Anthropology and the Bible

In an essay on atonement in the Jewish Studies Quarterly from the early 1990s, Mary Douglas explains “the social origins of ideas about defilement in general”:

Living in a community, people find themselves wanting to keep apart from some dissimilar others and to be included within sets of similars. To justify their wish to be separate they tend to invoke theories of contagion, religious or secular, aesthetic or scientific. Religious purity generally emerges as an ordering principle which a community has spontaneously evolved to sort and sanction its social relations. In the process the ministers of religion may try to monitor the effects of social inclusions and exclusions. They may also have a pastoral interest in softening the harsh divisions which the laity would like to enforce. The result may be some codification of impurity (111).

Well, no, that’s not how it works. It’s not even internally coherent. Douglas’s starting point—that people like similars and dislike dissimilars—presupposes that there is a classification system already in place. Classification into likes and unlikes doesn’t “emerge.” It’s always already there.

Perhaps her point is that this primordial classification system isn’t yet a purity system, but we’d need to see some evidence for that. After all, social classification systems often play off concerns of purity and impurity of one sort of another. Do we know of systems that are completely innocent of purity? Even if we do, how do we know that these are more basic, original, fundamental than purify classifications. And even if the original classification into like/dislike isn’t explicitly framed in purity terms, it functions as a purity system: It designates who gets included and who gets excluded. Purity, in short, is always already there. It is constitutive of the community, not merely a subsequent “emergent” property.

Perhaps her point is that purity systems are not “codified” at the beginning. That may be, but it would only suggest that purity systems might exist in the absence of formal “codification.”

Douglas’s explanation is a classic “sociological” explanation, and exhibits as clearly as we could hope the question-begged character of sociological explanation. “Social relations” explain other things only when we have stripped social relations of the things that make them possible; but such denuded social relations cannot cause anything. 

Better to accept the aporia: Social relations are not fundamental, cultural classification epiphenomenal; social relations exist only by virtue of cultural classification, and so social relations cannot provide an explanation for those classifications.