IS BAD LAW AND BAD POLICY
By John Weston Parry, J.D.
In a relatively recent New York Times article Michigan law professor, Sonja B. Starr, wrote an excellent op-ed piece on what she termed “sentencing, by the numbers.” She joined Attorney General Eric Holder in criticizing the “growing trend of evidence based sentencing” in which judges determine length of detention based on “data-driven predictions of defendants’ future crime risk….” This subjective and impressionistic judicial practice is being incorporated more widely into American criminal law in the guise of objectivity and greater reliability. Supporters of this sentencing trend reflexively claim “it’s a no brainer. Who could oppose `smarter’ sentencing?” Twenty states already use this approach and “some…require risk scores to be considered in every sentencing decision.” As many additional states contemplate adopting this practice in more formal ways, Congress and the U.S. Sentencing Commission will decide whether it should be introduced into the federal courts, and if so how.
Unfortunately, much of this sentencing “data” is nothing more than profiling based on socio-economic factors that unfairly target those who society devalues, including “poor people and minorities.” Professor Starr points out that this approach “sends the toxic message that … certain groups of people are dangerous based on their identity.” It “confirms the widespread impression that the criminal justice system is rigged against the poor,” a group which includes a disproportionately high percentage of persons with mental disabilities. Starr also identifies serious constitutional issues in using this impressionistic information in sentencing determinations. “The Supreme Court has consistently held that... impermissible discrimination cannot be justified by statistical generalizations about groups….”
With very few exceptions, it is a fallacy in logic to use even empirically-vetted generalizations about groups in order to draw specific conclusions about individuals. It is much worse when the generalizations prove false or misleading. Until recently in the legal system the lone exceptions to this constitutional and logical limitation have been generalizations that apply to groups of people with mental impairments when such judicial profiling is cloaked as risk assessments and other expert and judicial impressions about future dangerousness. These dangerousness generalizations have been employed in every state and every federal jurisdiction against the interests of defendants and respondents with mental impairments in a wide variety of criminal, quasi-criminal, and civil proceedings. Rather than curtailing such intellectually bankrupt practices, it appears that legislatures and courts are poised to dramatically expand their reach in the name of progress.
Evidence-based sentencing threatens to employ a variety of physical, mental, attitudinal, and socio-economic factors, in addition to criminal history and mental disability, in order to predict whether or not convicted defendants will create future risks to society. This brings us much closer to the science fiction nightmare in which risk predictions are employed to sentence people for crimes before they happen. Much of what are misleadingly characterized as “evidence-based” conclusions—as a practical matter—are predominantly biased opinions or presumptions with little or no scientific or empirical basis. In essence, we will punish, monitor, and intrude upon the lives of defendants based not on what they have done, but what they might do or be more likely to do.
Over many years, empirical and other objective measures have demonstrated that using psychiatrists, clinical psychologists, other experts, judges, and/or juries to predict the dangerousness of people who have mental impairments based on various risk and related factors is highly misleading, unreliable, and inaccurate. Nonetheless, this type of flawed, unfair, and discriminatory evidence and testimony has been deemed admissible in criminal and civil courts because the judiciary has created “legal fictions” that allow them to be used, despite their obvious inadequacies and inequities. These risk assessments and other unreliable impressions about future human behaviors are politically popular because it is conveniently presumed, without empirical justification, that they enhance public safety and their targets tend to be individuals that our society devalues.
What until recently was limited to discriminatory legislative, regulatory, and judicial practices intended only for people with mental disabilities may soon include everyone in America who shares specified characteristics that federal, state, or local governments believe—without substantial proof—is likely to increase the risk of future crimes, violence, or other antisocial behaviors. Not surprisingly, given how our society tends to function, those who possess mainstream characteristics or values will be favored, while those who have characteristics or values that are presumed to be undesirable or less desirable will be in jeopardy. In the process, the United States judicial system will have successfully transformed racial and ethnic profiling, which is generally illegal, into permitted socio-economic, mental, and attitudinal profiling that promises to keep most of America’s “troublemakers” incarcerated longer and with highly intrusive monitoring and supervision should they be released. It also will promise to grant earlier freedom to those who are more highly valued in order to depopulate our overcrowded prisons and jails.
Our judicial system will be used to try to ensure that those offenders, who share these unwanted characteristics—such as having family members with criminal histories, being a high school drop-out, being unemployed, not being legitimately married, being raised in a “bad” neighborhood, following a potentially terroristic religion or belief system, being or having been a member of a group which is designated as a “gang,” or viewing what are considered to be unacceptably pornographic, violent, or antisocial programs or materials—will be treated like offenders with mental disabilities. The criminalization of person with mental impairments will be expanded to become the criminalization of anyone who is socially devalued or deemed to be deviant.
While the “slippery slope” argument often is misused and misapplied, it seems especially appropriate here as we embark on this sentencing evolution. Do we really want unreliable and inaccurate predictions of risky or dangerous future behaviors—based on physical, mental, social, economic, and attitudinal characteristics—to largely govern what happens to offenders in our criminal justice system? Evidence-based sentencing is little more than a promotional slogan. The emperor has no clothes. Any evidence based on this type of prediction should be deemed inadmissible in the courtroom (or as the basis of any policy decision) until its specific relevance, validity, and reliability can be established with clear and convincing empirical and other social science-based findings. Also, any type of profiling of this kind should be closely scrutinized to ensure that it does not discriminate based on constitutionally protected personal characteristics. What we confidently accept as common sense too often proves to be nonsense or improper discrimination when it affects people we devalue, fear, or distrust.
The many flaws in using this type of evidence-based sentencing have been repeatedly demonstrated with respect to people with mental disabilities. Its use in this way has resulted in a criminal justice system in which offenders with mental disabilities serve longer sentences than other prisoners, despite their reduced culpability, and are deprived of the very treatment, rehabilitation, and other services they need to improve and avoid becoming worse. How this travesty of justice was created and operates in this country is detailed in Mental Disability, Violence, Future Dangerousness: Myths Behind the Presumption of Guilt (John Weston Parry, Rowman & Littlefield, October 2013).
One can only hope that law- and policymakers will scrutinize how deficient risk assessments and other subjective impressions about dangerousness have been, when used against people with mental impairments, before we decide to apply similar methods to almost everyone who enters the criminal justice system. Using predictions of future dangerousness to incarcerate and deprive individuals with mental disabilities of their fundamental rights is a model that deserves condemnation, not replication and expansion.