Adjudicating Identity

University of South Carolina University of South Carolina

Scholar Commons Scholar Commons

Faculty Publications Law School

2022

Adjudicating Identity Adjudicating Identity

Laura Lane-Steele

University of South Carolina School of Law

, [email protected]

Follow this and additional works at: https://scholarcommons.sc.edu/law_facpub

Part of the Law Commons

Recommended Citation Recommended Citation

Laura Lane-Steele, Adjudicating Identity, 9 TEX. A&M L. REV. 267 (2022).

This Article is brought to you by the Law School at Scholar Commons. It has been accepted for inclusion in Faculty

Publications by an authorized administrator of Scholar Commons. For more information, please contact

[email protected].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 1 10-MAR-22 14:18

ADJUDICATING IDENTITY

by: Laura Lane-Steele*

BSTRACT

Legal actors examine identity claims with varying degrees of inten-

sity. For instance, to be considered “female” for the U.S. Census, self-

identification alone is sufficient, and no additional evidence is neces-

sary. To change a sex marker on a birth certificate to “female,” how-

ever, self-identification is not enough; some states require people to

show that they do not have a penis to be considered “female.” Similar

examples of discrepancies in the type and amount of evidence consid-

ered for identity claims abound across identities and areas of law. Yet

legal actors rarely acknowledge that they are adjudicating identity in the

first place, much less explain or justify the varying levels of scrutiny

exacted upon identity claims. This Article attempts to make sense of

identity adjudication by providing a taxonomy that explains why some

identity claims are interrogated more than others. Taking a broad view

of identity adjudication, it examines three types of laws (data-collection,

anti-discrimination, and benefit laws) as well as four identity categories

(religion, sexual orientation, sex, and race) and concludes that both the

type of law at issue and the identity category affect how an identity

claim is adjudicated. It then argues that across identities and types of

laws, legal actors are often adjudicating identity without proper atten-

tion to the particular legal context and examining the wrong type of

identity evidence in light of the specific law at issue. This context-de-

tached approach to identity adjudication produces inconsistent and in-

coherent results; it can also impinge on privacy interests and reinforce

problematic stereotypes. This Article calls for a context-informed ap-

proach to identity adjudication, where the question of identity is linked

to the function of the specific law rather than treated as an independent

and stable “truth” about an individual.

https://doi.org/10.37419/LR.V9.I2.1

* Forrester Fellow, Tulane Law School. For helpful discussions, I am grateful to

Noa Ben-Asher, Meghan Boone, Jessica Clarke, Onnig Dombalagian, Elizabeth

Emens, Adam Feibelman, D. Wendy Greene, Jancy Hoeffel, Sonia Katyal, Holning

Lau, Ann Lipton, Kenneth Mack, Saru Matambanadzo, Angela Morrison, Emily

Prifogle, Stacy Seicshnaydre, Meili Steele, Emily Stolzenberg, and Mary Yanik. Addi-

tional thanks to the participants of the Intellectual Life Speaker Series at Tulane Law

School, the 2021 Summer Feminist Legal Theory Series, the Law & Society Annual

Conference, the Law & Society Early Career Workshop, and the Michigan Law Jun-

ior Scholars conference. I thank Katie McCarty and Meredith Clement for excellent

research assistance and Spencer Lockwood, Tayler Berlin, Nhu-Thinh Nguyen,

Madison Feyrer-Melk, and the other editors at the Texas A&M Law Review for their

outstanding comments and edits.

267

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 2 10-MAR-22 14:18

268 TEXAS A&M LAW REVIEW [Vol. 9

ABLE OF

ONTENTS

I. I

NTRODUCTION

.......................................... 268

II. D

EFINITIONS AND

UTABILITY

......................... 275

A. Types of Identity Evidence .......................... 275

B. Mutability ........................................... 277

III. A T

AXONOMY OF

DENTITY

DJUDICATION

............ 278

A. Data-Collection Laws ............................... 280

B. Anti-Discrimination Laws ........................... 284

1. Sexual Orientation .............................. 284

2. Religion ........................................ 286

3. Race ............................................ 288

4. Sex ............................................. 292

C. Benefit Laws ........................................ 298

1. Sexual Orientation .............................. 299

2. Religion ........................................ 302

3. Race ............................................ 304

4. Sex ............................................. 310

D. Possible Explanations ............................... 316

IV. C

ONTEXT

-D

ETACHED

& C

ONTEXT

-I

NFORMED

DJUDICATION

.......................................... 320

A. Data-Collection Laws ............................... 320

B. Anti-Discrimination Laws ........................... 322

1. Identity Adjudication and Black Letter Law.... 322

2. Identity Adjudication and Normative Theories

of Anti-Discrimination Law..................... 326

3. Problems Caused by Context-Detached

Adjudication .................................... 329

C. Benefit Laws ........................................ 331

1. Sex ............................................. 331

2. Sexual Orientation .............................. 337

3. Race ............................................ 339

4. Problems Caused by Context-Detached

Adjudication .................................... 343

V. C

ONCLUSION

............................................ 345

I. I

NTRODUCTION

Identity has been adjudicated in the court of public opinion with

increasing frequency over the past decade. Rachel Dolezal’s claim to a

biracial identity was generally accepted in her community because of

her appearance, but when her ancestry came to light, most people re-

jected her racial self-identification and deemed her to be white.

Pro-

fessor Andrea Smith self-identifies as Native American, but because

she has been unable to show that neither she nor any of her relatives

1. See

OGERS

RUBAKER

RANS

: G

ENDER AND

ACE IN AN

GE OF

NSET-

TLED

DENTITIES

1–3 (2016).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 3 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 269

are citizens of a tribe, her identity claim has been largely rejected.

Sex is also publicly adjudicated. In 2015, Caitlyn Jenner came out as a

transgender woman,

and Demi Lovato recently disclosed that they

identify as non-binary.

Most people treat these changes in sex self-

identification as legitimate, and failing to accept these individuals’ sex

identities is viewed as discriminatory in many circles.

Whether an in-

dividual’s identity claim is accepted by society can have tremendous

effects on that individual’s life—Dolezal has become a social pariah

and lost her jobs as the chapter president for the Spokane NAACP

and as an instructor of Africana Studies at Eastern Washington Uni-

versity. Jenner, on the other hand, ran for governor of California.

Identity adjudication is also happening in the mundane moments of

our everyday lives. Problems can arise when others adjudicate our

identities improperly, unnecessarily, or in ways that conflict with our

self-identification. Two examples from my life follow. On multiple oc-

casions, when I have entered the women’s restroom, employees of the

establishment or other patrons have adjudicated my sex to be male

based on my appearance. They then followed me into the restroom to

inform me that I was in the wrong restroom. At the airport security

checkpoint, the full body scanner machines require the TSA agents to

indicate whether the subject being scanned is male or female. TSA

agents frequently adjudicate my sex as male, which causes the ma-

chine to alert and sometimes results in me going through additional

screening.

2. See Sarah Viren, The Native Scholar Who Wasn’t,

N.Y. T

IMES

https://

www.nytimes.com/2021/05/25/magazine/cherokee-native-american-andrea-smith.html

(May 28, 2021) [https://perma.cc/TW7Q-JB3F]. Other examples of racial self-identifi-

cations being adjudicated include the controversies over Natasha Bannan and Jessica

Krug. See Raul A. Reyes, Who’s Really Latina? A Recent Controversy Draws Outrage

Over Identity, Appropriation,

NBC N

EWS

(Feb. 28, 2021, 5:00 AM), https://

www.nbcnews.com/news/latino/who-s-really-latina-recent-controversy-draws-outrage-

over-identity-n1258141 [https://perma.cc/E35F-Q5RW] (discussing Natasha Bannan,

who identified as Latina, despite not having Latin American ancestry and Jessica

Krug, who previously identified as Black but then revealed that she had mispresented

her ancestry).

3. Buzz Bissinger, Caitlyn Jenner: The Full Story,

ANITY

AIR

(June 25, 2015),

https://www.vanityfair.com/hollywood/2015/06/caitlyn-jenner-bruce-cover-annie-lei-

bovitz [https://perma.cc/JW72-HAW7].

4. Demi Lovato Is Non-Binary and Is Changing Pronouns to They/Them, Singer

Announces,

BBC N

EWS

(May 19, 2021), https://www.bbc.com/news/newsbeat-

57169541 [https://perma.cc/U69Z-W2PF].

5. New Poll Shows Americans Overwhelmingly Oppose Anti-Transgender Laws,

PBS N

EWS

OUR

(Apr. 16, 2021, 5:00 AM), https://www.pbs.org/newshour/politics/

new-poll-shows-americans-overwhelmingly-oppose-anti-transgender-laws [https://

perma.cc/645S-ZUCC].

6. See

RUBAKER

, supra note 1, at 3–4 (contrasting negative reaction to Dolezal

with positive reactions to Jenner); Jennifer Medina & Maggie Haberman, Caitlyn Jen-

ner Announces Run for California Governor,

N.Y. T

IMES

(Apr. 26, 2021), https://

www.nytimes.com/2021/04/23/us/politics/caitlyn-jenner-california-governor.html (Apr.

26, 2021) [https://perma.cc/BHV4-4ZMK].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 4 10-MAR-22 14:18

270 TEXAS A&M LAW REVIEW [Vol. 9

As these examples show, identity adjudication is both ubiquitous

and consequential. Sometimes less obviously, but at least as fre-

quently, identity is formally adjudicated in law. In some contexts, peo-

ple are taken at their word that they are a particular identity, and

other times, they have to put forth more evidence to prove their iden-

tity claim. For example, self-identification is enough for the Census;

you check a box indicating your race, sex, marital status, etc., and you

don’t need to provide any other evidence.

But if you seek asylum in

the United States based on your sexual orientation, self-identification

would not be sufficient; you would need to provide additional evi-

dence, like proof of prior same-sex relationships, corroborating testi-

mony from friends and family, or evidence that you faced homophobic

harassment.

Some identity claims are closely scrutinized while other

are not.

Yet legal actors do not usually acknowledge that they are

adjudicating identity in this way, much less explain why some identity

claims are examined more closely than others.

This Article seeks to systematically understand and make visible

identity adjudication in the law. It brings to the surface the implicit,

unidentified, and unconscious factors legal actors use to adjudicate

identity and provides a taxonomy of identity adjudication. Specifi-

cally, it develops this taxonomy by examining three types of laws—(1)

data-collection laws (like the Census), (2) anti-discrimination laws

(like Title VII), and (3) “benefit” laws, which involve individuals ask-

ing or petitioning a state actor to affirmatively do something on their

behalf

—and four identity categories: (1) race, (2) sex, (3) sexual ori-

entation, and (4) religion.

Two patterns emerge that tend to affect how identity claims are ad-

judicated. The first pattern goes to the type of law at issue. For data-

collection laws, the evidentiary burden for identity claims is low, and

self-identification alone is generally sufficient.

This burden increases

with anti-discrimination laws and is at its highest with benefit laws.

For example, self-identification is typically sufficient to prove racial

identity in the context of a data-collection law;

for a benefit law,

however, legal actors require more, such as ancestral evidence or prior

consistent racial self-identifications.

The second pattern goes to the

type of identity at issue. Identities the law treats as immutable (race

and sex) are more intensely scrutinized than the identities the law

7. E.g., About Race,

U.S. C

ENSUS

UREAU

, https://www.census.gov/topics/popu-

lation/race/about.html (Dec. 3, 2021) [https://perma.cc/E6UZ-PZJ9].

8. See generally infra Part III(C)(1).

9. See generally infra Part III(C)(1).

10. This Article’s definition of “benefit” law is explained more fully infra in Part

III(C).

11. See infra Part III(A).

12. See generally infra Part III.

13. E.g., About Race, supra note 7.

14. See infra Part III(C).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 5 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 271

treats as more mutable (sexual orientation and religion). For example,

sex adjudication for a benefit law generally requires proof of certain

biological traits.

But religious identity for a benefit law is usually

accepted if the individual can provide some corroborating evidence of

their identity claim.

Thus, (1) the immutability of the identity cate-

gory and (2) the type of law both affect the way legal actors treat

identity claims.

Making explicit this taxonomy of identity adjudication reveals a

problematic pattern across identities and areas of law—namely, that

legal actors tend to adjudicate identity in what I call a “context-de-

tached,” rather than a “context-informed,” manner. A context-de-

tached approach neither examines why the particular law asks about

the identity in the first place nor considers how the purpose or func-

tion of the law may influence how identity should be adjudicated.

Context-detached identity adjudication is not disciplined or regulated

by the applicable law; rather, it is guided by whatever legal actors

think they “know” about particular identities. Subjective and unar-

ticulated definitions of identity—formed by the particular legal actor’s

social, cultural, and political positionalities—dominate the context-de-

tached approach to identity adjudication. For example, a legal actor

who adopts this approach may define a “woman” as someone with

breasts who ascriptively appears to be a woman (at least to that legal

actor) and then applies that definition across different areas of law,

regardless of whether breasts or ascriptive factors matter to the law.

A context-detached approach to identity adjudication is problem-

atic for at least two reasons. First, and most obviously, it can produce

results that are doctrinally inconsistent or at odds with the purpose of

the law. Consider an example from anti-discrimination law. In racial

discrimination cases, some courts adjudicate the plaintiff’s “actual”

race.

In these cases, courts may examine the plaintiff’s accent, skin

color, or ancestry to determine her racial identity for themselves.

they may dismiss a claim from a plaintiff who was mistreated based on

a misperceived racial identity.

But when a plaintiff brings a claim for

sexual orientation discrimination, courts are not typically concerned

with determining whether the plaintiff is “actually” gay or not; they

don’t care whether the plaintiff self-identifies as gay, is in a same-sex

relationship, or conforms to stereotypes associated with gay people.

15. See infra Part III(C)(4).

16. See infra Part III(C)(2); see also Davis v. Fort Bend Cnty., 765 F.3d 480, 487

(5th Cir. 2014) (example of a court accepting a party’s religious identity based solely

on their testimony).

17. D. Wendy Greene, Categorically Black, White, or Wrong: “Misperception Dis-

crimination” and the State of Title VII Protection, 47

U. M

ICH

. J.L. R

EFORM

87, 90

(2013).

18. See infra Part III(B)(3).

19. See infra Part III(B)(3).

20. See infra Part III(B)(1).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 6 10-MAR-22 14:18

272 TEXAS A&M LAW REVIEW [Vol. 9

Instead, courts focus on the content of the discrimination or percep-

tion of the discriminator and ask whether the discriminatory conduct

was motivated by sexual orientation.

Thus, a plaintiff who self-iden-

tifies as straight but who was misperceived as gay by his discriminator

could probably sustain a claim, but a Hispanic plaintiff misperceived

as Black may have her claim dismissed based on her identity.

There is no doctrinal reason to treat these claims differently; the

same statute and the same analytical framework apply to both race

discrimination claims and sexual orientation discrimination claims.

As explained further in Part IV, the applicable anti-discrimination

laws do not require courts to determine the plaintiff’s “true” iden-

tity—rather, the discriminator’s perception of the plaintiff’s identity

or the cause of the discriminatory behavior should guide the identity

inquiry. Therefore, when courts attempt to determine the “actual”

race of a plaintiff, they are engaging in context-detached identity adju-

dication. Such an approach creates doctrinal inconsistency within anti-

discrimination law because some courts, like those addressing sexual

orientation claims, are not concerned with the “actual” identity of the

plaintiff. Moreover, this approach can harm individual plaintiffs and

undercut the purposes of anti-discrimination law on the whole. Anti-

discrimination laws are meant to provide remedies to plaintiffs who

were mistreated based on a protected category.

When courts deny

these plaintiffs’ claims based in whole or in part on the court’s adjudi-

cation of the plaintiff’s “true” identity, the goals of anti-discrimination

law are not served.

The second problem with a context-detached approach is that it can

result in the over-interrogation of identity by requiring too much iden-

tity evidence to fulfill the function of the law. I fully explore the con-

sequences of over-interrogation in other work,

but I summarize two

primary harms here. First, making identity claims dependent on un-

necessary identity evidence may infringe upon individuals’ privacy and

dignity interests. Consider some examples of highly sensitive identity

evidence legal actors either require or examine when adjudicating

identity: ancestral background from services like 23andMe, the

makeup of someone’s primary or secondary sex characteristics, past

sexual and romantic history, and the hormone treatments someone is

receiving or has received. Using this identity evidence to adjudicate

identity, when this evidence is not necessary considering the law at

issue, needlessly impinges on privacy and dignity interests.

21. See infra Part III(B)(1).

22. Kindall James, Thomas McGoey, II & Courtney Harper Turkington, U.S. Su-

preme Court Rules That Federal Anti-Discrimination Law Protects Gay and Trans-

gender Workers

, JDS

UPRA

(June 16, 2020), https://www.jdsupra.com/legalnews/u-s-

supreme-court-rules-that-federal-94054/ [https://perma.cc/5EEZ-RZGU].

23. See infra Part IV(B)(2).

24. See Laura Lane-Steele (manuscript in progress) (on file with author) [herein-

after Working Manuscript].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 7 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 273

Second, over-interrogation of identity can reinforce problematic

definitions of identity that unnecessarily essentialize identity. For ex-

ample, when a court concludes that someone is a particular race based

on the way they look and speak, that court is using the power of the

law to declare that certain ways of looking and speaking are essential

qualities of a particular race.

Similarly, a legal actor who denies

someone’s claim that she is a lesbian based on her prior marriage to a

man is helping construct an overly rigid and false definition what it

means to be a lesbian.

This Article calls for a context-informed approach to identity adju-

dication. Unlike a context-detached approach, a context-informed ap-

proach understands the identity question to depend on why that

particular law is asking the identity question in the first place. It be-

gins the identity inquiry by locating the function or purpose of the

applicable law. It then asks whether identity adjudication is necessary

at all, and if so, what definition of identity would best serve those

purposes.

Other work has identified how law determines identity claims and

has critiqued specific flaws in how legal actors determine identity, but

the scholarship tends to focus on one area of law, one identity, and/or

one model of identity adjudication.

This Article expands on this

25. See infra Part III(B)(3).

26. See infra notes 169–78 and accompanying text. My future work examines this

problem of over-interrogation in more depth and explores how identity adjudication

does not merely describe or identify pre-existing identity categories—but that it is

helping produce the categories and the cultural meanings associated with those cate-

gories. See Working Manuscript, supra note 24. Other scholars have explored how

legal classifications of race, sex, and sexual orientation construct those identities. See,

e.g.,

ANEY

OPEZ

, W

HITE BY

EGAL

ONSTRUCTION OF

ACE

(1996);

AURA

E. G

OMEZ

, M

ANIFEST

ESTINIES

: T

AKING OF THE

EXICAN

MERICAN

ACE

54, 63

(2007) (analyzing how law defined racial categories in the

nineteenth century); Dean Spade, Documenting Gender, 59

ASTINGS

L.J. 731, 747

(2008) (arguing that legal rules that govern sex classifications “do not simply discover

and describe maleness and femaleness but instead produce two populations marked

with maleness and femaleness”); Stefan Vogler, Legally Queer: The Construction of

Sexuality in LGBQ Asylum Claims, 50

L. & S

’

856, 884 (2016) (arguing that

legal process by which people seek asylum based on sexual orientation “is both regu-

latory and generative of sexual identities and that the law plays a powerful role in

shaping sexual identity”); see also

EOFFREY

C. B

OWKER

& S

USAN

EIGH

TAR

ORTING

HINGS

: C

LASSIFICATION AND

ONSEQUENCES

(1999).

27. See, e.g., Nancy Leong, Judicial Erasure of Mixed-Race Discrimination, 59

U. L. R

469, 471 (2010) [hereinafter Judicial Erasure] (“[A]ntidiscrimination juris-

prudence [is] inhospitable to claims brought by individuals who allege that they were

discriminated against because they were perceived as multiracial.”); D. Wendy

Greene, Categorically Black, White, or Wrong: “Misperception Discrimination” and

the State of Title VII Protection, 47

U. M

ICH

. J.L. R

EFORM

87, 129 (2013) (arguing that

courts’ rejection of misperception claims in Title VII cases “is categorically wrong”);

Jessica A. Clarke, Protected Class Gatekeeping, 92

N.Y.U. L. R

. 101, 104, 132

(2017) [hereinafter Protected Class] (critiquing courts’ refusal “to consider discrimina-

tion claims by plaintiffs who have not proven membership in a protected class” in

anti-discrimination law); Leora F. Eisenstadt, Fluid Identity Discrimination, 52

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 8 10-MAR-22 14:18

274 TEXAS A&M LAW REVIEW [Vol. 9

scholarship by examining legal identity adjudication across areas of

law and identity categories and by extracting patterns that exist

outside of any one doctrinal or identity silo. It also expands on the

scholarship’s critiques of identity adjudication and demonstrates how

context-detached identity adjudication is a systemic problem in law

rather than a problem limited by category of law or identity.

This Article is both broad and narrow. The broad scope allows me

to build the taxonomy of identity adjudication and identify the prob-

lem of context-detached identity adjudication. But the taxonomy also

raises many issues about identity adjudication that I do not address in

this Article.

For instance, the taxonomy sheds light on dominant so-

cietal discourses regarding identity and how those discourses are im-

plemented through law. It can also elucidate how legal identity

adjudication enforces identity-based hierarchies.

This Article lays

the groundwork for, but does not take on, these questions. Also, this

Article does not claim to be an exhaustive examination of all types of

identity adjudication in the law. It does not address, for example, disa-

bility, citizenship, or familial identity categories or the related laws

that adjudicate those identities.

Future work can build on, and im-

. L.J.

789, 791 (2015) (discussing how courts struggle to adjudicate fluid race and

sex identity claims in the context of anti-discrimination law); Jessica A. Clarke, Iden-

tity and Form, 103

ALIF

L. R

747, 750 (2015) [hereinafter Identity and Form]

(discussing formal identity in the content of race, sex, family, and citizenship); Sonia

K. Kayal & Jessica Y. Jung, The Gender Panopticon: AI, Gender, and Design Justice,

68 UCLA

L. R

692, 692 (2021) (documenting how legal and non-legal actors em-

ploy AI technologies to determine sex identity); Russell K. Robinson, Masculinity as

Prison: Sexual Identity, Race, and Incarceration,

99 C

ALIF

. L. R

1309, 1309 (2011)

(explaining how jail officials determine the sexuality and transgender status of

inmates).

28. Other scholars have argued that legal definitions of identity, and whether the

law should use identity in the first place, depend on the context, i.e., the function of

the law and how identity serves that purpose. See, e.g., Jessica A. Clarke, They, Them,

and Theirs, 132

ARV

. L. R

. 894, 902 (2019) (suggesting a “contextual approach to

debates over sex and gender regulation” within the context of non-binary gender and

providing three potential models that regulate sex differently depending on the con-

text); Lauren Sudeall, Identity as Proxy, 115

OLUM

. L. R

1605, 1609–11 (2015)

(arguing that equal protection doctrine should abandon identity as determinative of

the level of scrutiny courts apply in favor of an approach that better serves the pur-

poses and values of equal protection law). This Article begins the work of expanding

these arguments across identities and legal areas, work that I am continuing in my

works in progress.

29. This Article is the first in a series of articles about identity adjudication. In my

future work, I plan to address many questions this Article leaves unanswered.

30. Other work has identified how the law’s creation and maintenance of certain

categories has been crucial in maintaining inequality. See, e.g.,

OPEZ

, supra note 26,

at 139 (showing how legal adjudication of whiteness served to uphold white

supremacy); Dean Spade, Mutilating Gender, in

RANSGENDER

TUDIES

EADER

315

, 320 (Susan Stryker & Stephen Whittle eds., 2006) (arguing that law’s

reliance on the bio-medical model of sex identity privileges cisgender identity as natu-

ral and transgender identity as illness that needs to be cured).

31. Regarding disability, scholars have argued that legal actors over-interrogate

disability claims. That is, legal actors require more identity evidence than necessary in

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 9 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 275

prove upon, this taxonomy by including additional laws and identity

categories. Finally, this Article argues for a context-informed ap-

proach to identity adjudication, where identity inquiries are informed

by the function or purpose of the law at issue; however, it does not

take a normative stance on what the proper function or purpose of a

law should be. These debates are too expansive and complex for this

Article to tackle. This Article makes an argument about the proper

process of identity adjudication—i.e., a context-informed approach.

The remainder of the Article is structured as follows. Part II defines

the types of identity evidence used to prove an identity claim and pro-

vides background on immutable versus mutable identities. Part III de-

tails the taxonomy of identity adjudication that I have introduced

here. Specifically, it examines identity claims in the context of data-

collection laws and explains how self-identification is almost always

sufficient for these laws. Next, it addresses anti-discrimination laws

and provides a descriptive account of how courts adjudicate identity in

the context of these laws. It then addresses benefit laws and highlights

how identity claims are typically the most highly scrutinized, particu-

larly with race and sex. Lastly, it offers possible explanations for why

the type of law and the mutability of identity affect how closely an

identity is examined. Part IV turns to the normative critique of the

taxonomy and argues that identity is often adjudicated without proper

attention to the purpose or function of the particular law at issue. The

Article ends with a brief conclusion.

II. D

EFINITIONS AND

UTABILITY

A. Types of Identity Evidence

To make the following discussions of identity adjudication more co-

herent and efficient, this Article uses the following terms to refer to

certain categories of identity evidence. By “identity evidence,” I mean

the documents, traits, or information legal actors use to determine

someone’s identity.

1. Ascriptive evidence. Ascriptive evidence refers to both physical

appearance and the performative aspects of identity. Based on

one or both of these, individuals are involuntarily assigned or

ascribed an identity by third parties. For instance, a third party

may assign someone a racial identity when she “interprets an-

other person’s visible, physical features to correlate with a set of

features she identifies with a certain race or ethnic group.”

light of the purpose of the ADA. See, e.g., Katherine Macfarlane, Disability Without

Documentation, 90

ORDHAM

L. R

59 (2021) (arguing that requiring employees to

provide medical documentation of their disability before receiving an accommodation

conflicts with the ADA’s purpose).

32. Camille Gear Rich, Performing Racial and Ethnic Identity: Discrimination by

Proxy and the Future of Title VII, 79 N.Y.U.

L. R

. 1134, 1145 (2004) [hereinafter

Performing Racial and Ethnic Identity].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 10 10-MAR-22 14:18

276 TEXAS A&M LAW REVIEW [Vol. 9

Her racial assignment also may take into account certain “dia-

lects, aesthetics, and mannerisms” that she associates with a cer-

tain race.

Sex, sexual orientation, and religion

may also be

ascribed based on some combination of physical appearance

and behavior.

2. Documentary evidence. Documentary evidence refers to prior

articulations of someone’s identity, typically, but not exclu-

sively, in written form. This category includes an individual’s

prior self-identification on a form, like an employment applica-

tion or insurance form, and recordings by third parties of an

individual’s ascriptive identity, like a birth certificate designat-

ing someone’s race.

3. Ancestral evidence. Ancestral evidence focuses on evidence of

familial lineage and includes things like DNA evidence, ances-

try, and genealogies.

4. Biological evidence. Biological evidence refers to traits like

chromosomes, primary and secondary sex characteristics, and

other things tied to the body that are not encompassed under

ascriptive evidence.

33. Id. at 1158. This brief discussion does not capture the complexity of identity

performance or ascriptive identity, which has been widely discussed in the scholar-

ship.

EVON

ARBADO

& M

ITU

ULATI

, A

CTING

HITE

?: R

ETHINKING

ACE IN

“P

OST

-R

ACIAL

” A

MERICA

(1st ed. 2013) (discussing how racial judgments are often

based on the extent to which a person conforms their behavior to that which is stere-

otypically associated with a certain race); John O. Calmore, Random Notes of an Inte-

gration Warrior, 81

INN

. L. R

1441, 1450 (1997) (“Even dark-skinned, nappy-

headed African Americans like me can pass sociologically and culturally if we have

the right history of socialization, the right credentials, a respectable job, an affluent

income, and a proper street address or zip code.”); Devon W. Carbado & Mitu Gulati,

Working Identity, 85

ORNELL

L. R

. 1259, 1289 (2000);

ICHARD

HOMPSON

ORD

, R

ACIAL

ULTURE

: A C

RITIQUE

(2005) (explaining how a third party often

performs the subject’s identity back to the subject; for instance when a white depart-

ment store clerk follows a Black man around the store, she is performing his identity

back to him).

