Kosilek v. Spencer

United States Court of Appeals
For the First Circuit

No. 12-2194

MICHELLE KOSILEK,

Plaintiff, Appellee,

v.

LUIS S. SPENCER, Commissioner of the
Massachusetts Department of Correction,

Defendant, Appellant.

APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF MASSACHUSETTS

[Hon. Mark L. Wolf, U.S. District Judge]

Before

Torruella, Thompson, and Kayatta,
Circuit Judges.

Richard C. McFarland, Legal Division, Department of
Correction, with whom Nancy Ankers White, Special Assistant
Attorney General, was on brief, for appellant.
Frances S. Cohen, with whom Jeff Goldman, Christina Chan,
Bingham McCutchen LLP, Joseph L. Sulman, David Brody, and Law
Office of Joseph L. Sulman, were on brief, for appellee.
Andrew D. Beckwith on brief for the Massachusetts Family
Institute, amicus curiae in support of appellant.
Cori A. Lable, Daniel V. McCaughey, Kristin G. Ali, and Ropes
& Gray LLP on brief for World Professional Association for
Transgender Health, Mental Health America, Callen-Lorde Community
Health Center, Whitman-Walker Health, GLMA: Health Professionals
Advancing LGBT Equality, and Mazzoni Center, amici curiae in
support of appellee.
Joshua Block, Matthew R. Segal, and David C. Fathi on brief
for American Civil Liberties Union, American Civil Liberties Union
of Massachusetts, Legal Aid Society, Harvard Prison Legal
Assistance Project, Prisoners' Legal Services of New York, and
Prisoners' Legal Services of Massachusetts, amici curiae in support
of appellee.
Jennifer Levi and Bennett H. Klein on brief for Gay & Lesbian
Advocates & Defenders, EqualityMaine, Human Rights Campaign,
MassEquality, Massachusetts Transgender Political Coalition,
National Center for Transgender Equality, National Gay & Lesbian
Task Force, and Transgender New Hampshire, amici curiae in support
of appellee.

January 17, 2014

-2-
 THOMPSON, Circuit Judge. Twenty years after prison

inmate Michelle Kosilek first requested treatment for her severe

gender identity disorder, the district court issued an order

requiring the defendant, Luis S. Spencer, Commissioner of the

Massachusetts Department of Correction (the "DOC"),1 to provide

Kosilek with sex reassignment surgery. The court found that the

DOC's failure to provide the surgery — which was said by a group of

qualified doctors to be medically necessary to treat Kosilek's

condition — violated Kosilek's Eighth Amendment rights. The DOC

appeals the district court's order. Having carefully considered

the relevant law and the extensive factual record, we affirm the

judgment of the district court.

I. BACKGROUND

A. Gender Identity Disorder and Sex Reassignment Surgery

The concepts of gender identity disorder and sex

reassignment surgery sit center stage in this case and feature

prominently in this opinion. Therefore, before we go any further,

we provide a little context. As this court has explained, gender

identity disorder is "a psychological condition involving a strong

identification with the other gender." Battista v. Clarke, 645

F.3d 449, 450 (1st Cir. 2011). It is a disorder recognized by the

American Psychiatric Association, which describes gender identity

1
For ease of reference we will speak of the defendant as the
DOC, since the DOC's commissioner has changed multiple times during
the life span of Kosilek's lawsuits.

-3-
disorder as having two components. American Psychiatric Ass'n,

Diagnostic and Statistical Manual of Mental Disorders, Fourth

Edition - Text Revision 576 (2000) ("DSM-IV-TR").2 The first is

"evidence of a strong and persistent cross-gender identification,

which is the desire to be, or the insistence that one is, of the

other sex." Id. The second is "evidence of persistent discomfort

about one's assigned sex or a sense of inappropriateness in the

gender role of that sex." Id.

This current diagnosis of gender identity disorder grew

out of an earlier diagnosis of transsexualism, which first appeared

as an official disorder in the third edition of the DSM, published

in 1980. Judith S. Stern & Claire V. Merkine, Brian L. v. Admin.

for Children's Services: Ambivalence Toward Gender Identity

Disorder as a Medical Condition, 30 Women's Rts. L. Rep. 566, 567-

68 (2009). In the 1950s and 1960s, research began to show that a

"combination of psychotherapy, hormone treatment, and surgical

reconfiguration of the genitalia" could be used to treat gender

identity disorder. Id. at 571. This idea gained traction in the

1960s and 1970s, id., and as we will discuss more fully below, this

2
DSM-IV-TR, the version applicable in Kosilek's cases, uses
the term gender identity disorder. A newer edition, DSM-5, was
released in May 2013. DSM-5 replaces the term gender identity
disorder with gender dysphoria to avoid any negative stigma. See
American Psychiatric Ass'n, Gender Dysphoria,
http://www.dsm5.org/Documents/Gender%20Dysphoria%20Fact%20Sheet.pdf
(last visited Jan. 16, 2014). Because the term gender identity
disorder was used throughout Kosilek's cases and was the then-
appropriate nomenclature, we will use that term.

-4-
 triadic approach is still utilized by many practitioners today.

Sex reassignment surgery in particular has been performed in North

America since at least the 1950s, and it has been estimated that as

of January 2006, 30,000 sex reassignment surgeries have been

performed in the United States. Id. at 571-72.

With this bigger picture in place, we move on to the

facts of this case. We again note that this case has an over

twenty-year history. This has included two trials and two lengthy,

fact-intensive decisions issued by the district court, the latter

of which is the subject matter of this appeal. Because of this,

and because the district court's opinion was so fact-intensive, it

is necessary for us to lay out a good deal of background.

B. Kosilek's Conviction

Michelle Kosilek, née Robert, who is sixty-four years

old, was born and still is anatomically male. Kosilek suffers from

gender identity disorder. This has resulted in Kosilek's long-held

belief that she3 is a woman cruelly trapped in a man's body.

Kosilek, who spent some of her childhood in an orphanage,

suffered regular abuse as a child, in part because of her expressed

desire to live as a girl. As she grew older, she alternated

between living as a man and a woman. Kosilek's teenage and early

3
We will refer to Kosilek as her preferred gender of female,
using feminine pronouns.

-5-
adult years were marred by arrests, incarcerations, beatings, heavy

drinking, drug use, and a stint as a prostitute.

Sometime in the 1980s, Kosilek married Cheryl McCaul, a

volunteer counselor at a drug rehabilitation facility, who Kosilek

met while being treated there. McCaul thought she could cure

Kosilek's gender identity disorder, but Kosilek's desire to be

female did not go away. In 1990, Kosilek murdered McCaul. Kosilek

fled the area but was ultimately apprehended in New York.

Kosilek awaited trial at the Bristol County Jail. While

there, Kosilek, who had taken female hormones many years earlier,

started taking female hormones (in the form of birth control pills)

that she illicitly obtained from a guard. She also, at her own

expense, consulted with a gender identity disorder specialist,

though she was not allowed to undergo any treatment. While

awaiting trial, Kosilek twice tried to kill herself; one attempt

was made while she was taking the antidepressant Prozac. Kosilek

also attempted self-castration.

Kosilek was eventually tried and, in 1992, was convicted

of McCaul's murder and sentenced to life in prison without the

possibility of parole. She was turned over to the DOC and since

1994 has been residing in the general population at MCI-Norfolk, a

medium security male prison. There Kosilek started living, to the

extent possible, as a woman, legally changing her name from Robert

to Michelle.

-6-
 C. Kosilek's Lawsuits

In 1992, Kosilek filed a pro se complaint against the DOC

in the United States District Court for the District of

Massachusetts.4 See Kosilek v. Maloney, 221 F. Supp. 2d 156 (D.

Mass. 2002) ("Kosilek I"). Kosilek alleged the DOC was denying her

adequate medical care in violation of the Eighth Amendment.

Kosilek sought damages and an injunction ordering that she be

provided with sex reassignment surgery. The case was assigned to

Judge Mark L. Wolf, and proceeded for some years with the parties

engaging in discovery and motion practice.

Meanwhile, in December 2000, having not yet received the

relief she was seeking, Kosilek filed this case — a second pro se

lawsuit against the DOC and some of its medical providers. See

Kosilek v. Spencer, 889 F. Supp. 2d 190 (D. Mass. 2012) ("Kosilek

II"). Again the gravamen of Kosilek's complaint was that the DOC

was denying her adequate medical care in violation of the Eighth

Amendment by not providing her with sex reassignment surgery. This

case also went to Judge Wolf.

In February 2002, Kosilek's first lawsuit, Kosilek I,

finally proceeded to a non-jury trial. Due to some pretrial

skirmishing, only Kosilek's claim for injunctive relief remained

(her damages claim was gone). The trial lasted a couple of weeks.

4
Kosilek initially sued the Bristol County Sheriff, but later
amended the complaint to include the DOC after she was transferred
to its custody.

-7-
 D. The Kosilek I Decision

On August 28, 2002, Judge Wolf issued his decision. See

Kosilek I, 221 F. Supp. 2d at 156. The court explained that to

make out an Eighth Amendment violation, both an objective and a

subjective component must be satisfied. In short, the objective

component requires that the inmate has a serious medical need that

has not been adequately treated. The subjective piece, on the

other hand, focuses on the state of mind of the prison officials

and requires that they were aware the inmate was at risk for

serious harm.5

The court found that Kosilek suffered from a severe form

of gender identity disorder that caused her great mental anguish.

It went on to hold that this disorder was a serious medical need

within the meaning of the Eighth Amendment. To address the issue

of what types of treatment might be warranted, the court looked to

the Harry Benjamin Standards of Care (the "Standards of Care"),6

which it found to be the accepted protocols used by professionals

5
We will go into much greater depth regarding the Eighth
Amendment standard in our own analysis, but for now it suffices to
introduce the concept of the two-pronged test.
6
The operative version of the Standards of Care for both
Kosilek I and Kosilek II is the Sixth Version, issued in February
2001. See Harry Benjamin Int'l Gender Dysphoria Ass'n, Standards
of Care for Gender Identity Disorders, Sixth Version (2001). The
Seventh Version came out in 2011 under a new name. World
Professional Association for Transgender Health, Standards of Care
for the Health of Transsexual, Transgender, and Gender-
Nonconforming People, Version 7 (2011) (the "Standards of Care,
Version 7").

-8-
in the United States to treat gender identity disorder. The

Standards of Care indicated that, depending on the severity of an

individual's gender identity disorder, psychotherapy alone or

psychotherapy along with the administration of female hormones

could constitute sufficient treatment. In other instances,

however, sex reassignment surgery was medically necessary. The

court found that, despite DOC doctors recommending that Kosilek

receive female hormones and possibly surgery, the DOC, which at the

time was headed up by Commissioner Michael Maloney, had not

provided Kosilek with any of the treatment prescribed by the

Standards of Care.

The court chronicled the steps taken by Maloney to avoid

providing Kosilek with treatment. First, when an expert hired by

the DOC, Marshall Forstein, M.D., recommended psychotherapy and

hormones for Kosilek, and also, that she be allowed to consult with

a surgeon who specialized in sex reassignment surgery, the DOC

terminated its relationship with Dr. Forstein. Maloney then made

it clear to DOC doctors that he did not want to provide Kosilek or

any transgender prisoner with hormones or surgery, and the DOC

proceeded to find a Canadian doctor, Robert Dickey, M.D., who

believed inmates should never be considered for sex reassignment

surgery. Dr. Dickey advocated a "freeze-frame" policy whereby

transgender individuals would be frozen in the frame in which they

entered prison. For instance, only persons receiving hormones

-9-
before they were in prison would get hormones in prison. Despite

Maloney having little familiarity with the rationale behind Dr.

Dickey's philosophy on the treatment of gender identity disorder,

and not having read the Standards of Care, Maloney adopted Dr.

Dickey's freeze-frame policy for the DOC.

Dr. Dickey testified at trial, but the court did not find

him persuasive because he did not subscribe to the Standards of

Care, which the court found that prudent professionals follow, and

his approach did not allow for individual assessment. The court

found Kosilek's experts credible and relied on their testimony to

find that the objective component of the Eighth Amendment had been

satisfied, namely that Kosilek had a serious medical need that had

not been adequately treated.

But the court found Kosilek had fallen short of

establishing an Eighth Amendment violation because the subjective

component of deliberate indifference had not been satisfied.

Maloney, the court concluded, knew many facts from which he could

have inferred Kosilek would suffer serious harm if her gender

identity disorder was not treated, but he did not actually draw

that required inference. Instead Judge Wolf found Maloney's

refusal to allow Kosilek treatment was "rooted in sincere security

concerns, and in a fear of public and political criticism as well."

The end result: because there was no Eighth Amendment violation,

the court did not order the DOC to do anything.

-10-
 Nonetheless, Judge Wolf expected it would do something.

He wrote: "This court's decision puts Maloney on notice that

Kosilek has a serious medical need which is not being properly

treated. Therefore, he has a duty to respond reasonably to it.

The court expects that he will."

E. The After-Effects of Kosilek I

Following the court's dictate, lots of activity ensued at

the DOC. First, the DOC lifted its freeze-frame policy around

December 2002. In its place went a policy allowing inmates

suffering from gender identity disorder to receive a level of

treatment commensurate with that which they were receiving upon

entering prison (including receipt of hormones), but also providing

for increased or decreased treatment if it was determined to be

medically indicated by the University of Massachusetts Correctional

Health Program ("UMass"), the entity under contract with the DOC to

provide medical (including mental health) services to all inmates.

Prior to the implementation of any progressive or regressive

treatment changes, the DOC's Director of Health Services and the

Commissioner were required to consider whether the changes would

result in any safety or security concerns.

A couple of months later, in February 2003, the DOC

brought in a gender identity specialist, David Seil, M.D., to

-11-
evaluate Kosilek.7 Dr. Seil interviewed Kosilek and reviewed her

medical records, and then conveyed his findings and recommendation

in writing. Per DOC policy, which required any recommendations

involving gender identity disorder to go to both UMass's Medical

Director and its Mental Health Program Director, the report went to

Kenneth Appelbaum, M.D., who held the latter position.

Dr. Seil wrote the following. Like Drs. Forstein and

Dickey before him, he diagnosed Kosilek with gender identity

disorder. He found Kosilek's "gender dysphoria8 intense," and

though Kosilek had done what she could to obtain relief by living

as a woman, her basic disorder had been left untreated during her

incarceration. Dr. Seil, who noted that Kosilek had been living as

a woman in a male prison without security issues thus far,

indicated that the Standards of Care "need[ed] to be observed." He

further found that Kosilek was not currently suicidal because she

then felt some power over her pursuit of becoming a woman.

7
UMass's policy was to utilize consultants in matters that
went beyond the expertise of its direct staff.
8
Dr. Seil, and various other medical providers involved in
this case, sometimes use the term gender dysphoria. As indicated
previously, the American Psychiatric Association now uses this term
in place of gender identity disorder. The current version of the
Standards of Care says gender dysphoria is "broadly defined as
discomfort or distress that is caused by a discrepancy between a
person's gender identity and that person's sex assigned at birth."
Standards of Care, Version 7, at 2. It is not certain whether the
medical providers in this case were all using the term gender
dysphoria to mean precisely the same thing but, at a minimum, it
appears they were all using it similarly.

-12-
 Dr. Seil made several recommendations. First, Kosilek

should be provided with estrogen therapy, electrolysis

(specifically, permanent facial hair removal), and access to

gender-appropriate personal items, such as female clothing and

make-up. Additionally, Kosilek should be allowed to see the

therapist she had been treating with, master's-level psychologist

Mark Burrowes, more often than once a month. Further, feminizing

procedures, such as rhinoplasty (plastic surgery performed on the

nose) or breast augmentation, should be considered in the future.

Finally, Dr. Seil noted that sex reassignment surgery was the last

step in treating gender identity disorder. He explained that, as

provided in the Standards of Care, an evaluation of the necessity

of the surgery must wait until after a patient has lived as a woman

for at least a year.9 While Kosilek had been living as a woman for

many years, she had not had the benefit of hormone therapy and

electrolysis. Therefore Dr. Seil recommended that after Kosilek

had a year of hormone treatment under her belt, an experienced

gender identity specialist should evaluate her to determine whether

surgery was needed.

9
Dr. Seil was referring to the fact that the Standards of
Care require that prior to receiving sex reassignment surgery, a
person must live full-time for one year in the preferred gender
role. This requirement is commonly called the real-life
experience, and it is something we will discuss in more detail
later.

-13-
 The DOC began implementing Dr. Seil's recommendations.

In April 2003, Kosilek started treating with an endocrinologist to

develop a hormone therapy treatment plan. Concomitantly, per DOC

policy, the security implications of Kosilek receiving hormones

were assessed. On July 29, 2003, Luis Spencer, now the DOC

Commissioner but then the Superintendent of MCI-Norfolk, reported

to then DOC Commissioner Maloney that he did not believe there were

any current security concerns with Kosilek being provided estrogen

therapy, but once Kosilek began to exhibit physical changes

security concerns might have to be reevaluated.

And so on August 26, 2003, Kosilek began female hormone

treatment. Then, starting in October 2003, Kosilek was provided

with certain gender-appropriate items, such as female undergarments

and make-up. As neither Dr. Seil nor the endocrinologist Kosilek

was treating with had made specific recommendations as to whether

feminizing procedures were needed, none were provided. The DOC did

find a facility willing to provide Kosilek with electrolysis and,

after a security review of the facility was conducted, electrolysis

treatments were scheduled for November the following year.

In December 2003, Kathleen Dennehy, who had been the

DOC's Deputy Commissioner under Maloney, was elevated to

Commissioner. Right away she informed the staff she wanted to

-14-
"regroup on this GID stuff."10 And she wanted Kosilek reevaluated

before approving "laser hair removal or anything else."11

As of September 2004, Kosilek had been on hormones for a

year and, pursuant to the Standards of Care and Dr. Seil's

recommendation, was eligible to be evaluated for sex reassignment

surgery. The issue of conducting evaluations for prisoners with

gender identity disorder was taken up at executive staff meetings,

attended by UMass and DOC personnel, around this time. Those

typically in attendance from the DOC included Susan Martin, the

DOC's Director of Health Services, and Gregory Hughes, the DOC's

Director of Mental Health and Substance Abuse Services, as well as

a couple other DOC officials. Representing UMass was its Mental

Health Program Director Dr. Appelbaum, along with some additional

UMass personnel.

At one meeting, Dr. Appelbaum spoke about retaining the

Fenway Community Health Center (the "Fenway Center"), a Boston

healthcare facility focused on serving the lesbian, gay, bisexual,

and transgender community. Hughes, however, had some reservations;

he felt the Fenway Center might be too sympathetic to the prisoners

and too quick to recommend treatment. Dr. Appelbaum countered,

stating the Fenway Center's approach represented the norm rather

10
"GID" is an acronym for gender identity disorder.
11
It is unclear from the record what happened, if anything,
based on Dennehy's desires to regroup and reevaluate.

