6® SOCIAL SCIENTIST of rights over the means of production. Fhe political instance must therefore be explicitly and concretely conceptualised if we are td relate the historical process of change in social formations to our central concept of the mode of production. The community and the state are the two main structures we must analyse. It is quite clear that there are diverse forms of'community", which we have so far only described empirically .What is needed is a theoretical conceptualisation. It is surprising that despite their considerable critical efforts, me reviewers in Social Scientist have not discovered the main weakness of a concept such as the 'communal mode of power' as denned in my article in Subaltern Studies II. The weakness is that it remains an abstract concept, in the same way as the notion of the 'community" remains abstract in Marx's own writings. We can only describe some of the specific form in which the community seems to make its appearance in certain social formations. We cannot yet understand them as particular forms of a universal concept. This is a problem to which we must address ourselves explicidy. The reviewers have commended me for my brief description of the political aspect of the social relations of production in feudal Europe. The reason why this description seems more lucid than descriptions of other social formations is because the historians of medieval Europe have been able to arrange this historical material in a much more systematic form than historian of other countries or periods. But this is still at the empirical level. I do not think the problem of conceptualising the 'community* in the feudal mode of production has been thereby resolve. The problem can only be approached by a much more detailed examination of the forms of transition in historical social formations., (I have begun one such examination for the case of the late colonial period in India in Bengal 1920-194 7 : The Land Question, Calcutta, 1984). But it will not help if we continue to suffer from an undue suspicion about the term 'community^. After all, it was Marx himself who reminded his Russian correspondents in 1881 not to be afraid of the word 'archaic', for the crisis of'modern society* would have to end in 'a return of modern societies to an archaic type of communal property', only in a superior form. Yes, I do think that a framework for, the constructioji of a political theory of the kind I have indicated can be provided in terms of the opposition between community and capital. This can be a fundamentally Marxist project and I am convinced that this was the principal direction of Marx's own researches in* his last years. Obviously, no Marxist will claim that the history of transitions can be understood, as the reviewers describe my project, in terms of'the evolution of the political process alone'. The problem is how to conceptualise the political process at all. I can assure the reviewers that there is no trace in this project either of 'a Nietzschean concept... of political power isolated from the economic domain' or of'a profoundly Hegelian view of the totality which includes two paradigmatically opposed forces' (whatever that might mean !). PARTHA CHATTERJEE Centre for Studies in Social Sciences, Calcutta