<img alt="Brand" src="/images/eLaws/logo-elaws.png" class="img-responsive pull-left logo-head" />

. Craig
addressed an Oklahoma statute that outlawed the sale of 3.2% beer to young men
under the age of 21 and to young women under the age of 18, purportedly as a
means to promote traffic safety.

 at 191–92, 199,

97 S. Ct. at 454, 458

.
Considering the constitutionality of the statute, the Court first cast doubt on the
strength of the statistical evidence that young men drink and drive more frequently
than young women. See

id.

 at 200–01,

458–59. The Court concluded
that the statistical evidence did not demonstrate that gender was a “legitimate,
accurate proxy for the regulation of drinking and driving.”

Id. at 202, 204

459–60. Then, even setting aside the problematic correlation of gender and
drinking behavior, the Court observed that the statute as written did not even
prevent young men from driving under the influence. This was because the law
“prohibits only the selling of 3.2% beer to young males and not their drinking the
14
USCA11 Case: 18-13592          Date Filed: 07/14/2021       Page: 15 of 80
beverage once acquired (even after purchase by their 18-20-year-old female
companions).”

Id. at 204

. Thus, the Court decided that the statute
violated the Fourteenth Amendment because its terms did not achieve its statutory
objective.

The same is true here. We set aside for now that the policy treats
transgender students differently than non-transgender students. And we will
assume for the sake of argument that students’ privacy interests are advanced by
preventing transgender students from using the bathrooms corresponding to, like in
Mr. Adams’s case, their governmentally-recognized legal sex.3 The policy still
runs afoul of the Fourteenth Amendment because it does not even succeed in
treating all transgender students alike. It is arbitrary that some transgender
students—like Mr. Adams—are restricted by the bathroom policy, while others are
beyond its reach. Just as the statute in Craig did not prevent young men from
driving after drinking 3.2% beer, the bathroom policy fails to exclude every
transgender student from the restroom in the way the School District seeks to do.
The designation of a student’s sex on his school enrollment documents is not a
“legitimate, accurate proxy” for assigning a student to a particular bathroom to
protect student privacy. See

id. at 204

; Clark v. Jeter,

486 U.S. 3

The School District’s privacy concerns are not borne out by the record. The District
could not point to a single incident of a transgender student using a restroom acting in a manner
that invaded another student’s privacy. See also infra at 22–24.
15
USCA11 Case: 18-13592         Date Filed: 07/14/2021    Page: 16 of 80
456, 464,

108 S. Ct. 1910

, 1916 (1988) (invalidating a statute of limitations on
paternity actions under heightened scrutiny because the limitation prescribed was
“not substantially related to [the state’s] interest in avoiding the litigation of stale
or fraudulent claims”); Frontiero v. Richardson,

411 U.S. 677

, 689–91,

93 S. Ct. 1764

, 1772 (1973) (plurality opinion) (rejecting a gender-based policy as arbitrary
because the government did not show that the policy promoted “administrative
convenience” by actually saving any money or time (quotation marks omitted));
Grimm v. Gloucester Cnty. Sch. Bd.,

972 F.3d 586

, 620 (4th Cir. 2020) (Wynn, J.,
concurring) (agreeing in the judgment affirming that the challenged bathroom
policy violated equal protection, because the policy “is arbitrary and provides no
consistent reason” for assigning certain students to certain bathrooms), as amended
(Aug. 28, 2020), rehearing en banc denied,

976 F.3d 399

 (4th Cir. 2020), cert.
denied, ___ S. Ct. ___,

2021 WL 2637992

 (June 28, 2021); Whitaker, 858 F.3d at
1054 (noting that the school district’s documentation requirement behind its
challenged bathroom policy, with its internal inconsistencies, “demonstrates the
arbitrary nature of the policy”).
We also conclude the policy is impermissibly arbitrary in another way. The
policy requires, without justification, that the School District reject information on
current government records in favor of outdated information provided at the time
the student enrolled. The record shows that the School District “will not change
16
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 17 of 80
the official school records” based on any government documentation provided
after enrollment indicating the student’s sex. In other words, the policy requires
that a student’s enrollment package prevail over current government records, even
though those government-issued documents constitute controlling identification for
any other purpose. Presumably, federal and state governments allow for a process
for updating or correcting this type of personal information for a reason: to reflect
and promote accuracy.
The School District gives no explanation for why a birth certificate provided
at the time of enrollment takes priority over the same document provided at the
time the bathroom policy is applied to the student. And we have come up with no
explanation of our own. Mr. Adams has a birth certificate and a driver’s license
issued by the state of Florida stating that he is male. But the School District
refuses to accept for the purposes of the bathroom policy Mr. Adams’s sex listed
on those current government-issued documents. This kind of irrationality does not
satisfy intermediate review. See Reed,

404 U.S. at 76

92 S. Ct. at 254

 (“A
classification must be reasonable, not arbitrary.” (quotation marks omitted)); Craig,

429 U.S. at 215

97 S. Ct. at 466

 (“The [sex-based] disparity created by these
Oklahoma statutes amounts to total irrationality.”) (Stewart, J., concurring in the
judgment).
17
USCA11 Case: 18-13592          Date Filed: 07/14/2021        Page: 18 of 80
The School District failed to show a substantial, accurate relationship
between its sex classification and its stated purpose. And the Fourteenth
Amendment requires a substantial, accurate relationship between a gender-based
policy and its stated purpose. See Cleburne,

473 U.S. at 441

, 105 S. Ct. at 3255;
Craig,

429 U.S. at 198

97 S. Ct. at 457

; see also Village of Willowbrook v. Olech,

528 U.S. 562

, 564,

120 S. Ct. 1073

, 1074–75 (2000) (noting that the “purpose of
the equal protection clause of the Fourteenth Amendment is to secure . . . against
intentional and arbitrary discrimination, whether occasioned by express terms of a
[policy] or by its improper execution” (quotation marks omitted)). Because the
bathroom policy is arbitrary and does not do what it was designed to do, the School
Board cannot show the requisite substantial relationship.
Therefore, we conclude the School District’s bathroom policy violates the
Equal Protection Clause.4
4
We note that the Fourth and Seventh Circuits, the only other circuits to consider
challenges to bathroom policies brought by transgender students, along with the majority of
district courts that have addressed the issue, have also ruled that the challenged policies violate
equal protection, albeit on different or additional grounds. See Whitaker, 858 F.3d at 1051–54
(holding a transgender boy demonstrated likelihood of success on the merits of his equal
protection claim to use the boys’ restroom, because the policy was based on “sex-based
stereotypes,” because the policy was based on a privacy rationale founded on “sheer conjecture
and abstraction,” and because the policy was “arbitrary”); Grimm, 972 F.3d at 606–15 (affirming
grant of summary judgment to transgender boy on equal protection claim, because the policy
subjected him to sex discrimination for failing to “conform to the sex stereotype propagated by
the [p]olicy,” because transgender people constitute a “quasi-suspect class,” and because the
policy was “marked by misconception and prejudice” against the transgender student (quotation
marks omitted)); see also A.H. ex rel. Handling v. Minersville Area Sch. Dist.,

408 F. Supp. 3d 536

, 576–78 (M.D. Pa. 2019) (granting summary judgment to transgender girl on equal
protection claim for access to girls’ restroom, because the school district lacked an “exceedingly
18
USCA11 Case: 18-13592           Date Filed: 07/14/2021        Page: 19 of 80
D.
We now turn to the dissent. Insofar as the dissent suggests this opinion
opens the floodgates for vast societal change, it is only because the dissent has
decided to reframe the issue in this way. This case is only about Mr. Adams’s as-
applied challenge to the School District’s policy denying him the ability to use the
boys’ bathroom at Nease High School. The dissent also spills much ink over the
former (now vacated) panel opinion. The majority of the pages in the dissent are
directed at an opinion no longer in existence. 5 Indeed much of the dissent
continues to shadowbox with an opinion we never wrote. We view the dissent’s
recycling of outdated arguments as an apt metaphor for its analytical approach.
We begin by setting the record straight on a few issues. The dissent argues
that “relying on enrollment documents is not an arbitrary means of determining
sex.” Dissenting Op. at 41 (cleaned up). We understand the dissent to be making
three subsidiary claims in support of that assertion. First, the dissent says our
persuasive justification for th[e] sex-based discrimination”); Evancho v. Pine-Richland Sch.
Dist.,

237 F. Supp. 3d 267

, 293 (W.D. Pa. 2017) (holding transgender students showed
likelihood of success on equal protection claim to access restrooms consistent with their gender
identity, because “there is insufficient record evidence of any actual threat to any legitimate
privacy interests of any student by the [students’] use of [common] restrooms consistent with
their gender identity”); Bd. of Educ. of the Highland Local Sch. Dist. v. U.S. Dep’t of Educ.,

208 F. Supp. 3d 850

, 877 (S.D. Ohio 2016) (holding transgender girl showed likelihood of success on
equal protection claim to access bathroom matching her gender identity, because the defendants
failed to show that the “discriminatory policy is substantially related to their interests in privacy
or safety” and because the policy rested on hostility toward transgender students).
5
Specifically, only the first sixteen pages of the dissenting opinion address the now-
vacated majority opinion. See infra at 39–54.
19
USCA11 Case: 18-13592       Date Filed: 07/14/2021   Page: 20 of 80
opinion rules on the constitutionality of sex-segregated bathrooms and will have
far-reaching consequences.

Id.

 at 39–41, 53–54. Second, the dissent says the
challenged policy “does not exist” and that we misconstrue the parties’ arguments.

Id.

 at 48–53. Third, the dissent says the challenged policy is not arbitrary because
it is largely “accurate.”

Id.

at 42–48. None of these arguments is availing. We
address each in turn.
1. The Dissent Is Wrong to Say This Ruling Addresses the Constitutionality of
Sex-Segregated Bathrooms or Any Other Issue Involving Student Privacy
Contrary to the dissent’s assertion, this case is not about challenging sex-
segregated bathrooms. Mr. Adams does not challenge or even question the
ubiquitous societal practice of separate bathrooms for men and women. See
Appellee’s Br. at 41. Everyone in this case, including Mr. Adams, “agrees that
boys should use the boys’ restroom at Nease and that girls should use the girls’
restroom.” The School Board itself stated this point in its brief before us: “Indeed,
Adams did not challenge the School Board’s ability to separate boys and girls into
different bathrooms on the basis of sex, and the District Court did not hold that
such separation was impermissible.” Appellant’s Br. at 15. The District Court
emphasized—and we do as well—that the ruling in Mr. Adams’s case “will not
integrate the restrooms between the sexes,” because there is “no evidence to
suggest that [Adams’s] identity as a boy is any less consistent, persistent and
20
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 21 of 80
insistent than any other boy.” R. Doc. 192 at 47. 6
The dissent’s reliance on dictionary definitions, see Dissenting Op. at 42,
fails to provide an adequate response to the policy’s arbitrariness. Dictionaries
naturally capture the differing usages of the term “sex.” See, e.g., Sex, Black’s
Law Dictionary (11th ed. 2019) (“1. The sum of the peculiarities of structure and
function that distinguish a male from a female organism; gender.”). But we mean
“sex” the way the state of Florida does. And as a court of law, we cannot simply
ignore the legal definition of sex the state has already provided us, as reflected in
the official documentation of Mr. Adams’s sex as male on his driver’s license and
birth certificate. Contra Dissenting Op. at 41–42. In any event, the ruling in this
opinion does not turn on which definition of sex one adopts—ours or the
dissent’s—because the School District’s policy is arbitrary either way. As we’ve
6
It is unfortunate that the dissent appears to liken Mr. Adams’s use of the boys’
bathroom as somehow comparable to people who “commit tax fraud, evade the draft, and report
inaccurate information on the census,” Dissenting Op. at 44—in short, those who lie, deceive,
and demonstrate “evasion.”