-15-
than the exception and there really were not many other providers

in the area with whom to consult. Hughes said he had spoken with

a Johns Hopkins gender identity specialist, Cynthia Osborne, a

licensed social worker, who was working with the Virginia and

Wisconsin departments of corrections, which had also been sued by

transgender prisoners. It was noted on the meeting minutes that

Osborne "may do more objective evaluations," and was "[m]ore

sympathetic to DOC position."

Despite Hughes's qualms about the Fenway Center and the

possible option of using Osborne, the DOC went ahead with retaining

the Fenway Center. Kosilek was evaluated by Kevin Kapila, M.D.,

and Randi Kaufman, Psy.D., who conducted a ninety-minute interview

with her and reviewed her medical records. The doctors issued

their report with several findings on February 24, 2005 (the

"Fenway Report"). As had been documented in the past, Kosilek

"clearly fit[]" the diagnosis for gender identity disorder.

Kosilek had been on hormone therapy and living full-time as a woman

for seventeen months by that time. Kosilek's ability to live as a

woman, her good behavior, and her absence of conflict with others,

suggested, according to the doctors, an "intense motivation, as

well as a real adaptability to her environment." They noted her

favorable response to the use of hormones, electrolysis, and use of

feminine products. But, they noted, Kosilek was still "quite

distressed," and given Kosilek's "previous suicide attempts, her

-16-
ongoing distress, and lack of other goals in her life," the doctors

found it was quite likely Kosilek would attempt suicide again if

she was not provided with sex reassignment surgery. Kosilek had a

serious medical need, they felt, and there was a substantial risk

of harm if her disorder was left untreated.

The report went on to note that Kosilek had moved

successfully through the steps outlined in the Standards of Care,

there had not been any adverse reactions to Kosilek's feminized

appearance, and Kosilek had benefitted psychologically from her

changes. Drs. Kapila and Kaufman concluded Kosilek was likely

ready for sex reassignment surgery. The surgery, the doctors

opined, would provide Kosilek with full relief from the symptoms of

gender identity disorder and likely increase her chance of

survival. The ultimate recommendation of the specialists retained

by the DOC as advised by its own doctors: Kosilek should be given

the surgery.

Unhappy with the Fenway Report, the DOC turned to the

Johns Hopkins gender identity specialist, Cynthia Osborne, whose

name had been batted around at the earlier DOC meeting. The DOC

asked Osborne to conduct a peer review12 of the Fenway Report and

12
Peer review is the term used by the DOC — we presume because
social worker Osborne was tasked primarily with reviewing the
recommendation of her so-called peers, the Fenway Center doctors,
as opposed to interviewing Kosilek and conducting an independent
assessment.

-17-
she agreed. The report, along with other evaluations conducted of

Kosilek, were sent to Osborne on April 12, 2005.

In the meantime, on April 15, 2005, Dr. Appelbaum, who at

the time was a defendant in this lawsuit along with UMass and some

other doctors, filed (at the district court's request) a status

report. It advised the court of Drs. Kapila and Kaufman's sex

reassignment surgery recommendation for Kosilek. Dr. Appelbaum

also said he had advised the DOC of the doctors' recommendation and

had informed them he was unaware of any medical reason why Kosilek

should not receive the surgery.

The court responded with an April 25, 2005 order

directing the DOC to provide a report addressing potential security

concerns should Kosilek undergo the surgery. It also directed the

DOC to indicate whether it was going to provide Kosilek with the

surgery recommended by the Fenway Center doctors.

A few days later, on April 28, 2005, the DOC responded to

the UMass status report. The letter was penned by Susan Martin

(recall she was the DOC Director of Health Services) and sent to

Dr. Appelbaum and UMass's Medical Director, Arthur Brewer, M.D.

(also a defendant). Dennehy and other DOC officials were copied.

Martin maligned Dr. Appelbaum's status report, asserting that the

DOC did not consider it an adequate review of the Fenway Report or

a clear explanation of UMass's recommendation. Martin requested

that the UMass doctors make clear whether they thought Kosilek

-18-
should be operated on. She also wanted answers to various

questions relating to, among other things, the surgery's logistics,

such as who would perform it, the recovery process, and the success

rate.

Drs. Appelbaum and Brewer, in a May 10, 2005 response

letter to the DOC, clarified their stance on the Fenway Report. As

they explained, UMass deferred to the Fenway Center and stood

behind the doctors' sex reassignment surgery recommendation for

Kosilek. They were aware of no mental health barriers to Kosilek

being operated on and the next step was finding a surgeon. It

appeared there were no physicians in Massachusetts who could

perform the surgery, so some out-of-state practitioners were

suggested. The doctors also offered to look into the logistics of

providing the surgery.

In the interim, Osborne completed her review of the

Fenway Report.13 She chronicled her findings in a report dated May

20, 2005, which she sent to the DOC. It began with the caveat that

Osborne had not conducted a clinical evaluation of Kosilek and her

report was based solely on her review of the Fenway Report and some

of the other evaluations of Kosilek. Though she did not doubt

13
We pause to say a little more about Osborne's
qualifications. According to her C.V., at least as of 2006, she
held a master's degree in education and social work. Osborne was
also an assistant professor of psychiatry at Johns Hopkins
University School of Medicine and on the consulting faculty of the
University's Center for Sexual Health and Medicine.

-19-
Kosilek met the criteria for gender identity disorder, Osborne went

on to lodge numerous criticisms against Drs. Kapila and Kaufman's

approach.

First, Osborne disparaged the Fenway Report for not

addressing the issue of whether Kosilek was suffering from any

personality disorders. She opined that "[c]larity regarding the

presence, absence, nature and severity" of any personality

disorder, especially given that clinicians had at one point

diagnosed Kosilek with antisocial personality disorder, was

critical because its presence could complicate a gender identity

disorder diagnosis. Osborne also argued that threats of suicide

and self-harm signal serious mental illness apart from gender

identity disorder, which demands treatment and, in fact, counsels

against providing sex reassignment surgery. She thought the Fenway

Report had given short shrift to this issue.

Osborne then turned her focus to the Standards of Care,

expressing her concerns that they did not translate well into a

prison environment. She wrote: "In my view, providing surgery, or

even hormones, to incarcerated individuals, is an undeniable

lowering of the Standards, and an explicit violation of the

criteria regarding sociopathy and suicidality." While Osborne

recognized the Standards of Care as helpful, she noted they had no

regulatory authority. Also, there was no universal consensus in

the psychiatric community about what constituted medical necessity

-20-
in the treatment of gender identity disorder. She criticized the

Fenway Report for failing to address other possible treatment

options for Kosilek or to provide an adequate explanation for its

surgery recommendation. Kosilek, in Osborne's eyes, had an

unrealistic expectation that she was owed certain treatments and

Kosilek would instead benefit from a thorough assessment for

psychiatric disorders and treatment designed to address any such

disorders along with her gender identity disorder.

On May 25, 2005, the back and forth between the DOC and

UMass continued. Martin sent Drs. Appelbaum and Brewer another

criticism-laced letter. While the doctors had answered some of the

DOC's questions, they had not provided a comprehensive review of

the appropriateness or necessity of surgery for Kosilek. Citing

the DOC's Eighth Amendment obligations, Martin lamented that

neither UMass nor the Fenway Report had offered guidance on whether

surgery was "a medical necessity for Kosilek." Finally, Martin

informed them of the DOC's engagement of Osborne to review the

Fenway Report and enclosed a copy of Osborne's report. Martin

asked UMass to review the report and give the DOC its ultimate

recommendation as to the appropriateness of surgery for Kosilek.

The same day Martin fired off her letter, a news piece,

which was highly negative toward Kosilek and her quest for surgery,

appeared on a local television station. Commissioner Dennehy had

been interviewed for the piece on May 16, and one of her comments

-21-
made it on air.14 A Massachusetts state senator, who had filed

legislation seeking to prohibit the state from paying for sex

reassignment surgery for inmates, also spoke to the reporter.

"When you go to prison you lose some rights. You also lose your

rights to get a sex change operation," he stated. The senator, who

was an acquaintance of Dennehy, had called Dennehy on her cell

phone to advise her that he would be participating in the news

piece. The piece concluded with the reporter indicating: "Later

this week, the state will tell the federal court that sex surgery

for Michelle Kosilek would result in a security nightmare."15

Despite the security concerns the DOC expressed to the

media, Dennehy and her staff had yet (as of the time she was

interviewed) to officially convene to conduct their internal

security review. Not until May 19, three days after her interview,

did they actually meet. In attendance were Dennehy, DOC counsel,

Superintendent Spencer of MCI-Norfolk, and Superintendent Lynne

Bissonnette of MCI-Framingham, the women's prison where Kosilek

might be sent were she to undergo surgery. Dennehy did not have

the benefit of written reports from Spencer or Bissonnette, despite

14
She was quoted: "The courts are telling us that medical
professionals make medical recommendations and correctional
administrators assess the safety and security concerns."
15
It seems likely that the reporter learned this information
from Dennehy herself, though it is unclear whether it could have
come from someone on her staff. Either way the DOC does not
dispute that this tidbit came from within the DOC.

-22-
DOC policy calling for the superintendents to make such

recommendations, because (as Dennehy later testified) the security

concerns seemed self-evident.

On June 10, 2005, the DOC brought the court back into the

loop, finally filing the status report which the court had ordered

back on April 25. The DOC informed Judge Wolf it had decided not

to provide Kosilek with sex reassignment surgery. In support of

its decision, the DOC attached Osborne's report. It also attached

the court-ordered security report, which was principally authored

by DOC attorneys with input from Dennehy.

The security report, which said it was based on Dennehy's

thirty years of correctional knowledge and experience, as well as

the experience of colleagues she had spoken with, stated the

following. Allowing the surgery to go forward "would create

substantial safety and security problems for DOC." She claimed

that out-of-state surgery, which appeared to be the only option

based on a dearth of doctors in Massachusetts, would cause "complex

security and logistical issues" and might give Kosilek a chance to

escape custody. Kosilek's post-surgery confinement was also a

cause for concern. Dennehy argued Kosilek would be at risk for

sexual assault if she were to remain in a male prison, and housing

her in a female prison would come with its own host of problems.

Specifically, Dennehy anticipated "serious climate issues" since

the majority of the women at MCI-Framingham had histories of trauma

-23-
and it was well known that Kosilek had killed her wife. The women

might pose a threat to Kosilek and vice versa. Kosilek would need

to be isolated and restrictively confined in either prison and

this, Dennehy concluded, might be deleterious to Kosilek's mental

health. Citing these concerns, along with what she perceived as an

unclear stance from UMass on whether surgery was necessary,16 and

given Osborne's critiques of the Fenway Report, Dennehy indicated

that the DOC was denying Kosilek's request for surgery.

After the status report was filed with the court, the DOC

and UMass continued to clash over what the DOC perceived to be

UMass's equivocations. Drs. Appelbaum and Brewer directed a June

14, 2005 letter at the DOC. Calling the DOC's statements in its

previous letters and status report "disingenuous," the doctors

emphasized that it was not within their purview to decide whether

surgery is medically necessary for Kosilek as that term is

contemplated by the Eighth Amendment. UMass, they explained, had

consistently relied on the expertise of outside consultants, all of

whom had said the same thing. The doctors again underlined that

solely from a clinical perspective it appeared that Kosilek should

be offered surgery.

16
The report quoted the supposedly ambiguous language from
UMass's letter: "the treatment recommended in the Fenway
report . . . appears to be reasonable and appropriate, since the
patient has met criteria for the diagnosis of gender identity
disorder and has reached a point in clinical treatment where sexual
reassignment surgery, if desired, would be the next step."

-24-
 On October 7, 2005, Drs. Kapila and Kaufman prepared and

issued another report, this one a response to Osborne's critiques

of their Fenway Report, which they sent to UMass and which UMass

passed on to the DOC. Citing medical literature and studies, the

doctors took aim at Osborne's criticisms, explaining why her

challenges to their recommendation for surgery were clinically

unfounded or irrelevant. The report also argued that Kosilek was

stable, not currently suicidal, and a good candidate for surgery.

The doctors strenuously insisted that their recommendation of

surgery was an informed clinical judgment rooted in Kosilek's

gender identity disorder diagnosis, her marked mental health

improvement since being provided hormones and being allowed to live

as a woman, and the well-documented effectiveness of surgery. The

doctors reiterated their recommendation that Kosilek be provided

with the surgery. Dr. Appelbaum submitted this report to the DOC

on October 17, 2005, restating UMass's endorsement of the original

Fenway Report's recommendations and requesting that the DOC issue

a decision approving those recommendations.

As to the status of Kosilek's lawsuit, as of July 2005,

Kosilek had let all the medical provider defendants (Drs. Appelbaum

and Brewer, and UMass, among others) out of the case. That same

month she filed an amended complaint, this time with the benefit of

an attorney, with Dennehy as the sole named defendant. Kosilek no

longer sought damages, she only requested injunctive relief

-25-
requiring the DOC to provide her with adequate medical care,

including sex reassignment surgery. The case went to trial a

little less than a year later.

F. The Kosilek II Trial

1. Round One of Testimony

The non-jury trial, which like Kosilek I was presided

over by Judge Wolf (by then Chief Judge), began on May 30, 2006.

Testimony initially went until the end of June, with multiple

witnesses testifying.17

First the court heard from George Brown, M.D., a

practicing psychiatrist who helped author the Standards of Care,

and who testified as an expert on Kosilek's behalf (he also

testified in Kosilek I). As he had done back in 2001, Dr. Brown

evaluated Kosilek and reviewed her medical records. He prepared a

written report, and in it diagnosed Kosilek with chronic and severe

gender identity disorder. He did not think she met the criteria

for antisocial personality disorder. Dr. Brown also did not think

Kosilek was trying to game the system, writing that he was "hard

pressed to develop a rational explanation for why someone would

work so fervently to obtain this serious, painful surgery" other

than as a means to treat gender identity disorder. Dr. Brown also

opined that Kosilek had met or exceeded the readiness criteria for

17
We summarize the testimony in the order it was given, save
for Dr. Schmidt. He actually testified before Kosilek finished
putting on her witnesses (we assume for scheduling convenience).

-26-
sex reassignment surgery. Among other things, she had completed a

more than two-year monitored, real-life experience living as a

woman while incarcerated. Further, he wrote, Kosilek had "received

an unambiguous diagnosis of severe gender identity disorder from no

fewer than nine mental health professionals who have interviewed

her, many of whom are recognized international experts in the field

of gender identity disorder." Dr. Brown stressed, "[n]o further

treatment or real-life experience is necessary," and Kosilek should

receive the surgery, which Dr. Brown deemed "medically necessary."

Hammering the point home, Dr. Brown testified, consistent

with his written report, that the hormones and psychotherapy

Kosilek was receiving, though they had helped relieve her

dysphoria, were "[a]bsolutely not" sufficient to eliminate the

serious risk of harm Kosilek faced, up to and including suicide.

When asked whether sex reassignment surgery was medically

necessary, Dr. Brown stated: "Absolutely. If I can walk away from

these proceedings with one point being clear in people's minds,

it's that." Without surgery, he added, "the degree of likelihood

of [Kosilek] suffering serious medical consequences up to and

including suicide are exceedingly high." As for antidepressants,

they are for the treatment of patients with major depressive

disorder, which Dr. Brown stressed Kosilek did not have. With

regards to gender identity disorder, such antidepressant

medications had been shown to be "very ineffective" as they might

-27-
only slightly relieve some depressive symptoms but would not treat

the underlying gender identity disorder. Only surgery, Dr. Brown

testified, had the "significant potential" to cure Kosilek's

medical condition.

Next came Dr. Kenneth Appelbaum's testimony. The Fenway

Center doctors, he said, had significant experience in the area of

gender identity disorder and were well-trained, credentialed, and

knowledgeable. He thought their assessments were typically

reasonable and consistent with the approaches followed by most

other medical providers. In general, Dr. Appelbaum did not see why

the DOC would need to consult with Osborne (an out-of-state,

master's-level social worker) given that it had already received

the recommendations of the Fenway Center doctors (a local physician

and doctoral psychologist), who were, in Dr. Appelbaum's mind,

highly experienced in dealing with gender identity disorder. As

for Osborne, Dr. Appelbaum recalled that the DOC's Hughes had

commented that she would be more sensitive to the DOC's concerns

because she did not believe sex reassignment surgery was

appropriate in the corrections setting. Finally, Dr. Appelbaum

testified that from his conversations with Kosilek's treatment team

at the prison, it was his understanding Kosilek had shown good

adjustment being on hormones and receiving therapy.

Dr. Randi Kaufman from the Fenway Center then took the

stand. She reiterated the conclusion she made in her report, in

-28-
particular that Kosilek had successfully completed the real-life

experience contemplated by the Standards of Care. She also

testified that, to a reasonable degree of medical certainty, there

was a "very high likelihood" Kosilek would attempt suicide if

denied treatment for her gender identity disorder. Though Kosilek

benefitted from being on hormones, Dr. Kaufman felt Kosilek had a

level of gender identity disorder that could not be treated with

anything less than surgery. "She's done all the things that people

do to change their gender presentation," she said, and "[t]here

really isn't anything left except for surgery." Kosilek had a

medical need for the surgery.

Mark Burrowes, a licensed mental health counselor with a

master's degree in counseling psychology, who had been treating

Kosilek for four or five years, also testified. Kosilek, he

agreed, was ready for sex reassignment surgery. Not having the

surgery would be detrimental and could result in Kosilek making an

attempt on her life. Suicide monitoring would not be an adequate

alternative to surgery since Kosilek's male genitalia still caused

her distress.

Finally, Michelle Kosilek was called by her attorneys.

She spoke about her life at MCI-Norfolk. Generally she got along

quite well with most of the people there, though a small percentage

of the correction officers gave her a hard time. Kosilek had,

since she got to prison, worked on a daily basis. She testified

-29-
that the hormone treatments had made her a little less depressed

but she continued to feel distress over her body. Kosilek, who

said she was not currently suicidal, felt the hormones were not

enough and she needed surgery. She did not want to continue living

with her male genitalia and, if denied surgery, antidepressants and

psychotherapy would not help matters.

The DOC then put on its case. First, Chester Schmidt,

M.D., a psychiatrist at the Johns Hopkins School of Medicine, and

associate director of the Johns Hopkins Center for Sexual Health

and Medicine, testified as an expert for the DOC. He became

involved in this case through his connection with Cynthia Osborne.

As to his general approach for treating patients with gender

identity disorder, Dr. Schmidt stated he does utilize the Standards

of Care but thinks of them more as protocols or guidelines, as

opposed to actual standards of care. He does not agree with the

idea, set forth in the Standards of Care, that sex reassignment

surgery is medically necessary in patients with severe gender

identity disorder. In fact, Dr. Schmidt did not recall ever seeing

a case where he thought surgery was medically necessary. It was

his and his Johns Hopkins colleagues' practice to neither advocate

nor speak against a patient's desire for surgery, but to leave it

in the hands of the patient. He would not send a letter of

recommendation to a surgeon on behalf of a patient, but he would

-30-
release his file to the surgeon and simply indicate he saw no

contraindications to surgery.

