01:00 AM EDT on Monday, July 17, 2006

IN 1973, two Stanford investigators conducted a landmark experiment that came to be known as the Stanford Prison Experiment. A group of healthy, emotionally normal college students were randomly assigned to roles as mock inmates and mock guards in a mock prison exercise. The study, originally planned for two weeks, had to be aborted after six days because of significant pain and psychological trauma experienced by the participants.

The results surprised the investigators and their professional colleagues, and shocked the public. Otherwise emotionally strong students who were assigned to the inmate role suffered intense psychological pain, trauma, and breakdown. Normal and healthy students assigned to the guard role began mistreating their peers almost immediately, and devising sadistically innovative methods to harass and degrade the mock prisoners. The more aggressive guard subjects emerged as leaders; the more restrained mock guards held back, played along, and internalized. None of the less aggressive mock guards intervened or reported the abuse they witnessed.

The experiment demonstrated, in the investigator's words, "the extraordinary power of institutional environments to influence those who pass through them."

In reviewing the factors that led to the abuse and inhumane treatment of detainees by U.S. forces at Abu Ghraib, the panel led by former Defense Secretary James Schlesinger found this 33-year-old study relevant, noting that the Stanford Prison Experiment "underscored the degree to which institutional settings can develop a life of their own, independent of the wishes, intentions, and purposes of those who run them." The debate over the wishes, intentions, and purposes of those who were responsible for Abu Ghraib is ongoing, but this point about the influence of culture is worth noting, because it highlights the need for "those outside the culture to offer external perspectives on process and procedures."

For seven years, I was part of the culture at the Rhode Island Department of Corrections. In my last three years, I was medical-program director, and I resigned my post in the fall of 2004 in a disagreement with the director over the handling of an earlier abuse incident. While the state prison system is certainly not Abu Ghraib, as an institutional correctional system it is subject to the same social dynamics described at Stanford three decades ago.

Now, from the outside, I follow the coverage of the alleged assault of inmates at the Adult Correctional Institutions ("Three ACI guards charged with assaulting inmates," news, May 6) with great interest. The director and his staff are to be applauded for taking a strong public stand against the alleged abuses of inmates by correctional officers.

Yet I worry that when Corrections Director A.T. Wall characterizes the alleged abuses as the acts of "renegade staff," who are not representative of the majority of the prison staff, he minimizes the effect of prison culture on otherwise good human beings. I do agree with the director that the vast majority of correctional officers are honorable, decent individuals, performing a difficult and underappreciated public service.

But we are left with the following troubling facts, apparently corroborated by the state police. Multiple correctional officers were allegedly involved, with a wider circle implicated in a cover-up, and multiple incidents of alleged abuse occurred -- including the force-feeding of feces, striking inmates with a phone book, and punching them in the head. Inmates allege abuse over a period of time.

What is it about prison culture that leads to abuse? In my opinion, there are three key conditions: First, there is an absolute imbalance of power, with correctional staff controlling all aspects of the inmates' lives. Second, inmates are a socially ostracized group, and vulnerable to being dehumanized. Third, prisons are isolated from the communities they serve.

While I can't see quick remedy for the first two conditions, the third does have a viable solution: greater transparency and greater public oversight of the correctional system. Community involvement in the oversight of the Rhode Island correctional program is virtually nonexistent. An earlier effort by the legislature to appoint an oversight committee barely got off the ground, and was met by stonewalling from the Department of Corrections.

In a participatory democracy, the type of correctional system that the public gets is a function of how much the public participates in its management. Although correctional leaders have great opportunity to influence prison culture, and great responsibilities to do so, they cannot change cultures by themselves. Changing cultures requires transparency, community engagement, and the perspectives of those outside the culture. As the Stanford investigators demonstrated, prison culture left to run itself will run itself, developing a life of its own, independent of the wishes, intentions, and purposes of those who run them.

The Rhode Island Department of Corrections will release over 17,000 individuals back to the community this year. The level of social functioning of offenders released to the community is directly proportional to community engagement and involvement in the correctional-institution process and procedure. Like the schools we send our children to, the prison is a public institution whose quality of work affects the public.

It's time, then, for the public to take responsibility and become involved in management of the Rhode Island Department of Corrections, through an appropriately empowered oversight board appointed by the governor or the legislature.

Scott A. Allen, M.D., is a staff physician at the Eleanor Slater Hospital and a fellow of Physicians for Human Rights.