Id.

Mr. Adams has been nothing but honest. Mr. Adams says he is
a boy, and he has the legal documentation to support it. He has plainly and forthrightly self-
reported. Even if one does not take Mr. Adams at his word, surely one must defer to the status
given him by the state of Florida.
In response, the dissent goes so far as to say that Mr. Adams’s government-issued
documents provide “inaccurate information.” Id. at 47. The dissent again ignores that the
School District’s policy itself accepts and relies on government-issued documentation provided
at the time of enrollment. Meanwhile, the dissent argues without any apparent trace of irony that
a system of “self-reporting” and relying on “supporting legal documentation” should be an
“accurate method” for the schools to assign bathroom use by sex. Id. at 43–44, 48.
In other words, the dissent appears to say both that government-issued documents are
“inaccurate” while saying elsewhere that “legal documentation” is “accurate.” We submit that
the contradictory positions taken by the dissent are reflective of the policy’s arbitrariness.
21
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 22 of 80
explained, the policy fails to treat even all transgender students within the School
District the same. See supra at 12–16. The policy turns solely on the information
provided at the time of enrollment, and a transgender student who updates his
documents prior to enrollment would not be barred from using the bathroom
matching the sex on his legal documents. This, of course, is in contrast to the
treatment Mr. Adams received.
Despite the dissent’s imagined parade of horribles, this opinion does not
resolve any other issue of student privacy. First, we reiterate that this policy does
not do more than rule on whether Mr. Adams has shown that the policy was
arbitrary as applied to him. Second, we reject the dissent’s characterization of our
opinion as not acknowledging students’ interest in “avoiding the exposure of their
bodies to members of the opposite sex.” Dissenting Op. at 67. To the contrary, we
recognize student privacy interests as “[u]ndoubtedly . . . important.” See supra at
11. However, the point here is that the School District has never shown how its
policy furthers that interest, as this record nowhere indicates that there has ever
been any kind of “exposure” in the bathrooms at Nease High School, which all
contain separate stalls with doors that close and lock. See R. Doc. 151-17 at 12
(School Board admitting that “all multi-user boys’ restrooms and girls’ restrooms
at Nease High School have one or more stalls in them with doors that close and
lock”; that “all students who use a girls’ restroom at Nease High School must use a
22
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 23 of 80
stall when relieving themselves”; and that “any student who desires more privacy
can use a single-user restroom”); R. Doc. 137 at 26 (when using the restroom, Mr.
Adams “uses a stall to relieve himself, washes his hands, and leaves”). Again,
nothing in the record suggests Mr. Adams or any other transgender student ever
threatened another student’s privacy. Indeed, the School District confirmed it was
unaware of a single negative incident involving a transgender student using a
restroom, even as Mr. Adams used the boys’ bathroom for several uneventful
weeks. R. Doc. 162 at 16; R. Doc. 192 at 25. And at a recess during proceedings
before the District Court, at which Mr. Adams testified as a witness, Adams used
the men’s bathroom in the courthouse without incident.
Next, despite the dissent’s apparent obsession with locker rooms, see
Dissenting Op. at 41, 53–54, 65, this case is not a locker room case. Mr. Adams
did not bring a claim for access to the boys’ locker rooms at Nease High School
and has never asserted he is entitled to such access. This record shows that Mr.
Adams did not register for physical education classes; Nease High School locker
rooms are only available to students taking physical education classes; and, even
so, “no student at Nease High School is required to shower after physical education
classes.” We offer no opinion on any claims relating to locker rooms, which,
contrary to the dissent’s stark warnings, would entail a separate analysis of the
means-ends fit in light of the particular interests at stake. The dissent simply
23
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 24 of 80
manufactures a problem where none exists.
Our ruling is a narrow one that addresses only whether Mr. Adams’s as-
applied challenge to the School Board’s unwritten bathroom policy survives
intermediate scrutiny. In arguing that this opinion reaches further than it actually
does, the dissent relies on sheer conjecture and a rewriting of the record. But the
Supreme Court has reiterated time and time again that a sex-based “classification
must substantially serve an important governmental interest today,” not in some
past or hypothetical world. Sessions v. Morales-Santana, 582 U.S. __,

137 S. Ct. 1678

, 1690 (2017). In its attempts to meet this test, the dissent is forced to rewrite
the facts, because the evidence actually before us shows that Mr. Adams had used
the boys’ bathroom at Nease High School without any problems: student privacy,
safety, or otherwise.
2. The Dissent Wrongly Claims the Challenged Policy “Does Not Exist” and
Misconstrues the Parties’ Arguments
The dissent newly takes the position that a “policy does not exist” that defers
to enrollment forms. Dissenting Op. at 48. The dissent’s rewriting of the policy at
hand is inconsistent with both its prior position in this case and the record. The
dissent previously recognized the factual basis describing how the policy works.
See Previous Dissenting Op. at 58 (recounting without factual disagreement that
“the school district determines each child’s sex by looking to the enrollment forms
that the student provides when the student enrolls, which includes the student’s
24
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 25 of 80
birth certificate” and that “a transgender student who changed the sex on his birth
certificate before enrolling could use the bathroom matching” that updated
information).7
7
We also pause to correct the record on another distortion by the dissent. The dissent
says the School District bases its bathroom assignments on “a pre-enrollment physical
examination performed by the student’s doctor.” Dissenting Op. at 43. There are at least two
things wrong with the dissent’s assertion. First, there is no record support for the dissent’s
apparent assumption that Mr. Adams—or any student in the School District—is required to
provide a medical report of a physical examination. The dissent relies on Florida Statute
§ 1003.22(1), but that statute merely requires that “each child . . . present a certification of a
school-entry health examination performed within 1 year before enrollment in school.”
(Emphasis added). Here, the record shows that the School District merely tells enrolling students
to comply with Florida law’s requirement to undergo a “health examination” and self-report their
sex by writing it into a box on the enrollment health form. The record shows nothing more.
Instead, the evidence shows that self-reporting of medical information is all that is required. As
the School District’s lawyer who advised it on its bathroom policy testified, “As a statutory
condition of enrollment, a student is required to have a physical conducted by a doctor. So that
report is part of the enrollment package.” R. Doc. 162 at 50 (emphasis added); see also R. Doc.
192 at 4 n.6 (identifying Frank D. Upchurch as a long-time School Board attorney who advised
the Board on its “Best Practices Guidelines” and “who is well familiar with School Board
policies”). The “report” the School Board’s lawyer spoke of is a self-provided account of the
student’s medical history, which, when it comes to sex, requires only that a student self-report
that information. See R. Doc. 162 at 50 (Upchurch testifying: “There is a -- two boxes on that --
sort of the cover sheet of the form. And it says M/F. And the student checks -- the student’s
parent checks one.”). The dissent can point to no evidence saying otherwise.
Second, there is no record support for the dissent’s apparent assumption that the School
District actually bases its bathroom assignments on a health examination, as opposed to the self-
reporting of the student’s sex in the enrollment documents. The record merely shows that the
School District’s lawyer “speculated” that students could have had a physical, but he never said
the School Board enforces its bathroom policy using information from a physical. In any event,
we cannot discern from this record how the dissent got the idea that a certification of a routine
“health examination,” which is all that is required under Florida law, requires that a doctor
examine, much less report to a school about, a child’s genitalia.
We think the School District’s “health exam form speaks for itself, and contains no
indicator” for a doctor’s examination; “it simply includes a blank box for the child’s sex.” It is
undisputed that the School District does not inspect students’ anatomy before they use District
bathrooms. The School District relies solely on the information provided by the students to
enforce its bathroom policy, and the dissent has not pointed to any evidence to the contrary.
Rather than accept the record, the dissent continues to push its reimagining of the facts, which
even the School District did not advance. See Oral Arg. Recording at 5:43–6:22 (Dec. 5, 2019)
(Chief Judge Pryor: “I thought that there was also a state-mandated physical examination for
25
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 26 of 80
The dissent’s newfound view that the policy does not “exist” also ignores
the extensive work the District Court undertook to clarify the policy at issue. The
court documented, for instance, that it had been “repeatedly told that it was an
unwritten policy that prohibited [Mr. Adams] from using the boys’ bathroom,”
and, in order to hold the School Board liable, the court needed to know if the
Board stood behind that unwritten policy as its official policy. R. Doc. 198 at 8.
For that reason, the District Court requested that the chair of the School Board
appear in court, and asked him: “Is the rule that prohibits Drew Adams from using
the boys’ bathrooms at Nease High School the official policy of The School Board
of St. Johns County, Florida?” Id. at 11–12. The School Board chair responded,
“Yes, it is, Your Honor.” Id. at 12; see also id. at 82 (the District Court noting “the
school board has today, in open court, adopted” the unwritten policy). 8 Rather
enrolling students. Is that not true? And that that would also be evaluated as part of the policy?”
School District: “Your Honor, I’m not certain on that point. I’m not certain if that was part of
the record or not.” Chief Judge Pryor: “At trial Frank Upchurch testified, I thought, in addition
to the birth certificates, that the School Board also uses the state-mandated physical examination.
You don’t recall that?” School District: “No, Your Honor, I don’t recall that.”).
8
The School Board confirmed that the policy separates students solely on the basis of
information provided at the time of their enrollment. One discussion between the District Court
and Terry Harmon, counsel for the School Board, is particularly instructive:
District Court: “I thought that when [previous witnesses] were asked
what the reason for the policy was, I thought they cited things like
safety, privacy.”
Counsel: “The unwritten policy . . . has to do with a boy and what
bathroom a boy can use and a girl and what bathroom a girl can use.”
District Court: “[I]f you’re saying the whole policy is boys have to
use the boys’ restroom and girls have to use the girls’ restroom, the
26
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 27 of 80
than engage with the legality of the policy, the dissent rewrites the record and
argues against a policy that never existed.
The dissent ignores the record in an apparent attempt to save the School
District from itself. Dissenting Op. at 44–45. The School District told the District
Court and us on appeal that its policy ties a child’s bathroom designation to their
enrollment documents. And the School District repeatedly conceded that its policy
would allow a transgender male student to use the boys’ bathroom as long as he
provided documents at the time of enrollment that reflected his sex as male. The
dissent says, however, that these repeated admissions are of no moment because
the School Board also said they would “re-examine” the reliance on the “self-
identified” enrollment information if “abiding by [that] data became a problem.”
St. Johns County School Board has to be making a value judgment,
based on something, that Drew Adams is a girl, because otherwise
they’d let him use the boys’ bathroom, right?”
Counsel: “Yeah, and the value judgment comes in the form of the
enrollment materials, which is, what are you when you enroll in the
school district?”
District Court: “All right. Well, let’s talk about that. . . . What
happens if the person’s already transitioned before they come to
your school and so the paperwork says boy? . . . I think [a School
Board witness] said, ‘Well, then we would treat him like a boy until
we had reason not to,’ or something like that.”
Counsel: “Yeah. That’s correct.”
District Court: “But I’m not exactly sure what that means. So if the
value judgment is made at the time of the enrollment -- and what is
it based on? It’s based on --”
Counsel: “The enrollment material.”
R. Doc. 198 at 5, 82–83 (quotation marks omitted).
27
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 28 of 80
See id. But the School Board’s statement that it would seek to revisit the
arbitrariness of a policy if squarely confronted with it simply proves the point. A
policy that fails to accomplish what it purports to do is the definition of arbitrary.
See Craig,

