THE EPISTEMOLOGICAL FOUNDATIONS OF CULTURAL MATERIALISM [1] Guy Oakes For the last fifteen years, the attempt to develop a 'materialist' science of culture has been intimately associated with the work of Marvin Harris [2]. This paper is a critique of the three constitutive or definitive epistemologi- cal assumptions to which Harrisian cultural materialism is committed. In the ensuing, section I provides an analysis of these three principles: the Emic/Etic Dichotomy, the Principle of Infrastructural Determinism, and the Principle of Emic Mystification. Section II is devoted to their critique. The essay does not consider Harris's work, the specific positions he takes, and the difficulties they pose as an intrinsically significant body of research merit- ing investigation for its own sake, but rather as a collection of texts committed to these principles. Its interest lies in these three basic doctrines and their status as epistemological presuppositions of a possible sociocultural science. A. THE EMIC/ETIC DICHOTOMY "No aspect of a research strategy more de- cisiviely characterizes it," Harris claims, "than the way in which it treats the relationship be- tween what people say and think as subjects and what they say and think and do as objects of scientific inquiry" [3]. The purpose of Guy Oakes teaches Philosophy at Monmouth College, New Jersey, and Sociology at the Graduate Faculty, New School for Social Research. Harris's distinction between emics and etics is to develop a general principle for the analysis of this relationship. As Harris formulates it, the dichotomy is meant to establish a distinc- tion between two kinds of sociocultural pro- positions. The difference at stake in the dicho- tomy may be exposed by considering two different answers to one of the basic methodo- logical issues of the socioculturaI sciences: namely, under what conditions does a given concept or the ascription of a given predicate identify some datum as a sociocultural pheno- menon [4] ? Suppose that the ascription of a given predicate correctly identifies a socio- cultural datum within a specific society if, and only if, there is a sense in which it reproduces the native's perception of that datum. Put an- other way, suppose that the criterion for the legitimacy of this description is a sense in which it is equivalent to a native description of the datum. In that case, the proposition is emic. As Harris puts it, "the test of the ade- quacy of emic analyses is their ability to gene- rate statements the native accepts as real, meaningful or appropriate" [ 5 ]. Emic propo- sitions, therefore, depend upon the ascription of a certain complex of motives, beliefs, inten- tions, and purposes to the native. Suppose, on the other hand, that the identification of a given sociocultural datum is logically inde- pendent of any description of the native cons- ciousness. Suppose that the legitimacy of this description is a consequence of its status as a component of the theoretical apparatus em- ployed by a sociocultural scientist-observer. In that case, the proposition is etic. That is be- 0304-4092/81[0000-0000/$02.50 1981 Elsevier Scientific Publishing Company