The concept religion has been defined in many different ways. Some definitions are very broad, and include beliefs and practices that many people do not regard as religious (see cosmology and ecology, for example). Others are very narrow, and exclude beliefs and behaviors that most people would consider to be religious. The resulting definitions of religion are often “monothetic,” meaning that they operate with the classical assumption that every instance that accurately fits a given concept will share some defining property. For this reason, such monothetic definitions of religion are often criticized for being overly restrictive.
Some scholars have tried to move away from such narrow, substantive definitions of religion. One common approach is to drop the requirement that a practice or set of beliefs be based on a particular kind of reality, and define religion functionally instead: as whatever system of practices unites people into a moral community, or as the activities that provide orientation in life. Emile Durkheim’s definition of religion is a good example. Others have gone even further, and suggested that the whole notion of religion is an artificial construction. These critics argue that the modern semantic expansion of the word “religion” went hand in hand with European colonialism, and that we should therefore stop treating it as if it corresponded to something real.
It is also possible to define religion in a social genus sense, as a set of beliefs and practices that appear in all human cultures. Various philosophers have suggested such a definition, including Edward Tylor, who proposed that the minimal definition of religion be belief in spiritual beings; and Paul Tillich, who defined religion as whatever is the dominant concern of a person’s life. Such functional, or polythetic, definitions of religion are not universally accepted, however.
Defining religion in this way carries some philosophical issues, particularly the question of whether such an abstract concept can have an essence. For instance, if Herbert were aware that some people did not believe in a supreme deity, he might have argued that such beliefs should not be considered to belong to religion at all. Yet, other people might argue that those beliefs and practices should be regarded as part of a religion because they are of similar forms. This is a problem that is likely to occur for any abstract concepts used to sort cultural types, such as “literature” or “democracy”.