429 U.S. at

208–09,

97 S. Ct. at 463

(“[T]he principles embodied in the
Equal Protection Clause are not to be rendered inapplicable by statistically
measured but loose-fitting generalities concerning the [relevant] tendencies of
aggregate groups.”). And an arbitrary policy that is left to the whims of the
enforcer to correct is no less arbitrary. Though the School Board says it would
reexamine its policy once its arbitrariness becomes overwhelmingly obvious, we
also know the Board is not actively collecting the information that would allow it
to make that determination. As the School District’s witness testified: “The
district does not play bathroom cop.” In this case, the School District was alerted
to Mr. Adams’s otherwise uneventful use of the boy’s bathroom, not because any
boy or any boy’s parents complained. It only took action as a result of two
unidentified female students reporting that they had seen Mr. Adams entering the
boys’ restroom. But those female students were not asserting a privacy interest.
Nor could they. Mr. Adams was using the boys’ bathroom and not the bathroom
used by the girls. The School District even confirmed that neither of the female
students expressed privacy or safety concerns. R. Doc. 162 at 16–17. Forcing
students to suffer under an arbitrary policy until the School District decides, if at
28
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 29 of 80
all, to “re-examine” the policy is precisely the kind of arbitrary harm equal
protection prohibits.
The dissent also claims we “strike[] out on [our] own” and set aside the
parties’ presentation of the issues. Dissenting Op. at 50. But it is the dissent that
eschews party presentation by dispensing with the record and wrestling instead
with hypotheticals and new and entirely separate issues. Here, Mr. Adams has
consistently argued that the challenged policy violates the Equal Protection Clause
because it does not substantially fit with the School District’s purported goals. See
R. Doc. 137 at 10, 16–17 (Mr. Adams noting that he was instructed not to use the
boys’ bathroom “pursuant to an unwritten policy, and a written set of guidelines
entitled [‘Best Practices’]”; that the policy is based on students’ information
“designated in their enrollment paperwork, in the student’s birth certificate, and in
other school records”; that “[t]he school does not undertake any protocol or effort
to verify the student’s sex as it appears in the enrollment paperwork and the
student’s records”; and that “[t]he District does not routinely keep records of, or
ask students to identify, their chromosomes, external sex organs, or internal sex
organs”); see also R. Doc. 116 at 2 (Mr. Adams stating in the joint pretrial
submission that he challenged the policy that was based in part “on a Best
Practices Guideline”); Appellee’s Br. at 34–45 (arguing the policy excluding him
from the boys’ bathroom failed intermediate scrutiny). That the policy enforced
29
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 30 of 80
against Mr. Adams was never written down does not make it any less arbitrary—
and indeed may make it even more so. Cf. Whitaker, 858 F.3d at 1039 n.2 (“We
will refer to the School District’s decision to deny [the plaintiff] access to the boys’
restroom as a ‘policy,’ although any such ‘policy’ is unwritten and its exact
boundaries are unclear.”).
We relied on this reasoning regarding this issue in our now-vacated opinion
as well, see Previous Majority Op. at 15–18, and the dissent to that opinion never
claimed the issue was not presented. See generally Previous Dissenting Op. As
further indication that the issue we address on appeal was fully presented, we note
that both the District Court and this panel asked the School Board about the
arbitrariness of the policy, providing ample notice of the issue and affording the
School District ample opportunity to rebut it. See R. Doc. 198 at 82–83 (District
Court asking what happens if a person has already transitioned before enrolling in
the School District); Oral Arg. Recording at 12:01–12:43 (Dec. 5, 2019) (“There’s
. . . a bit of arbitrariness, because depending on when somebody arrives at your
school, they are treated differently . . . based on how far into the transition process
they are.”). But it did not. The School Board confirmed its unwritten bathroom
policy relies solely on the information provided by a student at the time of
enrollment. R. Doc. 198 at 83–84 (confirming this is how the policy works); Oral
Arg. Recording at 12:40–12:43 (Dec. 5, 2019) (same); see also id. at 4:34–4:52
30
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 31 of 80
(“Let’s suppose there was a transgender boy . . . who transfers into the school
district, moves from another state, in ninth grade, with the box checked as ‘M.’
The School District would accept that he is a boy for the purposes of restroom
usage, is that right?” School District: “That’s correct.”). It is not right for the
dissent to now make arguments the School Board never made for itself by claiming
the issues being decided today were not properly teed up. They were. They were
addressed in this litigation repeatedly through the standard course of record
development, briefing, and oral argument. Today’s ruling falls squarely within the
ordinary course of appellate review, and we may undoubtedly affirm the District
Court’s ruling based on the Equal Protection Clause after our independent review
of the law. See Tartell, 790 F.3d at 1257.
3. The Dissenting Opinion Wrongly Relies on Statistics for Its Equal Protection
Analysis
The dissent’s insistence that the policy passes heightened review because it
is “accurate,” Dissenting Op. at 45, misapprehends intermediate scrutiny.9 The
9
The dissent agrees that intermediate scrutiny applies if a policy treats people differently
on the basis of sex. Dissenting Op. at 50. To the extent the dissent now claims that rational
basis review applies because “the mere act of determining an individual’s sex, using the same
rubric for both sexes, does not treat anyone differently on the basis of sex,” id., the dissent
misunderstands the policy, which does more than “determine” an individual’s sex. It assigns
students to bathrooms on the basis of sex using the information provided at the time of
enrollment. Intermediate scrutiny undoubtedly applies in this case. Neither party has said
otherwise. As the School Board said, “Here, intermediate scrutiny applies.” Appellant’s Brief at
21; see also Oral Arg. Recording at 3:25–3:56 (Dec. 5, 2019) (Chief Judge Pryor: “Segregating
the bathrooms on the basis of sex is a sex-based classification, right?” School District: “Sure.
Sure, it is.” Chief Judge Pryor: “So the question is, . . . can that classification survive, for the
constitutional claim, can it survive intermediate scrutiny, right?” School District: “Correct, Your
31
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 32 of 80
dissent suggests the constitutionality of a policy turns on the soundness of the
statistical correlation supporting it, when the Supreme Court in Craig, following a
long line of precedent, rejected that precise premise.
In Craig, the Supreme Court invalidated a sex-based regulation despite the
fact that men in the relevant age group were over 11 times more likely than women
of the same age group to be arrested for driving under the influence of alcohol.

429 U.S. at 201

97 S. Ct. at 459

. Other statistics recounted in Craig showed that
78% of drivers in a relevant geographic area who were under the age of 20 were
male, and 84% of men expressed a preference for beer.

 at 203 n.16,

460 n.16. This too was not enough to save the challenged regulation. The Court
warned of the regulations that would be blessed “if statistics were to govern the
permissibility of state [action] without regard to the Equal [P]rotection Clause as a
limiting principle.”

 at 208 n.22,

463 n.22. “Indeed,” the Court
wrote, “prior cases have consistently rejected the use of sex as a decisionmaking
factor even though the statutes in question certainly rested on far more predictive
empirical relationships than” the one in Craig.

Id. at 202

459
Honor.”). Nor did the previous dissenting opinion ever raise the possibility of applying rational
basis review, even though our previous majority opinion stated exactly the same rationale in
determining the bathroom policy arbitrary. Compare Previous Majority Op. at 15–18, with
Previous Dissenting Op. at 47 (applying intermediate scrutiny).
In any event, the dissent cannot now claim on the one hand that the majority’s holding on
the School Board’s policy somehow decides the constitutionality of sex-segregated bathrooms
(which this appeal does not), while also asserting that we should be applying rational basis
review. The dissent cannot have it both ways. See Dissenting Op. at 50.
32
USCA11 Case: 18-13592       Date Filed: 07/14/2021   Page: 33 of 80
(emphasis added). For instance, the Court noted that in Reed, Idaho’s “premise
that women lacked experience in formal business matters (particularly compared to
men) would have proved to be accurate” in many cases, particularly in 1967 when
the facts giving rise to Reed came to pass.

 at 202 n.13;

459 n.13;
see also Reed,

404 U.S. at

76–77,

92 S. Ct. at 254

 (invalidating provision of Idaho
code reflecting the kind of “arbitrary legislative choice forbidden by the Equal
Protection Clause,” despite whatever statistical accuracy underlay it).
Supreme Court decisions “following Reed similarly have rejected
administrative ease and convenience as sufficiently important objectives to justify
gender-based classifications.” Craig,

429 U.S. at 198

97 S. Ct. at 457

. As Craig
recognized, the Supreme Court had in other Equal Protection cases “expressly
found appellees’ empirical defense . . . unsatisfactory.” Craig,

429 U.S. at

202
n.13,

459 n.13 (citing Frontiero,

411 U.S. at

688–91,

93 S. Ct. at

1771–
72; Weinberger v. Wiesenfeld,

420 U.S. 636

, 645,

95 S. Ct. 1225

, 1231–32
(1975)). Thus the Court invalidated sex-based regulations despite empirical
assertions of “the financial position of servicewomen and working women” even
before Craig.

 at 198–99,

457–58 (citation omitted); see also
Frontiero,

411 U.S. at

688–89,

93 S. Ct. at 1771

 (recounting how, in 1973, “the
Government maintain[ed] that, as an empirical matter, wives in our society
frequently are dependent upon their husbands, while husbands rarely are dependent
33
USCA11 Case: 18-13592         Date Filed: 07/14/2021        Page: 34 of 80
upon their wives”).
The Supreme Court in Craig acknowledged that statistics and numbers have
surface appeal. But the Court also knew it was “unrealistic to expect either
members of the judiciary or state officials to be well versed in the rigors of
experimental or statistical technique.”

429 U.S. at 204

. The easy
manipulation of statistics by parties, and the possibility of being fooled by them,
“illustrate[] that proving broad sociological propositions by statistics is a dubious
business, and one that inevitably is in tension with the normative philosophy that
underlies the Equal Protection Clause.”