After giving his general overview, Dr. Schmidt turned his

focus to Kosilek, whom he had personally evaluated in November

2005. Kosilek did meet the criteria for gender identity disorder,

he concluded. However, Dr. Schmidt did not believe surgery was

medically necessary as Kosilek, he theorized, had made an

"excellent adaptation" without surgery thus far. He also felt the

real-life experience contemplated in the Standards of Care was

virtually impossible to replicate in prison. But besides this

barrier, Dr. Schmidt admitted he did not see any contraindications

to surgery. Should Kosilek become depressed if she did not receive

the surgery, Dr. Schmidt thought she could be treated with

antidepressants and psychotherapy, and managed in a medical

facility should her suicidal desires become severe. This latter

scenario was a possibility, as Dr. Schmidt recognized Kosilek's

risk of suicide based on what she had said and done in the past.

But, Dr. Schmidt opined, psychotherapy and medications could

effectively reduce Kosilek's dysphoria to a level where she was no

longer at risk for serious harm.

Next Cynthia Osborne, the Johns Hopkins social worker and

DOC consultant, gave testimony echoing Dr. Schmidt. By the time of

trial, she had met with and interviewed Kosilek. Osborne agreed

with Kosilek's severe gender identity disorder diagnosis. Yet,

-31-
like Dr. Schmidt, in general, she did not believe the real-life

experience called for by the Standards of Care could happen in a

prison environment. Also, Kosilek was not, in her opinion, a good

candidate for sex reassignment surgery, nor was it medically

necessary, because Kosilek had responded very well to hormone

treatment. Rather, Osborne thought support groups or group therapy

could be used to treat Kosilek effectively. Kosilek's threats of

suicide if denied surgery did not change Osborne's mind on what

treatment was warranted. Indeed, she minimized concerns about

suicide by noting that any good mental health system would know how

to deal with a patient's suicidality. Osborne, again like Dr.

Schmidt, did not fully agree with the Standards of Care's statement

that sex reassignment surgery is medically necessary in cases of

severe gender identity disorder.

Luis Spencer, then still Superintendent of MCI-Norfolk,

also testified for the DOC. He explained the set-up at MCI-

Norfolk, describing it as having one of the more secure perimeters

in Massachusetts. It is surrounded by walls on all sides, an

electrified fence, and guard towers which are manned twenty-four

hours a day, seven days a week. Approximately one-third of the

inmates at MCI-Norfolk, Spencer explained, are serving a life

sentence and one-third have committed sexual felonies.

Spencer also testified about Kosilek. She had adjusted

fairly well to life at MCI-Norfolk and had not reported any threats

-32-
or harassment from other inmates. As of the time of trial, no

security concerns involving Kosilek had arisen; however, Spencer

had some apprehension going forward should Kosilek receive the

surgery. He would have "grave concerns" putting Kosilek back in

the general population with the full anatomy of a female. Spencer

worried that she could be raped or assaulted and he saw no

alternative but to house her in the high-security Special

Management Unit. This unit was a standalone secure building where

Kosilek would remain in her cell twenty-three hours a day and could

only leave when shackled and escorted by two guards. The

questioning also briefly touched on Kosilek's threats of suicide

were she not to receive the surgery. Spencer said his policy is

not to negotiate with inmates who threaten suicide, as to do so

would undermine his and the staff's authority. Rather, he would

implement the DOC's mental health policy and take the appropriate

steps to guard a suicidal inmate's safety.

Gregory Hughes, who was the DOC's Director of Mental

Health and Substance Abuse Services until 2005, and who holds a

master's degree in social work, was questioned next. Hughes

testified about the aftermath of Kosilek I and the DOC's efforts to

comply with the court's decision. His role included overseeing and

facilitating the services supplied to Kosilek, and in particular

making happen the independent evaluation Judge Wolf ordered in

Kosilek I. He testified about his dissatisfaction with the Fenway

-33-
Report, questioning its thoroughness and its heavy reliance on

Kosilek's self-reporting. He was concerned that the doctors had

not reviewed any of Kosilek's medical records or mental health

history or interviewed other people to verify Kosilek's self-

reports. And while Hughes knew the doctors suggested surgery, he

said he was uncertain about whether there were other

recommendations and what next step the DOC should take. Because he

was concerned with the Fenway Report's quality, Hughes had decided

to contact Osborne. Hughes also testified that he reported his

concerns about the report to Drs. Appelbaum and Brewer and that the

DOC sought their thoughts during staff meetings. Hughes recalled

the doctors saying there were no contraindications to surgery but

did not believe they used the term "medically necessary."

Robert Dumond, who was the Director of the DOC's Research

and Planning Division, and who had previously provided mental

health screenings at MCI-Framingham, testified on the DOC's behalf.

Dumond, who had a master's degree in psychology and experience in

the areas of victimization and sexual assault in the prison system,

was asked by the DOC to consider the risk factors for post-

operative individuals in prison. Placing Kosilek at MCI-Framingham

post-surgery, he responded, could "destabilize[] the safety and

security of the institution" and create a risk for Kosilek to

become a victim or victimizer.

-34-
 Arthur Beeler was also called by the DOC to testify about

security issues. Beeler, a thirty-year employee with the Federal

Bureau of Prisons, was warden at the Federal Correction Complex,

Federal Medical Center in Buckner, North Carolina. Beeler toured

MCI-Norfolk and MCI-Framingham, spoke with Superintendent Spencer,

and reviewed the DOC's mental health policies in preparation for

his testimony. In his opinion, Kosilek was currently safe and

secure at MCI-Norfolk, based in part on the set-up of the

facilities and her cell and the fact that Kosilek had never been

assaulted there. Beeler indicated that he "would be very

concerned" about placing Kosilek in the women's general population

at MCI-Framingham and was satisfied that the mental health

procedure at MCI-Norfolk was sufficient to address inmate suicide

ideation and behavior. Beeler cautioned against giving in to an

inmate's threats of suicide, likening it to opening Pandora's box.

Beeler was not permitted to testify specifically about whether

security concerns should preclude Kosilek from getting surgery

because, the court found, he was not sufficiently informed about

"the facts concerning Kosilek" because Beeler had not looked at her

disciplinary or medical records.

Susan Martin, the DOC's Director of Health Services

during the relevant time period, was next up. After speaking some

about the contractual relationship between the DOC and UMass,

Martin turned to the Fenway Center's evaluation of Kosilek. Like

-35-
Hughes, she was not satisfied with the evaluation. Martin did not

think the Fenway Report was very thorough and she raised this

concern to UMass at the time. Though Martin knew the Fenway Center

doctors were recommending surgery, she did not think they were

clear enough about what exactly needed to happen and when. Because

of these perceived omissions, Martin decided to have Osborne review

the report.

When asked about UMass's role, Martin indicated she

thought it was UMass's job to determine whether the Fenway Center's

recommendations "were medically necessary and clinically sound,"

and UMass, she said, would not do this. She considered it UMass's

responsibility to find a surgeon, and simply providing the DOC with

a list of possible surgeons was not sufficient.

Next came DOC Commissioner Kathleen Dennehy's testimony.

If faced with a court order compelling the DOC to provide Kosilek

with sex reassignment surgery, Dennehy said she would probably

retire before implementing something she considered unsafe. And

based "strictly [on] safety and security concerns" she said she

would still veto the surgery even if UMass told her it was

medically necessary and even if Kosilek would likely attempt

suicide if denied the surgery. (Dennehy claimed she was still

awaiting clear direction from UMass.)

When asked, Dennehy acknowledged her awareness of the

negative public attention that Kosilek's bid had been receiving.

-36-
She admitted she knew that a Massachusetts senator, who she was

friendly with and who had spoken in the news piece, vocally opposed

sex reassignment surgery for inmates and was pushing for

legislation to put a stop to its provision. She also said she knew

of the Massachusetts Lieutenant Governor's opposition to surgery

for transgender inmates.

Finally, Lynne Bissonnette, Superintendent of MCI-

Framingham, was called upon to discuss the feasibility of Kosilek's

post-operative placement there. She explained the set-up at MCI-

Framingham. The women lived in multi-person rooms or dormitory-

style housing units, with the exception of the women in the

segregation or medical units who had single cells. For the most

part the buildings at MCI-Framingham were surrounded by a single,

non-electrified twelve-foot-high fence. There had been,

Bissonnette indicated, no prisoner escapes during her three-year

tenure at Framingham.

As for the population at the facility, according to

Bissonnette, a large majority of the women there had been victims

of domestic violence or sexual assault. Also, a majority had

mental health issues with about a half receiving medication for

those issues. Bissonnette stated that if a woman could not

effectively be maintained by the mental health professionals at the

prison, she was sent for inpatient treatment at a Massachusetts

Department of Mental Health facility, where security was not the

-37-
equivalent of what was provided at the prison. And whereas

Massachusetts provides a special secure hospital for male

prisoners, Bridgewater State Hospital, she explained that no

comparative facility exists for women.

Bissonnette's testimony then turned to Kosilek

specifically, whose criminal record she had reviewed. Bissonnette

did not think Kosilek was an appropriate candidate to be housed at

MCI-Framingham post-operatively and noted several concerns. First,

Bissonnette considered Kosilek a flight risk based on the weak

perimeter at MCI-Framingham, the length of Kosilek's sentence, and

the fact that Kosilek had fled Massachusetts after killing her

wife.18 Second, Bissonnette worried about the lack of available

inpatient mental health care. Kosilek would have to be sent to a

less secure hospital for the general public should she require

care. Third, Bissonnette considered Kosilek both a potential

predator and victim within the inmate population — a predator

because Kosilek had strangled her wife and a victim because a large

proportion of MCI-Framingham's population had been the victims of

domestic violence and sexual abuse and might seek to harm her.

Further, and for essentially those same reasons, Bissonnette felt

Kosilek would have a negative effect on the prison population.

18
Bissonnette considered Kosilek's earlier flight, even though
the DOC classification manual says pre-custodial flight should not
be considered when classifying inmates.

-38-
 In light of all of these concerns, were Bissonnette

required to house Kosilek, she would put her in the Close Custody

Unit, the single cell segregation unit, where inmates cannot hold

any employment and are placed in restraints whenever they leave the

cell. Since Kosilek is serving a life sentence, Bissonnette

expressed concern that housing her in an environment this

restrictive for such an extended period of time would have a

negative impact on Kosilek's mental health.

On cross-examination Bissonnette made a few concessions.

She agreed that MCI-Framingham currently houses approximately forty

offenders who are serving life sentences for murder. Further, the

institution houses in the general population inmates who are

convicted of heinous crimes against children alongside prisoners

who are mothers. Bissonnette explained that if inmates are

perceived to experience trauma based on the presence of other

inmates, there are policies and procedures in place to follow (such

as simply ordering a cell transfer) and the mental health and

security staff would respond appropriately. Though Bissonnette

again acknowledged Kosilek could be housed in the Close Custody

Unit safely, she persisted that this was not the "best setting for

any inmate over a long period of time."19

19
During the first round of testimony some more negative press
came out. The Boston Globe ran an editorial on June 15, 2006,
which took the position that Kosilek should not receive the
surgery. The editorial referred to the "distastefulness of a wife
killer angling to serve out his sentence of life without parole in

-39-
 2. Responses to the First Round of Testimony

After the first bout of testimony, which concluded at the

end of June 2006, Judge Wolf then directed the UMass doctors to

review Dr. Schmidt's testimony and to inform the court whether the

latter's proposed approach was within prudent professional

standards. Drs. Appelbaum and Brewer responded in a report filed

with the court on September 18, 2006. They continued to endorse

the Fenway Center doctors' conclusion that Kosilek had a serious

medical need requiring surgery. In their opinion, Dr. Schmidt's

proposed alternative course of psychotherapy, medication, and

suicide watch fell "outside the bounds of acceptable professional

standards" and would not constitute "adequate medical care." Such

interventions would "likely do little" to reduce Kosilek's

dysphoria or the risk of harm to her. Only surgery, they

concluded, could do this, and there was "no good clinical reason to

withhold that treatment at this time."

The judge also ordered Dennehy to review the evidence

presented at trial and decide whether the DOC would reverse its

position on surgery for Kosilek. After doing so, Dennehy indicated

to the court that Drs. Appelbaum's and Brewer's testimony confirmed

what she suspected (but apparently did not previously find clear):

that the doctors believed surgery was medically necessary. Dennehy

stood firm though; she advised Judge Wolf that her decision not to

a women's prison."

-40-
allow sex reassignment surgery for Kosilek remained. Her safety

and security concerns with providing surgery, which Dennehy called

"alarming and substantial," had not changed.

3. Second Round of Testimony

Testimony resumed in early October 2006, with various

witnesses taking and re-taking the stand. First came Dr.

Appelbaum, who emphasized several points. Kosilek had a serious

medical need because there was a serious risk of harm if she was

not adequately treated; and he had informed Dennehy of this.

Antidepressants were unlikely to effectively treat Kosilek because

the source of her distress was her gender identity disorder and

medication would not target this underlying condition. Rather,

surgery, he explained, was the "recognized and appropriate

treatment" and the only treatment "likely to significantly relieve,

if not eliminate Miss Kosilek's distress."

Dr. Kevin Kapila from the Fenway Center also spoke. He

called Kosilek's gender identity disorder "one of the more severe

cases" he had ever seen and he testified about the report he and

Dr. Kaufman had written. In his opinion, Kosilek had a serious

medical need and there was a substantial risk of harm if she was

not treated with surgery. Dr. Kapila also testified that he

thought Dr. Schmidt's recommendation was unreasonable.

Specifically, Dr. Kapila (who had reviewed Dr. Schmidt's testimony)

opined to a reasonable degree of medical certainty that Dr.

-41-
Schmidt's plan to treat Kosilek with psychotherapy and medication

was inadequate. Pointing to his own evaluation of Kosilek after

she had been receiving psychotherapy and hormones for seventeen

months, Dr. Kapila noted her symptoms had still not resolved; "she

still was dealing with discomfort around having male genitalia."

Dr. Schmidt's approach would not deal with this core problem — only

surgery could — and, Dr. Kapila continued, one tries to treat the

problem, not merely the symptoms.

Dr. Kaufman then got recalled. While she agreed

Kosilek's dysphoria had improved as a result of receiving hormone

treatments and gender-appropriate items and clothing, Dr. Kaufman

persisted in opining that Kosilek still had a serious medical need

and if she did not receive surgery there was a substantial risk of

serious harm. Despite all the treatment Kosilek had received, she

still continued to experience severe dysphoria. Like Dr. Kapila,

Dr. Kaufman did not think the treatment advocated by Dr. Schmidt

was adequate. It is important to distinguish between depression

and dysphoria, she said, and Dr. Schmidt's treatment would not

sufficiently address the latter.

UMass's Medical Director, Dr. Arthur Brewer, was up next.

He had little to add to the testimony of the other medical

providers because, as medical director, he was not involved in

UMass's mental health program and had no role in Kosilek's

treatment. His position remained the same as in his and Dr.

-42-
Appelbaum's September 18, 2006 letter, which supported the Fenway

Center doctors and criticized Dr. Schmidt's approach.

Judge Wolf then again heard from Commissioner Dennehy.

She accepted that Kosilek had a serious medical need but said she

had previously been confused about whether surgery was medically

necessary. While she had now "deduced" UMass's opinion on the

appropriateness of surgery for Kosilek, she had not changed her

thoughts on the security and safety concerns related to providing

such surgery. Absent these concerns, Dennehy agreed she would have

no reason to interfere with any medical order for treatment.

The court questioned Dennehy about her awareness of any

publicity surrounding Kosilek's case. Dennehy was generally aware

of a couple of articles appearing in the Boston papers and admitted

skimming one. She also knew about an article on a national news

media website. In general she tried not to read newspaper articles

that involved her and she never thought about what the public and

political reaction would be if the DOC allowed Kosilek to have

surgery.

4. The Court-Appointed Expert

After hearing all the witnesses each side wished to

present, Judge Wolf decided to appoint an expert in order to help

him decide whether the care proposed by Dr. Schmidt was objectively

adequate. See Fed. R. Evid. 706 (providing that the court may on

its own initiative appoint an expert). After soliciting the

-43-
parties' thoughts on who to appoint, Judge Wolf selected Stephen

Levine, M.D., on October 31, 2006. Dr. Levine practiced at the

Center for Marital and Sexual Health in Ohio and was a clinical

professor of psychiatry at Case Western Reserve University School

of Medicine. The court informed Dr. Levine that Kosilek had been

living as a woman in prison and instructed him to treat this case

as if Kosilek were just another patient out in free society,

without all the issues attendant to her being incarcerated.

A month later, Dr. Levine, who did not interview Kosilek,

issued a written report. Commenting on the views held by Drs.

Brown and Schmidt, he noted they "reflect the current polarities

within psychiatry" with each of their respective positions having

merit. He wrote that "Dr. Schmidt's view, however unpopular and

uncompassionate in the eyes of some experts in GID, is within

prudent professional community standards."

In addition to submitting his report, Dr. Levine also

testified on December 19, 2006. At first, he reiterated that the

treatment recommended by Dr. Schmidt, though perhaps not popular,

was within prudent professional standards. He thought Kosilek had

obtained a good amount of relief from being on hormones and

dressing as a woman and was probably as feminine as she was going

to be; surgery would be "icing on the cake" he said. Dr. Levine,

like Dr. Schmidt, stopped short of saying that Kosilek would try to

kill herself if she was denied surgery. He thought it was just a

-44-
possibility and Kosilek's impulse could perhaps change over time.

And, he added, even if Kosilek received the surgery, it was

possible that, having nothing left to fight for, she could still

experience an emotional crisis and contemplate suicide. In

general, Dr. Levine had criticisms of both Dr. Schmidt's and Dr.

Brown's reports, but he thought they were both reasonable.

During questioning by Judge Wolf, it seemed Dr. Levine

had not followed some of the court's directives about not taking

into account the fact that Kosilek was in prison. Presuming

Kosilek would never be free, Dr. Levine did agree Kosilek had lived

a real-life experience as a woman in prison. He also believed she

essentially met all the eligibility and readiness requirements for

surgery under the Standards of Care. Dr. Levine opined that

providing Kosilek with surgery, assuming she had met the real-life

experience requirement, would be consistent with prudent

professional practice. And although Dr. Levine had earlier in his

testimony found Dr. Schmidt's approach to be prudent, he clarified

that putting aside issues such as cost and security, it would not

be within prudent professional standards to deny Kosilek surgery.

