Readers might recall that, at the close of the discussion on essentialism, you were told that a light switch could either be on or off, and never partially on or partially off. So too with essentialist definitions. In fact, the presumption that there is a distinct insider perspective as opposed to an outsider view--as opposed to seeing insides and outsides as continually changing and continually contested, all depending on where you stand--is itself a product of an essentialist viewpoint. Although the light switch imagery works to communicate the "either/or" nature of this approach to definition (whether defining who gets to count as a "patriotic citizen" or a "faithful" Christian, Jew, Hindu, Muslim, etc.), surely someone must have thought, "What about a dimmer switch?" Good point. In the study of definitons the dimmer switch example would be called the family resemblance approach.

The family resemblance approach to definition--sometimes called polythetic definitions--is thought by some to enable them to steer a middle path between essentialist and functionalist approaches. Although useful contemporary examples of this approach would be the "dimension theory of religion" advanced by the phenomenologist of religion, Ninian Smart, or Clifford Geertz's definition of religion as "moods and motivations," this approach is credited to the philosopher, Ludwig Wittgenstein (1889-1951), who, in trying to argue that there is no one defining characteristic that makes something "language," asked his readers to stop and consider how it is that they actually go about the activities of classifying, sorting, and distinguishing. If they did this, he suggested, if they actually considered what it was they did when they went about these activities, they would see that all members of a particular group more or less share a series of traits or characteristics of relevance to the classifier. To put it another way, all members of a common group overlap to varying degrees and in differing respects, just as no two members of a family are exactly alike (even identical twins); instead, they more or less share a delimited series of characteristics (such as name, hair color, temperament, height, favorite foods, blood type, etc.). Further, despite their best efforts to portray themselves as authoritative, no family member constitutes the definitive instance of the group--rather, all members share in the identity, to varying degrees. Group membership, Wittgenstein argued, is never a matter of yes or no (as in the essentialist approach) but always a matter of degree, a matter of "more or less."

To make his point, Wittgenstein uses the example of games, asking his readers to consider the variety of activities we commonly know by this name, challenging them to come up with the one essential trait that all of these activities share. "[T]he result of this examination," he conlcudes, "is: we see a complicated network of similarities overlapping and criss-crossing.... I can think of no better expression to characterize these similarities than 'family resemblances'; for the various resemblances between members of a family: build, features, color of eyes, gait, temperament, etc., etc., overlap and criss-cross in the same way. And I shall say: 'games' form a family."*

The acts of classification and definition, for Wittgenstein, were therefore activities of selection, of choice, not of merely passively recognizing qualities thought to reside in some object that catches our attention (such as the essence of beauty, of justice, or of religion); instead, something was only "more or less" this or that (hence the usefulness of the image of the dimmer switch on the ceiling light). So, if only two traits are shared in common between two objects, say those two things we call an apple and an orange (they are both round and can be eaten), we have no choice but to make a judgement call regarding whether these two things are the same. But are only two traits sufficient? If not, then what about three? Or four? Case in point: is Buddhism a religion, despite the lack of any belief in a god in some of its forms? Or what do we make of some Evangelical Protestants who assert that Roman Catholics are not Christians? They seem to be saying that their Roman Catholic counterparts are not part of the family known as "Christian," likely due to the latter group being judged by the former as not sharing a sufficient number of traits.

If we follow Wittgenstein, then it seems to fall to those who develop and use classification systems--such as those who attempt to define religion--not only to have what a recent anthropologist, Benson Saler, has termed a "prototypical definition," but also to be prepared to make judgment calls when a cultural artifact meets so few of their prototype's characteristics that it is questionable whether the artifact can productively be called a religion. That the prototype we use when we set about defining religion in this manner is often confused with being the ideal case or the norm is certainly a trouble of which scholars ought to be aware if they wish to avoid making but one social movement the norm. It is just such a confusion between prototype and ideal that has sometimes led European and North American scholars to use certain types of Christianity as the authoritative standard by which they measure the quality and legitimacy of other social movements.*

Contrary to essentialist and the functionalist scholars passively recognizing either some core feature or purpose/need served by a religion, Wittgensteinian scholars of religion see themselves as actively constituting a cultural practice as religious insomuch as it does or does not match their prototype. Moreover, they are prepared to adjust their prototype for it is merely a tool and a starting place. That the family resemblance definition widens in the case of more liberal scholars (either politically or theologically), and narrows in the case of those who are more conservative, should not go unnoticed.