Are Women’s Spaces Transgender Spaces? Single-Sex Domestic Violence Shelters, Transgender Inclusion, and the Equal Protection Clause

Are Women’s Spaces Transgender Spaces? Single-Sex Domestic Violence Shelters, Transgender Inclusion, and the Equal Protection Clause

Transgender survivors of intimate partner violence (IPV) face unique struggles in finding safe and inclusive housing as they seek reprieve from violence. Domestic violence shelters are often marked “women-only” with the goal of creating spaces for female empowerment, wherein women learn feminist principles of liberation and find a “sisterhood” of support by forging healthy female relationships. However, as a result, shelters frequently deny transgender women access because staff perceive them to be a threat to survivor comfort and to be disruptive to shelters’ female-empowerment model. Consequently, though transgender women face similar gender-based oppression and a relatively higher risk of violence as compared to cisgender women, shelters commonly deny transgender women equal protection. This Note conceptualizes how a Fourteenth Amendment equal protection challenge by transgender litigants to women-only shelters might proceed in federal courts. By situating transgender identity within the Supreme Court’s broader equal protection jurisprudence, it outlines three ways that the Court could analyze a transgender equal protection challenge: as an issue of first impression, as a sex-based discrimination claim, or as a sexual orientation claim.


“I once worked with a woman who was transgender, and whose partner had almost killed her. She had finally made the decision to leave the relationship and she went to a shelter in Massachusetts. When she got there, the counsellors were confused about her gender even though she had previously explained to them that she was transgender, and what that meant. The shelter staff asked her a set of intensive and grueling questions about her body including, ‘What is between your legs?’ . . . after this humiliating treatment, they told her that she could not be housed there because they decided that she was really a man. After being denied shelter, this woman went back to her batterer because she had no family, no friends and nowhere else to go.”––Emily Pitt, Director, Fenway Community Health’s Violence Recovery Program[1]

Feminist movements from the turn of the twentieth century have made essential reforms to domestic violence law and policy. But women of color and lesbian, gay, bisexual, and transgender (“LGBT”) activists have long called attention to the limitations of a movement that primarily focused on the needs of cisgender,[2] white women.[3] These critics advocate for a more intersectional feminist approach to IPV that takes into account the multiple sources of oppression faced by marginalized groups—including transgender[4] women. This tension reverberates in debates on how best to structure anti-violence shelters that house those seeking reprieve from abusive relationships. Domestic violence shelters are often marked “women-only” with the goal of creating spaces for female empowerment, wherein women learn feminist principles of liberation, engage with theories of male domination, and find a “sisterhood” of support by forging healthy female relationships. However, as a result, shelters frequently deny transgender women access because shelter staff perceive them to be a threat to survivor comfort and to be disruptive to shelters’ female-empowerment model. Consequently, though transgender women face similar gender-based oppression and a relatively higher risk of violence as compared to cisgender women, shelters commonly deny transgender women equal protection.

Federal agency policy and some state statutory provisions address this discrimination, but the Equal Protection Clause of the Fourteenth Amendment also presents a key opportunity to break down barriers to transgender inclusion. This Note conceptualizes how an equal protection challenge to women-only shelters might proceed in federal courts. By situating transgender identity within the Supreme Court’s broader equal protection jurisprudence, it outlines three ways that the Court could analyze a transgender equal protection challenge: as an issue of first impression, as a sex-based discrimination claim, or as a sexual orientation claim.

First, federal courts could conceptualize transgender identity as an issue of first impression and draw on the Carolene Products criteria to determine whether transgender people constitute a suspect class. Using the Court’s treatment of the poor and mentally disabled as a guide, this Note argues that courts will likely deny suspect class status to transgender people by reasoning that transgender identity is neither immutable nor discrete. Therefore, rational basis review, the most deferential standard of equal protection analysis, will likely apply and the women-only classification will likely pass constitutional muster. Given the strong likelihood of receiving rational basis review, transgender advocates should exercise caution in framing their case using the suspect class framework.

Second, courts might analyze the claim as a regular sex-discrimination claim. The Supreme Court’s sex-discrimination jurisprudence vacillates between striking down classifications that deny opportunities based on stereotypical gender roles and inadvertently using biological markers to justify some gender‑based classifications. Thus, if courts conceptualize a shelter’s exclusion of transgender women as discrimination against women, courts may either strike down the women-only classification because it rests on a stereotypical notion of biological womanhood or justify the classification on the basis that transgender women possess inherent biological differences.

Finally, courts might operationalize the “T” in LGBT, align gender identity with sexual orientation, and analyze the claim using the principles animating the Court’s gay rights cases. The Court’s LGB[5] jurisprudence does not follow traditional equal protection analysis. Instead, Justice Kennedy, the principal author of the Court’s gay rights cases, has used a conception of “animus” to strike down or uphold classifications that implicate the rights of gay and lesbian people. Within this framework, courts will not likely find that the exclusion of transgender people flows from animus if courts define that term as a bare desire to harm. If, however, courts adopt a more capacious understanding of animus—that is, one that problematizes the ways in which private biases and anti‑transgender stereotyping demean transgender people—they may be more likely to strike down a shelter’s women-only classification. Transgender rights advocates should thus encourage courts to adopt an expansive conception of animus to successfully invoke the Equal Protection Clause on behalf of their clients.

I. History and Background

Controversy around the inclusion of transgender women in women-only domestic violence shelters is not new. Rather, it reflects the ongoing tension between the early feminist movement and critics who argue that the theory and advocacy informing women’s organizing in the 1970s must reform to foreground the experiences of the most marginalized women. The transgender rights movement represents one such call for reassessment. Part I frames the debate between early second wave feminists, who stood at the forefront of the initial domestic violence movement in the United States, and their critics—intersectional feminists and LGBT activists—to illuminate the philosophical orientations and policy platforms of each side. It concludes that earlier strains of the domestic violence movement frequently lapsed into gender essentialism[6] because they (1) focused on male dominance as the primary cause of gender‑based oppression, but did not adequately take into account that women occupy the intersections of multiple oppressions; and (2) made law and policy with the needs of the heterosexual, cisgender survivor in mind. These limitations merit attention for the purposes of transgender women’s equal protection claims because, as this Note demonstrates, they undergird legal justifications for excluding transgender women from women-only shelters. To avoid lapsing into gender essentialism, the domestic violence movement must foreground the feminist principle of intersectionality and reform domestic violence law and policy in a manner that addresses the needs of transgender women. Similarly, legal advocates must rely upon intersectionality theory to demonstrate that women-only shelter policies rely upon generalizations and stereotypes about the female identity that the Equal Protection Clause typically does not tolerate.

A. Dominance Feminism: Philosophy, Tools, and Impact on Domestic Violence Laws

Contemporary domestic violence law and policy is heavily influenced by the philosophical tone and prescriptive tools advanced by a strand of the mid‑1970s feminist movement called “second wave feminism” or “dominance feminism.”[7] Though the movement has some variations, second wave feminists generally regard domestic violence as an outgrowth of patriarchal domination, wherein sexism or gender-based oppression primarily drives intra-family violence.[8] The most essentialist strain of dominance feminism attributes patriarchal domination to an “anatomical fiat,”[9] thereby assuming that the “superior physical strength of men[] allows them to use violence to capture and rape women.”[10] However, more critical veins attribute domestic violence to sexism produced by socially learned roles: whereas batterers are “over-socialized males who rigidly adhere to sexist patriarchal values,” battered women are “expected to be pretty and ladylike and grow up to marry nice young men who would care for them as their fathers had.”[11] Similarly, Professor Catherine MacKinnon, a leading legal scholar in the second wave feminist movement, argued that “our male-dominated society, aided by male dominated laws, had constructed women as sexual objects for the use of men.”[12] Thus, the focus on domestic violence as female subordination resulted in a movement that conceptualized IPV “primarily as a heterosexual, sociopolitical phenomenon with its basis in sexism, that is, gender.”[13]

Dominance feminism’s focus on oppressive gender roles was “tremendously effective” in highlighting the way in discrimination and violence against women in society were reinforced through battering, the “at-home version of that ‘gender oppression.’”[14] In other words, feminists succeeded in pushing domestic violence beyond the silence of the home by establishing the now well-known credo that “the personal is the political.”[15] A slew of reforms informed by dominance feminism’s focus on the heterosexual male-dominated household ensued, including practitioner guides that espoused theories of victimization to explain the dynamics of intra-family violence and legal protections for women experiencing IPV.[16]

Dominance feminism’s focus on aggressive males and subversive females resulted in models of IPV that focused on the effect of sex-role stereotyping on women and men. Professor Lenore Walker’s “Learned Helplessness” theory is informative here. Professor Walker argued that women in violent relationships cannot leave because they experience a psychological phenomenon called “learned helplessness,” wherein repeated battering makes women passive by convincing them that any response they mount will be unsuccessful.[17] In particular, drawing on dominance feminism’s focus on sex-role stereotyping, she posited that this passivity represents a response entrenched in women—and in particular, married women—through “parental and institutional conditioning” that restricts them from seeking out alternatives to the abusive relationship.[18] Professor Walker also proposed a “Cycle of Violence” model to understand domestic violence: a tension-building phase gives way to a battering incident and a subsequent honeymoon period in which the batterer displays his best behavior.[19] Absent external intervention, the cycle repeats itself.[20] Second wave feminist thought also influenced batterer interventions, relying on dominance feminism’s theory that domestic violence stems from male-specific forms of power and control. The Duluth Model exemplifies an early batterer intervention program premised on alerting men that they “choose violence towards women as a way to establish power and dominance, in response to societal pressure and acceptance.”[21]

State-based responses to IPV adopted dominance feminism’s focus on male dominance as the root of domestic violence.[22] Indeed, learned helplessness theory, by positioning women as passive, specifically prescribed state intervention as a way to extricate women from violent relationships, both through domestic violence shelters and law enforcement mechanisms.[23] Thus, while women themselves sought out domestic violence shelters, and, more often than not, called law enforcement on their own initiative, the policies that ensued after the woman entered the system overlooked these active help-seeking mechanisms. Instead, they still conceptualized the survivor as helpless, focusing largely on conditioning the survivor to leave the relationship, rather than engaging with the needs of individual women.[24] For example, in a study of domestic violence shelters in New Mexico, Professors Krishnan, Hilbert, McNeil, and Newman found that earlier shelters saw themselves primarily as places of “transition” (where women prepare to leave their batterers) rather than places of “respite” (where women can rest until they are ready to go back to their batterer stronger if they choose to do so).[25] The authors recommended that shelters “revisit their ideology and expand their definition of . . . service provision both from the point of view of transition as well as respite” to be “more realistic and effective in addressing the differential needs of all their residents.”[26] Mandatory arrest policies[27] and no-drop prosecutions[28] represented a more extreme example of this type of thinking in that they prioritized separating the abuser and survivor, irrespective of the survivor’s wishes.[29] Meanwhile, lawyers operationalized the “Battered Women’s Syndrome” defense to justify the homicidal acts of women who killed their batterers as an “irrational” response to repeated bouts of violence.[30] Similarly, legal actors learned to look for the dynamics identified by Walker’s Cycle of Violence when analyzing whether domestic violence occurred in a case.[31] Last, states integrated the Duluth Model into legislative fixes for domestic violence; as of 2007, it represented the “most commonly state-mandated model of intervention,” and in many states, it was the “only statutorily acceptable treatment model.”[32] In this way, state domestic violence policies adopted dominance feminist approaches in concrete ways.

Though the second wave feminist movement succeeded in making key gains for the domestic violence movement, its limitations also became clear. In particular, intersectional feminists and LGBT activists focused on the ways in which dominance feminist reforms paid little attention to other marginalized identities, including women of color and gender minorities. These tensions merit highlighting because they frame opposite sides of the debate regarding transgender inclusion in women-only domestic violence shelter.

B. Criticism from Within: Intersectional Feminist Response

As the second wave feminist movement matured, women of color and LGBT activists highlighted the limitations of the “dominance and control” approach to oppression and opened the movement up to meaningful criticisms. Intersectional feminism,[33] one such response, took issue with many of the second wave feminists’ core claims. First, it decentered male dominance as the sole cause of domestic violence and instead focused on the way in which women face several sources of oppression.[34] In this sense, it expanded the source of women’s oppression from merely sex-stereotyping to consideration of economic factors, gender identity, sexual orientation, race, disability, and nationality.[35] Intersectional movements further focused on movement building across issue areas. For example, Lee Jacobs Riggs, writing about rape reform, argued that a successful anti-rape movement will not conceptualize rape as only the product of male dominance, but also as reflecting myriad other oppressive systems, including white supremacy.[36]

Second, intersectional feminists and LGBT activists problematized the way in which a stereotype of the cisgender white woman and the male-female dyad informs much of second wave feminist organizing. These criticisms coalesced into a broader concern: that stereotypes regarding what the typical domestic violence dynamic looks like and who a typical batterer is will result in law and policy that fail to respond to the needs of individuals falling outside these molds. As Professor Donna Coker argued, law and policy informed by the experiences of “a generic category of ‘battered women,’ is likely to reflect the needs and experiences of more economically advantaged women and white women, and is unlikely to meet the needs of poor women and women of color.”[37] Accordingly, intersectional feminists rejected the “learned helplessness” approach, arguing that the theory lapses into essentialism in two ways: first, by painting all women with the victimhood brush and, second, by discounting the differences in how women respond to such violence. As Professor Leigh Goodmark argued, “Domestic violence does not transform every woman who experiences it into a stereotypical victim, nor should this victim stereotype shape domestic violence law and policy.”[38] Instead, Professor Goodmark insisted that domestic violence analysis should “delve into the complexities of the lives of individual women and consider the totality of who they are” rather than just reducing them to their lowest common denominator—women who experience domestic violence.[39] To this end, Professors Gondolf and Fisher rejected the idea that those in intimate partner relationships remain “psychologically paralyzed” and instead argued that women respond to abuse with a diverse range of “help-seeking efforts that are largely unmet.”[40] The “survivor theory” accordingly suggested that the state need not “drag” people out of domestic violence. Rather, survivors themselves, responding to an escalation in violence, failed intervention efforts, or new financial resources, will actively seek to leave.[41] Under this account, the state needs to respond to that call through flexible services that prioritize the individual needs of the survivor,[42] instead of depending on broad generalizations regarding those experiencing IPV.[43]

These criticisms have particular relevance to the LGBT community. Scholars studying the intersection of sexual orientation, gender identity, and IPV have exposed the way in which the second wave feminist focus on the male‑female dyad and heterosexuality excludes the unique forms of IPV that proliferate in the LGBT community. The sparse research addressing IPV within the LGBT community complicates the simplistic male-perpetrator and female‑victim dyad that informs second wave feminist organizing.[44] Yet the male-abuser and female-victim narrative remains so strong that it sometimes blinds even trained advocates to signs of LGBT IPV.[45] For example, a lesbian survivor claimed that she struggled to realize she was in a violent relationship because she adopted the societal belief that women are passive and nonviolent—a perception that her training at a domestic violence shelter only reinforced during.[46] Similarly, gay men experiencing abuse may feel pressured to paint their partner in hyper-masculine terms to have their domestic violence claims taken seriously.[47] Second wave feminists’ focus on patriarchy as male domination and control similarly leaves the movement unable to account for the unique types of abuse faced by LGBT survivors. For example, Calton, Gabbard, and Catteaneo argued that “LGBTQ IPV can include power and control tactics that are specific to minority sexual orientation or gender identity, including threats of disclosure and the use of homophobia.”[48] Research that surveyed domestic violence providers confirms this narrative, finding that more than 50 percent of respondents did not receive training on same-sex IPV and did not advertise LGBT services in community outreach.[49]

Second wave feminist responses to domestic violence render transgender people particularly vulnerable to erasure. Very little research addresses IPV among transgender people. Most studies exclude transgender people or rely on survey takers with binary gender categories of man and woman.[50] However, the studies that do exist show that transgender people face a significantly greater risk of experiencing IPV. One study, for example, indicated that whereas 20.4 percent of cisgender people reported experiencing IPV, the number rose to 31.1 percent among transgender people.[51] Other studies put the latter estimate at closer to 50 percent.[52] As Professor Goodmark argued, “[i]f intimate partner abuse is in large part about controlling and enforcing gender norms within relationships, transgender people, by virtue of their failure to conform to such norms, are particularly vulnerable to abuse.”[53] Indeed, while development of transgender identity differs among individuals, the process itself subjects many transgender people to abuse from intimate partners.[54] For example, transgender people may identify as transgender with a partner without coming out publicly and, as a result, the abuser may threaten to out them to others before they feel ready to disclose their identity and make associated changes, such as using their preferred pronouns, changing their name, changing their outward appearance or expression, and undergoing physical procedures.[55]

The larger social context within which transgender people experience abuse, however, suggests that they face extreme difficulty in accessing resources. Negative attitudes towards transgender people are even more severe than towards lesbian and gay couples.[56] Transgender people also face a much higher risk of assault and physical violence: according to Professor Stotzer’s review of the literature on violence against transgender individuals, 48–69 percent of transgender respondents had been harassed and 20–86 percent had been physically assaulted or beaten as a result of their gender identity or gender expression.[57] Ample literature further indicates that transgender people suffer a greater risk of unemployment and homelessness.[58] This context suggests that transgender people may hesitate to reach out to state resources and, even when they do, the state may not adequately respond to their help-seeking behavior.

