Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. The University of Chicago Press and Wenner-Gren Foundation for Anthropological Research are collaborating with JSTOR to digitize, preserve and extend access to Current Anthropology. http://www.jstor.org "nobody seems to know what [religion] is." Nadel (1954), Eister (1974), and Machalek (1977), among others, doubt the possibility of a definition. Eister (1974:2), for example, says that religion has defied social scientific consensus and may "not be definable in general terms." Of theories themselves, Evans-Pritchard wrote in 1965 that to date "either singly or taken together, [they do not] give us much more than common-sense guesses, which for the most part miss the mark [and] of the many attempts [none] is wholly satisfactory" (1965: 120-21). Geertz said the following year that the anthropology of religion was in a "general state of stagnation" and that there was no "theoretical framework. .. to provide an analytic account of religion" (1966:1-4). Despite Geertz's contribution and others, there is some consensus that anthropological theory of religion has "lagged" (Saliba 1976: 189) and that there still is no theory for "normal" research in the science of religion (Buchdahl 1977). I shall not review present theories here (Saliba 1976 gives a recent short review), but shall mention relevant aspects of several major ones. I agree with Tylor, Durkheim, and Freudotherwise sharply divergent on religion-that it anthropomorphizes the world in some significant way; with Geertz (1966), Bellah (1964), and Spiro (1966) that the use of symbols, characteristically human, is especially characteristic of religion; and with Horton (1960,1967, 1973a) and Spiro (1966) that belief in the human-like beings of religion is based in experience. I disagree with Malinowski (1948) that religion is primarily wish fulfillment. Tylor, who defined religion substantively as the "belief in Spiritual Beings" (1979 [18731:10), was in my view right to define it as a kind of conception of the world. He was also right to say that religious conceptions may be reasonable attempts to understand the world at large and that they attribute humanlike features such as language and ethics to nonhuman natural phenomena. Tylor's weakness is not so much his intellectualism or even individualism as his overestimation of two phenomena (dreams and death) as topics for human thought and therefore as sources of religious notions. His "spiritual beings" are composed of the ideas of the "phantom" and the "life force" that respectively arise from each person's experience of dreams and of ...
