In the previous blog post I discussed several arguments put forward by legal scholar Sonja Starr against including socioeconomic factors such as unemployment, low or lack of education, and homelessness in risk assessment instruments used for informing sentencing decisions.
Here is another argument, put forward by legal scholar Michael Tonry in an article titled ‘Legal and Ethical Issues in the Prediction of Recidivism’:
Tonry reiterates widely supported normative and ethical rules such as ‘don’t treat people differently based on the basis of social class’, that are ‘largely incompatible’ with sorting people into risk categories. Tonry describes how, in the US, in the 1970s federal parole guidelines initially allowed variables such as employment, education, residential status and family characteristics, but that these factors were gradually abandoned because ‘they are heavily correlated with race’. The 1991 parole guidelines do not include education, employment or family characteristics.
A number of ethical and normative propositions emerged from the sentencing and parole initiatives of the 1970s, among which
5-Personal characteristics such as education, employment, residential stability, and family circumstances should not ordinarily be included among aggravating sentencing and parole criteria because their use systematically disadvantages black and other minority offenders (although they may be used as mitigating factors).
It is unclear to me why it apparently is permissible that including these factors systematically disadvantages people who have no or little education and people who are jobless or homeless (or in instable jobs or housing, depending on how the criteria are formulated).
However, Tonry goes on to denounce the inclusion of socioeconomic factors and ‘factors that in effect punish lawful life-style choice that in a free society people are entitled to make’ such as ‘decisions to marry, have a family, take further education, have a regular job, and live in stable residential circumstances’ (p.8). In addition to concern about racial disparities, legal scholars argued that
it is unjust and undesirable in a society that celebrates Enlightenment values of freedom and personal autonomy to penalize citizens for making quintessentially personal choices about such things as marriage, education, work, and living arrangements. (references to Andrew Von Hirsch, 1976, 1985)
In so far as they are indeed life-style choices (and not at least partly the result of structural disadvantages) they are legitimate choices to make. Therefore, no-one should be punished for making these choices.
It is at least illiberal to attach incremental penal consequences to those choices.
While it is problematic to argue that they are purely life-style choices (Tonry indeed acknowledges that many offenders do not choose to be poorly or unemployed, poorly housed, or poorly educated), this is actually the strongest argument to make against including socioeconomic marginality in risk assessment instruments: it is not a crime to not work, not pursue higher education, or move around a lot, and thus such characteristics should have no place in sentencing decisions.
Image by Birdman Photos on Flickr