(Fair Warning- this is long.)
I had no idea this was going on, but it is terrifying:
IN a recent letter to the United States Sentencing Commission, Attorney General Eric H. Holder Jr. sharply criticized the growing trend of evidence-based sentencing (.pdf), in which courts use data-driven predictions of defendants’ future crime risk to shape sentences. Mr. Holder is swimming against a powerful current. At least 20 states have implemented this practice, including some that require risk scores to be considered in every sentencing decision. Many more are considering it, as is Congress, in pending sentencing-reform bills.
Risk-assessment advocates say it’s a no-brainer: Who could oppose “smarter” sentencing? But Mr. Holder is right to pick this fight. As currently used, the practice is deeply unfair, and almost certainly unconstitutional. It contravenes the principle that punishment should depend on what a defendant did, not on who he is or how much money he has.
The basic problem is that the risk scores are not based on the defendant’s crime. They are primarily or wholly based on prior characteristics: criminal history (a legitimate criterion), but also factors unrelated to conduct. Specifics vary across states, but common factors include unemployment, marital status, age, education, finances, neighborhood, and family background, including family members’ criminal history.
Such factors are usually considered inappropriate for sentencing; if anything, some might be mitigating circumstances. But in the new, profiling-based sentencing regimen, markers of socioeconomic disadvantage increase a defendant’s risk score, and most likely his sentence.
Advocates of punishment profiling argue that it gives sentencing a scientific foundation, allowing better tailoring to crime-prevention goals. Many hope it can reduce incarceration by helping judges identify offenders who can safely be diverted from prison.
While well intentioned, this approach is misguided. The United States inarguably has a mass-incarceration crisis, but it is poor people and minorities who bear its brunt. Punishment profiling will exacerbate these disparities — including racial disparities — because the risk assessments include many race-correlated variables. Profiling sends the toxic message that the state considers certain groups of people dangerous based on their identity. It also confirms the widespread impression that the criminal justice system is rigged against the poor.
It is naïve to assume judges will use the scores only to reduce sentences. Judges, especially elected ones, will face pressure to harshly sentence those labeled “high risk.” And even if risk scores were used only for diversion from prison, it would still be wrong to base them on wealth and demographics, reserving diversion for the relatively privileged.
Evidence-based sentencing also raises serious constitutional concerns. The Supreme Court has consistently held that otherwise-impermissible discrimination cannot be justified by statistical generalizations about groups, even if those generalizations are on average accurate. People have a right to be treated as individuals, and individuals often do not conform to group averages.
Quite obviously, this is insane. What they are doing is essentially using a statistical analysis to determine who is most likely to commit another crime, and then sentencing them accordingly. Score well on the statistical analysis, and by scoring well, show positive signs in the so-called objective criteria that they have chosen to review, and your sentence will be lighter. Score poorly, your sentence will be harsher because the analysis shows you might have a propensity for recidivism and must be incarcerated longer. And you thought pre-crimes were just for sci-fi books and movies!
Many, many moons ago when I was a young Political Science major, I spent a summer interning at a probation office. Like all aspects of the judicial system, it was radically underfunded, and there were nowhere near enough probation officers for the number of “clients” we had to attend to and the requirements passed down for contact with the PO after sentencing. Additionally, the Probation Office had a number of other tasks that we performed for the court, and one the most laborious tasks was conducting and writing pre-sentence investigations. Since I liked to write, was the only male and military (a lot of these guys were rapists and abusers and molesters and they scared the younger women in the office), I took over those duties for the summer. I was actually so good at it that they hired me on in the fall to keep doing them because they had such a backlog.
At any rate, what I would do is grab my notebook and head next door to the County Jail, and I would interview people who had been convicted of a crime but were yet to be sentenced. The interviews could take up to two hours, and I would ask them about their background, their education, their family life, how they found themselves in the situation and why they did what they did, past histories of sexual, physical, and substance abuse, and so on. I would then head back to the office, type it all up (they were usually 15-20 page documents, sometimes shorter, sometimes longer, depending on how much the inmates would open up to me), keep a copy for our files, pass one off to the Judge, the DA’s office, and the defense team. They would then use that information to get to know more about the person and their past and state of mind, and the judge would use that to help him formulate the sentence.
Quick sidebar- one of the judges I worked closely with was Larry Starcher, who later was a member of the state Supreme Court. He was a liberal Democrat, and I was at the time a blossoming
fascist Republican, so we had differing opinions on a lot of things, but I really, really liked the guy and it’s funny how much I learned from him and still remember. I remember one case in which I did the pre-sentence investigation, and it was a child molester who had been convicted of multiple accounts of abuse on victims ranging from infant to pre-teen. He was a wretched human being- everything about him exuded filth and evil and bad intent, and if his body odor had not been so horrific I bet I would have smelled sulfur. The interview was particularly contentious, with him issuing various threats and just generally being an asshole. I have always been against the death penalty, but this guy tested my principles.
