; 2 Linguistic Survey of India, iii. 3, p. 19. 3 Ibid., iii. 3, p. 184. 198 HOW NATIVES THINK too, with the form of objects: “ fam, which is used for round things ; pérr, for loads or bundles. Thus, mat pum kat means one pumpkin ; thing porr kat, a load of wood.’’* Sometimes there are special prefixes for definite classes of things. ‘“‘ Thus sak is used when human beings are counted, ge when inanimate things are counted, mang when animals, and bol when trees. These nouns are prefixed to the numerals. Mande sak gut signifies two men. The prefix ge is also employed in simple counting: gé sa, gé gui, gé gitam—1, 2, 3. After 20, these particles are added between the tens and the units.” ? In the Mikir language of the Naga group of the Tibeto-Burman family, generic prefixes are used with numbers, such as— With persons .. ne % Be .. hang With animals .. ey al ORE With trees and things siahatis g up ae in. TOME With houses or : 3 .. hum With flat things . a i is a Or With globular things oe pum With parts of the body, as APA rings, _ bangles and other ornaments .. as hong Finally, according to observations quoted by Conant, the same multiplicity prevails, ‘to a certain extent,’’ with the Aztecs. It is in current use by the Japanese, and Crawfurd found fourteen different classes of numbers without exhausting the list.3 According to our view these facts are traceable to the general trend of primitive mentality, for as its abstractions are always specializing rather than generalizing, it does, at a. certain degree of development, form numerals; but they are not abstract numbers such as we use. They are invariably the number-names of certain classes or persons or things, and these classes most frequently depend upon the conformation, attitude, position and movement of the objects. Now we have already seen how much importance these primitive languages accord to everything that expresses the contour and the relative position and movement of objects in space. This is carried to such a point that it would frequently be possible to superpose the detail of that which the words signify, upon 1 Linguistic Survey of India, iii. 3, p. 118. a Abid: il 0, ps Fis 3 The Number Concept, p. 89. p PRIMITIVES’ NUMERATION 199 the delineation which translates it into a reality for the eyes, and the gesture-language which expresses it by movement. In this way we can account for a phenomenon which occurs fairly frequently, and which is closely connected with the preceding. In certain languages numeration consists not only of numerals, more or less differentiated, but also of auxiliary terms, which are added to certain numbers to mark or range the stages of the numeration. English and American authors call such terms “ classifiers.’’ ‘‘ These verbs,” says Powell, “express methods of counting and relate to form : that is, in each case they present the Indian in the act of counting objects of a particular form and placing them in groups of ten.’”’! Boas has collected many examples of this kind among the dialects of British Columbia. They clearly show that the function of these auxiliaries is to make visible: as it were, the successive stages of the arithmetical process. “The appended verbs,’’ says Powell, ‘used as classifiers signify to place; but in Indian languages we are not apt to find a word so highly differentiated as place, but in its stead a series of words with verbs and adverbs undifferentiated, each signifying to place with a qualification, as I place DA I lay alongside of, I stand up, by, etc.” _ Thus these appended verbs are doubly specialized : firstly, in that which concerns the movements executed by the subject counting, and secondly in respect of the form of the objects to be counted. “The verbs serving as classifiers,’ says Gatschet, “‘ differ according to the shape of the counted objects, but all agree as to their common signification of depositing, placing on the top of.’ And he adds: ‘‘ The fact that the units from one to nine are not accompanied by these terms must be explained by some peculiarity of the aboriginal mode of counting. . . . The first ten objects counted (fish, baskets, arrows, etc.) were deposited on the ground in a file or row; and with the eleventh a new file was commenced : . . or a new pile.” Moreover, we are told, ‘‘ these appended verbs are not used for 10 or for multiples of 10. Such suffixes classify only the unit or units following the 10, not Io itself. This detail “ The Evolution of Language,” E. B. Rept., i. p. xxi. 2 mr Gatschet, The Klamath Language, p 534. 200 HOW NATIVES THINK throws light upon their origin and the reason for using them. The very number which follows directly after the 10 or its multiples, II, 31, 71, 151, etc., is sometimes accompanied by other classifiers than the numbers 32 to 39, 72 to 79, etc. ; because in the first case it applies to a single object, whilst in the others it relates to plurality. When I say 21 fruits— lip ni ta unepanta na’sh lutish likla—it literally means: above the 20 fruits one is placed on the top. When I say 26 fruits—ldpéna ta unepanta na’dshkshapta lutish péula—I understand: upon the top of twice 10 fruits I place 6 more. (The words /ékla and péula are only used in speaking, of globular objects.) But the 20 fruits which had been pre- viously counted are not recalled by the classifier, which relates only to the units mentioned by the number. The classifying verb may be rendered by the indefinite expression ‘ counted, numbered’; and before it the pronoun is omitted, but not before its participles Jiklato, péulatko. The simple verbal form, absolute or distributive, is used when the person speaking, or another person, is about to count the objects: the past participle placed in the direct or oblique cases in its absolute or distributive forms, is used when the objects have been previously counted, or the number is recalled.” It should be added that these appended verbs are not always used correctly by the Indians, and they often omit them, seeming to per- ceive, as Gatschet says, that they are “‘ a useless and cumber- some addition.’’ But they are nof an addition at all. Nothing favours the belief that the prelogical mind would have been more sparing in its counting than in expressing in language the sum-totals of its ideas. Its numeration merely presents the same quality of minute specialization and “ pictorial ” description that we found in the general construction of the languages of primitives. Codrington made a very careful study of numeration in the Melanesian languages. I have already tried to explain some of the data which he collected. Here I shall call attention to the following fact: the same term may signify different numbers successively. Codrington is thinking of what may be termed the number-limit, the point at which enumeration stops short. ‘A word,” he says, “ which in itself, though PRIMITIVES’ NUMERATION 201 we may not be able to trace its original meaning, is used to signify the end of the counting, naturally rises as the process of counting advances, to the signification of a higher number than it expressed at first. Thus in Savo fale or sale is Io, which in the Torres Islands = 100; the word, no doubt, is the same. As fini may possibly have signified the complete numeration as 3 in Mengone, and have advanced to Io in Fiji, and even to 10,000 in Maori, so tale may have signified the end of counting when no number beyond Io was counted, and have retained the meaning of 10 in Savo, while it has been advanced, as numeration improved, to signify 100 in Torres Islands. ‘Many’ means more in a later generation than in an earlier; the Lakona gapra (10) means nothing but “many”; tar, which in some languages is vaguely ‘ many,’ is in one 100, in several I,000.’’ ! Clearly in its original form this number-limit is not a number, and the word expressing it is not a numeral either. It is a term containing the more or less vague idea of a group of objects exceeding the “ number-totals,”’ of which the primi- tive’s mind has a clear and familiar intuitive grasp. But as numeration advances this term becomes a number, and more- over one that increases in value. When at last numeration is carried on by means of abstract numbers such as our own, the number-series is regarded as indefinite and the limiting term ceases to exist. The number is now absolutely differentiated from the things numbered, and the processes of logical thought supersede the functionings of prelogical mentality. III The result of all that we have just learnt seems to be a transformation of the traditional problems and a new method of treating them. Conant, for instance, after having collected the numerals employed by many primitive peoples in different parts of the world, is frankly puzzled by the diversity he finds in the numerical systems. Whence have the systems in use, differing so widely from each other, derived their bases ? How is it that the quinary system—the most natural 1 Codrington, Melanesian Languages, p. 249. 202 HOW NATIVES THINK of all, and one which would seem to be suggested to, and even imposed upon, man as soon as he begins to count— has not been universally adopted? What is the reason of there being so many binary, quaternary, vigesimal, mixed and irregular systems ? Counting on his fingers, would not man have inevitably been led to 5 as a basis of reckoning ? Conant finds the basic 4, which is met with fairly often, the most puzzling of all. It seems incredible to him that men capable of counting up to 5 (with the help of their fingers) and beyond 5, could have gone back to 4, to make it the basis of their numerical system. It is an enigma to which, he frankly admits, he has found no clue. Yet it is an artificial enigma. In formulating it one has to assume that individual minds like our own—that is, func- tioning in the same way, and accustomed to the same logical processes—have manufactured a system of numeration with these processes in view, and that they have been obliged to choose for it the basis which was most in accordance with their . experience. Now such a supposition is unwarranted. As a matter of fact, numerical systems, like the languages from which they must not be disjoined, are social phenomena which depend upon the collective mentality. In every aggre- gate this mentality is strictly dependent upon the type of the aggregate and its institutions. In primitive societies the mentality is mystic and prelogical; it expresses its thought in languages in which abstract concepts, such as we employ, are practically unknown. These languages do not possess numerals, properly so called, either, or at least, hardly ever. © They make use of words “ functioning as numbers,” or else they have recourse to ‘‘ number-totals,” concrete representa- tions in which the number is still undifferentiated. In short, however paradoxical the statement may appear, it is never- theless true that for long ages primitive man counted before he had any numbers. If this be so, how can we regard one special basis of a numerical system as more “‘ natural ” than any other? Every basis actually adopted has been founded upon the collective representations of the social group in which we discover it. In the lowest class in which we are able to make observations, where the numeration is almost wholly concrete, there is no PRIMITIVES’ NUMERATION 203 basis at all, neither is there any numerical system, The succession of movements going from the little finger of the left hand to the little finger of the right, traversing successively the fingers. of the left hand, then ascending the wrist, elbow, etc., to descend again in the inverse order on the right side of the body, has no one time that is more stressed than any other. It does not pause longer on the part of the body which corresponds with 2 or 5 or 10 than with any other part, for instance. Therefore Haddon is quite right in saying that the words pronounced are the names of the parts of the body and not the names of numbers. Numerals do not appear until a Lo periodicity begins to inform the series. This periodicity is in fact most frequently regulated by the number of fingers and toes : in other words, the quinary system is the most usual. It is not yet certain that, wherever we meet with it, its origin is that which seems so natural to us. Nearly all primitives use their fingers to count with, and often too, those who know nothing of the quinary system as well as those who use it. On this point the study of the ‘* manual concepts ”” is very instructive. Here is the method of counting by a Dènè-Dindjié in Canada, for instance. “Extending the left hand, the palm turned towards his face, he bends his little finger, saying, 1; then he bends the ring finger, saying 2, and bends the end again. The middle finger is bent for 3. He bends the index and says, showing the thumb, 4; there are no more but this. Then he opens the hand and says 5 ; it is finished with my (or one, or the) hand. The Indian, holding his left hand stretched out, three fingers of it fastened together, separates the thumb and index, which he brings near to the thumb of the right hand and says 6; i.e. there are three on each side, three by three. He joins four fingers of the left hand, brings the left thumb near the thumb and index finger of the right hand, and says 7; (on one side there are four, or else, there are still three folded, or again, three on each side and the point in the middle). He brings the three fingers of the right hand to touch the left thumb, and thus obtaining two sections of four fingers, he says 8; (four on four, or four on each side). Then showing the little finger of the right hand, which alone remains folded, 204 HOW NATIVES THINK he says 9; (there is still one below, or—one is still wanting, or—the little finger is lying down). Finally, clapping the hands in joining them, the Indian says 10 : i.e. each side is finished or, it is counted, reckoned, it is a count. Then he begins the same manceuvre once more, saying: one filled plus one, one counted plus one, etc.” ! Thus the Déné-Dindjié, whilst using his fingers to count with, has no idea of a quinary basis. He does not say, as we often find in other cases, 6 is a second one; 7 is a second two; 8 is a second three, etc. On the contrary, he says: 6 is 3 + 3, coming back to the hand whose fingers he has exhausted, and separating them to add two to the thumb of the other hand. This proves that in counting 5, in “finishing a hand,’ he has not marked any time more strongly than in counting 4 or 6. Therefore in this case and in the many others which resemble it, it is not in the method of counting itself, it is not in the movements accom- plished that we find the principle of periodicity, that is, that | which will be the basis of the numerical system. A basis may be imposed for reasons which have absolutely nothing to do with convenience in reckoning, and the idea of the arithmetical use of numbers may not enter the question at all. Prelogical mentality is mystic, and oriented differ- ently from our own. Accordingly it is often indifferent to the most evident objective qualities, and contrariwise con- cerned with the mysterious and secret properties of all kinds ' of entities. For instance, it may happen that the basic 4, and the quaternary system of numeration, arise out of the fact of the four cardinal points and the four winds, of the |! four colours and four animals, etc., which participate in these four points, and play an important part in the collective representations of the peoples under consideration. Therefore we do not need to tax our psychological insight in speculating why this basis should have been chosen by men who never- theless reckoned with their five fingers. Where we find it used, it has not been selected from choice. It, as well as numbers, had a pre-existence, in that long period when they were as yet undifferentiated, and when the “‘ number-totals ” took the place of numeration proper. It is a mistake to picture 1 Petitot, Dictionnaire de la Langue Dènè-dindjié, p. lv. PRIMITIVES’ NUMERATION 205 the human mind making numbers for itself in order to count, for on the contrary men first of all counted, with much effort and toil, before they conceived of numbers as such. IV When the numbers have names given to them, and a group has a numerical system at command, it does not follow that the numbers are ipso facto conceived abstractly. On the contrary, they usually remain connected with the idea of the objects most frequently counted. The Yorubas, for instance, have a somewhat noteworthy numerical system, to judge by the use of subtraction in it. T1 112,113. 14, 15 = 10 1, +2, 49, 4) ts: 16, 17, 18, 19 = 20 — 4, — 3, — 2, —1; 70 = 20 X 4 — 10; 130 = 20 X 7 — 10, etc. This phenomenon is explained by the Yorubas’ constant use of cowrie shells as money, and these are always arranged in parcels of 5, 20, 200, etc. ‘‘ Numerals,” says the observer who has noted this fact, ‘‘convey to the Yoruba ear and mind two meanings: (x) the number, and (2) the thing the Yorubas especially count, and this is money (shells). . Other objects are only counted in comparison with an equal number of cowries, for a nation without literature and without | a school knows nothing of abstract numbers’?! This con-. clusion is equally true of all social groups of the same level | of development. The number, despite its being named; still adheres more or less to the concrete representation of a certain class of objects which are, par excellence, the objects counted, and other objects are counted only, as it were, by super- position on these. But while admitting that this adherence yields by degrees, and numbers unconsciously come to be represented for them- selves, this does not yet occur in a way that is abstract, and it is precisely because each has its name. With primitive * Mann, “ On the Numeral System of the Yoruba Nation,” J.A.I., xvi. Ol: 206 HOW NATIVES THINK peoples nothing, or at any rate, scarcely anything, is perceived in the way that seems natural to us. To their minds there is no physical phenomenon which is purely a phenomenon, no image which is nothing but an image, nor form that is wholly form. Everything perceived is compassed about by a complexity of collective representations in which the mystic elements predominate. In like manner, there is no name which is purely and simply a name; neither is there any numeral which is nothing but a numeral. Let us disregard the practical use the primitive makes of numbers when, for instance, he reckons what is due for so many hours’ work, or how many fish he has caught on a certain day. Every time he imagines a number gua number, he necessarily pictures it with the mystic property and value appertaining to that number, and to it alone, by virtue of participations which are equally mystic. The number and its name are indifferently the vehicle of these participations. Thus each number has its own individual physiognomy, a kind of mystic atmosphere, a “ field of action ” peculiar to itself. Every number, therefore, is imagined—we might also say, felt—especially for itself, and without comparison with the others. From this standpoint numbers do not constitute a homogeneous series, and they are accordingly quite unsuited to the simplest logical or mathematical operations. The mystic personality contained in each makes them unable to be added, subtracted, multiplied or divided. The only pro- cesses they admit of are themselves mystic processes and not, like arithmetical operations, subject to the law of contra- diction. In short, we might say that in the primitive’s mind from two standpoints, number is undifferentiated, to a varying extent. In its practical use it still more or less adheres to the objects counted. In the collective representations the number and its name still participate so closely in the mystic qualities of the ensembles represented that they are indeed mystic realities themselves, rather than arithmetical units. It is to be noted that the numbers which are thus enveloped in a mystic atmosphere hardly exceed the first ten. They are the only numbers known to primitive peoples, and to which they have given a name. In peoples who have risen to an abstract conception of number, the value and mystic power PRIMITIVES’ NUMERATION 207 of numbers may indeed be preserved for a very long time, when it is a case of those which formed part of the very earliest collective representations: but they are not extended to their multiples, nor as a rule to the higher numbers. The reason for this is evident. The earlier numbers, to about ten or twelve, which are familiar to the prelogical, mystic, mentality, participate in its nature, and become purely arithme- tical numbers at a very late epoch. Possibly there is not even yet any aggregate in which they bear that aspect alone, save to mathematicians. On the contrary, the higher numbers, very slightly differentiated in the primitive’s mind, have never, with their names, been merged in its collective repre- sentations. They at once became arithmetical numbers and, with some exceptions, they have been nothing else. This limits the extent to which I subscribe to the conclu- sions reached in Usener’s fine book entitled Dreiheit (or Trinity).t After having adduced exhaustive evidence estab- lishing the mystical nature of the number three, and the mysterious. value and power attributed to it from remote antiquity, Usener accounts for it by concluding, as Diels does, that this mystic character had its origin in times when human societies counted no further than three. Three, then, would have signified the ultimate number, the absolute total, and thus for a long period it would have held the place given to infinity in more advanced aggregates. It is un- doubtedly possible that with certain primitive races the number three may have enjoyed such prestige, but Usener’s inter- pretation cannot be accepted as entirely satisfactory. As a matter of fact, in the first place, we do not find that numeration actually stops short at three anywhere. Even in the social groups inhabiting Australia, Torres Straits, and New Guinea, who have no numerals above one, two and occasionally three, prelogical mentality has methods of its own which permit of counting more than this. Nowhere is three the “ ultimate ”’ number. Moreover, the list of numbers named or used never stops short at a certain number which is distinctly the last, and expresses totality. On the contrary, all the data collected, not only in the primitive races just mentioned, but in Melanesia and both the Americas, as well as among the Dravidians ™ Rheinisches Museum, N.F., lviii. pp. 1-48, 161-208, 324-364. 208 HOW NATIVES THINK of India, etc., prove that the number-series always terminates in a vague term meaning ‘“ many” or “ plenty,” or “ multi- tude,” which afterwards becomes a definite numeral—five, six, or whateverit may be. Finally, as Mauss rightly observes," if Usener’s theory were correct—-if, for many centuries, the human mind, stopping short at the number three, had impressed an almost indelible character of mysticism upon it, such a character would pertain to the number in all social groups everywhere. Now there is nothing of this sort with the peoples of North and Central America. The numbers four and five, and the multiples of these, are constantly met with in their group ideas, and the number three plays a very insignificant part, or none at all. These objections are not only effective against Usener’s theory, but at the same time they overthrow all similar attempts at interpretation. The otherwise extremely ingenious theory of MacGee,? for instance, which is based on the observa- tion of North American natives, cannot account for data which have been collected among other primitive peoples. — The common mistake of all such hypotheses is that of general- izing from a psychological process which their advocates believe they have analysed in this or that milieu, and which serves” to account for the mystic worth attributed to certain numbers in these social groups. Facts do not bear out this generaliza- tion, and this sort of ‘“‘ explanation’’ leads nowhere. Must we not rather regard the collective representations of primitive races as prelogical and mystic by virtue of their constitution and their mental solidarity, and is not this true of the numbers they imply as well as of their other content ? There is no number which possesses a name and appears in their repre- sentations which has not a mystic value. That being granted, why should it be here the three, or there the four, or elsewhere the two or seven or any other number, which assumes a special importance and has a wholly individual virtue? The reason must be sought, not in purely psychological motive, for these would apply to all human aggregates, whatever their nature, but in the conditions peculiar to the group or col- lection of groups under consideration. In this respect there 1 Année Sociologique, vol. vii. p. 310 (1904). 2 “ Primitive Numbers,” E. B. Rept., xix. pp. 821-51. PRIMITIVES’ NUMERATION 209 is nothing more instructive than the facts revealed by Dennett in his book entitled At the Back of the Black Man’s Mind. The classification of the various types of social groups is not yet advanced enough to provide the guiding thread we need here. But what we can even now establish is that there is no number among the first ten that does not possess supreme mystic importance for some social group or other. It is quite unnecessary to bring forward evidence of this, as far as the first three numbers are concerned. Even among the most advanced nations, there are traces of this mystic character still discernible both in religion and metaphysics. The “ unit ’’ has maintained a prestige upon which the mono- theistic religions and the monistic philosophies plume them- selves. ‘“‘ Duality” is often the antithesis of unity by qualities which are diametrically opposite, since it signifies, implies, and produces the exact contrary of that which unity signifies, implies and produces. Where unity is a principle of good, order, perfection and happiness, duality is a principle of evil, disorder, imperfection ; a sign, that is, a cause, of misfortune. Many languages still preserve in their vocabulary traces of this opposition ; and we speak of “a double life,” “duplicity,”” etc. I shall not lay further stress upon the mystic character of the number three: it is sufficient to remind you of Usener’s monograph on the subject, mentioned above. I will confine myself to recalling some facts relative to the number four and those which follow it. These facts, naturally, cannot be ascertained among the most primitive peoples we know, for such have not yet given names to the number four and those above it. The majority of the Indian groups of North America, however, attach to the number four a mystic virtue that surpasses that of any other. “ Amongst almost all the Red Indian tribes, four and its multiples had a sacred significance, having special reference to the cardinal points and to the winds which blow from them, the sign and symbol of this quadruple nature-worship being the Greek or equal-armed cross.”3 In the great Navajo epic, “‘ the gods are all four in number, and all range them- * London, 1906. > MacGee, “‘ Primitive Numbers,” E. B. Rept., xix. pp. 821-51. 3 Buckland, ‘ Four as a Sacred Number,” J.A.I., xxv. pp. 96-9. I4 210 HOW NATIVES THINK selves one at each cardinal point, being painted in the colour appropriate to that point. There are four bear gods, four porcupines, four squirrels, four long-bodied goddesses, four holy young men, four lightning birds, etc. The hero is allowed four days and four nights to tell his story, and four days are employed in his purification.” So, too, we constantly find evidence of the mystic functioning of the number four in the Zuñi myths which Cushing has so admirably annotated for us, as well as in their rites and customs, as described by Mrs. Stevenson. ‘‘ Choose then, four youths, so young that they have neither known nor sinned aught of the flesh... Them four ye shall accompany. . . . Ye shall walk about the shrine four times, once for each region and the breath and season thereof. ... They carried the painted arrows of destiny, like the regions of men, four in number.’’? Among the Sioux, ‘‘ Takuskanskan, the moving deity, is supposed to live in the four winds, and the four black spirits of night do his bidding. .. . The four winds are sent by the ‘something that moves.’’’2? Again, with them there are four thunder- beings, or at any rate, ‘‘ four varieties of their external mani- festation. In essence, however, they are but one.” (In this we recognize the effect of the law of participation.) ‘‘ One is black, another yellow, scarlet, blue, etc. They live at the end of the world upon a high mountain. The dwelling opens towards each of the four quarters of the earth, and at each doorway is stationed a sentinel: a butterfly at the entrance, a bear at the west, a deer north, and a beaver south.”’ 3 It is nowadays the fashion to give a psychological inter- pretation to facts of this kind, which are innumerable. An association between the number four and the cardinal points which are exactly four, the winds coming from these four regions, the four gods presiding over them, the four sacred animals which dwell there, and the four colours symbolizing them, is supposed to have been set up. The prelogical mind, however, was never conversant with these as isolated ideas. It did not first of all conceive of the north as a spatial region having the east on the right and the west on the left, and 1 F, H. Cushing, ‘‘ Zufii Creation Myths,”’ E. B. Rept., xiii. p. 442. 2 Dorsey, “ Siouan Cults,” E. B. Rept., xi. p. 446. 3 Dorsey, ibid., p. 442. PRIMITIVES’ NUMERATION 211 then combine with it the idea of a cold wind, snow, the bear, the colour blue. . . . All such ideas on the contrary were originally enveloped in a complex representation which is of a collective and religious nature, in which the mystic ele- ments conceal those which we should consider the real ones. As one such element we find the number four, the vehicle on mystic participation, playing in this way a very important part, and one which, though indispensable to the prelogical mind, is very difficult for logical thought to reconstruct. When mystic participations are no longer felt, they leave behind, as it were, a residuum composed of these associations which still obtain to some extent everywhere. They are no longer anything but associations, because the internal bond which held them together has disappeared: but they were originally something very different. Such, for instance, are the asso- ciative correlation between cardinal points, seasons, colours, etc., so frequently met with in China. De Groot gives us the following : East Spring Blue Dragon South Summer Red Bird West Autumn White Tiger North Winter Black Tortoise ! The mystic participation realized by means of the number four in the minds of the North American tribes, is brought * out in many instances. Catlin tells us that with the Mandans “there were also four articles of great veneration and im- portance lying on the floor of the lodge, which were sacks containing in each some three or four gallons of water . . objects of superstitious regard, made with great labour and much ingenuity; . . . sewed in the form of a large tortoise lying on its back, with a bunch of eagle’s quill appended to itasa tail. . . . These four sacks of water have the appearance of great antiquity, and by inquiring . . . the medicine-man very gravely told me that ‘ those four tortoises contained the waters from the four quarters of the world—that those waters had been contained therein since the settling down of the world,” an explanation which amused Catlin very much. He tells us, too, that the buffalo dance (intended to oblige 1 J. J. M. de Groot, The Religious System of China, i. p. 317. 212 HOW NATIVES THINK the buffaloes to come near the hunters) is repeated four times during the first day, eight on the second day, twelve on the third and sixteen on the fourth day, and that the dance is given once to each of the cardinal points, and the medicine-man smokes his pipe in those directions. On the second day it is given twice to each, three times on the third day, and four times on the fourth.! We note the same mystic character attributed to the number four in the magic formulas of the Cherokees. Mooney lays considerable stress upon this. ‘The Indian,” he says, ‘has always four as the principal sacred number, with usually another only slightly subordinated. The two sacred numbers of the Cherokees are four and seven. . . . The sacred four has direct relation to the four cardinal points, while seven, besides these, includes also ‘above,’ ‘ below,’ and ‘here, in the centre.’ In many tribal rituals, colour and sometimes sex are assigned to each point of direction. In the sacred Cherokee formulas the spirits of the East, South, West and North are respectively Red, White, Black and Blue, and its colour has also its symbolic meaning of Power (War), Peace, Death and Defeat.’”’2 Mooney speaks too of “the veneration which their physicians have for the numbers four and seven. They say that after man was placed upon the earth four and seven nights were instituted for the cure of diseases in the human body. . . .” 3 In British Columbia, with the StatlumH tribe, four is par excellence the sacred number. After birth “ the mother and child remained in the lodge for at least four days, and if the weather permitted, this period would be extended to eight or twelve or twenty days, or to some other multiple of four, the Salish mystic number.” 4 In Vancouver, during the initiation ceremonies of the medicine-man, ‘‘ when he rises, he must turn round four times, turning to the left. Then he must put forward his foot four times before actually making a step. In the same way, he has to make four steps before t The North American Indians, i. pp. 185-6. a ‘‘ Myths of the Cherokee,” E. B. Rept., xix. P. 431. 3 Heywood, quoted by J. Mooney, “* The Sacred Formulas of the Chero- kee,” E. B. Rept., vii. p. 322. 4 Hill Tout, ‘‘ The Ethnology of the StatlumH of British Columbia,” J.A.I.,, XXXV. p. 140. PRIMITIVES’ NUMERATION 213 going out of the door. . . . He must use a kettle, dish, spoon and cup of his own, which are thrown away at the end of four months. . . . He must not take more than four mouthfuls at one time, etc.’ ! This same number four appears to form the basis of the complicated and obscure mysticism of numbers which became manifest in the southern and western parts of North America, and in Central America. ‘The nine days of ceremony... . have a nomenclature suggestive of divisions into two groups of four each. . . . On this basis it will be seen that the number four, so constant in Pueblo ritual, is prominent in the number of days in the Snake ceremonial. I will call attention also to the fact that the nine days of ceremonies plus the four days of frolic make the mystic number thirteen. It may likewise be borne in mind that the period of twenty days, the theoretical length of the mo:t elaborate Tusayan ceremony, was also characteristic of other more cultured peoples in Mexico, and that thirteen ceremonials, each twenty days long, make a year of 260 days, a ceremonial epoch of the Maya and related peoples.””2 I shall not enter into a discussion of this com- plicated question, but am content to have pointed out the significant place occupied by the number four, as we find it again in the Agrarian rites of the Cherokees.3 Lastly, I shall quote a remark made by Hewitt, apropos of an Iroquois myth, in which four children—two boys and two girls—are mentioned. ‘The use of the number four is here remarkable. It seems that the two female children are introduced merely to retain the number four, since they do not take any part in the events of the legend.”’ 4 The mystic number thus assumes the aspect of a category in which the content of the collective representations must be arranged. This is a feature which is found, well marked, in the Far East. ‘European languages,” says Chamberlain, “have such expressions as ‘the four cardinal virtues’ or ‘the seven deadly sins’; but it is no part of our mental dis- position to divide up and parcel out almost all things visible * F. Boas, “‘ The North-west Tribes of Canada,” Reports of the British Association, p. 618 (1890). 2 Fewkes, “ Tusayan Snake Ceremonies,’ E. B. Rept., xvi. p. 275. 3 J. Mooney, “ Myths of the Cherokee,” E. B. Rept., xix. p. 423. + Hewitt, “ Iroquoian Cosmology,” E. B. Rept., xxi. p. 233 (note). 214 HOW NATIVES THINK and invisible into numerical categories fixed by unchanging custom, as is the case among the natives from India east- wards.” In North America this “‘ category’’ seems very closely bound up with the cardinal points or the spatial regions. We must not imagine, however, that prelogical mentality imagines these points or regions in any abstract way, and that it detaches the number four from its idea of these, for its mystic purposes. In this, as in all else, such a mentality obeys the law of participation ; it imagines spatial directions, cardinal points and their number only in a mystic complex to which the number four owes its categorical character, which is not logical, but mystic. ‘The breath-clouds of the gods are tinted with the yellow of the North, the blue-green of the West, the red of South, and the silver of the East.” 2 This complex naturally contains elements which are of social origin. The division of space into regions corresponds with the division of the tribe into groups. Durkheim and Mauss opine that the latter determine the former, and that it is the general principle of what they call classifications.3 They quote facts observed in Australia especially, but also in China, and from the Pueblos of North America, particularly the Zuñis. I have already laid stress upon that which Spencer and Gillen term ‘“ local relationship,” kinship by community of position, and participation between a group and a given region. When a tribe stays in a place, for instance, whether its stay be provisional or permanent, the different clans or totems do not take up their positions at their own discretion. Each has its predetermined site, settled by virtue of the mystic connection between the clans or totems and the points of the compass. We have found that facts of the same kind _ were noted in North America, and other observations made betoken this mystic connection. The Kansa, for instance, as Dorsey tells us, were accustomed to cut out the heart of an enemy they had killed, and throw it on the fire as a sacrifice to the four winds. The Yata men, i.e. those who camped on the left side of the tribal circle, would raise their left hands, and bow to the east, south, west, and north winds succes- 1 Things Japanese, pp. 353-4. 2 Stevenson, ‘‘ The Zufiis,”’ E. B. Rept, xxiii. p. 23. 3 Année Soctologique, vi. pp. I-72. PRIMITIVES’ NUMERATION 215 sively.t The ritual order is regulated by the mystic tie which unites the clans with the position in space which they occupy, and Dorsey further says that every time the Osage and Kansa tribes permanently established themselves in a village, there was a certain consecration of the dwellings, before the people could install their belongings, and this was associated with the cult of the four winds. ‘The symbol of the earth, U-ma-ne, in Dakota, has never been absent from any religious exercise I have yet seen or learned of from the Indians. It is a mellowed earth space, and represents the unappropriated life or power of the earth. . . . The square or oblong, with the four lines standing out, is invariably interpreted to mean the earth or land with the four winds standing towards it. The cross, whether diagonal or upright, always symbolizes the four winds or four quarters.”’ 2 The numbers five, six and seven also sometimes possess a sacred character among the North American tribes, although not so constantly so as the number four. Gatschet writes, for instance: “ Here we have again the sacred number five occurring so often in the traditions, myths and customs of the Oregonian tribes.” 3 ‘‘ Many of the deified animals appear collectively, as five to ten brothers, or five sisters, sometimes with their old parents.” 4 The number of cardinal points or spatial regions is not necessarily four ; with the North American tribes it is sometimes five (reckoning the zenith), six (then adding the nadir), and lastly seven, terminating in the centre or the place occupied by the one counting. With the Mandans, for instance, the medicine-man “took the pipe, and after presenting the stem to the North, to the South, to the East and the West—and then to the sun that was over his head . . 5—With the Sia, “the priest, standing before the altar shook his rattle for a moment, and then waved it in a circle over the altar. He repeated this motion six times, for the cardinal points. . . . The circle indicated that all the cloud people of the world were invoked to water the earth. . . t Dorsey, ‘‘ Siouan Cults,’’ E. B. Rept., xi. p. 380. 2 Tbid.,) p. 451. 3 The Klamath Language, p. 86. 4 Ibid., p. ror. § Catlin, The North American Indians, i. p. 258. 216 HOW NATIVES THINK This sprinkling of the cardinal points was repeated four times.”’ ! —‘‘ The Omahaand Ponka used to hold the pipe in six directions while smoking towards the four winds, the ground, and the upper world.” »—‘‘ The snake chief made a circle of sacred meal about twenty feet in diameter . . . and drew in it six meal radii corresponding to the six cardinal points.’ 3 Finally, among the Cherokees, the sacred number four signifies the cardinal points, and the sacred seven signifies them also, by adding the zenith, nadir and centre.4 We find these numbers five, six and seven, involved in the same complex mystic participations as the number four. Among the Zuñis, Mrs. Stevenson has collected several examples relating to six. To quote but one of them, “ these primitive agriculturists have observed the greatest care in developing colour in corn and beans to harmonize with the six regions : yellow North, blue West, red South, white East, variegated zenith, and black nadir.’’ 5 Similar phenomena are to be found throughout the Far East, to say nothing of the Indo-European and Semitic peoples. In China the complexity of correspondences and participations involving numbers is infinite. They intersect and even con- tradict each other without the Chinese sense of logic being at all disturbed. In Java the native week lasts five days, and the Javanese believe that the names of these days bear a mystical relation to colours and to the divisions of the horizon. ‘The 1 means white and East ; the 2, red and South ; the 3, yellow and West; the 4, black and North; the 5, mixed colour and forms the centre. . . . In an ancient manu- script found in Java, the week of five days is represented by five human figures, two female and three male.” 6 In India the number five is lucky or formidable, according to the district, or according to the special participation concerned. “In 1817 a terrible epidemic of cholera broke out at Jessore. The disease commenced its ravages in August, and it was at once discovered that the August of this year had five Saturdays s Stevenson, “ The Sia,” E. B. Repi., xi. pp. 79, 93. 2 Dorsey, ‘‘ Siouan Cults,’’ E. B. Rept., xi. p. 375. 3 Fewkes, ‘‘ Tusayan Snake Ceremonies,” E. B. Rept., xvi. pp. 285, 295. ¢ J. Mooney, ‘‘ Myths of the Cherokee,” E. B. Rept., xix. p. 431. 5 Stevenson, “ The Zufiis,”’’ E. B. Rept., xxiii. p. 350. 6 Skeat, Malay Magic, p. 545 (note). PRIMITIVES’ NUMERATION 217 (a day under the influence of the ill-omened Sani). The number five being the express property of the destructive Siva, a mystical connection was at once detected, the infallibly baleful influence of which it would have been sacrilege to question.” In another place the number five possesses favourable powers. ‘‘ The peasant digs up five clods of earth with his spade. This is a lucky number, as it is a quarter more than four. . .. He then sprinkles water five times into the trench with the branch of the sacred mango... . Then a selected man ploughs five furrows. . . . In Mirzapur, only the northern part of the field, that facing the Himalaya, is dug up in five places with a piece of mango wood.” 2 Agra- rian rites and practices of this nature occur very frequently. There are peculiarly important mystic virtues attaching to the number seven, especially in places where the in- fluence of Chinese or Assyro-Babylonian beliefs obtains.3 In Malaya, ‘every man is supposed . . . to possess seven souls in all, or perhaps, I should more accurately say, a sevenfold soul. This ‘septenity in unity’ may perhaps be held to explain the remarkable importance and persistence of the number seven in Malay magic (seven twigs of the birch, seven repetitions of the charm in soul abduction, seven betel leaves, seven blows administered to the soul, seven ears cut for the Rice soul in reaping).’”’4 The animism which inspires Skeat’s work evidently suggests this theory: I incline to think that it presents the matter the wrong way round. It is not because they conceive of seven souls or a sevenfold soul for every one that the Malays use seven everywhere. It is on the contrary because the number seven possesses magic virtues on their eyes that it becomes a kind of “category,” upon which not only their magic practices, but also their ideas, not excepting their conception of the soul, are regulated. This is so true that Skeat himself adds’: ‘‘ What these seven souls were it is impossible without more evidence to determine.”’ If each of the seven souls is so little differentiated that we may * Crooke, The Folklore of Northern India,” i. p. 130. a Ibid., ii. p. 288. 3 Vide Von Adrian, ‘‘ Die Siebenzahl im Geistesleben der Volker,’’ Mitteil- ungen, pp. 225-271 (Vienna, 1901).; W. H. Roscher, ‘ Die Siebenzahl,”’ Philologus, pp. 360-74 (1901). 4 Skeat, Malay Magic, pp. 50, 509. 218 HOW NATIVES THINK speak equally well of one sevenfold soul, it is difficult to admit that the value attached to the number seven in general has its origin in this idea. ‘When Hindus have removed the ashes from a burning- ground they write the figures 49 on the spot where the corpse — was cremated. The Pandits explain this by saying that when written in Hindi the figures resemble the conch-shell or wheel of Vishnu ; or that it is an invocation to the 49 winds of heaven to come and purify the ground. It is more probably based on the idea that the number seven, as is the case all over the world, has some mystic application.” —“‘ In India, the water of seven wells is collected on the night . . . of the feast of lamps, and barren women bathe in it as a means of procuring children. . . . Hydrophobia, all over Northern India, is cured by looking down seven wells in succession.’’ 2—‘‘ The goddess of smallpox, Sitala, is only the eldest of a band of seven sisters by whom the pustular group of diseases is supposed to be caused. . . . Similarly in the older Indian mythology we have the seven mâtris, the seven oceans, the seven Rishis, the seven Adityas, and Danavas, and the seven horses of the sun, and numerous other combinations of this mystic num- ber.’’ 3—In Japan, ‘‘ seven, and all the numbers into which seven enters (seventeen, twenty-seven, etc.), are unlucky.” 4 It is the same with the Assyro-Babylonian people, who consider the seventh, fourteenth, twenty-first and twenty-eighth days ill-omened.s—Among the Hindus, medical prescriptions, like all magic formulas in general, attach the greatest import- ance to numbers on account of their mystic virtues. For example, ‘one favourite talisman is the magic square, which consists in an arrangement of certain numbers in a special way. In order to cure barrenness, for instance, it is a good plan to write a series of numbers which added up make 73 both ways, on a piece of bread, and with it feed a black dog. . . . To cure a tumour a figure in the form of a cross is drawn, with three cyphers in the centre and one at each of the four ends. This is prepared on a Sunday and tied round the left arm. The number of these charms is t Crooke, The Folklore of Northern India, ii. p. 51. 2 Ibid., i. pp. 50-1. 3 Ibid., 1. ‘p.' 2318, 4 Chamberlain, Things Japanese, p. 439. 5 Jastrow, The Religion of Babylonia and Assyria, p. 377. PRIMITIVES’ NUMERATION 219 legion,’ * adds Crooke; but it is not only in India that this is so. An infinite number of similar ones can be found in the magic and medicine of antiquity in Arabia and in the Middle Ages, in Europe and among all peoples who have numerals at command. The study of folk-lore affords abun- dant evidence of this. Among more developed peoples, with whom large numbers are of current usage, certain multiples of numbers of mystic value participate in their peculiar properties. In India, for instance, ‘ when the new moon falls on Monday, pious Hindus walk one hundred and eight times round it (the fig-tree).”’ 2 Possibly 108 possesses special virtue as a common multiple of 9 and 12, themselves multiples of 3 and 6. In the North- west Provinces, the numbers 84 and 360 are of extraordinary importance. Chaurasi (84), for instance, is the subdivision of a parganah or district, amounting to 84 villages. ‘‘ But it is not with respect to the occupation of land only that the numbers 84 and 360 are regarded with such favour. We find them entering into the whole scheme of the Hindi, Buddhist and Jain religions, cosmogonies, rituals, and legendary tales ; so much so, as to show that they are not taken by mere chance, as arbitrary numbers to fill up some of their extravagant fictions, but with a designed purpose of veiling a remote allusion under a type of ordinary character.” 3 The use of such mystic numbers is more systematic with the Buddhists than with the Hindus. May not this arise out of the fact that 84 is a multiple nt both 7 and 12; and 360 a multiple of 4,6,9,5 and12? There would thus be a combination in 84 and 360, in which all the properties of the respective numbers would participate. On several occasions Bergaigne has laid stress upon the nature of the mystic numbers in Vedic poetry, and the mystic processes applied to these. Multiplication seems to be effected chiefly by applying to the different parts of a whole a system of division first applied to itself. For instance, the division of the universe into three—heaven, earth, and atmosphere—may be repeated for each of these three— t Crooke, op. cit., i. pp. 159 —60. 2 Ibid., p. 100. 3 Elliot, Memoirs. of the Races in the North- west PAN of India, ii. pp. 47 et seq. 220 HOW NATIVES THINK three heavens, three earths, and three atmospheres—that is, nine worlds in all. Then, too, various systems of division having been applied to the universe, the figures given by two of them may also be multiplied together, thus: 3 x 2 = 6 worlds, three heavens and three earths.t Or again, to form a new magic number, we can add the unit to a given mystic number: 3 +1, 6 + 1, 9 + 1, etc. ‘‘ The usual object of this is to introduce into any system of the universe whatever the idea of an invisible world, or into any group whatever, of persons or things the idea of a person or thing of the same kind, but distinguished from the rest by a sort of mystery which surrounds him or it.”’2 The number 7, for instance, may possibly have independent mythological value; but it is certain that the Rishis have at least divided it into 6 + x, the addition of the unit to the number of the six worlds. These mythological numbers derive their virtue from their mystic relation to the spatial regions: the septenary division of the universe, for example, (seven worlds, i.e. 6 + 1) coincides with the mythological heptads, the seven places, races, ocean ~ depths, rivers, and so forth. That prelogical mentality is at work in these already systematized collective representations can be proved by the way in which the one and the many are identified. It is thus, says Bergaigne, that “‘ most groups of mythological beings or objects may be reduced to a single being or object with many forms, which sums up the group as a whole. The elements of each group are in this way revived as so many manifesta- tions of a single principle; and the multiplicity of these manifestations is accounted for by the multiplicity of the worlds. . . . The seven prayers are only the seven forms of the prayer which, considered both in its unity and in its different manifestations, becomes the prayer or hymn with seven heads. . . . The seven cows of the master of the prayer are naturally the seven prayers issuing from his seven mouths. . . . A male has two or three mothers, two or three wives, Ber 8 This leads to a conclusion which at first appears extra- ordinary ; different numbers are nevertheless equal numbers. 1 Bergaigne, La Religion Védique, ii. p. 115. 4 Thid., ti. pp: 123 et seq 3 Ibid., ii. pp. 147-8. PRIMITIVES’ NUMERATION 221 ‘“ The simultaneous and impartial use of three and seven . . . proves but one thing: their complete equivalence... . The various numbers we have found used for one another, because they all express, in different systems of division, the sum of the parts of the universe, have for the same reason been capable of being used, by a kind of pleonasm, side by side with each other. In fact, this has frequently been done. In this way, three is the same as seven, or as nine.” This equivalence, an absurdity to logical thought, seems quite natural to prelogical mentality, for the latter, preoccupied with the mystic participation, does not regard these numbers in an abstract relation to other numbers, or with respect to the arithmetical law in which they originate. The primitive mind considers each as a reality grasped by itself, and not needing for its definition to be regarded as a functioning of other numbers. Thus every number has an inviolate indi- viduality which allows it to correspond exactly with another number, itself equally inviolate. ‘‘ Most of the mythological numbers of the Rig-Veda, especially 2, 3, 5 and 7, express, not merely an indefinite plurality, but a totality, and this totality answers in principle to the ensemble of the worlds.” 1 Let it be, for instance, the mythical bull, having “‘ four horns, three feet, two heads, seven hands: bound with a threefold cord, the bull bellows, etc.” (Here we have two, three, seven, worlds, and four cardinal points.) The different characteristics in the description all indicate, by their allusion to the different divisional systems of the universe, that the personage in question is present everywhere.? We know from other sources that the idea of omnipresence, or of ‘‘ multipresence,”’ accord- ing to Leibnitz’ expression, is absolutely familiar to prelogical, mystical mentality. Finally, to complete the delineation of these mystic numbers, Bergaigne says further: .