Proceedings of a Colloquium to further a National
Consensus on Ending the Over-Use of Extreme Isolation
in Prisons

John Jay College of Criminal Justice
September 30-October 1, 2015

Prisoner Reentry Institute
With support from
The Jacob and Valeria Langeloth Foundation

A Colloquium to Further a National Consensus on Ending the Over-Use of Extreme
Isolation
Table of Contents
Executive Summary………………………………………………………………………………………………………...2
Introduction…………………………………………………………………………………………………………………...5
Major Presentations……………………………………………………………………………………………………...13
Discussion Following Presentation and Comments from Bernie Warner and Jack
Beck…………………………………………………………………………………………………………………..13
Discussion Following Presentation and Comments from Joe Ponte and Ron
Honberg…………………………………………………………………………………………………………….14
Discussion Following Presentation and Comments from Rick Raemisch and Laura
Markle……………………………………………………………………………………………………………….14
Work Group Discussions……………………………………………………………………………………………….15
Group 1……………………………………………………………………………………………………………...15
Group 2……………………………………………………………………………………………………………...17
Group 3……………………………………………………………………………………………………………...24
Recommendations………………………………………………………………………………………………………..38
Next Steps…………………………………………………………………………………………………………………….43
Acknowledgements………………………………………………………………………………………………………45
Appendix A…………………………………………………………………………………………………………………...46
Appendix B…………………………………………………………………………………………………………………...57
Appendix C…………………………………………………………………………………………………………………..63

1

Executive Summary
With support from the Jacob and Valeria Langeloth Foundation, on September 30,
2015 John Jay College of Criminal Justice convened a colloquium including 15 corrections
agency heads and a like number of attorneys, academics and experts from the community
of those seeking to reform the use of social isolation, often called “solitary confinement,” in
U.S. Prisons and Jails. The purpose of the colloquium was to determine if consensus might
be achievable about ways to achieve the sought after reforms by common agreement and
without resort to litigation. To facilitate discussion, all participating agreed to be bound by
the “Chatham House Rule,” that provides that the outcomes and discussion would be
limited to the participants and that the report would not identify discussants by name or
affiliation and that in the future the participants would not make reference to statements or
admissions by other participants.
What resulted was a remarkable two day experience that generated a great deal of
argument and debate as well as an equally exciting degree of agreement and consensus.
One attorney representing inmate interests called the gathering “worthwhile” in a
subsequent email to the organizer, and a corrections administrator said it was “great.”
Another participant, a medical expert commented after the colloquium, “Thanks so much
for including me in the colloquium; I had no idea how many folks in Corrections we could
trust, admire, and have a feeling of friendship with. What a transformation since I first got
involved with these issues 30-odd years ago! I was delighted to realize it did not have to be
"us versus them", but instead a shared mission and goal, but with different intelligences
about the pathways to and barriers resisting change. That is certainly how it felt in my
work group.”
The gathering provided a first opportunity for many to meet with “adversaries,”
often for the first time and to listen to and consider the other side’s point of view. New
friendships and alliances were forged and the basis for constructive conversation was
established. It would be difficult to say that any minds were changed but it can be said
without equivocation that emerging from the meeting was an energized group united in the
belief that the United States can do better in how it employs social isolation and ameliorate
many of the most damaging results from the overuse of extreme social isolation.
As a result of the deliberations several clear themes and areas of agreement are
apparent:


The use of social isolation is greater than it has to be, in large measure because
prisons have been called upon to do things they were never intended to do and
inadequately resourced to do it.



Persons with mental illness and other vulnerable populations who do not
require imprisonment should be treated elsewhere.

2



The only criterion for confining a person to social isolation within prison should
be behavior; persons should not be confined based upon their affiliation or
status.



When it is used at all, separation from general population should be for the least
amount of time necessary and in the least restrictive conditions.



Separation from general population must always provide for adequate living
conditions and for meaningful routine, and periodic medical and mental health
assessments.



Transparency and accountability in the use of segregated housing is essential.



Decisions about the use of social isolation in prison for disciplinary reasons
should be made using an appropriate due process procedure.



Prison discipline should incorporate a continuum of measures to hold
incarcerated persons proportionately accountable for their behavior and the use
of isolated confinement should be the last resort. Alternatives to isolated
confinement as punishment are desirable and should be found.



Decisions about the use of segregation in prison for other reasons should be
made by multi-disciplinary teams with a view toward improving outcomes.



The confinement of individuals in isolated confinement other than for
disciplinary reasons should not feel punitive to the affected individual.



The purpose of isolated confinement must be to improve the outcome for the
affected individual and to make the prison and the community safer, to that end
there must be meaningful interventions designed to address the reasons for the
confinement and attainable means for the individual to transition back to the
general population of the prison.



Wherever and whenever possible opportunities to relive the social isolation of
the confined individual should be employed.



Corrections administrators and advocates for incarcerated persons must work
together to obtain political and financial support for the changes needed.



Line corrections staff can and must be made to see the utility of reform.

These themes are reflected in the 23 specific recommendations contained at the end of this
report. Our intent in convening this colloquium was to try to establish a “roadmap” for
reform. What is clear is that the road to a state of affairs we can all be proud of is long and
3

there remains much to be done. But as the proverb says, a journey of a thousand miles
begins with a single step and these recommendations can serve as the map to the first turn.

4

Introduction
Over the last three decades, corrections systems have increasingly relied on
isolation and segregation as a prison management tool, even building entire “supermax”
prisons, where prisoners are held in extreme social isolation, often for years or even
decades. The massive increase in the numbers of men, women, and children behind bars
placed extreme stress on existing facilities, corrections systems, and criminal justice
budgets which struggled to accommodate the unprecedented growth in population without
the resources or political support necessary to create rehabilitative environments. Prison
administrators were left with few tools to keep their prisoners safe and to enforce
reasonable prison rules. This fostered an environment where the use of extreme social
isolation and segregation became the default approach to addressing many of the complex
challenges faced in operating places of detention and incarceration. On an average day in
2011–12, up to 4.4% of the state and federal inmates and 2.7% of jail inmates were held in
administrative segregation or extreme social isolation. Based on current prisoner populations
this translates into 660,000 state and federal prisoner and 201,000 jail inmates. 1 Both legal
and medical professionals have come to criticize extreme social isolation as unconstitutional
and inhumane, pointing to the well-known harms associated with placing people, especially
those with mental illness in extreme social isolation.2And now, research is emerging which
suggests that extreme social isolation can actually have a negative effect on public safety.3 The
results of this questioning of the status quo can be seen in administrative and legislative reform
now occurring in jurisdictions around the country.4 Corrections organizations are engaging on
the topic and beginning to develop guidelines for the field.5 Civil rights and human rights
organizations, faith community leaders, lawyers, and mental health organizations, have called
for reforms in policy and practice. There is a broad and emerging sense that the use of solitary
Beck, Allen J. Use of Restrictive Housing in U.S. Prisons and Jails, 2011-2012. Washington, D.C.: Bureau of
Justice Statistics, U.S. Department of Justice, and October 2015. Also see, Carson, E. Ann. Prisoners In 2014.
Washington, D.C.: Bureau of Justice Statistics, U.S. Department of Justice, September 17, 2015, and Milton,
Todd and Zeng, Zhen, Jail Inmates at Midyear 2014. Washington, D.C.: Bureau of Justice Statistics, U.S.
Department of Justice, June 2015.
2 In re Medley, 134 U.S. 160, 168 (1890) (“[Prisoners subject to solitary confinement] fell, after even a short
confinement, into a semi-fatuous condition, from which it was next to impossible to arouse them, and others
became violently insane; others still, committed suicide; while those who stood the ordeal better were not
generally reformed, and in most cases did not recover sufficient mental activity to be of any subsequent
service to the community.”). SeeSe also, “Solitary Confinement is Cruel and Ineffective,” Scientific American,
vol. 309, issue 2 July 2013.
3 See, e.g., KERAMET REITER, PAROLE, SNITCH, OR DIE: CALIFORNIA’S “SUPERMAX” PRISONS & PRISONERS, 1987-2007 4751 (2010); MAUREEN L. O’KEEFE, COLO. DEP’T OF CORRECTIONS, ANALYSIS OF COLORADO’S ADMINISTRATIVE SEGREGATION
25 (2005).
4 See for example, “Prison officials target solitary confinement, mental health treatment in effort to avoid
another Nikko Jenkins,” Omaha.com. http://www.omaha.com/news/legislature/prison-officials-targetsolitary-confinement-mental-health-treatment-in-effort/article_361583dd-f22d-5ec6-89b3d4242b0d7782.html. November 8, 2015.
5 Mohr, Gary and Raemisch, Rick,”Restrictive Housing: Taking the Lead.” Corrections Today. March 20, 2015.
http://www.aca.org/ACA_PROD_IMIS/Docs/Corrections%20Today/2015%20Articles/March%202015/Gues
t%20Editorial.pdf , see also, Association of State Correctional Administrators, Resolution # 24- Restrictive
Status Housing Policy Guidelines, adopted September 2, 2013.
1

5

confinement has gone too far in American correctional facilities. Promising approaches are
emerging in some jurisdictions, and political space is opening up for reform in numerous states.
While opinions on how to reform isolation differ across the board, the remarkable
common ground that exists between the corrections field, experts, and advocates on the need
for reform provides a unique opportunity to form unlikely partnerships and explore alternative
approaches to public engagement on one of the most pressing challenges to the safe, effective,
and humane management of our prisons and jails: the over-use of prolonged social isolation.
The moment is ripe for correctional reform. Experts across the political spectrum have
reached consensus that America’s incarceration levels are unsustainably high.6 Many attack
the morality of mandatory minimums, while others highlight the budgetary impact of
maintaining large swathes of the population behind bars.
On September 30, 2015 John Jay College of Criminal Justice with support from the
Langeloth Foundation convened a remarkable two day meeting between 15 corrections leaders,
including state corrections directors and leading officials from the major urban jails; and 15
experts, academics, and leaders in the advocacy community working for reform of isolation
practice to leverage this common ground. The meeting was, by agreement of all participants,
subject to the “Chatham House Rule,” which states, “When a meeting, or part thereof, is held
under the Chatham House Rule, participants are free to use the information received, but neither
the identity nor the affiliation of the speaker(s), nor that of any other participant, may be
revealed.”7 The goal was to bring together these leaders to plan next steps, forge greater
connections and collaborations, to identify “lessons learned” from efforts to date, and formulate
plans to ensure sustainability for a reform effort that still has much to accomplish. Seizing on the
increasingly recognized need for reform, the Colloquium facilitated a candid and productive
discussion among key figures of the prison policy community regarding the use of extreme
social isolation. From top correctional officials to the ACLU, the Center for Constitutional Rights
and the National Alliance for the Mentally Ill, many contrasting perspectives were brought to
the table. While disagreements remain, a single thread united the two days of spirited
discussion: we have arrived at an extraordinary moment in time where real change is possible.
The outcome of the convening was intended be a written reform agenda bearing the
imprimatur of the participants and carrying “weight” in each of their communities of practice.
This is that agenda. In subsequent reports John Jay will document the changes and
improvements resulting in the participating jurisdictions, and identify the barriers that impeded
reform.
The participants are listed below:
See for example, Jeremy Travis and Bruce Western, Editors; Committee on Causes and
Consequences of High Rates of Incarceration; Committee on Law and Justice; Division of Behavioral and Social
Sciences and Education; National Research Council. The Growth of Incarceration in the United States:
Exploring
Causes and Consequences. Washington, D.C.: The National Academies Press, 2014.
7 See more at: http://www.chathamhouse.org/about/chatham-house-rule#sthash.NhS71S3u.dpuf
6

