Reiyn Keohane v. Florida Department of Corrections Secretary

USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 1 of 46

IN THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
________________________

No. 18-14096
________________________

D.C. Docket No. 4:16-cv-00511-MW-CAS

REIYN KEOHANE,

Plaintiff - Appellee,

versus

FLORIDA DEPARTMENT OF CORRECTIONS SECRETARY,

Defendant - Appellant.

________________________

Appeal from the United States District Court
for the Northern District of Florida
________________________

Before WILLIAM PRYOR, Chief Judge, WILSON, MARTIN, JORDAN,
ROSENBAUM, JILL PRYOR, NEWSOM, BRANCH, GRANT, LUCK, LAGOA,
and BRASHER, Circuit Judges.

BY THE COURT:

A petition for rehearing having been filed and a member of this Court in

active service having requested a poll on whether this appeal should be reheard by

the Court sitting en banc, and a majority of the judges in active service on this
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 2 of 46

Court having voted against granting rehearing en banc, it is ORDERED that this

appeal will not be reheard en banc.

2
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 3 of 46

WILLIAM PRYOR, Chief Judge, joined by BRANCH, Circuit Judge, statement
respecting the denial of rehearing en banc:

I voted with the majority not to rehear this appeal en banc. I write separately

to explain why my dissenting colleague is wrong to assert that a grant of en banc

review is somehow objectively “demand[ed]” or is “an obligation,” Dissenting Op.

at 23, 45–46, in this appeal or any other. No statute, precedent, rule, or internal

operating procedure imposes such an obligation. The decision to grant en banc

review is always discretionary and disfavored.

No source of law obligates us to hear any appeal en banc. To be sure, a

statute grants us the authority to hear appeals en banc. See 28 U.S.C. § 46(c). And

a rule elucidates some procedural aspects of en banc review. See Fed. R. App. P.

35. We have added details of our own. See 11th Cir. R. 35-1–35-10; Fed. R. App.

P. 35, IOP 1–9. But none of those rules requires us to hear any appeals en banc.

Precedent points in the same direction. The Supreme Court long ago

explained that the statute permitting en banc review “vests in the court[s of

appeals] the power to order hearings en banc.” W. Pac. R.R. Case, 345 U.S. 247,

250 (1953). But “[i]t goes no further. It neither forbids nor requires each active

member of a Court of Appeals to entertain each petition for a hearing or rehearing

en banc.” Id. Ten years later, the Supreme Court reaffirmed this view: “the rights

of the litigant go no further than the right to know the administrative machinery

3
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 4 of 46

that will be followed and the right to suggest that the en banc procedure be set in

motion in his case.” Shenker v. Balt. & Ohio R.R. Co., 374 U.S. 1, 5 (1963). And

more recently, the Supreme Court acknowledged yet again that “[r]ehearing [e]n

banc is a discretionary procedure employed only to address questions of

exceptional importance or to maintain uniformity among Circuit decisions.”

Missouri v. Jenkins, 495 U.S. 33, 46 n.14 (1990) (emphasis added); see also Bryan

A. Garner et al., The Law of Judicial Precedent § 61, at 496 (2016) (“The decision

to grant a petition for hearing or rehearing en banc, or to initiate en banc review on

the court’s own motion, is discretionary.”); 16AA Charles A. Wright et al., Federal

Practice and Procedure § 3981.1, at 496 (5th ed. 2020) (“Consideration en banc

rests in the discretion of the court of appeals.”). The Supreme Court has described

this process as “essentially a policy decision of judicial administration.” Moody v.

Albemarle Paper Co., 417 U.S. 622, 627 (1974).

The grant of en banc review is and should be rare. The Federal Rules of

Appellate Procedure say so: “An en banc hearing or rehearing is not favored . . . .”

Fed. R. App. P. 35(a). Practical considerations confirm why: “[T]he institutional

cost of rehearing cases en banc is extraordinary. . . . It is an enormous distraction

to break into [our regular] schedule and tie up the entire court to hear one case en

banc.” Bartlett ex rel. Neuman v. Bowen, 824 F.2d 1240, 1243 (D.C. Cir. 1987)

(Edwards, J., concurring in denial of rehearing en banc). After all, a panel of three

4
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 5 of 46

judges has already spent considerable resources deciding the appeal once. For that

reason, we and our sister circuits have said again and again that the “heavy

artillery” of en banc review should be used rarely. United States v. Hogan, 986

F.2d 1364, 1369 (11th Cir. 1993); see, e.g., Mitts v. Bagley, 626 F.3d 366, 369–71

(6th Cir. 2010) (Sutton, J., concurring in denial of rehearing en banc); Kane County

v. United States, 950 F.3d 1323, 1324 (10th Cir. 2020) (Phillips, J., concurring in

denial of rehearing en banc); Church of Scientology of Cal. v. Foley, 640 F.2d

1335, 1339–42 (D.C. Cir. 1981) (en banc) (Robinson, J., dissenting).

Because en banc review is both discretionary and disfavored, reasonable

minds can differ about whether it is appropriate in a particular case. Indeed, the

problem of deciding whether to grant en banc review is evergreen; a judge

wrestling with the decision decades ago remarked that sometimes “one judge’s

case of ‘exceptional importance’ is another judge’s ‘routine or run-of-the-mill’

case.” Bartlett, 824 F.2d at 1242 (Edwards, J., concurring in denial of rehearing en

banc). Judges can reasonably disagree about the best way to allocate our judicial

resources. And, of course, I never take any colleague’s disagreement personally.

Cf. Dissenting Op. at 24 n.1. For the same reason, disagreements about whether to

grant rehearing do not warrant attacks on the integrity of judges or their

commitment to the rule of law nor, good grief, on the legitimacy of this Court. See

id. at 23–24 & n.1, 28–29, 41–42, 45.

5
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 6 of 46

NEWSOM, Circuit Judge, joined by LUCK, Circuit Judge, concurring in the denial

of rehearing en banc:

I offer the following pre-buttal to Judge Rosenbaum’s dissent from the

denial of rehearing en banc.

Before jumping into the merits, let me say this by way of introduction:

More often than not, any writing’s persuasive value is inversely proportional to its

use of hyperbole and invective. And so it is with today’s dissental—which, rather

than characterizing, I’ll let speak for itself.1 Among other things, the dissental

accuses me—as the author of the panel opinion—of “inaccurately purport[ing]”

(and alternatively “claiming”) “to apply the governing prior precedent” in Thomas

v. Bryant, 614 F.3d 1288 (11th Cir. 2010), “reimagin[ing]” Thomas’s holding,

construing Thomas “as [I] pleased,” “pretending” that Thomas sanctioned a

standard of appellate review that it “demonstrably did not,” “distort[ing] beyond

recognition” this Court’s prior-panel-precedent rule and “remold[ing]” it into an

“unrecognizable and dangerous form,” and now, in this opinion, of engaging in

“distraction tactics.” Rosenbaum Dissenting Op. at 23, 24, 26, 28, 30, 32, 42, 43,

44.

1
For the most part, I’ll use the term “dissental” to refer to Judge Rosenbaum’s dissent from the
denial of rehearing en banc, thereby distinguishing it from Judge Wilson’s panel-stage dissent.
See Alex Kozinski & James Burnham, I Say Dissental, You Say Concurral, 121 Yale L.J. Online
601 (2012).

6
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 7 of 46

And there’s so much more where that came from. The dissental saves its

most biting criticism—and its most soaring rhetoric—for the seven judges who

voted against rehearing. All of us, the dissental not so subtly implies, cast our

votes simply because we “agree[d] . . . with the ultimate outcome” of the panel

opinion. Id. at 24. In declining to rehear the case, the dissental charges, we have

blessed a “rogue interpretation of the prior-precedent rule,” sanctioned a “critical

threat to the stability and predictability of the law,” and thereby unleashed

“potentially devastating consequences.” Id. at 23, 45.

