6§ SOCIAL SCtE^TtSf history is concerned",4 i.e., those socio-economic changes which generate urban forms. It has been reiterated that the "proper concern should not be with cities as such but with complex societies, in which cities and their hinterlands are interwoven into tight political and economic webs."5 In studying the evolution of urban forms, there is the additional hazard of taking a single factor as universal and attributing absolute primacy to it or overstressing one aspect as innovative or catalytic at the expense of others. The hazards of searching for a general framework notwithstanding, research available on the ancient and early historical periods in India has shown that there were two major periods of urbanisation, to which a third may now be added for the early medieval period.6 It is also generally accepted that the first urbanisation, i.e., the Indus Valley urbanisation of the proto-historic period, left no legacy beyond the middle of the second millennium B.C.7 The second or the early historical phase, represents a long period of urban growth stretching from the sixth century B.C. t0i the third century A.D. With its epicentre in the Gangetic valley it spread over the whole of North India by the third century B.C. and over Central India, Deccan and the Andhra region between the second century B.C. and thir^ century A.D. It must be stressed, however, that it is the latter part of this long chronological span that witnessed the most clearly visible manifestations of this urbanism. The end of the second urban phase is a time-market for the early his-^ torical period and provides a starting point for the protagonists of the theory of "Indian feudalism59,8 i.e., a new socio-economic formation, sometimes termed as "land-grant economy". A new set of urban centres is associated with this period and with the decline of feudalism in about the thirteenth century A.D., this period is brought to a close, coinciding with the beginnings of Turkish rule in North India.9 These time brackets may serve as a useful working framework within which urban phases may be distinguished and overall patterns may be worked out. They may not, however, bring us any nearer to arriving at a general analytical framework, for precise chronological limits sometimes break down or become irrelevant, when the actuality of regional evidence is taken into account. Thorough investigations at a regional level wouM therefore become a pre-requisite, before the commonality of elements involved in urban growth can be isolated. Keeping in view the elusive character of urban theory, the limited scope of concepts to be applied to a variety of social contexts and given the present state of empirical research on the urban history of India, my work aims at providing a regional perspective for South India mainly as an indicator of a field where a great deal of fundamental work needs to be done.