Harvard Law Review Harvard Law Review Harvard Law Review

Privacy

Constitutional Privacy and the Fight Over Access to Sex-Segregated Spaces

The battle over transgender rights is raging on many fronts. Despite some local1×1. For example, at least 225 cities and counties now protect employees from discrimination on the basis of gender identity. See Cities and Counties with Non-discrimination Ordinances that Include Gender Identity, Hum. Rts. Campaign, https://www.hrc.org/resources/cities-and-counties-with-non-discrimination-ordinances-that-include-gender [https://perma.cc/PB5D-PV4H]. and national2×2. Under President Barack Obama, federal agencies instructed schools to “treat a student’s gender identity as the student’s sex for purposes of Title IX.” See Dear Colleague Letter on Transgender Students from Catherine E. Lhamon, Assistant Sec’y for Civil Rights, U.S. Dep’t of Educ., & Vanita Gupta, Principal Deputy Assistant Att’y Gen. for Civil Rights, DOJ 2 (May 13, 2016), https://www.ed.gov/ocr/letters/colleague-201605-title-ix-transgender.pdf [https://perma.cc/9WLB-TSPK] [hereinafter Dear Colleague Letter]. They also allowed transgender people to serve in the military. Memorandum from Ashton Carter, Sec’y of Def., to the Sec’ys of the Military Dep’ts et al. 2 (June 30, 2016), https://dod.defense.gov/Portals/1/features/2016/0616_policy/DTM-16-005.pdf [https://perma.cc/J4YK-ZB4W]. victories for transgender activists in the past few years, the battle is far from over. States like North Carolina have tried to bar transgender people from using bathrooms and locker rooms not matching their birth-assigned sexes.3×3. See Dave Philipps, North Carolina Bans Local Anti-Discrimination Policies, N.Y. Times (Mar. 23, 2016), https://nyti.ms/22GJ011 [https://perma.cc/HM9E-7N46]. Transgender rights activists later reached a settlement with North Carolina that “permanently barred” it from preventing transgender people from using bathrooms consistent with their gender identities in state government buildings. See Dan Levin, North Carolina Reaches Settlement on “Bathroom Bill,” N.Y. Times (July 23, 2019), https://nyti.ms/2YeWnN1 [https://perma.cc/ZBY5-K7JD]. At the federal level, President Donald Trump’s Department of Justice and Department of Education have revoked Obama-era guidance protecting transgender students in schools.4×4. See Dear Colleague Letter from Sandra Battle, Acting Assistant Sec’y for Civil Rights, U.S. Dep’t of Educ., & T.E. Wheeler II, Acting Assistant Att’y Gen. for Civil Rights, DOJ 1 (Feb. 22, 2017), https://www2.ed.gov/about/offices/list/ocr/letters/colleague-201702-title-ix.pdf [https://perma.cc/K8NE-L48E]. President Trump also rolled back protection for transgender servicemembers: on July 26th, 2017, he tweeted that transgender people should be barred from serving in the U.S. military,5×5. Donald J. Trump (@realDonaldTrump), Twitter (July 26, 2017, 5:55 AM) https://twitter.com/realDonaldTrump/status/890193981585444864 [https://perma.cc/SS84-CNRG]; Donald J. Trump (@realDonaldTrump), Twitter (July 26, 2017, 6:04 AM), https://twitter.com/realDonaldTrump/status/890196164313833472 [https://perma.cc/UT3N-PJSH]. and a version of his ban went into effect in 2019.6×6. See Nicole Gaouette, Pentagon Transgender Ban Goes into Effect, CNN (Apr. 12, 2019, 5:15 PM), https://www.cnn.com/2019/04/12/politics/transgender-troop-ban-starts/index.html [https://perma.cc/7FLP-DJY8]. This Term, the Supreme Court will determine whether Title VII of the Civil Rights Act of 1964’s7×7. 42 U.S.C. §§ 2000a to 2000h-6 (2012). prohibition of employment discrimination “because of . . . sex”8×8. Id. § 2000e-2(a)(1). prevents employers from firing transgender workers because of their gender identity.9×9. Adam Liptak, Supreme Court to Decide Whether Landmark Civil Rights Law Applies to Gay and Transgender Workers, N.Y. Times (Apr. 22, 2019), https://nyti.ms/2IAuaZi class=”small-caps”> [https://perma.cc/T8JC-V3NA].

This Note zooms in on a specific fight in the war over transgender rights: the existence and scope of a federal constitutional right to bodily privacy in one’s partially clothed body.10×10. This Note focuses on a federal constitutional right to privacy. Of course, some states have recognized a similar right to privacy in their constitutions. See Privacy Protections in State Constitutions, Nat’l Conf. St. Legislatures (Nov. 7, 2018), https://www.ncsl.org/research/telecommunications-and-information-technology/privacy-protections-in-state-constitutions [https://perma.cc/4HMX-CGRV]. State constitutional rights may have a different scope than federal constitutional rights. See Adam Hickey, Note, Between Two Spheres: Comparing State and Federal Approaches to the Right to Privacy and Prohibitions Against Sodomy, 111 Yale L.J. 993, 997–1002, 1007–19 (2002) (comparing the scope of state and federal constitutional privacy rights in the context of sodomy laws before the Supreme Court overturned Bowers v. Hardwick, 478 U.S. 186 (1986), in Lawrence v. Texas, 539 U.S. 558 (2003)). This right has played a prominent role in recent litigation over school districts’ bathroom and sex-segregated space policies. For example, when a Chicago-area school district created a trans-friendly bathroom and locker room policy, a coalition of cisgender students and their parents sued.11×11. See Students & Parents for Privacy v. Sch. Dirs. of Twp. High Sch. Dist. 211, 377 F. Supp. 3d 891, 893–94 (N.D. Ill. 2019). As has become typical in suits with similar facts,12×12. For a case with almost identical facts and claims, see Doe ex rel. Doe v. Boyertown Area School District, 897 F.3d 518, 524–25 (3d Cir. 2018). the students and parents brought a potpourri of claims: they argued that the school district violated Title IX of the Civil Rights Act of 1964’s prohibition of sex discrimination, parents’ “fundamental right . . . to direct the upbringing of their children,” Illinois’s version of the Religious Freedom Restoration Act, the Free Exercise Clause of the U.S. Constitution, and — most relevantly — cisgender students’ constitutional right to bodily privacy in their partially clothed bodies.13×13. Students & Parents for Privacy, 377 F. Supp. at 891, 900. The Boyertown plaintiffs did not bring religious discrimination claims or a claim based on parents’ fundamental rights to raise their children; instead, they brought claims under Pennsylvania tort law in addition to Title IX and bodily privacy claims. Boyertown, 897 F.3d at 525. Though the district court refused to recognize this constitutional right,14×14. See Students & Parents for Privacy, 377 F. Supp. at 902 (“Although it would not shock the Court if the Seventh Circuit or Supreme Court one day recognizes the right to bodily privacy that the plaintiff seeks to enforce, this Court is not at liberty to expand the substantive rights protected by the Due Process Clause.” (footnote omitted)). another court has done so.15×15. See Boyertown, 897 F.3d at 527 (“[A] person has a constitutionally protected privacy interest in his or her partially clothed body.”). Understanding the history of this right and its content is important to resolving these increasingly common constitutional claims. The stakes are high: bathroom access affects the approximately 150,000 high school–age students who identify as transgender in the United States,16×16. Jody L. Herman et al., The Williams Inst., UCLA Sch. of Law, Age of Individuals Who Identify as Transgender in the United States 2 (2017), https://williamsinstitute.law.ucla.edu/wp-content/uploads/TransAgeReport.pdf [https://perma.cc/Z9H5-V9M9]. many of whom avoid sex-segregated spaces because they feel unsafe or uncomfortable or because they face discrimination.17×17. Joseph G. Kosciw et al., GLSEN, The 2017 National School Climate Survey: The Experiences of Lesbian, Gay, Bisexual, Transgender, and Queer Youth in Our Nation’s Schools 93 (2018), https://www.glsen.org/sites/default/files/2019-10/GLSEN-2017-National-School-Climate-Survey-NSCS-Full-Report_0.pdf [https://perma.cc/RB4W-YWUC]; see also id. at 97 fig. 3.10 (showing that 77.9% of transgender students in the sample reported experiencing discrimination at school). Uncertainty around how schools will treat them may exacerbate their fears; in one study from 2017, more than half of LGBTQ student respondents reported that their schools did not have any policies in place related to transgender students.18×18. Id. at 62 fig. 2.7. And while which bathroom transgender students can use may seem like a narrow issue, people — even Supreme Court Justices — can’t stop talking about bathrooms when discussing transgender rights.19×19. See Masha Gessen, The Supreme Court Considers L.G.B.T. Rights, but Can’t Stop Talking About Bathrooms, New Yorker (Oct. 9, 2019), https://www.newyorker.com/news/our-columnists/the-supreme-court-considers-lgbt-rights-but-cant-stop-talking-about-bathrooms [https://perma.cc/SGA5-G8KD] (“[T]he Justices wanted to talk about bathrooms. Justice Roberts wanted to talk about bathrooms. Justice Gorsuch wanted to talk about bathrooms. . . . Then Justice Sotomayor wanted to talk about bathrooms. . . . [T]hen Justice Alito wanted to talk about bathrooms, too. . . . And Justice Elena Kagan wanted to talk about bathrooms.”).

