(RELIGION)^* 
I  -  ANTHROPOLOGICAL STUDY 

(by Clifford Geertz)

The anthropological study of religion has been highly sensitive to changes in the general intellectual and moral climate of the day; at the same time, it has been a powerful factor in the creation of that climate. Since the early discussion by Edward Tylor, interest in the beliefs and rituals of distant, ancient, or simpler peoples has been shaped by an awareness of contemporary issues. The questions that anthropologists have pursued among exotic religions have arisen from the workings--or the misworkings--of modern Western society, and particularly from its restless quest for self-discovery. In turn, their findings have profoundly affected the course that quest has taken and the perspective at which it has arrived.

Perhaps the chief reason for the rather special role of comparative religious studies is that issues which, when raised within the context of Western culture, led to extreme social resistance and personal turmoil could be freely and even comfortably handled in terms of bizarre, presumably primitive, and thus--also presumably--fanciful materials from long ago or far away. The study of "primitive religions" could pass as the study of superstition, supposedly unrelated to the serious religious and moral concerns of advanced civilization, at best either a sort of vague foreshadowing of them or a grotesque parody upon them. This made it possible to approach all sorts of touchy subjects, such as polytheism, value relativism, possession, and faith healing, from a frank and detached point of view. One could ask searching questions about the historicity of myth among Polynesians; when asked in relation to Christianity, these same questions were, until quite recently, deeply threatening. One could discuss the projection of erotic wishes found in the "totemic" rites of Australian aborigines, the social roots and functions of Mrican "ancestor worship," or the protoscientific quality of Melanesian "magical thought," without involving oneself in polemical debate and emotional distress. The application of the comparative method--the essence of anthropological thought--to religion permitted the growth of a resolutely scientific approach to the spiritual dimensions of buman life.

Through the thin disguise of comparative method the revolutionary implications of the work of such men as Tylor, Durkheim, Robertson Smith, Freud, Malinowski, and Radliffe-Brown soon became apparent--at first mainly to philosophers, theologians, and literary figures, but eventually to the educated public in general. The meticulous descriptions of tribal curiosities such as soul loss, shamanism, circumcision, blood sacrifice, sorcery, tree burial, garden magic, symbolic canilibalism, and animal worship have been caught up in some of the grander intellectual battles of the last hundred years--from those over evolutionism and historicism in the late nineteenth century to those over positivism and existentialism today. Psychoanalysts and phenomenologists, Marxists and Kantians, racists and egalitarians, absolutists and relativists, empiricists and rationalists, believers and skeptics have all had recourse to the record--partial, inconsistent, and shot through with simple error as it is--of the spiritual life of tribal peoples to support their positions and belabor those of their opponents. If interest in "primitive religion" among savants of all sorts has been remarkably high, consensus concerning its nature and significance has not.

At least three major intellectual developments have exercised a critical influence on the anthropological study of religion: (1) the emergence, in the latter half of the nineteenth century, of history as the sovereign science of man; (2) the positivist reaction against this sovereignty in the first decades of the twentieth century and the radical split of the social sciences into resolutely psychological approaches, on the one hand, and resolutely sociological ones, on the other; and (3) the growth, in the interwar period, of a concern with the role of ideational factors in the regulation of social life. With the first of these came an emphasis on the nature of primitive reasoning and the stages of its evolution into civilized thought. With the second came an investigation of the emotional basis of religious ritual and belief and the separate examination of the role of ritual and belief in social integration. The concern with value systems and other features of the ideational realm led to an exploration of the philosophical dimensions of religious ideas, particularly the symbolic vehicles in terms of which those ideas are expressed.

Evolutionism and its enemies

Like so much else in anthropology, the study of the religious notions of primitive peoples arose within the context of evolutionary theory. In the nineteenth century, to think systematically about human affairs was to think historically--to seek out survivals of the most elementary forms and to trace the steps by which these forms subsequently developed. And though, in fact, Tylor, Morgan, Frazer, and the rest drew more on the synthetic social-stage theories of such men as Comte and Hegel than on the analytic random-variation and natural-selection ideas of Darwin, the grand concept of evolution was shared by both streams of thought: namely, that the complex, heterogeneous present has arisen, more or less gradually, out of a simpler, more uniform past. The relics of this past are still to be found scattered, like Galapagos turtles, in out-of-the-way places around us. Tylor, an armchair scholar, made no "voyage of the Beagle." But in combing and organizing the reports of missionaries, soldiers, and explorers, he proceeded from the same general premise as did Darwin, and indeed most of the leading minds of the day. For them a comprehensive, historically oriented comparison of all forms of a phenomenon, from the most primitive to the most advanced, was the royal road to understanding the nature of the phenomenon itself.