34. See Kenji Yoshino, Covering, 111

ALE

L.J.

769, 779–80, 929 (2002).

35. Some scholars have included individuals’ biological and ancestral makeups in

their definitions of ascriptive identity. See Identity and Form, supra note 27, at 757.

For this Article, I have found it more useful to create separate categories for those

aspects of identity. Biology and ancestry often affect physical appearance, but the two

do not always correlate. For instance, an ascriptive model of identity (as I have de-

fined it) may result in a transgender person’s sex being adjudicated as female, while

under a biological one, it may not be. And multi-racial individuals are often not per-

ceived to be of the racial category that matches their ancestral makeup. See infra Part

(III)(B).

36. Jessica Clarke has discussed a closely related model of identity, called formal

identity, which she defines as an identity that “comes into being through the execu-

tion of a formality by the parties laying claim to a particular identity.” Identity and

Form, supra note 27, at 770. Her article focuses on “formalities that constitute identity

statuses for legal purposes, rather than those thought to merely reflect a status.” Id.

This Article understands both as documentary evidence—for example, an individual’s

sex marker on their birth certificate can reflect their sex, and when an individual

changes that sex marker under state law, they become that sex under that state’s law.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 11 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 277

5. Self-identification. Self-identification refers to the identity to

which the individual has laid claim to at the time of the litiga-

tion, controversy, point of data-collection, or other event in

question. This category does not include a person’s prior repre-

sentations of their identity, unless otherwise noted.

These forms of identity evidence are not mutually exclusive, and in

some cases, they may completely overlap. For example, a family gene-

alogy can be considered both ancestral evidence and documentary evi-

dence. A birth certificate can be considered either documentary

evidence or ascriptive evidence because it represents the sex a third

party assigned to the individual at birth. Moreover, not all types of

evidence apply to all categories of identity. For example, biological

and ancestral evidence are not relevant to the adjudication of sexual

orientation and most types of religious identity claims. Finally, I am

not committed to these definitions of identity evidence; these defini-

tions are not the point of this Article. Nor am I arguing that any of

these types of identity evidence are normatively constitutive of any

particular identity. In other words, having XY chromosomes does not

make someone “male” as a normative matter. Indeed, claiming that

certain types of identity evidence are definitional qualities of an iden-

tity would be antithetical to a project that advocates for a context-

informed approach to defining identity. Rather, I find that these terms

promote efficiency and consistency when discussing identity adjudica-

tion in such a broad manner.

B. Mutability

Because the taxonomy introduced in this Article finds that “immu-

table” identities (race and sex) tend to receive closer scrutiny than

more “mutable” identities (sexual orientation and religion), a brief ex-

planation of what I mean by immutability is in order. Mutability of

identity in this Article means the degree to which an individual can

chose, or validly change, their identity in the eyes of the law.

There

are different tiers of immutability, and identities are not simply immu-

table or mutable. Identities that are perceived to be rooted in nature

or biology are generally deemed strictly immutable, i.e., the identity

cannot be changed.

Race and ethnicity fall into this category of im-

mutability.

For many decades, sex was also strictly immutable;

37. This Article focuses on mutability of identity in law as opposed to societal

perceptions of immutability. Societal and legal understandings of immutability no

doubt inform each other, but societal definitions are too context-dependent, unstable,

and contested to conduct this cross-identity, cross-law analysis.

38. See Jessica A. Clarke, Against Immutability, 125

ALE

L.J.

2, 14–15 (2015).

However, immutable identities or traits do not have to be connected to biology or

ancestry. For example, the law considers illegitimacy to be an immutable trait. Weber

v. Aetna Cas. & Sur. Co., 406 U.S. 164, 175–76 (1972).

39. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 360 (1978) (Brennan, J.,

concurring in part and dissenting in part) (“[R]ace, like gender and illegitimacy . . . is

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 12 10-MAR-22 14:18

278 TEXAS A&M LAW REVIEW [Vol. 9

however, this has started to change with the gradual judicial and legis-

lative recognition that an individual can change their sex,

usually by

changing their primary or secondary sex characteristics.

The next tier of immutability, which some call the “new immutabil-

ity,” includes identities or traits that the law understands as capable of

being changed but that law should not require or incentivize people to

change.

These traits “are so central to a person’s identity that it

would be abhorrent for government to penalize a person for refusing

to change them, regardless of how easy that change might be physi-

cally.”

Courts have placed sexual orientation in this category of im-

mutability.

Religion also falls in this category. People can, and often

do, change their religion, but the law does not require or incentivize

changing one’s religion.

This discussion of the varying degrees of mutability of identity is not

an endorsement of their legitimacy. Many scholars have convincingly

critiqued the stability of the immutable/mutable distinction and its rel-

evance to equality law.

Rather, this immutable/mutable distinction

in the law is relevant because of its role in explaining the taxonomy

for identity adjudication.

III. A T

AXONOMY OF

DENTITY

DJUDICATION

This Part develops the taxonomy of identity adjudication identified

in the Introduction and reveals how both the type of law and the iden-

an immutable characteristic which its possessors are powerless to escape or set aside.”

(citation omitted)).

40. Frontiero v. Richardson, 411 U.S. 677, 686 (1973) (plurality opinion).

41. See Lisa Mottet, Modernizing State Vital Statistics Statutes and Policies to En-

sure Accurate Gender Markers on Birth Certificates: A Good Government Approach

to Recognizing the Lives of Transgender People, 19

ICH

. J. G

ENDER

& L.

373, 376

(2013).

42. The mutability of sex is limited, though, in the sense that it is generally only

able to be changed once in order to align the body with the mind or soul. In the eyes

of law, the “actual” sex of the individual resides in the mind or soul and is still under-

stood to be immutable. See Yoshino, supra note 34, at 922–23 (2002); Paisley Currah,

Gender Pluralisms Under the Transgender Umbrella, in

RANSGENDER

IGHTS

3, 4

(Paisley Currah, Richard M. Juang & Shannon Price Minter eds., 2006).

43. Clarke, supra note 38, at 4–5.

44. Watkins v. U.S. Army, 875 F.2d 699, 726 (9th Cir. 1989) (Norris, J.,

concurring).

45. See Kerrigan v. Comm’r of Pub. Health, 957 A.2d 407, 436–39 (Conn. 2008)

(holding that sexual orientation is immutable because of its centrality to a person’s

identity and can be “altered[,] [if at all,] only at the expense of significant damage to

the individual’s sense of self” (alteration in original) (citation omitted)).

46. Indeed, the First Amendment prevents the government from punishing or in-

centivizing changing religions. See, e.g., Larson v. Valente, 456 U.S. 228, 244 (1982).

47. See, e.g., Kenji Yoshino, Assimilationist Bias in Equal Protection: The Visibility

Presumption and the Case of “Don’t Ask, Don’t Tell”, 108

ALE

L.J

. 485, 487 (1998);

Clarke, supra note 38, at 10; Janet E. Halley, Sexual Orientation and the Politics of

Biology: A Critique of the Argument from Immutability, 46

TAN

. L. R

. 503, 519–21

(1994).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 13 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 279

tity category affect how an identity claim is examined. First, it ad-

dresses data-collection laws and demonstrates how, currently, they

almost exclusively rely on self-identification. For race, one of the more

immutable identities, there are some exceptions. Specifically, a third

party’s ascriptive assessment of someone’s race can trump self-identi-

fication, but it is not clear how often this happens in practice, or if it

happens at all. Then, this Part turns to identity adjudication in anti-

discrimination cases. It illustrates how courts are more likely to focus

on a plaintiff’s “actual” identity in cases involving race and sex. Fi-

nally, it turns to benefit laws and shows how legal scrutiny of identity

claims is highest in this category of laws and also how the proof re-

quirements for immutable identities (race and sex) are particularly on-

erous. The following is a visual representation of how identity

category and type of law affect identity interrogation.

48. As currently constituted, this taxonomy does not completely account for how

other intersecting identities affect how an identity claim is adjudicated—i.e., how

scrutiny of one identity (identity A) is affected by occupation of another privileged or

marginalized identity (identity B). For instance, the current version of this taxonym

does not fully address how race might affect sex identity or religious identity determi-

nations or how sex and race might affect sexuality determinations. See Kimberle

Crenshaw, Demarginalizing the Intersection of Race and Sex: A Black Feminist Cri-

tique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Policies, 1

. L

EGAL

139, 139–40 (1989) (arguing that different identities intersect to pro-

duce different experiences and forms of discrimination). Other scholars have dis-

cussed how intersecting identities affect identity determinations in discrete legal

contexts. E.g.,

EATH

OGG

AVIS

, B

EYOND

RANS

OES

ENDER

ATTER

112–13 (2017) (discussing how women of color are more likely to have their sex iden-

tity questioned than white women in the context of sex-segregated sports); Adeel

Mohammadi, Note, Sincerity, Religious Questions, and the Accommodation Claims of

Muslim Prisoners, 129

ALE

L. J.

1836, 1841–42 (showing how courts tend to require

more than just “religious sincerity” when Muslims seek religious accommodations in

prison, even though the First Amendment technically bars more searching in-

quires); see also Ariela J. Gross, Litigating Whiteness: Trials of Racial Determination

in the Nineteenth-Century South, 108

ALE

L.J.

109, 157, 163–66 (1998) (showing how

performative evidence of white identity took different forms depending on sex—for

men, for example, acting white meant exercising rights like voting). I am still studying

how intersecting identities affect identity determinations in a systemic way. In my

current work in progress specifically, I am exploring how occupation of one or more

marginalized identities can exacerbate the harm caused by the context-detached ap-

proach to identity adjudication.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 14 10-MAR-22 14:18

280 TEXAS A&M LAW REVIEW [Vol. 9

Race

Sex

Sexual Orientation

Religion

Data Collection BenefitAnti-discrimination

Less Scrutiny

More Scrutiny

A. Data-Collection Laws

“Data-collection” laws are just that: mechanisms that govern or

conduct the collection of data. The primary goal of data-collection

laws is to collect data, but they all have secondary functions, depend-

ing on how the identity information is used. Census data, for example,

is used for many purposes, including determining how a wide variety

of federal resources are allocated and tracking patterns of race and

sex discrimination.

The Equal Employment Opportunity Commis-

sion (“EEOC”) also collects data on race and sex to identify work-

place discrimination.

But no matter the purpose for which the

identity data is used, identity is determined almost exclusively by self-

identification. For instance, the Census gathers information on racial,

ethnic, and sex identity based on self-identification alone.

Sexual ori-

49.

U.S. D

’

., Q

UESTIONS

LANNED FOR THE

2020 C

ENSUS AND

MERI-

CAN

OMMUNITY

URVEY

U.S. C

ENSUS

UREAU

11, 41, 71 (2018), https://

www2.census.gov/library/publications/decennial/2020/operations/planned-questions-

2020-acs.pdf [https://perma.cc/EN2Z-SMYM].

50.

U.S. E

QUAL

. O

PPORTUNITY

OMM

’

2019

AND

2020 EEO-1 C

OMPO-

NENT

1 D

ATA

OLLECTION

NSTRUCTION

OOKLET

, EEOC

DATA

ORG

10–12 (2019),

https://eeocdata.org/pdfs/EEO-1%20Compo-

nent%201%20Instruction%20Booklet.pdf [https://perma.cc/V4LC-WUBV].

51. Id. at 11–15. The Census treats Native American status as a racial category and

allows participants to write the name of their enrolled or principal tribe. Id. at 11. The

Census Bureau does not cross-reference these responses against tribes’ official enroll-

ment lists. See id.;

U.S. D

’

TOF

., T

RIBAL

ONSULTATION

ANDBOOK

U.S.

ENSUS

UREAU

24 (2015), https://www.census.gov/content/dam/Census/library/pub-

lications/2015/dec/2020_tribal_consultation_handbook.pdf [https://perma.cc/9CSK-

VGGC].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 15 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 281

entation data is also captured via self-identification in both the Cen-

sus

and the Centers for Disease Control and Prevention (“CDC”)’s

telephone survey.

Religious identity is also determined through self-identification.

The Census does not currently ask about religious affiliation, but

when it used to, respondents checked the box corresponding to their

religious identities.

Other countries’ censuses collect data on relig-

ious practices based on self-identification.

In the United States,

think-tanks and other non-governmental research organizations col-

lect this information, and self-identification is always sufficient.

This is not to say that data-collection mechanisms permit unre-

strained freedom of identity self-determination. Many of these mecha-

nisms limit the options within each identity category. The Census, for

example, has only two options for sex: male or female.

People who

52. The 2020 Census did not ask respondents to check a box corresponding to

their sexual orientation; it asked respondents to define their relationship with their

cohabiting partner as an opposite sex husband/wife/spouse, opposite sex unmarried

partner, same-sex husband/wife/spouse, or same-sex unmarried partner

. U.S. D

’

., T

EST

UESTIONNAIRE FOR

2020 C

ENSUS

, U.S. C

ENSUS

UREAU

1, 3 (2018),

https://www.documentcloud.org/documents/4425959x-Test-Questionnaire-for-2020-

Census.html#document/p3/a414347 [https://perma.cc/FG8S-FMMR].

53. Behavioral Risk Factor Surveillance Systems (BRFSS): Years Survey Included

Sexual and Gender Minority (SGM)-Related Questions 2014 - Present,

TRS

FOR

EDICARE

& M

EDICAID

ERVS

, https://www.cms.gov/About-CMS/Agency-Informa-

tion/OMH/resource-center/hcps-and-researchers/data-tools/sgm-clearinghouse/brfss

(Dec. 1, 2021, 7:02 PM) [https://perma.cc/RXD9-9HPV].

54. Congress passed a statute in 1976 that prohibited the Census Bureau from

requiring disclosures of religious affiliation. 13 U.S.C. § 221(c). Although the Bureau

could ask people an optional question about their religious affiliations, it has chosen

not to due to separation of church and state concerns. Anne Farris Rosen, A Brief

History of Religion and the U.S. Census,

SCH

. C

(Jan. 26, 2010), https://

www.pewforum.org/2010/01/26/a-brief-history-of-religion-and-the-u-s-census/ [https://

perma.cc/3NEA-ED22] (quoting a former Census Bureau director who stated that

“[t]he decision not to add this question is based essentially on the fact that asking such

a question in the decennial census, in which replies are mandatory, would appear to

infringe upon the traditional separation of church and [s]tate”).

55. See, e.g.,

VANS

& M

IKE

ELSBY

FOR

’

TAT

., R

ELIGION

DUCATION AND

ORK IN

NGLAND AND

ALES

(2020), https://www.ons.gov.uk/

peoplepopulationandcommunity/culturalidentity/religion/articles/religioneducation-

andworkinenglandandwales/february2020 [https://perma.cc/VQX2-M7EP] (discussing

the United Kingdom’s census).

56. For instance, the Pew Research Center’s Religious Landscape Study asked

participants: “What is your present religion, if any? Are you Protestant, Roman Cath-

olic, Mormon, Orthodox such as Greek or Russian Orthodox, Jewish, Muslim, Bud-

dhist, Hindu, atheist, agnostic, something else, or nothing in particular?”

SCH

2014 R

ELIGIOUS

ANDSCAPE

TUDY

: M

AIN

URVEY OF

ATIONALLY

EPRE-

SENTATIVE

AMPLE OF

DULTS

INAL

UESTIONNAIRE

7 (2014), https://as-

sets.pewresearch.org/wp-content/uploads/sites/11/2016/10/25142557/RLS-II-

Questionnaire-for-5th-release.pdf [https://perma.cc/9DD9-EE7K].

57.

U.S. D

’

TOF

., 2010 C

ENSUS

UESTIONNAIRE

, U.S. C

ENSUS

UREAU

(2010), https://www.census.gov/content/dam/Census/programs-surveys/decennial/tech-

nical-documentation/questionnaires/2010questionnaire.pdf [https://perma.cc/ZC9T-

4XBN].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 16 10-MAR-22 14:18

282 TEXAS A&M LAW REVIEW [Vol. 9

do not fit neatly into either category cannot accurately claim their

identity.

Additionally, until 2000, the Census instructions limited re-

spondents to one racial category, so individuals who identified as mul-

tiracial could not select more than one race.

Thus, data-collection

laws do not allow people to identify however they want, but within the

options these laws create, identity claims are generally not questioned.

However, data-collection laws have not always relied predomi-

nately on self-identification, especially for racial identity. When the

Census Bureau began collecting information on racial identity, Census

administrators would designate a respondent’s race based on their ap-

pearance.

In other words, ascriptive evidence, not self-identification,

determined racial identity. Similarly, before 2006, the EEOC in-

structed employers to assign their employees a racial category based

on employees’ visual appearance.

Even today, there are some excep-

tions to the general rule that racial self-identification is sufficient for

data-collection laws. For instance, although self-identification is the

EEOC’s preferred method of collecting identity data, the agency per-

mits “the person attempting to secure information regarding race, sex,

or ethnic affiliation” to reclassify an employee’s identity “where the

declaration by the applicant or employee is patently false.”

How-

58. See Sonia K. Katyal, The Numerus Clausus of Sex,

84 U. C

. L. R

389,

400–01 (2017) (describing how law limits the options for sex designations to only two

categories (male and female) and comparing this to the numerus clauses principle in

property that limits the categories of property).

59. Alaina R. Walker, Choosing To Be Multiracial in America: The Sociopolitical

Implications of the “Check All That Apply” Approach to Race Adopted in the 2000

U.S. Census, 21

ERKELEY

AZA

L.J.

61, 64–66 (2011).

60. Id. at 64.

61. See Joseph Z. Fleming, I Believe There Is Something Out There Watching Us;

Unfortunately, It’s the Government: An Analysis of the EEOC’s “EEO-1” and

OFCCP Reporting Requirements,

ALI-ABA C

OURSE OF

TUDY

: A

DVANCED

M-

PLOYMENT

AW AND

ITIGATION

, Nov. 30–Dec. 2, 2006, at *1225, SM027 ALI-ABA

1209. Camille Gear Rich has discussed this relatively recent shift in the law’s prefer-

ence for self-identification over ascriptive evidence for racial identity determinations.

See Camille Gear Rich, Elective Race: Recognizing Race Discrimination in the Era of

Racial Self-Identification, 102

. L.J.

1501, 1505 (2014) [hereinafter Elective Race].

62. U.S.

QUAL

. O

PPORTUNITY

OMM

’

OMPLIANCE

ANUAL

§ 632.3(b)(2)(iii) (2006), 2006 WL 4672862. The EEOC does not explain how an em-

ployer should go about deciding when an employee’s declaration is “patently false,”

but considering that the EEOC’s pre-2006 policy used ascriptive race, it is reasonable

to assume that employers can reclassify employees when employers’ perception of

their employees’ identity does not match the employees’ self-identification. EEOC

guidance also instructs employers to use “observer identification” “if an employee

declines to self-identify his or her race and/or ethnicity.”

U.S. E

QUAL

. O

PPORTU-

NITY

OMM

’

, 2019

AND

2020 EEO-1 C

OMPONENT

1 D

ATA

OLLECTION

NSTRUC-

TION

OOKLET

, EEOC

DATA

ORG

11 (2019–2020), https://eeocdata.org/pdfs/EEO-

1%20Component%201%20Instruction%20Booklet.pdf [https://perma.cc/V4LC-

WUBV]. In this latter example, it is not that self-identification is insufficient to prove

an identity claim, but rather that the person failed to self-identify. See also Final Gui-

dance on Maintaining, Collecting, and Reporting Racial and Ethnic Data to the U.S.

Department of Education, 72 Fed. Reg. 59266, 59268 (Oct. 19, 2007) (calling for “ob-

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 17 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 283

ever, there does not appear to be any information on how often this

identity reassignment occurs in practice or if it occurs at all.

Both the modern-day and historical exceptions to the general rule

that self-identification is sufficient for data-collection laws involve im-

mutable identities, and race in particular. This is in line with the gen-

eral pattern that immutable identities tend to be more closely

interrogated than mutable ones. But for the most part, self-identifica-

tion is currently sufficient to establish an identity claim for a data-

collection law for both mutable and immutable identities. This is true

regardless of why the data is being used—whether it’s used to track

discrimination, allocate resources, or for some other purpose.

However, self-identification may not always be the best type of

identity evidence, depending on why the data is being collected. For

instance, if the identity data is being collected to track patterns of dis-

crimination, asking someone how they racially self-identify may not

actually track patterns of racial discrimination. This is because there

are many instances where someone’s racial self-identification does not

align with the race are assigned by someone who discriminates against

them. Said differently, if a multiracial or racially ambiguous individual

self-identifies as “white” on a data-collection form but is perceived as

Black and discriminated against on that basis, the data will not reflect

this instance of discrimination since the racial identity data does not

track the basis upon which they were discriminated. The same logic

applies for other identities, not just race. A gay person who is not out

to their employer or community may self-identify as gay on the data-

collection form. But if no one knows they are gay, they are unlikely to

be discriminated against on that basis. Thus, depending on why the

data is being collected, identity adjudication may need to take other

forms of identity evidence into account, beyond just self-

identification.

server identification” of the race and ethnicity of elementary and secondary school

students when neither the student nor the parent has elected a racial/ethnic identity).

63. Some legal scholars have made a similar point about data collection, self-iden-

tification, and measuring discrimination, noting that racial discrimination and hierar-

chies are more often produced by the racialized individual’s involuntary ascription to

a racial group rather than that person’s self-identification. See, e,g.,

ORD

, supra note

33, at 8 (describing “physical appearance as the primary marker of racial difference”);

Naomi Mezey, Erasure and Recognition: The Census, Race and the National Imagina-

tion, 97

. U.L. R

1701, 1753 (2003) (arguing that in the context of Census data-

collection efforts, racial self-identification makes it more difficult to accurately mea-

sure racial discrimination); Lauren Sudeall, Undoing Race? Reconciling Multiracial

Identity with Equal Protection, 102

ALIF

. L. R

1243, 1291–92 (2014) (arguing that

internal understandings of identity are not effective when the purpose of a data-col-

lection law to is address racial hierarchies, which are based on ascribed racial

identity).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 18 10-MAR-22 14:18

284 TEXAS A&M LAW REVIEW [Vol. 9

B. Anti-Discrimination Laws

On the whole, identity is adjudicated more intensely in anti-discrim-

ination law as compared to data-collection laws. Some courts focus on

discovering a plaintiff’s “actual” identity and require more than just

self-identification to prove an identity claim. This search for a plain-

tiff’s “true” identity tends to manifest in two ways. First, some judges

attempt to adjudicate the plaintiff’s identity for themselves based on

whatever the judge deems “objective” evidence of a particular iden-

tity. For instance, a judge may determine a plaintiff’s racial identity

based on her skin color, accent, and facial features. Second, judges

sometimes impose on plaintiffs what legal scholar D. Wendy Greene

calls an “actuality requirement.”

The actuality requirement man-

dates alignment between the plaintiff’s self-identification and the

identity their discriminator perceived them to be.

For instance, if a

Black man’s employer misperceived him as Hispanic and subjected

him to harassment based on that misperception, he would not be able

to bring a claim if a court imposed an actuality requirement. These are

also often called “misperception” cases.

On the other hand, some courts do not interrogate the plaintiff’s

“actual” identity at all. The relevant inquiry for these courts is not the

plaintiff’s identity per se; rather, they ask whether the alleged discrim-

ination was based on race, sex, religion, or sexual orientation. For ex-

ample, courts in this category do not care whether a plaintiff who

alleges sexual orientation discrimination is gay or not; what matters is

whether they were mistreated because of sexual orientation, whether

actual, perceived, or otherwise.

Whether a court is likely to adjudicate a plaintiff’s identity, or look

to the employer’s perception of the plaintiff’s identity, appears to be

highly dependent on the identity involved. Courts addressing claims

that involve more immutable identities (race and sex) are more likely

to adjudicate the “actual” identity of the plaintiff, and cases involving

more mutable identities (sexual orientation and religion) generally do

not.

1. Sexual Orientation

Courts addressing sexual orientation discrimination claims gener-

ally do not closely interrogate the plaintiff’s identity, if they do at all.

The relevant question in these cases is not what the plaintiff’s sexual

orientation actually is, but instead, whether the discrimination was

based on sexual orientation. This means that straight plaintiffs who

are misperceived as gay can typically bring claims even though their

64. Greene, supra note 17, at 90–91.

65. Id.

66. This Subsection focuses primarily on the adjudication of identity in disparate

treatment cases versus disparate impact cases.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 19 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 285

sexual orientation does not align with the sexuality upon which they

are basing their discrimination claim. Indeed, many states’ anti-dis-

crimination statutes protect actual or perceived sexual orientation,

meaning that plaintiffs whose sexuality is misperceived can bring

claims. For example, in a suit brought under a statute that protects

both actual and perceived sexual orientation, the plaintiff—who iden-

tified as heterosexual and was married to a woman—alleged that he

was subjected to anti-gay bullying because his coworkers thought he

was gay.

The plaintiff’s heterosexuality was irrelevant to the ultimate

adjudication of his claim, and after a trial, a jury awarded him millions

in damages.

Statutes like these, however, do not apply to all instances of anti-

gay discrimination. When straight plaintiffs bring claims based on

homophobic bullying, they lose when their harassers know that they

are straight. For example, one court dismissed a plaintiff’s harassment

claim because his coworkers testified that they did not think he was

gay, even though they teased him for liking “gay boy music” and re-

ferred to him as a “dick smoker.”

Thus, when state law protects both

actual and perceived sexual orientation, the plaintiff’s identity is not

relevant in the sense that a plaintiff of any sexual orientation can

bring a discrimination claim based on anti-gay bias. The plaintiff’s

identity only becomes relevant when it informs the discriminator’s

perception or knowledge of the plaintiff’s sexual orientation. When

the discriminator becomes aware of the plaintiff’s sexual orientation,

the plaintiff can only bring a claim based on that specific sexuality.

Federal law mirrors state law on this question. Sexual orientation

was not protected under Title VII in some jurisdictions until the Su-

preme Court’s Bostock v. Clayton County decision,

but the jurisdic-

67. See, e.g.,

. G

’

ODE

§ 12926(o) (West 2021) (“[S]exual orientation . . .

includes a perception that the person has . . . [that] characteristic[ ] . . . .”).

68. Pearl v. City of Los Angeles, 248 Cal. Rptr. 3d 508, 511–12 (Ct. App. 2019).

69. See id. at 510; see also Matt Hamilton, L.A. Sanitation Worker Taunted Over

Perceived Homosexuality Wins $17.4-Million Verdict

L.A.

IMES

(June 15, 2017, 9:30

PM), https://www.latimes.com/local/lanow/la-me-ln-city-discrimination-verdict-

20170615-story.html [https://perma.cc/P967-KZJK] (noting that a heterosexual L.A.

sanitation worker was awarded $17.4 million after he endured repeated harassment

by his supervisors who perceived him as gay); 1212 Rest. Grp., LLC v. Alexander, 959

N.E.2d 155, 158 (Ill. App. Ct. 2011) (upholding the trial court’s finding that the plain-

tiff’s harassers perceived the plaintiff to be gay based on their anti-gay insults and

their statements to coworkers that they thought the plaintiff was gay).

70. Glinski v. Radioshack, No. 03-CV-93OS, 2006 WL 2827870, at *12 (W.D.N.Y.

Sept. 29, 2006). In a similar case, the plaintiff’s coworkers “called him names such as

‘gay,’ ‘homosexual,’ ‘sissy,’ and ‘woman’” and subjected him to “unwanted touching,

sexual jokes and comments.” The court held that the plaintiff was not protected under

California’s anti-discrimination statute because his “coworkers knew he had been

married and had a child” and testified “that they did not perceive him to be gay.”

Akoidu v. Greyhound Lines, Inc., No. B147046, 2002 WL 399476, at *5 (Cal. Ct. App.

Mar. 15, 2002).

71. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1737–38 (2020).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 20 10-MAR-22 14:18

286 TEXAS A&M LAW REVIEW [Vol. 9

tions that did include sexual orientation under Title VII’s protections

pre-Bostock allowed misperception claims. In a Southern District of

New York case, for instance, the defendant argued that the plaintiff’s

claims failed because he didn’t allege that he was “not heterosexual,”

but the court rejected this argument and reasoned that heterosexual

plaintiffs can bring claims based on their misperceived sexual orienta-

tion.