“‘ The numbers three and seven, in the general system of Vedic mythology, should be regarded as frameworks prepared beforehand, independently of the personalities which may be summoned to occupy them.’"’3 Frameworks prepared beforehand: that is to say, categories, according to the expression used by Chamberlain, * Bergaigne, La Religion Védique, ii. P. 156. TID Pu pie kar, 3 Ibid., iii. p. gg. 222 HOW NATIVES THINK quoted above, and dealing with this precise subject. There is no better way of emphasizing the difference between these mystic numbers and those which serve the purpose of arith- metical calculation. Instead of the number depending on the actual plurality of the objects perceived or pictured, it is on the contrary the objects whose plurality is defined by receiving its form from a mystic number decided upon beforehand. Thus the properties of numbers predetermine, as it were, what the multiplicity will be in the collective representations. How does it happen, we may ask, that the mystic nature of numbers is not most clearly manifest where the representa- tions are themselves most profoundly mystic, that is, in the peoples of the most primitive type familiar to us? How is it that this mystic character seems, on the contrary, more strongly marked in cases where logical thinking is already somewhat developed, and knows how to use numbers in a really arithmetical method—in the races of North America : and the Far East, for instance, whilst it is not noticed in the Australian aboriginals, or in the South American or Indian primitives? It may seem as if our theory does not take all the facts of the case into account, and that if we are to explain the mystic virtues attributed to numbers, we must have recourse to other principles than the participations of which these numbers are the vehicles in the collective representations. This objection may be answered in the following ways : (1) Among peoples still wholly primitive, numbers (above two or three) are as yet undifferentiated, and consequently they do not figure as actual numbers in the collective repre- sentations. As they have not been the object of an abstraction, not even of that isolating but not generalizing abstraction peculiar to prelogical mentality, they are never imagined per se. And above all, having no names, they can never act as condensers of mystic virtues, a part attributed to them in the collective representations of more advanced peoples. (2) Above all, however, it is possibly in this undifferentiated and unnamed state that the mystic efficacy of the number is greatest. The divisions of the social group into totems, clans, phratries, which are themselves subdivided, although not expressed numerically, nevertheless comprehend definite PRIMITIVES’ NUMERATION 223 numbers ; and have we not found that these divisions and their numbers extend to all reality represented, to animals, plants, inanimate objects, stars, spatial directions? Institu- tions, beliefs, religious and magical practices—-do not all these constantly imply, through these same divisions and “ Classifications,” the numbers which are comprised therein without being expressed? Yet it is precisely because the mentality which is mystic and prelogical moves thus in an element natural to it, that we find it so difficult to reconstruct it. Whatever the effort we may make, a number which is purely intricate, undifferentiated, felt and not concelved, is unimaginable to us. A number is not a number to us unless we imagine it and as soon as we picture it we imagine it logically and with a name. There is no doubt that, once named, we can very well conceive of it, either from the abstract point of view, without any qualifying attribute, and absolutely homogeneous with other numbers, or as a sacred vehicle of mystic qualities. Our religions, and sometimes our meta- physics, still tell us of such numbers, and our myths, legends, and folklore have familiarized us with them. But it is far more difficult to go back to a number which has no name, and to discover the function it fulfils in the collective representa- tions of primitive peoples. à A hy Le Ya ? 18 : À Meur ANT arr eo ; RU TE AR ANNE AT RCE. FAUNE NU : : j ‘ |: FE (RAR CT NT Fi CHU OU PART “Tit 15 CHAPTER VI INSTITUTIONS IN WHICH COLLECTIVE REPRE- SENTATIONS GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (I) INTRODUCTION Our study of the collective representations and their inter- relations among primitives has led us to conclude that these people possess a mystic, prelogical mentality differing in many essentials from our logical thinking. This conclusion, moreover, seems to have been confirmed by our examination of certain characteristics in the languages spoken by primi- tives and the system of numeration used. A counter-proof, however, is necessary. We therefore have to show that primi- tives’ ways of acting do indeed correspond with their ways of thinking, as analysed in this volume, and that their collective representations express themselves in their institutions with the mystic, prelogical characteristics already noted. Such a demonstration would not only yield a valuable verification of my theory, but would lead the way to a better interpre- tation of these institutions than any of the psychological and merely probable “explanations’”’ so often urged, can afford, for this interpretation must above all take into account the mentality that is pos to the social groups under consideration. By way of example I have selected a certain number of institutions, preferring either the simplest or else those which, as far as our knowledge goes at present, seem best suited to show the nature of prelogical mentality most clearly. In no case, however, have I ventured to suggest an “‘ explana- tion,” nor to reduce these examples to a general principle which would account equally well for all. To ‘“ explain” 228 HOW NATIVES THINK these customs or institutions properly would require a detailed monograph in each case, and I need hardly say that not even the first faint outline will be found here. My object is some- thing quite different, and its scope is much more general. I only desire to show that if we are to understand these insti- tutions and these practices we must refer them to that prelogical, mystic mentality, the chief laws of which I have endeavoured to determine, the mentality which is peculiar to primitives. Admitting these laws to have been established, the savant has still to inquire into the conditions in which each of these special institutions and practices has made its appearance and maintained its existence in any given com- munity, but at any rate he will henceforth have at his disposal, to minimize the risk of his going astray, a clue which only too often was lacking to his predecessors I Let us first of all consider the operations by means of which the social group procures its sustenance—hunting and fishing, more particularly. Here, success is dependent upon a number of objective conditions: the presence of the game or the fish in a certain place, the necessary precautions in approaching it unperceived, the traps set to entice it to the desired spot, the projectiles required to bring it down and so forth. To the mind of the primitive, however, these con- ditions, although necessary, are by no means sufficient, and unless others be fulfilled, all the means employed will fail, whatever the skill of the hunter or fisherman. These means must possess magic virtue, must be endowed, so to speak, through the performance of special rites, with mystic power, just as the objective elements which are perceived are sur- rounded by a mystic complex. Without such magic perform- ance, the most experienced hunter or fisherman will fail to find game or fish, or if he does perceive it, it will escape the snare or the hook. Or again, his bow or his gun will miscarry. He may even attack his prey, and prove it invul- nerable ; or, having wounded it he may fail to findit. Thus the mystic operations are not mere preliminaries of the chase or the fishing, like St. Hubert’s Mass; for example, when the PRIMITIVES’ PRACTICES 229 effectual pursuit of game or fish remains the essential feature of the occasion. On the contrary, this effectual pursuit is not the most important factor to the primitive mind; the really essential part is the mystic process which alone can bring about the he presence, and secure the capture, of the prey. If such a process does not take place, there is no object in making any effort in the matter. Processes of such a kind are manifold and often complex. For the sake of clearness I shall distinguish between those which must be carried out before, during, and after the chase ; and also those exercised upon the agent, (or the members of his group) to ensure his striking a successful blow, and those which have his prey as their aim, either to render it unable to escape or to defend itself, or else to pacify it and obtain pardon for its death. Accordingly we shall see action domin- ated by an ensemble of definite mystic relations, dependent upon the collective representations of the social group and, like these collective representations themselves, governed by the law of participation. A. In hunting, the very first essential is to exert a magic influence upon the game pursued, to secure its presence, willingly or unwillingly, and to constrain it to come from a distance if need be. With most primitive people this process is considered indispensable. It consists mainly of dances, incantations and fasts. Catlin has given a detailed descrip- tion of the buffalo dance, ‘‘ held for the purpose of making ‘buffalo come’, as they term it... . About ten or fifteen Mandans at a time join in the dance, each one with the skin of the buffalo’s head (or mask) with the horns on, and in his hand his favourite bow or lance, with which he is used to slay the buffalo... . These dances have sometimes been con- tinued for two or three weeks without stopping an instant, until the joyful moment when buffaloes make their appear- ance.’’ (They represent the capture and the killing of the buffalo.) ‘‘ When an Indian becomes fatigued of the exercise, he signifies it by bending quite forward, and sinking his body towards the ground; when another draws a bow upon him and hits him with a blunt arrow, and he falls like a buffalo, is seized by the bystanders, who drag him out of the ring by the heels, brandishing their knives about him; and having dérotssiiiet 230 HOW NATIVES THINK gone through the motions of skinning and cutting him up, they let him off, and his place is at once supplied by another, who dances into the ring with his mask on; and by this taking of places, the scene is easily kept up night and day, until the desired effect, that of making ‘ buffalo come,’ has been produced.”’ ! It is a kind of drama, or rather pantomime, representing the prey and the fate that awaits him at the hands of the Indians. Since the primitive mind does not recognize an image, pure and simple ;—since to him the image participates of the original, and the original of the image, to be conversant with the image already, to some extent, ensures the possession of the original. This mystic participation constitutes the virtue of the operation. In other places, this participation assumes a slightly different form. To secure the presence of the animal, it is essential to conciliate it. With the Sioux, for instance, ‘the bear-dance . . . is given several days in succession, previous to their starting out, and . . . they all join in a song to the Bear Spirit which, they think, holds somewhere an invisible existence, and must be consulted and conciliated before they can enter upon their excursion with any prospect of success. . . One of the chief medicine-men placed over his body the entire skin of a bear. . . . Many others in the dance wore masks on their faces, made of the skin from the bear’s head ; and all, with the motions of their hands, closely imitated the movements of that animal, some representing its motion in running, and others the peculiar attitude, and hanging of the paws, when it is sitting up on its hind feet, and looking for the approach of an enemy.” ? Occasionally, the operations have undergone a kind of simplification of an abstract kind, which nevertheless allows their real nature to be perceived. ‘In order to cause the deer to move towards the locality where they may be desired, the shaman will erect, on a pole placed in a favourable position, an image of some famous — hunter and conjurer. The image will represent the power of the person as conjurer, and the various paraphernalia attached to the image assist in controlling the movements 1 Catlin, The North American Indians, i. pp. 127-8. 2 Ibid., 1. pp. 245-6. PRIMITIVES’ PRACTICES 231 of the animals.’ :—In West Africa, Nassau tells us, the native appeals to his fetich in hunting, warfare, trading, love-making, fishing, tree-planting, and starting upon a journey. ‘The hunter or hunters start out, each with his own fetich hanging from his belt or suspended from his shoulder . . . or if the hunt includes several persons, a temporary charm may be performed by the witch-doctor, or even by the hunters themselves. Such ceremonies preliminary to the chase are described by W. H. Brown as performed in Mashonaland,” ? and their object appears to be that of obliging the animals to appear. Conversely, certain acts are forbidden, because they would produce the contrary effect. ‘To name even the word deer when searching for one is ‘mali’ or ‘ tabooed,’ and now they thought it was useless my going to look for them any more. . . . They listen to omens religiously whenever on a hunting or fishing expedition, and never name the animal for fear that the spirits should carry information to the object of pursuit.” 3 In British Columbia, “ should a pubescent girl eat fresh meat, it was believed her father’s luck as a hunter would be spoiled thereafter. The animals would not permit him to kill them; for it was held that no animal could be killed against its own wish and will. Indeed the Indian looked upon all his food, animal and vegetable, as gifts volun- tarily bestowed upon him by the ‘spirit’ of the animal or vegetable, and regarded himself as absolutely dependent upon their good will for his daily sustenance.’’ 4 B. The magic operations to which the hunter himself must submit aim at securing for him mystic power over the game. These are often lengthy and complicated. In Canada, “the hunters observe a week’s fast, during which time they may not drink a drop of water . . . they do not cease their incantations while the daylight lasts . . . many of them cut themselves in various parts of the body . . . and all this is done to induce the spirits to reveal where bears are to be found in large numbers. . . . With the same end in view, they offer prayers to the nae of the beasts they have slaughtered in previous hunts. . . . Before they set out, all, t Turner, “‘ The Hudson Bay Eskimo,” Æ. B. Rept., xi. pp. 196-7. 2 Nassau, Fetichism in West Africa, p. 173. 3 Brooke, Ten Years in Sarawak, ii. pp. 90-1. 4 Hill Tout, “The Ethnology of the StatlumH,”’ J.Ad., xxxv. p. 136. 232 HOW NATIVES THINK or at any rate, most of them, must have seen bears in the same district in their dreams. . . . Later on, they bathe, whatever the weather may be, and a feast is given by the head hunter, who eats nothing himself, but recounts his hunting exploits. There are renewed invocations to the spirits of the dead bears, and then they begin their march, besmeared with black, equipped as for war, and acclaimed by the whole village.” t So too, Nicolas Perrot relates: ‘‘ This festival is always preceded by a week’s fast, neither food nor drink being allowed, so that the bear may be favourable to him and to. his party, and this means that he desires to find and kill bears without any disastrous consequences to him or any of his people. . . . When the day of departure arrives, he calls them all together, and they blacken their faces with charcoal as he does ; acy. all fast until evening, and then eat but very little.” \lt is qu universal among primitive peoples that the hunter, about to start upon an expedition, should abstain from sexual relations, take careful note of his dreams, cleanse his person, fast, or at any rate, eat certain foods only, paint and adorn himself in a special manner; and all of these are practices which will have mystic effects upon the game he desires to capture. Boas tells us that “the mountain-goat hunter fasts and bathes for several nights. Then early in the morning he paints his chin with red paint, and draws a red line over his forehead down to the point of his nose. Two tail feathers of the eagle are fastened to his hair. These ornaments are believed to enable him to climb well. The elk hunter adorns his hair with coal, red paint, and eagle down, etc.” 3 The special aim of such adornment is to gain the animal’s favour. A bear-hunter assured Boas that he him- self had received instructions from a bear regarding what he was to wear. Mooney describes the magic preparations for the hunt made by the Cherokees, and in explaining them he reproduces the formulas which reveal their signification. He tells us that t Charlevoix, Journal d'un Voyage dans l Amérique Septentrionale, iii, pp. T15~16 (1721). 2 Mémoire sur les Meurs . de l'Amérique Septentrionale, pp. 66-7. 3 F. Boas, ‘ The North- west Tribes of Canada,’ Reports of the British Association, p. 460 (1894). PRIMITIVES’ PRACTICES 233 the night before the departure the hunter “ goes to the water ”’ and recites the appropriate formula. In the morning he leaves without breaking his fast, and he neither eats nor drinks during the day’s march. At sunset he again goes to the water and repeats his incantation, and then he encamps, lights his fire, eats his supper and, after having rubbed his breast with ashes, he lies down for the night. Next morning his quest begins . . . and in all hunting expeditions it is the rule, and almost a religious injunction, to abstain from food until sunset. . . . The hunter addresses his supplications to the fire whence he draws his omens, to the reeds of which he makes his arrows, to Tsu’lkala the great lord of the game, and finally he addresses in songs the very animals he is seeking to slay.? In British Guiana, ‘‘ before an Indian sets out to hunt, he goes through one or more strange performances to ensure success. Round his house he has planted various sorts of ‘beenas’ or plants, generally caladiums, which he supposes to act as charms to make the capture of game certain. These are for his dogs, which are made to swallow pieces of the roots and leaves. Sometimes the poor brutes have to undergo more painful operations. . . . The hunter inflicts tortures on himself . . . he submits to the bite of ants of a large and venomous kind . . . and rubs himself with caterpillars which irritate his skin, etc.2 Finally, the weapons and implements of the chase must also be endowed with special virtue by means of magical operations. To give but one example only, —it is customary among the Fang in West Africa to make, previous to the expedition, biang nzali, which is a charm for the guns, and to place the weapons upon it: this will enable them to go direct to their mark.”’ 3 C. Let us now assume that'all the operations have attained their end and the prey is in sight : will it now suffice to attack and strike it down? By no means; here again, everything depends upon the practice of magic. For instance, with the Sioux, “on coming in sight of the herd, the hunters talk kindly to their horses, applying to them the endearing t J. Mooney, “The Sacred Formulas of the Cherokee, IY pa 8 Pall oe. 2) vii. EP 370, 372, 342 2 Im Thurm, Among the Indians of Guiana, pp. 228-31, 3 Bennett, ‘“‘ Ethnographical Notes on the Fang,” /.4.1., xxix. p. 94. 234 HOW NATIVES THINK names of father, brother, uncle, etc. . . . The party having approached near to the herd, they halt to give the pipe bearer an opportunity to perform the ceremony of smoking, which is considered necessary to their success. He lights his pipe and remains a short time with his head inclined, and the stem of the pipe extended towards the herd. He then smokes and puffs the smoke towards the bisons, and the earth, and finally to the cardinal points successively.’’! This rite is evidently intended to bring the animals into mystic relation with the hunters and with certain spatial positions which will prevent their escaping.| It is magic enchantment, designed to make sure of them. We find something similar in Malaya. ‘When the hiding-place (of the deer) is discovered, all the young men of the kampong assemble, and the following cere- mony is performed.” . . . (A description is here given.) “It is believed that the absence of this ceremony would render the expedition unsuccessful, the deer would prove too strong for the ropes. . . .’2 So, too, in South Africa, Livingstone relates of one of his men that he was considered the leader of the hunting expedition, by virtue of his having a knowledge of “‘ elephant medicine.” He would go in front of the rest, and examine the animals, everything depending upon his verdict. ‘If he decided to attack a herd, the rest went boldly on, but if he refused, none of them would engage. A certain part of the elephant belonged to him by right. . . .”’ 3 Lastly, in South Australia, in hunting emus, it is not enough to have discovered their whereabouts ; they must be rendered powerless by means of magical operations. ‘‘ A mineral stone about the size of a pigeon’s egg . . . is found in the quarries and called ‘emu eye’ by the natives. These stones are © wrapped in feathers and fat, and when within a few hundred yards of the birds they commence throwing them towards the emus, believing there is a charm about the stones and that it prevents the emus from running.” 4 In New South Wales, ‘“ when a man went out hunting, he took with him a charmed wommera or spear-lever, the hook t Dorsey, “‘ Siouan Cults,” £. B. Rept., xi. pp. 375-6. 2 Skeat, Malay Magic, p. 172. 3 Livingstone, Missionary Travels, pp. 599-600. 4 Gason, “‘ Manners, Customs, etc., of the Dieri, Auminie Tribes,” J.4.7., XXIV. p. 172 PRIMITIVES’ PRACTICES 235 of which consisted of a bone from a dead man’s arm, ground to a point. The fat of the corpse was mixed with the gum used in lashing the hook to the shaft of the weapon. When the hunter espied an emu, kangaroo, turkey or similar game, he held up the wommera in sight of the animal, which would thereby be spellbound, and unable to run away. . . . When a clever man is out hunting and comes across the tracks of, say, a kangaroo, he follows them along and talks to the foot- prints all the time for the purpose of injecting magic into the animal which made them. He mentions in succession all the parts of the foot, and then names the different parts of the leg right up to the animal’s back. As soon as he reaches the backbone, the creature becomes quite stupid and is an easy prey.... Or again, “‘a hunter takes some fat, or skin, or piece of bone, of a dead man, and puts it into a little bag. He then goes to some place in the bush frequented by kanga- roos, emus, etc. . . . Here he selects a tree belonging to the proper phratry, and hangs his little bag on one of the spreading branches. When an animal gets within ‘shooting distance ’ of this magical artillery, it becomes stupid, and wanders about heedlessly until the hunter gets an opportunity of spearing it.” Or yet again, ‘as soon as some emus or kangaroos appeared in sight, the men commenced chewing human hair and spitting towards the animals, accompanied by magical incantations. This was expected to work a charm on the game and cause them to remain quiet and sluggish, so that a man could steal upon them. . . . In following along the tracks of an emu, kangaroo, wild dog, etc., if the hunter at intervals drop hot coals in the footmarks of the animal, this will have the effect of making it hot and tired, or induce it to come round again towards its pursuer.’’! Similarly, near Port Lincoln, ‘‘ the superstitious simplicity of these natives is strikingly apparent in their manner of hunting. . . . There have been transmitted to them, by their early ancestors, several short rhymes of two lines, which now are known to the adults only, and these, on pursuing an animal, or when on the point of spearing it, they constantly repeat with great 1 Mathews, “‘ Aboriginal Tribes of New South Wales and Victoria,” Jownal and Proceedings of the Royal Soctety of New South Wales, vol. xxxviii. PP- 254-7 (1905). 236 HOW NATIVES THINK rapidity. The literal meaning is totally unknown to them, and they are quite unable to give an explanation of them, . . but they faithfully believe them to possess the power, either to strike with blindness the animal which they are pursuing, or to create in it such a feeling of security and carelessness, that it cannot perceive its enemies or to weaken it so that it cannot effect its escape.’’ ! These are very significant facts. They show clearly that hunting is an essentially magical operation, and in it every- thing depends, not on the skill or strength of the hunter, but on the mystic power which will place the animal at his mercy. With many peoples success depends, too, upon certain inhibitions to be observed during the hunters’ absence by those who did not accompany them, and particularly by their wives. In Indo-China, for instance, “the Laotian hunters . set out after having recommended their women to practise strict abstinence during their absence ; they forbade them to cut their hair, anoint themselves with oil, to place the pestle and mortar for the rice out of doors behind the house, or to make light of the marriage contract, for such practices would prejudice the result of the hunt... . If the captured elephant in its struggles succeeds in overthrowing the mounts engaged in subduing it, it is because the wife left at home is unfaithful to her husband. If the rope that binds it breaks, she must have cut her hair; and if the rope slips so that the beast escapes, she must have smeared herself with oil.” 2 “When setting out they make offerings of rice, spirit, ducks, fowls, to the spirits of the long ropes with a running noose which are to be used in capturing the elephants. Yet more, the hunters command their wives to abstain from cutting their hair or offering the hospitality of their homes to astranger. Should these commands be disobeyed, the animals captured would escape, and the incensed husband would have the right to a divorce on his return. On his side, the hunter must abstain from sexual relations and, in accordance with a custom very common in Further India, must give conven- t Wilhelmi, ‘‘ Manners and Customs of the Australian Natives, in Par- ticular of the Port Lincoln District,’ Royal Society of Australia, v. p. 176 1860). 2 Aymonier, Voyage dans le Laos, i. pp. 62-3. PRIMITIVES’ PRACTICES 237 tional names to all ordinary things, which has the effect of creating a special ‘hunter’s language.’ In the hunting-field the chief chants certain formulas, handed down from father to son.”’1 So, too, with the Huichols during the hunting of the deer, a matter of the greatest significance to them, “the. important thing is that the chiefs, and the woman who has officiated, do not infringe the law of fasting. They follow the hunters in their thoughts all the time, and pray to the fire and the sun and all the other gods to give success, and thus, happiness to all. ... Now and then some of the fasters would rise and pray aloud, and with so much fervour that they and all the rest were moved to tears.”? From Schoolcraft we learn that if an Indian is unlucky in the chase he at once declares that someone must have broken their laws.3 D. Even when the game has been shot down and retrieved, all is not over. Fresh magical operations are necessary to conclude the cycle that the initial ones have begun; just in the same way as in the sacrifice, the opening rites, as Hubert and Mauss have demonstrated, correspond with the concluding rites. These operations are of two kinds, and it is sometimes difficult to distinguish between them. The object of the earlier ones is to stultify the animal’s vengeance—and at the same time that of the spirit which represents all the animals of the species ; for prelogical mentality, governed by the law of participation, does not recognize any distinct difference between the individual and the substance of the species. The later ones tend to pacify the victim (or its spirit). Death does not bring about entire disappearance in the case of animals any more than it does with men. On the contrary they continue to live, that is, to participate in the existence of their group, although in slightly different conditions, and, again like men, they are destined to be born again. Accord- ingly it is of the greatest importance to remain on friendly terms with them. Mooney tells us that the Cherokees have formulas for appeasing the animals which have been slaughtered, and that the hunter, when returning to camp, lights a fire on the road t Aymonier, Voyage dans le Laos, i. p. 311. 2 C. Lumholtz, Unknown Mexico, ii. p. 43. 3 Schoolcraft, Information, etc., ii. p. 175. 238 HOW NATIVES THINK behind him so that the deer chief cannot follow him to his home,? (and inflict him with illness, particularly a rheumatic affection). In Canada, “ when a bear has been killed, the hunter puts the bowl of his pipe between the animal’s teeth, blows down it, and thus filling its mouth and throat with smoke, entreats its spirit not to harbour resentment for what its body has suffered, and not to injure his prospects in any succeeding chase he may undertake.” ?—Among the Indians of Nootka Sound, “‘ after well cleansing the bear from the dirt and blood with which it is generally covered when killed, it is brought in and seated opposite the king, with a chief’s bonnet wrought in figures on its head, and its fur powdered over with white down. A tray of provision is then set before it, and it is invited by words and gestures to eat.” 3 Nothing is more common than the rendering of such honours to animals slain in the chase. Sometimes the ceremony assumes a mysterious character, and must be carried out away from the presence of the unbeliever. In West Africa, for instance “even my presence,” writes Nassau, “ was objected to by the mother of the hunter ; (he, however, was willing). After the animal had been decapitated, and its quarters and bowels removed, the hunter, naked, stepped into the hollow of the ribs, and kneeling in the bloody pool contained in that hollow, bathed his entire body with that mixture of blood and excreta, at the same time praying the life-spirit of the hippo that it would bear him no ill-will for having killed it, and thus cut it off from future maturity ; and not to incense other hippo- potami that they should attack his canoe in revenge.’ 4 It was evidently rites of this kind to which Du Chaillu was referring when he wrote: ‘‘ Before the manga (a kind of mamatee) was cut up, the manga doctor went through some ceremonies which I did not see, and nobody was permitted to see the animal while he was cutting it up.’’5 In the case of the Huichols, it is a public ceremony, and a very complicated one. “The animal was laid so that its legs were turned toward the east, and all sorts of food and bowls of tesvino 1 J. Mooney, “‘ The Sacred Formulas of the Cherokee,” E. B. Rept., vii. P. 347- 2 Charlevoix, op. cit., iii, p. 118. 3 Jewitt, Adventures and Sufferings, P. 133. 4 Nassau, Fetichism in West Africa, p. 204. 5 Equatorial Africa, pp. 402-3. PRIMITIVES’ PRACTICES 239 | were placed in front of it. Everyone in turn stepped up to the deer, stroking him with the right hand from the snout to the tail, and thanking him because he had allowed himself to be caught. ‘ Rest thyself, elder brother’ (if it be a doe they call her elder sister), A shaman may talk to the dead animal for a long time. . . . ‘ Thou hast brought us plumes, and we are profoundly thankful.’’”’: We know that antlers are akin to plumes. By means of such ceremonies normal relations between the social group to which the hunter belongs and the group of the slaughtered animal, are re-established once more. The murder is cancelled, and there is no longer _ any reason to dread vengeance. Fresh hunting expeditions _ again become possible in the future, provided that they are accompanied by the same mystic practices. II The primitive peoples whose main sustenance is derived from fishing proceed in just the same way as those who depend upon hunting for their food supply. They exercise magic influence upon the fish by means of dances similar to those I have just cited. ‘‘ The dances of the Torres Straits islanders are practised at night, and have for their object success in hunting and fishing. It is on these occasions that the extra- ordinary masks of tortoise-shell are used, and I assume that the forms of the masks to be worn would have relation to the particular sport to be engaged in ; for example, in the dance to ensure success in fishing, the mask would represent a fish, and so on.”’2—In the Nicobar Isles, ‘‘ during the whole of the day, the people were engaged in preparing torches for the ceremony of Ki-alah for ‘ multiplying the fish in the sea,’ and then started their fishing at night.” 3 The fisherman, like the hunter, must previously undergo _ fasts, rites of purification, periods of abstinence, in short, he must submit to mystic preparation. ‘‘ When a man was going to hunt sea-otter he fasted and kept away from his wife for a month. He kept his chamber box behind the door, always 1 C. Lumholtz, Unknown Mexico, ii. p. 45. 2 C. H. Read, “‘ Stone Spinning Tops from Torres Straits,” Fa TS Xvi ak yf 3 Solomon, “ Diaries kept in Car Nicobar,’”’ /.A.J., xxxii. p. 2 8 240 HOW NATIVES THINK urinated into it, and let no one else touch it. At the end of the month he started out after an eagle, and having killed one cut off the foot and tied a certain flower to it. Then he made a miniature canoe with figures of himself and perhaps others inside, and he represented himself in the act of aiming at a sea-otter. He made the eagle’s talon clasp the seat so that he could have a sure aim and secure the animal. When at length he went out and was beginning to approach the sea- otter he blew some of his urine towards it. This would confuse it, so that it would swim in his direction. Sometimes he tied a piece of wood to the eagle’s talon so that the sea-otter would stand right up in the water like a buoy, and be easily shot.’’! ‘“Those who go to catch sturgeon bathe in a pond early in the morning.” 2 Jewitt says of the Indians of Nootka Sound, ‘‘ the king makes a point of passing a day alone on the mountain, whither he goes very privately early in the morning, and does not return till late in the evening. This is done for the purpose of singing and praying to his god for success in . whaling the ensuing season. . . . The next two days he appears very thoughtful and gloomy, scarcely speaking to anyone, and observes a most rigid fast. . . . In addition to this, for a week before commencing their whaling, both himself and the crew of his canoe observe a fast, eating but very little, and going into the water several times in the course of each day to bathe, singing and rubbing their bodies, limbs and faces, with shells and bushes . . . they are likewise obliged to abstain from any commerce with their women for the like period.” 3 Sometimes these mystic observances are concentrated upon a single individual who thus becomes, as it were, the vehicle of the magic relation established between the human social group and the fish group. A very fine example of partici- pation of this sort is to be found among the New Guinea natives. ‘‘ The preparation for dugong and turtle fishing are most elaborate, and commence two months before the fishing is started. A headman is appointed who becomes belaga (holy). On his strict observance of the laws of the dugong t Swanton, ‘* The Tlingit Indians,”’ E. B. Rept., XXvi. P. 447 2 F, Boas, “ The North-west Tribes of Canada,” Reports of the British AU p. 460 (1894). 3 Jewitt, Adventures and Sufferings, pp. 154-5. PRIMITIVES’ PRACTICES 241 net depends the success of the season. He lives entirely secluded from his family, and is only allowed to eat a roasted banana or two after the sun has gone down. Each evening at sundown he goes ashore and bathes on the point of land overlooking the dugong feeding-ground . . . meanwhile throw- ing into the sea some mula-mula, i.e. medicine to charm the dugong. While he is undergoing these privations all the able- bodied men of the village are employed in collecting bark . . . and making nets.” Among the Ten’a Indians of the Yukon territory, when the fishing season is beginning, a medicine-man is reputed to penetrate beneath the ice to the place where the salmon, in large numbers, spend the winter. This is evidently with the purpose of securing their goodwill.z So too with the Hurons, ‘in each fishing hut there is generally a ‘fish preacher,’ accustomed to exhort the fishes by a sermon: if these are * clever men, they are greatly sought after, because the Hurons believe that the exhortations of an able man have much power to entice fish into their nets. The one we had was held to be one of the best; and it was a fine sight to witness his efforts, both by word of mouth and by gesture when he preached, as he used to do every evening after supper, after having first enjoined silence, and arranged everyone in his place, lying at full length around him. . . . His argument was that the Hurons never burned fish-bones, and then he proceeded, in the most affectionate terms, exhorting the fish, entreating and conjuring them to appear, to allow themselves to be caught, to take courage and not be afraid, since they were serving their friends, who honoured them, and never burned their bones. . . . To secure a good haul, they also burn snuff sometimes, using certain expressions which I do not understand. They throw some, too, with a like intention, into the water for particular spirits whom they believe to be _ presiding there, or rather, for the soul of the water, (since they believe all material and inanimate things to possess a soul that can hear) and they entreat it, in their customary fashion, to be of good courage, and allow them to take a goodly number of fish.’’ 3 * Guise, ‘“‘ Wangela River Natives,” J-A.T., xxviii. p. 218. + Fr, Jetté, ‘ On the Medicine-men of the Ten’a,’”’ /.A.1., xxxvii. D.°174. 3 Fr. Sagard, Le Grand Voyage au pays des Hurons,” PP. 257-9 (1632). 16 242 HOW NATIVES THINK The fisherman’s canoe and all his implements must, like his own personality, be invested with magic virtue to ensure success. In Malay, ‘“‘each boat that puts to sea has been medicined with care; many incantations and other magic observances having been had recourse to. . . . After each take the boat is ‘swept’ by the medicine man with a tuft of leaves prepared with mystic ceremonies which is carried at the bow for the purpose’: Just as the hunter prays to his horse and his weapons, so too, does the fisherman entreat the good- will of his nets. ‘‘One day, when I was about to burn at the camp fire the skin of a squirrel which a native had given me, the Hurons would not suffer me to do so, but made me burn it outside, because there were some nets in the camp at that time, saying that if I burned it there, these would tell the fish. I told them that the nets could not see, but they assured me they could, and that they also saw and ate: 1 I was once scolding the children of the district for some very bad behaviour, and the next day it happened that the men caught very few fish; they attributed the fact to this repri- mand, which the nets had reported to the fishes.”’ ? Many investigators have thought, too, that the form given to the fishing apparatus was designed to the same end as the ceremonies already described. ‘“‘ It is extremely probable that the carvings formerly inserted at the stern of most Torres Straits canoes had a magical significance . . . such were... . the head of the frigate-bird, and occasionally that of the sea- eagle . . . the tail of a king-fish. . . . All these creatures are voracious fishers. The indication of a head at the butt-end of the dugong harpoons . . . was doubtless magical in signi- ficance.’’3 In British Columbia, ‘‘ almost all the clubs that I have seen,” says Boas, “‘ represent the sea-lion or the killer- whale—the two sea-animals which are most feared by the Indians, and which kill those animals that are to be killed by means of the club. The idea is . . . to give it a form appro- priate to its function, and perhaps, secondarily, to give it, by means of its form, great efficiency.” 4 With the Cherokees, ‘the fisherman must first chew a 1 Skeat, Malay Magic, p. 193. 2 Fr. Sagard, op. cit., p. 256. - 3 The Cambridge Expedition to Torres Stratis, v. p. 338. 4 F. Boas, op. cit., p. 679. PRIMITIVES’ PRACTICES 243 small piece of the Venus’ flytrap and spit it upon the bait and also upon the hook. Then, standing facing the stream, he recites the formula and puts the bait upon the hook... . This . . . will enable the hook to attract and hold the fish as the plant itself seizes and holds insects in its cup. . . . The prayer is addressed directly to the fish, who are repre- sented as living in settlements.”’ ! During the fishing magical observances must be practised, and these correspond closely with those performed by the Sioux when the game comes in sight. For instance, in the case of the Baganda, ‘ when the net is let down the chief fisherman takes some of the herbs they have obtained from the priest of Mukasa, which are kept in a special pot, and smokes them in a clay pipe; the smoke he pufis from his pipe over the water, and it causes the fish to get into the net. . . . This pot has a special place where it resides; it is supposed to be animate, and resents being put anywhere but in its place of honour, and vents its anger by causing the fish to escape. . . . The canoes, too, which are used in fishing, have fish offered to them.” 2—In New Zealand, we are told, ‘the religious ceremonies connected with fishing were very singular. The day before they went to sea, the natives arranged all their hooks around some human excrement, and used a karakia (magic incantation) which will not bear repetition. In the same evening there were incantations. . . . When they reached the sea, and all the hooks were duly arranged, the tohunga set apart for fishing commenced to pray... standing up and stretching out his arms. . . . The first fish caught was returned to the sea, a karvakia being previously uttered over it, to cause it to bring abundance of fish to their hooks.” 3 Finally, after fishing as after hunting, mystic observances are necessary for the purpose of appeasing the “ spirit ”” of the creature (and of its tribe) of stultifying its resentment and regaining its goodwill. ‘‘ As soon as a sturgeon is caught,” says Boas, “the ‘sturgeon hunter’ sings, and by means of his song pacifies the struggling sturgeon, who allows himself to be killed.” 4 From his study of the Salish and other tribes, 1 J. Mooney, ‘The Sacred Formulas of the Cherokee,” E. B.Rept., vii. p. 83. + Roscoe, “ Manners and Customs of the Baganda,”’ J .AL,, xxxii, pp. 55-6. 3 R. Taylor, Te Ika a Maui, pp. 83-6. ' 4 F. Boas, op. cit., p. 460 (1894). 244 HOW NATIVES THINK Hill Tout has been led to the conclusion that these rites are always propitiatory. “They were intended to placate the spirits of the fish, or the plant, or the fruit, as the case may be, in order that a plentiful supply of the same might be vouchsafed to them. The ceremony was not so much a thanksgiving as a performance to ensure a plentiful supply of the particular object desired ; for if these ceremonies were not properly and reverently carried out, there was danger of giving offence to the ‘spirits’ of the objects, and of being deprived of them.” ! III Is it necessary to demonstrate that most of the customs relating to war betray the same characteristics as those we have already studied ? To the primitive mind there is no essential difference between fighting and hunting. In this pursuit also, we find the opening and the closing rites, the mystic ceremonies which mark the beginning of the campaign, dances, fasts, abstinence, purification, the consultation of dreams, the inhibitions imposed upon non-combatants, incan- tations directed against the enemy, charms, amulets, fetishes, medicines of all kinds to secure immunity from wounds, prayers to entreat the favour of the spirits; then, when the action commences, supplications addressed to the horses and weapons, the guardian spirits of the individual and the group, magical operations and formulas to bring about the blindness of the enemy, to incapacitate him from defending himself and paralyse his efforts; and lastly, after the battle, the ceremonies, often very complicated, by which the conqueror endeavours either to ward off the vengeance of his slaughtered foes (by mutilating or destroying their corpses) or else to conciliate their spirits,? to cleanse himself from the contamina- tion he may have suffered during the struggle, and finally to 1 “ Ethnological Reports of the Halkomelem,” J.A.J., xxxiv. p. 331. 2 They adopt the same attitude with regard to living prisoners. The North American Indians would either torture and kill their prisoners of war, or else they would adopt them. In the latter case they gave them the name of some dead warrior, whom they would represent henceforth. Thus they became merged in the very being of the social group, and from that time forward were subject to the same duties and enjoyed the same privileges as its other members. Cf. Catlin, The North American Indians, ii. p. 272. PRIMITIVES’ PRACTICES 245 make his superiority a lasting reality by the possession of trophies, such as heads, skulls, jawbones, scalps, weapons, etc. From the instant that war is seriously contemplated until the moment when it becomes a thing of the past there are always mystic relations to establish or to rupture, as the case may be, and upon these relations and the necessary operations involved the success of the campaign depends above all. Gallantry, finesse, superior weapons, numbers, tactics are certainly not matters of indifference, but they remain second- ary conditions only. If the birds of augury refused to eat and the Roman soldiers knew it, they believed that their army would be beaten, but among primitive peoples, if the dreams are unfavourable they do not even think of fighting. Again, our method of presenting these facts, a method necessarily in accordance with our mental habit of thought and subject to the laws of a language which reflects this habit, distorts them in the attempt to explain them. We cannot avoid putting the mystic operations on the one side, and the actions which combine to bring about the desired results on the other. But it is inherent in the very nature of prelogical mentality (and this makes it so difficult for us to reconstruct it), that there is no difference between these. The operations of both kinds form one inseparable method of acting. On the one hand, all actions, even those of a most positive char- acter, are mystic in their essence. The bow, gun, fishing-net, the horse of the hunter or warrior, all these participate in mysterious forces brought into action by these ceremonies. And on the other hand, these ceremonies are not only the indispensable preliminaries of the chase or of the fight : they are, themselves, both hunting and fighting. In short, in these forms of activity as in perception, primitive mentality has another orientation than ours; its character is essentially mystic, and its collective representations are governed and regulated by the law of participation. IV We understand, or at any rate we believe we understand, without any difficulty the customs of the primitive relating to hunting, fishing, and fighting, because we recognize that 246 HOW NATIVES THINK there are, among our own peoples, customs which are apparently somewhat similar. Such are the agrarian rites still so persistently practised. Such, again, are those of a religious rather than mystical character, which intercede with a benevolent and powerful intermediary, for the success of an undertaking. The Icelandic fishing fleet, for instance, does not leave Paimpol without having received a priestly blessing ; without it, many of the sailors would dread return- ing with a scarcity of fish, or that they might nor return at all. So, too, a Spanish admiral, before setting out to sea, consecrates his fleet to the Virgin, and his crews believe that the Mother of Jesus will assure them the victory. But among primitive peoples other ceremonies of a similar kind take place, and these we do not understand, because the effect expected is not among those we should look for. Then we miss the analogy, and we realize that the explanation it fur- nished was superficial and inadequate. These are precisely the ceremonies which best enable us to penetrate the real character peculiar to a mentality which is prelogical and mystic. Among them are the ceremonies which have for their aim the securing the regularity of the seasons, the normal pro- duction of the harvest, the usual abundance of fruit and of animals used for food, etc. Here we recognize one of the essential differences between their mentality and our own. To our way of thinking, “‘ Nature ” constitutes an objective order which is unchangeable. The savant, no doubt, has a clearer and more rational idea of it than the unlettered : but as a matter of fact, the idea obtrudes in familiar fashion on all minds even without their reflecting upon it. And it matters little whether this order is conceived as having been created and maintained by God, since God Himself is regarded as unchanging in His decrees. Action is regulated therefore by the idea of an order of phenomena subject to certain laws and exempt from any arbitrary interference. To the primitive’s mind, however, ‘“‘ Nature,’ in this sense, is non-existent. The reality surrounding the social group is felt by him to be a mystic one: in it everything relates, not to laws, but to mystic connections and participations. Doubt- less these do not, in general, depend any more upon the PRIMITIVES’ PRACTICES 247 goodwill of the primitive than the objective order of Nature in our case depends upon the individual who imagines it. Nevertheless, this objective order is conceived as based upon a metaphysical foundation, whilst the mystic connections and participations are simultaneously imagined and felt as solidary with the social group, and dependent upon it as it depends upon them. We shall not be surprised, therefore, to find this group concerned in maintaining what is, to us, the natural order, and that it may succeed in this, making use of cere- monies similar to those which secure for it its supply of game or fish. The most characteristic of these are undoubtedly the intichiuma ceremonies, which Spencer and Gillen have des- cribed in detail, and which they define thus: “‘ sacred ceremonies performed by the members of a local totemic group with the object of increasing the number of the totem animal or plant.” 1 I cannot give even a brief analysis of them here. Asa rule, they consist of a series of very compli- cated rites: dances, paintings, special ornaments of the members of the totem exclusively concerned with the cele- bration of the ceremony, the imitation of the movements of the totemic animal and efforts to realize their communion with it. To minds of this kind, the individuals forming part of a totemic group, the group itself, and the totemic animal, plant, or object, are all the same thing. We must understand “the same ” by virtue, not of the law of identity, but of the law of participation. We have already seen proofs of this.? The British Scientific Expedition to the Torres Straits affords others. “A... mystic affinity... is held to obtain between the members of a clan and their totem. This isa deeply ingrained idea and is evidently of fundamental import- ance. More than once we were told emphatically: ‘ Augud (the totem) all same as relation, he belong same family.’ A definite physical and psychological resemblance was thus postulated for the human and animal members of the clan. There can be little doubt that this sentiment reacted on the clansmen and constrained them to live up to the traditional character of their respective clans. . . . We were told that 1 The Native Tribes of Central Australia, p. 650. 2 Vide Chap. II. pp. 9r et seq. 248 HOW NATIVES THINK the following clans like fighting: the cassowary, crocodile, snake, shark . . . the ray and sucker-fish are peaceable. . The Umai (dog) clan was sometimes peaceable and at other times ready to fight, which is very characteristic of actual dogs.’’ And, in fact, in nearly all the islands of the Straits, the inhabitants would celebrate, with the help of costumes, masks, and dances, etc., ceremonies which were absolutely similar to the snfichiuma ceremonies described by Spencer and Gillen.? In the snfichiuma ceremony of the witchetty grub totem, the actors reproduce the actions of the mythical ancestor, the intermediary through which the group participates in its totem, and which, as a consequence, exerts a mystic influence upon it.3 Sometimes, even, the physical modality of this mystic influence can be apprehended. For instance, in the course of the intichiuma ceremony of the kangaroo totem, it may happen that the men of the totem cause their blood to flow over a certain rock. The effect of this proceeding is to drive in all directions the spirits of the kangaroos which used to inhabit this ground, and consequently to increase the number of living animals, for the kangaroo spirit enters into the living kangaroo just as the spirit of the male animal enters into the body of the female (when she is impregnated), This close solidarity (which finds no satisfactory expression in our language, because it is something more than solidarity ; it is, so to speak, a mystic participation in the same essence) may be extended to all the members of a given social group, and thus realize that which we, for want of a better term, shall call ‘‘ mystic symbiosis.”” In the Tjingilli tribe, Spencer and Gillen noted the following ceremony. ‘To make both young men and women grow strong and well-favoured, the men perform, at intervals of time, a long series of ceremonies . . . dealing with the various totems. There is no special reference to the young men or women in them, but they are performed solely with the idea and object of increasing the growth of the younger members of the tribe who are not, of * The Cambridge Expedition to Torres Straits, v. p. 184. Cf. Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. p. 120. 2 Ibid., v. pp. 347-9. 