After hearing Dr. Levine's testimony, the court asked the

UMass doctors to prepare a written letter indicating what treatment

the DOC would propose as a possible alternative to surgery. In a

letter filed with the court on February 22, 2007, Dr. Brewer

indicated that, after consulting with Drs. Kapila and Kaufman, he

-45-
felt that, if denied surgery, Kosilek should receive psychotherapy

and should continue to receive hormones and feminine clothes and

items. However, Dr. Brewer hastened to add what Drs. Kapila and

Kaufman had repeatedly advised him: such treatment was "likely

going to be ineffective to relieve Ms. Kosilek's distress and may

well result in self harm or suicide."

5. Third Round of Testimony

Trial picked back up on March 15, 2007, with Kosilek

again calling Dr. Kaufman, as well as an additional witness, Dr.

Marshall Forstein of Harvard Medical School and the Cambridge

Health Alliance.

Dr. Forstein, who had evaluated Kosilek on behalf of the

DOC around the time of Kosilek I (and testified in that trial),

evaluated Kosilek again in 2005. Kosilek called on Dr. Forstein to

address whether surgery was appropriate and if psychotherapy would

be a reasonable alternative to surgery. Sex reassignment surgery,

Dr. Forstein opined, was the only reasonable treatment for Kosilek

for preventing the potential for self-mutilation or death. He

thought surgery was probably not appropriate for actively suicidal

patients, but he did not find Kosilek to be so. Nor did Dr.

Forstein think psychotherapy or antidepressants were reasonable

alternatives to surgery for Kosilek.

Dr. Kaufman again made her views clear during her

testimony: "We don't recommend psychotherapy. We recommend

-46-
surgery." Psychotherapy, she said, does not relieve gender

identity disorder, especially when it is severe: "rather than

trying to change the mind, we find that we need to change the

body." After this testimony, both parties rested.

6. Commissioner Clarke Enters the Fray

On May 2, 2007, Dennehy informed the court she was no

longer commissioner of the DOC and James Bender had taken her

place. Bender's tenure was short and he never testified. His

successor, Harold Clarke, became commissioner in November 2007.

In April 2008, the court ordered Clarke to review certain

trial transcripts, to submit a report of his conclusions, and to

offer testimony at a hearing to be held the following month.

Clarke submitted his report to the court on May 7, 2008.20 In it,

he took the same stance as Dennehy had before him.

"Insurmountable" is how Clarke characterized his concerns attendant

to providing Kosilek with sex reassignment surgery. Ticking off

several of those concerns, Clarke claimed that transporting Kosilek

out of the state for surgery would pose a flight risk based on

Kosilek's life sentence and her exodus from the state after killing

her wife.21 Next, he claimed there were flight risks with housing

20
In connection with preparing the report, Clarke read the
testimony of Kosilek, Dennehy, Spencer, and Bissonnette. He had
also reviewed several trial exhibits including the Fenway Report
and Dr. Seil's report.
21
Like Bissonnette, Clarke considered Kosilek's earlier
flight, even though the DOC classification manual says pre-

-47-
Kosilek post-surgery at MCI-Framingham, as it was not as secure as

MCI-Norfolk, and Kosilek would pose a risk to the female inmates

and vice versa. Clarke opined that the only viable post-operative

option would be to place Kosilek in the Special Management Unit at

MCI-Norfolk where she would be on lock down twenty-three hours a

day.

Clarke's written report highlighted other problems he

envisioned. An out-of-state transfer was not a viable option

because there was no guarantee another state would take Kosilek or,

if it did, that the state would keep her. Nor did Clarke think a

special unit for inmates with gender identity disorder would work

as such inmates vary greatly in their security and treatment needs.

Further, providing Kosilek with surgery in response to her threats

of suicide would be "contrary to well-established correctional

practices." Inmates should not "be permitted to manipulate the

system utilizing a 'do it or else' theory." All of these

conclusions, Clarke explained, were based on his own correctional

experience, and he expressly disclaimed reliance on any political,

media, or cost-related influences.

A few days after submitting his report, on May 12 and 13,

2008, Clarke came before Judge Wolf. He reiterated the security

concerns outlined in his report, but was forced to make a few

concessions on cross-examination. Clarke did not know that Kosilek

custodial flight should not be considered when classifying inmates.

-48-
was, at that time, fifty-eight years old; he had assumed she was

much younger. Prisoners in their fifties, he agreed, generally

cease to be aggressive and have a calming effect in the prison

environment. Clarke, who did not consult with any of the prison

superintendents in connection with his review, had also never read

Kosilek's recent classification report and was not aware of her

excellent disciplinary record. Yes, he was aware Kosilek had been

transported to scores of doctor's appointments without any attempt

at escape and he agreed that if need be the DOC could, "with some

degree of certainty," safely transport Kosilek for surgery out of

state. Clarke also acknowledged that his previous employer, the

State of Washington Department of Corrections (he had been the

secretary of this DOC), had housed without any issues a post-

operative female transgender inmate, who was serving a life

sentence for murdering his sister, in one of its women's prisons.

Clarke was also asked about some letters he had recently

received from Massachusetts lawmakers. In particular, a few days

after he was ordered by the court to testify, Clarke received a

letter signed by seventeen Massachusetts state senators voicing

their concerns over Clarke's review of Kosilek's case. The

senators "urge[d]" Clarke to deny Kosilek's request for surgery as

it would be an "affront to the taxpayers" and "raise a significant

security risk." A decision in favor of Kosilek would "send the

wrong message to the citizens of Massachusetts." Around that same

-49-
time Clarke received another letter signed by twenty-five state

representatives. Similar sentiments were expressed in that letter.

The representatives vented their "outrage" at Kosilek's request,

citing state budget concerns, and contended providing the surgery

would "set a bad precedent." At trial, Clarke said he did not

respond to the letters as he thought it would be inappropriate

given the review he was tasked with. Nonetheless, Clarke was aware

the lawmakers were drafting legislation seeking to limit the

state's ability to expend funds in this case. He was not however

(he said) influenced by the legislators' ire or any of the media's

anti-Kosilek coverage.

7. The Trial Comes to an End

After Clarke completed his testimony, the parties made

closing statements. Then, two years after it had begun, Kosilek's

trial came to an end. No more testimony was taken by the court

after May 2008. In total the court had sat for twenty-eight days

of trial. However time rolled on, and in 2009 and again in 2011

Judge Wolf heard additional arguments from the parties to address

recent judicial decisions and developments.

G. The Kosilek II Decision

On September 4, 2012, after confirming with the parties

they had nothing new to report, Judge Wolf issued his decision.

See Kosilek II, 889 F. Supp. 2d at 190. The decision was lengthy

(126 pages); it contained a thorough history of Kosilek's quest, a

-50-
detailed summary of the evidence adduced at trial, and loads of

factual findings. In order to prevail on her claim, Judge Wolf

found Kosilek had to prove five things: the first two

considerations as part of the Eighth Amendment's objective

component, the last three as part of its subjective prong. Kosilek

had to establish that: (1) she has a serious medical need; (2) the

need can only be adequately treated with sex reassignment surgery;

(3) the DOC knows that Kosilek is at a high risk for serious harm

if surgery is not performed; (4) the DOC did not deny Kosilek the

surgery based on a legitimate penological purpose, namely good

faith, reasonable security concerns; and (5) the DOC's

unconstitutional conduct will continue in the future. The decision

addressed each piece in turn.

First, with respect to the serious medical need

component, the court found that credible evidence at trial

established that Kosilek suffered from severe gender identity

disorder and would suffer serious harm if it was not adequately

treated. The court focused on Kosilek's threat of suicide

(determined to be credible and not manufactured) and the fact that

multiple highly qualified doctors employed by the DOC had diagnosed

Kosilek with a severe form of gender identity disorder, a diagnosis

confirmed by Drs. Brown and Forstein. The court's takeaway:

Kosilek had established a serious medical need.

-51-
 On point two — what treatment was adequate to address

this need — the court found the following. The Standards of Care

dictated a three-part sequence that called for hormone therapy, a

real-life experience living as the opposite sex, and then sex

reassignment surgery (though not all patients need or want all

these things). Kosilek of course had been on hormone therapy and

the court concluded the evidence showed that she underwent a real-

life experience living as a woman in prison. The court found Drs.

Brown, Kaufman, and Forstein credible when they testified,

consistent with the Standards of Care, that surgery was medically

necessary for some individuals with severe gender identity disorder

and, indeed, medically necessary for Kosilek herself. It also

concluded, relying in part on the opinions of Drs. Appelbaum and

Levine, that Dr. Schmidt's proposed alternative treatment of

psychotherapy, antidepressants, and suicide watch did not meet

prudent professional standards. More specifically, the court

determined that Dr. Schmidt's categorical views about sex

reassignment surgery, including his refusal to recommend sex

reassignment surgery for patients, were not supported by the

Standards of Care. Moreover, the treatment advocated by Dr.

Schmidt would not treat the cause of Kosilek's mental anguish (it

would diminish the symptoms at most) or reduce her suffering to the

point that she no longer had a serious medical need. Having

determined that Kosilek had proven she had a serious medical need

-52-
that had not been treated adequately, the court decided that

Kosilek had satisfied the objective prong of the Eighth Amendment

test. The court then turned its attention to the so-called

subjective prong: in essence, did the DOC know of and disregard an

excessive risk to Kosilek's health.

The court started with the third issue it had outlined —

whether the DOC knew that Kosilek was at risk for serious harm if

surgery was not performed. As a threshold matter, the court

considered who within the DOC should be its focus as far as who

knew what and when. The court determined, based in part on

stipulations from the parties, that former DOC Commissioner Dennehy

was the sole decision-maker on the issue of whether security

concerns should preclude Kosilek from getting surgery. And Dr.

Appelbaum of UMass, it decided, should be the focus of the court's

determination about whether the DOC was aware that Kosilek had a

serious need for surgery. Ultimately though, the court decided

this was a distinction without a difference because the evidence

established that both Dennehy and Dr. Appelbaum were aware of facts

from which they could infer — and did in fact draw the inference —

that "a substantial risk of serious harm to Kosilek existed." The

court focused on Dr. Appelbaum's numerous communications with

Dennehy in which he advised her of the serious risks faced by

Kosilek if the Fenway Center's proposed course of treatment was not

undertaken. It also looked to the fact that Dennehy had read the

-53-
Kosilek I decision, read testimony in this case, and that, in her

own testimony, she ultimately did not dispute that Kosilek had a

serious medical need.

The court's analysis proceeded to the fourth factor —

that is, whether the DOC denied Kosilek surgery based on a

legitimate penological purpose. First, the court explained it is

not legally permissible to deny a prisoner medical treatment based

on cost alone, but the court did not find cost to be the basis for

the DOC's decision here. Noting the DOC's provision of various

expensive medical treatments to many prisoners, the court

specifically declined to find that the DOC's denial of surgery to

Kosilek was motivated by the high cost of the treatment. The court

then moved on to the main bone of contention, which was whether the

DOC's proffered security concerns were its real reason for denying

Kosilek surgery. If the DOC's decision was made in good faith and

based on reasonable security concerns, the court said it would

defer to the DOC's decision. However, such deference was not to

be. Quite the contrary, the court found that, after a long period

of pretense and prevarication, Dennehy persisted in falsely

asserting that providing Kosilek with the surgery would present

insurmountable security concerns. The court found Dennehy's

security excuses nothing more than a pretext to deny Kosilek

surgery. Indeed, it was a "fear of controversy, criticism,

ridicule, and scorn" that was the real driving force behind the

-54-
DOC's decision to withhold surgery. In support of its conclusion,

the court cited Dennehy's pattern of delay, deviation from DOC

policy, feigned ignorance about what treatment DOC doctors were

recommending, failure to do a thorough security evaluation before

making up her mind, the unrealistic nature of the supposed security

risks, and the public and political outcry against Kosilek. It

summed things up this way: "Because there is no penological

justification for denying Kosilek the treatment prescribed for

[her], [s]he is now being subject to the 'unnecessary and wanton

infliction of pain' prohibited by the Eighth Amendment."

Expressing a reticence to tell the DOC how to discharge

its duty, and lamenting the fact that the DOC had not just made the

proper medical decisions on its own, the court moved on to the

final consideration, which was whether the DOC's unconstitutional

conduct, absent court intervention, would continue. Based on the

DOC's pattern of unconstitutional conduct as chronicled above, the

court determined the DOC's deliberate indifference would not cease

without judicial intervention. Kosilek was entitled to an

injunction. Because an injunction must be narrowly tailored, the

court declined to decide who should do the surgery, where it should

be done, or where Kosilek should be housed afterwards. Rather

Judge Wolf simply ordered the DOC to provide Kosilek with sex

reassignment surgery. The order read: "Defendant shall take

-55-
forthwith all of the actions reasonably necessary to provide

Kosilek sex reassignment surgery as promptly as possible."

The DOC quickly filed this appeal and moved the district

court to stay its order to provide Kosilek with surgery pending

resolution of the appeal, which the court did.

II. THE ISSUES ON APPEAL

After setting forth the extensive backdrop of Kosilek's

odyssey, we start by narrowing the issues before us. On appeal,

the DOC attacks the district court's decision but limits its

offensive to just a couple of issues. To start, it does not take

umbrage with many of the district court's findings, namely that

Kosilek has a serious medical need, that the DOC knows she has a

serious medical need, that surgery could appropriately treat this

need, or that injunctive relief (should the DOC's conduct be deemed

unlawful) is an appropriate vehicle for relief. And based on the

evidence adduced at trial, we think the DOC would not have had

strong arguments on any of these points.

The two basic criticisms the DOC levels at the district

court's decision are the following. First, it claims the court

erred in finding that the DOC's decision not to provide Kosilek sex

reassignment surgery constitutes inadequate medical care in

violation of the Eighth Amendment. Second, it contends the court

erred in finding the DOC deliberately indifferent to Kosilek's need

for treatment. We elaborate.

-56-
 On point one, the DOC argues that surgery is not

constitutionally required because the treatment Kosilek is already

receiving — psychotherapy, hormones, permanent hair removal, and

access to female clothing and cosmetics — is adequate. Kosilek, it

contends, is not entitled to the most sophisticated or desirable

treatment or to curative treatment, as opposed to treatment for her

mental illness. And should Kosilek engage in self-injurious

behavior (e.g., attempt to commit suicide) if denied surgery, the

DOC contends treating this behavior with methods like

psychotherapy, antidepressants, and a protective environment would

be adequate. In sum, the DOC thinks the court erred when it found

the DOC's current and proposed treatment course inadequate (i.e.,

failed to meet prudent professional standards) under the Eighth

Amendment and the sex reassignment surgery medically necessary.

Kosilek disagrees. She counters that the court correctly

found surgery to be the only adequate treatment for her serious

medical need — a finding supported by more than ample evidence.

Kosilek points out that almost all of the medical professionals who

testified at trial agreed this was the case, and only the DOC's

experts disagreed; Kosilek says the court was entitled to

disbelieve them. She claims the evidence also establishes that,

even with hormones and psychotherapy, her mental anguish is severe

and the DOC's proposed course of treatment would not sufficiently

-57-
reduce this. Sex reassignment surgery is medically necessary,

according to Kosilek.

As to the court's deliberate indifference finding, the

DOC contends the court erred in rejecting the safety and security

concerns presented by DOC officials at trial in support of their

decision to withhold surgery, which they say were based on their

best correctional judgment and were reasonable and not pretextual.

The DOC "vigorously disputes" that the public criticism of Kosilek

and her request for surgery had any impact on the decision-making

process of Commissioner Dennehy or her successor, Clarke. However,

other than this broad assertion it says little else about Dennehy

herself and her motives. Instead it homes in on Dennehy's ultimate

successor Clarke, claiming that he should be the focus of this

court's determination as to the validity of the security concerns

voiced by the DOC (despite the DOC's at-trial stipulation that

Dennehy was the operative decision-maker).22 There was an absence

of evidence in the record, says the DOC, that Clarke's proffered

security concerns were exaggerated or made in bad faith and so, the

court was required to accord him deference. Further, the DOC

points out Kosilek offered no counter correctional expert

testimony, and the district court did not find that Superintendent

22
At oral argument counsel for the DOC clarified that he does
not suggest that this court disregard Dennehy's testimony, only
that it should focus on Clarke's as it is more relevant being
closer in time.

-58-
Bissonnette or Spencer acted in bad faith or were tainted by a fear

of public criticism.

On the court's deliberate indifference finding, Kosilek

argues that the DOC had no legitimate security reason to withhold

surgery and importantly, more than sufficient evidence supported

the court's conclusion that the alleged security concerns were

pretextual. Kosilek points to several pieces of evidence: the

security decision was made before any security review occurred; a

subsequent security review did not follow DOC procedure; the

security report was hastily drafted by trial counsel; and the

security evaluation that was performed did not take into account

certain fundamental factors like Kosilek's age and good

disciplinary record. Kosilek adds that there was evidence that

Kosilek's transportation to surgery and post-operative placement

would not be actually impossible from a security perspective. As

such, no "mechanical deference" is owed to the DOC, according to

Kosilek, whose argument focuses on the illegitimacy of the

proffered security concerns rather than the supposed role public

criticism played in the DOC's decision. Kosilek also lambasts the

DOC for attempting to distance itself from Dennehy on appeal since

below the DOC stipulated that Dennehy's motivations should be the

focus and in fact objected to Clarke's testimony as unnecessary.

With the arguments delineated and the issues narrowed, we

turn to the operative law.

-59-
 III. EIGHTH AMENDMENT CRITERION

The Eighth Amendment provides the vehicle through which

courts scrutinize "the treatment a prisoner receives in prison and

the conditions under which he is confined." Farmer v. Brennan, 511

U.S. 825, 832 (1994) (internal quotation marks omitted). One way

prison officials violate the Eighth Amendment is when they fail to

provide an inmate with adequate medical care, such that "their

'acts or omissions [are] sufficiently harmful to evidence

deliberate indifference to serious medical needs.'" Leavitt v.

Corr. Med. Servs., Inc., 645 F.3d 484, 497 (1st Cir. 2011) (citing

Estelle v. Gamble, 429 U.S. 97, 106 (1976)). To prevail on an

Eighth Amendment inadequate medical care claim, a plaintiff must

satisfy two inquiries, one objective and one subjective. Id.

The objective component requires that "the deprivation

alleged must be, objectively, sufficiently serious." Farmer, 511

U.S. at 834 (internal quotation marks omitted); Leavitt, 645 F.3d

at 497. Thus an Eighth Amendment claim such as this one turns, in

part, on whether the prisoner has a "serious medical need," in

other words, "one that has been diagnosed by a physician as

mandating treatment, or one that is so obvious that even a lay

person would easily recognize the necessity for a doctor's

attention." Mahan v. Plymouth Cnty. House of Corr., 64 F.3d 14,

18-19 (1st Cir. 1995) (internal quotation marks omitted). A

prisoner is entitled to adequate medical care for that need, though

-60-
this does not necessarily mean the most sophisticated care

available. United States v. DeCologero, 821 F.2d 39, 42 (1st Cir.

1987). Rather, adequate care is "services at a level reasonably

commensurate with modern medical science and of a quality

acceptable within prudent professional standards." Id. at 43.