We view the dissent’s reliance on a figure of 99.96 percent as a prime
example of the folly of relying on statistics in equal protection analysis.
Dissenting Op. at 44–47, 58.10 The relevant inquiry in this case is not what
percentage of St. Johns students are transgender, but whether the challenged policy
furthers the important goal of student privacy. And on that point, the dissent offers
no statistics. Nor could it. When it comes to actual reported invasions of privacy
10
The dissent defends its reliance on its 99.96 percent figure by pointing to instances
where the Supreme Court recounted statistics outside of the context of an Equal Protection
analysis. These cases are inapposite. See Dissenting Op. at 46–47 (citing Birchfield v. North
Dakota, 579 U.S. __,

136 S. Ct. 2160

, 2193 n.10 (2016) (Sotomayor, J., concurring in part and
dissenting in part) (writing in a Fourth Amendment case); Cent. R.R. Co. of Pa. v. Pennsylvania,

370 U.S. 607

, 610 n.2,

82 S. Ct. 1297

, 1301 n.2 (1962) (recounting the number of railroad cars
located on out-of-state railroads in the facts section of a case concerning capital stock tax)). No
one is saying members of the judiciary are unable to do simple math. But the Supreme Court has
time and again warned against using arithmetic to supply pat resolutions to sex discrimination
cases.
34
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 35 of 80
in restrooms by a transgender student, the School Board conceded there were zero
such incidents. Thus, there is no evidence of a correlation between the incidents of
invasions of privacy at St. Johns School District and the sex of the student. If
anything, the dissent’s insistence that 99.96 percent of students in St. Johns are not,
like Mr. Adams, transgender, simply underscores how limited the ruling today is,
which again applies only to Adams.
IV
We affirm the District Court’s award of compensatory damages because
there is no doubt Mr. Adams suffered harm. Medical professionals explain that
many transgender people experience debilitating distress and anxiety from gender
dysphoria, which is alleviated by using restrooms consistent with their gender
identity. Medical opinion also concludes that “forc[ing] transgender people to live
in accordance with the sex assigned to them at birth” both “fail[s]” to change
transgender people from who they are and “cause[s] significant harm.”
True to medical opinion, Mr. Adams described that he suffered anxiety and
depression from walking past the boys’ restroom on his way to the single-stall
bathroom, knowing every other boy is permitted to use it but him. He testified to
feeling “alienated” and “humiliat[ed]” as a result of the policy. Mr. Adams
testified that, because of the policy, “I know that the school sees me as less of a
35
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 36 of 80
person, less of a boy, certainly, than my peers.” This is so, despite as the District
Court noted in its findings of fact supporting damages, that Mr. Adams
has undergone social, medical, and legal transitions to
present himself as a boy. Adams wears his hair short; he
dresses like a boy; his voice is deeper than a girl’s; his
family, peers, classmates and teachers use male pronouns
to refer to him; he takes hormones which suppress
menstruation and make his body more masculine,
including the development of facial hair and typical male
muscle development; he has had a double mastectomy so
his body looks more like a boy; the state of Florida has
provided him with a birth certificate and driver’s license
which state he is a male; and when out in public, Adams
uses the men’s restroom.
Adams, 318 F. Supp. 3d at 1326.11
Mr. Adams also suffered harm because he was separated from his peers in
single-stall restroom facilities. Mr. Adams had little choice in this matter.
The evidence at trial indicated that using the girls’ restroom at school hindered Mr.
Adams’s clinical treatment for gender dysphoria. As such, because of the
bathroom policy, Mr. Adams had to use single-stall restrooms at school. 12 Mr.
Adams explained it felt like a “walk of shame” when he had to walk past the
11
The dissent appears to overlook the fact that the District Court’s damages award was
based on its review of the full record developed at trial, which includes recognition of Mr.
Adams’s social, medical, and legal transition. See Dissenting Op. at 49.
12
The evidence at trial also showed that the School District expected this outcome. The
School District official who developed the bathroom policy testified that she would prefer a
transgender student to use the gender-neutral single-stall facilities instead of the restroom of his
sex assigned at birth. This employee also acknowledged that a transgender student who, like Mr.
Adams, presents as a boy, could face “safety, security, and privacy concerns” while using the
restroom of his sex assigned at birth.
36
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 37 of 80
communal restrooms for a single-stall, gender-neutral restroom. It heightened the
stigma he felt for being transgender.
Based on this evidence, the District Court found that Mr. Adams “suffered
emotional damage, stigmatization and shame from not being permitted to use the
boys’ restroom at school.” Adams, 318 F. Supp. 3d at 1327. The consequences
Mr. Adams suffered are well-recognized as injurious. See Whitaker, 858 F.3d at
1045–47 (affirming a finding of irreparable harm because excluding a transgender
student from the boys’ restroom “stigmatized” the student and caused him
“significant psychological distress” including “depression and anxiety” (quotation
marks omitted)); Dodds v. U.S. Dep’t of Educ.,

845 F.3d 217

, 221–22 (6th Cir.
2016) (per curiam) (affirming a finding of irreparable harm because excluding a
young transgender student “from the girls’ restrooms has already had substantial
and immediate adverse effects on the daily life of an eleven-year-old child (i.e.
multiple suicide attempts prior to entry of the injunction)”).
V
This record demonstrates that the School Board has not met its “demanding”
constitutional burden by showing a substantial relationship between its policy for
excluding transgender students from certain restrooms and student privacy. See
VMI,

116 S. Ct. at 2275

. We therefore affirm the District Court’s
grant of relief to Mr. Adams under the Fourteenth Amendment.
37
USCA11 Case: 18-13592         Date Filed: 07/14/2021   Page: 38 of 80
Having concluded that Mr. Adams prevails on his equal protection claim,
which fully entitles him to the relief granted by the District Court, see Adams, 318
F. Supp. 3d at 1326–27 & n.58, we decline to reach his Title IX claim.
*        *     *
The record developed in the District Court shows that the School Board
failed to honor Mr. Adams’s rights under the Fourteenth Amendment. The
judgment of the District Court is therefore AFFIRMED.
38
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 39 of 80
WILLIAM PRYOR, Chief Judge, dissenting:
Not long ago, a suit challenging the lawfulness of separating bathrooms on
the basis of sex would have been unthinkable. This practice has long been the
common-sense example of an acceptable classification on the basis of sex. And for
good reason: it protects well-established privacy interests in using the bathroom
away from the opposite sex. Although the Supreme Court recently considered the
relationship between transgender status and sex in the context of claims of
employment discrimination under Title VII, it declined to consider the
permissibility of sex-separated bathrooms. See Bostock v. Clayton County,

140 S. Ct. 1731

, 1753 (2020); see also id. at 1739 (assuming that “sex” refers “only to
biological distinctions between male and female”). After all, context matters. As
the late Justice Thurgood Marshall once put it, “A sign that says ‘men only’ looks
very different on a bathroom door than a courthouse door.” City of Cleburne v.
Cleburne Living Ctr., Inc.,

473 U.S. 432

, 468–69 (1985) (Marshall, J., concurring
in the judgment in part and dissenting in part).
Against this backdrop, the St. Johns County School Board has long enforced
a policy that separates the bathrooms in its elementary and secondary schools by
sex. Last year, the majority issued an opinion ruling that this policy violated
federal law regulating schools and was unconstitutional. As I pointed out then, its
opinion distorted the policy, misunderstood the legal claims asserted, and rewrote
39
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 40 of 80
well-established precedent. The majority now tacitly concedes that its opinion
could not withstand scrutiny.
The majority accordingly has withdrawn its earlier opinion. But its new
opinion fares no better than the last. It distorts the challenged policy in a brand-
new way, and it invents a legal claim the parties never presented. And its new
opinion still fails to identify a violation of the law.
The majority’s new position is as wrong as its old position. I first explain the
errors in its revised opinion. The parties litigated the case as a challenge to the
lawfulness of sex-segregated bathrooms, and the district court decided it on that
basis. But on appeal, the majority identifies a new policy as unconstitutional: the
schools’ practice of relying on self-reported sex and supporting documentation to
determine the sex of its students. The majority’s new reasoning relies entirely on
eliding that practice and the policy Adams actually challenged—the policy of
separating bathrooms by sex.
I then explain why the majority is right to retreat from its previous position.
By failing to address head-on the lawfulness of sex-separated bathrooms in
schools, the majority recasts the school policy as classifying students on the basis
of transgender status. And based on this recasting, it reached the remarkable
conclusion that schoolchildren have no sex-specific privacy interests when using
the bathroom.
40
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 41 of 80
The new majority opinion continues its earlier pretense that its reasoning
applies only to plaintiff Drew Adams and that it does not decide that sex-separated
locker rooms are unconstitutional. But the majority’s assurances ring hollow. The
logic of this decision, no different from the last opinion the majority issued, would
require all schoolchildren to use sex-neutral bathrooms and locker rooms. I dissent.
A. Relying on Enrollment Documents Is Not an Arbitrary Means of
Determining Sex.
The new majority opinion identifies a single problem with the actions of the
schools. It concludes that Adams should prevail because the policy of “assigning
students to bathrooms based solely on the documents the School District receives
at the time of enrollment” is arbitrary and fails to advance the schools’ interests in
privacy. Majority Op. at 12. It reaches this conclusion because a student could
evade the sex-separated bathroom policy by enrolling as a member of the opposite
sex, and because, “without justification,” the schools “reject information on current
government records in favor of outdated information.”

at 15–17. The majority
frames the issue incorrectly in at least two ways.
For one thing, the majority follows Adams in using the word “sex” as a
synonym for “gender identity.” E.g., id. at 6 (“Mr. Adams amended his legal
documents to reflect his male sex.”). As Adams put it, “[t]ransgender persons are
people whose gender identity diverges from the sex they were assigned at birth,”
and a “transgender boy’s sex is male (even though he was assigned the sex of
41
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 42 of 80
female at birth).” But the schools use “sex” in its ordinary, traditional sense, and
they use “biological sex” as a synonym. See, e.g., Sex, The American Heritage
Dictionary of the English Language (New College ed. 1979) (“The property or
quality by which organisms are classified according to their reproductive
functions.”); Sex, The American Heritage Dictionary of the English Language (4th
ed. 2006) (similar); Sex, The Random House College Dictionary (rev. ed. 1980)
(“either the male or female division of a species, esp. as differentiated with
reference to the reproductive functions”); Sex, Shorter Oxford English Dictionary
(6th ed. 2007) (“Either of the two main divisions (male and female) into which
many organisms are placed on the basis of their reproductive functions or
capacities”); see also Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of
Mental Disorders 451 (5th ed. 2013) (“This chapter employs constructs and terms
as they are widely used by clinicians from various disciplines with specialization in
this area. In this chapter, sex and sexual refer to the biological indicators of male
and female (understood in the context of reproductive capacity) . . . .”). I too use
the word “sex” in its ordinary, traditional sense.
The majority’s linguistic sleight of hand plays into its second framing
mistake. It misunderstands what the schools ascertain from enrollment documents.
The schools determine students’ sex based on what the students report on their
enrollment forms and on the supporting documentation the students submit in the
42
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 43 of 80
form of birth certificates or other government-issued identification, and a pre-
enrollment physical examination performed by the student’s doctor. See

Fla. Stat. § 1003.22

(1) (“Each district school board . . . shall require that each child who is
entitled to admittance . . . present a certification of a school-entry health
examination performed within 1 year before enrollment in school.”). The schools
then require each student to use the restroom matching the sex determined through
that process. They do not accept new documents to supersede the documents
provided at the time of enrollment.
There is nothing wrong with the schools’ approach. No evidence suggested
the schools’ method of identifying the sex of their students has ever produced an
inaccurate result. Indeed, school officials confirmed that their method of
determining students’ sex has not been a problem, and that they would change their
method if it did not work well. More generally, self-reporting is often a reliable
method for gathering information. It is good enough for filing tax returns, selective
service registration, and the census. And the schools supplement students’ self-
reported information by requiring support in the form of official documents.
Nor is there any reason for the schools to accept updates. A student’s sex
does not come with an expiration date, and it does not require periodic updates to
confirm its continuing accuracy. The object of the schools’ practice with respect to
the enrollment documents is to determine students’ sex, not their gender identity.
43
USCA11 Case: 18-13592        Date Filed: 07/14/2021   Page: 44 of 80
Even Adams admitted that the schools sought to sort students in this manner. In a
state that allows a minor to conform legal documents to the minor’s gender
identity, as Adams did, accepting updated documents would make the schools’
records less likely to reflect a student’s sex.
This correct understanding of the schools’ practices sweeps away the
majority’s newfound concerns about how the schools determine the sex of their
students. The majority first identifies a hypothetical situation that purportedly
proves the schools arbitrarily treat some transgender students differently than other
transgender students. It imagines that a student who identifies as a member of the
opposite sex, and who has conformed his legal documents to that identity, could
report his gender identity instead of his sex on the enrollment forms and could
provide his legal documents as support. Majority Op. at 13–14. The schools
acknowledged that if a student took this strategy, they would unwittingly allow the
student to use the restrooms of the opposite sex without realizing the student was
using the restroom of the opposite sex. But that hypothetical, minor shortcoming is
far from fatal.
The possibility of evasion does not render unconstitutional the schools’
reliance on self-reporting and legal documentation to determine students’ sex.
People can commit tax fraud, evade the draft, and report inaccurate information on
the census without creating constitutional infirmities in those systems. And keep in
44
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 45 of 80
mind that the district court found that only 16 of the school district’s 40,000
students were transgender. Even if every single transgender student successfully
enrolled as a member of the opposite sex, the school district would still be 99.96
percent accurate at identifying the sex of its students. This near-perfect result is
certainly enough to satisfy intermediate scrutiny, even if the majority is correct that
intermediate scrutiny applies. Intermediate scrutiny does not “require[] that the
[policy] under consideration must be capable of achieving its ultimate objective in
every instance.” Nguyen v. Immigr. & Naturalization Serv.,

