Therapeutic jurisprudence is not intended as a panacea for all that is wrong with the legal system. It suggests law reform in a step-by-step process, based on empirical behavioral research and legal analysis of therapeutic and antitherapeutic effects. Law reform can occur through changing a procedure or by actually changing laws. If laws are to be changed, then empirical research can serve as a partial basis for the change, based on how legal rules, legal procedures, and legal actors contribute to or detract from the emotional well-being of concerned parties. This is not anticipated to be an easy or clear-cut process. We can look to previously learned lessons in the social science literature for caveats to consider in this process.

The hope of law reform could be hindered by two extremes: either too quickly incorporating research findings without clearly understanding their implications or by not incorporating findings at all. Sherman and Berk's (1984) seminal work in arrest in domestic violence provides an apt example of research being translated into policy too quickly, based in large part on the findings of one study. Before the National Institute of Justice could complete six additional replication studies (Berk et al., 1992; Dunford, Huizinga and Elliott, 1989; Hirschel et al., 1991; Pate and Hamilton, 1992; Sherman et al., 1992), none of which fully reproduced Sherman and Berk's work, arrest policies across the United States had dramatically changed to incorporate various forms of mandatory arrest (Buzawa and Buzawa, 1996). The additional studies uncovered the complexities of arrest, including the heightened risk it posed to certain groups of domestic violence victims (NIJ Research Report, 1998). These findings emerged long after policy changes had already been implemented. Sherman and Berk's (1984) study was not solely responsible for a nationwide move to mandatory arrest, but it was interpreted as powerful empirical evidence advocating it, without fully considering the unintended consequences that emerged in the other studies. Could this now controversial influence of one study have been averted?

One answer can be found in the social science and law scholarship. Monahan and Walker (1986, 1998) set out four parameters for courts to evaluate social science research. They stress the importance of utilizing empirical studies that have been critically reviewed by the scientific community, that have utilized valid research methods, that have generalizable findings, and that are supported by a body of research. If lawmakers had followed these steps, when the mandatory arrest for domestic violence laws were changed, evaluation of the research might have been questioned perhaps more than it was on the grounds that the research was not supported by a body of evidence—meaning it had not been replicated prior to the decision to change the laws. It is true that several other political factors influenced the change in the domestic violence arrest laws and the research was clearly not the only factor (Buzawa and Buzawa, 1996). However, at least to the extent that the research was used as evidence for change in the laws, its connection might have been more carefully analyzed. Based on Monahan and Walker's recommendations, one study, even a rigorous one, is not enough to change policy. They actually make the point that they expect that less social science research will be integrated into law if their recommendations are followed precisely because only rigorous research that has withstood the test of time would be considered by judges. This approach risks erring on the side of conservatism, however, since replicating studies and publication of findings can take years. Meanwhile, judges must make decisions that impact defendants, victims, and communities on a day-to-day basis.

The mandatory arrest example reveals the ambiguities that must be faced as the integration of empirical research findings into law reform is considered. Timing and context become key considerations. Empirical work can occur prior to a legal reform, as Monahan and Walker suggest. However, it can also occur simultaneously with a policy change. Incorporating the context, or integrating the legal process in its entirety within one study, also becomes important, particularly for purposes of generalization. Studies on domestic violence arrest and coordinated court approaches provide examples of both of these issues. The arrest studies failed to produce a uniform conclusion as to whether arrest was a deterrent (NIJ Research Report, 1998). It was not until research designs incorporated additional components of the arrest process, including sentencing and probation (Thistlethwaite, Wooldredge and Gibbs, 1998) that a consensus began building in the research literature around the usefulness of arrest if it was followed by conviction and an order for court-ordered treatment (Steinman, 1990; Syers and Edleson, 1992; Tolman and Weisz, 1995). Considering this larger context becomes a key factor for generalizing findings to real-world practice settings. The arrest studies initially looked at only one piece of a domestic violence offender's experience—that is, the arrest—without considering how prosecution, jail time, probation, or offender characteristics may affect offender outcomes. Later work did a better job, incorporating a greater portion of the context of an offender's experience from arrest to post-conviction, and that research seems to be standing up to the test of time as far as consistent findings. In addition, these studies could not have occurred if court reforms had not already taken place. Coordinated community responses in domestic violence have been implemented around the nation, though not on a large scale basis (Keilitz, 2000). Conducting research in these settings simultaneously with their new implementation has permitted researchers to provide useful findings that can build on or confirm what judges in these settings are learning. This research can then be accessed by those about to embark on similar projects, and becomes useful prior to implementation. Empirical research does not have to pre-date law reform, but can work simultaneously with changes in the legal process, particularly when they occur on a smaller scale.

Turning to the opposite scenario of research being overlooked, the previous discussion has shown how research can be left out of the process of law reform through a conservative approach of waiting for consensus from multiple rigorous studies—a process that can take easily 10 years or more. It can also be left out if it contradicts a study already referenced by a higher court. To avoid this scenario, Monahan and Walker (1986) suggest that research be considered similar to legal precedent. Legal precedent changes, and so does research. This allows lower courts to introduce new research, or to bring in empirical studies that contradict existing research that a higher court might already have used. In this way the door remains wide open between research and courtrooms rather than funneling down to only those studies that are accessed by the higher courts.

The relationship between research and policy implementation can be a tenuous one, at best, particularly in criminal justice. This makes the need for integration of pertinent, meaningful research findings as early as possible that much more pressing. Having a process in place to do this can only improve the relationship between social science research and law.

0 0

Post a comment