Other courts have followed suit.

Whether this same principle

will be applied in other jurisdictions post-Bostock is not certain at this

point, but as of right now, there is little reason to think other courts

would apply a different principle in adjudicating sexual orientation.

2. Religion

Like sexual orientation, religious identity claims are also not closely

scrutinized in discrimination cases. Courts generally take plaintiffs at

their word that they are members of particular religions or that their

religious beliefs require them to do something. Moreover, plaintiffs

can usually sustain religious misperception claims, and one recent Su-

preme Court case suggests any inquiry into individuals’ actual relig-

ious beliefs is not necessary.

However, some courts do reject

misperception cases, though these appear to be in the minority.

Cases in which a plaintiff’s self-identification is sufficient to prove

their religious identity abound. In one paradigmatic case out of the

Fifth Circuit, Davis v. Fort Bend County, the plaintiff was fired for

attending a religious service rather than reporting to her job, so she

brought suit alleging religious discrimination.

The court held that the

plaintiff sincerely believed her attendance was compelled by her

faith.

To support its holding, the court relied exclusively on the

plaintiff’s own testimony and did not require any additional evi-

dence—such as testimony from church leaders or other congregation

members—to prove her religious identity.

72. Sanderson v. Leg Apparel LLC, No. 1:19-cv-08423, 2020 WL 3100256, at *9

(S.D.N.Y. June 11, 2020).

73. See Johnson v. City of New York, No. 18-CV-9600, 2020 WL 2036708, at *4

(S.D.N.Y. Apr. 28, 2020) (holding that allegations of mistreatment based on perceived

sexual orientation sufficient to state a claim under Title VII); see also Jamiel v. Maison

[email protected], No. 1:19-cv-01389, 2019 WL 9362541, at *4 n.7 (S.D.N.Y. Dec. 19,

2019) (same).

74. EEOC v. Abercrombie & Fitch Stores, Inc., 575 U.S. 768, 773 (2015).

75. Davis v. Fort Bend Cnty., 765 F.3d 480, 483–84 (5th Cir. 2014).

76. Id. at 487.

77. Id. Under similar circumstances, the Seventh Circuit concluded that the plain-

tiff’s attendance at his father’s funeral was required under his sincerely held religious

beliefs based only on the plaintiff’s testimony. See Adeyeye v. Heartland Sweeteners,

LLC, 721 F.3d 444, 452 (7th Cir. 2013) (looking to plaintiff’s deposition and declara-

tions). Other courts have followed suit. E.g., Fox v. Hoenigsberg & Duevel Corp., No.

1:18-CV-265, 2019 WL 10960442, at *4 (E.D. Tenn. Aug. 8, 2019) (relying solely on

the plaintiff’s testimony to determine that her religious belief was sincere); EEOC v.

Consol. Energy, Inc., 860 F.3d 131, 142–143 (4th Cir. 2017) (same); Baker v. Home

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 21 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 287

Davis and similar cases only require self-identification, but a 2015

Supreme Court case suggests that any inquiry into the plaintiff’s relig-

ious beliefs, including self-identification, may not be necessary. In this

case, EEOC v. Abercrombie & Fitch, Abercrombie & Fitch refused to

hire Samantha Elauf because the headscarf she wore violated Aber-

crombie’s dress code.

Although Elauf had not disclosed her religion

to Abercrombie, Abercrombie believed that she wore the headscarf in

connection with her Muslim faith.

When Elauf sued, Abercrombie

argued that she could not show disparate treatment because she had

never disclosed her religion to Abercrombie or explained that she

wore the headscarf for religious reasons.

The Court rejected this ar-

gument and held that the employer need not have actual knowledge

of someone’s religion to violate Title VII.

Rather, an “employer who

acts with the motive of avoiding accommodation may violate Title VII

even if he has no more than an unsubstantiated suspicion that accom-

modation would be needed.”

Elauf did identify as Muslim, but the

Court’s reasoning applies regardless of her actual religious identity.

The employer used ascriptive evidence (her headscarf) to assign a par-

ticular religious identity to Elauf and then discriminated against her

based on that ascription.

Elauf’s “actual” religious identity was irrel-

evant to the Court.

Like the sexual orientation cases, it was the em-

ployer’s perception that mattered.

Relatedly, courts are also generally willing to accept misperception

claims, focusing not on whether the plaintiff is in fact a member of a

particular religion but whether she was treated worse because of relig-

ion. In Smith v. Specialty Pool Contractors, for instance, the court de-

nied the defendant’s motion for summary judgment when a Catholic

plaintiff claimed that his employer thought he was Jewish and discrim-

inated against him on that basis.

Depot, 445 F.3d 541, 547 (2d Cir. 2006) (same); Jiglov v. Hotel Peabody, G.P., 719 F.

Supp. 2d 918, 927 (W.D. Tenn. 2010) (same).

78. Abercrombie, 575 U.S. at 770.

79. Id.

80. Id. at 772.

81. Id.

82. Id. at 773 (emphasis added).

83. Id.

84. Id.

85. See id.; see also Cole v. Cobb Cnty. Sch. Dist., No. 1:17-cv-01378, 2017 WL

9477374, at *7 (N.D. Ga. Dec. 6, 2017) (applying Abercrombie to religious mispercep-

tion claim and holding that Title VII applies when an “employer’s adverse decision is

based on ‘unsubstantiated suspicion,’” even when that unsubstantiated suspicion is

incorrect (citation omitted)).

86. Smith v. Specialty Pool Contractors, No. 02:07cv1464, 2008 WL 4410163, at

*5–6 (W.D. Pa. Sept. 24, 2008). In a related context, the Third Circuit approved of

religious misperception claims, stating that an “employer is still discriminating on the

basis of religion even if the applicant he refuses to hire is not in fact a Muslim.”

Fogleman v. Mercy Hosp., Inc., 283 F.3d 561, 571 (3d Cir. 2002). What matters, the

court explained, “is that the applicant, whether Muslim or not, was treated worse than

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 22 10-MAR-22 14:18

288 TEXAS A&M LAW REVIEW [Vol. 9

However, some courts do not follow this pattern and have rejected

misperception religious discrimination claims. For example, one fed-

eral district court dismissed the plaintiff’s discrimination claim in part

because his allegations were not based on his actual religion (Univer-

salist) but rather on the misperception that he was Muslim.

The

court reasoned that “Congress has not enacted laws that would make

adverse employment actions based on someone being perceived as

having a religious affiliation, be it Catholic, Muslim, or any other re-

ligion, actionable under Title VII.”

3. Race

Many courts exact a more demanding burden on plaintiffs’ racial

identity claims than with sexual orientation or religion. First, unlike

religion or sexual orientation, some courts attempt to determine for

themselves the plaintiff’s racial identity based on what they call “ob-

jective” indicators. These indicators are predominantly ascriptive fac-

tors, specifically appearance and identity performance. For instance,

in Bennun v. Rutgers State University, the court determined that the

plaintiff was Hispanic based on its own observations of the plaintiff’s

appearance, his speech and mannerisms, his “immersion in Spanish

ways of life,” and “that he speaks Spanish in the home.”

In other

words, the court adjudicated the plaintiff’s identity to be Hispanic not

because he said he was (which he did) but based on how he looked,

spoke, and lived.

As additional support for its finding, the court re-

lied on ancestral evidence, citing the fact that the plaintiff was born in

Argentina and that his father traced his ancestry to Spain.

Although

the court recognized that discrimination is generally triggered by how

the alleged discriminator perceives the plaintiff, it did not determine

the plaintiff’s race using the discriminator’s perception; instead, the

court relied on its own assessment of the plaintiff’s race.

he otherwise would have been for reasons prohibited by the statute.” Id.; see also

Kallabat v. Mich. Bell Tel. Co., No. 12-CV-15470, 2015 WL 5358093, at *4 (E.D. Mich.

June 18, 2015) (allowing claim to proceed based on coworkers’ perception that plain-

tiff was Muslim).

87. El v. Max Daetwyler Corp., No. 3:09cv415, 2011 WL 1769805, at *6 (W.D.N.C.

May 9, 2011), aff’d, 451 F. App’x 257 (4th Cir. 2011).

88. Id.; see also Le v. N.Y. State, Off. of State Comptroller, No. 1:16-CV-1517,

2017 WL 3084414, at *6 (N.D.N.Y. July 18, 2017) (rejecting misperception claim);

Lewis v. N. Gen. Hosp., 502 F. Supp. 2d 390, 401 (S.D.N.Y. 2007) (rejecting mis-

perception claim but stating that misperception was not dispositive). The Max

Daetwyler court also held that plaintiff’s claim failed because he did not establish that

his coworkers thought he was Muslim. Max Daetwyler, 2011 WL 1769805, at *6–7.

89. Bennun v. Rutgers, 941 F.2d 154, 173 (3d Cir. 1991).

90. Id. at 172–73.

91. Id.

92. Id.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 23 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 289

Similarly, in Perkins v. Lake County Department of Utilities,

the

court determined whether the plaintiff was Native American

based

on its own assessment of the plaintiff’s “physical appearance [as] an

Indian” and the plaintiff’s self-identification.

The court also refer-

enced the plaintiff’s potential Native American ancestry, noting “he

may have some Native American blood, albeit less than 1/16th.”

However, the court also made an alternative holding that focused on

the discriminator’s perception of the plaintiff. The court reasoned that

“[s]o long as there is some objective evidence reflective of a basis” for

the plaintiff’s employer to think that he was Native American, the

court did not need to “resolve whether the employee is one percent

Native American or one hundred percent Native American.”

This

“objective basis,” according to the court, included ascriptive evidence

such as the plaintiff’s “physical appearance, language, cultural activi-

ties, or associations” that “lead the employer to believe that a given

employee is a member of that protected class, and to deal with him/

her on that basis.”

Although this alternative holding shifts the focus to the employer’s

perception of the plaintiff’s identity and away from the court’s assess-

ment of the plaintiff’s identity, the employer’s perception is still sub-

ject to an “objectivity” test based on the court’s understanding of

racial characteristics.

Like Bennun and Perkins, other courts have

used ascriptive evidence such as linguistic characteristics,

100

“charac-

93. Perkins v. Lake Cnty. Dep’t of Utils., 860 F. Supp. 1262, 1276 (N.D. Ohio

1994).

94. “Native American” status is considered a racialized identity and/or a political

category, depending on the context. See Bethany R. Berger, Race, Descent, and Tribal

Citizenship, 4

ALIF

. L. R

. C

23, 33, 25–26 (2013). This Article’s discussion of

anti-discrimination laws includes Native American under the umbrella of “race” be-

cause courts treat claims alleging discrimination on the basis of Native American sta-

tus as race or national origin claims. See, e.g., Perkins, 860 F. Supp. at 1272–74

(discussing the lack of coherent distinction between race and national original dis-

crimination claims). In different contexts, “Native American” is not treated as a racial

category. See Morton v. Mancari, 417 U.S. 535, 553 n.24 (1974) (holding that the Bu-

reau of Indian Affairs’ preference for hiring Native Americans was not racial discrim-

ination because it applies to “members of ‘federally recognized’ tribes” and not a

racial group).

95. Perkins, 860 F. Supp. at 1276.

96. Id.

97. Id. at 1278.

98. Id.

99. Id.; see also Greene v. Swain Cnty. P’ship for Health, 342 F. Supp. 2d 442, 451

(W.D.N.C. 2004) (following Perkins and holding that the plaintiff could prove “she

was a member of a protected class” by showing her employer “had a basis to reasona-

bly believe” that she was Native American (emphasis added)).

100. See Harel v. Rutgers, State Univ., 5 F. Supp. 2d 246, 269 (D.N.J. 1998)

(“Harel’s objective appearance likely might lead the reasonable employer to deduce

that he was Israeli: he was raised in Israel; he speaks Hebrew, his first language, at

home; he speaks English with a discernible accent; he remains an Israeli citizen; and

he clearly identifies with his Israeli culture.”).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 24 10-MAR-22 14:18

290 TEXAS A&M LAW REVIEW [Vol. 9

teristics common to Jamaican[-]born persons,”

101

and skin color

102

adjudicate plaintiffs’ racial identities.

103

And in one case involving

anti-Native American discrimination, the court required ancestral evi-

dence to prove Native American identity.

104

At least one court has rejected a plaintiff’s race discrimination

claims because he did not meet the court’s “objective” indicators of

racial identity. In Nieves v. Metropolitan Dade County, the court ques-

tioned the plaintiff’s Hispanic self-identification based on the court’s

own observations about the plaintiff, stating that “neither from ob-

serving the Plaintiff nor from listening to his speech patterns, manner-

isms and pronunciation of the English language was it apparent that

Plaintiff was Hispanic.”

105

Because these ascriptive indicators of the

plaintiff’s racial background did not align with what the court thought

Hispanic people looked or sounded like, the plaintiff’s identity was

not adjudicated to be Hispanic, and his claim failed.

106

These so-called “objective” indicators of racial identity are gener-

ally pulled from thin air. Courts are typically not relying on any legal

authority, much less anything in Title VII itself, to justify their reliance

on these forms of identity evidence. Nor do they explain why these

indicators are “objective” as opposed to the judge’s own subjective

thoughts about the constitutive elements of a particular racial identity.

In addition to this “objective” indicators of race inquiry, the second

way courts’ increased scrutiny on racial identity claims manifests is in

the more frequent rejection of misperception claims as compared with

other identity claims. For example, in Lopez–Galvan v. Mens

Wearhouse, the plaintiff identified as Latino and brought a discrimina-

tion claim on the ground that his employer misperceived him to be

Black and mistreated him based on that misperception.

107

The court

101. Holness v. Penn State Univ., No. 98–2484, 1999 WL 270388, at *6 (E.D. Pa.

May 5, 1999).

102. See, e.g., id.

103. See also Perry v. Autozoners, LLC, 948 F. Supp. 2d 778, 787 (W.D. Ky. 2013)

(finding that mixed-race individuals belong to a protected class as long as they “visu-

ally appear[ ]” to belong in that class).

104. Leonard v. Katsinas, No. 05-1069, 2007 WL 1106136, at *13 (C.D. Ill. Apr. 11,

2007). On summary judgment, the court rejected the defendants’ invitation to impose

an actuality requirement on the plaintiffs and held the discriminator’s belief that the

plaintiffs were Native American was sufficient. Id. However, the court also explained

that at trial, the plaintiffs would need to provide proof of their Native American an-

cestry. Id. Based on the court’s descriptions of the plaintiffs, many of them would not

be able to meet this standard. Id. at *2. All the plaintiffs self-identified as Native

American, but only one of the plaintiffs was, according to the court, a “card-carrying”

Native American because he was able to produce his father’s tribal enrollment card.

Id.

105. Nieves v. Metro. Dade Cnty., 598 F. Supp. 955, 961 (S.D. Fla. 1984).

106. Id.

107. See, e.g., Lopez–Galvan v. Mens Wearhouse, No. 3:06cv537, 2008 WL 2705604,

at *7–8 (W.D.N.C. July 10, 2008) (the exact allegation was that he was misperceived

as “Negroid”).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 25 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 291

held that his claim was not viable because Title VII does not explicitly

include people who are perceived to be a particular race.

108

So, be-

cause the plaintiff was not Black, he could not bring a claim.

109

Nu-

merous other courts have come to the same conclusion regarding

misperception claims in the context of race.

110

At least one court has required the discriminatory conduct itself—

in addition to the discriminator’s perception—to align with the plain-

tiff’s self-identification.

111

Said differently, the plaintiff’s self-identifi-

cation, ascribed identity, and the racial identity that is the target of the

harassment must align. In this case, Burrage v. FedEx Freight, Inc., the

plaintiff’s co-workers initially misperceived him as Mexican and

harassed him on that basis.

112

Once he clarified that he self-identified

as biracial (half Black and half white), his co-workers still harassed

him using derogatory comments about Mexicans.

113

The court dis-

missed Burrage’s claim and reasoned that he did not “maintain that

his supervisors and co-workers began to use any terms that

were . . . related to his status as an African-American, upon learning

of his true race.”

114

Under the court’s reasoning, Burrage could only

sustain a claim if he were in fact Mexican or if the derogatory com-

ments were about biracial or Black people.

115

To be sure, many other cases do not impose this requirement on

plaintiffs and have openly critiqued the rationale behind the actuality

requirement.

116

According to one district court, the actuality require-

ment “is as offensive as it is incorrect.”

117

For these courts, the plain-

tiff’s identity is of little importance, and these courts focus on how the

108. Id.

109. Id. at *9.

110. See, e.g., Sears v. Jo-Ann Stores, Inc., No. 3:12-1322, 2014 WL 1665048, at *7–8

(M.D. Tenn. Apr. 25, 2014) (rejecting discrimination-based misperception that the

plaintiff was Black); Yousif v. Landers McClarty Olathe KS, LLC, No. 12-2788, 2013

WL 5819703, at *5 (D. Kan. Oct. 29, 2013); Guthrey v. Cal. Dep’t of Corr. & Rehab.,

No. 1:10-cv-02177, 2012 WL 2499938, at *15 (E.D. Cal. June 27, 2012); Adler v. Ev-

anston Nw. Healthcare Corp., No. 07 C 4203, 2008 WL 5272455, at *4 (N.D. Ill. Dec.

16, 2008); Uddin v. Universal Avionics Sys. Corp., No. 1:05-CV-1115, 2006 WL

1835291, at *6 (N.D. Ga. June 30, 2006); Butler v. Potter, 345 F. Supp. 2d 844, 850

(E.D. Tenn. 2004).

111. Burrage v. FedEx Freight, Inc., No. 4:10CV2755, 2012 WL 1068794, at *8

(N.D. Ohio Mar. 29, 2012).

112. Id. at *1.

113. Id.

114. Id. at *8 n.8.

115. Greene, supra note 17, at 110 (making a similar observation about Burrage).

116. See EEOC v. MVM, Inc., No. TDC-17-2864, 2018 WL 2197727, at *10 (D. Md.

May 14, 2018) (“Discrimination where the employer is mistaken in his belief that an

employee is of a particular national origin is just as insidious as discrimination where

the employer is correct.”); Arsham v. Mayor & City Council of Balt., 85 F. Supp. 3d

841, 845 (D. Md. 2015); Henao v. Wyndham Vacations Resorts, Inc., 927 F. Supp. 2d

978, 987 (D. Haw. 2013).

117. Boutros v. Avis Rent A Car Sys., LLC, No. 10 C 8196, 2013 WL 3834405, at *7

(N.D. Ill. July 24, 2013).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 26 10-MAR-22 14:18

292 TEXAS A&M LAW REVIEW [Vol. 9

discriminator perceived the plaintiff’s identity. For instance, in Wood

v. Freeman Decorating Services, Inc., the employer’s perception of the

plaintiff’s Native American identity was sufficient even though the

plaintiff had no documentary evidence that he was Native Ameri-

can.

118

Rather, both the plaintiff’s birth certificate and his prior em-

ployment records identified him as white/Hispanic.

119

However, his

employer thought he was Native American, the slurs directed at him

were about Native Americans,

120

and Wood stated that he self-identi-

fied as Native American.

121

In summary, courts interrogate race (an immutable identity) more

closely than religion and sexual orientation (more mutable identities),

which manifests in two ways: first, by determining whether the plain-

tiff’s claimed racial identity meets some “objective” standard adopted

by the court; and second, by rejecting misperception claims. Courts

addressing discrimination claims based on sexual orientation and re-

ligion, for the most part, did neither.

4. Sex

122

Courts adjudicating sex in anti-discrimination cases also focus on

the plaintiff’s “actual” sex. For most courts, a plaintiff’s “actual” sex is

118. Wood v. Freeman Decorating Servs., Inc., No. 3:08–CV–00375, 2010 WL

653764, at *4 (D. Nev. Feb. 19, 2010).

119. Id.

120. Wood alleged that he was called things like a “fuck’n injun,” “prairie ni****,”

“injun,” and “wagon burner.” Id. at *3. He also stated that he was “forced [ ] to use a

desk that had a poster attached to it depicting a wagon on fire and with the words

‘Wagon Burner’” and that his co-workers “introduced him to clients as ‘one of our

wagon burners’ or ‘stupid injun.’” Id.

121. Id. at *5 (“The alleged discrimination is ‘no less malevolent because it may

have been based on an erroneous assumption.’” (alterations incorporated)); see also

Estate of Amos ex rel. Amos v. City of Page, 257 F.3d 1086, 1094 (9th Cir. 2001)

(mistaken belief that individual was Native American was sufficient to maintain equal

protection claim); Eriksen v. Allied Waste Sys., Inc., No. 06-13549, 2007 WL 1003851,

at *6 (E.D. Mich. Apr. 2, 2007) (same). For these courts, perceived race can be power-

ful enough to rebut other identity evidence, such as self-identification and ancestral

evidence. Chaib v. GEO Group, Inc., 92 F. Supp. 3d 829, 837 (S.D. Ind. 2015) (dis-

missing claim brought by a plaintiff who self-identified as African American and was

of Algerian descent because she was not perceived as African American by her em-

ployer or co-workers), aff’d, 819 F.3d 337 (7th Cir. 2016).

122. This Article does not differentiate between “sex” and “gender.” Recognizing

that the definitions of these two terms have sparked much debate, this Article does

not engage with that debate. Formulating universal definitions of sex and gender for

this Article is at odds with one of its primary arguments—namely, that the definitions

of identity are highly dependent on context. See

UDITH

UTLER

ENDER

ROUBLE

EMINISM AND THE

UBVERSION OF

DENTITY

1–34 (Linda A. Nicholson ed., 1990)

[hereinafter

ENDER

ROUBLE

];

UDITH

UTLER

ODIES

HAT

ATTER

: O

N THE

ISCURSIVE

IMITS OF

“S

”

8–12 (1993) [hereinafter

ODIES

HAT

ATTER

]; see

also Clarke, supra note 28, at 933–34 (resisting universal definitions for sex and

gender for similar reasons and explaining that “[a]ttempts to settle metaphysical

debates about what sex and gender are distract from the question of how these

concepts should be defined in particular legal contexts, if at all”). What is typically

referred to as “sex” (chromosomes, hormones, secondary sex characteristics) this

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 27 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 293

based on biological evidence, including their sex assigned at birth,

chromosomes, sex characteristics, or hormones. Documentary evi-

dence (like birth certificates), ascriptive sex, and self-identification are

also relevant in certain cases, but biological evidence is of primary

importance.

Most sex discrimination cases involve cisgender plaintiffs, and in

these cases, the plaintiff’s sex is not explicitly adjudicated. Typically, a

cisgender plaintiff’s sex is not questioned by the court or the parties

because they assume they know the “true” sex identity of the plaintiff.

That is, if the plaintiff says she is a woman, she ascriptively appears to

be a woman, and there is no evidence to think otherwise, the parties

and the court will assume that she has other traits associated with wo-

manhood (XX chromosomes, a uterus, etc.).

123

However, as explained

further below, even though her sex was not explicitly adjudicated, her

claims still depend on her “actual” sex. Thus, although sex adjudica-

tion is most visible when it comes to transgender plaintiffs, the same

focus on “actual” biological sex applies to cis and trans plaintiffs.

Until fairly recently, discrimination on the basis of transgender

identity was not covered under anti-discrimination laws’ prohibition

against sex discrimination. Now, and particularly after Bostock v.

Clayton County, the opposite is true. Yet courts have adjudicated sex

in similar ways regardless of whether discrimination against trans-

gender plaintiffs is covered. Specifically, sex adjudication has de-

pended on biological evidence. In older cases, courts denied

transgender plaintiffs’ claims because biological evidence of their sex

did not align with their self-identification or their ascriptive sex.

124

more recent cases, courts still rely on biological evidence but generally

allow transgender plaintiffs to bring claims under anti-discrimination

statutes.

125

Older cases interpreted anti-discrimination statutes’ prohibition of

sex discrimination as prohibiting discrimination against “women be-

cause they are women and against men because they are men.”

126

These courts understood Title VII and related statutes to apply when

people who were assigned female at birth are discriminated against

because of their identification as a woman and/or their performance

of their womanhood. A trans woman, under this reasoning, is born

male but is discriminated against based on her identification as a wo-

Article calls biological evidence of sex. And the common understanding of “gender”

(masculinity, femininity) is encompassed in what this Article calls ascriptive evidence

of sex.

123. Genitals ascribed to someone based on more visible sex cues (appearance,

mannerisms, etc.) have been referred to as “cultural genitals.” See

UZANNE

J. K

ESS-

LER

& W

ENDY

ENNA

ENDER

: A

THNOMETHODOLOGICAL

PPROACH

153

(Univ. of Chi. Press ed., 1978).

124. See, e.g. Ulane v. E. Airlines, Inc., 742 F.2d 1081, 1085 (7th Cir. 1984).

125. See generally Bostock v. Clayton Cnty., 140 S. Ct. 1731 (2020).

126. Ulane, 742 F.2d at 1085.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 28 10-MAR-22 14:18

294 TEXAS A&M LAW REVIEW [Vol. 9

man (or as a transgender woman) and performance of her woman-

hood; accordingly, Title VII did not apply because the discrimination

is against a biological man because she is a “woman”—not against a

“man” because he is a “man.”

One of the most oft-cited cases following this line of reasoning is a

1984 case from the Seventh Circuit, Ulane v. East Airlines. This case

involved a plaintiff who transitioned from male to female on the job

and was subsequently terminated.

127

The court recognized that the

plaintiff self-identified as a woman and ascriptively appeared to be a

woman, noting that Ulane took hormones, developed breasts, had sex

reassignment surgery, and changed the sex marker on her birth certifi-

cate.

128

Yet, the court still found that Ulane was a “biological male”

because she did not “have a uterus and ovaries,” could not “bear ba-

bies,” and still had XY chromosomes.

129

According to the court,

Ulane was someone with a “male body who believes himself to be

female.”

130

Because Ulane was a “biological man” who failed to allege

discrimination based on her “manhood,” she could not bring a claim

on this basis under Title VII.

131

In other words, the biological evi-

dence of her sex did not match the ascriptive evidence of her sex or

her self-identification.

The Tenth Circuit echoed this reasoning over 20 years later in Et-

sitty v. Utah Transit Authority.

132

The court relied on Ulane’s holding

that Title VII’s “prohibition on sex discrimination means only that it is

‘unlawful to discriminate against women because they are women and

men because they are men.’”

133

Thus, according to the court, Title VII

did not recognize a claim based on a plaintiff’s “status as a transsex-

ual.”

134

It explained that Title VII “extends to transsexual employees

only if they are discriminated against because they are male or be-

cause they are female.”

135

Because the plaintiff was alleging discrimi-

nation based on the incongruity between her sex assigned at birth

(male) and her chosen sex (female or “transsexual”), she could not

state a claim.

136

More recent cases allow transgender plaintiffs to get relief under

Title VII, but courts’ adjudication of sex still mostly relies on plain-

127. Id. at 1082.

128. Id. at 1083.

129. Id. at 1083, 1087.

130. Id. at 1085.

131. Id. at 1087.

132. Etsitty v. Utah Transit Auth., 502 F.3d 1215 (10th Cir. 2007).

133. Id. at 1221 (quoting Ulane, 742 F.2d at 1085).

134. Id. at 1221.

135. Id. at 1222; see also Oiler v. Winn–Dixie La., Inc., No. 00-3114, 2002 WL

31098541, at *3 (E.D. La. Sept. 16, 2002) (holding that Title VII does not recognize

claims by “a person born with a male body who believes himself to be a female, or a

person born with a female body who believes herself to be male” (quoting Ulane, 742

F.2d at 1085)).

136. Etsitty, 502 F.3d at 1221.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 29 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 295

tiffs’ biological traits, and these biological traits still determine the

plaintiff’s “actual” sex. The only difference is that rather than requir-

ing plaintiffs’ biological traits to align with their ascriptive sex and

self-identification, it is the mismatch between these types of identity

evidence that allows them to state a claim.

This line of cases relies on a sex-stereotyping theory of sex discrimi-

nation—that requiring conformity between biological sex and ascrip-

tive sex/self-identification constitutes sex discrimination. Sex-

stereotyping finds its origins in Price Waterhouse v. Hopkins, a case

about a cisgender woman’s gender nonconformity.

137

In Price

Waterhouse, Ann Hopkins, a senior manager at a top accounting firm,

sued under Title VII, alleging that she had been denied a promotion

to partner because her “aggressive” demeanor did not align with the

partners’ understanding of femininity.

