« Election Law and Compromise: Reactions to President Obama's Election Commission | Main | The Angsting Thread (Law Review Edition, Spring 2013) »

Wednesday, February 13, 2013

Where Top-Down Meets Bottom Up

My focus on lower-level criminal courts tends to orient my thinking towards bottom-up views of the criminal justice system.  That's not to say that I neglect the top-down stuff.  But a great article by Fergus McNeill and his collaborators (including Cyrus Tata, whose work I featured yesterday) identifies the point at which both approaches close in on each other, and attempts to fill one of the gaps that currently exists between the two processes.

The top-down approach that I'm most interested in is the "new penology's"  explanation of mass incarceration and criminal justice more generally.  According to the proponents of the new penology (including Jonathan Simon, Malcolm Feeley, and Bernard Harcourt) is the identification of actuarial aggregation and categorization as a central feature of the contemporary criminal justice system.  Using McNeill et al.'s work, I'm going to suggest that the actuarial isn't always a stable concept, but is instead sometimes a site of contestation among low-level court officers.  Put differently, the high-level category of risk is translated into usable terms by the different members—judge, counsel, social worker—that comprise the courthouse workgroup doing sentencing on the ground. 

The new penology emphasizes the criminal justice system's use of actuarial instruments to engage in data-gathering and statistical analysis.  As Bernard Harcourt describes in his magisterial study, Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age, in the last thirty years or so, policy makers and criminal justice officials (judges, probation officers, parole boards, and so on) have increasingly turned to these types of instruments to identify and control spaces, behaviors, and subpopulations in terms of their contribution to the crime rate.  Actuarialism is focused on predicting which individuals are most likely to engage in particular categories of criminal conduct.  The criminal justice system is then supposed to manage them through various forms of control—incapacitation of high-risk offenders and supervision of low-risk offenders.

The new penology tends to take a top-down approach: it depends upon looking at the development of policy and its uptake by criminal justice officials.  It tends to be less sensitive to the manner in which individuals at the bottom do not merely use, but repurpose these policies as a means of protecting their own norms of professional identity and conduct.  Thus, in addition to Bernard Harcourt's attack on actuarialism as generating a series of perverse effects that undermine goals of accurate prediction and rational regulation of risk, the bottom-up approach calls into question the whether the policies have actually penetrated low-level practice by challenging the stability of the central actuarial concept of risk.

The McNeill et al. study of lower-level criminal courts in Scotland exemplifies this gap in the literature.  As mentioned yesterday, McNeill, along with Tata and others, studied the lower-court's use of pre-sentence reports prepared by criminal social workers (the Scottish equivalent of probation officers) and found that the courthouse workgroup of judge, attorney, and social worker had established a regime of interpretation and misinterpretation  strikingly similar to Richard White's discussion of the "middle ground" carved out by the clash of British, French, and Native American cultures in the "pays d'en haut" in 18th Century America. (My analogy, not Tata's). 

What's fascinating about both studies is the manner in which, in McNeill & co.'s case, a legislatively mandated document, the pre-sentence report, serves different functions dependent upon the purposes, professional interests and self-understandings of the individuals using the report.  In his description of the cross-purposes governing the understandings and uses of pre-sentence reports, McNeill & co.'s study in and around Glasgow, Scotland (they use two sites) is eerily reminiscent of Malcolm Feeley's work in New Haven, Connecticut. 

Of particular interest for me was the finding that judges and social workers translate the risk assessment protocol, one of the many described by Harcourt, into (in my crude terms) a trust-assessment protocol.  That is, the social workers (who, during the study, were being evaluated on their use of the protocol and so were intensely focused on it) measured risk in terms of their ability to have a meaningful relationship with the offender, one in which they were able to interact with, influence, and control the individuals under their charge.  In making their recommendations, the social workers recognized that they were putting their reputations on the line—and were sensitive to the marginal status of those reputations in the criminal justice system.  Rather than simply engage in a prediction of categories of re-offense, the social workers were more interested in which McNeill & co. call "responsibilization": the extent to which offenders could be rendered amenable to social control. 

Similarly, though to a lesser extent, the judges were concerned with their professional reputation as well, measured in terms of the consequences of letting these offenders back into the community.  But in maintaining a posture of authority and neutrality, they demanded that the pre sentence report present a strictly factual and non-adversarial account of the offenders' personal circumstances, and a detailed account of their criminal record.  The notion of risk that the judges were working with again appears modulated through their professional role, and different from both the announced policy and the messages that the social workers sought to impart through the pre sentence report document. 

This work is fascinating, especially for those of us who are interested in the notion of holistic legal practice.  It suggests that holism requires more than the inclusion of different professionals or practice styles into the workgroup or legal practice.  Rather, it requires skills of self-reflection, communication and negotiation among different sources of knowledge, with attention to the ways in which the lingua franca—in this case, of risk assessment in pre-sentence reports—is subject to misinterpretation and repurposing in order to maintain rather than eviscerate cultural differences between the participants. 

A central site for the discussion of holistic legal practice is in the wealth of literature produced by those most directly engaged in it: judges, clinical faculty, and so on.  It's also the sort of practice uncovered by ethnographies (sociological, criminological, anthropological and comparative) of the courtroom or the law firm—both in its "good" and "bad" forms.  This practice-generated or practice-focused scholarship is an essential part of developing a pluralistic understanding of what legal practice entails in the middle ground of the low-level court.

Posted by Eric Miller on February 13, 2013 at 12:03 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Where Top-Down Meets Bottom Up:


Thanks for the interesting comment; the instrument under discussion in both Harcourt and McNeill et al. is the LSI-R: "developed in Canada in the late 1970s and is used today in nearly all of the United States and the Canadian provinces at some point in the postconviction process." Bernard E. Harcourt. Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age. Harcourt describes this as measuring risk of reoffending.

Posted by: Eric J. Miller | Feb 13, 2013 12:50:03 PM

There may be a distinction between risk assessment instruments that predict ability to sucessfully complete probation (e.g., RANT) and those that predict dangerousness. Probation departments may use both, either or, rarely, neither.

Posted by: Judge Peggy Hora (Ret.) | Feb 13, 2013 12:45:17 PM

The comments to this entry are closed.