Unlike conventional economic theory, the paradigm of capitalist development contrasts peasant and capitalist production in terms of (i) the market-mediation of production and (ii) the relations of production and forms of exploitation. The structure of causation underlying the capitalist transformation of peasant production has been the object of much discussion and debate; however, there remain problems which are not entirely resolved or taken up. These have to do with the part-market economy of peasant agriculture, the complex and varying relationships between production and the market and the relative significance of market growth and class forces in propelling the process of capitalist development. The thesis aims at reconstructing some of these problems from first principles, drawing upon, in particular, the classical texts.

It is argued that in peasant agriculture, production and distribution are interlinked by market and non-market sectors with specifiable properties. Such properties are compatible with different relations of production, so that the latter cannot be deduced from the former. Capitalist development, therefore, is not simply a matter of the growth of the market, nor can such change in the relations of production be adequately explained at the level of circulation processes.

In fact, there are complex variations in the relationship between production and the market, manifested in distinct patterns of market participation by peasant classes. Analysis of the conditions of reproduction as a whole endows such patterns with a certain coherence and helps to identify the variable range of 'production problems' which differentiate classes of peasant producers. It is argued that capitalist production may not be undertaken either because it is not feasible, or because it is not systematically necessitated.

Furthermore, the reproduction of non-capitalist producers cannot be fully explained without taking account of the relations of production, and corresponding forms of exploitation, to which they are subject. In particular, relations of production which systematically retard the generation of wage labour can constrain capitalist development because of the peculiar properties of land as a means of production. The problem, therefore, has to be posed in terms of the transformation of the pre-existing relations of production rather than choices made by individual agents/enterprises to maximize surplus or to switch to avenues with higher rates of profit.

These illustrated arguments are with evidence developed in terms of a model which from Bangladesh, supplemented by that is from India. It is, however, not intended to be an empirical study of peasant production and capitalist development in Bangladesh.