3 Spencer and Gillen, The Native Tribes of Central Australia, pp. 171 et seq. PRIMITIVES’ PRACTICES 249 course, allowed either to see them or take any part in them.” ! There is some mystic participation between the rain and the members of the rain totem, between the water totem and its members, and as a result the intichiuma ceremonies will be performed in order to assure the regular rainfall, or the usual amount of water in the pools. These bear a striking resemblance, down to the most minute detail, to the cere- monies carried out, with the same end in view, by the Zuñis, Arapahos, and the pueblos of North America as a whole : ceremonies which the collaborators of the Ethnological Bureau in Washington have so carefully described. In Australia, as in New Mexico, we may see “a curved band supposed to represent a rainbow, a drawing also of one or more of these on his (the actor’s) own body, and a special one on a shield which he brings with him. . . . This shield is also decorated with zigzag lines of white pipeclay which are supposed to represent the lightning.’’2 Here, no doubt, we have another “ motif,” familiar to prelogical mentality: the mystic value residing in the likeness, and the power exercised over a person or thing by the possession of this likeness (a sympathetic magic which utilizes the participation between the one and the other). But in the infichiuma ceremonies, at any rate those celebrated by Australian aborigines, the appeal is simul- taneously and specially to another yet more profound parti- cipation, I mean the essential communion between the totemic group and its totem. As Spencer and Gillen tell us, “the members of a totem, such as the rain or water totem, will hold their snfichiuma when there has been a long drought, and water is badly wanted ; if rain falls in a reasonable time, then of course it is said to be influenced by the intichiwma. . . . The performance of these ceremonies is not associated in the native mind with the idea of appealing to the assist- ance of any supernatural being.’’3 We know, moreover, that these investigators were never able to discover, in the abori- ginals they studied, any idea of gods, properly so called, and ‘in no case is there the slightest indication or leaning to 1 The Northern Tribes of Central Australia, P. 295. + The Native Tribes of Central Australia, p. 170. 3 Ibid. 250 HOW NATIVES THINK anything which may be described as ancestor-worship. . . These Alcheringa ancestors are constantly undergoing rein- carnation so that this belief . . . practically precludes the development of anything like a worship.”’ ! | Here we are confronted with a method of acting in which the prelogical mystic mentality betrays itself in a different fashion from that seen in most of the similar practices noted in primitives of a more highly developed type than the Australian aborigines. : Nothing is more widespread than prac- tices having as their object the cessation of drought, and the assurance of rain: (we see this even yet in our own Rogations) but such practices usually take the form of supplications or prayers. Even when including, as they nearly always do, recourse to sympathetic magic, they are at the same time addressed to one. or more divine beings or spirits or souls whose intervention will produce the desired phenomenon. They feel themselves further removed from the rain than they do from souls or spirits or deities and they feel that they can influence these last and communicate with them, by means of prayers, fasting, dreams, sacrifices, dances, and sacred ceremonies of all kinds: and they do not feel that they can communicate directly with the rain in this way. In certain parts of China, for instance, as De Groot tells us, the people maintain the convents solely because they are con- vinced that they are able to regulate the winds and the rains, and thus assure the crops, since these are so exposed to drought in the treeless regions of China. . . . They subscribe liberally to erect and maintain the buildings, in return for which the monks are expected to bring about the cessation of the drought by means of their ceremonies, when occasion requires it.2 The monks, in their turn, address themselves to the divinities concerned, and regulate the fung-shui. Among the Australians, on the contrary, we find no priests or inter- mediaries of any kind. The intichiuma ceremony manifests the direct solidarity and mystic participation between the rain-totem and the rain, as it does between the members of the kangaroo totem and the kangaroos, and however strange it may seem to us, this solidarity and this participation are 1 The Northern Tribes of Central Australia, p. 494. * J. J. M. de Groot, La Code du Mahayana en Chine, p. 100. PRIMITIVES’ PRACTICES 251 not only imagined but fel¢ collectively by the members of the totemic group. But, we may ask, supposing the rain falls without the ceremonies having taken place? The aborigines have quite unconsciously provided for such an objection. They have given, beforehand, the most natural and decisive answer, regarded from the point of view of mystic mentality. The imhichiuma ceremony has not been celebrated by the totemic group, it is true, but since the rain has fallen, it has never- theless taken place, for it has been celebrated by well-disposed spirits (truniarinia). ‘The iruntarinia are supposed fre- quently, but not always or of necessity, to commend in dreams to the Aluntaja’’ (an old man who acts as religious chief) “of any group the time at which it is right to perform the ceremony of the snfichiuma. They themselves perform similar ceremonies, and after a plentiful supply of, say, witchetty grub or emu appears without the performance of imhchiuma by the peoples of the respective totems, then the supplies are attributed to the performance of intichiuma by friendly truntarinia.”’* At Mabuiag, in Torres Straits, a similar belief is to be found. ‘The madub was a wooden image of human shape. . . . The business of the madub was to take charge of the garden beside which it was placed, and to give good crops of yams, etc. . . . At night time the madub became animated and went round the garden, swinging the bull-roarers to make the plants in the garden grow, and they danced and repeatedly sang. .. Indeed, the mari (spirits) of the madub do what men do.’’? According to Bergaigne, the Vedic sacrifice is performed in heaven as on earth, and with the same results.3 There is an analogous belief which maintains that cere- monies having the same object as the intichiuma, are also performed by animals. ‘‘ As a matter of fact the Tarahumares assert that the dances have been taught them by the animals. Like all primitive peoples they are close observers of nature. To them, animals are by no means inferior creatures ; they understand magic and are possessed of much knowledge, and * Spencer and Gillen, The Native Tribes of Central Austvalia, pp. 519-21. ? The Cambridge Expedition to Torres Straits, v. pp. 345-6. 3 La Religion Védique, i. pp. 12-13. 252 HOW NATIVES THINK may assist the Tarahumares in making rain. In spring, the singing of the birds, the cooing of the dove, the chirping of the cricket, all the songs uttered by the denizens of the green- sward, are to the Tarahumares appeals to the deities for rain. For what other reason should they sing or call? And as the gods grant the prayers of the deer expressed in its antics and dances, and of the turkey in its curious playing, they easily infer that to please the gods they, too, must dance as the deer and play as the turkey. . . . Dance, with these people is a very serious and ceremonious matter, a kind of worship and incantation rather than an amusement... . The very word for dancing, noldvoa, means literally to work.” ! So too, according to a Hopi tradition, cockroaches dying of thirst dance to obtain rain.? In most primitive peoples rather more developed than the Australian aborigines, it is no longer the whole totemic group of the locality which secures, by the celebration of appropriate ceremonies, the desired result. Frequently one member of the group, possessing special qualifications, is the forced or the voluntary vehicle of the participation which is to be established. We noted acase of this kind in New Guinea, with respect to the dugong fishing. Sometimes it is an individual chosen expressly for this end, and doubtless indicated as the choice of the members of the group for some mystic reason or other. Sometimes he is qualified for this function by a special initiation which has rendered him more ready to receive and exercise magic influence than his fellows. And finally he is sometimes indicated beforehand by his birth and the participations it implies, since to these peoples a man actually zs the same as his ancestors have been, and he really is, at times, the reincarnation of a certain ancestor. Thus it is that chiefs and kings, by virtue of their origin, are very often the intermediaries required. By the rites which they perform, and which they alone are qualified to perform, they secure the regularity of natural phenomena and the very life of the group. It is in this way that we must interpret the numerous 1 C. Lumholtz, Unknown Mexico, i. pp. 330-2. 2 Voth, The Traditions of the Hopi, pp. 238-9 (Field Columbian Museum, Anthrop., viii. 1905). PRIMITIVES’ PRACTICES 253 instances collected by Frazer in The Golden Bough, which show the social group almost as solicitous about the person of their king as a hive of bees is about the safety of its queen. The welfare of the group depends upon the king’s welfare. To the Ba-Ronga, “‘ royalty is a venerable and a sacred insti- tution. Respect for the chief and obedience to his orders are universal, and it is not the display of wealth or power which maintains his prestige, but the mystic belief that the nation lives through him as the body does through the head.” 1 We noted that the Mandans, according to Catlin, were very uneasy at the thought that their chiefs’ portraits might be in the hands of strangers, and this might interfere with their repose in the grave. In Bengal, the Banjogis and Pankhos affirm that “in the time of one of the rajahs, Ngung Jungnung, they were the dominant and most numerous of all the tribes in this part of the earth. . . . They attribute the declension of their power to the dying out of the old stock of chiefs, to whom divine descent was attributed.” 2 The king, more particularly, often secures a regular rain- fall and the abundant harvest which is its natural result. “In the good old times, when there was still a Ma Loango, the Bafoti were certainly more prosperous. The king enjoyed considerably more power than others. He did not urge on the clouds or direct the winds: . . . he did better than that, for he made the rain fall direct from heaven, as soon as his subjects needed it.”’3 In Malaya, ‘ the king is firmly believed to possess a personal influence over the works of nature, such as the growth of the crops and the bearing of the fruit-trees.”’ (The same power is attributed to the British Residents.) “I have known (in Selangor) the success or the failure of the rice crops attributed to a change of district officers, and in one case I even heard an outbreak of ferocity which occurred among man-eating crocodiles laid at the door of a most zealous and able, though perhaps occasionally somewhat unsympathetic, representative of the government.”’ 4 The mystic power of the king sometimes persists after his death. Thus, with the Ba-Ronga people, ‘the death of 1 Junod, Les Ba-Ronga, p. 139. * Quoted by Grierson, Linguistic Survey of India, iii. p. 145. 3 Dr. Pechuël-Loesche, Die Loango-Expedition, iii. 2, p. 449. 4 Skeat, Malay Magic, p. 36. 254 HOW NATIVES THINK Mapunga must have taken place in the course of the year 1890, but no one ever spoke of it. When the news was made. public, however, Manganyeli, a young native from the Ribombo district said to me: ‘When the miphimbi trees were so full of fruit last season (about Christmas, 1890)—we had an excep- tional crop—we thought that the chief must be dead, and that he had sent us this wonderful abundance.’ ”’ ! It is, therefore, of the greatest importance that the tribe should remain in communication, or rather, in communion, with its dead kings. Junod has very clearly shown how such participation between the social group and its deceased chief is realized both physically and mystically, and the passage is worth quoting in full. “In every little Ba-Ronga clan . .. there is a sacred object which one might be inclined to take for an idol, but which is in reality something quite different. It is called the mhamba (a word which denotes any kind of offering or sacrifice, but which is especially applied to such an object as this). It even seems as if they had such a respect for it that they hesitate to give it an official name. They prefer to call it nhlengoué (treasure, wealth). But what is this mhamba, this holy and mysterious kind of ark? When a chief dies, the nails of his hands and feet are cut off, and his hair and beard, and all the parts of his body which can be preserved are hardened with the manure formed from the oxen killed at the time of his death. Thus there is formed a sort of pellet which is afterwards wrapped round with strips of hide.. When the successor of this chief dies, a second pellet is manufactured and added to the first, and thus it goes on for generations. At the present time the mhamba of the Tembé tribe is about a foot long, according to the account of one who has often seen it (since it is guarded by a cousin of his). The guardian of the sacred relic is very carefully selected. He must be a man of particularly calm tempera- ment, not addicted to bad language or to intemperance. As a kind of high priest . . . he stands in a very responsible relation to the whole country. . . . The mysterious object is preserved in a building erected for the purpose, behind the village to which the guardian belongs. When the keeper of the mhamba knows that he is to make use of it for a religious t Junod, Les Ba-Ronga, p. 128 (note). PRIMITIVES’ PRACTICES 255 ceremony, he lives in a state of entire continence for the previous month. As for the solemn sacrifice performed with the aid of this object, it is usually a goat. . . . The priest traces circles in the air with the mhamba. The prayer is then pronounced and it is of course addressed to all the former chiefs whose nails and hair the officiating priest holds in his hands: a strange and striking way of entering into com- munion with the gods. . . . The sacrifice with the mhamba is carried out (in the Tembé clan, at any rate), at the beginning of the bokagne season, before offering the firstfruits to the ghosts of the departed. In times of national danger, doubt- less, they have recourse to this solemn practice. “This amulet, therefore, of incalculable efficacy, is the nation’s most treasured possession. . . . Consequently it is the very last object which would be allowed to fall into the hands of an enemy. If the army is vanquished and put to flight, the depositary of the mhamba starts first of all... . The sacred emblem will not be wrested from him until the entire forces of the tribe have been destroyed. Such a mis- fortune did occur, it seems, some years ago, to the Tembé clan. ... Then a terrible drought ensued, and for a whole year the sky was of a fiery colour. ...’’ And the worthy pastor concludes: “ Is there not something profoundly logical in the idea of preserving some physical part of the mighty dead who have become the gods of their country, in order to exercise influence upon their will, and power over them? ”’1 This account of his shows very clearly that to the collective consciousness of the social group, its welfare, its very existence and the regular order of nature are connected, by means of a mystic participation, with the kings and chiefs of the group\ We know that to the prelogical mind, governed by the law of participation, part of a living whole is equal to the whole, is, in fact, the whole, in the prelogical sense of the verb ‘‘ to be,” and thus we can understand the part played by the nails and hair of the dead kings in the mhamba. This recognized power, whether in the totemic groups, the medicine-men or the chiefs, living or dead, of consolidating or even of effecting, through appropriate ceremonies, the order of nature and the regular recurrence of generation, has a certain 1 Junod, Les Ba-Ronga, pp. 398-401: 256 HOW NATIVES THINK analogy with the continuous creation which theologians and. certain metaphysicians have defined by saying that without the Divine intervention created beings could not subsist for a single instant. It is indeed a participation of the same kind that the prelogical mind imagines, though in a more gross fashion. According to its view the natural order only endures by virtue of periodical renewals, obtained through the special ceremonies performed by those possessing the requisite mystic powers. And the social order frequently disappears when the king dies, until his successor has assumed the power: every interregnum becomes anarchy. But there is this difference that, according to the doctrine of continuous creation, though the world subsists only through God, God would exist without the world if the latter were to disappear ; whereas in the case of prelogical mentality there is complete reciprocity. As a rule there is absolute action and reaction between the totemic group and its totem, and, in a higher type of society, between the nation and the succession of kings. It is the ‘‘ mystic symbiosis’? of which we have already spoken, and which our logical thought cannot clearly conceive save in a distorted form. V The community of essence, the mystic participation, is not conceived and sensed between the members of the same totemic group merely. It exists, too, in many primitive peoples, between the child and its father, the child and its mother, and the child and both parents; and, once its prin- ciple has been recognized, it is translated into customs which express it very clearly. Of these customs the couvade is often the only one which claims the attention of explorers. Its apparent strangeness has struck them. In reality, however, it is but part of a sum-total of taboos or precautions which are imposed, sometimes on the father, sometimes on the mother, and sometimes on both, which begin as soon as the pregnancy is announced, and continue long after the birth of the child. We shall note but the chief of these only. “ When the wife of a Braman is in the family way, as soon PRIMITIVES’ PRACTICES 257 as her husband knows it, he cleans his teeth, and eats no more betel nor trims his beard, and fasts until his wife gives birth to her child.’”’t ‘In Loango, the nganga (medicine- man) imposes upon the future parents, or upon the mother only, a taboo either simple or complicated, which they must observe until the child has uttered its first cry, or taken its first step, or even much later, until the child walks well, or has brothers and sisters. It may happen therefore that a father’s conduct is somewhat strange before, as well as after, the birth of his offspring.’”’ Du Chaillu tells us that “ while she (a female gorilla) was alive, no woman who was enceinte nor the husband of such woman dared approach her cage. They believe firmly that should the husband of a woman with child, or the woman herself, see a gorilla, even a dead one, she would give birth to a gorilla, and not to a man child. This superstition I have noticed among other tribes too, and only in the case of the gorilla.”’ 3 At Amoy in China, the husband must be extremely cautious in his movements during his wife’s pregnancy. “If the soil is disturbed, the repose and growth of the embryo in the womb of women is, by the law of sympathy, disturbed also... . Especially perilous is it to drive a nail into a wall, as it might nail down the earth-spirit that resides in it, and cause the child to be born with a limb stiff or lame, or blind of one eye ; or it might paralyse the bowels of a child already born, and give it constipation with fatal results. The dangers which threaten a pregnant woman increase as her pregnancy ad- vances. In the end, nothing that is heavy may be displaced in the house, it being well known that the earth-spirits are wont to settle preferably in things which, owing to their weight, are seldom moved. But even the shifting of light objects is a source of danger. Instances are known of fathers who had rolled up their bed-mat after it had long lain flat, being frightened by the birth of a child with a rolled-up ear. One day I saw a boy with a hare-lip, and was told by the father that his wife when pregnant of this child, had thought- lessly made an incision in an old coat of his, while repairing t Duarte Barbosa, ‘“‘ A Description of the Coast of East Africa and Malabar in the Beginning of the Sixteenth Century,” Hakluyt Society, xxxv. P. 1231 2 Dr. Pechuél-Loesche, Die Loango-Expedition, ili. 2, p. 402 3 Du Chaillu, Equatorial Africa, p. 262; Cf. p. 305. 17 258 HOW NATIVES THINK it.’ '—In New South Wales, “‘ certain foods are forbidden to women during portions of their pregnancy and lactation.”’ 2 —In New Guinea, during pregnancy, the Jabim women abstain from the flesh of iguanas, cuttlefish, dogs, in short of all fat and rich foods, “ for fear that the child may be born dead or deformed.’ Amanofthe Jabim tribe is equally bound by cer- taininhibitions. During his wife’s pregnancy heis forbidden to go to sea; “ the fish avoid him, and the sea becomes rough.”’ 3 In Brazil, ‘“‘ many Indian tribes have a custom that when a pregnancy is declared, the husband and wife must rigorously observe a fast. They must abstain from everything but ants, mushrooms, and water with which a little guarana powder has been mixed.” 4 In the Admiralty Isles, “when a pregnant woman feels her confinement approaching, she stays at home, and eats fish and sago only. She does not eat yams lest her child should be long and thin; she does not touch taro, lest he should be short and stocky, and she abstains from pork, for fear that he should have bristles instead of hair.” 5 During the accouchement certain practices show that the idea of the solidarity between father and child is still felt. For instance, in the tribes studied by Spencer and Gillen, when labour has begun, the father’s girdle is taken off and transferred to the mother. “Not a word is spoken, but if after a time the birth of the child is not announced, the husband, still quite unadorned, walks once or twice slowly, at a distance of about fifty yards, up and down past the erlukwirra (women’s camp) with a view to inducing the un- born child to follow him.’’® In the more northerly tribes, “when the father of the newborn child goes out into the scrub for three days, away from his camp, he leaves his waist- girdle and arm-bands behind him, so that he has nothing tied tightly round any part of his body, a state of affairs which is supposed to be beneficial to the lubra.”’ 7 1 J. J. M. de Groot, The Religious System of China, i. pp. 538-0. 2 Mathews, “ Aboriginal Tribes of New South Wales and Victoria,” Journal of the Royal Society of New South Wales, p. 219 (1905). 3 Hagen, Unter den Papuas, p. 229. 4 Von Martius, Bettrage zur Ethnographie Stid-Amerika’s, i. p. 402. 5 Parkinson, Dreissig Jahre in der Südsee, p. 398. 6 The Native Tribes of Central Australia, pp. 466-7. 7 The Northern Tribes of Central Australia, p. 607. PRIMITIVES’ PRACTICES 259 Now we come to the couvade practices, properly so called, which have often been described. Upon examining them in detail, however, we see that in the majority of cases, the inhibitions and the observances are imposed upon both parents. Explorers have laid more stress upon the father’s part in them, either because it is greater, as is frequently the case, or because the fact seems in their opinion more extraordinary and worthy of mention. And if the rules to be observed by the father are more important and more rigorous, it is because the community of substance between the child and himself are imagined and felt more vividly than that between the child and the mother. This participation has been well illustrated by Dobrizhofier, who has described the couvade in detail: the abstinence from food, from all violent exercise, etc. Otherwise, if the child dies, it is the father’s fault. An Indian refuses snuff, because he might injure his newborn child by sneezing. In short, the Abipones believe ‘ that any discomfort whatever, suffered by the father, influences the child, on account of the close sympathy existing between them.” 1—Von den Steinen, too, has given us a detailed description of the couvade, as noted by him in Brazil. ‘The married pair did not leave their hut save to relieve the neces- ‘sities of nature. They lived exclusively upon pogu boiled into a pap, and mandioc cakes crumbled and mixed with water. Any other food would have been injurious to the child ; it would have been as bad as if the child itself ate meat, fish, or fruit. . . . The Indians despise those who do not conform to this custom. ... The father is weak and sickly, since he feels himself to be one with the newborn child. . . . Among the Bororos, not only does the father fast, but if the child is ill, it is the father who swallows the medi- cines, as we learnt from the dispenser to the Brazilian Military Colony, who was immensely astonished at this.”’ 2 According to Von Martius, inhibitions are placed upon the father and mother alike. ‘ After the birth the father hangs his hammock beside his wife’s, to await, as she does, the fall of the umbilical cord. During this time, the mother is regarded as unclean, and the beds of the pair are separated 1 Dobrizhoffer, Historia de Abiponibus, ii. pp. 231 et seq. 2 K. von den Steinen, Unter den Naturvélkern Zentralbräsiliens, pp. 289-94. 260 HOW NATIVES THINK by a partition of palm-leaves, if they do not occupy separate huts. For the whole period, neither father nor mother can do any work, and the father cannot leave his hut save for a moment in the evening. His habitual bath is forbidden... .” (There are certain taboos with regard to food.) “Still more strange is the forbidding of the scratching of the head or the body with the finger-nails. . . . A breach of these regulations would bring about the death of the nursling, or at any rate make him sickly for the rest of his life.” 1 So too, in New Guinea, “ until the child can walk and begins to talk, the mother must follow the diet prescribed for her pregnancy. After her accouchement, she must not smoke tobacco, ‘‘ because the child would be all blackened within, and would die.’’ The father, too, must abstain from tobacco and betel-nut for a time ; however, according to what Vetter, who tells us this, says about it, this rule is not very strictly observed.2 With the people of Goa, ‘‘ during this period of three weeks, not only the mother but also the father is deemed to be impure, and is required to abstain from all his ordinary occupations.” 3—In South India, ‘it was noted by the Rev. S. Mateer, that, after the confinement of a Paraigan women in Travancore, the husband is starved for seven days, eating no cooked rice or other food, only roots and fruits, and drinking only arrack or toddy.” 4«—Among the Klamaths of Oregon, “on account of a childbirth, the father and the mother eat no meat for ten days.” 5 These facts, to which we might easily add many similar instances, will doubtless suffice to show that all the customs relating to the connection between the newborn child and its parents, including the couvade, as well as the taboos affecting pregnancy, imply, at any rate originally, the idea of a close participation between the child just born, or about to be born, and its father or mother, or both. The case of the Bororo taking his sick child’s medicine is perhaps the most characteristic, because it marks the parti- 1 Von Martius, op. cit., i. p. 643. 2 Hagen, Unter den Papua’s, p. 234. 3 Risley, Castes and Tribes of Bengal, i. p. 289. 4 E. Thurston, Ethnographic Notes in South India, p. 550. 5 A. Gatschet, The Klamath Language, p. 91. PRIMITIVES’ PRACTICES 261 cipation most clearly. But the other customs proceed, at least primitively, from the same collective representations. As certain of these customs persist—the actual couvade, for instance—while others perish, as their meaning is lost, one seeks, and naturally will find, an explanation of the extra- ordinary custom which obliges a man to go to bed as soon as his wife is confined. These more or less probable explanations do not hold good, however, when the couvade is once relegated to the sum-total of the practices to which it belongs. Even for a long time after the birth, even when at a dis- tance from each other, the participation between father and child can still be strongly felt. Thus, in Borneo, “ the war- coats of the men are often made of goat- or deer-skin, and any man may wear such a coat. But when a man has a young son he is particularly careful to avoid contact with any part of a deer, lest through such contact he should transmit to his son in any degree the timidity of the deer. On one occasion when we had killed a deer, a Kenyah chief resolutely refused to allow its skin to be carried in his boat, alleging the above reason.” ! It is at the time of the initiation of the young men—an initiation, as we shall presently see, which is a kind of new birth—that the participation between them and their respec- tive mothers becomes felt once more, and there are many customs which betray this. For instance, among the Aruntas, when the young man who has been circumcised is out in the bush, his mother ‘‘ may not eat opossum, or the larger lace lizard, or carpet snake, or any fat, as otherwise she would retard her son’s recovery. . . . Every day she rubs her body all over with grease, as in some way this is supposed to help her son’s recovery.” ? In other tribes, ‘while the boy is out in the bush, the mother wears alpita in her hair . . . and is careful also never to let her fire go out. The object of the former is to assist the boy to be watchful at night-time. .. . The alpita is the tail tip of the rabbit-bandicoot, a small animal which is very lively during the night, so that, of course, the wearing of the alpita is a sure stimulus to wakefulness. 1 Hose and Macdougall, ‘‘ Men and Animals in Sarawak,” J.A.J., xxxi. p. 187. 2 Spencer and Gillen, The Native Tribes of Central Australia, p. 250. 262 | HOW NATIVES THINK (The deprivation of sleep is one of the initiation tests.) Not only is it efficacious in the case of the actual wearer, but it is effectual when worn by someone closely related to the individual whom it is desired to influence in this particular way.’ ! In an Australian initiation ceremony described in detail by Mathews, the mothers of the novices are treated in a special way, much resembling that accorded to mourners, and later to those who have been confined—which confirms the idea that initiation comprises an apparent death and a new birth. “At the camp the mothers of the novices belonging to each contingent occupy quarters by themselves a little distance from the camp of their own tribe. Every mother has a fire of her own, and no one else is permitted to use it. . . . Their sisters, or mothers’ sisters, or some of the older women provide them with food, and attend to their wants generally. These women are collectively known as yanniwa, and none of the other women or the children are permitted to interfere with them. Each mother eats the whole of the food brought to her, as it would bring evil upon her son if she gave any portion of it to the other women present. All the mothers are, however, very abstemious with their food whilst their sons are away.’’? In the province of Victoria, “the mothers of the novices eat practically the same kind of food which is given to their sons in the bush, and must remain silent the same as their sons. They sing the pre- scribed songs every morning at dawn, and every evening at dusk ; and whilst standing singing they lift burning sticks from off the fire and wave them repeatedly towards the direc- tion in which they suppose the camp of the novices to be situated.” 3—Finally, in New South Wales, “ during the time a youth is out in the bush with the old men, going through the initiation ceremonies, he must only eat certain kinds of food, and his mother and father are restricted to the same diet as he. And when a novice is released from any taboo regarding food, his mother is freed at the same time.” 4 * Spencer and Gillen, The Northern Tribes of Central Australia, P. 344. ? Mathews, “The Burbong or Initiation Ceremony,” Journal of the Royal Society of New South Wales and Victoria, pp. 131-51 (1898). 3 Id., “‘ Aboriginal Tribes of New South Wales and Victoria,” Journal of the Royal Society of New South Wales and Victoria, PP. 317-18 (1905). 4 Id., Thurrawal and Thoorga Tribes, p. 259 CHAPTER VII INSTITUTIONS IN WHICH COLLECTIVE REPRE- SENTATIONS GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (II) INTRODUCTION THERE are certain practices, the meaning of which, at any rate as far as their object is concerned, does not admit of doubt: such are those which aim at treating the sick, pre- venting the disease from having a fatal issue, and restoring the sufferers to health. Here again we shall see, when we study the practices almost universally followed by primitive peoples, that they confirm the results yielded by an analysis of their collective representations. Once more we perceive the mystic orientation, the preconceived ideas which allow observation and experience but a very limited area of influence, the connections between entities and phenomena dominated by the law of participation. The facts vary greatly in detail, and yet when they are referred to the mental conditions upon which they depend, they will be found fairly uniform. I shall preferably select those which most clearly demonstrate these mental conditions. 1 À. In the first place, the very notion of illness is a mystic one. That is, illness and disease are always regarded as the product of an invisible, intangible agent, pictured, moreover, in many different ways. Observers are unanimous upon this point. “We must carefully avoid thinking that the Fijian conceives of illness as we do. To his mind, illness is like a wave, an outside influence which weighs down upon the sick 264 HOW NATIVES THINK man and seems to possess him. This wave, this influence can only proceed from the gods or the devils or from living beings ; but from natural causes, such as cold or heat, hardly ever... . There is no natural cause for sickness to the Fijians; they seek it practer naturum, 1.e. in an invisible world, existing side by side with this.’”’* The expressions which Father Rougier makes use of are noteworthy. To our minds, indeed, this invisible world can only be co-existent with, yet outside, that which we call nature. That which, on the contrary, characterizes prelogical mentality, as I conceive of it, is that in its collective representations these two worlds make but one. Its mystic elements are quite as natural as the others which participate in them. Junod the missionary helps us to perceive this participation beneath the distinction he is endeavouring to establish. ‘‘ The native regards diseases not only as physical disorders, but as the result of a kind of natural malediction of a more or less spiritual nature, and that is why the patient must not only be treated for such and such a symptom, but he must have the defilement which he has under- gone removed. When he effects this second cure, the healer has become what is popularly called a medicine-man. Hence all his efforts to appear a supernatural personage, (his costumes, paraphernalia, etc.). . . . All these accessories inspire both fear and confidence in his clients.’’2 The illness, however, does not require the separate, successive intervention of healer and sorcerer. It is the mystic conception the native has of it which involves the need of mystic methods of sub- duing it and escaping from its hold. This mystic idea may be of almost infinite variety. In Loango, for instance, the natives affirm that ‘something takes the man by surprise, enters into him and ill-treats him. This something may be powers or malign influences or poisons—which emanate from natural objects, places, solid or liquid foods—but also from fetishes, men, wizards. They may also be souls of any kind, which brush against the sick man and slip into him, or else definite souls which feed upon his vital forces, cause him pain, paralyse him, and trouble 1 Em. Rougier, ‘ Maladies et Médicins à Fiji Autrefois et Aujourd’hui,” Anthropos, ii. pp. 69, 999 (1907). 2 Junod, Les Ba-Ronga, p. 375. PRIMITIVES’ PRACTICES 265 his mind, etc.” In Laos, “all illnesses, whatever they may be, from the smallest to the most serious ailment, pro- ceed either from an angry spirit, or else from a dead person who is displeased. . . . Thay medicine knows scarcely any- thing of natural causes.” 2—In Bombay, we are told, that whatever be the illness which attacks man, woman or child, or even cattle, the Kolies imagine that it arises from the action of an evil spirit or an offended deity; and at the end of a certain period, after having vainly endeavoured to cure the malady by remedies known to them, they consult an exorcist able to expel evil spirits.3 At Bahr-el-Ghazal, “even where disease is not directly attributed to the machinations of enemies, the idea of ‘ possession’ seems to hold.” 4 In short, without laying further stress upon facts that are well known, the sick man is a prey to some malign power or influence. B. The practices relating to diagnosis are the direct out- come of this mystic idea of illness. It is important to discover what malign power or influence has taken possession of the sick man; what witchcraft is being exercised upon him; what being, living or dead, grudges him his life; and so on. This diagnosis, upon which everything else depends, can only be made by a man who is qualified to come in contact with mysterious forces and spirits, and who is powerful enough to fight them and expel them. The first step, therefore, is to appeal to the medicine-man, shaman, wizard, doctor, exorcist, or whatever he may be called; and if this person consents to undertake the cure, his first care will be to put himself into the special state necessary to be able to communi- cate with forces and spirits, and effectively exercise upon them his potential influence. This necessitates a whole series of preliminary operations which may last for several hours, or even a whole night. There must be fasting, or intoxication ; a special costume; magical adornments ; incantations ; dances to the extent of complete exhaustion and excessive perspiration ; until at last the “‘ doctor ’’ seems to lose con- t Dr. Pechuél-Loesche, Die Loango-Expedition, ili. 2, pp. 443-4. : A. Bourlay, ‘‘ Les Thay ’’ (Laos), Anthropos, ii. p. 620 (1907). 3 A. Mackintosh, ‘‘ An Account of the Tribes of Mhadeo-Kolies,’’ Trans. of Bombay Geo. Soc. (1836), i. p. 227 (1864 edit.). 4 Cummings, “‘ Sub-tribes of the Bahr el Ghazal Dinkas,” /.4.1., p. 156 (1904). 266 HOW NATIVES THINK sciousness or be “‘ beside himself.” Then there takes place what we should call the “doubling” of his personality. He has become unconscious of his entire surroundings, but on the other hand he feels himself transported to the world of intangible, invisible realities—the world of spirits—or at any rate he enters into communication with it. It is at this moment that the diagnosis is accomplished, intuitively, and consequently without any possibility or error: the patient and his entourage believe in it blindly. Here is one example among hundreds. “The most important among the para- phernalia of the shaman”’ (when about to treat a sick man) ‘‘is a head-dress made of a mat, which is worn in his incanta- tions. . . . Before putting it on they blow on it, and sprinkle it with water which has been poured over magic herbs. As soon as the shaman puts on the head-dress, he ‘ acts as though he was crazy,’ i.e. he puts himself into a trance by singing the song he had obtained from his guardian spirit at the time of his initiation. He dances until he perspires apa and finally his spirit comes and speaks to him.” 1 Since the diagnosis thus depends upon mystic price both necessary and adequate, very little attention is paid to physical symptoms. ‘‘In West Africa,” says Nassau, “ this diagnosis is not made by an examination and comparison of the physical and mental symptoms, but by drum, dance, frenzied song, mirror, fumes of drugs, and conversation with the spirit itself.” With the Cherokees, ‘ the description . . is always of the vaguest character, while in general the name given to the disease by the shaman expresses only his opinion as to the occult cause of the trouble. Thus they have definite names for rheumatism, toothache, boils, and a few other ailments of like positive character, but beyond this their description of symptoms generally resolves itself into a state- ment that the patient has bad dreams, looks black around the eyes, or feels tired, while the disease is assigned such names as “when they dream of snakes,’ ‘when they dream of fish,’ “when ghosts trouble them,’ ‘ when something is making something else eat them,’ ‘ when the food is changed,’ * F. Boas, “ The North-west Tribes of Canada,” Reports of British Association, pp. rats —6 (1890). 2 Nassau, Fetichism in West Africa, p. 215. PRIMITIVES’ PRACTICES 267 i.e. when a witch causes it to sprout and grow in the body of the patient or transforms it into a lizard, frog or sharpened stick.” 1 Moreover, this indifference concerning physical symptoms also proceeds from the mystic idea of illness. The seat of the malady is not within the body or the visible organs: it is the mind or spirit which is attacked. Thus, in West Africa, “the dogma that rules his (i.e. the doctor’s) practice is that in all cases of disease in which no blood is showing, the patient is suffering from something wrong in the mind.” 2—According to the Iroquois, “‘ every illness is a desire of the mind, and one does not die unless the desire is unfulfilled.—In Acadia, the sick man is refused nothing he asks, because in such a state his desires are orders from the guardian spirit,—and when wizards are called in, it is because they can best learn from the spirits the cause of the trouble and the suitable remedy. . . . It is the wizard’s duty to discover the sorcery which has caused the illness. He begins by putting himself into a perspiration, and when he is tired of shouting, writhing about, and invoking his guardian spirit, the first extraordinary thing that occurs to his mind is regarded by him as the cause of the trouble. Many of them, before beginning to sweat profusely, take a certain beverage which is calculated to help them to receive celestial impressions.” 3 C. The treatment. We can readily see that, whatever it may be, it is valuable solely on account of its mystic virtue. Its efficacy depends entirely upon connections and partici- pations of a spiritual or a magic kind. As a consequence, the therapeutics of white people are valueless. Their remedies may possibly do harm (on account of the unknown mystic proprieties in them); it is certain they can do no good, and they are quite ineffectual in dealing with the maladies of primitives. ‘‘ There is a woman here, ill for the last year. I went to her several times and asked her if she were willing to take medicine; her reply was ‘ the devil has caused this illness to me, and it cannot be cured by medicine. The tamtluanas (witch-doctors) only can cure me by driving the 1 J. Mooney, ‘ The Sacred Formulas of the Cherokee,” E. B. Rept., Vil. pp. 337; 368. ? M. Kingsley, West African Studies, p. 169. 3 Charlevoix, Journal d'un Voyage dans l’ Amérique Septentrionale, iii. p. 367. 268 HOW NATIVES THINK 212907 devil out of me. The strong repugnance felt to European remedies, of whatever kind they may be, is very general. From the prelogical point of view, this is inevitable, and we must remember that as a matter of fact the event often justifies it, especially in the case of individuals belonging to the most primitive types. In Victoria, for instance, “one doctor confessed that as a general rule, every time he had taken any sick native under his especial care, he succeeded only in killing him the sooner. . . . Taken to the bush, they rapidly recover.”” Why should this be so? “In the first place, in the hospital there is the feeling of loneliness; his spirits droop. Then... his hair, perhaps, has been cut, his old clothes taken away from him, and with them probably some valued possession on which his heart is set.” (He feels himself at the mercy of strangers who may exercise all sorts of sinister influences upon him without his being aware of it.) ‘ He fears the white man, dreads his medicines, and shrinks from the outward applications which may, for aught he knows, be possessed of secret properties that will cause his destruc- tion.’’2 Certain observers have clearly penetrated the reason for this dislike. Of New Zealand, Elsdon Best writes: “A great distrust of European doctors is manifest in this district. It is probable that this is not due to any disbelief in the medical knowledge of the said profession, but that the natives have an instinctive fear that a doctor will interfere with the state of tapu, that the life principle will be endangered by the methods of the Europeans being employed. A middle- aged woman of this district was taken seriously ill at Rotorua, — and it was proposed that she should be sent to the hospital. Her people strongly objected, urging her to adhere to native customs, saying that they would rather see her die than be operated upon by a European. However, she was taken to hospital by Europeans, was operated upon, and recovered. When she returned here, I heard an old woman asking her : ‘ In what state are you now?’ (meaning ‘is your fapu, your vital principle untouched ?’) The reply was: ‘O! every vessel of the white men has been passed over me.’ (Her body had been washed with water heated in a kitchen, x Solomon, ‘‘ Diaries Kept in Car Nicobar,” J.A.J., xxxii. pp. 231-2. 2 Brough Smyth, The Aborigines of Victoria, i. pp. 259-60. PRIMITIVES’ PRACTICES 269 and there is no stain worse than that.) Her fapu has gone, and she is clinging with great earnestness to European ways and customs as a means of protecting her vitality. But this is a rare case.’’! On the other hand, the Australian aborigines have infinite faith in the methods of treatment pursued by their native doctors, though admitting they could neither cure nor harm Europeans. “ To prove the imposture practised upon them by their Baangals, I have offered myself as a subject to be operated upon by any of them they might select, telling them that it would not be necessary to complete the process to prove their case, the mere fact of my being made slightly ill would be proof to me perfectly conclusive that their Baangals were all they claimed to be.—To them this offer of mine seemed so ridiculously absurd, they merely laughed at me, saying : ‘Stupid ears you. Too much you white fellow. Not that one Baangal belonging to you. What for you stupid head ?’”’ 2 Whatever may be the treatment prescribed for the sick man, the medicines he is to take, the regimen he must follow ; whether he is to have steam baths, phlebotomy, or, in certain cases, trepanning—it is the mystic forces alone to which their efficacy will be due. All investigators are agreed upon this point. Thus, the Dayaks “ don’t prize any drug, unless it be covered with mysterious passes, with numberless instructions of how to take it, in what position, and what incantation to repeat when looking at it. They can’t set a value or trust on anything, unless it is connected in some measure with the supernatural,’’3 To the Negritos of the Philippines, ‘“‘all disease is caused by spirits, which must be expelled from the body, before a cure can be effected.” #—Nassau is very definite about this. “To a sick native’s thought the alleviant medicinal herb used by the doctor and its associated efficiency-giving spirit evoked by the same doctor are insepar- able. . . . It is plain that the component parts of any fetich are looked upon by them as we look upon the drugs of t Elsdon Best, ‘‘ Maoiri Medical Lore,’’ Journ. of Polyn. Soc., xiii. pp. 223-4 (1904). 2 Beveridge, ‘‘ The Aborigines of the Lower Murray,”’ Journal of the Royal Society of New South Wales, p. 70 (1884). 3 Brooke, Ten Years in Sarawak, ii. pp. 228-9. W.A. Reed, The Negritos of Zambales, p. 68 (Manila, 1904). 270 HOW NATIVES THINK our materia medica. It is plain, also, that these drugs are operative, not, as ours, by certain inherent chemical qualities, but by the presence of a spirit to whom they are favourite media. And it is also clear that the spirit is induced to act by the pleasing enchantments of the magic doctor.’”’! Miss Kingsley expresses the same idea in a forcible way. “‘ Every- thing works by spirit on spirit; therefore the spirit of the medicine works on the spirit of the disease. Certain diseases are combatable by certain spirits in certain herbs. Other diseases are caused by spirits not amenable to herb-dwelling spirits ; they must be tackled by spirits of a more powerful grade.” 2 7 We must also take into account the fact that prelogical mentality does not deal with concepts as rigidly defined as those of our own thought. There is no clear-cut distinction between the mystic influence which, by means of certain processes, causes or cures a malady, and that which, without being of a medical nature, produces the same kind of effects, by altering the physical or moral disposition of men, animals, or invisible beings. Thus we have here a very general, though not abstract and truly conceptual, idea, similar to the con- ceptions of mana, wakan, orenda, mulungu, and others which | we have already studied. Some observers have remarked this particularly. ‘If the doctor and the magician are usually confused, the confusion is due to the fact that the concept of mori (medicine) is an extremely vague one. Mori does not stand for the medicinal plant, the healing herb alone; it means, too, magic methods of all kinds, among them those © which change the will-power. The heathen are convinced that if their children become Christians, it is because they have had a medicine, a mov1, administered to them; it is mort which makes abandoned girls attractive. Everything, even the powder with which the whites clean the rust off their weapons is mort.’ 3 Very frequently, when the “doctor” has finished his incantations and his magic practices, when he is in communi- cation with the spirits, he applies his lips to the diseased 1 Nassau, Fetichism in West Africa, pp. 81, 162. 2 West African Studies, p. 153. 3 Junod, Les Ba-Ronga, p. 468 (note). PRIMITIVES’ PRACTICES 271 part, and after more or less prolonged suction, triumphantly produces, before the patient and his friends, a small piece of bone or stone or coal or some other substance. They all believe that he has extracted it from the patient’s body, and that the cure is thereby assured, and even accomplished. We may compare this act to that of a surgeon showing his students a tumour which he has just removed, but the analogy is merely external. The fragment of bone or stone which the “doctor ”” takes out of his mouth is not the ill that the patient is suffering from ; it is not even the cause ; it is merely the vehicle of it. ‘The idea that pains are caused by foreign bodies embedded in the flesh of the sufferer is widely spread among uncivilized people throughout the world, as has long been known; but, as far as I know, it has not been noted that this foreign substance—at least among the Indians of Guiana —is often, if not always, regarded, not as simply a natural body, but as the materialized form of a hostile spirit.’’ : The true cause of the evil is the malign influence, the sorcery or witch-craft which has introduced into the body, with this fragment of bone or this stone, a principle of decay. Its extraction betokens the superiority of the ‘“ doctor’s”’ influence over this noxious principle: it is the visible sign of victory. But this victory is as mystic as the evilitself. It is always, as Miss Kingsley says, the working of spirit upon spirit. Nothing illustrates this characteristic of prelogical mentality better than the medical practices of the Cherokees, the formulas of which, with their explanation, Mooney received from the lips of the Indians themselves. Let us take, for instance, the formula respecting the treatment of rheumatism. It is composed of two parts: the first consists of invocations addressed in turn to the red, blue, black, and white dogs; the second gives a detailed prescription, showing how to prepare and to take the medicine.? This formula is a “particularly explicit”? one. It would, however, be quite unintelligible, at any rate in its first part, without the explana- tions afforded by the ideas, already noted, which Indians * Im Thurm, Among the Indians of Guiana, p. 333. ? J. Mooney, “ The Sacred Formulas of the Cherokee,” E. B, Rept., vii. pp. 346 et seq. 272 HOW NATIVES THINK have respecting maladies and the way in which they must be combated. The Cherokees’ most common belief with respect to rheumatism regards it as caused by the spirits of animals which have been slain, generally by those of deer, anxious to avenge themselves on the hunter. The malady itself, called by a figurative name signifying ‘‘ the one who enters,” is regarded as a living being. The verbs used in speaking of it demonstrate that this being is long and narrow, like a snake or a fish. It is brought by the chief of the deer, which makes it enter the hunter’s body (particularly his limbs and extremities) and immediately intense pain is felt. The ‘intruder ’’ cannot be chased away save by some animal- spirit stronger than, and a natural enemy of, the deer: as a rule, by the dog or the wolf. These animal-deities live in a higher clime, above the seventh heaven, and are the great prototypes, of which the animals on earth are but faint — reflections. As a rule, they dwell at the four cardinal points, each of which has a mystic name and a colour peculiar to all that pertains to it. (Here we recognize those complex participations which are always expressed in the collective representations of primitive mentality.)' Thus the east, north, west, and south are respectively the countries of the sun, the cold, the darkness, and wa’hd la’ ; and their colours are red, blue, black, and white. The white and red spirits are usually invoked to obtain peace, health, and other benefits of this sort ; red alone, the success of an enterprise ; blue, to unmask and defeat an enemy’s plans; black, to secure his death. The red and white spirits are regarded as the most powerful. That being so, in the formula for rheumatism the doctor first invokes the Red Dog in the country of sunshine, “ as if he were a long way off,” begging him to come quickly to the aid of the sufferer; then supplication gives place to the assertion that the Red Dog is there, and that he has removed part of the evil to the other end of the world. In the para- graphs which follow, the Blue Dog of cold, the Black Dog of darkness, the White Dog of wahala, are invoked in the same way, and each carries off part of the trouble. . . . t Vide Chap. V. p. 211. PRIMITIVES’ PRACTICES 273 As a rule, the formulas have four paragraphs. This one is exceptional, and has five. Then the physical part of the treatment is set forth. The remedy consists of a warm decoction made from the roots of four sorts of ferns, with which the patient is rubbed. This rubbing is done four times by the doctor, who at the same time recites the formula of invocation in a subdued voice: the first at dawn, the last at noon. Four is the sacred number which appears in all the details of these formulas—thus, four spirits are invoked in four paragraphs; the doctor breathes four times on the part attacked; there are four herbs in the decoction; four rubbings; and often taboos are imposed which last for four days. Another Cherokee formula is for the treatment of those whom “something is causing something else to eat them.” This malady especially attacks children of tender years. Its symptoms are excessive restlessness and disturbed sleep, the child waking suddenly and beginning to cry. What causes this illness? The birds. A bird has cast its shadow over the little one, or several birds have “‘ gathered in council ” in its body. (This latter is a favourite expression in these formulas to indicate the great number of the disease animals. They have “‘ formed a settlement or established a townhouse ” in the patient’s body.) ‘ The disease animal, being a bird or birds, must be dislodged by something that preys upon birds ; and accordingly the Blue Sparrow Hawk and the Brown Rabbit Hawk from above are invoked to drive out the intruders. . . . The remedy consists of a hot decoction of the bark and roots of certain plants. The bark is always taken from the tree on its eastern side, and the roots most fre- quently, if not always, selected from the same side. The bark and roots are not pounded, but merely steeped in hot water for four days. The child is then undressed and washed all over with this decoction, night and morning, for four days ME 1 It should be noted that in other places also we find a belief that the hunter’s illness is caused by the vengeance of the game. Thus, among the Bororo of Brazil, “a hunter falls ill or dies, who has done him this bad turn ? It is the spirit of an animal he has killed which is taking its revenge.” (K von den Steinen, Unter den Naturvôlkern Zenträlbrasiliens, p. 399.) Cf. Schoolcraft, Information, etc., ii. p. 180. 2 The Sacred Formulas of the Cherokee, pp. 355-0. 18 274 HOW NATIVES THINK D. The materia medica used by the “ doctors’”’ and the medicine-men of primitive communities suggests similar reflections. Their knowledge of simple remedies varies very much. In some cases their skill amazes explorers ; in others, like the Cherokees, for instance, their actual resources are very scanty. But even supposing that for a certain disease they should prescribe the same medicine as our doctors would do, they would do it in quite another spirit. Nearly always the chief consideration, for them, is to expel the influence or spirit whose presence causes the trouble, or to make the sufferer participate in some known or supposed virtue in the remedy which will enable him to surmount it. In the latter case it is ‘‘ sympathetic ’”’ therapeutics, practised almost everywhere, and which European practitioners were still employing three centuries ago. To take but one example only, in British Columbia, ‘“ decoctions of wasps’ nests or of flies are drunk by barren women, to make them bear — children, as both bring forth many young.” ! Facts of this kind, as we know, are very numerous. In any case, the healing virtues of the drug are as a rule governed by a good many conditions. If it is composed of plants, they must have been gathered by certain persons, at a given moment, with special incantations and instruments, the moon being in such and such a phase, etc.: if these conditions are not fulfilled, the remedy will not operate. In Canada, “before the departure of a war-like expedition . . the whole village being assembled, one of these medicine- men declares that he will communicate to the roots and the plants of which he has a large store, the power of curing all sorts of wounds, and even of raising the dead to life. They immediately begin to sing; other tricksters reply, and it is supposed that during the concert, which is accompanied by many contortions on the part of the actors, the healing virtues are bestowed upon the drugs.” »—Among the Chero- kees, the “ doctors’’ who go in search of the plants and bark and roots must conform with very complicated conditions, which Mooney has not been wholly able to give in detail. 1 F, Boas, “ The North-west Tribes of Canada,” Reports of the British Association, p. 577 (1890). * Charlevoix, Voyage dans l’Amérique Septentrionale, iii. pp. 219-20. PRIMITIVES’ PRACTICES 275 The shaman must be provided with a number of white and red beads, (of the kind which play an important part in magical operations, when they begin to move in the shaman’s hands, for then the Indians believe that they are alive). He must approach the plant from a certain direction, going round it from right to left one, or four, times, reciting certain prayers all the time. Then he pulls up the plant by the roots, drops one of his beads in the hole, and covers it over with soil. . . . Sometimes the shaman must leave the first three plants he sees, and take the fourth only, after which he may go back to the three others. The bark is always removed from the eastern side . . . because it will have received more healing power by being subjected to the sun’s rays.! If the sufferer is cured, all goes well, and the professional healer receives the promised guerdon and an expression of gratitude from the relatives. But if, in spite of his efforts, the issue is unfavourable, it rarely happens—though cases have been known—that he is held responsible. In certain communities where there already is some slight degree of political organization the task of looking after kings and great personages may be fraught with danger. In more primitive societies, failure is as a rule attributed to “the malignant action of a superior magic on the part of some hostile spirit or individual.’ The medicine-man will not be alarmed, but a fresh question will arise: what is the spirit, and above all who is the enemy whose witchcraft has proved so power- ful? But in a general way, since the idea of illness, its causes and its treatment, remains a mystic one, the failure of the efforts made for the patient are accounted for as easily as their success. It is the more powerful “‘force’’ or “influence” or “ spirit’? which triumphs, establishing or rupturing participations upon which life and death depend. According to the primitive’s mentality, nothing is more natural. Finally, certain observations made prove that the dis- tinction between maladies of mystic origin and those pro- ceeding from causes which we should call natural, has begun to be instituted: either the same malady arises from a 1 J. Mooney, The Sacred Formalas of the Cherokee, p. 339. 2 Spencer and Gillen, The Native Tribes of Central Australia, p. 531. 276 HOW NATIVES THINK mystic or non-mystic cause, according to circumstances, or else they recognize categories of maladies which are originally different. With the Kafirs, for instance, ‘when a witch- doctor has diagnosed a case of illness, there are three possible conditions: firstly, the malady may have arisen of itself; secondly, the spirits of his ancestors have caused it; thirdly, it is due to witchcraft.”* Among the Bahima, “sickness is accounted for in four ways: (1) It is thought to be caused by the departed king, who has been offended in some way ; the Mandwa (chief priest to the king) is the only person who can assist in such a case: paralysis is attributed to this source. (2) It is set down to witchcraft (kuloga), which is practised by a person with the desire to kill another secretly : (this may be any form of disease). (3) The fever is due to natural causes; no person is held responsible for it. (4) Illness is attributed to ghosts (muzimu), which take possession of people for various causes, and have to be exorcised.’ These classifications are somewhat confused, apparently ; yet they mark a transition from the wholly mystic idea of illness and the treatment which must be applied to it, to the methods of thought and action in which observa- tion and experience have a little more hold. IT The mystic idea of illness has a number of corresponding customs which testify to the same prelogical mentality. So, too, the equally mystic idea of death finds its expression in a number of customs relating to the dying and dead, which have been noted by investigators in most primitive peoples, and which would be altogether incomprehensible if we could not interpret them by such a mentality In the first place, death is never natural. This is a belief common to the Australian aborigines and to the but slightly civilized tribes in the two Americas, Africa and Asia. ‘ The native,” according to Spencer and Gillen, ‘‘is quite unable to realize death from any natural cause.’ 3—‘‘ To the mind 1 Fr, Aegidius Müller, “ Wahrsagerei bei den Kaffern,”’ Anthropos, p. 43 (1907). + J. Roscoe, ‘‘ The Bahima, a Cow Tribe of Enkole,”’ J.A.T., XXxXvii. p. 103. 3 The Native Tribes of Central Australia, p. 356. PRIMITIVES’ PRACTICES 277 of the Muganda,” says Roscoe, “ there is no such thing as death from natural causes. Both disease and death are the direct outcome of the influence of some ghost.’’ !—With the Fang, “‘ death is never considered due to natural causes. Disease followed by death is due to evus (witches).’’ ?—Du Chaillu says the same. “The greatest curse of the whole country is the belief in amemba (sorcery or witchcraft). The African family firmly believes death to be always a violence. He cannot imagine that a man who was well two weeks ago should now be lying at death’s door with disease, unless some potent wizard had interfered, and by witchcraft broken the thread of life and inflicted sickness.’’ 3—‘‘In ancient times the Cherokees had no conception of anyone dying a natural death. They universally ascribed the death of those who perished by disease to the intervention or agency of evil spirits and witches and conjurers who had connection with the . . . evil spirits... . . A person dying by disease, and charging his death to have been procured by means of witch- craft or spirits, by any other person, consigns that person to inevitable death.” 4 When observers tell us that “natives are quite unable to realize death from natural causes,” the dictum comprises two statements which it would be well to distinguish. The first of these means that the cause of death, like that of disease, is always represented as a mystic one; how could it be otherwise? If every malady is the deed of a “spiritual influence ”” or ‘‘force’’ or “spirit ’’ or “ ghost,” which is acting upon, or in possession of, the patient, why should not the same cause be assigned to the fatal issue of the malady ? That which would really be incomprehensible would be for the primitive mind to conceive of a “ natural death.” It would be a unique idea, quite unlike all the rest. It would mean that this one alone, the most impressive and mysterious of all phenomena, should, through some totally incomprehensible exception, have been released from the mystic sheath which still encloses all the rest. cs t J. Roscoe, “‘ Manners and Customs of the Baganda,” J.A.J., xxxii. p. 40. 2 Bennett, “ Ethnographical Notes on the Fang,” J.A.I., xxix. p. 95. 3 Du Chaillu, Equatorial Africa, p. 338. 4 Haywood (1823), quoted by J. Mooney, ‘“ The Sacred Formulas of the Cherokee,” E. B, Rept., vii. p. 322. 278 HOW NATIVES THINK In this respect, there is nothing more significant than the well-known cases in which death has overtaken a person who is aware that he has violated some taboo, even inadvertently. Frazer has quoted a number of these ;! and here is another characteristic one. ‘The lad was strong and healthy, until one day when Mr. MacAlpine found that he was ill. He explained that he had ‘stolen some female opossum’ before he was permitted to eat it; that the old men had found it out, and that he should never grow up to be a man. In fact, he lay down under the belief, so to speak, and never got up again, and he died within three weeks.’’2 Here we have a type of “natural death ” as the primitive’s mentality regards it, if we may venture to lend him the expression. No less “natural’’ is the death which will assuredly take place if the man has but been lightly scratched by a weapon which has been bewitched. ‘‘ There is no doubt whatever that a native will die after the infliction of even a most super- ficial wound if only he believes the weapon which inflicted the wound had been sung over and thus endowed with arungquiliha. He simply lies down, refuses food and pines away.” Spencer and Gillen have been witnesses of many cases of this kind.3 The only possible cure for a man wounded by a “charmed” spear is the exercise of strong counter- magic.4—Yet more, the mere fact of stimulating the practice of magic suggests to an old man that he himself may have become contaminated, and the travellers seem to have feared that this idea of his might cost him his life. ‘“ After the old man had vigorously jerked the pointing stick towards an imaginary enemy, he himself was evidently rather upset, and told us that some of the arungquiltha or evil magic had gone up into his head. The natives are people of the most wonderful imagination, and we thought at first it was going to affect him seriously.’’5 But this is not a case of even a wonderful imagination being affected. It is the expression of a fear, very natural from the primitive’s point of view. The case of the old Australian native might be compared with that of a surgeon who, in giving a demonstration on a * The Golden Bough (2nd edit.), i. pp. 321 et seq. ? Howitt, “On Australian Medicine-men,”’ J.A.I., xvi. Pp. 42 (note). 3 The Native Tribes of Central Australia, p. 537. 4 The Northern Tribes of Central Australia, Pp, 675. 5 Ibid., p. 462. PRIMITIVES’ PRACTICES 270 corpse, thinks he has pricked himself, and may have become infected. The second assertion to be distinguished implies that death is never due to natural causes, because it is always violent ; in other words, it is always a crime, an assassination desired, premeditated and accomplished by a certain person by means of magic practices. This leads to those terrible and only too frequent trials for witchcraft, specially common in Africa, of which travellers have given such striking accounts. Nassau even sees in them one of the causes of the depopula- tion of the Dark Continent. There is no district, however, where this belief applies to every kind of death without exception. Such actions are not instituted when it concerns the death of children of tender age, slaves, or people of no importance, as arule. Inquiries are only made when dealing with suspicious deaths, and persons about whom it is worth troubling. It is true, however, that among these peoples there are infinitely more suspicious deaths than there are among us. On the one hand, the practice of magic is current there, and everybody makes use of it more or less. No one can do without it, nor has he even any idea of avoiding it: everyone is more or less disposed to suspect his neighbour of practising it upon occasion, and each in his turn becomes the object of a similar suspicion. On the other hand, the very idea, common to all, that illness and death are always due to mysterious influences, leads easily to the conclusion that the death has been a violent one, in the sense that these forces have been set to work by the will of an enemy. Hence, among primitive peoples, it often happens that the deaths which are the most “natural’’ in our eyes, being referred to mystic causes, are regarded as violent, to the utter disregard of all that seems to be ‘the evidence. It is a point upon which the difference between our mental habits and those of primitives is most clearly attested. In the Torres Straits, for instance, ‘ death from snake bite is generally supposed to be due to the snake having been influenced by a sorcerer.’’ 1—“ The natives (of Port Lincoln) are not content even when the cause of death is sufficiently clear, but seek t Seligman, ‘“ The Medicine, Surgery, and Midwifery of the Sinaugolo (Torres Straits),” J.A.I., p. 299 (1902). we HOW NATIVES THINK to find a hidden cause. . . . A woman, while clearing out a well, was bitten in the thumb by a black snake. It began to swell immediately, and in the short space of twenty-four hours the woman was a corpse. Still it was asserted that it was not an accident, but that the deceased had pointed out a certain native as her murderer. Upon this evidence, which was heightened by the circumstance that no blood flowed from the wound, the woman’s husband and his friends challenged the accused and his friends to combat. Peace, however, was made; and upon the offensive side it was acknowledged that the woman was in error with regard to the guilty person. But still not satisfied that the snake bite should have been the cause of the death, another indivi- dual was suddenly discovered and accused.”’ 1 The same method is followed when an old man dies of senile decay: his relatives seek to discover whose witchcraft is responsible for his death, and take vengeance upon the supposed author. Here is a yet more characteristic case. ‘‘ The Melbourne natives lost a man of their tribe, generally supposed from natural causes. A number of the deceased’s friends resorted to the usual mode of trench-digging,? and strictly in accordance with the straw-pointing, proceeded to Joyce’s Creek, and there at midday attacked a party of natives who were at the time hunting, and killed a fine young man... . The friends of this young man, although eye-witnesses to his murder, and in the full knowledge of who the guilty party were, proceeded in the usual way to tie up the body and dig the trench. The straw pointing in the direction of the Goulbura, a strong party, consisting of eighteen men, were then equipped with spears, etc., and in about a week from the Joyce’s Creek tragedy a similar life was taken by this party in the locality named.’’ 3—However incredible such a fact may appear, Dobrizhofter quotes similar instances among the Abipones. ‘If an Abipone die from being pierced with many wounds, or from having his bones broken, or his strength exhausted by extreme old age, his countrymen all deny that wounds or weakness occasioned his death, and t Wilhelmi, ‘“ Manners and Customs of the Australian Natives,” etc. Roy. Soc., Vict., v. p. 191 (1860). 2 Vide p. 282. 3 J. Parker, in Brough Smyth’s The Aborigines of Victoria, ii. pp. 155-6. PRIMITIVES’ PRACTICES 281 anxiously try to discover by which of the jugglers and for what reason, he was killed.” ! Such customs as these are best suited to show clearly the extent to which the mentality of primitive peoples is oriented differently from our own. Both the Australian aborigines and the Abipones perceive, as we do, that very serious wounds will inevitably be followed by death. But their thought does not stop there, for their collective repre- sentations oblige them, as it were, by some form of pre- conception or preconnection, to refer the death to a mystic cause. The wound, then, can only be one of the ways in which this mystic cause fulfils its end: it might just as well have been a snake bite or a suffocation by drown- ing. It is no use stopping short at the means employed, The only thing that matters is the true cause, and among certain peoples at any rate, this cause is always mystic in its nature. The methods employed to discover this cause naturally correspond with the idea formed of it, and are no less characteristic of prelogical, mystic mentality. They have recourse to divination, and the guilty person indicated is executed on the spot. In Africa,—with the Kafirs, for instance—or in the French Congo, and in other parts of the Dark Continent, where trials for witchcraft are extremely common, the following is an abridged summary of the method of procedure. As soon as the death of an important person, or one that appears suspicious, has taken place, all the t An Account of the Abipones, ii. p. 84. 2 In peoples slightly in advance of these, we find certain transitional forms. Death, like illness, is still in certain cases attributed to mystic causes ; but in others it is considered natural, in the current meaning of the term, We are told that among the Nez-Percés, for instance, the chiefs say that they and their sons are too great to die of themselves. There is no doubt that they can become ill, grow weaker, and die like others, but it is because some person or some malign spirit incited by him has imperceptibly brought the death about. This is the reason that, when a chief or his son dies, the author of the supposed crime must be executed. (Parker, quoted by Bancroft, The Native Races of the Pacific States of North America, iii. p. 157.) So, too, the missionary Brun says: “‘ According to what our colleagues in Equatorial Africa tell us, many negro tribes believe that every man’s death is caused by sorcerers or spirits, Among the Malinkas who surround .us, this belief is not so absolute, and they do attribute many deaths to their true natural causes, such as illness, old age, starvation, or some kind of accident.” (Brun, ‘ Note sur les Croyances des Malinkes. Côte Occidentale Frangaise,”’ Anthropos, ii. p. 948.) 282 HOW NATIVES THINK relatives, servants, and often the whole population of the village, collect, and the ‘‘ medicine-man’”’ begins the magic practices which are to reveal the identity of the guilty person Miss Kingsley has furnished us with a striking picture of these dread assemblies, where the very bravest trembles at the thought of finding himself pointed at and, in a few seconds, condemned and done for,—the object of public execration and hatred, and without the slightest hope of establishing his innocence. Sometimes, indeed, he is executed on the spot. Sometimes he has to undergo an ordeal,—he must swallow a certain amount of poison, for instance—and those who prepare the dose have decided beforehand what its result shall be. Europeans who are witnesses of these dramas cannot but regard them as a horrible travesty of justice. But the tenacity with which the natives defend such practices proves that, in their eyes, at any rate, they are strictly bound up with highly important group ideas regarding illness, life, _ death, and social order. Thus, as they put it, they are “indispensable,” from the point of view of prelogical, mystic mentality, however ridiculous they may appear to our logical thought. In the least civilized peoples known to us, it is noteworthy that divination is frequently resorted to in order to discover the point of space in which the criminal may be found. There is a practice, very common in Australia, which consists of digging a trench in the place where the body has been buried, and noticing the side towards which an insect turned up by digging will betake itself. ‘‘ During the process of digging,”’ says Grey, “an insect having been thrown up, its motions were watched with the most intense interest, and as this little animal thought proper to crawl off in the direction of Guil- ford, an additional proof was furnished to the natives of the guilt of the boyl-yas of that place.’’! This is the practice of trench-digging recently mentioned by Brough Smyth, who has moreover provided us with a kind of synopsis of the methods employed. ‘The Western Port tribe in Victoria, and the tribes near Perth in Western Australia watch the movements of a living insect that may accidentally be turned 1 Grey, Journal of Two Expeditions of Discovery in North-West and West Australia, ii. pp. 325-6. PRIMITIVES’ PRACTICES 283 © up in digging the earth; the Melbourne tribe look for the track of a worm or the like; the Yarra black watch the direction which a lizard takes ; at Cooper’s Creek the corpse is questioned; the tribes at the mouth of the Murray and at Encounter Bay rely on the dreams of a wise man who sleeps with his head on the corpse; and on one part of the Murray they watch the drying of the damp clay that covers the grave, and see in the line of the principal fissure where they are to look for the sorcerer.’’'—We are told that in Central Australia a day or two after the death, there is a ceremonious visit to the exact spot where it took place, and the little mound raised there, as well as the soft earth surrounding it, are both examined very carefully to see whether any trace of the criminal is to be found. “If, for example, a snake trace were visible, then this would be regarded as a sure sign that a man of the snake totem was the culprit, and then there would remain the task of finding out which particular snake man was guilty.’’ The explorers further relate that if no traces have been seen, the relatives wait until the body begins to decompose, and then the widow’s father and brother very carefully examine the liquid that has flowed from the bier upon which the body has been laid, for they believe that the direction taken by the liquid indicates that from which the murderer has come. If the liquid has stopped some way off, the man is still quite near, but if it flows to a yet greater distance, the natives know that the guilty person belongs to a distant tribe.? In New Guinea, too, among other methods of divination, “according to Kunze, some betel ash and a crab are put into the dead man’s hand, and a string is attached to his little finger. When the grave has been closed, the string is pulled, and someone walks over the grave calling out ‘Arise!’ The movement of the string makes the crab uneasy, and it moves about. While moving it scatters the ash all around it, and from the direction in which the ash falls, the locality of the criminal can ultimately be found.’3 In Guiana, too, ‘‘ Schomburgk informs us that even the death of t Brough Smyth, The Aborigines of Victoria, p. 28. 2 Spencer and Gillen, The Northern Tribes of Central Australia, pp. 526-9. 3 Hagen, Unter den Papua’s, p. 256. 284 HOW NATIVES THINK a man who has been attacked by disease is imputed to an unknown kanaima, or sorcerer. He has seen the father of a child which had died of dropsy cut off its thumbs and toes and throw them into a vessel full of boiling water, which all the relatives watched most attentively, for it was on the side upon which the boiling water first threw up a finger that the unknown murderer was to be found.” ! All these practices demonstrate the peculiar importance attaching to spatial relations in the primitive mind. We have already had numerous proofs of this—in particular, of the care taken in most primitive languages to give the express distance, direction, and height, necessary to find the person or thing spoken of. In part, undoubtedly, this care may be explained by the piciorial character of languages which are still but slightly conceptual, but it is probably also derived from the attention paid by the primitive to spatial directions. . And this very attention arises out of the mystic value of the spatial directions (the cardinal points) and the manifold participations in which these directions are involved. Proofs of this have already been given,” such as the “ local relation- ship’’ of the Australian natives, the mystic symbolism of the Zuñis and the Cherokees, who attribute a colour, an animal, a value of its own, to each point of the compass. The practices we have just described rest upon the belief in a similar participation—so much so, that if the trace of a snake be discovered on soil that has recently been disturbed, it is assuredly a man of the snake totem who has caused the death ; so, too, if an insect thrown up by a digger’s spade crawls to the north, it is a sure sign that a man of a northern tribe is the criminal. If we try to regard this as a piece of deductive reasoning, we shall never arrive at any thing but an absurdity ; it is not a deduction or an argument at all, however; it is a mode of activity peculiar to prelogical mentality, and, for this very reason, almost incomprehensible to us. To such a mentality, there are no accidental circum- stances. The insect which took the northerly direction might just as well have crawled to the west or the south, or anywhere 1 Von Martius, Beiträge zur Ethnographie Stid-Ameyvika’s, i. p. 651. 2 Vide Chap. V. pp. 211 et seq. PRIMITIVES’ PRACTICES 285 else. If it selected the north, therefore, there must be some mystic participation between the spatial direction and that which the primitive mind is seeking to ascertain at that particular moment. That which logicians designate the fallacy of post hoc, ergo propter hoc, may help us to form some idea of this participation. An extraordinary vintage occurs in the autumn of a year in which a great comet has been visible during the summer; a war breaks out after a total eclipse of the sun. Even to the minds of peoples already civilized, such con- nections are not accidental. The relation of events to each other in time does not consist solely of succession ; a relation, difficult to analyse clearly, connects the abundance of wine with the comet and the war with the eclipse. Here we have an obstinate survival of that which we have called partici- pation. But minds of the most primitive type, which recognize no accidental relations, i.e. minds which assign a mystic signification to all the relations which may figure in their ideas say just as readily juxta hoc, ergo propter hoc, as they do post hoc, ergo propter hoc. Contiguity in space is just as much of a participation as contiguity in time—even more so, to the extent that such minds pay more attention to the spatial determination than to the temporal one. Accordingly, in those complex relations of mystic partici- pation which, in the mass, are to prelogical mentality what causality is to logical thought, juxtaposition sometimes plays the part which we should attribute to sequence. We are told by Gatschet, for instance, that it was formerly admitted by universal custom that Indians had the right to attack and slay their neighbours if an owl were heard screeching at night near the hut of the latter. And from a brief tale of the Klamath Indians we learn, too, that a dog having howled outside a certain cabin just as the sun went down, an Indian appeared, to attack, wound, and finally kill its owner.z Here is the juxta hoc principle at work: it is parti- cipation, manifested by contiguity in space, between the Indian at whose door the animal of evil omen is heard, and the misfortune announced—and consequently, in a certain t A, Gatschet, The Klamath Language, p. 89. a Ibid., p. 133. 286 HOW NATIVES THINK sense, caused by the animal. It is noteworthy that in nearly all the Indo-European languages the prepositions which denote ‘on account of,” ‘by means of,” are words which originally referred to spatial, and not temporal, relations.? Possibly the primitive mind at first conceived of relations of time as relations of position, or rather, juxtaposition, and, since it recognizes no relations save those which are of mystic significance, as participation by means of contiguity. It is to contiguity, accordingly, that the primitive mind clings, and whether the one factor of the relation precede the other in point of time, or follows it, is quite a secondary, and possibly indifferent, matter. In Torres Straits, ‘ mishaps or unlucky events are regarded as warnings or omens that something is going wrong somewhere, or wll shortly do so... In 1888, Nomoa, the then chief of Mabuiag, who has perhaps killed more dugong than any other man, one day boasted to me that he was invariably successful. Very shortly after . this, he went out to harpoon dugong, and had the misfortune not only to fail in attacking but also to break the dart of his dugong harpoon. I think he made an unsuccessful attempt the following day. Within three or four days, first a baby died in the village, and then two women. Nomoa at once told me that this accounted for his bad luck, and he was quite happy in the belief that it was not his fault that he had missed the dugong.” 2 There is then a mystic bond, appreciated by the primitive mind, between the unsuccessful fishing and the misfortunes which occur some days afterwards. But it would be difficult to say which is cause and which effect, if the cause is taken to mean antecedent, since the unsuccessful fishing, on the one hand, is accounted for by the deaths which occur afterwards, and, on the other hand, this unsuccessful fishing is the harbinger, and thus, in a certain sense, the cause of the deaths. So, too, in North America, “‘ they consider an eclipse as an augury of death or war or sickness: but this augury does not always precede the misfortune it prognosticates ; it sometimes follows it; for the savages, having seen the lunar eclipse in this year (1642) said that they were no longer 1 Verbal communication (A: Meillet). 2 The Cambridge Expedition to Torres Straits, v. p. 361. PRIMITIVES’. PRACTICES 287 surprised that the Iroquois had slain so many of their people during the winter; they saw in it the omen and the sign, but somewhat too late to be able to take precautions.” : We find similar beliefs elsewhere. In China, for instance, as De Groot informs us, the spirits of inanimate objects often manifest their baleful presence by predicting misfortunes, and to these simple, illogical minds, this is equivalent to preparing them and bringing them about. Books frequently tell of deaths, conflagrations or other calamities following upon the fact that things have tumbled over without any apparent cause. Here again, it is not the temporal relation of the two occurrences which interests the primitive mind, for that is mainly intent upon the mystic participation which unites them. The Lolos recognize three bad influences which inflict sicknesses and misfortunes. They are (1) the ghosts of those who have died unclean deaths; (2) demons; (3) the slo-ta, by which they designate any unusual appearances, phenomena contrary to nature, which not only portend but also cause disaster; they are monstrous births, hens that crow like cocks, etc.3—We find the same beliefs and the same practices in South Africa, where the negroes try to combat these “unusual phenomena’ by suppressing them. They call them “‘zlolo,”’ translated by Livingstone as “‘ transgression.” Albinos are generally put to death. We are told that a child that cuts its upper teeth before the lower was always dis- patched by the Bakaa, and probably by the Bakwains also. In some tribes, only one of twins was allowed to survive (though this may have been for other reasons also). If an ox, while lying in the pasture, beat the ground with its tail, it was treated in the same way, because it was believed that it had invited death to visit the tribe. When Livingstone bearers came _ through Londa, they brought a specially large breed of poultry with them, but if any of them crowed before midnight, it was guilty of #lolo, and killed.4 1 Relations des Jésuites, xxii. pp. 191-6. 2 J. J. M. de Groot, The Religious System of China, ii. p. 664. 3 A. Henry, “‘ The Lolos and other Tribes of West China,’”’ /.4.1., xxxiii. P. 104. 4 Livingstone, Missionary Travels, p. 577. Cf. Baumann, Usambara und seine Nachbargebicte, p. 43. 288 HOW NATIVES THINK III Accordingly, every unusual phenomenon is considered as a sign, and at the same time as a cause of the misfortune which will occur later; but, from another point of view, and just as correctly, this misfortune may be regarded as the cause of the unusual phenomenon. Therefore, if we interpret these collective representations by the law of causality, which implies an invariable and irreversible order in time between the antecedent cause and the consequent effect, we distort them. As a matter of fact, these representations obey the law of participation, a constitutional law of prelogical mentality. It is impossible to reduce to a logical analysis the mystic connection that unites any unusual phenomenon with the misfortune of which it is the sign. Now the phenomena which play the part, so often difficult to explain, of harbingers, occur but rarely, and the world. in which the primitive lives comprises an infinity of mystic connections and participations. Some of these are unvarying and recognized, such as the participation of the individual in his totem, the connections of certain animal or vegetable species with each other, and so forth. But how many others that it is of great, often of vital, interest to know, im- perceptibly establish themselves, or disappear. If these connections do not manifest themselves, their appearance must be incited, and this is the origin, or at least one of the main origins of divination. We must remember that to the primitive mind the external world had another orientation than ours; for the primitive’s perceptions are mystic, Le. the elements of perceptions which, to logical thought, are objective and alone real, are enveloped in his mind in an un- differentiated complex of mystic elements. And these mystic elements, as well as their connections, though invisible, intangible, and inaccessible to the senses, are nevertheless the most important of all. It is imperative that he should know all there is to know about them, and it is this end which is served by divination. To the primitive, divination is an added perception. Just as we employ instruments to enable us to perceive things which are too fine to see with the naked eye, or to supply PRIMITIVES’ PRACTICES 289 senses in which we are deficient, the primitive mind makes use of, first and foremost, dreams, then the conjurer’s wand, the crystal, the ossicles, mirrors, the flight of birds, ordeals, and an almost infinite variety of other devices, to enable him to apprehend mystic elements and connections which do not otherwise reveal themselves. His desire for know- ledge is even more imperative than our own, for our general idea of the world could, if need be, do without the elements which the instruments of modern science have furnished. In its essential features it was constituted before these were discovered, whilst to primitive mentality, from its very con- struction, divination is absolutely indispensable. The more the mystic elements and connections predominate in their collective representations, the greater the need of mystic processes to enable them to be informed about them. As a matter of fact, no practices are more widely prevalent than those of divination. I do not believe that there is a single primitive community in which they are altogether wanting. Undoubtedly it is in peoples already somewhat civilized that we find divination as a complex and refined art, with a commonalty and a hierarchy of diviners. But even in the very lowest types we know, divination is already practised, at any rate by means of dreams. It is familiar to the Australian aborigines and to the most primitive social groups of the two Americas. To quote but one example only, we are told of a tribe of Eastern Brazil where a chief made it his business to harangue his followers on the eve of an engagement with the enemy, telling them that they must all recall the dreams they would have that night, and endeavour to have only those of good omen.! To represent the practices of divination as merely designed to reveal the future, is to define them in too limited a sense. It is true that among the mystic connections that these people seek to discover, those which determine that which is to be are of special importance and greater interest. It is these which decide whether one is to act or to refrain from acting, according to the circumstances of the case. But divination, considered in itself, refers to the past as well as to the future, 1 “ The Captivity of Hans Stade in Eastern Brazil, 1547-1555," Hakluyt Soctety, li. p.108 3) Cf, D: “152, 19 290 HOW NATIVES THINK and this is proved by the important part it plays in criminal investigations. In the trials for witchcraft, for instance, the tribe and the individual offender are nearly always pointed out by means of divination. It is the same, too, if they seek to discover who has bewitched a sick man, or what evil spirit possesses him, where a missing article is to be found, whether a man who has not recently been heard of be still alive, etc. “The more one gets to know these tribes intimately,” says Junod, “‘ the more surprised one is to see how important a part the ossicles play in their affairs; they intervene of necessity in all events, however unimportant, both in the career of individuals and in the national life.” 1 De Groot remarks the same about the Chinese, and explorers in general do not fail to state that ‘‘ savages ”” are very ‘‘ superstitious.”’ That means, in our view of the matter, that they act in con- formity with their prelogical, mystic mentality. It would be surprising, and even inconceivable, ii they were not ‘‘ superstitious. ’’ . With such a mentality, indeed, the consultation of the diviner is an almost indispensable preliminary to every enterprise, frequently even to those which a European would consider the most commonplace, such as resuming the march after a night passed at a stopping-place. It may happen that the native porters manifest the greatest unwillingness, or even, if they dare, refuse to start. The white traveller, (as Miss Kingsley remarks,) unless he knows his men’s mentality thoroughly, will perceive in this nothing but laziness, insub- ordination, failure to keep a promise, incorrigible bad faith, whereas it may possibly be something altogether different. Perhaps when they awoke, one of the negroes perceived a sign which augured ill either for him or for the entire company ; hence their stubbornness. The consultation in this case is offered of itself ; when not offered, it is induced. For if, by virtue of these invincible mysterious relations, that which they are about to undertake should turn out ill, it is as | unreasonable to risk it as it would be for us to contravene natural laws, the law of gravity, for instance. And how are they to know this, save by divination ? Moreover, it will not suffice to resort to it to make sure of 1 Junod, Les Ba-Ronga, pp. 455 et seq. PRIMITIVES’ PRACTICES 201 the success of the enterprise as a whole ; the omens must be consulted and the dreams interpreted at each stage, and almost at every moment. This is a matter which many explorers have demonstrated. In war, hunting, in nearly every case in which individual or collective activity has some end in view, nothing is done without:the favourable opinion of the diviner, the medicine-man, the witch-doctor. If success is achieved it is to the strict observance of his injunctions that it is due. ‘ Yes,” said a Dyak chief to Rajah Brooke one day, “‘all my people are well off this year for padi, because we have paid every attention to the omens . . . and appeased the Antus by taking alligators, killing the pigs and examining their hearts, and we have judiciously interpreted our dreams. The consequence is a good harvest, but those who have neglected to do this are still poor, and must pay more atten- tion in the future.’’* When on a campaign, all the move- ments made by these same Dyaks depend upon omens. They cannot advance nor retreat, nor attack nor change their position until the auguries are known. “I have known a chief who lived in a hut for six weeks, partly waiting for the twittering of birds to be in a proper direction, and partly detained by his followers. . . The white man who commands the forces is supposed to have an express bird and lucky charm to guide him always; and to these the Dyaks trust considerably. ‘ You are our bird, we follow you.’ . . . Besides the whole way in adventuring, their dreams are religiously interpreted and adhered to. . .. I well know the names and can distinguish the songs of their birds... the effect of these signs on myself was often very marked. . A maia’s head (the orang-utan) was hanging in my room, and oe was thought to be my director to successful expeditions.” ae to Cushing, many of the Zuñi games were a kind of divining. For instance, the game of the hidden ball would be played by two camps, representing east and west, or north and south, and each camp consisted of members of the corresponding clans. The indication would be given by the result of the game. And since the war-dance was like a preliminary or a souvenir of the battle, represented 1 Ten Years in Sarawak, i. p. 203. 2 Ibid., ii. pp. 234-5. 292 HOW NATIVES THINK beforehand as a kind of drama, to determine which side would be victorious—so too this game, celebrating a mythical decision made by the gods (especially the gods of the wind and the gods of the water), was a means of interrogating fate as to which party would prevail. It would show, too, whether the gods of the wind or of the water would dominate, whether the season would be dry or rainy, and, according to the number of points gained by each camp, would reveal the extent of this drought or humidity. The players on one side would represent the north and the winter, the windy, barren season; those on the other side, the south and summer, fertilizing showers; the former stood for drought and the latter for moisture. Consequently they could regulate their plans according to the points gained by the prospective camps, and if the camp representing the wind won, they would take care to plant their seed deeper, and in well-watered spots.! | This interpretation of Cushing’s is valuable in more ways than one. Not only does it “‘illustrate”” the idea that games have a prophetic significance, but it shows how divination serves to obtain exact indications how to proeed, as well as a revelation of the future. Itis certainly important, above all, that the Zuñis should know whether they may expect rain, or not. For them it is almost a question of life or death, and many of their games, among other practices which are now fairly familiar in their details, have the religious, and at the same time, magic object of obtaining rain. They need to know also, however, to what extent, at what time, and for how long, the rain will fall. This is what divination tells them, when they interpret, at the same time as the result, properly so called, the turns of fortune in a game mystically undertaken between the powers of drought and of rain, the number of points marked to the credit of each camp, etc. Thus here, too, divination is an added perception. Strictly speaking, it is an anticipation of it, to which prelogical mentality trusts as much as to perception itself. Their confidence rests on the participations imagined and felt between the players, their respective clans, the spatial regions _* Quoted by Culin in “‘ Games of the North American Indians,” E. B. Rept., XXIY4/D.3 74: PRIMITIVES’ PRACTICES 293 \ corresponding with them, their mythical animals, colours, gods, winds, and finally, the rain and drought themselves. The transition from divination to magic is almost im- perceptible. Both are founded upon the same collective representations of mystic relations: divination is primarily concerned to discover these relations, and magic to utilize them. But in actual practice, the differing aims unite once more, since to be able to exercise magic influence it is neces- sary to know these mystic relations, and, on the other hand, if divination seeks to discover them it is for the purpose of deriving advantage from the knowledge. We may even go further and say that all the practices we have considered hitherto, those relating to hunting, fishing, war, sickness, death, etc., and in a general way, many practices correspond- ing with the collective representations of primitive peoples are magic in their nature. I have preferred to call them ‘“ mystic,” because this emphasizes the rigid solidarity with the mentality which I call prelogical and mystic, whereas the word ‘ magic ”” takes on meanings which vary according to the state of the social type studied. Thus, among Australian natives, or among those of South America, found in Brazil, Terra del Fuego, and other parts, for instance, the majority of the practices which correspond with their most important group-ideas are magical. This is very clearly brought out in Spencer and Gillen’s two works. In dealing with peoples who are more differentiated as to type, however, such as most of those in Equatorial and South Africa, the practices which are really magical are different from religious rites, and cannot be designated by the same term. There arises a differentiation of function which is clearly present to the collective consciousness of the group." Junod tells us, for instance, that there is continual con- fusion between “ diviner, wizard, doctor, exorcist, caster of lots, etc. (among the Ba-Ronga). . . . In my opinion, this is a serious mistake, and African ethnography ought to be very careful to guard against it. Undoubtedly the same person may be both priest, doctor, diviner, etc. But each 1 Cf. Hubert et Mauss, ‘‘ Esquisse d’une Théorie Générale de la Magie,” Année Sociologique, vii. pp. 1-147. 294 HOW NATIVES THINK of these functions is a different one, and the native language gives each a special name. . . . The most usual term is that of mongoma, which denotes the doctor, especially in his character of wonder-worker, as well as the ‘ possessed’ who have been cured, have undergone initiation and can look after others. ‘“‘ Nganga is the doctor, but the doctor curing by means of the more or less secret remedies he employs. He is the man of medicine, wa-mori. It is his task to prepare, too, the miraculous plants which will confer invulnerability in war. As we see, nganga and mongoma are very closely related. “ Gobela is the exorcist, who expels the Zulu or Ndjao spirits. There are even two different categories of these, according to whether the north or the south is referred to. ‘© Wa-bula is the bone-thrower, the diviner par excellence, the family counsellor, the interpreter of fate. “The chinusa is the personage who divines by visions or . by ecstasy. ““ As for the word sorcerer, that must be reserved for the baloyi, those who bewitch, the magicians, if you prefer, those who go by night to place their charms, and kill by their witchcraft. These individuals have no connection with the preceding. “ Finally, the priest, that is, the muhahli, and this, in his own family, is what every father is. He may be only this, but it is also possible that he occasionally has other functions.” ! In this stage of development, if we persist in calling all these practices by the name of “‘ magic,” there would seem to be an official and public magic, as it were, a private and legitimate magic, and an occult and criminal practice of magic. It is evidently better not to include in one and the same term ideas which have become mutually exclusive. Among the Kafirs, for instance, “‘if an tsangoma (diviner) were saluted as a umtakati (sorcerer), it would be the most deadly insult that could be offered. It would be just the same as calling a European police agent a thief. To the Kafir mind, the diviner on the contrary is the protector of social order, and his task consists in discovering criminals and sorcerers and 1 Junod, Les Ba-Ronga, pp. 467-8. PRIMITIVES’ PRACTICES 295 putting justice on their track. Whilst the sorcerer exercises his art from motives of personal interest, to attain an end which is prohibited, the diviner must work in a legitimate way for the public good, as a kind of official, and for this reason he enjoys considerable importance among the Kafirs.”’ 7 When a religion, properly so called, has been established, with forms of worship and organized clergy, there is additional reason for the strengthening of the contradistinction between religious ceremonies, either public or private, and magical operations which are more or less secret and malevolent. I do not propose to follow this differentiation further. It is sufficient to have shown that prelogical, mystic representations can but have corresponding methods and traditions which are equally mystic in their nature. The orientation of both is necessarily the same. Their correlation is the more manifest when observation is brought to bear upon the most primitive types, and therefore, to prove my case, I have selected facts drawn from communities of this kind. Nevertheless, beneath this contradistinction there is a certain relationship. Practices which, from the social point of view, have become very different, none the less imply collective representations of the same nature. That is they relate to a mentality which is prelogical and mystic, still recognizable under the modifications it may have undergone. For instance, the actually religious rites, the prayers and ceremonies by means of which the priests hope to soften the gods who are masters of the rainfall, imply collective repre- sentations of the same order as those which form the basis of the intichiuma ceremonies. The intermediate stages are to be observed among the Zuñis, for instance. The antagonism between the actually magical practices and the legitimate and religious practices in most aggregates which are already civilized to a certain extent, does not therefore signify that prelogical, mystic mentality is the source of the one and not of the other. On the contrary, we find it more or less unadul- terated at the basis of all these practices, and possibly it is this common origin which accounts for the variety and extent of the meaning given to the word ‘“‘ magic.” If by ‘ magic ” ia Aegidius Müller, ‘‘ Wahrsagerei bei den Kaffern,’’ Anthropos, p. 762 1906). 