Despite the obvious vulnerability of transgender people to abuse, the women’s movement has not been very receptive to calls for inclusion. Responses range from open-minded reflection, defensiveness, fear, and outright hostility. Professor Judith Halberstam observed, for example, that some lesbian feminist groups perceive female-to-male (FTM) transgender people as traitors.[59] As a concrete consequence, FTM transgender people that already used the services of women’s organizations prior to transitioning risk losing their employment or access to services as a result of identifying as male.[60] Transgender women face similar hostility: for example, the Michigan Womyn’s Music Festival, an international cultural event during which the debates of the larger Western feminist communities played out and informed future organizing efforts, explicitly excluded transgender women by creating a space for “womyn born womyn.”[61] Prominent feminists, including Janice Raymond, Mary Daly, and, in the 1970s, Gloria Steinem, took the position that the existence of transgender women merely reinforced gender normativity because they were constructed by the medical establishment, which itself was a patriarchal institution.[62] Sheila Jeffereys similarly argued that “transsexualism pathologizes the body and subsequently mutilates it, in order to satisfy societal codes of gender, sex, and sexuality.”[63]

These trends demonstrate the way in which second wave feminism falls subject to its own essentialism. Specifically, in creating a singular male-abuser and female-victim narrative, it reinforces the very gender stereotypes it seeks to dismantle. As a service provider, Allison Cope, observed,

[A]ll that we’ve talked about social construction of gender seems to go right out the window and it all comes down to estrogen and ovaries and breasts and that’s about all; that’s who we are. So we need to get beyond that; we need to get beyond biological determinism which seems to have rooted itself in feminism and once we do that we’re going to really be freed up.[64]

As Cope highlighted, overcoming this lapse into biological determinism and gender essentialism means staying true to feminist principles of inclusion and intersectionality by including transgender women in women-only spaces. Yet because early versions of dominance feminism have informed so much of domestic violence law and policymaking, practices are only now catching up to incorporate criticisms advanced by LGBT activists and women of color. Domestic violence shelters, the cornerstone of IPV services, exemplify this process of reform.

II. Domestic Violence Shelters & The Case For Transgender Inclusion

IPV shelters are more than just temporary housing spaces. The typical shelter allows women to stay for thirty to sixty days, conceals their locations from abusive partners, and provides counselling and referrals that respond to women’s help-seeking behaviors. Most shelters offer services that include accompanying a survivor to a hospital, helping her find a new apartment, applying for welfare, or getting a civil protection order.[65] Because IPV shelters represent such an integral part of IPV services, it is especially important for transgender litigants to gain constitutional protections that ensure more equitable access to these spaces.

Part II of this Note outlines the historical background of the shelter movement and, within that context, elaborates upon the arguments provided by some service providers for excluding transgender women from women-only shelters. The Note concludes that the arguments provided in favor of excluding transgender women may be rooted in genuine concerns for survivor safety and comfort, but ultimately rest on gender essentialism assumptions about the typical IPV survivor. First, society, perhaps unintentionally, conflates MTF transgender experiences with those of cisgender males, while discounting MTF transgender experiences as “women’s experiences.” Because shelters’ empowerment models for survivors are focused on tapping into an authentic female experience, this conflation allows shelters to rationalize the exclusion of transgender women on the grounds that it disrupts attempts to help survivors. Yet the material and social conditions of cisgender women and transgender women actually suggest that they are somewhat similarly situated to women in their experiences of gender marginalization. Second, shelters argue that allowing transgender women to enter women-only shelters would pose a safety threat to women who are traumatized by the presence of males in the wake of battering. However, this rationale rests on the stereotype that transgender survivors are inherently threatening, and cisgender women are inherently vulnerable. Indeed, as the experiences of transinclusive shelters demonstrate, there were no safety issues with including transgender people in shelters. The seriousness with which the Court considers shelters’ rationales for excluding transgender women will affect its decision to uphold the exclusion or strike it down as violating the Equal Protection Clause.

A. The Rise of the “Shelter Movement”

Early strains of the feminist movement saw women-only shelters as critical to helping IPV survivors reassert control over their lives after experiencing abuse. In the early 1970s, second wave feminist organizing at the grassroots level established the right for women to extricate themselves from abusive relationships and, accordingly, established the first network of shelters for women facing IPV.[66] As Professor Haaken noted, the shelter movement had deep roots in the broader feminist movement and, thus, “[s]helters became the ‘womb’ of feminism—maternal spaces of protection, guided by egalitarian principles rather than hierarchical or paternalistic models.”[67] For example, the first Boston shelter, Transition House, was “influenced by women’s liberation ideas.”[68] The founders, survivors of IPV, joined with members of Boston’s earliest radical feminist groups to establish a space that “encouraged [women] to explore their personal lives” and examine “the political parameters of ‘private’ problems.”[69] In so doing, they aimed to reframe battering from an “isolated fact of daily existence” into the larger phenomenon of “women’s oppression.”[70] Relatedly, advocates argued that the women-only environment created healthy female relationships.[71] As psychologist Cheryl Beardsley argued, “[f]or many [survivors], bonding with other women is a novelty; abused women are often isolated by their abusive partners and taught to be jealous of other women.”[72] Studies showed that these relationships became sources of support as survivors worked through trauma and navigated everyday practicalities of living with IPV. For example, an anthropological account of a Minneapolis women’s shelter documented how residents “discuss[ed] the importance of getting an apartment above the ground floor, so he won’t come through the window; of the best kind of window bars; of varying their routes to work each day.”[73] In this sense, second wave feminists saw women-only shelters as crucial tools that helped women reassert control over their lives in the wake of battering.[74]

B. Transgender Exclusion & The Case for Inclusion

Domestic violence shelters have faced the same criticisms levied against the broader IPV movement. First, though second wave feminists instituted shelters with the vision of “sisterhood” in mind, some Black feminists, including bell hooks, criticized this idea of “sisterhood,” and pointed out the historical exclusion of women of color from mainstream feminist spaces.[75] In the context of shelters, this dynamic plays out in subtle ways, since women of color are not actively excluded but struggle to find shelters that meet their language and cultural needs. For example, Professors Fisher, Gondolf, and McFerron argued that Latina and Hispanic women were, at times, not able to find shelters that could accommodate them due to language difficulties, fears of having their immigration status outed, limited mobility because of larger families, lower personal incomes, and experiences with discrimination within the shelter itself.[76] Indeed, in her seminal article Mapping the Margins, Professor Kimberle Crenshaw recounted multiple accounts of Latina survivors being denied access to shelters in New York during the 1990s because they lacked English proficiency.[77] The shelters stated that the survivors could not adequately partake in mandatory counselling programs at the shelter, but refused to provide translators or even have children translate for their mothers. The shelter reasoned that using the son “further victimized the victim” and that past experiences indicated that these women tended to leave early anyway.[78] Similarly, Professor Lisa Martinson observed a perception “among African-American women that ‘shelters and institutions established to help battered women are only for the needs of white women.’”[79] Indeed, as Professor Haaken and Yragui’s interviews with domestic violence shelter providers revealed, Black women have briefer stays than white women and are sometimes “asked to leave for behavior that is perceived by white staff and residents as violent,” including “talking loud or yelling.”[80] Accordingly, Black survivors may determine that “it will be better for them to stay in the abusive situation rather than face racial discrimination in a shelter.”[81]

Sometimes, well-intentioned policies also negatively affected women of color survivors. For example, early shelters were concealed to prevent batterers from finding and stalking survivors, but this prevented women of color from accessing them.[82] Some women of color saw the secretiveness of the shelter as “going into hiding” in a way that meant the loss of “more primarily cultural alliances.”[83] A practitioner argued that “secret shelters don’t work for women of color. . . . They are too isolating, and they cut you off from your community.”[84] In establishing their own domestic violence shelter called “House of Hope” after facing “stereotyped expectations” and language barriers at other shelters, a group of Latina women specifically left the shelter location open to the public and enlisted the help of the community to ensure safety.[85]

Whereas the shelter movement was somewhat receptive to these criticisms levied by women of color, shelters still struggle to accommodate the LGBT community and, particularly, transgender people. According to attorney Wayne Thomas, the inability to access secure shelter remains the single most important issue facing transgender people subjected to abuse.[86] Indeed, access to legal help often depends on finding a shelter first because the shelter provides information about legal options and provides emotional support to IPV survivors as they exercise their legal rights. Despite this importance, the 2011 National Transgender Discrimination Survey uncovered rampant discrimination in shelter access: of those who sought shelter, 55 percent suffered harassment by shelter staff and 29 percent were turned away because of their gender presentation.[87] Some shelters either flatly refused to admit transgender people or would admit only transgender people who passed as a woman successfully.[88] As activist Viviane Namaste recounted, “I was informed that an MTF transsexual would be accepted into some shelters ‘if the person doesn’t come across as too terribly masculine.’ Staff people claimed that the physical appearance of transsexual women was related to their ability to ‘fit in.’”[89] Shelters thus operate a “hierarchy of inclusion,” wherein the degree to which the survivor matches hegemonic constructions of woman or man within traditional gender classifications determines whether they can access space.[90]

These access problems have devastating effects on transgender survivors. Feminist Ryka Aoki argued, for example, that denying shelter space to transgender people “tacitly condone[s]” violence by forcing transgender survivors to stay in abusive relationships out of the fear that they will not be able to find welcoming shelter space.[91] This narrative of a transgender woman accessing shelter space illuminates this point:

The shelter staff asked her a set of intensive and grueling questions about her body including, ‘What is between your legs?’ . . . after this humiliating treatment, they told her that she could not be housed there because they decided that she was really a man. After being denied shelter, this woman went back to her batterer because she had no family, no friends and nowhere else to go.[92]

Victoria Cruz, a transgender woman, similarly explains,

If I am a victim of domestic violence and need to go someplace, I have no place to go, because male-to-female [transgender] survivors are funneled into the men’s shelter system. I don’t have to tell you what would happen there. My most vulnerable episodes there would be when I need to take a shower or go to the bathroom. I would be revictimized then not only by the residents, but also by the service providers.[93]

This fear is very concrete for many transgender people who have already been previously marginalized by other facets of the state system. Attorney Morgan Lynn, for example, recounted that one of her MTF transgender clients was sent to a men’s hotel when attempting to access an emergency shelter space.[94] She received a call her from her client, who was crying that she would have been better off with her abuser.[95] Because access to legal help often depends on access to shelter space and all the resources offered therein, exclusion from shelter space impairs transgender people’s help-seeking behavior.[96]

Despite this stark reality, shelter operators offer several rationales for excluding transgender women, often citing genuine concerns for survivor comfort and safety. For instance, a group of service providers writing on domestic violence policies offered the following justifications for maintaining ciswomen-only shelter spaces: (1) MTF transgender people can never be women because they were socialized as boys and men and, consequently, exercise male privilege; (2) male privilege is intrusive to women-only spaces; (3) there are valid differences between transgender women and cisgender women that justify differential treatment; (4) women’s organizations are still trying to meet the needs of lesbians, women of color, Aboriginal women, and women with disabilities; and (5) the admission of transgender women will inevitably lead to the inclusion of men who will simply claim that they are “female in a man’s body” and use that justification to stalk and harm the survivor.[97] Other IPV service providers have, in refusing to grant transgender women access, cited a “fear” that transgender women would disrupt “the safety and comfort of other residents.”[98] Drawing from her own experiences with male abuse, Professor Nicki argued that to some survivors, the penis is weaponized and appears threatening.[99] She elaborated:

When I was a child and was being raped by my father I thought the penis was a sword, a potential instrument of death. Penises are not inherent instruments of rape, and pre-operative transsexuals are not simply potentially threatening people with penises. However, our culture is rampant with rape, with penises used for destruction. . . . As much as they would like to forget the trauma, the lack of avenues that lead away from it and the worldwide persistence of male domination and sexual violence keep survivors circling the trauma. Survivors may feel cognitively and emotionally unable to make a distinction between a benign penis and one used for harm.[100]

Thus, while reducing transgender people to their body parts represents exactly the type of essentialism that intersectional feminists seek to avoid, personal experiences and socialization processes will often leave survivors unable to disentangle a benign penis from one used for harm.[101] Even if one were to reject this argument of biological determinism, Professor Nicki proffered another reason why some transgender people may not belong in women-only shelters—because they have not “grown up with the publicly affirmed gender identity of a woman and with all the experiences and repercussions that this brings.”[102] She concludes that transgender and women’s spaces must maintain boundaries of respect for these reasons.[103] Overall, these accounts reveal that transgender women are seen as “invaders” in these gender-segregated spaces.[104]

Tellingly, the justifications used by Professor Nicki and others closely parallel arguments advanced by domestic violence shelters facing equal protection challenges from cisgender men seeking access to women-only shelters. In Blumhorst v. Jewish Family Services of Los Angeles, for example, amici that filed briefs on behalf of a domestic violence shelter advanced the following arguments in favor of maintaining a women-only space. First, they argued that introducing men into these shelters would violate the privacy rights of women because battered women are “conditioned to fear all men” and thus, men’s presence would prevent the shelters from providing a “safe, private, and restorative” location for survivors.[105] This concern is particularly acute in the context of a shelter, since limited resources mean that residents “sleep together, bathe together, and eat together.”[106] Second, amici argued that introducing men would threaten survivors’ safety by enabling “batterers to claim to be victims to gain entrance into the shelter and use the opportunity to stalk, further abuse, or even kill their victims.”[107] Third, amici suggested that battered women’s shelters centered on the idea that “women should come together to heal and to provide a network of other women that understand the plight of the others around them.”[108] Thus, men would “detract from the women’s right to develop relationships privately that can contribute to an improved sense of self-identity and worth.”[109] Finally, amici argued that integrating men into women’s shelters would prompt women to turn away from shelters altogether, leaving them at a high risk of homelessness.[110] Given that these arguments tracked some feminists’ concerns about including transgender women in women-only spaces, similar justifications will likely arise when transgender litigants challenge their exclusions from women-only shelters.[111]

The rationales offered to exclude both cisgender men and transgender women closely parallel each other and ultimately collapse into three overarching concerns. First, transgender women and cisgender men lack the socialized experiences of cisgender women and would thus disrupt a shelter space and divert resources from already underserved cisgender women. Second, the presence of individuals with body parts associated with maleness—the penis, for example—will re-traumatize survivors. Third, the inclusion of transgender women and cisgender men poses a safety threat to survivors. It is dubious whether these rationales properly justify the exclusion of cisgender men,[112] and they certainly lapse into gender essentialism when used to justify the exclusion of transgender women.

First, the idea that battered women are “conditioned to fear men” and will be prone to “bond” with other women as a result of shared socialization processes lapses into gender essentialism because it assumes a singular, authentic “woman” experience or shared sisterhood. Similarly, the second rationale, that all survivors are conditioned to fear maleness, is a broad generalization. Both assumptions implicitly position women as ladylike victims and all men as aggressive, taking after the simplistic analysis of domestic violence that informed early variants of dominance feminism. As Part I.B discussed, LGBT activists and women of color have criticized such singular narratives for ignoring the experiences of people who do not conform to traditional conceptions of “woman.” That concern arises here, too: that nearly identical rationales have emerged to exclude both men and transgender women from shelters suggests that cisgender men’s identities somehow subsume those of transgender women, even though many transgender women identify, present, and have lived experiences as women. This reality complicates Woo’s argument that all transgender women possess “male privilege” and are “socialized as men.”[113] Nevertheless, because transgender women may deviate from what it means to be a cisgender woman in a male‑female dyad, their experiences are not seen as “women’s experiences.” Some feminists, therefore, consider transgender women to have an innate maleness that precludes them from being perceived as women.

This perhaps-inadvertent conflation between cisgender males and transgender women explains why service-providers privilege cisgender women’s concerns, while turning a blind eye to the needs of transgender women. They associate cisgender women with victimhood, but may implicitly regard transgender women as perpetrators rather than survivors. Interrogating that assumption, however, renders exclusionary rationales less airtight. In reality, transgender women and cisgender women survivors are similarly situated (though certainly not identical) by their struggles to cope with violence and to access appropriate services. For example, the well-established gender asymmetry in domestic violence, which demonstrates that women much more frequently suffer domestic violence than do men, applies with even greater force to transgender women: transgender people are much more likely to be subject to IPV than cisgender women.[114] A study of transgender people in Chicago revealed high rates of intimate partner abuse among those surveyed: 66 percent of respondents reported suffering violence in their homes.[115] Moreover, transgender women and cisgender women face similar struggles when attempting to access shelter space. This similar material realities of transgender and cisgender female survivors suggest that shelters could conceivably justify excluding cisgender men to better serve women, a population relatively more vulnerable to domestic violence, the same rationale cannot extend to exclude transgender women. Indeed, if anything, shelters should privilege the concerns of transgender women, given that statistics demonstrate they most acutely need IPV services.

Finally, shelters posit survivor discomfort, safety, and privacy as concerns that justify the exclusion of transgender people, but these, too, rest on gender essentialist notions. Yet if shelters adopt an intersectional feminist approach—that is, one that foregrounds the needs of women at the margins and structures services to respond to the needs of individual women—they can adequately address these concerns without resorting to blanket exclusions. First, safety has rarely been a problem for those shelters admitting transgender women: transgender women are no more prone to committing acts of violence than are cisgender women.[116] Second, instead of responding to genuine survivor discomfort of the kind Professor Nicki expressed through blanket exclusions, shelters could take a more intersectional feminist approach by making reasonable accommodations for individual survivors who feel traumatized by the presence of transgender women. This proposal responds to the needs of individual women, without stigmatizing an entire socially vulnerable group.

There thus exists an urgent call for law and policy to accommodate the needs of transgender women survivors of IPV. Applying legal pressure by creating a constitutional prohibition against transgender discrimination can encourage women-only state-sponsored shelters to make reasonable accommodations for transgender women. Part III discusses how potential legal claims based on the Equal Protection Clause may contribute to that end.

