Pennsylvania is on the precipice of becoming one of the first states to base criminal sentences not only on the crimes people have been convicted (at the present and in the past), but also on the likelihood that they may commit additional crimes in the future. As soon as 2016, Pennsylvania judges could receive risk assessments on a particular convicted criminal to help the court decide how long of a prison term- if any- should be ordered.
Risk assessments try to predict if a criminal will repeat offend or break the rules of parole or probation by using statistical probabilities based on factors such as sex, age, prior criminal record and employment history. Such assessments are now used at some point in the criminal justice process in almost every state. Up until now, however, assessments have never been used in the sentencing itself. Pennsylvania aims to do just that.
Obviously, very basic philosophical questions are raised when considering whether to sentence someone based on a crime he/she has not yet committed. However, Pennsylvania has managed to largely sidestep the issue. Mark Bergstrom, the executive director of Pennsylvania’s Sentencing Commission, says it is not up to him whether the state should use a risk assessment analysis when sentencing criminals, since the legislature has already voted to do so. Bergstrom’s job, he says, is to figure out what the assessment will look like and how it will be implemented.
Pennsylvania’s Sentencing Commission is trying to stave off inevitable lawsuits filed by defense attorneys by designing its risk assessment tool to use only information that comes from databases, as opposed to interviews. Both prosecutors and defense lawyers alike can see all of the information that goes into the assessment’s scoring system and will have an opportunity to verify its accuracy and to request changes if something is incorrect.
Even many supporters of risk assessment tools believe they can be problematic in practice. Official records can contain mistakes and many tools include subjective questions that require the person filling out the questionnaire to characterize the offender’s attitudes and feelings. Such a process can introduce error.
The central debate surrounding the use of risk assessment tools in the sentencing phase is what the goal of criminal justice reform actually is. Some supporters see reducing jail time as the primary goal; others want to focus on reducing repeat offenses; and others want to reduce and eliminate racial disparities. In theory, risk assessments can accomplish these goals. However, once they are used in the real world, it is unknown how well they will actually work.
Bergstrom knows he is walking a delicate line. The commission’s research has found that prior arrests are actually a better predictor of repeat offenses rather than prior convictions. However, using arrests would almost certainly draw constitutional challenges. Public defenders will likely point to the racial disparities in arrest rates and claim that it is illegal to presume someone is guilty just because of a prior arrest.
As of now, Bergstrom says that he is leaning towards using a risk assessment tool to identify criminals on the extreme ends of the spectrum: sentence no prison for low-risk individuals and flag high-risk individuals for extra treatment or prison time. While it would be a limited approach to utilizing the assessment tool, it still would not avoid the question of whether criminals should spend a longer time in jail simply because a statistical tool says they will behave a certain way in the future.