Strong words. Not a one of them true. Allow me to turn down the volume

and provide a little perspective.

I

I begin with a brief factual summary.

Reiyn Keohane is a Florida inmate currently serving a 15-year sentence for

attempted murder. Keohane was born anatomically male, but she began to identify

as female sometime during her preadolescent years. Beginning at age 14—and up

until the time she was incarcerated at 19—Keohane wore women’s clothing,

makeup, and hairstyles. At 16, she was formally diagnosed with gender dysphoria.

About six weeks before the arrest that eventually landed her in prison, Keohane

began hormone therapy under the care of a pediatric endocrinologist. See Keohane

v. Fla. Dep’t of Corr. Sec’y, 952 F.3d 1257, 1262 (11th Cir. 2020).

7
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 8 of 46

Upon her incarceration in a Florida prison, Keohane requested (as relevant

here) two forms of treatment. First, she sought to continue hormone therapy. That

request was initially denied for reasons that have no real bearing on my colleague’s

dissental and that I therefore won’t belabor. See id. at 1262–63. In short, though,

not long after Keohane filed suit, the Florida Department of Corrections reversed

course and referred Keohane to an outside endocrinologist, who immediately

prescribed her hormone therapy. See id. at 1263. The panel majority held that the

FDC’s decision to grant Keohane’s hormone-therapy request mooted her challenge

to the initial denial of that treatment. See id. at 1270–72. The dissental doesn’t

take issue with the panel’s mootness determination, so for present purposes we can

leave the hormone-therapy requests to the side. See Rosenbaum Dissenting Op. at

30 n.3.

Second, and separately, Keohane requested the ability to engage in “social

transitioning”—in particular, she asked to wear female undergarments and

makeup, and to grow out her hair in a long, feminine style. Keohane, 952 F.3d at

1263. The FDC refused Keohane’s social-transitioning requests on the grounds

that they violated prison policy—which required male inmates to wear “[u]nder

shorts” and to “have their hair cut short to medium uniform length at all times with

no part of the ear or collar covered,” Fla. Admin. Code r. 33-602.101(2), (4)—and

that they posed a security risk. Keohane, 952 F.3d at 1263. Most notably, the

8
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 9 of 46

FDC expressed concern that an inmate wearing makeup and female undergarments

would inevitably become a target in an all-male prison, thereby endangering not

only the inmate but also the prison employees who would have to step in to protect

her. Id. The FDC also (and relatedly) determined that there were clear advantages

to maintaining uniformity in a prison setting, including the ability to more readily

detect contraband. Id.

As relevant to the concerns raised in the dissental, the district court held that

by refusing Keohane’s social-transitioning requests, Florida prison authorities were

“deliberately indifferent” to Keohane’s serious medical needs in violation of the

Eighth Amendment. Id. at 1264–65. Accordingly, the court entered an injunction

ordering prison officials to “permit Ms. Keohane to socially transition by allowing

her access to female clothing and grooming standards.” Id. at 1265. On appeal,

the panel held (again, as relevant here) that the FDC did not violate the Eighth

Amendment by refusing to accommodate Keohane’s social-transitioning-related

requests, and we therefore vacated the district court’s injunction. See id. at 1272–

80. In so doing, we reviewed de novo the district court’s ultimate determination

that there was an Eighth Amendment violation, and we reviewed subsidiary issues

of fact for clear error. See id. at 1272 & n.8 (citing Thomas v. Bryant, 614 F.3d

1288 (11th Cir. 2010)).

9
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 10 of 46

II

Next, a bit of necessary legal background: A deliberate-indifference claim

entails two components, the latter of which entails three sub-components. See

Hoffer v. Sec’y, Fla. Dep’t of Corr., 973 F.3d 1263, 1270 (11th Cir. 2020). First,

the inmate must establish “an objectively serious medical need”—that is, “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”—that, “if left unattended, poses a substantial risk of serious harm.” Id.

(quotation marks omitted). Second, the inmate must prove that prison officials

acted with deliberate indifference to that need by showing (1) that they had

“subjective knowledge of a risk of serious harm” and (2) that they “disregarded”

that risk (3) by conduct that was “more than gross negligence.” Id.

Here, there’s no debate about the first component—everyone agrees that

Keohane’s gender dysphoria constitutes a serious medical need. Rather, the

parties’ dispute (and the dissental’s concern) hinges on the application of—and in

particular our review of—the second component.

III

Today’s dissental is predicated on an assertion that the panel only

“purport[ed] to follow,” but instead strategically “reimagine[d],” this Court’s

earlier decision in Thomas v. Bryant, 614 F.3d 1288 (11th Cir. 2010), when we

10
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 11 of 46

held that the de novo—rather than the clear-error—standard of review applied to

the district court’s determination that prison officials violated the Eighth

Amendment in refusing Keohane’s social-transitioning requests. See Rosenbaum

Dissenting Op. at 26, 29, 43. And because the panel turned its back on Thomas,

the dissental asserts, it violated this Court’s prior-panel-precedent rule, pursuant to

which “a prior panel’s holding is binding on all subsequent panels unless and until

it is overruled or undermined to the point of abrogation by the Supreme Court or

by this court sitting en banc.” United States v. Archer, 531 F.3d 1347, 1352 (11th

Cir. 2008).

With respect, the panel didn’t “reimagine” Thomas—let alone reimagine it

“as [we] pleased.” Rosenbaum Dissenting Op. at 26, 32. To the contrary, and the

dissental’s shade-throwing notwithstanding, the panel followed Thomas—or, to be

more precise, it followed the breadcrumbs left in Thomas’s various (and sometimes

conflicting) passages. Because the panel followed (rather than “flout[ed],” see id.

at 23) Thomas, it didn’t violate the prior-panel-precedent rule. And because the

panel didn’t violate the prior-panel-precedent rule, the basis for the dissental

evaporates.

In an effort to paint a picture of lawless, result-oriented judging, the

dissental gives the misleading impression that Thomas is pellucidly clear—that it

just says, over and over and over, that the entirety of an Eighth Amendment claim,

11
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 12 of 46

from stem to stern, is subject only to clear-error review, and that the panel, in the

face of all that clarity, willfully bulled ahead with de novo review instead. As is so

often the case, the reality is more complicated.

It is true, as the dissental says, that Thomas states that “[a] prison official’s

deliberate indifference is a question of fact which we review for clear error.” 614

F.3d at 1312; see also id. at 1302 (“Subsidiary issues of fact are reviewed for clear

error.”). In our opinion, the panel candidly admitted as much. See Keohane, 952

F.3d at 1272 n.8. But Thomas also holds—and I’ll just quote it directly—that an

appellate court must review the ultimate determination “that there was an Eighth

Amendment violation warranting equitable relief . . . de novo.” Thomas, 614 F.3d

at 1303.

For better or worse, then, it fell to the panel to try to synthesize those

competing directives. And for what it’s worth—and totally unsurprisingly, I’m

sure—I think the panel got it exactly right. Let me explain.

A

Despite all the adverbs that the dissental uses to describe the Thomas

opinion—it says, in turn, that Thomas “expressly,” “demonstrably,”

“unmistakably,” “unambiguously,” and “repeatedly” prescribed across-the-board

clear-error review—everyone recognizes that some synthesis of Thomas’s mixed

messages is necessary. I recognize it. Judge Wilson’s panel dissent recognized it.

