The Methods of Ethnology Franz Boas STOR ® American Anthropologist, New Series, Vol. 22, No. 4 (Oct. - Dec, 1920), 311-321. Stable URL: http://links.jstor.org/sici?sici=0002-7294%28192010%2F12%292%3A22%3A4%3C311%3ATMOE%3E2.0.CO%3B2-5 American Anthropologist is currently published by American Anthropological Association. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/anthro.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support@jstor.org. http://www.j stor.org/ Tue Feb 15 12:56:32 2005 American Anthropologist NEW SERIES Vol. 22 October-December, 1920 No. 4 THE METHODS OF ETHNOLOGY By FRANZ BOAS DURING the last ten years the methods of inquiry into the historical development of civilization have undergone remarkable changes. During the second half of the last century evolutionary thought held almost complete sway and investigators like Spencer, Morgan, Tylor, Lubbock, to mention only a few, were under the spell of the idea of a general, uniform evolution of culture in which all parts of mankind participated. The newer development goes back in part to the influence of Ratzel whose geographical training impressed him with the importance of diffusion and migration. The problem of diffusion was taken up in detail particularly in America, but was applied in a much wider sense by Foy and Graebner, and finally seized upon in a still wider application by Elliot Smith and Rivers, so that at the present time, at least among certain groups of investigators in England and also in Germany, ethnological research is based on the concept of migration and dissemination rather than upon that of evolution. A critical study of these two directions of inquiry shows that each is founded on the application of one fundamental hypothesis. The evolutionary point of view presupposes that the course of historical changes in the cultural life of mankind follows definite laws which are applicable everywhere, and which bring it about that cultural development is, in its main lines, the same among all races and all peoples. This idea is clearly expressed by Tylor in the introductory pages of his classic work "Primitive Culture." As soon as we admit that the hypothesis of a uniform evolution 21 311 312 AMERICAN ANTHROPOLOGIST [n. s., 22, 1920 has to be proved before it can be accepted, the whole structure loses its foundation. It is true that there are indications of parallelism of development in different parts of the world, and that similar customs are found in the most diverse and widely separated parts of the globe. The occurrence of these similarities which are distributed so irregularly that they cannot readily be explained on the basis of diffusion, is one of the foundations of the evolutionary hypothesis, as it was the foundation of Bastian's psychologizing treatment of cultural phenomena. On the other hand, it may be recognized that the hypothesis implies the thought that our modern Western European civilization represents the highest cultural development towards which all other more primitive cultural types tend, and that, therefore, retrospectively, we construct an ortho-genetic development towards our own modern civilization. It is clear that if we admit that there may be different ultimate and coexisting types of civilization, the hypothesis of one single general line of development cannot be maintained. Opposed to these assumptions is the modern tendency to deny the existence of a general evolutionary scheme which would represent the history of the cultural development the world over. The hypothesis that there are inner causes which bring about similarities of development in remote parts of the globe is rejected and in its place it is assumed that identity of development in two different parts of the globe must always be due to migration and diffusion. On this basis historical contact is demanded for enormously large areas. The theory demands a high degree of stability of cultural traits such as is apparently observed in many primitive tribes, and it is furthermore based on the supposed correlation between a number of diverse and mutually independent cultural traits which reappear in the same combinations in distant parts of the world. In this sense, modern investigation takes up anew Gerland's theory of the persistence of a number of cultural traits which were developed in one center and carried by man in his migrations from continent to continent. It seems to me that if the hypothetical foundations of these two extreme forms of ethnological research are broadly stated as boas] THE METHODS OF ETHNOLOGY 313 I have tried to do here, it is at once clear that the correctness of the assumptions has not been demonstrated, but that arbitrarily the one or the other has been selected for the purpose of obtaining a consistent picture of cultural development. These methods are essentially forms of classification of the static phenomena of culture according to two distinct principles, and interpretations of these classifications as of historical significance, without, however, any attempt to prove that this interpretation is justifiable. To give an example: It is observed that in most parts of the world there are resemblances between decorative forms that are representative and others that are more or less geometrical. According to the evolutionary point of view, their development is explained in the following manner: the decorative forms are arranged in such order that the most representative forms are placed at the beginning. The other forms are so placed that they show a gradual transition from representative forms to purely conventional geometric forms, and this order is then interpreted as meaning that geometric designs originated from representative designs which gradually degenerated. This method has been pursued, for instance, by Putnam, Stolpe, Balfour, and Haddon, and by Verworn and, in his earlier writings, by von den Steinen. While I do not mean to deny that this development may have occurred, it would be rash to generalize and to claim that in every case the classification which has been made according to a definite principle represents an historical