III. Transgender Identity & The Equal Protection Clause

Progress has been slow and fragmented, but policy makers are beginning to address the exclusion of transgender women from IPV services. As recently as September, 2016, the Department of Housing and Urban Development (HUD) published a final rule mandating shelter providers to ensure equal access to services. This mandate requires ensuring that “an individual is placed, served, and accommodated in accordance with the gender identity of the individual” and ensuring that the “individual is not subjected to intrusive questioning or asked to provide anatomical information or documentary, physical, or medical evidence of the individual’s gender identity.”[117] Notably, the rule puts the onus on the service provider to make reasonable accommodations based on health and privacy concerns.[118] The rule defines gender identity as “the gender with which a person identifies, regardless of sex assigned to that person at birth and regardless of the person’s perceived gender identity.”[119] Statutory and policy protections are also emerging at the state level. Nineteen states and the District of Columbia have public accommodation laws that prohibit discrimination on the basis of sexual orientation and gender identity.[120]

Yet these gains, while significant, still exhibit key gaps. First, not all shelter programs receive HUD funding, as many are funded by their state and local governments. Second, whereas some states have constitutional protection from discrimination on the basis of gender identity, others are silent on the subject, and some are hostile towards the rights of transgender people.[121] Third, though HUD’s most recent policy favors transgender people, agency interpretations have no precedential value and can thus change easily with shifts in executive power. Indeed, the administration of President Donald J. Trump has rolled back progress on transgender rights by banning transgender people from serving in the military and reversing course on education policy that allowed transgender students to use the school bathroom that corresponded with their gender identity.[122] This drastic departure from the previous presidential administration’s policies[123] highlights the limitations of protecting key transgender rights through administrative policy.

Where political will and differences in administrative policy frustrate the achievement of transgender rights, constitutional protections for transgender people present a key opportunity to enshrine uniform changes that transcend shifts in administrative policy and legislative responses to discrimination. The Fourteenth Amendment’s Equal Protection Clause represents a significant tool in this regard. This Part discusses below the opportunities the Equal Protection Clause offers transgender litigants seeking to secure permanent access to domestic violence shelters.

Though LGBT rights movements are slowly making incursions into the legal landscape, courts rarely discuss the “T” separately in legal opinions. Thus far, no transgender litigant has challenged or succeeded in striking down a facially discriminatory federal policy under the Equal Protection Clause.[124] As a result, it remains unclear how the Court will analyze equal protection claims by transgender people. In this sense, the civil rights movements of LGB people have largely subsumed transgender rights in the courts.[125] But after the recent bout of Supreme Court jurisprudence affording civil rights to LGB people culminated with its marriage decision in Obergefell v. Hodges,[126] and after the Court accepted a key transgender civil rights case for its 2016 term,[127] scholars observed a “passing of the torch from ‘LGB’ to ‘T’” underway, and argued that “[t]he next civil rights frontier belongs to transgender people.”[128]

Prior cases illuminate three key pathways for transgender rights advocates seeking to challenge transgender exclusions. First, the Court may decide that transgender people are sufficiently unique that they constitute a category unto themselves. Under this approach, petitioners can seek suspect class status. Second, the Court may, in line with prior precedent, attempt to fit transgender people into bifurcated categories of “man” and “woman” based on biology or dominant gender expression. Under such an interpretation, the Court would analyze transgender people’s claims just as it does those involving classifications between men and women. Finally, the Court may conceive of transgender people as part of the “LGBT” umbrella and consider transgender claims as it does those involving sexual orientation—namely, by invoking the “animus” principle.

Part III of this Note will first provide an overview of the Court’s suspect class, sex, and sexual orientation jurisprudence. It will then analyze how the Court may reason through transgender litigants’ challenges to women-only shelter policies in light of this precedent.

A. Review Under the Equal Protection Clause Generally

The Fourteenth Amendment’s Equal Protection Clause prevents states from denying anyone within their jurisdiction “equal protection of the laws.”[129] In other words, the Equal Protection Clause reflects the principle that similarly situated people should be treated alike under the eyes of the law. The Fifth Amendment imposes similar obligations on the federal government.[130] Equal protection analysis under the Fourteenth Amendment questions the legitimacy of the governmental interest in differential treatment and the extent to which the classification—the aspect of the government policy that differentiates between social groups—advances that interest.[131] Courts apply one of three standards of review to make this determination: strict scrutiny, intermediate scrutiny, or rational basis review.[132]

Under strict scrutiny, the Court asks whether the classification implicates a compelling state interest and whether the classification is narrowly tailored to achieve those ends. Because strict scrutiny is exceedingly demanding and exists to “smoke out illegitimate purposes,” the Court rarely upholds laws reviewed under this standard.[133] Under intermediate scrutiny, the Court engages in a less rigorous review of the challenged law. The Court reviews the classification to determine whether it serves an important government interest and whether the law substantially advances that interest.[134] The intermediate scrutiny inquiry requires showing that nondiscriminatory alternatives are not viable means of achieving the government interest.[135]

If the rational basis test applies, the Court inquires only whether the stated government purpose is legitimate, rather than compelling or important.[136] Even if the state does not proffer a legitimate rationale, the Court can hypothesize a rational basis for the law, even if it stands apart from the one articulated by the legislature.[137] Moreover, all the Court requires to uphold classification is a loose rational connection between means and ends. Therefore, under the rational basis test, because the Court does not scrutinize the government’s rationales, it tolerates under- and overinclusiveness in the law and allows the legislature to take a step-by-step approach to legislation by addressing the concerns of one group, while ignoring similarly situated others.[138]

Thus, how the Court decides to conceive of transgender identity will affect the level of review it uses to analyze the case, which will in turn dictate the level of protection transgender people will enjoy. If the Court considers transgender people as a separate group, transgender people will have to petition for suspect or quasi-suspect class status to secure strict or intermediate scrutiny, respectively. The court first articulated the concept of suspect class status in United States v. Carolene Products, reasoning that some social groups may face prejudice as “discrete and insular minorities,” and, as such, may not be able to avail themselves of political processes to vindicate their rights.[139] In such cases, the Court conducts a more searching inquiry to smoke out illegitimate classifications. If transgender people cannot establish that they are a suspect class, the Court will apply rational basis review.

On the other hand, if the Court analyzes transgender people’s claims through its regular sex-based classification framework, intermediate scrutiny will apply. Finally, if the Court classifies transgender identity as something akin to sexual orientation, Supreme Court precedent dictates that traditional standards of review will not apply. Instead, the Court will invoke an animus principle to strike down laws that it perceives to be motivated by hatred for transgender people or, in some cases, private biases against transgender people. This Note analyzes each of these possibilities in turn.

B. A Separate Transgender Identity: The Dim Possibility of Suspect Class Status

As gender identity can be considered a distinct analytical category from the biological sex of a person or their sexual orientation, there is a chance the Court will treat a transgender litigant’s equal protection claim as an issue of first impression. A theory of intersectionality demonstrates that sexual orientation and gender identity coalesce to form part of a person’s identity: for example, a transgender person can identify as heterosexual, homosexual, bisexual, queer, or another sexual orientation. Further, a transgender person can present as either a man or woman. But because transgender people defy gender norms in a way that separates them from the rigid biological categories the Court has used to analyze classifications based on sex and sexual orientation, the Court may consider transgender people as distinct from cisgender men and women, and gay and lesbian people. If the Court treats transgender litigants as a class of first impression, then it must decide what standard of review it will use to analyze their equal protection claims. To make this determination, the Court will consider whether the group constitutes a “suspect class” that warrants application of heightened scrutiny.[140] This Note argues that the Court will likely conclude that though transgender people face a history of rampant discrimination, the range of expressions transgender identity can take prevent them from being a suspect class. Thus, the Court will not likely declare transgender people to be a suspect class and will likely engage in rational basis review.

1. The Supreme Court and Suspect Classifications

To analyze how the Court may consider a transgender claim for suspect class status, a review of its Carolene Products jurisprudence with respect to other classes is instructive. The Court typically uses four factors to determine whether a group constitutes a discrete and insular minority that renders them a suspect class: (1) the group faced a history of discrimination; (2) the group lacks political power; (3) the classification is based on an immutable characteristic; and (4) the characteristic has no bearing on the group’s ability to contribute to society.[141] Relying on these factors, the Court has held sex, race, and nationality classifications to be inherently suspect, but it has explicitly denied suspect class status to people with disabilities and the poor.[142]

Precedent shows that demonstrating a history of discrimination does not suffice to secure suspect class status when the Court does not consider a social group to be discrete with immutable characteristics. For example, in San Antonio School District v. Rodriguez, the Court held that poverty does not give rise to a suspect classification and that discrimination against the poor should receive only rational basis review.[143] In so holding, the Court particularly highlighted the difficulty of defining the poor, arguing that the category is fluid enough that many reasonable lines could be drawn.[144] Reasoning that the judiciary stood ill equipped to draw those lines, it held that the poor were not a sufficiently discrete group to constitute a suspect class.[145] Similarly, in City of Cleburne, Texas v. Cleburne Living Center, Inc., the Court engaged in a Carolene Products analysis to hold that people with disabilities do not represent a suspect class.[146] The Court viewed disabled people as a “large and diversified” class, rather than discrete and insular.[147] In addition, the Court saw no evidence of discrimination in the laws passed to protect disabled people because it reasoned that differential applications of the law were needed to protect the mentally disabled.[148] Moreover, the robust legislative response protecting those with mental disabilities showed that they were not politically powerless, since they succeeded in attracting the attention of lawmakers.[149] These precedents demonstrate that though many groups face a history of discrimination, discrimination as such does not suffice to gain a group suspect class status. It also shows that the Court uses rigid conceptions of discreteness and immutability to deny groups suspect class status.

Though unsettled, lower courts’ analysis of LGB people as a potential suspect class helps predict how the Court will treat a transgender claim. There is some confusion among courts as to whether sexual orientation is entirely foreclosed as a suspect classification or remains open to contest. Whereas some lower courts have afforded LGB people suspect class status using the Carolene Products analysis,[150] a Supreme Court majority has never engaged in suspect class analysis to analyze LGB petitioners’ claims.[151]

Many lower courts do not apply heightened scrutiny to LGB equal protection claims, and the Court may extend this logic to transgender people to deny them suspect class status and, therefore, heightened scrutiny. Professor Evan Gerstmann noted that “the appellate courts have consistently rejected the argument that gays and lesbians are a suspect class.”[152] Courts have typically problematized the “immutability,” “discreteness,” and “political power” components of the Carolene Products analysis. In High Tech Gays v. Defense Industrial Security Clearing House, for example, the Ninth Circuit reasoned that “[h]omosexuality is not an immutable characteristic; it is behavioral and hence is fundamentally different from traits such as race, gender, or alienage, which define already existing suspect and quasi-suspect classes.”[153] Moreover, some courts and judges reject the idea that gay and lesbian people lack political power, since legislatures have taken steps to protect LGB people through antidiscrimination legislation.[154] In Romer, for example, Justice Scalia’s dissent suggests that some members of the Court internalized the “pernicious wealthy gay stereotype” and used that sociological portrait to deny sexual orientation heightened scrutiny based on perceived political or social power.[155]

However, other cases indicate that courts do consider sexual orientation to be a suspect class, and the Court may find their reasoning persuasive when applied to transgender litigants’ claims. Before the Court considered Windsor v. United States, the Second Circuit held that gay people constituted a quasi-suspect class. The court reasoned that gay people have suffered a history of persecution, sexual orientation has no relation to ability to contribute to society, gay people represent a discernable group, and gay people remain politically weak.[156] Therefore, whereas some courts have denied sexual orientation heightened scrutiny on the grounds that sexual orientation is not an immutable characteristic and that LGB people do not lack political power, other courts have expanded traditional conceptions of discreteness and immutability to afford LGB people quasi-suspect class status. The unsettled landscape suggests that the Court has multiple routes to follow in deciding this issue.

2. Are Transgender People Discrete & Immutable? The Problem with Suspect Class Status

In light of the mixed landscape described above, it is not entirely clear how the Court will consider transgender classifications in an equal protection analysis. Out of the four Carolene Products factors, showing a history of discrimination towards transgender people and demonstrating that transgender people are politically powerless represent perhaps the least controversial factors. Transgender people suffer from discrimination and harassment when accessing basic needs, such as employment, housing, and healthcare, which in turn translates into high rates of poverty among transgender people.

As for employment troubles, in a national survey of transgender Americans, 97 percent of respondents had experienced harassment or mistreatment on the job and 47 percent had lost their jobs, been denied a promotion, or been denied a job as a result of being transgender.[157] Not surprisingly, research indicates that unemployment among transgender people is twice the rate of the general population and four times higher than the general population for transgender people of color.[158] Even those transgender people who were employed on average earned less than $25,000 per year[159] or 42 percent of the national median income for 2017.[160] Young transgender people experience discrimination in schools, with 78 percent of respondents indicating that they had been harassed and 35 percent reporting physical assault at school.[161]

Homelessness also plagues a significant number of transgender people. The same national survey indicates similar results for homelessness: 19 percent of respondents had been homeless at some point in their lives, a rate 2.5 times higher than the national population.[162] As a result of homelessness and unemployment, transgender people spend more time on the streets and in the underground economy as compared to cisgender people, making them highly likely to experience police brutality.[163] This persecution has a devastating impact on transgender lives: the American Psychological Association has concluded that “discrimination and lack of equal civil rights is damaging to the mental health of transgender and gender variant individuals.”[164] This has led to 41 percent of transgender respondents reporting attempted suicide compared to just 1.6 percent of the general population.[165]

In addition to experiencing daily marginalization, transgender people often lack political power. Though various administrative agencies slowly incorporated transgender people’s needs into regulatory policy under the Obama administration, the Trump administration has largely reversed these gains. For example, in February 2017, the Department of Education issued a letter withdrawing the Obama administration’s position that denying transgender students the right to use the bathroom of their choice violates federal sex discrimination law.[166] Moreover, in July 2017, President Trump signaled a shift in military personnel policy by tweeting that the US government “will not accept or allow transgender individuals to serve in any capacity.”[167] The President subsequently revised the ban to bar anyone diagnosed with gender dysphoria from serving, but made an exception for those willing to hide their transgender identity or those willing to forgo making any transitions to their gender identity, and for current transgender groups.[168] Whereas advocates of the law argue that this appropriately focuses the law on military readiness, opponents still maintain that it is effectively a total ban.[169]

The state legislative front appears similarly hostile to transgender people. States are passing overtly anti-transgender legislation to frustrate efforts at transgender inclusion. In North Carolina, for example, the state legislature passed a law to nullify local ordinances protecting transgender people who use public restrooms based on their gender identity.[170] The statistics on transgender unemployment and homelessness as well as the examples of overt antitransgender legislation demonstrate the ways in which transgender people continue to face political hostility and daily discrimination.

Courts recognize this troubled history. In Brocksmith v. United States, the D.C. Circuit recognized that “[t]he hostility and discrimination that transgender individuals face in our society today is well-documented.”[171] Similarly, a court in the Southern District of New York stated that “this history of persecution and discrimination [against transgender people] is not yet history,” and proceeded to apply heightened scrutiny in a §1983 civil rights claim involving a transgender plaintiff who alleged police brutality.[172] Thus, even if an obscured sociological portrait of gay and lesbian people has distorted sexual orientation jurisprudence in the past by leading the Court not to explicitly declare LGB persons a suspect class, the incontrovertible evidence that transgender people suffer from societal and, at times, state-sponsored harassment and discrimination suggests a well‑established history of discrimination, likely satisfying the first Carolene Products factor.

The requirements of immutability and discreteness, however, present more complicated considerations. The Cleburne Court held that the “large and diversified” status of people with intellectual disabilities does not make them a discrete class. The Court went on to reason, “[n]or are they all cut from the same pattern: as the testimony in this record indicates, they range from those whose disability is not immediately evident to those who must be constantly cared for.”[173] Similarly, transgender people express their gender identity in multiple ways, and transgender identity is not always immediately evident. For example, while some transgender people undergo physiological changes, others choose to express their identity solely through aesthetic statements. Some openly discuss their transgender identity, and others choose to be more private. To illustrate the diversity of expressions of transgender identity, consider Thomas Beatie, a transgender man who kept his biologically female reproductive organs in order to be able to bear children, choosing to do so because his wife at the time was infertile.[174] Beatie’s experience suggests that the Court may look to the diverse experiences of transgender people and find them to be a “large and differentiated class,” similar to those with disabilities.

This diversity poses a strategic dilemma for transgender activists. It is desirable to express the diversity among transgender people and present gender identity as a fluid spectrum because it comports with the lived realities of the transgender community. But diversity in representation also limits transgender people’s legal protections under equal protection analysis. Limiting transgender identity to apply only to those, for example, who undergo physiological changes may result in a neat category that provides courts with a principled basis to limit the suspect group. Yet as Part II discussed, failing to recognize diverse forms of gender expression that results in a “hierarchy of inclusion” that is detrimental to transgender survivors’ needs.[175]

Some courts recognize this complexity and attempt to arrange a discreteness analysis around it. For example, the Second Circuit in Windsor reasoned that “[w]hat seems to matter is whether the characteristic of the class calls down discrimination when it is manifest.”[176] That court offered the example of a person of illegitimate birth (a child of unmarried parents), a status that invokes heightened scrutiny, applying for Social Security benefits, making their status manifest, and experiencing discrimination on that basis even though their status not readily discernible in the first instance.[177] In Adkins, a New York district court, considering a claim that a police officer discriminated against a protestor on the basis of the protestor’s transgender identity, applied that scenario specifically to transgender people:

[M]any forms of identification required for asserting legal rights, such as birth certificates, indicate the bearer’s gender. A mismatch between the gender indicated on the document and the gender of the holder calls down discrimination, among other problems. Document troubles aside, transgender people often face backlash in everyday life when their status is discovered. For instance, plaintiff alleges that, upon learning that he was transgender, police officers gawked and giggled at him and asked him what he had “down there.”[178]

The Adkins court confronted a scenario very similar to those of transgender survivors attempting to access IPV shelter spaces. As Victoria Cruz’s account demonstrates, shelters often ask transgender survivors invasive questions about their bodies and deny them access to space on that basis.[179] In this sense, transgender people can be considered a discrete group because as soon as they make their transgender identity manifest to an IPV service provider, discrimination can ensue.[180] If the Court follows the Second Circuit in Windsor, rather than earlier conceptions of discreteness as articulated in Cleburne, the Court may apply heightened scrutiny.

