80 SOCIAL SCIENTIST mode of production. They discovered more than one stage of structural changes in the socio-economic and political orders. By making systematic application of the Marxist methodology for social analysis, theoretical formulations advanced by D.D. Kosambi6 ('Feudalism from Above' and 'Feudalism from Below') and by R.S. Sharma7 in his monograph on 'Indian Feudalism' (stressing on changes from above) regarding the crucial changes resulting in the new political formulations in the post-Kusana/Gupta period proved to be important contribution to Indian history in general and to the study of the early medieval period in particular. Decline of trade and commerce and of towns, and the- absence of a centralised bureaucracy have been held to be the most important historical forces of change in the post-Gupta era in the 'Indian Feudalism' formulation. These changes forced the artisans, merchants and traders to immigrate into the rural areas which in turn augmented population pressure in the latter. In the absence of coins, salaries to the officers of regional kingdoms were paid in the form of land grants/revenue grants which served more than one purpose. Many land grants or village grants were also made to religious monasteries, individual brahmins or groups of brahmins in the barren/uncultivated or forest areas. In some cases land/village grants were to resident peasants, artisans and farm workers. As a resirit of these developments agricultural space expanded but the loss of mobility among artisans, merchants, etc. tied them to their master's soil and they were reduced to the status of tenants. The process of localisation or ruralisation of the economy fostered a closed economy and generated a sense of localism.8 Proliferation of feudal social ranks, increasing caste hierarchy as a result of the proliferation of castes, the process of decline of the vaishyas and the ascendancy of the sudras are said to be the other important features of early medieval society. In the last three decades many American historians/anthropologists have churned out an impressive volume of works on the polity and society of medieval South India. They have applied the 'segmentary state' model which had first appeared in the context of anthropological study of the Alur society of Africa,9 in order to examine the nature of the socio-political order of this period. In this approach the state is seen as a segmentary structure in which political power is diffused asymmetrically in heads of clans or lineages of different nadus. Nadus, in their view, were a 'quasi autonomous local focus of power' and an ethnically and administratively coherent peasant micro region. Subbarayalu10 views the nadu as the most crucial unit in local agrarian rural space. It has been argued that the nadu existed as a totally autonomous agrarian rural space much before the coming of Cholas. But even when their power equations altered with the Chola state or when, as Stein believes, the Chola date established its ritual sovereignty over macro-regions, its influence on