development. The order might as well be reversed and we might begin with a simple geometric element which, by the addition of new traits, might be developed into a representative design, and we might claim that this order represents an historical sequence. Both of these possibilities were considered by Holmes as early as 1885. Neither the one nor the other theory can be established without actual historical proof. The opposite attitude, namely, origin through diffusion, is exhibited in Heinrich Schurtz's attempt to connect the decorative art of Northwest America with that of Melanesia. The simple fact that in these areas elements occur that may be interpreted as eyes, induced him to assume that both have a common origin, without 314 AMERICAN ANTHROPOLOGIST [n. s., 22, 1920 allowing for the possibility that the pattern in the two areas—each of which shows highly distinctive characteristics—may have developed from independent sources. In this attempt Schurtz followed Ratzel who had already tried to establish connections between Melanesia and Northwest America on the basis of other cultural features. While ethnographical research based on these two fundamental hypotheses seems to characterize the general tendency of European thought, a different method is at present pursued by the majority of American anthropologists. The difference between the two directions of study may perhaps best be summarized by the statement that American scholars are primarily interested in the dynamic phenomena of cultural change, and try to elucidate cultural history by the application of the results of their studies; and that they relegate the solution of the ultimate question of the relative importance of parallelism of cultural development in distant areas, as against worldwide diffusion, and stability of cultural traits over long periods to a future time when the actual conditions of cultural change are better known. The American ethnological methods are analogous to those of European, particularly of Scandinavian, archaeology, and of the researches into the prehistoric period of the eastern Mediterranean area. It may seem to the distant observer that American students are engaged in a mass of detailed investigations without much bearing upon the solution of the ultimate problems of a philosophic history of human civilization. I think this interpretation of the American attitude would be unjust because the ultimate questions are as near to our hearts as they are to those of other scholars, only we do not hope to be able to solve an intricate historical problem by a formula. First of all, the whole problem of cultural history appears to us as a historical problem. In order to understand history it is necessary to know not only how things are, but how they have come to be. In the domain of ethnology, where, for most parts of the world, no historical facts are available except those that may be revealed by archaeological study, all evidence of change can be BOAS] THE METHODS OF ETHNOLOGY 315 inferred only by indirect methods. Their character is represented in the researches of students of comparative philology. The method is based on the comparison of static phenomena combined with the study of their distribution. What can be done by this method is well illustrated by Dr. Lowie's investigations of the military societies of the Plains Indians, or by the modern investigation of American mythology. It is, of course, true that we can never hope to obtain incontrovertible data relating to the chronological sequence of events, but certain general broad outlines can be ascertained with a high degree of probability, even of certainty. As soon as these methods are applied, primitive society loses the appearance of absolute stability which is conveyed to the student who sees a certain people only at a certain given time. All cultural forms rather appear in a constant state of flux and subject to fundamental modifications. It is intelligible why in our studies the problem of dissemination should take a prominent position. It is much easier to prove dissemination than to follow up developments due to inner forces, and the data for such a study are obtained with much greater difficulty. They may, however, be observed in every phenomenon of acculturation in which foreign elements are remodeled according to the patterns prevalent in their new environment, and they may be found in the peculiar local developments of widely spread ideas and activities. The reason why the study of inner development has not been taken up energetically, is not due to the fact that from a theoretical point of view it is unimportant, it is rather due to the inherent methodological difficulties. It may perhaps be recognized that in recent years attention is being drawn to this problem, as is manifested by the investigations on the processes of acculturation and of the interdependence of cultural activities which are attracting the attention of many investigators. The further pursuit of these inquiries emphasizes the importance of a feature which is common to all historic phenomena. While in natural sciences we are accustomed to consider a given number of causes and to study their effects, in historical happenings we are compelled to consider every phenomenon not only as effect but 3i6 AMERICAN ANTHROPOLOGIST [n. s., 22, 1920 also as cause. This is true even in the particular application of the laws of physical nature, as, for instance, in the study of astronomy in which the position of certain heavenly bodies at a given moment may be considered as the effect of gravitation, while, at the same time, their particular arrangement in space determines future changes. This relation appears much more clearly in the history of human civilization. To give an example: a surplus of food supply is liable to bring about an increase of population and an increase of leisure, which gives opportunity for occupations that are not absolutely necessary for the needs of every day life. In turn the increase of population and of leisure, which may'be applied to new inventions, give rise to a greater food supply and to a further