Economic sociology has developed into an important area of research over the last two decades. This is especially true for the United States, but increasingly also for some European countries. In the center of this development stands the notion of "embeddedness" as a common denominator for very heterogeneous research strands. However, "embeddedness" is rather a programatic concept than the basis for a coherent research program since it does not define a core problem for economic sociology.

In this paper I will argue that many of the difficulties of economic sociology to develop a clear theoretical foundation have to do with its relationship to economics. Economic sociology stands in competition with the arguably most successful discipline in the social sciences and hence has to legitimate itself in relation to economics. Very often the sociological critique focuses on the unrealistic assumption of individual rationality as developed in the concept of homo economicus. I will argue that this is an insufficient starting point for economic sociology since the strength of economics lies in its normative assumptions. That is, it gives prescriptions how actors should act in order to promote their interests in the most efficient way. A sociological critique has to take this point much more seriously than sociological accounts usually do. However , economic sociology will find a theoretically defined vantage point if it can be shown that the action-theoretic model of homo economicus is not consistent with the general achievement of optimal outcomes. The two core questions are:

1. Do we always reach pareto-efficient results if actors base their decisions on the recommendations of economic theory for rational action? And if not, how can we reach optimal outcomes?
1. Is it always possible for actors to recognize optimizing strategies? And if not, how do we act then?

I will argue that we can identify exactly three situations, wich are central for economic processes, in wich we have to answer "no" to at least one of the two questions. These situations are cooperation, uncertainty and innovation. The justification for this claim is developed in the paper . The three situations form the most important systematic starting point for a sociological investigation of the economy. They are not only decisive to determining the relation between sociology and economics, but especially for the empirical understanding of economic structures and processes in market economies. Moreover the systematic approach to economic sociology allows to achieve a clearer understanding of "embeddedness". Modern market economies are "embedded" not least because actors cannot always recognize optimizing strategies; or rational strategies do not always generate optimal outcomes. Political, cultural, institutional, and social embeddedness achieves its significance as "substitute rationality" in situations in which optimizing strategies are not available to actors.