Primitive Religion

The evolutionary character of theories of primitive religion is present in the sociological literature from the beginning. It is evident, for example, in the writings of the so called founding father of sociology, Auguste Comte (1798-1857), who believed that religion originated in fetishism or the worship of inanimate things, then developed into polytheism which in turn developed into monotheism (Comte 1853). The view that religion evolved from polytheism to monotheism is, of course, much older than the formal beginnings of sociology and anthropology as academic disciplines. It is present in the Scottish philosopher David Hume’s The Natural History of Religion (1759). The nineteenth century theorists – they would today be classified as armchair anthropologists and sociologists – most closely associated with the construction of the concept of primitive religion were less concerned about religion per se and its nature and more about finding proof with which to discredit the so called higher religions and in particular Christianity. Their intention was to discover the origins of primitive religion or religion in its most basic or elementary form in order to show that it was profoundly mistaken and arose from ignorance or some emotional need and that the so called higher religions which derived from such erroneous ideas and behavior did not therefore merit the assent and commitment of rational and emotionally mature and balanced people. In fact, religion held society back.

Among the better known of the nineteenth century theorists of primitive religion was the Orientalist and authority on mythology Max Muller (1823-1900) who claimed (1893) that religion was grounded in an intuitive sense of the divine which everyone possessed and which was awakened by the wonder and power of nature. Religion began in this way as metaphor and symbol and eventually the natural objects that evoked thoughts and feelings of the infinite were personified as gods in their own right. Others like the sociologist Herbert Spencer (1820-1903), who defined progress and the development of the heterogeneous out of the homogeneous, maintained that religion began with ancestor worship, or more precisely with belief in the continued existence of the souls or ghosts of remote ancestors which in time were deified (Spencer 1901-7). The process of deifying ancestors exists in many societies, including the Yoruba society of southwestern Nigeria, but there is no evidence that the religion of these people began in this way. Spencer then asserts that the notion of soul or ghost developed into that of god or divinity. Thus, it is assumed without any supporting evidence that the notion of ghost is the first ever notion of divinity devised by humans.

The anthropologist Edward Tylor (1832­1917), whose main interest was in the evolution of society and its institutions, opposed theories of religion that reduced the phenomenon to the psychological immaturity of early human beings, and sought to base his own theory on reason, hence the description of his approach as intellectualist. Like Spencer, Tylor also traces the origins of religion to the development of the idea of the soul, which he contended originated in dreams. So called primitive people were alleged to believe that the soul left the body during sleep and actually experienced what they had been dreaming about while asleep, and inferred that this behavior would continue after death. Hence also the idea of immortality. Tylor went further, however, claiming that such people not only personified all other beings like themselves, but also natural phenomena, and endowed the latter also with souls. Tylor’s use of the term soul in preference to ghost or spirit led to his being regarded as the founder of the animist theory of the origins of religion. As for his approach to the question of religion’s origins, this is described as intellectualist principally for the reason that it insists on offering a rational basis for the belief in the soul, which it argues did not derive from fear or superstition or psychological immaturity on the part of the primitive, but resulted instead from a deductive and logical process, even if the reasoning was mistaken. While, as Evans Prichard (1965: 25) has pointed out, it is possible that the notion of the soul developed in this way, there is no evidence to say it did.

Tylor’s successor at Oxford, the barrister and anthropologist Robert Marett (1866-1943), is credited with being the father of the pre-animist theory of religion (Marett 2001). He argued that in terms of its beginnings, a rudimentary religion, a form of supernaturalism consisting of awe of the mysterious, existed prior to ideas of soul, ghost, and spirit. It was this attitude of mind that provided religion with its raw material and could exist apart from animism and indeed might well have been the basis for animistic beliefs. This theory was based on Marett’s interpretation of studies on Melanesian religious life and in particular its concept of mana. Religion, he insisted, was something that was lived or acted out: it helped the primitive to live, it provided the necessary assurance of being in touch with a higher power, and it offered hope and induced fear. Marett reduces both magic and religion to psychological states and suggests that they function most effectively in situations of emotional stress.