increase in the amount of leisure, so that a cumulative effect results. Similar considerations may be made in regard to the important problem of the relation of the individual to society, a problem that has to be considered whenever we study the dynamic conditions of change. The activities of the individual are determined to a great extent by his social environment, but in turn his own activities influence the society in which he lives, and may bring about modifications in its form. Obviously, this problem is one of the most important ones to be taken up in a study of cultural changes. It is also beginning to attract the attention of students who are no longer satisfied with the systematic enumeration of standardized beliefs and customs of a tribe, but who begin to be interested in the question of the way in which the individual reacts to his whole social environment, and to the differences of opinion and of mode of action that occur in primitive society and which are the causes of far-reaching changes. In short then, the method which we try to develop is based on a study of the dynamic changes in society that may be observed at the present time. We refrain from the attempt to solve the fundamental problem of the general development of civilization until we have been able to unravel the processes that are going on under our eyes. Certain general conclusions may be drawn from this study even now. First of all, the history of human civilization does not boas] THE METHODS OF ETHNOLOGY 317 appear to us as determined entirely by psychological necessity that leads to a uniform evolution the world over. We rather see that each cultural group has its own unique history, dependent partly upon the peculiar inner development of the social group, and partly upon the foreign influences to which it has been subjected. There have been processes of gradual differentiation as well as processes of leveling down differences between neighboring cultural centers, but it would be quite impossible to understand, on the basis of a single evolutionary scheme, what happened to any particular people. An example of the contrast between the two points of view is clearly indicated by a comparison of the treatment of Zuni civilization by Frank Hamilton Cushing on the one hand, on the other by modern students, particularly by Elsie Clews Parsons, A. L. Kroeber and Leslie Spier. Cushing believed that it was possible to explain Zuni culture entirely on the basis of the reaction of the Zuni mind to its geographical environment, and that the whole of Zuni culture could be explained as the development which followed necessarily from the position in which the people were placed. Cushing's keen insight into the Indian mind and his thorough knowledge of the most intimate life of the people gave great plausibility to his interpretations. On the other hand, Dr. Parsons' studies prove conclusively the deep influence which Spanish ideas have had upon Zuni culture, and, together with Professor Kroeber's investigations, give us one of the best examples of acculturation that have come to our notice. The psychological explanation is entirely misleading, notwithstanding its plausibility, and the historical study shows us an entirely different picture, in which the unique combination of ancient traits (which in themselves are undoubtedly complex) and of European influences, have brought about the present condition. Studies of the dynamics of primitive life also show that an assumption of long continued stability such as is demanded by Elliot Smith is without any foundation in fact. Wherever primitive conditions have been studied in detail, they can be proved to be in a state of flux, and it would seem that there is a close parallelism between the history of language and the history of general cultural development. Periods of stability are followed by periods 318 AMERICAN ANTHROPOLOGIST [n. s., 22, 1920 of rapid change. It is exceedingly improbable that any customs of primitive people should be preserved unchanged for thousands of years. Furthermore, the phenomena of acculturation prove that a transfer of customs from one region into another without concomitant changes due to acculturation, are very rare. It is, therefore, very unlikely that ancient Mediterranean customs could be found at the present time practically unchanged in different parts of the globe, as Elliot Smith's theory demands. While on the whole the unique historical character of cultural growth in each area stands out as a salient element in the history of cultural development, we may recognize at the same time that certain typical parallelisms do occur. We are, however, not so much inclined to look for these similarities in detailed customs but rather in certain dynamic conditions which are due to social or psychological causes that are liable to lead to similar results. The example of the relation between food supply and population to which I referred before may serve as an example. Another type of example is presented in those cases in which a certain problem confronting man may be solved by a limited number of methods only. When we find, for instance, marriage as a universal institution, it may be recognized that marriage is possible only between a number of men and a number of women; a number of men and one woman; a number of women and one man; or one man and one woman. As a matter of fact, all these forms are found the world over and it is, therefore, not surprising that analogous forms should have been adopted quite independently in different parts of the world, and, considering both the general economic conditions of mankind and the character of sexual instinct in the higher animals, it also does not seem surprising that group marriage and polyandrous marriages should be comparatively speaking rare. Similar considerations may also be made in regard to the philosophical views held by mankind. In short, if we look for laws, the laws relate to the effects of physiological, psychological, and social conditions, not to sequences of cultural achievement. In some cases a regular sequence of these may accompany the development