In Tylor's view, the elementary form out of which all else developed was spirit worship--animism. The minimal definition of religion was "a belief in spiritual beings." The understanding of religion thus came down to an understanding of the basis upon which such a belief arose at its most primitive level. Tylor's theory was intellectualistic. Belief in spirits began as an uncritical but nonetheless rational effort to explain such puzzling empirical phenomena as death, dreams, and possession. The notion of a separable soul rendered these phenomena intelligible in terms of soul departure, soul wandering, and soul invasion. Tylor believed that the idea of a soul was used to explain more and more remote and hitherto inexplicable natural occurrences, until virtually every tree and rock was haunted by some sort of gossamer presence. The higher, more developed forms of "belief in spiritual beings," first polytheism, ultimately monotheism, were founded upon this animistic basis, the urphilosophy of all mankind, and were refined through a process of critical questioning by more advanced thinkers. For this earnest Quaker the religious history of the world was a history of progressive, even inevitable, enlightenment. [See TyLOR.]

This intellectualistic, "up from darkness" strain has run through most evolutionist thought about religion. For Frazer, a nineteenth-century figure who lived for forty years into the twentieth century without finding it necessary to alter either his views or his methods, the mental progress involved was from magic to religion to science. Magic was the primordial form of human thought; it consisted in mistaking either spatio-temporal connection ("sympathetic magic," as when drinking the blood of an ox transfers its strength to the drinker) or phenomenal similarity ("imitative magic," as when the sound of drumming induces thunderheads to form) for true scientific causality. For Durkheim, evolutionary advance consisted in the emergence of specific, analytic, profane ideas about "cause" or "category" or "relationship" from diffuse, global, sacred images. These "collective representations," as he called them, of the social order and its moral force included such sacra as "mana," "totem," and "god." For Max Weber, the process was one of "rationalization": the progressive organization of religious concern into certain more precisely defined, more specifically focused, and more systematically conceived cultural forms. The level of sophistication of such theories (and, hence, their present relevance) varies very widely. But, like Tylor's, they all conceive of the evolution of religion as a process of cultural differentiation: the diffuse, all-embrac!ng, but rather unsystematic and uncritical religious practices of primitive peoples are transformed into the more specifically focused, more regularized, less comprehensively authoritative practices of the more advanced civilizations. Weber, in whom both intellectualism and optimism were rather severely tempered by a chronic apprehensiveness, called this transformation the "disenchantment (Entzauberung) of the world." [See DURKHEIM; FRAZER; WEBER, MAX.]

On the heels of evolutionism came, of course, antievolutionism. This took two quite different forms. On one side there was a defense, mainly by Roman Catholic scholars, of the so-called degradation theory. According to this theory , the original revelation of a high god to primitive peoples was later corrupted by human frailty into the idol worship of present-day tribal peoples. On the other side there was an attack, mainly by American scholars of the Boas school, upon the "armchair speculation" of evolutionary thinkers and a call for its replacement by more phenomenological approaches to the study of tribal custom.

The first of these reactions led, logically enough, to a search among the most primitive of existing peoples for traces of belief in a supreme being.. The resulting dispute, protracted, often bitter, and stubbornly inconclusive as to the existence of such "primitive monotheism," turned out to be unproductive--aside from some interesting discussions by Lang ( 1898) concerning culture heroes and by Eliade (1949) concerning sky gods--and both the issue and the theory that gave rise to it have now receded from the center of scholarly attention. The second reaction has had a longer life and great impact on .ethnographic methodology, but it too is now in partial eclipse. Its main contributions--aside from some devastating empirical demolitions of evolutionist generalization--came in the field of cultural diffusion. Leslie Spier's study of the spread of the Sun Dance through the Great Plains (1921) and A. L. Kroeber's application of the age-area approach to aboriginal religion in California are gqod examples of productive diffusion studies. However, apart from their importance for culture history, the contribution of such distributional studies to our understanding of religious ideas, attitudes, and practices as such has not been great, and few students now pursue these studies. The call of the Boas school for thorough field research and disciplined inductive analysis has been heeded; but its fruits, insofar as religious studies are concerned, have been reaped Qy others less inhibited theoretically.

Psychological approaches

The major reaction against the intellectual tradition of the cultural evolutionists took place not within anthropology, however, but in the general context of the positivist revolt against the domination of historicist modes of thought in the social sciences. In the years before World War I the rise of the systematic psychologism of psychoanalysis and of the equally systematic sociologism of the Annee sociologique forced evolutionist theorizing into the background, even though the leaders of both movements--Freud and Durkheim--were themselves still very strongly influenced by it. Perhaps even more relevant, it introduced a sharp split into anthropological studies of religion which has resolved into the militantly psychodynamic and the militantly social-structural approaches.