533 U.S. 53

, 70
(2001); see also Michael M. v. Super. Ct. of Sonoma Cnty.,

450 U.S. 464

, 473
(1981) (plurality opinion) (“The relevant inquiry . . . is not whether the statute is
drawn as precisely as it might have been . . . .”). Nor does it “require that [the
policymaker] elect one particular mechanism” to accomplish its goals, “even if that
mechanism arguably might be the most scientifically advanced method.” Nguyen,

533 U.S. at 63

. This point also undermines the majority’s reliance on the decision
of the Supreme Court in Craig v. Boren,

429 U.S. 190

 (1976). See Majority Op. at
14–16. In that decision, the Supreme Court held that a correlation of two percent
between maleness and drunk driving made Oklahoma’s higher drinking age for
men “an unduly tenuous ‘fit’” with its goal of reducing drunk driving. Craig,

429 U.S. at

201–02. It takes no more explanation to see that Craig does not control
here.
45
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 46 of 80
The majority fails to understand this analysis or to appreciate the privacy
interests at stake. As the majority notes, the Supreme Court has cautioned that it is
“unrealistic to expect either members of the judiciary or state officials to be well
versed in the rigors of experimental or statistical technique.”

Id. at 204

. The
majority contends that “rel[ying] on a figure of 99.96 percent [is] a prime example
of the folly of relying on statistics in equal protection analysis.” Majority Op. at 34.
To be clear, the figure of 99.96 percent comes from subtracting the number of
transgender students in the district (16) from the total number of students (40,000)
and then dividing that number by the total number of students (40,000).
Respectfully, basic subtraction and division are not “rigor[ous] . . . experimental or
statistical technique[s]” beyond the capabilities of federal judges. Craig,

429 U.S. at 204

; see, e.g., Birchfield v. North Dakota,

136 S. Ct. 2160

, 2193 n.10 (2016)
(Sotomayor, J., concurring in part and dissenting in part) (“Seven thousand annual
arrests divided by 82 judges and magistrate judges is 85.4 extra warrants per judge
and magistrate judge per year. And 85.4 divided by 52 weeks is 1.64 extra warrants
per judge and magistrate judge per week.”); Cent. R.R. Co. of Pa. v. Pennsylvania,

370 U.S. 607

, 610 n.2 (1962) (“If [605,678 car days are] divided by 365, the
quotient (1,659) represents the average number of cars located on [out-of-state]
railroads on any one day during [that year].”). And contrary to what the majority
asserts, the Supreme Court has never said that simple arithmetic cannot be used to
46
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 47 of 80
evaluate claims brought under the Equal Protection Clause. Majority Op. at 34
n.10. The majority also contends that the figure of 99.96 percent “simply
underscores how limited the ruling today is, which . . . applies only to Adams.” Id.
at 35. To the contrary, all students—not only Adams or the 0.04 percent of students
who are transgender—are potentially affected by a ruling that forces them to share
bathrooms with individuals of the opposite sex.
The majority’s feigned outrage on behalf of Adams likewise misses the
point. The majority asserts that comparing Adams’s attempted use of the boys’
restroom to committing tax fraud, evading the draft, or reporting inaccurate
information on the census is “unfortunate” because “Adams has been nothing but
honest.” Id. at 21 n.6. But Adams’s honesty has nothing to do with the issue at
hand. The school district asks students to report their sex, not their gender identity
or “governmentally-recognized legal sex.” Id. at 15. So if Adams reports gender
identity or “governmentally-recognized legal sex” instead of sex, then Adams has
reported inaccurate information. And it is irrelevant that Adams has supporting
legal documentation because the school district is not asking students to report
their “governmentally-recognized legal sex.”
The majority’s second reason that the schools’ method of determining
students’ sex is impermissibly arbitrary is similarly unpersuasive. “[W]ithout
justification,” it says, the schools prefer older documents to newer documents,
47
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 48 of 80
even though newer documents are “[p]resumably” more accurate. Id. at 16–17. The
majority says it can “come up with no explanation” for the schools’ preference for
students’ older documents. Id. at 17.
The majority is bewildered only because it closes its eyes to the record, not
to mention biological reality. The schools separate restrooms by sex, not (as the
majority thinks) by “governmentally-recognized legal sex.” Id. at 15. Because a
student’s sex does not change over time, the schools have no need to accept
updates. They instead determine sex once, using the ordinarily accurate method of
self-reporting and supporting legal documentation. The majority’s assertion that
newer documents are “[p]resumably” better evidence of sex than older documents,
id. at 17, makes sense only on the view that “sex” means the same thing as “gender
identity.” But the terms do not share the same meaning.
The majority’s reasoning depends on distorting two policies by eliding them
into one. It says the “unwritten bathroom policy” requires “that a student use either
a designated single-stall restroom or the bathroom corresponding to the sex listed
on the student’s enrollment documents.” Id. at 7–8. That policy does not exist. The
parties were clear that the schools require a student to use a gender-neutral
restroom or the restroom corresponding to his sex, as opposed to his gender
identity. The district court reached the same conclusion, finding that “the unwritten
School District bathroom policy was that boys will use the boys’ restrooms at
48
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 49 of 80
school and girls will use the girls’ restrooms at school, using those terms as
traditionally defined based on biological traits.” The majority is also wrong when it
says the school had a policy of requiring students to use the restroom associated
with their “governmentally-recognized legal sex.” Id. at 15. The schools use
government documents as evidence of the students’ actual sex, and they require
each student to use the bathroom associated with his sex. Even the majority
understood this point in its last opinion. Vacated Majority Op. at 12. The majority
transforms the schools’ sensible way of ascertaining sex and unassailable way of
separating restrooms into a single, irrational policy that is unsupported by the
record and would be unrecognizable to the parties.
Another point further undermines the majority’s reimagination of the school
policies. Under the majority’s second theory, the school district did not begin to
violate Adams’s rights until it refused to accept documents designating Adams as
male. But nothing about this lawsuit turns on when Adams obtained an updated
birth certificate or driving permit or when the school district refused to accept
them. Adams obtained both documents and provided them to the school district
well after the events that prompted the lawsuit. And Adams obtained damages
reflecting “emotional damage, stigmatization and shame from not being permitted
to use the boys’ restroom at school,” not limited to the time after the school district
refused to accept the documents identifying Adams as male.
49
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 50 of 80
Only by framing the means of ascertaining students’ sex as being the goal of
the schools’ separation of bathrooms does the majority turn the schools’ reliance
on enrollment papers into a sex-based classification to which intermediate scrutiny
could apply. Intermediate scrutiny applies only if the policy treats people
differently on the basis of sex. United States v. Virginia,

518 U.S. 515

, 533 (1996).
Otherwise, rational-basis review would apply. Armour v. City of Indianapolis,

566 U.S. 673

, 680–81 (2012). Separating bathrooms by sex treats people differently on
the basis of sex: it instructs boys to use one set of facilities and girls to use another.
By contrast, the mere act of determining an individual’s sex, using the same rubric
for both sexes, does not treat anyone differently on the basis of sex. To be clear,
the distinction should make no difference here. The schools’ practice of relying on
self-reported sex and supporting documentation survives either form of review.
The distinction matters only for the majority’s decision to affirm the judgment of
the district court by applying the wrong kind of constitutional scrutiny.
The majority’s theory not only misrepresents the record as it stands and
applies the wrong form of scrutiny. It also was not litigated by the parties before
the district court or before us. In this important case, involving some of the most
pressing legal issues of the day, the majority eschews the briefs and strikes out on
its own. But cf. United States v. Sineneng-Smith,

140 S. Ct. 1575

, 1579 (2020) (“In
50
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 51 of 80
our adversarial system of adjudication, we follow the principle of party
presentation.”).
Adams was clear throughout the proceedings in the district court that the
policy under review was the schools’ policy of separating bathrooms by sex, as
opposed to by gender identity. Adams’s complaint represented that the lawsuit
“challenge[d] . . . [the schools’] policy of excluding transgender students from the
single-sex facilities that match their gender identity.” Adams employed that
understanding of the complaint in later filings seeking to exclude the schools’
evidence as irrelevant, too, representing that “the issue in this case is narrow,
limited solely to whether [the schools’] policy prohibiting transgender students . . .
from using the restroom consistent with their gender identity is discriminatory,”
and that the “actual case being tried” was about “whether [the schools’] policy
excluding [Adams] . . . from using the restroom associated with his gender identity
is discriminatory.” The joint pretrial statement likewise represented that Adams
contended that the schools violated the law by not allowing transgender students to
“access facilities that match their gender identity.” And after trial, Adams framed
the issue in precisely the same way. Adams’s proposed findings of fact and
conclusions of law again challenged the schools’ “policy [that] requires students to
use restrooms that match their ‘biological sex’” and “bar[s] transgender students
from the restrooms consistent with their gender identity.” And Adams argued the
51
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 52 of 80
school discriminated by treating Adams “differently (i) from other boys, who can
use restrooms that match their male gender identity; and (ii) from non-transgender
students, since the policy in effect relegates him to a gender neutral restroom.”
Adams never relied on the sole rationale on which the majority now rests its
conclusion: that the school acts arbitrarily because it treats transgender students
differently from each other and because it prefers older documents to newer ones.
For all these reasons, the majority’s new rationale does no better than its old one.
The majority distorts what I have said both here and in my earlier dissent.
For example, the majority wrongly insists that I previously agreed with its
understanding of the schools’ policy. Majority Op. at 24–25. Although I
acknowledged the finding that the school district determines each student’s sex by
looking at enrollment forms, Vacated Dissenting Op. at 58, I never asserted or
implied that the school district requires students to use the bathroom corresponding
to their “governmentally-recognized legal sex,” Majority Op. at 15. I have always
understood the school district’s policy to be that students are required to use the
bathroom corresponding to their sex. The majority also asserts that I stated,
without record support, that newly enrolled students must receive a physical
examination in which a doctor examines their genitalia.