of the psychological or social status. This is illus- boas] THE METHODS OF ETHNOLOGY 319 trated by the sequence of industrial inventions in the Old World and in America, which I consider as independent. A period of food gathering and of the use of stone was followed by the invention of agriculture, of pottery and finally of the use of metals. Obviously, this order is based on the increased amount of time given by mankind to the use of natural products, of tools and utensils, and to the variations that developed with it. Although in this case parallelism seems to exist on the two continents, it would be futile to try to follow out the order in detail. As a matter of fact, it does not apply to other inventions. The domestication of animals, which, in the Old World must have been an early achievement, was very late in the New World, where domesticated animals, except the dog, hardly existed at all at the time of discovery. A slight beginning had been made in Peru with the domestication of the llama, and birds were kept in various parts of the continent. A similar consideration may be made in regard to the development of rationalism. It seems to be one of the fundamental characteristics of the development of mankind that activities which have developed unconsciously are gradually made the subject of reasoning. We may observe this process everywhere. It appears, perhaps, most clearly in the history of science which has gradually extended the scope of its inquiry over an ever-widening field and which has raised into consciousness human activities that are automatically performed in the life of the individual and of society. I have not heretofore referred to another aspect of modern ethnology which is connected with the growth of psycho-analysis. Sigmund Freud has attempted to show that primitive thought is in many respects analogous to those forms of individual psychic activity which he has explored by his psycho-analytical methods. In many respects his attempts are similar to the interpretation of mythology by symbolists like Stucken. Rivers has taken hold of Freud's suggestion as well as of the interpretations of Graebner and Elliot Smith, and we find, therefore, in his new writings a peculiar disconnected application of a psychologizing attitude and the application of the theory of ancient transmission. While I believe some of the ideas underlying Freud's psycho- 320 AMERICAN ANTHROPOLOGIST [n. s., 22, 1920 analytic studies may be fruitfully applied to ethnological problems, it does not seem to me that the one-sided exploitation of this method will advance our understanding of the development of human society. It is certainly true that the influence of impressions received during the first few years of life have been entirely underestimated and that the social behavior of man depends to a great extent upon the earliest habits which are established before the time when connected memory begins, and that many so-called racial or hereditary traits are to be considered rather as a result of early exposure to a certain form of social conditions. Most of these habits do not rise into consciousness and are, therefore, broken with difficulty only. Much of the difference in the behavior of adult male and female may go back to this cause. If, however, we try to apply the whole theory of the influence of suppressed desires to the activities of man living under different social forms, I think we extend beyond their legitimate limits the inferences that may be drawn from the observation of normal and abnormal individual psychology. Many other factors are of greater importance. To give an example: The phenomena of language show clearly that conditions quite different from those to which psychoanalysts direct their attention determine the mental behavior of man. The general concepts underlying language are entirely unknown to most people. They do not rise into consciousness until the scientific study of grammar begins. Nevertheless, the categories of language compel us to see the world arranged in certain definite conceptual groups which, on account of our lack of knowledge of linguistic processes, are taken as objective categories and which, therefore, impose themselves upon the form of our thoughts. It is not known what the origin of these categories may be, but it seems quite certain that they have nothing to do with the phenomena which are the subject of psycho-analytic study. The applicability of the psycho-analytic theory of symbolism is also open to the greatest doubt. We should remember that symbolic interpretation has occupied a prominent position in the philosophy of all times. It is present not only in primitive life, but the history of philosophy and of theology abounds in examples boas] THE METHODS OF ETHNOLOGY 32I of a high development of symbolism, the type of which depends upon the general mental attitude of the philosopher who develops it. The theologians who interpreted the Bible on the basis of religious symbolism were no less certain of the correctness of their views, than the psycho-analysts are of their interpretations of thought and conduct based on sexual symbolism. The results of a symbolic interpretation depend primarily upon the subjective attitude of the investigator who arranges phenomena according to his leading concept. In order to prove the applicability of the symbolism of psycho-analysis, it would be necessary to show that a symbolic interpretation from other entirely different points of view would not be equally plausible, and that explanations that leave out symbolic significance or reduce it to a minimum, would not be adequate. While, therefore, we may welcome the application of every advance in the method of psychological investigation, we cannot accept as an advance in ethnological method the crude transfer of a novel, one-sided method of psychological investigation of the individual to social phenomena the origin of which can be shown to be historically determined and to be subject to influences that are not at all comparable to those that control the psychology of the individual. Columbia University, New York City