Freud's major work in this field is, of course, Totem and Taboo, a book anthropologists in general have had great difficulty in evaluating--as Kroeber's two reviews of it, the first facilely negative, the second, two decades later, ambivalently positive, demonstrate. The source of the difficulty has been an inability or an unwillingness to disentangle Freud's basic thesis--that religious rituals and beliefs are homologous with neurotic symptoms--from the chimerical ethnology and obsolete biology within which he insisted upon setting it. Thus, the easy demolition of what Kroeber called Freud's "just so story" concerning primal incest, parricide, and guilt within some prqtohumanhorde ("in the beginning was the deed") was all too often mistaken for total rejection of the rather more penetrating proposition that the obsessions, dreams, and fantasies of collective life spring from the same intrapsychic sources as do those of the isolati:id individual.

For those who read further in Freud's writings, however--especially in "Mourning and Melancholia" and "Obsessive Acts and Religious Practices"--it became apparent that what was at issue was the applicability of theories concerning the forms and causes of individual psychopathology to the explanation of the forms and causes of public myth and group ritual. RÛheim (1950) analyzed Australian circumcision rites against the background of orthodox Freudian theories of psychosexual development, especially those clustered around the Oedipal predicament. However, he explicitly avoided recourse to speculations about buried memories of primordial occurrences. Bettelheim (1954) adopted a similar, though more systematic and less orthodox, approach to initiation practices generally, seeing them as socially instituted symbolic mechanisms for the definifion and stabilization of sexual identity. Kardiner (1945) , taking a neo-Freudian position, sought to demonstrate that the religibus institutions of tribal peoples were projections of a "basic personality structure," formed not by the action of an unconsciously remembered historical trauma but by the more observable traumas produced by child-training practices, an approach later extended and cast into quantitative form by Whiting (Whiting & Child 1953). Erikson (1950), drawing upon developments in ego psychology which conceived the emergence of the adult personality to be a joint product of psychobiological maturation, cultural context, and historical experience, interpreted the religious notions of the Yurok and the Sioux in terms of certain basic modes of relating to the world. These relationships gradually developed during the whole course of childhood and adolescence. Othersnotably Devereux (1951)--have attempted to use the 'autobiographical, case-history approach to determine the relations between personality dynamics and religious orientation in particular individuals; still others--notably Hallowell (1937-1954 )--have employed projective tests, questionnaires, reports of dreams, or systematic interviews toward similar ends.

In all such studies, even when individual authors have dissented from many of Freud's specific views, the basic premise has been Freudian: that religious practices can be usefully interpreted as expressions of unconscious psychological forces--and this has become, amid much polemic, an established tradition of inquiry. In recent years, however, responsible work of this type has come to question the degree to which one is justified in subjecting historically created and socially institutionalized cultural forms to a system of analysis founded on the treatment of the mental illnesses of individuals. For this reason, the future of this approach depends perhaps more upon developments within psychoanalysis, now in a somewhat uncertain state, than within anthropology. So far, perhaps only Kluckhohn's pioneering Navaho Witchcraft ( 1944 ) has attempted to systematically relate psychological factors to social and cultural aspects of primitive religion. The great majority of psychoanalytic studies of tribal beliefs and rites remain willfully parochial.

In any case, not all psychological approaches to religion have been Freudian. Jungian influences have had a certain impact, especially on studies of myth. Campbell (1949), for example, has stressed the continuity of certain themes both cross-culturally and temporally. These themes have been interpreted as expressions of transpersonal constancies in unconscious mental functioning which are at the same time expressions of fundamental cosmic realities.

Simple emotionalist theories have also been extremely popular. There have been two main varieties of these: awe theories and confidence theories. Awe theories have been based on some usually rather vague notion of "religious thrill" experienced by human beings when brought face to face with CbSmic forces. A wide range of ethnologists, from' Max Muller through Lang and Marett to Lowie and Goldenweiser, have accepted such theories in ope form or another. However, awe theories remail1 mere notations of the obvious-that religious experience is, in the nature of the case; touched with intense feelings of the grandeur of the universe in relation to the self and of the' vulnerability of the self in relation to the universe. This is not explanation, but circular reasoning.