Id.

 at 25 n.7. But I said no
such thing. Both Florida law and the record make clear that students who enroll in
a Florida school system are required to present a certification of a physical
52
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 53 of 80
examination performed within one year before enrollment. See

Fla. Stat. § 1003.22

(1). The majority imagines assertions that I never made.
It is bad enough that the majority twists both the record and my dissent, but
it compounds its errors by failing to consider the ramifications of its ruling. I
warned in my previous dissent that the majority’s view would have consequences
far beyond the confines of this appeal. The logic of its retracted opinion would
invalidate all government policies that separate bathrooms—or locker rooms and
showers, for that matter—by sex. To be sure, the majority “assume[d]” that the
government can promote privacy interests by separating bathrooms by sex, and it
insisted that the lawfulness of sex-separated locker rooms was not before it.
Vacated Majority Op. at 8 n.3, 14. The majority offers similar cold comfort in its
new opinion. But anyone can take advantage of the majority’s demolition of sex-
specific bathroom privacy.
Moreover, the new majority opinion continues its earlier pretense of
invalidating the policy it considers—then, the policy of separate bathrooms for the
sexes; now, the schools’ method of determining students’ sex—only as it applies to
Adams. Its earlier attempt at limiting itself to an as-applied challenge changed
nothing. See Bucklew v. Precythe,

139 S. Ct. 1112

, 1127 (2019). The earlier
majority opinion and the new one alike do not offer a meaningful way to
distinguish this appeal from one that challenges sex-separated bathrooms and
53
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 54 of 80
locker rooms. That high school teenagers are not required to shower after physical
education classes is not an adequate response to concerns about student privacy.
Majority Op. at 23.
The majority insists that these other issues are not before it. “Do not believe
it.” Lawrence v. Texas,

539 U.S. 558

, 604 (2003) (Scalia, J., dissenting). Future
plaintiffs can still leverage the majority’s narrow view of privacy. Ultimately, if
the privacy interest at stake is untethered from using the bathroom away from the
opposite sex or from biological differences between the sexes, then no justification
exists for separating bathrooms—or any related facility—by sex.
B. The Statutory and Constitutional Challenges Raised by Adams Fail.
The majority has now retreated from its agreement with the statutory and
constitutional challenges Adams presented. Indeed, the majority now puzzlingly
asserts that its previous opinion “no longer . . . exist[s],” Majority Op. at 19, even
though that opinion—though vacated—can still be found in the Federal Reporter,
see Adams v. Sch. Bd. of St. Johns Cnty.,

968 F.3d 1286

 (11th Cir. 2020).
Regardless, I address the statutory and constitutional challenges Adams presented
because, unlike the majority’s newly invented theory of the case, those issues were
actually decided by the district court. I begin with the equal-protection arguments
challenging the schools’ policy of separating bathrooms on the basis of sex. Then I
turn to Adams’s Title IX challenge.
54
USCA11 Case: 18-13592         Date Filed: 07/14/2021     Page: 55 of 80
1. The Bathroom Policy Does Not Violate the Equal Protection Clause.
Because the Equal Protection Clause “does not make sex a proscribed
classification,” a policy that classifies on the basis of sex is constitutional if it
survives the two requirements of intermediate scrutiny. Virginia,

.
The government first must prove that the “classification serves important
governmental objectives.”

 (internal quotation marks omitted). For an objective
to be “important,” it cannot stem from “overbroad generalizations about the
different talents, capacities, or preferences of males and females.”

 The objective
must also be “genuine, not hypothesized or invented post hoc in response to
litigation.”

 In addition to proving that its policy serves important objectives, the
government must prove that “the discriminatory means employed are substantially
related to the achievement of those objectives.”

 (internal quotation marks
omitted).
Under this well-established standard, this appeal is not complicated.
Although the schools’ sex-separated bathrooms policy classifies on the basis of
sex, it serves the important objectives of protecting the interests of children in
using the bathroom away from the opposite sex and in shielding their bodies from
exposure to the opposite sex. The policy also fits tightly with both interests in
privacy. By requiring students to use the bathroom away from the opposite sex, the
policy directly protects the first interest and eliminates one of the most likely
55
USCA11 Case: 18-13592         Date Filed: 07/14/2021    Page: 56 of 80
opportunities for a violation of the second interest. In short, it easily satisfies
intermediate scrutiny, and even if questions remained, the Supreme Court has long
required that we defer to the judgment of public-school officials in this context.
The schools’ first objective—to protect students’ interest in using the
bathroom away from the opposite sex—is important. As then-Professor Ruth Bader
Ginsburg explained, “Separate places to disrobe, sleep, [and] perform personal
bodily functions are permitted, in some situations required, by regard for individual
privacy.” Ruth Bader Ginsburg, The Fear of the Equal Rights Amendment, Wash.
Post, Apr. 7, 1975, at A21. Indeed, “[a]cross societies and throughout history, it
has been commonplace and universally accepted to separate public restrooms,
locker rooms, and shower facilities on the basis of biological sex in order to
address privacy and safety concerns arising from the biological differences
between males and females.” G.G. ex rel. Grimm v. Gloucester Cnty. Sch. Bd.,

822 F.3d 709

, 734 (4th Cir. 2016) (Niemeyer, J., concurring in part and dissenting in
part), vacated,

137 S. Ct. 1239

 (2017). Separating bathrooms based on sex “dates
back as far as written history will take us.” W. Burlette Carter, Sexism in the
“Bathroom Debates,” 37 Yale L. & Pol’y Rev. 227, 287–88 (2018); see also

id.

 at
258–61 (documenting sex-separated bathrooms in feudal Japan and in ancient
Egypt, Greece, and Rome).
56
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 57 of 80
Unsurprisingly, the Supreme Court and our sister circuits have long
acknowledged a privacy interest in using the bathroom away from the opposite sex.
Even as it ordered the Virginia Military Institute to enroll women, the Supreme
Court acknowledged that “[a]dmitting women to VMI would undoubtedly require
alterations necessary to afford members of each sex privacy from the other sex in
living arrangements.” Virginia,

518 U.S. at

550 n.19 (citing Brief for Petitioner at
27–29 (arguing that integrating the institute would not require the sexes to be
together “when sleeping, dressing and using the bathroom”)). Our sister circuits
have likewise accepted that “the law tolerates same-sex restrooms . . . to
accommodate privacy needs.” Chaney v. Plainfield Healthcare Ctr.,

612 F.3d 908

,
913 (7th Cir. 2010); accord Faulkner v. Jones,

10 F.3d 226

, 232 (4th Cir. 1993);
Cumbey v. Meachum,

684 F.2d 712

, 714 (10th Cir. 1982); see also, e.g., Women
Prisoners of the D.C. Dep’t of Corr. v. District of Columbia,

93 F.3d 910

, 926
(D.C. Cir. 1996) (“[S]egregation of inmates by sex is unquestionably
constitutional.”). This privacy interest has long been “appropriately harmonized”
with the principle of equality. Ginsburg, Equal Rights Amendment, supra.
The schools’ policy is also “substantially related to the achievement” of its
objective to protect this privacy interest. Virginia,

 (internal
quotation marks omitted). Indeed, the policy is a mirror image of its objective—it
protects students from using the bathroom with the opposite sex by separating
57
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 58 of 80
bathrooms on the basis of sex. The policy “is not a means to some greater end, but
an end in itself.” Barnes v. Glen Theatre, Inc.,

501 U.S. 560

, 572 (1991) (plurality
opinion). Intermediate scrutiny is satisfied when a policy directly achieves the
objective itself. See

id.

This conclusion would stand even if the district court were correct that
gender identity, not biology, determines a person’s sex—that is, if the school
policy should have assigned the school district’s 16 transgender students to the
bathroom that aligned with their gender identity. The policy would still assign the
rest of the district’s 40,000 students to the right bathrooms, so it would still be
99.96 percent accurate in separating bathrooms by sex. As discussed already, this
near-perfect result is certainly enough to satisfy intermediate scrutiny. Nguyen,

533 U.S. at 70

; see also Michael M.,

450 U.S. at 473

 (plurality opinion).
Nor does it matter that Adams brings an as-applied challenge to the
bathroom policy. “[C]lassifying a lawsuit as facial or as-applied . . . does not speak
at all to the substantive rule of law necessary to establish a constitutional
violation.” Bucklew, 139 S. Ct.at 1127. And, to reiterate, intermediate scrutiny
does not “require[] that the [policy] under consideration must be capable of
achieving its ultimate objective in every instance.” Nguyen,

533 U.S. at 70

.
Demanding that the policy satisfy its privacy interests as to each plaintiff who
brings an as-applied challenge would disregard intermediate scrutiny by
58
USCA11 Case: 18-13592       Date Filed: 07/14/2021   Page: 59 of 80
demanding a perfect fit between the sex-based classification and the government
interest at issue.
The school policy also substantially advances its objective to protect
children from exposing their unclothed bodies to the opposite sex. Courts have
long understood that the “special sense of privacy” that individuals hold in
avoiding bodily exposure is heightened “in the presence of people of the other
sex.” Fortner v. Thomas,

983 F.2d 1024

, 1030 (11th Cir. 1993) (internal quotation
marks omitted); accord, e.g., Cookish v. Powell,

945 F.2d 441

, 446 (1st Cir. 1991);
Harris v. Miller,

818 F.3d 49

, 59 (2d Cir. 2016); Doe v. Luzerne County,

660 F.3d 169

, 177 (3d Cir. 2011); Strickler v. Waters,

641 F.2d 1375

, 1387 (4th Cir. 1993);
Moore v. Carwell,

168 F.3d 234

, 236–37 (5th Cir. 1999); Brannum v. Overton
Cnty. Sch. Bd.,

516 F.3d 489

, 494 (6th Cir. 2008); Canedy v. Boardman,

16 F.3d 183

, 185 (7th Cir. 1994); Byrd v. Maricopa Cnty. Sheriff’s Dep’t,

629 F.3d 1135

,
1141 (9th Cir. 2011) (en banc); Hayes v. Marriott,

70 F.3d 1144

, 1146 (10th Cir.
1995). And separating bathrooms by sex eliminates one of the most common
opportunities for exposure to the opposite sex. The district court acknowledged the
undisputed testimony that students at Adams’s school change clothing outside
bathroom stalls and that bathrooms are ordinarily unsupervised. By separating
bathrooms by sex, the policy eliminates the risk of bodily exposure where it is
most likely to occur, which satisfies intermediate scrutiny. See Nguyen,

533 U.S. at

59
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 60 of 80
70 (holding that a policy that “seeks to foster the opportunity” for a government
objective “has a close and substantial bearing on” that objective).
Even if any doubt remained about whether the bathroom policy survives
scrutiny, we must resolve that doubt in favor of the Board because the policy
governs student conduct in public schools. The Supreme Court has long held that
the constitutional rights of students, including “Fourteenth Amendment rights, are
different in public schools than elsewhere.” Vernonia Sch. Dist. 47J v. Acton,

515 U.S. 646

, 656 (1995). Schools have a “custodial and tutelary” power over minor
students, “permitting a degree of supervision and control that could not be
exercised over free adults.”

Id. at 655

. For that reason, the Supreme Court has long
deferred to the decisions of school districts in a variety of constitutional contexts,
including when determining whether a suspicionless drug search was reasonable
under the Fourth Amendment,

id. at 665

, whether the censorship of certain speech
was acceptable under the First Amendment, Morse v. Frederick,

551 U.S. 393

,
403–06, 409–410 (2007) (collecting decisions), and whether corporal punishment
was cruel and unusual under the Eighth Amendment, Ingraham v. Wright,

430 U.S. 651

, 681–82 (1977).
The bathroom policy falls squarely within the Board’s authority to
“prescribe and control conduct” in its schools.

