A longer paper that grew out of this essay can be found here.
What might be an appropriate and useful way to talk about something like “religion” across times, locales, and disciplines? This is a question shared by classical approaches to the study of religion like those of Max Weber, Émile Durkheim, and Max Müller, and more recent scholars like Clifford Geertz, Talal Asad, Tomoko Masuzawa, and Jonathan Z. Smith. What Smith, Asad, Geertz, and Masuzawa share in common is a sense of the inadequacy of the classical approaches. Asad, Smith, and Masuzawa in particular offer as alternatives paying more attention to reflexivity, disciplinary history, and the relation between power and knowledge. In this post, I want to offer some comments and a question based primarily on Asad’s chapter “The Construction of Religion as an Anthropological Category” in his seminal book Genealogies of Religion: Discipline and Reasons of Power in Christianity and Islam (Baltimore: Johns Hopkins University Press, 1993).
Before looking at Asad’s proposal in more detail, it is worth commenting on Geertz’s work, focusing on his chapter “Religion as a Cultural System” in The Interpretation of Cultures (New York: Basic Books, 1973). Geertz has the advantage that he proposes his own clear definition of religion: essentially, a symbolic system that is part of a culture, which he understands as a semiotically organized system of meaning. Geertz’s definition is also the springboard for Asad’s critique. Before moving there, I want to comment on one of Geertz’s central ideas: the symbol. Funnily enough, “symbol” is itself a fundamentally religious term. In 250 CE, Cyprian, Bishop of Carthage, used the Latin symbolum to refer to the Apostles’ Creed, which was the “mark” or “sign” that distinguished the Christian from the heathen. (Cyprian was adapting a classical Greek word which had meant “any token serving as proof of identity” (LSJ), also meaning “contract, covenant, bond, treaty”.) “Creed” itself comes from the first word of the Latin version, credo; the famous Nicene Creed is still referred to in Greek today as the Σύμβολον τῆς Πίστεως, symbolon tis pisteos. According to the Oxford English Dictionary, the word “symbol” was first used in something like the modern sense in English in 1590, in Edmund Spenser’s Faerie Queene. In today’s usage, just as the Apostles’ Creed stands in for Christian faith, so too can e.g. the golden bear stand in for the state of California. It struck me while reading Geertz how strange it is to propose a supposedly secular, objective definition of religion using terms that are basically religious and specifically Christian in origin.
My small observation about Geertz is in line with the larger thesis Talal Asad advances by taking issue with Geertz’s definition of religion. Asad argues that
there cannot be a universal, transhistorical definition of religion, not only because its constituent elements and relationships are historically specific, but because that definition is itself the historical product of discursive processes. (29)
These “discursive processes” include, most prominently, the Christian heritage in which definitions of religion like Geertz’s are steeped. As Asad puts it:
what appears to anthropologists today to be self-evident, namely that religion is essentially a matter of symbolic meanings linked to ideas of general order (expressed through either or both rite and doctrine), that it has generic functions/features, and that it must not be confused with any of its particular historical or cultural forms, is in fact a view that has a specific Christian history. (42)
Why is this a problem? Asad thinks that by forgetting the specific history of the definition of religion, we also forget how closely intertwined religion has been with power. Thus, Asad says, the entire “theoretical search for an essence of religion invites us to separate it conceptually from the domain of power” (29). Although Foucault is nowhere cited in this essay, his influence is clear: Asad proposes that we undertake a genealogy of the discourse on the study of religion so as to excavate the entanglement of this particular knowledge formation with historically shifting forms of power.
Where does that leave us? Although Asad does not offer his own definition of religion, he does — to his credit — give us some pointers on what we can do with the study of religion beyond critique. In the conclusion of the essay, Asad writes that “the anthropological student of particular religions should therefore begin … [by] unpacking the comprehensive concept which he or she translates as ‘religion’ into heterogeneous elements according to its historical character” (54). In other words, before studying any particular religious practices we must take stock of what concepts of “religion” we are operating with. Thus, it is not enough to locate religion within the context of a culture. For Asad, religious symbols cannot be understood on the analogy with words as “vehicles for meaning” (53). Instead, meaning comes from use: Asad thinks that “the possibility and authoritative status” of “the meanings of religious practices and utterances … are to be explained as products of historically distinctive disciplines and forces” (54). (Put in even more starkly Wittgensteinian terms, the meaning of religious symbols belongs to the forms of life in which they are used.) Interestingly, I think Asad would agree with Geertz that the meaning of particular religious practices and symbols is tied to their cultural context. Thus, Asad accepts that “religious symbols … cannot be understood independently of their historical relations with nonreligious symbols or of their articulations in and of social life, in which work and power are always crucial” (53). Yet Asad wants us to go beyond Geertz by recognizing that these concepts themselves, things like “symbol” and “religion,” form grammars that are bound to historically and socially constituted language-games.
Asad says that the basic point of his essay has been “to problematize the idea of an anthropological definition of religion by assigning that endeavor to a particular history of knowledge and power … out of which the modern world has been constructed” (54). My worry is that this alone is too thin of ground for us to have meaningful conversations across times, locales, and disciplines. This is a common worry expressed with regards to critical projects like those proposed by Asad, Smith, and Masuzawa. Yet I do think Asad is pointing us somewhere beyond just reflexivity. He suggests that we place both religion (as observed practices) and “religion” (as a constructed object of academic study) within their respective forms of life. The question that remains, for me, is how to use concepts to communicate across forms of life.