Confidence theories also begin with a notion of man's inward sense of weakness, and especially of his fears--of disease, of death, of ill forrilne of all kinds--and they see religious practices as designed to quiet such fears, either by explaining them away, as in doctrines of the afterlife, or by claiming to link the individual to external sources of sti'ehgth, as in prayer. The best-known confidence theory was that set forth by Malinowski. He regarded magic as enabling man to pursue uncertain but essential endeavors by assuring him of their ultimate success. Confidence, or anxiety-reduction, theories, like awe theories, clearly have empirical foundation but do not adequately explore the complex relationship between fear and religious activity. They are not rooted in any systematic conceptualization of mental functioning and so merely point to matters desperately in need of clarification, without in fact clarifying them.

Sociological approaches

The sociological approach to the analysis of the religions of nonliterate peoples proceeded independent of, and even at variance with, the psychoanalytic approach, but it shared a concern with the same phenomenon: the peculiar "otherness," the extraordinary, momentous, "set apart" quality of sacred (or "taboo") acts and objects, as contrasted with the profane. The intense aura of high seriousness was traced by Freud to the projection of unacceptable wishes repressed from consciousness onto external objects. The dramatic ambiva.lence of the sacred--its paradoxical unification of the commanded and the forbidden, the pure and the polluted, the salutary and the dangerous-was a symbolic expression of the underlying ambivalence of human desires. For Durkheim, too, the extraordinary atmosphere surrounding sacred acts and objects was symbolic of a hidden reality, but a social, not a psychological one: the moral force of the human community.

Durkheim believed that the integrity of the social order was the primary requisite for human survival, and the means bywhich that integrity superseded individual egocentricity was the primary problem of sociological analysis. He saw Australian totemism ( which he, like Freud, made the empirical focus of his work) as a mechanism to this end. For example, the collective rituals involving the emblems of the totemic beings--the so-called bull roarers--aroused the heightened emotions of mass behavior and evoked a deep sense of moral identification among the participants. The creation of social solidarity was the result of the common public veneration, by specific groups of persons, of certain carefully designated symbolic objects. These objects had no intrinsic value except as perceptible representations of the social identity of the individuals. Collective worship of consecrated bits of painted wood or stone created a moral community, a "church," upon which rested the viability of the major social units. These sanctified objects thus represented the system of rights and obligations implicit in the social order and the individual's unformulated sense of its overriding significance in his life. All sacred objects, beliefs, and acts, and the extraordinary emotions attending them, were outward expressions of inward social necessities, and, in a famous phrase, God was the "symbol of society." Few anthropologists have been able to swallow Durkheim's thesis whole, when put this baldly. But the more moderate prop. osition that religious rituals and beliefs both reflect and act to support the moral framework underlying social arrangements (and are in turn animated by it) has given rise to what has become perhaps the most popular form of analysis in the anthropological study of religion. Usually called "functionalism"--or sometimes, to distinguish it from certain variants deemed objectionable, "structuralism"--this approach found its champion in Radcliffe-Brown and its major development in Great Britain, though its influence has now spread very much more widely.

Radcliffe-Brown ( 1952) agreed with Durkheim's postulate that the main role (or "function") of religion was to celebrate and sustain the norms upon which the integration of society depends. But, unlike Durkheim (and like Freud), Radcliffe-Brown was concerned with the content of sacred symbols, and particularly with the reasons why one object rather than another was absorbed into rite or woven into myth. Why here stones, there water holes, here camp circles, there personified winds?

Durkheim had held this to be an arbitrary matter, contingent upon historical accident or psychological proclivity, beyond the reach of and irrelevant to sociological analysis. Radcliffe-Brown considered, however, that man's need for a concrete expression of social solidarity was not sufficient explanation of the structure of a people's religious system. Something was needed to tie the particular objects awarded sacred status (or, in his terminology, "ritual value") to the particular social interests they presumably served and reflected. Radcliffe-Brown, resolute empiricist that he was, chose a solution Durkheim had already magisterially demolished: the utilitarian. The objects selected for religious veneration by a given people were either directly or indirectly connected to factors critical to their collective well-being. Things that had real, that is, practical, "social value" were elevated to having spiritual, or symbolic, "ritual value," thus fusing the social and the natural. into one overarching order. For primitives at least (and Radcliffe-Brown attempted to establish his theory with regard to the sanctified turtles and palm leaves of the pre-agricultural Andaman Islanders and, later on, with regard to Australian totemism) , there is no discontinuity, no difference even, between moral and physical, spiritual and practical relationships and processes. These people regard both men and things as parts of a single normative system. Within that system those elements which are critical to its effective functioning (or, sometimes, phenomena empirically associated with such elements, such as the Andaman cicada cycle and the shifting monsoons) are made the objects of that special sort of respect and attention which we call religious but which the people themselves regard as merely prudential.