For the subjective prong to be satisfied, prison

officials must have had "a sufficiently culpable state of mind";

that is, they showed deliberate indifference to an inmate's health

and safety. Farmer, 511 U.S. at 834; Leavitt, 645 F.3d at 497.

"Deliberate indifference means that a prison official subjectively

must both be aware of facts from which the inference could be drawn

that a substantial risk of serious harm exists, and he must also

draw the inference." Ruiz-Rosa v. Rullán, 485 F.3d 150, 156 (1st

Cir. 2007) (internal quotation marks omitted). To establish a

subjective intent, "a deliberate intent to harm is not required,"

rather "it is enough for the prisoner to show a wanton disregard

sufficiently evidenced 'by denial, delay, or interference with

prescribed health care.'" Battista v. Clarke, 645 F.3d 449, 453

(1st Cir. 2011) (quoting DesRosiers v. Moran, 949 F.2d 15, 19 (1st

Cir. 1991)).

If a court finds that the Eighth Amendment's objective

and subjective prongs have been satisfied, it may grant appropriate

injunctive relief. Farmer, 511 U.S. at 846.

-61-
 IV. STANDARD OF REVIEW

Here Kosilek's Eighth Amendment claim was heard

exclusively by a judge. When this court decides an appeal from a

judgment following a bench trial, different standards of review are

at play. Wojciechowicz v. United States, 582 F.3d 57, 66 (1st Cir.

2009).

First, questions of law engender de novo review. Id.

Findings of fact, however, are reviewed only for clear error. Id.

This means we accept the court's factual findings unless the

evidence compels us to conclude contrarily that a mistake was made,

"keeping in mind that the district judge had the opportunity to

assess the credibility of the witnesses." Janeiro v. Urological

Surgery Prof'l Ass'n, 457 F.3d 130, 138 (1st Cir. 2006). "'This

deferential standard extends . . . to inferences drawn from the

underlying facts,' and 'if the trial court's reading of the record

[with respect to an actor's motivation] is plausible, appellate

review is at an end.'" Id. at 138-39 (alteration in original)

(quoting Smith v. F.W. Morse & Co., 76 F.3d 413, 420 (1st Cir.

1996)). Finally, with mixed questions of law and fact, there is a

continuum. Johnson v. Watts Regulator Co., 63 F.3d 1129, 1132 (1st

Cir. 1995). The more fact-intensive the question, the more

deferential our review. Id. Conversely, the more law-dominated

the question, the more likely our review moves toward a de novo

look. Id.

-62-
 Before going any further we address a point of dispute

among the parties. While everyone agrees that the above standard

of review controls, the parties disagree on how to apply it.23

Kosilek argues that the district court's determinations on the

adequacy of the medical care and the viability of the purported

security rationale are findings of fact and therefore this court's

review is for clear error only. The DOC, however, claims that the

court's conclusion regarding the appropriateness of Kosilek's

health care is a conclusion of law engendering de novo review; it

does not address at all what standard should apply to the court's

deliberate indifference finding. Neither side gives us any legal

support for the respective positions they take in this debate.

This court has not provided clear guidance; however, we

are not left totally in the dark. For one, we have explained that

a "state-of-mind issue such as the existence of deliberate

indifference usually presents a jury question," Torraco v. Maloney,

923 F.2d 231, 234 (1st Cir. 1991), in other words an issue for the

finder of fact.24 Further, when reviewing a trial judge's

23
In violation of our rules, the DOC did not include a
standard of review in its opening brief. See Fed. R. App. P.
28(a)(9)(B). Not until its reply brief did we learn the DOC's
stance on this important issue.
24
The Eighth Circuit has held that "whether an official was
deliberately indifferent to [an] inmate's serious medical need" is
a question of fact. Schaub v. VonWald, 638 F.3d 905, 915 (8th Cir.
2011). On the other hand, the Fifth Circuit has characterized it
as a legal conclusion resting on certain facts. Walker v. Butler,
967 F.2d 176, 178 (5th Cir. 1992).

-63-
determination on the adequacy of medical treatment following a

bench trial, this court has applied the deferential clearly

erroneous standard. DesRosiers, 949 F.2d at 19-20.25 And it is

well established that "elusive issues of motive and intent" (at

play here in connection with the subjective prong) are typically

fact-bound ones subject to the clearly erroneous rule. Fed.

Refinance Co. v. Klock, 352 F.3d 16, 27-28 (1st Cir. 2003); see

also McIntyre ex rel. Estate of McIntyre v. United States, 545 F.3d

27, 40 (1st Cir. 2008).

On top of all this, the district court here "engaged in

a careful and close analysis of the trial evidence" and therefore,

given the fact-intensive nature of the court's inquiry, we are

satisfied that the court's findings — that the DOC refused to

provide medically necessary treatment for a serious, life-

threatening medical condition that could be accommodated without

security risk — should be reviewed for clear error only. See,

e.g., Turner v. United States, 699 F.3d 578, 584 (1st Cir. 2012).

To the extent that any of the court's findings could be viewed as

more akin to a mixed question of law and fact or even a legal label

(the most arguable one being whether the DOC's conduct amounted to

deliberate indifference), some deference is still appropriate. See

25
The Ninth Circuit seems to take a similar approach. It
described a dispute about whether "any option other than surgery
was medically acceptable" for an inmate alleging an Eighth
Amendment violation as an "issue of fact." Snow v. McDaniel, 681
F.3d 978, 988 (9th Cir. 2012).

-64-
Battista, 645 F.3d at 454 ("The legal labels applied to facts are

reviewed on appeal more closely than a district court fact-finding,

but often with some deference to the district judge."); Fed.

Refinance Co., 352 F.3d at 27 (explaining that the more fact-

intensive the question the more deferential our review).26

The dissent first diverges from this opinion on the

standard of review issue. While our dissenting colleague agrees

that a continuum (anchored on one side by de novo review and on the

other by clear error review) is at play, he is not convinced that

the district court's controlling findings should be reviewed on the

clear error end of the spectrum. However, the dissent does not

explain how any of the findings made by the district court, which

26
Our precedent indicates that where a district court's
decision threatens to intrude on a party's First Amendment rights,
we must "make an independent examination of the whole record" in
reviewing a decision that there is no such intrusion. Bose Corp.
v. Consumers Union of U.S., Inc., 466 U.S. 485, 499 (1984) (quoting
New York Times Co. v. Sullivan, 376 U.S. 254, 285 (1964)). But we
are aware of no precedent for importing this elevated review into
other contexts not implicating a need to guard against judicial
restriction of First Amendment rights. Supreme Court decisions
certainly signal no license to extend Bose Corp. beyond First
Amendment cases. See Maine v. Taylor, 477 U.S. 131, 145 (1986)
("[N]o broader review is authorized [under Bose Corp.] simply
because this is a constitutional case."); Hernandez v. New York,
500 U.S. 352, 367-69 (1991) (refusing to apply elevated review
under Bose Corp. and instead applying clear error review to a claim
under the Equal Protection Clause); Thornburg v. Gingles, 478 U.S.
30, 78-79 (1986) (doing the same in a claim of vote dilution under
the Voting Rights Act). And as our dissenting colleague
perceptively explained in another opinion, it is not clear that a
heightened standard would apply even if this were a First Amendment
case, because the district court did not reject but sustained the
assertion of a constitutional right. United States v. Frabizio,
459 F.3d 80, 96 (1st Cir. 2006) (Torruella, J., concurring).

-65-
led to the result we are affirming, are anything more than

quintessentially factual findings. And while the dissent argues

that we must review de novo the district court's conclusion that

the facts demonstrate an Eighth Amendment violation, where a legal

conclusion flows directly from factual findings, our review of

those factual findings — and thus the overall conclusion — remains

deferential. For example, in a recent Eighth Amendment case

similar to this one, we reviewed the district court's determination

that the state had been deliberately indifferent to the needs of a

prisoner suffering from gender identity disorder by asking whether

the court had a "reasonable basis" for its conclusion, not whether

the district court actually reached the right decision. Battista,

645 F.3d at 454-55. Here, the success of Kosilek's claim depends

almost entirely on questions of credibility (in assessing the

state's motives) and on questions of medical care (in assessing

Kosilek's medical needs). These are exactly the kinds of questions

that we must review deferentially, especially where, as here, there

is ample testimony precisely and directly supporting the district

court's answers to such questions.

With the parameters of our review cleared up, we move on

to the merits of this case.

V. THE OBJECTIVE PRONG

As we said, the DOC does not dispute that Kosilek has a

serious medical need. The real issue is whether the district court

-66-
erred in finding that the care the DOC has provided Kosilek with

(i.e., hormones, psychotherapy, and feminine items) and proposes to

provide her with to relieve any self-injurious behavior (i.e.,

psychotherapy, antidepressants, and a protective environment) is

adequate. Said another way, is, as the district court found, sex

reassignment surgery medically necessary and the only appropriate

treatment for Kosilek?

Our review starts off with the DOC's expert Dr. Schmidt,

who the district court concluded was not a prudent professional.27

This finding was not clearly erroneous. First, Dr. Schmidt

expressed a good deal of disagreement with the Standards of Care.28

However, the Standards of Care were widely relied upon and trusted

by the other medical providers who testified at trial and have been

cited by the courts as generally accepted. See, e.g., De'Lonta v.

Johnson, 708 F.3d 520, 522-23 (4th Cir. 2013) (describing the

27
In its decision, the court noted it had considered Osborne's
testimony but was not discussing it separately because she
testified consistent with her colleague Dr. Schmidt and there was
a question as to whether she should even be considered in the
prudent professional debate because she is a social worker and not
a medical doctor. We take the same tack and focus on Dr. Schmidt.
28
The Seventh Version of the Standards of Care, which, as
indicated, came out in 2011, contains a new section addressing the
applicability of the standards to persons living in institutional
environments such as prison or long-term care facilities.
Standards of Care, Version 7, at 67. It indicates that, for those
persons, health care "should mirror that which would be available
to them if they were living in a non-institutional setting" and
that "[a]ll elements of assessment and treatment as described in
the [Standards of Care] can be provided to people living in
institutions." Id.

-67-
Standards of Care as "the generally accepted protocols for the

treatment of GID"); Soneeya v. Spencer, 851 F. Supp. 2d 228, 231

(D. Mass. 2012) (noting that the "course of treatment for Gender

Identity Disorder generally followed in the community is governed

by the 'Standards of Care'"); O'Donnabhain v. Comm'r of Internal

Revenue, 134 T.C. 34, 65 (U.S. Tax Ct. 2010) (indicating that the

Standards are "widely accepted in the psychiatric profession, as

evidenced by the recognition of the standards' triadic therapy

sequence as the appropriate treatment for GID and transsexualism in

numerous psychiatric and medical reference texts").

More specifically, the Standards of Care, which the

parties agree are applicable in some regard to the treatment of

gender identity disorder (though the DOC would certainly like to

see them given less weight than Kosilek would), provide for a

triadic approach: a real-life experience in the preferred gender

role, hormones of the desired gender, and surgery to alter one's

genitalia. The Standards of Care indicate that for persons with

severe gender identity disorder, sex reassignment surgery is

effective. Standards of Care at 18. In such persons, surgery

paired with hormone therapy and a real-life experience is

"medically indicated and medically necessary." Id. Surgery is

not, the Standards of Care say, "experimental, investigational,

elective, cosmetic, or optional in any meaningful sense." Id. Dr.

Schmidt, however, disagreed with the proposition that surgery is

-68-
medically necessary in cases of severe gender identity disorder.

In fact, he did not recall seeing a case of gender identity

disorder serious enough to require surgery in the approximately 300

patients he has evaluated. Dr. Schmidt's views regarding the need

for surgery for patients with severe gender identity disorder was

not only unsupported by the Standards of Care but also contradicted

by the testimony of the other medical providers at trial.

As for what treatment was appropriate for Kosilek in

particular, Dr. Schmidt opined that surgery was not necessary as

Kosilek had done well just using hormones thus far. The main

barrier Dr. Schmidt saw was that, in his opinion, a real-life

experience living as the opposite gender could not be effectively

replicated in prison. Rather, Dr. Schmidt's proposed course of

treatment was to utilize psychotherapy and medication to reduce

Kosilek's dysphoria to a point where she was no longer at risk for

serious harm. Should her suicidal desires persist, Dr. Schmidt

thought Kosilek could be effectively managed in a medical facility.

However, a majority of the other medical providers who testified at

trial regarding Dr. Schmidt's proposed course of treatment thought

it was unreasonable (namely, Drs. Kapila, Kaufman, Appelbaum, and

Forstein).29 The consensus was that Dr. Schmidt's approach would

29
We do not list Dr. Levine among these providers because, as
the reader will recall, there was some change in his testimony.
Initially he opined that Dr. Schmidt's view, though unpopular, was
reasonable. However, Dr. Levine apparently had disregarded the
court's order to treat Kosilek as if she were a patient out in free

-69-
not effectively treat the real issue, which was Kosilek's gender

identity disorder, or significantly relieve her distress to a level

where she was not at risk for serious harm. As Dr. Kapila

explained, Kosilek's symptoms and distress had not been resolved

though she had been on hormones for some time.

Ultimately, the court was confronted with two

diametrically opposed opinions, both given by qualified medical

professionals. Given the contradictory evidence, which was heavily

weighted against the DOC, we are far from left with the impression

that the trial court made a mistake when it determined that Dr.

Schmidt was not a prudent professional or that his approach was

unreasonable. See Janeiro, 457 F.3d at 138. The court did not

clearly err.

The question that remains is what treatment is medically

adequate to treat Kosilek. The DOC twists the district court's

holding, claiming that it impermissibly held that the Eighth

Amendment requires treatment that actually "cures" the inmate's

condition. The court said no such thing. It simply found that the

only adequate treatment in this case was sex reassignment surgery.

society. After being reminded of those parameters, Dr. Levine
clarified his opinion. He said, putting aside a variety of
concerns about the surgery, including cost, security, and the
requirement of a "real life experience," prudent professionals do
not ordinarily prevent a patient from receiving sex reassignment
surgery. But even were we to disregard Dr. Levine's opinion
entirely, there was ample evidentiary support for the court's
conclusion that Dr. Schmidt's approach was not prudent.

-70-
And though, according to the experts, surgery did have the

potential to cure or at least greatly alleviate Kosilek's gender

identity disorder, this does not translate to a finding by the

court that only curative treatment passes constitutional muster.

To give an example, if Kosilek had cancer and the court found

chemotherapy to be the only adequate treatment under the Eighth

Amendment, as opposed to, say, an aspirin, it would not necessarily

follow that the court held that the Eighth Amendment requires an

inmate to be cured. The DOC's argument misses the mark.

However, as the district court found, there is a

difference between treating the underlying disorder and treating

only its symptoms. The Seventh Circuit Court of Appeals drew a

similar distinction when it held that a Wisconsin statute which

prohibited the state's department of corrections from providing

transgender inmates with hormones and sex reassignment surgery was

unconstitutional. Fields v. Smith, 653 F.3d 550, 552-53, 559 (7th

Cir. 2011). The court, discussing how some patients require

hormone therapy, noted that although the defendant department could

"provide psychotherapy as well as antipsychotics and

antidepressants, defendants failed to present evidence rebutting

the testimony that these treatments do nothing to treat the

underlying disorder." Id. at 556. The defendants failed to show,

the court concluded, that another treatment could be an adequate

stand-in for hormone therapy. Id.

-71-
 And providing some treatment is not the same as providing

adequate treatment. The Fourth Circuit Court of Appeals, in

De'Lonta, 708 F.3d at 520, reversed the district court's dismissal

of a transgender inmate's Eighth Amendment claim. The court found

that the inmate, who sought sex reassignment surgery because her

gender identity disorder symptoms persisted despite receiving

hormones, stated a plausible deliberate indifference claim. Id. at

522, 525. Notably the court found that just because the Virginia

Department of Corrections had provided the inmate with some

treatment for her gender identity disorder (hormone therapy and

psychological counseling) consistent with the Standards of Care,

"it does not follow that they have necessarily provided her with

constitutionally adequate treatment." Id. at 522, 526. The court

added that total deprivation of care is not a prerequisite for a

constitutional violation. Id. at 526.

Here there was testimony from the medical providers at

trial that the preferred approach is to treat the underlying

problem (the gender identity disorder) and not just the symptoms,

as Dr. Schmidt proposed. "[Y]ou treat the primary problem, you

don't treat symptoms," Dr. Kapila testified. And while the

evidence, as recognized by the district court, was that Kosilek had

shown improvement being on hormones and had obtained some measure

of stability holding on to the hope of receiving surgery, all the

medical providers, save Dr. Schmidt, agreed that this was not

-72-
enough. As the Fenway Center doctors concluded early on, all signs

indicated that Kosilek was still quite distressed and likely to

suffer serious harm in the form of suicide if not operated on.

The take-away from the evidence: surgery was the next

logical adequate treatment step. Kosilek had been on hormones for

some time and the evidence, in the form of reports and testimony

from the Fenway Center doctors and Drs. Forstein and Brown,

supported the notion that Kosilek had undergone a real-life

experience living as a woman in prison. In fact, Dr. Brown

testified that in his opinion Kosilek had not only met the minimum

real-life experience but had exceeded it. Dr. Brown focused on the

amount of information (presumably the significant medical and

disciplinary records kept in a prison setting) that existed

regarding Kosilek's time in prison. Dr. Brown would not, he

explained, have such a thorough record with a patient in the

outside world. And Dr. Schmidt's opinion that a real-life

experience could never be replicated in prison did not take into

account Kosilek's situation in particular, or more generally the

different realities of transgender prisoners. As the district

court found, "[f]or someone like Kosilek who is serving a sentence

of life without the possibility of parole, prison is, and always

will be, [her] real life."

With the prerequisites for surgery satisfied, Drs. Brown,

Kaufman, Forstein, Kapila, and Appelbaum all testified

-73-
unequivocally: sex reassignment surgery was medically necessary and

the only appropriate treatment for Kosilek. And they all agreed

that a very likely consequence of Kosilek not receiving the surgery

was a serious risk of harm, predominantly suicide. Kosilek herself

testified, and the court found credibly so, that though hormone

treatments had helped, she still suffers intense mental anguish

over her male genitalia and believed she needed surgery. As

Kosilek explained, she did not want to continue living with her

male genitalia and antidepressants and psychotherapy would not

change that.

We are bound by certain well-established adages: a party

challenging findings of fact after a bench trial "faces a steep

uphill climb" and this court "is not at liberty to start afresh."

Monahan v. Romney, 625 F.3d 42, 46 (1st Cir. 2010) (internal

quotation marks omitted). Here there was ample evidence to support

the district court's conclusion that sex reassignment surgery was

the only adequate treatment for Kosilek. In fact, substantial

evidence, notwithstanding Dr. Schmidt's testimony, pointed in that

direction. The court did not clearly err in deciding to credit the

evidence and testimony offered by Kosilek and in concluding that

the objective component of the Eighth Amendment inquiry had been

satisfied. In sum, where at least three eminently qualified

doctors testify without objection, in accord with widely accepted,

published standards, that Kosilek suffers from a life-threatening

-74-
disorder that renders surgery medically necessary, and the

factfinder is convinced by that testimony, we are at a loss to see

how this court can properly overrule that finding of fact.