296 HOW NATIVES THINK we understand every process which supposes that mystic relations and occult forces are at work, there is scarcely any act, even of peoples who are comparatively advanced, which may not be of magic character to some extent atleast. In so far as the law of participation more or less governs the mentality, it also more or less determines the modality of its action. There would seem to be nothing less mystic than for the primitive to satisfy his hunger when food is within his reach. Nevertheless we see him, nearly everywhere, voluntarily abstaining from certain foods which are forbidden. In almost all the aggregates known to us at present, there are a great many taboos with respect to food. In particular, where totemic institutions prevail, on no account—save on certain definite occasions—would a man consent to feed upon his totem. On the other hand, partaking of a being is in a certain sense participating in it, communicating and identifying oneself with it: for this reason, therefore, there are certain foods which should be sought after, and others which should | be avoided. This is, as we know, the origin of a certain kind of anthropophagy. The heart, liver, fat, and brain of enemies killed in war are eaten so that their courage and their intelligence may be appropriated; just as, with us, tuber- culous subjects eat raw meat that they may be better nourished. Other foods are rejected for the opposite reason. ‘ They ’’ (the Abipones) ‘all detest the thought of eating hens, eggs, sheep, fish, and tortoises, imagining that those tender kinds of food engender sloth and languor in their bodies and cowardice in their minds. On the other hand, they eagerly devour the flesh of the tiger, bull, stag, boar... having an idea that, from continually feeding on these animals, their strength, boldness and courage are increased.” 1—In the north-eastern provinces of India, “‘ the owl is the type of wisdom, and eating the eyeballs of an owl gives the power of seeing in the dark.’’2 In New Zealand, ‘“ a good orator was compared to the korimako, the sweetest singing bird of New Zealand ; to enable the young chief to become one, he was fed upon that bird, so that he might the better acquire its qualities.”’ 3 The Cherokees had the same idea. ‘‘ He who t Dobrizhoffer, An Account of the Abipones, i. p. 258. 2 W. Crooke, Folklore of Northern India, i. p. 279. 3 KR. Taylor, Te Ika a Maui, p. 353. PRIMITIVES’ PRACTICES 207 feeds on venison is swifter and more sagacious than the man who lives on the flesh of the clumsy bear or helpless dunghill fowls. . . . Formerly their great chieftains observed a con- stant rule in their diet. . . . A continuous adherence to the diet commonly used by a bear will finally give to the eater the bear nature, if not also the bear form and appearance. A certain term of white man’s food will give the Indian the white man’s nature, so that neither the remedies nor the spells of the Indian doctor will have any effect upon him.’”’! What is true of men and animals applies also to the gods. ‘The idol was very dirty and smeared with blood, but in his right side was a hole showing the natural white colour of the material, contrasting strangely with the dusky appearance of the rest of the figure. This hole owes its existence to the belief that the power of healing and the knowledge of mysterious things are acquired by eating a little of the god’s holy body.’ ? Facts of this kind are extremely common, and the reader will find many of them analysed in Robertson Smith’s Lectures on the Religion of the Semites.”’ It is the same with clothing as with food: the mystic element plays an important part, and in certain cases pre- ponderates over utilitarian considerations. Many “ savage ” tribes, before coming into contact with white men, wore no garments of any kind, but there were none among whom there were no adornments: feathers, beads, tattoo marks, painting, and the like. Now, as we know, these adornments are not mainly, nor solely, decorative. They have a mystic character, and possess magic power. The eagle’s feather assures him who wears it the strength, piercing sight and wisdom of the bird, and it is the same with other things. Conversely, if the attention of investigators had been directed to this point, they would doubtless have noted taboos relating _ to clothing as to food: we have already seen the case of the Malay chief who refused to carry a deerskin in his canoe, lest the timidity of the deer should be communicated to his young son. As a general rule, just as one participates in the quali- ties of that which he eats, so does one secure the qualities of that which he wears. Here is one example among many. : Adair, quoted by Mooney, ‘‘ Myths of the Cherokee,” E. B. Rept., xix. P. 472. 2 C. Lumholtz, Unknown Mexico, ii. p. 170. 298 HOW NATIVES THINK ‘ One morning I shot a hyena in my yard. The chief sent one of his executioners to cut off its nose and the tip of its tail, and to extract a little bit of brain from the skull. The man informed me that those parts are very serviceable to elephant- _hunters, as securing for them the cunning, tact, and power to become invisible which the hyena is supposed to possess. I suppose that the brain would represent the cunning, the nose the tact, and the tip of the tail the vanishing quality.” ! Here we have that ‘“‘ sympathetic magic ” so fully described by Tylor, Frazer, and their disciples among the English school of anthropology. May I be allowed to refer readers to their works for facts such as these, to be met with in hundreds ? In them we see how various qualities are com- municated by contact and transference, how the whole is attained by means of a part, such as holding a man in one’s power when one has his hair, his nail-parings, saliva, water, name, or likeness, and finally how to produce like by means. of like (to incite the rain to fall by sprinkling water, etc.). What matters here is to show that these practices of a ‘““sympathetic magic,” which are often akin to those which I have analysed, refer equally to the collective representations of the primitive mind and to the law of participation which governs them. In this sphere, too, behaviour is oriented in the same direction as the mental activity. Prelogical, mystic mentality, perceiving everywhere secret relations, actions and reactions which are external and at the same time intimate, between all entities, in short, finding partici- pations everywhere, can only hope to influence nature by establishing, or by rupturing, similar participations. Among the Baganda, for instance, “a sterile wife is generally sent away because she prevents her husband’s garden from bearing fruit . . . on the contrary, the garden of a prolific woman bears plentifully.2 Here the husband of a sterile wife is merely combating a troublesome participation ; in another case, one will try to induce favourable ones. Thus, in Japan, ‘trees must be grafted only by young men, because of the special need of vital energy in the graft.” 3 t Arnot, quoted by Nassau, Fetichism in West Africa, p. 204. 3 J. Roscoe, ‘“‘ Manners and Customs of the Baganda,”’ J.A.J., xxxii. pp. 38, 56. 3 Chamberlain, Things Japanese, p. 440. PRIMITIVES’ PRACTICES 299 Contact, also, in certain definite cases, establishes partici- pation. Here is a very clear instance. In Loango, “the ba nganga teach that a new fetish sucks in strength, as it were, from contact with other fetishes of proved vigour, next which it is placed—provided only, that it is destined to serve the same end as theirs. For this reason they agree, if the price offered be satisfactory, to put new and untried charms into their collection of effective and well-tested magical objects, and to leave them there for weeks or months. So too, if a fetish has become somewhat doubtful and feeble, its strength is renewed by this process. It is a method of restoring its vigour. If a fetish proves excellent, a duplicate is made, and left for a long time in contact with the first. The duplicate is called the child of the original fetish.”’ ! Finally, it is participation again, that the well-known practices by whose means “‘sympathetic magic” seeks to further the principle of “like acting upon like” aim at establishing. If we study them in communities which are already fairly developed, we shall perhaps be inclined to regard as adequate the interpretation which refers them to the association of ideas and the confusion of the objective and the subjective. In China, for example, innumerable practices of this kind may be noted, and sometimes they seem to be like the translating of a punning joke into action. Thus, at a certain period in the funeral ceremonies, ‘‘ the son of the deceased . . . in concert with most of the attending kinsmen, hastens to swallow a few mouthfuls of cooked vermicelli, wisely inferring that the long threads of this food must greatly counteract, nay totally neutralize the life-shortening influences which the grave-clothes may have exercised over his person.” 2 It seems as if here we have one of those cases of association of ideas which the somewhat abstract subtlety of the Chinese affects: the underlying idea, however, is that of rupturing a dreaded participation. But here is a circumstance reported from South India, the interpretation of which is yet more definite. ‘“‘ Carved wooden figurines, male and female, repre- sented in a state of nudity, are manufactured at Tirupati and sold to Hindus. Those who are childless perform on 1 Dr. Pechuél-Loesche, op. cit., ill. 2, pp. 366, 380. 2 De Groot, The Religious System of China, i. pp. 68, 208. 300 HOW NATIVES THINK them the ear-boring ceremony, in the belief that, as the result thereof, issue will be born to them. If there be grown-up boys and girls in a family who remain unmarried, the parents celebrate the marriage ceremony between a pair of dolls, in the hope that the marriage of their children will speedily follow. They dress up the dolls in clothes and jewellery, and go through the ceremonial of a real marriage. Some there are who have spent as much money on a doll’s wedding as on a wedding in real life.’”’! This expenditure, sometimes very considerable, attests the faith of the Hindus in the efficacy of the practice. Is it sufficiently accounted for by an appeal to association by similarity ; is it enough to say that they imagine like will produce like? It is a “probable” explanation, but it is difficult to maintain when we know that with primitive peoples practices like these proceed, not from an association of ideas in the individual, but from the participations imagined and felt in the group ideas. The Hindu who has a doll’s wedding is. acting just like the Redskin who dances “‘ to make the buffaloes come,” and the rain-maker who sprinkles his neighbours. It is a wholly mystic dramatization, designed to secure for the actors an equally mystic power over the person or the action imitated, creating between them and him a bond which is doubtless unintelligible to logical thought, but quite consistent with the law of participation which governs the collective representations of the prelogical primitive mind. Where prelogical mentality is purest, in aggregates of the most primitive type, the customs reflect it most clearly. We have noted this in the Australian aborigines and in certain groups in North and South America. In the more advanced types modality of action becomes more complicated, and a variety of divergent motives determines it. In nearly every case, however, when we probe this modality to its lowest depths, we discover traces which bear witness to the predominance formerly exercised by the law of participation. We shall find as many proofs of this as we can desire in the great Eastern civilizations, or even nearer home, in the folklore of European nations. 1 Thurston, Ethnographic Notes in Southern India, p. 347. CHAPTER VIII INSTITUTIONS IN WHICH COLLECTIVE REPRE- SENTATIONS GOVERNED BY THE LAW OF PARTICIPATION ARE INVOLVED (III) THERE are certain practices which I have reserved for con- sideration by themselves, as much on account of the important place they occupy in the life of primitives as of the light they throw upon their mentality. These are the practices relating to the dead, or rather, to the connection between the living and the dead, They are found everywhere: there is hardly any social group, of whatever type it may be, in which observers have not noted customs, taboos, rites which are obligatory at the moment of death, and for a shorter or longer period afterwards. i Everybody knows how, from time immemorial, considera- tion for the dead has imposed a burden on the living, among the Chinese.t In Canada, “if a fire should break out in a village where there are any dead bodies, they will be the first to be placed in safety. The people will despoil themselves of their most precious possessions to adorn the dead ; from time to time they open their tombs to change their garments, and they deprive themselves of food that they may carry it to their graves and to the places where they imagine their souls to be wandering. . . . In the grave they are careful to cover their bodies in such a way that no soil will touch them; the corpse is, as it were, in a little cell lined with skins, much richer and more ornate than a hut.”’ 2 From the primitives’ point of view we can easily under- 1 J. J. M. de Groot, The Religious System of China, i. p. 658 et passim. 2 Charlevoix, Journal d’un Voyage dans l'Amérique Septentrionale, iii. pp. 372 et seq. 302 HOW NATIVES THINK stand that practices of this kind would be universal. To them, there is no insuperable barrier separating the dead from the living. On the contrary, they are constantly in touch with them. They can do them good or harm, and they can also be well or badly treated by them. To the primitive there is nothing stranger in communicating with the dead than in being connected with “‘spirits,’ or with any occult force whose influence he feels, or which he flatters himself he is subduing. Miss Kingsley relates that she once heard a negro talking - aloud, as if conversing with an interlocutor unseen by her. Upon inquiry she found that the negro was talking to his dead mother who, according to him, was present. To the primitive the reality of the objects he perceives does not in the least depend upon his being able to verify this reality by what we call experience ; indeed, as a rule, it is the intangible and invisible that is most real in his eyes. Moreover, the dead do not remain without revealing their presence even to the senses. To say nothing of the dreams in which they appear (and which are, as we know, perceptions which are matter of privilege and solicitation), the dead appear to sight as spectres and ghosts, and to hearing also. Sometimes they afford the living indescribable but very vivid sensations of a contact which has nothing material about it. Occasionally they are heard in the wind. ‘It cannot be seen; it is like the wind; in fact, the gentle rustling of the plantain leaves is said to be caused by the ghosts, and a whirlwind which carries up dust, leaves and straws is said to be the ghosts at play.” 1 It is the same in Brazil; ‘the Ges thought that the murmuring of the wind betrayed the presence of the dead.’’? In short, without insisting on well-known facts, the primitive lives with his dead as he does with the living who surround him. They are members, and very important members, of a society with manifold participations, a social symbiosis in which the collec- tive representations of his group give him his place. Hence we derive one of the most distinct differences dis- cernible between the mentality of primitives and logical thinking. Assuming that the latter develops gradually (a t J. Roscoe, ‘ Manners and Customs of the Baganda,”’ J.A.J., xxxii. p. 73. 2 Von Martius, Beitrage zur Ethnographie Stid-Amertka’s, 1. p. 291. PRIMITIVES’ IDEA OF DEATH 303 hypothesis which we shall examine later), its ideas relating to the external world have finally become organized in “‘ nature,”’ i.e. In an order which is fixed and immutable, subordinated to laws which the subject may be aware of, but which appear to be wholly independent of him. Ideas with regard to the dead have not provided any such. They have only constituted that nebulous ensemble which we significantly call “the other world.” To the primitive mind, on the contrary, the other world and this one make one and the same reality only, a reality both imagined, felt and lived. Even to such a mind, however, ideas relating to the dead, and the customs connected with them, are distinguished by a more markedly prelogical character. However mystic other collective representations relating to data furnished by the senses may be, however mystic the customs pertaining to them, (such as hunting, fishing, warfare, illness, divination, - and so on) it is still necessary, if the desired end is to be attained,—the enemy conquered, the game brought down, etc.—that in some essential points the representations should coincide with objective reality, and the customs, at a given moment, be actually adapted to effect that end. Thus a minimum of order, objectivity and coherence is guaranteed. But in the representations and the customs relating to the dead, this external standard is not in force. The indifference to the law of contradiction which is peculiar to primitive mentality, will accordingly be manifested without let or hindrance. In this domain, therefore, we shall find practices which most strikingly attest the prelogical nature of this mentality. The law of participation governs in despotic fashion the collective representations upon which these practices depend ; and permits of the most flagrant contradictions in these. We are already aware that, to minds like this, there is nothing exactly corresponding with that which we term soul or per- sonality. They regard a soul as both one and multiple, and believe that it can be present in two places at the same time, etc. In their customs, therefore, we must be prepared to find what we, from the logical point of view, regard as inextricable confusion. Our efforts must be directed, not to dispersing this confusion by establishing a logical order in the output of 304 HOW NATIVES THINK a mystic mentality which cares nothing about it, but to show- ing how it is that such confusion is the natural result of the law of participation which governs the primitive mind. First of all, then, generally speaking, the primitive finds no difficulty in imagining the dead as sometimes constituting a community in the other world quite distinct from living communities, and again, as intervening on all occasions in the life of those here. For instance, among the natives of Torres Straits, “soon after a man dies, his mari goes to Kibu, where on his arrival the mari of a previously deceased friend takes the newcomer and hides him. At the first night of a new moon the mart is introduced by his friend to all the other markaz, each of whom takes stone-headed clubs and hits him on the head and then he is a true markai. They then teach the new markat how they spear fish, and how to do whatever they themselves do. . . . There was a general agreement that mari or markat behaved in every way as do men, and they could marry mortals.’’! In China, where the ideas and practices relating to the dead have been preserved from the very earliest times, we find the same contradictions. On the one hand, there is a world of spectres which imitates the world of men. Its society is organized in the same way, and its clan life continues. Each one maintains the rank that he had pre- viously held in life, and he continues to practise ancestor- worship with the same rites as before. The dead have their armies and their fights, their burial places and their funeral ceremonies. Men are as formidable to spectres as spectres are to men ; the bad influence they exert is reciprocal, and on both sides this is exorcized by means of sacrifices. De Groot relates a very significant legend of men penetrating into the land of spectres, their presence there causing terror. Sacrifices are offered to them and they are conducted to the frontier with infinite precautions.3 On the other hand, however, according to the same author, “it is an inveterate conviction of the Chinese people, a doctrine, an axiom, that spirits exist, keeping up with the living a most lively intercourse, as intimate almost as that among men. 1 The Cambridge Expedition to Torres Straits, v. p. 357. > J. J. M. de Groot, The Religious System of China, i. pp. 48, 924. 3 Ibid; ii. pp. 8002-71. PRIMITIVES’ IDEA OF DEATH 305 There exists, in fact, a line of separation between the dead and the living, but it is a very faint line, scarcely discernible. In every respect their intercourse bears an active character. It brings blessing, and evil as well, the spirits thus ruling effectually man’s fate.’’! These expressions show that the dead are believed to be alive in their graves. “‘ Throughout the whole range of Chinese literature, coffins with a corpse inside are designated ‘ animated encoffined corpses’ or ‘ ani- mated coffins.’’’+ A young girl, widowed as a bride, “‘ when she has acquired the consent of her own parents and those of her deceased bridegroom may renounce conjugal life for ever. She is as a rule allowed to settle for good in the mortuary house, and is then formally united with the dead in marriage.”’ 3 We are told that “in ancient China there existed the curious custom of placing deceased females in the tombs of lads who died before they were married.” 4 Public opinion inclines so favourably to the sacrifice of wives who follow their husbands to the tomb, and the act reflects so much honour on the families that widows often desire it, or at least resign them- selves to it, or they may even be constrained to it by their family circle. The negroes of West Africa hold that when a man dies he only gets rid of his corporeal body and changes his abode, for everything else remains as before.s In North America, among the Sioux, as Dorsey tells us, “‘the dead in all respects resemble the living. . . . They are not always visible to the living. They are sometimes heard but not seen, though in the lodge with a mortal. Occasionally they become materialized, taking living husbands or wives, eating, drinking, and smok- ing, just as if they were ordinary human beings.” 6—" A young Dakota dies just before marrying a girl whom he loved. The girl mourned his death. . . . The ghost returned and took her for his wife. Whenever the tribe camped for the night, the ghost’s wife pitched her tent at some distance from the others, and when the people removed their camp, the woman and her husband kept some distance behind the main body. The ghost always told the woman what to do; and he brought 1 J. J. M. de Groot, The Religious System of China, ii. p. 464. 2 [bid., i. p. 348. 3 Ibid., i. p. 763. 4 Ibid., pp. 802 et seq. 5 A. B. Ellis, The Ewe-speaking Peoples, p. 106. 6 Dorsey, ‘‘ Siouan Cults,” Æ. B. Rept., xi. p. 485. 20 306 HOW NATIVES THINK her game regularly. . . . The people could neither see nor hear the ghost, but they heard his wife address him. He always sent word to the tribe when there was to be a high wind or heavy rain.’ An Iroquois legend tells of a dead man who talks to his daughter and gives her advice.? Facts of this kind are, as we know, very numerous. If, therefore, we are to interpret aright the primitives’ conceptions and customs with regard to their dead, it will be necessary to rid ourselves, as far as we can, of the ordinary concepts of ‘‘life’’ and ‘‘ death,” just as it is wise not to make use of the concept “‘soul.’’ It is impossible for us to define these terms save by elements that are physiological, objective, and experimental, whilst the ideas which correspond with them among primitives are mystic in their essence, and to such an extent that they ignore a dilemma which logical thought regards as insurmountable. To us, a human being is either alive or dead: there is no middle course, whereas to the prelogical mind he is alive in a certain way, even though he be dead. While participating in the life of human beings, he at the same time makes one of the company of the dead. To put it more precisely, he is living, or dead, to the extent in which such and such a participation exists or does not exist, for him. The way in which the living conduct themselves with regard to him depends entirely upon whether these participa- tions continue or have been ruptured or are about to be ruptured. We are thus faced by collective representations and customs which are of extreme complexity. We cannot study their almost infinite detail here, for even in the same country they vary in the different tribes. ‘ No particular description of burial ceremonies can bé held applicable to all tribes, or even to eny one tribe, if the age, character, or position of the deceased was such as to procure for him more than ordinary respect.’ 3 Just because the dead continue to live, each one of them is treated according to his rank, sex, and age,— circumstances which observers, when describing these cere- monies, very often neglect to specify. Moreover, the customs I Dorsey, “ Siouan Cults,’ E. B. Rept., xi. p. 490. 2 Hewitt, ‘ Iroquoian Cosmoiogy,” Æ. B. Rept., xxi. pp. 147 et seq. 3 Brough Smyth, The Aborigines of Victoria, i. p. 114. PRIMITIVES’ IDEA OF DEATH 307 relating to death and the dead are possibly the most persistent of all. Consequently, whilst the social milieu, institutions, creeds are modified, these customs change but slowly. They remain in use even when their significance has gradually become obscured and been lost. Interpreted in accordance with new ideas and sentiments,—that is, in a contrary sense very frequently—it may happen that they are carried out in such a way as to become self-contradictory in the portion retained. The funeral rites in most social aggregates, if not in all, thus present an irregular stratification in which customs which date back to remote antiquity and relate to a very primitive type of mentality are blended with practices depend- ing upon more recent conceptions, incompatible with such a mentality. Finally, in a great many aggregates the funeral rites and ceremonies vary once more according to the nature and cause of the decease, and whether the individual has departed this life by a ‘‘ good ” or ‘‘ bad ”’ death. | It is the province of ethnography to give the most detailed and circumstantial description possible of these ceremonies and their very varied nature. It will suffice here to show that it is the mystic and prelogical mentality which inspires them ; that to such a mentality the different degrees of life and death consist in participations or the lack of them ; and finally, that the same man does not pass through two stages of being only, one when he is alive, and the other when he is dead, but goes through a series of states and a cycle of phases in which he participates more or less in what we call life and death. The plan I am about to sketch for you can only be a rough pro- visional one, and I regard it as applicable (to the extent in which it may be accepted) to communities of the most primi- tive type only,—communities in which totemistic organization — is still recognizable, even if not in its full vigour—such aggre- gates as the Australian aborigines, for instance. I warn you also that it is impossible to imagine man as an integral whole going through all the phases of this schematic cycle in succes- sion. On the contrary, that which characterizes certain of these phases is that participations have no part in them, and without these, according to our logical thought, there can be no human being. But it is the nature of prelogical mentality to move freely among these participations or these isolations, 308 pe HOW NATIVES THINK whilst logical thought feels disconcerted and lost there, for lack of ability to operate upon concepts that are clearly defined and conformable with its own laws. With such exceptions, our sketch, starting from the state of the adult man who has undergone the initiation ceremonies and has married within his social group, would differentiate the following stages : (x) Death, and the longer or shorter period which elapses between the last breath and the funeral ceremonies ; (2) The period which intervenes between the funeral cere- monies and the end of the days of mourning ; i.e. the rite which definitely ruptures the relations between the dead man and those who were closely associated with him in the social group; (3) The period, indefinite in its duration, but yet finite, in which the dead is awaiting his reincarnation ; (4) Birth, and the longer or shorter time which elapses between the birth and the giving of the name ; (5) The period which intervenes between the naming and the initiation ; (6) The life of an adult initiate man, which ends with death, when the cycle begins once more. In our language, then, we should say that death, like birth, is accomplished in stages. Death begins with the first period, and is not complete until after the ceremony which ends the second. Birth, too, begins with the actual confine- ment, and man is not complete until after the initiation cere- monies. Here again our mental habits and our language oblige us to divide into clearly distinct periods that which to prelogical mentality is represented in the form of multiple participations, either beginning or ceasing to be co-existent. II According to primitive ideas, death always implies a mystic cause, and nearly always, a notion of violence. It is an abrupt rupture of the ties binding an individual to the social group. New relations, accordingly, are established between this group and himself. Far from being henceforth a negligible quantity, he who has just died is an object of pity, fear, respect, and a number of varied and complex feelings. PRIMITIVES’ IDEA OF DEATH 309 The funeral rites reveal group ideas inseparably connected with these feelings. There are communities in which these rites begin even before the patient has breathed his last, so great is the primi- tives’ haste to remove the dead from the living. Among the Abipones, for instance, ‘if the respiration of the dying man be not heard at a distance .. . and if his breath stop even for a moment, they proclaim . . . that he has given up the ghost . .. The first business of the bystanders is to pull out the heart and the tongue of the deceased . . . and give them to a dog to devour, that the author of his death may soon die also. The corpse, while yet warm, is clothed according to the fashion of his country, wrapped in a hide . . . and conveyed on ready horses to the grave. . . . I strongly suspect that the heart is sometimes cut out when they are half alive.” 1 Among the Cape Flattery Indians, “‘ I have known several cases where considering the circumstances, there is scarcely any doubt but that people have been buried during a swoon, or when in a comatose state merely. I have often told them how foolish it was to bury these people without having tried to revive them; but I have never been able to persuade them to wait a single instant after they believe that the last breath has been drawn. . . . A woman had just lost her husband. She was remarkably robust and in good health. I saw her seated beside the brook bemoaning the death of her husband, and I went on to the village a quarter of a mile away, where I had some sick people to see. . . . Suddenly I heard women’s lamentations announcing a death. I hastened to return and found that it was the woman I had seen a few moments before. . . . Before I reached her hut she was wrapped in her grave-clothes and squeezed into a box, ready to be buried. Her relatives would not listen to a single word, nor allow me to do anything whatever.” 2 “ Frequently,” said a Jesuit missionary, speaking of the Canadian Indians, ‘‘ they were buried according to their own rites before they had actually expired.” 3 Von den Steinen testifies to the same precipitancy among 1 Dobrizhoffer, An Account of the Abipones, ii. pp. 266-8. 2 Swan, The Indians of Cape Flattery, pp. 84-5. 3 Relations des Jésuites, p. 260 (1636). 310 HOW NATIVES THINK the Bororos.t It is the same thing with the Bakwains of South Africa. ‘‘ Scarcely is the breath out of the body,” says Livingstone, ‘“‘ when the unfortunate patient is hurried away to be buried. An ant-eater’s hole is often selected, in order to save the trouble of digging a grave. On two occasions while I was there this hasty burial was followed by the return home of the men, who had been buried alive, to their affrighted relatives.’”’2 ‘In a general way, in the many communities where they hold that a living man has several souls, it is enough that one of them should depart (provided that this departure is a definite one) for him to be considered dead: it matters little that the soul of his body be there still, if that which makes his personality has disappeared without intention of returning. We found this belief prevalent among the North American Indians. It exists elsewhere also, among the Dravidian tribes of Bengal. When the sick man is in his last agony, barely conscious, with the death-rattle in his throat, they say: ‘ His body is still moving, but his soul (roa) . has already departed.’ ”’ 3 Most frequently, however, the funeral rites and the inter- ment do not take place immediately, precisely because they do not know whether death is a definite fact, or whether the soul (I use this term, in default of a better,) will return to the body as it does after dreams, fainting-fits, etc. They accordingly wait a while, making use of all the means likely to bring back the departed. Hence arises that widely spread custom of calling aloud on the dead, imploring and beseeching him not to leave those who love him. “The Caribs lament loudly, their wailings being interspersed with . . . questions to the dead as to why he preferred to leave this world, having every- thing to make life comfortable. They place the corpse on a little seat in a ditch or grave four or five feet deep, and for ten days they bring food, requesting the corpse to eat. Finally, being convinced that the dead will neither eat nor return to life, they . . . fill up the grave.’’4 So, too, do the natives of Unter den Naturvôlkern Zeniralbräsiliens, pp. 350, 397. 2 Livingstone, Missionary Travels, p. 129. 3 “‘ Sagen, Sitten und Gebrauche der Munda Kolhs in Chota Nagpore,”’ Zeitschrift fr Ethnologie, p. 371 (1871). 4 Bruhier, in Yarrow’s “‘ Mortuary Customs of the North American Indians,” E. B. Rept., i. p. 166. PRIMITIVES’ IDEA OF DEATH 311 the west coast of Africa appeal aloud to the dead, begging him not to leave them, except in the case of his having been deaf when alive (for then they believe the ghost or soul to be deaf also). ‘“‘ Generally speaking, it is only when the corpse begins to become corrupt, and the relatives thereby become certain that the soul does not intend to return, that it is buried.’’? Finally, in China, the custom of ‘‘calling back the dead ”’ has existed from the very earliest times, and is still tenaciously maintained to-day.? | | Among the tribes of Eastern Sumatra, ‘‘ when a man dies, the corpse remains a day and a night in the house....A grave is dug; and the body, with a bottle of water and a fowl . . . is placed therein and the grave closed. (If the corpse be that of a woman, her ornaments are included.) Then bonfires are lighted, and the family live and sleep near the tomb for a period of three days ; (this is extended to seven in the case of a chief). This time is necessary, they believe, for the dead to be really dead, and until that period has elapsed they must keep him company.”’ 3 At this moment, then, the dead man’s: participation in his social group is suspended rather than actually ruptured. They believe, or they wish to believe, that nothing is irre- parable, and that former conditions may be re-established. If the soul does not return and the dead awake, however, other ideas supervene, and with them other feelings are aroused. Of the Australian aborigines we are told that they believe that the spirit of a man, especially if he died a violent death, is extremely miserable and malevolent ; in his anger he is always ready to become exasperated on the slightest pretext, and to vent his spite on the living. . . . They seem to think, too, that the deceased, for a certain number of days after his death, has not received his spiritual body, which is slow in forming, and that during this period of transition he is, like a child, particularly petulant and revengeful.4 The dead man finds himself in a state that is painful for himself and dangerous for others, who henceforward shun contact with‘ 1 A. B. Ellis, The Ewe-speaking Peoples, pp. 106-8. | 2 J. J. M. de Groot, The Religious System of China, i. p. 253. 3 Moskowski, Siämme von Ost Sumatra, p. 644. 4 J. Fraser, “ Some Remarks on the Australian Languages,” Journal of the Royal Society of New South Wales, pp. 235 (1890). 312 HOW NATIVES THINK him. Since he no longer is an integral part of the group like others, he must get away from them. This feeling is often expressed in the most naive and vivid fashion. ‘‘ When a man is dying his friends take food to him, and say: ‘Be good; if you leave us, go altogether.’’’* As for the Igorots of the Philippines, “ during the first days the old women and again the old men sang at different times the following song: ‘ Now you are dead ... We have given you all things necessary, and have made good preparation for the burial. Do not come to call away any of your relatives or friends.’ ”’ 2—Again, in West Africa, we find from Nassau's account that the feelings of survivors with respect to the dead are ‘‘ very mixed. The outcry of affection, pleading with the dead to return to life is sincere, the survivor desiring the return to life to be complete; but almost simultaneous with that cry comes the fear that the dead may indeed return, not as the accustomed, embodied spirit, helpful and companion- able, but as a disembodied spirit, invisible, estranged, perhaps’ inimical.’’3 Miss Kingsley has noted the same fear, and it was explained to her that it was due not to malevolence but to loneliness; the dead did not wish to hurt the members of his own family, particularly the children, but he desired to have their company, and ‘“ this desire for companionship is of course immensely greater, and therefore more dangerous, in the spirit that is not definitely settled in the society of spirit- dom.’ 4 This it is which makes the first few days after death peculiarly critical. Not only is the relation of the dead with his own group only just ruptured, but his connection with the spirit-group is not yet established. North American natives manifest the same ideas and sentiments. To the Tarahumares, ‘‘death means only ... a change of form. ... They are afraid of the dead and think that they want to harm the survivors. This fear is caused by the supposition that the dead are lonely and long for the company of their relatives. The dead also make people ill, that they too may die and join the departed. . . . When a person dies his eyes are closed, his hands crossed over his t Basil Thomson, ‘‘ The Natives of Savage Island,” 7.4.1. xxxi., p. 139. 2 Jenks, The Bontoc Igorot, p. 75 (Manila, 1905). 3 Fetichism in West Africa, p. 223. 4 West African Studies, p. 113. PRIMITIVES’ IDEA OF DEATH 313 breast, and . . . the relatives talk to him one by one, and bid him goodbye. The weeping widow tells her husband that, now that he has gone and does not want to stay with her any longer, he must not come back to frighten her or his sons or daughters or anyone else. She implores him not to call any of them off, nor do any mischief, but to leave them all alone. A mother says to her dead infant: ‘Now go away! Don’t come back any more, now that you are dead!’ And the father says to the child: ‘Don’t come back to ask me to hold your hand, or to do things for you. I shall not know you any more.’ ’’! These fears are so much the more vivid because during the early days one imagines the dead, that is, his soul or spirit, living in the hut or cabin in which he died, in any case not far away from his corpse, or else roaming about the neighbourhood, especially during the night. This belief is almost universal; even were it not implied in the group ideas a psychic process would arouse it in indi- viduals. Do not we ourselves, when death has robbed us of one of our dear ones, at first expect to see him returning home at his wonted hour and coming out of his room to take his place at table? With primitives, however, there is something beyond these poignant recollecticns which amount almost to hallucination: it is the visible presence of the body which, in their collective representations, brings in its train the invisible presence of the soul. ‘‘ After death, the soul remains for some time in the vicinity of the corpse before beginning its journey to the bura kure.”2 Hence it follows that in disposing of the body the fate of the soul can be settled at the same time. The dead man is assigned to the resting- place henceforth to be his, and the living are free from the terror caused by his presence during the intermediate period. Thus, whatever may be the form of these funeral cere- monies, in whatever fashion the corpse is disposed of—whether by interment, cremation, placing on a platform or in a tree, etc.—the essential characteristic of such practices is mystical, or, if you prefer it, magical, like the rites we have already studied. Just as the essential feature of the chase is con- 1 C. Lumholtz, Unknown Mexico, i. pp. 380-2. 2 Hagen, Unter den Papua’s, p. 266. 314 HOW NATIVES THINK tained in rites which force the prey to appear, which paralyse the animal’s flight, make him blind, etc. ; just as the essential part of healing consists of practices designed to reveal the malevolent cause of the malady and give the medicine-man power to exorcise it; so too the essential features of the funeral rites which take place in the first days after death are the practices which definitely divide the dead from the com- pany of the living. They prevent him from mingling with them henceforward, and bring about his admission to the society of which he will hereafter form a part. Not that all relations between the living and the dead are broken off; we shall shortly discover that such is not the case. Hence- forward, however, these relations are to be controlled. In return for the observance of the regulations made, the dead man, being pacified, will demand nothing more, and the survivors on their side have nothing more to fear from him. On the other hand, it is absolutely necessary that these ceremonies should take place. We know that the Athenian and Roman citizens used to think respecting this much as do the Chinese and most of the peoples actually known at the present time. On the west coast of Africa, for instance, ‘ when a person dies abroad, the family try to obtain some- thing that appertained to him, such as pieces of his hair, or nail-cuttings, over which the funeral ceremonies are then performed ; for the general belief is that the ghost or soul lingers near the remains until these ceremonies are performed ; and either cannot or will not depart to Dead-land before them. Hence, to declare to a criminal that, after his execution, no funeral rites will be held over his body, is to him more terri- fying than death itself; for the latter merely transmits him to another sphere, where he continues his ordinary avocations, while the former opens to his imagination all kinds of ill- defined terrors.”’ 1 Therefore without going into the extraordinarily varied details of the ritual which takes place from the moment the sufferer expires until the more or less distant day when the obsequies, properly so called, are carried out, and above all, without disputing the fact that by the way in which these are performed, various distinct ends are often desired ; (such, for 1 A. B. Ellis, The Ewe-speaking Peoples, p. 159. PRIMITIVES’ IDEA OF DEATH 315 example, as avoiding contamination from the proximity of the corpse, helping the weakness of the deceased, who is not com- pletely in possession of his human body and has as yet not attained to his ghostly one, safeguarding the survivors from his violence, interrogating the dead to ascertain who killed him, and so on,) we may admit that the general tendency of these customs is to what-is mystical, and that they aim at determining, with or without his concurrence, the position of the dead man to his own satisfaction and the safety of the survivors. From a shaman who had been resuscitated, Boas received the following very interesting account relating to the experiences he had undergone in the few days immediately after his death. ‘‘ When I was dead, I did not feel any pain. I sat by my body, and saw how you prepared it for burial, and how you painted my face with our crest. . . . After four days I felt as though there was no day and no night.” (Thus, in the course of these four days, the deceased was by degrees removed from the ordinary conditions of life, which were still his immediately after his death; during this period, which extends from the moment of death until the celebration of the burial rites, everything that happens is calculated to weaken the connection between the deceased and his living state, and to prepare him for other relations.) ‘I saw you carrying away my body and felt compelled to accompany it, although I wished to stay in our house. I asked every one of you to give me some food, but you threw it into the fire, and then I felt satisfied. At last I thought, ‘I believe I am dead, for nobody hears me, and the burnt food satisfies me,’ and I resolved to go into the land of the souls.”’ 1 This shaman, like those to whom he is speaking, has no doubt that the soul wants to dwell among the living, and that it would have remained there, indeed, if the funeral ceremonies had not forced it to follow the corpse. | It may also happen that the soul of the dead does not retire to a distance immediately after the celebration of the obsequies. Among the Zuñis, for instance, although the interment takes place at once, the soul haunts the village for four days after death, and does not start on its long journey 1 F. Boas, ‘ The North-west Tribes of Canada,” Reports of the British Association, p, 843 (1889). 316 HOW NATIVES THINK until the morning of the fifth.t Conversely, we often read of practices designed to expel the spirit of the dead even before the obsequies. The Baidyas of South India believe that the ghost of a dead person haunts the house until the fifth day. ‘* Before retiring to bed on the evening of this day, the inmates sprinkle the portico with ashes from the spot where the de- ceased breathed his last, and take great care to abstain from walking thereon, or approaching the sprinkled spot, lest the ghost should strike them. Early next morning they examine the ashes, to see if the marks of the cloven foot of the ghost are left thereon. If the marks are clear, it is a sign that the ghost has departed ; otherwise a magician is called in to drive it out.”’ 2—‘‘ Among the Tyans of Malabar, on the morning of the third day . . . the nearest relative brings into the room a steaming pot of savoury funeral rice. It is immediately removed, and the spirit, after three days’ fasting, is understood to greedily follow the odour of the tempting food. They at once close the door and shut out the spirit.’’3 Lastly, that we may not multiply facts that are similar, we are told that among the Iban or Dyaks of Sarawak, when night has come, the manang or medicine-man celebrates a ceremony that is called Baserara, i.e. separation. . . . They believe that this ceremony separates the soul of the dead from those of the living ; thus they cause him to forget the living, and deprive him of the power to return and carry away with him the souls of his relatives and friends.4 Have we not here as clear a symbol as possible of the rupture of the relation between the soul of the dead man and his social group? Nevertheless, these more or less ingenious artifices, and these infinitely varied magical practices by means of which the soul of the dead is expelled or excluded, by depriving him of the desire to return to the place he occupied when alive, would not be a sufficient guarantee if the funeral rites, properly so called, i.e. the ceremonious performance of the obsequies, did not settle the dead in the state henceforth to be his, at least for a time. 1 Stevenson, The Zuñis, p. 307. 2 Thurston, Ethnographic Notes in Southern India, p. 307. 3 Ibid., P. 218. 4 Dunn, The Religious Rites and Customs of the Iban or Dyaks of Sarawak,” Anthropos, i. p. 170 (1906). PRIMITIVES’ IDEA OF DEATH 317 III The period which elapses between the first funeral cere- monies and that which puts an end to the time of mourning varies considerably in length. It may be a few weeks, or a few months, or even longer. It sometimes happens, as Hertz has demonstrated, that the final ceremony tends to be confined to certain rites only, or even to be confounded with the first obsequies. With most primitive peoples, however, there is still a clear distinction. This period differs from the pre- ceding one in that the terror inspired’ by the dead man has subsided. No longer is he, a dread and unhappy spectre, to be felt prowling around, ready to inflict injuries upon the survivors, and drag them, too, to their death. The mystic force of the funeral ceremonies has severed, to some extent, his participation in the group of living beings. It has secured for him an assured position, and at the same time it has ensured peace to the survivors. Certain relations persist, however: the custom of not neglecting the dead, but of bringing him at regular intervals food and other offerings ; of conciliating him ; above all, of not irritating him ; —(a custom which is general everywhere,) all these prove that even yet the dead man retains some power over the living. On both sides there are duties to fulfil and rights to maintain, and though the dead man is outside the group of the living, he is not yet a complete stranger. In the Arunta tribe, for instance, ‘‘ within a very short time of death the body . . . is interred . . . in a round hole in the ground, the earth being piled directly on to the body, so as to make a low mound with a depression on one side. This is always made on the side which faces towards the direction of the dead man or woman’s camping ground in the Alcheringa, that is, the spot which he or she inhabited whilst in spirit form. The object of this is to allow of easy ingress and egress to the Ulthana or spirit which is supposed to spend part of the time until the final ceremony of mourning has been enacted in the grave, part watching over near relatives, and part in company of its Arumburinga, that is, its spiritual double. . . ."’? 1 “ La Représentation Collective de la Mort,” Année Sociologique, X. Pp. 120, 2 Spencer and Gillen, The Native Tribes of Central Australia, p. 497. 318 HOW NATIVES THINK Even when the obsequies have taken place, the dead man enjoys perfect liberty to come and go, and he pays great attention to the conduct of his relatives with regard to him. In the Northern tribes, ‘‘ the spirit of the dead person, called ungwulan, hovers about the tree (where the body is placed), and at times visits the camp, watching, if it be that of a man, to see that the widows are mourning properly. . . . It is consulted as to the time when the final ceremony shall take place.” ! | | It is quite natural that during this period the needs of the dead shall be provided for. In New Guinea, according to Edelfelt’s evidence, for some time great care is taken of the grave; trees are planted there, and feasts given at regular intervals in honour of the dead,?—feasts in which he himself naturally takes part. So too, with the Bororos of Brazil. “The first interment takes place on the second or third day. . . . The corpse is buried in the forest near water ; after about two weeks there is no flesh remaining, and the final festival, of which the aim is to decorate and wrap up the skeleton, is then celebrated. In the interval relations are maintained with the dead during the day, and above all during the night, by means of funeral songs. After the second ceremony nobody troubles about the dead again.’’3 In California the Yokaia are accus- tomed “to feed the spirits of the dead, for the space of one year, by going daily to places which they were accustomed to frequent while living, where they sprinkle food upon the ground. A Yokaia mother who has lost her babe goes every day to some place where her little one played while alive, or to the spot where its body was buried, and milks her breasts into the air. . . . Like the Yokaia and the Konkau, the Senel believe it necessary to nourish the spirits of the departed for the space of a year.” 4 This very rigid obligation is imposed for a limited time only. Whilst the dead man remains in the vicinity, and comes and goes as he will, whilst he exercises surveillance over the group of which he once formed part, he has a right to many things, and he demands that all that is his due shall be * The Northern Tribes of Central Australia, p. 530. > Edelfelt, Customs and Superstitions of New Guinea Natives, p. 20. 3 K. von den Steinen, Unter den Naturvôlkern Zentralbräsiliens, pp. 390-6. + Powers, Tribes of California, pp. 166, 167, 171. PRIMITIVES’ IDEA OF DEATH 319 rendered him. With the second ceremony all this is at an end. It is permissible to think that the chief, if not the only, object of this ceremony is definitely to break the bond which allows the dead person, despite all, to participate to some extent in the life of the social group. The second or, accord- ing to the circumstances, the final ceremony perfects the death and makes it complete. Henceforth the soul of the dead man has no personal influence upon his social group, at any rate for a period of indefinite duration in which it is awaiting reincarnation. Therefore, after a minute descrip- tion of the final ceremony, in which the most important rite is the breaking of a bone, Spencer and Gillen add : “ When once _ the ceremony of breaking the bone . . . has been performed, and the bone deposited in its last resting-place, the spirit of the dead person, which they describe as being about the size of a grain of sand, goes back to its camping-place in the Win- gara, and remains there in company with the spirit parts of other members of its totem until such time as it undergoes reincarnation.” In Hertz’s monograph, La Représentation Collective de la Mort, he has cited a great number of cases, principally taken from the peoples of the Malay Archipelago, in which we find that death takes place in two parts, and is only considered to be completed after the second of these. I shall confine myself to relating some similar facts noted among the social groups of American and African primitives. With the Sioux, ‘‘ when a son dies, the parents with a knife cut off some hair from the top of the head, just above the forehead, placing the hair in a deerskin cover... . (They bring it offerings at regular intervals.) . . . At a certain moment the bag is opened, and the hair or ghost is taken out and buried. From this time the parting with his parents is absolute. They think that, until the hair is buried, the deceased is really present with the household, and that when this burial takes place, he dies a second time.” ?