Sir James Frazer (1854-1941), author of the monumental and widely known work on primitive superstitions The Golden Bough (1920), also differed from Tylor in propounding a developmental theory of religion in which he introduced a pre-religious stage in the form of a magical phase. His theory bears a striking resemblance (in the way it describes religion’s evolution from magic to religion and religion to science) to Auguste Comte’s three phases of intellectual development: the theological, meta physical, and positive. Relying on ethnographical data of poor quality, Frazer, the last of the great armchair evolutionists, was to claim that magic characterized simple societies, and as they became more complex they also became less  superstitious  and  more  scientific and rational, a line of argument for which there is no proof worthy of the name. Moreover, like Lucien Levy Bruhl, he wrongly viewed magic as an elementary form of modern science, but differed from the latter by mistaking ideal for real connections between things.

Intellectualist theories of the kind advanced by Tylor and Frazer were in turn opposed by thinkers concerned to locate the origins of “primitive” religion in social structure rather than logic and emotion. The best known and most influential theorists to adopt this approach were Robertson Smith (1846-94) and Emile Durkheim (1858-1917). According to Smith, the clan cult or totemism was the earliest and most elementary form of religion and was best accounted for and understood by reference to its social character, an idea for which he was indebted to Fustel de Coulange (1830-99) and in particular to his major work The Ancient City (1980), a work that also influenced Durkheim.

Durkheim was unimpressed by the kind of animist, intellectualist, emotionalist, and action based theories of primitive religion advanced by Spencer, Tylor, Frazer, Marett, and others, and indeed by any theory that suggested religion was false or an illusion. Primitive religion was the earliest in the sense of simplest form of religion. It was the form that was practiced when human society was passing through its simplest form. Moreover, far from being an illusion, it was as much a thing or social fact, in the sense that it was a reality external to the individual, as any other thing or social fact. It enjoyed the same degree of reality as any material thing. It rested on a permanent underlying reality that could be uncovered if studied objectively: that is, society. Religion belonged, Durkheim argued, to the class of social facts that includes established beliefs and practices that were the product of the collectivity and/or a group within society. By treating religion in this way Durkheim believed he had given it a foothold in reality and made it accessible to scientific analysis. As to its earliest, most elementary or primitive form, Durkheim maintained that this was to be found in totemism, an idea he borrowed from Robertson Smith. He used the available ethnographic material on the Australian aboriginals and in particular the Arunta to demonstrate his thesis, published as the The ElementaryForms of the Religious Life (1915). What bound clan members together, he argued in this study, was that each and every one of the members had the same totem, which in several senses was regarded as sacred, including the sense that it symbolized the totemic principle in the form of an impersonal religious force which he referred to as mana.

In Durkheim’s evolutionist understanding of it, religion would regress – regression was also a notion used by Spencer in respect to civilizations and their institutions – as social institutions developed. As an example he pointed to religion’s loss of control in modern society over men’s (sic) minds compared with primitive society (where he contended religion dominated everything) and went on to argue that to become intensely religious again it would be necessary for society to return to the beginning. In his own words religion could not regain its domination over people’s minds as in primitive society ”unless the great societies crumble and we return to small social groups of long ago, that is unless humanity returns to its starting point, religion will never be able to exert deep or wide sway over consciousness” (Durkheim 1952: 430).

Although   Durkheim’s   highly speculative account of the origins of religion has been heavily criticized on methodological, logical, ethnographic, and other grounds, it nevertheless contains many valuable insights and remains one of the most thought provoking and influential studies of the sociological character of religion, particularly that part of it that treats the purposes and functions of ritual.