VI. THE SUBJECTIVE PRONG

The basis for the district court's deliberate

indifference conclusion was its findings that the DOC had no valid

penological reason to deny surgery. As indicated earlier, the

judge primarily focused on Dennehy (as the parties stipulated),

concluding that she delayed treatment, deviated from policy,

inadequately reviewed security ramifications, and manufactured

security concerns. The court also assessed Clarke's motivations

and concluded that Clarke's failure to do a thorough security

review suggested he did not operate with an open mind. Also found

to be important on both the Dennehy and Clarke front: the

unrealistic nature, in the court's mind, of the supposed security

risks and the public and political outcry against Kosilek. The

issue for this court to decide is whether the record supported

these underlying factual conclusions.

First, let us get a few things out of the way. Though a

lot of the public and political opposition to Kosilek receiving

surgery involved a rally cry that taxpayers should not have to foot

the bill for the procedure, the district court did not think that

cost considerations played any role in the DOC's decision to deny

surgery. And both the parties agree with this point on appeal;

-75-
there was no evidence that the DOC withheld surgery because it was

too expensive. Therefore we have no cause to consider the cost

issue.30 Second, as alluded to, the parties disagree about whether

Dennehy or Clarke or both should bear the brunt of this court's

attention on the validity-of-the-security-concerns issue. In suits

like this one, where injunctive relief is sought "to prevent a

substantial risk of serious injury from ripening into actual harm,"

we consider deliberate indifference "'in light of the prison

authorities' current attitudes and conduct,' [and] their attitudes

and conduct at the time suit is brought and persisting thereafter."

Farmer, 511 U.S. at 845 (quoting Helling v. McKinney, 509 U.S. 25,

36 (1993)). Therefore we will look at both Dennehy and Clarke.

With those preliminary points dealt with, we proceed to the merits.

30
Although the DOC has not argued the cost issue, much media
commentary focused on the cost of sex reassignment surgery for
Kosilek, and a prior commissioner of the DOC claimed that the
surgery was an inappropriate use of taxpayer funds. As the
district court noted in Kosilek II, one Boston Globe column from
2000 griped that Kosilek was "demanding that the state, meaning you
and me, pay the $25,000 for a sex-change operation." 889 F. Supp.
2d at 215. While the record does not indicate if that number
reflects the exact cost of the surgery today, it appears to be in
the ballpark. Yet, the cost of sex reassignment surgery pales in
comparison to the amount of money it seems the state will be
expending to defend this lawsuit. Around a year ago, Kosilek's
attorneys filed a motion in the district court seeking to have the
DOC pay them over $800,000 in fees and costs. In a tentative
ruling issued from the bench, Judge Wolf indicated that he is
planning to order the DOC to pay around $700,000, though it does
not appear that this amount has yet been formalized in a written
order.

-76-
 Clearly, when decisions about medical care are made in

the prison system, there are certain considerations not present in

society at large. "Any professional judgment that decides an issue

involving conditions of confinement must embrace security and

administration and not merely medical judgments." Battista, 645

F.3d at 455. And because "security considerations . . . matter at

prisons" and conflicting demands must be balanced by officials, the

deliberate indifference test "leave[s] ample room for professional

judgment." Id. at 453, 454.

However, as we explained in Battista, at some point a

defendant forfeits the advantage of deference. Id. at 455. In

Battista, which involved some of the same players as this case,

this court considered the claim of another transgender inmate suing

the Massachusetts DOC. Id. at 450. This court found the record

supported the district court's determination that the DOC had shown

deliberate indifference by refusing to provide the inmate with the

hormone treatment that doctors had recommended for her. Id. at

450, 455. As it does in this case, the DOC cited security concerns

for denying the recommended treatment. Id. at 452. Nonetheless,

in reaching its result, the Battista court refused to give the DOC

the advantage of deference. Id. at 455. It cited a "pattern of

delays, new objections substituted for old ones, misinformation

and other negatives," including an initial failure to take the

inmate's diagnosis of gender identity disorder and request for

-77-
hormones seriously, the passage of years before a substantial

security justification was made, and the DOC's portrayal of the

only options as withholding hormones or placing the inmate in

severely constraining protective custody. Id.

Here Judge Wolf found a very similar pattern of behavior

on the DOC's part and the record supports this finding. On the

delay front, it has indisputably been many years since medical

providers started considering the propriety of surgery for Kosilek.

Dr. Forstein, back in then Commissioner Maloney's tenure (the years

1997 to 2003) during the days of Kosilek I, had recommended that

Kosilek be allowed to consult with a surgeon who specialized in sex

reassignment surgery. In 2003, Dr. Seil made the same

recommendation, indicating that Kosilek should be allowed to meet

with a specialist after a year on hormones. Then in February 2005,

the Fenway Center doctors indicated after evaluating Kosilek that

she should be allowed to have surgery. Dennehy herself was Deputy

Commissioner during Kosilek I and was involved in the decisions

made in connection with that case. And right when she started as

Commissioner, Dennehy slowed things down. She took the unusual

step in assuming an active role in a de novo blanket reassessment

of the treatment of those inmates suffering from gender identity

disorder, including Kosilek, despite the fact that the DOC's

contract with UMass provided that the UMass medical professionals

would make the decisions about the medical care for inmates with

-78-
this disorder and the Commissioner would only step in at the end to

assess any security concerns.

The DOC explains away this delay by claiming that for a

long time it did not understand that UMass recommended surgery for

Kosilek, but the district court did not buy it. And, in reality,

the issue seems to be one of semantics. While there was testimony

from Hughes and Martin that they knew UMass supported the Fenway

Center doctors' recommendation for surgery, they did not think

UMass was clear enough on the logistics or whether surgery was

"medically necessary" (as opposed to medically optional). However,

Drs. Appelbaum and Brewer made pellucid in their June 14, 2005

letter that they did not think it was within their purview to

decide as a legal matter whether surgery for Kosilek was medically

necessary. Thus, in the end, there was evidence that the DOC knew

that Kosilek's medical providers were recommending surgery, and in

response, the DOC dallied and disregarded. This behavior is

significant, as in order to establish a subjective intent, "it is

enough for the prisoner to show a wanton disregard sufficiently

evidenced 'by denial, delay, or interference with prescribed health

care.'" Battista, 645 F.3d at 453 (quoting DesRosiers, 949 F.2d at

19); see also Johnson v. Wright, 412 F.3d 398, 404 (2d Cir. 2005)

(A "deliberate indifference claim can lie where prison officials

deliberately ignore the medical recommendations of a prisoner's

treating physicians.").

-79-
 While the DOC maintains that it opposed surgery based on

security concerns, the district court did not clearly err in

finding that these concerns followed hasty, results-driven

evaluations. Written procedure adopted by the DOC required that,

subsequent to a medical provider's recommendation that an inmate

receive treatment for gender identity disorder, the Superintendent

in the relevant prison would conduct a security assessment and make

a recommendation to the Commissioner. Despite this policy, the DOC

concluded, as was reported in the news piece Dennehy was

interviewed for, that surgery for Kosilek "would result in a

security nightmare" before it ever conducted such a review.

Once the security review was eventually undertaken, the

DOC did not give itself much time to complete it. Although the DOC

was ordered by the court back on April 25, 2005 to conduct the

review, Dennehy, the attorneys, and the superintendents did not

meet until May 19 to produce the report, which was due to the court

on May 27. The report was then penned predominantly by trial

counsel and reviewed by Dennehy only a day or two before it went to

Judge Wolf. This pattern of haste continued. On November 23,

2005, just nine days before expert disclosures were due, Dennehy

contacted the director of the Federal Bureau of Prisons looking for

a recommendation for a security expert to testify at trial. When

the DOC experts did testify, it was apparent, as the court

explained, that both Dumond and Beeler "failed to consider material

-80-
aspects of Kosilek's history and personal characteristics in

forming their opinions," such as her good disciplinary record or

her medical records. As a result, the court gave their testimony

"little weight."

As for Clarke's security review, the court criticized him

for not consulting with Spencer, then still the Superintendent of

MCI-Norfolk, and for apparently not reviewing any of the DOC's

expert security testimony offered at trial, when deciding whether

he agreed with Dennehy's stance on surgery. The DOC calls this

finding disingenuous since the court's order only called for Clarke

to review certain testimony and Clarke testified that accordingly

this is what he did. It is unclear whether Clarke shirked his

review responsibilities or genuinely felt that his review was

limited in scope. But both views are permissible and when faced

with two permissible views of the evidence, the district judge's

choice of one of them cannot be clearly erroneous. See Monahan,

625 F.3d at 46. Plus the thoroughness of Clarke's review is

undercut by the fact that he did not know some important pieces of

information, such as Kosilek's age and excellent disciplinary

record, when he advanced his security concerns.

Another denouement of the district court was that the

security concerns the DOC ultimately proffered (following its hasty

review) were "largely false" and "greatly exaggerated." This

conclusion was not clearly erroneous. The DOC repeatedly protested

-81-
that transporting Kosilek to surgery out of state would pose an

insurmountable security risk. On its face, the concern seems

patently unrealistic. First, the DOC undoubtedly has a large

amount of experience transporting prisoners within and outside of

Massachusetts. Further, the likelihood of Kosilek, who has been

transported to multiple doctor's appointments without issue,

fleeing while traveling to receive the surgery that she has

dedicated decades of her life to obtaining is improbable enough

that we need say nothing more. Almost equally as unlikely is the

idea that a now sixty-four year old, post-surgical, recovering

Kosilek would be able to escape when being transported back to

prison. Even Clarke conceded that it was near certain that the DOC

could safely transport Kosilek to and from surgery. On top of all

this, there was evidence that the surgery might be able to be

performed in Massachusetts. An Illinois doctor testified at

deposition that he would be willing to evaluate Kosilek for surgery

and travel to Massachusetts to operate on her if all the

appropriate arrangements, such as licensing requirements, could be

made.

The DOC's argument that Kosilek's post-operative housing

would create a security risk is more plausible than its out-of-

state-transport one, but not enough for us to think the court

clearly erred in not crediting it as a bar to surgery. Ultimately

there was evidence of viable housing options for Kosilek.

-82-
 Though the DOC claims that housing Kosilek in the general

population at MCI-Norfolk or MCI-Framingham is not feasible, the

evidence did not unequivocally support this proposition. First,

Kosilek has been housed safely while living as a woman — wearing

female clothing, using female cosmetics, and taking female hormones

that caused her to develop breasts and a feminine body shape — in

the general population at MCI-Norfolk for many years. Moreover, in

connection with Kosilek I, then Commissioner Maloney was adamant

that there were serious security concerns surrounding Kosilek

remaining at MCI-Norfolk while receiving hormones. He reasoned

that many inmates were sex offenders and a prisoner living as a

female with female attributes such as breasts would create a risk

of violence. But once an actual security review was done, then

Superintendent Spencer reported that there were no current security

concerns with Kosilek being provided estrogen therapy. And no

security issues ended up cropping up after that. The DOC's about-

face calls into question their present stance with regard to the

impact of surgery.

Second, with regard to housing Kosilek in the general

population at MCI-Framingham, there was evidentiary support for the

court's conclusion that the DOC's concerns were bogus or at least

overblown. For one, the DOC claimed men are stronger than women.

But Kosilek's perceived superior strength as a man did not jibe

with her advanced age, physically slight frame, or the fact that

-83-
she has been on female hormones for years. Moreover, the DOC's

claim that she would be an escape risk based on the weak perimeter

at MCI-Norfolk is also questionable when one considers Kosilek's

excellent disciplinary record and the fact that she would, at least

initially, be a post-operative patient. And while the DOC pointed

to Kosilek's life sentence as a factor compounding the risk of her

escape, Bissonnette testified that MCI-Framingham already housed

around forty life offenders and gave no explanation why Kosilek

should be viewed differently from these other lifers. Finally, in

rejecting the DOC's heightened flight risk contentions if Kosilek

were housed at MCI-Framingham, the court noted that according to

the DOC's classification manual, Kosilek's post-crime, one-time

pre-arrest flight (a flight embarked on twenty-three years ago)

should not even have been a consideration, even though Bissonnette

claimed it was.31

The potential for Kosilek causing inmate climate issues

at MCI-Framingham due to the fact that she murdered her wife was

another theory the DOC floated in support of its security concern

31
According to both Spencer and Bissonnette, Kosilek also
could be safely housed in their respective high-security units,
though of course that would come with trade-offs for Kosilek as far
as how restricted her life would be. Obviously Kosilek is aware of
those trade-offs and persists in her request for surgery.
There was also evidence of another scenario — an out-of-state
transfer for Kosilek. The DOC's only counter to this was the
speculative ground advanced by Clarke that there was no guarantee
another state would take or keep Kosilek. But there was no
evidence that either of these scenarios were explored or probable.

-84-
argument. But the possibility of one inmate being offensive to

another based on the crime the other inmate committed is not a new

phenomenon, and the evidence was that there were procedures in

place to deal with these types of situations at MCI-Framingham,

such as moving prisoners around. Not to mention, as Clarke

recognized, his former employer, the Washington Department of

Corrections, housed a post-operative female transgender inmate,

also serving a life sentence for murdering a female relation,

without security or climate issues.

Aside from the DOC's purported security concerns, the

court pointed to other evidence which it thought suggested the

DOC's denial of surgery was not prompted by valid penological

concerns but rather a deliberate indifference to Kosilek's medical

needs. For instance, there was evidence that the DOC did not leave

things up to chance when it sought an opinion about whether an

operation for Kosilek was even warranted. The DOC knew before it

retained Osborne that she was assisting other departments of

corrections in defending litigation filed by transgender prisoners.

In fact, Hughes specifically noted that Osborne would be more

sympathetic to the DOC's concerns and that she did not believe that

sex reassignment surgery was appropriate in the corrections

setting. It was not a stretch for the court to disbelieve

Dennehy's testimony that Osborne's very predictable opposition to

-85-
providing Kosilek with surgery did not play a role in her

selection.

The public disapproval of Kosilek's quest was another

piece of the puzzle. Even though Dennehy and Clarke denied being

motivated by avoidance of public controversy, the district court

found this testimony lacking in credibility and concluded that

Dennehy and Clarke were keenly aware of and in fact motivated by

the outcry. Evidence supporting the court's finding included

Dennehy's press appearance in the news piece featuring her senator

acquaintance who opposed the surgery; Dennehy's testimony that she

was aware that some politicians were against Kosilek being provided

with surgery and that she was generally aware of the negative news

coverage; and Clarke's admission that he received the two letters

from the seventeen unhappy state senators and twenty-five unhappy

representatives.

This evidence could be conceivably viewed as not

overwhelming in amount. However it was up to the district court to

make a credibility call, and Judge Wolf did not believe Dennehy's

or Clarke's testimony on the impact of public opinion on their

decisions. Credibility calls are something we seldom second guess

in this context. Rather we give due regard to the judge's

opportunity to assess witness credibility. Fed. R. Civ P.

52(a)(6); Monahan, 625 F.3d at 46. This deference extends to

"inferences drawn from the underlying facts, and if the trial

-86-
court's reading of the record [with respect to an actor's

motivation] is plausible, appellate review is at an end." Janeiro,

457 F.3d at 138-39 (internal quotation marks omitted) (alteration

in original). Here the district court's impression of Dennehy and

Clarke's motivations was certainly plausible. Furthermore,

adequate record support for a court's conclusion that "'deliberate

indifference' has been established — or an unreasonable

professional judgment exercised" — can exist even though that

indifference "does not rest on any established sinister motive or

'purpose' to do harm." Battista, 645 F.3d at 455. It is enough

that the district court had a reasonable basis for its perception

that the DOC had shown a pattern of "delays, poor explanations,

missteps, changes in position and rigidities." Id. And as we

chronicled above, there was ample evidentiary support for this

finding. Finding no clear error, we defer to the district court's

assessment of Dennehy's and Clarke's testimony and the other

evidence on the issue.

Finally, the DOC offered one last argument to counter the

court's finding of deliberate indifference. Besides the various

security concerns it alleged, there was a good amount of testimony

from DOC officials and experts that it is not wise to give in to

inmate threats of suicide. The fear seems to be that other inmates

will mimic Kosilek's threats of suicide in order to receive some

benefit (let's say, desirous medical treatment or a preferable

-87-
housing assignment) based on Kosilek's threats in this case. As a

general proposition we agree that the DOC should not have to yield

to an inmate's threats. However, we do not see Kosilek's

particular suicide issue as quite the concern the DOC makes it out

to be. First, the evidence was that Kosilek, who had previously

attempted both suicide and self-castration, did not manufacture a

suicide threat to game the system. Sadly, it is not unheard of for

inmates suffering from gender identity disorder to engage in self-

destructive behavior. See, e.g., Konitzer v. Frank, 711 F. Supp.

2d 874, 879 (E.D. Wis. 2010) (transgender inmate cut open and

wounded his scrotum on multiple occasions while incarcerated);

De'Lonta v. Angelone, 330 F.3d 630, 632 (4th Cir. 2003)

(transgender inmate stabbed or cut his genitals on more than twenty

occasions after the department of corrections stopped his hormone

regimen). Second, as far as other inmates are concerned, the DOC

admittedly employs competent mental health professionals who can no

doubt assess whether an inmate's threats of suicide are real or

manufactured. And if they are real, they can be addressed

appropriately on an individualized basis. Finally, and more

fundamentally, even though deterring other inmates from potentially

engaging in undesirable behavior may be a valid penal objective, it

is not a reason to withhold medical care that has been deemed

medically necessary for a particular inmate. Our ultimate

conclusion: the district court did not err in finding the security

-88-
rationale submitted by the DOC for not providing Kosilek with

surgery largely false and greatly exaggerated.

VII. CONCLUSION

We are assuredly mindful of the difficult tasks faced by

prison officials every day. But as the Supreme Court has

cautioned, while sensitivity and deference to these tasks is

warranted, "[c]ourts nevertheless must not shrink from their

obligation to 'enforce the constitutional rights of all 'persons,'

including prisoners.'" Brown v. Plata, 131 S. Ct. 1910, 1928

(2011) (quoting Cruz v. Beto, 405 U.S. 319, 321 (1972) (per

curiam)). And receiving medically necessary treatment is one of

those rights, even if that treatment strikes some as odd or

unorthodox.

Here the trial judge had the opportunity to preside over

two lawsuits involving the same players and similar allegations, to

hear evidence in this case over the course of a twenty-eight day

trial, to question witnesses, to assess credibility, to review a

large volume of exhibits, and, in general, to live with this case

for twelve years (twenty years if you count Kosilek I). The judge

was well-placed to make the factual findings he made, and there is

certainly evidentiary support for those findings. Those findings

— that Kosilek has a serious medical need for the surgery, and that

the DOC refuses to meet that need for pretextual reasons

unsupported by legitimate penological considerations — mean that

-89-
the DOC has violated Kosilek's Eighth Amendment rights. The court

did not err in granting Kosilek the injunctive relief she sought.