—In British Columbia, “a year after the death of a person his relatives collected a large amount of food and clothes, and gave a new feast on the grave. This was the end of the mourning period, and thence- forth they tried to forget the deceased. At this feast his son 1 The Northern Tribes of Central Australia, p. 542. 2 Dorsey, ‘‘ Siouan Cults,” E. B. Rept., xi. pp. 487-8. 320 HOW NATIVES THINK adopted his name,’’! This last is a significant trait, for the name of a person forms a part of his individuality. In Mexico the Tarahumares celebrate three successive feasts. At the first, which takes place less than a fortnight after the death, all the mourners, headed by the shaman, speak to the dead man, and he is entreated to leave the sur- vivors in peace. . . . The second is held six months later. . . . Three men and three women take food and drink to the tomb ; the relatives remain at home. The third is the final effort to get rid of the deceased, and the proceedings terminate with a race between the young men. “They all come back rejoicing, and show their satisfaction by throwing into the air their blankets, tunics, and hats, because now the dead is at last chased off. . . . According to the names which the Tara- humares apply to the three functions, the main idea of the first is to give food, of the second to replenish the first supply, of the third to give drink. Each generally lasts one day and one night, and begins at the hour at which the dead breathed his last. . . . There are three feasts for a man and four for a woman. She cannot run so fast, and it is harder therefore to chase her off. Not until the last function has been made, will a widower or widow marry again, being more afraid of the dead than are other relatives.’’ 2 But these same Tarahumares, having once celebrated the final ceremony, know that they have nothing to fear, and act accordingly. “The Tarahumares,’’ says Lumholtz, ‘‘ had no great scruples about my removing the bodies of their dead, if the latter had died some years before and were supposed to have been properly dispatched from this world. One Tarahumare sold me the skeleton of his mother-in-law for one dollar.” 3 “The Huichols are not afraid of dead who passed — out of life long enough ago.” 4 Knowing that Lumholtz was in search of human skulls they readily brought him some. It is the same thing in Equatorial Africa. The negroes ‘‘ fear the spirits of the recently departed, and besides placing furniture, dress and food at their graves, return from time to time with other supplies of food. During the season appointed for 1 F. Boas, “ The North-west Tribes of Canada, Reports of the British Assoctation, p. 643 (1890). 2 C. Lumholtz, Unknown Mexico, i. pp. 384-7. 3 Ibid., p. 390. bids a> Dress. PRIMITIVES’ IDEA OF DEATH 321 mourning, the deceased is remembered and feared, but when once his memory grows dim, the negro ceases to believe in the prolonged existence of the departed spirit, . . . Ask a negro about the spirit of his brother who died yesterday, and he is full of terror ; ask him about the spirit of those who died long ago, and he will tell you carelessly: ‘It is done,’ that is to say, it has no existence.” ! In Ceylon, ‘‘ the Weddahs no longer betray any fear of the skeleton of a man who died long ago. We never had the least difficulty in collecting Weddah skeletons. The natives readily showed us the place where, according to the instructions of the English inspector, they had buried them. Whilst we were extracting the skeletons from the ground, they nearly always watched us with interest, betraying not the slightest emotion ; and if it were necessary to search in the dust for the little bones of the hands and feet, they were always ready to help us.” ? Hertz has clearly shown why the ceremony which closes the mourning period is divided from the early obsequies by an interval which varies in length, but which is, as a rule, fairly long. In order that the final ceremony may be celebrated and the dead be relegated to a sufficient distance to preclude him from any intention to return, as well as to enable him to join other spirits awaiting their reincarnation, he himself must be absolutely discarnate. The flesh must have entirely dis- appeared from his bones, and the process of decomposition have been accomplished. This stands out clearly in the many cases related by Hertz and also in the detailed description of the funeral ceremonies of the Australian aborigines furnished by Spencer and Gillen. The decomposing corpse is visited from time to time, and the spirit which haunts its vicinity is asked at what moment it considers the bones to be sufficiently bare for the final ceremony to take place. I find, too, a confirmation of Hertz’ theory in a belief which _ is very common among primitives, and which is still main- tained in China. Certain ghosts are particularly dangerous and malevolent ; they are homicidal, and the source of horror and affright when they appear. It is found, moreover, that Du Chaillu, Equatorial Africa, p. 330. 2 P, and F. Sarasin, Ergebnisse naturwissenschaftlicher Pen auf Ca. iil. Pp. 494, 21 322 HOW NATIVES THINK when the graves are opened, in order that these terrible spectres may be “ laid,” they are seen not to have undergone decomposition at all. In Loango, “when, on opening the tomb, the corpse is found intact, his eyes open, .. . it is destroyed by burning.” —In East Africa, “a near relative of the deceased, wife, husband or sister, dreaming of the departed night after night, for some weeks after death, wakes up in terror, goes and looks out of the hut, and finds the ghost of the deceased sitting near the door of the hut ; or it often happens that the ghost is seen sitting on the children’s playing-ground close to the village where the deceased was wont to play when a child. The ghosts are always much bigger than life-size. . . Then the grave is opened by one of the near relatives of the deceased, generally a brother, and the corpse is invariably found to be quite undecomposed and white. Itis then taken up and burnt, and the ashes carefully re-interred. This procedure lays the ghost.’’2 De Groot tells of many similar cases.3 Thus the dead man whose body does not decay is peculiarly to be dreaded. He is abnormal, because he cannot proceed to that complete death which will definitely separate him from the living. He haunts and persecutes them by reason of his inability to become discarnate, and pass gradually from the first death to the second. Does not this collective representa- tion, too, make manifest the mystic, prelogical character of primitive mentality ? The law of participation which governs this mentality causes the primitive to regard the mystic relation which binds spirit to flesh and bones as perfectly simple, incomprehensible as it appears to logical thought. In one sense, the dead man 7s this flesh and these bones; in another sense, he is something quite different, and the two propositions are not mutually exclusive, because, to the primitive mind, “to be’”’ means ‘to participate of.” The decay of the dead man’s flesh is what logical thought would define as the sign, the condition, cause, and the very fact of the second death. When it has taken place, death, too, has been accomplished, i.e. the bond between the individual and his social group is definitely ruptured. * Dr. Pechuél-Loesche, Die Loango-Expedition, iii. 2, Dw Atm ? Hobley, ‘“ British East Africa: Kavirondo and Nandi,” J-A.T., xxxiii. PP: 339-40. <0 LÉ 3 The Religious System of China, i. pp. 106, 127. PRIMITIVES’ IDEA OF DEATH 323 IV Among the customs accompanying either the first or the second funeral ceremony, or even both, I shall refer, by way of example, to a series in which the mystic, prelogical nature of primitive mentality is clearly to be seen. As a general rule it consists of burying with the dead, or else simply destroying, everything that has belonged to him. Such a custom is maintained everywhere among undeveloped peoples, almost without exception. We learn that the Aruntas, in the case of a man’s death, cut off his hair, but his neck and arm ornaments and the fur string used for winding round his head are all carefully preserved to be used afterwards. (These are indeed objects possessed of highly magic value which the Australian has himself received either from ancestors or other relatives. He has merely been, as it were, enjoying the usufruct thereof, and they somewhat resemble the churinga. . . .) But as soon as the interment has taken place, the camp of the man or woman is immediately burnt, and everything in it destroyed. In the case of women, absolutely nothing is preserved.1 . In South Australia, too, ‘‘ all that belonged to the dead man, his weapons, fishing-nets, etc., are placed in the grave with his body.”’? In the Victoria district, the medicine-man throws into the grave all the personal effects of the dead man that he has been able to collect . . . he then inquires if there is any- thing else, and if he hears of anything he has it brought and put with them. Every single thing that has belonged to the living must be placed near his corpse.3—In the Bismarck Archipelago, “‘ all the deceased’s movable property is placed on the tomb; and only after the expiration of three weeks is it burnt.” 4—The Bororo suffer considerable loss when a member of the family dies, for everything he has used is burnt or thrown into the river, or buried’ with him.s Dobrizhoffer tells us the same about the Abipones. “All the utensils belonging to the lately deceased are burnt on a bier. Besides : Spencer and Gillen, The Native Tribes of Central Australia, p. 497. 2 Beveridge, “ The Aborigines of the Lower Murray,” Journal of the Royal Society of New South Wales, p. 29 (1884). ; 3 Brough Smyth, The Aborigines of Victoria, 1. p. 104. 4 Parkinson, Dreissig Jahre in der Stidsee, p. 441. s K. von den Steinen, Unter den Naturvolkern Zentralbräsiliens, pp. 384-9. 324 HOW NATIVES THINK the horses killed at the tomb, they slay his small cattle if he have any. The house which he inhabited they pull entirely to pieces. His wife, children and the rest of the family remove elsewhere, and having no house of their own, reside for a time in that of some other person, or lodge miserably on mats. To utter the name of a lately deceased person is reckoned a nefarious vice amongst the Abipones.’’! In California the Komacho sacrifice everything belonging to the deceased, even to his horse. With the Nishinam, as soon as life is extinct the body is burned, and with it everything the deceased possessed. The Wintus throw into the grave everything that they possibly can of a man’s former belongings, knives, rakes, old whisky bottles, oyster-cans, etc., and everything that cannot be buried is burnt. When an Indian of high rank dies his wig- wam is burned down. ‘The name of the dead is never mentioned more.” —‘‘ The Hurons either bury or else enclose with the corpse cakes, oil, skins, axes, kettles, and other implements so that the soul of their relative may not dwell in poverty and distress for lack of these things.” 3 In Vancouver, everything belonging to the dead is placed near his body; otherwise he would return to take it away. Sometimes even > his house is razed to the ground.4 The Zuñis, too, destroy or burn nearly everything that has belonged to a dead man. The same custom is widely prevalent throughout Africa, and even where it no longer obtains, we still find traces of it. On the Slave Coast, for instance, “‘ the children are not the sole inheritors of their parents: the uncles on both sides having also proprietorial rights. For this reason the children take all the valuables from their father’s house when they see his end approaching. And his brothers hasten likewise in the last hours to take as much as they can of his movable possessions.’ The same missionary, however, informs us: ‘‘ The dead man has been placed in his grave with many coverings, but nothing else is placed inside or upon it. Formerly pots containing fatty substances were broken over it, but this custom is no longer observed.” 6—In the Ba-yaka tribe, ‘when a man dies all his 1 Dobrizhoffer, op. cit., ii. pp. 273-4. 2 Powers, Tribes of California, pp. 173, 239, 328. 3 Fr. Sagard, Le Grand Voyage au Pays des Hurons, p. 233 (1632). ‘ F. Boas, op. cit., p. 575. 5 Spieth, Die Ewe-stämme, p. 120. 6 Ibid., p. 256. PRIMITIVES’ IDEA OF DEATH 325 vessels are shattered, and the pieces left on his tomb.’ In South Africa, as we learn from Macdonald, when the funeral solemnities have been completed and the party of mourners has dispersed, the house which the deceased was occupying when his death occurred is burnt, together with all its contents, even valuable things, corn, implements, weapons, ornaments, charms, as well as furniture, beds and bedding: everything must be destroyed by fire.? In South India, as soon as a Savara is dead, a gun is fired at his door to help the spirit to escape. The corpse is washed, taken to the family burying-ground, and burnt. Everything a man possesses—bows, arrows, hatchets, daggers, necklets, clothes, rice, etc.—is burned with his body.3 And lastly, not to prolong the list unduly, we have De Groot’s testimony that in China at one time a man’s death entailed the complete ruin of his family. By degrees the custom of burying all his valu- ables with the deceased fell into disuse, though it has not entirely disappeared. At the same time, he tells us, the ao, or filial obligations, became more and more imperative. Accordingly children, while avoiding the actual renunciation of their parents’ estate, were all the more zealous in preserving the outward aspect of such a renunciation . . . by dressing in the cheapest garments and eating the simplest food possible.4 The current interpretation of such practices as these refers them to the following general motives: to provide the dead with all that he requires, so that he may not be unhappy in his new state, and, if it is a case of some important personage, to furnish him with the means of maintaining his rank ; —to rid the living of objects which death has rendered unclean and therefore unusable; (this would explain, for instance, the almost universal custom of burning or pulling down the house in which the death has occurred) ;—to avoid any risk that the dead man, exercising jealous supervision over the survivors, may be tempted to return and seek his possessions. That such motives, or at any rate one or other of them, should influence : Torday and Joyce, “‘ Notes on the Ethnography of the Ba-yaka,” J 7: 19 XXXVI. P. 43. a2 Macdonald, ‘‘ Manners and Customs... of South African Tribes,’’ Punt, xix... 27G, 3 E. Thurston, Ethnographic Notes in Southern India,” p. 206. 4 The Religious System of China, i. p. 474. 326 HOW NATIVES THINK those who practise these observances, cannot be doubted in a great number of instances. Very often both explorers and missionaries expressly mention them. Occasionally, however, they ask themselves whether these motives are really sufficient to account for a custom so extra- ordinary, and apparently so subversive of the evident interests of the survivors. In the Congo, for instance, ‘‘ the dead man is first adorned with all his ornaments, and everything he possesses that is valuable, for all this must perish with him. Why should this be so? If it were from motives of avarice, that he should not be separated from his wealth even in the grave, the same sentiment would influence his heir, who would refuse to be thus mulcted. From all that I have been able to gather, they act in this way from blind obedience to the Kissy who enjoins it upon them, and they are too ignorant to be able to argue about their religion, to which they implicitly submit.”’ 1 It is true that it occasionally happens that observers, in relating these facts, confuse them with what seems to them the most natural explanation. We do find, however, that such motives are explicitly attributed to them. ‘“‘ They think and believe that the spirits of these utensils, axes, daggers and all that they dedicate to them (especially at the great Feast of the Dead) proceed to the other world to be of use to the spirits of the dead, whilst the bodies of these skins, utensils, axes, etc., dwell in the bier and in the grave with the bones of the dead. This was their invariable reply when we told them that mice ate the bread and oil, and that moth and rust attacked the skins and the weapons that they buried in the graves of their relatives and friends.”’ 2 Yet very often these customs are maintained even when their original meaning has been lost. Those who still observe them never fail to interpret them in accordance with their ideas and sentiments at the time, just as myths may be hidden under many layers of additional illustrative matter of a con- trary meaning, when the collective representations in which they originated have been modified by the social milieu. We may indeed admit that with primitive peoples among whom — * Degrandpré, Voyage à la Côte Occidentale d’Afrique, i. pp. 147-8 (1801). > Fr. Sagard, Le Grand Voyage au Pays des Hurons, pp. 233-4 (1632). PRIMITIVES’ IDEA OF DEATH 327 the custom of destroying all that belongs to the dead still obtains, the motives we have indicated above are those which natives attribute to themselves, and yet ask ourselves whether, as a matter of fact, these customs ought not to be referred to other collective representations which are peculiar to mystic, prelogical mentality. According to our view, these customs imply a special participation both imagined and felt. The things that a man has used, the clothes he has worn, his weapons, ornaments, are part of him, are his very self, (construing the verb “ to be ” as ‘‘to participate’’), just like his saliva, nail-parings, hair, excreta, although to a lesser extent. Something has been communicated to them by him which is, as it were, a con- tinuance of his individuality, and in a mystic sense these objects are henceforward inseparable from him. By virtue of a kind of polarization, they are not weapons and ornaments in general: they are the weapons and ornaments of So-and-so, and they cannot be deprived of this distinguishing character- istic, or become the weapons and ornaments of any other. Now primitive mentality regards the mystic features of things and their occult properties as by far the most important. Oriented differently from our logical thinking, this mentality tends to methods of functioning which are quite unlike our own, and, from the utilitarian point of view, often irrational. For instance, the chief may have resolved upon a hunting expedition early the next day, and instead of going to rest so that they may be thoroughly refreshed and ready, the Bororo will spend the night in singing and dancing: Von den Steinen, who is greatly astonished at such a proceeding, does not realize that in the opinion of the Bororo the capture of the game depends far more upon the mystic influence exercised upon it by the songs and dances, than upon the skill and agility of the hunters. Again, when we consider the immense amount of work which the manufacture of their weapons, canoes, and implements often entails upon the natives, we feel inclined to ask how they can sacrifice the product of so much effort and patience every time that a death occurs. But to them the efficacy and utility of weapons and implements is quite a secondary matter, compared with the mystic bond 1 Unter den Naturvôlkern Zentralbräsiliens, p. 367. 328 HOW NATIVES THINK which unites them with the individual who has made and used and possessed them. If this man should die, what is to be- come of his belongings ? Such a question does not even occur to the primitive mind.1 There are not several possible alter- natives, for so strong is the participation between him and all that belongs to him that the idea of utilizing these things apart from him cannot present itself. These things must of course go with him. They will be placed near his body, and since they are usually regarded as animate, they, too, will pass into the neighbouring region whither death (in its first stage) has transferred him. We can readily perceive that the motives just appealed to do not contradict the mystic origin of such practices. It would be dangerous to appropriate these things, for instance, for he who made use of them would run the risk of arousing the dead man’s anger, and might be the object of his ven- geance. Or again, the dead man may be grateful for the care that has been taken in seeing that his belongings accompany him and, as a recompense, he will refrain from disturbing the peace of the survivors, etc. These motives are only secondary, however. They are to be accounted for by the original mystic bond between the dead man and his possessions, whilst this bond is not to be accounted for by them, but arises directly out of collective representations familiar to prelogical mentality. To it, at this stage, possession, property, use, are not distinguishable from participation. This mystic bond cannot be destroyed by death, seeing that the dead man continues to live, and his relations with the social group are not ruptured ; far from it. As we should express it, he remains the proprietor of everything that is his, and the customs we are studying merely attest that this property is recognized as his. We must simply understand that, at the moment, this property constitutes a mystic relation, a participation between the possessor and the things possessed. It is not even per- missible to conceive of the property passing into other hands, or, if it does pass, that it may be of any use whatever,—to say 1 “ I asked him again why they buried the clothing of the dead with them. ‘They belong to them,’ answered the Indian, ‘why should we take them away ?’”’ (Relations des Jésuites, v. p. 130 (1633).—"" ‘ We never speak about the dead again,’ he said to me; ‘even the relatives of a dead man never use the things which he used when alive.’ ”’ (Ibid., v. p. 134.) PRIMITIVES’ IDEA OF DEATH 329 nothing of the dire consequences which the very violation of this mystic bond would entail. Although investigators are, as a rule, very far from thinking thus, it often happens that their language more or less clearly confirms the idea. In the first place, they often tell us that the things buried, shattered, or sacrificed are the personal possessions of the deceased, i.e. those things he had himself made, or had exclusively used. The destruction does not usually extend to objects which would be the property (speak- ing in the mystic sense) of other members of his family. ‘A widow retains all the baskets and trinkets made by herself.”’ 7 We are told that in New Guinea, some of the bows and arrows belonging to a dead man and most of the things he has used, are broken to pieces on his grave and left there to witness to the fact of his incapability to use them more. A similar practice takes place with regard to cooking utensils, and the materials for women’s work ; her skirt is also left on the tomb, with everything she was wearing at the time of her death.? Sometimes, too, the observer points out the mystic nature of the bond uniting the dead with the objects he possessed. ‘ The dead man’s garments, weapons, utensils are buried with him . . . his canoe of bark is placed upside down on his tomb or else set adrift on the stream. All the objects which have belonged to him and which cannot be concealed with him are sacrificed. They are burned, thrown into the water, or else they are hung in the trees, because they are etn’ary étay, i.e. ‘anathema.’ This is a new species of taboo, the use of which is often found elsewhere.’’3 The comparison is, in my opinion, a very just one; seeing that this taboo is imposed with regard to those who form part of the same social or religious group, but is not binding on others. So, too, with the Ba-Ronga, when a man is dead, “ his clothes and every- thing else he wore are thrown into his deserted hut. His dishes and drinking-vessels are broken on his grave; nobody dare touch them again . . .” Junod adds in a note, how- ever: ‘except the Christians. . . . One of our converts at Rikatla, Lois, told me smilingly that she had been well able 1 Powers, Tribes of California, p. 249 (Shastika). a Edelfelt, ‘‘ Customs and Superstitions of New Guinea Natives,” Proc. of Roy. Geo. Soc. of Australia, vii. p. 20 (1891). 3 Petitot, Dictionnaire de la Langue Déné-dindjié, p. xxvi. 330 HOW NATIVES THINK to complete her store of crockery by buying that of a dead man, his heirs having parted with it for almost nothing.” ! This last is a very significant fact. The mystic bond between a dead man and his possessions is doubtless no longer powerful enough in the Ba-Ronga people for it to be necessary to reduce them to the same condition as his present one. They are, however, taboo for the survivors, who destroy them rather than make use of them. But the proof that they do not destroy them so that the dead shall not use them, or for fear he may come to look for them, is seen in the fact that they do not scruple to sell them to natives who have become Christians. In the same way, among the Hos, the Dravidians of Bengal, on the death of a man all his personal effects are destroyed. This is not with the idea that the dead derives any advantage from it, however. ‘ The Hos have always given a negative answer to this question. They gave me the same explanation as the Chulikata Mishmis of Upper Assam had given: namely, that they are unwilling to derive any immediate benefit from the death of a member of their family, so they commit to the flames all his personal effects, the clothes and vessels he had used, the weapons he carried, and the money he had about him. But new things that have not been used are not treated as things that he appropriated, and they are not destroyed. It often happens that venerable old Hos abstain from wearing new garments which they have become possessed of, to save them from being wasted at the funeral.” 2 We could hardly find an instance that more clearly shows that the very essence of property is a mystic bond established between the owner and the objects which participate in it in some way, because they have been used or worn by him, and that if these things are destroyed when their owner dies, it is because death does not break the mystic bond. The participation continues to exist, and on the one hand it is opposed to any use being made of the things, and on the other, it decides the customs which put them, as it were, at the disposal of the dead. It is not even indispensable that the things should be destroyed. A dead man may remain the owner of living wealth. ‘ There are spirits (dead) who become very rich in cattle and slaves, © x Junod, Les Ba-Ronga, p. 58. 2 Risley, Tribes and Castes of Bengal, i. p. 334. PRIMITIVES’ IDEA OF DEATH 331 by reason of the continual offerings made to them: these cattle are regarded as sacred, and the parents of the dead man to whose spirit they belong guard them carefully.” ! Possibly it is in this persistent participation between the dead and that which is mystically united with him as property, that we must seek for the reason of a good many customs relating to mourning, especially to customs which are often so cruel, so complicated, and so prolonged as those imposed on widows among certain primitive peoples. There are many facts which would tend to such a belief. Firstly, as a rule, the widow ceases to observe these practices at the precise moment when the ceremony which ends the mourning has taken place, i.e. the moment when the second time of death is over ; when, death having been perfected, the relations between the deceased and the social group are definitely ruptured. But during the time that elapses between the death and the final ceremony the dead man, even after the first obsequies, pays special attention to the behaviour of his widow. He watches over her, ready to intervene if she does not observe every detail of the mourning. The dead man’s near relatives under- take to see that she does not avoid any of them: otherwise, they may chastise her, and sometimes even kill her. There- fore the bond between her and her dead husband must be a very strong one. In the second place, during his lifetime this bond in many respects resembled that which we have termed property, in the mystic sense in which prelogical mentality conceives of it. With many primitive peoples, from the day of her marriage a woman who until then had been allowed the greatest sexual liberty, becomes taboo for the members of any group except her husband’s.2 She only belongs to him because he has acquired her, sometimes at great cost, and adultery is accordingly a kind of theft. Between him and her a participation has been established which no doubt makes her dependent upon him, but at the same time makes her actionsreflect uponhim. If heis hunting or fighting, for instance, t J. Roscoe, “‘ The Bahima, a Cow Tribe of Enkole,”’ J.A.J., xxxvii. p. 109. 2 With the Maoris of New Zealand, for instance, formerly every woman . could select as many companions as she liked, without being thought guilty of any impropriety, until given away by her friends to someone as her future master; she then became tapu to him, and was liable to be put to death if found unfaithful’? (R. Taylor, Te Ika a Maui, p. 167). 332 HOW NATIVES THINK his success and his very life may be endangered by some im- prudence or unconsidered act of his wife’s. If he repudiates her, the mystic tie is ruptured; but if he has not repudiated her, and he dies, the participation between her and the dead man still subsists with all the consequences it entails. _. Strictly speaking, these should involve the death of the widow. We find innumerable instances of such a custom, even among peoples already somewhat civilized, especially when the dead husband was a person of some importance. At the death of more than one petty African king, his wives, or at least some of them, sacrificed themselves. Formerly, Ellis tells us, the king had hardly breathed his last than the women of the palace would begin to break the furniture, ornaments, vessels, and lastly, kill themselves.1 We know that in India, in the Far East, above all in China, the suicide of widows uponstheir husbands’ graves is still very common. De Groot speaks of this in very significant terms. “ The most numerous class,” he says, “is that which comprises the suicides perpetrated by widows wishing to escape the chance of being remarried or of being in some other way deprived of their chastity. Indeed, being the property of her husband after his death, a woman of good principles cannot but consider it an act of the highest injustice towards his manes, nay, of theft, to surrender herself up to another . . . and so rejoin him in the life hereafter in a state less pure than that in which he left her behind. These considerations are obviously very old, being traceable to a certain tribe . . . which was in the . habit of casting out many a widowed wife into the wilderness because she was now wife to a spirit, treating her in fact as the Chinese of the present day generally do the inanimate personal effects of the deceased.”’ 2 But the widow is “ the wife of a spirit ” for a time only, until the ceremony which perfects death and breaks the last tie between the dead man and his social group can be cele- brated. She may therefore be allowed to live, provided that she does so in a way which will not anger the spirit of the dead who is still her master, and cause him to return to vindi- cate his rights, and thus disturb the peace of the social group. 1 A. B. Ellis, The Ewe-speaking Peoples, p. 128. 2 J. J. M. de Groot, The Religious System of China, i. p. 744. PRIMITIVES’ IDEA OF DEATH 333 This accounts for most of the practices enjoined upon the widow, even though those who impose them are frequently obeying other impulses, such as avoiding contamination (for a widow is rendered unclean by the death of her husband, and might communicate this impurity to the survivors, etc.). Moreover, in these utilitarian motives there still remains a trace of the mystic bond whence the original meaning of such practices is derived. But these motives are secondary ones, whilst the mystic bond is the main point. Without entering into the very complex details of the mourning customs which it is not my object to explain here, a few examples will suffice to show that until the ceremony which makes death complete, the woman in a mystic sense remains the property of her dead husband, and that special rites are necessary to bring this participation to an end. In the northern tribes of Central Australia, ‘the za (younger brother) of the dead man cuts off the widow’s hair, and afterwards burns it. . .. We must add that sooner or later the woman will become the property of the za. The woman’s hair is thrown into the fire, and she covers her body with ashes from the camp fire and continues to do so during the period of her mourning. If she did not do this, the atn- rinja or spirit of the dead, which follows her everywhere, would kill her and strip her bones of their flesh.””! The mourning over, and the dead man departed for his camp in the Alcheringa, the woman is given to one of his younger brothers, but this is not accomplished without much ceremonial. Finally, ‘‘ one night, the /ubra comes to the ttia’s camp, but the two sleep on opposite sides of the fire. On the next day, the itia hands her over to men who stand in the relationship of unkalla, okilia, itia, gammona and oknia, i.e. to men repre- sentative of all classes. All of them have access to her and make her presents of alpita, red ochre, fur-string, etc., which she carries to the itia’s camp, where he decorates her with the string. Previously he has sent by the /wbra, an offering of spears and shields to each man—a gift which is necessary, or else, later on, they might kill him if he had taken possession of the widow without both offering them the present and allowing them to have access to her. If one ztza does not 1 Spencer and Gillen, The Northern Tribes of Central Australia, p. 507. 334 HOW NATIVES THINK want her, then she is simply passed on to another.’’! It is not enough, therefore, that the bond between the woman and her first husband should be broken by death, for the mystic relationship to be established between her and her new master. She must be, as it were, disappropriated, in order to be appro- priated anew, and this can only be done by the brother of her dead husband. We think immediately of the levirate, and without failing to recognize the utilitarian or legal character which this custom has assumed among many peoples, we are inclined to think that it may have originated in collective representations of the same kind as those which induce the Australian aborigines to act thus. Among the Ba-Ronga, the persistence of the tie between the widow and her dead husband is well shown by the follow- ing customs. ‘In the weeks which follow the death of the husband two preliminary acts are carried out: (1) that which is called the escape into the bush. The widow leaves the village of the dead secretly. She goes far away, to a neigh-- bourhood where she is unknown, and there she establishes a connection with somebody or other, a man of loose character, to whom she yields herself. In any case he will not make her a mother. She escapes from him and takes flight once more. (2) She returns home, assured that ‘she has managed to get rid of her trouble.’ She is free from the curse or the stain which became hers on account of her husband’s death. Shortly afterwards, one of the dead man’s relatives, he who is the heir presumptive of this woman, comes to her, bringing her a present . . . begging her to take it to her parents and tell them that So-and-so has ‘come for her.’ Henceforward this man will watch over her. He will look after her harvest, will take her to visit the dead man’s village. But she will not leave her home for a whole year. She will till the land yet once more for the dead husband, and it is only when the last award of the inheritance is made that she goes to her new master and becomes his wife. . . . ‘ But the wife thus obtained by inheritance is not the property of the heir to the same extent as a wife he would have purchased. She remains, at heart, the property of the 1 Spencer and Gillen, The Northern Tribes of Central Australia, pp. 509-10 PRIMITIVES’ IDEA OF DEATH 335 eldest son of the dead. She is only a wife pour le sommeil. The children she had by her first husband will not belong to the second, but to the eldest son of the first. Nor will any of those which may be born of her new relationship belong to the heir. He is deemed to be still working for his dead brother (or his uncle on the mother’s side) and the offspring of this kind of semi-marriage, too, will belong to the real hereditary heir, the eldest son. One of the daughters who may be born to him will belong to him alone. On the other hand, he has the advantage of receiving his bowl of cooked food every evening from this new wife.’’! In this case, the disappro- priation of the widow does not actually become complete and, since among the Ronga tribe property has already been moulded into a legal form, the lasting participation between a woman and her dead husband is interpreted in terms expres- sive of the conditions governing individuals. Some at least, therefore, of the customs relating to mourn- ing would seem to refer to that permanent participation which, in the prelogical mind, corresponds with that which logical thinking regards as the concept of property. At this stage, however, it is not a concept : it is still one of the repre- sentations which are both general and concrete, many examples of which we have already studied, which never appear save enveloped in a complex of mystic ideas and sentiments. The object possessed participates in the nature of the possessor ; the objects possessed by a dead man participate in his nature, at any rate until the closing ceremony of the period of mourning, and they inspire the same feelings as the deceased. For similar reasons the property of a living person is no less inviolable. The objects a man possesses participate in their owner to such an extent that no other would care to take them from him. With the Macusis of Guiana, for instance, “each individual’s property, whether it be his hut, his utensils, or the patch of ground he cultivates, is sacred. Viola- tion of this property, save in case of war, is scarcely possible, and disputes respecting mine and thine are accordingly extremely rare”? It is therefore sufficient to indicate by an outward sign that a person possesses a thing for it to be- t Junod, Les Ba-Ronga, pp. 67 2 Richard Schomburgk, Run A ie “British Guiana, il. Pp. 321. 336 HOW NATIVES THINK come inviolate. ‘‘ We are told that the Cumana natives used to surround their plantations by a simple woollen thread or by a creeper two feet above the ground, and that their property was thus absolutely protected; for it would have been a serious crime to cross this barrier, and all believed that he who tried to do so would shortly die. The same belief still obtains among the Indians of the Amazon.” ! In New Zealand, “a person often leaves his property in exposed places, with merely this simple ¢ohu or sign (a piece of drift timber with something tied round it) to show that it is private, and generally it is allowed to remain untouched, however many may pass that way. . .. with a simple bit of flax on the door of a man’s house, containing all his valuables, his line, or his food store ; they are thus rendered inviolable and no one would meddle with them.’’? In other words, “if anyone wanted to pre- serve his crop, his house, his garments or anything else, he made them fapu ; a tree which had been selected in the forest for a canoe (which he could not yet use) . . . he rendered tapu by tying a band round it with a little grass in it.”’ 3 Since the property is thus made “‘ sacred ” there is no need to defend it by external means, and this is true of that which belongs to groups as of that which is personal. According to Brough Smyth, the area of ground owned by each tribe was definitely known to all its members, and as carefully drawn as if a land-surveyor had defined its limits and reckoned its superficies.4 It is the same with the aborigines of Central Australia, and Spencer and Gillen tell us that these limits are never disputed and that one tribe would not think of appro- priating territory belonging to another. What would it do with it, if it had it? The idea of such territories is above all a mystic one, and that which dominates it is not the quantity of game or of water to be found there; it is its division into “local totem centres ”’ in which the spirits which are awaiting their reincarnation in the tribe dwell. How would it avail another tribe to expel the one whose ancestors have lived there and who still inhabit it in mystic fashion ? The partici- pation between the social group and the soil is so close that 1 Von Martius, Beitrage zur Ethnographie Stid-Amerika’s, i. p. 86. 2 R. Taylor, Te Ika a Maui, p. 63. 3 Ibid., p. 58. 4 Brough Smyth, The Aborigines of Victoria, i. p. 139. PRIMITIVES’ IDEA OF DEATH 337 the idea that the soil may be disappropriated does not even occur to them. In such conditions the property is, as Schomburgk puts it, “ sacred ” : it is inviolable and inviolate as long as the collective representations which we have noted maintain their empire over the mind. In certain cases this mystic sentiment of property may even become an obstacle to exchange. To give away a thing one possesses is giving something of oneself, and therefore it means giving to another a power over oneself. Barter is a process which involves mystic elements, and however advantageous or tempting it may be, a primitive often starts by refusing altogether. Lumholtz tells us “they are chary of selling to strangers. When a Mexican wants to buy a sheep, or some corn, or a girdle, the Tarahumare will first deny that he has anything to sell. . . . A purchase however establishes a kind of brotherhood between the two negotiants, who afterwards call each other navagua and a confidence is established between them almost of the same character as that which exists between compadres among the Mexicans.” 1 V When the period of mourning has been ended by the final ceremony, the death of the individual is complete, in the sense that his relations with the social group of which he was a member when alive are entirely ruptured. If everything that belonged to him has not been destroyed, it is disposed of, and his widow may become the wife of another. His very name, which it was forbidden to pronounce, may be heard once more among certain primitives. Does this imply that all reciprocal influence has disappeared? From the logical point of view, not admitting of contradictory standards, this would seem to be the necessary consequence. It is not so to the prelogical mind, which finds no embarrassment in contra- diction, at any rate in its collective representations. On the one hand, now that the final ceremony has taken place, there is nothing either to be feared or to be hoped from the dead. On the other hand, however, the social group feels that it lives in very close dependence with respect to its dead, in 1 C, Lumholtz, Unknown Mexico, i. p. 244. 22 338 HOW NATIVES THINK general. It exists and subsists only through them. Firstly, it is necessarily recruited from among them. Then there is the respect with which the churinga of the Australian abori- gines, for instance, is surrounded,—for these churinga which represent the ancestors, ave indeed these ancestors, in the meaning of the word according to the law of participation— the totem ceremonies periodically celebrated by. the tribes, upon which their welfare depends, and finally the other insti- tutions testifying to a participation between the living group and its dead, which is most intensely realized. It is not merely a question of the dead who are still near, whose bodies have not yet entirely decayed, but above all of the dead who have started for their camp in the Alcheringa, who are no less present in their churinga, and also in their nanja, in the place in which their mythical ancestor disappeared beneath the soil. Here we have difficulties which would be insurmountable to logical thinking, which cannot admit the multipresence of individuals, and their simultaneous inhabitation of several. different localities. We find it hard enough to conceive that in the period previous to the closing ceremony, which marks the definite departure of the dead, a dead man can at one and the same time dwell in his grave with his body, and yet be present, as a kind of tutelary deity, in the house that was his home; yet to the Chinese, for instance, there is nothing inconceivable in the idea. Still less can we reduce to a thoroughly intelligible schema the collective representations of Australian aborigines with respect to those who are “entirely ” dead. We can neither define in one clear concept the personality of these dead, nor form any satisfactory idea of the way in which the living group participates in their existence, and in which it is ‘ participated ’’ by them, accord- ing to Malebranche’s use of the term. What is actually certain is that this reciprocal participation is real, as we have already seen, that it is not to be confused with that which in other peoples is called ancestor-worship, and that it relates to the characteristics which are peculiar to prelogical mentality. When a child is born a definite personality reappears, or, to put it more precisely, is formed again. All birth is a rein- carnation. “There is a vast ensemble of peoples, negroes, t Vide Chap. II. pp.. 90. et seq. PRIMITIVES’ IDEA OF DEATH 339 Malays, Polynesians, Indians (the Sioux, Algonquin, Iroquois, Pueblo groups of the north-west), Esquimaux, Australians, among whom the reincarnation of the dead and the inheri- tance of the individual name in the family or the clan is the rule. With the peoples of North-western America, the indi- vidual is born with his name, his social functions, and his coat-of-arms ready made. ... The number of individuals, of names, spirits, and rôles in the clan is limited, and its existence is but an ensemble of deaths and rebirths of identical beings. With the Australian aborigines and the Negritos, the same phenomenon obtains, though not so clearly marked . . as a necessary institution. From its beginning the clan is conceived as attached to a certain spatial district, the ancestral home of the totem-spirits, the rocks in which the ancestors are buried, whence the children which are to be conceived issue, and finally, whence are derived the spirits of the totem animals, the reproduction of which is secured by the clan.” ! Birth, just like death, then, is merely the transition from _ one form of life to another. Just as, for the individual, for the first momeats at any rate, the latter is but a change of con- dition and of residence, all the rest being as before, so birth is but the transference, through the medium of its parents, of the child to the light of day. “~ The child is not the direct result of the intercourse ; it may come without this, which merely, as it were, prepares the mother for the reception, and birth also, of an already-formed spirit-child who inhabits one of the local totem centres.’’2 Seeing what the general orienta- tion of prelogical mentality is, and the paramount interest which the mystic factors in every phenomenon claim, would not the physiological aspect of birth disappear from the primi- tive’s field of vision, lost in the infinitely more important idea of the totemic ties subsisting between child and parents ? Birth, like life, death, disease, must inevitably be represented in mystic fashion, and in the form of participation. In the tribes of Northern Australia, ‘‘ descent both of class and totem are strictly paternal. A spirit-child is not supposed to go into any woman unless she be the wife of a man belonging to the same moiety and totem as the spirit. As a general rule, 1 Mauss, Année Sociologique, vol. ix. p. 267. 2 Spencer and Gillen, The Northern Tribes a Central Australia, p. 265. / 340 HOW NATIVES THINK / the spirit is supposed to enter a woman whose children are born into the class to which it, the spirit, itself belongs.” These spirits are, as it were, lying in wait for a possible mother, in the various totem centres in which they reside, and each deliberately chooses the mother it desires, without making any mistake. ‘If the wife of a snake-man were to conceive a spirit at the spot inhabited by bee-spirits, it would simply mean that a snake-spirit had followed the father up from his own place and had gone inside the woman.” A woman who does not desire a child carefully avoids passing these local totem centres, and if she finds it inevitable she runs quickly by, begging the spirit-children not to enter her.! The same idea of conception occurs elsewhere. Among the Baganda, for instance, ‘‘ should a child be still-born, or die in infancy, it is buried at four cross-roads, and thorns are placed upon the grave. Every woman who passes by throws a few blades of grass upon the grave to prevent the ghost from entering into her and the child being reborn.’ In the French Congo, children whose mothers have died in childbed are thrown into the bush, but near the road, “in order that their souls may choose a new mother from the women who pass by.” 3 A belief that is widely prevalent among the tribes of different parts of Australia clearly shows that the reincarna- tion of spirits, or that which we call birth, is conceived of as independent of physiological conditions. White men are thought to be ancestors who have returned to the world. ‘“ Miago assured me that this was the current opinion, and my own personal observation subsequently confirmed his state- — ment. At Perth, one of the settlers, from his presumed like- — ness to a defunct member of the tribe of the Murray river, was visited by his supposed kindred twice every year, though in so doing they passed through sixty miles of what was not infre- quently an enemy’s country.” 4#—“ Since they have found out the existence of the race of white people, they have adopted the notion that their souls will hereafter appear in the bodies of such white people. . . . It may be instanced as a proof of how firmly they do believe, or rather, have believed this, that t Spencer and Gillen, The Northern Tribes of Central Australia, p. 170. 2 Roscoe, ‘‘ Manners and Customs of the Baganda,” J.A.J., xxxii. p. 30. 3 Mary Kingsley, Travels in West Africa, p. 478. 4 Stokes, Discoveries in Australia, i. p. 60 (1846). PRIMITIVES’ IDEA OF DEATH 341 _ in the idea they had recognized in some of the settlers natives _ long ago departed from life, they actually gave them the names | which these had gone by when alive. This notion is not confined to the Port Lincoln blacks, but prevails also with those of Adelaide and Victoria.” '—‘‘In the south-east portions of Australia, the old men used to say that the forms or spirits of the dead went to the westward, towards the setting sun; and the natives of West Australia had the same belief. When therefore they saw white men coming over the sea from that quarter, they at once took them to be their deceased relatives reincarnated.” 7—‘‘I found the belief among the Yantruwunter natives, that white men were once blacks. I was once asked by some old men how long it was since I was a blackfellow. . . . I was told that I had once been one of the Mungalle family.” 3—‘‘ Buckley, the white man who spent so many years with the wild natives of Port Philip, Victoria, is said to have owed his life to their assuming - that he was one of them who had come to life again. A similar belief was discovered at Port Lincoln, South Australia, in 1846, by Mr. Schiirmann, who says ‘ they certainly believe in the pre-existence of the souls of black men.’ ” 4—" On one occasion, Mr. Bland, in endeavouring to refute their belief, said: ‘Nonsense! I was never here before!’ and was answered, by an intelligent lad, ‘ Then how did you know the way here?’ .... This belief, however, began to die out as they saw that children were born to the white people.” 5 This latter circumstance clearly proves that the natives had formerly held that the whites were the reincarnation of their dead relatives, who had not been through the process of actual birth. This is also the interpretation accorded to it by an excellent observer, Dr. W. E. Roth. He points out that in many of the North Queensland dialects the same word is found to do duty for a European and a deceased aboriginal’s spirit, ghost, etc., and he is satisfied ‘that instead of a return of the t Wilhelmi, ‘‘ Manners and Customs of Australian Natives,”’ etc., Trans Roy Soc., Vict., v. p. 189. a Chauncy, cited by Brough Smyth, The Aborigines of Victoria, ii. p. 22. 3 Howitt, Notes on the Aborigines of Cooper’s Creek, cited by Brough Smyth, The Aborigines of Victoria, ii. p. 307. | 4 Mathews, ‘‘ Aboriginal Tribes of New South Wales and Victoria,” Journal and Proceedings of Royal Society of New South Wales, p. 349 (1905). 5 Chauncy, cited by Brough Smyth, The Aborigines of Victoria, pp. 270, 274. 