This Note does three things. First, Part I provides a descriptive account of the history of the constitutional right to bodily privacy frequently at issue in court battles over transgender students’ access to appropriate sex-segregated spaces. It observes that circuit courts began recognizing this constitutional right in the prison context at a time when the Supreme Court was weakening the enforceability of prisoners’ rights. This right soon spread to other contexts and is now a centerpiece of litigation challenging trans-friendly school bathroom policies. Part II delves deeper into the content of the right, exploring its scope and the various factors courts consider when determining whether infringing it is permissible. Part III examines how and why courts have interpreted the right to be sex-based and argues that a sex-based notion of the right impractically essentializes sex and harms transgender children.

I. The History of the Right

To better understand the constitutional right to bodily privacy in one’s partially clothed body — where it comes from, who’s invoked it, how courts have analyzed it, and how opponents of trans-friendly policies have wielded it — a brief history is in order. Rather than suggest that the right is legitimate or illegitimate because of its historical pedigree, this Part serves mainly as an observation that the ease with which courts have found a constitutional right in one context has affected their holdings in future contexts.

A. Origins

We begin in 1963 — two years after the Supreme Court incorporated the Fourth Amendment against states via the Fourteenth Amendment in Mapp v. Ohio,20×20. 367 U.S. 643, 660 (1961) (“Having once recognized that the right to privacy embodied in the Fourth Amendment is enforceable against the States, and that the right to be secure against rude invasions of privacy by state officers is, therefore, constitutional in origin, we can no longer permit that right to remain an empty promise.”). but two years before Justice Douglas’s penumbral reasoning in Griswold v. Connecticut21×21. 381 U.S. 479, 484 (1965) (“[S]pecific guarantees in the Bill of Rights have penumbras, formed by emanations from those guarantees that help give them life and substance. Various guarantees create zones of privacy.” (citation omitted)). — when the Ninth Circuit first recognized a constitutional right to bodily privacy in one’s partially clothed body in York v. Story.22×22. 324 F.2d 450 (9th Cir. 1963).

In York, a woman sued three local police officers for “taking and distributing photographs of her in the nude”23×23. Id. at 451. after she went to the police station to report an assault.24×24. See id. at 452. She argued that taking these photographs constituted “an unreasonable search within the meaning of the Fourth Amendment,” that “the Fourth Amendment is premised upon a basic right of privacy,” and that the police officers’ acts were so invasive that they “amount[ed] to a deprivation of liberty, without due process of law, as guaranteed to her by the Due Process Clause of the Fourteenth Amendment.”25×25. Id. at 454. The court separated the act of photographing the woman — which “may or may not constitute an unreasonable search in the Fourth Amendment sense” — from distributing the photographs, which “could hardly be characterized as unreasonable searches.”26×26. Id. Thus, to grant relief for the distribution of the photographs, the court found it necessary to decide whether all of the officers’ acts — the photography and the distribution — “constitute[d] an arbitrary invasion upon the security of one’s privacy in [the] Due Process sense.”27×27. Id. at 455. The Fourth Amendment protects citizens against unreasonable (and, consequently, arbitrary) police searches of their homes because these searches violate citizens’ privacy, so the court reasoned that people must have a constitutional right to bodily privacy.28×28. See id. (“We do not see how it can be argued that the searching of one’s home deprives him of privacy, but the photographing of one’s nude body, and the distribution of such photographs to strangers does not.”). There is no “more basic subject of privacy than the naked body,” and “[t]he desire to shield one’s unclothed figure from view of strangers, and particularly strangers of the opposite sex, is impelled by elementary self-respect and personal dignity.”29×29. Id.

York is an important case. Although it involves neither students, bathrooms, nor transgender people, courts and litigants continue to cite it in transgender bathroom cases for the proposition that students in schools enjoy a constitutional right to privacy in their partially clothed bodies.30×30. See, e.g., Doe ex rel. Doe v. Boyertown Area Sch. Dist., 897 F.3d 518, 527 & n.53 (3d Cir. 2018); Carcaño v. McCrory, 203 F. Supp. 3d 615, 641 (M.D.N.C. 2016); G.G. ex rel. Grimm v. Gloucester Cty. Sch. Bd., 132 F. Supp. 3d 736, 751 (E.D. Va. 2015), rev’d in part, vacated in part, 822 F.3d 709 (4th Cir. 2016), vacated, 137 S. Ct. 1239 (2017); see also Parents for Privacy v. Dall. Sch. Dist. No. 2, 326 F. Supp. 3d 1075, 1096 (D. Or. 2018) (“Plaintiffs argue that their asserted privacy right finds its genesis in . . . York v. Story . . . .”). For thirty-nine years, it was the only case to recognize and apply this right outside of the prison context.31×31. In 2002, the Second Circuit became the next court to recognize the right outside of the prison context. See Poe v. Leonard, 282 F.3d 123, 139 (2d Cir. 2002). It also represents the theoretical maximum power of the right; unlike cases involving students or prisoners — whose constitutional rights are more limited — York involved an adult plaintiff who was not imprisoned.

B. Prisons

After York, the right lay dormant for many years. Perhaps because the facts in York were so unusual, no other circuit had a reason to address whether people might have a constitutional right to privacy in their partially clothed bodies. However, beginning in the 1970s — in part because of the rise of Title VII equal employment suits against prisons that facilitated cross-sex surveillance of male prisoners by female guards32×32. See Teresa A. Miller, Sex & Surveillance: Gender, Privacy & the Sexualization of Power in Prison, 10 Geo. Mason U. C.R. L.J. 291, 297 (2000) (identifying this trend).  — courts began to grapple with a slew of cases involving prisoners claiming that their constitutional rights were violated when their nude bodies were viewed by guards of the opposite sex. Litigants and courts in transgender rights cases find it difficult to discuss the constitutional right to privacy in one’s partially clothed body without citing cases involving prisoners.33×33. See, e.g., Students & Parents for Privacy v. Sch. Dirs. of Twp. High Sch. Dist. 211, 377 F. Supp. 3d 891, 901–02 (N.D. Ill. 2019); Carcaño, 203 F. Supp. 3d at 641 (citing Lee v. Downs, 641 F.2d 1117 (4th Cir. 1981), for the proposition that “[t]here is no question that the protection of bodily privacy is an important government interest and that the State may promote this interest by excluding members of the opposite sex from places in which individuals are likely to engage in intimate bodily functions”).

