It is important to note that each of these cases is a very extreme case where “standardized” segregation policies are unable to directly address or provide a best solution. As such they are not useful for shaping more generic segregation policies and conditions of confinement. However, if we look at these situations from a number of angles a range of options emerge. The first and foremost issue is safety to other inmates and staff.

Vignette I: Tom Silver

Clearly, Tom Silver has demonstrated three times his capacity to murder fellow inmates and staff. As strange as it seems, with age Silver’s propensity to murder again drops significantly. The vignette does not provide the dates of the murders while incarcerated, which would be critical to the risk assessment determination. However, if the last murder occurred over ten years ago and his conduct has been non-threatening since then, the option of reducing the conditions of segregation, while not eliminating them, becomes a viable option.

A floor must be established in terms of the level of security to which this inmate must be assigned. That “floor” would be a maximum security prison, probably for the rest of his life. The issue is whether this inmate can ever be released from a segregation unit into maximum security but within the general population.

The other consideration is litigation—not on behalf of this inmate but in case of the remote possibility that this inmate would commit yet a fourth murder or serious assault against another inmate or staff. If this were to occur, it would be difficult for the state to argue that it was not culpable in allowing this inmate to once again murder or assault other people. For most states this is a risk, no matter how remote, that the correctional administrator would simply be unwilling to take.

So what to do? This inmate clearly would be a good candidate for a segregation “step-down” program that would allow the inmate and staff to observe under close security his ability to handle progressively less secure confinement. The first step would be double celling within the segregation unit with increased out of cell time for recreation and programs. Clearly, regular assessment of his mental and medical health status would be required. At some point, based on his behavior, the inmate would progress to the least restrictive phase of the step-down program, which operates very much like a maximum security general population unit. After some time (one to two years) of suitable behavior he would be transferred to the general population but to a very specific housing unit where staff are very familiar with his background.

Returning, then, to the original question of whether the 35 years of solitary confinement was warranted—the answer is clearly no. But just when he could return to general population and only to a very specific general population unit for maximum security inmates would depend on the professional judgment of medical, mental health, and security staff.

As an example, let’s move from the hypothetical to the real world. I recently encountered an older inmate in the Kentucky prison system who had also murdered three people—two on the outside and one on the inside (his cell mate). After his last murder he requested a life without the possibility of parole from the judge because “the way I was going I was going to murder someone I really cared for.”

He spent about 10 years in segregation and for several years never communicated with anyone. Now in his 60s with no hope of release from prison, he was released from segregation to the general population after demonstrating positive behavior to the staff. Today he is in general population in maximum security and is assigned to the facility’s service dog training program for disabled military veterans. He said he likes the program because “the dogs don’t care about what I have done in the past.”

Vignette II: Tony Menace

This vignette is just the opposite of the first one in that (assuming the gang intelligence is correct) this inmate clearly poses a serious threat to the safety and welfare of staff and inmates. So the decision to segregate is certainly appropriate. The remaining issues are for how long and under what conditions. While most inmates will have much shorter sentences and shorter periods of segregation, this inmate will remain in segregation until he demonstrates he is willing to cease his efforts to organize and carry out assaults and murders.

Here again this case shows the value of the step-down program where the duration and conditions of segregation vary based on conduct. In this case, upon admission to the program, he would be directly confronted with his identified gang activities and told that he will remain in segregation until his gang activities cease. Given his age and sentence it may take some time for him to get the message and begin to show positive behavior, but that will occur and when it does the inmate needs to be rewarded by less severe restrictions (more recreation time, double celling, more visits, more telephone calls, and more access to programs). The initial phase of segregation would be the most restrictive (single celled and minimal out of cell time) but that would last for 90 days unless his threatening gang behavior persists. Moreover, one would expect him to leave segregation well before the end of his sentence. Was the 20 year sentence needed? Absolutely not. And the initial conditions of confinement are clearly excessive.

Again, moving on from the hypothetical to the real, during my early years as a correctional sociologist at Stateville and Joliet prison I encountered a high-ranking member of the Black P. Stone Nation street gang who happened to be in my caseload. He was very much like the inmate in this vignette (young, very long sentence, and considered dangerous). He too was quickly placed in segregation for many years but eventually, over time, his behavior changed for the better. The last time I saw him he was in a minimum security unit teaching bible classes. So people do change for the better; it’s a matter of when and under what conditions. This is why policies that allow states to sentence prisoners to long periods of segregation need to be replaced with progressive step down programs that allow prisoners to be safely returned to the general population within a few months and not a few years.

Vignette III: Sunny Bayou

The use of any long-term isolation for juveniles is especially troublesome. Bayou’s lawyer is correct that if the youth is simply isolated for an extensive period of time, even if the cell is “specially-equipped,” there may well be long-term consequences to his mental health as well as his ability to attend critically needed education and other services.

For these reasons it is increasingly less the case that long-term isolation is used or allowable in juvenile correctional facilities and occurs more frequently in adult jails and prisons. This was not always the case, but with the help of major standard setting organizations like the Office of Juvenile Justice and Delinquency Prevention, the American Bar Association, and the Annie E. Casey Foundation, the use of isolation as described above must to limited to only a few days—not weeks or months. It also must be remembered that the period of secure confinement is much shorter for most juveniles than adults, largely due to the age of jurisdiction for juveniles. Simply isolating a youth for extensive periods of time will likely serve to impair the youth’s ability to transition back to his or her family, peers, and community.

For this particular case the initial decision to place Bayou in a restricted cell should be used only for a short period of time to stabilize and control his aggressive behavior. The plan should be to get him back to a more normal albeit secure setting where he can interact with staff and attend school as required by federal and state laws. This transition from the “specially-equipped” cell needs to occur within a few days with a well developed re-entry plan. I find it all too common that juveniles (and adults) are released from long periods of segregation with no strategy on how to house, supervise, and manage them in the prison’s general population. Security staff who must supervise them are often ill-informed on the past history and special needs of these people. Case managers often make no particular effort to monitor situation by having daily contacts for the first week with the person once released to the general population. These common sense case management approaches are needed to help ensure that the cycle of repeated fights and violence is broken.

*James Austin is the President of JFA Associates, Washington, D.C.