ABSTRACT
While working with Terry Irving on Class Structure in Australian History, I had been taking for granted the superiority, indeed the necessity, of a historical approach to class. The common currency of academic sociology, and hence of discourse in the semi-professions that drew on it (welfare, teaching, and the like), was an ahistorical stratificationism. Apart from the brilliant but brief introduction to the second edition of Thompson’s Making of the English Working Class, there was nothing much to which one could point that would explain what the difference between historical and ahistorical class analysis actually was.
I had long felt that the difference was not mainly in subject-matter, was not even a question of duration at all (for you can get cross-sectional stratificationist studies that talk about time via repeated cross-sections, as in ‘panel’ surveys like Lazarsfeld, Berelson and Gaudet’s The People’s Choice ). It was a matter of the inner logic of conceptions of class, which determined whether or not they could grasp class dynamics. And thus it was the logic of class analyses that ultimately decided their political meaning. But how to explain the different logics? Some years earlier I had begun to read Chomsky’s linguistics, and it seemed that the distinctions he made in Syntactic Structures between different kinds of grammars had some bearing on the problem I was facing. An adaptation (rather than an application) of his models helped to formulate the distinction between dimensional and generative theories that is the basis of this paper. A short first version was written in April 1976, a much expanded one for the Sydney Class Analysis conference in 84August that year, and another revision was tried on my long-suffering students in a course on Class Analysis in early 1977.