12
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 13 of 46

And, yes, even Judge Rosenbaum’s dissental ultimately recognizes it. Here, it

seems to me, are the contenders:

1. The Panel Opinion. In tackling the case, the panel took seriously

Thomas’s dual directives (1) that an appellate court must review de novo the

district court’s ultimate determination “that there was an Eighth Amendment

violation warranting equitable relief” and (2) that “[s]ubsidiary issues of fact are

reviewed for clear error.” 614 F.3d at 1302. Accordingly, we held (1) that the

clear-error standard governs what we (echoing the Supreme Court) called

“historical facts—e.g., what happened, who knew what, how did they respond?”—

but (2) that “what the Eighth Amendment means—and requires in a given case—is

an issue squarely within the core competency of appellate courts” and is thus

subject to de novo review. Keohane, 952 F.3d at 1272–73 n.8. The panel opinion

therefore gives meaningful roles to both the de novo and clear-error standards, both

of which Thomas prescribes.

2. The Dissental’s Proposal. Today’s dissental suggests (without quite

saying) that, perhaps, de novo review applies only to the first, objective component

of a deliberate-indifference claim: “Thomas’s precise statements applying clear-

error review to all components of the subjective inquiry . . . are entirely

harmonious with Thomas’s statement that de novo review applies to the

overarching question of deliberate-indifference. The overarching standard of

13
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 14 of 46

review is necessarily de novo because it incorporates within it de novo review of

the objective inquiry of the deliberate-indifference analysis.” Rosenbaum

Dissenting Op. at 33 (emphasis added); see also id. at 34 n.5 (“[De novo review]

extends to the ultimate conclusion because the ultimate conclusion necessarily

includes within it a determination based on de novo review (the objective

inquiry).” (emphasis added)).

With respect, that can’t be correct. The Thomas opinion prescribes de novo

review in two separate places—with respect to two different issues. In one place, it

states that the constituent determination whether the deprivations suffered by the

inmate “are objectively ‘sufficiently serious’ to satisfy the objective prong” of the

deliberate-indifference standard “is a question of law” subject to de novo review.

Thomas, 614 F.3d at 1307. In another, it states, separately, that the district court’s

ultimate determination “that there was an Eighth Amendment violation warranting

equitable relief”—i.e., the entirety of that determination—“is reviewed de novo.”

Id. at 1302. So de novo review unquestionably applies to more than just the first

prong of the deliberate-indifference standard—it applies, somehow or another, to

the whole enchilada.

3. The Panel Dissent. Which brings me to the panel dissent’s reading of

Thomas. It seemed to appreciate that de novo review applies to the entirety of the

deliberate-indifference claim—to the ultimate determination that the Eighth

14
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 15 of 46

Amendment was violated—but it left the de novo standard only a vanishingly

small role:

[I]f the district court, despite checkmarks in both the objective and
subjective boxes, still concluded that there was no Eighth Amendment
violation, we would lend no deference to this error. We would review
it de novo, and would no doubt reverse. And if the district court,
despite holding that one of the elements was not met, still concluded
that there was an Eighth Amendment violation, we would do the
same. We would review this error de novo, and no doubt
reverse. That is the ultimate conclusion that we review de novo.

Keohane, 952 F.3d at 1288 (Wilson, J., dissenting).

With respect, that can’t be right, either. As the panel majority explained in

our opinion, it is inconceivable that “the de novo standard’s sole office is to ensure

that the district court puts ‘checkmarks’ in the right boxes, and then doesn’t make a

truly boneheaded, asinine mistake.” Id. at 1273 n.8.2

B

Among the available alternatives, it won’t surprise you to learn that I think

the panel’s synthesis of Thomas’s mixed messages is clearly correct. I say so for

2
The dissental suggests that the panel “g[ave] itself permission to reimagine what Thomas held
because it conclude[d] that what Thomas expressly said ‘cannot possibly be what we’ve meant.’”
Rosenbaum Dissenting Op. at 26 (quoting Keohane, 952 F.3d at 1273 n.8). No. What the panel
said in the passage that the dissental snatch-quotes is that the panel dissent’s “checkmark”
interpretation of de novo review “cannot possibly be what we’ve meant when we have
repeatedly held that de novo review applies to the district court’s determination whether ‘there
was an Eighth Amendment violation warranting equitable relief.’” Keohane, 952 F.3d at 1273
n.8 (quoting Thomas, 614 F.3d at 1303).

15
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 16 of 46

reasons that I have already explained and that neither of my dissenting colleagues

has even engaged, let alone rebutted.

1

First, the panel’s interpretation gives meaningful roles to both the de novo

and clear-error standards—both of which, again, Thomas expressly prescribes. By

contrast, neither of my dissenting colleagues has any viable explanation of what

role de novo review should play in Eighth Amendment deliberate-indifference

cases. As between the courts of appeals and the district courts, who decides what

the Eighth Amendment ultimately means and requires in a given case? On their

theories, the district courts do—at which point the appellate courts’ hands are

pretty much tied. What an odd state of affairs. In what other circumstance do the

courts of appeals effectively cede to district courts the job of determining the

meaning and proper application of the Constitution?3

2

Second, “meaningful appellate review of a district court’s ultimate

constitutional holding follows straightaway from Supreme Court precedent

3
I note that while a holding that clear-error review applies to the entirety of the deliberate-
indifference analysis—and effectively binds us to the district courts’ determinations—might
serve Keohane well in this particular case, it would be cold comfort to the multitude of prisoners
who appeal from district court orders rejecting deliberate-indifference claims. And of course,
the vast majority of deliberate-indifference cases that appellate courts see arise on appeal by
inmates who have lost below.

16
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 17 of 46

prescribing de novo review of other application-of-law-to-fact questions—

including those arising under the Eighth Amendment.” Keohane, 952 F.3d at 1273

n.8. In United States v. Bajakajian, 524 U.S. 321 (1998), for instance, the

Supreme Court rejected the contention that an appellate court should defer to a

district court’s determination whether a fine is excessive for Eighth Amendment

purposes. As the Supreme Court explained there, while the district court’s factual

findings in conducting the excessiveness inquiry “must be accepted unless clearly

erroneous,” “whether a fine is constitutionally excessive calls for the application of

a constitutional standard to the facts of a particular case”—and thus calls for “de

novo review.” Id. at 336–37 & n.10.

Similarly, in Ornelas v. United States, the Supreme Court held “that as a

general matter determinations of reasonable suspicion and probable cause should

be reviewed de novo on appeal.” 517 U.S. 690, 699 (1996). The Court

acknowledged—precisely as our opinion did—“that a reviewing court should take

care both to review findings of historical fact only for clear error and to give due

weight to inferences drawn from those facts by resident judges and local law

enforcement officers.” Id. (emphasis added); see Keohane, 952 F.3d at 1273 n.8.

But it insisted that the ultimate determination—the application of the constitutional

standard to those facts—demands de novo review. Significantly, the Court gave

three reasons to support its holding, all of which apply equally here: (1) the

17
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 18 of 46

constitutional standards at issue involve “fluid concepts that take their substantive

content from the particular contexts in which the standards are being assessed”; (2)

the applicable “legal rules . . . acquire content only through application,” and

“[i]ndependent review is therefore necessary if appellate courts are to maintain

control of, and to clarify, the legal principles”; and (3) “de novo review tends to

unify precedent” and “stabilize the law.” Ornelas, 517 U.S. at 696–98.

Finally, applying Bajakajian and Ornelas—and repeating the same

considerations—the Supreme Court held in Cooper Industries, Inc. v. Leatherman

Tool Group, Inc., “that courts of appeals should apply a de novo standard of review

when passing on district courts’ determinations of the constitutionality of punitive

damages awards” under the Fourteenth Amendment. 532 U.S. 424, 436 (2001).