Before examining these prisoners’ rights cases in more detail, it’s worth noting that, around the same time courts decided them, the Supreme Court began significantly limiting the constitutional protection afforded to the privacy rights of prisoners. In 1979, in Bell v. Wolfish,34×34. 441 U.S. 520 (1979). the Supreme Court assumed that pretrial detainees “retain some Fourth Amendment rights upon commitment to a corrections facility,”35×35. Id. at 558. but permitted strip searches and visual body-cavity inspections on less than probable cause as long as they were conducted in a reasonable manner.36×36. Id. at 560. It explained that searches are reasonable if “the significant and legitimate security interests of the institution” outweigh “the privacy interests of the inmates.”37×37. Id. Five years later, in Hudson v. Palmer,38×38. 468 U.S. 517 (1984). the Court held that “society is not prepared to recognize as legitimate any subjective expectation of privacy that a prisoner might have in his prison cell and that, accordingly, the Fourth Amendment proscription against unreasonable searches does not apply within the confines of the prison cell.”39×39. Id. at 526. Explaining that “[t]he recognition of privacy rights for prisoners in their individual cells simply cannot be reconciled with the concept of incarceration and the needs and objectives of penal institutions,” the Court greatly restricted the privacy rights of prisoners.40×40. Id.; see also Teresa A. Miller, Keeping the Government’s Hands Off Our Bodies: Mapping a Feminist Legal Theory Approach to Privacy in Cross-Gender Prison Searches, 4 Buff. Crim. L. Rev. 861, 862–63 (2001) (explaining how the effect of Hudson was “to limit drastically the degree of privacy to which prisoners could lay claim,” id. at 863). Finally, in Turner v. Safley,41×41. 482 U.S. 78 (1987). the Court announced a rational basis–like test for all violations of prisoners’ constitutional rights.42×42. Id. at 89 (“[W]hen a prison regulation impinges on inmates’ constitutional rights, the regulation is valid if it is reasonably related to legitimate penological interests.”). This trio of cases — Wolfish, Hudson, and Safley — made it increasingly easy for the government to justify violations of prisoners’ rights.43×43. See Miller, supra note 32, at 311–22. While there is not necessarily a causal relationship between the weakening of the standard of review applied to prisoners’ rights cases and more widespread recognition of the right outside of prisons, it’s possible that this doctrinal shift lowered the stakes around courts’ decisions to recognize a constitutional right to bodily privacy in one’s partially clothed body. The right is easier to recognize if it need not be enforced.

As women gained entry to all-male prisons and began serving as guards, prisoners began claiming that cross-sex surveillance violated their constitutional rights.44×44. See Rebecca Jurado, The Essence of Her Womanhood: Defining the Privacy Rights of Women Prisoners and the Employment Rights of Women Guards, 7 Am. U. J. Gender Soc. Pol’y & L. 1, 30–31 (1998). Courts were forced to determine whether male prisoners had a right to not be viewed by female guards.45×45. There were also cases involving male guards and female prisoners. Id. at 31. For an argument that courts have prioritized the privacy rights of female prisoners over those of male prisoners in these cases, see id. at 42–52. Lower courts had assumed without deciding that they did,46×46. See, e.g., Timm v. Gunter, 917 F.2d 1093, 1097 n.4 (8th Cir. 1990) (“We proceed in our analysis of this case on the assumption that inmates possess a constitutional right to privacy. For purposes of this case, it is not necessary for us to decide whether this unenumerated right exists or to define its constitutional foundation or scope, and we expressly decline to do so.”); see also Kent v. Johnson, 821 F.2d 1220, 1227 (6th Cir. 1987); Hardin v. Stynchcomb, 691 F.2d 1364, 1373 (11th Cir. 1982). but as the Supreme Court weakened the standard of review applicable to violations of prisoners’ rights, courts began recognizing the right explicitly.47×47. For examples of circuits that had recognized the right from the start, see Lee v. Downs, 641 F.2d 1117, 1120 (4th Cir. 1981); and Grummett v. Rushen, 779 F.2d 491, 494 (9th Cir. 1985). For an example of a court that revisited its assumption in an earlier case — Kent, 821 F.2d at 1227 — and later recognized the right, see Cornwell v. Dahlberg, 963 F.2d 912 (6th Cir. 1992), in which the Sixth Circuit stated that it had “joined other[ circuits] in recognizing that a convicted prisoner maintains some reasonable expectations of privacy while in prison, particularly where those claims are related to forced exposure to strangers of the opposite sex,” id. at 916. When doing so, they frequently found that this right to privacy turned at least in part on the sex of the guard viewing a prisoner’s partially clothed body,48×48. See, e.g., Canedy v. Boardman, 16 F.3d 183, 185 (7th Cir. 1994) (“[W]hile all forced observations or inspections of the naked body implicate a privacy concern, it is generally considered a greater invasion to have one’s naked body viewed by a member of the opposite sex.”); see also Cornwell, 963 F.2d at 916; Gunter, 917 F.2d at 1102; Kent, 821 F.2d at 1227; Grummett, 779 F.2d at 494; Cumbey v. Meachum, 684 F.2d 712, 714 (10th Cir. 1982); Lee, 641 F.2d at 1119. a nuance that later courts seized on when analogizing these prison cases to other contexts.

While only three circuits had recognized the right by 198749×49. See Kent, 821 F.2d at 1226–27 (first citing Lee, 641 F.2d at 1119; then citing Grummett, 779 F.2d at 493; and then citing Gunther v. Iowa State Men’s Reformatory, 612 F.2d 1079, 1086 (8th Cir. 1980), abrogated on other grounds by Hickman v. Elec. Keyboarding, Inc., 741 F.2d 230 (8th Cir. 1984)). (when the Supreme Court delivered its final blow to prisoners’ rights in Safley), the trend in the circuits was toward greater recognition of the right.50×50. Cookish v. Powell, 945 F.2d 441, 446 (1st Cir. 1991) (“[T]he trend, if not the clearly established law [in 1987], was that an inmate’s constitutional right to privacy is violated when guards of the opposite sex regularly observe him/her engaged in personal activities, such as undressing, showering, and using the toilet.”). They’ve found it in many places; courts have recognized that the First, Fourth, Eighth, or Fourteenth Amendments could serve as the basis for the right.51×51. See Miller, supra note 40, at 861–62, 861 n.2. However, as the Seventh Circuit observed, “[a]fter Wolfish and Hudson[,] monitoring of naked prisoners is not only permissible . . . but also sometimes mandatory.”52×52. Johnson v. Phelan, 69 F.3d 144, 146 (7th Cir. 1995). Even if the right to bodily privacy in one’s partially unclothed body exists, cross-sex monitoring is constitutional under certain conditions.53×53. Id. at 150–51. Today, it is well established that cross-sex monitoring of prisoners is allowed.54×54. See, e.g., Story v. Foote, 782 F.3d 968, 972 (8th Cir. 2015); Oliver v. Scott, 276 F.3d 736, 745 (5th Cir. 2002).

C. Beyond Prisons

The right soon spread outside the prison context. In 2002, the Second Circuit became the first court since York to recognize it outside of prisons. In Poe v. Leonard,55×55. 282 F.3d 123 (2d Cir. 2002). the court cited a collection of prison cases and discussed them in depth56×56. Id. at 137–38. before recognizing a “constitutional right to privacy in [one’s] unclothed body.”57×57. Id. at 139. The court also cited York. Id. at 138. Similarly, in Brannum v. Overton County School Board,58×58. 516 F.3d 489 (6th Cir. 2008). the Sixth Circuit recognized that children, too, have a constitutionally protected right to bodily privacy in their unclothed bodies that protects them from being viewed59×59. Although students retain Fourth Amendment privacy rights to be free from unreasonable searches and seizures, school officials can conduct searches of children if the officials have individualized, reasonable suspicion. See New Jersey v. T.L.O., 469 U.S. 325, 341–42 (1985) (“[T]he legality of a search of a student should depend simply on the reasonableness, under all the circumstances, of the search . . . [, and a search will be permissible when] the measures adopted are reasonably related to the objectives of the search and not excessively intrusive in light of the age and sex of the student and the nature of the infraction.”). It “requires the support of reasonable suspicion of danger or of resort to underwear for hiding evidence of wrongdoing before a search can reasonably make the quantum leap from outer clothes and backpacks to exposure of intimate parts.” Safford Unified Sch. Dist. #1 v. Redding, 557 U.S. 364, 377 (2009); cf. Doe v. Renfrow, 631 F.2d 91, 92–93 (7th Cir. 1980) (per curiam) (“It does not require a constitutional scholar to conclude that a nude search of a thirteen-year-old child is an invasion of constitutional rights of some magnitude.”). by members of the opposite sex.60×60. See Brannum, 516 F.3d at 495. When discussing the various prison cases in which the Sixth Circuit had recognized the right, the court wrote that the difference between students’ rights and prisoners’ rights is “one of degree, rather than of kind.”61×61. Id. Thus, by citing to both Leonard and various prisoners’ rights cases,62×62. Id. at 494–95 (first citing Cornwell v. Dahlberg, 963 F.2d 912, 916 (6th Cir. 1992); and then citing Kent v. Johnson, 821 F.2d 1220, 1226 (6th Cir. 1987)); id. at 499 (citing Leonard, 282 F.3d at 138–39). the Sixth Circuit became the third circuit after the Ninth Circuit in York and the Second Circuit in Leonard to recognize the right outside of the prison context. As more courts grapple with the right outside of prisons, the need to cite cases within the prison context wanes. For example, when the Third Circuit considered the issue in 2018, it didn’t even need to cite a prison case; it relied only on Leonard, Brannum, and the by-then-ancient York.63×63. Doe ex rel. Doe v. Boyertown Area Sch. Dist., 897 F.3d 518, 527 n.53 (3d Cir. 2018). As the court explained, id., it had hinted that the right existed in an earlier case — Doe v. Luzerne County, 660 F.3d 169 (3d Cir. 2011) — but Luzerne, too, cited only Leonard, Brannum, and York. See id. at 176.