6

Corrections Officials:
John Baldwin: Illinois Department of Corrections
Jeffrey Beard: California Department of Corrections and Rehabilitation
Leann Bertsch: North Dakota Department of Corrections and Rehabilitation
Marshall Fisher: Mississippi Department of Corrections
Carol Higgins O'Brien: Massachusetts Department of Corrections
Gary Lanigan: New Jersey Department of Corrections
Gregg Marcantel: New Mexico Corrections Department
Terri McDonald: Los Angeles County Sheriff's Department
Shirley Moore Smeal: Pennsylvania Department of Corrections
Joe Ponte: City of New York Department of Correction
Rick Raemisch: Colorado Department of Corrections
Scott Semple: Connecticut Department of Corrections
Bryan Stirling: South Carolina Department of Corrections
Bernie Warner: Washington State Department of Corrections
Heidi Washington: Michigan Department of Corrections
Advocates:
Sarah Baumgartel: Yale Law School, New Haven CT
Jack Beck: Correctional Association of New York, New York NY
Brett Dignam: Columbia University School of Law, New York NY
Jamie Fellner: Human Rights Watch, New York, NY
Amy Fettig: American Civil Liberties Union, Washington DC
Robert Fleischner: Center for Public Representation, Northhampton MA
7

Stuart Grassian, MD: Chestnut Hill MA
Ron Honberg: National Alliance on Mental Illness, Arlington VA
Terry Kupers, MD: Institute for Global Communications, Oakland CA
Jules Lobel: CCR & University of Pittsburgh, Pittsburgh PA
Joe Luppino-Esposito: Right on Crime/Texas Public Policy Foundation, Washington DC
Laura Markle Downton: National Religious Campaign Against Torture, Washington DC
Taylor Pendergrass: New York Civil Liberties Union, New York NY
Heather Rice Minus: Prison Fellowship/Justice Fellowship, Landsowne PA
Margo Schlanger: University of Michigan School of Law, Ann Arbor MI
Facilitators:
Brian Fischer: NYS Department of Corrections and Community Supervision, Retired
Andie Moss: The Moss Group Inc., Washington DC
Michael Mushlin: Pace University School of Law, White Plains NY
Invited Observers:
Andre Bethea: U.S. Department of Justice’s Bureau of Justice Assistance, Washington DC
Andrea Fionda: Jacob & Valeria Langeloth Foundation, New York NY
Marie Garcia: National Institute of Justice, Washington DC
Lila Kazmeian: John Jay College of Criminal Justice, New York NY
Jeff Mellow: John Jay College of Criminal Justice, New York NY
Scott Moyer: Jacob & Valeria Langeloth Foundation, New York NY
Bettina Muenster: John Jay College of Criminal Justice, New York NY
Fred Patrick: Vera Institute of Justice/ Center on Sentencing and Corrections, New York NY
Sara Sullivan: Vera Institute of Justice/ Center on Sentencing and Corrections, New York NY
8

Judith Resnik: Yale School of Law, New Haven CT
Reporters:
Erica Danielsen: Pace University School of Law, New York NY
Sarah Lusk: Pace University School of Law, New York NY
Abigail Marion: Columbia University School of Law, New York NY
Jennifer Peirce: John Jay College of Criminal Justice, New York NY
Nick Reck: Columbia University School of Law, New York NY
Susruta Sudula: John Jay College of Criminal Justice, New York NY.
The first day of the Convening began with presentations setting forth three case studies
of reform, such as: reduction of segregation in Washington; removal of the seriously mentally ill
from administrative segregation in Colorado; and reforms for youth at Rikers’ Island in New
York City. This opening session framed the successes and challenges of these approaches and
built in time for lively discussion and comparative analysis between participants. This initial
discussion raised several themes that would recur throughout the Colloquium, namely,
definitional difficulty; goals for reform, including interventions that provide positive incentives;
challenges to reform, including staff culture, insufficient political support, constrained
resources, and the challenge of responding to the large number of incarcerated persons with
mental illness.
Following the case studies, participants moved into small groups designed to confront
the core areas where reform is necessary but hard questions about achieving real and
meaningful change remain. These key areas include the following questions:


What does meaningful, effective and humane order maintenance within correctional
facilities without the use of isolation look like?



How can correctional institutions deal with prison gangs and dangerous predatory
individuals effectively and safely without the use of segregation/isolation housing?



How can vulnerable individuals, such as the seriously mentally ill, pregnant women
and adolescents be protected adequately without the use of isolation?

For each key area, the following three questions were to be addressed:

9



What strategies and programs can be used to ameliorate social isolation effectively
where segregated housing is necessary for the safety and security of an
individual/institution?



What is necessary to effectuate reform?



What are the barriers to reform and how can we overcome them?

While the full group had different perspectives and agendas to some extent, the goal of
the meeting was to establish specific areas of commonality to forward a national roadmap for
supporting the reduction in the use of long-term isolation. Progress towards realizing these
principles will be self-reported by participants to the John Jay Coordinator (Martin F. Horn) over
the course of the next two years. John Jay will periodically publish a report on the progress of
these reform efforts and lessons learned.
Emerging from the Colloquium are advocates and administrators unified in the belief
that we must seize the momentum of the day to address and reduce the overuse of extreme
social isolation—across its myriad incarnations. The Colloquium demonstrated that while
different constituents have different priorities, they share the overarching goal of creating a
safe, measured, and humane correctional environment. Just as importantly, a reasoned
discussion is not only possible, but also essential to progress.
At the outset of the Colloquium, one correctional administrator recalled a conversation
with a good friend and academic who said, “Solitary confinement is torture.” The administrator
then wondered aloud, “If that is true, does that make me a torturer?” Though disconcerting for
some, this first animating question held the seed of hope for collaboration. No one in the room
wants to be a torturer. No participant wished to do harm. Rather, all participants shared the
goal of an efficient and humane correctional regime. This commonality served as the bedrock
on which these recommendations were built.
One of the first challenges encountered was the lack of a uniform definition of solitary
confinement. Administrative, disciplinary, and protective segregation are defined and treated
differently across jurisdictions. Despite these distinctions, a common definition is key to reform,
as failure to reach definitional consensus may lead to the continued misuse of extreme social
isolation—albeit under a different name. The Colloquium largely settled on the Mandela Rules
definition—22 hours of social isolation a day—as the base point for discussion but was unable
to reach consensus on an absolute limit to the length of that confinement.8

“For the purpose of these rules, solitary confinement shall refer to the confinement of prisoners for 22 hours
or more a day without meaningful human contact. Prolonged solitary confinement shall refer to solitary
confinement for a time period in excess of 15 consecutive days.” 44 United Nations Standard Minimum Rules
for the Treatment of Prisoners, proposed, Commission on Crime Prevention and Criminal Justice Twentyfourth session Vienna, 18-22 May 2015.
8

10

The participants also discussed the goals of reform. Curbing the negative effects of social
isolation on the mental and physical wellbeing of incarcerated persons is a primary concern.
Concurrently, the physical safety and mental health of the correctional staff is of paramount
importance, and must be a central focus for any reforms on the use of social isolation. One
participant offered the following formulation:


When we talk about “segregated confinement” we need to distinguish between two
concepts: “Separation” versus “Deprivation.”
o Separation is the need to keep an incarcerated person separated from
some or all others because of danger. (This would include, for example,
isolation for infectious disease, restricted confinement to prevent harm to
a suicidal patient, isolation of vulnerable persons from would-be
aggressors, etc.).
o Deprivation is the restriction of meaningful (anchoring, as opposed to
aversive/noxious) perceptual, social, and occupational stimulation. This
leaves the individual with an inadequate basis upon which to maintain an
adequate state of attention and alertness (thus resulting in the specific
solitary confinement syndrome, basically in the realm of stupor and
delirium, along with a multiplicity of other burdens (loneliness, freefloating anxiety, etc.).



In practice, these two concepts are often not distinguished; an individual who needs
to be separated is often exposed without cause to extensive deprivations. Moreover,
imposing extensive deprivation, especially over long periods of time, creates
problems, and danger, rather than reducing them.

Participants frequently noted that social isolation is a form of intervention. Many people
agreed that a shift from punitive, isolation-dependent models might well be displaced by
positive reinforcement and incentive-based models. Such programs have shown positive
outcomes, including decreased institutional violence, and improved facility safety9. There is no
real penological justification to put an individual in segregation for an administrative (as
opposed to punitive or investigsative) purpose without attempting to enrich his environmental
opportunities as much as possible. (Conjoint recreation, education, and religious worship;
books, writing material, letters, phone calls, books, visits, etc.)
And even in the case of punitive confinement, do the deprivations - especially when
imposed for more than a short period of time - serve any purpose in keeping the prison safe, or

9

See for example, presentations of Bernie Warner, Joe Ponte and Rick Raemisch infra.

11

in preparing the individual for return to general population or to release to the larger
community after his prison sentence ends? Or, does it just create more danger for everyone?10
Numerous impediments to reform were also identified. Staff can be resistant to
decreasing the use of isolation, for fear that prison safety might be compromised if incarcerated
persons are placed in less restrictive housing. Prison administrators noted that correctional
professionals could be won over by highlighting that, in many instances, reducing the use of
social isolation is correlated with lower rates of violence.
Colloquium participants also noted that political support for reform, though growing,
must be expanded significantly. The public is frequently resistant to increased expenditures on
prisons. People might be more willing to invest in prison reform if advocates remind them that
more than 90% of all incarcerated persons return to the community: prison policy directly
affects public safety, as the treatment prisoners experience while incarcerated will determine
their abilities and behavior after release.11 Advocates can be partners in communicating the
financial and safety benefits of a responsible policy to both the general public and interest
groups, such as correctional staff and labor unions.
Resource constraints are another significant barrier to change. Even where prison
administrators are able to cut costs, state legislatures frequently will not allow correctional
officials to reallocate those funds within the correctional system. Instead, state legislature will
reclaim the newly freed funds by reducing correctional budgets.12
Participants also noted that current prison architectural infrastructure can be an
impediment to reform. Supermax prisons were not designed for any useful purpose beyond
detention, and correctional administrators strain to re-appropriate them usefully as institutions
that facilitate social interaction.
Lastly, participants noted that correctional facilities were never meant to be mental
health care facilities. Yet, the 20th century deinstitutionalization of the mentally ill has led too
often to correctional re-institutionalization.13 Reform must address the abysmal shortcomings
of the patchwork American mental health care system and the counter therapeutic environment
of prisons.