Except to say that they aren’t deliberate-indifference cases, neither of my

dissenting colleagues has offered any explanation why the rationale of Bajakajian,

Ornelas, and Cooper Industries doesn’t apply here. District courts are

undoubtedly better situated than appellate courts to make findings of what the

panel (echoing the Supreme Court in Ornelas) called “historical facts,” and their

determinations with respect to those facts are accordingly entitled to deference.

But what the Eighth Amendment means—and requires in a given case—is, as I

have said, an issue squarely within the core competency of appellate courts. And

to be clear, it’s no answer to say, as the panel dissent did—citing Justice Scalia’s

18
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 19 of 46

solo dissent in Ornelas—that some issues underlying a deliberate-indifference

claim may be “fact-specific and not easy to generalize.” Keohane, 952 F.3d at

1291 n.13 (Wilson, J., dissenting). The Supreme Court recognized as much

regarding the “mixed questions” in Bajakajian, Ornelas, and Cooper Industries—

and yet applied de novo review anyway. Just so here.

3

Finally, the panel’s synthesis of Thomas’s standard-of-review conundrum

squares precisely with the First Circuit’s en banc decision in the factually similar

Kosilek v. Spencer, 774 F.3d 63 (1st Cir. 2014). The court there rejected the very

arguments that my dissenting colleagues have made and held that de novo, rather

than clear-error, review governed a district court’s determination that the Eighth

Amendment required prison authorities to accommodate a transgender inmate’s

medical-treatment requests. In Kosilek, an inmate alleged that the Massachusetts

Department of Correction’s refusal to provide sex-reassignment surgery to treat the

inmate’s gender-identity disorder constituted deliberate indifference. Id. at 68–69.

Sitting en banc, the First Circuit explained that “[t]he test for establishing an

Eighth Amendment claim of inadequate medical care encompasses a multitude of

questions that present elements both factual and legal”—and, therefore, that

“[r]eview of such ‘mixed questions’ is of a variable exactitude,” such that “the

more law-based a question, the less deferentially we assess the district court’s

19
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 20 of 46

conclusion.” Id. at 84. Citing our opinion in Thomas, the Kosilek court held that

“[t]he ultimate legal conclusion of whether prison administrators have violated the

Eighth Amendment is reviewed de novo.” Id. In so holding, the court rejected the

dissenting judges’ argument that “the ultimate constitutional question is

inextricably tied up with the factual details that emerged at trial,” which, according

to them, “counsels against pure de novo review.” Id. at 99 (Thompson, J.,

dissenting). While acknowledging—again, just as the panel did here—that

appellate courts “award[] deference to the district court’s resolution of questions of

pure fact and issues of credibility,” the Kosilek majority stood by its conclusion

that the ultimate Eighth Amendment question is reviewed de novo. Id. at 84–85.

Notably, the court buttressed its holding with citations to decisions from several

other circuits reaching the same conclusion. 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 demonstrate an Eighth Amendment violation is a question of law that we

review de novo.”); Hickey v. Reeder, 12 F.3d 754, 756 (8th Cir. 1993) (“Whether

conduct, if done with the required culpability, is sufficiently harmful to establish

an Eighth Amendment violation is an objective or legal determination which we

decide de novo.”); 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

20
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 21 of 46

violation of constitutional rights is a question of law that may be assayed anew

upon appeal.”).

* * *

So in short, the panel here didn’t “reimagine” this Court’s earlier decision in

Thomas but, rather, synthesized Thomas’s mixed messages in accordance with

Supreme Court and other circuits’ precedents.

IV

I’ll conclude where I began. For all its rhetorical flourish, today’s dissent

from denial simply doesn’t make a compelling argument that this case warranted

en banc reconsideration. The panel was faced with the vexing question of how the

de novo and clear-error standards of review map onto the various elements and

sub-elements of an Eighth Amendment deliberate-indifference claim—a question

made all the more vexing by Thomas’s (let’s just say) imprecise discussion of that

issue. Faced with all that ambiguity, the panel did its level best—both to apply

Thomas and to faithfully and correctly decide the case before it. I, for one, think

the panel got it exactly right. But even if I’m wrong about that—and reasonable

minds can disagree—the worst that can be said of the panel opinion is that it

“misappli[ed the] correct precedent to the facts of the case.” 11th Cir. R. 35-3. In

this Circuit, that is not a ground for en banc rehearing. Id.

21
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 22 of 46

While the dissental’s spicy rhetoric doesn’t enhance its argument—but

rather pretty severely diminishes it, to my mind—it does, I fear, corrode the

collegiality that has historically characterized this great Court. Here’s hoping for

better—and more charitable—days ahead.

22
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 23 of 46

ROSENBAUM, Circuit Judge, joined by WILSON, MARTIN, and JILL PRYOR,

Circuit Judges, dissenting from the denial of rehearing en banc:

This is not an easy dissent to write—not because the legal issue involved in

the merits of this case is complex or difficult (it’s not), but because our denial of

rehearing en banc here is not—or at least should not be—normal. We are denying

en banc rehearing in a case that objectively qualifies for it under the Federal Rules

of Appellate Procedure and that indeed demands it to preserve the sanctity of the

prior-precedent rule and the important policies of stability and predictability that that

rule serves.

Our failure to hold en banc review in a case where the panel opinion

contradicts our holdings in opinions earlier panels issued yet claims nonetheless to

comply with the prior-precedent rule introduces uncertainty and confusion into the

law of our Circuit. And worse, it undermines the prior-precedent rule, “the

foundation of our federal judicial system.” Smith v. GTE Corp., 236 F.3d 1292 (11th

Cir. 2001) (quoting Jaffree v. Wallace, 705 F.2d 1526, 1533 (11th Cir. 1983)). We

as a Court must reckon with these potentially devastating consequences of our

actions if we continue to allow opinions that flout the prior-precedent rule while

claiming they comply with it to issue unchecked.

A case like this one, where the opinion distorts beyond recognition the prior-

precedent rule—a fundamental mechanism by which this Court ensures the

23
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 24 of 46

predictability and stability of the law—is exactly the kind of case for which the en

banc tool was designed. For that reason, whether any individual judge agrees or

disagrees with the ultimate outcome of Keohane should be irrelevant to the question

of whether this case warrants en banc review. Too much is at stake.

I am sure each of us believes that we are applying the appropriate standards

in determining whether to vote for en banc rehearing.1 But an objective analysis

suggests we are not. So we need to recalibrate. I urge our Court—and each of us

individually—to carefully and objectively reexamine this vote and to truly reflect on

the dangers of condoning panel opinions that contradict our prior precedent while

nonetheless claiming to follow the prior-precedent rule.

I divide my discussion into three substantive sections. In Section I, I review

the law governing the limited circumstances in which en banc review is appropriate.

In Section II, I show that this case warrants en banc rehearing. And in Section III, I

1
I am truly sorry that Chief Judge Pryor and Judge Newsom seem to have taken my
concerns personally. I do not believe this dissent to be personal. I have great respect for all my
colleagues, and I value this Court’s collegiality. But I also have great respect for the rule of law
and the need for our Court to maintain its legitimacy. And I don’t agree that defending these things
or pointing out what I think is wrong with Keohane and explaining why I view it as such a big
problem makes me “[un]collegial[]” and “[un]charitable,” see Newsom Op. at 22, or is an “attack[
on] . . . the integrity of judges or their commitment to the rule of law . . . [or] the legitimacy of this
Court,” W. Pryor Op. at 5. Nor do the labels and characterizations the W. Pryor and Newsom
Opinions feel a need to impose provide a good enough reason to remain silent in the face of the
threat Keohane represents to our judicial norms. I am aware of no other way to oppose what I see
as the failure of our Court to require the Keohane panel to comply with the prior-precedent rule,
other than by writing a dissent that candidly discusses that problem and its significance.
24
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 25 of 46

explain why we must insist on strict adherence to the prior-precedent rule by every

panel.

