Let me begin by acknowledging my debt for practical suggestions on these three cases to Washington State colleagues, inside and outside corrections: Kristin Cloyes, Anne Deacon, Scott Frakes, Dan Pacholke, and Eldon Vail.1 Coming from different perspectives and expressed in different terms, their observations on diagnostic, treatment, and security issues were remarkably consistent with each other. Discussion of individual cases, however, poses the risk that the existence of people such as Mr. Silver may seem to support prison management policies that have led to supermax facilities mushrooming across the U.S. over the past 30 years.

The problem is not precisely whether the cases are representative, although many prisoners find themselves assigned to extended segregation for less evident or compelling reasons than appear in these cases. The problem is rather that, once we establish the authority to respond to dangerous prisoners through long-term segregation, it is likely to be applied far too freely. Perhaps power corrupts, but corrupt or not, in prisons the power to segregate is likely to be overused because of the liability of not using it when the lives and limbs of staff and other prisoners are at stake. Furthermore, many of the conditions that prevail in such units, though rationalized in terms of security, may be better explained by the attempt to let the system serve a general deterrence function within prisons, i.e., as a form of punishment. Whether or not deterrence is a legitimate function for extended solitary confinement, we may note in passing that the growth of supermax populations over the past 30 years raises questions about the effectiveness of this strategy. So from the real threats exemplified in these vignettes, it is critical not to infer that anything like our current practices can be justified.

Notorious prisoners like Mr. Silver can be found in every system, and were there long before the modern supermax regime was inaugurated at Marion in the early 1980s. When someone has killed staff or other prisoners, it would be foolhardy to deny the need to place him in a setting that prevents him from assaulting others. Indeed, it is precisely the obligation to prevent people from harm, while they are wards of the state, that supports the security measures that characterize modern prisons as well as the obligation to provide medically necessary treatment to prisoners, including those with serious mental illness. For notorious prisoners, the primary obstacle to developing alternatives is that even to imagine them may seem a moral failure: Does he really deserve another chance, after what he did? Perhaps not, but building an entire policy on the basis of undeserving cases is another story. In these comments I suggest that solutions to individual cases can be considered rationally only when we take responsibility for carrying out policy principles, and that the jurisprudential concept of strict scrutiny provides a model for those principles.2 Ethics first, then risk assessment.

What does it mean to begin with ethics? Differences of opinion can plausibly be maintained about the effects of long-term confinement on symptoms of mental illness, prison conditions, or later recidivism, but one should begin by stipulating that locking someone up around the clock—with solitary breaks five times a week in a small concrete yard or fresh-air ventilated room—interferes with the fundamental capacities of humans as social animals, not to mention more advanced moral claims such as dignity and respect.Such a deprivation is not to be imposed without rigorous justification.3 May we do so as a form of punishment, either because some people are so evil they deserve it or to scare off others who might otherwise emulate them? Some may be tempted to affirm punishment in this form, thinking of a particular case such as that of Mr. Silver; but whatever we think he in particular deserves, the question is not about him but about how one might justify a system in which some people are made to live forever in a concrete box, and others made to keep them there, for the sake of retribution or deterrence.

Solitary confinement in the present-day supermax setting is represented as administrative and preventive, rather than as fair punishment for what the prisoner has done. What he has done may provide the primary evidence that locking him up is needed, for example, to prevent further assaults by members of one group upon members of another. The length of administrative solitary confinement, however, may far exceed what is normally allowed for disciplinary segregation, in effect conceding the premise with which many of us begin: this is something that prison systems have no right to do except for the most exceptional reasons and under strict limits.

The jurisprudential concept of strict scrutiny provides a model for the relationship of ethical principles and empirical findings required to justify, not our current system, but the occasional practice of separating some inmates under conditions that prevent unshackled encounters with others in the same space. Under strict scrutiny, extended solitary confinement could be imposed only if necessary to serve a compelling state objective; for example, preventing Mr. A from trying to kill Mr. B. The mandatory character of the objective—something one has to do, once the risk is detected—distinguishes it from other important objectives, such as making it easier for corrections officers to do their jobs: the latter is important but not compelling enough to justify a systematic deprivation of the most basic elements of life as a social, language-using animal. Furthermore, with respect to the means employed, one may plausibly argue that the compelling objective of saving a man’s life is most readily served, at the moment the risk appears, by locking up Mr. A, Mr. B, or both. To maintain prison order by making it easier for corrections officers to do their jobs, other means are available, though not always as quick: uniforms, observation, orders, gates, fences, towers, authorized weapons, infractions, hearings, loss of privileges, disciplinary segregation, and extension of prison terms, as well as positive programs and amenities. Even a compelling objective wouldn’t justify placing someone in solitary confinement if, for example, the same objective could be met, maybe less conveniently, by separating Mr. A and Mr. B in different quadrants, facilities, or prison systems.