At any rate, his sentencing came, and I was sitting in court (another PO obligation is one of us had to be in court for sentencing), and a seemingless endless stream of family members testified about what a wonderful man the convicted was and that the Judge should be lenient. I just sat there chuckling, wondering if we were talking about the same person. Judge Starcher sentenced him (harshly, and Starcher was always a judge who would be willing to work with alternate sentencing and had a reputation among the right as being too lenient, but he nailed this guy’s balls to the walls). After the sentence, we were chatting, and I mentioned that I just could not believe that all those family members had the lack of shame to testify about how swell he was when he had molested kids in the family. Starcher paused, looked at me, and said “If your family isn’t going to come to bat for you, who will,” basically challenging me to look at the convicted from a different angle that maybe I had not seen.
That always stuck with me.
Back to the point. I have no problem with looking at a person’s history and who he is and what he has been and using that information to coming to an appropriate sentence. But it is subjective, and each person is treated as an individual and a human being, not merely the sum of collective data points. That’s what is happening here, and it is quite terrifying.
Additionally, as AG Holder noted (.pdf) (and should be blatantly obvious to everyone), this is problematic in more ways than one:
First, most current risk assessments – and in particular the PCRA, which is specifically mentioned in the pending federal legislation – determine risk levels based on static, historicaloffender characteristics such as education level, employment history, family circumstances and demographic information. We think basing criminal sentences, and particularly imprisonment terms, primarily on such data – rather than the crime committed and surrounding circumstances – is a dangerous concept that will become much more concerning over time as other far reaching sociological and personal information unrelated to the crimes at issue are incorporated into risk tools. This phenomenon ultimately raises constitutional questions because of the use of groupbased characteristics and suspect classifications in the analytics. Criminal accountability should be primarily about prior bad acts proven by the government before a court of law and not some future bad behavior predicted to occur by a risk assessment instrument.
Second, experience and analysis of current risk assessment tools demonstrate that utilizing such tools for determining prison sentences to be served will have a disparate and
adverse impact on offenders from poor communities already struggling with many social ills. The touchstone of our justice system is equal justice, and we think sentences based excessively on risk assessment instruments will likely undermine this principle.
Third, use of risk assessments to determine sentences erodes certainty in sentencing, thus diminishing the deterrent value of a strong, consistent sentencing system that is seen by the community as fair and tough. Our brothers and sisters in the defense and research communities have repeatedly cited research to the Commission about the value and efficacy of certainty of apprehension and certainty of punishment in deterring crime. Swift, certain and fair sanctions are what work to deter crime, both individually and across society. We know that certainty in sentencing – certainty in the imposition of a particular sentence for a particular crime, and certainty in the time to be served for a sentence imposed – simultaneously improves public safety and reduces unwarranted sentencing disparities. We are concerned that excessive reliance on risk tools will greatly undermine what has been achieved around certainty of sentencing in the federal system.
Determining imprisonment terms should be primarily about accountability for past criminal behavior. While any effective sentencing and corrections policy will take account of
future behavior to some extent – incapacitating those more likely to recidivate and utilizing effective reentry efforts to reduce the likelihood of recidivism – we believe the length of imprisonment terms should mostly be about accounting for past conduct. As analytics evolve, we are concerned about the implications of sentencing policy moving away from this
It’s really hard to believe that prison “reformers” could come up with something worse than mandatory minimums or three strikes you’re out, but they did. This is a problematic process for any number of reasons.
1. It separates punishment from the crime, which should be the focus of sentencing. Again, IANAL, but I thought we punished people for the crimes they committed, not the crimes they might commit or the sins of their family and friends. Additionally, the old axiom “You do the crime, you do the time” is now “You do the crime, you do the time, depending on who you are.” This has always been the case in some form or other, and that has always been a bad thing. This, however, takes the previous subjective disproportionate sentencing that we have tried to eliminate (racial disparities in sentencing (.pdf), etc.) and essentially codifies it, giving a veneer of objectivity.
2. The veneer of objectivity is just that- a facade. Anyone who has ever worked with social science data sets knows full well that the results that come out of any analysis depend on the data you enter, so depending on what variables you use to determine the propensity to commit future crime, you could come up with wildly varying results. There is even a saying for it- garbage in, garbage out,” and you all have heard “Lies, damned lies, and statistics.” It’s why even when using the simplest form of survey questionnaire all the items are pored over and why even the simplest measure, say something like the PRCA-24 (Personal Response of Communication Apprehension), is refined and refined and all the items are factor analyzed so that you are sure you are measuring what you actually think you are measuring and that all the items of the measure correlate internally (long time since I have done this, so my terminology may be off).