But first, a word of caution: this dissent is not about what the substantive law

that governs Keohane’s case should or should not be. And to avoid any possible

misunderstanding on that, I begin by stating expressly that I take no position on that

in this dissent. Even assuming without deciding that Keohane arrived at the

objectively legally correct rule for the appropriate standard of review in Eighth

Amendment deliberate-indifference claims, the problem here is that on the way to

doing so, it issued a new rule that is contrary to our binding precedent while

nonetheless attributing that new rule to that same precedent. The proper procedure

for overruling binding precedent in this Circuit requires the Court sitting en banc to

set it aside; a panel is not free to overrule binding precedent on its own. To be clear,

then, this dissent is solely about the importance to the stability and predictability of

the law of ensuring every panel strictly follows our prior-precedent rule.

I.

Rule 35, Fed. R. App. P., anticipates that en banc rehearing will be ordered

only when it is “necessary to secure or maintain uniformity of the court’s decisions”

or the case “involves a question of exceptional importance.” Fed. R. App. P. 35(a).

25
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 26 of 46

This is one of those cases.2 First, this case objectively warrants en banc review under

Rule 35(a) because the panel majority opinion here, Keohane v. Florida Department

of Corrections, 952 F.3d 1257 (11th Cir. 2020), creates confusion and inconsistency

in our Eighth Amendment Circuit jurisprudence. But second and more urgent is

what the Keohane panel’s interpretation and application of the prior-precedent rule

and our refusal to take this case en banc do to that rule. In trying unsuccessfully to

avoid running afoul of our prior-precedent rule and raising a conflict with our earlier

precedent known as Thomas v. Bryant, 614 F.3d 1288, 1312 (11th Cir. 2010), the

Keohane majority opinion remolds the prior-precedent rule into an unrecognizable

and dangerous form: it gives itself permission to reimagine what Thomas held

because it concludes that what Thomas expressly said “cannot possibly be what

we’ve meant,” Keohane, 952 F.3d at 1272.

Our “firmly established” prior-precedent rule strictly requires later panels to

follow the precedent of earlier panels unless and until the prior precedent is overruled

or undermined to the point of abrogation by the Supreme Court or this Court sitting

2
Chief Judge Pryor notes that “[t]he grant of en banc review is and should be rare.” W.
Pryor Op. at 4. Of course, as I recognize above, that’s true, when we consider the total number of
cases we review every year. But that’s also an oversimplification of what happened here. By my
count, in the most recent nearly two-year period (since January 1, 2019), we have voted for en
banc rehearing in twelve cases. During that same period, we have voted against en banc review
only seven times when a member in active service on this Court has requested an en banc poll. So
once a member of this Court in active service has sought an en banc poll, we have granted en banc
rehearing at a rate of 63%—a majority of the time. Presumably, that is because we exercise
extreme discretion in requesting an en banc poll in the first place.
26
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 27 of 46

en banc. United States v. Steele, 147 F.3d 1316, 1318 (11th Cir. 1998) (en banc);

see also Smith, 236 F.3d at 1303 n.11. We have described ourselves as “emphatic”

in our strict adherence to this rule, see Steele, 147 F.3d at 1318 (quoting Cargill v.

Turpin, 120 F.3d 1366, 1386 (11th Cir. 1997)), and have said that this Court “take[s]

[the prior-precedent] rule seriously,” Atl. Sounding Co., Inc. v. Townsend, 496 F.3d

1282, 1286 (11th Cir. 2007) (Ed Carnes, J., concurring).

We have gone so far as to hold that under that rule, a later panel cannot

overrule a prior one’s holding “even though convinced it is wrong.” Steele, 147 F.3d

at 1317-18. Indeed, we have held that the prior-precedent rule binds later panels

even when the prior panel’s decision failed to mention controlling Supreme Court

precedent and reached a holding in conflict with that precedent. Smith, 236 F.3d at

1302-03. So strong is the prior-precedent rule that under it, a later panel is bound

by the earlier panel’s “reasoning and result,” even when the prior panel does not

explicitly state its rule. See id. at 1304.

No exceptions to the prior-precedent rule exist. See id. at 1302. That is so,

we have explained, because if an exception applied, “it could end up nullifying the

well-established prior panel precedent rule that is an essential part of the governing

law of this Circuit.” Id. (emphasis added). Not only that but the prior-precedent rule

“helps keep the precedential peace among the judges of this Court, and it allows us

to move on once an issue has been decided.” Townsend, 496 F.3d at 1286 (Ed

27
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 28 of 46

Carnes, J., concurring). Without it, as Judge Ed Carnes has cautioned, “every sitting

of this court would be a series of do-overs, the judicial equivalent of the movie

‘Groundhog Day.’” Id.

The Newsom Opinion takes issue with my discussion, describing it as full of

“hyperbole and invective” because of my concerns that continued disregard of our

prior-precedent rule jeopardizes the rule of law. See Newsom Op. at 6. But those

concerns are not overblown. Indeed, our own Court has emphasized that the prior-

precedent rule “serves as the foundation of our federal judicial system[, as]

[a]dherence to it results in stability and predictability.” Smith, 236 F.3d at 1303

(quoting Jaffree v. Wallace, 705 F.2d 1526, 1533 (11th Cir. 1983)). For these

reasons, if we continue to allow panels to skirt the prior-precedent rule, we certainly

ask for trouble.

Under our prior-precedent rule, if a panel vehemently disagrees with a prior

precedent, its only option is to apply it, anyway, and call for en banc rehearing. It

may not, under any circumstances, create its own conflicting rule and inaccurately

purport to apply the governing prior precedent. And if it does, en banc review is in

order—either to correct the panel opinion and make it comply with binding

precedent, or to overrule the prior precedent. But in any case, changing prior

precedent is not something that a panel may do.

II.

28
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 29 of 46

Yet that’s precisely what the Keohane panel did: while insisting it was

following Thomas, it instead created a new rule diametrically opposed to Thomas’s

holding. To show that this is necessarily the case, I must first briefly review the

relevant facts of Keohane’s case (Section A) and the law governing claims of

deliberate indifference to prisoners’ medical claims (Section B). Then in Section C

I point out how the Keohane panel opinion is at war with Thomas, the precedent it

purports to follow.

A.

Reiyn Keohane was assigned male at birth, but she has identified as female

since she was about eight years old. ECF No. 171 at 1. She was formally diagnosed

with gender dysphoria when she was sixteen. Id. at 2. At that time, Keohane began

a hormone-therapy regimen. Id.

After her arrest, she was cut off from her treatment, including hormone

therapy and the ability to dress and groom as a woman. Id. at 2. She complained,

but the prison did not respond. Id. Keohane’s untreated dysphoria caused her such

extreme anxiety that she attempted to kill herself and castrate herself while in

custody. Id.

These facts and others led Keohane to sue the Florida Department of

Corrections (the “FDC”) under 42 U.S.C. § 1983, alleging violations of her Eighth

Amendment rights and seeking declaratory and injunctive relief. As the panel

29
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 30 of 46

opinion explained, after a bench trial, the district court entered a 61-page order

awarding Keohane relief. As relevant to the concerns I raise in this dissent,3 the

district court directed the FDC to permit Keohane “to socially transition by allowing

her access to female clothing and grooming standards.” 952 F.3d at 1262 (quoting

district court order).

The panel opinion reversed. Claiming to follow Thomas, the Keohane panel

applied the de novo standard of review to certain components of the district court’s

findings for which the Thomas Court had instructed a clear-error standard governs.

In so doing, the panel simultaneously injected conflict into our Eighth Amendment

deliberate-indifference jurisprudence and stretched interpretation of our prior-

precedent rule beyond recognition.

B.