342 HOW NATIVES THINK deceased native’s actual body after death in the form of a European, the meaning intended to be conveyed was that the vital principle (spirit, etc.) is reincarnated in the white man.”’1 In any case, as a rule, the reincarnation takes place through the medium of pregnancy, and the spirit-child reincarnated already stands in a definite relation to the father and mother who engender it. It virtually formed part of the class or totem of one or other of them. But just as the man who has just . expired is scarcely dead, so the child which has just seen the light of day is only partially born. As we should put it, birth, like death, is accomplished in successive stages. As the © dead is only “ perfectly ’’ dead after the final ceremony ending the mourning period, and by virtue of that ceremony, so the newborn individual will not be perfected until after the final initiation ceremonies, and by virtue of these ceremonies. Perhaps the best way to explain these representations which are so familiar to prelogical mentality is to consider the com- plex system of participations. To the extent that the indivi- dual is engaged in, or apart from, certain of these, he is more completely born or more entirely dead. Most of the customs relating to the dead are intended to rupture their participation in the social group of the living, and to establish their partici- pation in that of deceased members of the same group. So, too, most of the practices relating to the newly-born, to children, and to novices, are designed to effect their increasing participation in the life of the social group to which they belong. The period immediately following the accouchement is mutatis mutandis identical with that which directly follows upon the drawing of the last breath. Like the latter, it is characterized by an extreme sensibility on the part of the subject. Undoubtedly, a newborn child does not inspire the same “‘ mixed ” sentiments as a person who has just died: one has no fear of it, nor does one feel a very lively affection for it. But one imagines it as frail and exposed to many dangers, just like the recently deceased. Practices relating to couvade, which we have already studied,? sufficiently prove the care of * The North Queensland Ethnography Bulletin, No. 5, pz 16 (1903), cited by The Cambridge Expedition to Torres Straits, v. p. 355. ? Vide supra, Chap. VI. pp. 256-260. PRIMITIVES’ IDEA OF DEATH 343 it which is regarded as essential. Its participation in the living social group is still very limited. It has scarcely entered it, just as, at the moment of transition, a man is hardly more dead than alive. Nothing is yet decisive. The friends recall the spirit which has just left the body, they entreat it not to forsake those who love it, they feel it near to them. Similarly, the newborn being uttering its first cry is rather a candidate for life in the social group than a living entity. In this case, too, nothing is yet decisive. If there be any reason, however slight, for not admitting it, they will not hesitate to reject it. It seems that it is in these collective representations that we must seek the main origin of the infanticide which, in various forms, is so common among many primitive peoples. Sometimes it is the children of the female sex who are sacri- ficed, sometimes the opposite. Occasionally both twins are put to death, and again it may be only one; if they are of different sex, it may be in one case, the boy, and in another the girl; in certain aggregates, moreover, twin births are regarded as a happy event. Westermark has collected a vast amount of data on these practices. They are usually explained in the utilitarian sense: the mother who suckles one child cannot properly feed a second. In the Australian aborigines studied by Spencer and Gillen, infanticide is prac- tised everywhere for this reason. But should the mother have given the breast to the newborn child but once only, it is never killed.2 However, this is not the only motive alleged. Among the Abipones, for instance, ‘‘ the mothers suckle their children for three years, during which time they have no carnal inter- course with their husbands who, tired of this long delay, often marry another wife. The women, therefore, kill their unborn babies for fear of repudiation. . . . I have known some who killed all the children they bore, no one either preventing or avenging these murders. The mothers bewail their children, who die of a disease, with sincere tears; yet they dash their newborn babes against the ground, or destroy them in some t Origin and Development of Moral Ideas, i. pp. 394 et seq., 458 et seq. 2 Spencer and Gillen, The Northern Tribes of Central Australia, pp. 608-9. 344 HOW NATIVES THINK other way, with calm countenances.” 1 In other places it is not the desire to retain the husbands, but economic reasons that make the appeal. Hawtrey tells us of the Lengua Indians of Paraguay that it is the woman’s task to bring the foodstuffs from garden and field, and that she does all the carrying. The Lenguas are a nomadic race, often covering ten to twenty miles in a day, and the woman carries all the household belongings, pots, water-jars, skins and blankets in a large string bag on her back. She may have an iron bar, or perhaps some domestic animal or bird to attend to with one hand, and she carries the child upon her shoulders, whilst the man walks ahead, carrying nothing but his bow and arrows. In such case it would be quite impossible for a mother to have more than one child to carry and look after. The importance of such divers motives is undeniable, and the influence they exert may, in the circumstances described, prove irresistible. But on the one hand, we do not find. infanticide always confined to cases in which the mother was already suckling one child, or indeed to those in which she fears that her husband may take another wife. On the other hand, these motives would scarcely suffice if the collective representations did not make infanticide—practised at the very moment of birth: this latter point is of capital import- ance—almost a matter of indifference, because the newborn child has but an infinitely small participation in the life of the social group. Thus the Gallinomero of California ‘‘ do not seem to have limited themselves to killing twins, or having made any distinction of sex, but cut off boy and girl alike, especially if deformed. When resorted to, the act was imme- diate. . . . If the child were allowed to live three days its life was thenceforth secure. They did not call it a ‘ relation ’ until they had decided to spare its life.’’ 3 Moreover, the newborn who is suppressed does not die like the adult. The latter, having accomplished the cycle of participations in the world of the living, enters upon the first stage in the life of the dead, and must traverse it entirely before being born again. But the newborn, who is hardly 1 Dobrizhoffer, An Account of the Abipones, ii. pp. 97-8. ? Hawtrey, “ The Lengua Indians of the Paraguayan Chaco,” J.A.I., XXXI. P. 295. 3 Powers, Tribes of California, p. 177. PRIMITIVES’ IDEA OF DEATH 345 alive, in the sense that he only slightly participates in the life _ of the social group, will remain, if his birth is not completed, at the portals of life, in the terminal period which abuts upon reincarnation. Death to him is scarcely a regression, and he remains an immediate candidate for the next life. Hence there is but slight scruple in disposing of him.# He is not suppressed, but delayed: possibly he will enter, even the following year, the very same mother. Spencer and Gillen expressly state this. “‘It must be remembered that the natives believe that the spirit part of the child returns at once to the Alcheringa home” (hence, without undergoing the ordinary stages) ‘and that it may very soon be born again, entering, very likely, the same woman.’’! This latter belief seems to make the indifference of the mothers in sacrificing their children less strange. The child is only taken from them for a time; they will find it again, and it will return to them. The Khonds of India were accustomed to rid themselves of female children at their birth, and the British had considerable trouble in inducing them to abandon this practice. It pro- ceeded from collective representations, the substance of which has been happily preserved for us. ‘‘ They believe that the reincarnation of a soul in the tribe, when it is first seen to animate a human form is completed only on the performing of the ceremony of naming the infant on the seventh day after its birth ; and they hold the curious doctrine that Boora sets apart a certain quantity of soul to be distributed amongst each generation of men. Thence they believe that if an infant die before it is named, its soul does not enter into the circle of tribal spirits, to be reborn as often as Dinga wills, but rejoins the mass of spirits set apart for the generation to which it belongs.” Here we perceive the reasons which make the Khonds act in this way. ‘‘ Thus by the destruction of a female infant, either the addition of a new female soul to the number of souls attached to a tribe is prevented, and the chance of getting a new male soul in its place is gained, or the return of a female soul by rebirth in that tribe is postponed.”’ 2 But in any case the controlling idea is that the newborn child, since it is but an imperfect being, has not to die as the adult t The Northern Tribes of Central Australia, p. 609. 2 Macpherson, Memorials of Service in India, p. 131. : a 346 HOW NATIVES THINK does. Forit, there is no question of rupturing participations, since they have not yet been established. If the child lives, that is, if for some good reason it is not ‘‘ postponed,’ its welfare depends (by virtue of a mystic participation) upon the actions of its parents, the food they take, their labour and their repose, etc. ; and we already know how stringent are the regulations to which parents must submit. But above all, in order that the child may be released from the period in which his life is still uncertain, (as the death of the newly-dead is still uncertain) it is necessary for him to receive his name, and this in a ceremony which is more or less complicated. In other words, it must be decided who he is. Now it is not a case of choosing a name for him. The child who comes into the world is the reincarnation of a certain ancestor: he therefore has his name already, and this name it is essential to know. Sometimes it may be revealed by an external sign, a mark upon the body. E. M. Gordon tells us that he has found among the Chamars a practice known as Botlagana. Before burying any important member of the family they are accustomed to make some sign upon the corpse, either with ghee or oil or soot, and when a child is born into the family they examine its body to see whether the mark reappears. If they find it, they consider the child to be the reincarnation of that particular ancestor.! More frequently, however, they have recourse to divina- tion. The parents send for a witch-doctor or a medicine-man or a shaman—in short, for some person capable of discovering the mystic participations. On the West Coast of Africa, as Ellis tells us, the dead often return to earth and are born again into the family to which they belonged in a previous existence. A mother will send for a babalowo to tell her which ancestor dwells in the newborn babe, a fact he never fails to reveal, and when this important matter has been settled, he gives advice as to its bringing-up, so that it may conform in every way with its proper character, since the parents are often taken unawares.2—In New Zealand, ‘when the navel-string came off . . . the child was carried to the rt E. M. Gordon, “ People of Mungeli Tahsil, Bilaspur District,” Journal of Asiatic Society of Bengal, iii. pp. 48-9 (1902). 2 A. B. Ellis, The Yoruba-speaking Peoples, pp. 128-9, 152. PRIMITIVES’ IDEA OF DEATH 347 priest. . . . The end of the waka pakoko rakau (idol) was placed in the child’s ear, that the mana . . . of the god might be transferred to him, and the following karakia was repeated. “Wait till I pronounce your name. What is your name ? ’ The priest repeated a long list of names, and when the child sneezed, that which was then being uttered was the one selected.’’ 1—So, too, with the Khonds. ‘‘ Khond births are celebrated on the seventh day after the event by a feast given to the priest and to the whole village. To determine the best name for the child, the priest drops grains of rice into a cup of water, naming with each grain a deceased ancestor. From _ the movement of the seed in the fluid, and other observations made on the person of the infant, he pronounces which of his ancestors has reappeared in him, and the child generally, at least among the Northern tribes, receives the name of that ancestor.” 2 This name is not the only or the most important one which the individual will bear. Among many primitive peoples, at each stage of a man’s life he receives a name which is the sign, the mystic medium of a new participation set up for him: at the time of his initiation, his marriage, his first slaying of an enemy, his acquiring a scalp, or securing certain game, his entry into a secret society or receiving a higher rank in it, and so on. The first name given him, as a rule shortly after his birth, is thus merely a kind of mystic registration; it marks the beginning of definite existence. Henceforth he will have a recognized place in the familial and social group. In it he represents a member who has been in complete partici- pation with it in the past, and he is qualified to participate in it in the same way in the future, when he has undergone the necessary ceremony of initiation. VI During the long period which follows, lasting usually from earliest infancy till puberty, or at least until initiation, the growing children are left almost entirely to their mothers. The men do not worry about their daughters, and take no 1 R. Taylor, Te Ika a Maui, p. 74. 2 Macpherson, op. cit., pp. 72-3. 348 HOW NATIVES THINK trouble with respect to their sons except to teach them, by means of games, that which will later be their actual business : the fabrication and the handling of weapons and implements. These children, moreover, much loved and not a little spoilt, are not yet “ perfect’? members of the social group. They are in the period which corresponds with that elapsing between the first obsequies and the ceremony which terminates the mourning, when the dead is not yet “‘ perfectly ’’ dead, since his body, or at least its fleshy part, has not entirely decayed, and still adheres to his bones. In the same way, whilst the child’s body is growing and developing, he is not definitely “born.” His individuality is not yet complete, and many characteristic features show that this is clearly understood. Among the Ba-Yaka, for instance, ‘even men must observe certain restrictions with regard to the eating of fowls; if the bird be a hen, it may be shared by several, but a cock must be ‘eaten by one man alone, or illness results. He may, however, give some to his son if not yet circumcised. This fact is particularly interesting since it seems to show that a male child before circumcision is not supposed to possess an indivi- duality apart from the father, although it is regarded as belonging to the village of the mother.”’ 1 For the child to reach the state of “ perfect ’’ man, it is not enough to be fully grown, or to have arrived at puberty. His corporal maturity is a necessary, but not an all-sufficing condition. It is not even the essential one. That which matters most, in this as in other cases, governed by the orien- tation peculiar to prelogical mentality, is the mystic elements, the mysterious practices, rites and ceremonies which will enable the young man to participate in the very essence of tribe or totem. If he have not passed through this initiation, whatever his age may be, he will always be ranked among the children. We have many facts in proof of this: here are some, borrowed from a collection published by Webster.? Fison, speaking of the Fijians, noted that an old Wainimala man made no difference between the men who had not been initiated and the children, but spoke of them all as koirana * Torday and Joyce, ‘‘ Notes on the Ethnography of the Ba-Yaka,”’ PP ALT CROC th Ae > Primitive Secret Societies, pp. 25 et seq., pp. 205-6. PRIMITIVES’ IDEA OF DEATH 349 (children).:—An old native of West Kimberley told another observer that until they had undergone subincision (five years after circumcision), boys were like dogs or any other animals.? —Howitt bears witness to a significant fact which occurred during the ceremony known as kadjawalung, at which he was present. At that time in the native camp there were two or three men of the Biduelli tribe with their wives and children, and also a Krauatun Kurnai man with his wife and child. When the ceremony began, they all went away, except one, because those tribes had no initiation ceremonies, and there- fore the visitors had never been “ made men.” The only one who remained was the old Biduelli patriarch, “ and he was now driven crouching among the women and children. The reason was self-evident he had never been ‘ made a man,’ and therefore was no more than a mere boy.” 3—In Savage Island, ‘a child not so initiated (through mata pulega, a ceremony akin to circumcision) is never regarded as a full-born member of the tribe.” 4 This state of minority, which lasts until initiation has taken place, is accompanied by many disabilities and dis- advantages. In Samoa, until tattooed, a boy was in his minority. ‘He could not think of marriage, and he was constantly exposed to taunts and ridicule, as being poor and of low birth, and as having no right to speak in the society of men.’’5 In most Australian tribes, ‘‘ there are various kinds of meat which he must not eat; he cannot enter into any argument in camp; his opinion on any question is never asked, and he never thinks of giving it; he is not expected to engage in fights, and he is not expected to fall in love with any of the young women. In fact, he is a nonentity; but when he has gone through the initiatory process of being made a young man, he takes his proper place among the members of the tribe.’® In South Africa, ‘at puberty the life of an African may be said to begin.” 7 One observer 1 The Nanga, or Sacred stone enclosures of Wainimala, /.4.1., xiv. p. 18. 2 Froggart, Proceedings of the Linnean Society of New South Wales, p. 652 1888). ; 3 The Native Tribes of South-east Australia, P. 530. 4 Thomson, “‘ The Natives of Savage Island,” J.A.J., xxxi. p. 140. 5 Turner, Samoa, p. 88. 6 Brough Smyth, The Aborigines of Victoria, i. p. 83. 7 Macdonald, ‘‘ Manners, Customs . . . of South African Tribes,”’ J.4.I., xix. p. 268. 350 HOW NATIVES THINK has summed all this up in a very striking passage. ‘“‘ Like the dead, the children not arrived at puberty may be compared to seed not yet sown. The child not yet adult finds himself in the same state as this seed, that is, a state of inactivity, of death, but of death with potential life within it.’’ ? As long as initiation has not taken place, marriage is for- bidden. The man who does not yet participate in the mystic essence of the social group is incapable of begetting children able to participate in it one day. Spencer and Gillen tell us : ‘Every man without exception throughout the central area, in all tribes in which the rite is practised, is subincised . . . before he is allowed to take a wife, and infringement of this rule would simply mean death to him if found out.’’2 In East Africa, ‘no man can marry unless he had entered the galo, or if he married, his children would be killed.”’3 As a matter of fact, these children, even as adults, could never be “full” members of the tribe, their father not having been one at the moment of their birth. With certain peoples, however, where the initiation ceremonies can only be celebrated at very long intervals, this principle has had to be abrogated, and married men, fathers of families, have been seen to undergo the tests at the same time as quite young men. “ Bonifaz gave us as an example his uncle, who was already married, but was initiated at the same time as himself, then but eleven — years old ; because it was such a long time since the festival © had been held.” (The case of such individuals is a very special one, and has lasting effects.) ‘‘ A man married under such conditions has no right to enter the house of the spirits, nor to take part in ceremonies from which women and children are excluded. If he has no children as yet, it will be possible to « initiate him when the next public ceremony takes place, but if he already be the father of a family, he will be taken unawares and circumcised, perhaps during a journey. Never- theless, since he has not been ‘made a man’ in a public cere- _ mony, so that the women might know of it, he can never * Passarge, ‘‘ Okawangosumpfland und seine Bewohner,” Zeitschrift für Ethnologie, v. p. 706 (1905). 2 Spencer and Gillen, The Native Tribes of Central Australia, p. 264. 3 Dale, ‘* Customs of the Natives Inhabiting the Bondei Country,” /.4.1, XXV. pp. 188 et seq. PRIMITIVES’ IDEA OF DEATH 351 frequent the house of spirits save by stealth, and without the women and children hearing of it.” ! The initiation ceremonies, therefore, are designed to “perfect ’’ the individual, to render him capable of all the functions pertaining to a legitimate member of the tribe, to complete him as a living being, as the ceremony ending the mourning period renders him “‘ perfectly’”’ dead. It is thus that Spencer and Gillen, who have given a most minute description of these ceremonies, characterize them. “The Engwura . . . is in reality a long series of ceremonies con- cerned with the totems, and terminating in what may be best described as ordeals by fire, which form the last of the initia- tory ceremonies. After the native has passed through these, he becomes what is called Urliara, that is, a perfectly developed member of the tribe... .”’2 I shall not lay stress upon these practices, possibly the best known of all that are habitually met with among primitive peoples: numerous instances of them will be found in Frazer’s Golden Bough,3 or in Webster’s Primitive Secret Societies.4 I shall not enter upon a discussion of the theories which have been proposed as an explanation of them either. I will confine myself to drawing attention to the fact, once again, that the attempt to make such practices ‘intelligible’ is often likely to prove contradictory. If it attains its end it is a failure. In fact, that which is “ intel- ligible ’’ to logical thought is very unlikely to coincide with the idea of the prelogical mind. Without pretending to “ explain ”’ these practices, I have merely tried to show precisely how they, as well as so many other customs in use by primitive peoples, are related to the collective representations of such peoples, and the laws governing these collective representations. Now the general idea of such practices is as follows. An aim which we should consider a positive one, such as the capture of game, the cure of a sick person, is pursued by a variety of means in which those of a mystic character largely predominate. Hunting is not possible unless a mystic partici- pation between hunter and quarry be established, and accord- ingly a whole system of practices designed to secure this is 1 P. W. Schmidt, “ Die geheime RE er der Karesau-Insulaner,” Se EPEOPOS, ii. pp. 1032, 1037-8 (19 The Native Tribes of Central PR #27 3 Vol. ni. pp. 422 et seq. 4 Pp. 21-58, 352 HOW NATIVES THINK inaugurated. [Illness is due to the action of a spirit and no therapeutic methods will have any likelihood of success until the “ doctor ” is in communication with this spirit, to subdue and expel it, by force if necessary. Let us apply this idea to initiation. The novices are separated from the women and children with whom they have lived until this time. As a rule, the separation occurs sud- denly, and often takes them unawares. Confided to the superintendence and care of a particular adult, to whom they often are definitely related, the novices must submit passively to everything imposed upon them and bear pain uncomplain- ingly. The tests they undergo are long and difficult, and often they are actual tortures: deprivation of sleep, of food, being whipped with cords or rods, cudgellings upon the head, pulling out the hair and extracting the teeth, branding, circumcision, subincision, bleeding, the stings of poisonous insects, suffocation by smoke, being suspended by means of hooks fastened in their flesh, ordeal by fire, etc. The secondary motive of these practices may no doubt be to ascertain the novices’ courage and powers of endurance, and to test their virility, by seeing whether they are capable of bearing pain and of keeping a secret. But the primal aim sought after is a mystic effect which in no way depends upon their will-power: the important matter is to establish a participation between them and the mystic realities which are the very essence of the social group, the totems, the mythic or human ancestors, and to give them, by means of this parti- cipation, a ‘‘ new soul,’ as it has been termed.} Herein we perceive difficulties which appear insurmountable to our logical thought, since they raise the question of the unity or of the multiplicity of the soul, whilst the prelogical mind finds no difficulty in imagining that which we call soul as at the same time one and multiple. Just as the North American Indian hunter, by fasting for a week, establishes between himself and the spirit of the bear a mystic bond which will enable him to find and kill bears, so do the tests imposed upon the novices establish between them and the mystic beings in whom they must participate, a relation which is indispensable to the spiritual fusion which is desired. It is not the material aspect of these tests that is of importance; that, in itself, has as PRIMITIVES’ IDEA OF DEATH 353 little to do with the case as has the pain of the patient, with respect to the success of a surgical operation. The means employed by primitives to induce in novices the required condition of receptivity are, as a matter of fact, exceed- ingly painful, but it is not because they are painful that they make use of them, nor would they give them up for that reason, either. Their attention is fixed upon one single point, and it is the only one that matters: this is the condition of special receptivity in which the novices must be placed if the desired participation is to be realized. This condition of receptivity mainly consists in a kind of loss of personality, and of consciousness, induced by fatigue, pain, feebleness, privations,—in short, in an apparent death followed by a new birth. The women and children, (who are excluded from these ceremonies, on pain of the direst penalties) are made to believe that the novices actually die. The novices themselves are induced to believe it, and perhaps the older men share this belief, to a certain extent. ‘The colour of death is white, and the novices are painted white.’’! There are numerous features of this kind and, as Frazer has so clearly shown, the testimony respecting it 1s unanimous. But if we remember what death and birth mean to the pre- logical mind, we shall see that such a mind must imagine thus the condition which permits the participation aimed at in the initiation of the young men. Death is in no case the pure and simple suppression of all the forms of activity and existence which constitute life. The primitive has no idea whatever of such annihilation. That which we call death is never, in his eyes, absolute. The dead live and they die, and even after this second death they continue to exist, while awaiting another reincarnation. That which we call death is accom- plished in successive stages. The first stage, that which the initiation tests strive to imitate, is nothing but a change of dwelling, a transference of the soul which has momentarily abandoned its body, while remaining in the immediate vicinity. It is the beginning of the rupture of participation. It places the individual in a special state of susceptibility and receptivity, akin to the dream-states, catalepsy, or ecstasy 1 Passarge, op. cit., v. p. 706. 23 354 _ HOW NATIVES THINK which, in all primitive peoples, are the invariable condition of communication with the invisible world. Thus by its attachment, in respect to initiation, to prac- tices which produce a kind of death (in the primitive’s sense of the word) primitive mentality has followed its usual course. As ever, it has translated into action and reality its collective representations. VII In most of the primitive peoples we know, there are per- sonages who undergo an additional initiation. These are the wizards, medicine-men, shamans, doctors, or whatever else they may be called. At the period of puberty, they pass through the tests imposed upon all young men; and in order to become capable of fulfilling the important functions they will be called upon to exercise, they have also to undergo a- further novitiate, which lasts for months or even years, and is carried on under the superintendence of their masters, i.e. witch-doctors or shamans in actual practice. Now the resem- blance between the initiation rites of witch-doctors or shamans, and those of the novices of the tribe in general, is a striking one. Still, the general initiation of novices is imposed on all; it is comparatively public, (except as far as women and chil- dren are concerned); and it necessarily takes place at fairly regular intervals. The initiation of witch-doctors, shamans, and medicine-men, on the contrary, is reserved for certain individuals who have a “ vocation’’; it is surrounded by mystery, and only takes place when these persons meet. As to the details of the tests, and the effect obtained by them, (the apparent death and the new birth), however, analogy occasionally becomes identity. ‘‘ When they are being made, the candidates are not allowed to have any rest, but are obliged to stand or walk about until they are thoroughly exhausted and scarcely know what is happening to them. They are not allowed to drink a drop of water or taste food of any kind. They become, in fact, dazed and stupefied.” 1 When this condition has reached its height, one may say that they are dead. In other words, the spirits (7untarinia) who t Spencer and Gillen, The Northern Tribes of Central Australia, p. 485. | | wf PRIMITIVES’ IDEA OF DEATH 355 | preside at the initiation, kill them and then cause them to be born again. ‘“ At break of day, one of the iruntarinia comes to the mouth of the cave, and finding the man asleep, throws at him an invisible lance which pierces the neck from behind, | passes through the tongue, making therein a large hole, and | then comes out through the mouth. ...A second lance _ thrown by the iruntarinia pierces the head from ear to ear, and the victim falls dead and is at once carried into the depths _ of the cave ’’—(where the spirits dwell). | ‘Within the cave, the truntarinia removes all the internal organs and provides the man with a completely new set, ! after which operation has been successfully performed, he _ presently comes to life again ; but in a condition of insanity.? . For several days the man remains more or less strange in his appearance and behaviour until one morning it is noticed that he has painted with powdered-charcoal and fat a broad band across the bridge of his nose. All signs of insanity _ have disappeared, and it is at once recognized that a new medicine-man has graduated.” 3 It is the same in South America. ‘The pajé (witch- _ doctor) becomes such of his own free will. From his youth up he must prepare himself for his sinister rôle. He must with- draw to some solitary and inaccessible spot, fast, maintain silence, and practise all sorts of deprivations for many years he must dance wild and obscene dances until he falls exhausted, and he must even, like the young men undergoing the initiatory tests, expose himself to the sting of the large poisonous ants.’’ 4 The same rites, resulting in the same apparent death, are often the indispensable condition of initiation, not to the function of shaman or witch-doctor, but merely to member- ship of a secret society the candidate desires to enter. For instance, with the Abipones, when a man desires to be raised to the status of hocheri, ‘they make a previous trial of his fortitude . . . to sit down at home for three days, and 1 A ceremony identical with that undergone by the tribal novices during their apparent death. 2 Again like the novices. 3 The Native Tribes of Central Australia, pp. 524-5; The Northern Tribes of Central Australia, pp. 480-4. 4 Von Martius, Beitvage zur Ethnographie Stid-Amerika’s, i. p. 558. 356 HOW NATIVES THINK during that time to abstain from speaking, eating and drink- ing. . . . On the evening preceding the military function, all the women flock to the threshold of his tent. Pulling off their clothes from the shoulder to the middle, and dishevelling their hair, they stand in a long row (signs of mourning) and lament for the ancestors of him who is, next day, to be adorned with a military dignity. ... The next day the initiated man directs his course towards all the points of the compass in turn, and then there is a ceremony in which an old woman shaves off his hair, and he receives a new name.’’! He has clearly passed through death and a new birth.—Among the Clallams, a coast tribe on the mainland opposite the south end of Van- couver, ‘the novice of the order’”’ (a secret society) ‘‘ must for three days and three nights fast alone in a mysterious lodge prepared for him, round which, during all that time, the brethren already initiated sing and dance. This period elapsed, during which it would seem that the old nature has been killed out of him, he is taken up as one dead, and soused with the nearest cold water, where he is washed till he revives, which thing they call ‘washing the dead.” When his senses are sufficiently gathered to him, he is set on his feet ; upon which he runs off into the forest, whence he soon reappears a perfect medicine-man, rattle in hand and decked out with the various trappings of his profession.” 2 Lastly, in the Lower Congo territory, we have the Nkimba, “‘ an institution which has a wide range among the Lower Congo tribes,” (in which) “the initiatory rites are in charge of the mganga, or fetish man, who lives with his assistants in an enclosure near each village. The candidate for this order, having previously imbibed a sleeping potion, swoons in some public assembly and is at once surrounded by the nganga and his assistants, who take him to the enclosure. It is given out that he is dead, and has gone to the spirit world, whence, by the power of the great nganga, he will subsequently be restored to life. The novice remains with the nganga for a prolonged period, some- times for several years, learning a new language, probably an archaic Bantu, and receiving instructions in the mysteries of | the order. No woman is allowed to look on the face of one of t Dobrizhoffer, op. cit., ii. pp. 441-5. a Bancroft, The Native Races of the Pacific Coast of North America, iii. p. 155. PRIMITIVES’ IDEA OF DEATH 357 the Nkimba, who daily parade through the woods, or the surrounding country, singing a strange, weird song to warn the uninitiated from their approach. When brought back to the village, and introduced by his new name, he affects to treat everything with surprise as one come to a new life from another world ; to recognize no one, not even his father or mother, while his relatives receive him as raised from the dead ; and for several days the newcomer is permitted to take anything he fancies in the village, and is treated with every kindness until it is supposed that he has become accus- tomed to his surroundings. . . . He then decides whether he will become a fetish-man or return to his ordinary life.” 1 We might quote many similar instances, but these will doubtless suffice to demonstrate that the initiatory rites for witch-doctors, shamans, medicine-men, fetish-men, etc., or members of some secret society, in their general procedure as, frequently, in their most insignificant details, reproduce the public ceremonies of initiation imposed upon the youths of the tribe when they reach puberty. Now there can be no manner of doubt as to the end sought by the former : they are designed to enable the candidates to participate in mystic realities, to put them in communication, or rather, in communion, with certain spirits. Does not the power of witch-doctor or shaman proceed from the privileges he possesses of entering into relation, when he pleases, by means of which he holds the secret, with occult forces known only to the ordinary man by the effect they produce? There is therefore no doubt, either, as to the end aimed at by the practices which constitute the usual initiatory rites for the youths of the community. They are magic operations designed to place them in a condition of ecstasy, unconsciousness, ‘‘ death,” indispensable if they are to participate in the essentially mystic reality of their tribe, their totem and their ancestors. Once this participation has been effected, the novices are ‘‘ full’’ members of the tribe, for to them its secrets have been revealed. From this moment, these full members, these complete men, are the depositaries of all that the social group holds most sacred, and the sense of their responsibility never leaves them. A man’s life is, as it : Glave, Six Years of Adventure in Congoland, p. 80, quoted in Webster's Primitive Secret Societies, pp. 173-4. 358 HOW NATIVES THINK were, “ sharply marked out into two parts . . . first, ordinary life, common to all the men and women... second, what gradually becomes of greater and greater importance to him, the portion of his life devoted to matters of a secret or sacred nature. As he grows older he takes an increasing share in these, until finally this side of his life occupies by far the greater part of his thoughts.” ! At last, death supervenes, and the cycle, of which I have essayed to portray the princi- pal stages, begins once more. * Spencer and Gillen, The Northern Tribes of Central Australia, Pp. 33 RAR Ga DV: an Pers LA 5 AA hag ty : CHAPTER IX THE TRANSITION TO THE HIGHER MENTAL RY PES INTRODUCTION ANALYSIS of the facts studied in the preceding chapters seems to bear out the essential theses which this books aims at establishing : (x) The institutions, customs and beliefs of primitives imply a mentality which is prelogical and mystic, oriented differently from our own. (2) The collective representations and interconnections which constitute such a mentality are governed by the law of participation and in so far they take but little account of the logical law of contradiction. It is the natural consequence of such a position that I have been endeavouring to demonstrate, It is useless to try and explain the institutions and customs and beliefs of undeveloped peoples by starting from the psychological and intellectual analysis of “the human mind” as we know it. No inter- pretation will be satisfactory unless it has for its starting-point the prelogical and mystic mentality underlying the various forms of activity in primitives. | But it is not only the study of inferior races that com- prehension of this prelogical, mystic mentality helps. Subse- quent mental types derive from it, and cannot avoid repro- ducing, in forms more or less apparent, some of its features. To understand these, therefore, it is necessary to refer back to a type which is comparatively ‘‘ primitive.” A vast field for positive research into the mental functioning of aggre- gates of various kinds, as well as into our own laws of thought, 362 HOW NATIVES THINK is thus laid open to us. In conclusion, I should like to show, by referring to certain important points, that this research may even now prove fertile of result, if we accept as a working hypothesis the idea of prelogical mentality as defined in this book. I In aggregates of the type furthest removed from our own, the collective representations which express the mentality of the group are not always, strictly speaking, representations. What we are accustomed to understand by representation, even direct and intuitive, implies duality in unity. The object is presented to the subject as in a certain sense distinct from himself ; except in states such as ecstasy, that is, border states in which representation, properly so called, disappears, since the fusion between subject and object has become com- plete. Now in analysing the most characteristic of the primi- tive’s institutions—such as totemic relationship, the intichiuma and initiation ceremonies, etc.—we have found that his mind does more than present his object to him: it possesses it and is possessed by it. It communes with it and participates in it, not only in the ideological, but also in the physical and mystic sense of the word. The mind does not imagine it merely ; it livesit. In a great many cases the rites and ceremonies have the effect of giving reality to a veritable symbiosis, that between the totemic group and its totem, for instance. At this stage, therefore, rather than speak of collective representations, it would be wiser to call them collective mental states of extreme emotional intensity, in which representation is as yet undifferentiated from the move- ments and actions which make the communion towards which ~ it tends a reality to the group. Their participation in it is so effectually dived that it is not yet properly imagined. We shall not be astonished, therefore, that Spencer and Gillen should have discovered in the Australian aborigines they studied ‘ not the slightest trace of anything that might be described as ancestor-worship”’ . . .ï very few traditions about the origin of animals; few myths; and no objects of worship, properly so called; no personification of natural 1 The Northern Tribes of Central Australia, P. 494. TRANSITION TO THE HIGHER MENTAL TYPES 363 forces or of animal or vegetable species.t Similar paucity in this respect has been noted by Ehrenreich 2 in the most primitive races of South America—peoples who are unfortunately far less known to us than those of Australia. This fact demon- strates that the prelogical and mystic collective mentality is still actively predominant in the social group. The feeling of symbiosis effected between the individuals of the group, or between a certain human group and one which is animal or vegetable in substance, is directly expressed by institutions and ceremonies. At the moment it needs no symbols other than those used in the ceremonies. Such are the churinga, the decorations and ornaments with which the actors in the ceremonies adorn themselves, the dances, masks, gestures and traditions relating to the ancestors of the Alcheringa, among Australian aborigines ; or again, among the Indians of Brazil (the Bororo, Bakairi, and others), the entire group of customs known as couvade, in which a participation, both mystic and physical, between parents and child is so evidently felt and realized. This form of mental activity, which differs so notably from the forms which our own aggregates afford us the opportunity of studying, is not yet seeking to understand or explain its object. It is oriented in quite another direction; it cannot be dissociated from the mystic practices which give effect to its participations. The ubiquity or multipresence of exist- ing beings, the identity of one with many, of the same and of another, of the individual and the species—in short, every- thing that would scandalize and reduce to despair thought which is subject to the law of contradiction, is implicitly admitted by this prelogical mentality. It is, moreover, im- permeable to what we call experience, i.e. to the lessons which may be learnt by observation of the objective relations between phenomena. It has its own experience, one which is wholly mystic, much more complete and exhaustive and decisive than the ofttimes ambiguous experience, the censorship of which thought, properly so called, knows that she must accept. It is entirely satisfied with this. In this respect there is nothing more significant than the The Northern Tribes of Central Australia, p. 442. Die Mythen und Legenden der Süd-Amerthanischen Urvôlker, pp. 12, 15. 364 HOW NATIVES THINK primitive classifications I have already cited, to which Durk- heim and Mauss have drawn attention, for in the primitive mentality these, to a certain extent, occupy the position held by the categories in logical thought. The participations felt by the members of the social body and expressed in their divisions and groupings, are extended to all the entities which such a mentality imagines. Animal and vegetable orders, the heavenly bodies, inorganic matter, directions in space, are all fitted into some division or other of the social frame- work. To give but one example only: “ In this tribe,” says Howitt, ‘‘ the two main classes and the four sub-classes divide, so to speak, the whole universe into groups. The two main classes are Mallera or Wuthera ; consequently all other objects are Mallera or Wuthera. This custom is carried to such an extreme that a medicine-man who is a Mallera, for instance, in his magical operations, can only use things which belong to his own class. Moreover, when he dies, the bier on which his body rests must perforce be made of the wood of a tree of the Mallera class.” 2 À certain community of being is thus immediately felt, not only between members of the same totemic family, but between all entities of any kind whatsoever which form part of the same class and are linked together in mystic fellowship. And this feeling, which environs a representation still undiffen- tiated, is necessarily accompanied by the feeling (and the undifferentiated representation) of a non-participation with beings and objects belonging to other classes. To a mentality of this kind, the feeling of not being bound by any mystic relation to another being in the vicinity, is not merely a nega- ‘tive sentiment: in certain cases it may be a very definite and positive feeling. We may reconstruct it, to a certain extent, by conjuring up what is now called racial antagonism, and the sentiments which that which is “ foreign’? may arouse, even among civilized people. From the standpoint of action, therefore, there arises a need to resort to certain individuals or to the members of a certain group, who alone possess the mystic qualifications enabling them to carry out a ceremony or execute a dance or a rite, or merely to be 1 Vide Chap. III. p. 128. * Howitt, “ Notes on Australian Message Sticks,” J.4.I., xviii. p. 326. TRANSITION TO THE HIGHER MENTAL TYPES 365 present at it. The result to be obtained depends above all upon the mystic participations between the classes of beings and of objects. II In his recent work on Animism in the Indian Archipelago, Kruijt believes it necessary to distinguish two successive stages in the evolution of primitive communities: one in which individual spirits are reputed to inform and inspire every being and every object (animals, plants, boulders, stars, weapons, tools, and so forth), and another and earlier one, in which individualization has not as yet taken place, in which there is a diffused principle capable of penetrating everywhere, a kind of universal and widespread force which seems to animate persons and things, to act in them and endow them with life.t Here we recognize Marett’s “‘ pre-animistic”’ stage, upon which Durkheim and Mauss have also insisted. Kruijt adds—and this remark of his has a very important bearing upon the subject with which we are concerned—that the differentiation of these two periods corresponds with a differ- ence in the mentality of the social group. At the time when souls and spirits are not yet individualized, the individual consciousness of every member of the group is and remains strictly solidary with the collective consciousness. It does not distinctly break away from it ; it does not even contradict itself in uniting with it; that which does dominate it is the uninterrupted feeling of participation. Only later, when the human individual becomes clearly conscious of himself as an individual, when he explicitly differentiates himself from the group of which he feels himself a member, do beings and objects outside himself also begin to appear to him as provided with individual minds or spirits during this life and after death. Thus, when the relations between the social group and the individuals composing it are evolved, the collective repre- sentations, the group ideas, are modified at the same time. In its purest form primitive mentality implied a participation which was felt and lived, both by individuals with the social group, and by the social group with the surrounding ones. t Kruijt, Het Animisme in den Indischen Archipel, pp. 66-7 (1906). 2 Ibid., pp. 2-5. 366 | HOW NATIVES THINK Both these participations are solidary, and the modifications of the one reflect accordingly upon the other. In proportion as the individual consciousness of each member of the group tends to declare itself, the feeling of a mystic symbiosis of the social group with surrounding groups of beings and objects becomes less intimate and direct and less constant. Here as there bonds which are more or less explicit tend to take the place of the feeling of direct communion. In a word, parti- cipation tends to become ideological. For instance, as soon as individual consciousness begins to grasp itself as such and consequently to distinguish individuals as such in the sur- rounding groups of beings, these ideas also define, more or less distinctly, that of the groups as such, and as a further consequence, an idea of the mystic relations uniting the indi- viduals of a group, and the different groups in their turn. The communion which is no longer actually lived, the need for which still appears just as pressing, will be obtained by means of. intermediaries. The Bororo tribe will no longer declare that they ave araras. They will say that their ancestors were araras, that they are of the same substance as the araras, that they will become araras after death, that it is forbidden to kill and eat araras, except under conditions which are rigidly defined, such as totemic sacrifice, etc. Following on the paucity I recently noted in the Aruntas, the Bororo and other aggregates of a very primitive type, we shall find, in those more advanced, such as the Huichols, the Zuñis of New Mexico, the Maoris of New Zealand, an in- creasing wealth of collective representations properly so called, and of symbols. In the former the feeling of mystic symbiosis is still intense and permanent. To express itself it need but resort to the very organization of the social group and the ceremonies which assure its prosperity and the rela- tions with surrounding groups. In the latter, the need of participation is perhaps no less active. But as this participa- tion is no longer directly felt by every member of the social group, it is obtained by means of an ever-increasing display of religious or magic practices, of sacred and divine beings and objects, by rites performed by priests and members of secret societies, by myths, etc. F. H. Cushing’s admirable work on the Zuñis, for instance, shows us how a prelogical and mystic TRANSITION TO THE HIGHER MENTAL TYPES 367 mentality of an already exalted type expands into a magni- ficent efflorescence of collective representations destined to express, or even to produce, pateicipaons which are no pes directly felt. The “vehicles ” of these participations are very diverse in their nature. In many communities we find, concurrent with representations similar to those of the Melanesian mana,' col- lective representations of more or less individualized spirits, of souls more or less distinctly conceived, mythical beings with an animal or human or semi-human form, heroes, genii, gods. The observers find names for them easily enough. But the difficulty is not to allow ourselves to be deceived by these names, but to reconstruct under them the mystic and pre- logical collective representations which no longer exist for us. This difficulty never appears so great as when it is a question of defining the “‘religion’’ of the most primitive peoples we are acquainted with. For we might say equally well that the mentality which expresses itself in their collective representa- tions is wholly religious, or, in another sense of the word, that it is hardly at all so. In so far as a mystic communion with, and actual participation in, the object of the religious senti- ment and ritual practice is of the very essence of religion, primitive mentality must be declared religious because it does realize a communion of such a nature, and indeed to the highest degree it is possible to imagine. But in other respects it does not seem correct to speak of it as “‘ religious,’’ at least to this extent, that by reason of the direct character of this parti- cipation it does not recognize as an ideal outside and above itself the beings with whom it feels itself united in mystic and intimate communion. We may recall the definite statements made by Spencer and Gillen on this point, As a matter of fact, the primitives’ “ religious ’’ ideas are a constant source of error and confusion to us. Our own way of thinking makes us imagine the objects of their thought in the attitude of divine beings or objects, and that it is by virtue of this divine character of theirs that homage, sacrifice, prayer, adoration and all actual religious belief is directed towards them. But to the primitive mind, on the contrary, these 1 Cf. Hubert and Mauss, “ Mélanges d’Histoire des Religions,” Année Soctologique, pp. xx et seq. 368 HOW NATIVES THINK objects and these beings become divine only when the parti- cipation they guarantee has ceased to be direct. The Arunta who feels that he is both himself and the ancestor whose churinga was entrusted to him at the time of his initiation, knows nothing of ancestor-worship. The Bororo does not make the parrots, which are Bororo, the objects of a religious cult. It is only in aggregates of a more advanced type that we find an ancestor-worship, a cult of heroes, gods, sacred animals, etc. The ideas which we call really religious are thus a kind of differentiated product resulting from a prior form of mental activity. The participation or communion first realized by mystic symbiosis and by the practices which affirmed it is obtained later by union with the object of the worship and belief called religious, with the ancestor, the god. The personality of these objects comprises, as we know, an infinite variety of grades, from mystic forces of which we cannot say whether they are single or manifold, to divinities clearly defined by physical and moral attributes, such as those of the Melanesian or the Greek deities. It depends above all on the degree of development attained by the group studied, i.e., upon the type of its institutions as well as its mental type. Ii] When we consider myths in their relation to the mentality of the social groups in which they originate, we are led to similar conclusions. Where the participation of the indi- vidual in the social group is still directly felt, where the parti- cipation of the group with surrounding groups is actually lived—that is, as long as the period of mystic symbiosis lasts— myths are meagre in number and of poor quality. This is the case with the Australian aborigines and the Indians of Northern and Central Brazil, etc. Where the aggregates are of a more advanced type, as, for instance, the Zuñis, Iroquois, Melanesians, and others, there is, on the contrary, an in- creasingly luxuriant outgrowth of mythology. Can myths then likewise be the products of primitive mentality which appear when this mentality is endeavouring to realize a parti- cipation no longer directly felt—when it has recourse to inter- mediaries, and vehicles designed to secure a communion which ~ TRANSITION TO THE HIGHER MENTAL TYPES 369 has ceased to be a living reality ? Such a hypothesis may seem to be a bold one, but we view myths with other eyes than those of the human beings whose mentality they reflect. We see in them that which they do not perceive, and that which they imagine there we no longer realize. For example, when we read a Maori or Zufii or any other myth, we read it trans- lated into our own language, and this very translation is a betrayal. To say nothing of the construction of the sentences, which is bound to be affected by our customary habits of thought, if only in the very order of the words, to primitives the words themselves have an atmosphere which is wholly mystic, whilst in our minds they chiefly evoke associations having their origin in experience. We speak, as we think, by means of concepts. Words, especially those expressive of group-ideas, portrayed in myths, are to the primitive mystic realities, each of which determines a champ de force. From the emotional point of view, the mere listening to the myth is to them something quite different from what it is to us. What they hear in it awakens a whole gamut of harmonics which do not exist for us. Moreover, in a myth of which we take note, that which mainly interests us, that which we seek to understand and interpret, is the actual tenor of the recital, the linking-up of facts, the occurrence of episodes, the thread of the story, the adventures of the hero or mythical animal, and so forth. Hence the theories, momentarily regarded as classic, which see in myths a symbolic presentment of certain natural pheno- _ mena, or else the result of a “ disease of language ”” ; hence the classifications (like that of Andrew Lang, for instance) which arrange myths in categories according to their content. But this is overlooking the fact that the prelogical, mystic mentality is oriented differently from our own. It is un- _ doubtedly not indifferent to the doings and adventures and vicissitudes related in myths; it is even certain that these interest and intrigue the primitive’s mind. But it is not the positive content of the myth that primarily appeals to him. He does not consider it as a thing apart ; he undoubtedly sees it no more than we see the bony framework beneath the flesh of a living animal, although we know very well that it is there. 