Immutability presents similar concerns. Immutability represents a key aspect of suspect class status because the law draws distinctions between what it perceives as circumstances arising as a result of choices or personal responsibility, on the one hand, and those determined solely by the accident of birth, on the other.[181] Under this logic, differential treatment based on an immutable ground is unfair because the person has no control over that circumstance.[182]

Recognizing this distinction between acts and identity in courts’ reasoning, transgender advocates have brought claims that attempted to flip the sex-based discrimination analysis: whereas sex—one’s biological assignment—is mutable through medical procedures, one’s gender identity is immutable.[183] For example, in a New Jersey case challenging the dismissal of a physician who came out as a transgender woman in her workplace, litigants used gender dysphoria to attempt to establish immutability.[184] They argued that transgender people are subject to gender dysphoria and face a “cognitive realization that [they were] born in the ‘wrong’ body.”[185] Advocates thus developed narratives that pathologized the status of transgender people: advocates argued, for example, that “being transgender is not a lifestyle choice” but “a condition or syndrome in which one’s identification or desire to live as a member of the other sex is deep-seated, unavoidable, and overwhelming.”[186] Legal scholars advocating for transgender rights have also put forward this narrative of immutability, explaining that “[t]he formation of one’s gender identity begins even earlier, likely within the first two years . . . . This, of course, well precedes a child’s ability to choose.”[187] In these ways, advocates have advanced the notion that gender identity is prefixed to satisfy the Court’s concern with immutability.

This tactic may allow for piecemeal gains for transgender rights in sympathetic courts, but it nevertheless puts transgender litigants in the same bind as does the discreteness requirement. Namely, to gain rights and protection, transgender litigants must tread “dangerously close to essentialism” by denying the social construction of gender and forwarding the idea of gender identity as a “presocial fixed category.”[188] This reversion to determinism ultimately hurts similarly situated gender minorities, including transgender people who are more ambivalent about their gender identity and may regard themselves as gender nonconforming people. Because embracing gender identity as an immutable trait disavows gender as an expressive act and instead focuses on gender as a sense of self, aspirational notions of “man” and “woman” are re-inscribed into the law. The difficulty of preserving transgender people’s diversity while still satisfying the Court’s suspect class factors of immutability may result in a deferential standard of review for transgender litigants’ claims.

These difficulties suggest that the Court will not likely embrace the idea that gender identity is an immutable trait. If its reasoning regarding the poor serves as any guide, it will likely reason that drawing rigid lines between what gender expressions qualify as “male” or “female” lie beyond the expertise of the Court, just as it raised the impossibility of drawing lines between the poor and nonpoor.[189] Moreover, it will reason that affording heightened review on the basis of gender identity will open the door to a flood of claims seeking to strike down classifications that regulate identity in any way.[190] On these grounds, the immutability requirement will likely defeat transgender people’s petitions for suspect class status.

Given these complications with showing discreteness and immutability, transgender people will not likely gain suspect class status, particularly given that the Court rarely invokes the Carolene Products analysis in the first place.[191] The suspect class framework was originally intended to recognize present-day biases and afford particularly close judicial scrutiny to laws that may be based in these prejudices. In practice, however, it has served as a rigid gatekeeping mechanism for the Court and is perceived as a disciplining force that reins in lower courts’ attempts to expand the suspect class category to new groups.[192] Despite facing a slew of claims implicating new forms of discrimination, it has been over forty years since the Court has recognized a new suspect classification.[193] Notably, even though the Court has confronted claims implicating lesbian and gay rights and lower courts have afforded sexual orientation quasi-suspect class status, the Court has not followed suit.[194] This reluctance to find new suspect classes, when combined with the problems of discreteness and immutability discussed above, suggests that if the Court reviews transgender identity as a distinct category from claims based sexual orientation and biological sex, it will likely afford transgender litigants rational basis review.

3. Rational Basis Review: The Likely Result of a Failed Attempt to Achieve Suspect Class Status

If the Court employs traditional rational basis review, it will almost certainly consider the exclusion of transgender people from women-only shelters to be constitutional. Under the rational basis review framework, a state‑sponsored domestic violence shelter need only articulate a “rational” reason for the exclusion. Given the highly-deferential nature of rational basis review, a shelter’s argument that the classification promotes administrative efficiency by ensuring safety and privacy for survivors[195] or that a single-sex female environment promotes empowerment will likely pass muster.[196] Moreover, the shelter will almost certainly pass the tailoring test, the requirement that the women-only policy should be rationally related to the shelter’s asserted goals. Under rational basis review, a step-by-step approach to legislation is tolerated, meaning that the Court does not critically question the government’s choice to include some social groups but not others in the program at issue. As a result, even though shelters that exclude transgender survivors are underinclusive in their protections by failing to include transgender women, the Court can defer to the shelter’s decision to restrict protections to cisgender women without questioning whether that population is in fact most vulnerable to IPV. This type of cursory reasoning is typical of the Court’s rational basis review analysis. As a result, a shelter’s women-only rule will likely pass constitutional scrutiny.

C. “Real Differences”: Overcoming Gender Essentialism in The Equal Protection Clause

Transgender litigants attempting to challenge their exclusion from women‑only domestic violence shelters can argue that these policies violate the Equal Protection Clause’s prohibition of sex-based discrimination. This section parses the Court’s prior gender jurisprudence to argue that the Court strikes down sex-based classifications when they rely on harmful stereotypes that prescribe certain roles to men and women, but not if the policies rest on what the Court perceives to be real differences between the genders. In light of this principle, if the Court adopts the sex-based classification framework to analyze transgender litigants’ claims, it may apply strict scrutiny to strike down shelter exclusion policies, which frequently rest on stereotypes of what it means to be male and female.

1. The Search for “Real Differences”: Biological Determinism & the Supreme Court’s Sex-based Equal Protection Jurisprudence

If the Court relies on its sex classification equal protection analysis, it will attempt to slot transgender identities into the existing bifurcated gender model in equal protection jurisprudence. The Court’s equal protection analysis often uses the terms “gender-based” and “sex-based” classifications interchangeably. However, prior jurisprudence has focused on biological sex rather than gender identity, and the Court has reasoned from the assumption that it is comparing biologically stable categories of “man” and “woman.” In Regents of University of California v. Bakke, for example, Justice Powell claimed that “[w]ith respect to gender there are only two possible classifications.”[197] This statement reveals that, for the Court, there exist only two intelligible genders: biological “man” and biological “woman.”

This pivot to biology drives the Court’s decision to apply intermediate rather than strict scrutiny to gender classifications. The Court sees race as an immutable category that does not implicate biological differences, and thus rarely finds a valid basis for drawing racial lines in the law. However, the Court applies intermediate scrutiny to gender-based claims because it views gender as a hybrid category—that is, while the Court recognizes that cultural assumptions shape what it means to be a man or woman, it also reasons that gender is, in part, biologically determined. Even longstanding women’s rights advocates on the Court appear to subscribe to this hybrid view of gender as part socially constructed, part biologically determined. In United States v. Virginia, for example, Justice Ginsburg reasoned: “‘Inherent differences’ between men and women, we have come to appreciate, remain cause for celebration, but not for denigration of the members of either sex or for artificial constraints on an individual’s opportunity.”[198] Thus, where the Court deems some classifications to be legitimate in light of “inherent differences” between men and women, it still desires some form of heightened scrutiny to root out those classifications based on socially constructed roles. Accordingly, it adopts intermediate scrutiny as the appropriate test.

In analyzing sex-based equal protection claims, the Court attempts to ascertain whether the classification in question rests on harmful stereotypes or so-called real differences between men and women. It relies on a few tools to make this determination. First, it ascertains whether the government’s stated purpose for the classification during litigation appears to comport with the sociohistorical circumstances.[199] Second, if gender is being used as predictor of certain behavior, the Court questions whether this causal link is supported by statistics.[200] Third, it examines whether there exist sex-neutral alternatives to achieving the government’s goal that have not been tried.[201]

A long line of cases illustrates the Court’s deployment of these tools to strike down gender-based classifications on the grounds that they did not substantially advance a legitimate government interest. In Reed v. Reed, the Court, for the first time, struck down an Idaho statute that automatically made men estate administrators.[202] The decision may have been rooted in a concern that the statute relied on a gender-based assumption that men controlled capital. In that case, Idaho argued that the automatic gender assignment advanced administrative efficiency, but the Court rejected the argument: “[t]o give a mandatory preference to members of either sex . . . merely to accomplish elimination of hearings on the merits . . . [or] avoid[] intrafamily controversy, the choice in this context cannot be mandated on the basis of sex.”[203] Similarly, in Frontiero v. Richardson, the Court struck down an armed services rule that granted servicemen benefits to support their wives, but did not afford female soldiers the opportunity to support their husbands in the same way.[204] The Court rejected the argument that such a rule promoted administrative efficiency when it rested on a male-as-breadwinner stereotype.[205] In a now famous line, Justice Brennan reasoned that “[s]uch discrimination was rationalized by an attitude of ‘romantic paternalism’ which, in practical effect, put women, not on a pedestal, but in a cage.”[206]

Along the same lines, in Craig v. Boren, the Court struck down a classification in a state law designed to deter driving under the influence by prohibiting the sale of beer to men under the age of twenty-one and women under the age of eighteen.[207] Though the Court recognized that the state had a legitimate interest in promoting traffic safety, it reasoned that statistical evidence showed that males were only slightly more likely to get into car accidents than were females and, thus, the numbers did not suffice to draw a “gender line.”[208] According to Professor Keith Cunningham-Parmeter, the Court has struck down such laws not only on the basis of a commitment to formal equality, but also following an anti-subordination principle aimed at “dismantl[ing] gender hierarchies” that stereotyped men and women.[209]

Another line of cases shows that the anti-subordination streak in the Court’s sex-based equal protection jurisprudence extends to both men and women. In VMI, Hogan, and Hibbs,[210] the Court examined gender classifications with a critical eye toward smoking out stereotypes. It ultimately struck down laws that perpetuated harmful generalizations about the appropriate roles of men and women and upheld a Congressional law that sought to curtail state employers’ use of harmful gender stereotypes as an appropriate prophylactic remedy for preventing equal protection clause violations.

In VMI, the Court ruled that Virginia Military Institute (The Institute), an all-male, single-sex military institution of higher learning, could not deny admission to women.[211] The Institute argued that it advanced its mission of training “citizen-soldiers” who went on to achieve success in politics, the military, and business through the “adversative method,” a demanding educational technique that included “[p]hysical rigor, mental stress, absolute equality of treatment, absence of privacy, minute regulation of behavior, and indoctrination in desirable values.”[212] The living quarters of the school were also “Spartan barracks” where students were constantly watched and privacy was nonexistent.[213] The state’s interest in VMI went beyond mere administrative efficiency; it also proffered an interest in promoting a diversity of educational opportunities, including single-sex schools, and asserted that including women would require changing the Institute’s unique character.[214] While the Court recognized that “diversity among public educational institutions can serve the public good,” the majority opinion explained that the Institute was not established with that diversity purpose in mind.[215] Moreover, the Court found insufficient evidence that the Institute could not maintain its “adversative method” if it admitted women, reasoning that such “self-fulfilling prophecies . . . [are] routinely used to deny rights or opportunities.”[216] Though the Court recognized that “admitting women to [the Institute] would undoubtedly require alterations necessary to afford members of each sex privacy from the other sex in living arrangements, and to adjust aspects of the physical training programs,” it nevertheless concluded that the fundamental character of the school would remain intact in spite of these accommodations.[217] As Justice Ginsburg wrote, “[G]eneralizations about ‘the way women are,’ [and] estimates of what is appropriate for most women, no longer justify denying opportunity to women whose talent and capacity place them outside the average description.”[218] This case demonstrates that the Court’s sex jurisprudence is marked with a desire to dismantle stereotypes that prevented women from being given full educational opportunities.

In Mississippi University for Women v. Hogan, the Court examined whether Mississippi had a legitimate state interest in excluding men from a nursing program at a state-sponsored school or whether the statutory objective “reflect[ed] archaic and stereotypic notions” and “fixed notions concerning the roles and abilities of males and females.”[219] In Hogan, the Court rejected the idea that the statute served a compensatory or remedial purpose for women’s employment, holding that such an interest was legitimate only when “members of the gender benefited by the classification actually suffer a disadvantage related to the classification.”[220] Here, women were overrepresented in nursing fields.[221] Thus, “excluding males from admission to the School of Nursing tends to perpetuate the stereotyped view of nursing as an exclusively woman’s job.”[222] Moreover, the policy did not substantially advance the state’s objective, since men already audited classes without affecting women’s enrollment or the school’s teaching methods.[223]

In Nevada Department of Human Resources v. Hibbs,[224] the Court recognized that allowing employers to grant women more family-care leave than men would perpetuate the stereotype that “women are mothers first, and workers second” and that “caring for family members is women’s work.”[225] In that case, the Court held that the Family and Medical Leave Act (FMLA) was a valid exercise of Congress’ Section 5 powers to enforce the Fourteenth Amendment because providing guaranteed leave to men and women prevented employers from using stereotypical thinking to discretionarily grant family leave to women (as primary caregivers), but deny it to men. This case demonstrates that the Court considers gender stereotypes to be sufficiently prevalent that Congress can proactively enact laws to counteract them.

Taken together, VMI, Hogan, and Hibbs demonstrate that the Court is capable of striking down policies that penalize plaintiffs for not conforming to a socially constructed stereotype of their gender. But the Court has not been consistent in its ability to recognize gender essentialist stereotypes, nor in its commitment to strike down laws that may perpetuate those stereotypes.

In particular, the Court failed to strike down gender classifications in two different circumstances. First, when a state uses a benign classification to correct past discrimination against women, the Court typically upholds the classification. Justice Ginsburg explicitly created space for remedial sex-based classifications in her VMI opinion, stating that “[s]ex classifications may be used to compensate women ‘for particular economic disabilities [they have] suffered,’ . . . to ‘promot[e] equal employment opportunity,’ . . . [and] to advance full development of the talent and capacities of our Nation’s people.”[226] In Califano v. Webster, the Court approved of sex-based classifications to combat “the socialization process of a male-dominated culture.”[227] Whereas classifications based on harmful stereotypes fail under intermediate scrutiny analysis, Professor Cunningham-Parmeter has argued that “the Court has approved of special treatment laws that favor one sex over the other when they effectively confront age-old gender stereotypes and advance an antisubordination agenda.”[228]

Second, when the Court identifies what it perceives to be “real differences” between men and women, it tends to uphold sex-based classifications. In Michael M. v. Superior Court of Sonoma County, the Court rejected an equal protection challenge to a statute that criminalized sex with a minor only for males but not females.[229] The Court based its reasoning in biology: because women get pregnant and men do not, nature burdens the act of underage sex for women.[230] Thus, the Court suggested, the legislative penalty on men merely evened the deterrent scales.[231] In this way, a biological marker, pregnancy, prevents a deeper analysis into whether the law is motivated by harmful gender based stereotypes, such as, for example, the assumption that men are more sexually aggressive or predatory than women.

Similarly, in Nguyen v. Immigration and Naturalization Services, the Court upheld a gender classification that automatically gave citizenship to children with US-citizen mothers, but required those seeking citizenship through their fathers to meet several tests.[232] These tests included proving that a blood relationship existed; that the father had US citizenship status at the time of the child’s birth; that the father had agreed to provide financial support until the child reached eighteen; and either that the child lived with the father, the father had acknowledged paternity through oath, or paternity had been adjudicated in court.[233]

The Court rationalized the classification on the basis that the government had an interest in ensuring that a biological parent-child relationship existed and that the parent had demonstrated an opportunity to develop an everyday relationship with the child.[234] The Court held that the classification substantially achieved that end because a biological relationship between the mother and child is verifiable from the birth itself through a birth certificate or hospital records and witnesses.[235] Moreover, the Court reasoned that while mother and child would inevitably develop a relationship through the birth process, no similar “biological inevitability” existed with respect to “unwed father[s].”[236] The Court thus justified the classification based on the notion that, here, “the use of gender specific terms takes into account a biological difference between the parents” to account for the “unique relationship of the mother to the event of birth.”[237] As in Michael M., here, the Court used the biological marker of pregnancy to curtail a deeper analysis into whether the government was using a sex-based generalization (that mothers are innately nurturing and fathers are absent) to differentiate between men and women.

These cases demonstrate the way in which the Court’s sex jurisprudence vacillates between striking down gender classifications based on harmful sex stereotypes and upholding classifications based on what the Court perceives as “real differences.” In cases where the Court recognized and struck down sex classifications based on harmful stereotypes—Frontiero, Hogan, Craig, Reed, and VMI—it seemed sympathetically aligned with Katherine Franke’s observation that “biology operates as the excuse or cover for social practices that hierarchize individual members of the social category ‘man’ over individual members of the social category ‘woman.’”[238] Yet the Court’s pivot to biological difference lapses back into gender essentialism and prevents a deeper analysis into whether the government’s logic is based on sex stereotypes. For example, though the Court used pregnancy as a biological marker to justify differentiating between men and women in Michael M. and Nguyen, it in fact may have relied on stereotypes of men as sexually deviant people[239] or absent fathers[240] and women as innately nurturing mothers[241] to justify sex classifications. The Court’s valorization of women as fragile beings and selfless nurturers also reverts back to the romantic paternalism that the Frontiero court explicitly disavowed.[242] In the end, this analysis shows that the presence of any biological differences between transgender women and cisgender women could cause the Court forgo a deeper analysis that smokes out harmful sex stereotypes that undergird the government’s policy. As a result, the Court may uphold the exclusion of transgender women from shelters by arguing that transgender women are biologically different that cisgender women.