The Eighth Amendment prohibits the government from inflicting “cruel and

unusual punishments” on inmates. Wilson v. Seiter, 501 U.S. 294, 296–97 (1991).

That prohibition encompasses “deprivations . . . not specifically part of [a] sentence

but . . . suffered during imprisonment.” Id. at 297. An inmate who suffers

“deliberate indifference” to her “serious medical needs” may state a claim for a

violation of the Eighth Amendment. Estelle v. Gamble, 429 U.S. 97, 104 (1976);

3
Appellant’s motion for rehearing also seeks rehearing on whether the Majority Opinion
correctly applied our mootness exception for voluntary cessation. For purposes of this dissent
from the denial of rehearing en banc, I express no views about the propriety of that holding.
30
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 31 of 46

Brown v. Johnson, 387 F.3d 1344, 1351 (11th Cir. 2004). Under our precedent, a

deliberate-indifference claim has two components: an objectively serious medical

need and subjective deliberate indifference by the official to that need. Brown, 387

F.3d at 1351.

Thomas holds that the objective inquiry (whether a serious medical need

exists) includes both questions of fact subject to clear-error review and a question of

law subject to de novo review. Thomas, 614 F.3d at 1307, 1308. I am not concerned

with that aspect of Keohane because the panel opinion had no occasion to comment

on or apply the Thomas standard of review to the objective inquiry, since “all

agree[d] that Keohane’s gender dysphoria” satisfies that requirement. 952 F.3d at

1273. Instead, I focus on the subjective component of Keohane’s deliberate-

indifference claim.

Before Keohane, we had described the subjective inquiry (whether the

defendant was subjectively deliberately indifferent to the plaintiff’s serious medical

need) to require the plaintiff “to prove three facts: (1) subjective knowledge of a

risk of serious harm; (2) disregard of that risk; and (3) by conduct that is more than

mere negligence” (the “subjective-inquiry components”). Brown, 387 F.3d at 1351

(emphasis added). Of course, in characterizing as “facts” the three things that the

plaintiff must prove to satisfy the subjective inquiry, we suggested that the district

court’s resolutions of the three subjective-inquiry components were findings of fact.

31
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 32 of 46

Then, as Judge Wilson plainly showed in his dissent in Keohane, 952 F.3d at

1289-92 (Wilson, J., dissenting), we said precisely that a few years later in Thomas.

There, several inmates sued FDC employees, alleging that their use of chemical

agents on certain mentally ill inmates violated the Eighth Amendment. 614 F.3d at

1293. After a five-day bench trial, the district court determined that the FDC’s non-

spontaneous disciplinary use of chemical agents on inmates who, at the time, were

unable to conform their behavior to prison standards because of their mental

illnesses, violated the Eighth Amendment. Id. at 1294. The FDC employees

appealed this conclusion (among other district-court actions not relevant here). Id.

In analyzing the appeal, we noted that we review de novo the “legal

conclusion—that there was an Eighth Amendment violation[,]”—but that

“[s]ubsidiary issues of fact are reviewed for clear error.” Id. at 1303. Had we

stopped there and both said nothing more about the standard of review and not

applied the standard of review in a way that demonstrated what we meant by this

division of the standard of review, perhaps the Keohane panel would have been free

to construe those two propositions as it pleased, without running afoul of the prior-

precedent rule.4

4
That, too, is questionable (though less so than Keohane’s characterization of Thomas), in
light of our prior description in Brown, 387 F.3d at 1351, that a plaintiff had to prove three “facts”
to establish all three aspects of the subjective part of a deliberate-indifference claim.
32
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 33 of 46

But that was not the end of our discussion and application of the standard of

review. Rather, later in Thomas, our statements concerning the appropriate standard

of review, as well as our application of that standard of review, unmistakably show

that the phrase “legal conclusion . . . that there was an Eighth Amendment violation”

refers to “[t]he ultimate legal conclusion” of whether the defendants violated the

Eighth Amendment, Kosilek v. Spencer, 774 F.3d 63 (1st Cir. 2014). And the phrase

“[s]ubsidiary issues of fact” refers to the issues of fact that one of the objective-

inquiry components and all the subjective-inquiry components of the deliberate-

indifference analysis constitute.

Contrary to the Newsom Opinion’s characterization of Thomas, there is

nothing “conflicting” or “competing,” see Newsom Op. at 11, 12, about Thomas’s

direction. Nor does Thomas send “mixed messages.” Id. at 12. Thomas’s precise

statements applying clear-error review to all components of the subjective inquiry—

outlined in detail below—are entirely harmonious with Thomas’s statement that de

novo review applies to the overarching question of deliberate-indifference. The

overarching standard of review is necessarily de novo because it incorporates within

it de novo review of the objective inquiry of the deliberate-indifference analysis.5

5
The Newsom Opinion attempts to alter the focus from my reason for seeking en banc
review—Keohane’s failure to follow Thomas and abide by the prior-precedent rule—by arguing
that daylight exists between Judge Wilson and me concerning Thomas’s holding on the applicable
standard of review. See Newsom Op. at 13-15. It doesn’t. I fully agree with Judge Wilson that
Thomas applies the clear-error standard to all aspects of the subjective-inquiry prong of the
deliberate-indifference test. See Keohane, 952 F.3d at 1287-90 (Wilson, J., dissenting). I likewise
33
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 34 of 46

And nothing in Thomas supports the Keohane majority opinion’s suggestion

that, by “[s]ubsidiary issues of fact,” Thomas meant only what the Keohane majority

opinion deemed “historical facts.” See Keohane, 952 F.3d at 1272 n.8. Tellingly,

the Keohane majority opinion cited nothing in Thomas for its contention. Nor did

Thomas ever use the term “historical facts.” On the contrary, as Judge Wilson’s

Keohane dissent and this dissent demonstrate in detail, see infra at 35-38, Thomas

unambiguously applied the clearly erroneous standard of review to each of the three

components of the subjective inquiry. The Keohane majority opinion was not free

to stray from the clear-error standard of review that Thomas held governs the

components of the subjective inquiry. Yet that is what it did.

understand Thomas’s statement that we review “the district court’s . . . conclusion [] that there was
an Eighth Amendment violation warranting equitable relief [] . . . de novo,” Thomas, 614 F.3d at
1303, to refer in context to the notion that “[o]ur de novo review extends only to questions of law
(i.e., the objectively-serious-need element) and to the district court’s ultimate conclusion whether
the objective and subjective elements of a deliberate indifference claim state an Eighth
Amendment violation.” Keohane, 952 F.3d at 1287 (Wilson, J., dissenting). As I have noted, it
extends to the ultimate conclusion because the ultimate conclusion necessarily includes within it a
determination based on de novo review (the objective inquiry). And we’re not the only ones to
understand Thomas’s clear analysis this way. Indeed, in Kosilek v. Spencer, 774 F.3d 63 (1st Cir.
2014), the First Circuit cited Thomas for the proposition that “[t]he ultimate legal conclusion of
whether prison administrators have violated the Eighth Amendment is reviewed de novo,” while
quoting from Thomas in the supporting parenthetical only the following: “Whether the record
demonstrates that [the prisoner] was sprayed with chemical agents . . . and that he suffered
psychological injuries from such sprayings are questions of fact. Whether these deprivations are
objectively ‘sufficiently serious’ to satisfy the objective prong, is a question of law . . . .” Kosilek,
774 F.3d at 84 (quoting Thomas, 614 F.3d at 1307) (quotation marks omitted) (alterations by the
Kosilek Court) (emphasis added). Tellingly, Kosilek never refers to the subjective inquiry to
provide an example of presenting any legal questions to show why Thomas refers to the
overarching question of deliberate-indifference liability as subject to de novo review. Nor does
Kosilek cite Thomas for the proposition that any components of the subjective inquiry are subject
to de novo review.
34
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 35 of 46

C.