Affirmed.

-Dissenting Opinion Follows-

-90-
 TORRUELLA, Circuit Judge, Dissenting. Lest we lose sight

of the rule that we are called upon to enforce, stretching it

beyond the bounds of its intended purpose, it is perhaps

appropriate to begin by reciting the text of the Eighth Amendment:

"Excessive bail shall not be required, nor excessive fines imposed,

nor cruel and unusual punishments inflicted." U.S. Const. amend.

VIII. In applying any rule, precise demarcation of its reach is

both appropriate and necessary. Where these boundaries are

ignored, the results are most often unforeseen, unintended, and

unwarranted. It is only through careful attention to the

countervailing interests that prescribe the sweep of a rule that we

are best able to identify both those situations that fall clearly

within its bounds, and those complexities that skirt along its

outermost edges. Such limitations serve more than an exclusionary

purpose; by establishing a rule's proper scope, they ensure the

effectiveness of its protections when correctly and adeptly

applied.

With due respect to the majority, I am forced to dissent

because I cannot support what is, in my view, an outcome that

proceeds with little recognition of such boundaries. Instead, it

allows to stand a decision that, finding its foundation in several

erroneous assumptions, reaches a result beyond the limits of our

established Eighth Amendment jurisprudence.

-91-
 I. Kosilek's History and Treatment

Michelle Kosilek ("Kosilek") is an anatomically male

prisoner who suffers from severe gender identity disorder ("GID").

In 1992, Kosilek was convicted of murdering her32 wife and sentenced

to life imprisonment without the possibility of parole. While in

prison, Kosilek legally changed her name from "Robert" to

"Michelle" and began living as a woman to the extent possible

within a male prison environment. Kosilek has previously sought

legal redress for what she alleged were constitutional shortcomings

in the Commonwealth of Massachusetts's treatment of prisoners with

GID. See Kosilek v. Maloney, 221 F. Supp. 2d 156 (D. Mass. 2002).

Resolved in 2002, this litigation failed to substantiate any Eighth

Amendment violations but ultimately contributed to changes in the

care and treatment of GID prisoners, including Kosilek.

Today, the Massachusetts Department of Corrections

("DOC") provides Kosilek with a bevy of ameliorative measures aimed

at treating her GID. These measures, as recommended by the DOC's

medical advisors, include: psychotherapy, hormone therapy,

electrolysis for facial hair removal, and access to female clothing

and personal items (including underwear and cosmetics) such as

32
The district court adopted masculine pronouns in reference
to Kosilek's anatomical gender. See Kosilek v. Spencer, 889 F.
Supp. 2d 190 (D. Mass. 2012). Kosilek, however, self-identifies as
female and has undertaken significant efforts, including through
treatment provided by the DOC, to formalize this gender
presentation. I therefore follow the majority's practice of using
female pronouns.

-92-
those provided to inmates at MCI-Framingham, Massachusetts's only

female prison. The DOC's medical providers, Kosilek's

psychiatrist, and Kosilek herself testified as to the positive

impact these measures have had on her mental state and self-esteem.

Nonetheless, the district court, validated by the

majority in this appeal, has now ordered that the DOC provide

Kosilek with sex reassignment surgery ("SRS") to change her male

sex organs to female. According to the district court, this

surgery is the only adequate medical treatment for the serious risk

posed by Kosilek's GID; although Kosilek is not now suicidal, a

failure to provide the surgery could result in the deterioration of

her mental state and the potential for future self-harm.33

Consequently, in the district court's mind, any other treatment --

namely, the continued provision of psychotherapy, hormonal

treatments, and female attire, in addition to treating any

potential suicidality through antidepressants and increased therapy

-- is violative of the Eighth Amendment.

II. The Eighth Amendment

Fundamental to our understanding of criminal sentencing

and penological standards is the requirement that "cruel and

unusual punishments [not be] inflicted" upon those convicted of a

crime. U.S. Const. amend. VIII. In adopting this prohibition,

33
Kosilek has previously attempted suicide and self-castration
while in custody. These attempts were made prior to 1992, while
Kosilek was awaiting trial on murder charges.

-93-
"Americans . . . feared the imposition of torture and other cruel

punishments not only by judges acting beyond their lawful

authority, but also by legislatures engaged in making the laws by

which judicial authority would be measured." Ingraham v. Wright,

430 U.S. 651, 665 (1977) (citing Weems v. United States, 217 U.S.

349, 371-73 (1910)).

Later courts made apparent that the Eighth Amendment's

restrictions on criminal punishment also governed the treatment to

which prisoners were entitled when they became sick or injured

while in custody. After all, where "society takes from prisoners

the means to provide for their own needs[,] . . . [a] prison's

failure to provide sustenance [and care] for inmates may actually

produce physical 'torture or [] lingering death.'" Brown v. Plata,

131 S. Ct. 1910, 1928 (2011) (quoting Estelle v. Gamble, 429 U.S.

97, 103 (1976)). The Eighth Amendment therefore proscribes medical

care that does not rise to the level of "the evolving standards of

decency that mark the progress of a maturing society." Estelle,

429 U.S. at 102 (quoting Trop v. Dulles, 356 U.S. 86, 101 (1958)).

That appropriate medical care must be provided does not,

however, mean that inmates may seek and receive the care of their

choosing. United States v. DeCologero, 821 F.2d 39, 42 (1st Cir.

1987). Rather, this worthy pledge of protection is made

practicable through the creation of a floor below which the

standard of care must not fall. Prison officials commit no

-94-
violation so long as the medical care provided is minimally

adequate. See id.; Leavitt v. Corr. Med. Servs., Inc., 645 F.3d

484, 497 (1st Cir. 2011) (stating that an Eighth Amendment

violation occurs when the medical care provided is "so inadequate

as to constitute an unnecessary and wanton infliction of pain or

[is] repugnant to the conscience of mankind" (quoting Estelle, 429

U.S. at 105-06)). "[T]his obligation is met in full measure by the

provision of . . . services at a level reasonably commensurate with

modern medical science and of a quality acceptable within prudent

professional standards." DeCologero, 821 F.2d at 43. This limit

on the scope of the Eighth Amendment's protection is clear: care

need not be ideal, so long as it is both diligent and within the

bounds of prudence.

Neither do all instances of inadequate care constitute

constitutional violations. To substantiate a constitutional claim

there must be proof that the government was "deliberately

indifferent" to this lack of treatment. See Battista v. Clarke,

645 F.3d 449, 452 (1st Cir. 2011) (citing Farmer v. Brennan, 511

U.S. 825, 837 (1994); Estelle, 429 U.S. at 104-05). A finding of

deliberate indifference requires two showings. First, a prisoner

must prove his or her medical need is objectively serious. Mahan

v. Plymouth Cnty. House of Corr., 64 F.3d 14, 17-18 (1st Cir.

1995). A serious medical need is "one that has been diagnosed by

a physician as mandating treatment, or one that is so objectively

-95-
obvious that even a lay person would easily recognize the necessity

for a doctor's attention." Gaudreault v. Municipality of Salem,

923 F.2d 203, 208 (1st Cir. 1990) (citations omitted). Second, a

prisoner must prove subjective intent to deny care on the part of

prison officials. Farmer, 511 U.S. at 837 (reasoning that the

Eighth Amendment's prohibition of punishment means that it reaches

only the intentional provision of inadequate medical care).

Therefore, a prisoner must establish both that an official was

"aware of facts from which [an] inference [of a substantial risk of

serious harm] could be drawn" and that the official in fact drew

the inference. Id. It follows that a showing of ordinary

negligence is insufficient to establish a constitutional violation;

deliberate indifference requires a level of obstinacy akin to

criminal recklessness. Giroux v. Somerset Cnty., 178 F.3d 28, 32

(1st Cir. 1999).

A final boundary delimiting the Eighth Amendment's scope

of protection is one founded in the recognition that "security

considerations also matter at prisons . . . and administrators have

to balance conflicting demands." Battista, 645 F.3d at 454.

Therefore, deference is given to the reasonable judgments of prison

officials "so long as [those] balancing judgments are within the

realm of reason and made in good faith." Id. Although prison

officials may "forfeit[] the advantage of deference" when their

stated rationales for the rejection of medically prescribed

-96-
treatment are pretextual, id. at 455, an assessment of deliberate

indifference must still "embrace security and administration, []

not merely medical judgments." Cameron v. Tomes, 990 F.2d 14, 20

(1st Cir. 1993).

III. Clear Error Review

It is beyond argument that our standard of review in

cases such as this one falls upon a continuum, ranging from clear

error for questions of pure fact, to de novo for questions of pure

law. See, e.g., United States v. Mariano, 983 F.2d 1150, 1158 (1st

Cir. 1993). Moreover, resolving Kosilek's claim certainly requires

the careful and thorough consideration of innumerable factual

findings, including the weighing and assessment of expansive

testimony provided by both medical personnel and prison officials.

That this element of our task includes a grant of deference to the

district court is not in dispute.

Starting from this shared presumption, however, my path

quickly diverges from that of the majority. Namely, I take issue

with the majority's conclusion that all issues in this case fall

squarely on the factual end of our spectrum, and that,

consequently, clear error review applies to all elements of the

district court's decision, including its ultimate conclusions.

Recognizing, as the majority does, that our precedent is far from

crystallized on the matter, see ante at 63, I would not be so quick

to leave such legal determinations bereft of searching appellate

-97-
review.34 Neither, I believe, is that the intended result of our

standards.

At a minimum, our court should carefully apply a more

critical eye to the district court's distillation of factual

findings into legal conclusions, reviewing those ultimate

conclusions with significantly less deference. See, e.g., Hallett

v. Morgan, 296 F.3d 732, 744 (9th Cir. 2002) ("The district court's

factual findings regarding conditions at the Prison are reviewed

for clear error. However, its conclusion that the facts do not

34
While "no broader review is authorized . . . simply because
this is a constitutional case," see Maine v. Taylor, 477 U.S. 131,
145 (1986), I believe that where such rights are implicated in
cases presenting closely intertwined questions of law and fact our
court would be wise to tread carefully before applying, in toto, a
clearly erroneous standard of review. Although the Supreme Court
has rejected the application of a higher standard of review in
constitutional cases where the question at hand was one purely of
"historical fact," the Court has not expressly foreclosed
heightened review to other questions involving "legal, as well as
factual, elements." Hernandez v. New York, 500 U.S. 352, 366-67
(1991). The majority makes much of the fact that I have previously
explained the existence of a circuit split -- in which the First
Circuit has not taken a side -- regarding whether heightened review
of underlying facts would appropriately apply in cases in which a
right is protected below. United States v. Frabizio, 459 F.3d 80,
96 (1st Cir. 2006) (Torruella, J., concurring). Critically,
however, I do not now sound a call for plenary review of what are
wholly factual findings. Cf. Bose Corp. v. Consumers Union of
U.S., Inc., 466 U.S. 485, 499 (1984). Rather, I intend a more
general point, cautioning that where factual determinations and
constitutional standards are closely related we ought to carefully
ensure that such subordinate determinations do not erroneously cast
the die of our legal conclusions. After all, our ultimate
conclusions derive unquestionably from legally operative standards,
and it is undoubtedly the duty of appellate courts to "to clarify[]
[such] legal principles." Ornelas v. United States, 517 U.S. 690,
697 (1996).

-98-
demonstrate an Eighth Amendment violation is a question of law that

we review de novo." (citing Campbell v. Wood, 18 F.3d 662, 681 (9th

Cir. 1994) (en banc)));35 Alberti v. Klevenhagen, 790 F.2d 1220,

1225 (5th Cir. 1986) ("[O]nce the facts are established, the issue

of whether these facts constitute a violation of constitutional

rights is a question of law that may be assayed anew upon

appeal."). Any deference must also admit of exception where the

trial court bases its findings on an "erroneous interpretation of

the standard to be applied," Vinick v. United States, 205 F.3d 1,

7 (1st Cir. 2000) (quotation marks omitted) (quoting United States

v. Parke, Davis & Co., 362 U.S. 29, 44 (1960), for even under

deferential review we have a duty to "look carefully . . . to

detect infection from legal error," Sweeney v. Bd. of Trs. of Keene

State Coll., 604 F.2d 106, 109 n.2.

While recognizing that the delineation between questions

of law and fact is often less than pristine, see, e.g., Miller v.

Fenton, 474 U.S. 104, 113-14 (1985), the inherent difficulty in our

task cannot lead to the abdication of our responsibility to

identify and strenuously review a district court's conclusions of

law, even where those conclusions are not easy to parse from their

35
The majority is correct to note that the Ninth Circuit has
found the question of adequate medical alternatives to be one of
fact. Snow v. McDaniel, 681 F.3d 978, 988 (9th Cir. 2012). Yet,
Ninth Circuit practice includes plenary review of a district
court's eventual Eighth Amendment holding. Thus, I seriously
question the majority's proclamation that our sister circuit "takes
a similar approach" to the one they now advocate.

-99-
factual underpinnings. I cannot agree, therefore, with the

majority's failure to undertake any inquiry more searching than

that provided by clear error review.36

IV. The District Court's Conclusions

In its review, the district court undertook to answer

five distinct questions which, when answered in the affirmative, it

found substantiated a constitutional violation deserving of remedy.

These were: (1) whether Kosilek had a serious medical need; (2)

whether SRS was the only adequate treatment for that need; (3)

whether the DOC knew Kosilek was at high risk of serious medical

harm absent SRS; (4) whether the DOC's denial of treatment was made

in bad faith or for pretextual reasons; and (5) whether the DOC's

conduct, if found to be unconstitutional, would continue in the

future.

In my view, by parsing the issue into such discrete,

hermetic questions, the district court's opinion artfully shielded

from review the complex and oft-interrelated nature of our Eighth

36
The majority notes that our court has previously upheld a
finding of deliberate indifference where the district court had a
"reasonable basis" for its finding. Battista, 645 F.3d at 455.
That same opinion, however, made clear that a finding of deliberate
indifference was appropriately "reviewed on appeal more closely
than [] district court fact-finding." Id. at 454 (citations
omitted). While the majority admits of this subtlety, see ante at
64, its review then appears to abdicate such nuance and apply
maximum deference throughout. In any case, where, as here, the
district court's determinations were infected by various errors as
described below, I believe the majority presumes too much regarding
their reasonableness.

-100-
Amendment inquiry. See Leavitt, 645 F.3d at 498 ("[T]he subjective

deliberate indifference inquiry may overlap with the objective

serious medical need determination; similar evidence, including

evidence of adverse effects, may be relevant to both components."

(internal quotation marks and citation omitted)); DesRosiers v.

Moran, 949 F.2d 15, 18-19 (1st Cir. 1991) (recognizing that "[i]n

practice" the objective and subjective components of our deliberate

indifference standards "may overlap or merge"). In treating

Kosilek's contentions, therefore, I adopt our court's past practice

of assessing the district court's "several subordinate findings" in

a more holistic manner.37 Battista, 645 F.3d at 452. This approach

gives due recognition to the fact that "[m]edical need' in real

life is an elastic term," id. at 454, and acknowledges that any

determination of a treatment's adequacy must carefully balance the

many competing concerns faced by prison officials. I begin with a

discussion of the district court's errors, as I see them.

A. Prudent Medical Care

The district court faced a question about the practice of

prudent medical professionals that, at its crux, hinged on whether

the DOC's preferred treatment plan -- advocated by Dr. Schmidt --

was a medically adequate response to Kosilek's GID. Ultimately,

37
That GID is a serious medical condition is not contested.
Further, the fifth question need only be reached upon establishing
a constitutional violation, for the purposes of crafting a remedy.
Therefore, I focus on only the second, third, and fourth questions
presented.

-101-
the district court, in a decision now upheld by the majority,

determined that Dr. Schmidt was not a prudent professional, based

largely on his statements of equivocation regarding use of the

Harry Benjamin Standards of Care (the "Standards of Care" or the

"Standards"). Indeed, Dr. Schmidt testified that he viewed the

Standards as "guidelines." He also made clear that he found the

protocol laid out on the Standards of Care "very useful for

patients," and that he "referr[ed] [patients] to the protocol and

ask[ed] them to become familiar with them." As to SRS, Dr. Schimdt

stated that he "neither advocate[s] for nor . . . speak[s] against

the decision[]." Instead, he "leaves[s] the decision-making in the

hands of the patients." This is far from what the district court,

now affirmed by the majority, characterizes as an outright

rejection of the Standards' applicability.

Still, the district court took particular issue with

Dr. Schmidt's practice of not writing letters of recommendation for

patients seeking surgery. The district court's concern was

predicated on its belief that letters of recommendation are

required by the Standards of Care, and "[a]ccordingly, prudent

professionals . . . write such letters." Kosilek, 889 F. Supp. 2d

at 233. This reasoning contains an inferential leap. That there

is a predicate requirement to a medical procedure does not lead

inexorably to the conclusion that prudence mandates assisting

patients to meet that requirement. For instance, if a surgery

-102-
could not be conducted on an individual under a particular age

without several letters of recommendation, a medical professional

who refused to write such a letter based on their understanding of

that treatment's appropriateness for youths would not be,

necessarily, imprudent.

In affirming the district court's finding that

Dr. Schmidt was not prudent, the majority also assigns significant

weight to the fact that, despite having treated approximately 300

individuals with GID, Dr. Schmidt does not appear to believe

surgery was ever "medically necessary." The majority disparages

this belief as clearly contrary to the Standards of Care, and

therefore clearly imprudent. Yet, again, Dr. Schmidt admits to

using the standards for guidance and to "maintain[ing] a neutral

position" on surgery. At the request of his patients, he also

released medical files to surgeons and wrote letters indicating,

where appropriate, that there were no contra-indications to

surgery. His testimony regarding his disagreement that surgery was

medically necessary stemmed from his belief that patients

exhibiting particularly high levels of distress often suffer from

co-morbid conditions that require treatment in their own right.

Moreover, the Standards of Care themselves admit of just

this sort of flexible application, not simply strict adherence.

The first page of the Standards of Care states unequivocally that,

"[t]he Standards of Care are Clinical Guidelines;" it continues on

-103-
to make clear that the Standards are "intended to provide flexible

directions" and that "[a]ll readers should be aware of the

limitations of knowledge in the area." Standards of Care at 1

(emphasis added). But see Kosilek, 889 F. Supp. 2d at 236 (citing

O'Donnabhain v. Comm'r, 134 T.C. 34, 45 (U.S. Tax Ct. 2010))

(relying on O'Donnabhain's rejection of any characterization of the

Standards of Care as "guidelines" as imprudent). The Standards of

Care further provide that "[i]ndividual professionals and organized

programs may modify them." Standards of Care at 2. This much was

made clear in Dr. Levine's testimony:

[T]he "Standards of Care" was a consensus
document from people from seven different
countries or something, you know, who come
from different systems, and it was a political
process that forged together a set of
standards . . . . So "prudent" is a wonderful
word, but it's not like it has one simple
definition.