138

The Supreme Court held that

discrimination based on sex included discrimination based on a failure

to conform to sex stereotypes—i.e., an employer cannot fire a cis-

gender woman because she fails to perform her femininity in accor-

dance with societal expectations of womanhood.

139

Using this analytical framework, courts have extended the Price

Waterhouse rationale to include discrimination claims from trans-

gender plaintiffs, reasoning that the discrimination stemmed from a

mismatch between the plaintiff’s “actual” sex (based on biological

traits) and the plaintiff’s sex presentation and performance (i.e., their

ascriptive sex). Two examples follow.

First, in Smith v. City of Salem, the Sixth Circuit held that a trans-

gender plaintiff could bring a discrimination claim under Title VII,

140

reasoning that if “an employer who discriminates against women be-

cause, for instance, they do not wear dresses or makeup, is engaging in

sex discrimination,” then so are “employers who discriminate against

men because they do wear dresses and makeup, or otherwise act femi-

ninely.”

141

Thus, the court adjudicated the plaintiff’s “actual” sex to

be “male” because she was assigned male at birth.

142

Because her “ac-

tual” sex did not align with her ascriptive sex, the court applied Price

Waterhouse.

143

137. Price Waterhouse v. Hopkins, 490 U.S. 228, 233–36 (1989).

138. Id.

139. Id. at 252; see also Back v. Hastings on Hudson Union Free Sch. Dist., 365 F.3d

107, 113 (2d Cir. 2004) (holding that employer could not fire an employee based on

the stereotype that mothers cannot devote sufficient time to their jobs).

140. Smith v. City of Salem, 378 F.3d 566, 568, 572 (6th Cir. 2004).

141. Id. at 574; see also Glenn v. Brumby, 663 F.3d 1312, 1316 (11th Cir. 2011)

(applying the mismatch theory to an equal protection claim brought by a transgender

plaintiff and stating that “[t]he very acts that define transgender people as trans-

gender are those that contradict stereotypes of gender-appropriate appearance and

behavior”).

142. Smith, 378 F.3d 566.

143. Id. at 574–75.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 30 10-MAR-22 14:18

296 TEXAS A&M LAW REVIEW [Vol. 9

About 15 years later, the Sixth Circuit again held that an employer

who discriminated against a transgender woman violated Title VII be-

cause the discrimination stemmed from a mismatch between the plain-

tiff’s biological traits and her self-identification/ascriptive sex.

144

And

again, the court’s holding was based on a determination that the plain-

tiff’s “actual” sex was “male” and biologically based.

145

Specially, the

court reasoned that “discrimination on the basis of transgender status

necessarily entails discrimination on the basis of sex” because “an em-

ployer cannot discriminate against an employee for being transgender

without considering that employee’s biological sex.”

146

These cases suggest that transgender plaintiffs’ “real” sex is the one

they were assigned at birth and ascriptive evidence of their sex merely

does not align with that sex. Because these courts fit these cases into

the Price Waterhouse rationale, they treat transgender plaintiffs as

simply more “gender nonconforming” than the cisgender plaintiff in

Price Waterhouse—a difference in degree rather than in type.

147

The recent Supreme Court decision Bostock v. Clayton County—

which held that Title VII’s prohibition on the basis of sex includes

“transgender status”—does not rely on Price Waterhouse or sex-stere-

otyping theory specifically. Yet the Court’s rationale still rests on

identifying biological evidence of sex, and specifically a plaintiff’s sex

assigned at birth. First, the Court adopts a definition of “sex” that

refers “only to biological distinctions between male and female.”

148

Then, it reasons that to discriminate against someone for being trans-

gender, the discriminator must take the “sex” of the individual into

account.

149

The Court explained that if an employer fires someone

“who was identified as a male at birth but who now identifies as a

female” but “retains an otherwise identical employee who was identi-

144. EEOC v. R.G. &. G.R. Harris Funeral Homes, Inc., 884 F.3d 560, 574–75 (6th

Cir. 2018).

145. Id. at 578 (reasoning that “discrimination on the basis of transgender status

necessarily entails discrimination on the basis of sex—no matter what sex the em-

ployee was born or wishes to be”). Schoenbaum calls this shift to “the view that trans-

gender persons are ‘inherently’ gender non-conforming, and thus that being

transgender is, by definition, gender nonconformity to sex” the “per se” approach.

Naomi Schoenbaum, The New Law of Gender Nonconformity, 105

INN

. L. R

831, 854 (2020).

146. Harris, 884 F.3d at 578 (emphasis added).

147. As other commentators have pointed out, this reasoning “effectively erases

transgenderism as an identity,” because “in order to avail herself of the gender-stere-

otyping theory[,] [the transgender woman] had to take on a male identity, namely, as

a man who wanted to participate in the workplace dressing and looking like a wo-

man.” Zachary A. Kramer, The New Sex Discrimination, 63

UKE

L.J.

891, 916 (2014)

(emphasis added); see also

NNA

IRKLAND

, F

IGHTS

: D

ILEMMAS OF

IFFER-

ENCE AND

ERSONHOOD

86 (2008) (“Transsexuals or transgender people per se do

not really exist in the Smith opinion; there just happen to be some men out there who

want to wear dresses.”).

148. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1739 (2020).

149. Id. at 1741.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 31 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 297

fied as female at birth, the employer intentionally penalizes a person

identified as male at birth for traits or actions that it tolerates in an

employee identified as female at birth.”

150

Thus, Bostock still relies on

biological traits to determine the “actual” sex of a plaintiff.

151

Although this Subsection has focused on transgender plaintiffs, this

same focus on biological evidence of sex applies to cisgender plain-

tiffs—the difference is that the plaintiff’s “actual” sex is not contested.

Said differently, a cisgender plaintiff’s biological traits still dictate her

“sex,” but the composition of those biological traits is assumed. Ann

Hopkins did not have to prove that she had a vagina, a uterus, and XX

chromosomes because she was presumed to have them. But those bio-

logical traits were still constitutive of her “actual” sex and formed the

basis of the Court’s sex-stereotyping holding.

152

Legal scholar Naomi Schoenbaum’s analysis of Price Waterhouse

elucidates the role “actual,” biological-based sex plays in sex stere-

otyping cases involving cisgender people. She argues that courts em-

ploy a three-step analysis when applying the Price Waterhouse sex-

stereotyping doctrine.

153

The very first step is determining the plain-

tiff’s “actual” sex (i.e., Ann Hopkins is a woman).

154

Then, courts

evaluate how the employer perceived the plaintiff’s gender (i.e., “the

employer viewed Hopkins as ‘masculine’”).

155

Last, courts “compar[e]

the expected gender performance based on the plaintiff’s sex with her

perceived gender performance: [I]f the employer acted because these

do not ‘match,’ the employer has engaged in unlawful sex stereotyp-

ing.”

156

Adjudicating the plaintiff’s “actual” sex, therefore, is founda-

tional to courts’ analyses.

Sex is the identity most rigorously scrutinized in anti-discrimination

law for two reasons. First, “actual” sex appears to be key to most dis-

crimination cases. For racial identity, only some courts adjudicate the

“actual” identity of the plaintiff, and very few do so in the context of

sexual orientation and religion. Second, courts rely more heavily on

biological evidence in sex adjudication than any other identity, and

biological evidence of sex is either impossible to change (like chromo-

somes) or difficult to change (like having surgery to change sex char-

acteristics). No other identity category analyzed in this Article is

adjudicated with as heavy of a reliance on biological evidence as sex.

150. Id.

151. Not all courts define sex according to sex assigned at birth, or other biological

traits, like Bostock. See, e.g., Harris, 884 F.3d at 578; Fabian v. Hosp. of Cent. Conn.,

172 F. Supp. 3d 509, 526 (D. Conn. 2016) (rejecting a biologically based, binary defini-

tion of sex).

152. See generally Price Waterhouse v. Hopkins, 490 U.S. 228 (1989).

153. Schoenbaum, supra note 145, at 842.

154. Id.

155. Id.

156. Id. at 842–43.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 32 10-MAR-22 14:18

298 TEXAS A&M LAW REVIEW [Vol. 9

*************************

This cross-identity examination of identity adjudication in anti-dis-

crimination law has revealed a large degree of inconsistency. For in-

stance, some courts accept racial misperception claims while others

don’t; courts are more likely to reject racial misperception claims as

compared with sexual orientation and religious misperception claims;

a plaintiff’s “actual” race and sex are more likely to matter than a

plaintiff’s “actual” sexual orientation or religion. Additionally, legal

actors sometimes adjudicate plaintiffs’ racial identity claims them-

selves based on “objective” standards. Yet with sexual orientation and

religion, courts do not require plaintiffs’ identity claims to meet some

supposedly “objective” criteria. They do not, for instance, require

plaintiffs bringing a sexual orientation discrimination claim to be in a

same-sex relationship, be a member of an LGBT organization, “look”

gay, or even self-identify as gay. For religious identity, requiring some-

one’s religious beliefs to meet some “objective” standard would likely

violate the First Amendment.

157

But there are no doctrinal justifications for why courts are adjudi-

cating these identity claims differently. The applicable anti-discrimina-

tion statutes and case law instruct courts to apply the same analytical

framework regardless of the specific type of discrimination alleged.

Moreover, courts themselves are not explaining why certain identity

claims receive different treatment—indeed, they may not even be

aware of this differential treatment. Thus, not only is there inconsis-

tent identity adjudication in anti-discrimination law, this inconsistency

is not made visible or explained.

C. Benefit Laws

Generally speaking, benefit laws involve forward-facing state action

and may involve individuals asking a state actor to affirmatively do

something on their behalf—like deem them eligible for something,

grant them a certain status, or classify them in a particular way. Some

examples include: qualifying for government contracting programs

targeting businesses owned by racial minorities; meeting legal require-

ments to have the state change sex markers on government docu-

ments; and being eligible for immigration relief based on a fear of

identity-based persecution. To be sure, this definition is fairly broad.

But to conduct a coherent cross-identity analysis of identity adjudica-

tion, the definition of “benefit” law needs to be broad. Unlike anti-

discrimination laws, which apply to all of the identity categories dis-

cussed in this Article, identity adjudication in other legal contexts is

not as doctrinally confined.

157. See infra note 185 and accompanying discussion of the religious question

doctrine.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 33 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 299

In spite of the relative capaciousness of this category, identity adju-

dication tends to operate somewhat similarly within this category.

Specifically, benefit laws tend to adjudicate identity more intensely

across identity categories than anti-discrimination laws and data-col-

lection laws. Self-identification alone is insufficient across the board to

prove identity, and various combinations of ancestral, biological, doc-

umentary, and/or ascriptive evidence of identity are necessary, de-

pending on the identity.

This increased identity interrogation for benefit laws is exhibited

not only through the amount of identity evidence required but also, at

times, through “identity estoppel.”

158

Identity estoppel occurs when

the identity an individual is currently claiming is different from one

that they have previously claimed, and this prior inconsistent self-

identification estops them from claiming their current identity. For ex-

ample, if an individual previously identified as X on a form and now

make a claim as Y, identity estoppel would prevent them from claim-

ing to be Y. Like with anti-discrimination laws, race and sex, the more

immutable identities, tend to be more closely interrogated than relig-

ion or sexual orientation, and identity estoppel is more frequent in

race and sex identity claims than with other identities.

1. Sexual Orientation

One area of law that provides a benefit based on sexual orientation

is asylum law. Individuals are eligible for asylum if they can establish

that they have either experienced past persecution or have a “well-

founded fear of persecution” based on, among other things, “member-

ship of a particular social group,” which has been construed to include

sexual orientation.

159

In other words, asylum functions to prevent peo-

ple from being harmed in their home country based on sexual orienta-

tion. Petitioners first bring their asylum claims before an immigration

judge (“IJ”) and then may appeal to the Board of Immigration Ap-

peals (“BIA”) if they lose. They are then permitted to appeal to the

United States Courts of Appeal. The Courts of Appeal may reverse

the BIA’s decision, but reversal is rare due to the deferential standard

under which the courts review these decisions.

160

Thus, they often af-

firm IJ’s/BIA’s adjudication of sexual orientation.

In order for petitioners to prove their sexual orientation in this con-

text, they cannot rely on mere self-identification and generally need to

158. See Christopher A. Ford, Administering Identity: The Determination of “Race”

in Race-Conscious Law, 82

ALIF

. L. R

1231, 1258–59 (1994) (discussing this con-

cept of identity estoppel in the context of race adjudication).

159. See In re Toboso-Alfonso, 20 I. & N. Dec. 819, 822 (B.I.A. 1990).

160. 8 U.S.C. § 1252(b)(4)(B) (requiring the reviewing court to yield unless “any

reasonable adjudicator would be compelled to conclude” that the IJ (or the Board)

erred); see also Tawuo v. Lynch, 799 F.3d 725, 727 (7th Cir. 2015) (explaining this

highly deferential standard).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 34 10-MAR-22 14:18

300 TEXAS A&M LAW REVIEW [Vol. 9

put forth some sort of documentary evidence to support their identity

claim. Ascriptive evidence is also important—courts have looked to

the petitioner’s appearance and mannerisms, the sex of their romantic

and sexual partners, and whether they are a member of an LGBT or-

ganization in their home country.

Courts’ focus on romantic and sexual relationships as proof of sex-

ual orientation played out in Fuller v. Lynch.

161

In this case, the IJ and

the BIA found that the asylum petitioner was not bisexual, as he

claimed to be.

162

They decisions relied on the fact that he was married

to a woman, had fathered children with a woman, was previously con-

victed of sexual assault on a woman, and lacked any corroborating

evidence from his former boyfriends.

163

They also pointed to inconsis-

tencies in his testimony related to his sexual orientation,

164

as well as

testimony unrelated to his claim to bisexuality.

165

On appeal, the Sev-

enth Circuit affirmed the IJ’s determination that the petitioner was

not bisexual.

166

As the dissenting opinion noted, IJ and the BIA ignored additional

evidence that supported Fuller’s claim that he was bisexual, including

his in-depth testimony “about his being bisexual and having had nu-

merous sexual relationships with both men and women” and about

the repeated physical and verbal abuse he endured because of his sex-

uality.

167

The dissent also critiqued the IJ’s reliance on inconsistencies

in Fuller’s testimony that had nothing to do with his bisexuality and

the IJ’s “refus[al] to believe the seven letters from Fuller’s children

and friends attesting to his bisexuality.”

168

For these legal actors, the fact that Fuller had sexual and romantic

relationships with women trumped all this other evidence indicating

that he was bisexual. In other words, his heterosexual relationships

estopped him from making a claim to bisexuality—even though his

relationship with a woman did not contradict his claim that he was

bisexual. As the dissent noted in its critique of the IJ’s rationale, “Ap-

161. Fuller v. Lynch, 833 F.3d 866, 869–72 (7th Cir. 2016).

162. Id. at 871–72.

163. Id. at 869–70.

164. Inconsistencies in testimony that he was shot by an “anti-gay mob” when he

was at a party with his then boyfriend. Id. at 868, 870.

165. These inconsistences included “confus[ing] his sisters’ names, mix[ing] up a

sister with his mother, and g[iving] different figures for the number of sisters that he

had.” Id. at 869.

166. Id. at 871.

167. Id. at 872 (Posner, J., dissenting). The dissent stated:

He testified that in college he was stoned by other students on several occa-

sions and a few years later taunted as gay by a group of men who sliced his

face with a knife. On another occasion he was robbed at gunpoint by a man

who called him a “batty man,” which is a Jamaican slur for a homosexual.

Id.

168. Id. at 873.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 35 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 301

parently, the immigration judge does not know the meaning of

bisexual.”

169

A similar example involving a lesbian identity claim comes from the

Eleventh Circuit. In this case, Ingrida Mockeviciene sought asylum

based on her sexuality.

170

She submitted substantial evidence that she

had been persecuted in her home country because she was a lesbian,

including evidence that she was assaulted by the police, evicted from

her house, and beaten and raped by her husband while his friends held

her down.

171

She also submitted letters from her friends stating that

they knew she was a lesbian and that the police harmed her because of

her orientation.

172

Her daughter’s testimony also corroborated her

allegations.

173

However, the IJ found that Mockeviciene’s claim that she was a

lesbian was not credible.

174

He based his finding primarily on the fact

that she never had a same-sex partner and was “[a]t best . . . a non-

practicing lesbian.”

175

He also noted that she had “not joined any

groups while being here in the United States for four years that in-

volve[d] lesbian activities” and had no documents supporting her alle-

gations regarding the police violence she endured.

176

The BIA did not

explicitly adopt the IJ’s reasoning, but it affirmed the decision based

on Mockeviciene’s marriage to a man because the marriage undercut

her lesbian identity claim.

177

The Eleventh Circuit was “skeptical”

about the IJ’s reasoning but held that “[g]iven Mockeviciene’s recent

marriage, the evidence does not compel reversal of the BIA’s credibil-

ity determination.”

178

Just like in Fuller, Mockeviciene’s marriage to a

man negated the other evidence supporting her lesbian identity claim,

i.e., it estopped her from claiming a lesbian identity.

179

These two cases show courts’ heavy, and sometimes illogical, reli-

ance on prior sexual relationships as a proxy for sexual orientation.

Even though Fuller and Mockeviciene had been in relationships with

people of a different sex, they put forth other evidence showing that

169. Id. at 874 (emphasis in original).

170. Mockeviciene v. U.S. Att’y Gen., 237 Fed. App’x 569, 570 (11th Cir. 2007).

171. Id. at 570–71.

172. Id. at 571.

173. Id.

174. Id. at 572.

175. Id.

176. Id.

177. Id. at 573.

178. Id. at 574 (emphasis added).

179. Id.; see also Eke v. Mukasey, 512 F.3d 372 (7th Cir. 2008). In Eke, the court

held that the BIA and IJ did not err by determining that Eke was not gay because he

did not provide any “documentation indicating his sexual preferences, such as letters,

affidavits, photographs, or other forms of corroborative evidence,” or his sexual ori-

entation; the courts reached this conclusion even though Eke testified that he was in a

long term relationship with a man, was harassed due to his sexual orientation, and was

pressured to marry a woman who, after discovering that he was gay, ended the rela-

tionship. Eke, 512 F.3d at 375–76, 381.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 36 10-MAR-22 14:18

302 TEXAS A&M LAW REVIEW [Vol. 9

they would face persecution in their home country based on their sex-

ual orientation. These courts, however, seemed to care more about

whether they were “really” the sexuality they claimed to be, as op-

posed to than the legally relevant question about likelihood of perse-

cution based on sexual orientation.

180

Other courts take a different approach and adjudicate sexual orien-

tation by looking to the petitioner’s perceived sexual orientation in

their home country. This approach is not focused on determining the

petitioner’s actual sexual orientation but rather, examines how the pe-

titioner’s potential persecutors will understand or perceive their sexu-

ality—which is similar to the anti-discrimination cases that focus on

how the alleged discriminator perceived the plaintiff’s identity. The

Third Circuit took this approach in Amanfi v. Ashcroft.

181

In Amanfi,

the petitioner had sex with another man to avoid being ritually sacri-

ficed, but after he was spared, he was harassed because people be-

lieved he was gay.

182

The court held that this constituted imputed

membership in the LGBT community and remanded the case for fur-

ther proceedings.

183

Other courts of appeal have also held that per-

ceived sexuality can be grounds for immigration relief under similar

reasoning.

184

2. Religion

Religious identity adjudication in the context of benefit laws has

constitutional limits, unlike other identity claims.

185

Under the “relig-

180. While courts treat the sex of someone’s sexual partners as highly probative of

sexual orientation in this context, courts generally do not permit IJs to consider ap-

pearance and mannerisms (i.e. ascriptive evidence) when adjudicating sexual orienta-

tion. See Shahinaj v. Gonzales, 481 F.3d 1027, 1029 (8th Cir. 2007) (reversing the IJ’s

finding that Shahinaj was not gay based on the IJ’s “personal and improper opinion

Shahinaj did not dress or speak like or exhibit the mannerisms of a homosexual” and

“Shahinaj’s lack of membership in any Albanian homosexual organizations”);

Razkane v. Holder, 562 F.3d 1283, 1288 (10th Cir. 2009) (rejecting IJ’s determination

that the petitioner was not gay based on the fact that the petitioner did not “dress in

an effeminate manner or affect any effeminate mannerisms” and reasoning that the

IJ’s reliance on his own views of the appearance, dress, and affect of a homosexual

“elevate[d] stereotypical assumptions,” which could “lead to unpredictable, inconsis-

tent, and unreviewable results”).

181. Amanfi v. Ashcroft, 328 F.3d 719 (3d Cir. 2003).

182. Id. at 722–23.

183. Id. at 730.

184. See, e.g., Velasquez-Banegas v. Lynch, 846 F.3d 258, 259–60, 264 (7th Cir.

2017) (holding that petitioner was likely to be regarded as gay in his home country

because he was HIV positive and had never married and vacating BIA’s denial of his

withholding of removal); Pozos v. Gonzales, 141 F. App’x 629, 632–33 (9th Cir.

2005) (holding that the petitioner had a well-founded fear of persecution based on his

perceived sexual orientation, even though he claimed that he was not actually gay).

185. This doctrine can also come up in anti-discrimination laws, or any laws, not

just benefit laws. But because religious identity claims are not very rigorously interro-

gated in other contexts, and are more intensely examined in benefit laws, this doctrine

is most relevant in the benefit law analysis.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 37 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 303

ious question doctrine,” the First Amendment prevents courts from

evaluating the accuracy of someone’s religious beliefs; however,

courts may adjudicate the sincerity of those beliefs.

186

In other words,

courts cannot determine whether someone is Catholic based on their

belief in or adherence to some set of “objective” Catholic tenants—

someone can be wrong about what it means to be Catholic and still be

Catholic. But courts can determine whether someone sincerely identi-

fies with Catholicism and whether they actually believe that their faith

mandates a particular belief or action. This doctrine prevents religious

identity from being as closely interrogated as other identity claims be-

cause individuals do not need to meet a certain set of predetermined

criteria to prove their identity. Despite the religious question doctrine,

the law requires religious identity adjudication for benefit laws in a

number of legal contexts.

187

This Article focuses on asylum to limit

this Article’s scope and to have some doctrinal consistency in this

section.

Asylum seekers can generally prove their religious identity by sub-

mitting documentary evidence and testifying about their beliefs. For

instance, one petitioner sought asylum on the grounds that he had

converted from Islam to Christianity.

188

He submitted documentary

evidence of his conversion, including a letter from his current pastors

confirming his faith and baptism, and a certificate of his baptism.

189

This evidence was sufficient to assure both the BIA and the Ninth

Circuit that this was “not a case that presents any palpable risk of

admitting any fraudulent claims on the basis of a phony religious con-

version.”

190

Similar cases abound.

191

186. United States v. Ballard, 322 U.S. 78, 86–88 (1944); see also Nathan S. Chap-

man, Adjudicating Religious Sincerity,

92 W

ASH

. L. R

. 1185,

1187–88 (2017); Ben

Adams & Cynthia Barmore, Questioning Sincerity: The Role of the Courts After

Hobby Lobby, 67

TAN

. L. R

NLINE

59, 59–60 (2014) (“There is a long tradition

of courts competently scrutinizing asserted religious beliefs for sincerity without delv-

ing into their validity or verity.”).

187. See, e.g., Ballard, 322 U.S. at 86–87 (1944) (criminal law); Witmer v. United

States, 348 U.S. 375, 378–82 (1955) (conscientious objector to war); Andreola v.

Glass, No. 04-C-282, 2006 WL 2645186, at *2 (E.D. Wis. Sept. 13, 2006) (prisoner’s

rights); Dobkin v. District of Columbia, 194 A.2d 657, 659 (D.C. 1963) (relief from

criminal conviction); Ideal Life Church of Lake Elmo v. Cnty. of Washington, 304

N.W.2d 308, 310–13 (Minn. 1981) (tax exemptions for churches).

188. See Kazemzadeh v. U.S. Att’y Gen., 577 F.3d 1341, 1345 (11th Cir. 2009).

189. Id. at 1346.

190. Id. at 1356.

191. See Rizal v. Gonzales, 442 F.3d 84, 89–93 (2d Cir. 2006) (holding that the peti-

tioner sufficiently proved that he converted to Christianity when he submitted his

certificate of baptism, an unsworn affidavit from a priest, a letter from his congrega-

tion confirming his membership, and a letter from his sister about his conversion);

Jiang v. Gonzales, 485 F.3d 992, 995, 997–98 (7th Cir. 2007) (reversing IJ’s decision

that petitioner was not a “real Christian” based on his limited knowledge of Christian-

ity); Cosa v. Mukasey, 543 F.3d 1066, 1070–71 (9th Cir. 2008) (reversing IJ’s determi-

nation that petitioner was not a Millenist when petitioner submitted declarations from

the leader of her Millenist group and fellow Millenists and IJ’s finding was based on

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 38 10-MAR-22 14:18

304 TEXAS A&M LAW REVIEW [Vol. 9

Like with sexual orientation, some courts focus on whether the peti-

tioner will be persecuted based on an imputed/perceived religious be-

lief or identity rather than whether the petitioner’s religious claim is

sincere. These courts examine how the petitioner’s identity would be

understood by those likely to persecute them and are “not as con-

cerned with the heart of the convert.”

192

For example, in Bastanipour

v. I.N.S., the BIA found that the petitioner had not presented suffi-

cient evidence that he had converted to Christianity from Islam in part

because he was never baptized, was not a member of a church, and

had requested a pork-free diet in prison.

193

The Seventh Circuit re-

versed and reasoned that the BIA’s focus should not have been on

whether BIA itself thought the petitioner had converted to Christian-

ity, but on whether the petitioner had converted to Christianity in the

eyes of those set to punish him for his conversion if he returned to his

home country.

194

The petitioner’s actual religion was not important in this case.

Rather, even if he was lying about his religious beliefs, under the

court’s reasoning, he could still be eligible for immigration relief as

long as he would be persecuted based on his claimed religious be-

lief.

195

Under this approach, both a petitioner who actually subscribes

to her asserted religious belief and one who is faking religious conver-

sion for opportunist reasons are treated the same way for purposes of

asylum and similar forms of religion-based immigration relief.

3. Race

Many scholars have examined the history of racial adjudication in

the context of what this Article defines as benefit laws. This important

work elucidates law’s significant role in drawing and enforcing strict

boundaries around whiteness, which, in turn, ensured that individuals

who the courts did not understand to be white did not gain access to

the benefits reserved for those who were “actually” white.

196

petitioner’s doctrinal knowledge of the religion). There are cases where petitioners’

religious claims are rejected even when they provide documentary evidence to sup-

port their identities; the documentary evidence in these cases, however, appears weak.

See Mendoza v. I.N.S., 28 F. App’x 586, 587 (8th Cir. 2002) (The documentary evi-

dence was a statement from the mayor of petitioner’s former town, and the petitioner

testified that he did not know the mayor.); Mejia-Paiz v. I.N.S., 111 F.3d 720, 724 (9th

Cir. 1997) (affidavit from the petitioner’s brother did not provide any basis for his

knowledge that petitioner was a Jehovah’s Witness).

192. Najafi v. I.N.S., 104 F.3d 943, 949 (7th Cir. 1997).

193. Bastanipour v. I.N.S., 980 F.2d 1129, 1132–33 (7th Cir. 1992).

194. Id. at 1133.

195. Id. at 1132–33. The court was open to the argument that “a nonbelieving

Christian is not really a Christian and thus does not fall within the scope of the asylum

statute.” Id. But the BIA did not deny the asylum application on those grounds, so

the court did not determine the validity of that argument. Id.

196. Until fairly recently, many laws explicitly required individuals to either be

white or to not be a disfavored racial minority to gain access to a variety of legal

benefits. Racial identity determined individuals’ eligibility to enter the United States

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 39 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 305

This Article, on the other hand, focuses on modern race adjudica-

tion in the context of benefit laws.

197

The benefit laws discussed in this

Subsection include judicial adjudication of eligibility for government-

ordered affirmative action programs, programs for contracting with

businesses owned by racial minorities, and the distribution of settle-

ment funds from a race discrimination lawsuit to racial minorities.

