Resemblances Among Religions (continued)

So what might a family resemblance definition of religion look like and do such definitions also have shortcomings? Consider the following two examples, the first from the philosopher, William Alston, and the second from the historian of religions, Bruce Lincoln:

Alston defines religion by means of what he characterizes as "religion-making characteristics":

1. Belief in supernatural beings (gods).
2. A distinction between sacred and profane objects.
3. Ritual acts focused on sacred objects.
4. A moral code believed to be sanctioned by the gods.
5. Characteristically religious feelings (awe, sense of mystery, sense of guilt, adoration), which tend to be aroused in the presence of sacred objects and during the practice of ritual, and which are connected in idea with the gods.
6. Prayer and other forms of communication with gods.
7. A worldview or a general picture of the world as a whole and the place of the individual therein. This picture contains some specification of an overall purpose or point of the world and an indication of how the individual fits into it.
8. A more or less total organization of ones life based on the worldview.
9. A social group bound together by the above.*

And Bruce Lincoln defines religion as follows: "A proper definition must therefore be polythetic and flexible, allowing for wide variations and attending, at a minimum, to these four domains":

1. A discourse whose concerns transcend the human, temporal, and contingent, and thus claims for itself a similarly transcendent status....
2. A set of practices whose goal is to produce a proper world and/or proper human subjects, as defined by a religious discourse to which these practices are connected....
3. A community whose members construct their identity with reference to a religious discourse and its attendant practices....
4. An institution that regulates religious discourse, practice, and community, reproducing them over time and modifying them as necessary, while asserting their eternal validity and transcendent value.

Although hardly the same--Alston emphasizes rather traditional aspects associated with religions whereas Lincoln focuses on religion's role in establishing authority and order--both definitions provide a series of characteristics, or domains, that one would expect to find when one goes looking for a religion. It is held that the advantage of this way of defining is that it is thought to avoid reducing religion to some essential trait or function; for, as Lincoln--citing the work of Talal Asad--warns, "[a]ny definition that privileges one aspect, dimension, or component of the religious necessarily fails, for in so doing it normalizes some specific traditions (or tendencies therein), while simultaneously dismissing or stigmatizing others" (5).*

But, just as with the essentialist and the functionalist approaches, a few criticisms are possible here as well. First, family resemblance approaches often appear to be circular definitions (what we might call tautological or repetitive); religion, for Alston, is defined in light of such words as "sacred" and "profane"--words that are themselves generally defined in light of religion, suggesting that we already need to know something about religion in order to understand the very words used to define religion--hardly a helpful approach to defining something. For instance, Alston's fifth point defines religions as having "characteristically religious feelings," and Lincoln's definition defines religions as being comprised of "religious discourses"--defining the noun "religion" by means of qualifiers that employ the adjective "religious" likely does not help us too much.

A second, related difficulty involves the role played by ones prototype; when one reads a family resemblance definition, it often seems to be no more than a description of the thing one is defining--for example, it is not difficult to see Christianity lurking behind Alston's definition. It is as if the scholar was looking at a religion, describing its main features, and then generalizing from these to establish a definition of what one would expect other religions to possess. As Benson Saler makes clear, we have no choice but to employ prototypes--after all, we never experience the concept of "table," but, rather, we seem to experience actual tables, one or more of which we seem to use as the model for identifying those other things we classify as tables. But how did we know that the thing we had first encountered--that which became our prototype--was a religion? How, for instance, did Alston know that Christianity was a religion and that its features could also be found elsewhere? And what if something else had constituted his prototype? What then would his definition look like and how useful would it be to us in carrying out cross-cultural research?

This is the criticism developed in some detail by the British scholar of religion, Tim Fitzgerald*: despite the fact that the family resemblance approach is portrayed as more inclusive and therefore capable of recognizing the variety of actual religions, as with Penner's critique of functionalism, there is an odd sense in which an essentialist definition yet remains at the very core of the family of traits by which other religions are identified. For Alston's definition seems to be saying: Christianity is a religion; Christianity has these various components; therefore anything else that is a religion will also have some or all of these components. In the midst of all this, the essentially religious identity of the thing we call Christianity is simply assumed and thereby reinforced. Why it gets to count as a religion--rather than, say, a mass political movement--is never explored.

The problem is that, as already identified with the insider/outsider problem, such an approach merely takes what one group of participants already understand as a religion (what we might call their folk knowledge) and extends that (often in a rather vague manner) to other cases, as if generalization was all that was required to turn folk knowledge into a scholarly theory. In so doing, the scholar implicitly authorizes one among many (potentially competing) emic perspectives.

< Previous
Next >