Radcliffe-Brown focused upon the content of sacred symbols and emphasized the relation between conceptions of the moral order of existence and conceptions of its natural order. However, the claim that the sanctity of religious objects derives from their practical social importance is one of those theories which works when it works and doesn't when it doesn't. Not only has it proved impossible to find even an indirect practical significance in most of the enormous variety of things tribal peoples have regarded as sacred (certain Australian tribes worship vomit), but the view that religious concerns are mere ritualizations of real-life concerns leaves the phenomenon of sacredness itself--its aura of mystery, power, fascination--totally unexplained.

More recent structuralist studies have tended to evade both these questions and to concentrate on the role played by religion in maintaining social equilibrium. They attempt to show how given sets of religious practices ( ncestor worship, animal sacrifice, witchcraft and sorcery, regeneration rites) do in fact express and reinforce the moral values underlying crucial processes (lineage segmentation, marriage, conflict adjudication, political succession) in the particular society under investigation. Arnold van Gennep's study of crisis rites was perhaps the most important forerunner of the many analyses of this type. Although valuable in their own right as ethnography and as sociology, these structural formulations have been severely limited by their rigid avoidance, on the one side, of the kind of psychological considerations that could account for the peculiar emotions which permeate religious belief and practice, and, on the other, of the philosophical considerations that could render their equally peculiar content intelligible. [See GENNEP.]

The analysis of symbolic forms

In contrast to other approaches--eyolutionary, psychological, sociological--the field of what we may loosely call "semantic studies" of religion is extremely jumbled. There is, as yet, no well-established central trend of analysis, no central figure around whom to order debate, and no readily apparent system of interconnections relating the various competing trends to one another.

 Perhaps the most straightforward strategy--certainly the most disarming--is merely to accept the myriad expressions of the sacred in primitive societies, to consider them as actual ingressions of the divine into the world, and to trace the forms these expressions have taken across the earth and through time. The result would be a sort of natural history of revelation, whose aim would be to isolate the major classes of religious phenomena considered as authentic manifestations of the sacred--what Eliade, the chief proponent of this approach, calls hierophanies--and to trace the rise, dominance, decline, and disappearance of these classes within the changing contexts of human life. The meaning of religious activity, the burden of its content, is discovered through a meticulous, wholly inductive investigation of the natural modalities of such behavior (sun worship, water symbolism, fertility cults, renewal myths, etc.) and of the vicissitudes these modalities undergo when projected, like the Son of God himself, into the flux of history.

Metaphysical questions (here uncommonly obtrusive) aside, the weaknesses of this approach derive from the same source as its strengths: a drastic limiting of the interpretations of religion to the sort that a resolutely Baconian methodology can produce. On the one hand, this approach has led, especially in the case of a scholar as erudite and indefatigable as Eliade, to the uncovering of some highly suggestive clusterings of certain religious patterns with particular historical conditions--for example, the frequent association of sun worship, activist conceptions of divine power, cultic veneration of deified heroes, elitist doctrines of political sovereignty, and imperialist ideologies of national expansion. But, on the other hand, it has placed beyond the range of scientific analysis everything but the history and morphology of the phenomenal forms of religious expression. The study of tribal beliefs and practices is reduced to a kind of cultural paleontology whose sole aim is the reconstruction, from scattered and corrupted fragments, of the "mental universe of archaic man."

Primitive thought. Other scholars who are interested in the meaningful content of primitive religion but who are incapable of so thoroughgoing a suspension of disbelief as Eliade, or are repelled by the cultic overtones of this somewhat mystagogic line of thought, have directed their attention instead toward logical and epistemological considerations. This has produced along series of studies that view "primitive thought" as a distinctive mode of reasoning and/or a special body of knowledge. From LÈvy-Bruhl through LÈvi-Strauss, and with important contributions from members of the evolutionary, psychoanalytic, and sociological schools as well, this line of exploration has persisted as a minor theme in anthropological studies of religion. With the recent advances in linguistics, information theory, the analysis of cognition, semantic philosophy, modern logic, and certain sorts of literary investigation, the systematic study of symbolic activity bids fair to become, in a rather thoroughly revised form, the major theme for investigation. The "new key" Susanne K. Langer heard being struck in philosophy in the early 1940s--"the concern with the concept of meaning in all its forms"--has, like the historicist and positivist "keys" before it, begun to have its echo in the anthropological study of religion. Anthropologists are increasingly interested in ideational expression, increasingly concerned with the vehicles, processes, and practical applications of human conceptualization.