Two factors may explain the Court’s vacillation between stereotypes and real differences. Elucidating these factors will help predict whether the Court will regard a classification distinguishing between women and transgender people as stereotypical or justified by “real differences.” First, differences in what the Court views as a stereotype may explain the disconnect in jurisprudence. Compare, for example, Justice Kennedy’s contention in Nguyen that a stereotype is “a frame of mind resulting from irrational or uncritical analysis”[243] with Justice O’Connor’s view in Hogan that a stereotype is an impermissible “proxy for other, more germane bases of classification.”[244] Justice Kennedy’s focus on irrationality suggests that the classification must be outrageous or have no relationship whatsoever to differences in in-group behavior. This high bar seems to condone overbroad generalizations about gender and does not foreclose the specter of romantic paternalism that troubled the Frontiero court.[245]

On the other hand, Justice O’Connor’s conception of stereotype focuses on how social constructions of what is appropriate for each gender might normalize what are in fact illusory distinctions. For example, whereas Justice Kennedy uses pregnancy as a proxy for a long-term relationship between mother and child,[246] Justice O’Connor might suggest that this assumption relies on the stereotype that women are inherently nurturing and that every pregnant woman has a long-term relationship with her child. The bar to finding an impermissible stereotype is clearly lower in Justice O’Connor’s conception than in Justice Kennedy’s. Indeed, Justice O’Connor’s position focuses on the imprecision of the proxy, which suggests that the exceptions—that is, those that do not fit with an otherwise generally accurate classification—will bear on the Court’s decision. The Court’s lapses into biological essentialism may reflect, in part, a narrow vision of what a stereotype entails. On the other hand, when the Court strikes down classifications by criticizing them as overbroad generalizations or by questioning whether gender represents the best proxy for an exclusion or inclusion, it appears to adopt the more capacious understanding of stereotype. The difference between irrational and imprecise is quite broad and can explain the Court’s vacillation between smoking out stereotypes to strike down classifications, and justifying certain classifications on the basis of biological essentialism.

Second, the Court’s sex jurisprudence suggests that the presence of any biological marker—such as the capacity to get pregnant—forecloses analysis of the way in which these so-called biological facts are mediated through the socialization process. In other words, as soon as it identifies a biological marker, the Court forgoes actual engagement in intermediate scrutiny by inquiring into the classification’s stated purposes and looking for viable sex-neutral alternatives. Instead, the Court appears to assume that socialization and stereotypes do not compound biological differences to create barriers for men and women.

This tendency for biology to obscure the Court’s perception of unequal treatment cuts across its jurisprudence. For example, when confronted with abortion rights cases, the Court pivoted to due process and privacy concerns instead of grounding its analysis in equal protection doctrine because it perceived pregnancy and the reproductive capacity of women as biological markers that justified the differential treatment of men’s and women’s bodies.[247] Professor Franke’s observation that biology reifies differences that are actually quite illusory seems very apt here since physical markers are used to uncritically justify imposing burdensome regulations on women, but not men.[248] Whether the Court adopts a capacious or narrow understanding of stereotypes and whether it focuses on the biological differences between transgender women and cisgender women will decide whether it upholds transexclusionary shelters.

The Court’s tendency to uphold remedial classifications and classifications based on biological differences between men and women has important implications for transgender litigants claims. Women-only shelters will likely attempt to argue that there focus on cisgender women is a remedial classification. They will posit that the majority of IPV survivors are cisgender women; focusing solely on them corrects a gender imbalance. This rationale is unlikely to pass muster, since transgender litigants can show that transgender women are equally likely—if not more likely—to experience IPV. Shelters will likely also argue that transgender women are biologically different than cisgender women and, thus, differential treatment is justified. This claim will be tougher to parse for the Court. Its reasoning will depend, in part, on how capacious it conceptualizes the concept of stereotype to be and whether it can move past its biological essentialism to focus on the lived experiences of transgender people, rather than reducing them to their biological make up.

2. Applying the Supreme Court’s Sex Jurisprudence to Transgender Women: Are Distinctions based on “Real Differences” or Stereotypical Womanhood?

The Court’s sex jurisprudence carries important implications for transgender women seeking to challenge their exclusion from women-only domestic violence shelters. As a threshold matter, the biological determinism undergirding the Court’s jurisprudence means the Court will likely bifurcate transgender identities into male or female, rather than contend with the notion that many transgender people occupy an interstitial space between man and woman on a fluid scale of gender identity.[249] If the Court analyzes transgender women’s claims under its bifurcated sex-based discrimination framework, it will regard transgender women as a subset of women or, where transgender women have not undergone transition surgery, as men. Using the Court’s former sex jurisprudence as a framework, this Section considers how each of these biological classifications may affect the type of scrutiny the Court affords transgender women’s claims. It then proceeds to consider how shelters’ arguments for exclusion may fare under the Court’s sex jurisprudence.

The Court’s categorization of transgender women may affect the level of scrutiny their claims receive. If the Court analyzes transgender women as a subset of women, it may conclude that a gender-based classification does not exist at all. This conclusion may be fatal to transgender women’s claims because, in the absence of a sex-based classification, the Court typically applies the rational basis test, an extremely deferential form of review that almost always upholds the government’s policy.

In Geduldig v. Aiello, the Court found that California’s disability insurance system did not contain a sex-based classification even though it excluded pregnancy-based disabilities from coverage.[250] In reaching this conclusion, the Court ignored the gendered nature of pregnancy and emphasized that, of the protections given, there were no distinctions between men and women.[251] The conclusion that program benefits accrued equally to both sexes may reflect the reasoning that nonpregnant women faced the same risk levels under the program as men, and thus there was no basis for a gender line.[252] In other words, because the policy did not disadvantage all women but only a subset of women with a gender-related disability (pregnancy), the Court did not find a sex-based classification. As a result, the Court applied rational basis review and upheld the pregnancy-based exclusion based on the state’s legitimate interest in limiting coverage to keep the program economically viable.[253]

Applied here, the Geduldig logic may inhibit transgender plaintiffs from receiving heightened scrutiny. Just as not all women face pregnancy-related health risks, the Court could reason that not all women face the physiological and psychological experiences of transgender women.[254] And, because some members of the group (cisgender women) can access shelters, the Court may fail to find sex-based discrimination against women more generally.

On the other hand, it remains unclear how far Geduldig extends,[255] and transgender advocates might highlight distinguishing factors to render Geduldig inapposite. For example, the Geduldig Court analyzed a disability insurance system that extended the same set of coverage to men and women.[256] As the Court emphasized, “California does not discriminate with respect to the persons or groups which are eligible for disability insurance protection under the program.”[257] In this sense, though women may have been exposed to higher risk levels than men, both genders enjoyed equal eligibility for the coverage offered by the program. However, women-only shelters do not base eligibility for entry on even ostensibly gender-neutral criteria, like pregnancy. Instead, men are explicitly ineligible from participating in the shelters altogether as a result of their sex. In this sense, advocates could posit that the assumption of neutrality given to California’s insurance system in Geduldig is not necessarily justified here because there exists an overarching sex-based classification that warrants higher scrutiny. If this argument proves successful at distinguishing Geduldig, transgender women seeking access to shelters might still be able to receive intermediate scrutiny even if the Court analyzes their claim as one brought by only a subset of women.

If the Court classifies transgender women as male for the purposes of its sex jurisprudence, ironically, transgender women’s claims become more straightforward. Because women-only shelters exclude all men, transgender advocates can use Hogan to argue that a women-only policy arises from overbroad generalizations about men not being victims of IPV.[258] Because a women-only shelter policy operates as a total ban on male access, the Court will almost certainly apply intermediate scrutiny in this context.

Ultimately, this discussion reveals that the Court’s current sex jurisprudence, with its emphasis on biologically stable categories of men and women, cannot accurately capture the often-fluid nature of transgender identity. However, because the Court may be unwilling to disturb its assumption regarding the existence of only two genders, it remains fruitful to consider how the Court might conduct equal protection analysis within its sex jurisprudence framework. Though there is a possibility that the Court may classify transgender women as male, this Note proceeds under the assumption that the Court will regard transgender women as women. The remainder of the Section considers state-interest rationales for excluding transgender women, analyzes potential responses that transgender advocates can make, and assesses the Court’s likely conclusions.

Shelters will likely argue that the exclusion of transgender women serves three important government interests: (1) that the exclusion represents a classification attempting to serve those most likely to face IPV—cisgender women; (2) the inclusion of transgender women would disrupt the shelter’s ability to provide feminist education and empowerment to survivors; and (3) the inclusion of transgender women would cause survivors discomfort or threaten their safety. The Court will consider these arguments under the intermediate scrutiny framework: (1) does the government have an important interest in excluding transgender people? And, if so, (2) does a total ban on transgender women substantially further that government interest?

First, the shelters will likely argue that a women-only shelter policy furthers an important government interest because cisgender women comprise the majority of IPV survivors. Ample studies show the gender asymmetry among IPV survivors[259] and thus, the exclusion serves important government interest by functioning as a remedial classification that allows shelters to focus on the most likely victims of IPV.[260] Because women are substantially more likely to face IPV, the shelters could emphasize the priorities that Woo’s argument embraced, that “women’s organizations are still trying to meet the needs of lesbians [and] women of colour.”[261] More cisgender women than transgender women experience IPV, which justifies a “gender line.”[262] In other words, statistically speaking, being female functions as a good proxy for facing IPV. Thus, the shelter will argue its women-only policy constitutes a remedial classification. The Court has alluded to the legitimacy of remedial classifications in Hogan[263] and Virginia,[264] and it has explicitly condoned such a classification in Califano.[265] Thus, the shelters can argue that reserving services for the affected populations represents a recognized important state interest that survives intermediate scrutiny. This arises from the Court’s prior jurisprudence protecting those whose prior needs the state historically marginalized and who face the greatest risk of needing the services. [266]

In response, transgender advocates can argue that classifications with a remedial purpose serve important government interests only in circumstances when they serve a disproportionately burdened group.[267] Thus, it might serve an important state interest to privilege a gender in certain policies if it was shown to have been historically disproportionately affected.[268] However, the statistics on the gender-asymmetry between men and women are often drawn from studies of heterosexual and cisgender men and women.[269] Therefore, whereas a remedial purpose may justify the different treatment of cisgender, heterosexual men and women, transgender women have a greater propensity to experience IPV.[270] Thus, transgender advocates can suggest that the state cannot offer an important remedial justification for the classification.

The Court will likely find that the remedial nature of the classification does not form an adequate basis for differential treatment. The Court’s logic in Hogan directly applies here. Even if women were generally underrepresented in higher education, in Hogan they were overrepresented in nursing education.[271] Similarly, here, though cisgender women face more violence in the home than cisgender men, transgender women are even more likely to face such violence than cisgender women.[272] The Hogan court found the remedial classification to be untenable because it merely perpetuated the stereotype that nursing was a woman’s job.[273] Likewise, here, having women-only shelter policies may only perpetuate the stereotype that they are the only social group prone to IPV.

If the Court finds that the remedial justification that the shelter advances is not an important government interest, then it will strike down the classification and end its inquiry. However, if it concludes that the state interest is important, it will ask whether the ban on transgender women at issue substantially furthers the goal preventing IPV among disadvantaged survivors. This entails a consideration of whether the classification at issue is based on overbroad generalizations or stereotypes.

To that end, transgender women litigants can argue that a total ban on transgender women does not substantially further the goal of serving disadvantaged survivors because it is based on a generalization or stereotype that (1) cisgender women are the only social group at risk for IPV; and (2) relatedly, despite their identifications otherwise, transgender women have an inherent maleness that prevents them from experiencing IPV at the same rate as ciswomen. To dispel these assumptions, transgender advocates can cite statistics laid out in Part II.B to demonstrate that transgender people are proportionally more likely to experience IPV than are heterosexual women. Moreover, transgender advocates can point out the ways that the stereotype of the male‑abuser, female-victim has prevented domestic violence service providers from recognizing and responding to IPV among nonheterosexual, noncisgender populations.[274] These arguments underscore how using cisgender women as a proxy for IPV survivor is underinclusive and imprecise in light of the high rates at which both transgender women and gay men experience IPV.

Whether these arguments convince the Court to strike down the classification will depend, in part, on how it conceptualizes stereotypes. If Justice Kennedy’s limited conception of a stereotype holds sway, using cisgender women as a proxy for IPV survivor would have to be irrational. Given that a gender asymmetry does exist in rates of violence between cisgender, heterosexual men and women, the Court could conclude that the differential treatment is justified without considering the exceptions—including transgender women and gay men. Moreover, Nguyen reveals the Court’s tendency to use biological markers—such as pregnancy—to justify sex-based differences and condone generalizations.[275] Applying that trend here, the Court may use the fact that some transgender women have not transitioned or fully present as women to justify distinguishing them from the type of woman that will face risks of IPV.

However, if the Court foregrounds Justice O’Connor’s conception of stereotype as an imprecise proxy, the shelter’s rationale may not pass muster. The Court could reason that a stereotype that cisgender women are the sole demographic at risk from domestic violence gives rise to the exclusion of transgender people or that transgender women have an inherent maleness that prevents them from facing the same risks of IPV as ciswomen. In this sense, just as the Court overturned legislation based on stereotypical notions of women as managers of homes[276] and women as naturally inclined toward care work,[277] here, the Court could reject the underlying stereotype that women are vulnerable, passive victims of domestic violence, and males—irrespective of gender identity and sexual orientation—are invariably aggressors. The Court may find the language of VMI particularly applicable here: “[g]eneralizations about ‘the way women are’” and “estimates of what is appropriate for most women[] no longer justify denying opportunity” to those who fall outside the “average description.”[278] Under this logic, transgender women cannot be denied access to services just because they fall outside the average description of an IPV survivor. Like Craig, where the Court concluded that drawing a gender line on the basis of a 2 percent difference in accident rates among men and women was impermissibly overbroad, here, the factual finding that transgender women are just as likely, if not more likely, to experience violence as cisgender women may convince the Court that the blanket exclusion of transgender women in fact relies on an impermissible stereotype. [279]

In summary, the Court will likely find that the shelter’s interest in preserving their resources for cisgender women is not justified as a remedial classification because transgender women are just as likely to be disadvantaged by IPV as cisgender women. If the Court treats the remedial classification as an important state interest, however, transgender advocates must convince the Court that the exclusionary policy does not substantially further the shelter’s remedial purpose because it rests on a stereotype that cisgender women are the only victims of IPV.

As a second important state interest that justifies upholding the classification, the shelter may argue that operating a cisgender women-only space is crucial to creating a site of empowerment for survivors. Using the example of shelters created in the 1970s,[280] this argument links female-only shelters to the creation of a “maternal space” for survivors, so they can foster healthy female relationships and provide support to each other.[281] As Professor Nicki argued, the addition of transgender people will disrupt that sense of solidarity because transgender women have not “grown up with the publicly affirmed gender identity of a woman and with all the experiences and repercussions that this brings.”[282] Woo further argued that transgender women carry “male privilege” that would disrupt the egalitarian principles around which shelters are organized.[283] Relatedly, as mentioned above, women-only shelter advocates may argue that because some female survivors may feel “cognitively and emotionally unable to make a distinction between a benign penis and one used for harm,” they will regard transgender people who present as male and those with penises as weaponized and “threatening.”[284] This will prevent survivors from staying in or accessing spaces because they do not feel safe.

These claims will be tougher to parse for the Court. A few considerations will likely come to bear on its analysis. As a threshold matter, are these rationales truly important government interests or do they reflect post hoc rationalizations, unsupported by the socio-historical context of the shelter movement? If the Court concludes that there exists an important government interest, it will move on to consider whether the exclusion of transgender women substantially furthers the government interest. To this end, two key considerations arise. First, do real differences between cisgender and transgender women support differential treatment, or does exclusion rest merely on impermissible stereotypes, where other more germane bases of classification are available? Second, are sex neutral alternatives available—in other words, can reasonable accommodations be made within domestic violence shelters to accommodate transgender women without hurting the state interest at issue?

Given the history of the women’s movement in this case,[285] transgender advocates will have trouble arguing that a shelters’ interest in creating a site of feminist education and empowerment represents a post hoc rationalization. Thus, the Court will likely find that shelters have an important interest in empowering survivors through fostering healthy female relationships. In Virginia, the Court acknowledged that diversity in educational approaches constituted a legitimate and important government goal, but nevertheless dismissed the military institute’s interest because, after an examination into the sociohistorical context of the institute’s founding, no such rationale reasonably emerged.[286] But here, research into the sociohistorical circumstances of the “shelter movement” reveals that shelters were indeed founded as refuges for women and rooted in notions of promoting sisterhood and women’s liberation through female solidarity.[287] Moreover, the shelter movement took affirmative steps since its inception to ensure that shelters remained secret from male batterers.[288] The Court will likely use this history to conclude that shelters’ perceived need to keep women safe and comfortable, specifically through excluding those presenting as male, is genuine and not merely a post hoc rationalization devised upon litigation. However, even though the state interest in protecting spaces of women’s empowerment may be genuine, it can nevertheless be invalid under the Court’s sex jurisprudence if it rests on a harmful stereotype as opposed to a real difference.

As the Court will likely consider the empowerment of survivors to be an important state objective, it will move to consider whether the exclusion of transgender women from shelters substantially furthers the important governmental interest of empowering survivors in practice. This reasoning entails a consideration of (1) whether the exclusion rests on a stereotype or overbroad generalization about what it means to be a woman; and (2) whether transinclusive alternatives are available. Here, to strike down the classification, transgender advocates must show that the inclusion of transgender individuals into the feminist curriculum of shelters does not disrupt the empowerment process of survivors.