Other courts, however, have been more cautious in recognizing the right. When deciding a challenge to a trans-friendly school policy, a district court in Oregon found that the plaintiffs’ citation to York and Ninth Circuit prison cases, among others, was “unpersuasive precedent that fails to establish their purported privacy right.”64×64. Parents for Privacy v. Dall. Sch. Dist. No. 2, 326 F. Supp. 3d 1075, 1096 (D. Or. 2018). Similarly, a district court in Illinois discussed various prison cases before finding that they do not “compel[] a conclusion that the right [asserted] in this case is within the bounds of substantive due process; nor do they synthesize to support a substantive due process right.”65×65. Students & Parents for Privacy v. Sch. Dirs. of Twp. High Sch. Dist. 211, 377 F. Supp. 3d 891, 902 (N.D. Ill. 2019); see also id. at 901–02.

Relying on the right as it developed in the prison context isn’t inherently good or bad, correct or incorrect. Since most courts have suggested that prisoners have a constitutional right to bodily privacy in their unclothed bodies — and because it would be odd to hold that prisoners have this right but students and members of the general population do not — courts must reckon with the application of this right outside of prisons. The cases above suggest two approaches for doing so: the Brannum and Leonard courts embraced the right as it developed in prisons, while the district courts in Oregon and Illinois cited prison cases but distinguished them, effectively limiting the applicability of the right and walking back some of their circuits’ earlier, sweeping language. The next Part explores the scope of this right.

II. The Scope of the Right

Until now, this Note has largely bracketed any discussion about what, exactly, the constitutional right to bodily privacy protects. Section II.A explores what conduct implicates the right, while section II.B details under what conditions courts have found violations of the right permissible. Finally, section II.C argues that, under the various factors courts have considered, trans-friendly policies are permissible infringements of the right.

A. Defining the Scope of the Right

For clarity, and although it is elementary, it is worth disentangling two moving parts in cases involving the constitutional right to bodily privacy in one’s partially clothed body: (1) defining the scope of the right itself and (2) determining whether infringing the right was permissible.66×66. Courts frequently fail to make this distinction explicitly. For example, in Johnson v. Phelan, 69 F.3d 144 (7th Cir. 1995), the Seventh Circuit wrote that in a prior case, the Eighth Circuit “conclude[d] that ‘opposite-sex surveillance of male inmates, performed on the same basis as same-sex surveillance,’ is constitutionally permissible.” Id. at 147 (quoting Timm v. Gunter, 917 F.2d 1093, 1102 (8th Cir. 1990)). This characterization obscures whether the court’s reasoning hinged on the absence of a prisoner’s right not to be viewed by members of the opposite sex or whether an invasion of the right was justified.

To define the right, courts must first make threshold determinations about (a) whether a constitutional right to bodily privacy in ones’ partially clothed body exists and (b) the contours of this right. For example, a court could hold — like the court in Brannum — that “the constitutional right to privacy . . . includes the right to shield one’s body from exposure to viewing by the opposite sex.”67×67. Brannum, 516 F.3d at 494. This holding means that (a) a person has a constitutional right to privacy in their partially clothed body and that (b) it would be infringed if their body was viewed, through some sort of state action, by members of the opposite sex. On the other hand, a court could hold — like the district court in Illinois — that the “right to be free from government-enforced, unconsented risk of exposure to the opposite sex when [students] . . . are partially or fully unclothed” does not exist.68×68. Students & Parents for Privacy, 377 F. Supp. 3d at 900; see id. at 900–02. This holding could be viewed in two ways: either as a rejection of a right to bodily privacy in one’s partially clothed body,69×69. Such a holding would be in deep tension with Johnson, which assumed that there is a constitutional right not to be viewed by members of the opposite sex when it analyzed a prisoner’s claim of a Fifth Amendment due process violation under Turner’s “reasonably related” standard. See Johnson, 69 F.3d at 146. If the right did not exist, the Johnson court would not have needed to apply Turner. or as a determination that a right not to be viewed exists but is not violated when there is only a risk of exposure (that is, the rights violation occurs only when there is a viewing).

The prison cases discussed in Part I featured a range of claims by prisoners. Ordered roughly from the most invasive to the least invasive practices, prisoners have sought protection from body-cavity searches,70×70. See, e.g., Hardin v. Stynchcomb, 691 F.2d 1364, 1373 (11th Cir. 1982). strip searches,71×71. See, e.g., id. deliberate viewings by members of the opposite sex,72×72. See, e.g., Johnson, 69 F.3d at 148 (“Some cases say that the Constitution forbids deliberate cross-sex monitoring (as opposed to infrequent or accidental sightings).”); Cornwell v. Dahlberg, 963 F.2d 912, 916–17 (6th Cir. 1992). and incidental viewings by members of the opposite sex.73×73. See, e.g., Timm v. Gunter, 917 F.2d 1093, 1098 (8th Cir. 1990). In the school context, cisgender plaintiffs have attempted to argue that their constitutional rights are violated not just when their partially clothed bodies are incidentally viewed by transgender students but also when transgender students are present in sex-segregated spaces (creating the risk of incidental viewing).74×74. See, e.g., Doe ex rel. Doe v. Boyertown Area Sch. Dist., 897 F.3d 518, 531 (3d Cir. 2018); see also Recent Case, Doe ex rel. Doe v. Boyertown Area School District, 897 F.3d 518 (3d Cir. 2018), 132 Harv. L. Rev. 2058, 2060 (2019) (explaining how the court rejected the “presence” right). These activities are invasive to varying degrees; a strip search or body-cavity viewing infringes one’s right to bodily privacy significantly more than an incidental viewing does. Courts, then, must articulate the exact contours of the constitutional right to bodily privacy in one’s partially clothed body; what, exactly, does it protect?

Courts should find that people have a right not to be viewed in a state of undress. This right is not only most consistent with existing case law but also recognizes that people have an expectation that their nude bodies will not be viewed by others. On the other hand, a right protecting against the risk of exposure — which no court has recognized — seems far too broad; under such a right, the presence of a female guard patrolling a male prison or a boy in a girl’s bathroom (if only to wash his hands) would amount to an infringement of a constitutional right.