Once the criteria of necessity and compelling state objective are met, the most consequential implication of the strict scrutiny model is that the administrative use of solitary confinement be narrowly tailored to the objective, imposed only to the extent and for as long as required to prevent a violent assault. To limit isolation so strictly requires individualized, selective procedures—as opposed to routine supermax assignment of classes of prisoners—that provide sound empirical evidence that a violent outcome is likely if this particular inmate is left in the presence of others. If punishment is not a legitimate objective, many standard features of supermax regimes would fail to meet the narrowly tailored standard; for example, barring communication, interaction, daylight, outdoor exercise, and unshackled movement from place to place for all isolated prisoners regardless of what brought them there.4 In addition to limiting how readily isolation is used and challenging the routine severity of isolation settings, the strict scrutiny model requires that isolation last only as long as necessary to mitigate the threat that warranted the initial classification decision. To do so requires a shift of focus within supermax units to returning people to general population as quickly and safely as possible. Once this conclusion is reached on moral grounds, along with the urgency of acting on it, we may expect that administrators and staff will find creative solutions to the problems posed in our three vignettes.

What began happening over 15 years ago in Washington, initially through the courageous leadership of Gary Jones, was not a matter of following evidence-based prescriptions, because there was then, as now, no solutions manual for people like the subjects of our vignettes. Rather, it was a matter of giving losing combatants in a struggle with prison authorities the opportunity to lay down arms and settle upon dignified terms of peace. Without such a recognition, we may fail to engage the motives upon which many maximum custody prisoners have acted:

They wanted me to take anger management. But I don’t fight because I’m angry: I fight because I’m supposed to.

We assume here that the mental health issues mentioned in the Silver vignette are likely a consequence of long-term isolation; while these establish a claim to modify harmful conditions, they do not place Mr. Silver in a specially designated protected class, i.e., the seriously mentally ill, whom courts have attempted to exclude from extended segregation in several important cases.5 Acknowledging the political or (very loosely speaking) ethical framework by which assaultive prisoners such as Mr. Silver rationalize their conduct, means, first, coming up with a program to help them understand how much has been lost and how little is to be gained by continuing conflict. Two stages in this process are reflected by the following two program veterans with allegiances like Mr. Silver’s:

I thought I was a courteous, respectful person, opening doors for ladies and things like that, but when I put together the timeline, I realized that I had lived my life as an immoral person, a dirtbag.6

---

The struggle for power in prisons makes no sense. You tell people, if you don’t do what we want, we’re going to hurt you. That kind of thinking always leads to a cop; you encounter the resistance of those who want to live in peace.

Without defending the severity of most contemporary supermax regimes, we must also acknowledge that specific deterrence plays a role here: We interviewed one solitary confinement prisoner on a life-without-parole sentence, whose unapologetic stance about his record of lethal violence exemplified why maximum custody exists; but years later we encountered him in a program to reintegrate into general population:

I realized that if I kept going like this I was going to spend the rest of my life in a cage, and that’s just fucking retarded.7

The behavioral component of programs for defeated warriors is needed both to develop and to test the authenticity and sustaining power of the hard-won understandings cited here; in particular, practicing skills of self-control and respectful communication in the presence of others.

Vignette I: Tom Silver

In Mr. Silver’s case, the prospects are slim indeed for a return to the relative liberty of general population, until he’s too infirm to mount an assault. Regardless of how convinced we may be that he’s turned himself around, the stakes are too high for administrators to take a chance no matter how improbable the risk. None of our correspondents recommended an immediate change in custody; all of them, however, saw prospects for cognitive-behavioral program intervention (including monitored or physically restrained interaction with other prisoners) and mitigation of conditions (e.g., outdoor exercise) for as long as some form of segregation is required to keep others safe from Silver.

Vignette II: Tony Menace

With Tony Menace, the specific content of program intervention will depend on diagnosis, but none of our correspondents believed a 20-year assignment to extreme isolation was warranted. The extremity of the conditions cannot be justified on such a long-term basis for anyone (even Mr. Silver), and, extreme or not, isolation would be warranted for only as long as the threat persists. Anyone who has raised sons knows that the reckless behavior characteristic of many 19 year old young men tends to moderate as they mature. Furthermore, there is the question of mental illness which raises particular ethical and legal concerns about isolation, absent intensive treatment to mitigate the threat and allow eventual reintegration into general population. While it is difficult to reconcile severe mental illness with Tony’s evident ability to lead or control the behavior of other inmates, our correspondents agreed that the significant mental or behavioral disorder must be addressed.