This Subsection only focuses on affirmative action in the context of an

adversarial proceeding, where a court or other adjudicative body de-

termines someone’s eligibility for a particular program. It does not

include affirmative action in contexts like university admissions,

where a court is not adjudicating race through an adversarial

process.

198

Modern racial adjudication interrogates claims of racial identity

very closely, particularly when an individual’s racial identity claim

does not align with how others typically perceive their racial identity.

Legal actors consider ancestral and documentary evidence of race, as

well as proof that people in the individual’s community perceive them

as their claimed racial identity. Identity estoppel is another common

thread through these cases, and courts look to prior inconsistent racial

self-identification as evidence that the current racial identity claim is

false.

and be naturalized as United States citizens, whom they could marry, whether they

could vote, and whether they could be legally enslaved.

OPEZ

, supra note 26, at

11–13; see also Ariela J. Gross, Litigating Whiteness: Trials of Racial Determination in

the Nineteenth-Century South, 108

ALE

L.J

. 109, 120–23 (1998) (examining the racial

determination cases in the nineteenth century).

197. The goals of the modern benefit laws are markedly different than the ones at

issue in older race-litigation cases. While the latter were intended to maintain strict

racial boundaries in service of white supremacy, the former, at least on their face, are

generally about remediation for prior discrimination or racial diversity.

198. Racial adjudication in university admissions and similar contexts is often not

very transparent and therefore difficult to evaluate. Decisionmakers likely consider

some combination of self-identification, membership in affinity groups, physical ap-

pearance (through an Internet search or a personal interview), and evidence in the

personal essay. However, there is usually no formal racial adjudication in this context,

and the weight decisionmakers give to these various forms of evidence is not clear.

Additionally, as scholars have noted, universities and employers who are interested in

having racial diversity at their institutions have little incentive to adjudicate race. The

broader their racial definitions and the less they scrutinize racial identity claims, the

better their chances of claiming more diversity. See Camille Gear Rich, Affirmative

Action in the Era of Elective Race: Racial Commodification and the Promise of the

New Functionalism, 102 G

. L.J. 179, 196 (2013) [hereinafter Racial Commodifica-

tion and the Promise of the New Functionalism] (“The institution or employer is incen-

tivized to accept even the most weak racial-identity claims as it will get the benefit of

an employee it can categorize as a minority worker for diversity-reporting purposes

without having to do the work necessary to reach out to more acutely subordinated

workers from heavily racialized minority communities.”); Nancy Leong, Racial Capi-

talism, 126

ARV

. L. R

2151, 2156 (2013) (explaining how employers commodify

their employees’ race to their advantage).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 40 10-MAR-22 14:18

306 TEXAS A&M LAW REVIEW [Vol. 9

One of the most frequently discussed cases in the context of a race-

conscious hiring program is Malone v. Civil Service Commission.

199

Malone, two brothers applied to the Boston fire department and self-

identified as white on their applications.

200

They were not hired be-

cause their scores on the entrance exam were too low.

201

The brothers

applied a second time but, this time, identified themselves as Black on

their applications.

202

Under the fire department’s court-ordered af-

firmative action program, the passing score on the entrance exam for

Black applicants was lower than for white applicants.

203

The Malone

brothers were subsequently hired.

204

About ten years later, the com-

missioner of the fire department suspected that the brothers had lied

about their race on their applications, and a hearing officer was ap-

pointed to determine whether the commissioner’s suspicions were

correct.

205

After a two-day evidentiary hearing, the officer concluded that the

brothers were not Black and were not eligible for the affirmative ac-

tion program.

206

There was no pre-existing test to determine racial

identity, so the officer developed the following three-part test herself:

“(i) visual observation of physical features; (ii) documentary evidence

establishing black ancestry, such as birth certificates; and (iii) evidence

that the Malones or their families held themselves out to be black and

are considered black in the community.”

207

Or in the terms of this

Article, (1) ascriptive race as determined by the adjudicator, (2) an-

cestral evidence, and (3) ascriptive race as determined by third

parties.

208

As to the first factor, the officer found that the Malones did not

appear to be Black to her because they had “fair skin, fair hair color-

ing, and Caucasian facial features.”

209

For the second factor, the

brothers submitted a picture of a woman who they claimed was their

Black great-grandmother.

210

However, the officer discounted this evi-

dence because they could not prove that the woman in the picture was

Black.

211

The officer instead relied on other evidence indicating that

199. Malone v. Civ. Serv. Comm’n, 646 N.E.2d 150 (Mass. App. Ct. 1995).

200. Id. at 151 n.3.

201. Id.

202. Id.

203. See id.

204. Id.

205. Id. at 151.

206. Id.

207. Id. at 151, 151 n.4.

208. Racial Commodification and the Promise of the New Functionalism, supra note

198, at 199 (citing Malone v. Haley, No. 88-339, slip op. at 16 (Mass. Sup. Jud. Ct.

Suffolk Cnty. July 25, 1989)). If the Malone brothers could not satisfy these criteria,

they could still prevail if they genuinely believed they were Black. See id. at 205.

209. Id.

210. Malone, 646 N.E.2d at 152 n.5.

211. Racial Commodification and the Promise of the New Functionalism, supra note

198, at 206.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 41 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 307

they were not Black, including (1) the brothers’ and their family mem-

bers’ birth certificates stating that they were all white, and (2) the

brothers’ self-identifications as white in their first job applications and

a subsequent application for a promotion to lieutenant.

212

Finally, the

officer found that the brothers were not considered part of the Black

community, pointing out that they had not joined the African-Ameri-

can firefighters group and that they did not hold themselves out to be

Black.

213

Compared to sexual orientation and religion adjudication, the level

of scrutiny exacted on the brothers’ racial identity claims was high.

Over a two-day hearing, the officer took into account ascriptive evi-

dence, multiple forms of documentary evidence, and ancestral evi-

dence.

214

She evaluated their physical appearances, various birth

certificates, and how they represented their race to others.

215

Identity

estoppel also played a role in this case—the hearing officer considered

their previous identifications as white as an indication that their Black

racial identity claim was false.

216

The Malone brothers’ claims that

they were Black, which may have been supported by some documen-

tary evidence, could not overcome their prior and subsequent incon-

sistent self-identifications, conflicting documentary evidence, and

their physical appearances.

217

More than 20 years after Malone, another racial identity claim was

interrogated with similar intensity. This case, Orion Insurance Group

v. Washington State Office of Minority & Women’s Business Enter-

prises, involved the Small Business Administration’s 8(a) program.

This program provides government contracting opportunities to small

businesses that are “owned and controlled by socially and economi-

cally disadvantaged individuals,” also known as Disadvantaged Busi-

ness Enterprises (“DBE”).

218

“Socially disadvantaged individuals are

those who have been subjected to racial or ethnic prejudice or cultural

bias” within American society “because of their identity as member of

a group without regard to their individual identity.”

219

The relevant

regulation provides that most racial minorities are presumptively “so-

cially disadvantaged individuals.”

220

212. Id. at 206–07.

213. Id. at 208–09.

214. Malone, 646 N.E.2d at 151.

215. Id. at 151–52.

216. Racial Commodification and the Promise of the New Functionalism, supra note

198, at 207.

217. Id. at 205–10.

218. Orion Ins. Grp. v. Wash. State Off. of Minority & Women’s Bus. Enters., No.

16-5582, 2017 WL 3387344, at *1 (W.D. Wash. Aug. 7, 2017), aff’d sub nom. Orion Ins.

Grp. v. Wash.’s Off. of Minority & Women’s Bus. Enters., 754 F. App’x 556 (9th Cir.

2018); 15 U.S.C. § 637(a)(1)(A)–(D).

219. 15 § 637(a)(5); see also 49 C.F.R. § 26, App. E (2020).

220. 49 C.F.R. § 26.67(a) (2020).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 42 10-MAR-22 14:18

308 TEXAS A&M LAW REVIEW [Vol. 9

Ralph Taylor sought a DBE designation for his company on the

basis that he was presumptively socially disadvantaged because he was

Black and Native American.

221

In his initial application, Taylor sub-

mitted the following evidence to support his asserted identity claims:

(1) a genetic ancestry test stating that he was 90% European, 6% In-

digenous American, and 4% Sub-Saharan African; (2) a copy of his

driver’s license with his picture on it; (3) his birth certificate stating

that his parents are “Caucasian” but not providing his race; (4) a letter

from his father asking that his birth certificate be changed to reflect

that he is “Caucasian, African, and American Indian;” (5) his father’s

genetic results, estimating that he was 44% European, 44% Sub-

Saharan African, and 12% East Asian; and (6) a death certificate from

a Virginia woman who was presumably related to Taylor identifying

her as a “Negro.”

222

The law does not require DNA or other ancestral

evidence to show membership in an identity category. Taylor presum-

ably provided this evidence because he looked white in his

photograph.

223

The relevant regulation permits the agency to ask for additional evi-

dence of an identity claim if the agency has a “well[-]founded reason

to question the individual’s claim of membership in that group.”

224

Based on Taylor’s appearance and his DNA test results, the agency

asked for additional evidence that Taylor was Black and/or Native

American.

225

In response, Taylor submitted the following: (1) evi-

dence connecting the Virginia woman to his mother’s side of the fam-

ily; (2) a birth certificate from a relative, born in 1914, whose father’s

race is listed as “white?”; (3) a statement that he “considered himself

to be Black based on his DNA test results, that he joined the NAACP,

subscribed to Ebony magazine, and has ‘taken a great interest in

Black social causes;’” and (4) letters from individuals stating that they

considered him a person of “mixed race” or “mixed heritage.”

226

When considering this additional evidence, the agency “must con-

sider whether the person has held himself out to be a member of the

group over a long period of time prior to application for certification

and whether the person is regarded as a member of the group by the

relevant community.”

227

The agency may also “require the applicant

to produce appropriate documentation of group membership.”

228

The

agency found that he was not Black because none of the evidence he

221. Orion Ins. Grp., 2017 WL 3387344, at *2.

222. Id.

223. Ex. Selection of Admin. Rec. at 1–2, Orion Ins. Grp. v. Wash. State Off. of

Minority & Women’s Bus. Enters., No. 16-5582, 2017 WL 3387344 (W.D. Wash. Aug.

7, 2017) (No. 16-5581), 2017 WL 3387344 [hereinafter Ex. Selection of Admin.].

224. 49 C.F.R. § 26.63(a)(1) (2020).

225. Orion Ins. Grp., 2017 WL 3387344, at *3.

226. Id. at *10–11.

227. 49 C.F.R. § 26.63(b) (2020).

228. Id.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 43 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 309

produced showed that he “ever held himself out to be Black, that he

ever was perceived in the business community as Black, or that he

ever suffered any adverse consequences because people with whom he

wished to engage in business refused to do business with him because

they perceived him as Black.”

229

Taylor appealed, and the district court affirmed, holding that the

agency did not act arbitrarily or capriciously when it found that Taylor

provided insufficient evidence that he was either Black or Native

American.

230

Interestingly, Taylor also applied for the state version of

the federal DBE program.

231

Washington State initially denied his ap-

plication, but after he appealed, the state voluntarily reversed its deci-

sion and classified Taylor’s business as a Minority Business Enterprise

(“MBE”).

232

Compared to the Malone brothers, Taylor provided more evidence

of his claim to Blackness, including ancestral evidence and evidence

that at least some individuals regarded him as multiracial.

233

But like

the Malone brothers, he did not ascriptively appear to be Black to the

adjudicator.

234

Also like the Malone brothers, he had previously iden-

tified as white (before receiving his DNA test results),

235

and in both

cases, the adjudicators took these prior inconsistent racial identifica-

tions as evidence that they were not Black.

236

That is, Taylor and the

Malone brothers were estopped from claiming a Black identity based

on prior inconsistent self-identifications.

Courts have invoked identity estoppel in other benefit cases involv-

ing racial identity claims. For instance, the United States settled a law-

suit against an employer based on the employer’s discrimination

against certain racial minorities, and the benefits of the settlement

were reserved for the racial minorities that the employer had discrimi-

nated against.

237

Membership in a racial minority for the purpose of

229. Ex. Selection of Admin., supra note 223, at 3–4.

230. Orion Ins. Grp., 2017 WL 3387344, at *8.

231. Id. at *2.

232. Id.

233. Compare id. at *3, with supra discussion accompanying notes 209–13 (describ-

ing the evidence presented by the Malone brothers).

234. Compare Orion Ins. Grp., 2017 WL 3387344, at *10, with supra discussion ac-

companying note 209–13 (describing the adjudicator’s findings regarding the Malone

brothers’ ascriptive appearances). The agency stated that Taylor appeared to be white

according to his driver’s license picture. Ex. Selection of Admin., supra note 223, at 1.

235. Id.; Orion Ins. Grp., 2017 WL 3387344, at *2 (stating that Taylor received his

DNA results on August 25, 2010, and subsequently applied for his company’s certifi-

cation as an MBE under Washington State law on April 19, 2013); see also Racial

Commodification and the Promise of the New Functionalism, supra note 198, at 207.

236. Orion Ins. Grp., 2017 WL 3387344, at *8; Malone v. Civ. Serv. Comm’n, 646

N.E.2d 150, 151 (Mass. App. Ct. 1995).

237. See United States v. N.Y.C. Bd. of Educ., 85 F. Supp. 2d 130, 133, 135–36

(E.D.N.Y. 2000) (summarizing the central terms of the parties’ Agreement, including

providing “permanent civil service status to 43 identified blacks, Hispanics, Asians,

and women” who had been serving as provisional employees and further providing

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 44 10-MAR-22 14:18

310 TEXAS A&M LAW REVIEW [Vol. 9

distributing the settlement funds was determined by how individuals

had identified their race or national origin on their job applications.

238

Individuals who marked “white” on their job applications were ex-

cluded from the settlement, and some of those individuals subse-

quently objected to their exclusion, claiming that they were not

white.

239

The employer reexamined whether these objectors were mis-

classified as white for purposes of the settlement and reclassified

some, but not all, of them.

240

The remaining objectors sought relief

from the court, but the court refused to second-guess the employer’s

racial classifications and held that their “self-identifications as white

[were] sufficient to exclude them from the benefits of the

settlement.”

241

The court appeared wary of engaging in any identity adjudication of

its own, stating that it would be “unnecessary, and would be extremely

unwise, for this court to . . . attempt to engage in the dubious task of

defining or categorizing any person’s race or national origin.”

242

How-

ever, by allowing prior inconsistent racial self-identification to estop

these objectors from asserting a different identity claim, the court was

adjudicating identity by assuming that the racial identity asserted on

the employees’ application was accurate for the purposes of this

settlement.

In sum, racial identity claims can be highly scrutinized in the con-

text of a benefit law—especially when someone’s self-identification

does not align with other identity evidence (like appearance or prior

inconsistent self-identifications). Prior inconsistent racial identifica-

tion is also salient in racial adjudication and may estop individuals

from making a successful racial identity claim.

4. Sex

This Subsection examines two legal contexts in which the law deter-

mines eligibility for something or access to something based on

whether someone is sufficiently male or female.

243

Those contexts are

(1) being able to change a sex marker on a birth certificate and (2)

“retroactive seniority, including retroactive pension relief, to 54 identified black, His-

panic, Asian, and female incumbent” employees (emphasis in original)).

238. Id. at 153.

239. Id.

240. See id. The filings do not explain what criteria the employer used to reclassify

the employees.

241. Id.

242. Id.

243. I recognize that using the term “benefit law” in the context of sex adjudication

feels like a misnomer more so than with other identities. Not allowing someone to

change their sex designation because of their sex seems more like discrimination than

a benefit. These laws fall under this category because, in contrast to anti-discrimina-

tion laws, these laws are not about addressing prior instances of discrimination;

rather, they are forward looking and require proof of “sex” for the government to

take a certain action.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 45 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 311

participating in sex-segregated sports.

244

For someone to prove their

sex in these contexts, the burden of proof is usually relatively high.

Self-identification is not enough, and people generally need to show

biological and/or documentary evidence of their sex. Sex adjudication

in benefit laws, like with anti-discrimination laws, is most visible in

cases involving transgender people because transgender people are

more likely to need the state to adjudicate their sex.

Laws regulating changes to sex markers on birth certificates may

seem more like data-collection laws than benefit laws, at least at first

blush. Unlike data-collection laws, birth certificates are a primary le-

gal mechanism through which the state exercises its power to assign a

sex classification, and this legal sex designation is extremely conse-

quential for individuals.

245

A birth certificate is often used to obtain

most other identity documents, including driver’s licenses and pass-

ports.

246

These identity documents facilitate access to a wide variety of

institutions and programs, including schools, insurance, financial insti-

tutions, air travel, and much more.

247

Thus, when someone’s sex

marker on these documents is incorrect, their ability to access the ser-

vices they need is likely to be impaired. So when the state regulates

changing sex markers on birth certificates, the state is not collecting

data about sex—it is adjudicating whether someone is sufficiently

“male” or “female” for the state to recognize them as such and grant

them access to a correct birth certificate.

248

States regulate birth certificates, and many states require that an

individual modify their anatomy in some way to amend their sex

244. Other legal contexts in which sex is adjudicated include determining who is

sufficiently “male” or “female” to access a particular bathroom. See, e.g., Grimm v.

Gloucester Cnty. Sch. Bd., 972 F.3d 586, 619 (4th Cir. 2020) (finding that Grimm was

a “boy” and could use the boy’s restroom). As discussed, addressing all laws that

adjudicate identity is beyond the scope of this Article, and future work can explore

these areas.

245. On a more abstract level, a sex designation is required for personhood. Some-

one cannot operate in current society without a legal sex classification. As Judith But-

ler has posited, “the moment in which an infant becomes humanized is when the

question, ‘is it a boy or a girl?’ is answered.”

ODIES

HAT

ATTER

, supra note 122,

at 111.

246. Lisa Mottet, Modernizing State Vital Statistics Statutes and Policies to Ensure

Accurate Gender Markers on Birth Certificates: A Good Government Approach to

Recognizing the Lives of Transgender People, 19

ICH

. J. G

ENDER

& L.

373, 391

(2013).

247. See id. at 392 (“Birth certificates establish initial gender designation for other

government identify documents, such as driver’s licenses, passports and Social Secur-

ity records. The birth certificate, as well as these other government documents, in turn

breed many other identity documents, such as school records, college ID cards, work

identification, and commercial licenses.”).

248. See Katyal, supra note 58, at 412 (describing how “being ‘sexed’ or classified

by the state . . . confer[s] the benefits of recognition on individuals who fit the mor-

phological model and den[ies] certain entitlements, particularly recognition, to those

who transgress or who do not fulfill the regulatory requirements for transition”).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 46 10-MAR-22 14:18

312 TEXAS A&M LAW REVIEW [Vol. 9

marker.

249

Lisa Mottet surveyed all 50 states in 2013, and at the time,

27 states required surgery by statute, regulation, or policy.

250

Other

states do not have a codified policy that regulates sex marker changes,

but in practice, individuals must submit proof of some form of sex

reassignment surgery.

251

More recent 50-state surveys confirm that

about half of states require some form of surgery to change sex mark-

ers on birth certificates.

252

A handful of states have broader language

but still require proof of some clinical treatment.

253

Vermont, for in-

stance, requires that the individual has “undergone surgical, hormo-

nal, or other treatment appropriate for that individual for the purpose

of gender transition.”

254

For some states that require proof of surgery, not just any surgery

will do. According to Mottet’s research, some agencies “have strict

(generally unwritten) rules that a particular surgery must be

shown.”

255

And in other jurisdictions, a judge determines what type of

surgery is required.

256

Individuals who have had surgeries that change

their genitalia are more likely to successfully change their sex

marker.

257

For instance, approximately 80% of trans men who had

surgery to create a penis received a corrected birth certificate, com-

pared with 56% who just had top surgery.

258

Similarly, 74% of trans

women who had a vaginoplasty or had their scrotum and testes re-

moved received a corrected birth certificate, whereas 32% of those

who had only breast surgery received a corrected birth certificate.

259

These statutes adjudicating sex for birth certificates can be under-

stood as extreme examples of the law’s use of identity estoppel. Prior

identification as one sex (or perhaps more accurately, assignment at

birth to one sex by a third party) makes it very difficult to claim a

different sex.

260

If someone claiming to be a man was assigned female

249. This Article focuses on birth certificates, as opposed to other documents, be-

cause a birth certificate is required to obtain most other identity document (driver’s

licenses, Social Security cards, passports) or change sex markers on those identity

documents. Mottet, supra note 245, at 391–93.

250. Id. at 400–01.

251. Id. at 401.

252. See Changing Birth Certificate Sex Designations: State-by-State Guidelines,

AMBDA

EGAL

https://www.lambdalegal.org/know-your-rights/article/trans-chang-

ing-birth-certificate-sex-designations (Sept. 17, 2018) [https://perma.cc/4U4F-6GNH]

[hereinafter

AMBDA

EGAL

253. Mottet, supra note 246, at 400.

254.

. S

TAT

. A

tit. 18, § 5112(b) (West, Westlaw through Acts 1 through 76

(end) & M-1 through M-6 (end) of Reg. Sess. & Act 1 (end) of Special Sess. of

2021–2022 Vt. Gen. Assemb. (2021)).

255. Mottet, supra note 246, at 401.

256. Id.

257. See id. at 390 nn. 63–65.

258. Id. at 390 n.61.

259. Id.

260. See, e.g., Sandy E. James et al., The Report of the 2015 U.S. Transgender Sur-

vey,

TRANSEQUALITY

ORG

86 (Dec. 2016), https://transequality.org/sites/default/files/

docs/usts/USTS-Full-Report-Dec17.pdf [https://perma.cc/4K8X-FRJ3] (“More than

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 47 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 313

at birth and previously identified as a girl or woman, these laws estop

him from claiming a male sex identity, unless he makes substantial

changes to his body. Unlike racial identification, where prior inconsis-

tent identification was just one of many factors that adjudicators con-

sidered,

261

prior inconsistent sex identifications make claiming a

different sex identity in this context very costly.

262

The second context examined in this Subsection is the adjudication

of sex for sex-segregated sports. How sex is determined for sports var-

ies widely across the country, depending on the sport, level (youth/

collegiate/professional), and jurisdiction. Thus, for scope purposes, I

focus on two examples: one that is considered more inclusive of trans-

gender athletes (the NCAA rule) and one that essentially bans trans-

gender women and girls from participating in sex-segregated sports

(an Idaho law).

263

From 2011 until the end of January 2022, the

NCAA rule deemed a transgender woman sufficiently “female” to

participate on a women’s team if she had “complet[ed] one calendar

year of testosterone suppression treatment.”

264

A transgender man

was able to participate on a men’s team, but not a women’s team, if

he was taking testosterone.

265

If a transgender athlete was not taking

hormones, their sex was determined based on their sex assigned at

birth.

266

While the presence or absence of hormones, and testosterone

in particular, was key to sex determinations for the NCAA, hormones

were not the only thing at play.

267

The NCAA’s policy only addressed

the testosterone levels of transgender athletes—it did not require cis-

gender women to have their testosterone levels checked to join a wo-

two-thirds (68%) of respondents did not have any ID or record that reflected both the

name and gender they preferred.”).

261. See, e.g., Malone v. Civ. Serv. Comm’n, 646 N.E.2d 150, 151, 155 (Mass. App.

Ct. 1995).

262. See, e.g., Megan Brodie Maier, Altering Gender Markers on Government Iden-

tity Documents: Unpredictable, Burdensome, and Oppressive, 23

U. P

. J.L. & S

HANGE

203, 229 (2020) (summarizing barriers to changing documentation due to

cost).

263. Unlike other identity adjudicators in this Article, the NCAA is not a state

actor. Tarkanian v. NCAA, 488 U.S. 179, 196-97 (1988). Nevertheless, I chose to ex-

amine the NCAA’s rule for two reasons. First, the NCAA rule applies nationwide;

how state actors adjudicate sex in other contexts, like K–12 sex-segregated sports, for

instance, is dependent on the particular school district and thus not applicable to as

many athletes as the NCAA rule. In a forthcoming article, legal scholar Scott Skinner-

Thompson provides a detailed analysis of how sex is regulated in K–12 sports in each

state and each state’s largest school district. See Scott Skinner-Thompson, Identity by

Committee (manuscript in progress) (on file with author). Second, the NCAA’s pol-

icy, before its recent amendment, was considered one of the most trans-inclusive poli-

cies for sex-segregated sports; therefore, the forthcoming critiques of this policy

would apply with at least equal force to most other policies.

264.

RIFFIN

& H

ELEN

ARROLL

NCAA, NCAA I

NCLUSION OF

RANS-

GENDER

TUDENT

-A

THLETES

13 (2011), https://www.ncaapublications.com/

DownloadPublication.aspx?download=11INCL.pdf [https://perma.cc/P5TL-67FP].

265. Id.

266. Id.

267. See id.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 48 10-MAR-22 14:18

314 TEXAS A&M LAW REVIEW [Vol. 9

men’s team.

268

The only athletes who had their sex adjudicated were

transgender athletes whose sex assigned at birth was different from

the sex of the team they wish to play on.

269

Thus, sex adjudication in

this context was dependent on two things: (1) transgender status, and

(2) presence or absence of testosterone.

On January 19, 2022, the NCAA changed its rule and adopted a

sport-by-sport approach to sex adjudication.

270

Specifically, “the up-

dated NCAA policy calls for transgender participation for each sport

to be determined by the policy for the national governing body

[(“NGB”)] of that sport.”

271

If there is no “NGB policy for a sport,

that sport’s international federation policy [is] followed,” and “[i]f

there is no international federation policy, previously established IOC

policy criteria [are] followed.”

272

Some NGBs do not currently have

rules addressing requirements for participation in sex-segregated

sports and will likely be adopting such policies in the coming weeks

and months.

273

Although this rule change has created much uncertainty regarding

sex determinations in collegiate sports, there is reason to believe that

sport-specific rules will mirror the NCAA’s former rule and use tes-

tosterone levels to determine whether an athlete is sufficiently “fe-

male” to participate in women’s sports. First, many NGBs that

currently have applicable rules use testosterone levels to determine

eligibility for women’s sports.

274

Additionally, the NCAA seems to

anticipate that testosterone levels will continue to play a major role;

its press release stated that “[t]ransgender student-athletes will need

to document sport-specific testosterone levels beginning four weeks

before their sport’s championship selections.”

275

While the NCAA’s rule will likely continue to rely mostly on testos-

terone levels, the “Fairness in Women’s Sports Act” Idaho recently

passed imposes a far more rigorous test for sex determination in

268. See id.

269. See id. at 13, 22.

270. Press Release, NCAA Bd. of Governors, Board of Governors Updates Trans-

gender Participation Policy (Jan. 19, 2022, 8:41 PM), https://www.ncaa.org/news/2022/

1/19/media-center-board-of-governors-updates-transgender-participation-policy.aspx

[https://perma.cc/RLM3-P754] [hereinafter NCAA Updated Policy].

271. Id.

272. Id.

273. For example, USA Swimming, the NGB for college swimming, released a

statement after the NCAA rule change stating that although there is currently no

policy on the books regarding transgender athletes, it is working with the Interna-

tional Swimming Federation to adopt one and expects to release it soon. Sarah

Berman, USA Swimming Releases Statement on Transgender Athlete Inclusion,

WIM-

WAM

(Jan. 20, 2022), https://swimswam.com/usa-swimming-releases-statement-on-

transgender-athlete-inclusion/ [https://perma.cc/28FR-CXG7].

274. Policies by Organization,

TRANSATHLETE

COM

, https://www.transathlete.com/

policies-by-organization [https://perma.cc/WR24-G5ZZ] (listing rules for many NGBs

and other organizations).

275. NCAA Updated Policy, supra note 270.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 49 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 315

sports. The Act bars “students of the male sex” from participating in

women’s and girls’ sports and limits participation to students whose

“biological sex” is female.

276

Its scope is extremely broad and applies

to all women/girl-only sports at most ages and levels, including intra-

mural teams and teams sponsored by “primary schools.”

277

The Act

permits an unnamed set of individuals to “dispute” the “biological

sex” of any student on an all-women’s team. A student whose sex is

challenged can prove their “biological sex” by providing documenta-

tion “signed by [their] personal health care provider” verifying “the

student’s reproductive anatomy, genetic makeup, or normal endoge-

nously produced testosterone levels” (i.e. testosterone naturally pro-

duced by the body and not through hormone treatment).

