From the remarkable evolutionist conceptions at the turn of the 19th to the 20th century, religious ethnology was born, based on what Albert Marie Bros, in L’Ethnologie religieuse: introduction à l’étude comparée des religions primitives, calls “religious facts observable in primitive peoples” (1923: 15). This evolutionist view of societies remained for quite some time. Lewis Morgan, regarded as one of the pioneers of religious anthropology, in La Société Archaique, 1877, states that religious beliefs are linked to the domain of the imagination and affectivity and are therefore marked by the principle of uncertainty, which is why “primitive” religions are “grotesque” and “unintelligible”.
Despite having freed itself from the methodological and conceptual shackles of evolutionism – a determining obstacle to the construction of an equitable vision of cultures – the discipline remains, even if already under the designation of Anthropology of Religions, devoted to the study of religious forms in primitive societies (Radcliffe-Brown 1945). Nevertheless, British functionalism (with Malinowski, Radcliffe-Brown and Evans-Pritchard) would mark the discipline by establishing the central place of systematic and rigorous field studies, changing the focus from the origin of religion and its contribution to the formation of human societies to the social functions of religion. As Radcliffe-Brown would later affirm, in response to Lewis Morgan, “the social function of religion is independent of its truth or falsity, and religions which we find false or even absurd and repugnant, as of savage tribes, may play an important and effective role in the social mechanism” (1945: 231). Functionalism, despite its relevance in the formation of an ethnographic model as the basis of analysis, failed to focus, above all, on the social function of religion, neglecting other aspects beyond its dimension of social production.
At the turn of the post-war period, Lévi-Strauss’ structuralism centred the debate on religion on the terrain of the psychological and social aspects of human symbolism. Unfortunately, it reduced the significance of religion within anthropology by arguing that the importance of its study boils down to inscribing it within the larger framework of understanding more general human thought and its “social translations”, i.e., rites, institutions, etc., from which to understand cultural universals.
The 1950s and 1960s marked the exhaustion of interest in the study of religions, leading to a discourse of “exhaustion” of the subject. Claude Lévi-Strauss even claimed (1974) that anthropology seemed to have moved away from religious facts. Evans-Pritchard (1965), argues that religion no longer has the same (central) importance it had for anthropology in its early days, justifying this disinterest with an accompaniment of social changes, with anthropology following society’s disinterest in religion.
However, Roger Bastide (1978), taking up the approach of Claude Lévi-Strauss (1974b), in an entry in the Encyclopaedia Universalis, considers that the principle of religious ethnology is to study the diversity of beliefs or practices of ethnic groups in relation to each other, with the knowledge it produces “remaining located in a cultural area, as a type of religion, animist, for example, or polytheistic”, whereas the principle of the anthropology of religion is to discover the general laws of man as a religious subject. This means, therefore, observing religion from empirical materials in order to produce theoretical models capable of explaining human nature from its religious nature.
As a result of the geography of the discipline, i.e. the fact that it was formulated in the West, between Europe and the United States, the illusion is established that there is only one anthropology of religions. This is why Luís Batalha writes, “there is American anthropology, another British, another French, another Chinese, another Japanese, another German, etc. Or, perhaps, in each of those countries there are even several anthropologies” (1998: 320). This is understandable insofar as, as Kishimoto (1961) states, each cultural context produces its own definition of “religion,” which demands localized methodologies and conceptualizations that are globalized in the broad sense of generating tools for the discipline rather than localized globalized conceptions that tend to sin by their inability to fit the realities under study.