DISCUSSION 65 In this context, it was pointed out in the earlier review that anthropological studies which have analysed pre-capitalist societies using the concept of modes of production were totally ignored by Partha Chatterjee. Now we are told that though these Marxist anthropologists "have made a number of observations" with regard to the community, these are "specific, ad hoc and limited" {Social Scientist No. 141, p. 59); they had been left out "primarily for reasons of space" (ibid, p.58). Even a discussion of Engels' The Origin of the Family, Private Property and the State was left out for "reasons of space" {Subaltern Stitdies-11, p.318). On the other hand the functionalist anthropologists did not face this problem of space and were summarised in over twelve pages {Subaltern Studies-11, p. 315-330). This is especially strange because Partha Chatterjee now seems to be violently opposed to this school—"I find it incomprehensible that this body of theory should seem consistent with my very clearly stated position....." {Social Scientist No. 141, p.58). As far as the concept of the Asiatic Mode of Production and the configuration of property specific to it is concerned, it should be pointed out that in Mughal India also there are records of sale of Zamindaris or land right. These are similar to the sale of land in Medieval Europe where the sale did not mean absolute ownership in the capitalist sense but only the right to extract rents. The development of capitalist legal property went apace with the development of capitalist relations of production as can be seen clearly in the conjunction of the enclosure acts with the expropriation of the peasants and the primitive accumulation of capital in England. The crucial fact was that the development of capitalist legal property went hand in hand with the development of capitalism. Its absence in the non-European societies should not be taken to mean that the transition to capitalism in the non-European countries was not possible. Partha Chatterjee's positivism can be seen clearly in his assertion that "in spite of their considerable critical efforts" the reviewers have failed to pinpoint the main weakness of the concept of the communal mode of power— "that it remains an abstract concept." {Social Scientist No. 141, p.60). Now it is in the empiricist epistemology that abstraction is considered a weakness. If we accept that a mode of power is a regional instance of the wider concept of the mode of production it necessarily has to be an abstract concept. A mode of production does not correspond to any empirically given society—but is a theoretical construct which has to be abstracted from the empirically given social structure. The twin articulation of the labourers, means of production and non-labourers along the axes of means and relations of production for instance is a theoretical construct. So are the various levels or instances of a mode of production—economic, political, ideological—in terms of which the mode of power has been defined. The invariant elements of a mode of production can be combined theoretically in order to generate modes of production which may not have existed historically. To sum up, while Partha Chatterjee considers the concepts of modes of production and modes of power to be compatible, the precise relation of