The purpose of this Note is two-fold. First, it shows that Schmitter implicitly includes in his definitions of systems of interest intermediation faulty statements concerning the causal link from institutional forms to power relations between the state and interest groups. Second, it argues that in order to avoid such errors, and benefit in other ways, students of systems of interest representation should remove implicit statements about power from their definitions and explicitly state their hypotheses connecting institutional forms and power relations involving the two actors of those systems.

In the March 1988 issue of this Review, Virginia Gray and David Lowery presented a respecification of Mancur Olson's model of economic growth and tested the revised model with U.S. state data. A special feature of the Gray-Lowery analysis is its more thorough measurement of interest group effects. These investigators found interest group influences quite different from those anticipated by Olson's model. Paul Brace and Youssef Cohen argue that the Gray-Lowery model misspecifies the determinants of state economic growth, overstating the role of interest group size and failing to incorporate crucial exogenous variables. Gray and Lowery join the issue and defend their specification.

The argument of this paper is that the emergence of military dictatorships, such as the Brazilian regime of 1964, is not caused by an economic crisis of dependent capitalist development. Rather, it results from a polarization and radicalization of the democratic regime by which it is preceded. Democracies handed down from above, like that in Brazil and other South American democracies, lend themselves to polarization and radicalization. They therefore favor the emergence of modern forms of autocracy.

From the onset of industrial capitalism in Latin America, urban workers have been dependent on the state. This is particularly true in Brazil, where urban workers are still in a situation of extreme dependence on the state. Thus, the Brazilian setting provides an ideal opportunity for the study of the consciousness of dependent workers as well as an important point of reference for comparative study of working-class consciousness. This paper outlines the general characteristics of the situation of state-dependent workers in Latin America, with special attention to the Brazilian worker, in order to show how their dependence is reflected in their consciousness. The evidence for our interpretation is based on the attitudes of 617 urban workers who are part of a larger probability sample of the population in central and southeast Brazil.

The central argument of this paper is developed as a criticism of a widely accepted interpretation of collective violence in new states. It is shown that instead of indicating political decay, violence in these states is an integral part of the process of accumulation of power by the national state. To the degree that this power accumulation is necessary for the imposition or maintenance of order, collective violence also indicates movement towards political order on a new scale. Admittedly, our evidence is far from definitive. Nevertheless, it consistently contradicts the interpretation of violence as political decay and supports our interpretation of violence as a usual feature of the process of primitive accumulation of power.