As we observe the various states trying to get handles on their current and even worse looming corrections sentencing problems, considering in particular sentencing commissions and guidelines, we don’t hear much about the original reason for development of guidelines and commissions—ending sentence disparity, making sure that like offenders got like offenses and unlike got unlike. It’s there a little, but the overwhelming concern right now is to match policy with increasingly finite resources.
I don’t consider that decline of concern a horrible thing. That doesn’t mean I’m a fan of sentence disparity. I’m the opposite. It’s because I haven’t been convinced for a long time that we accomplish what we’re after in most cases by institutionalizing structures to inhibit disparity when we do guidelines. Most guidelines systems rely heavily if not totally on criminal history as the “offender score” in their grid systems and to apply enhancers and mandatory minimums. I’ve worked for over a dozen years now with criminal history records, I’ve seen how they’re developed, and I think there is, even today, too much wrong with the collection and reporting to justify the faith that guidelines put in them.
I’m not saying I disagree that crim hist is the most important factor in most cases predicting future crim hist in the general case. What I’m saying is that I’m not confident that we end sentence disparity with the criminal history records we have and on which we base our sentencing decisions. What I’m saying is that it’s still too possible for two identical offenders with identical lifetime criminal behavior to be sentenced in the identical system differently, disparately.
A couple of quick reasons. One, the record collection and reporting, while better than a decade ago, are still too ambiguous. I talk regularly to people who still laugh and shake their heads at the idea that a rap sheet can be interpreted as a science rather than an art or that they believe it’s completely accurate when read correctly. You need good disposition reporting and maintenance to make them real, and people trained to get it done right and also to even care. Once the guy goes away, why worry about updating anything? Two, some DAs charge everything and stay with it, some charge everything and bargain away some, some charge just one or two things to keep everything straight and simple. So three identical guys doing the same criminal things could conceivably end up with three different prior records. And, the next time they come through for, again, the same thing and those prior records are applied to a grid or a mandatory, they will get different sentences even though they have never in their lives been or done anything but exactly the same. And this doesn’t even get into differential charge bargaining.
Your statistics showing that more priors predict future crim behav best would still be good, assuming, reasonably, that the problems I describe are spread evenly across the offender pool. Yet, in application of guidelines or mandatories based on them in individual cases, you could end up with significant disparity, even as your analysis told you the disparity didn’t exist. Minorities have argued for years that that’s one of the chief ways they get hammered disproportionately in guidelines systems, and, absent better studies of what happens in DA offices, I’m hard pressed to tell them they’re wrong. Guidelines can definitely give the appearance (and reality, I’m afraid) of bias, and that undermines their legitimacy, support, and value.
That’s why I would argue, if guidelines are being done, for a more inclusive offender score, one that counts criminal history but goes beyond that to score points for mitigators and aggravators and for risk assessment tools (not risk of violating probation, but real risk of future re-offending) that so many judges are now advocating upfront before sentencing occurs. A more inclusive score might be more of a hassle to calculate, but it’s already done in some states, like MD, and can incorporate a lot more of the factors judges and the public consider necessary to actually get at the proper sentence. Later evaluation and statistical analysis can be used to refine and tailor the resulting sentencing system to ensure that like sentences yada yada while also testing their predictive ability of future recidivism and public safety risk.
So, if you’re thinking commission and guidelines, do your best to get the best info on how well your crim hist system is actually capturing reality before you start using it to form those guidelines. And please use this open window of opportunity to bring in other measures that can improve and enhance the offender scores you apply. In the end, you may just end up with more public safety, less disparity, and more dollars to put back into other criminal justice needs all at the same time.