Labrecque, Ryan M, “The Effect of Solitary Confinement on Institutional Misconduct: A Longitudinal
Evaluation.” A dissertation submitted to the Division of Research and Advanced Studies, University of
Cincinnati, 2015.
11 Beijersbergen, Dirkzwager, & Nieuwbeerta, , “Reoffending after release: Does procedural justice during
imprisonment matter?” Criminal Justice and Behavior, October 2015.
12 One participant, a corrections administrator observed,” We are closing 15 ad. seg. and shu units and
converting them to general population, this will result in a savings of about 250 staff or an average of 16 staff
per unit.” But the administrator noted that State budget officials were refusing to allow the corrections
department to reinvest the savings into better operations.
13 Doris J. James and Lauren E. Glaze, Mental Health Problems of Prison and Jail Inmates. Washington, D.C.:
Bureau of Justice Statistics, U.S. Department of Justice. (revised December 2006).
10

12

Major Presentations
Bernie Warner, Secretary of Corrections, State of Washington—Appendix A
Joe Ponte, Commissioner of Correction, City of New York—Appendix B
Rick Raemisch, Executive Director, Colorado Department of Corrections—Appendix C
Discussion Following Presentation and Comments from Bernie Warner and Jack Beck
Several participants expressed common concern on the following points:


Additional research needs to be conducted on the impact of various isolation
methods on mental health;



A common definition for isolation or solitary confinement is necessary but difficult to
pin down;



A common definition for disciplinary, administrative, and long term segregation; Are
such distinctions useful?



Some programs purporting to curtail isolation practices might continue severe
isolation under euphemistic titles;



Trauma suffered by corrections officers is weighty and must be addressed;



Culture change among corrections officers is necessary to effect change;



Hiring practices may need to be significantly changed



“Merit” based approaches to getting out of social isolation/segregation may not
properly capture what is not a given individual’s fault;



Opportunities for additional transparency and access to prisons for outside groups;



The reluctance of unions to embrace change where there is perceived risk;



The “danger” of releasing individuals directly from segregation into the general
population, or to the community.

With regard to culture within facilities, several participants voiced their opinion that telling
corrections officers that new policies will promote officer safety would be, and has been,
effective in reforming facility culture.

13

Discussion Following Presentation and Comments from Joe Ponte and Ron Honberg
Participants raised concerns, including:


The importance of corrections officials expressing how the failures of the American
mental health system has impacted corrections;



Lack of commitment from elected officials to recognize the problem in the American
mental health system and a lack of understanding how such reform can be
institutionally vital to corrections;



The “deinstitutionalization [of the American mental health system] led to a different
kind of institutionalization”



The importance of understanding different categories of responses to and impacts of
isolation: social, perceptual, occupational;



Definitions of isolation are not “black and white”

Discussion Following Presentation and Comments from Rick Raemisch and Laura Markle
Participants raised many concerns, including:


Prisons must respect an incarcerated person’s inherent dignity



Disparate racial outcomes in use of social isolation must be addressed;



Incentive structures are more effective in reforming problematic behavior, even
among people with mental illness, and should be broadly implemented.



Recognizing gender differences: women very rarely need and rarely respond
positively to isolation.



The difficulty of re-purposing supermax prisons for any other purpose



Perhaps culture is not as much of a barrier as is often discussed; see Colorado’s quick
turnaround as an example of what energized leadership can accomplish.



Therapy dogs are an example of an effective intervention as a means of reducing
isolation.



Simply noting the trouble with America’s mental health system may serve to pass the
buck.

14

Work Group Discussions
Following these presentations, participants were organized into three groups of ten, to
facilitate detailed discussion. Each group focused on questions about social isolation in a
particular context:


What does meaningful, effective and humane order maintenance within correctional
facilities without the use of isolation look like?



How can correctional institutions deal with prison gangs and dangerous predatory
individuals effectively and safely without the use of segregation/isolation housing?



How can vulnerable individuals, such as the seriously mentally ill, pregnant women
and adolescents be protected adequately without the use of isolation?

These groups were tasked with reaching consensus on as many specific reform
recommendations as possible. The discussions of each are laid out in turn.
Group 1:


What does meaningful, effective and humane order maintenance within correctional
facilities without the use of isolation look like?

Maintaining the Colloquium’s commitment to voicing a diverse range of opinion a robust
discussion ensued on the role of segregation as a correctional tool. The discussion was generally
limited to disciplinary segregation – participants explicitly did not focus on administrative
segregation or other forms of isolation. Discussion was focused on four strands: the criteria for
using segregation, viable alternatives to segregation, the role of transparency, and the barriers
to reform.
Criteria
Consensus was reached on several key points. First, all participants agreed that
segregation should be used for the minimum time and in the least restrictive conditions
necessary to resolve the condition that led to isolation. Participants further agreed that all
isolation should have an incentive component, which would restore certain privileges if the
individual is able to reach certain behavioral goals. Ideally, these incentive programs would
operate on relatively short timeframes—e.g., two days of good behavior earns a reward—so
that the individual would quickly begin to see their good behavior rewarded. Participants also
agreed that isolation should have a goal of changing specific behavior and an individualized
achievable path to reach it.
As to the conditions of social isolation, consensus emerged that segregation must include
mental health rounds, health care rounds, and basic adequate conditions (physical space, light,
air etc.) Participants also agreed that there should be a minimum amount of family contact
15

allowed while individuals are in segregation, as the loss of family contact can be extremely
agitating for both the incarcerated person and the family; increasing family contact and visits in
response to improved behavior can provide a strong incentive.
Group 1 also reached consensus that there must to be due process protections in place.
These must include procedural safeguards for placement in segregation, periodic review during
segregation, and an exit mechanism. This process should consider the severity of the offense,
length of time spent in segregation, fairness, and the ability of individual to comply with
imposed conditions.
About half of the participants agreed that disciplinary segregation must be for a
determinate length of time, and recommended that disciplinary segregation not exceed 15 days,
unless extenuating circumstances otherwise dictate. Of those who disagreed with a hard 15day upper limit, several participants indicated that 15 days would be an acceptable limit, if the
15 days would only trigger a procedural review for extending the time rather than a hard upper
limit. Other participants also suggested that the 15-day limit might be used to trigger a different
type of segregation with increased access to, for example, television or some type of congregate
activity with other individuals. Others felt that a year would be an acceptable limit for serious
offenses, such as rape or seriously assaulting a guard.
Alternatives to Segregation
The use of alternatives to the default use of segregation was raised. These participants
agreed that punishment should be imposed on a continuum, with segregation as the tool of last
resort. Less serious punishments might include loss of commissary or personal property. Loss
of programming, social contact and family contact should be reserved for more serious
infractions or after other punishments have proved ineffective. It was also agreed that the loss
of privileges needs to be proportionate to the infraction, and must include a prosocial incentive
system to restore the privileges.
The Role of Transparency
Both advocates and administrators were quick to acknowledge the importance of
transparency in furthering reform. That is, transparency increases awareness and trust for the
public, prison staff, and incarcerated people. Advocates tended to focus on the importance of
granting faith-based, academic, civil rights, and rehabilitation organizations internal access to
correctional facilities and data collection. Correctional administrators were more interested in
internal feedback loops between prison administrators, the staff, and the incarcerated regarding
behavior and punishment expectations. There was agreement that such communication
between advocates and correctional administrators may avoid needless litigation, assure the
responsible stewardship of funds, and help both correctional staff and the public at large to
understand reform in public safety terms. It was agreed that statistics would support reform. In
turn, transparency of statistics and narrative will educate all interested parties about public
safety and other benefits of reducing long-term isolation.

16

Barriers to Reform
Consensus was most easily reached on the multitude of problems both advocates and
administrators face in effecting change. It was agreed that efforts must be made to get staff “buy
in” on reforms from the outset. Correctional management should find ways to “celebrate
courage in the service of public safety through small victories.” In other words, corrections
officers must find ways to celebrate the .1% improvement in some small way and, when the
inevitable tragic but isolated incident occurs, we must resist the impulse to abandon all reform.
If staff are invested, change will be more collaborative than totalitarian. Absent such
involvement, their staff will have no incentive to implement change amid a perceived risk to
safety and opposition from unions, victims’ rights advocates, and other interest groups.
All were quick to agree that resource limitations are a significant barrier to reform.
Supermax prisons do not easily lend themselves to rehabilitative programming spaces.
Legislatures have little political cover or incentive to lead reform. One administrator noted that
when he implemented new efficiencies to save money, the legislature decreased his budget
rather than allowing for a re-appropriation of funds. In many jurisdictions, correctional staff is
grossly underpaid and have little incentive to see themselves as part of the rehabilitative
process rather than “just a guard.”
The difficulty of promoting the public safety narrative as a means of fighting inertia was
acknowledged. Correctional staff, —who face considerable mental and emotional trauma in
addition to physical danger—the media, general public, and immediate community around a
prison need to understand how changes that might appear risky in fact generate safer
environments at every juncture of contact. That is, the “tough on crime” opposition to reform is
really being “tough on the community.” It is clear that a partnership between advocates and
correctional administrators can be a vital part in fostering reform through a public safety
narrative.
Group 2:


How can correctional institutions deal with prison gangs and dangerous predatory
individuals effectively and safely without the use of segregation/isolation housing?

Participants addressed eight questions about segregation and predatory persons. The
early conversation demonstrated the range of perspectives on some of the major themes of
discussion, such as who should be put in segregation, for what reasons, under what rules, with
what types of transparency, etc. The answers arrived at demonstrate that there was significant
agreement on some questions, particularly the categories of people and/or behaviors that
should or should not be eligible for segregation. There was also significant divergence of
opinion, though, notably on the appropriate conditions of confinement.
The scope of the questions sparked one participant to ask a more fundamental question:
Is segregation ever necessary? There was a clear division here. Some participants argued that
segregation does not achieve its intended results in terms of behavior change, deterrence, or
safety; therefore, they argued that it should not be used except for logistical reasons (e.g. during
17

an investigation). Other participants suggested that it is appropriate to use segregation as
punishment for egregious transgressions and/or to manage individuals who are seriously
dangerous. It was agreed to discuss the various dimensions of segregation in detail, since the
most “realistic” intermediate goal is to reduce the use of segregation – not to eliminate it. Still,
some participants maintained that the longer-term goal should be to eliminate its use. The
group agreed that the prison system’s goal should be to get those in segregation out of
segregation – or to less restrictive housing – as soon as possible.
Several participants argued that there is no evidence that segregation deters either the
individual or the general population. It was agreed that segregation often makes people worse,
and that reducing the use of segregation creates a safer institution. Additionally, several
participants said that there is little evidence – research or anecdotal – about “what works” to
change behavior during segregation – that is, a theoretically “different way” of practicing
segregation. Nevertheless, everyone agreed that since segregation is widely used and can cause
harm, it is important to develop rules and guidelines to reduce its use and its harshness as much
as possible
Based on this discussion, consensus was reached that could be converted into
recommendations. There was robust debate about the other themes that eventually led to a
“compromise” recommendation. On a few topics, there were strongly divided views and
agreement was not reached. Time ran out before every topic on the list could be addressed, and
so, for a few areas, further discussion is recommended.
Behaviors as the Criteria for Segregation
First, it was agreed that the criteria for segregation should be for behaviors only, and not
for other “status” reasons. Therefore, it is recommended that segregation should never be used
because of an person’s gang affiliation. On the question of behaviors that do merit segregation,
there was consensus only on including the most serious violent behaviors: murder, rape, or
assault. After some discussion, it was agreed that “threats of violence” – such as one individual
ordering another to commit violence – should also be included in this category. Everyone
agreed that segregation is not appropriate for minor rule infractions, but there was less clarity
on what counts as a major infraction, as well as whether segregation is an appropriate response
to a major infraction. Some participants argued that major “disturbances” – such as
participating in a riot, attempting escape, or trafficking drugs inside prison (not simply
possession) – justify segregation, while others argued that these actions merit a disciplinary
response, but not necessarily segregation, since the individual does not pose a risk of harm to
others. Another area of disagreement was what constitutes “assault” on a staff member. Most
participants agreed that prison staff can abuse their discretion and can call a minor slight (such
as profanity or spitting toward a staff) an “assault” – and that segregation is not an appropriate
response to such incidents. However, no agreement was reached on the level of serious violence
that constitutes the type of assault that merits segregation. Moreover, some participants noted
that prison officials used harsher discipline – including segregation – for lesser levels of violence
toward a staff member, while an equivalent level of violence against another person would not
result in segregation. Responding to this assertion, other participants contended that an assault
against a staff member implies a level of “disturbance” to the prison environment and is thus
18