According to the Center for Sentencing Initiatives, there are a number of different instruments being employed currently, and this is a sample of the criteria considered in so-called evidence-based sentencing are:
The COMPAS Core assessment for adult offenders contains both static and dynamic factors. Content may be individually tailored based on jurisdictional needs and resources, but can include 4 risk and 4 need scales:
Risk: failure to appear, non-compliance (technical violations), general recidivism, violent recidivism
Criminogenic needs: cognitive, behavioral, criminal associates/peers, criminal involvement, criminal opportunity, criminal personality, criminal thinking (self-report), current violence,
family criminality, financial problems, history of non-compliance, history of violence, leisure/boredom, residential instability, social adjustment, social environment, social
isolation, socialization failure, substance abuse, vocation/education
How smart you are, who your friends are, who your family members are, how much education you have, how much money you have, your employment history, etc.
As we know, poor and minorities are likely to have less educational attainment, fewer job prospects, lower incomes and less accumulated wealth, etc., so using these variables as indicators of future crimes can come up with wildly different outcomes depending on the person’s background, and is far more likely to advocate for harsher prison sentences for those less well off. Consider two cases:
A 35 year old white male auto mechanic has a gambling problem and is in enormous debt. He has a steady job, completed High School, owns a house in a good neighborhood and a car, is married, and has two kids. There is no history of criminal activity in his past other than a drunk and disorderly, a couple speeding tickets, and an altercation in his 20’s. No one in his immediate family is incarcerated or has a felony conviction. He is at the end of the rope and completely desperate and afraid of those he owes his gambling debts, so in order to pay off his gambling debt, he robs a convenience store at gunpoint, is caught and convicted. He shows up in court wearing his Sunday suit.
A 35 year old black male, a HS dropout with a spotty employment record and few opportunities for employment has two kids and lives in government subsidized housing in a not so good neighborhood and relying on 200 bucks a month in food stamps. His only criminal record are a couple petty larcenies as a young adult and a possession of marijuana charge. his brother is currently in jail for assault, and several of his extended family members have records. His kids are starving, so at his wits end, he robs a convenience store at gunpoint. He shows up in court wearing his best outfit- a borrowed button down and out of style tie and fraying slacks.
Using what we know about the data they are entering into this analysis, which one do you think is going to come back flagged as more likely to commit another crime in the future? Exactly. The data falsely gives a veneer of objectivity to what is actual subjective, because we are choosing what data to enter into our formula. That isn’t even getting into the proprietary nature of some of these measures and that the intellectual property is owned by private enterprise, not the public sector.
3. All of this allegedly objective evidenced based sentencing analyses are even before the openly subjective decisions that the judge can and will make, and should anyone question the judge for racial disparities in his sentencing, he can simply point to the analysis and say he used the data that was given to him. In addition to the other problems with this, it provides a codified excuse for racist judges and make the work of the Civil Rights division of the DOJ damned near impossible.
4. It takes away the humanity of the criminal and makes him just another number. Dehumanizing people makes it easier (NPR podcast on the subject here, .mp3) to be casually cruel and indifferent about how you treat them. See also, nip, Kraut, Red, untermensch, and all the other examples you an think of like this.
5. What we are essentially doing is we now live in a society where there is structural financial (.pdf) and educational inequality between disparate ethnic and racial populations, and we have created basically a permanent underclass that is rarely talked about other than in liberal publications and sociology texts. With this type of approach to sentencing, we are merely exacerbating the existing cycle. Everyone on the right says that a stable home and a mother and father are some of the best things to keep people from run-ins with the laws and growing up as productive members of society, but with this we are creating an unbreakable chain. Already, hundreds of thousands of minority children have parents in the system, and this will just make things worse, with poor and minorities spending more time in prison and less time doing the things that most of us agree will help them and their family.
I’m sure there are a lot more points I am overlooking, but these are just the basics that I can think of off the top of my head. This is an awful, awful development, and I would not be surprised if the prison/industrial complex is a behind the scenes pushing this. I’d bet my bottom dollar that a group like ALEC is behind this.
Oh, and the whole name “evidence-based sentencing” is a load of horseshit. There are rules for evidence in the courtroom, and they are strictly controlled, and the only evidence that should be considered during deliberations and sentencing is actual evidence, not subjective horseshit (and this is subjective, because it is based on the data you input, a subjective decision) masquerading as fact.
*** Update ***
Edited to add links I forgot to add.