The development of this approach has come in two fairly distinct phases, one before and one after World War II. In the first phase there was a concern with "the mind of primitive man" and in particular with its capacity for rational thought. In a sense, this concern represented the evolutionists' interest in primitive reasoning processes detached from the historicist context. In the second phase, which is still in process, there has been amove away from, and in part a reaction against, the subjectivist emphasis of the earlier work. Ideational expression is thought of as a public activity, rather like speech, and the structure of the symbolic materials, the "language," in whose terms the activity is conducted becomes the subject of investigation.

The first, subjectivist; phase was animated by a protracted wrangle between those who used the religious beliefs and practices of tribal peoples as eyidence to prove that there was a qualitative difference between the thought processes of primitives and those of civilized men and the anthropologists who considered such religious activity as evidence for the lack of any such differences. The great protagonist of the first school was the French philosopher LÈvy-Bruhl, whose theories of "ptelogical mentality" were as controversial within anthropology as they were popular outside it. According to LÈvy-Bruhl, the thought of primitives, as reflected in their religious ideas, is not governed by the immanent laws of Aristotelian logical reasoning, but by affectivity--by the vagrant flow of emotion and the dialectical principles of "mystical participation" and "mystical exclusion." [See LŠVY-BRUHL.]

The two most effective antagonists of LÈvy-Bruhl's theories concerning primitive religion were Radin and Malinowski. Radin, influenced by Boas' more general attacks on theories of "primitive mentality," sought to demonstrate that primitive religious thought reaches, on occasion, very high levels of logical articulation and philosophical sophistication and that tribal society contains, alongside the common run of unreflective doers ("men of action"), contemplative intellectuals ("men of thought") of boldness, subtlety, and originality. Malinowski attacked the problem on an even broader front. Using his ethnographic knowledge of the Trobriand Islanders, Malinowski argued that alongside their religious and magical notions (which he, too, regarded as mainly emotionally determined) the "savages" also had a rather well developed and, as far as it went, accurate empirical knowledge of gardening, navigation, housebuilding, canoe construction; and other useful arts. He further claimed that they were absolutely clear as to the distinction between these two sorts of reasoning, between mystical-magical and empirical-pragmatic thinking, and never confused them in actual practice. Of these two arguments, the former seems to be today nearly universally accepted and was perhaps never in fact really questioned. But with respect to the latter, serious doubts have arisen concerning whether the lines between "science," "magic," and "religion" are as simple and clear-cut in the minds of tribal peoples (or any peoples) as Malinowski, never one for shaded judgments, portrayed them. Nevertheless, between them, Radin and Malinowski rather definitively demolished the notion of a radical qualitative gap between the thought processes of primitive and civilized men. Indeed, toward the end of his life even LÈvy-Bruhl admitted that his arguments had been badly cast and might better have been phrased in terms of different modes of thinking common to all men. (In fact, Freud, with his contrast between primary and secondary thinking processes, had already made this distinction.) [See MALINOWSKI; RADIN.]

Thus, the debate about what does or does not go on in the heads of savages exhausted itself in generalities, and recent writers have turned to a concern with the symbolic forms, the conceptual resources, in terms of which primitives (and nonprimitives) think. The major figure in this work has been Claude LÈvi-Strauss, although this line of attack dates back to Durkheim and Mauss's influential 1903 essay in sociological Kantianism, Primitive Classification. The writings of E. E. Evans-Pritchard on Zande witchcraft, Benjamin Whorf on Hopi semantics, and Gregory Bateson on Iatmul ritual and, among nonanthropologists, works by Granet, Cassirer, and Piaget have directed attention to the study of symbolic formulation.

Symbolic systems. LÈvi-Strauss, whose rather highly wrought work is still very much in progress, is concerned with the systems of classification, the "homemade" taxonomies, employed by tribal peopIes to order the objects and events of their world (see LÈvi-Strauss 1958; 1962). In this, he follows in the footsteps of Durkheim and Mauss. But rather than looking, as they did, to social forms for the origins and explanations of such categorical systems, he looks to the symbolic structures in terms of which they are formulated, expressed, and applied. Myth and, in a slightly different way, rite are systems of signs that fix and organize abstract conceptual relationships in terms of concrete images 1 and thus make speculative thought possible. They permit the construction of a "science of the concrete"--the intellectual comprehension of the sensible world in terms of sensible phenomena which is no less rational, no less logical, no more affect-driven than the abstract science of the modern world. The objects rendered sacred are selected not because of their utilitarian qualities, nor because they are projections of repressed emotions, nor yet because they reflect the moral force of social organization ritualistically impressed upon the mind. Rather, they are selected because they permit the embodiment of general ideas in terms of the immediately perceptible realities--the turties, trees, springs, and caves--of everyday experience; not, as Levi-Strauss says, apropos of Radcliffe-Brown's view of toems, because they are "good to eat," but because they are "good to think."

