Bloodletting

To the Editors:

In his review of my Comparative Functionalism: An Essay in Anthropological Theory, [May 26, 1966] Edmund Leach accuses me of merely bowdlerizing Malinowski, giving no consideration to the meaning of the phrase “comparative functionalism” or to what the book intends to set forth. Since his review stems out of the cultural limitations that have come to dominate British social anthropology, it seems only fair to your readers (and to my publisher) that I elucidate the approach that I develop in my brief essay.

The two personalities that drew British anthropology out of sterile antiquarianism were Malinowski and Radcliffe-Brown, and the segmental opposition that has characterized social anthropology has stemmed from this original pair. Their approaches were alike in that both tried to make sociological sense out of ethnographic fact, but in style the two men were as disparate as an East European nobleman and a middle-class Englishman—the one brilliant, witty, and wide-ranging in his intellectual interests; the other dry, mechanistic, and narrowly logical. It is one of the minor enigmas of recent intellectual history that British anthropology turned away from the holistic understanding of Malinowski, with its recognition of the intricacies of the psychobiological in man’s behavior, to the formalized structuralism of Radcliffe-Brown. It seems too facile to lay this to national character—to one with kippers and cold toast and the suppression of the grosser appetites—but it certainly cannot be credited to the fact that the one has greater explanatory powers than the other. Be that as it may, in the further segmentation of the lineal descendants of the original pair, there has also come into being a new group which denies not only the psychological involvement of Malinowski, but also the scientific interests of Radcliffe-Brown, and has turned to a humanistic approach as the only permissible role for anthropological inquiry into the nature of human behavior.

It is in the context of these considerations that one must read Leach’s review of Comparative Functionalism. For, on the one hand, I have accepted the Malinowski notion of functional needs and, on the other, the Radcliffe-Brown position that the comparative study of human societies can lead us to verifiable understandings of human behavior in its diversity. Both these widely held assumptions are rejected by Leach.

Malinowski never engaged in comparative studies. In fact, he was unable to do so because, though he wrote about needs, he focused on institutions. The structuralists, who do endeavor comparisons—or say they do—are caught by the fact that institutions, each of them growing out of the culture in which it is found, are not actually comparable entities. Comparative Functionalism seeks a way out of this dilemma, by focusing on the requisite functions for all social systems and seeking the institutions that perform these functions, building thereby a systematic understanding of the elements necessary to any society. Let me illustrate this. A Malinowskian approach to the law examines the internal meaning of legal actions—resolution of conflict and the definition of rights and obligations—in terms of the culture under investigation. In such an approach there is no one institution of law, separate from the remainder of the society; rather, one finds the law jobs performed by such institutions as kin groups, magical practices, and the like. On the other hand, the structuralists would have us compare legal institutions, and thus reduce the actuality of legal behavior to preconceived units of social action, thereby doing violence to the ethnographic data. The result is a reification of concepts and endless quibble over terminology. But if we focus on functions—on the law jobs to be performed—we can discover which of these are in fact universal and which are limited, and we can discover the diverse ways in which a particular functional need may be met. We can empirically test a functional, as opposed to a structural, model of society. Though the idea may seem simplicity itself, it was not a part of Malinowski’s schema and it does constitute a step forward in the theory of anthropological investigation. I must add that, though it seems a simple idea, it forces the anthropologist to face his own assumption about the universal character of man and the inherent needs of social systems—something many of my colleagues on both sides of the Atlantic are loath to do. But, since such assumptions can be formulated as premises to be tested by empirical data, it is really not so formidable to make them explicit…. We are beginning to discover that when societies are categorized into ecological types, they show a high degree of similarity in their basic organizational features. The social structure of peasant communities in politically organized socities—as Mr. Leach himself discovered in Ceylon—tend to have certain specific features. For instance, they characteristically have monogamous marriages, a patrilineal bias, and a tendency toward three-generation household systems. Such a point is not merely of theoretical interest; we find in West Africa, as a result of the formulation of modern states, a strain toward monogamy both in urban and peasant areas.

But these are “scientific” explanations and in Mr. Leach’s world anthropologists must either be humanist or scientist—good guys or bad guys, with the hallmark of the bad guys being the denial of free will to the human subjects. This is arrant nonsense. If we are endeavoring to explain human history rather than merely recount it, to explain social behavior rather than merely record it, we must put forth such efforts as we can to make our explanations verifiable. Scientists normally do this by testing hypotheses with as large a sample of controlled variables as can be mustered. Social scientists emulate this procedure as best they can with their sparse and recalcitrant data, not because science is good and humanities are bad, but because that is the only way to verify a causative statement other than by an appeal to common sense (i.e., folk attitudes). It may be that no science of human behavior is possible and we are wasting money on the behaviorial sciences; but so long as we (Mr. Leach and the British social anthropologists included) are making causative statements about human behavior, we are in fact making the metaphysical assumption that regularities exist to be discovered and verified.

In view of these considerations, the self-conscious apologia for British arrogance with which Mr. Leach opened his review takes on special poignancy.

Walter Goldschmidt

University of California