First, with respect to stereotypes, transgender advocates could argue that an underinclusive and stereotypical notion of womanhood undergirds a shelter’s concerns that all transgender women will inevitably interrupt the shelter’s feminist orientation. Transexclusionary women-only shelters perpetuate a policy in which purely biological markers (having a vagina and breasts) give rise to notions of womanhood, rather than focusing on the way that garb, aesthetic appearance, biological make up, names, and other such identity markers inform what it means to be a woman. In other words, using biological markers to police the definition of woman functions as a stereotype about womanhood. As such, transgender advocates can argue that the shelter policy is underinclusive of women in a manner that falls below the precision required for intermediate scrutiny.

Marking out stereotypes on the basis of biology rather than gender-based expression breaks from the Court’s prior jurisprudence striking down gender classifications, since those cases assumed that the category “woman” was biologically stable and instead focused on stereotypes as rigid gender roles ascribed to those residing within this category. However, there may be room to think about biological stereotypes as prohibited by equal protection analysis in a manner that the Court finds persuasive. First, Justice Ginsburg professed a concern for the exceptional woman in her Virginia opinion by reasoning that “generalizations about ‘the way women are’ [and] estimates of what is appropriate for most women” cannot justify the denial of opportunities for those who fall “outside the average description.”[289] The Court could thus conclude that transgender women represent such a class of women, falling outside the description of the average woman. Similarly, under Justice O’Connor’s conception of a stereotype, the Court could conclude that genitalia serve as an impermissible “proxy” for womanhood, and other more germane features (such as gender presentation) function as better proxies for classification.[290] If these capacious understandings of impermissible stereotypes hold sway, then advocates might succeed in challenging the exclusion of transgender women on the basis of biological difference alone.

However, Justice Kennedy’s conception of a stereotype renders such a showing more difficult. Though feminist scholars have shown the way that bifurcated gender categories based on biological genitalia are neither natural nor inevitable, they are nevertheless naturalized as a method of social organization by the state.[291] In this way, Justice Kennedy’s naturalization of biological difference suggests that the Court will not likely perceive classification on this basis to be “a frame of mind resulting from irrational or uncritical analysis.”[292]

Moreover, this approach risks prompting the Court to pivot to biology in order to justify a distinction between cisgender women and transgender women, as it did in Nguyen. Just as the Court focused on reproductive capacity and pregnancy to justify differential treatment in Michael M. and Nguyen, here, the Court may focus on the physical aspects of the female body as indicative of lived experiences that separate cisgender women from transgender women.[293] Just as Justice Kennedy suggested that pregnancy and birthing inevitably lead to bonds between mother and child, Professor Nicki argued that transgender women cannot bond, profess solidarity, and feel safe in the presence of each other in the same way as cisgender women, because they have not “grown up with the publicly affirmed gender identity of a woman and with all the experiences and repercussions that this brings.”[294] Woo, similarly, expressed concerns that MTF transgender people can never be women because they were socialized as boys and men and consequently exercise male privilege and that male privilege is intrusive to women-only spaces.[295] Because of the bind that discarding biology in favor of gender identity presents, the Court will likely adopt a similar view to Professor Nicki and affirm the “safety” theory as an important state interest.

If the Court were to endorse the idea that expression and not biology is gender-determinative, such a recognition would help those transgender litigants who present most visibly as a woman. But it nevertheless leaves the “hierarchy of inclusion”[296] intact—that is, IPV shelters will still face the task of policing the boundaries of womanhood based on feminized stereotypes. Consider, for example, that a shelter may deny access to a transgender woman dressed in women’s clothes but who still has facial hair, because they do not look like a woman. Here, too, the shelter is acting with a stereotypical aesthetic representation of what a feminized woman looks like. Indeed, the Court has problematized this type of stereotyping in other contexts: when confronted with a Title VII claim in which a woman CEO was penalized for not acting more feminine, through, for example, talking softly or wearing makeup, the Court held that such stereotyping violated Title VII.[297] If the Court concludes that a stereotype of womanhood guides transexclusionary policy, then it will likely conclude that the policy does not substantially further the goal of empowering women, and it will strike the classification down.

An analysis of whether the transexclusionary shelter policy substantially furthers the goal of empowering survivors also entails a consideration of whether transinclusive alternatives are workable. An examination of the initial shelter movement in Part II.A revealed three key themes: (1) a commitment to nonhierarchical and egalitarian principles; (2) a feminist education founded in notions of female liberation and power and control patriarchy; and (3) the fostering of healthy bonds through female relationships. These themes will be considered in turn with an eye toward whether the inclusion of transgender survivors allows for their preservation.

As to the first theme, transgender advocates can argue that commitments to nonhierarchical principles can transcend gender lines, since they are methods of organization rather than innate to one gender or another.[298] Scholars such as Professor Nicki argued that those principles might not be innate to one gender or another and that socialization makes women more prone to embrace those principles than men. Whether or not that proposition has merit, transgender advocates can respond that the transgender experience complicates that socialization presumption because transgender individuals experience life as men and women through a combination of physiological and aesthetic expressions. This complication casts doubt on whether excluding transgendered women substantially furthers the state’s interest in providing female empowerment models in shelters. Transgender advocates can also argue that the second theme, an education in female liberation and power and control patriarchy, is possible in spite of transgender inclusion. Many transgender women experience the classical feminist notion that patriarchy is male domination because they are disciplined for, and harassed as a result of, expressing their female identity.[299] Moreover, their inclusion in that space enhances the curriculum by speaking to the ways that classic feminist notions of “sisterhood” are underinclusive towards LGBT people. Finally, as to the last theme, transgender advocates can argue that accounts of relationships between women at shelters demonstrate that the relationships do not form out of a deep-seated notion of “womanhood,” [300] but out of shared experiences with abuse.[301] In that sense, there is nothing to stop transgender women from contributing to those relationships.

The Court may be receptive to these arguments. Consider Justice Ginsburg’s repudiation of the state’s interest in VMI: just as the petitioners in that case marshaled insufficient evidence to suggest that they could not maintain its “adversative method” by admitting women, here, shelters would probably marshal insufficient evidence to suggest that they cannot maintain their feminist space by admitting transgender women.[302] Because the themes of egalitarianism and equality cut across gender lines, and because many transgender women do grow up experiencing the type of discrimination and violence that cisgender women suffer, the Court may find that the inclusion of transgender women does not significantly alter the shelter’s empowerment programs. Instead, the Court may conclude that refusals to admit transgender women to women-only shelters rest on raising “self-fulfilling prophec[ies] . . . routinely used to deny rights or opportunities” to excluded groups.[303] Finally, the Court may strike down the classification even if shelters have to make small changes to their curriculum to accommodate transgender women. Just as the Court required the inclusion of women in VMI even if it necessitated “adjust[ing] aspects of the physical training programs,” here, shelter curricula can change to include intersectional feminism and queer theory without decentering the fundamental thrust of empowerment.[304]

In summary, although the Court will likely conclude that empowering survivors is an important state interest, transgender advocates might be able to successfully convince the Court the transexclusionary policy does not substantially further that empowerment goal. In particular, transgender advocates can suggest that teaching feminist principles and fostering healthy relationships are pedagogical approaches that can include transgender women because they face similar sources of gender-based marginalization. Moreover, the Court can tolerate even small rearrangements.

Finally, the third important state interest posited in favor of exclusionary shelters—addressing survivor discomfort and safety—may pose the most difficulty for the Court. Two rationales inhere in shelters’ concerns over safety: first, that the presence of transgender people may affect the safety and comfort of other residents. This is because those who appear male threaten women survivors with whom they have to share close quarters, including bathrooms and sleeping spaces.[305] Second, males who may pose as survivors to gain access to shelter spaces and victimize women threaten that safety.

Transgender advocates will not have much luck arguing that addressing survivor discomfort and ensuring survivor safety are not an important state interest. But transgender advocates can focus on the dearth of data available to support the claim that transgender women disrupt survivor safety and comfort, as well as the presence of transinclusionary alternatives, to suggest that a blanket exclusionary policy toward transgender women is not the most effective way to address discomfort or safety. Under intermediate scrutiny, these arguments could demonstrate that a blanket transexclusionary policy would not substantially further the shelter’s goals and would result in the classification being struck down.

First, transgender advocates can argue that evidence regarding transinclusionary accommodations suggests that fears regarding privacy and safety are overblown. With regard to safety, there is a dearth of data on whether males have posed as transgender women survivors to hurt shelter residents. However, the successful integration of transgender women in many domestic violence shelters belies this fear. Moreover, transgender advocates can point out that vague fears about the privacy and comfort of cisgender women are routinely leveled in favor of allowing transgender women to access public services, including the bathrooms of their choice. The argument in the bathroom context is familiar to domestic violence shelters and thus merits examining. In response to antidiscrimination legislation protecting transgender people, an outcry erupted that allowing transgender people to use the bathroom corresponding with their gender identity would cause assaulters to pose as transgender women and harm those using the female bathroom.[306] Yet as the public debate on this matter grows, little evidence has emerged to support the belief that the presence of transgender people will cause such discomfort or harm safety in practice.[307]

Transgender advocates should use this dearth of evidence to show that a stereotype that transgender people are deviant or more prone to violence undergirds fears regarding safety. As directors of transgender-inclusive shelters in San Francisco and Boston have emphasized, the point is not so much that transgender women are incapable of committing violence, but that transgender women are no more likely to commit acts of violence than cisgender women.[308] Gunner Scott, a nonprofit director who has trained Boston-area domestic violence shelters that accept transgender women, suggests that the specter of transgender women threatening the physical safety of women in domestic violence shelters in fact reflects a stereotype:

Stereotypes of transgender people attacking women come from movies and television shows that inaccurately portray transgender people as dangerous and abusive. This is far from the truth. When it comes to transgender people, the more serious risk is that violence will be committed against transgender people by others. Also, shelters need to learn that it is a myth that a woman-only space is always safe. The occurrence of woman-to-woman abuse by both straight and lesbian women is real, and shelters need clear rules against it. By enforcing these rules for all residents, transgender and nontransgender, these spaces can become truly safe.[309]

Thus, advocates can advance two key safety arguments: first, that the lack of violent incidents in transgender inclusive shelters suggests that the presence of transgender women does not particularly threaten survivors’ physical safety; and, second, that justifications based on physical threats are grounded in legally impermissible stereotypes regarding transgender women.

To further bolster their case regarding the workability of transinclusionary alternatives, transgender advocates can highlight the successful integration of many transgender women into shelters. Recognizing the workability of transinclusionary alternatives, the aforementioned HUD Equal Access rule puts the onus on service providers to provide reasonable accommodations for transgender survivors: first, the agency rule does not police gender categories, but gives the transgender person autonomy to decide their gender identity; second, it asks the service provider to make reasonable accommodations to ensure safety, privacy, and health for all those involved; and third, it prohibits withholding shelter access absent answers to humiliating and intimate questions.[310]

Advocates can also point out that shelter providers have proposed and implemented workable plans. For example, UNITY, an organization that runs Domestic Violence and Homeless Shelters in New Orleans rents apartment buildings throughout the city and houses survivors in these units rather than in large collective groups.[311] In Antelope Valley, the Valley Oasis domestic violence shelter operates from a small resort and offers services to male, female, LGBT, and transgender people by providing small cabins for each survivor.[312] These nondiscriminatory alternatives can easily incorporate transgender survivors. Thus, transgender advocates can argue that the existence of these gender-neutral alternatives suggests that the exclusion of transgender people does not survive intermediate scrutiny analysis.

The lack of evidence demonstrating that transgender women pose a danger to shelters and the workability of transinclusive alternatives may lead the Court to strike down exclusionary policies. Under the Court’s reasoning in Hogan, that some shelters have successfully accommodated transgender women into formerly women-only shelters, amounts to strong proof that gender-neutral alternatives exist and that a gender classification that excludes transgender women is illegitimate.[313] Moreover, in Craig, the Court recognized traffic safety as an important state interest, but nevertheless struck down a gender-based traffic safety rule because statistics demonstrated that there was no big difference between propensity for car accidents between men and women.[314] Under this logic, the absence of evidence demonstrating that transgender women perpetrate violence in transinclusionary shelters may lead the Court to reject drawing a gender line here.[315] The workability of transinclusionary alternatives and the lack of statistics showing that transinclusionary shelters have undermined survivor safety should convince the Court that fears regarding survivor safety and privacy merely reflect a front for impermissible prejudices against transgender women. In particular, they help demonstrate that the purported real differences between cisgender women and transgender women that form the justification for exclusionary shelters remain, in fact, illusory. As such, the Court will more likely recognize that a stereotype persists here.

Furthermore, the Court will likely be receptive to the gender-neutral alternatives that some shelters implement even if they involve making changes to shelter spaces. Justice Ginsburg, when confronted with many of the same privacy concerns associated with allowing women into a single-sex institute, reasoned that accommodations that do not change the fundamental character of the institution are possible: just as admitting women into an all-male military institute would require alterations necessary to “afford members of each sex privacy . . . in living arrangements,” here, gender-neutral bathrooms and separated sleeping quarters can provide sex-neutral alternatives to excluding transgender survivors without fundamentally altering the shelter’s functionality.[316] While it may be more administratively convenient not to make such alterations, that justification alone has never been enough for the Court to uphold a gender classification when using intermediate scrutiny.[317]

This analysis demonstrates that challenging transgender women’s exclusion from domestic violence shelters by bringing sex-based equal protection claims could encourage reform. Such a legal challenge highlights how rationales invoked to exclude transgender women are often based on gender essentialism and on implicit stereotypes of transgender women. Several, gender-neutral, alternatives exist that accommodate the needs of individual survivors without invoking categorical exclusions. Though using binary categories to litigate transgender people’s claims presents certain limitations, the above analysis demonstrates that sex-based equal protection claims represent at least a useful starting point for transgender activists.

D. Operationalizing the “T” in LGBT: Unconstitutional Animus

Although scholars commonly group transgender people under the broad LGBT umbrella, the Court has not explicitly reasoned through the relationship between discrimination based on sexual orientation and gender identity. On the one hand, lower courts considering transgender discrimination claims have compared the discrimination transgender people face as a result of their gender identity to discrimination on the basis of sexual orientation. In Adkins v. City of New York, for example, a New York District Court reasoned that, “[w]hile transgender people and gay people are not identical, they are similarly situated.”[318] On the other hand, Supreme Court jurisprudence on sexual orientation has never explicitly included the “T” in LGBT, instead focusing only on the experiences of LGB people. In light of this mixed record, it is at least conceivable that when confronted with a case implicating transgender identity for the first time, the Court may analyze the issue using the same reasoning it has employed in its jurisprudence on sexual orientation. In that case, Justice Kennedy’s opinions in Romer v. Evans, Lawrence v. Texas, Windsor v. United States, and Obergefell v. Hodges may inform the Court’s analysis.

1. The Supreme Court and Animus: When is Prejudice Unconstitutional?

The Court’s sexual orientation jurisprudence has been markedly different from its jurisprudence regarding race and gender. Instead of decisively invoking the traditional Carolene Products analysis to declare LGB people a suspect class deserving heightened scrutiny, or to decide that LGB people do not constitute a suspect class, thus triggering rational basis review, the Court has put forward a line of precedent, authored by Justice Kennedy, that invokes other concepts to afford rights on the basis of sexual orientation. In particular, the Court analyzes whether animus motivated the classification in question and decides whether to strike down or uphold the classification on that basis. In prior equal protection cases, the Court has deviated from traditional strict, intermediate, or rational basis scrutiny when it perceived that animus motivated laws, with animus representing the Court’s conception of impermissible prejudice. The Court has invoked animus sporadically to strike down laws, and its description of animus has vacillated over time.

In one line of cases, the Court has conceptualized animus to merely mean “spite” or “rank hatred.” In Cleburne and Moreno, for example, the Court characterized animus as a “bare desire to harm a politically unpopular group” and struck down laws that marginalized people with intellectual disabilities and the poor, respectively, without declaring them a suspect class.[319] However, the Court’s conception of animus has not been static; it has evolved over time and among different Justices.

A more capacious understanding of animus appeared first in Palmore v. Sidotti, where a trial court granted custody of a child to the father instead of the mother because the mother was engaged in an extramarital interracial relationship, and the court worried that this would stigmatize the child.[320] However, concerned that the court’s order would give those private racial biases effect, the Court reversed the lower court’s opinion. In a now-famous line, reasoned that, “Private biases may be outside the reach of the law, but the law cannot, directly or indirectly, give them effect.”[321] This concern with private bias and stigmatization, even in the presence of a compelling state interest in protecting the child, suggests that the Court had endorsed a more sweeping conception of animus that went beyond condemning hatred toward a group.[322]

The concept of animus appeared once again in Board of Trustees v. Garett, but the splintered opinion reveals that Supreme Court justices do not share one mind regarding what counts as unconstitutional bias: the dissenting opinion argued that differential treatment based on “negative attitudes, fear, or irrational prejudice”[323] necessarily violated the Equal Protection Clause; the majority opinion, written by Chief Justice Rehnquist, responded that “[a]lthough such biases may often accompany irrational . . . discrimination, their presence alone does not a constitutional violation make.”[324] Occupying a middle ground, Justices Kennedy and O’Connor wrote, “[p]rejudice . . . rises not from malice or hostile animus alone,” but also from “insensitivity caused by simple want of careful, rational reflection” or “some instinctive mechanism to guard against people who appear to be different in some respects from ourselves.”[325] The Court’s sexual orientation jurisprudence repeats these differing conceptions of animus.