1 Encylopedia Britannica. Mythology (9th ed.), xvii. pp. 156-7. 24 370 HOW NATIVES THINK That which appeals to him, arouses his attention and evokes his emotion, is the mystic element which surrounds the positive content of the story. This element alone gives myth and legend their value and social importance and, I might almost add, their power. It is not easy to make such a trait felt nowadays, precisely because these mystic elements have disappeared as far as we are concerned, and what we call a myth is but the inanimate corpse which remains after the vital spark has fled. Yet if the perception of beings and objects in nature is wholly mystic to the mind of the primitive, would not the presentation of these same beings and objects in myths be so likewise? Is not the orientation in both cases necessarily the same? To make use of a comparison, though but an imperfect one, let us hark back to the time when in Europe, some centuries ago, the only history taught was sacred history. Whence came the supreme value and importance of that history, both to those who taught and those who learnt ? Did it le in the actual facts, in the knowledge of the sequence of judges, kings © or prophets, of the misfortunes of the Israelites during their strife with the neighbouring tribes? Most certainly not. It is not from the historical, but from the sacred, point of view that the Biblical narrative was of incomparable interest. It is because the true God, perpetually intervening in the story, : makes His presence manifest at all times and, to the Christian — idea, causes the coming of His Son to be anticipated. In short, it is the mystic atmosphere which surrounds the facts and . prevents them from being ordinary battles, massacres or revolutions. Finally it is because Christendom finds in it a witness, itself divine, of its communion with its God. Myths are, in due proportion, the Biblical narrative of primitive peoples. The preponderance of mystic elements, however, in the group ideas of myths, is even greater than in our sacred history. At the same time, since the law of parti- cipation still predominates in the primitive mind, the myth is accompanied by a very intense feeling of communion with the mystic reality it interprets. When the adventures, exploits, noble deeds, death and resurrection of a beneficent and civilizing hero are recounted in a myth, for instance, it is not the fact of his having given his tribe the idea of making . TRANSITION TO THE HIGHER MENTAL TYPES 371 a fire or of cultivating mealies that of itself interests and espe- cially appeals to the listeners. It is here as in the Biblical narrative, the participation of the social group in its own past, it is the feeling that the group is, as it were, actually living in that epoch, that there is a kind of mystic communion with that which has made it what it is. In short, to the mind of the primitive, myths are both an expression of the solidarity of the social group with itself in its own epoch and in the past and with the groups of beings surrounding it, and a means of maintaining and reviving this feeling of solidarity. Such considerations, it may be urged, might apply to myths in which the human or semi-human ancestors of the social group, its civilizing or its protecting heroes, figure ; but are they valid in the case of myths relating to sun, moon, stars, thunder, the sea, the rivers, winds, cardinal points, etc.? It is only to an intellect such as ours that the objection appears a serious one. The primitive’s mind works along the lines that are peculiar to it. The mystic elements in his ideas matter considerably more to him than the objective features which, in our view, determine and classify beings of all kinds, and as a consequence the classifications which we regard as most clearly evident escape his attention. Others, which to us are inconceivable, however, claim it. Thus the relationship and communion of the social group with a certain animal or vegetable species, with natural phenomena like the wind or the rain, with a constellation, appear quite as simple to him as his communion with an ancestor or a legendary hero. To give but one instance, the aborigines studied by Spencer and Gillen regard the sun as a Panunga woman, belonging to a definite sub-class, and consequently bound by the ties of relationship to all the other clans of the tribe. Let us refer again to the analogy indicated above. In he sacred history of primitives natural history forms a part. If this view of the chief significance of myths and of their characteristic function in aggregates of a certain mental type be correct, several consequences of some importance will ensue. This view does not render the careful and detailed study of myths superfluous. It provides neither a theory for classifying them in genera and species, nor an exact method of interpreting them, nor does it throw positive light upon 372 HOW NATIVES THINK their relations with religious observances. But it does enable us to avoid certain definite errors, and at any rate it permits of our stating the problem in terms which do not falsify the solution beforehand. It provides a general method of pro- cedure, and this is to mistrust “‘ explanatory ’’ hypotheses which would account for the genesis of myths by a psychological and intellectual activity similar to our own, even while assuming it to be childish and unreflecting. The myths which have long been considered the easiest to explain, for instance, those regarded as absolutely lucid, such as the Indian nature-myths, are on the contrary the most intriguing. As long as one could see in them the spontaneous product of a naive imagination impressed by the great natural phenomena, the interpretation of them was in fact self-evident. But if we have once granted that the mentality which generates myths is differently oriented from ours, and that its col- lective representations obey their own laws, the chief of which is the law of participation, the very intelligibility of these myths propounds a fresh problem. We are led to believe that, far from being primitive, these myths, in the form in which they have reached us, are something absolutely arti- ficial, that they have been very highly and consciously elabo- rated, and this to such an extent that their original form is almost entirely lost. On the other hand, the myths which may possibly seem the easiest to explain are those which most directly express the sense of the social group’s relation- ship, whether it be with its legendary members and those no longer living, or with the groups of beings which surround it. For such myths appear to be the most primitive in the sense that they are most readily allied with the peculiar prelogical, mystic mentality of the least civilized aggregates. Such, among others, are the totemic myths. If, however, the aggregates belong to a type even slightly more advanced, the interpretation of their myths very soon becomes risky and perhaps impossible. In the first place, their increasing complexity diminishes our chances of correctly following up the successive operations of the mentality which produces these myths. This mentality not only refuses to be bound by the law of contradiction—a feature which most myths reveal at first sight, so to speak—but it neither abstracts TRANSITION TO THE HIGHER MENTAL TYPES 373 nor associates, and accordingly it does not symbolize as our thought does. Our most ingenious conjectures, therefore, always risk going astray. If Cushing had not obtained the interpretation of their myths from the Zuñis themselves, would any modern intellect have ever succeeded in finding a clue to this prehistoric laby- rinth? The true exposition of myths which are somewhat complicated involves a reconstruction of the mentality which has produced them. This is a result which our habits of thought would scarcely allow us to hope for, unless, like Cushing, a savant were exceptionally capable of creating a ‘primitive ’’ mentality for himself, and of faithfully trans- cribing the confidences of his adopted compatriots. Moreover, even in the most favourable conditions, the state in which the myths are when we collect them may suffice to render them unintelligible and make any coherent inter- pretation impossible. Very frequently we have no means of knowing how far back they date. If they are not a recent product, who is our authority for assuming that some frag- ments at any rate have not disappeared, or, on the other hand, may not myths which were originally quite distinct, have been mingled in one incongruous whole? The mystic elements which were the predominant feature at the time when the myth originated may have lost some of their importance if the mentality of the social group has evolved at the same time as their institutions and their relations with neighbouring groups. May not the myth which has gradually come to be a mystery to this altered mentality have been mutilated, added to, transformed, to bring it into line with the new collective representations which dominate the group? May not this adaptation have been performed in a contrary sense, without regard to the participations which the myth originally ex- pressed ? Let us assume—an assumption by no means un- reasonable—that it has undergone several successive trans- formations of this kind: by what analysis can we hope ever to retrace the evolution which has been accomplished, to find once more the elements which have disappeared, to correct the misconceptions grafted upon one another? The same problem occurs with respect to rites and customs which are often perpetuated throughout the ages, even while they 374 HOW NATIVES THINK are being distorted, completed in a contrary sense, or acquiring a new significance to replace that which is no longer understood. IV When the participations which matter most to the social . group are secured by means of intermediaries or ‘‘ vehicles,’ instead of being felt and realized in more direct fashion, the change reacts upon the mentality of the group itself. If, for instance, a certain family or a certain person, a chief, a medicine- man in any tribe is represented as “‘ presiding ” over the — sequence of the seasons, the regularity of the rainfall, the conservation of species which are advantageous—in short, — the periodic recurrence of the phenomena upon which the 1 existence of the tribe depends—the group-idea will be pecu- — liarly mystic, and it will preserve the characteristic features proper to prelogical mentality to a very high degree. Partici- pation, concentrated, as it were, upon the beings who are its media, its chosen vessels, thus itself becomes ideological. By force of contrast, other families, other individuals of the social group, the neighbouring groups not interested in this participation, are represented in a more indifferent and im- partial way, a fashion less mystic and therefore more objective. This means that a more and more definite and permanent distinction tends to be established between sacred beings and objects on the one hand, and profane beings and objects on the other. The former, inasmuch as they are the necessary vehicles of participation, are essentially and eternally sacred. The latter only become so intermittently by virtue of their communion with the former, and in the intervening periods — they present no more than faint, derivative mystic features. This leads to two connected consequences. In the first place, since the beings and the objects among which the social _ group lives are no longer felt to be in direct communion with ) it, the original classifications by which this communion was 4 expressed tend to become obliterated, and there is a redistri- bution of less mystic nature, founded upon something other … than the ramifications of the social group. Ideas of animal and plant life, the stars, etc., are doubtless still impregnated . with mystic elements, but not all of them to the same extent. TRANSITION TO THE HIGHER MENTAL TYPES 375 Some of them are markedly so, others to a far lesser degree, and this difference brings about fresh classifications. The beings and the objects represented as ‘‘ containers ’’ of mystic virtue, the vehicles of participation, are inevitably differentiated from those which do not possess this supreme interest for the social group. The latter are beginning to be ranged according to an interest of another order; their distinguishing features are less mystic, but more objective. In other words, the collective representations of these beings and objects is begin- ning to tend towards that which we call “concept.” It is still remote from this, but the process which is to bring it nearer has already begun. $ Moreover, the perception of these entities at the same time loses some of its mystic character. The attributes we term objective, by which we define and classify entities of all kinds, are to the primitive enveloped in a complex of other elements much more important, elements exacting almost exclusive attention, at any rate to the extent allowed by the necessities of life. But if this complex becomes simpler and the mystic elements lose their predominance, the objective attributes ipso facto readily attract and retain the attention. The part played by perception proper is increased to the extent in which that of the mystic collective representations diminishes. Such a modification is favourable to the change of classifica- tion of which we have spoken, and in its turn this change reacts upon the method of perceiving, as an inducted stream reacts upon the main current. Thus, as by degrees the participations are less directly felt, the collective representations more nearly approach that which we properly call “‘idea’’—that is, the intellectual, cognitive factor occupies more and more space in it. It tends to free itself from the affective and motor elements in which it was at first enveloped, and thus arrives at differen- tiating itself. Primitive mentality, as a consequence, is again modified in another respect. In aggregates in which it is least impaired, in which its predominance is at its maximum, we have found it impervious to experience. The potency of the collective representations and their interconnections is such that the most direct evidence of the senses cannot counter- act it, whilst the interdependence of the most extraordinary fay NAD Cpe 376 HOW NATIVES THINK kind between phenomena is a matter of unwavering faith. But when perception becomes less mystic, and the precon- nections no longer impose the same sovereign authority, sur- rounding nature is seen with less prejudiced eyes and the collective representations which are evolving begin to feel the effect of experience. Not all at the same time, nor to the same extent: on the contrary, it is certain that these are unequally modified, in accordance with a good many diverse circumstances, and especially with the degree of interest felt by the social group in the object. It is on the points in which participation has become weakest that the mystic preconnections most quickly yield, and the objective relations first rise to the surface. At the time when the mentality of primitive peoples grows more accessible to experience, it becomes, too, more alive to the law of contradiction. Formerly this was almost entirely a matter of indifference, and the primitive’s mind, oriented — according to the law of participation, perceived no difficulty at all in statements which to us are absolutely contradictory. A person is himself and at the same time another being; he is in one place and he is also somewhere else : he is individual as well as collective (as when the individual identifies himself with his group), and so on. The prelogical mind found such statements quite satisfactory, because it did more than perceive and understand them to be true. By virtue of that which I have called a mystic symbiosis, it felt, and lived, the truth of them. When, however, the intensity of this feeling in the collective representations diminishes, the logical diffi- culty in its turn begins to make its presence felt: Then by degrees the intermediaries, the vehicles of participation, appear. They render it representable by the most varied methods—transmission, contact, transference of mystic qual- ities—they secure that communion of substance and of life which was formerly sensed in a direct way, but which runs the risk of appearing unintelligible as soon as it is no longer lived. Properly speaking, the absurdities to which the primitive mind remains insensible are of two kinds, undoubtedly closely connected with each other, but yet appearing very different to our way of thinking. Some, like those we have just in- stanced, arise out of what seems to us an infringement of the — TRANSITION TO THE HIGHER MENTAL TYPES 377 logical law of contradiction. These manifest themselves gradually, as the participations formerly felt are “‘ precipi- tated ’’ in the form of definite statements. Whilst the feeling of participation remains a lively one, language conceals these absurdities, but it betrays them when the feeling loses some of its intensity. Others have their source in the preconnec- tions which the collective representations establish between persons, things, occurrences. But these preconnections are only absurd through their incompatibility with the definitely fixed terms for these persons, things, occurrences—terms which the prelogical mind has not at its command in the beginning. It is only when such a mind has grown more cognizant of the lessons taught by experience, when the attributes we term ‘ objective ” get the better of the mystic elements in the col- lective representations, that an interdependent relation between occurrences or entities can be rejected as impossible or absurd. In the earlier stage the dictum deduced from Hume's argument, that ‘anything may produce anything,’ might have served as a motto for primitive mentality. There is no metamorphosis, no generating cause, no remote influence too strange or inconceivable for such a mentality to accept. A human being may be born of a boulder, stones may speak, fire possess no power to burn, and the dead may be alive. We should refuse to believe that a woman may be delivered of a snake or a crocodile, for the idea would be irreconcilable with the laws of nature which govern the birth even of monstrosities. But the primitive mind, which believes in a close connection between a human social group and a snake or crocodile social group would find no more difficulty in this than in conceiving of the identity of the larva with the insect, or the chrysalis with the butterfly. Moreover, it is just as incompatible with ‘the laws of nature ”” that.a corpse, whose tissues have become chemically incapable of sustaining life, should arise again ; nevertheless, there are millions of cultivated persons who believe implicitly in the resurrection of Lazarus. It is enough that their representation of the Son of God involves His having the power to effect miracles. To the primitive mind, however, everything is a miracle, or rather, nothing is; and therefore everything is credible, and there is nothing either impossible or absurd. 378 HOW NATIVES THINK As a matter of fact, however—and in this sense the dictum is only partially applicable to the prelogical mind—the pre- connections involved in its collective representations are not as arbitrary as they appear. While indifferent to that which we call the real and objective relations between entities and manifestations, they express others much more important to such a mind, to wit, the mystic participating relations. It is these relations, and no others, which are realized in the preconnections, for these are the only ones about which the primitive mind troubles. Suggest to a primitive that there are other relations, imaginary or actual, between persons, things and occurrences: he will set them aside and reject them as untrue or insignificant or absurd. He will pay no attention to them, because he has his own experience to guide him, a mystic experience against which, as long as it con- tinues to exist, actual experience is powerless. It is not only therefore because, in itself and in the abstract, any relation whatever between entities and occurrences is just as acceptable © as any other; it is above all because the law of participation admits of mystic preconnections that the mind of the primitive seems undeterred by any physical impossibility. But, granted that in a certain community the mentality evolves at the same rate as the institutions, that these pre- connections grow weaker and cease to obtrude themselves— other relations between persons and things will be perceived, representations will tend to take on the form of general, abstract concepts, and at the same time a feeling, an idea of that which is physically possible or impossible will become more definite. It is the same then with a physical as with a logical absurdity, for the same causes render the prelogical mind insensible to both. Therefore the same changes and the same process of evolution cause it to be alive to the impossi- bility of affirming two contradictory statements at the same time, and the impossibility of believing in relations which are incompatible with experience. Such a concomitance cannot be merely adventitious. In both cases the impossibility is felt only in a condition common to both: it is necessary, and it is enough, that the collective representations tend towards conceptual form. On the one hand, in fact, participations expressed in such a form can TRANSITION TO THE HIGHER MENTAL TYPES 379 only be preserved, as we have already seen, by transforming themselves in order to avoid contradiction. And, on the other hand, it is when sufficiently definite concepts of beings and objects have been formed that the absurdity of certain mystic preconnections is first felt to obtrude. When the essential features of stone are, as it were, registered and fixed in the concept “stone,” which itself forms one among other concepts of natural objects differing from stone by properties no less definite and constant than its own, it becomes inconceivable that stones should speak or boulders move of their own accord or procreate human beings, etc. The more the concepts are determined, fixed and arranged in classes, the more con- tradictory do the statements which take no account of such relations appear. Thus the logical demand made by the intellect grows with the definition and determination of con- cepts, and it is an essential condition of such definition and determination that the mystic preconnections of the collective representations become impaired. It grows then simul- taneously with the knowledge acquired by experience. The progress of the one helps the other and vice versa, and we cannot say which is cause and which effect. V The process which is going on does not necessarily present itself as progressive, however. In the course of their evolu- tion concepts do not submit to a kind of “ finalité interne ” which directs them for the best. The weakening of the mystic preconnections and elements is not inevitable nor always continuous. The mentality of primitive peoples, even whilst becoming less impervious to the teaching of experience, long remains prelogical, and most of its ideas preserve a mystic imprint. Moreover, there is nothing to prevent abstract and general concepts, once formed, retaining elements which are still recognizable as vestiges of an earlier stage. Precon- nections, which experience has been unable to dissolve, still remain; mystic properties are yet inherent in beings and objects. Even in aggregates of the most advanced type, a concept which is free of all admixture of this kind is exceptional, and it is therefore scarcely to be met with in any others. The 380 HOW NATIVES THINK concept is a sort of logical “ precipitate’ of the collective representations which have preceded it, and this precipitate nearly always brings with it more or less of a residuum of mystic elements. How can it be otherwise? Even in social aggregates of a fairly low type abstract concepts are being formed, and while not in all respects comparable with our own, they are never- theless concepts. Must they not follow the general direction of the mentality which gives rise to them? They too, then, are prelogical and mystic, and it is only by very slow degrees that they cease to beso. It may even happen that after having been an aid to progress, they constitute an obstacle. Forif the determination of the concept provides the rational activity of the mind with a lever which it did not find in collective repre- sentations subject to the law of participation; if the mind inures itself to reject as impossible statements which are incompatible with the definition of the concepts, it very often pays dearly for the privilege when it grows used to regarding, as adequate to reality, conceptual ideas and relations which are very far removed from it. If progress is not to find itself arrested, concepts of entities of all kinds must remain plastic and be continually modified, enlarged, confined within fixed limits, transformed, disintegrated and reunited by the teaching of experience. If these concepts crystallize and become fixed, forming themselves into a system which claims to be self- sufficing, the mental activity applied to them will exert itself indefinitely without any contact with the reality they claim to represent. They will become the subject of fantastical and frivolous argument, and the starting-point of exaggerated infatuation. Chinese scientific knowledge affords a striking example of this arrested development. It has produced immense encyclo- pædias of astronomy, physics, chemistry, physiology, pathology, therapeutics and the like, and to our minds all this is nothing but balderdash. How can so much effort and skill have been expended in the long course of ages, and their product be absolutely nil? This is due to a variety of causes, no doubt, but above all to the fact that the foundation of each of these so-called sciences rests upon crystallized concepts, concepts which have never really been submitted to the test TRANSITION TO THE HIGHER MENTAL TYPES 381 of experience, and which contain scarcely anything beyond vague and unverifiable notions with mystic preconnections. The abstract, general form in which these concepts are clothed allows of a double process of analysis and synthesis which is apparently logical, and this process, always futile yet ever self-satisfied, is carried on to infinity. Those who are best acquainted with the Chinese mentality—like De Groot, for instance—almost despair of seeing it free itself from its shackles and cease revolving on its own axis. Its habit of thought has become too rigid, and the need it has begotten is too imperious. It would be as difficult to put Europe out of conceit with her savants as to make China give up her physicians and doctors and Fung-shui professors. India has known forms of intellectual activity more akin to our own. She has had her grammarians, mathematicians, logicians and metaphysicians, Why, however, has she pro- duced nothing resembling our natural sciences ? Undoubtedly, among other reasons, because there, too, concepts as a rule have retained a very considerable proportion of the mystic elements of the collective representations whence they are derived, and at the same time they have become crystallized. Thus they have remained unable to take advantage of any later evolution which would gradually have freed them from such elements, as in similar circumstances Greek thought fortunately did. From that time, even in becoming conceptual, their ideas were no less destined to remain chiefly mystic, and only with difficulty pervious to the teaching of experi- ence. If they furnished matter for scientific knowledge, the sciences could only be either of a symbolical and imagina- tive kind, or else argumentative and purely abstract. In peoples of a less advanced type, even although already fairly civilized, such as in Egypt or Mexico, for instance, even the collective representations which have been “ precipitated ” as concepts have distinctly retained their prelogical, mystic features. Finally let us consider the most favourable case, that of peoples among whom logical thought still continues its progress, whose concepts remain plastic and capable of con- tinual modification under the influence of experience. Even in such circumstances logical thought will not entirely supersede 382 HOW NATIVES THINK prelogical mentality. There are various reasons for the per- sistence of the latter. Firstly, in a large number of concepts there are indelible traces which still remain. It is far from being all the concepts in current use, for instance, which express the objective features and relations of entities and of phenomena solely. Such a characteristic is true of a very small number only, and these are made use of in scientific theorizing. Again, these concepts are, as a rule, highly abstract, and only express certain properties of phenomena and certain of their relations. Others, that is our most familiar concepts, nearly always retain some vestiges of the corresponding collective representations in prelogical mentality. Suppose, for example, that we are analysing the concepts of soul, life, death, society, order, fatherhood, beauty or anything else you like. If the analysis be complete it will undoubtedly comprise some relations dependent upon the law of partici- pation which have not yet entirely disappeared, Secondly, even supposing that the mystic, prelogical elements are finally eliminated from most concepts, the total disappearance of mystic, prelogical mentality does not neces- sarily follow. As a matter of fact the logical thinking which tends to realize itself through the purely conceptual and the intellectual treatment of pure concepts, is not co-extensive with the mentality which expressed itself in the earlier representations. The latter, as we know, does not consist of one function merely, or of a system of functions which are exclusively intellectual. It undoubtedly does comprise these functions but as still undifferentiated elements of a more complex whole in which cognition is blended with motor and above all emotional elements. If then, in the course of evo- lution the cognitive function tends to differentiate itself and be separated from the other elements implied in collective representations, it thereby achieves some kind of independ- ence, but it does not provide the equivalent of the functions it excludes. A certain portion of these elements, therefore, will subsist indefinitely outside and side by side with it. The characteristic features of logical thought are so clearly differentiated from those of prelogical mentality that the progress of the one seems, ipso facto, to involve the retro- gression of the other. We are tempted to conclude that in the TRANSITION TO THE HIGHER MENTAL TYPES 383 long run, when logical thought imposes its laws on all mental operations, prelogical mentality will have entirely disappeared. This conclusion is both hasty and unwarranted, however. Un- doubtedly the stronger and more habitual the claims of reason, the less tolerant it is of the contradictions and absurdities which can be proved as such. In this sense it is quite true to say that the greater the advance made by logical thought, the more seriously does it wage war upon ideas which, formed under the dominance of the law of participation, contain implied contradictions or express preconceptions which are incompatible with experience. Sooner or later such ideas are threatened with extinction, that is, they must be dissolved. But this intolerance is not reciprocal. If logical thought does not permit contradiction, and endeavours to suppress it as soon as it perceives it, prelogical, mystic mentality is on the contrary indifferent to the claims of reason. It does not seek that which is contradictory, nor yet does it avoid it. Even the proximity of a system of concepts strictly in accordance with the laws of logic exerts little or no influence upon it. Consequently logical thought can never be heir to the whole inheritance of prelogical mentality.\ Collective representations which express a participation intensely felt and lived, of which it would always be impossible to demonstrate either the logical | contradiction or the physical impossibility, will ever be main- | tained. Ina great many cases, even, they will be maintained, sometimes for a long time, 7m spite of such a demonstration. The vivid inner sentiment of participation may be equal to, and even exceed, the power of the intellectual claim. Such, in all aggregates known to us, are the collective representa- tions upon which many institutions are founded, especially many of those which involve our beliefs and our moral and religious customs. The unlimited persistence of ccd collective representa- tions and of the type of mind of which they are the witness, among peoples in whom logical thought is most advanced, enables us to comprehend why the satisfaction which is de- rived from the most finished sciences (exclusive of those which are purely abstract) is always incomplete. Compared with ignorance—at least, conscious ignorance—knowledge undoubtedly means a possession of its object ; but compared aay 384 HOW NATIVES THINK with the participation which prelogical mentality realizes, this possession is never anything but imperfect, incomplete, and, as it were, external. To know, in general, is to objectify ; and to objectify is to project beyond oneself, as something which is foreign to oneself, that which is to be known. How intimate, on the contrary, is the communion between entities participating of each other, which the collective representations of prelogical mentality assures! It is of the very essence of participation that all idea of duality is effaced, and that in spite of the law of contradiction the subject is at the same time himself and the being in whom he participates. To appreciate the extent to which this intimate possession differs from the objectifying apprehension in which cognition, pro- perly so called, consists, we do not even need to compare the collective representations of primitive peoples with the content of our positive sciences. It will be sufficient to consider one object of thought—God, for instance, sought after by the logical thought of advanced peoples, and at the same time assumed in the collective representations of another order. Any rational attempt to know God seems both to unite the thinking subject with God and at the same time to remove Him to a distance. The necessity of conforming with the claims of logic is opposed to a participation between man and God which is not to be represented without contradiction. Thus knowledge is reduced to a very small matter. But what need is there of this rational knowledge to the believer who feels himself at one with his God? Does not the con- sciousness which he possesses of his participation in the Divine essence procure him an assurance of faith, at the price of which logical certainty would always be something colourless and cold and almost a matter of indifference ? This experience of intimate and entire possession of the object, a more complete possession than any which originates in intellectual activity, may be the source and undoubtedly is the mainspring of the doctrines termed anti-intellectual. Such doctrines reappear periodically, and on each reappear- ance they find fresh favour. This is because they promise that which neither a purely positive science nor any theory of philosophy can hope to attain: a direct and intimate contact with the essence of being, by intuition, interpenetration, the TRANSITION TO THE HIGHER MENTAL TYPES 385 mutual communion of subject and object, full participation and immanence, in short, that which Plotinus has described as ecstasy. They teach that knowledge subjected to logical formulas is powerless to overcome duality, that it is not a veritable possession, but remains merely superficial. Now the need of participation assuredly remains something more im- perious and more intense, even among peoples like ourselves, than the thirst for knowledge and the desire for conformity with the claims of reason. It lies deeper within us and its source is more remote. During the long prehistoric ages, when the claims of reason were scarcely realized or even per- ceived, it was no doubt all-powerful in all human aggregates. Even to-day the mental activity which, by virtue of an intimate participation, possesses its object, gives it life and lives through it, aspires to nothing more, and finds entire satisfaction in this possession. But actual knowledge in con- formity with the claims of reason is always unachieved. It always appeals to a knowledge that protracts it yet further, and yet it seems as if the soul aspires to something deeper than mere knowledge, which shall encompass and perfect it. Between the theories of the ‘“ intellectualists ’’ and their opponents the dialectic strife may be indefinitely prolonged, with alternating victories and defeats. The study of the mystic, prelogical mentality of undeveloped peoples may enable us to see an end to it, by proving that the problems which divide the two parties are problems which are badly couched. For lack of proceeding by a comparative method, philosophers, psychologists and logicians have all granted one common postulate. They have taken as the starting-point of their investigations the human mind always and everywhere homo- geneous, that is, a single type of thinker, and one whose mental operations obey psychological and intellectual laws which are everywhere identical. The differences between institutions and beliefs must be explained, therefore, by the more childish and ignorant use which is made of principles common to all aggregates. Accordingly a reflective self-analysis carried out by a single individual ought to suffice to discover the laws of mental activity, since all subjects are assumed to be con- stituted alike, as far as mind is concerned. Now such a postulate does not tally with the facts revealed 25 386 HOW NATIVES THINK by a comparative study of the mentality of the various human aggregates. This teaches us that the mentality of primitive peoples is essentially mystic and prelogical in character ; that it takes a different direction from our own—that is, that its collective representations are regulated by the law of participation and are consequently indifferent to the law of contradiction, and united, the one to the other, by connections and preconnections which prove disconcerting to our reason. It throws light, too, upon our own mental activity. It leads us to recognize that the rational unity of the thinking being, which is taken for granted by most philosophers, is a desideratum, not a fact. Even among peoples like ourselves, ideas and relations between ideas governed by the law of participation are far from having disappeared. They exist, more or less independently, more or less impaired, but yet ineradicable, side by side with those subject to the laws of reasoning. Understanding, properly so called, tends towards ~ logical unity and proclaims its necessity ; but as a matter of fact our mental activity is both rational and irrational. The prelogical and the mystic are co-existent with the logical. On the one hand, the claims of reason desire to impose themselves on all that is imagined and thought. On the other hand, the collective representations of the social group, even when clearly prelogical and mystic by nature, tend to subsist indefinitely, like the religious and political institutions of which they are the expression, and, in another sense, the bases. Hence arise mental conflicts, as acute, and sometimes as tragic, as the conflict between rival duties. They, too, proceed from a struggle between collective habits, some time-worn and others more recent, differently oriented, which dispute the ascendancy of the mind, as differing moral claims rend the conscience. Undoubtedly it is thus that we should account for the so-called struggle of reason with itself, and for that which is real in its antinomies. And if it be true that our mentality is both logical and prelogical, the history of religious dogmas and systems of philosophy may henceforth be explained in a new light. INDEX ABIPONES, 22, 31, 75, 100, I5I, 157, 160, 178, 183, 259, 280-1, 296, 309, 333, 334, 343-4 abstraction, 116-21 ADAIR, 297 AFRICA, CENTRAL, 48, 130 AFRICA, EAST, 50, 75, IOI, 257, 322, 359 AFRICA, EQUATORIAL, 65, 95, 257, 277, 281, 293, 321 AFRICA, SOUTH, 39, 157, 164, 234, 287, 293, 310, 325, 349 AFRICA, WEST, 54, 66-7, 128, 134, 231, 238, 266, 267, 298, 305, 312, 314, 326, 346 alcheringa, 90, 92, 179, 317, 333, 338, 345, 303 Aleutian language, 146, 175 AMERICA, CENTRAL, 88, 208, 213 AMERICA, NORTH, 16, 31, 40, 47, 50, 213, 222, 281 (v. NORTH AMERI- CAN INDIANS) AMERICA, NORTH, INDIANS OF, 56-7, 70, 73, 85, 96, 97, 103, 112, 113, 131, 133, 140, 160, 180, 196, 197, 207, 208, 209, 211, 213-15, 232, 244, 249, 292, 310, 352 AMERICA, SOUTH, 102, 160, 181, 207, 222, 284, 293, 355, 363 ANDAMAN ISL., 22, 190, IQI animism, 18-27, 81-101, 102, 130, 365 ARISTOTLE, 126 ARNOT, 298 arungquiltha, 133-4, 278 ARUNTAS, 90, 92, 93, 119, 120, 133, 261, 317, 323, 306 ASHANTI, 64 associationism, 26 ASSYRO-BABYLONIANS, 218 AUSTRALIA, CENTRAL (v. EYRE & SPENCER & GILLEN) AUSTRALIAN ABORIGINES, 22, 40, 40, G27700, (FIO, 012, (hisieteG.. 190; 172, 181, 183, 207, 222, 262, 269- 112, 172, 76, 282, 289, 293, 307, 311, 321, 323, 338, 339-40, 341, 368 Australian languages, 143, 144, 146, 149, 158, 159, 160, 162, 174, 178, 179, 192, 280, 311, 341, 343 AYMONIER, 236-7 AZTECS, 198 badi, 40 BAFIOTI, 40, 101 BAGANDA, 243 (v. ROSCOE) BakiMma (v. ROSCOE) BAINING, 108 BANCROFT, 88, 97, 157, 281 Ba-RONGA, 102, 252-4, 270, 290, 329-39, 334 BA-YAKA (v. TORDAY & JOYCE) BENGAL, 65, 66, 252, 260, 309, 330 BENNETT, 65, 233, 277, 322 BERGAIGNE, 219-22 BEST (ELSDON), 70, 172, 268-9 BEVERIDGE, 269, 323 birth, a reincarnation, 338-42 birth, “‘ complete,”’ 342-3 BISMARCK ARCHIPELAGO, 170, 323 BOAS (Fr.), 52, 86, 160, 175, 194, 196, 199, 213, 232, 240, 242, 243, 266, 274, 315, 320, 324 BODDING, 51 BonNTOC, 43, 312 Bororo, 48, 77, 80, 91, 259, 260, 273, ,309, 318, 327, 363, 307 BOURLAY (A.), 265 BOWDITCH, 64 BRAZIL, 48, 77, III-I2, 115, 118, 160, 164, 169, 171, 258, 273, 289, 293, 302, 309, 318, 323, 327, 303 BRIDGES, (T.), 150, 157 BRITISH COLUMBIA, 52, 85, 94, 95, 160, 163-4, 175, 196, 199, 212, 231, 242, 274, 319 BROOKE (Rajah), 95, 189, 231, 269, 291 144, 169, 388 BROUGH SMYTH, 159, 170, 174, 268, 280, 283, 306, 323, 336, 341, 349 BROWN (W. H.), 231 BRUHIER, 310 BRUN, 281 BUCKLAND, 209 BULWER, 170 burial customs, 21, 22, 308-16 burial, premature, 309-10 CALABAR, 88, 89 CALIFORNIA, 171, 318, 324, 329, 344 CANADA (v. BOAS) cardinal points, 211-12, 215-16 CATLIN, 47, 80,\119, 180, 211, 215, 227, 244, 252 CEYLON, 321 CHALMERS (J.), 186, 190 CHAMBERLAIN, 51, 213, 218, 298 CHARLEVOIX, 59, 96, 112-13, 133, 157, 192 282,207. 274) 30% CHAUNCY, 341 CHEROKEES, 38, 39, 50-I, 57, I4I, 212 243. 290.) 23a2—%, 1237) 2320, 266-7, 271, 273, 274, 275, 284, 297 CHINA, CHINESE, 22, 23, 42, 46-7, 52, 54, 128, 211712, 214,’ 216, 250, 257, 287, 290, 299, 301, 304, 305, 311, 314, 321, 322, 325, 332, 338, 380-1 churinga, 92, 93, 98, 117, 120, 134, 323, 338, 363 ** classifiers,’ 199 CODRINGTON (R. H.), 31, 134, 143, 144, 193, 200-1 collective representations, 13, 35-68, 70) 1227610: COMTE (4.), 15, 16 CONANT, 192, 194, 196, 197, 198, 201 concepts, 76-00, 116-21, 127, etc. CONGO, 70, 71, 281, 326, 340, 356, 357 continuity of life, 103, 131 couvade, 256-60, 342 CRAWFURD, 198 CROOKE (W.), 217, 296 CULIN, 292 CUMMINGS, 265 CUNY, 145 CUSHING (F. H.), 41, 52, 70, 101, I6I-2, I9I, 210, 291, 292, 366, 373 HOW NATIVES THINK DACOTA, 132 DALE, 350 DAYAKS, 188, 316 death, primitive idea of, 276-82, 301-22, 353 DEGRANDPRE, 326 DE GROOD (los MN 22 ea, ae, 46-7, 48, 52, 54-5, 66-7, 211, 250, 287, 290, 299, 301, 304, 305, 311, 322, 325, 344, 381 DÈNÈ-DINDJIÉ (v. PETITOT) DENNETT, 128, 209 divination, 282-4 DOBRIZHOFFER, 31, 75, 100, 145, 146, 151, 157, 160, 178, 183, 184, 259, 280-1, 296, 309, 323, 324, 356 DORSEY, 52, 59-60, 85, 98, 132, 133, 210, 214, 215, 216, 234, 305, 306, 319 dreams, significance of, 56-61 DUARTE BARBOSA, 257 DU CHAILLU, 64-5, 95, 239, 257, 277, 321 DURKHEIM (EMIL), 18 DURKHEIM & MAUSS, 128, 364, 365 EDELFELT, 72, 318, 329 EGYPT, 381 EHRENREICH (P.), 102, 363 ELLIS (A. B.), 48, 82-4, 305, 311, 314, 322, 346 BELIOT, 219 ESKIMO, 231, 339 EWE-LANGUAGE (v. ELLIS (A. B.) & SPIETH) experience, native impermeability to, 63-5, 374 EVRE, AG) LEP LIo fady, 178 FANG, 65, 233, 277 fetich, fetish, 65,67, 290 (v. NASSAU) fetish-man, 65, 356, 357 FEWKES, 213, 216 fighting, 244-5 Fiji, FIJIANS, 54, 86, 179, 193, 201, 348-9 FISCHER (Dr. G. A.), 160 fishing, 239-44 FISON, 348 FITZROY, 114 FLETCHER (Miss A.), 103, 131 INDEX FRASER (J.), 311 FRAZER (Sir JAS.), 17, 18, 20, 21, 22, 32, 56, 68, 102, 177, 278, 298, 351, 353 FROGGART, 349 FUEGIANS, 22, I14 GAGE, 88 GALLATIN, 141, 168, 177 GASON (S.), 160, 234 GATSCHET (A.), 57, 62, 141, 142, 147, 148, 149, 150-6, 168, 173, 180, 199, 215, 260, 285 gesture language, 159-64 GLAVE,(E. J.), 357 GORDON (E. M.), 346 GOTTSCHLING (E.), 172 GRA Y,-72 GREECE, GREEKS, 314, 368, 381 GREY (GEO.), 111, 172, 282 GRIERSON, 169, 173, 197-8, 253 GUARANIS, 183 GUIANA, 41, 233, 271, 283, 335 GUISE {(R. E.), 73, 241 HADDON (A.), 84-5, 98, 160, 185, 189, 190, 193 HAGEN, 160, 258, 260, 283, 313 HARTLAND (S.), 17 HAWTREY, 192, 344 HAYWOOD, 277 HENNEPIN (Father), 49 HENRY (4.), 287 HENRY (V.), 146, 175 HERTAUK,); 317, 319, 321 HETHERWICK, 48, 130, 135 HEWITT, 213, 306 HEYWOOD, 212 HIDATSA, 59, 85 hikuli, 39, 122, 123, 124, 125, 128 HILL TOUT, 52, 94-5, 164, 212, 231, 244 HOBLEY (C. W.), 75, 322 HOFFMAN, 180 Hopt, 252 HOSE & MACDOUGALL, 58, 103, 261 HOWITT (A. W.), 56, 61, 185, 278, 341, 364 HUBERT & MAUSS, 99, 134, 135, 237, 293, 307 HUICHOLS, 38, 39, 63, 100, 122-5, 128, 131, 237, 3006 389 HUME, 377 HUNT, 185 hunting, practices relating to, 228-39 HURONS, 241, 242, 324, 326 IGOROTS, 43, 312 illness, how regarded, 263-76 image, mystic idea of, 46-7, 79-80 impermeability to experience, 63-8 IM THURM, 233, 271 INDIA, INDIAN TRIBES, 16, 40, 41, 47, 50; | §6, 120, 128) 169, 173,170, 180, 197-8, 217-19, 233, 240, 241, 253, 260, 265-6, 296, 299-300, 309, 310, 325, 332, 339, 344, 345, 368 INDIAN ARCHIPELAGO, 365 INDO-CHINA, 236 infanticide, 343-6 initiation, 260-2, 347-54 intichiuma, 39, 98, 247-51, 295, 362 IROQUOIS, 213, 306 iyuntarinia, 251, 355 JAPAN, JAPANESE, 51, 198, 214, 218, 298 JASTROW, 218 JAVA, 216 JENSKS, 43,312 JETTE (Fr.), 241 JEWITT, 179, 180, 239, 240 JONES (G. H.), 66 FONOD LH LAN MIO TASIUTO 7 252-4, 264, 270, 290, 293-4, 329- 39, 335 KAFIRS, 276, 281, 295 karakia, 243, 347 KEANE, 173 KINGSLEY (MARY), 54-5, 66-7, 70, 88, 89, 267, 271, 282, 290, 302, , 312, 340 KiTcH1 GAMA, 59, 160 KLAMATH, 57, 62, 141-56, 173, 180, 215, 260, 285 KOHL, 59, 160 KOREA, 66 KRUIJT (A. C.), 103, 365 kwet, 66 LANG (A.), 17, 20, 369 language by gesture, 159-64 Laos, 263-7, 265 390 LAPPS, 173 LAST, 178 Lautbilder, 164—6 law of participation, chs. II, VI, VII, VIII LAWES (W.G.), 144 LEIBNITZ, 61 LEJEUNE, 85 LEROY (Ch.), 182 lisoka, 130-1 LIVINGSTONE (D.), 64, 112, 157, 165, 172-3, 174, 234, 287, 310 LoanGo, 40, 43, 48, 65, 72, 73, 87, TOL, "113, 107,252; 257, 203~4, 299, 322 local relationship, 121, 129, 284 LUMHOLTZ (C.), 38, 62-3, 122-5, 239, 252, 297, 313, 329, 337 MACDONALD, 325, 349 MACDOUGALL (v. HOSE) MacGEE, 208, 209 MACKINTOSH (4.), 265 MACPHERSON, 345-7 MADAGASCAR, 178 MATER, H., 14 maladies, 263-76 MaALAYA, MALAYS, 39, 40, 57, IOI, 178, 216, 234, 248, 252, 319, 339 MALEBRANCHE, 99, 388 MALLERY, 160 mana, 14, 135, 176, 270, 347, 307 MANDANS, 47, 80, 85, I19, 215, 227, 252 MANN, 205 manual concepts, 161-2, I91I MAORIS, 70, 99, 172, 174, 201, 260, 331, 367, 369 MARETT, 365 MARSHALL ARCHIPELAGO, 169 masoka, 48 MATHEWS, 144, 146, 149, 177, 178, 235, 262, 341 MAUSS, 208, 339 (vu. DURKHEIM & MAUSS, HUBERT & MAUSS medicine-men, 61-2, 65, 278 medicine-men, initiation of, 354-8 MEILLET (4.), 173-286 memory, 109-16, 182-3 MExIco, 38, 39, 62-3, 122, 123, 237, 239, 252, 297, 313, 320, 337, 360, 381 MICRONESIA, 169 HOW NATIVES THINK MIKHAILOVSKI, 62 MOFFAT (R.), 172 | MOONEY (J. J.), 39, 50-1, 57, 212, 213, 216, 232-3, 238, 243, 266-7, 271, 275, 277; 297 mort, 270 MORICE, 194 MOSKOWSKI, 311 mourning practices, 308-22 MÜLLER (ÆGIDIUS), 276, 295 mulungi, 130-1, 133, 270 mystic abstraction, 117-21 mystic perceptions, 38-63, 288 myths, 368-74 names, naming, 50-4, 346-7, 350, 357 NASSAU (R. H.), 67, 231, 239, 266, 270, 279, 298 NEW FRANCE, 131 New GUINEA, 72, 73, 142, 143, 144, 160, 179, 186, 187, 188, 190, 207-8, 252, 258, 283, 318, 329 NEW HEBRIDES, 142, 144 NEw MEXICO, 249 NEW POMERANIA, 193-4 New SOUTH WALES, 144, 145, 146, 177, 234, 258 NEW ZEALAND, 70, 172-73, 243, 268, 296, 331, 336, 346-7, 307 nganga, 356 NICOBAR ISL., 42, 64, 239, 268 NIUÉ (v. SAVAGE ISLAND) number-concepts, 192-4 numbers, mystic power of, 205-23 numeration, systems of, 181-92 nurtunja, 116, I17, 119 ondaqut, 47-8 ovenda, 135, 176, 270 Papua, PAPUANS, 142; 160, 169, 258, 260, 283, 313 Paraguay, 192 PARKER (J.), 280-1 PARKINSON (R.), 118, 169, 171, 258, 323 participation, law of, ch. II PASSARGE, 350, 353 PECHUEL-LOESCHE, 40, 43, 48, 65, 71-2, 73, 74, 87, 101 113, 114, 167, 252, 257, 263-4, 299, 322 PETITOT, 145, 204, 329 PHILIPPINES, 43 PHILLIPS, 7% INDEX plurals, 142-5 PORTLAND, 191 portraits, mystic idea of, 47-8 POWELL (Major), 30, 58, 70, 141, 150, 151, 168-9, 199 POWERS, 171, 318, 324, 329, 344 practices, primitives’, 227-358 prefixes and suffixes, 152-8 PREOGSS (Keath. 39 primitives’ languages, 139-80 primitives’ mentality prelogical, 78, 295 primitives not given to analysis, 107-9 primitives’ objections to change, 41-3 primitives’ perceptions, 35-68, 288 QUEENSLAND, 72, III PIS OH); 170 READ (C. H.), 239 REED (W. A.), 269 RIBOT, 14, 15 RISLEY, 65-6, 260, 330 RIVERS (W. H. R.), 51 ROBERTSON SMITH, 297 ROHDE, 84 ROMANS, 254, 314 Mui Cs), 243, 276; 277, (295, 302, 331, 340 ROTH (W. E.), 62, 111, 158, 159, 162, | 341 ROUGIER (E.), 264 DACGARDN(FTr.),1 73,- 74, ZAI, 242, 324, 326 SAMOA, 31, 349 SANTALS, 51 SARASIN (P. and F.), 321 SARAWAK, 57-8, 95, 103, 189, 231, 261, 269, 291, 316 SAVAGE ISLAND, 312, 349 SCHMIDT (P. W.), 144, 351 SCHOMBURGK (R.), 283, 335, 337 SCHOOLCRAFT, 197, 237, 273 SELIGMAN, 279 shadows, 54-5 shaman, initiation of, 354-8 shamanism, 62 shen, sien, 22, 66-7 SIEROSHEWSKI, 87 Stoux, SIOUANS, 52, 59-60, 85, 98, 210, 243, 305-6, 319 391 SKEAT (W. W.), 40, 101, 102, 178, SLGWaigy 2348249 SMITHSONIAN INSTITUTE, 16, 30 SOCRATES, 126 SOLOMON, 42, 64, 239, 268 SOLOMON ISLES, 144, 193 soul, bush, 88 soul, double, 87 soul, multiple, 84 soul, primitive ideas of, 81-go space, conception of, 120-1 SPENCER (HERBERT), 21, 26, 27 SPENCER & GILLEN, 51, 58, 62, 74, 84, 90, OI, 92, 93, 94, 99, 110, 116, 117, 119-21, 133-4, 158, 159, 170,102, 214,247, 245240, 251) 258, 261, 262,1275,1278, 283, 284, 317, 318, 319, 321, 323, 333-4, 330, 339-49, 345, 350, 351, 354, 355, 358, 362, 363, 367, 371 SPIETH (J.), 324 SPIX & MARTIUS, 160, 164, 171 STEPHAN (Dr.), 194 STEVENSON (Myrs:), 316 STOKES, 114, 340 SUMATRA, 311 sun, mystic idea of, 129, 371 SWAN, 309 “sympathetic magic,’’ 298 210) ar elo: TAINE, 61 TANNA, 72, 179 tapu, taboo, 176, 268-9 TARAHUMARES (v. MEXICO) TASMANIANS, 170 TAYLOR (R.), 174, 296, 331, 336 THOMPSON (B. A.), 86 THOMSON (Basil), 312, 349 THURSTON (E.), 260, 300, 316, 325 THWAITES (E.), 48 TORDAY & JOYCE, 325, 348 TORREND, 146, 164 TORRES STRAITS, 84-5, 98, 134, 143, 160, 171, 185-6, 187, 189, 193, 201, 207, 239, 247-8, 279, 286, 304, 342 totem, 90-8, 120 totemic pilgrimages, 121 totemic relationship, 362 transition to higher types, 361-36 TUT Eleva 8) 251/240 392 PY LOR (EB) B.))46;.17, 20,247,305 48, 56, 81, 83, 84, 89, 102, 134, 298 USENER, 207, 209 VICTORIA (v. AUSTRALIAN ABORI- GINES) VON MARTIUS (C. F.), 258, 259, 260, 284, 302, 336, 355 VON DEN STEINEN, 48, 77, 91, III-I2, 114, 115, 118, 169, 259, 273, 318, 323, 327 VON DE WALL, 40 VOTH, 252 wakan, wakanda, 47, 103, 131, 132, 133, 176, 270 WEBSTER, 348, 351, 357 HOW NATIVES THINK WESTERMANN, 165-6 WESTERMARCK, 343 WIDENMANN (A.), 160 * widows, treatment of, 331-5 WILHELMI, 159, 236, 280, 341 witchcraft, 289-300 WOODS (J. D.), 160 WYATT GILL (Dr.), 185 YAHGANS, 149, 150 Yaxkuts (v. SIEROSHEWSKI) Yaos, 48, 130, 133 YARROW, 310 YORUBA, 205 ZAMBESI, 158, 174 ZuRIs, AI, 70, 161-2, 192, 210, 214, 249, 284, 291, 292, 295, 315, 334, 367, 368-9, 373 ad hee 4 | ENT Date Due RO AS ae 4 e | a2 | à : j PRINTED|IN U.S.A, Pe GN451 .L643 How natives think —Speer Library Princeton Theological Seminary 1 1012 00072 3348 | | | wh | iM ss mise 1e MN RENTE EE ages vine