This "goodness" exists inherentiy in sacred objects because they provide the raw materials for analogical reasoning. The relationships perceived among certain classes of natural objects or events can be analogized, taken as models of relationships--physical, social, psychological, or moral--obtaining between persons, groups, or other natural objects and events, Thus, for example, the natural distinctions perceived among totemic beings, their species differentiation, can serve as a conceptual framework for the comprehension, expression, and communication of social distinctions among exogamous clans--their structural differentiation. Thus, the sharp contrast between the wet and dry seasons (and the radical zoological and botanical changes associated with it) in certain regions of Australia is employed in the mythology of the native peoples. They have woven an elaborate origin myth around this natural phenomenon, one that involves a rainmaking python who drowned some incestuous sisters and their children because the women polluted his water hole with menstrual blood. This model expresses and economizes the contrasts between moral purity and impurity, maleness and femaleness, social superiority and inferiority, fertilizing agent (rain) and that which is fertilized (land), and even the distinction between "high" (initiate) and "low" (noninitiate) levels of cultural achievement.

Levi-Strauss contends that primitive religious systems are, like all symbolic systems, fundamentally communications systems. They are carriers of information in the technical Shannon-Weaver sense, and as such, the theory of information can be applied to them with the same validity as when applied to any physical systems, mechanical or biological, in which the transfer of information plays a central regulative role. Primitives, as all men, are quintessentially multichanneled emitters and receivers of messages. It is merely in the nature of the code they employ--one resting on analogies between "natural" and "cultural" distinctions and relationships--that they differ from ourselves. Where there is a distinguishing difference, it lies in the technically specialized codes of modern abstract thought, in which semantic properties are radically and deliberately severed from physical ones. Religion, primitive or modern, can be understood only as an integrated system of thought, logically sound, epistemologically valid, and as flourishing in France as in Tahiti.

It is far too early to evaluate LÈvi-Strauss's work with any assurance. It is frankly incomplete and explorative, and some parts of it (the celebration of information theory, for example) are wholly programmatic. But in focusing on symbol systems as conceptual models of social or other sorts of reality, he has clearly introduced into the anthropology of religion a line of inquiry which, having already become common in modern thought generally, can hardly fail to be productive when applied to tribal myth and ritual.

Whether his own particular formulation of this approach will prove to be the most enduring remains, however, rather more of a question. His rejection of emotional considerations and his neglect of normative or social factors in favor of an extreme intellectualism which cerebralizes religion and tends to reduce it yet again to a kind of undeveloped (or, as he puts it, "undomesticated") science are questionable. His nearly exclusive stress on those intellectual processes involved in classification, i.e., on taxonomic modes of thought (a reflex of his equally great reliance on totemic ideas, as type cases of primitive beliefs), at the expense of other, perhaps more common, and certainly more powerful styles of reasoning, is also doubtful. His conception of the critical process of symbolic formulation itself remains almost entirely undeveloped--hardly more than a sort of associationism dressed up with some concepts from modern linguistics. Partly as a result of this weakness and partly as a result of a tendency to consider symbol systems as entities functioning independently of the contextual factor, many of his specific interpretations of particular myths and rites seem as strained, arbitrary, and oversystematized as those of the most undisciplined psychoanalyst.

 But, for all this, Levi-Strauss has without doubt opened a vast territory for research and begun to explore it with theoretical brilliance and profound scholarship. And he is not alone. As the recent work of such diverse students as Evans-Pritchard, R. G. Lienhardt, W. E. H. Stanner, Victor W. Turner, Germaine Dieterlen, Meyer Fortes, Edmund R. Leach, Charles 0. Frake, Rodney Needham, and Susanne K. Langer demonstrates, the analysis of symbolic forms is becoming a major tradition in the study of primitive religion--in fact, of religion in general. Each of these writers has a somewhat different approach. But all seem to share the conviction that an attempt must be made to approach primitive religions for what they are: systems of ideas about the ultimate shape and substance of reality.

Whatever else religion does, it relates a view of the ultimate nature of reality to a set of ideas of how man is well advised, even obligated, to live. Religion tunes human actions to a view of the cosmic order and projects images of cosmic order onto the plane of human existence. In religious belief and practice a people's style of life, what Clyde Kluckhohn called their design for living, is rendered intellectually reasonable; it is shown to represent a way of life ideally adapted to the world "as it 'really' ('fundamentally,' 'ultimately') is." At the same time, the supposed basic structure of reality is rendered emotionally convincing because it is presented as an actual state of affairs uniquely accommodated to such a way of life and permitting it to flourish. Thus do received beliefs, essentially metaphysical, and established norms, essentially moral, confirm and support one another .