Four cases grappling with sexual orientation, all written by Justice Kennedy, have offered different conceptions of animus. In Romer v. Evans, the Court ruled that a state constitutional amendment in Colorado that prevented statutes from being passed to protect discrimination towards LGB people violated the Fourteenth Amendment.[326] The Court reasoned that this law raised the “inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected.”[327] Such a “bare . . . desire to harm a politically unpopular group” could not be a legitimate state interest.[328] Professor Robinson argued that Romer represents the “thin” version of animus because the opinion problematizes only open hostility toward LGB people.[329] Such animus, Professor Robinson wrote, amounts to the “sexual orientation analogue to the racism embodied in Jim Crow discrimination.”[330] This discussion of animus parallels the Court’s opinions in Cleburne and Moreno, where the Court considered that laws violate equal protection principles only if they rest on rank hatred toward a certain group.

In later sexual orientation cases, however, the Court seems to have slowly expanded its conception of animus. Though commentators often describe Lawrence v. Texas as a due process case, scholars have argued that it expanded the Court’s understanding of animus by standing for the principle that “bare moral disapproval of homosexual conduct or homosexual identity is not a valid basis for a law.”[331] The shift in language from “bare . . . desire to harm”[332] in Romer to “moral disapproval”[333] in Lawrence signals a more capacious understanding of unconstitutional animus because the Court problematized not only open hostility but also private prejudices rooted in notions of appropriate sexuality. Similarly, United States v. Windsor, the opinion in which the Court struck down the Defense of Marriage Act (DOMA), cited the Romer opinion’s language on “bare . . . desire to harm” and, like Lawrence, admonished the state for passing laws based on a “moral disapproval” of homosexuality.[334] But the opinion also discussed the ways that the DOMA regime “demeans the [homosexual] couple” and “humiliates tens of thousands of children now being raised by same-sex couples.”[335] This concern expands impermissible animus and parallels the Palmore opinion’s warning that laws may inadvertently give effect to private biases—even when based neither in rank hatred, nor a bare desire to harm, nor moral disapproval.[336] This move is particularly important because it signals a shift from the Court’s traditional focus on discrimination as malignant intent on the part of those instituting the contested law or policy to a focus on the stigmatizing effects of the law on a politically unpopular group. In centering effects rather than intent, the Court thus opens the door to the possibility that even laws passed in good faith may violate the Constitution, when they stigmatize or denigrate a group.

These disparate threads converge in Obergefell, where the principle of animus is at its most capacious. In Obergefell, the Court held that the prohibition of same-sex marriage violates the Fourteenth Amendment.[337] Rather than characterizing those who oppose gay marriage as bigots, Justice Kennedy reasoned that the view that marriage is confined to a union between man and woman has “long has been held—and continues to be held—in good faith by reasonable and sincere people here and throughout the world.”[338] Professor Robinson pointed out that in 2012, the majority of officials in the Democratic Party opposed gay marriage and that many parents, despite loving their gay or lesbian family members, declined to endorse gay marriage.[339] Thus, that “muddle of discomfort, religious concern, and likely political calculation” is at odds with the idea that opposition to gay marriage is borne out of a bare desire to harm.[340] Nonetheless, as Justice Kennedy wrote, when “sincere, personal opposition becomes enacted law and public policy, the necessary consequence is to put the imprimatur of the State itself on an exclusion that soon demeans or stigmatizes those whose own liberty is then denied.”[341] In such cases the law cannot be upheld.[342] In this sense, the Obergefell decision went beyond conceptualizing prejudice as a “bare desire to harm” and instead recognized the ways that “antigay stereotyping and implicit and structural biases against LGBT sexuality and identity” are also a form of prejudice that violates equal protection principles.[343] Notably, Obergefell also decentered the intent behind the exclusionary law at issue and instead problematized the way that laws based on good faith and by “reasonable and sincere people” may inadvertently “demean[] or stigmatize[]” a group in a manner that violates the Constitution’s equal protection principles.[344] This shift has important implications for transgender litigants seeking to challenge exclusionary shelter policies.

2. Transgender Litigants and the Animus Principle

What does this mean for transgender litigants? First, upon invoking the animus principle, the Court does not purport to apply heightened scrutiny or declare lesbian, gay, and bisexual persons a suspect class. Nevertheless, it takes a closer look at legislative motives and requires a tighter “fit” between means and ends than traditional rational basis review requires. Some scholars have conceptualized this type of review as “rational basis with bite.”[345] Professor Pollyogt opined that such an approach reflects a “micro suspect classification analysis,” wherein the Court does not “reify[] . . . categories of concern” by definitively designating suspect class status to a group.[346] Nevertheless, it looks at “the validity of the classification in light of the interests at stake in that particular case.”[347] This context-specific analysis may be particularly suited to dealing with claims by transgender litigants because it does not base its analysis of discrimination on membership in some readily identifiable sociopolitical group, but rather entails a context-specific inquiry that asks whether the law is using impermissible stereotypes to—explicitly or implicitly—stigmatize the plaintiff’s identity. This malleable approach can accommodate the diversity of transgender identity in a way that rigid classifications inherent to the suspect class framework cannot.[348]

Whether transgender litigants will prevail in a claim against a shelter will depend largely on how the Court conceptualizes animus when it reviews shelter policies that exclude transgender women. If the Court applies the version in which unconstitutional animus is simply rank hatred, and nothing more, it will be difficult for transgender litigants to show that shelters excluding transgender people were motivated out of a “bare . . . desire to harm” transgender people.[349] Certainly, some of the earlier feminist rhetoric on transgender people evinces the sentiment that the mere existence of transgender people is the product of illegitimate patriarchal institutions and that an irrational hatred undergirds that view.[350]

More often, however, opponents of transgender inclusion also have genuine concerns that women survivors who have repeatedly faced violence at the hands of a male intimate partner will have traumatic experiences or will be retraumatized by sharing close quarters with those presenting as male or those who may have body parts associated with male identity.[351] At the beginning of her essay, Professor Nicki explicitly affirmed the importance of transgender lives and the important contributions transgender activists have made to breaking down barriers.[352] Nevertheless, she expressed distinct reservations on the matter of transgender inclusion because, as a survivor of intimate partner abuse, she is “cognitively and emotionally unable to make a distinction between a benign penis and one used for harm.”[353] These reflections parallel Professor Robinson’s observations regarding opponents of gay marriage: just as the constituency there faced a “muddle of discomfort, religious concern, and likely political calculation,” the same “muddle” of concern for cisgender survivor discomfort and the threat that the inclusion of transgender survivors might pose to classical feminist curricula exists in the case of transgender inclusion.[354] Because this fear does not manifest itself as rank hatred or even bare moral disapproval of a class as was evident in Romer, Lawrence, and Windsor, the Court might not strike down the exclusion if it limits its conception of unconstitutional animus.

But if the Court embraces the more capacious understanding of animus that runs through the Garrett concurrence and dissent, the Palmore opinion, and, in the context of sexual orientation jurisprudence, the Obergefell opinion, even these “muddled” concerns would violate the Equal Protection Clause. Because the more expansive view of animus problematizes laws that have stigmatizing effects on politically unpopular groups, even if passed in good faith, the Court will not likely consider transexclusionary shelter policies to be considered constitutional under this framework. Domestic violence service providers that advocate for cisgender-women-only shelter spaces may have good faith concerns for exclusion that have nothing to do with moral disapproval, hatred, or a desire to harm transgender people. As Part II discusses, however, because shelter’s exclusionary policies have a demeaning, stigmatizing, and, in some cases, life-threatening effect on transgender women,[355] they can still be found unconstitutional. Transgender litigants can encourage the Court to apply this more capacious understanding of animus to their cases


This Note has exposed the limits of second wave feminist approaches to domestic violence law and policy making. Reviewing the tensions between dominance feminism and intersectional feminism, it has argued that earlier strains of the domestic violence movement, informed by dominance feminism, unfortunately tend to lapse into a gender essentialism that marginalizes women at the intersections of multiple oppressions. In particular, earlier strains of the domestic violence movement regarded IPV as an outgrowth of male dominance over biological females who had been socially trained to be helpless. Those earlier strains of the movement used this vision of the unequal male-female dyad to inform their interventions, and they largely overlooked the needs of LGBT populations. To overcome this lapse into gender essentialism, the domestic violence movement must decenter the cisgender woman and instead structure interventions to focus on the diverse needs of each survivor. Reforming domestic violence law and policy in a manner inclusive of transgender women represents a key step forward in this regard.

The Note has also analyzed the rationales that scholars and some domestic violence service providers have used to exclude transgender women from women-only domestic violence shelters. The concerns that emerge closely parallel those used in legal battles to exclude cisgender men from domestic violence shelters: first, that transgender women have “male privilege” that disrupts the feminist education offered at most shelters; second, that survivors will be retraumatized by having transgender women that present as male in close proximity; third, that allowing transgender women threatens survivors’ safety because male batterers, posing as transgender women, will stalk and kill the survivors; fourth, that because resources are already very slim, they should not be taken away from the cisgender women in need. Although these rationales carry weight when applied to the exclusion of cisgender men, they do not hold up as justifications to exclude transgender women survivors. Many transgender women grow up presenting as female and, thus, face similar gender-based oppressions as cisgender women. Privacy and security concerns have not proven to be a problem in shelters that already include transgender women, which have successfully made reasonable alterations to accommodate survivor discomfort and fear. Finally, the gender asymmetry between cisgender men and women that justifies privileging resources for cisgender women does not carry over to the transgender women context, as transgender women face far higher rates of abuse than do cisgender women. Resource-saving rationales therefore also do withstand scrutiny. Ultimately, this analysis shows that justifications provided in favor of exclusion often rest on stereotypes of transgender survivors as inherently threatening on one hand and cisgender women as inherently vulnerable on the other. Not only do these stereotypes fail to cohere with reality, but the critical material need for shelter space among transgender women far outweighs the discomfort female survivors may feel as a result of including transgender women.

Legal and policy responses must be swift, uniform, and long lasting. In this regard, the Equal Protection Clause represents a key a tool to encourage women‑exclusive, state‑sponsored shelters to make reasonable accommodations to bring transgender women into their ranks. This Note has outlined three possible paths that such a claim might take. First, the Court could decide that transgender people are sufficiently unique that they constitute a category unto themselves. Under this approach, because petitioners will not likely gain suspect class status, the Court will review the exclusion of transgender women under a highly deferential rational basis review test, likely upholding the exclusion of transgender women from women-only shelters. Second, the Court could, in line with prior precedent, attempt to fit transgender people into bifurcated categories of “man” and “woman” based on biology or dominant gender expression. Within this framework, the Court would most likely strike down shelter policies excluding transgender women because transexclusionary policies rest on impermissible stereotypes of what it means to be a woman, and many reasonable gender-neutral alternatives have proven workable for domestic violence shelters. Finally, the Court could conceive of transgender people as part of the “LGBT” grouping, and consider transgender claims as it does those involving sexual orientation, that is, by invoking the “animus” principle. But the Court’s conception of animus has vacillated. If it understands animus to be merely a condemnation of rank hatred, then it will uphold the transexclusionary shelter policies because domestic violence shelters are unlikely to exclude transgender women based on hatred. If, on the other hand, the Court adopts an expansive conception of animus that condemns implicit biases and structural discrimination, irrespective of the intent of those doing the excluding, the Court will likely strike down exclusionary shelters that ban transgender women.

In light of this landscape, transgender litigants seeking to change law around shelters should bring equal protection claims by arguing that transexclusionary shelter policies violate the Fourteenth Amendment’s prohibition on sex-based discrimination and that they constitute a form of unconstitutional animus that reify antitransgender biases.

Rishita Apsani: J.D., University of California, Berkeley, 2018.

           [1].     GLBT Domestic Violence Coal. & Jane Doe Inc., Shelter/Housing Needs for Gay, Lesbian, Bisexual and Transgender (GLBT) Victims of Domestic Violence (2005), [].

           [2].     This Note uses the term cisgender to describe someone who exclusively identifies as their sex assigned at birth. See LGBTQ+ Definitions, Trans Student Educational Resources 2016, []. Consistent with modern usage, the Note also uses the gender-neutral pronouns they and their where appropriate. See Merriam‑Webster’s Dictionary 1298 (11th ed. 2009) (“The use of they, their, them, and themselves as pronouns of indefinite gender and indefinite number is well established in speech and writing, even in literary and formal contexts.”).

           [3].     See, e.g., Sharon Angella Allard, Rethinking Battered Woman Syndrome: A Black Feminist Perspective, 1 UCLA Women’s L.J. 191, 194 (1991) (arguing that the definition of “woman” that undergirds theories like “battered woman syndrome” rests on “limited societal constructs of appropriate behavior for white women” but not for women of color); Michele Bograd, Strengthening Domestic Violence Theories: Intersections of Race, Class, Sexual Orientation, and Gender, 25 J. Marital & Fam. Therapy 275 (1999) (highlighting the ways in which family therapy practices that seek to rehabilitate survivors focus singularly on gender and exclude discussions of race, class, and sexual orientation-based oppressions); Donna Coker, Shifting Power for Battered Women: Law, Material Resources, and Poor Women of Color, 33 U.C. Davis L. Rev. 1009, 1014 (2000) (arguing that domestic violence law has a “tendency to ignore or undervalue the significance of race or ethnicity in shaping the efficacy of universal intervention strategies”); Diana Courvant & Loree Cook-Daniels, Trans and Intersex Survivors of Domestic Violence: Defining Terms, Barriers, and Responsibilities, The Survivor Project, [] (arguing that the domestic violence movement of the 1970s consistently neglects the “growing class of survivors who transcend stereotypes of gender expression or physical sex”); Adele M. Morrison, Changing the Domestic Violence (Dis)Course: Moving from White Victim to Multi-Cultural Survivor, 39 U.C. Davis L. Rev. 1061, 1068 (2006) (“Domestic violence legal discourse is racialized as white and thus fails to adequately respond to the needs of women of color who are victimized by intimate abuse.”); Natalie J. Sokoloff & Ida Dupont, Domestic Violence at the Intersections of Race, Class, and Gender, 11 Violence Against Women 38 (2005) (providing a comprehensive review of the ways in which domestic violence policy has historically ignored race, class, gender, and sexual orientation analysis); Beth E. Richie, A Black Feminist Reflection on the Antiviolence Movement, 25 Signs: J. Women in Culture & Soc’y 1133 (2000) (describing the ways in which antiviolence movements have ignored the needs of Black women).

           [4].     In this Note, the term transgender refers to people who do not identify or exclusively identify with their sex assigned at birth. Accordingly, the term transgender woman refers to those who were not born biologically female, but identify as female.

           [5].     Notably, none of the Supreme Court cases on sexual orientation have addressed the status of transgender people. Thus, this Note uses the term LGB to highlight this absence.

           [6].     Gender essentialism refers to the belief that there exists a monolithic women’s experience that encompasses the marginalization of all women. As a result, in describing the experiences of women, gender essentialists treat differences in race, class, and sexual orientation as unimportant. See, e.g., Angela P. Harris, Race and Essentialism in Feminist Legal Theory, 42 Stan. L. Rev. 581 (1990).

           [7].     See Martha Chamallas, Introduction to Feminist Legal Theory (2003) (referring to “dominance feminism” as a strand of second-generation feminism).

           [8].     See generally R. Emerson Dobash & Russell Dobash, Violence against wives: A Case Against the Patriarchy (1979) (identifying intrafamily violence as an outgrowth of patriarchy).

           [9].     Susan Brownmiller, Against Our Will: Men, Women and Rape 7–8 (1975).

         [10].     Pilar Rodriguez Martinez, Feminism and Violence: The Hegemonic Second Wave’s Encounter with Rape and Domestic Abuse in USA (1970–1985), 23 Cultural Dynamics 147, 149 (2011).

         [11].     Patrick Letellier, Gay and Bisexual Male Domestic Violence Victimization: Challenges to Feminist Theory and Responses to Violence, in Domestic Partner Abuse 4 (L. Kevin Hamberger & Claire Renzetti eds., 1996) (internal quotations omitted).

         [12].     Leigh Goodmark, A Troubled Marriage: Domestic Violence and the Legal System 2 (2012).

         [13].     Letellier, supra note 11, at 2.

         [14].     Id.

         [15].     Second wave feminists in the 1960s used this slogan to push back against the notion that feminist empowerment and consciousness-raising among women was “navel-gazing” or therapeutic rather than challenge oppressive power relationships. See Carol Hanisch, The Personal is Political (2006), [].

         [16].     See, e.g., Phyllis B. Frank & Beverly D. Houghton, Confronting the Batterer (1987) (batterer treatment policy); Daniel J. Sonkin & Michael Durphy, Learning To Live Without Violence (1985) (batterer treatment policies, legal laws); Susan Schechter, Women and Male Violence: The Visions and Struggles of the Battered Women’s Movement (1982) (victimization theory); Del Martin, Battered Wives (1976) (victimization theory).

         [17].     Lenore E. Walker, The Battered Woman 45–54 (1979).

         [18].     Id. at 52.

         [19].     Id.

         [20].     Id.

         [21].     Johnna Rizza, Comment, Beyond Duluth: A Broad Spectrum of Treatment for a Broad Spectrum of Domestic Violence, 70 Mont. L. Rev. 125, 130 (2009).

         [22].     Leigh Goodmark, Reframing Domestic Violence Law and Policy: An Anti-Essentialist Proposal, 31 Wash. U. J.L. & Pol’y 39, 41 (2009).

         [23].     Walker, supra note 17.

         [24].     Goodmark, supra note 22, at 41.

         [25].     Satya P. Krishnan, Judith C. Hilbert, Keith McNeil & Isadore Newman, From Respite to Transition: Women’s Use of Domestic Violence Shelters in Rural New Mexico, 19 J. Fam. Violence 165, 165–69 (2004).

         [26].     Id. at 172-73.

