Anemic and Emetic Analyses in Social Anthropology

Abstract

A delivered before the Southwestern Anthropological Association meetings, spring 1965, a gentleman came to me and suggested helpfully that I might use a framework for the comparative study of caste which would be based on emic, etic and allo concepts.This experience, on brief reflection, inspired me to the title and topic of this essay, which I presented orally before the Kroeber Anthropological Society's ninth annual meeting at Berkeley, on May 8, 1965.On scarcely more reflection, I thought of the spate of recent and announced journal articles, special issues, and other works by "ethnoscientists" or "componential analysts" who treat culture in a linguistic idiom in the avowed hope of discovering structure and enhancing rigor in ethnology.I especially recalled those in a special edition of this journal entitled Transcultwal Stzcdies of Cognition (AMERICAN ANTHROPOLOGIST, 1964).I then recalled a statement, whose author shall remain anonymous, in which that edition was described as ('an attempt to understand the mood and temper of man through empty words.Vacuous, sterile, inconclusive, programmatic, hyper-professionalized, and the product of apolitical, asexual, amoral, asocial anthropology."A major aim of the ethnoscientists is cogently and readably put by Frake (1962) in an article which should be required reading for all commentators on componential analysis: that by utilizing the linguistic analogy and seeking "contrast sets" of "cultural segregates" we can derive superior ethnography.Superior ethnography is a concern of mine too.I see the concern of those advocating a linguistic analogy to achieve this end, and my own concern to achieve this end, as deriving from a common source: the problem of how to treat human situations scientifically.I t can be phrased, and has often been treated, as a dilemma: how to be scientific and at the same time retain the humanistic insights-the human relevance-without which no account of human beings makes sense.There are two extreme but common reactions t o this dilemma, namely, to ignore one or the other of its two horns.That is, to be scientific at the expense of insight, or to be insightful at the expense of being unscientific.The first is what I would term the "scientistic" refuge.It involves a retreat to, or preoccupation with, such things as quantification, abstract models, simulation, and highly formal methods of data collection and analysis.I t rejects intuitive