Men of ideas : a sociologist's view

why particular systems change or endure. In other words it is the latter interpretation of his question he wishes to discuss. But to answer this question demands not simply a treatise on the internal mechanics of political systems but also the consideration of why (in his terms) the authoritative allocation of values in society is always necessary. The societal sources of politics are of at least equal importance to the explanation of the persistence of political systems as the activities of the authorities. Easton is in the position Pareto would have been if, having listed the variables of the political system, instead of going on to try to demonstrate their usefulness in explaining political changes, he had assumed he had answered the Spencerian problem of the functions of political systems for society. (3) But this raises the further issue of the usefulness of Easton's conception of systems analysis for political scientists. The most successful sections of the book, conceptual discussions of authority and community, owe nothing to it, and some of the claims he makes for it, that it is operational, or that empirical generalizations emerge as by-products, are extraordinary. For the terminology of inputoutput-feedback does not derive from political analysis but from a language developed to describe any adaptive system whatsoever. In consequence it tells us nothing about politics, but can only be applied to what is already known. Any empirical generalizations stem not from systems analysis but from quite independent research into politics. The usefulness of this language can be demonstrated if it is successful in