We hope Washington State is not alone in having established an intensive treatment unit in a maximum security setting, with rich clinical staffing, close observation, and the flexibility to ease restrictions on movement and exercise as behavior improves. In such a setting, clinical and security staff collaborate on detailed, individualized plans to gradually improve the behavioral skills and attitudes of inmates whose violent or obstreperous behavior reflects traumatic brain injury, severe personality disorders, or major mental disorders. Such plans are characterized by approximately a dozen short-term incremental steps or phases (thereby minimizing the complexity of objectives at each stage), with increasing privileges at each one to reinforce participants’ progress, beginning with mitigation of the habits and behaviors that interfere with treatment. The observations and suggestions of custody staff who interact many times a day with the inmates are critical to this process, as are the diagnostic and behavioral treatment skills of clinical staff. Once Mr. Menace achieves clinical stability and the ability to interact coherently with staff and other inmates, it will be time enough to address the role of gang allegiance in his conduct.

Vignette III: Sunny Bayou

Like mental illness in the case of Tony Menace, the juvenile status of Sunny Bayou raises serious questions about the fairness as well as the likely effectiveness of extended solitary confinement. Our correspondents agreed that the solution proposed in the vignette is both damaging and unwarranted. Of course measures must be taken to minimize Sunny’s opportunities to assault other inmates, whom the juvenile authorities are obliged to protect just as they are obliged to protect Sunny from the damage solitary confinement may cause to his maturing capacities. His impulsivity warrants assessment for traumatic brain injury or other conditions that may underlie his behavior. Modification of a few desks in a classroom setting to restrict Sunny’s ability to assault seems both less expensive as well as less damaging than designing a special cell for individual instruction. The presence of others is required for Sunny to learn how to behave more peaceably as well as for others to assess and intervene to promote his social learning. An older counselor who can serve as a parental mentor may encourage him to work on his educational and character development. For the community’s sake, releasing him from isolation to the streets sounds like a terrible idea, providing yet another argument for helping him change first. More generally, there is nothing in the vignette to indicate that less extreme measures have been tried without success, and that the proposed solution is really a last resort.

The cases of Tony Menace and Sunny Bayou pose challenges that can be addressed, without resort to extreme long-term isolation, given procedures that promote collaboration rather than mutual misunderstanding among custody and clinical staff in prisons. What the three cases have in common is the necessity of an organizational commitment: whatever their clinical status or however vicious their behavior, we must find alternatives to making people live in a cage with no prospect for release while they serve out their terms. We have proposed a strict scrutiny model in order to exhibit the case that such an organizational commitment is ethically required. n

End Notes

1. Others should be mentioned from whom I have learned much, though not specifically about these cases: David Allen, Eric Janus, Gary Jones, Steve Ramsey, Lorna Rhodes, and Hans Toch.

2. Model, because we are speaking here of an ethical relationship, not a legal one. A legal rights model is described by Fred Cohen (2008). Penal isolation: Beyond the seriously mentally ill, Criminal Justice and Behavior, 35(8), 1017-1047. Cohen also distinguishes, as we do here, the temporary need to insulate some inmates from contemporary practices of solitary confinement

3. The empirical difficulties of providing such a justification, given ethical concerns, are canvassed in detail in Mears, Lawrence (2013). Supermax prisons: The policy and the evidence, Criminology and Public Policy, 12(4), 681-719.

4. Similar conclusions are defended by Cohen, cited above. With respect to uniformity of rules across supermax prisoners, see Lovell, David (2008). Criminal Justice and Behavior, 35(8), 985-1004, which distinguishes between individualized flexibility in classification (programs, privileges and return to general population) and uniformity in “unit security procedures such as the management of gates, cell doors, yard access, and movement” (p. 1001).

5. Madrid v. Gomez, 889 F. Supp. 1146 (N.D. Cal. 1995); Jones’El v. Berge, 164 F. Supp. 2d 1096 (W.D. Wis. 2001).

6. The participant is referring to a “self-analysis timeline” constructed as part of a reintegration program for maximum custody inmates, following a principle put forward by Hans Toch: teaching people in trouble to become students of their own behavior. Among off-the-shelf cognitive behavior programs recommended for use in programs for offenders, one finds parallels to this injunction: not surprising, given the common source, in social learning theory, of Toch’s approach as well as more formulaic program manuals. Toch, H., & Adams, K. (2002). Acting out: Maladaptation in prisons. Washington, D.C.: American Psychological Association.

7. Lovell, supra note 4.

*David Lovell, Ph.D., is Research Associate, Professor Emeritus, University of Washington.