more serious. There was no clear consensus on defining a “serious” assault or distinguishing
assaults on staff versus on another incarcerated person. Additionally, there was not agreement
on how to determine whether a seemingly “minor” altercation or fight could count as a “major”
infraction if the staff has reason to believe it will escalate.
Defining Types of Segregation
A few of the participants held the view that segregation is almost never justifiable,
except in a very few cases of extremely violent or predatory individuals – what was colloquially
termed “the Hannibal Lecters.”14 The discussion on how to handle these few individuals is
addressed below. More broadly, those who favored the most extreme limits on the use of
segregation acknowledged that, since segregation remains a widely used tactic, they would
support talking about a spectrum of “types” of segregation, with the goal of using the harshest
type the least often. All participants agreed that this “spectrum” is a useful concept, since this
can reflect the different rationales for segregation, such as punitive versus risk-management.
Since this group was addressing segregation only for predatory or dangerous individuals
(i.e. not for protection from others), the participants agreed to consider the types of segregation
in this framework: short-term for investigative purposes, short-term for disciplinary reasons,
short-term for “cooling-off” reasons, and longer-term for reasons of risk of violence or harm. It
was agreed that more austere conditions are acceptable for short-term segregation, but not for
longer-term segregation – but there was rigorous debate about the amount of time constituting
“short term” and about acceptable minimum conditions.
It was generally agreed that too many people are in the “longer-term” category than
should be. One participant offered the example of prison systems that put condemned persons
in long-term segregation because of the conventional view that they are risky due to their “I
have nothing to lose” position. Yet, according to this participant, when condemned persons
were housed in non-restricted housing, they did not exhibit more violent or risky behavior, and
in fact showed improvements in mental health and social interactions.
Conditions for Short-Term Segregation
The most common “definition” of the harshest form of segregation is being confined to a
cell for 22 hours every day. It was agreed that setting a specific number of hours as a minimum
standard was not as important as defining the minimum standards for various dimensions:
physical, food, services, staff interaction, interaction with other people, programming, etc. After
some debate, agreement was reached on a specific phrasing for these basic conditions – which
is set out in the recommendations.
It was agreed that in order for short-term segregation to serve a punitive purpose, it
would have to have more restrictions and more austere conditions than the conditions in the
14

Hannibal Lecter (born 1933) was a serial killer, notorious for his habit of consuming his victims, earning him the
nickname "Hannibal the Cannibal." http://hannibal.wikia.com/wiki/Hannibal_Lecter

19

general population or any separated population. These aforementioned minimum conditions
are meant to represent the “floor” for what constitutes humane confinement. Some participants
mentioned examples of privileges that should not be offered in short-term segregation,
including additional recreation, TV, phone calls, and contact visits. Participants explained that in
order to create positive incentives, prison officials would need to have the ability to offer
privileges. Thus, prison officials cannot offer all the privileges as a minimum standard. Other
members agreed with this premise but maintained that this should not mean that persons in
segregation do not get any of these privileges.
Conditions of Longer-Term Segregation
It was agreed that only people posing more serious or ongoing risks of violence to others
should be in longer-term segregation, and that this risk should be determined by professional
clinicians through regular inter-disciplinary assessments. It was also agreed that there should
be a clear decision to “transfer” a person from short-term to longer-term segregation– rather
than simply shifting them automatically after a certain period of time.
Some participants argued that the risk of harm posed by longer-term segregation is only
justifiable if the facility provides interventions or programs that attempt to address the
underlying cause of the risk of violence – that is, if there is an effort to give the individual a path
out of the longer-term segregation. The other participants agreed with this view in principle, but
expressed concerns about resource requirements for such programs. The group agreed that, in
theory, this approach could also involve a graduated program, in which behavior improvements
could lead to more time out of the cell. It was acknowledged that such interventions would not
necessarily work for everyone, but that it is important to offer them. All the participants agreed
that these interventions should address behavior and not merely be recreational in nature.
Group 2 frequently mentioned having a minimum standard of 20 hours per week of
programming activities for those in longer-term segregation.15 A few participants noted that
some of this programming could occur with some level of restraint, if necessary, rather than
denying it completely to those needing restraint.
It was generally agreed that the term “step down” could be unhelpful. They briefly
discussed how the “levels system”, which is meant to be a path out of segregation, could be a
trap for people who have underlying behavioral or mental health problems. There was no
agreement explicitly on whether to recommend eliminating it.
Several participants stressed that the conditions in longer-term segregation needed to
be more restrictive than the conditions in the general population. The rationale for this was that,
if individuals had nearly equal access to services and privileges in segregation and the general
population, but a “private cell” in segregation, they might try to get into longer-term segregation

Sometimes referred to as “10 and 10” meaning 10 hours a week of outside recreation and 10 hours or
therapeutic intervention, averaging about 3 hours daily.
15

20

on purpose. Given this, there was agreement to add the caveat that conditions in longer-term
segregation should typically be more restricted than in general population.
Maximum Length of time for Short-Term/ Punitive Segregation
The group agreed that the current length of time used for disciplinary segregation is too
long – many years, in some cases – and that those in segregation for “risk” reasons tend to be reassessed too infrequently.
The notion of a 15-day limit on short-term segregation, for disciplinary or cooling-off
reasons, under the most austere of conditions was discussed next. The 15-day limit is a clear
reference point because it is defined as “prolonged solitary confinement,” in the Mandela
Rules16 and recommended by the UN Rapporteur. All conceded that this is a relatively arbitrary
number, without a clear rationale from social science. Some members strenuously argued that
the maximum amount of time allowed in the harshest of segregation should be less than 15
days, regardless of whether the individual is suffering harm (e.g. mental health deterioration).
Other participants suggested that the maximum amount of time should be short but could
depend on how resilient the individual is to the harms of segregation, based upon clinical
assessments. Views on an appropriate maximum length of time for the harshest segregation
ranged dramatically – from one day to more than a year. There was agreement to recommend
the “briefest possible period” – a phrase that conveys the urgency of a short time, without
prescribing a number of days. The issue of a segregation time clock being “suspended” when a
person misbehaves during segregation was also discussed. Some participants argued that this
tactic is not effective as a deterrent, and can make a person more angry and violent. Although no
explicit recommendation on this issue emerged, there was agreement on the notion that “time
suspended” is not a helpful or effective practice.
Some members repeatedly said that imposing an absolute 15-day limit on short-term
segregation would be “too much, too fast” for some prison officials, and that having such a
number risks causing a backlash. Given this and the fact that other members would not accept a
higher number of days, no consensus on a maximum time limit was reached.
Alternatives to Segregation
Agreement was easily reached on the recommendation that alternatives to segregation
should be used more widely and that segregation should be used as a last-resort. A main
alternative to segregation is separation. In the case of predatory individuals, one example is
gang members who are predatory towards specific individuals, but not toward all, and staff. All
the participants agreed that gang members could be kept in a separate area, with others to
whom they are not a risk, and with access to a communal space, programs, etc. The main
obstacle to this alternative is that many facilities do not have the structure or space for this type
of physical separation. Some participants also pointed out that classification systems attempt to

16

see fn 6 supra.

21

do this to some extent, but that prison staff and buildings often do not have the resources to
build a separation system with enough levels and categories.
Other types of separation, including transferring individuals from segregation to the
general population of another facility or another state system were suggested. Participants
noted that in many cases, the individual’s behavior problems were contextual, and thus stopped
after the transfer. It was agreed that such transfers should be promoted as another version of
separation, which is an alternative to segregation.
Due Process
Issues of due process in the hearings on incidents of misconduct and disciplinary
decisions were briefly discussed. Several participants commented that because the
administrative systems are overloaded with cases, hearings are often delayed for weeks or
months, and sometimes officials do not have time to consider each case in sufficient detail.
Participants underlined the dilemma that incarcerated persons face: If they accept a short
disciplinary segregation “sentence” without a full hearing, they de facto incriminate themselves.
However, if they insist on a hearing, they can be in segregation for several months, waiting for
the disciplinary hearing. Some members suggested that the hearings are not fair, because most
of the accused are found guilty. A few people suggested that individuals are pleading guilty –
regardless of actual guilt – simply to reduce their time in segregation. Overall, there were many
concerns about due process, but all agreed that additional external oversight or interventions
might not solve the problem. Due to a lack of time, a consensus or recommendation on this topic
was not reached, and therefore listed it as requiring further discussion in the future.
Transparency and Accountability
There is agreement that there is clear need for more data to be collected and analyzed.
Some noted that states have different capacities for in-house data analysis. Therefore, there was
consensus that there should be flexibility in whether the capacity for data analysis exists via the
agency or through partnerships with universities. Many participants pushed for more detail as
to the type of data that is made available pertaining specifically to segregation. It was agreed to
recommend that agencies should collect and make data available on a specific set of issues
related to segregation.
Staff Buy-In
In the initial exercise, several participants listed lack of staff buy-in as an obstacle to
change. During the discussion, administrators argued that even if everyone agrees on the end
goal in terms of how to reduce the use of segregation, the implementation should be gradual, as
staff is often resistant to change. Members suggested that changes in recruiting and hiring of
staff – for example, selecting people for their social-work skills and orientation – could shift the
level of buy-in. A few participants commented several times that even the best-designed policies
require consistent understanding and implementation by staff, and that in some facilities, high
staff turnover is more of an obstacle than staff resistance to change. A concrete
22

recommendation on this topic was not agreed upon, although this was more because there was
not a clear suggestion than because there was explicit disagreement.
Strategies for Change
All participants agreed that sustainable change to segregation policies requires some
change in legislation. They also agreed that there should be a concerted effort to focus on this
arena.
Another strategy dilemma related to resources. Everyone recognized that reducing the
number of people in segregation and the length of segregation may save money – but some
members noted that, typically, state governments reabsorb any savings resulting from changes
in prison expenditures, rather than reinvesting them in other prison services or expanding
buildings or staff. Removing people from isolation may rquire new resources and incur new
costs.
It was agreed that advocating for the reduced use of segregation on a cost-savings
rationale alone is unhelpful, as it contradicts the equally important need for additional
resources for more restorative services. Several participants noted that the problem of the overuse of segregation is in some ways self-generated by the under-resourcing of prisons: when
incarcerated persons are under pressures due to overcrowding and lack of services, they are
more likely to lash out, which leads to discipline and the over-use of segregation, which in turn
drains resources away from better conditions and services for the general population.
With regard to how to “sell” changes in segregation policies to the public and to
politicians, some members cautioned against using the advocacy strategy that reducing
segregation results in reduced recidivism within prison. Given the punitive public attitudes,
they suggest that the only argument that resonates with the public is that reducing segregation
has clear public safety outcomes.
Finally, the tradeoffs of prescriptive recommendations versus general guidance were
debated. Some members suggested that when guideance too directive, many prison officials
might balk. Rather, they contend that general guidance backed up by examples of successful
outcomes of new initiatives or changes in policy would be more persuasive.
All participants agreed that this meeting was a unique and crucial opportunity for
generating real change – and that if this change process is not done quickly and well, the
window will vanish.
It was frequently noted during the discussion that the key point or conclusion would
only apply to prisons and not jails, particularly on topics of clear differentiations of physical
space and programs for different categories of individuals, given that there is less space and
resources, and more flux in jails and that different types of infractions and incentives are
common in jails versus prisons. The group recommended that separate discussions be held on
the topic of the over-use of segregation with specific reference to jails and (separately) to the
juvenile justice system.
23

Group 3:


How can vulnerable individuals, such as the seriously mentally ill, pregnant women
and adolescents be protected adequately without the use of isolation?