B. Was Infringing the Right Permissible?

As the Third Circuit has explained, the inquiry into whether a constitutional violation has occurred is contextual. For example, a case where transgender students receive permission from the school to use their preferred sex-segregated spaces “implicates different privacy concerns than . . . a case involving an adult stranger sneaking into a locker room to watch a fourteen year-old girl shower.”75×75. Boyertown, 897 F.3d at 533. The court’s implication is not that one of these situations involves the infringement of a constitutional right and the other does not; in the Third Circuit, both acts infringe the right of the person being watched because both are viewings. Instead, its argument must be that the context of the viewing determines whether the infringement is permissible.76×76. See id. at 527–28 (finding that in the former context, a viewing “would not give rise to a constitutional violation because the School District’s policy served a compelling interest — preventing discrimination against transgender students”). Depending on various other factors — notably, whether the constitutional right to bodily privacy is a fundamental right or, as we have seen, whether the person whose rights are violated is a prisoner — a court might apply strict scrutiny, intermediate scrutiny, or rational basis review. The level of scrutiny applied affects how persuasive the government’s justification must be to allow an infringement of a constitutional right. While a broad constitutional right might not have any practical effect when recognized in the prison context because courts apply rational basis review when prisoners’ privacy rights are infringed, it could be outcome determinative if a court applies strict scrutiny when the right-enforcer is not imprisoned.

Courts have identified at least five factors that go to whether a viewing is justified: the viewer’s intent (motive), what the viewer saw (extent of exposure), how long the viewer looked (length of exposure), the cost-effective allocation of resources (efficiency), and compliance with laws like Title VII (antidiscrimination).

1. Motive. — In the prison context, some courts have made a distinction between deliberate viewings by prison guards and incidental ones;77×77. See Johnson, 69 F.3d at 148 (explaining how the other circuits have made this distinction). this distinction suggests that the motive of the rights-invader is relevant. Considering motive allows courts to screen out bad actors: some viewings are unconstitutional because the viewer acted wrongfully, while others are permissible in part because the rights-infringers had no ulterior motive.78×78. See Lee v. Downs, 641 F.2d 1117, 1121 (4th Cir. 1981) (“It should be noted that aside from the restraints imposed by the male guards upon her limbs, there is no claim of unprofessionalism in their conduct during the removal of her undergarments or the later search of her vagina. There were no offensive words or gestures or any other indecency.”).

2. Extent of Exposure. — Other courts have held that when a guard’s view of a prisoner is obstructed or distant — for example, if a guard is standing above the prisoners and thus her view is “obscured by the angle and distance” of her location — then the viewing is not a violation of the prisoner’s constitutional rights.79×79. Grummett v. Rushen, 779 F.2d 491, 495 (9th Cir. 1985) (finding that cross-sex surveillance is permissible when “the observations by the female correctional officers stationed on the gunrails overlooking the tiers and the yard areas are obscured by the angle and distance of their locations”). Taking the extent of the exposure into account allows courts to distinguish between viewings of varying invasiveness: obstructed viewings are easier to justify than unobstructed ones.

3. Length of Exposure. — Still other courts have focused on the temporal element of a violation, holding that infrequent and temporary viewings are permissible80×80. See, e.g., Michenfelder v. Sumner, 860 F.2d 328, 334 (9th Cir. 1988) (“Our circuit’s law respects an incarcerated prisoner’s right to bodily privacy, but has found that assigned positions of female guards that require only infrequent and casual observation . . . are not so degrading as to warrant court interference.” (citing Grummett, 779 F.2d at 494–95)). while taking photographs or recordings of someone’s unclothed body is not.81×81. See, e.g., Brannum v. Overton Cty. Sch. Bd., 516 F.3d 489, 498 (6th Cir. 2008) (finding that secretly videotaping students is impermissible); Poe v. Leonard, 282 F.3d 123, 126 (2d Cir. 2002) (“By 1993, it was clearly established that a police officer violates a person’s Fourteenth Amendment right to bodily privacy when that officer views, photographs or otherwise records another’s unclothed or partially unclothed body, without that person’s consent.”). Considering the length of the viewing enables courts to distinguish viewings that might have lasting privacy implications (that is, documented viewings where the length of exposure is limitless) from less invasive, temporary, unrecorded viewings.

4. Efficiency. — Courts have also considered efficiency.82×82. See, e.g., Oliver v. Scott, 276 F.3d 736, 746 (5th Cir. 2002) (“[R]equiring only male guards to supervise inmates at night and in the showers would have the ripple effect of forcing [the prison] to reassign a high percentage of its prison staff.”). In the prison context, one court noted that cross-sex viewings are justified because it “is more expensive for a prison to have a group of guards dedicated to shower and toilet monitoring . . . than to have guards all of whom can serve each role in the prison.”83×83. Johnson v. Phelan, 69 F.3d 144, 147 (7th Cir. 1995).

5. Antidiscrimination. — Finally, courts have recognized that a policy to prevent discrimination could justify a violation of the right not to be viewed in a state of undress. For example, the Seventh Circuit upheld a cross-sex monitoring policy in part because protecting a male prisoner’s right not to be viewed in a state of undress by a female guard would be possible “only by relegating women to the administrative wing, limiting their duties . . . , or eliminating them from the staff.”84×84. Id. at 148. Because allowing cross-sex monitoring — and, consequently, otherwise-unconstitutional viewings — would “reduce[] the potential for conflict with Title VII and the equal protection clause,” the court allowed it.85×85. Id. at 147.

Recognizing a constitutional right not to be viewed in a state of undress, then, allows courts to recognize the genuine interest people have in shielding their nude bodies from view. Under certain conditions, however, courts have found that this right may be infringed. The next section applies these five factors to trans-friendly sex-segregated space policies.

C. The Right in the School Context

All of the factors discussed above suggest that whatever rights-infringing viewings might occur because of trans-friendly policies are permissible.

1. Motive. — Viewings that occur while students use the bathroom are likely incidental, not intentional. Even if viewings are intentional, the motives of transgender students are likely no worse than those of an “overly curious student of the same biological sex who decides to sneak glances at his or her classmates performing their bodily functions,” as the Seventh Circuit has observed.86×86. Whitaker ex rel. Whitaker v. Kenosha Unified Sch. Dist. No. 1 Bd. of Educ., 858 F.3d 1034, 1052 (7th Cir. 2017).

2. Extent of Exposure. — This justification also points toward upholding trans-friendly policies and becomes stronger as bathrooms become more private. People are increasingly concerned with their bodily privacy generally; they want to shield their bodies from everyone, not just transgender students.87×87. See J.Y. Chua, Bathing in Controversy, The Atlantic (June 2, 2017), https://www.theatlantic.com/health/archive/2017/06/school-bathrooms-history/528978 [https://perma.cc/LGC6-T5EV]. Partially in response to this concern, school bathrooms are becoming more private,88×88. See, e.g., Jeremy P. Kelley, Transgender Debate Spurs Bathroom Changes in Kettering Schools, Dayton Daily News (Jan. 12, 2017), https://www.daytondailynews.com/news/transgender-debate-spurs-bathroom-changes-kettering-schools/rOPL57Ow597oDBe8F92DYO [https://perma.cc/HA9F-KEBW]. and schools have continued to move away from once-common communal showers, limiting and obstructing the views students have of each other’s partially clothed bodies.89×89. See Chua, supra note 87 (noting the “demise of communal showers”).

3. Length of Exposure. — Because allowing transgender students into bathrooms aligning with their gender identities results in only temporary, relatively infrequent viewings with short-term privacy implications, this factor also helps justify any rights violation. Presumably, students are not taking permanent photos or videos of each other in sex-segregated spaces.

4. Efficiency. — Trans-friendly bathroom policies could also be justified on efficiency grounds: allowing these policies would not only reduce the need or incentive for schools to provide single-use facilities for students (and thus maintain three separate sets of bathrooms) but could also pave the way for more cost-effective all-user facilities. All-user facilities could be designed to be more space-efficient and more private; they could also reduce maintenance and teacher supervision costs.90×90. Andrew LaRowe & Mike Raible, A Moment of Privacy, spaces4learning (Apr. 1, 2018), https://webspm.com/articles/2018/04/01/restrooms [https://perma.cc/8LM2-HVBY].

5. Antidiscrimination. — Finally, because school districts have enacted trans-friendly policies to prevent discrimination against transgender students (or, in a pre-Trump world, to comply with Title IX91×91. See Dear Colleague Letter, supra note 2. ), this justification also supports trans-friendly policies. Indeed, the Third Circuit has recognized that school districts have a compelling interest in not discriminating against transgender students.92×92. Doe ex rel. Doe v. Boyertown Area Sch. Dist., 897 F.3d 518, 527–28 (3d Cir. 2018).