In fact, Dr. Levine, who was an independent expert hired

by the district court, expressly stated in his initial report that,

while not popular, Dr. Schmidt's view was within "prudent

professional standards." In its opinion, however, the district

court took significant pains to recast this finding, dismissing it

as erroneous based on Dr. Levine's purported refusal to testify, at

least initially, as to how a medically prudent professional would

act if all countervailing interests were set aside. In other

words, the court required Dr. Levine to presume that a patient had

fully met all the readiness criteria in the Standards of Care and

-104-
faced no other extrinsic obstacles to surgery (such as money,

safety, or external pressure). The district court then hung its

hat on the fact that, "[e]liminating these considerations and any

security concerns, Dr. Levine opined that a prudent professional

would not deny Kosilek sex reassignment surgery." Kosilek, 889 F.

Supp. 2d at 235 (emphasis added).

Medical prudence, however, does not exist in a bubble,

and a standard of minimal adequacy must inherently admit of

conditions that are less than ideal. See Rhodes v. Chapman, 452

U.S. 337, 367 (1981) (Brennan, J., concurring) ("[C]ases are not

decided in the abstract. A court is under the obligation to

examine the actual effect of challenged conditions . . . ."). In

fact, Dr. Levine's testimony recognizes just such nuance -- even if

the district court's reading of it does not. Dr. Levine stated

that while prudent professionals would not deny SRS to eligible

individuals, "life, [and] reality" sometimes would. In those

instances, prudent professionals "bring to bear" the same methods

described by Dr. Schmidt to otherwise alleviate the individual's

symptoms of GID.38

38
In my reading, Dr. Schmidt never counseled for denying
surgery. His testimony suggested deference to a patient's choice
and willingness to release medical records to qualified surgeons.
He then expressed concern as to whether a prisoner could ever meet
readiness criteria for surgery, noting disagreement with the
district court's presumption that a real-life experience could
necessarily occur behind bars. Although admonishing Dr. Schmidt
for purportedly ignoring the Standards of Care, the district court
discredited his testimony in part based on this expression of

-105-
 Nonetheless, the district court suggested that portions

of Dr. Levine's testimony might be properly "disregard[ed]" based

on the purported change in his opinion. Kosilek, 889 F. Supp. 2d

at 234 n.15. I see no merit in this assertion, and moreover

believe that it evidences the district court's troublesome practice

of rejecting testimony -- even the testimony of an impartial,

court-appointed expert -- where it explored the very real nuances

implicit in defining prudent care. For one, the reason for any

such "change" is clearly evidenced in the record: the district

court demanded it. For another, Dr. Levine's testimony, even after

he was admonished for undertaking an assessment recognizing the

realities in which GID patients live, was not inconsistent.

He began, in his written report, by stating that

Dr. Schmidt's method, while not preferred, was prudent. In later

testimony, after specifically predicating his statement with an

acknowledgment that "the 'Standards of Care' [] have to be

interpreted . . . by the life of the environment in which Michelle

Kosilek is going to live," Dr. Levine again concluded that Dr.

Schmidt's proposed treatment was not "imprudent." The following

exchange then occurred:

concern regarding whether a real-life experience -- a key component
of those Standards -- could occur in prison. See Kosilek, 889 F.
Supp. 2d at 235. In fact, in combination with the district court's
insistence that Dr. Levine ignore questions regarding this real-
life experience and instead presume that the experience necessarily
can and did occur, this suggests a purposeful tipping of the
testimonial scales away from an area of potentially worthy inquiry.

-106-
 THE COURT: But is this an area in which you
think prudent professionals can reasonably
differ as to what is at least minimally
adequate treatment for this condition?
[DR. LEVINE]: Yes, and do.

Therefore, in addition to finding no internal inconsistencies in

Dr. Levine's expert testimony, see Mitchell v. United States, 141

F.3d 8, 17 (1st Cir. 1998), I disagree that he ever testified to

Dr. Schmidt's proposed method of care being outside professional

standards. At his most negative, Dr. Levine stated that

Dr. Schmidt's proposal would be "uncompassionate" and "unpopular."

At his most mincing, he referred to Dr. Schmidt's proposed method

of care as "not exactly imprudent."39 As such, the district court's

proffered conclusion that Dr. Levine found Dr. Schmidt's proposal

unreasonable is unsupported by the record. Insofar as the majority

now affirms the same, erroneous reading, I find their conclusion to

be equally flawed.

B. Serious Risk

It is undisputed that surgery for Kosilek would be an

appropriate option for treating her GID. This fact is far from

determinative, however, of whether a choice not to provide the

surgery gives rise to a deprivation of constitutional magnitude.

See Cameron, 990 F.2d at 20 (finding that prison officials are not

39
This admittedly unenthusiastic endorsement referenced
Dr. Schmidt's proposed treatment presuming that there was a total
absence of countervailing factors to consider in developing a
treatment plan.

-107-
"bound to do what the doctors say is best . . . even if the doctors

are unanimous"). If an alternative short of surgery is still

sufficient to address, with minimal adequacy, Kosilek's medical

need, no constitutional claim can arise. See DeCologero, 821 F.2d

at 43.

The district court reasoned, however, that any treatment

except surgery is necessarily inadequate, given that Kosilek's

medical providers testified to a likelihood that a denial of

surgery would significantly increase Kosilek's risk of severe

emotional distress, potentially manifesting in self-harm. Moreover

both the district court and majority rejected, as violative of the

Eighth Amendment, the DOC's plan to treat any symptoms of

heightened distress and suicide ideation with additional

psychotherapy and the possible use of antidepressants.

This conclusion rests on an artful -- and in my mind

erroneous -- compartmentalization of the DOC's preferred treatment

plan. The district court seeks to draw a clear line between the

cause of Kosilek's distress (GID) and her symptoms (emotional

distress and possible suicide ideation).40 In support of the same

reasoning, the majority cites a Seventh Circuit case for the

40
It is rare, to my understanding, that medical treatments may
so neatly and completely delineate between symptom and disease.
Certainly, were a patient to present with signs of both obesity and
severe hypertension, it is an uncommon doctor that would disparage
a peer for prescribing blood-pressure medication, although designed
no doubt to treat a symptom.

-108-
proposition that "psychotherapy as well as antipsychotics and

antidepressants . . . do nothing to treat the underlying disorder

[of GID]." See Fields v. Smith, 653 F.3d 550, 556 (7th Cir. 2011).

Critically, however, in Fields the prisoner was denied any hormonal

treatment, meaning that the court was called on to resolve a

question of whether psychotherapy and antidepressants alone could

sufficiently treat GID. In contrast, here the question was whether

the continued provision of all ameliorative measures currently

afforded Kosilek in addition to antidepressants and psychotherapy

would be constitutionally adequate.

Indeed, the DOC's proposed method of treating Kosilek's

distress and desire to self-harm cannot be assessed piecemeal, but

must be addressed in light of Kosilek's entire course of treatment.

Were surgery not provided, the provision of psychotherapy,

hormones, electrolysis, and female clothing and cosmetics would

continue to represent a very real and direct treatment for

Kosilek's GID. Moreover, although she remains distressed, Kosilek

admits that the DOC's current treatment regimen has led to a

significant stabilization in her mental state. Kosilek's doctors

testified to the same, highlighting her "joy around being

feminized." This claim is also borne out by the long passage of

time since she exhibited symptoms of suicide ideation or attempted

to self-castrate. The provision of additional, supplemental care

specifically targeting her risk of suicide cannot, in my reading,

-109-
render that treatment, which has successfully mitigated her

symptoms for nearly a decade, suddenly inadequate.41

What is clear from the record is that the DOC has

provided Kosilek with care sufficient to decrease her levels of

distress and manage her desires to self-harm. On the whole, this

suggested course of treatment appears tailored to Kosilek's current

symptoms and adequately prepared to address her future ones in a

manner that is in no way "so inadequate as to shock the

conscience." Torraco v. Maloney, 923 F.2d 231, 235 (1st Cir. 1991)

(internal quotation marks omitted) (quoting Sires v. Berman, 834

F.2d 9, 13 (1st Cir. 1987); see also DeCologero, 821 F.2d at 43

(finding that care is adequate where it is "reasonably commensurate

with modern medical science").

C. Security Concerns

In issues of security, "[p]rison administrators . . .

should be accorded wide-ranging deference in the adoption and

41
The Massachusetts DOC has recently undertaken a significant
effort to ensure it is well-prepared to address the needs of
prisoners exhibiting symptoms of suicidality. See Disability Law
Ctr. v. Mass. Dep't of Corr., C.A. No. 07-10463-MLW, 2012 WL
1237760 (D. Mass. Apr. 12, 2012). I see nothing to suggest such
care, if provided, would not itself be thorough and adequate.
Moreover, it bears consideration that Kosilek is not currently
suicidal and that although her medical providers suggest a
likelihood that suicide ideation will reemerge if SRS is not
provided, there is no indication as to the severity, duration, or
even sole causal factors of this potential result. As Dr. Levine
testified, the presumption that Kosilek may become suicidal must
also recognize the potential that this impulse is not stagnant, but
might naturally -- and with the assistance of therapy -- dissipate
or "evolve over time."

-110-
execution of policies and practices that in their judgment are

needed to preserve internal order and discipline and to maintain

institutional security." Bell v. Wolfish, 441 U.S. 520, 547

(1979). Although we cannot "abdicate our responsibility to ensure

that the limits imposed by the Constitution are not ignored,"

Blackburn v. Snow, 771 F.2d 556, 562 (1st Cir. 1985), we do not sit

to substitute our own judgment for that of prison administrators.

Nonetheless, believing that the DOC's cited security concerns were

mere pretext, the district court declined to afford them weight.

Kosilek, 889 F. Supp. 2d at 247. The majority affirmed this

denial; a decision that I believe ignores the very real security

issues presented by the DOC.

That various security concerns might arise in the context

of a prison setting in which a post-operative, male-to-female

transsexual is housed with male prisoners takes no great stretch of

the imagination.42 In fact, nearly every case to consider the

provision of medical care to prisoners at some point relies on the

Supreme Court's 1994 decision in Farmer v. Brennan, 511 U.S. 825

(1994). In that case, an Eighth Amendment claim was predicated on

42
I find the DOC's concerns regarding Kosilek's post-operative
housing significant. I do not, however, dispute the district
court's finding, affirmed by the majority, that any security
concerns regarding Kosilek's ability to escape custody while being
transported for surgery are, at best, extremely minimal. Prison
officials have significant experience transporting prisoners, and
both Kosilek's age and history of good behavior counsel in favor of
safe transport.

-111-
prison officials' failure to provide (in part through segregation)

for the safety of a pre-operative transsexual. Id. at 847. The

court reasoned that by knowingly allowing the petitioner to remain

in general custody at a male prison despite his feminine body

shape, clothing, and slight stature, prison officials could

illustrate the sort of subjective indifference necessary to sustain

an Eighth Amendment claim. Id.

Despite the obviousness of such risks, the majority

reasons that no clear error occurred, in part because the DOC's

security review of MCI-Norfolk was started, completed, and

submitted in a matter of weeks. With speed, it suggests, comes

inadequacy. I am not so ready to adopt that presumption. The

record shows that all involved parties met for the first time on

May 19, 2005, to discuss a report that was due by May 27, 2005.

That this was their first meeting, however, does not necessarily

mean that it was the first instance in which the various

individuals considered the issues and questions implicit in

ensuring a safe environment for prisoners undergoing treatment for

GID.

The district court and the majority also highlight the

fact that experts retained by the DOC were not wholly knowledgeable

about Kosilek's personal characteristics, such as age and record of

good behavior. Although this shortcoming in their knowledge was

not ideal, I cannot credit any presumption that these lapses

-112-
rendered the experts unable or unqualified to speak to the general

security concerns created by housing a post-operative transsexual

in a prison's general population. Kosilek's record of good

behavior, for instance, has no bearing on an assessment of whether

other prisoners might threaten or harm her based on her post-

operative anatomy and gender presentation.

Further, in reaching its conclusion the district court

stated that "the DOC [could] reasonably assure the safety of

Kosilek and others after sex reassignment surgery by housing

Kosilek in a segregated protective custody unit." Kosilek, 889 F.

Supp. 2d at 243. Yet, the court also warned that "it may

foreseeably be argued that keeping Kosilek in segregation is

unnecessary and a form of extrajudicial punishment that is

prohibited by the Eighth Amendment." Id. at 245. The tension

between these statements is clear, and the district court's proffer

that we disregard security concerns based on the existence of a

possibility for segregated housing appears unreasonable when, in

short turn, they assert that such a course of action would violate

the Constitution.

The majority defends the district court's determination

in part by noting that Kosilek may continue to be housed in MCI-

Norfolk's general population where no security issues have arisen

during her tenure. The fact that no such issues have arisen in the

past, however, does not necessarily render inappropriate or

-113-
unreasonable the DOC's concerns that issues might present

themselves in Kosilek's post-operative future. Certainly, courts

cannot and should not strip from prison officials the ability to

consider and implement prophylactic solutions to foreseeable issues

reasonably within the scope of their security expertise. In fact,

such a retroactive style of administration would, in itself, seem

to amount to just the sort of indifference to credible threats of

harm that might constitute a constitutional violation. See Cortés-

Quiñones v. Jiménez-Nettleship, 842 F.2d 556, 558 (1st Cir. 1988)

(stating that prison officials have a duty to take reasonable

measures to protect prisoners from harm).

Ultimately, in a feat of conclusory reasoning, the

district court overlooked the legitimacy of the DOC's concern based

on its belief that the decision to deny SRS was a response to

"public and political criticism."43 Kosilek, 889 F. Supp. 2d at

240. The evidence on record tending to support this theory

includes a press appearance by Commissioner Dennehy, negative news

coverage regarding Kosilek's request for surgery, and letters

received by the DOC from members of the Massachusetts legislature.

43
Perhaps cognizant of the inferential leap made by the
district court, see ante at 85, the majority places greater
emphasis on other rationales mentioned by the district court. The
district court's opinion, however, makes clear that its conclusion
rested predominantly on concern about public opinion. Kosilek, 889
F. Supp. 2d at 240 ("[T]he defendant has refused to provide . . .
[SRS] in order to avoid public and political criticism. This not
a legitimate penological purpose. Therefore, the defendant's
conduct . . . violates the Eighth Amendment.").

-114-
Surely, this evidence provides ample support for the fact that

public criticism existed and was leveled at the DOC, both by the

media and politicians. It in no way, however, proves the DOC's

reasons for denying Kosilek's request or shows that this denial was

motivated specifically by the public outcry.

In any case, even if the district court's finding that

public criticism played a role in shaping the DOC's decision is

accepted wholesale,44 this finding might at most counsel for the DOC

to lose "the advantage of deference." Battista, 645 F.3d at 455

(emphasis added). It cannot, however, suddenly render superfluous

the very real concerns the DOC expressed about housing Kosilek

after her operation. I find no license in the record for the

district court to have wholly dismissed the validity of these

concerns.

V. Kosilek's Eighth Amendment Claim

Having set forth my disagreements with the district

court's conclusions regarding the scope of medical prudence, the

potential for adequate treatment short of surgery, and the DOC's

security concerns, I turn to the task of determining whether

Kosilek has proven deliberate indifference to a serious medical

need. Cameron, 990 F.2d at 20 ("Indeed, when it comes to

44
As the majority notes, credibility determinations of this
type are given particular deference by our court. See ante at 85-
86. Thus, while I see no extrinsic support in the record, I
recognize I cannot equal the district court's ability to hear and
weigh testimony.

-115-
constitutional rights, none of the professionals has the last word.

Professional judgment, as the Supreme Court has explained, creates

only a 'presumption' of correctness; welcome or not, the final

responsibility belongs to the courts." (citing Youngberg v. Romeo,

457 U.S. 307, 323 (1982)). As a starting point, this review must

embrace the many competing concerns, including those relevant to

prison administration, that are inherent in our constitutional

inquiry, for there is "[n]othing in the Constitution [that]

mechanically gives controlling weight to one set of professional

judgments." Id.

What is clear is that the DOC has, for several years,

provided Kosilek with significant treatment for her GID. Equally

clear is that this treatment has resulted in marked improvement in

Kosilek's mental state and contentment. She is not currently

suicidal, and all reported instances of self-harm occurred two

decades ago, long prior to her current course of treatment. The

DOC also stands prepared to offer additional psychiatric services

should Kosilek begin exhibiting signs of suicidality. I can see no

violation on these facts. Not performing surgery is not the most

compassionate solution to Kosilek's GID. Neither, however, does it

fall outside the scope of clear professional standards, DeCologero,

821 F.2d at 43, or illustrate severe obstinacy and disregard of

Kosilek's medical needs, DesRosiers, 949 F.2d at 19 ("[T]he

-116-
complainant must prove that the defendants had a culpable state of

mind and intended wantonly to inflict pain." (citations omitted)).

Kosilek is receiving, and would continue to receive, a

regimen of treatment that mitigates the severity of her GID. This

treatment is far from the proverbial "aspirin" doled out to a

cancer patient in lieu of chemotherapy. See ante at 71. Rather,

the DOC has for years ensured an individualized treatment plan for

her physical and mental needs as well as consistent access to a

team of specialists. I do not see in this treatment, nor does the

district court or majority make clear, any "reasonable basis," see

ante at 66, for a finding of wanton disregard. Giroux, 178 F.3d at

32 (requiring a level of "excessive risk" like that of criminal

recklessness); DesRosiers, 949 F.2d at 19. Rather, giving due

consideration to countervailing security concerns and based on a

review of the record that shows the DOC's proposed care was not

outside the realm of professionalism, I cannot say that the DOC has

failed to adequately care for Kosilek's GID or callously ignored

her pain. Feeney v. Corr. Med. Servs., Inc., 464 F.3d 158, 162

(1st Cir. 2006) ("The care provided must have been 'so inadequate

as to shock the conscience.'" (quoting Torraco, 923 F.2d at 235)).

Facing litigation that was equally protracted and

passionate, the district court's task was by no means a simple one.

The complexities of this case were many, and the testimony

considerable. I am convinced, however, that the district court

-117-
ultimately erred in several key respects, skewing its factual

conclusions towards a result, now upheld by the majority, that is

beyond the boundaries of our accepted legal precedent.

The Eighth Amendment proscribes punishment, including

punishment in the form of medical care so unconscionable as to fall

below society's minimum standards of decency. See Wilson v.

Seiter, 501 U.S. 294, 300 (1991); Estelle, 429 U.S. at 102. Its

boundary simply does not reach, however, to instances of care that,

although not ideal, illustrate neither an intent to harm nor the

obstinate and unwarranted application of clearly imprudent care.

Respectfully, I would reverse.

-118-