is a failure, we may wish to excuse Kant on the grounds that he knew much less about the
full extent of religious diversity outside the West than we do. And we may think that the
increased knowledge of religious diversity we have acquired since Kant's time provides
us with a novel opportunity to work out an adequate definition of religion. As a result of
this expanded knowledge, we have many more examples than Kant did that can serve as
data against which to test a proposed definition of religion. Framing definitional
proposals and testing them against such data might be regarded as one of the main tasks
of a comparative philosophy of religion. In an essay that proposes an agenda for this kind
of philosophical work, Paul J. Griffiths counts the strategy of definition and classification
as one
end p.406
of “three kinds of intellectual enterprise that have title of some kind to be called
comparative philosophy of religion” (1997, 616).
We owe much of our increased knowledge of religious diversity to the work of
anthropologists in the field. There is something important at stake for anthropology in the
enterprise of defining religion; a definition will circumscribe the data that must be
covered by proposed empirical generalizations. So we would expect anthropologists with
a theoretical cast of mind to have a keen interest in defining religion. Rival definitions of
religion have been proposed by Clifford Geertz and Melford E. Spiro. Comparing their
definitions will enable us to explore some of the problems that confront the project of
formulating a definition of religion.
According to Geertz, “A religion is: (1) a system of symbols which acts to (2) establish
powerful, pervasive, and long-lasting moods and motivations in men by (3) formulating
conceptions of a general order of existence and (4) clothing these conceptions with such
an aura of factuality that (5) the moods and motivations seem uniquely realistic” (1966,
4). To facilitate understanding of his definition, Geertz offers some commentary on each
of its five parts. A symbol, on his view, is anything that serves as a vehicle for a
conception. Moods and motivations are dispositions to perform certain sorts of actions
and experience certain sorts of feelings in certain sorts of situations. Conceptions of a
general order of existence provide a framework of cosmic order to help deal with threats
of chaos at the limits of human analytical abilities, the limits of human endurance, and
the limits of human moral insight. Ritual is the chief instrument by which the conviction
that these conceptions are veridical is generated. And they alter the landscape presented
to common sense in such a way that the moods and motivations seem supremely
practical, the only sensible ones to have, given the way things really are.
An obvious objection to this definition is that it is too broad. Systems of symbols that
characterize secular ideologies such as Nazism satisfy the defining conditions it proposes
but are not religions. Hence, it does not provide a sufficient condition for being a religion.
One might respond to the objection by biting the bullet at this point and granting that
Nazism is a religion. Thus, for example, John Rawls claims that “Hitler's demonic
conception of the world was, in some perverse sense, religious” (1999, 20). Citing the
work of Saul Friedländer, Rawls attributes to Hitler a redemptive anti-Semitism, “born
from the fear of racial degeneration and the religious belief in redemption” (Friedländer
1997, 87). But many people will insist that Nazi symbols belong to a secular political