Id. at 682

 (internal quotation marks
omitted). “[I]n a public school environment . . . the State is responsible for
60
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 61 of 80
maintaining discipline, health, and safety.” Bd. of Educ. of Indep. Sch. Dist. No. 92
of Pottawatomie Cnty. v. Earls,

536 U.S. 822

, 830 (2002); see also

Fla. Stat. § 1001.42

(8)(a) (making school boards responsible for the “control of students at
school, and for proper attention to health, safety, and other matters relating to the
welfare of students”). This responsibility is so weighty that school districts can be
liable for sexual assault and harassment between students. See Miami-Dade Cnty.
Sch. Bd. v. A.N.,

905 So. 2d 203

, 203–04, 206 (Fla. Dist. Ct. App. 2005)
(upholding a jury verdict against a school for failing to protect a student who was
sexually assaulted in a bathroom by another student); see also Williams v. Bd. of
Regents of Univ. Sys. of Ga.,

477 F.3d 1282

, 1299 (11th Cir. 2007). And
bathrooms, one of the few areas in a school that are unsupervised, plainly pose
risks to student “discipline, health, and safety.” Earls,

536 U.S. at 830

; see also
A.N.,

905 So. 2d at

203–04, 206. The Board’s assessment of the privacy risks its
students face and the effectiveness of its policy in mitigating those risks deserves
deference.
The since-withdrawn majority opinion elided this entire analysis by
misunderstanding both the classification and privacy interests at issue. It contended
that the policy triggers heightened scrutiny not because it separates bathrooms by
sex but because it purportedly imposes “differential treatment” on transgender
students. Vacated Majority Op. at 12. In doing so, the majority misstated the
61
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 62 of 80
school policy, conflated sex-based classifications with transgender-based
classifications, and contravened Supreme Court precedent. Compounding its
errors, the majority then ignored fundamental understandings of why bathrooms
are separated on the basis of sex by rejecting long-standing privacy rationales for
sex-separated bathrooms. This conclusion led it to fault the objective underlying
the school policy as both hypothetical and based on impermissible stereotypes.
After misconstruing both the classification and the privacy interests at issue—the
only two ingredients of intermediate scrutiny—the majority opinion then
concluded that the schools’ classification does not substantially advance a valid
objective. I take each of these errors in turn.
The majority’s now-vacated conclusion that the school policy classifies on
the basis of sex because it “singles out transgender students” was both central to its
analysis and wrong.

 The previous majority opinion reached this incorrect
conclusion by pointing to a provision of the school policy that does not have that
effect. The former majority opinion said the school policy targets transgender
students because of the following provision: “Transgender students will be given
access to a gender-neutral restroom and will not be required to use the restroom
corresponding to their biological sex.”

 But this provision only offers the option
of using gender-neutral bathrooms as an alternative to the bathroom that matches a
child’s sex. It is an accommodation for transgender students, not a special burden.
62
USCA11 Case: 18-13592        Date Filed: 07/14/2021   Page: 63 of 80
What Adams actually challenges is the requirement that students cannot use the
bathrooms of the opposite sex, which long predates the provision that
accommodates transgender students. It is this policy that “prohibit[s] . . .
transgender students from using the restrooms matching their gender identity.” Id.
at 14. Because this policy divides bathrooms by sex, not transgender status, it does
not facially classify on the basis of transgender status.
The decision of the Supreme Court in Geduldig v. Aiello,

417 U.S. 484

(1974), is instructive. There, the Court held that a state insurance policy that
excluded coverage for pregnancies did not classify on the basis of sex.

 at 495–
97. It explained that the classification at issue created two groups—pregnant and
nonpregnant people.

 at 496 n.20. Although “the first group is exclusively
female,” the Court explained, “the second includes members of both sexes,” which
revealed a “lack of identity” between pregnancy and sex. Id.; see also Bray v.
Alexandria Women’s Health Clinic,

506 U.S. 263

, 271 (1993) (reaffirming this
reasoning). This analysis applies with equal force here. The bathroom policy
creates two groups—students who can use the boys’ bathroom and students who
can use the girls’ bathroom. Both groups contain transgender students and non-
transgender students, so a “lack of identity” exists between the policy and
transgender status. Geduldig,

417 U.S. at

496 n.20.
63
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 64 of 80
At most, the policy has a disparate impact on transgender students, which is
not enough to create a sex-based classification. Facially neutral policies trigger
intermediate scrutiny only if “invidious gender-based discrimination” motivated
them. Pers. Adm’r of Mass. v. Feeney,

442 U.S. 256

, 274 (1979); see also Vill. of
Arlington Heights v. Metro. Hous. Dev. Corp.,

429 U.S. 252

, 265 (1977) (“Proof of
. . . discriminatory intent or purpose is required to show a violation of the Equal
Protection Clause.”). Adams cannot argue that a discriminatory purpose against
transgender students motivated the policy. Although the district court found that
the Board acted with discriminatory intent because it failed to update its policies
when it became “aware of the need to treat transgender students the same as other
students,” the Supreme Court has repeatedly held that “[d]iscriminatory purpose
. . . implies more than . . . awareness of consequences.” E.g., Bray,

506 U.S. at 760

(internal quotation marks omitted); Feeney,

442 U.S. at 279

 (internal quotation
marks omitted). It instead requires the Board to act “at least in part ‘because of,’
not merely ‘in spite of,’ . . . adverse effects upon an identifiable group.” Feeney,

442 U.S. at 279

. The district court found that the “School Board did not have
transgender students in mind when it originally established separate multi-stall
restrooms for boys and girls,” which precludes a finding of discriminatory intent.
The decision of the Supreme Court in Bostock v. Clayton County,

140 S. Ct. 1731

, is not to the contrary. To be sure, Bostock clarified that “discrimination
64
USCA11 Case: 18-13592        Date Filed: 07/14/2021    Page: 65 of 80
based on homosexuality or transgender status necessarily entails discrimination
based on sex” in the context of employment discrimination under Title VII. Id. at
1747; see also Glenn v. Brumby,

663 F.3d 1312

, 1318 (11th Cir. 2011) (holding
that discrimination based on gender nonconformity constitutes sex discrimination
regardless of whether the victim is transgender or not). But this appeal concerns
the converse question: whether discrimination on the basis of sex necessarily
entails discrimination based on transgender status. Of course, a policy can classify
on the basis of sex without also classifying on the basis of transgender status. See,
e.g., Nguyen,

533 U.S. at 60

. Indeed, Bostock expressly disclaimed reaching any
conclusion on the permissibility of sex-separated bathrooms and locker rooms. See
140 S. Ct. at 1753.
The now-withdrawn majority opinion’s misunderstanding of the
classification at issue infected its constitutional inquiry. Intermediate scrutiny turns
on the relationship between the classification at issue and the government’s
objectives—that is, whether a sex-based classification is substantially related to the
government’s objectives. See Nguyen,

533 U.S. at 60

. So the relevant question is
whether excluding students of one sex from the bathroom of the other sex
substantially advances the schools’ privacy objectives. The question is not, as the
majority framed it, whether excluding transgender students from the bathroom of
their choice furthers important privacy objectives. The majority’s
65
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 66 of 80
misunderstanding of the classification as based on transgender status
gerrymandered its analysis to the second question.
In addition to misunderstanding the classification at issue, the now-vacated
majority erroneously redefined the privacy interests at stake. Although the majority
conceded that protecting bathroom privacy, in some abstract sense, is an important
objective, it rejected both of the privacy interests that the school policy protects.
For the first interest, the majority asserted that the Board incorrectly decided that
its students had any privacy interest in using the bathroom away from “students
who do not share the same birth sex.” Vacated Majority Op. at 21 (internal
quotation marks omitted). And although the former majority opinion appeared to
acknowledge that students have a privacy interest in not exposing their bodies, it
did not accept that this interest can be sex-specific—that the interest is heightened
when exposure is to the opposite sex. Instead, it asserted not only that Adams’s
“anatomical differences” from boys are “irrelevant” to bathroom privacy, id. at 24,
but also that thinking otherwise is an unconstitutional stereotype, id. at 25–28.
The new majority opinion repeats the second mistake. It devotes less than a
page to the legitimacy of the government interests at stake. Majority Op. at 11–12.
It acknowledges an abstract interest in “protecting the bodily privacy of young
students,” and it “recognize[s]” that the government can permissibly protect
privacy by “maintaining separate bathrooms for boys and girls, men and women.”
66
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 67 of 80
Id. It does not explain what it means by those terms, and it again does not
acknowledge that students can have a heightened interest in avoiding the exposure
of their bodies to members of the opposite sex.
The majority’s understanding of each interest contravenes precedent. Its
decision to limit students’ privacy interest to bodily exposure ignores that children
also have a distinct privacy interest in using the bathroom away from the opposite
sex. See, e.g., Virginia,

518 U.S. at

550 n.19; Chaney,

612 F.3d at 913

; Faulkner,

10 F.3d at 232

; Cumbey,

684 F.2d at 714

; see also Skinner v. Ry. Labor Execs.’
Ass’n,

489 U.S. 602

, 626 (1989) (concluding that urination is “an excretory
function traditionally shielded by great privacy”). Similarly, the majority’s
continued failure to acknowledge that the privacy interest in avoiding bodily
exposure is heightened when children of the opposite sex are present ignores
longstanding precedent, not to mention common sense. See, e.g., Fortner,

983 F.2d at 1030

; Harris, 818 F.3d at 59; Luzerne County,

660 F.3d at 177

; Brannum,

516 F.3d at 494

; Canedy,

16 F.3d at 185

.
Finally, the former majority opinion’s alternative contention that the privacy
interests at issue are invalid because they rest on impermissible sex stereotypes
remains incorrect. According to the former majority opinion, the school policy
“presumed every person deemed ‘male’ at birth would act and identify as a ‘boy’
and every person deemed ‘female’ would act and identify as a ‘girl.’” Vacated
67
USCA11 Case: 18-13592         Date Filed: 07/14/2021    Page: 68 of 80
Majority Op. at 26. It also faulted the policy for indulging in the purportedly
unconstitutional stereotype that “one’s gender identity and expression should align
with one’s birth sex.”