I begin with what Thomas had to say about how the standard of review applies

to the evaluation of a district court’s determination on the subjective inquiry of an

Eighth Amendment deliberate-indifference claim—that is, whether the defendant (1)

subjectively knew of a risk of serious harm; (2) nonetheless disregarded that risk;

and (3) did so by conduct that is more than mere negligence. Thomas, 614 F.3d at

1312. Thomas explained that this inquiry requires us to determine whether “the

evidence . . . demonstrate[s] that with knowledge of the infirm conditions, the official

knowingly or recklessly declined to take actions that would have improved the

conditions.” Id. (internal quotation marks and alterations omitted). This summary

of what the subjective inquiry requires encompassed all three prongs of the

subjective inquiry: “knowledge of the infirm conditions” means the defendant

“subjectively knew of a risk of serious harm”; “declined to take actions that would

have improved the conditions” means the defendant “disregarded the risk”; and

“knowingly or recklessly” did so means the defendant “did so by conduct that is

more than mere negligence.”

Immediately after we summarized what the three components of the

subjective inquiry require, we unambiguously stated that “[a] prison official’s

deliberate indifference is a question of fact which we review for clear error.” Id.

(emphasis added). Based on the placement and content of the remark, it is clear that

35
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 36 of 46

with this statement, we were referring to the entirety of the three-part subjective

inquiry. Not only did this quotation appear immediately following our summary of

all the elements of the subjective component of an Eighth Amendment violation, but

significantly, by definition, “[a] prison official” can engage in “deliberate

indifference” only if all three subjective-inquiry components are satisfied.

We need not review Thomas any further than this to know that we

unambiguously held in Thomas that the district court’s rulings on the entire

subjective inquiry—including its rulings on all three of the subjective inquiry’s

components—are subject to the clearly erroneous standard of review.

But there’s more. In addressing each of the three parts of the subjective

inquiry in Thomas, we again said and demonstrated that each component is to be

reviewed for clear error.

With respect to the first question—whether the defendant subjectively knew

of a risk of serious harm—we relied in Thomas on Farmer v. Brennan, 511 U.S. 825,

842 (1994), citing to Farmer’s proposition that “[w]hether a prison official had the

requisite knowledge of a substantial risk is a question of fact subject to

demonstration in the usual ways . . . .” 6 See Thomas, 614 F.3d at 1313 (quoting

Farmer, 511 U.S. at 842) (alteration omitted) (emphasis added).

6
Though in Thomas we omitted the phrase “question of fact” from the Farmer quotation,
we did so only to avoid repetition of the phrase “question of fact,” which appears in the sentence
immediately preceding the Farmer quotation. As I’ve noted above, that sentence states that the
36
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 37 of 46

As to the second and third parts of the test—whether the defendant

disregarded this risk by more than mere negligence—we concluded that “[t]he

record . . . supports the district court’s finding that the Secretary of the [FDC] and

the Warden . . . recklessly disregarded the risk of psychological harm to inmates like

[the Thomas plaintiff].” Id. at 1315 (emphasis added). We then described the

evidence in the record that underpinned the district court’s finding. Id. For example,

we opined that “the [FDC]’s refusal to modify its non-spontaneous use-of-force

policy provides support for the district court’s finding of more than mere or even

gross negligence on the part of the [FDC].” Id. (emphasis added). The repeated

references to evidence in the record and uses of the words “support for the district

court’s finding” further unmistakably demonstrate that we viewed the district court’s

“findings” as factual findings, and we reviewed them for clear error.

But you need not take my word for it. Thomas says as much. Thomas began

its analysis by invoking the clear error-test and by summarizing its conclusion that

the defendant failed to meet that standard: “[O]ur review of the district court’s

voluminous uncontested factual findings as they relate to the defendants’ deliberate

indifference does not leave us with the definite and firm conviction that a mistake

has been committed. Accordingly, the defendants have failed to satisfy their burden

entire subjective inquiry is a question of fact we review for clear error. See Thomas, 614 F.3d at
1312.
37
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 38 of 46

of demonstrating the district court’s clear error.” Thomas, 614 F.3d at 1313

(emphasis added).

That was no mistake. Thomas also ended its discussion of the second and

third components of the subjective inquiry by summing up that it could not

“conclude that the district court was clearly erroneous in finding that the record

demonstrates that FDC officials turned a blind eye to [the plaintiff’s] mental health

needs and the obvious danger that the use of chemical agents presented to his

psychological well-being.” Id. at 1316 (internal quotation marks omitted and

emphasis added). In our very next sentence, we explained that “[t]urning a blind eye

to such obvious danger provides ample support for the [district court’s] finding of

the requisite recklessness.” Id. (emphasis added). Finally, we held that “an

examination of [the] entire record demonstrates that the district court did not commit

clear error in finding the defendants’ deliberate indifference.” Id. at 1317 (emphasis

added).

In short, our discussion in Thomas of the subjective inquiry of the deliberate-

indifference claim repeatedly shows that we characterized and treated each of the

three components as factual ones, governed by the clearly erroneous standard of

review on appeal.

Whether each of us personally agrees or disagrees that a clear-error standard

of review is a good idea for each of the components of the subjective inquiry, see

38
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 39 of 46

Newsom Op. at 16 n.3, is irrelevant. It is beyond dispute that Thomas held that clear-

error review governs.

In contrast, the Keohane majority opinion reviewed de novo the district

court’s findings on the second and third parts of the subjective inquiry.7 See

Keohane, 952 F.3d at 1274-78. In fact, except in a footnote dismissing the notion

that clear-error review applies to each of the three components of the district court’s

factual findings on the subjective inquiry, the majority opinion never once employed

the term “clear error” in conducting its analysis. See id. And even in that footnote,

the Keohane panel opined only that even assuming the clear-error standard of review

governs review of the district court’s findings on each of the three subparts of the

subjective inquiry, “we would have little trouble formulating the required firm

conviction that a mistake had been committed.” Id. at 1272 n.8 (citation and internal

quotation marks omitted). The majority opinion said so, though, without any

corresponding analysis other than a throwaway reference to its legal-error analysis

in the text. See id.

Perhaps most disturbingly, though, despite the Keohane majority opinion’s

use of the de novo standard of review to review the second and third subparts of the

7
The Keohane majority opinion concluded that it did not need to evaluate whether the
district court correctly determined that the prison officials had actual knowledge of a risk of serious
harm because, in any case, Keohane did not establish the second and third parts of the subjective
inquiry. 952 F.3d at 1274. For that reason, the Keohane majority opinion did not expressly review
the first part of the district court’s ruling on the subjective inquiry.
39
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 40 of 46

district court’s subjective inquiry on Keohane’s deliberate-indifference claim, the

Keohane majority opinion asserted that it followed Thomas. But in support of this

proposition, the Keohane majority opinion relied solely on Thomas’s statements that

we review de novo whether “there was an Eighth Amendment violation” and that

“[s]ubsidiary issues of fact are reviewed for clear error.” See Keohane, 952 F.3d at

1265 n.2. Likewise, that is the sole “breadcrumb[]” from Thomas that the Newsom

Opinion cites in justifying Keohane’s application of de novo review to all

components of the subjective inquiry. See Newsom Op. at 11-12.

But as I have already discussed, that one “breadcrumb[]” is part of a trail that

leads inescapably to the conclusion that Thomas holds that the clear-error standard

governs the subjective inquiry.

Yet that one statement deprived of its proper context is the only thing that

Keohane and the Newsom Opinion point to from Thomas to justify Keohane’s

conclusion that Thomas required de novo review of the subjective inquiry. Indeed,

neither Keohane nor the Newsom Opinion responds to any of the numerous

quotations Judge Wilson’s dissent and I have cited from Thomas that show that

Thomas held that clear-error review governs the subjective inquiry. Keohane and

the Newsom Opinion just ignore them. But ignoring Thomas’s words does not make

Thomas’s holding go away.