It is this mutual confirmation that religious symbols express and celebrate and that any scientific analysis of religion must somehow contrive to explain and clarify. In the development of such an analysis historical, psychological, sociological, and what has been called here semantic considerations are all necessary, but none is sufficient. A mature theory of religion will consist of an integration of them all into a conceptual system whose exact form remains to be discovered.

CLIFFORD GEERTZ

BIBLIOGRAPHY

BETTELHEIM, BRUNO 1954 Symbolic Wounds: Puberty Rites and the Envious Male. Glencoe, Ill.: Free Press.

CAMPBELL, JOSEPH 1949 The Hero With a Thousand Faces. New York: Pantheon.

DEVEREUX, GEORGE 1951 Reality and Dream: Psychotherapy of a Plains Indian. New York: International Universities Press.

ELIADE, MIRCEA (1949) 1958 Patterns in Comparative Religion. New York: Sheed & Ward. -> First published as Traite d'histoire des religions.

ERIKSON, ERIK H. (1950) 1964 Childhood and Society. 2d ed., rev. & enl. New York: Norton.

HALLOWELL, A. IRVING (1937-1954) 1955 Culture and Experience. Philadelphia: Univ. of Pennsylvania Press.

KARDINER, ABRAM 1945 The Psychological Frontiers of Society. New York: Columbia Univ. Press. -+ A paperback edition was published in 1963.

KLUCKHOHN, CLYDE 1944 Navaho Witchcraft. Harvard University, Peabody Museum of American Archaeology and Ethnology, Papers, Vol. 22, No.2. Cambridge, Mass. : The Museum.

LANG, ANDREW (1898) 1900 The Making of Religion. 2d ed. New York: Longmans.

LESSA, WILLIAM A.; and VOGT, EVON Z. (editors) (1958) 1965 Reader in Comparative Religion: An Anthropological Approach. 2d ed. New York: Harper. -> Contains a comprehensive bibliography.

LŠVI-STRAUSS, CLAUDE (1958) 1963 Structural Anthropology. New York: Basic Books. -> First published in French. See especially the chapters on myth and religion, pages 167-245.

LŠVI-STRAUSS, CLAUDE ( 1962) 1966 The Savage Mind. Univ. of Chicago Press. -> First published in French.

RADCLIFFE-BROWN, A. R. (1952) 1961 Structure and Function in Primitive Society: Essays and Addresses. London: Cohen & West; Glencoe, Ill. : Free Press.

R²HEIM, GŠZA 1950 Psychoanalysis and Anthropology: Culture, Personality and the Unconscious. New York: International Universities Press.

SPIER, LESLIE 1921 The Sun Dance of the Plains Indians: Its Development and Diffusion. American Museum of Natural History, Anthropological Papers, Vol. 16, part 7. New York: The Museum.

WHITING, JOHN W.; and CHILD, IRVIN L. 1953 Child Training and Personality: A Cross-cultural Study. New Haven: Yale Univ. Press. -> A paperback edition was published in 1962.

________________

(*)  I. Anthropological Study Clifford Geertz;

II. The Sociology of Religion Robert N. Bellah;

III. Psychological Study James E. Dittes.

Religion: anthropological study, in: Sills, David L. (ed.): International Encyclopedia of the Social Sciences (16 volume edition). New-York/N.Y./USA 1968: Macmillan & Co., vol. 13 (Psyc-Samp), pp. 398-406

source: paper version as indicated above.

Using this text is also subject to the general HyperGeertz-Copyright-regulations based on Austrian copyright-law (2001), which - in short - allow a personal, nonprofit & educational (all must apply) use of material stored in data bases, including a restricted redistribution of such material, if this is also for nonprofit purposes and restricted to a specific scientific community (both must apply), and if full and accurate attribution to the author, original source and date of publication, web location(s) or originating list(s) is given ("fair-use-restriction"). Any other use transgressing this restriction is subject to a direct agreement between a subsequent user and the holder of the original copyright(s) as indicated by the source(s). HyperGeertz@WorldCatalogue cannot be held responsible for any neglection of these regulations and will impose such a responsibility on any unlawful user.

Each copy of any part of a  transmission of a HyperGeertz-Text must therefore contain this same copyright notice as it appears on the screen or printed page of such transmission, including any specific copyright notice as indicated above by the original copyright holder and/ or the previous online source(s).