An almost exact contemporary of Durkheim, the German sociologist Max Weber (1864-1920), regarded as one of the founders of modern sociology and most widely known for his thesis linking Protestantism to the rise of modern capitalism and his comparative sociology based on the principle of verstehen or empathic understanding, also began his analysis of religion from an evolutionist perspective by looking at its most elementary forms. These, he believed, were to be found in the religions of tribal societies in which, he contended, questionably, people were so preoccupied with meeting their everyday needs that they had little alternative but to practice magic rather than religion. Such people were largely concerned with attempting to manipulate and coerce the gods, whom they conceived as being part of this world and immanent, rather than as in religion, which has a more transcendental conception of their status, with worshipping them. Thus, according to Weber (1965), elementary or primitive religion tends toward the magical and out of this emerges religious conceptions, as human society evolves. Magic begins to develop into religion when the extraordinary qualities or mystical powers (referred to as mana by Durkheim, Marett, and others, and as charisma by Weber) that are believed to inhere in objects are attributed less to the objects themselves and increasingly to a reality behind them, as it were, such as a soul, spirit, or demon. Thus, once the source of this power came to be perceived as being outside the material world, and the spirits behind it came to be regarded as being more and more removed from this world, the way was open, Weber maintained, for ethical rationalization to begin to dominate religious attitudes. At this stage of religious evolution the gods become increasingly bound up with ethical considerations, and values and principles replace self interest as the core concerns of religion. Thus, Weber suggests, religion only truly begins with the appearance of ethical rationalization, and coterminous with this development is the demise of the central role of the magician and the rise of a priesthood that concerns itself with intellectual matters such as the formulation of doctrinal and ethical systems.

Primitive Religion as Monotheistic

Andrew Lang (1844-1912) and the Catholic priest Wilhelm Schmidt (1868-1954) challenged the traditional understanding of the content of primitive religious belief. The latter contended that primitive monotheism predated the technological advances some believe led to it. Schmidt, founder of the journal Anthropos (1906), sought to establish a chronology of primitive cultures from circumstantial evidence. Like Lang, he maintained that people who were on the lowest rung of the ladder of social and cultural development were monotheists. Schmidt, though he was concerned to discredit the evolutionary kind of ethnography and ethnology prevalent in his day, did not escape their influence. He also believed he had been able to identify the ethnologically oldest people whom he claimed belonged to the most primitive culture. However, this was a culture in which totemism, fetishism, magic, and belief in ghosts or spirits were absent. Instead, these were a people who, by observation and inference, had come to believe in one, eternal, all knowing, all powerful, beneficent God who satisfied all their desires and wants. Once again, no strong evidence is supplied in support of this thesis on the origins of religion and primitive thinking about God and the supernatural order. As a theory, however, it was not without influence, shaping as it did the thinking of some missionaries working in various part of the world about the beliefs of so called primitive people.

Lévy Bruhl and the Prelogical Primitives

The notion of primitive religion is closely linked not only to that of primitive society but also to the concept of primitive mentality, especially as it was developed by the French philosopher Lucien Levy Bruhl (1857-1939), who in his La Mentalité primitive (1922) set about describing its attributes. He was not convinced by the theories of Tylor and Frazer, which assumed that the only difference between primitive and more advanced people was not one of intellect but of ignorance, the former being more ignorant than the latter. In terms of intellect, both were the same.

Levy Bruhl saw things differently. Starting from the assumption that each type of society has its own distinctive mentality he contended that, broadly speaking, there were two types of society: primitive and civilized. He then proceeded to argue that there existed two types of mentality: primitive and civilized. These modes of thought, Levy Bruhl maintained, were collective in the sense of being all pervasive, taken for granted ways of thinking to which there were no exceptions. The primitive mode he characterized as essentially prelogical and/or mystical, and the civilized as logical. The primitive mentality is prelogical and/or mystical in the sense that primitive people do not make the distinctions between the natural and the super natural order that so called modern, civilized people make. The latter are capable thus of seeking the causes of things in natural processes and explaining them scientifically. Primitive thought is guided by what he termed the law of participation and does not concern itself with contradiction or the rule of logic, but is rather held together by links or connections that do not conform to the logical thought patterns of more advanced peoples.

