Empirical tests of the deterrence doctrine have suffered from two deficiencies: (1) they have rarely been conducted within the context of a more general theory of social control, and (2) they have confounded the effects of sanction threats on different kinds of offending decisions. In this paper, deterrence variables (perceived certainty and severity) were incorporated into a rational choice model of common (minor) delinquent offending. The rational choice perspective presumes that crime and delinquency are the products of imperfectly informed choice and that decisions to offend are made, assessed, and remade. It was assumed here that decisions to offend and cease offending would in part be affected by the contemplation of sanction threats in addition to material, affective, and moral considerations. The data in support of the inclusion of deterrence variables in a rational choice model were equivocal. Although perceptions of certain punishment had a significant effect on the decision to commit and desist from some offenses, the observed effects were marginal and far less consequential than were the effects of social costs and other considerations. The data did indicate that certainty of punishment (but not severity) affects some offending decisions.

Lawyers in cases involving injunctions against picketing represent clients in situations of great immediacy. A significant number of injunction actions are settled with reductions in picketing despite a seemingly restrictive statute and a highly organized workforce. This study of legal culture examines the role of lawyers in striving to create predictability, especially in regard to judges and the police, and in transforming conflicts of value into disputes over interests that can be resolved without resort to formal adjudication.

The research presented here examines the effect of legal and extralegal resources on the likelihood of female-perpetrated homicides. It looks at the relationship between gender-specific rates of partner homicide in the United States for 1976–79 and 1980–84, and the availability of legal and extralegal resources for abused women, using states as the unit of analysis. Results indicate that the availability of such resources is associated with a decline in the rates of female-, but not male-, perpetrated homicides in 1980–84, compared to the 1976–79 time period. These findings suggest that legal and extralegal interventions can provide nonviolent alternatives for victims of male partner abuse.

Recent studies by Weitzman (1985) and Peters (1986) suggest that no-fault divorce has harmed women. The research presented here tests these findings by examining the effects of no-fault divorce on the financial situation of women using data on salary and wage income, home ownership, and child support from the National Longitudinal Surveys (NLS) of Labor Market Experience (Ohio State University Center for Human Resource Research [OSU], 1986) young women's cohort. The analyses do not support the hypothesis that no-fault divorce produces adverse financial effects. Rather, it appears that no-fault has had little effect on women's finances. Alternative models that might explain how divorce law affects the financial condition of women are suggested.

Experimental research has recently shown a powerful impact on legal policy. An experiment demonstrating a deterrent effect of arrest on domestic violence has shaped public policy. Stimulated by efforts to publicize the results of the Minneapolis Domestic Violence Experiment (Sherman and Berk, 1984a), police departments were persuaded to adopt an arrest policy for misdemeanor domestic violence. Over one-third of respondents from U.S. police departments in 117 cities said their policy had been influenced by the experiment, although respondents from some departments that adopted an arrest policy did not recognize the experiment or its results. Lempert (1987, 1984), citing medical research as precedent, suggests that this impact is premature and inappropriate until replications are completed. However, we find no indication that medical research employs a standard of delaying adoption of research results prior to replication. Our analysis suggests that publicity can encourage replication of legal research at other sites and thus improve the knowledge base for policy recommendations.

In their paper in this issue Sherman and Cohn (1989) respond to arguments I made in this journal about when research is ripe for publicity. This paper continues the conversation by pointing out that policy-relevant research may be publicized at three levels: (1) in professional journals, (2) directly to those practitioners whose practice decisions might be informed by the research results, and (3) through the mass media. I then argue that the reliability of the results, the ability to communicate main findings precisely, and the likely effects of publicity are keys to responsible publicization at all levels. In weighing these factors researchers must regulate themselves, but to guide them the law and society community should seek to develop professional norms. I comment on some of the considerations that should guide such norms, using the Minneapolis Domestic Violence Experiment and the publicity it received as an example.