DOES REGRESSION WORK?

EVERYONE wants to stop crime. But John Hubner is a man on a mission. An editor at the San Jose Mercury News, Hubner believes that rehabilitating criminals is possible. His book, "Last Chance in Texas," is a plea to use age-regression therapy and psychoanalysis to turn youthful offenders' lives around.

To make his case, Hubner focuses on two young offenders at the Giddings State School, a last resort stop for the worst of the worst juvenile offenders in Texas (hence the book's title). The school is interesting because of its remarkably low recidivism rate.

A warning to readers: Read this book only if you have an extremely high tolerance for a lot of psychobabble. Because we are told that teenagers cannot really be held completely responsible for their actions.

According to Hubner, the "most crucial parts of the brain develops 95 percent of its capacity in adolescence. It doesn't even begin to develop until you hit 11 or 12 and doesn't finish until you're 19 or 20." But, if that were true, then it would be reasonable to ask if society should even want those under 21 voting if the "most crucial parts" of their brains aren't functioning.

For his part, Hubner argues that if the right experts get access to these troubled young people while their brains are still forming, we can change their lives for the better.

For example, he reports on the life stories of two young offenders at the school and describes the treatment they receive.

One of the offenders, Ronnie, has had a rough life. He is from a broken family. His aunt beat him. His mother left him to be raised by his grandparents but in her defense, she also had a rough life when she grew up.

Hubner argues that one positive adult influence in a child's life can be enough to keep him from a life a crime. And yet Ronnie had two loving grandparents on his mother's side and an uncle who seems to have been wise beyond his years. So why did he turn out so badly?

One of the school's techniques for rehabilitating these young lives involves making them tell their life story combined with role-playing and age regression. The school also uses group therapy. Hubner describes a role-playing session, as Ronnie watches the psychologists and some fellow students put on a play reenacting part of his life.

The fact that Ronnie reacts by throwing himself to the floor crying is a little hard to understand. Hubner briefly mentions that this type of therapy is controversial, but despite his assurances that no one could fake such strong emotional responses, it is not clear why watching what appears to be a quickly devised amateur play about one's life should elicit such a strong reaction. Or convince Ronnie that he shouldn't return to a life of crime.

And yet, the school has been very successful at preventing its charges from committing more crimes. Hubner is reporting on a school that is actually succeeding where many other programs fail — a 2004 study reported that only 10 percent of graduates of Giddings' Capital Offenders group have been rearrested for a violent crime after three years on parole. So he's describing a success story but can't seem to get his analysis right.

He clearly agrees with the recent Supreme Court decision that struck down the death penalty for 17-year-olds who committed murder because they didn't believe that they would be severely punished, let alone put to death. But his analysis runs afoul of economics research that shows juveniles are smart enough to be as deterred by higher penalties or a higher probability of being convicted as adults.

The book would have been more convincing if it actually delved into the debate among psychologists over age regression and its effectiveness. Or faced the hard questions of whether treating 16- and 17-year-olds more leniently encourages more crime, when criminologists have long known that crime rates drop as soon as one reaches the age that the court system considers offenders adults.

John R. Lott Jr., a resident scholar at the American Enterprise Institute, is the author of "More Guns, Less Crime" (University of Chicago, 2000).