80 SOCIAL SCIENTIST productive forces in the region are under developed and that yields can be substantially increased, the figures quoted understate the existing level of productivity. The authors point to the changes in productive relations that have occurred since the prc-colonial period, though it appears that there is far too much idealization of the pre-colonial situation in the region. For example, available documentary evidence does not suggest that the jajmani system was so much the centre of socio-economic relations as they make it out to be. There was even then considerable buying and selling (both for cash and for paddy) in prc-British towns and weekly markets. Even new agricultural tools could often be bought in local markets, though the village artisans were responsible for their maintenance, especially during the peak period when the cultivators had to work daily in their fields. The information provided on land ownership is quite interesting, but it is unfortunate that the lowest land-owning category has not been broken down further. Specifically, it would have been useful not to lump together those owning less than ten cents of land with those owning close to U acres. It is striking that there is a low incidence of landlcssness in this village. Yet the distribution of land holdings in the higher categories is quite consistent with most other reports available for Tamil Nadu, with a considerable concentration of land in the hands of a very small number of households. Fitting Data to Theory? Too much time is spent on arguing theoretical points instead of putting down the data straight. Though one can agree with much of their theoretical argument, there arc some problems. For one thing, there is much cross-cultural data that should have been considered, before arriving at generalisations. Thus, they state: ^Sharecropping for example, is a form usually associated with feudalism", and then they ask whether it can be termed a feudal relation when a middle farmer takes land as a share cropper from a small landowner. But the issue is more complex. Sharecropping is a common pattern found in many parts of the American south. Is it feudal? It clearly emerged as one way of dealing with developing capitalism in the nineteenth century, and is closely related there to certain crops like tobacco. Likewise, in Tamil Nadu, a landowner who prefers to spend his time on other things can manage to give his land to a sharecropper if he is cultivating paddy, but not if he is cultivating sugarcane. Sharecropping is often an effective way of exploiting labour and diversifying productive assets in the case of paddy. However in the case of sugar, only one or two men are needed to look after the land most of the year, and only during the harvest season need they be supervised carefully. Indeed, many have solved the problem even more simply, by selling the standing sugarcane crop to a