Levy Bruhl is not suggesting by this that primitives are innately incapable of reasoning or thinking logically or that they are a logical, illogical, or anti logical. He is describing the categories in which they reason, their collective representations and the mystical realities in which they move and which shape their thought. It is for this reason that primitives reason incorrectly and not because, as Tylor and Frazer suggest, their logical processes were mistaken. However, like the rest of the above mentioned theorizing about primitive religion, there is no evidence to support Levy Bruhl’s argument that ”primitive thought” differs in quality from ”civilized thought.” Neither is there any basis for classifying en bloc whole peoples who differ so much from each other socially, culturally, and economically as either primitives or civilized, nor can it be automatically assumed that there is a contradiction between a scientific, causal explanation and a mystical one, or that because something is thought of in mystical terms it cannot also be understood scientifically (and the converse).

Modern Usage

While realizing its limitations and controversial character, some scholars have nevertheless offered a robust defense of their use of the term primitive religion. Douglas (1966: 81-2) considers her use of the term in the more general context of a discussion of the distinguishing features of the notion of primitive worldview, which she suggests be characterized by non differentiation. She also describes this world view as subjective and personal, one in which different modes of existence are confused, and one in which the limitations of man’s (sic) being are not known. It is anthropomorphic and resembles, Douglas maintains, a pre-Copernican worldview. The belief of the !Kung Bushmen of the Kalahari Desert in N!ow provides a good example in her view of belief in anthropocentric powers.

Regarding the question of the use of the term primitive and whether it should be abandoned, Douglas expresses the hope that its use will not be discontinued, on the grounds that if this concept can be given a valid meaning in art and technology, and possibly economics also, then presumably it can also be given a similar sense when used of a certain kind of culture. While accepting that it can have a pejorative sense when used of religious beliefs, she is not convinced that because of this the label should be abandoned and suggests that to do so could well amount to an inverted form of superiority.

Evans Pritchard (1965) defends his use of the term primitive on the following grounds: that he uses it in a value free sense, is obliged to use the language of those he is critiquing, that it is too firmly established to be dropped, and that etymologically it is unobjectionable. He also points out that the term primitive can be used in both a chronological sense (as it is used by him) and a logical sense, both of which should be kept distinct.

Others do not go to such lengths to defend their use of the term. Bellah (1964) in his treatise on religious evolution – in which one of the categories used is that of primitive religion – is more anxious to make clear what he means by evolution in this case. Evolving religion as he understands it is a symbol system that develops from a compact or primitive form or state to a more differentiated or modern one, the latter not necessarily being better or truer or more beautiful than the former. Thus, religious evolution involves a process of increasing differentiation and complexity of organization. The outcome of this process is to endow the particular system in question, in this case religion, with the greater capacity to adapt to its environment, thus becoming more autonomous in relation to that environment than was the case in its less complex stage. This is the underlying assumption on which Bellah constructs his evolutionary scheme of religion, which begins with primitive religion and evolves into archaic religion, followed by historic religion,   early  modern  religion,   and modern religion. These types are not seen as completely distinct, nor does Bellah suggest that this kind of evolution is either inevitable or irreversible.

Confining comment to Bellah’s primitive religion, this owes much to Levy Bruhl’s (1922) notions of the mythical world and to research on the mythical world of Australian religion – in particular, anthropological interpretations of the core concept of Dreaming. Bellah’s world of primitive religion is a world in which the actual and mythical worlds are closely related to each other, and one which -in terms of its organization – is extremely fluid. He characterizes primitive religion itself in language reminiscent of Marett: it is given over not to worship or sacrifice, but is characterized by identification, participation, or acting out. It is ritual based. As to primitive religion’s social implications, Bellah suggests in line with Durkheim that they consist in reinforcing social solidarity and in socializing the young into the norms of tribal society. These goals and the fluidity and flexibility of primitive religion militate against any kind of radical change.

