Skip to content
Who's in the Video
Laurence Steinberg is the Distinguished University Professor of Psychology at Temple University. An internationally renowned expert on psychological development during adolescence, he is the author of more than 250 articles[…]

Juvenile recidivism is far from inevitable, it’s just a matter of employing techniques that actually work.

Question: How should we change the juvenile justice system?

Laurence Steinberg: I think there are a couple of changes that would make a big difference. The first, and I think most important, is that we don't use, generally speaking, the kinds of interventions and treatments that have scientifically proven to be effective and cost-effective. We commissioned a study when I was working on this project for the MacArthur Foundation in which we looked at what works for juvenile offenders. And there's good news and bad news. The good news is that there are programs and interventions that work; so family-based programs tend to be pretty effective in reducing recidivism. The bad news is that only about 5 percent of juvenile offenders who are eligible to be in these kinds of programs get that kind of treatment. So the first change that we need to make is to really move toward evidence-based practices within the juvenile justice system. When people look at the data and they say, look at this high recidivism rate; obviously nothing works -- well, that's not how I read the data. What I read the data as saying is that we don't use the things that work, and that's why the juvenile justice system has the track record that it does. So that's one important change.

A second change is, we need to keep kids from penetrating more deeply into the system than they need to penetrate. So we don't need to lock up as many kids as we lock up. We lock up a lot of kids for nonviolent offending. We don't need to transfer as many kids to the adult system as we do. We transfer a lot of kids who are first-time offenders; we transfer a lot of kids who are nonviolent offenders. So one of our recommendations in our book, Rethinking Juvenile Justice, is that we limit the eligibility for transfer to the adult system to violent repeat offenders who are at least 15 years old. That would significantly diminish the number of kids that are transferred into the adult system.

And then the third change is to recognize that a lot of kids in the juvenile justice system have significant mental health problems. A lot of them have substance abuse problems. A lot of them suffer from psychiatric disorders like PTSD or depression. And if we don't get those kids services, when they come of the justice system they're going to be likely to offend again because many times their offending is related to the mental illnesses that they have. So substance abuse is a perfect example. In our studies we find that if you give juvenile offenders with substance abuse problems substance abuse treatment, they're less likely to offend, even though you haven't done anything directly to address the offending. But by diminishing their addiction, or by eliminating it, or by controlling it, you're engaging in crime control.

Question: How would a family-based prevention program work?

Laurence Steinberg: Okay, so a family-based program, something like functional family therapy, would take place in the community, so we wouldn't incarcerate the juvenile. We would work with the family, with the parents or the guardians, to improve the quality of their parenting, which has a big -- so we would work with the family or the guardians in the community to improve the quality of the parenting that the juvenile gets, because parenting and bad parenting has been shown to be related to juvenile offending. We would try to strengthen the links between the family and the school, because being engaged in school is also a protection against juvenile offending. And we would work with the adolescent as well. The idea here is that you can't just change the kid; you have to change the context that the kid lives in. And that's the logic behind these family-based interventions.