Of the making of “structures” there is no end. So might any onlooker think when surveying the intellectual fashions that have enlivened the development of the social sciences during the last 75 years. The term has been used for a confusing succession of notions and concepts that have shared very little with one another except the label “structure” itself. Anthropology has been powerfully influenced by Radcliffe-Brown, Parsons, and Levi-Strauss, each offering his own distinctive structural road to intellectual enlightenment. Parsons has had most of his following in sociology, but so too have Blalock and Duncan, with their very different understanding of what is meant by structure. The liveliest structuralist controversies have erupted on the borders of social science, as conventionally defined, over the structural Marxism of Althusser, structural linguistics, and above all, structural analyses of literature.

Our onlooker, if he or she is tidy-minded and likes ideas and practices to be pigeonholed unambiguously, might regret that so many different people, without consulting one another, have taken on the role of Humpty Dumpty and declared the word “structure” to mean whatever each wanted it to mean. Intellectual discourse, he or she might well think, would be much more effective, and much less frustrating, if the humanities and social sciences were to use a technical vocabulary closer akin to that of the natural sciences, with no ambiguity and just the right amount of redundancy. Then there would be no doubt about what people were trying to say.

Alas, this vision of an immaculate natural science is only a naive onlooker's mirage.