The Future of Violence Risk Management
Injustice and inefficiency invariably flow from any blending of the criminal-law and mental health powers of the state. Each is sufficient unto itself to achieve a just balance between freedom and authority; each has its own interested constituency; when they are mixed together, only the likelihood of injustice is added. (Morris 1982, p. 31)
Despite this prescient warning, a blending of the criminal justice and mental health systems is exactly what is happening in the United States, and what may be on the verge of happening in the United Kingdom as well. On the mental health side, the blending has taken the form of a resurgent emphasis on the state's police power as the principal justification for the civil commitment of people with serious mental disorder who have committed no crime but who are believed to be “dangerous to others.” The result has been the first significant loosening of the 1960s-era libertarianoriented standards for involuntary treatment as a hospital inpatient, and the first meaningful laws ever that allow for involuntary treatment in the community. On the criminal justice side, the blending has come as statutes have proliferated, and been upheld, authorizing the indefinite civil commitment in a mental hospital of convicted offenders who are believed to have a “mental abnormality”—which need not rise to the level of a “mental disorder”—and are found likely to recidivate. To date, these statutes have been limited to offenders convicted of sexually violent offenses, but proposals have been introduced in the United States and in the United Kingdom to expand them to the perpetrators of any form of violent act.
Both of these moves toward convergence have been facilitated by developments in the professional technology that enables their operation: violence risk assessment. After two decades of largely depressing research on the abilities of mental health professionals at predicting violence, the journals are suddenly replete with guardedly optimistic studies indicating that actuarial instruments may succeed where clinicians have failed. For the first time, relatively accurate assessments of the risk of violence that