The various notions of symbolism developed in the social sciences constitute different responses to a central, apparently inevitable, tension in a scientific description of culture. Most cultural productions convey some meaning, yet a social science cannot readily accept these overt meanings as a sufficient explanation for those manifestations. To close this gap is the main point of developing a notion of symbolism. The point is argued forcefully by Durkheim; consider for instance his famous statement, that a human institution like religion ‘cannot be founded on an error and a lie…. One must know how to go underneath the symbol to the reality which it represents’ (Durkheim 1947 [1915]: 14). This statement also introduces the ambiguity, to be found in most discussions of symbolism, between three different understandings of the term. Cultural manifestations are called symbolic, first, because they can be interpreted as indices of underlying social realities by social scientists, or second, because they are expressive of particular concerns of the actors, or, finally, in the sense that they are prima-facie irrational. These different understandings are at the foundation of three major approaches to cultural symbolism—sociological, hermeneutic and psychological respectively.
The first, sociological stance is principally associated with the name of Durkheim, although the notion of cultural productions as signs or symptoms of social relations was forcefully articulated in Hegelian and Marxist approaches to culture. In Marx in particular, the uneasy definition of ideology reflects a central ambiguity in the study of symbolism. Ideology is often described as an exercise in camouflage, as the representations whereby a social class explicitly portrays the social order in which it dominates as the only possible one. But ideology is also part of a broader, historically determined form of consciousness, which includes all the representations that make social interaction possible. While Marx concentrated his activity as a decoder of cultural forms on the first, narrower domain of ideology, Durkheim extended the notions of symbol and referent to the whole of human culture. The main thrust of this approach was a rejection of Tylorian intellectualism (Skorupksi 1976). The contrast is particularly evident in the domain of religion, seen by Tylor as a misguided attempt at explaining the natural world, while for Durkheim it is a ‘figurative expression’ of social structure. In the Durkheimian system, however, symbols cannot be seen as just projections of pre-existing social forms. Religion in particular creates social cohesion by enforcing its ideal counterpart, conceptual cohesiveness.
One may argue that the Durkheimian symbolist tradition tends to confuse two quite different aspects of cultural symbols, namely, their use as a source of information for the sociologist and their meaning for the social actors. Consider for instance Leach’s statement, that ‘the various nats [i.e. spirits] of Kachin religious mythology are, in the last analysis, nothing more than ways of describing the formal relationships that exist between real persons and real groups in ordinary human Kachin society’ (Leach 1954:182). The explanation proposed may well exhaust the sociological significance of the various spirit-notions of the Kachin. But this leaves untouched the particular meanings they articulate for Kachin actors. Such considerations found the second, hermeneutic stance in the approach to symbolism, in which the main problem is to translate the meanings of cultural symbols, not to explain their occurrence. For Geertz, for instance, the study of culture is ‘not an experimental science in search of law but an interpretive one in search of meaning’ (Geertz 1973:5). In such authors as Turner (1967; 1974) or Fernandez (1986), the emphasis lies on the figurative power of cultural symbols, on their effects as both expressive figures and organizing principles for inchoate or unstructured feelings and thoughts.
It is possible, however, to approach meanings without taking this hermeneutic stance, and consider the production of meaning as a psychological process, amenable to scientific investigation like any other such process. This leads us to the third, cognitive understanding of symbolism, intrinsically related to considerations of rationality. From Levy-Bruhl or Rivers down to modern cognitive approaches, anthropology has tried to provide some description of the cognitive processes whereby people can be led to hold beliefs for which rational justifications seem either impossible or unavailable. So symbolism could be construed as the product of a special ‘mode of thought’, with particular functional properties. An echo of this conception can be found in Lévi-Strauss’s notion of a pensée sauvage. Here symbols are produced by the application of universal, formal operations, such as binary opposition and analogy, to a set of ‘concrete’ categories: nature and culture, male and female, the raw and the cooked, etc.
(Lévi-Strauss 1966; see also Leach 1976). In the absence of precise psychological models, however, the idea of particular modes of thought leads to a conception of symbolism as a residual category, justifying Gellner’s comment that ‘in social anthropology if a native says something sensible it is primitive technology, but if it sounds very odd then it is symbolic’ (Gellner 1987:163). An important attempt to go beyond this characterization can be found in Sperber’s (1975) cognitive account of symbolism. For Sperber, certain cultural phenomena are ‘symbolic’ to particular actors if their rational interpretation does not lead to a limited and predictable set of inferences. This triggers a search for possible, generally conjectural representations which, if true, would make a rational interpretation possible. This conception has two interesting corollaries for the social scientist. First, it implies that it is futile to provide keys or translations for cultural expressions. The fact that a phenomenon is treated symbolically precisely excludes the possibility of a single, exhaustive interpretation. Second, while symbolism exists as a psychological process, there are no such things as ‘symbols’, as a particular class of cultural products. Any conceptual or perceptual item can become symbolic, if there is some index that a rational interpretation is unavailable or insufficient. In this framework, a proper account of cultural symbolism will be found, not in the social sciences as such, but in the empirical and theoretical developments of cognitive science.
Pascal Boyer
University of Cambridge
References
Durkheim, E. (1947 [1915]) The Elementary Forms of the Religious life, London.
Fernandez, J.W. (1986) Persuasions and Performances: The Play of Tropes in Culture, Bloomington, IN.
Geertz, C. (1973) The Interpretation of Culture, New York.
Gellner, E. (1987) Culture, Identity and Politics, Cambridge, UK.
Leach, E.R. (1954) The Political Systems of Highland Burma: A Study of Kachin Political Systems, London.
——(1976) Culture and Communication: The Logic by which Symbols are Connected, Cambridge, UK.
Lévi-Strauss, C. (1966) The Savage Mind, London.
Skorupski, J. (1976) Symbol and Theory: A Philosophical Study of Theories of Religion in Social Anthropology, Cambridge, UK.
Sperber, D. (1975) Rethinking Symbolism, Cambridge, UK.
Turner, V. (1967) The Forest of Symbols: Aspects ofNdembu Ritual, Ithaca, NY.
——(1974) Dramas, Fields and Metaphors, Ithaca, NY.