Thus, recognizing a constitutional right to privacy in one’s partially clothed body does not mean that trans-friendly sex-segregated space policies are doomed. All of the justifications courts have relied on to uphold nude viewings in certain contexts also point toward allowing school districts to craft trans-friendly policies.

One additional hurdle to allowing trans-friendly bathroom policies, however, is that courts have typically viewed the right as sex-based. They have required more persuasive justifications when someone is viewed by a member of the opposite sex than when someone is viewed by a member of the same sex. As the next section argues, the right should not be sex-based. The extent to which the right is sex-based has important implications for the strength of the right: if it is not sex-based, then even a broad constitutional right not to be viewed in a state of partial undress is relatively weak given the prevalence of same-sex viewings historically,93×93. See Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 657 (1995) (“Public school locker rooms . . . are not notable for the privacy they afford. The locker rooms in Vernonia are typical: No individual dressing rooms are provided; shower heads are lined up along a wall, unseparated by any sort of partition or curtain; not even all the toilet stalls have doors.”). and school districts would not need particularly strong justifications to infringe on this right.

III. The Sex-Based Nature of the Right

In addition to determining the scope of the right, courts must also reckon with the fact that, in the prison context, courts have recognized a right to bodily privacy in one’s partially clothed body that depends on the viewer’s sex; the right is sex-based.94×94. See sources cited supra note 48. This sex-based conception of the right, like the right itself, has spread outside of the prison context.95×95. See Brannum v. Overton Cty. Sch. Bd., 516 F.3d 489, 494 (6th Cir. 2008) (defining the right in terms of being viewed by a member of the opposite sex). Even when courts do not explicitly articulate the sex-based nature of the right, they must implicitly believe that any constitutional right not to be viewed in a state of undress is sex-dependent.96×96. The court in Boyertown never explicitly defined the right in gendered terms. See 897 F.3d at 527. However, the Third Circuit precedent the Boyertown court relied on suggests that the right is sex-dependent. See id. at 527 n.53 (citing Doe v. Luzerne County, 660 F.3d 169, 177 (3d Cir. 2011) (finding that people have “a reasonable expectation of privacy . . . particularly while in the presence of members of the opposite sex”)). For example, in the school context, if sex were irrelevant to the right, how could a transgender boy in the boys’ bathroom have different constitutional privacy concerns than a cisgender boy in the boys’ bathroom? Section III.A explores the two main arguments in favor of a sex-based constitutional right. Section III.B argues that the constitutional right to privacy should not be sex-based.

A. The Justifications of a Sex-Based Right

Proponents of a sex-based conception of the right have offered two main justifications: the “nudity taboo”97×97. Johnson v. Phelan, 69 F.3d 144, 152 (7th Cir. 1995) (Posner, C.J., concurring and dissenting). and public safety.

1. The “Nudity Taboo” Justification. — One basis of this sex-based conception seems to be the “nudity taboo.” As Chief Judge Posner explained, “[t]he nudity taboo retains great strength in the United States. It should not be confused with prudery. It is a taboo against being seen in the nude by strangers, not by one’s intimates. . . . The taboo is particularly strong when the stranger belongs to the opposite sex.”98×98. Id. He also noted that not all Americans share the nudity taboo but that it is strongest among Christians.99×99. Id. He concluded that it should be constitutionalized: “the interest of a prisoner in being free from unnecessary cross-sex surveillance has priority over the unisex-bathroom movement.”100×100. Id. Closely related to the nudity taboo is the social convention that private acts like undressing in front of members of the opposite sex often “support the possibility of intimate private actions.”101×101. Vincent J. Samar, The Right to Privacy and the Right to Use the Bathroom Consistent with One’s Gender Identity, 24 Duke J. Gender L. & Pol’y 33, 54 (2016). More bluntly, people undress to have sex; they don’t like doing it in front of strangers, particularly members of the opposite sex. Scholars and judges have criticized this “nudity taboo” logic along many dimensions: accepted practices like cross-sex doctor and nurse examinations undermine the taboo’s natural law appeal,102×102. See Miller, supra note 32, at 354 (“[A]ccepted practices such as medical examination by a physician or nurse of the opposite sex, co-ed hot tubs and saunas and the constitutionally protected practice of nude dancing belie the strength of such a norm in the United States.”). Posner’s statements about Christianity are inaccurate,103×103. See Mary Anne Case, Essay, All the World’s the Men’s Room, 74 U. Chi. L. Rev. 1655, 1661 (2007) (arguing that the nudity taboo is far stronger in orthodox Judaism and Islam, that early Christians were far more opposed to luxurious dress than nudity, and that “taboos on nudity in Western culture not only extend to, but often actually focus on, having the nudity of others displayed to one”). and the taboo — to the extent it is rooted in sexual desire — fails to account for nonheterosexual viewers.104×104. See Canedy v. Boardman, 16 F.3d 183, 185 n.1 (7th Cir. 1994) (“We note that York, many of the cases discussed below involving cross-gender observations and strip searches, as well as [petitioner’s] brief on appeal . . . assume that all of the relevant actors are heterosexual.”); see also Johnson, 69 F.3d at 147 (“There are too many permutations to place guards and prisoners into multiple classes by sex, sexual orientation, and perhaps other criteria, allowing each group to be observed only by the corresponding groups that occasion the least unhappiness.”).

2. The Public Safety Justification. — Opponents of trans-friendly policies have also attacked them on public safety grounds. The fear is usually articulated in gendered terms: people don’t want men going into women’s bathrooms. For example, in 2015, Houston tried to enact an antidiscrimination ordinance designed to protect transgender residents. Opponents of the ordinance ran a television ad featuring a young girl in a bathroom; the voiceover warned that if the ordinance were enacted, “[e]ven registered sex offenders could follow women or young girls into the bathroom.”105×105. Russell Berman, How Bathroom Fears Conquered Transgender Rights in Houston, The Atlantic (Nov. 3, 2015), http://www.theatlantic.com/politics/archive/2015/11/how-bathroom-fears-conquered-transgender-rights-in-houston/414016 [https://perma.cc/PR4R-3GZS] (quoting Campaign for Hous., Campaign for Houston — TV Spot 1, YouTube (Oct. 13, 2015), https://www.youtube.com/watch?v=D7thOvSvC4E [https://perma.cc/6NHH-Q8WC]). “Prevent danger,” it advised, and “[v]ote no” on the bathroom ordinance.106×106. Id. (quoting Campaign for Hous., supra note 105). The ordinance did not pass.107×107. Id.

Despite the prevalence of safety concerns, there is little empirical evidence justifying these fears. A 2018 study found that “reports of privacy and safety violations in public restrooms, locker rooms, and changing rooms were exceedingly rare” after legislatures or voters enacted antidiscrimination policies like the one Houston failed to pass; it concluded that fears of increased privacy and safety violations “are not empirically grounded.”108×108. Amira Hasenbush et al., Gender Identity Nondiscrimination Laws in Public Accommodations: A Review of Evidence Regarding Safety and Privacy in Public Restrooms, Locker Rooms, and Changing Rooms, 16 Sexuality Res. & Soc. Pol’y 70, 80 (2019). The concern also seems irrational. Men who want to sexually assault women in bathrooms would probably be no more deterred by trans-exclusionary laws than by the criminal penalties that already apply.109×109. See Samar, supra note 101, at 57. Further, due to the stigmatized nature of trans identities, it seems unlikely that people would be willing to abuse antidiscrimination policies by claiming genders that are not their own.110×110. See Jessica A. Clarke, They, Them, and Theirs, 132 Harv. L. Rev. 894, 972 (2019).

B. Against a Sex-Based Right

On the other hand, there are many reasons why the right should not be sex-based. Sex and gender are not binary, and a sex-based right essentializes sex, is difficult to administer, and disproportionately harms transgender children.