The group was charged with discussing issues involving reforming the use of social
isolation for vulnerable individuals such as the mentally ill, youth, and protective custody
populations. Currently, many facilities in the United States use extreme social isolation to
manage these vulnerable individuals. Is there consensus on proposals to reform this system and
if so how best can we implement change?
There was initial disagreement on how to interpret the questions posed by Conference
organizers. One participant raised concern about whether to start the discussion with
identifying individuals who need protection or whether to start the discussion with a discussion
of individuals who need services. Also mentioned was that not all individuals who may be
labeled vulnerable will be dangerous to others. Vulnerable individuals can be dangerous to
others but they also may be vulnerable, dangerous, and need protection from themselves. Some
members interpreted these questions as asking “who is vulnerable and who should be placed in
a special environment” while others interpreted the questions as asking “which categories of
individuals should we worry about when taking those individuals out of general population and
placing them in a separate unit.”
A participant suggested that the discussion should not begin with labeling and
discussing protection for certain individuals but rather the discussion should begin with
determining what services and programs individuals should be provided. It was suggested that
inadequately treated individuals are the ones who could potentially be vulnerable and if those
individuals were treated adequately then they would not need protection.
To focus the group back to its goal, the facilitator clarified that there are two
conversations to have based on the interpretations of the questions posed. The questions
would address:
(1) Whether, if individuals are treated with services and programs, would they still need
to be isolated; and
(2) If those individuals do need to be isolated, what conditions of confinement and
services should they have?
Additionally, the questions could also be interpreted by asking whether an individual is:
(3) Vulnerable to the general population; and/or
(4) Vulnerable to the problems of segregation.

24

Who are vulnerable?
The old DSM17, AXIS I, and AXIS II diagnoses relating to manic depression, bipolar, and
functional impairments that define serious mental illness (SMI) is in litigation. However, the
new DSM has moved away from a pure diagnostic approach to a functional approach. Some
believed SMI is too broad a description and that there are circumstances where an individual
may not fall under SMI but may still require protection. It was also suggested that the specific
individual needs must be managed and not the specific diagnosis. Suggestions were again raised
that rather than using separation and isolation for the mentally ill, programs to treat the
mentally ill should be implemented in order to allow persons with mental illness to become
functional in general population.
Agreement was reached on the following categories of individuals who are potentially
vulnerable in the general population:


Serious Mental Illness “SMI”



People with intellectual disabilities



Juveniles



Old/Elderly



Infirm



Intakes



LGBT18



Protective custody



Pregnant



Chronically ill



Sex offenders

However, one size does not fit all. Individuals need to have personalized and individual
treatment plans and not merely be placed in populations solely based on being labeled as being
The Diagnostic and Statistical Manual of Mental Disorders (DSM), published by the American Psychiatric
Association (APA), offers a common language and standard criteria for the classification of mental disorders.
The DSM is now in its fifth edition, DSM-5, published on May 18, 2013.
18
Although LGBT was included as a potential vulnerable population Group 3 did not come to a consensus for
LGBT and therefore LGBT was taken off the list of vulnerable populations.
17

25

in a vulnerable category. Placing an individual from a vulnerable group solely based on status is
not a solution to the problem. Corrections officials should not be forced to place these
individuals in segregation in order to protect them but rather these individuals should be
provided opportunities to find ways to live in general population.
Should Vulnerable Individuals Be Held In Separate Groups Or Can They Live In The General
Prison Population?
It was suggested that the default analysis is that all incarcerated persons should be in
general population irrespective of their classification, whether they are LGBT, an informant, etc.
The job of corrections officials is to maintain a safe environment for all. Corrections officials
could create separate units that place vulnerable individuals with other informants, sex
offenders, etc. but this can get dicey. For example, would that just create an LGBT ghetto?
Several participants agreed that isolation is not a solution to the problem and should not
be used to as a means to protect people. An individual in segregation, merely because of their
status and not because of their behavior, should not be isolated but rather should be provided
programs. Therefore, the question becomes how do we create living units for vulnerable
individuals to live, which are separate from the general population, but are not isolated and do
not create ghettos? No one should be isolated based on status.
It was further suggested that the analysis be as follows:
(1) What is the person’s status? Based on status alone, a person should not be placed in
segregation;
(2) What special needs this individual has that correction officials must address?
(3) If the individual has a vulnerable status and also has behavioral issues, how do we
respond? If the individual is mentally ill, that person should receive therapeutic
programs that travel with the individual;
(4) What do we do with people who cannot be in the general population because of their
status or because of what they have done; and
(5) What resources will we need to do this?
Discipline for Vulnerable Populations
A participant then posed the question of how to handle an individual who is not isolated
based on status but rather isolated because of disruptive behavior. Another suggested that this
type of situation be addressed with therapeutic intervention, which goes back to the discussion
of these individuals needing services. It was emphasized that status is different than behavior
and something has to be done with regard to disruptive behavior. It was suggested that the
service travel with the incarcerated person wherever that person is.
26

It was also recommended not to base the discussion upon individuals who are out of
control. One participant strongly advocated what is called an “intermediate care unit.” For
example, if an individual with mental illness gets in trouble for hitting an officer, that
individual’s disciplinary plan should stay within the intermediate care unit. Another asked
would this just be segregation within segregation? The response was no, because individuals in
an intermediate care unit can still receive their services and join the general population.
The facilitator clarified the question asked, “what happens to a vulnerable person who
does something in violation of prison rules?” There was agreement that if an individual is
mentally ill that person will at some point do something in violation of prison rules.
It was noted that in the State of Washington, discipline is capped at 30 days. It was
suggested that there is a difference in isolation based on status and isolation based on discipline.
If a person is separated because of their need for protection that person’s experience should not
feel punitive. There is no reason to deny that person visitation or anything else that they would
otherwise have if they were in general population.
Resources
In order to change this process and to do it right we will require resources. A
comparison was made to aging individuals outside of the prison context. When an individual
ages and cannot live in their home any longer the question arises as to where they should go.
Prisons can be compared to a town or city that builds facilities that can house smaller
populations. Just as when a person ages and is placed in different facilities, individuals in prison
should be put in different facilities.
However, resources are an issue. For example, in Illinois there are few doctors and
social workers that are willing to travel to southern Illinois where the prisons are located.
Southern Illinois cannot provide services that are normal or typical to a bigger city because it
lacks the means and resources.
One participant suggested that it might be that we cannot run our current prison
systems the way we want them to run with the current resources and funding we have. States
will have to invest in creating new prison environments in locations that will support the needs
of vulnerable populations.
Extreme Social Isolation (22-Hour A Day Lockdown) Of Vulnerable Prison Populations
Next discussed was, “to what extent should people in these categories be held in cells for
22 hours a day because of their status?” A consensus was quickly reached that no one in a
vulnerable category should be held in their cell for 22 hours a day solely because of their status.
It was suggested that vulnerable populations should have as close to the level of amenities that
the general population has, in a setting as similar to the general population as possible, while
still being afforded the rights and privileges an individual would otherwise have in the general
population. The accommodation should be customized to the risk of harm presented, with the
27

same range of services as the general population. It was even suggested that these populations
should receive more amenities than the general population.
However, this raises the issue of some individuals wanting to be placed in these units
because they will receive a quiet, single cell while still receiving all the same amenities as the
general population. Certain individuals prefer to be placed in special needs units because it
makes prison life easier for them. We need to be careful in extending social interactions and
services beyond what is offered in general population because if it becomes better in these
separate units then individuals may seek to get these additional privileges.
Services and Programs for Vulnerable Populations
Overall, services available in the general population are insufficient for vulnerable
populations. Units with vulnerable populations need additional services, socialization directed
towards a goal, and progress should not be based on a behavior plan, which is impossible for the
individual to meet. A negative reinforcement and punishment-based system should be opposed
because individuals with mental disabilities cannot meet these requirements. Rather, a positive
incentive based program should be used.
Additionally, individuals with mental illness can be mixed with the general population as
long as they are receiving services. The goal should be to figure out programs that work for the
majority of people and not the outlier for now, as special programs will be needed for those
individuals. One participant cautioned that designing a program for the special outlier
prisoners was how we ended up with supermax prisons in the first place. These comments
however drew the question of how then do we manage the 2% of the outliers?
The question was posed, “how do you meet the needs of individuals who are in a
vulnerable population who want to participate in programs that are only available to the
general population such as congregate religious activities or school?” It was suggested:
(1) Using escorts to take the individuals to the services or programs;
(2) Allowing the individuals to interact with other incarcerated people they trust and are
in their same status; and
(3) Allowing the individuals out of cell time with staff and outside personnel who come
to fill the day of the individual (art, conversation, passive recreation, e.g.).
Consensus was reached that the mere fact an individual is in a vulnerable population,
they should not deprive them of the same services that are provided to the general population.
Individuals who are Vulnerable to Being Placed in Segregation
The discussion changed to individuals who, if placed in segregation, can become
dangerous to themselves and who can be vulnerable to the conditions of isolation. It was also
agreed upon that juveniles should never be held in an adult facility.
28