40
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 41 of 46

Rather than explaining how Keohane’s holding can possibly be consistent

with Thomas’s numerous quotations, the Newsom Opinion takes a different tack: it

appears to attempt to distract the reader from its inability to demonstrate that

Keohane does not violate Thomas. Indeed, careful readers can’t help but notice that

the Newsom Opinion spends nearly all its pages trying to change the subject.

For example, it defends at length the correctness of the outcome of Keohane

and the new rule that the Keohane panel imposed contrary to Thomas’s rule. See,

e.g., Newsom Op. at 8-9 (arguing the facts of Keohane’s case—including defending

the FDC’s decision not to allow Keohane to dress and groom herself as a woman—

that have no bearing on whether Keohane followed Thomas), 16 (opining that, as a

matter of law, it makes better sense for appellate courts to “decide[] what the Eighth

Amendment ultimately means and requires in a given case” because anything else

would be “an odd state of affairs.”8), 16-19 (contending that United States v.

Bajakajian, 524 U.S. 321 (1998), Ornelas v. United States, 517 U.S. 690 (1996),

Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 532 U.S. 424 (2001)—none

of which Thomas cites, by the way—support the new Keohane rule), 19-21

(discussing Kosilek v. Spencer, 774 F.3d 63 (1st Cir. 2014), to support the notion

8
Most respectfully, I disagree that district courts are somehow not equipped to make
capable rulings on the application of the subjective-inquiry components of the Eighth Amendment
deliberate-indifference standard in any given case. And even if a district court erred, it would not
be the last word on the matter, since a party could always appeal to the circuit court. Clearly
erroneous review does not mean no review.
41
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 42 of 46

that de novo review is the correct legal answer to the standard of review that should

govern the components of the subjective inquiry9).

The Newsom Opinion also tries to divert attention from its failure to show

how it is consistent with Thomas by grading my writing and that of the Thomas

panel. See, e.g., Newsom Op. at 3, 4, 12, 21, 22.

These distraction tactics miss the point. Whether de novo review of the

components of the subjective inquiry is or is not a better answer than Thomas’s clear-

error review is not the issue here. And whether I use too many adverbs in my writing

or whether the Newsom Opinion likes how Thomas is written is similarly irrelevant

to the issue before the Court.

The only question here is whether Keohane is faithful to Thomas. The

Newsom Opinion’s failure to show how the numerous quoted statements from

Thomas can possibly be consistent with Keohane’s new rule applying the de novo

standard of review to the subjective inquiry answers that question with a resounding

“no.”10

9
Kosilek’s sole citation of Thomas for the proposition that “[t]he ultimate legal conclusion
of whether prison administrators have violated the Eighth Amendment is reviewed de novo.”
Kosilek, 774 F.3d at 84, also does not show that Thomas applied de novo review to the components
of the subjective inquiry. In fact, as I have explained in note 5, supra, it demonstrates the opposite.
So to be clear, Kosilek did not purport to read Thomas to hold that the components of the subjective
inquiry are subject to de novo review.
10
The W. Pryor Opinion’s silence on this issue likewise speaks volumes: the W. Pryor
Opinion doesn’t even try to show that Keohane is consistent with Thomas or that it didn’t violate
the prior-precedent rule. Nor does it defend Keohane’s interpretation of the prior-precedent rule,
which allows a later panel to reinvent the holding of a prior panel.
42
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 43 of 46

The Keohane majority opinion reached another conclusion only because it

viewed Thomas’s plain language to require a “mindless, mechanical box-checking

assessment.” Id. So the Keohane majority opinion reasoned that what Thomas

unambiguously said “cannot possibly be what we’ve meant[.]” Id. That language

from Keohane, in and of itself, gives up the game and implicitly concedes that the

Keohane majority opinion did not follow Thomas.

III.

Under our prior-precedent rule, it was not up to the Keohane panel to

reimagine the meaning of Thomas’s unmistakable language holding that the clearly

erroneous standard of review applies to the second and third components of the

subjective inquiry. The Keohane panel was bound by Thomas, whether it agreed

with it or not and whether it found Thomas’s standard of review to be consistent with

“meaningful appellate review” or not. Id. If the Keohane panel had a problem with

the standard of review that Thomas requires, as Judge Wilson pointed out in his

Keohane dissent, 952 F.3d at 1292 (Wilson, J., dissenting), its only option under the

prior-precedent rule was to apply the Thomas standards and call for en banc review.

There was no option to recast Thomas as having held that de novo review applies

when Thomas in fact and unmistakably held that clearly erroneous review governs.

Because the Keohane panel’s holding on the applicable standards of review

conflicts directly with the Thomas panel’s, the Keohane panel introduced conflict in

43
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 44 of 46

our precedent. Under our earliest-precedent rule, “[w]hen we have conflicting

[precedents], we follow our oldest precedent.” CSX Transp., Inc. v. Gen. Mills, Inc.,

846 F.3d 1333, 1338 (11th Cir. 2017) (citation and internal quotation marks

omitted). So the earliest-precedent rule requires later panels and district courts to

follow Thomas. But because Keohane claims to be consistent with Thomas,

Keohane purports to render the earliest-precedent rule inapplicable. To do that,

though, it violates the prior-precedent rule by failing to abide by Thomas’s true

holding imposing the clearly erroneous standard of review and by instead pretending

that Thomas sanctioned the de novo standard of review when it demonstrably did

not.

Ultimately, our refusal to hear Keohane en banc creates a mess with respect

to the current state of the law concerning the correct standards of review governing

the components of the subjective inquiry on an Eighth Amendment deliberate-

indifference claim: should district courts and later panels follow Thomas, as our

prior-precedent rule requires, or should they follow Keohane, which holds the

opposite of Thomas while claiming to have followed it?

But the real problem is that our refusal to hear Keohane en banc sows

uncertainty as to the meaning and strength of our prior-precedent rule. This may be

no big deal to the W. Pryor Opinion (though that opinion never tells us why). But

as I have noted, see supra at Section I, for good reason, we as a Court have

44
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 45 of 46

historically viewed anything that erodes the prior-precedent rule as a critical threat

to the stability and predictability of the law. See Smith, 236 F.3d at 1303; see also

Steele, 147 F.3d at 1317-18. Keohane’s rogue interpretation of the prior-precedent

rule certainly qualifies as such a threat.

If we are willing to accept Keohane as compliant with our prior-precedent

rule, then our prior precedent means only what the last panel to have reconstrued

what the deciding panel held says it means—no matter how inconsistent the most

recent panel’s interpretation may be with what the deciding panel actually held. As

a result, the practical effect is that no later panel will be bound by anything an earlier

panel said.

For these reasons, regardless of what any individual judge on this Court

believes the correct standard of review here to be, a bigger issue is at stake: the rule

of law imposes an obligation to rehear Keohane en banc and reaffirm our

“emphatic[ally]” strict adherence to the prior-precedent rule. See Steele, 147 F.3d

at 1318 (citation omitted). Then, if a majority of judges on the Court thinks Thomas

got it wrong, the Court can say so and change our precedent. But a panel cannot and

should not be allowed to do that. And a panel certainly should not be permitted to

do so by reinterpreting our prior-precedent rule to the point where it allows precisely

what it has always prohibited: a later panel to issue a holding that directly conflicts

with an earlier panel’s precedent.

45
 USCA11 Case: 18-14096 Date Filed: 12/03/2020 Page: 46 of 46

By any recognized measure, Keohane demands en banc review. We must do

better in the future.

46