1. A Sex-Based Right Presupposes a Binary Notion of Sex and Gender. — As currently conceived by courts, the constitutional right to bodily privacy in one’s unclothed body does not reflect the complexities of sex and gender. While courts typically conflate “sex” and “gender,” using “gender” when they mean “sex” and vice versa,111×111. Jessica A. Clarke, Identity and Form, 103 Calif. L. Rev. 747, 763 n.78 (2015). the concepts are distinct: sex is determined by a mix of “chromosomes, hormones, internal and external reproductive organs, and secondary sex characteristics” and typically assigned at birth,112×112. GLAAD Media Reference Guide — Transgender, GLAAD, https://www.glaad.org/reference/transgender [https://perma.cc/4YTE-EBB5]. while gender is both socially constructed and performed.113×113. See Judith Butler, Gender Trouble: Feminism and the Subversion of Identity 7–8, 137–41 (1990). Neither is necessarily binary.114×114. See Clarke, supra note 110, at 897–98. Someone’s “gender identity is their subjective, deep-core sense of self as being a particular gender.”115×115. Doe ex rel. Doe v. Boyertown Area Sch. Dist., 897 F.3d 518, 522 (3d Cir. 2018). These distinctions between sex and gender are not new in legal academia,116×116. For a 1995 example of legal scholarship arguing for a “postmodern account of gender,” drawing heavily from the work of Judith Butler, see Note, Patriarchy Is Such a Drag: The Strategic Possibilities of a Postmodern Account of Gender, 108 Harv. L. Rev. 1973 (1995). but they’re sufficiently foreign to legal practice that the Third Circuit found it necessary to define “sex,” “gender,” and “gender identity” in 2018.117×117. Boyertown, 897 F.3d at 522. A sex-based right — first conceived in 1963,118×118. See York v. Story, 324 F.2d 450, 455 (9th Cir. 1963). when sex and gender were widely considered both interchangeable and immutable119×119. For a history of the evolution of sex, gender, and gender identity, see Milton Diamond, Sex, Gender, and Identity over the Years: A Changing Perspective, 13 Child & Adolescent Psychiatric Clinics N. Am. 591, 591–95 (2004). — ignores a modern understanding of identity.

2. A Sex-Based Right Essentializes Sex. — If courts interpret a right to bodily privacy in one’s partially clothed body as a sex-based or gendered one, they essentialize and aggrandize one form of identity over others.120×120. See generally Clarke, supra note 111, at 754–69 (explaining that two predominant theories of identity are ascriptive and elective, that ascriptive identity includes “any definition that assigns identity labels based on whether an individual meets certain biological, social, or cultural standards that are considered objective,” id. at 757, that an elective model of identity “would require legal recognition of all choices by individuals to identify as men or women,” id. at 763, and that many scholars and advocates support an “elective sex,” id. at 764); Janet E. Halley, Sexual Orientation and the Politics of Biology: A Critique of the Argument from Immutability, 46 Stan. L. Rev. 503, 547 (1994) (“Essentialism assumes at minimum that a pure and perfect definition of a particular thing can be found. . . . [But i]f this is what an essence is, clearly nature cannot be the only source of essences.”); Martha Minow, Identities, 3 Yale J.L. & Human. 97, 127 (1991) (“[U]nderestimating individuals’ latitude for choice despite their assigned identities, and failing to acknowledge the constraints on individuals despite the powers to choose, are two central mistakes in legal assessments of identity.”). Essentialization isn’t bad because, in this context, it harms transgender people; it’s bad because it requires the court to wade into a messy debate about who is — and who is not — a man or woman, and because it requires the court to enforce a gender binary that is increasingly questioned in society and the law.121×121. See Clarke, supra note 110, at 897, 900 (contemplating “what American law would look like if it took nonbinary gender seriously,” id. at 900, and noting that, as of 2019, eight states (and four foreign countries) have recognized nonbinary identities through legal documents, id. at 897). Scholars have also argued that the gender binary reinforces sex stereotypes. See, e.g., Butler, supra note 113, at 25 (“[G]ender proves to be performative — that is, constituting the identity it is purported to be.”); Kenji Yoshino, Covering, 111 Yale L.J. 769, 933 (2002) (“Protecting particular traits as constitutive of particular identities thus risks essentializing those identities as always embracing those traits.”). By choosing to define the right in terms of the biological sex of the viewer, courts discount the importance of gender identity. And, even if courts defined the right in terms of gender identity rather than sex, they’d still be constitutionalizing and essentializing some form of identity (and would thus be discounting the views of people who — like the cisgender students who don’t want to risk being viewed by transgender students in bathrooms — believe that biological sex is the only relevant identity). By jettisoning the sex-based nature of the right, courts can sidestep the problems inherent in embedding one particular identity into a constitutional right.122×122. See Clarke, supra note 111, at 830 (“Defining identities contextually . . . is less likely to pigeonhole individuals into categories that do not match their lived experiences.”).

3. A Sex-Based Right Is Difficult to Administer. — Beyond the theoretical messiness of embedding sex into a constitutional right, there’s also a practical difficulty in a sex-based notion of the right. A sex-based right requires a more precise definition of sex than the law is willing or prepared to offer.123×123. Cf. Olga Tomchin, Comment, Bodies and Bureaucracy: Legal Sex Classification and Marriage-Based Immigration for Trans* People, 101 Calif. L. Rev. 813, 856 (2013) (“Trans* people’s legal sex is entirely positional and dependent on who is asking, where they are, and where they were born.”). Does sex mean sex assigned at birth? If not, is sex mutable? If so, at what point does someone switch from being one sex to another sex — when, if ever, is a transition complete? And what about people who are nonbinary? Once the right is sex-based, courts must either assume or determine the sex of the purported rights-infringer and the sex of the person whose rights are infringed. Courts might not want to get into this business.

Perhaps the most administrable definition of sex is sex assigned at birth. Some courts have used this definition, in accord with an English case from 1970 holding that biological, birth-determined sex is supreme and immutable.124×124. See Corbett v. Corbett [1971] P 83 at 104 (Eng.) (“It is common ground between all the medical witnesses that the biological sexual constitution of an individual is fixed at birth (at the latest), and cannot be changed, either by the natural development of organs of the opposite sex, or by medical or surgical means. The respondent’s operation, therefore, cannot affect her true sex.”); see also In re Estate of Gardiner, 42 P.3d 120, 133–37 (Kan. 2002) (discussing Corbett, id. at 133, before holding that despite undergoing “electrolysis, thermolysis, tracheal shave, hormone injections, extensive counseling, and reassignment surgery,” the transgender appellant “remains a transsexual, and a male for purposes of marriage,” id. at 137); In re Declaratory Relief for Ladrach, 513 N.E.2d 828, 832 (Ohio Prob. Ct. 1987); Littleton v. Prange, 9 S.W.3d 223, 227–31 (Tex. Ct. App. 1999) (citing Corbett for the proposition “once a man, always a man,” id. at 227, before holding that a transgender woman was male as a matter of law because “[t]here are some things we cannot will into being. They just are,” id. at 231); Sonia K. Katyal, The Numerus Clausus of Sex, 84 U. Chi. L. Rev. 389, 427 (2017) (“[M]any courts still cling to the presumption that sex cannot be changed.”). But this definition fails to recognize that the sex someone is assigned at birth may not be correct, and it locks people into this assignment forever. Opponents of transgender rights might not like this definition either. Most of the rhetoric against allowing transgender people to use bathrooms consistent with their identity raises the specter of men in the women’s bathroom.125×125. See, e.g., supra p. 1699. But interpreting sex to mean sex assigned at birth might find a constitutional violation when a stereotypically masculine transgender man who’s undergone sex reassignment surgery enters the men’s room, but no violation when that man enters the women’s room. To the extent that the safety and nudity taboo arguments against trans-friendly policies are fundamentally driven by a fear of people who look like men using the women’s room and vice versa,126×126. Concern with appearance does seem to be a driving factor; backlash against trans-friendly policies has led cisgender people to harass other cisgender people whom they perceive as trans. See, e.g., Matt DeRienzo, Woman Mistaken for Transgender Harassed in Walmart Bathroom, NewsTimes (May 16, 2016, 3:50 PM), https://www.newstimes.com/local/article/Woman-mistaken-for-transgender-harassed-in-7471666.php [https://perma.cc/ZU2R-ALCS]. relying on sex assigned at birth does not address the problem.