278

That is,

the Act excludes from the category of “female” any students (1) who

don’t have a vagina, uterus, ovaries (and other sex characteristics as-

sociated with a “female” sex), (2) whose chromosomes are not XX,

and (3) whose testosterone levels are naturally below a certain level,

regardless of whether they are taking testosterone-suppressing

hormones.

279

The Act essentially excludes intersex individuals and transgender

women and girls from participating in girl’s/women’s sports. First,

most transgender students do not have the “reproductive anatomy”

associated with a “female sex” because they have not had genital sur-

gery; even with surgery, transgender girls and women won’t have ova-

ries or a uterus. Second, transgender girls and women cannot change

their chromosomes. And third, by focusing on endogenously produced

hormones, the Act bars transgender athletes who are taking testoster-

one suppressants and therefore have circulating testosterone levels

like cisgender women. A federal district court has enjoined the Act on

the grounds that it likely violates both transgender and cisgender wo-

men’s rights under the Equal Protection Clause.

280

The case is cur-

rently pending before the Ninth Circuit.

281

Idaho is not alone in its

desire to exclude transgender students from single-sex sports and to

impose such strict and invasive methods for sex determinations; many

other states are currently considering similar legislation.

282

276.

DAHO

ODE

§ 33-6203(2).

277. Id. § 33-6203(1).

278. Id. § 33-6203(3).

279. See id.

280. Hecox v. Little, 479 F. Supp. 3d 930, 989 (D. Idaho 2020).

281. See Brief for Appellees Lindsay Hecox and Jane Doe, Hecox v. Little, 479 F.

Supp. 3d 930 (D. Idaho 2020) (Nos. 20-35813, 20-35815), 2020 WL 7634923, at *1.

282. See, e.g., S.B. 1046, 55th Leg., 2d Reg. Sess. (Ariz. 2022) (determining “sex”

using the same criteria as the Idaho law); see also Take Action!: 2022 Sports Bills,

TRANSATHLETE

COM

(Jan. 27, 2022), https://www.transathlete.com/take-action [https://

perma.cc/TLA3-NMNW] (listing similar bills in other states). The South Dakota legis-

lature passed a very similar bill on February 1, 2022, and the governor could sign it

into law any day. Morgan Matzen, South Dakota House Passes Bills Limiting Trans-

gender Students’ Access to Sports, Bathrooms,

MSN

(Feb. 1, 2022), https://

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 50 10-MAR-22 14:18

316 TEXAS A&M LAW REVIEW [Vol. 9

Thus, sex adjudication in the context of sports, like with birth certif-

icates, is determined based on biological evidence. Some rules adjudi-

cate “female” sex based on current hormone levels, (like the NCAA’s

former policy) while others involve much more searching inquires that

include examination of genitals, chromosomes, and hormones, which

essentially limit the category of “female” to cisgender women.

D. Possible Explanations

The previous discussion showed how both the type of law and the

mutability of the identity category affect how closely identity claims

are scrutinized. It also exposed significant inconsistency in identity ad-

judication both across and within different areas of law. Yet courts

and other legal actors do not address or explain the principles they

employ when adjudicating identity or acknowledge these inconsisten-

cies. Indeed, legal actors may not even be self-consciously aware that

they are adjudicating identity. This Section provides some possible

reasons why the level of interrogation into identity claims seems to be

correlated with the type of law and the mutability of the identity.

One reason legal actors interrogate identity differently based on the

type of law at issue may have something to do with balancing (1) a

perceived incentive to misrepresent an identity with (2) the cost to the

state of getting it wrong and (3) the cost of acquiring additional evi-

dence. This framework does a good job explaining the different levels

of scrutiny in data-collection laws versus some benefit laws. For data-

collection laws, individuals typically have little incentive to misrepre-

sent their identity. For instance, someone’s decision to check “His-

panic” or “non-Hispanic” on the Census does not have any direct

effect on that individual.

283

While that decision may affect how re-

sources are distributed to the community at large, the link between

the individual choice and the effect on the group is attenuated.

284

for the cost of error to the state, even if a small percentage of respon-

dents did misstate their identity, these misstatements would not make

much of a difference in the aggregate.

285

Combine this low incentive

to misrepresent and the low cost of misstatements with the massive

administrative burden of confirming large numbers of identity claims,

and data collection’s reliance on self-identification makes sense.

For many benefit laws—like asylum and affirmative action—iden-

tity adjudicators may think the incentives to “lie” about an identity

www.msn.com/en-us/news/politics/south-dakota-house-passes-bills-limiting-trans-

gender-students-access-to-sports-bathrooms/ar-AATnlVm [https://perma.cc/EWD2-

9NRT].

283. See Mezey, supra note 63, at 1745.

284. See N

’

SCH

. C

OUNCIL

, R

ESEARCH AND

LANS FOR

OVERAGE

EA-

SUREMENT IN THE

2010 C

ENSUS

: I

NTERIM

SSESSMENT

2-2

(Robert Bell & Michael

Cohen eds., 2007).

285. Id. at 2-3.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 51 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 317

are high.

286

Avoiding deportation or securing a government contract

are generally understood to be desirable, and unlike data-collection

laws, the identity claim does have a direct, and sometimes substantial,

effect on the individual.

287

The cost to the government of getting it

wrong can also be higher than data-collection laws, but not always. If

a limited resource is involved (like a limited number of government

contracts), getting the identity claim wrong may mean that someone

who should have access to that resource, will not, which undermines

the benefit law’s function. Moreover, the administrative burdens of

adjudicating identity claims are lower than with data-collection laws

because of the reduced number of identity claims involved.

But for anti-discrimination laws, incentives to misrepresent an iden-

tity, the harms of error, and/or administrative burdens fail to ade-

quately explain why courts interrogate these identity claims more

closely than data-collection identity claims. There is no apparent in-

centive to misstate an identity, like with benefit laws.

288

The anti-dis-

crimination laws in this Article are symmetrical, meaning that anyone

of any race, sex, religion, or sexuality can bring a claim, so plaintiffs

need not misstate their identity to bring a claim. For this same reason,

the cost of a court incorrectly adjudicating an identity claim is not

high. And like benefit laws, the costs of additional evidence are fairly

low.

Some scholars have hypothesized that legal actors (incorrectly)

think that the costs of getting an identity claim wrong is high—that is,

some judges may want to protect only those plaintiffs who are “actu-

ally” minorities.

289

Under this view, anti-discrimination laws are

meant to protect minorities, so only those plaintiffs who are “actually”

members of the group the law intended to benefit should be allowed

to bring those claims. Consequently, judges with this perspective may

be more likely to interrogate an identity claim closely to ensure that

the law is being used to address discrimination against subordinated

or minority identities. I explain in the next Section why this concern is

unjustified both doctrinally and normatively, but this type of “pro-

tected class gatekeeping”

290

may explain, at least partially, why iden-

286. See Kazemzadeh v. U.S. Att’y Gen., 577 F.3d 1341, 1345 (11th Cir. 2009);

Tseming Yang, Choice and Fraud in Racial Identification: The Dilemma of Policing

Race in Affirmative Action, the Census, and a Color-Blind Society, 11

ICH

. J. R

ACE

& L.

367, 387 (2006).

287. Not all benefit laws come with such incentives, like laws governing sex mark-

ers on birth certificates. See Paisley Currah & Lisa Jean Moore, “We Won’t Know

Who You Are”: Contesting Sex Designations in New York City Birth Certificates, 24

YPATIA

113, 114 (2009).

288. See supra Part III(C).

289. See Protected Class, supra note 27, at 156–57 (“It is perhaps this anti-subordi-

nation instinct, generally unstated, that motivates protected class gatekeeping, and

leads progressive jurists to continue to include language inviting protected class

gatekeeping in proposals to reform discrimination doctrine.”).

290. I borrow this term from Jessica Clarke. See id. at 101.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 52 10-MAR-22 14:18

318 TEXAS A&M LAW REVIEW [Vol. 9

tity is adjudicated more intensely in anti-discrimination laws than

data-collection laws.

Why mutability of identity affects how closely an identity is scruti-

nized evades easy explanation. But the perception that immutable

identities remain stable over time and are easily ascertainable based

on objective traits may help explain some patterns from the taxon-

omy.

291

First, if legal actors understand immutable identities to be sta-

ble over time, they may be more likely to view inconsistent self-

identification as an indication of fraud rather than a legitimate change

in identity. In other words, if legal actors believe an identity cannot be

changed, and the individual previously said they were X, and now they

say they are Y, legal actors may think the individual is lying. This may

explain why identity estoppel tends to be more prevalent in the con-

text of race and sex adjudication than religion or sexual orientation.

Consider the following example. Anna is a multiracial individual who

converted to Judaism from Christianity after getting married. Anna

marked her race as “white” when she applied for a job. Two years

later, her employer starts an affirmative action program for racial mi-

norities, and Anna applies, stating that she self-identifies as multira-

cial. Her employer deems her ineligible because her prior self-

identification as white estops her from identifying as non-white, as

race is perceived to be immutable. This time, Anna sues her employer

for discriminating against her based on her Jewish identity. Her prior

self-identification as a Christian is not fatal to her assertion that she is

Jewish because religious identity is mutable, and therefore, her cur-

rent assertion is not seen as fraudulent.

Indeed, fraud narratives about race and sex are both prevalent and

historically rooted. As for race, there is increasing societal concern

about college applicants identifying as minorities, even if that claim is

tenuous or false,

292

or declining to indicate a racial identity,

293

hopes of benefiting from the university’s affirmative action policy.

And historically, racial minorities have claimed a white identity to

avoid enslavement, become a citizen, or gain access to the numerous

other benefits that came with a legal declaration of whiteness.

294

for sex, transgender individuals have long been deemed “gender

frauds.”

295

For example, the “transsexual panic” defense to murdering

291. See, e.g., John Tehranian, Changing Race: Fluidity, Immutability, and the

Evolution of Equal-Protection Jurisprudence, 22

U. P

. J. C

ONST

. L.

1, 7 (2019)

(showing how equal protection doctrine assumes race is immutable when, in fact, ra-

cial identity is “malleable and can change” due to a variety of factors).

292. Yang, supra note 286, at 369. Yang finds that this “appears to be most preva-

lent with respect to American Indian and Hispanic/Latino identities.” Id.

293. See, e.g., Camille Gear Rich, Decline to State: Diversity Talk and the American

Law Student,

S. C

. R

. L. & S

. J

UST

539, 540 (2009) [hereinafter Decline to

State] (addressing law school applicants specifically).

294. See

OPEZ

, supra note 26, at 3, 27, 38, 44.

295. Currah & Moore, supra note 287, at 128.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 53 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 319

a transgender woman relied on the idea that the perpetrator was de-

frauded into thinking the victim was a woman based on her physical

appearance and then “justifiably” reacted violently when he discov-

ered that she was “really” a man.

296

More recently, opponents of laws

enabling transgender youth to participate in sex-segregated sports and

to use their chosen restroom have cited a concern that “boys” might

fraudulently claim that they are “girls” so that they can go into girls’

bathrooms or join girls’ sports teams.

297

Although some of these socie-

tal narratives around race and sex fraud are problematic, they are sali-

ent and may help explain legal actors’ willingness to view inconsistent

identification as fraudulent.

Second, if legal actors think immutable identities are easily ascer-

tainable through visual observation, or other “objective” evidence,

they might not be able to understand how someone could misperceive

another person’s identity.

298

This may explain why courts are more

willing to reject discrimination claims involving racial misperception

claims than sexual orientation or religious misperception claims. D.

Wendy Greene has made this argument—that this fixed view of racial

identity may lead legal actors to reject a misperception claim because

they “may simply be unable to rationalize the proposition that a plain-

tiff’s race was mistaken” or “might outright disbelieve that a plaintiff’s

race could be mistaken, based on her own preset conceptualizations of

race and sex as fixed, indisputable constructs.”

299

Third, and relatedly, if immutable identities are defined by “objec-

tive” traits, legal actors may interrogate these identities more closely

to ensure that an individual meets the legal actor’s “objective” criteria

for a particular identity. So if the law understands “objective” evi-

dence of a “woman” or a “female” to include breasts and a vagina,

people without those traits will not be able to, for example, change

their sex marker on their birth certificate. This may also shed some

light on why some courts adjudicate racial identity claims based on so-

called “objective” criteria but don’t do the same for the more mutable

identities, religion and sexual orientation, because unlike race, courts

may not have “objective” criteria that an individual must meet.

296. Id. at 120.

297. See e.g., Grimm v. Gloucester Cnty. Sch. Bd., 972 F.3d 586, 599 (4th Cir. 2020)

(noting community members’ justifications for opposing a student’s ability to use the

restroom matching their gender identity).

298. See Elective Race, supra note 61, at 1506 (“Because we are in a period of tran-

sition, many Americans still are wedded to fairly traditional attitudes about race. For

these Americans, race is still an objective, easily ascertainable fact determined by the

process of involuntary racial ascription—how one’s physical traits are racially catego-

rized by third parties.”).

299. Greene, supra note 17, at 135.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 54 10-MAR-22 14:18

320 TEXAS A&M LAW REVIEW [Vol. 9

IV. C

ONTEXT

-D

ETACHED

& C

ONTEXT

-I

NFORMED

DJUDICATION

This Part identifies a problematic pattern across the taxonomy of

identity adjudication. Specifically, it argues that legal actors tend to

adjudicate identity claims without considering why the identity is rele-

vant to the law at issue. I call this a context-detached approach to

identity adjudication. A legal actor employing a context-detached ap-

proach does not adequately consider whether someone’s identity mat-

ters to the law in the first place. Nor does this legal actor use the

purpose of the law to inform what types of identity should be used to

adjudicate identity. A context-informed approach to identity adjudica-

tion, on the other hand, understands the identity question to depend

on why that particular law is asking the identity question in the first

place. It identifies the function or purpose of the applicable law first.

It then asks whether identity adjudication is necessary at all, and if so,

what definition or model of identity would best serve those purposes.

This Part addresses each type of law in turn and shows how legal

actors often fail to adjudicate identity with proper attention to the

function or purpose of the law. It also discusses what a context-in-

formed approach might look like and summarizes the harms of con-

text-detached adjudication, which tend to fall in two categories. First,

context-detached adjudication produces outcomes that are at odds

with the purpose of the law and often internally inconsistent with each

other. Second, context-detached adjudication can cause over-interro-

gation of identity, where legal actors examine more identity evidence

than necessary to fulfill the purpose of the law. Over-interrogation of

identity can harm individuals’ privacy and dignity interests and may

reinforce problematic definitions of identity through the power of the

law.

A. Data-Collection Laws

Data-collection laws almost exclusively rely on self-identification to

adjudicate identity. Yet as discussed, self-identification alone might

not be the right type of identity evidence, depending on the purpose

of the law at issue. If identity data is being collected to track discrimi-

nation, for instance, asking someone how they self-identify will only

track discrimination if their self-identification aligns with their ascrip-

tive identity, or how their identity is understood or perceived in the

relevant community. A self-identified straight woman may be mis-

perceived as a lesbian based on her appearance or mannerisms and

then discriminated against on that basis. The data-collection mecha-

nism would not necessarily track this type of discrimination, however,

because she self-identified as straight. Thus, the current approach to

adjudicating identity for data-collection mechanisms used to track dis-

crimination fails to adequately consider the purpose of the law.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 55 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 321

What might a context-informed approach to identity adjudication

for a data-collection mechanism meant to identify discrimination look

like? One option is to revert to older models where third parties as-

signed someone an identity based their appearance and/or identity

performance. This option has the advantage of relying on ascriptive

identity, which may be able to capture patterns of discrimination more

reliably than self-identification. But reverting to a system where third

parties assign identities to individuals does not seem like a desirable

or practical solution.

300

First, this would not work for all identities—

for instance, how would a Census worker assign someone a sexual

orientation based on ascriptive evidence? Second, the identity would

be based on one person’s subjective determination, which may not re-

flect the identity the individuals is usually ascribed, and in turn, may

not be more accurate or reliable than self-identification. Third, and

most relevant to this discussion, third-party ascription does not neces-

sarily align with the purpose of the law either. One reason federal

agencies abandoned third-party ascription in favor self-identification

was “[r]espect for individual dignity.”

301

Dignity interests are not

served when individuals are involuntarily assigned a race, sex, sexual-

ity, or religion.

A context-informed approach to identity adjudication would need

to reflect the anti-discrimination function of the law and the dignitary

interests the law seeks to protect. One approach that aligns with both

of these purposes is to change the substance of the questions asked to

reflect ascriptive identity. Rather than (or in addition to) asking peo-

ple how they identify, ask people how others identify them. That is,

ask individuals how people in their community view their race, sex,

sexuality, etc. The question is not “what is your sexual orientation,”

for example, but rather, something like “what does your boss (or co-

workers or community, depending on the context) understand your

sexual orientation to be.”

302

This method of identity adjudication can

track discrimination patterns better than self-identification while

avoiding the dignity harms of involuntary identity ascription.

Data-collection mechanisms that have multiple purposes may need

to ask a variety of questions about someone’s identity to align with

300. See Mezey, supra note 63, at 1753 (explaining how very few racial minorities

want to return to a system where Census workers collect identity data through racial

ascription).

301. Revisions to the Standards for the Classification of Federal Data on Race and

Ethnicity, 62 Fed. Reg. 58,782, 58,788–90 (Oct. 30, 1997). The Office of Management

and Budget (“OMB”) issued a directive to federal agencies stating that racial-data-

collection efforts should facilitate self-identification to the greatest extent possible.

See id.; see also Elective Race, supra note 61, at 1505–06 (discussing how institutions

have increasingly valued dignity, privacy, and autonomy interests and how those in-

terests are best served by a model that uses self-identification to collect identity data).

302. See

AVID

A. H

OLLINGER

OSTETHNIC

MERICA

179–82 (2006) (proposing

similar amendments to the census in the context of racial identity); Sudeall, supra

note 63, at 1296.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 56 10-MAR-22 14:18

322 TEXAS A&M LAW REVIEW [Vol. 9

those various purposes. They could ask questions that seek informa-

tion about ascriptive identity, self-identification, ancestry, or prior

self-identifications, depending on the purposes for which the data is

being used.

Thus, data-collection laws’ exclusive reliance on self-identification

reflects a context-detached approach. Self-identification is not always

the best measure of identity because it doesn’t always align with the

identity they are generally ascribed by others. Other types of identity

evidence may also conflict with their self-identification—someone’s

“biological” sex might not match their self-identification or their as-

criptive sex, or someone’s ancestry may not influence their racial self-

identification or their ascriptive race. These various forms of identity

evidence may be more or less useful depending on why the identity

data is being collected.

B. Anti-Discrimination Laws

This Section argues that courts adopt a context-detached approach

to identity adjudication when they focus on determining the “actual”

identity of the plaintiff. Determining a plaintiff’s “actual” identity is

not relevant under black letter anti-discrimination law or the two pre-

vailing theories of anti-discrimination law (anti-classification and anti-

subordination). Rather, this Section argues that the focus of analysis

should be the content of the discrimination and the perspective of the

discriminator. First, I argue that neither black letter anti-discrimina-

tion law nor the two prevailing normative theories of anti-discrimina-

tion law call for courts to determine “actual” identity. Then, I turn to

the harms of context-detached identity adjudication and show how

this approach can violate privacy interests and needlessly reinforce

problematic stereotypes about certain identities.

1. Identity Adjudication and Black Letter Law

Determining a plaintiff’s “actual” identity is not required under the

anti-discrimination laws discussed in this Article for a fairly straight-

forward reason—Title VII, and similar state laws, are symmetrical,

meaning that they protect all races, religions, sexes, etc., not just one

sex or one race.

303

So regardless of whether someone is Black or

white, Muslim or Christian, or male or female, they are still protected

under Title VII because they have a race, sex, or religious identity. If

Title VII were asymmetrical, like the Age Discrimination in Employ-

ment Act (“ADEA”), which only applies to people over 40,

304

deter-

mining the plaintiff’s true identity (age in this case) might be

303. For a fuller discussion regarding the distinction between symmetrical and

asymmetrical laws, see Naomi Schoenbaum, The Case for Symmetry in Antidis-

crimination Law, 2017

. L. R

69, 69 (2017) [hereinafter Case for Symmetry].

304. 29 U.S.C. § 631(a).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 57 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 323

necessary. But Title VII and similar state laws apply to everyone, re-

gardless of their specific identity. Nevertheless, some courts mistak-

enly construe anti-discrimination case law and statutes to require a

determination of the plaintiff’s identity. The following discussion sets

forth different doctrinal rationales courts offer to justify their reliance

on “actual” identity and then explains why they are flawed.

First, Title VII prohibits discrimination “because of such individ-

ual’s race, color, religion, sex, or national origin,” and some courts

interpret this language to require a determination of the individual’s

identity.

305

But this language does not mean that the court needs to

figure out if a plaintiff is Black or white, Muslim or Christian, or male

or female. Rather, this language indicates that Title VII protects cer-

tain identity categories and is focused on the treatment of individuals,

not groups. It does not require courts to engage in identity fact-

finding.

306

Another frequent justification is the first prong of the McDonell

Douglas framework for analyzing disparate treatment cases—namely

that the plaintiff must be a member in a protected class or group.

307

the McDonell Douglas case itself, this meant that the plaintiff needed

to show “that he belongs to a racial minority.”

308

But in setting forth

this framework, the Supreme Court explicitly stated that this frame-

work was not required in all disparate treatment cases and that Title

VII protected all members of identities categories.

309

Lower courts

therefore misapply Supreme Court precedent when they construe this

prong “as an absolute bar against misperception discrimination

claims.”

310

Moreover, this interpretation of McDonell Douglas con-

flicts with the symmetrical nature of Title VII.

Some courts that dismiss misperception cases base their reasoning

on the fact that another anti-discrimination statute—namely, the

Americans with Disabilities Act (“ADA”)—protects individuals “re-

305. 42 U.S.C. § 2000e-2(a)(2) (emphasis added).

306. Moreover, as Jessica Clarke argues, in cases where the discrimination is based

on a misperceived identity, such discriminatory behavior generally still flows from

“some protected trait of the plaintiff, such as skin color, or stereotypes about that

trait” and is “generally inextricable from those plaintiffs’ own racial, sexual, or other

protected identities.” Protected Class, supra note 27, at, 113. For example, when a

plaintiff is discriminated against based on the mistaken belief that she is Black, that

discrimination is usually based on her skin color or some other racially correlated

trait. The statutory language, therefore, does not require a determination of the plain-

tiff’s actual identity.

307. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 (1973) (holding that

the plaintiff, who identified himself as Black, established the initial element of a prima

facie case of intentional race discrimination by demonstrating that he belonged to a

“racial minority” group).

308. Id. at 802.

309. Id. at 802 n.13.

310. For other scholars who have made similar points, see Greene, supra note 17, at

119–120; see also Protected Class, supra note 27, at 117–19; Eisenstadt, supra note 27,

at 807–08; Judicial Erasure, supra note 27, at 548.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 58 10-MAR-22 14:18

324 TEXAS A&M LAW REVIEW [Vol. 9

garded as” disabled, not just disabled individuals.

311

Under this rea-

soning, because Title VII does not include the “regarded as” language,

Title VII does not protect individuals who are “regarded as” (or mis-

perceived as) a particular identity.

312

But unlike the anti-discrimina-

tion laws discussed in this Article, the ADA is asymmetrical, i.e., it is

focused on remedying discrimination against individuals with disabili-

ties—not able-bodied people.

313

The ADA therefore needs to include

the “regarded as” language to protect people perceived to have a disa-

bility even if they do not actually have that disability.

314

This is not

true for Title VII and related laws because they are symmetrical and

protect all races, religions, etc. Thus, including the “regarded as” lan-

guage in Title VII would be redundant and unnecessary.

315

The inconsistency with which courts interrogate identity in discrimi-

nation cases only further shows that anti-discrimination law does not

require identity adjudication. As the previous Section demonstrated,

there is inconsistency both within and between identity categories.

Some courts impose actuality requirements in race discrimination

cases, while others do not;

316

same story for religious discrimination

claims.

317

Moreover, courts rarely, if ever, throw out sexual orienta-

tion discrimination claims based on the plaintiff’s actual sexual orien-

tation but are much more likely to do so for race discrimination

claims.

318

Courts are coming to these inconsistent conclusions even though

neither Title VII nor corresponding state laws allow courts to use dif-

ferent tests for different identity categories. In other words, the same

anti-discrimination statutes cannot allow misperception claims in the

context of religion and then reject them in the context of race—and

311. See 42 U.S.C. § 12101(a)(1) (“The Congress finds that . . . many people with

physical or mental disabilities have been precluded from [fully participating in all as-

pects of society] because of discrimination; other who have a record of a disability or

are regarded as having a disability also have been subjected to discrimination . . . .”

(emphasis added)).

312. Butler v. Potter, 345 F. Supp. 2d 844, 850 (E.D. Tenn. 2004) (“Congress has

shown, through . . . the Americans with Disabilities Act, that it knows how to enact

legislation that protects persons who are wrongly perceived to be in a protected

class.”).

313. See 42 U.S.C. § 12101(b)(1) (“It is the purpose of the chapter to provide a

clear and comprehensive national mandate for the elimination of discrimination

against individuals with disabilities.”).

314. See Protected Class, supra note 27, at 115–16 (making the same point).

315. See id. (arguing that the addition of “regarded as” to Title VII is not only

unnecessary but also would be a bad idea); see also Greene, supra note 17, at 153–162

(presenting a survey of cases and EEOC directives to show that rather than adding

“regarded as” language to Title VII, “all individuals alleging categorical discrimina-

tion on the basis of Title VII’s proscribed characteristics are already entitled to statu-

tory protection and have standing to maintain a claim of discrimination regardless of

whether their identity was correctly perceived”).

316. See discussion supra Part III(B)(3).

317. See discussion supra Part III(B)(2).

318. See discussion supra Parts III(B)(1), III(B)(3).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 59 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 325

the same statute cannot require a determination of a plaintiff’s “ac-

tual” identity in sex discrimination cases and then not require such a

determination in sexual orientation discrimination cases.

319

To be sure, identity is relevant to an anti-discrimination claim. The

discriminatory conduct must be connected to a protected identity cat-

egory to fall under Title VII and related laws. But courts can conduct

this inquiry under the causation prong of its analysis rather than inter-

rogating the plaintiff’s identity.

320

In other words, instead of asking

what a plaintiff’s identity is, they should ask what the cause of the

discrimination was—rather than ask whether the plaintiff is Hispanic,

gay, Muslim, etc., ask whether the alleged discriminatory conduct was

motived by race, sex, religion, etc.

321

To answer this causation ques-

tion, courts can and should focus on how the discriminators under-

stood or perceived the plaintiffs’ identities and their motivations

behind the discrimination.

322

Taking all of this together, identity adjudication is not doctrinally

necessary. The viability of a racial discrimination claim does not de-

pend on the “true” racial identity of the plaintiff. Determining a plain-

tiff’s “actual” identity is not doctrinally necessary for sex

discrimination cases either. Recall that the cases discussed in Part III

used plaintiffs’ biological traits to determine their “actual” sex,

323

but

determining a plaintiff’s “actual” sex, using a plaintiff’s biological

traits or otherwise, is not necessary to adjudicate sex discrimination

claims. Said differently, courts do not need to know the makeup of

plaintiffs’ internal and external reproductive organs, the composition

of their chromosomes, the sex that they were assigned at birth, or any

other sensitive biological information to adjudicate a sex discrimina-

tion claim. Rather, just like with other identities, the relevant inquiry

is the discriminator’s perception of the plaintiff’s sex identity.

Consider Ann Hopkins, the plaintiff in Price Waterhouse. The dis-

crimination Hopkins faced was based on the fact that her perceived

sex (not her “actual” sex) did not align with how her employer ex-

pected people of that identity to perform their identity.

324

Her em-

319. Identity interrogation is also not necessary under other anti-discrimination

laws addressed here. Section 1981 is symmetrical and applies to all persons regardless

of their actual races. McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 286–87

(1976) (holding that § 1981 applies to all persons, including white people). State level

anti-discrimination laws are generally modeled after federal laws, so the arguments

from Title VII and § 1981 also applies to those laws. See, e.g., D

. C

ODE

. tit. 19,

§§ 710–711 (2021);

. S

TAT

§ 111.32 (2021).

320. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 (1973).

321. Naomi Schoenbaum has also argued that by considering identity in the causa-

tion prong, the court’s analysis focuses on the conduct of the discriminator, as op-

posed to the identity of the plaintiff. Case for Symmetry, supra note 303, at 138–40.

322. For more on why this approach would benefit multiracial plaintiffs in particu-

lar, see Judicial Erasure, supra note 27, at 547–50.

323. See, e.g., Ulane v. E. Airlines, Inc., 742 F.2d 1081, 1087 (7th Cir. 1984).

324. Price Waterhouse v. Hopkins, 490 U.S. 228, 234–35 (1989).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 60 10-MAR-22 14:18

326 TEXAS A&M LAW REVIEW [Vol. 9

ployer perceived her to be a woman and did not think women should

be masculine, competitive, or aggressive.

325

Nothing about Hopkins’

case or the Court’s reasoning would change if Hopkins had a penis or

XY chromosomes, as long as her discriminators did not know about

these biological traits and/or they did not affect the discriminators’

perception of her sex.

326

The relevant inquiry was not whether Hop-

kins had biological traits associated with a female sex identity but

which sex her employer understood her to be.

The same is true if Hopkins did have penis or XY chromosomes,

and her employer fired her when he found out about these biological

traits. If her employer understood her sex to be male (because of

these biological traits), she could win under a sex-stereotyping theory

of sex discrimination by arguing that the discriminatory treatment was

based on the employer’s stereotype that males are not supposed to be

feminine or identify as a woman. If her employer perceived her sex to

be female (because of her self-identification and ascriptive factors),

sex-stereotyping also applies because the discrimination stemmed

from a stereotype that females are not supposed to have penises. Ei-

ther way, interrogating Hopkins’ “actual” sex is not necessary.

327

In sum, courts who focus on determining the “actual” identity of

plaintiffs are engaged in context-detached identity adjudication be-

cause these particular anti-discrimination laws do not require a deter-

mination of identity in the first place.

2. Identity Adjudication and Normative Theories of Anti-

Discrimination Law

In addition to not being required under black letter anti-discrimina-

tion law, interrogating plaintiffs’ identities does not serve the overall

purposes of anti-discrimination law. The purposes of anti-discrimina-

tion laws have been widely debated,

328

but here, I address the two

most prominent theories: anti-classification and anti-subordination.

329

Under the “anti-classification,” also known as “anti-differentia-

tion,” school of thought, anti-discrimination laws are meant to protect

325. Id. at 235.

326. Katherine Franke has also argued that sex discrimination is rarely, if ever,

based on biological traits like chromosomes and that equality law should abandon its

commitment to biological definitions of sex and to a truth of sexual difference. See

Katherine M. Franke, The Central Mistake of Sex Discrimination Law: The Disaggre-

gation of Sex from Gender, 144

U. P

. L. R

1, 5, 98 (1995).

327. Bostock’s holding would still stand under this conception of sex adjudication.

The only change would be in its rationale. Rather defining “sex” as the employee’s

sex assigned at birth, Bostock could have focused on how the employer perceived the

plaintiff’s sex as the starting point of analysis.

328. See, e.g., Ruth Colker, Anti-Subordination Above All: Sex, Race, and Equal

Protection, 61

N.Y.U. L. R

1003, 1005–06 (1986).

329. Jessica Clarke argues why protected-class gatekeeping is also not justified

under two additional theories of anti-discrimination law—anti-essentialism and anti-

balkanization. Protected Class, supra note 27, at 145–55.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 61 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 327

individuals from mistreatment based on a protected category (sex,

race, religion, etc.).

330

This position is focused on individual fairness

and favors prohibiting decisionmakers from considering any protected

identity category, regardless of whether that identity is privileged or

subordinated in society.

331

To an anti-classificationist, it is equally

harmful to discriminate against men as it is women. Litigating the

plaintiff’s “actual” identity isn’t in line with this view because the dis-

criminator should be punished for taking a protected category into

account no matter what the plaintiff’s “real” identity is. That is, re-

gardless of whether an individual is discriminated against based on

their actual or perceived identity, they were still discriminated against

based on a protected identity. Thus, a discriminator violates these fair-

ness principles of anti-discrimination law regardless of the plaintiff’s

“actual” identity and even when a discriminator is wrong about the

plaintiff’s identity.

332

Anti-subordination theory also does not support identity adjudica-

tion for anti-discrimination laws. Under an anti-subordination ap-

proach, anti-discrimination law should seek to prevent “practices that

enforce the secondary social status of historically oppressed

groups.”

333

Unlike anti-classification theory, which views the harms of

discrimination on an individual level, anti-subordination theory fo-

cuses on systemically disadvantaged groups. Under this view, discrimi-

natory practices may harm privileged groups—straight people, white

people, men, etc.— but not in the same way that discrimination

against marginalized groups “compound[s] a pattern of stigma and

bias going back generations and reaching across domains of social

life.”

334

An anti-subordination theorist might be in favor of different

liability standards for claims challenging anti-Black, anti-woman, or

anti-queer discrimination compared to claims alleging anti-white, anti-

male, or anti-heterosexual discrimination. It may seem that under this

theory, adjudicating the plaintiff’s identity is necessary to determine

which legal standard should apply.

330. Colker, supra note 328, at 1005.

331. See Bradley A. Areheart, The Anticlassification Turn in Employment Discrimi-

nation Law, 63

. L. R

955, 994–95 (2012); Johnson v. Ry. Express Agency,

Inc., 421 U.S. 454, 457–58 (1975); McKennon v. Nashville Banner Publ’g Co., 513 U.S.

352, 358 (1995).

332. Craig Robert Senn, Perception over Reality: Extending the ADA’s Concept of

“Regarded As” Protection Under Federal Employment Discrimination Law, 36

. U. L. R

. 827, 856 (2009) (arguing that “‘erroneous discriminators’ are—to be

blunt—still discriminators”).

333. Jack M. Balkin & Reva B. Siegel, The American Civil Rights Tradition: An-

ticlassification or Antisubordination?, 58

U. M

. L. R

9, 9 (2003). For some addi-

tional accounts of anti-subordination theory, see

ERRICK

ELL

AVED

: T

LUSIVE

UEST FOR

ACIAL

USTICE

(1987); Owen M. Fiss, Groups and

the Equal Protection Clause, 5

HIL

. & P

. A

FFS

107, 108, 157 (1976).

334. Protected Class, supra note 27, at 156.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 62 10-MAR-22 14:18

328 TEXAS A&M LAW REVIEW [Vol. 9

However, this position misses the distinction between interrogating

a plaintiff’s identity and what qualifies as illegal discrimination.

335

Ad-

judicating whether the plaintiff is white or Black, gay or straight, etc.

is a separate question from whether anti-white discrimination should

be cognizable under anti-discrimination law or litigated under the

same standards as discrimination against racial minorities. Said differ-

ently, discrimination claims that attack anti-Black racist structures in

the workplace support an anti-subordination theory of anti-discrimi-

nation law regardless of the identity of the plaintiff who brings those

claims. Under anti-subordination theory, “those harmed by discrimi-

natory dynamics such as racism, sexism, religious intolerance, and

homophobia should have recourse to the law, whatever their

identities.”

336

Another reason interrogating plaintiffs’ identities does not serve

anti-subordination goals is exposed through the multi-identity lens of

this Article. As discussed, race discrimination claims are more likely

to be dismissed based on the identity of the plaintiff than other types

of discrimination claims. This applies to discrimination cases that fur-

ther anti-subordination goals—i.e., cases attacking racial hierarchies—

not only cases of “reverse” discrimination. On the other hand, in the

sexual orientation context, discrimination cases that further anti-sub-

ordination purposes—i.e., cases disrupting heteronormativity and

homophobia—are generally not dismissed based on the plaintiff’s

identity. So while both straight and non-straight plaintiffs are permit-

ted to bring claims based on anti-gay harassment, in some cases, only

plaintiffs of a particular race can bring claims based on racial subordi-

nation. This result is not in line with anti-subordination goals, particu-

larly considering that the origins of the principle are rooted in

addressing racial subordination.

337

Shifting the focus away from adjudicating a plaintiff’s “actual” iden-

tity allows courts to direct their attention to the discriminatory con-

duct the plaintiff encountered. Dismissing cases based on the

plaintiff’s identity, and/or determining whether a plaintiff’s identity

claim meets some court-adopted “objective” standard is not sup-

ported by either the black letter law of anti-discrimination law or the

two dominant theories of anti-discrimination law.

335. Jessica Clarke makes this argument when explaining why protected class in-

quiries are not required by an anti-subordination approach to anti-discrimination law.

See id. at 158.

336. Id. (emphasis added); see also Case for Symmetry, supra note 303, at 86–98

(arguing that a symmetrical approach to anti-discrimination laws serves anti-subordi-

nation goals).

337. See Fiss, supra note 315, at 157.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 63 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 329

3. Problems Caused by Context-Detached Adjudication

The problems with context-detached identity adjudication have

been raised throughout this Article, but I put a point on them here.

First, courts who adopt a context-detached approach and focus on de-

termining a plaintiff’s “real” identity create inconsistency in the case

law. Currently, the success or failure of plaintiffs’ misperception

claims does—but should not—depend on the particular judge as-

signed to the case or the identity category involved. This approach can

also produce results that are at odds with the purpose of these anti-

discrimination laws. Title VII and related statutes are intended to pro-

vide a remedy to people who were mistreated based on a protected

category regardless of their “real” identity. So when plaintiffs lose

claims in whole or in part based on their “real” identity, these laws are

not being implemented to fidelity.

Additionally, when legal actors try to determine a plaintiff’s “ac-

tual” identity, they are relying on more identity evidence than neces-

sary to adjudicate the identity—that is, they are over-interrogating

identity. I fully explore the consequences of over-interrogation in

other work,

338

but I summarize two primary harms here. One, making

identity claims dependent on unnecessary identity evidence may in-

fringe upon plaintiffs’ privacy. As other scholars have explained, indi-

viduals have legally protected privacy interests in their identities and

should be able to exercise control over how, when, and to whom vari-

ous aspects of their identities are disclosed.

339

So when courts examine

things like the make-up of someone’s chromosomes,

340

ancestral back-

ground,

341

and blood quantum,

342

when the law doesn’t require them

to do so, they may be impinging on plaintiffs’ privacy. To be sure, the

plaintiffs themselves disclosed this identity information to the court in

the first instance, but the court then rebroadcast that information in

publicly available court opinions. Moreover, if courts made clear that

the identity inquiry was not about determining plaintiffs’ “real” identi-

ties, plaintiffs would probably not reveal this information, or at least

not as much of it, to the court in the first place. By relying on this

information to determine “actual” identity and treating “actual” iden-

tity as important to the success of plaintiffs’ claims, courts are encour-

aging disclosure.

338. See Working Manuscript, supra note 24.

339. See Scott Skinner-Thompson, Outing Privacy, 110

. U. L. R

159, 161–62

(2015) (arguing that information about someone’s sexuality, sex, and related health

and medical information have heightened constitutional importance under the Due

Process Clause); Elective Race, supra note 61, at 1505–07 (arguing that individuals

have a legally recognized interest in exercising control over the terms of disclosing

their racial identity).

340. See Ulane v. E. Airlines, Inc., 742 F.2d 1081, 1085 (7th Cir. 1984).

341. See Bennun v. Rutgers State Univ., 941 F.2d 154, 172–73 (3d Cir. 1991).

342. See Leonard v. Katsinas, No. 05-1069, 2007 WL 1106136, at *2 (C.D. Ill. Apr.

11, 2007).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 64 10-MAR-22 14:18

330 TEXAS A&M LAW REVIEW [Vol. 9

The other harm from over-interrogation involves the power of the

law unnecessarily reifying the essential characteristics of an identity

and reinforcing stereotypes about that identity.

343

For example, when

the Bennun v. Rutgers State University court found that the plaintiff

was Hispanic based on his appearance, speech, cultural affiliation with

“Spanish life,” and ability to speak Spanish, the court was suggesting

that some of these things (all of these things?) are definitional quali-

ties of being “Hispanic.”

344

In turn, the court reinforced stereotypes

about what it means to be a “real” Hispanic (someone who speaks,

dresses, and looks a certain way).

Additionally, courts’ reliance on “actual” sex in sex discrimination

cases, and determination of “actual” sex based on biological charac-

teristics, supports the dominant discourse that sex has innate, fixed,

physical components. First, as previously argued, there is no legal

need to adopt any definition of “actual” sex in this context.

345

Moreo-

ver, defining “actual” sex as sex assigned at birth, chromosomes, sex

characteristics, or other biological identity evidence supports a model

of sex that does not align with many people’s lived experiences and

can cause a myriad of harms, including: promoting nonconsensual sur-

gery on babies born with genitals that are not clearly a penis or a

vagina,

346

rendering invisible nonbinary individuals who do not sub-

scribe to binary models of sex,

347

and stigmatizing and pathologizing

transgender people,

348

to name a few.

To a certain extent, all identity adjudication runs the risk of essen-

tializing identity—when legal actors determine whether someone is a

particular race, sex, etc., whatever identity evidence is dispositive of

that determination becomes linked to that identity. But as long as the

identity inquiry is tied to the function of the law, no one type of iden-

tity evidence will ever be determinative of identity across all legal con-

343. See

ORD

, supra note 33, at 78 (arguing that when legal actors make certain

traits or characteristics essential to identity, they may “underwrite destructive self-

images and misguided commitments with the force of law and the intractability of

precedents”).

344. See Bennun, 941 F.2d at 172–73 (finding that the plaintiff was Hispanic due to

his “birth in a Latin American country where Hispanic culture predominates, his im-

mersion in Spanish ways of life, and the fact that he speaks Spanish in the home,” as

well as his “appearance, speech and mannerism” and other traits that indicate his

“objective appearance to others”). In other words, the court is giving “Hispanic” a

“substantive, cultural content [that] is a way of freezing the definition of that identity

in place and time, or declaring that it contains universal characteristics, or certain

parameters within which anyone who truly belongs must fall.” Cristina M. Rodr´ıguez,

Against Individualized Consideration, 83

. L.J.

1405, 1412 (2008).

345. See supra Part III(C)(4).

346. See, e.g., Chinyere Ezie, Deconstructing the Body: Transgender and Intersex

Identities and Sex Discrimination—the Need for Strict Scrutiny, 20

OLUM

. J. G

ENDER

& L.

141, 150–51 (2011).

347. See, e.g., Spade, supra note 30, at 322.

348. See, e.g., Dean Spade, Resisting Medicine, Re/modeling Gender, 18

ERKELEY

OMEN

’

L.J.

15, 25–26 (2003).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 65 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 331

texts. Rather, identity definitions will be flexible and context-specific.

For instance, some laws may require biological evidence to determine

sex, but other times, ascriptive evidence or self-identification may be

more appropriate. This way, no single trait becomes an essential ele-

ment of sex.

C. Benefit Laws

Legal actors also, at times, engage in context-detached identity ad-

judication with benefit laws, and, in turn, examine the wrong kind of

evidence to achieve the purpose of the law. Just like the analysis of

anti-discrimination laws did not take a position on what the goal, pur-

pose, or function of anti-discrimination law should be, this analysis

also does not tackle what the purpose of a particular law should be;

that is beyond this Article’s scope. The proper normative function of

any one of these laws could be and is the topic of entire law review

articles. Rather, this Section uses the function of the law as stated in

the case law, statutes, regulations, lawmakers, or other primary

sources. This Section provides examples of context-detached adjudica-

tion and shows how, in many instances, the purpose or function of

these laws can be served with a less strict, or different, standard for

identity claims.

1. Sex

Recall that many states require people to have some sort of surgical

procedure to change the sex marker on their birth certificate.

349

Man-

dating invasive and expensive surgery is burdensome and arguably

more taxing than any other form of identity proof discussed in this

Article.

350

These requirements do not align with the underlying goals

of these laws. Here’s why.

States’ interests in regulating sex markers on identity documents are

generally not made explicit in the governing statutes or regulations,

but some courts, agencies, and legislators have cited (1) the regulation

of sex-segregated spaces and (2) concerns about fraud as potential

349.

AMBDA

EGAL

, supra note 252.

350. Vaginoplasty, for example, is a six hour surgery with a recovery time of up to

18 months. Erin Larowe, What to Expect: Vaginoplasty at the University of Michigan

Health System,

U. M

ICH

. H

EALTH

, http://www.med.umich.edu/pdf/Vagi-

noplasty.pdf (Dec. 2016) [https://perma.cc/ZPW5-YS5V]. Phalloplasty can take as

long as eight hours with a similar recovery time. What to Expect: Phalloplasty at Mich-

igan Medicine,

ICH

. M

, http://www.med.umich.edu/1libr/Surgery/PlasticSurgery/

GenderConfirmationSurgery/Phalloplasty.pdf (Apr. 2019) [https://perma.cc/3GXW-

4LX2]. Chest masculinization surgery, or “bilateral mastectomy,” typically takes 3 to

5 hours to perform, and patients recover over a period of 6 to 12 months. Ctr. for

Transgender Health, FAQ: Chest Masculinization,

OHNS

OPKINS

, https://

www.hopkinsmedicine.org/center-transgender-health/services-appointments/faq/top-

surgery [https://perma.cc/6F3L-M8EW].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 66 10-MAR-22 14:18

332 TEXAS A&M LAW REVIEW [Vol. 9

state interests.

351

But neither of these interests support such an oner-

ous proof requirement for a sex identity claim. First, any interest the

state has in ensuring that sex-segregated spaces (bathrooms, locker

rooms, etc.) only include people of the same sex does not support the

surgical requirement for sex marker changes. When made, this argu-

ment is generally tied to a privacy or safety interest—i.e., people (and

cisgender women in particular) will not be safe and/or will have their

privacy interests violated in the restroom if people of another sex are

permitted to enter that space.

352

One obvious reason that requiring

surgery to change sex markers on identity documents does not sup-

port this interest is because these identity documents are usually not

required to access sex-segregated spaces.

353

But putting that aside, safety and privacy concerns do not support

the burdensome identity proof standard for sex marker changes. First,

there is little to no empirical evidence suggesting that allowing trans-

gender people access to sex-segregated spaces raises safety concerns

for cisgender people.

354

In fact, transgender individuals are more

likely to experience—not commit—violence in the restroom than cis-

gender individuals.

355

Moreover, it is unclear how requiring surgery to change a sex

marker helps ensure privacy or safety. As for the safety concern, re-

quiring surgery suggests that someone who has had sex-reassignment

351. See, e.g., Mottet, supra note 246, at 413–20; see also supra notes 247–48 and

accompanying text.

352. See, e.g., Gwen Aviles, Kentucky Bill Would Let Students Sue Over Trans-

gender Bathroom Use,

NBC N

EWS

(Dec. 13, 2019, 1:21 PM), https://

www.nbcnews.com/feature/nbc-out/kentucky-bill-would-let-students-sue-over-trans-

gender-bathroom-use-n1101651 [perma.cc/76E6-8KSU] (describing a bill in Kentucky

regarding bathroom use stating that “[c]hildren and young adults have natural and

normal concerns about privacy while in various states of undress, and most wish for

members of the opposite biological sex not to be present in those circumstances”);

Aamer Madhani, Battle Brewing Over Transgender Bathroom Laws in State Capitals,

USA T

ODAY

, https://www.wkyc.com/article/news/battle-brewing-over-transgender-

bathroom-laws-in-state-capitals/57900588 (Feb. 27, 2016, 9:14 AM) [https://perma.cc/

3F38-8T7F] (finding that opponents of a nondiscrimination bill in Houston argued

that transgender women are sexually perverse men who would use the ordinance to

enter the women’s restrooms and prey on wives, mothers, and daughters).

353. Mottet, supra note 246, at 419.

354. See, e.g., Carlos Maza & Luke Brinker, 15 Experts Debunk Right-Wing Trans-

gender Bathroom Myth,

EDIA

ATTERS

(Mar. 19, 2014, 4:06 PM), https://

www.mediamatters.org/sexual-harassment-sexual-assault/15-experts-debunk-right-

wing-transgender-bathroom-myth [https://perma.cc/UFT9-G97S] (compiling reports

from law enforcement officials, government employees, and advocates for victims of

sexual assault from 12 states and concluding that concerns about violence have no

factual basis).

355. See Katy Steinmetz, Why LGBT Advocates Say Bathroom ‘Predators’ Argu-

ment Is a Red Herring,

IMES

(May 2, 2016, 4:29 PM), https://time.com/4314896/trans-

gender-bathroom-bill-male-predators-argument/ [https://perma.cc/L87D-HSNB].

ANDY

E. J

AMES ET AL

’

FOR

RANSGENDER

QUAL

EPORT OF

THE

2015 U.S. T

RANSGENDER

URVEY

226–27

(2016), https://transequality.org/sites/

default/files/docs/usts/USTS-Full-Report-Dec17.pdf [perma.cc/7S34-VLUX].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 67 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 333

surgery is less prone to violent behavior than someone who has not

had sex-reassignment surgery. But a self-identified woman with a va-

gina is not less likely to harm someone than a self-identified woman

with a penis.

356

There is simply no evidence indicating that pre-opera-

tive transgender people pose a greater safety risk than post-operative

transgender people or cisgender people.

357

As for the privacy concern, some have asserted that cisgender indi-

viduals’ privacy interests are violated when transgender individuals

with different genitals or secondary sex characteristics enter a sex-seg-

regated space.

358

But as Professor Susan Hazeldean has convincingly

argued, none of the six primary theoretical justifications for privacy

rights support this claimed privacy violation.

359

Instead, she shows

how it is transgender individuals’ privacy rights that are at stake when

they are excluded from sex-segregated spaces.

360

Even assuming cisgender individuals’ privacy interests are legiti-

mate, these interests do not require transgender people to have sur-

gery to access a sex-segregated space. Many sex-segregated spaces

(like bathrooms and dressing rooms) provide ample privacy options

such that individuals do not have to expose their body or view anyone

else’s body. For other spaces, like locker rooms—where privacy may

be harder, but usually not impossible, to come by—the privacy inter-

est can be addressed by adding additional private spaces, which is

something that most people generally desire.

361

Additionally, by re-

quiring surgery, the law suggests that an individual’s subjective feel-

ings of discomfort when a woman with a penis sees their exposed body

(compared to their discomfort when a woman with a vagina sees them

naked), is so great that the law should require people to have expen-

sive and evasive surgery. Privacy or safety concerns, therefore, cannot

justify such high standards for sex marker change laws.

A second interest often invoked in this context is fraud prevention,

or the concern that people will fraudulently change their sex markers

356. See generally Maza & Brinker, supra note 354 (suggesting that because al-

lowing transgender people access to sex-segregated spaces has not resulted in in-

creased crimes, a transgender person is not more likely to commit a crime than a

cisgender person).

357. Id.

358. See, e.g., Wolf Continues Its Fight for Girls’ Privacy and Safety in Boyertown,

OMEN

’

IBERATION

RONT

(Jul. 12, 2018), https://www.womensliberationfront.org/

news/wolf-continues-its-fight-for-girls-privacy-and-safety-in-boyertown?rq=birth

%20certificate [https://perma.cc/W7MU-UNR5] (statement of Lierre Keith) (“A

male person cannot become female. Male people who wish to enter the intimate

spaces that are reserved for female people are invading women’s spaces, which vio-

lates the human rights of women and girls.”).

359. Susan Hazeldean, Privacy as Pretext, 104

ORNELL

L. R

1719, 1769–70

(2019).

360. Id. at 1770; see also Ruth Colker, Public Restrooms: Flipping the Default Rules,

HIO

. L. J.

145, 171–76 (2017) (debunking the privacy and safety justifications

for sex-segregated bathrooms).

361. Clarke, supra note 28, at 982–83.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 68 10-MAR-22 14:18

334 TEXAS A&M LAW REVIEW [Vol. 9

to avoid law enforcement or creditors or to engage in some other bad

act.

362

But the extent to which changing a sex marker helps someone

commit fraud is questionable. Changing other information—like

name, address, financial institution, and hair color—is more helpful in

facilitating fraud than changing a sex marker.

363

Yet changing this in-

formation is comparatively much easier than changing a sex marker.

Moreover, making it difficult to change sex markers on identity docu-

ments can ring false alarms for fraud and divert law enforcement at-

tention away from legitimate fraud. For instance, consider a person

who ascriptively appeared to be a woman to most observers. The pic-

ture on her license is updated and matches her current appearance,

but her sex marker still reads “M” because she did not meet the law’s

surgical requirements. Because of this mismatch between her appear-

ance and her sex marker, she is subjected to extra security screenings

at airports, additional barriers when applying for loans, and increased

scrutiny when purchasing items that require proof of identification.

Resources are spent chasing down suspected fraud because her sex

marker raises red flags.

364

There is no empirical evidence that less onerous standards to

change sex markers are correlated with an increase in fraud.

365

But

even assuming that the fraud concerns were legitimate, requiring inva-

sive medical procedures to change a sex marker would still be too high

of a burden. An individual seeking to commit fraud by changing their

birth certificate can be deterred through less rigorous standards—like

a statement from a doctor or other professional attesting to the verac-

ity of their desire to change sex markers.

Some scholars have convincingly argued that the government does

not have any legitimate interest in regulating sex markers on birth

certificates and that sex markers should be eliminated on birth certifi-

cates.

366

But even assuming arguendo that there is some reason the

state needs to have sex markers on birth certificates in the first place,

or needs to erect any barriers to changing them, requiring biological

evidence of sex, usually in the form of surgery, does not align with the

362. See, e.g., Kenji Yoshino, Sex and the City: New York City Bungles Transgender

Equality, S

LATE

(Dec. 11, 2006, 2:43 PM), https://slate.com/news-and-politics/2006/12/

new-york-city-bungles-transgender-equality.html [https://perma.cc/F8BY-W752]

(describing the concern that “[l]owering the barriers to sex reassignment increases the

incentive for individuals who have no sincere desire to change their sex to do so for

opportunistic reasons.”); Clarke, supra note 28, at 947–48 (discussing this concern and

other concerns in the context of non-binary gender).

363.

AVIS

, supra note 48, at 51 (arguing that sex-identity markers are “poor prox-

ies for personal identification” and that governments should use other techniques,

such as biometrics, instead).

364. See id. at 2–3.

365. See Yoshino, supra note 362 (discussing the unlikeliness of sex fraud).

366. See, e.g.,

AVIS

, supra note 48, at 51 (“[R]elevant [government] agencies

[should] remove sex markers from the identity documents they issue.”).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 69 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 335

purported functions of these laws (even assuming these functions are

legitimate).

Turning to sex-segregated sports, both the former NCAA rule (the

focus of which on testosterone levels is likely to be perpetuated

through its new sport-specific approach) and the Idaho law determine

sex in ways that fail to align with their stated purposes, though the

latter much more so than the former. As for the NCAA rule, before

the rule change in January 2022, the self-proclaimed purpose behind

the policy was to ensure that no one has an unfair advantage in wo-

men’s sports due to high testosterone levels.

367

However, the way the

NCAA adjudicates sex (requiring trans women to be on testosterone

blockers for one year

368

) does not necessarily align with this purpose.

To be sure, there is evidence linking higher testosterone levels to in-

creased muscle mass, strength, and other traits that can give athletes

in certain sports a competitive advantage.

369

However, if testosterone

levels produce an unfair advantage, why only regulate the testoster-

one of transgender women? Cisgender women could have higher

levels of testosterone than a transgender woman on testosterone

blockers.

370

In this way, the NCAA’s former policy, and NGBs that

have adopted, or will adopt, a similar rule, are context-detached be-

cause they fail to account for testosterone in cisgender women. Rather

than ensuring people who compete in women’s sports have similar

levels of testosterone, this policy singles out transgender people and

their bodies as objects of regulation.

The purported purposes of Idaho’s law are similar to the fairness-

based goal of the NCAA rule—namely, “promoting sex equality, pro-

viding opportunities for female athletes to demonstrate their skill,

strength, and athletic abilities, and by providing female athletes with

opportunities to obtain college scholarship and other accolades.”

371

However, the Act’s approach to determining who is “female” for sex-

segregated sports has very little to do with these purposes.