It was agreed that vulnerable individuals should not be placed in extreme social isolation
just because of their characteristics or vulnerabilities. It was also agreed to use the term
“solitary confinement” as defined by the Mandela Rules. This rule defines solitary confinement
as “22-hour lockdown without meaningful human contact.” Furthermore, the group
determined that in its deliberations it considered two types of vulnerable persons:
(1) Vulnerable individuals who are susceptible to harm in the general population; and
(2) Vulnerable people who are in special danger of harm when they are placed in
solitary confinement.
When, if ever, can isolation be imposed on vulnerable individuals?
The first proposal considered dealt with when, if ever, 22-hour a day lockdown should
be imposed for vulnerable populations. The various risks and resources available to prison
officials to deal with vulnerable populations were discussed. There were special considerations
raised such as separate units for pregnant women. Small county jail facilities, which do not have
sufficient resources to adequately care for mentally ill people, were also raised. A correctional
administrator commented that until he can assess and stabilize an individual who has slashed
someone, he knows that the individual is going to present a risk, and he struggles with how to
allow this person contact with others while still protecting others. . An advocate conceded that
the current practice is to put these individuals in isolation but the advocate wants to see a more
flexible system that acknowledges that prolonged confinement is not the answer. The advocate
stressed that the risk to the individual needs to be balanced with the risk to others and that even
in violent situations these individuals should not be placed in isolation. An administrator also
suggested that ignoring the behavior of an incarcerated person who slashed someone does not
help the situation.
At this point how long an individual could be held in isolation before services are
provided and he or she is evaluated was considered. Placing a vulnerable individual in 22-hour
a day lockdown for an indefinite period is not the answer. Rather, there should be an initial
period of isolation to calm and address the threat and thereafter isolation cannot extend to a
longer period of time. These individuals can be contained until the issue is addressed but then
they need to be transferred to another unit that will address the vulnerable person’s needs. One
advocate proposed that individuals could stay in isolation for a few days while the situation is
calmed and addressed. However, a strategy needs to be developed and then the individual
should be placed in a specialized unit as soon as practicable. Another advocate countered that
an individual should be placed in isolation only when all other alternatives to de-escalation of
the immediate dangerous situation fail. The restraint needs to end when the emergency ends
and there needs to be time limits for when the individual should be released, specifically when
the individual is no longer dangerous to themselves or others.
An issue that arose with these limitations and time limits is that, as a practical matter,
there is a possibility of a lack of hospital beds being available, limited resources in smaller or
more rural facilities, or the lack of other alternatives to remove the individual from isolated
29

confinement. In such circumstances a suggestion was made to assess the individual within 24
hours of being placed in isolation, to have access to mental health and medical services,
immediately if possible while in isolation, and to make an assessment to place the individual,
which will all mitigate the effects of isolation. However, it was recognized that there may not be
another facility, unit, or alternative placement available because of a lack of resources especially
in areas which have small prison populations. Adding on to this suggestion, one advocate
commented that the individual should be seen by a physician within an hour of being placed in
isolated confinement, then they must be reassessed a certain number of hours later, followed by
a disciplinary team meeting to determine a treatment plan, and a reassessment once they are
out of isolation.
This suggestion met some disagreement with one advocate who asserted that
vulnerable individuals in isolation should not just be assessed only when something goes wrong
but rather they should be assessed immediately. If the individual is assessed and held for longer
than 7 to 10 days, two weeks, etc. then an alternative needs to exist because these vulnerable
populations cannot stay in isolated confinement without being especially harmed. Next debated
was what the alternative to isolation should be, acknowledging that a higher standard or
threshold should exist for vulnerable individuals to put them in isolated confinement in the first
place. Most participants suggested 10/1019, which averages to about 3 hours out of cell per day.
What should be done once the incarcerated persons are in isolation and how isolation
should be used was then addressed. A corrections official suggested that when an individual
presents a clear risk to themselves or others and the individual needs to initially be placed in
isolation, the goal should be to address the needs of the individual within a specialized unit or a
restricted environment, with a plan to get the individual out and back to general population
while appropriately addressing the individual’s need for services.
An advocate suggested that no disciplinary segregation should ever be imposed on
these individuals, but rather a mental health standard should be used. Thus, only
individuals who are dangerous to themselves or others should be placed in isolation.
The decision to place an individual in an Intensive Treatment Unit can be made by
custodial staff but then must be validated by a physician. The discharge from isolation is up to
the physician. There ensued debate about the procedures to be followed, with some believing
that only mental health workers and doctors should be allowed to order physical restraints on
the vulnerable individuals and others who wanted to allow the prison official to be able to
physically restrain the individual.
Multiple placements and the frequency of isolated confinement were next considered.
Many ideas were suggested including a limit of only 15 out of 30 days a month, dependent on
the circumstances, and no more than 15 days at a time without at least 7 days of non-isolation
between being released and before placing the individual back in again. One corrections official
10/10 – 10 hours of structured therapy or education and 10 additional hours of recreation and other forms
of “unstructured activities” – comes out to 3 hours a day of out-of-cell time
19

30

did not agree with these set number of days and gave an example that an individual could be
released and then officials will know that in, for example 3 more days, they can put that
individual back into isolation.
A corrections official was concerned about an offender who is put through an incentive
based program, and within one hour of being released back into general population assaults
another person and is placed back in solitary. It needs to be recognized that there are some
individuals who do not want to be placed in general population, and who will act out to avoid
being placed there. There needs to be provisions for recognizing that these individuals do exist
and to develop standards for how to address their behavior. One corrections official pointed out
that the goal is not to craft standards that will cover every conceivable circumstance, but instead
to figure out what will work for 99% of the population and not consider the outliers at this
point. It was argued however that there should be an exigent circumstances provision, or a
"carve out" for these individuals who are the tough cases.
The discussion then turned to the specificity and precision of language used to define
extreme isolation and also how to mitigate the effects of isolation on an individual. Extreme
isolation needs to be defined (example of solid doors vs. bars or glass on doors) as these
distinctions make an important difference.
Is there a Limit to the Number of Days a Vulnerable person can be placed in Isolation?
Whether 15 days was workable for vulnerable people to be held in 22-hour lockdown
was then considered. One administrator argued that 15 days is not workable, but he thought
that 30 days was an acceptable upper limit, and that 45 days was excessive. Another
commented that any system we devise should allow flexibility and we should not set up the
system to fail. Another participant suggested that 10/10 could be the plan but not 22-hour
lockdown. Several states do exclude mental illness from isolation and give them 10/10 but the
question then arose, if the person is in segregation but is allowed out of his or her cell for 20
hours a week is this still isolation?
When should a Vulnerable Person in Isolation be assessed?
The practical issues of how fast medical attention or assessment can occur and how
detailed or specific a rule should be was considered next. Connecticut uses a 30-day rule and it
was thought that 15 days is not enough time. It was suggested that 15 days should be the
baseline but with a procedure to allow for another 15 days with a mental health assessment if
necessary. Many seemed to think that it would only take 2 or 3 hours after a vulnerable person
is put in isolation to have that person assessed by a medical professional or a corrections official
and that a treatment team should meet with the individual on day one. However, this
assumption runs counter to the current standard of 72 hours to assess and formulate a plan for
the person. A concern were raised about what information a treatment team can get on the first
day to understand what the behavior is about and how long is necessary for medication to take
effect?

31

Who is Especially Susceptible to Harm in Isolated Confinement?
The group then resumed the discussion from Wednesday about which vulnerable
people are particularly vulnerable to the effects of isolated confinement. The issue of the elderly
and infirm was addressed first. Agreement was reached that the elderly and infirm should be
included under the vulnerable to isolated confinement list but did not want to get into age
specifics. No consensus emerged whether LGBT, Protective Custody, the chronically ill or sex
offenders should be included in this particular list.
The issues of pregnant women were addressed and it was argued that women do not get
what they need (healthcare, exercise, and nutrition) to keep their babies healthy when they are
in isolation. There was some push back to this notion because it was argued that women in
prison might receive better healthcare than they would on the street and that prisons are
exceptional care givers for some mothers and babies. It was clarified that we were only
discussing pregnant women in isolated confinement and not in prison generally. The
conversation continued with one correction official sharing the story of a pregnant woman who
was jumping and diving off of her prison bed in order to abort her baby. The state has a moral
obligation to protect an unborn child from such attempts to harm it and the woman was placed
in isolated confinement in order to protect her life and the life of her unborn child. It was
acknowledged that if someone is harming herself or others then the prison needs to protect her,
however, she should not be in 22-hour isolated confinement for more than 30 days. Following
this discussion consensus was reached that pregnant women are a population that is especially
vulnerable to the harms of isolated confinement.
In sum, it was agreed that the following persons are significantly vulnerable to the effects of
isolation and the above principles should apply to them:


Serious Mental Illness



Intellectual Disabilities



Serious Cognitive Limitation/Impairments



Juveniles (18 and under/defined by state law)



Elderly/Infirm (without a specific age)



Pregnant Women

Group 3 also reached consensus quickly on populations that are vulnerable in general
population. This category was defined as including the following people:


Serious Mental Illness



Intellectual Disabilities/Developmental Disabilities
32



Juveniles (18 and under/defined by state law)



Elderly/Infirm (without a specific age)



Protective Custody



Chronically Ill



Sex Offenders

Although there was discussion, there was no consensus on LBGT people.
Should there be a Classification Appeal for Vulnerable Individuals?
Group 3 reached consensus that a multi-disciplinary approach should be used but there
was disagreement as to whether the incarcerated person should have a say in his or her
classification and placement into a separate unit. One advocate argued that the individual
should have input, not necessarily a vote, but input into the decision citing PREA. There was
disagreement from a corrections official that he was not in favor of such input and did not want
to put it into the principles because it would set an expectation that he did not believe
warranted. It was also suggested that classification moving people into separate units is a
decision that is made at the facility level by mental health professionals. It was decided, that:
The procedures for determining whether to place a vulnerable person in the general
population or in a separate unit should be reached through a multidisciplinary process, which
includes input from the individual, regarding which special unit they should be placed in. The
procedure need not be a formal procedure such as an Individual Education Plan or a
disciplinary due process hearing.
This question arises when an individual wants to be housed in general population but
the decision is to place him in a special unit and it also arises when the individual is placed in
general population but feels that he should be held in a special unit for vulnerable persons. No
consensus was reached on this issue. There was a discussion over liberty interests, due process
issues, and the fact that in many systems classification is not grounds for a grievance. One of the
corrections administrators stated that there are procedures for an appeal can be pursued in
order to challenge one’s classification. The conversation ended with no consensus of what the
appeals process would be for an individual wanting to challenge a classification decision of
whether or not to place him in a special unit.

33

General Approach and Step-Down Program for these Separate Units
Group 3 briefly discussed this issue and commented that a step down program20 is a
very good idea and should be the goal but that some prisons have small populations of
vulnerable individuals and therefore do not always have the resources for a step-down
program. A correction official stated he would endorse a step down program but asked
how we can do it when it is not always suitable. It was also noted that before attempting to
initiate a step down program there should be improvements to amenities and incentives
that will lead to the next step with good behavior. Since time was running out the it was
agreed that they would endorse that step-down programs were a good idea but that it
should be indicated in the notes of their deliberations that there was not enough time to
discuss the details of this program. There was an understanding that resources or certain
circumstances may not allow for a step-down program. One advocate wanted to note that
incarcerated persons should never be serving “dead time”21 and there should always be a
next step where they receive services and mentioned that the mental health treatment
programs function similarly to the step-down program.
Accountability and Transparency
The members of Group 3 were unanimous in their belief that accountability and
transparency is essential to reform efforts. It was suggested that anyone and everyone should
be brought into the fold, people should be allowed into the prisons, critical advisory boards
should be in place, outside monitoring, and that everything should be public. Transparency and
accountability is important for the overall effort. An additional point was that in the process of
accountability more data should be collected as to the performance of correctional institutions’
treatment of vulnerable populations. This is because over time we must be able to show that the
implementation of our recommendations makes a difference and that these units are successful.
We cannot just focus on the bad data that comes out; rather, the good data will prove right more
than wrong. It was agreed that:


Transparency is mission critical, ensures ethical and moral appropriateness, and
ensures a commitment to positive performance.