Some courts unwilling to tie a person’s sex to the sex that person was assigned at birth have instead relied on the concept of legal sex. For example, a district court in the District of Columbia was willing to find that a transgender woman was “legally a female” and that “the law on cross-gender searches [was] relevant” when male prison guards strip searched her.127×127. Shaw v. District of Columbia, 944 F. Supp. 2d 43, 55 (D.D.C. 2013). The court explained that the terminology the defendants used to refer to the plaintiff — they said she was “biologically male, but ‘legally female,’”128×128. Id. at 55 n.26. — was meaningless; the court treated her as a woman because she had undergone gender affirming surgery and “had her sex legally changed to female.”129×129. Id. at 47. But legal sex is jurisdiction dependent and isn’t always clear; people may have a cornucopia of government documents with conflicting sex classifications.130×130. See Katyal, supra note 124, at 412 (“Gender classification is a primary power of the state, but as scholars have shown, it is an inordinately messy, shifting, complex, and contradictory set of rules, demonstrating a near total absence of coherence.”); see also Tomchin, supra note 123, at 856 (“Trans* people often have an assortment of documentation with mismatched sex markers.”). While some states allow people to change their legal sex, others — like Tennessee — don’t, even after people undergo sex reassignment surgery.131×131. See Transgender Law Ctr., State-by-State Overview: Rules for Changing Gender Markers on Birth Certificates (2017), http://transgenderlawcenter.org/wp-content/uploads/2016/12/Birth-Cert-overview-state-by-state.pdf [https://perma.cc/NG7B-YTZ9] (detailing the cacophony in state laws and explaining that, in Tennessee, a statute “specifically forbids the correction of sex designations on birth certificates for transgender people,” id. at 4). Some scholars have criticized the requirement that people undergo sex reassignment surgery before changing their legal sex because it reduces autonomy, conditions access to care on a potentially stigmatizing diagnosis, and could restrict access to the rich. See Dean Spade, Commentary, Resisting Medicine, Re/modeling Gender, 18 Berkeley Women’s L.J. 15, 18, 24 (2003). Legal sex, then, is too indeterminate a foundation on which to base the constitutional right to privacy in one’s partially clothed body. Courts might instead conduct their own investigation.

The inquiry a sex-based right would require is reminiscent of the inquiry into race that courts performed to enforce miscegenation laws in the nineteenth century. As Professor Peggy Pascoe has written, although “[m]ost Americans are sure they know race when they see it, . . . very few can offer a definition of the term.”132×132. Peggy Pascoe, What Comes Naturally: Miscegenation Law and the Making of Race in America 111 (2009). When miscegenation laws made race classification necessary — forcing judges, juries, and marriage license clerks to “assign people to racial categories” — these categories became increasingly arbitrary.133×133. Id. at 111, 113. To prove race, judges and juries relied heavily on physical markers; for example, one woman was asked to “partially disrobe before a jury in an attempt to uncover supposedly persuasive physical evidence of race.”134×134. Id. at 113. Appellate courts — deprived of the opportunity to conduct visual inspections of the parties — often overturned lower court rulings; they were better able to “notice the discrepancies and contradictions that marked race-making in local courts.”135×135. Id. at 115 (noting that while appellate courts initially upheld convictions, they eventually began to “diverge from local court judgments”).

If courts were to enforce a constitutional right not to be viewed by a member of the opposite sex, they’d be forced to make similar judgments. They seem wary of doing so; when sex is unclear, many courts have been cautious with their language, referring to transgender people by the correct pronouns but explaining that this does not reflect a determination of their sex.136×136. See, e.g., Diamond v. Owens, 131 F. Supp. 3d 1346, 1353 n.1 (M.D. Ga. 2015) (“All parties refer to Diamond with feminine pronouns. Consequently, the Court does the same. This should suggest nothing other than consistency.”). Requiring courts to determine sex also risks reducing transgender people to their appearance or “a specific set of behaviors and anatomical differences.”137×137. Katyal, supra note 124, at 429. A vivid example of this dynamic comes from a recent study of how immigration courts determine transgender identity in asylum cases. There, courts forced to make determinations about sex or gender typically based their decisions on whether claimants presented a clear narrative of their transitions from one sex to another.138×138. Stefan Vogler, Determining Transgender: Adjudicating Gender Identity in U.S. Asylum Law, 33 Gender & Soc’y 439, 447 (2019) (“[C]onsistent, linear identity narratives that suggest a fixed gender identity and clear movement from one gender to its assumed opposite continue to predominate as the preferred ‘proof’ of transgender status.”). When writing their opinions, these courts focused on factors such as “a claimant’s choice of name, hairstyle, dress, and body modifications.”139×139. Id. at 454. Relying on these factors risks turning courtrooms into stages on which transgender people must act out their gender identities.

4. A Sex-Based Right Disproportionately Harms Transgender Children. — In states where sex reassignment surgery is necessary to change one’s legal sex, children will be effectively barred from having a sex other than their birth sex legally recognized.140×140. See Eli Coleman et al., World Prof’l Ass’n for Transgender Health, Standards of Care for the Health of Transexual, Transgender, and Gender-Nonconforming People 21 (7th ed. 2012) (“Genital surgery should not be carried out until (i) patients reach the legal age of majority to give consent for medical procedures in a given country, and (ii) patients have lived continuously for at least 12 months in the gender role that is congruent with their gender identity.”). Navigating legal sex might also be intermediated through a parent; some states require doctors’ notes or other medical documents before legal sex can be recognized, and some transgender children — particularly those who have been rejected by their families — may be unable to access care.141×141. If transgender children must start hormone therapy before doctors will provide a note to the government, family rejection could present an even greater barrier; hormone therapy typically requires parental consent. Emily Ikuta, Note, Overcoming the Parental Veto: How Transgender Adolescents Can Access Puberty-Suppressing Hormone Treatment in the Absence of Parental Consent Under the Mature Minor Doctrine, 25 S. Cal. Interdisc. L.J. 179, 189 (2016). For an argument that children might still be able to access hormone therapy absent parental consent, see Ikuta, supra. Transgender children may also be unsure of their sex or gender identity; medical interventions for transgender adolescents are often designed to give them “more time to explore their gender nonconformity.”142×142. Coleman et al., supra note 140, at 19.

Declining to recognize a sex-based notion of the right won’t necessarily lead to bathroom anarchy. In some sense, a non-sex-based version of the right is broader than a sex-based one; it is more protective of people’s privacy. Instead of protecting people’s privacy only when they are viewed by members of the opposite sex, a non-sex-based right would provide people with the same constitutional protections regardless of who views them. If existing sex-segregated spaces are insufficiently private to protect such a broad right, then the proper response is to increase privacy in these spaces (which would benefit everyone who’s ever felt uncomfortable in communal facilities, not just transgender people), not to bar transgender people from facilities aligning with their gender identity and force courts to adjudicate their sex.

Conclusion

The acceptance of a constitutional right to bodily privacy in one’s partially clothed body was not inevitable. Until 2002, only a single circuit court had affirmatively recognized the right. However, a combination of increased cross-sex monitoring in prisons — largely due to sex discrimination litigation under Title VII — and a steady weakening of prisoners’ rights coincided with a widespread acceptance of the premise that people have a constitutional right not to be viewed in a state of undress by members of the opposite sex. More recently, courts have relied on the acceptance of this premise in prisoners’ rights cases to find that the right does, indeed, exist. And cisgender students have tried to harness the right to prevent school districts from allowing transgender students into sex-segregated spaces aligning with their gender identities.

The increasing adoption of the right leads to interesting questions about what, exactly, the right entails. While courts have been somewhat vague about the scope of the right, most seem to agree that it is fairly broad and is implicated whenever one person views another in a state of undress. While a broad right sensibly recognizes the genuine privacy interest people have in their partially clothed body, courts have articulated a number of factors that go to whether invading this right is permissible; all of these point toward allowing trans-friendly bathroom policies. Most courts have also, either implicitly or explicitly, found that this right is sex-based. A sex-based right is problematic not only because it constitutionalizes a nudity taboo, essentializes sex at the expense of other forms of identity, and disproportionately harms transgender children, but also because judicial policing of sex is messy and difficult. Courts should not get in the business of saying who is — and who is not — a man or woman. By rejecting the notion of a sex-based constitutional right to privacy in one’s partially clothed body, they won’t have to.