 Neither of these arguments has merit.
The majority never explained how the school policy “presume[d] every
person deemed ‘male’ at birth would act and identify as a ‘boy’ and every person
deemed ‘female’ would act and identify as a ‘girl.’” Nor could it. The policy does
not turn on how students “act and identify.” It assigns bathrooms by sex, which is
not a stereotype. See Nguyen,

533 U.S. at 73

 (“Mechanistic classification of all our
differences as stereotypes would operate to obscure those misconceptions and
prejudices that are real.”); Etsitty v. Utah Transit Auth.,

502 F.3d 1215

, 1224 (10th
Cir. 2007) (“Use of a restroom designated for the opposite sex does not constitute a
mere failure to conform to sex stereotypes.”).
The withdrawn majority opinion’s other contention—that believing “one’s
gender identity and expression should align with one’s birth sex” is an
unconstitutional stereotype—fares no better. The Supreme Court has long
grounded its sex-discrimination jurisprudence in reproductive biology. See, e.g.,
Nguyen,

533 U.S. at 73

 (“The difference between men and women in relation to
the birth process is a real one . . . .”); Geduldig,

417 U.S. at

496 n.20 (“[I]t is true
that only women can become pregnant . . . .”); see also Virginia,

(“Physical differences between men and women, however, are enduring: [T]he two
68
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 69 of 80
sexes are not fungible . . . .” (internal quotation marks omitted)). Indeed, the
Court’s justification for giving heightened scrutiny to sex-based classifications
makes sense only with reference to physiology.
In one of its foundational sex-discrimination decisions, the Court justified
heightened scrutiny this way: “since sex . . . is an immutable characteristic
determined solely by the accident of birth, the imposition of special disabilities
upon the members of a particular sex because of their sex would seem to violate
the basic concept of our system that legal burdens should bear some relationship to
individual responsibility.” Frontiero v. Richardson,

411 U.S. 677

, 686 (1973)
(plurality opinion) (internal quotation marks omitted). In other words, the Court
endorsed heightened scrutiny because laws “distributing benefits and burdens
between the sexes in different ways very likely reflect outmoded notions of the
relative capabilities of men and women.” Cleburne,

473 U.S. at 441

. To say that it
is an unconstitutional stereotype to believe that “one’s gender identity and
expression should align with one’s birth sex,” the majority must not only rewrite
the Supreme Court’s physiological rationale for heightened scrutiny of sex-based
classifications, but also hold that many of the Court’s sex-discrimination decisions
turned on an impermissible stereotype. See Nguyen,

533 U.S. at 73

; Virginia,

; Geduldig,

417 U.S. at

496 n.20; Frontiero,

411 U.S. at 686

. We, of
course, cannot take either of those actions.
69
USCA11 Case: 18-13592       Date Filed: 07/14/2021   Page: 70 of 80
The majority’s narrow construction of bathroom privacy skewed the
intermediate-scrutiny analysis in favor of Adams. Policies that separate bathrooms
on the basis of sex arise from the understanding that privacy interests are
sometimes sex specific. By failing to acknowledge any sex-specific privacy
interest, the majority demands the impossible: a justification for sex-separated
bathrooms that does not involve sex. To be sure, the since-withdrawn majority
suggested that a different trial record—one that contained evidence that Adams or
other transgender students “harass[ed] or peep[ed] at” other students in the
bathroom—might support the bathroom policy. Vacated Majority Op. at 21. But
that evidence would not justify a sex-based classification. If voyeurism is equally
problematic whether it occurs between children of the same or opposite sex, then
separating bathrooms by sex would not advance any interest in combatting
voyeurism. Only single-stall bathrooms could address that concern. Further, under
intermediate scrutiny, an invidious stereotype about members of a suspect class
cannot justify a discriminatory policy “even when some statistical support can be
conjured up for the generalization.” J.E.B. v. Ala. ex rel. T.B.,

511 U.S. 127

, 139
n.11 (1994). So evidence that children of a particular sex—or transgender students
under the majority’s perspective—are likely to “harass or peep at” members of the
opposite sex could not justify sex-separated bathrooms.
70
USCA11 Case: 18-13592      Date Filed: 07/14/2021    Page: 71 of 80
Only after it replaced both of the inquiries relevant to intermediate
scrutiny—the “discriminatory means employed” by the policy and the privacy
interests at issue, Virginia,

 (internal quotation marks omitted)—did
the defunct majority opinion conclude that the school policy does not substantially
advance its objective to protect privacy. The end result was an opinion on sex
discrimination that looked nothing like an intermediate-scrutiny inquiry into
whether a sex-based classification satisfies the Equal Protection Clause. The
majority’s impulse to try again is understandable.
When shorn of misunderstandings of the school policy and the legal
standards that govern sex-based classifications, this appeal is straightforward. The
school policy protects longstanding privacy interests inherent in using the
bathroom, and it does so in an ancient and unremarkable way—by separating
bathrooms on the basis of sex. That policy is not unconstitutional.
2. The Bathroom Policy Does Not Violate Title IX.
The majority no longer addresses Adams’s statutory challenge to the
schools’ policy, but I explain why the policy is permissible. Title IX mandates that
no person “shall, on the basis of sex, be excluded from participation in, be denied
the benefits of, or be subjected to discrimination under any education program or
activity receiving Federal financial assistance.”

20 U.S.C. § 1681

(a). But an
important qualification tempers this mandate: “nothing contained herein shall be
71
USCA11 Case: 18-13592        Date Filed: 07/14/2021     Page: 72 of 80
construed to prohibit any educational institution receiving funds under this Act,
from maintaining separate living facilities for the different sexes.”

Id.

 § 1686. The
implementing regulations clarify that institutions “may provide separate toilet,
locker room, and shower facilities on the basis of sex, but such facilities provided
for students of one sex shall be comparable to such facilities provided for students
of the other sex.”

34 C.F.R. § 106.33

.
Whether the Board violated Title IX turns on the answer to one question:
what does “sex” mean in Title IX? Regardless of whether separating bathrooms by
sex would otherwise constitute discrimination “on the basis of sex,”

20 U.S.C. § 1681

(a), the bathroom policy does not violate Title IX if it falls within the safe
harbor for “separate toilet . . . facilities on the basis of sex.” 34 C.F.R § 106.33.
And if the school policy is valid under Title IX, then Title IX also permits the
schools to require all students, including Adams, to follow that policy. Cf. Antonin
Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts § 30, at
192–93 (2012) (“[W]henever a power is given by a statute, everything necessary to
making it effectual or requisite to attaining the end is implied.” (quoting 1 James
Kent, Commentaries on American Law *464)).
Contrary to Adams’s arguments, the Supreme Court did not resolve this
question in Bostock. Far from it. Not only did the Court “proceed on the
assumption that ‘sex’ . . . refer[s] only to biological distinctions between male and
72
USCA11 Case: 18-13592       Date Filed: 07/14/2021   Page: 73 of 80
female,” it disclaimed deciding whether Title VII allows for sex-separated
bathrooms. Bostock, 140 S. Ct. at 1739, 1753. And any guidance Bostock might
otherwise provide about whether Title VII allows for sex-separated bathrooms
does not extend to Title IX, which permits schools to act on the basis of sex
through sex-separated bathrooms. See

20 U.S.C. § 1686

34 C.F.R. § 106.33

.
Turning to the provisions at issue, this question is not close. As used in Title
IX and its implementing regulations, “sex” unambiguously is a classification on
the basis of reproductive function. We must, of course, give words in statutes the
ordinary meaning they conveyed when Congress enacted them. See New Prime
Inc. v. Oliveira,

139 S. Ct. 532

, 539 (2019); Scalia & Garner, Reading Law §§ 6–7,
at 69–71, 78–79. And “sex” has never meant gender identity. See, e.g., Sex, The
American Heritage Dictionary of the English Language (New College ed. 1979)
(defining “sex” by reference to reproductive functions); Sex, The Random House
College Dictionary (rev. ed. 1980) (same); see also Am. Psychiatric Ass’n,
Diagnostic and Statistical Manual of Mental Disorders 451 (5th ed. 2013)
(similar).
That “sex” did not mean gender identity is unsurprising. When Congress
enacted Title IX in 1972, psychiatric literature conflated sexual orientation with
gender identity. See Jack Drescher, Transsexualism, Gender Identity Disorder and
the DSM, 14 J. Gay & Lesbian Mental Health 109, 111 (2010). And as with
73
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 74 of 80
homosexuality, a common belief among psychiatrists was that “trans people [were]
severely mentally disturbed.” See id. at 114, 116–17. Indeed, the American
Psychiatric Association first classified “Gender Identity Disorders” as
psychosexual disorders in which a person’s internal sense of gender did not align
with his or her anatomy. See Am. Psychiatric Ass’n, Diagnostic and Statistical
Manual of Mental Disorders 261 (3d ed. 1980). Consistent with this view,
“[m]ainstream medical thinking” when Title IX became law was firmly opposed to
sex-reassignment surgery. Drescher, supra, at 111–12. Even among its proponents,
“[s]ex reassignment was . . . considered not a cure, but a palliative treatment.”
Dallas Denny, A Selective Bibliography of Transsexualism, 6 J. Gay & Lesbian
Psychotherapy 35, 38 (2002). It is untenable to construe transgender status, which
even the medical community saw as a departure from the norm, as altering the
norm itself among the general public.
In deciding otherwise, the vacated majority opinion erroneously concluded
that the safe harbor for bathrooms does not apply because Title IX and its
regulations do not “declare” whether “sex” as applied to Adams is the “sex
identified at birth”—female—or the sex listed on Adams’s amended birth
certificate and driver’s license—male. Vacated Majority Op. at 40–41 (quoting
Bostock, 140 S. Ct. at 1746). But the ordinary meaning of “sex” in the safe-harbor
provision does not change when a plaintiff is transgender. See Cochise
74
USCA11 Case: 18-13592       Date Filed: 07/14/2021    Page: 75 of 80
Consultancy, Inc. v. U.S. ex rel. Hunt,

139 S. Ct. 1507

, 1512 (2019) (“In all but the
most unusual situations, a single use of a statutory phrase must have a fixed
meaning.”). And as explained above, the ordinary meaning of “sex” when
Congress enacted Title IX turned on reproductive function. That Congress did not
define “sex” does not change this conclusion. See United States v. Sepulveda,

115 F.3d 882

, 886 n.9 (11th Cir. 1997) (“[A] statute is not ambiguous merely because it
contains a term without a statutory definition.”). And under the unambiguous
meaning of “sex” in the safe-harbor provision, the Board did not violate Title IX
when it prohibited Adams from using the boys’ bathroom.
Instead of grappling with the meaning of “sex,” the now-vacated majority
opinion abdicated its duty to interpret the law. According to that opinion, it is
unnecessary to delve into the meaning of “sex” in Title IX because the safe harbor
“does not dictate how schools should approach transgender students’ restroom
use.” Vacated Majority Op. at 42. But courts regularly apply general standards of
law to particular facts, and the Board asks this Court to apply the safe-harbor
provision to the facts in this appeal. By declaring it not “necessary” to interpret the
safe-harbor provision, the majority abandoned statutory interpretation in favor of
legislating a transgender exception to the safe-harbor provision. This approach
offends basic principles of statutory interpretation. See Scalia & Garner, Reading
Law § 8, at 93 (“The principle that a matter not covered is not covered is so
75
USCA11 Case: 18-13592 Date Filed: 07/14/2021 Page: 76 of 80
obvious that it seems absurd to recite it. The judge should not presume that every
statute answers every question . . . . Nor should the judge elaborate unprovided-for
exceptions to a text . . . .”); EEOC v. Abercrombie & Fitch Stores, Inc.,

135 S. Ct. 2028

, 2033 (2015) (“We construe [a statute’s] silence as exactly that: silence.”).
Indeed, the previous opinion turned Title IX on its head by requiring a clear
statement from Congress that the safe harbor protects the Board. Because Congress
enacted Title IX under its Spending Clause power, U.S. Const. art. I, § 8, cl. 1, the
Board’s violation must be unambiguous to trigger liability. Although the Spending
Clause allows Congress to “attach conditions on the receipt of federal funds,”
South Dakota v. Dole,

483 U.S. 203

, 206 (1987), “[t]he legitimacy of Congress’
power to legislate under the spending power . . . rests on whether the State
voluntarily and knowingly accepts the terms of the ‘contract.’” Pennhurst State
Sch. & Hosp. v. Halderman,

451 U.S. 1

, 17 (1981). In other words, “if Congress
intends to impose a condition on the grant of federal moneys, it must do so
unambiguously.”