Religion is a concept with a very broad semantic range. Its sense has shifted over time and across cultures, and it seems likely that the word will continue to evolve, as have all abstract concepts that sort cultural types such as “literature” and “democracy”. The variety of practices now said to fall within this category raises two philosophical issues that are relevant for any analysis: (1) whether one can understand this social taxon as having an essence; and (2) if so, what that essence is.
Most of the discussions of religion have centered on these questions, and they have been dominated by the debate over monothetic and polythetic definitions. These arguments have tended to confuse the issue, as the debate over what it is that distinguishes “religion” from other forms of life has been inextricably linked with the debate over what it is that makes this term a “religion”.
The most familiar signpost in the anthropology of religion has been the work of Clifford Geertz (1926-2006). His definition defines religion as a system of symbols that establishes powerful and long-lasting moods and motivations by fashioning conceptions of a general order of nature and clothing them with such an aura of factuality that they seem uniquely realistic.
While this approach provides a useful frame for anthropological study of religion, it tends to obscure the fact that religion is not only a mental state or a set of beliefs, but also a way of living. Consequently, other scholars have sought to develop more robust definitions that incorporate a sense of what it is that makes something a religion rather than just a belief or set of beliefs.
For example, many researchers have adopted functional definitions of religion. Emile Durkheim defined religion as whatever system of practices unite a group of people into a moral community (whether or not these systems involve belief in unusual realities). Paul Tillich took a similar approach, defining religion as whatever dominant concern organizes a person’s values.
These definitions have been valuable, but they are not without problems. For example, they have the consequence of treating all forms of life that are organized as religions in some way the same. This is a logical error, since it is possible for some forms of life to have very different functions, and it is important to make a distinction between them.
Finally, functional definitions tend to overstate the role of religion in society. They assume that religious behavior is inherently beneficial for individuals, families, societies, and states, and can reduce the incidence of such problems as out-of-wedlock births, crime, delinquency, addictions, health problems, anxiety disorders, etc. For these reasons, it is essential that anthropologists be careful not to use functional definitions as a basis for judgments about whether a form of religion is “real” or not.