Review by Kenneth L. HALE

''This short but interesting essay is partly a case study-it is a description of the methods that its author used while doing linguistic field work with speakers of TeleCfodl in the Territory of New Guinea. The issues discussed seem, however, to be universally valid for linguistic field work among nonliterate peoples. The essay is apparently aimed at those who have done little or no linguistic field work, since those who have worked extensively with informants will find the methods described highly familiar. Nonetheless, it is convenient, for pedagogical purposes, to have these field practices discussed in a short volume such as this, and the 63-item bibliography accompanying the discussion is quite helpful. After an introduction, the essay is divided into seven sections: general approach, linguistic surveys, investigating phonology, pair testing, eliciting grammar mono-lingually, determining meaning, and using tape recorders. The first section discusses the initial contact with informants, and the final section is concerned with various technical and social problems associated with the use of tape recorders. The remaining sections are concerned with specifically linguistic problems. The discussions are short and quite general, but a large number of points essential to any successful program of field work are made clearly and well. The only technique described in great detail is the so-called “pair test” for phonological contrasts. Perhaps the most interesting discussions are those concerned with the dangers involved in reliance on bilingual elicitation. A number of extremely useful techniques are suggested for counteracting the distortions that may be introduced through the use of a trade language. Some of these techniques are monolingual in the sense that they make use of the language under study, but the discussions make it quite clear that monolingual elicitation in this sense is quite different from the “monolingual method” needed in situations where no trade language is available. The matter of informant training is not discussed in this paper despite its seemingly obvious relevance. Experience with informants in the context of research on generative grammars has convinced me that the deepest insights into particular grammatical problems are gained when the informant is, in some sense, functioning as a linguist. It does not, in principle, seem at all unreasonable to suppose that informants could be systematically trained to view language in the way a linguist does. In fact, some efforts in this direction have already shown promise. Of course, when we speak of informants who are trained in this sense, we are no longer speaking of “unsophisticatedinformants.” The question then is, should a linguist persist in working with unsophisticated informants throughout the field work period? Or should a large part of his effort be devoted to removing his informants from the ranks of the linguistically unsophisticated? ''

In American Anthropologist