Behavior Science Goes to Prison

Janet Ellis

The most difficult inmates (those who incite riots, fight with other prisoners, or attack guards), end up on what is called administrative segregation – Ad Seg, for short. I was asked to set up and run a program for managing 20 Ad Seg prisoners in the Beto 1 prison, one of the facilities in the Texas Department of Criminal Justice.

These prisoners were the worst of the worst. They banged on bars with metal objects, shouted and cursed, stopped up their toilets, set fire to their mattresses, and threw urine on passersby – including not only guards, but nurses, social workers, and chaplains. Most had been in trouble all their lives. All 20 were mentally retarded. Many were completely illiterate, and had to sign the required consent form for participating in the program by making an “X.”

We knew that working with these prisoners would be a real challenge, but if we could change their behavior for the better, perhaps our program would be adopted throughout the prison. It might even provide a model for prison reform throughout the nation. Such was our hope.

Our program consisted of providing privileges and other rewards for useful and appropriate behavior such as cleaning their cells, interacting with guards and other people politely, attending literacy classes, working on their school work in the cells, and, eventually, painting the unit, scrubbing floors, starting a garden to provide vegetables for the prison, and distributing laundry to other inmates. They learned the kinds of skills people need in order to live together, and they learned that behaving well pays off.

The results exceeded our expectations: Aggressive and other anti-social behavior decreased, and good behavior increased. Guards liked working in the program and resisted being transferred from the wing. Our unit was the first such unit to be allowed off monitoring by the federal judge, a sign of our success. And with this success the program grew: from 20 prisoners to 100.

Unfortunately, despite these results, the program was not extended to the prison as a whole. We were told the program took too much time for the prison psychologists to monitor, that it was too expensive to use with all prisoners, that the program pampers prisoners who are there to be punished.

Nor is the program likely to spread to other prisons throughout the country. Today, there is little public interest in rehabilitation. The result is that little is being done to train the academic, social, and work-related skills prisoners need in and out of prison. And with prison populations growing, there is increasing pressure to cut “frills,” including programs that dramatically improve prisoner behavior.

Our study does not prove, of course, that prisoners who behave better during confinement will behave better when they are released. But at the very least we have demonstrated that even the most difficult prisoners can be treated in a humane way, can learn to behave as appropriately as any other member of society, and can learn skills that are useful beyond prison walls. Surely such prisoners have a better chance of staying out of trouble on the outside.

Humane prison programs such as ours are not popular with taxpayers, many of whom consider them forms of coddling and a waste of money. But street crime and high taxes are not popular either, and if we do not change the way we treat prisoners, we are going to continue to have more of both.

Janet Ellis, Ph. D. is Associate Professor,
Department of Behavior Analysis,
University of North Texas,
Denton, Texas.

Return Home