The Impact of Federalism on National Party Cohesion in Brazil

This article explores the impact of federalism on national party cohesion. Although credited with increasing economic growth and managing conflict in countries with diverse electorates, federal forms of government have also been blamed for weak party systems because national coalitions may be divided by interstate conflicts. This latter notion has been widely asserted, but there is virtually no empirical evidence of the relationship or even an effort to isolate and identify the specific features of federal systems that might weaken parties. In this article, I build and test a model of federal effects in national legislatures. I apply my framework to Brazil, whose weak party system is attributed, in part, to that country's federal form of government. I find that federalism does significantly reduce party cohesion and that this effect can be tied to multiple state‐level interests but that state‐level actors' impact on national party cohesion is surprisingly small.     

Echoes of Butler? The Conservative Research Department and the Making of Conservative Policy

Although it remains a comparatively little known institution, the Conservative Research Department (CRD) has played an important role in modern British political history. The last few years have seen a remarkable revival in its fortunes. In several important respects, this revival echoes that of the Department under the chairmanship of Rab Butler following the election defeat of 1945. This article examines some of the parallels to be drawn between the experience of the CRD between 1945 and 1951 and between 2001 and the present. It concludes by drawing some general conclusions about how the Conservative party should develop policy when in opposition.     

THE SOCIAL STRUCTURE OF VENGEANCE: A TEST OF BLACK'S MODEL*

The current research tests Black's (1990) model of vengeance through interviews with prison inmates. Drawing on a case‐control method, each respondent describes a matched pair of conflicts: the crime of vengeance and a similar nonviolent conflict from the same time period. The quantitative data are used to estimate the impact of conflict structure on conflict management. Conditional logistic regression models offer mixed support: relational distance and functional independence increase the odds of vengeance, but the remaining theoretical variables do not. The qualitative data are used to illustrate the quantitative patterns and unearth parallels between premodern and modern violence.     

SOCIAL SCIENTISTS AS EXPERT WITNESSES: Their Use, Misuse, and Sometimes Abuse

Upon entering the legal realm, the expert witness must begin to define his or her role and establish a working relationship with the attorneys. The authors recommend that relationships between the social scientist and the attorneys be subject to clear and explicit agreements and that the role of the expert witness be defined so that the social scientist maintains control of any needed research. Potential expert witnesses should be aware of the unique research problems they may encounter and be prepared to deal with these obstacles. Throughout the research, the expert should be cognizant of the adversarial setting in which the results will be presented, and must carefully document and substantiate the research accordingly. Finally, the social scientist should be aware of the potential consequences of participating as an expert witness.     

PERCEPTUAL DETERRENCE AMONG ACTIVE RESIDENTIAL BURGLARS: A RESEARCH NOTE*

The research reported here is based on a comparison of active residential burglars and a matched control group regarding their willingness to commit a burglary at varying levels of certainty of arrest, severity of penalty, and anticipated reward. Initial analyses revealed that few controls were willing to offend regardless of risk, penalty, or reward and that offenders were not influenced by penalty on its own. Consequently, responses of the offenders only were further analyzed in relation to the impact of risk, penalty, and reward. The results of a logit analysis indicated that both risk of being caught and prospect of increased gain had a significant influence on the offenders' decision making.