Transparency, external and internal accountability, and robust data supporting
measuring outcomes, are essential and critical to the success of these units and
should be publicly available.

Some measures that would help guarantee success were agreed upon. These measures
included data on institutional force and violence, suicide attempts, grievances, disciplinary
tickets, assaults on staff, and cell extraction. Internal accountability gives people the awareness
A step down program is an incentive-based, multi-step process that provides those placed in isolation the
opportunity to earn enhanced privileges by refraining from participation in Security Threat Group (STG)
affiliations and behaviors. The ultimate goal of SDP is to release the persons from the isolation unit.
21 Meaning time with no intervention or opportunity to improve one’s condition or term of imprisonment.
20

34

that outside people are watching their behavior and their behavior will be considered when
decisions are made about them.
Barriers to Reaching these Goals
The last question tackled by was the barriers to reform measures. Everyone said lack of
resources is a barrier. It was discussed that many of those most at risk present an expensive
problem for the system therefore it was suggested that money and resources need to be
reinvested in separation units to address these vulnerable individuals. The more problematic an
individual, the more resources they need. To bolster that point it was suggested that a needy
person will often act out as a way to get attention and this would not happen as often in a
healthy system that would address the incarcerated persons needs before they act out. To
express these thoughts it was agreed that:


There are barriers to achieving reform that will be distinct based on the facility and
jurisdiction.



Resources will vary by the system and state and each jurisdiction is ultimately going
to have to come up with a solution for adequate resources that will work for their
system through new funding or redistribution.

The group contemplated two different vulnerabilities that we used as the main goal to
approach our discussion. The first vulnerability occurs when an individual is vulnerable to the
harm associated with isolation and the second vulnerability is when an individual is vulnerable
to other prisoners. As soon as someone in the first category, vulnerable to isolation, is placed in
isolation, the goal is to work on getting him or her out as soon as possible. When an individual
from the second category is placed in isolation we need to work on alternatives.
It was also suggested that for those individuals who are separated, but not in isolation,
that separation should not deprive those individuals of habilitative, rehabilitative, educational,
similar opportunities, or even additional opportunities to those available to the general
population. Separation should not result in a lack of access to opportunities.
Number of Days that Vulnerable Individuals can be held in Isolation
Group 3 recommended that vulnerable populations should not be placed in isolation for
longer than 30 days. Some participants were not willing to come up with an exact time limit but
rather suggested that it should be the shortest amount of time necessary.
It was stated that in litigation involving individuals with Serious Mental Illness (SMI) a
complete prohibition on their placement in isolation has resulted. However, others were not
aware of case law that says short-term isolation is unconstitutional. Some people stated that
case law allows for the use of short-term isolation for these individuals. For example,
Massachusetts and New York allow for short-term isolation up to 30 days. This short term
isolation provides corrections officials with time to figure out the right medications Colorado
35

provides for no segregation for individuals with SMI, however, they would receive 10/10, while
escorted, in restraints, at restraint tables, etc. until they are safe.
Strong opposition was voiced to the 30-day limit for SMI and reference was made to
the American Psychiatric Association’s Position Statement on Segregation of Prisoners with
Mental Illness:
Prolonged segregation of adult inmates with serious mental illness, with rare exceptions,
should be avoided due to the potential for harm to such inmates. If an inmate with serious
mental illness is placed in segregation, out-of-cell structured therapeutic activities (i.e.,
mental health/ psychiatric treatment) in appropriate programming space and adequate
unstructured out-of-cell time should be permitted. Correctional mental health authorities
should work closely with administrative custody staff to maximize access to clinically
indicated programming and recreation for these individuals.22
The consensus proposal was that the recommendation should be much less than 30
days. Opinions were that 30 days is punishment; nothing is served by 30 days; and that 30 days
should be the limit but less is better.
Discussion as to Services
The services available in prison do not compare to the services of a psychiatric hospital.
We need to recognize the limitations of correctional resources that are available during the time
an individual is in isolation. Individuals with mental illness need to be helped in order to regain
control when they act out rather than just placing them in isolation. Most prison systems do not
have enough psychiatric beds and sometimes there are worse conditions of confinement in
psychiatric cells than in isolation. There should be daily mental health services that involve
contact with mental health staff, interaction, and programs. Individuals with SMI need to receive
treatment and out of cell time. Individuals in isolation should receive treatment immediately. If
a person placed in isolation has a mental illness there should be an immediate evaluation. For
an acute incident, it should not take more than an hour after the person enters isolation to
evaluate them.
Discussion of Discipline
When an individual with SMI also has a behavioral issue that person may need to be
separated to keep everyone safe, however, this issue is separate and apart from what services
the person receives from the first day he is separated. Group 3 did not contemplate isolation as
being used in any way to punish a vulnerable person such as an incarcerated person with SMI \.
In Colorado, if an individual with SMI commits an act caused by their mental illness, it is
believed that punishment is not appropriate and that intervention to address the underlying
American Psychiatric Association Position Statement on segregation of prisoners with Mental Illness,
approved by Board of Trustees, December 2012.
22

36

mental illness is called for. In Colorado, even for discipline, they use a 10/10 plan and some
individuals stay in 10/10 forever. However, there should be caution in trying to determine
whether the mental illness caused the person’s behavior. There is no easy way to determine
whether mental illness is the cause of the behavior. Disruptive behavior should be thought of
separately and not a reason for punishment. We need to find alternatives to isolation if
punishment is the goal. We should not punish someone in a way that will cause him or her
serious harm. For example, we cannot punish a diabetic by not giving them insulin. This does
not mean that you can never discipline someone with a mental illness. Rather, it needs to be
clarified that an incarcerated person can be held accountable without the use of social isolation.

37

RECOMMENDATIONS
1

Segregation should be used for the minimum time and in the least restrictive
conditions necessary to resolve the condition that led to the segregation.
1.1.1 For those in segregation or restricted housing, the goal should be to get them
into the least restrictive housing possible. If they are separated from the
general population, it should be for the shortest amount of time necessary.
We urge correctional officials to consider alternatives to segregation or
restricted housing.

2

Separation is one alternative to segregation or restricted housing. This can be
through moving someone to a different area of a facility, a different facility, or a
different prison system.

3

Positive incentives should be incorporated into the management of all incarcerated
people, including those in segregation or restricted housing.
3.1

4

All isolation should have an incentive component, which would restore
certain privileges if the individual were able to reach certain behavioral
goals. Ideally, these incentive programs would operate on relatively short
timeframes—e.g., two days of good behavior earns a reward—so that
incarcerated persons would quickly begin to see their good behavior
rewarded. Participants also agreed that isolation should have a goal of
changing specific behavior and an individualized achievable path to reach it.

Even for the most restrictive segregation, the conditions should be humane. These
conditions should include, at a minimum: access to natural light, control of light in
cells, basic sanitary and safe environmental conditions including adequate space,
ventilation and temperature, adequate nutrition, adequate medical and mental
health services, and reading materials. There should be initial and subsequent
periodic mental health evaluations of those in segregation or restricted housing, to
determine whether changes in conditions of confinement are warranted for mental
health reasons.
4.1

Segregation must include meaningful mental health rounds, health care
rounds, and adequate basic conditions.

4.2

Apart from the briefest possible initial period, all incarcerated persons in
segregation or restricted housing should have some access to out of cell time,
congregate activity, meaningful social interaction,
programming/interventions, phone calls, and visits, recognizing that the
extent of these privileges may be more limited than in general population.
The most restrictive segregation should be for the shortest amount of time
necessary.
38

4.3
5

Segregation or restricted housing for investigation purposes should be brief
and may require a brief period of restricted contact with others.

We recognize that there is a small number of people who will require prolonged
separation from the general population because they pose a threat of violence to
incarcerated persons or staff.
5.1

Their separation from the general population is not punitive and should not
be experienced that way. For these people, the conditions should be humane
and as close to general population conditions as possible (in addition to the
basic conditions listed in item 4 above).

5.2

These people should be provided with interventions to address their needs
and to promote their safe transition back to less restrictive settings.

6

All people in segregation or restricted housing should be periodically reviewed to
determine whether they could be released to a less restrictive environment (e.g.
having met treatment goals).

7

Responses to disciplinary infractions should be imposed on a continuum, with
segregation as the tool of last resort.

8

7.1

Segregation or restricted housing for disciplinary or management purposes
should be used only for the most serious behavioral offenses, such as
violence or threats of violence.

7.2

It should not be used for problems such as gang affiliation, status or political
belief or minor infractions, except for a brief segregation period for
investigation or cooling-off purposes.

There must to be due process protections in place.
8.1

These must include procedural safeguards for placement in segregation,
periodic review during segregation, and an exit mechanism.

8.2

This process should consider the severity of the offense, length of time spent
in segregation, fairness, and the ability of individual to comply with imposed
conditions.

9

The loss of privileges needs to be proportionate to the infraction, and must include a
pro-social incentive system to restore the privileges.

10

There should be family contact allowed while incarcerated people are in
segregation, as the loss of family contact can be extremely agitating for both the
incarcerated person and the family; increasing family contact and visits for
improved behavior can provide a strong incentive.
39

11

Loss of programming, social contact and family contact should be reserved for more
serious infractions or after other punishments have proved ineffective.

12

Anyone who is in segregation or restricted housing for more than a brief period of
time should be provided with interventions to address their needs and promote
their safe transition back to less restrictive settings.

13

Incarceration should be avoided whenever possible to prevent bringing vulnerable
populations23 into the prison system in the first place (e.g. juveniles should be in
youth systems and should never be in adult prisons; people with mental illness
should receive treatment and services elsewhere; elderly and infirm released on
parole; etc.).

14

Where incarceration cannot be avoided, every reasonable effort should be made to
manage the vulnerable individual within the general population environment and
provide adequate services to meet their needs while in the general population.
14.1

15

The determination to place a vulnerable person in the general population or
in a separate unit should be through a multidisciplinary process, which
includes input from the prisoner, regarding which special unit they should be
placed in. The procedure need not be a formal procedure such as an IEP or a
disciplinary due process hearing.

Where general population placement cannot be effectively managed without posing
an unacceptable risk, vulnerable populations should be assigned to separate living
units where their needs can be appropriately met with a goal of maximizing
congregate activity, habilitative, rehabilitative, and programmatic opportunities.
15.1

The separation accomplished in these living units is separation from the risk
posed by general population, not separation or isolation from all other
individuals. The conditions of confinement in these separate units should
never be punitive.

Vulnerable individuals who are so vulnerable to 22 hour lockdown that regardless of their behavior they
should have additional limitations on solitary confinement.
Significantly Vulnerable Individuals who are at high risk of harm in solitary:
SMI (serious mental illness)
Intellectual disabilities
Serious cognitive impairments
Juveniles
Infirm (elderly without specific age)
Pregnant women
23

40

16

For significantly vulnerable individuals at high risk of harm in extreme isolation, it
should be imposed only as a very temporary emergency measure, for no more than
15 days, when absolutely necessary to address immediate serious safety needs. No
later than 72 hours following placement in extreme isolation:
16.1

Measures to reduce social isolation, to ameliorate the risks from extreme
isolation, and to soften the environment should begin. E.g., for prisoners
with serious mental illness, structured therapeutic activities (i.e., mental
health/psychiatric treatment) in appropriate programming space and
adequate unstructured out-of-cell time; measures to allow pregnant women
adequate access to large-muscle exercise; etc.

