ABSTRACT
Proponents of formal semantics have promised far more than a distinction between coins and bills or between books and magazines. They have held out the hope for analyses in other domains which would approach in elegance those which have already been done on kinship terms. I yield to no one in my admiration for
Goodenough's analysis of Trukese kinship terminology, but Goodenough showed that his methods were possible by giving us an analysis of a whole system, and not by simply suggesting methods and iiiustrating them on scattered and artificially simple examples. In discussing my fear of the residue of indeterminacy, I said, "I can be proved wrong by analyses which admit to no alternatives." But is it too much to ask for demonstrations of the technique on whole systems other than kinship or pronouns? One need not go off to exotic parts of the world to do this. English would really be a better test case, since readers could more easily judge the results. How about a full analysis of American folk terminology for trees? I have the nagging fear that the reason full analyses have not been given is because the· methods advocated are not equal to the goals. I would really like to be proved wrong, for I think componential analysis is lots of fun, but I wiii only be persuaded by substantive descriptions, not by methodological arguments.
