AGRARIAN CHANGE 13 State—through agrarian reform legislations, massive pumping of credit, provision of subsidised irrigation and procurement and price policies—has been extremely important in strengthening commoditization, developing productive forces and promoting capitalist relations of production. Our material also shows that, while as a consequence of both State intervention and class struggle, the obstacles posed by landlordism and usury have been partially overcome in the study area, the process is both fragile and not without its contradictions/' The fragility of the process is especially evident in the sphere of credit where large sections of the peasantry faced with a crisis brought about by sharply increased relative prices of manufactured inputs and crop failures, lent their support to the "farmers's agitation" demanding moratorium on debt and distraint proceedings, and higher prices for produce. Similarly, while in our wet area where class strugles as well as State-initiated land reforms have led to the considerable strengthening of owner cultivation and of tenants in relation to the landlords, the latter still control two fifths of land, and land-rent remains an important element of surplus. The analytical resolution of these contradictory aspects is obviously not to be sought in agriculture, but in the economy as a whole, and specifically in the contradictions of capitalist development under the aegis of a bourgeois-landlord State, and against the background of its complex relations with imperialism. We thank our excellent field assistants : Arasu, Ayyavoo, Brindhuvahini, Chelliah, Guruswamv, Jothi, Krishnamuthv, Mariasoosai, Natarajan, Rajagopal, Sampath, Vidyasagar. The project has received support from the Swedish Agency for Research Cooperation with Developing Countries (SAREC) and from Madras Institute for Development Studies. Corresspondence to Lindberg, Department of Sociology, Bom 114, 221 00 Lund, Sweden. 1 A recent review of the debate can be found in Thorner, 1982. 2 See especially Bhaduri 1973, Banaji 1977, and Patnaik 1976. 3 One finding which is not immediatelv related to the debate on the mode of production, but is relevant to it is that our material throws considerable doubt on the alleged inverse relationship between farm size and productivity asserted by many writers (see for instance Berry and Cline 1979). Instead, our data demonstrate that intensity of input use and advantages of class are decisive mfluences on productivity. For a detailed discussion see Athreya et al1980b. 4 The point should be stressed that in any comprehensive—i.e. economic, political and ideological—conceptualization of class, the distinction between rich peasants and landlords would of course be important. 5 Prices received by big capitalist farmers in the wet area for paddy were, on the average, 25 per cent higher in the case of local and improved varieties, and 10 per cent higher in the case of HYV's, than the prices received bv the poor and middle peasants. In the case of some banana varieties, big capitalist farmers received prices which were higher by almost two thirds. Similary, in the dry area, capitalist tamers enjoyed a 15-20 per cent price advantage