Stark and Bainbridge (1987) construct a general theory of religion that is also evolutionist after a fashion. Again, it suggests that religion develops from more compact to more differentiated forms. It is argued that when societies reach a certain size and level of complexity specific social organizations emerge, including religious organizations. While in its early stages religion was closely related to magic, Stark and Bainbridge suggest that as society becomes more complex they become increasingly differentiated in terms of specialists and organizations. With increasing complexity the idea of gods emerged, who, though considered to be supernatural beings, were believed to share with humans the attributes of consciousness and desire and come to be seen as supernatural exchange partners who bestow upon humans rewards in return for the fulfillment of certain obligations. As society became even more complex the number of gods decreased – in other words polytheism has tended to give way to monotheism – and religious specialists have emerged to provide explanations of how rewards can be obtained, or if not the actual rewards themselves then how general “compensators” of a supernatural kind can be guaranteed and the costs involved assessed.

Prior to all of this in a simpler, less differentiated world people had resort to magic, which in the opinion of Stark and Bainbridge differs from religion in that it offers very specific compensators that are easily disconfirmed. Therefore, unlike religion, it does not have the capacity to foster long term exchange relationships and as a consequence does not develop into an organization such as a church.


The reasons behind the largely pointless search for the origins of religion in the nineteenth century and first part of the twentieth century have been outlined above. We have also seen that well into the late twentieth century some scholars – in their attempt to construct a refined evolutionist theory of the development of religion – have continued to apply the term primitive to what they consider to be its earliest form. Despite the caveats and qualifications offered for the continuing use of this term, the question whether it should be retained remains. It is such a highly controversial term as to suggest that there is a strong need for a replacement, such as traditional or early forms of religion, although these two terms also have their limitations. While the label early forms is virtually free of any pejorative meanings, the limitations associated with the term primitive are especially evident in relation to the term traditional, which can convey the sense of stagnation and imply that a society where this type of religion is the norm is a society that is unchanging and lacking in dynamism and creativity. Although the term traditional is hardly value free either and raises serious methodological problems, it is less pejorative than the concept of primitive and of greater value analytically. As Evans Pritichard (1965) pointed out, the term primitive as used of religion creates only confusion and is likely most of the time to generate in the mind of the reader a negative stereotype about the religious beliefs, worldview, and practices to which it is applied and by implication about those who adhere to them.


  1. Bellah, R. (1964) Religious Evolution. American Sociological Review 29: 358 74.
  2. Clarke, P. B. & Byrne, P. A. (1993) Religion Defined and Explained. Macmillan, Basingstoke.
  3. Comte, A. (1853) The Positive Philosophy, 2 vols. Trans. H. Martineau. Triibner, London.
  4. Douglas, M. (1966) Purity and Danger. Routledge & Kegan Paul, London.
  5. Douglas,   (1973) Natural Symbols. Penguin, London.
  6. Durkheim, IE. (1915) The Elementary Forms of the Religious Life. George Allen & Unwin, London.
  7. Dürkheim, E. (1952 [1897]) Suicide: A Study in Routledge & Kegan Paul, London.
  8. Evans-Pritchard, E. E. (1965) Theories of Primitive Religion. Clarendon Press, Oxford.
  9. Frazer, J. (1920) The Golden Bough. Macmillan, London.
  10. Fustel de Coulange, N.-D. (1980) The Ancient City. Johns Hopkins University Press, Baltimore.
  11. Hume, D. (1977) The Natural History of Religion and Dialogues Concerning Natural Religion. A. W. Culver & J. V. Price. Clarendon Press, Oxford.
  12. Levy-Bruhl, L. (1922) La Mentalité primitive. Alcan, Paris.
  13. Levy-Bruhl, L. (1926) How Natives Think. George Allen & Unwin, London.
  14. Marett, R. (2001 [1909]) The Threshold of Religion. Elibron, London.
  15. Müller, M. F. (1893) Introduction to the Science of Religion. Longman, London.
  16. Runciman, W. G. (Ed.) (1978) Weber: Selections in Translation. Cambridge University Press, Cam­bridge.
  17. Schmidt, W. (1931) The Origin and Growth of Religion. Cooper Square Publishers, New York.
  18. Spencer, H. (1901 7) Principles of Sociology.D. Appleton, New York.
  19. Stark, R. & Bainbridge, W. S. (1987) A Theory of Religion. Lang, New York.
  20. Weber, M. (1965) The Sociology of Religion. Methuen, London.

Back to Sociology of Religion