16.2

Efforts to assess the prisoner’s behaviors and the best strategies towards a
goal of alternative long-term housing should begin.

17

In extraordinary cases in which a stay of longer than 15 days is essential any
extension must be based on an authorization by medical or mental health care
professionals in the exercise of their independent professional judgment, with
additional review each 7 days thereafter or more often if needed and in no case shall
extreme isolation for significantly vulnerable individuals extend beyond 30 days.

18

There is a consensus on endorsing step down programs for vulnerable individuals,
but no exact procedures for step down were agreed upon.

19

We acknowledge the importance of transparency in furthering reform and believe
transparency and accountability further the goals of public safety. Transparency
increases awareness and trust for the public, prison staff, and incarcerated persons.
Transparency is mission critical and ensures ethical and moral appropriateness and
ensures the commitment to positive performance.

20

Every agency should have data on the use of segregation or restricted housing,
including:
20.1

Demographics of individuals in segregation/restricted housing,

20.2

The nature of segregation/restricted housing,

20.3

Length of time in segregation/restricted housing,

20.4

Where individuals were released (internally or to the community).

20.5

Agencies should track the outcomes of those who are released from
segregation.

41

20.6

It is essential there be robust data collection which measures the outcomes
critical to the success of these units. This should be publicly available. Data
should include:
20.6.1 Institutional violence
20.6.2 Cell extraction
20.6.3 Suicide attempts
20.6.4 Grievances
20.6.5 Disciplinary tickets
20.6.6 Assaults on staff

20.7

Agencies should also:
20.7.1 Have the capacity to undertake research and data collection.
20.7.2 Make data publicly available on their website on a regular basis.
20.7.3 To the extent possible, be open to outside research projects for both
external and internal accountability.

21

Communication between advocates and correctional administrators may avoid
needless litigation, assure the responsible stewardship of funds, and help both
correctional staff and the public at large to understand reform in public safety
terms.

22

Advocacy should also focus on legislation to ensure sufficient resources, including
reallocation of resources saved by reducing segregation or restricted housing.
There are barriers to achieving reform that will be distinct based on the facility and
jurisdiction. There is consensus that adequate resources will be needed to meet the
recommendations set out above either through new funding or reallocation of
savings.

23

Efforts must be made to get staff “buy in” on reforms from the outset. Correctional
management should find ways to “celebrate courage in the service of public safety
through small victories.

24

We call for a separate conversation and set of recommendations on segregation or
restricted housing for jails and juvenile justice facilities.

42

Next Steps
Although the colloquium resulted in the consensus recommendations above, much
remains to be done and areas of disagreement and questions remain. The participants in
the colloquium specifically recommended continuing discussions like this specifically
directed to the unique challenges facing jails and juvenile corrections institutions.
The first step that should be taken is to further publicize and write about the
recommendations herein. Jurisdictions that have not participated will require assistance to
understand and implement these recommendations. Opportunities to incorporate the
ideas emanating from this colloquium exist in the work of the VERA Institute’s “Safe
Alternatives to Segregation,” initiative to provide assistance to state and local jurisdictions
interested in implementing some of these ideas. We have already discussed and shared
these recommendations with them and are exploring opportunities to support their work
and incorporate these recommendations into their practice.
As important as the recommendations themselves, what emerged from this
colloquium was the beginning of meaningful and respectful dialogue between parties on
both sides of the issue and with competing views of the problem. The dynamic that was
begun here can and should be continued by bringing the group together again both to hear
about progress that continues to be made, learn from the research being done by VERA and
others about “what works,” and to advance the discussion of outstanding issues and
narrow the range of disagreement.”
Learning from the medical profession, this colloquium can serve as the beginning of
a “consensus development” effort within corrections to address the use of social isolation.
Consensus methods have been used productively to solve problems in medicine and health.
Their main purpose is to define levels of agreement on controversial subjects. Advocates
suggest that, when properly employed, consensus strategies can create structured
environments in which experts are given the best available information, allowing their
solutions to problems to be more justifiable and credible than otherwise. The challenge
moving ahead will be selecting problems, choosing members for consensus panels,
specifying acceptable levels of agreement, properly using empirical data, obtaining
professional and political support, and disseminating results.24
Examples for further discussion include the best ways to manage the process of
“stepping down” an individual from social isolation in prison; specific time limits on the use
of extreme social isolation; definitions of “serious” disciplinary infractions as distinguished
from minor infractions that do not warrant the use of social isolation as punishment. The
goal would be to identify common definitions and “best practices.”

“Consensus Development at the NIH: Improving the Program”.
http://www.ncbi.nlm.nih.gov/books/NBK235517/
24

43

Finally, the strong recommendation from the colloquium for transparency and
accountability requires further work to determine accurately the state of knowledge and
available data in each jurisdiction regarding the prevalence and frequency with which
different forms of social isolation are being employed, how they are defined, and what the
outcomes are. Efforts have begun including the ground breaking work by the Liman Center
at Yale Law School and the Association of State Correctional Administrators to quantify the
use of “Administrative Segregation.”25 Studies such as that must be continued and
supported. A first step would be to survey jurisdictions to determine where their
segregation policies meet, exceed, or fall short of the articulated consensus principles. And
next to assess what policies re-write to at least meet the principles looks like. Subsequent
reports could analyze how changes based on the principles have impacted policy and
program outcomes. Following on that, there may be an opportunity to revisit the
principles in a few years to see where greater consensus or new principles have emerged
as a result of implementation experience on the ground.
John Jay and the Prisoner Reentry Institute look forward to continuing our efforts to
advance these efforts.

Sarah Baumgartel, Corey Guilmette, Johanna Kalb, Diana Li, Josh Nuni, Devon Porter, and Judith Resnik,
“Time in Cell: the ASCA-Liman 2014 National Survey of Administrative Segregation in Prison,” New Haven:
Liman Center, Yale Law School, September 2015. http://nicic.gov/library/030161
25

44

Acknowledgements
First, and most importantly we wish to thank the Jacob and Valeria Langeloth
Foundation whose commitment to promote and support effective and creative programs,
practices and policies related to healing from illness, accident, physical, social or emotional
trauma, and to extend the availability of programs that promote healing to underserved
populations allows efforts like this to pursued. The Foundation has taken a leading
position in supporting efforts to ameliorate the effects of extreme social isolation and
through their efforts have moved the conversation forward in important ways.
This colloquium could not have been possible without the assistance and support of
the staff of the Prisoner Reentry Institute (PRI) at John Jay College of Criminal Justice. PRI
takes seriously John Jay’s mission of Education for Justice and works hard to support our
“fierce advocates for justice.” Particularly we wish to thank Daonese Johnson-Colon and
and Aimee Baker whose skill and perseverance ensured that the colloquium would be a
success.
Finally, the outcome of the colloquium, these discussions, and our recommendations
are the result of the commitment and hard work of our facilitators, Brian Fischer retired
Commissioner New York State Department of Corrections and Community Supervision,
Andie Moss President of Moss Associates, and Professor Michael Mushlin of Pace University
Law School. Their leadership was supported and documented by our able team of
reporters Susruta Sudula and Jennifer Peirce of John Jay, Abigail Marion and Nick Reck of
Columbia Law School, and Erika Danielsen and Sarah Lusk of Pace University Law School.

45

Appendix A
Presentation by Bernie Warner, Secretary of Corrections, State of Washington

46

47

48

FROM
Suppression and containment

Use as punishment

Managing different types of
prisoners the same

IMS structured as a
time-driven system

TO
Intensive programming

Use as a management tool
when they cannot be safely
managed in population

Mission-specific housing to
target risk, need, responsivity

Behavior change through
programming and
congregate activity
49

50



Motivating Offender Change (MOC) Program
◦ Targets Security Threat Group
(STG/Gang) prisoners
◦ General population STG units
co-located at WSP
◦ Anger Control Training
◦ Four phases of behavior change
and development
◦ Incremental reinforcers to
encourage behavior change

51



Reintegration and Regression
Programming (RAPP)
◦ Targets mentally ill prisoners
◦ Co-location of Intensive
Treatment Unit, mental health
facility at MCC
◦ Cognitive Behavior Therapy
◦ One mental health professional
per 50 prisoners
◦ Individualized Treatment/
Behavior Management Plan

52

Number of Uses of Force over Fiscal Year in WSP - IMU
14

Number of Uses of Force

12
10
8
6
4
2
0
J

A

S

O

N

D

J

F

M

A

M

J

J

FY2012

A

S

O

N

D

J

F

M

A

M

J

FY2013
Fiscal Year

53



Intensive Transition Program (ITP)
◦ Targets chronic IMU recidivists
◦ Provides prisoners pro-social skills to successfully live in
general population
◦ Includes mixed cognitive-behavioral therapy curriculum
with phases and congregate activity



80% success rate
◦ Of the 131 program graduates ITP; 107 have not returned

54



Give staff professional development tools
◦ Core Correctional Practices
◦ Motivational interviewing



Engage staff in the change process

◦ Encourage interaction between prisoners and staff
through physical setting and interactive tools
◦ Having staff build programs, set up classroom,
etc.

55



Missioned Housing
◦
◦
◦
◦
◦
◦

Skill Building unit
Crossroads transitional mental health program
Sage unit - intermediate nursing care
Senior living unit
Veterans living units
Treatment units – sex offender; chemical
dependency; mental health

56

Appendix B
Presentation by Joe Ponte, Commissioner of Correction, City of New York

57

58

The Ending of Punitive Segregation
for 16-17 year-olds
October 1, 2014
• 50 adolescents in Punitive Segregation (25 in RHU + 25 in
regular P-Seg)
• 257 in custody
• 1,477 days owed total
December 31, 2014
• 0 adolescents in Punitive Segregation
• 176 in custody
January/February 2015
• 0 days owed
September 29, 2015
• 0 adolescents in Punitive Segregation
• 205 in custody
59

CAPS - Clinical Alternative to Punitive
Segregation
• Specialized mental health treatment of
seriously ill inmates who have committed
violence.
• DOC established CAPS in 2013 (Opened
Oct 17, 2013, CAPS began with 4 inmates
at AMKC)
• CAPS has 30 inmates (September 28, 2015)
• The Use of Force in CAPS was 40% lower
than the rate on Restricted Housing Unit
(RHU) during the first 6 months of 2015

60

PACE (Program to Accelerate
Clinical Effectiveness)
• Non-punitive model
• Created in January 2015 to build
on the CAPS.
• 57 inmates (September 28, 2015)
• Designed to encourage adherence
to treatment.
• Continuity of care and a teambased approach.

61

Commissioner Joseph Ponte
at the
New York City Department of
Correction
April 2014
• Commissioner Ponte Appointment
Summer 2014
• CAPS Program Expansion
December 2014
• Punitive Segregation for Adolescent Inmates ends
January / February 2015
• Enhanced Supervision Housing Created
• Elimination of Time Owed
April to July 2015
• PACE Units 1 & 2 Open
September 2015
• DOC Punitive Segregation Population declines
2/3
February 2016
• Punitive Segregation for Young Adult Inmates to
End

62

Appendix C
Presentation by Rick Raemisch, Executive Director, Colorado Department of Corrections

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89