Doubt as methodology and object in the phenomenology of religion

Abstract

The Holy Grail in the phenomenology of religion (and, to a lesser extent, the sociology of religion) is a definition of religion that actually works, but, so far, this seems to have been elusive. Classical definitions of religion—substantive (e.g. Tylor) and functionalist (e.g. Durkheim)—fail, in part because they attempt to be in three places at once, as it were: they attempt to distinguish religion from non-religion; they attempt to capture what religions have in common; and they attempt to grasp the 'heart', or 'core', of religion. Consequently, family resemblance definitions of religion replace certainty and precision for its own sake with a more pragmatic and heuristic approach, embracing doubt and putting forward definitions that give us a better understanding (Verstehen) of religion. In this paper, I summarise some 'new' definitions of religion that take this approach, before proposing and defending another one, defining religion as non-propositional and 'apophatic', thus accepting that doubt is central to religion itself, as well as to the analysis of religion.