抄録
Anthropologists and linguists share the term “informant” to discribe the man from whom they learn. The “investigator” of some other social sciences may lead neglect the subject matter that inheres in ratturally occurring contexts, because his thought plays a major role throughout the investigations. An “informant” is regarded as a collaborator in the activity of providing structural description. Ethnographic and linguistic description, which does not present a material phenomenon, requires specialized techniques of processing observed phenomena such that they can construct a theory of how their “informant” has organized the same phenomena. It is the theory constructed by “investigator” and “informant”.
In this paper, I propose a thumbnail sketch of the theoretical and methodological background in the handling of data as model, not as phenomenon itself, with particular focus on the relationship between “investigator” and “informant”, by tracing from Levi-Strauss' general theory of kinship to W.H. Goodenough's monumental declaration stimulated a development which is called “The new ethnography”.