372

As dis-

cussed, the Act relies on “reproductive anatomy,” chromosomes, and

naturally occurring hormone levels. Yet these traits do not affect ath-

letic performance. That is, whether a student was born with a vagina

or a penis, or XY or XX chromosomes, in and of themselves, have

nothing to do with that student’s athletic capabilities. Because these

traits do not affect students’ bodies in ways that could improve their

physical capabilities, using them to determine “sex” does nothing to

367. See

RIFFIN

& C

ARROLL

, supra note 263, at 10, 13.

368. Id. at 13.

369. See Hecox v. Little, 479 F. Supp. 3d 930, 982 (D. Idaho 2020).

370. See David J. Handelsman, Angelica L. Hirschberg & Stephane Bermon, Circu-

lating Testosterone as the Hormonal Basis of Sex Differences in Athletic Performance,

NDOCRINE

EVS

. 803, 807 (2018) (showing varying levels of testosterone in cis-

gender women).

371.

DAHO

ODE

§ 336202(12).

372. Little, 479 F. Supp. at 984.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 70 10-MAR-22 14:18

336 TEXAS A&M LAW REVIEW [Vol. 9

ensure fairness in women’s sports or promote sex equality. The fair-

ness purpose of the Act would only arguably be advanced by some-

thing it excludes—circulating testosterone, the primary driver of

relevant physiological differences between male and female athletic

performance.

373

Instead, the Act relies on endogenous testosterone,

which depending on the individual, may not reflect their current tes-

tosterone levels or their athletic capabilities. Moreover, on its face,

Act does not bar transgender men who are taking testosterone from

participating in women’s sports as long as they have not had surgery

to their genitals. In other words, a student assigned female at birth

who has testosterone levels similar to those of a cisgender man are

“biologically female” under the terms of the Act even though they

arguably have a competitive advantage over cisgender women in some

contexts.

In these ways, and in others,

374

the Act’s definition of “sex” is com-

pletely context-detached because it fails to advance any of the Act’s

purported purposes. In its decision enjoining the Act, the district court

relied on the complete lack of fit between the Act’s stated goals and

the means used to achieve those goals when holding that the Act

likely violated the Equal Protection Clause.

375

Due to this gross lack

of fit between the Act’s stated goals and its means of achieving those

goals, the district court held that the purported purposes of the Act

were not the Act’s actual goal; rather, it held that the Act’s actual goal

is to “exclude[e] transgender women and girls from women’s sports

entirely, regardless of their physiological characteristics.” The Act’s

definition of sex is not context-detached at all if transgender exclusion

is its purpose; rather, it effectively serves its goal.

376

373. See id. at 984 (“[T]he Act’s definition of ‘biological sex’ intentionally excludes

the one factor that a consensus of the medical community appears to agree drives the

physiological differences between male and female athletic performance.”).

374. For instance, the Act also fails to make any distinctions based on age, skill

level, or sport. All three of these factors affect whether sex distinctions are relevant or

ensure fairness for women’s sports. For instance, sex distinctions are not related to

athletic performance until puberty; yet the Act applies to youth sports. Id. at 979

(“[B]efore puberty, boys and girls have the same levels of circulating testosterone.”).

Additionally, sex distinctions (post-puberty) might make a difference in sports like

basketball or soccer, but for sports like long-distance running and swimming or golf,

there is evidence that testosterone levels do not necessarily produce a competitive

advantage. See Nancy Leong, Against Women’s Sports, 95

ASH

. U. L. R

1249,

1275–76 (2018) (compiling evidence that for many sports, sex-segregation is not justi-

fied and that testosterone levels do not give people a competitive advantage).

375. Little, 479 F. Supp. 3d at 979–84 (“That the Act essentially bars consideration

of circulating testosterone illustrates the Legislature appeared less concerned with en-

suring equality in athletics than it was with ensuring exclusion of transgender women

athletes.”).

376. In my future work, I expand on how context-detached identity determinations

can reveal the actual purpose of a particular law, like how the context-detached na-

ture of the Act led the district court to conclude that the Act’s stated goals were not

its actual goals.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 71 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 337

2. Sexual Orientation

Context-detached adjudication of identity also occurs with sexual

orientation and asylum. The function of asylum and similar forms of

immigration relief is to determine whether a petitioner will be perse-

cuted if forced to return to their home country and, for our purposes,

whether that persecution is motivated by sexual orientation.

377

Ac-

cordingly, what people in the petitioner’s home country know or think

about the petitioner’s sexuality, or will know or think about the peti-

tioner’s sexuality if they return, is crucial to this question of future

persecution.

378

If those likely to persecute the petitioner think the pe-

titioner is not straight, the petitioner will probably be persecuted.

But sometimes, this persecution-prevention purpose is not guiding

the identity inquiry. As discussed, some adjudicators reject non-

straight identity claims when petitioners have a history of different-sex

relationships or lack a history of same-sex relationships; however, re-

quiring alignment between petitioners’ current self-identification and

past sexual practices does not adequately take into account the goals

of asylum for at least three reasons. First, sexuality is not rigid and

stagnant; rather, it changes depending on context and over time.

379

a petitioner’s prior different-sex relationships do not necessarily shield

her from future persecution because her sexuality could have

changed. Second, being in a different-sex relationship does not mean

that someone identifies as heterosexual. Bisexual and queer people

are often in different-sex relationships and can still be persecuted

based on their non-straight orientation. Third, because these petition-

ers come from countries where not being straight is, at best, socially

stigmatized and, at worst, criminalized,

380

it is reasonable, and perhaps

expected, that these petitioners would have been in different-sex rela-

tionships. For this same reason, petitioners often have difficultly solic-

377. 8 U.S.C. § 1101(a)(42); see also Cruz v. Sessions, 853 F.3d 122, 127 (4th Cir.

2017).

378. In this way, these laws mirror anti-discrimination laws in the sense that with

both laws, the crucial identity question is how people’s identities will be perceived by

their discriminators/persecutors. However, unlike with anti-discrimination laws,

where I argued that the “actual” identity of the plaintiff is irrelevant, the petitioner’s

sexual orientation may be relevant. Anti-discrimination law looks backward in time to

adjudicate prior discriminatory conduct, but asylum law is forward-looking because it

attempts to predict how the petitioner will be treated. If a petitioner is, in fact, not

straight, they are likely to be persecuted in the future based on their non-straight

identity. Thus, the petitioner’s actual identity is strong evidence of how they are likely

to be perceived in their home country.

379. See Christine E. Kaestle, Sexual Orientation Trajectories Based on Sexual At-

tractions, Partners, and Identity: A Longitudinal Investigation from Adolescence

Through Young Adulthood Using a U.S. Representative Sample, 56

J. S

SCH

811,

812 (2019) (describing fluidity of sexual orientation).

380. See

RAN

SLAMIC

ENAL

ODE

arts. 233–40 (criminalizing same-sex relations,

punishable by a range of various sentences ranging from flogging to the death pen-

alty);

UNISIA

ENAL

ODE

, art. 230 (“Sodomy. . . is punished by imprisonment for

three years.”).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 72 10-MAR-22 14:18

338 TEXAS A&M LAW REVIEW [Vol. 9

iting corroborating testimony from their same-sex partners in their

home country because it may subject their prior partners to

persecution.

To be sure, past sexual and romantic relationships can be probative

of whether petitioners will be persecuted based on their sexual orien-

tation. But when petitioners have other identity evidence showing that

their sexuality is known in their country, or will be known upon their

return, allowing prior different-sex relationships to trump this evi-

dence does not serve the purpose of the law—i.e., to prevent

persecution.

For instance, recall that in Mockeviciene, the petitioner had evi-

dence that people in her home country thought she was a lesbian; in-

deed, while in her home country, the police harassed her, and her

husband beat her when he discovered her sexuality.

381

In other words,

she had sufficient evidence that she would likely have been persecuted

on the basis of her sexual orientation if she returned to her home

country. Yet her claim for relief was denied because she had been

previously married to a man and not had a same-sex partner.

382

Thus,

the court failed to adjudicate her sexual orientation in a context-in-

formed manner, guided by the persecution-preventing purpose of

asylum.

The court’s error was even worse in Fuller—where the court re-

jected the petitioner’s claim to bisexuality because he was married to a

woman, had kids with a woman, and was convicted of sexually assault-

ing a woman.

383

Not only does this case suffer from the same context-

detached approach as Mockeviciene, it also makes it nearly impossible

for a petitioner to convince the court that they are bisexual. If Fuller,

someone who had romantic and sexual relationships with men and

women, was not bisexual to the court, who would be? The rejection of

Fuller’s bisexuality claim may reflect a view that bisexuality is not a

legitimate sexual orientation, which is a fairly common narrative, par-

ticularly as applied to men.

384

Thus, Fuller’s reasoning suggests that

proving a bisexual identity may be more difficult than proving a gay or

381. Mockeviciene v. U.S. Att’y Gen., 237 F. App’x 569, 570–71 (11th Cir. 2007).

382. Id. at 574.

383. Fuller v. Lynch, 833 F.3d 866, 868 (7th Cir. 2016).

384. See Elissa L. Sarno et al., Bisexual Men’s Experiences with Discrimination, In-

ternalized Binegativity, and Identity Affirmation: Differences by Partner Gender, 49

RCHIVES

EXUAL

EHAV

. 1783, 1783–84 (2020) (discussing how bisexual individuals

are stigmatized to be attention-seeking, untrustworthy, confused about their sexual

orientation, or feigning to be trendy); Katherine M. Hertlein et al., Attitudes Toward

Bisexuality According to Sexual Orientation and Gender, 16

J. B

ISEXUALITY

339, 341

(2016) (noting that female bisexuality is considered more socially acceptable than

male bisexuality); Yoshino, supra note 365, at 362 (arguing that homosexuals and het-

erosexual share an interest in bisexual erasure).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 73 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 339

lesbian identity, depending on the adjudicators’ biases against bisexual

identity.

385

Not all courts adjudicating identity for asylum claims share this

problem. Indeed, in the context of both religion and sexual orienta-

tion, some courts ask how the petitioner’s identity will be understood

in their home country.

386

Thus, identity adjudication in asylum is not

consistent and appears dependent on the particular legal actors

involved.

3. Race

Race adjudication also, at times, is insufficiently attentive to the

purposes of the law at issue. The function of the laws addressed in this

section have been widely debated, discussed, and litigated. Currently,

the Supreme Court has limited the constitutionally permissible pur-

poses of race-conscious benefit laws to (1) remedying past or present

discrimination

387

and (2) promoting institutional diversity.

388

For the

laws discussed in this Article, their purposes are primarily remedial, so

this analysis will therefore focus on that function.

Supreme Court precedent has greatly limited the scope of the reme-

dial function of these laws. Under this precedent, remedying societal

discrimination at large is too broad of a function.

389

Rather, the reme-

dial focus needs to be narrower, like remedying racial exclusion from

a particular industry or job within a certain geographical area (e.g.,

the construction industry in Atlanta).

390

Considering this function,

how then should eligibility for these programs be determined? In

other words, how can the law figure out who has likely been excluded

from a particular industry based on their race? One option is to ex-

385. This increased burden to prove sexual orientation may also apply to other sex-

uality categories that are less politically powerful and culturally salient than “gay” and

“lesbian,” like, for instance, asexuality. See Elizabeth F. Emens, Compulsory Sexual-

ity, 66

TAN

. L. R

303, 306, 347–48, 374 (2014) (discussing how our legal system

assumes the existence of sexuality and how asexuality has not yet been included in

legal understandings of sexual orientation).

386. Amanfi v. Ashcroft, 328 F.3d 719, 726 (3d Cir. 2003).

387. See, e.g., Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 286 (1986) (O’Connor,

J., concurring in part and concurring in judgment).

388. Grutter v. Bollinger, 539 U.S. 306, 343 (2003); Regents of the Univ. of Cal. v.

Bakke, 438 U.S. 265, 320 (1978); Gratz v. Bollinger, 539 U.S. 244, 275 (2003).

389. City of Richmond v. J.A. Croson Co., 488 U.S. 469, 502 (1989); O’Donnell

Const. Cor. v. Dist. of Columbia, 963 F.2d 420, 427 (D.C. Cir. 1992); Adarand Con-

structors, Inc. v. Pena, 515 U.S. 200, 227 (1995).

390. J.A. Croson Co., 488 U.S. at 496–97 (1989) (citation omitted) (comparing the

“the ‘focused’ goal of remedying ‘wrongs worked by specific instances of racial dis-

crimination’” with “the remedying of the effects of ‘societal discrimination,’ an amor-

phous concept of injury that may be ageless in its reach into the past” and explaining

that only the former is permissible). Moreover, remedial measures need to be sup-

ported by a “strong basis in evidence” of a racial disparity, which generally requires

that the government show a disparity between the number of eligible racial minorities

in the relevant market and the number of racial minorities in that market. See id. at

502.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 74 10-MAR-22 14:18

340 TEXAS A&M LAW REVIEW [Vol. 9

amine how the applicant may be racialized in the context of that in-

dustry. In other words, if the law’s purpose is to remedy

discriminatory treatment of Black or Hispanic applicants in a certain

industry, then eligible individuals should include those who are racial-

ized as Black or Hispanic in that industry. They are the ones most

likely to have been excluded under the discriminatory hiring regime.

If racialization is the most accurate way to identify who has been

excluded from these industries, then adjudicators’ goals should be to

determine whether someone has been voluntarily or involuntarily

racialized in a particular way.

391

Some aspects of racial adjudication in

the Malone and the Orion cases deviate from this purpose. First, in

both cases, the adjudicators examined the applicants’ physical appear-

ances and racial phenotypes to determine whether they appeared to

be the race they claimed. And in both cases, the applicants appeared

white to the adjudicators.

To be sure, physical appearance is highly relevant to how someone

is racialized. But relying exclusively on the adjudicators’ opinions of

whether the applicants’ physical appearances matched their self-iden-

tification may be the wrong inquiry. Whether someone appears to be a

member of a particular racial group is highly dependent on the

viewer—their experiences with other racial groups, their own race,

etc.

392

So the race adjudicators assign to applicants based on their

physical features may not be indicative of how these individuals are

racialized most of the time. Moreover, if the remedial purpose is fo-

cused on the context in which the discriminatory treatment has oc-

curred, it makes more sense for individuals within that particular

context to assign race based on an applicant’s physical appearance,

rather than adjudicators who are not in that industry or job.

Another example of context-detached race adjudication comes

from Malone, when the court considered whether the Black commu-

nity considered the Malone brothers to be Black.

393

Here again, this

inquiry does not help elucidate how the applicant was racialized

within the industry the benefit law is targeting. Whether the Malone

brothers were racialized as Black by the group they were claiming to

be a part of (the Black community) matters less here than how they

391. This discussion evaluates race adjudication under the theory that racialization

within an industry best captures eligibility based on the remedial purpose of these

laws. See Racial Commodification and the Promise of the New Functionalism, supra

note 198, at 210 (using racialization to determine eligibility for affirmative action pro-

grams). There may be a better mechanism to determine eligibility that is also constitu-

tionally permissible. The subsequent evaluation of whether racial adjudication is

overly onerous or not would likely change depending on the particular mechanism for

determining eligibility. The larger point this Article is making, however, remains the

same—that the identity evidence required for an identity claim should be informed by

what is needed to fulfill the function of the law.

392. See Greene, supra note 17, at 90.

393. See Malone v. Civ. Serv. Comm’n, 646 N.E.2d 150, 151 (Mass. App. Ct. 1995).

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 75 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 341

were racialized by the potential discriminators within the job or indus-

try who were not Black.

394

In contrast, in Orion, the analysis centered

on how Taylor’s race was perceived in the proper context, i.e., the

business community.

395

Thus, requiring that applicants be racialized as

the race they are claiming to be in contexts beyond the remedial focus

of the benefit law may also be overly burdensome and not in line with

the function of the law.

Another potentially problematic aspect of racial adjudication is the

assumption that prior inconsistent identifications are indicative of

fraud and then using those prior identifications to estop an identity

claim. In Malone, the brothers’ prior identification as white played a

role in their ultimate exclusion from the affirmative action program,

as did Taylor’s prior self-identification as white (before he received his

ancestry results) in Orion.

396

And in United States v. New York City

Board of Education, employees’ self-identification as white on their

job applications was enough to exclude them from the settlement

funds.

397

Using prior inconsistent racial identification to bar membership in a

claimed racial group does not necessarily align with the law’s function

and may exclude people the law intended to include. First, inconsis-

tent identification does not necessarily indicate fraud. Multiracial indi-

viduals sometimes self-identify as monoracial, or identify with

different monoracial groups, depending on the timing, the context, the

form of the inquiry about their racial status, and how they think their

responses will be used.

398

These same individuals may be racialized as

non-white and discriminated against on that basis; thus, in light of the

purpose of these laws, their prior identifications should not estop

them from asserting a different identity. Second, changes in self-iden-

tification may be associated with discovering new information about

one’s background, like the plaintiff in Orion.

399

If someone reveals

this new information or their new self-identification to others, it could

394. Other scholars have made related points about affirmative action and the Ma-

lone case. Racial Commodification and the Promise of the New Functionalism, supra

note 198, at 208–09; see also Ford, supra note 158, at 1281–82.

395. See Orion Ins. Grp. v. Wash. State Off. of Minority & Women’s Bus. Enters.,

No. 16-5582, 2017 WL 3387344, at *2–3 (W.D. Wash. Aug. 7, 2017), aff’d sub nom.

Orion Ins. Grp. v. Wash.’s Off. of Minority & Women’s Bus. Enters., 754 F. App’x 556

(9th Cir. 2018).

396. See Malone, 646 N.E.2d at 151; Orion Ins. Grp., 2017 WL 3387344, at *2.

397. United States v. N.Y.C. Bd. of Educ., 85 F. Supp. 2d 130, 153 (E.D.N.Y. 2000).

398. Elective Race, supra note 61, at 1533–35; see also Racial Commodification and

the Promise of the New Functionalism, supra note 198, at 192 (2013) (citing studies

showing that factors such as “class, history of imprisonment, and other experiences of

social marginalization can trigger multiracials to ‘choose’ to claim a minority iden-

tity”); see also Tehranian, supra note 291, at 3, 7–8 (explaining that racial identities are

not stable both because the categories themselves change and because individuals’

racial identities may change in relation to other aspects of their identity, such as their

political affiliation).

399. See Orion Ins. Grp., 2017 WL 3387344, at *2.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 76 10-MAR-22 14:18

342 TEXAS A&M LAW REVIEW [Vol. 9

affect how they are racialized. Indeed, a police officer sued for racial

discrimination based on the anti-Black workplace harassment he

faced after sharing his Ancestry.com results, which revealed he was

18% African.

400

Thus, changes in racial self-identification are not nec-

essarily indicative of fraud and do not mean that individuals should be

deemed ineligible for benefit laws.

A final common thread in both Orion and Malone that merits dis-

cussion is their use of ancestral evidence to determine eligibility for

these benefit laws. In Malone, the hearing officer affirmatively asked

the brothers to provide this evidence, and in Orion, Taylor provided it

on his own to support his racial claims. But these adjudicators did not

appear to use this ancestral evidence to determine how these appli-

cants were racialized. Although ancestry certainly plays a role in

racialization through the inheritance of certain racialized features, like

skin color, hair color, etc., neither Malone nor Orion used ancestral

evidence as a proxy for physical appearance. In both cases, physical

appearance and racial phenotype were considered via other forms of

evidence. Ancestral evidence could also have affected these appli-

cants’ racialization if they revealed their ancestral backgrounds and

were subsequently assigned a racial identity based on that informa-

tion. But that’s not how ancestral evidence was used in these cases

either. Rather, the ancestral evidence seemed to provide these appli-

cants with an alternative way to prove their racial claim—a path sepa-

rate and apart from how they were racialized. In Malone, if the

brothers had been able to prove that their grandmother was Black,

then they might have been eligible for the affirmative action program,

regardless of how they were racialized. And in Orion, Taylor’s case

may have come out differently if he was more than 4% African. An-

cestral evidence could have also been treated as an additional require-

ment for eligibility, that is, Taylor and the Malones would need to

have a certain percentage of African ancestry to qualify, regardless of

other identity evidence.

If the remedial purpose of these laws was permitted to be broader

and capable of addressing how the historical subordination of racial

groups across education, employment, healthcare, housing, etc., has

led to current racial disparities, ancestral evidence could be necessary

to fulfill the purpose of these laws. If the law were able to recognize

how, for example, descendants of slaves have inherited hundreds of

years of systemic disadvantage, then examining whether someone may

be a descendant of an enslaved person would be crucial. But as dis-

400. See Complaint at 5, Brown v. City of Hastings, No. 17-cv-0033 (W.D. Mich.

2018). The case was settled after the officer filed a complaint. Nicole Rojas, White

Police Officer Receives $65,000 Settlement from City in Racial Discrimination Suit,

EWSWEEK

(Aug. 1, 2018, 12:05 PM), https://www.newsweek.com/cleon-brown-has-

tings-police-michigan-racial-discrimination-settlement-1052977 [https://perma.cc/

66SN-XDDQ].

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 77 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 343

cussed, the Supreme Court has barred broader and more historically

conscious remedial functions of benefit laws. Thus, examining ances-

tral evidence does not necessarily align with the purposes of these

laws.

4. Problems Caused by Context-Detached Adjudication

The harms from context-detached adjudication in benefit laws mir-

ror those from anti-discrimination laws. First, a context-detached ap-

proach causes doctrinally inconsistent outcomes. For instance, the

identity inquiry in asylum claims is sometimes based on how the peti-

tioner’s identity will be understood in their home country and, other

times, based on whether they “really” are the identity they claim to

be. Context-detached adjudication also produces results inconsistent

with the purpose of asylum law—asylum law is meant to protect peti-

tioners, like Fuller and Mockeviciene, who were likely to be perceived

as not straight in their home country and persecuted on that basis.

Another example of inconsistency arises with sex adjudication—

whether someone is legally male or female for the purposes of a legal

sex designation on a birth certificate depends on the state in which

someone lives. Someone could be legally male in one state and legally

female in the next. Similarly, an athlete’s sex could be “female” under

the NCAA sport-specific rule, and “male” under Idaho’s law.

Context-detached identity adjudication also leads to over-interroga-

tion of identity and raises concerns similar to those raised in the con-

text of anti-discrimination laws. First, using more identity evidence

than called for to adjudicate identity can raise privacy concerns. For

example, if ancestral evidence for racial adjudication isn’t necessary

for a particular law, as suggested above, examining someone’s ances-

tral background, via their results from service like 23andMe or other-

wise, is unnecessarily intrusive. Relatedly, dignity interests may be

threatened when a state actor conducts a visual inspection of an indi-

vidual and then ascribes them a racial identity; if the relevant question

is not how a single state actor perceives someone’s race, there is no

need to risk dignity harms.

401

In the context of sex, requiring invasive and expensive surgery and/

or hormone usage to change a legal sex designation on identity docu-

ments has obvious implications for people’s privacy and dignity.

402

401. See Racial Commodification and the Promise of the New Functionalism, supra

note 198, at 193 (warning that census officials using third-party observations to deter-

mine a person’s race may result in “racially categorizing an individual in a way that

fundamentally contradicts the individual’s own understanding of her race”); see also

Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 797 (2007)

(Kennedy, J., concurring in part) (arguing that forcing individuals to “live under a

state-mandated racial label” is “inconsistent with the dignity of individuals in our soci-

ety” because it assigns labels “that an individual is powerless to change”).

402. Other scholars have made similar arguments. See, e.g., Holning Lau, Gender

Recognition as a Human Right in

AMBRIDGE

ANDBOOK ON

UMAN

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 78 10-MAR-22 14:18

344 TEXAS A&M LAW REVIEW [Vol. 9

Rules mandating hormone therapy to play sex-segregated sports in

college raise similar questions, and Idaho’s definition of “female” for

sports subjects individuals to particularly invasive inspections. When

any student on an all girl’s/women’s team has their sex challenged,

they must agree to pelvic examinations to “verify” their “reproductive

anatomy,” genetic testing to determine their chromosomes, and/or

hormone testing to remain on the team.

403

This law also harms trans-

gender students’ dignity and autonomy interests by assigning them a

sex identity that does not align with their self-identification and forc-

ing them into spaces and activities that do not reflect their identity.

404

This is not to say that privacy and dignity interests should always

outweigh the interests underlying the particular law at issue. But when

identity adjudication does not serve the interests of the underlying law

(i.e., when identity adjudication is context-detached), these interests

are being harmed for no reason.

Also, over-interrogation can lead to legal actors reinforcing prob-

lematic definitions of identity. For example, both the Mockeviciene

and Fuller courts constructed factually incorrect and overly restrictive

definitions of lesbian and bisexual identities—the former by holding

that the petitioner was not a lesbian because she had been married to

a man and hadn’t been in a relationship with a woman and the latter

by holding that the petitioner was not bisexual because of his relation-

ships with women.

405

In the context of sex, requiring that someone has

certain sex characteristics to change their legal sex designation or play

sex-segregated sports helps construct and reinforce the dominant nar-

rative that is sex binary, fixed, and defined by certain body parts. As

discussed, this construction of legal sex has many harmful conse-

quences for transgender and intersex people, but also for cisgender

people through its role in reinforcing restrictive sex stereotypes.

406

IGHTS

: R

ECOGNITION

, N

OVELTY

, R

HETORIC

194 (Andreas von Arnauld, Kerstin

von der Decken & Mart Susi eds., 2020).

403.

DAHO

ODE

§ 33-6203(3); Hecox v. Little, 479 F. Supp. 3d 930, 985–87 (D.

Idaho 2020) (holding that the Act subjects students who try out for the girls’ soccer

team “to the possibility of embarrassment, harassment, and invasion of privacy

through having to verify her sex” and that “[s]uch violations are irreparable”).

404. See, e.g., Evancho v. Pine–Richland Sch. Dist., 237 F. Supp. 3d 267, 294 (W.D.

Pa. 2017) (describing how exclusion of transgender students from restrooms that

matched their identity “caus[ed] them genuine distress, anxiety, discomfort[,] and

humiliation”).

405. See generally Mockeviciene v U.S. Att’y Gen., 237 F. App’x 569, 570 (11th Cir.

2007); Fuller v. Lynch, 833 F.3d 866, 872 (7th Cir. 2016); see also Robinson, supra note

27, at 1314 (making a similar argument about how jail officials’ determinations re-

garding inmates sexuality reflect and create narrow and problematic definitions of gay

identity).

406. See supra Subsection IV.B.3. As mentioned, I will explore these harms, and

others flowing from over-interrogation of identity, in my future work.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 79 10-MAR-22 14:18

2022] ADJUDICATING IDENTITY 345

V. C

ONCLUSION

This Article understands identity adjudication as a legal phenome-

non operating across identities and areas of law and has attempted to

make sense of this phenomenon by developing a taxonomy through

which to understand the factors legal actors use to determine some-

one’s identity—factors that these actors are not making explicit and/or

even consciously using. Examining identity adjudication in this broad

way reveals systemic problems with how legal actors understand and

adjudicate identity. This Article has examined one of those

problems—a context-detached approach to determining identity—

and offered an alternative—a context-informed approach. Rather

than approaching identity as a stable fact or characteristic about

someone, legal actors should begin by asking why identity matters to

the law in the first place. Perhaps a particular legal rule does not re-

quire identity adjudication at all (like anti-discrimination law). And if

the legal rule does require determining someone’s identity, the defini-

tion of identity and the proof requirements of an identity claim should

be informed by the purpose of the specific law at hand. A context-

informed approach would also bring identity adjudication to the sur-

face and force legal actors to make the fact of, and their reasoning

behind, identity adjudication explicit. Additionally, because the law

would be disciplining the identity inquiry, there would be less room

for legal actors’ subjective or conscious ideas, biases, and stereotypes

about particular identities to taint the identity determination.

This is not to say that context-informed identity adjudication will

always produce just, fair, or equitable results. The normative value of

context-informed identity adjudication is tied to the normative value

of the law at issue. For example, a context-informed approach to adju-

dicating identity for a law that limits citizenship to only those individu-

als who qualify as “white” would produce outcomes that further

entrench racial subordination. However, adopting a context-informed

approach would limit the normative assessment to the purpose of the

law and would eliminate many of the problems that flow from legal

actors making identity determinations based on their own subjective

ideas about identity rather than the particular law at issue.

\\jciprod01\productn\T\TWL\9-2\TWL206.txt unknown Seq: 80 10-MAR-22 14:18