         [27].     Mandatory arrest policies remove police officer discretion when responding to domestic violence by mandating that officers arrest the offender if they find probable cause that a domestic violence offense occurred. See Miriam H. Ruttenberg, Note, A Feminist Critique of Mandatory Arrest: An Analysis of Race and Gender in Domestic Violence Policy, 2 Am. U. J. Gender & L. 171, 171 n.3 (1994).

         [28].     Under a no-drop prosecution policy, the state will not drop a domestic violence case after formal charges have been filed, even if the survivor wishes to withdraw the complaint. See Angela Corsilles, Note, No-Drop Policies in The Prosecution of Domestic Violence Cases: Guarantee to Action or Dangerous Solution?, 63 Fordham L. Rev. 853, 856 (1994).

         [29].     See, e.g., Leigh Goodmark, Autonomy Feminism: An Anti-Essentialist Critique of Mandatory Interventions in Domestic Violence Cases, 37 Fla. St. U. L. Rev. 1 (2009); Ruttenberg, supra note 27; Linda G. Mills, Killing Her Softly: Intimate Abuse and the Violence of State Intervention, 113 Harv. L. Rev. 550, 561 (1999).

         [30].     See Anne M. Coughlin, Excusing Women, 82 Calif. L. Rev. 1 (1994).

         [31].     Goodmark, supra note 22, at 41.

         [32].     Lynette Feder & David B. Wilson, A Meta-Analytic Review of Court-Mandated Batterer Intervention Programs: Can Courts Affect Abusers’ Behavior?, 1 J. Experimental Criminology 239, 245 (2005). See also John R. Barner & Michelle Mohr Carney, Interventions for Intimate Partner Violence: A Historical Review, 26 J. Fam. Violence 235 (2011); Rizza, supra note 21.

         [33].     Intersectional feminism refers to a movement of scholars and activists arguing that a person’s social location—including race, economic class, sexual orientation and other identity markers—informs how they experience gender. See Stephanie A. Shields, Gender: An Intersectionality Perspective, 59 Sex Roles 301, 301 (2008).

         [34].     See generally Katharine T. Bartlett, Gender Law, 1 Duke J. Gender L. & Pol’y 1 (1994); Kimberlé Crenshaw, Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics, U. Chi. Legal F. 139 (1989).

         [35].     Chamallas, supra note 7, at 78–79.

         [36].     Lee Jacobs Riggs, A Love Letter from an Anti-Rape Activist to Her Feminist Sex-Toy Store, in Yes Means Yes: Visions of Female Sexual Power and a World Without Rape 107, 117 (Jaclyn Friedman & Jessica Valenti eds., 2008). See also Angela Davis, The Color of Violence Against Women, 10 Colorlines: Race, Action, Culture (2000), []. (exploring the possibilities of linking a movement demanding remedies for women of color survivors with the abolition of the prison system).

         [37].     Donna Coker, Crime Control and Feminist Law Reform in Domestic Violence Law: A Critical Review, 4 Buff. Crim. L. Rev. 801, 811–12 (2001).

         [38].     Goodmark, supra note 22, at 41.

         [39].     Id.

         [40].     Edward W. Gondolf & Ellen R. Fisher, Battered Women as Survivors: An Alternative to Treating Learned Helplessness 11–18 (1998).

         [41].     Id.

         [42].     Id.

         [43].     Id.

         [44].     Dena Hassouneh & Nancy Glass, The Influence of Gender Role Stereotyping on Women’s Experiences of Female Same-Sex Intimate Partner Violence, 14 Violence Against Women 310 (2008).

         [45].     See, e.g., Christine E. Murray, A. Keith Mobley, Anne P. Buford & Megan M. Seaman‑DeJohn, Same-Sex Intimate Partner Violence: Dynamics, Social Context, and Counselling Implications, 1 J. LGBT Issues in Counselling 7 (2007) (highlighting the ways in which mental health counsellors miss signs of IPV violence in LGBT patients); Kevin L. Ard & Harvey J. Makadon, Addressing Intimate Partner Violence in Lesbian, Gay, Bisexual, and Transgender Patients, 26 J. of Gen. Internal Med. 630 (2011) (discussing the ways in which doctors miss signs of IPV violence in LGBT patients).

         [46].     Alysondra Duke & M. Meghan Davidson, Same-Sex Intimate Partner Violence: Lesbian, Gay, and Bisexual Affirmative Outreach and Advocacy, 18 J. Aggression, Maltreatment & Trauma 795 (2009).

         [47].     Id.

         [48].     Jenna M. Calton, Lauren Bennett Catteaneo & Kris T. Gebhard, Barriers to Help Seeking for Lesbian, Gay, Bisexual, Transgender, and Queer Survivors of Intimate Partner Violence, 17 Trauma, Violence & Abuse 585 (2015).

         [49].     See, e.g., Chandra L. Ford, Terra Slavin, Karin L. Hilton & Susan L. Holt, Intimate Partner Violence Prevention Services and Resources in Los Angeles: Issues, Needs, and Challenges for Assisting Lesbian, Gay, Bisexual, and Transgender Clients, 14 Health Promotion Prac. 841 (2013).

         [50].     Id. at 841.

         [51].     Taylor N.T. Brown & Jody L. Herman, Intimate Partner Violence And Sexual Abuse Among LGBT People, The Williams Institute (Nov. 2015), [].

         [52].     Id. at 3 (“Findings of lifetime IPV among transgender people from purposive studies range from 31.1% to 50.0%.”).

         [53].     Leigh Goodmark, Transgender People, Intimate Partner Abuse, and the Legal System, 48 Harv. C.R.-C.L. L. Rev. 51, 55 (2013).

         [54].     Nicola Brown, Holding Tensions of Victimization and Perpetration: Partner Abuse in Trans Communities, in Intimate Partner Violence in LGBTQ Lives 153, 156 (Janice L. Ristock ed., 2011).

         [55].     Id.

         [56].     Calton, supra note 48.

         [57].     Rebecca L. Stotzer, Violence Against Transgender People: A Review of United States Data, 14 Aggression & Violent Behav. 170 (2009).

         [58].     Jaime M. Grant, Lisa A. Mottet & Justin Tanis, Injustice at Every Turn: A Report of the National Transgender Discrimination Survey 3, Nat’l Gay and Lesbian Task Force & Nat’l Ctr. for Transgender Equality, [].

         [59].     Judith Halberstam & C. Jacob Hale, Butch/FTM Border Wars: A Note on Collaboration, 4 GLQ: J. Lesbian & Gay Stud. 283, 283–85 (1998).

         [60].     Implications of 2004–2005 Transgender Sexual Violence Survivor Research, Forge (Aug. 1, 2005), []; Caroline White, Re/Defining Gender and Sex: Educating for Trans, Transsexual, and Intersex Access and Inclusion to Sexual Assault Centres and Transition Houses (2002) (unpublished M.A. Thesis, University of British Columbia) (on file with author).

         [61].     See Michigan Womyn’s Music Festival, Human Rights Campaign (July 30, 2014), [].

         [62].     Janice G. Raymond, The Transsexual Empire: The Making of the She-male xxiv (1979).

         [63].     White, supra note 60 (summarizing Jefferey’s work) (internal quotations omitted).

         [64].     See id.

         [65].     See generally Emilio C. Viano, Intimate Violence: Interdisciplinary Perspectives (1992).

         [66].     See, e.g., Sara R. Epstein, Glenda Russell & Louise Silvern, Structure and Ideology of Shelters for Battered Women, 16 Am. J. Community Psychol. 345 (1988).

         [67].     Janice Haaken & Nan Yragui, Going Underground: Conflicting Perspectives on Domestic Violence Shelter Practices, 13 Feminism & Psychol. 49, 53 (2003).

         [68].     Susan Schechter, Building Bridges Between Activists, Professionals, and Researchers, in Feminist Perspectives on Wife Abuse 299, 302 (Kersti Yllö & Michelle Bograd eds., 1988).

         [69].     Id.

         [70].     Id.

         [71].     See, e.g., Micheline Beaudry, Battered Women (1985); Nancy Janovicek, No Place to Go: Local Histories of the Battered Women’s Shelter Movement (2007).

         [72].     Merle H. Weiner, From Dollars to Sense: A Critique of Government Funding for the Battered Women’s Shelter Movement, 9 L. & Ineq. 185, 190 (1991).

         [73].     Id. at 189.

         [74].     See id. See also Beaudry, supra note 71, at 88; Wiener, supra note 72, at 191 n.23 (“It conveys a message to society of the strength of women working together.”).

         [75].     See Bell Hooks, Ain’t I a Woman: Black Women and Feminism (1981).

         [76].     Edward Gondolf, Ellen Fisher & Richard McFerron, Racial Differences Among Shelter Residents: A Comparison of Anglo, Black, and Hispanic Battered Women, 3 J. Fam. Violence 39 (1988).

         [77].     Kimberle Crenshaw, Mapping the Margins: Intersectionality, Identity Politics, and Violence Against Women of Color, 43 Stanford L. Rev. 1241, 1262–65 (1991).

         [78].     Id. (emphasis omitted).

         [79].     Lisa M. Martinson, An Analysis of Racism and Resources for African-American Female Victims of Domestic Violence in Wisconsin, 16 Wis. Women’s L.J. 259, 269 (2001).

         [80].     Haaken & Yragui, supra note 67, at 50.

         [81].     Martinson, supra note 79, at 269. See also Zanita E. Fenton, Domestic Violence in Black and White: Racialized Gender Stereotypes in Gender Violence, 8 Colum. J. Gender & L. 1, 24–25 (1998).

         [82].     Haaken & Yragui, supra note 67, at 65.

         [83].     Id. at 65; see also Crenshaw, supra note 77, at 93–118; Jenny Rivera, Domestic Violence Against Latinas by Latino Males: An Analysis of Race, National Origin, and Gender Differentials, 14 B.C. Third World L.J. 231 (1994); Susan B. Sorenson, Violence Against Women: Examining Ethnic Differences and Commonalities, 20 Evaluation Rev. 123 (1996).

         [84].     Haaken & Yragui, supra note 67, at 65 (internal quotations omitted).

         [85].     Id.

         [86].     Goodmark, supra note 53.

         [87].     Grant, supra note 58, at 4.

         [88].     Viviane K. Namaste, Invisible Lives: The Erasure of Transsexual and Transgendered People 180 (2000).

         [89].     Id.

         [90].     Goodmark, supra note 53, at 69.

         [91].     Ryka Aoki, On Living Well and Coming Free, in Gender Outlaws: The Next Generation 143, 148 (Kate Bornstein & S. Bear Bergman eds., 2010).

         [92].     Shelter/Housing Needs for Gay, Lesbian, Bisexual and Transgender (GLBT) Victims of Domestic Violence, supra note 1.

         [93].     Valerie B et al., Conference, Lesbian, Gay, Bisexual, and Transgender Communities and Intimate Partner Violence, 29 Fordham Urb. L.J. 121, 147 (2001).

         [94].     Goodmark, supra note 53, at 51.

         [95].     See id.

         [96].     Id.

         [97].     Virginia Woo, Jeanne d’Arc, Maria N. Penn & Clara, The Struggle for Women Only Space, Kinesis 15 (1998).

         [98].     Goodmark, supra note 53, at 69.

         [99].     A. Nicki, Women’s Spaces are not Trans Spaces: Maintaining Boundaries of Respect, in Trans/forming Feminisms: Trans/feminist Voices Speak Out 154 (Krista Scott-Dixon ed., 2006).

      [100].     Nicki, supra note 99, at 157.

      [101].     Id.

      [102].     Id. at 155.

      [103].     Id.

      [104].     Dean Spade, Documenting Gender, 59 Hastings L.J. 731, 753 (2008).

      [105].     Brief for Haven Hills, Inc. et al. as Amici Curiae Supporting Respondents, Blumhorst, v. Haven Hills, Inc. et al., No. B170904, 2004 WL 3218210, at *2–31 (Cal. App. 2 Dist. 2004).

      [106].     Id.

      [107].     Id.

      [108].     Id.

      [109].     Brief for Haven Hills, Inc., supra note 105.

      [110].     Id.

      [111].     Id.

      [112].     For example, some sub-classes of men may be as equally underserved as cisgender women when it comes to domestic violence services, including gay and bisexual men. See, e.g., Catherine Finneran, Anna Chard, Craig Sineath, Patrick Sullivan & Rob Stephenson, Intimate Partner Violence and Social Pressure among Gay Men in Six Countries, 13 West. J. Emerg. Med. 260, 260 (2012) (“Recent research suggests that men who have sex with men (MSM) experience intimate partner violence (IPV) at significantly higher rates than heterosexual men.”). Moreover, some IPV workers argue for gender-integrated sexual assault groups under the reasoning that being around nonviolent men could help survivors overcome trauma. See Part of the Solution: Gender-Integrated Sexual Assault Support Groups, Forge (July 20, 2017), [].

      [113].     Woo, supra note 97, at 15.

      [114].     See, e.g., Michael P. Johnson, Conflict and Control: Gender Symmetry and Asymmetry in Domestic Violence, 12 Violence Against Women 1003 (2006); Bograd, supra note 3.

      [115].     Shelter/Housing Needs for Gay, Lesbian, Bisexual and Transgender (GLBT) Victims of Domestic Violence, supra note 1.

      [116].     See Lisa Mottet & John M. Ohle, Transitioning our Shelters: A Guide to Making Homeless Shelters Safe for Transgender People, Nat’l Gay & Lesbian Task Force Pol’y Inst. & Nat’l Coalition for the Homeless (2003), [].

      [117].     24 C.F.R. § 5.106 (2017).

      [118].     Id.

      [119].     Id.

      [120].     Non-Discrimination Laws, Movement Advancement Project (June 4, 2018) []

      [121].     See, e.g., National Equality Map, Transgender L. Ctr. (June 4, 2018), [] (providing an overview of discriminatory laws against transgender people by state).

      [122].     See infra notes 166–67 and accompanying text.

      [123].     See 24 C.F.R. § 5.106 (2017).

      [124].     Id.

      [125].     Shannon Price Minter, Do Transsexuals Dream of Gay Rights?, in Transgender Rights 141, 142 (Paisley Currah, Richard M. Juang & Shannon Price Minter eds., 2006).

      [126].     135 S. Ct. 2584 (2015).

      [127].     The court was poised to hear arguments from transgender litigant Gavin Grimm, who claimed, among other things, that his school’s refusal to allow him to use the bathroom that corresponded with his gender identity violated the Fourteenth Amendment’s Equal Protection Clause. See Gloucester Cty. Sch. Bd. v. G.G. ex rel. Grimm, 137 S. Ct. 369 (2016) (granting cert petition). However, the Court ultimately remanded the case to the Fourth Circuit in light of new executive guidance on the issue. Robert Barnes, Supreme Court Sends Virginia Transgender Case Back to Lower Court, Wash. Post (Mar. 6, 2017), [].

      [128].     Kevin M. Barry, Brian Farrell, Jennifer L. Levi & Neelima Vanguri, A Bare Desire to Harm: Transgender People and the Equal Protection Clause, 57 B.C. L. Rev. 507 (2016).

      [129].     Erwin Chemerinsky, Constitutional Law: Principles and Policies 668 (4th ed. 2011).

      [130].     Id. at 669.

      [131].     Id.

      [132].     Id. at 677.

      [133].     See, e.g., City of Richmond v. J.A. Croson Co., 488 U.S. 469, 493 (1989).

      [134].     See United States v. Virginia, 518 U.S. 515 (1996).

      [135].     See Miss. Univ. for Women v. Hogan, 458 U.S. 718 (1982).

      [136].     Id.

      [137].     Id.

      [138].     Id.

      [139].     304 U.S. 778, 784 n.4 (1938).

      [140].     See generally Johnson v. California, 543 U.S. 499 (2005).

      [141].     See Frontiero v. Richardson, 411 U.S. 677, 685–88 (1973) (using these factors to decide that women constitute a suspect class deserving of heightened scrutiny).

      [142].     Compare Johnson, 543 U.S. 499 (applying strict scrutiny to a classification based on race) and United States v. Virginia, 518 U.S. 515 (1996) (applying intermediate scrutiny to gender-based classifications) with San Antonio Indep. Sch. Distr. v. Rodriguez, 411 U.S. 1 (1973) (denying suspect class status to the poor) and City of Cleburne, Texas v. Cleburne Living Ctr., Inc., 473 U.S. 432 (1985) (denying suspect class status to the intellectually disabled).

      [143].     Rodriguez, 411 U.S. at 19 (“[D]isadvantaged ‘poor’ cannot be identified or defined in customary equal protection terms.”).

      [144].     See id.

      [145].     Rodriguez, 411 U.S. at 19.

      [146].     Cleburne Living Ctr., Inc., 473 U.S. at 432.

      [147].     Id. at 472.

      [148].     Id. at 473.

      [149].     Id.

      [150].     See, e.g., Windsor v. United States, 699 F.3d 169, 176 (2d Cir. 2012) (“Section 3 of DOMA is subject to intermediate scrutiny under the factors enumerated in City of Cleburne v. Cleburne Living Center . . . .”)

      [151].     See Dale Carpenter, Windsor Products: Equal Protection from Animus, 2013 Sup. Ct. Rev. 183 (2013).

      [152].     Evan Gerstmann, The Constitutional Underclass: Gays, Lesbians, and the Failure of Class-Based Equal Protection 60 (1999).

      [153].     High Tech Gays, v. Def. Indus. Sec. Clearance Office, 895 F.2d 563, 573 (9th Cir. 1990).

      [154].     See, e.g., Darren Lenard Hutchinson, “Not without Political Power”: Gays and Lesbians, Equal Protection and the Suspect Class Doctrine, 65 Ala. L. Rev. 975, 995-1003 (2014) (discussing cases that regard LGBT persons as having political power).

      [155].     Romer v. Evans, 517 U.S. 620, 646–46 (1996); Darren Lenard Hutchinson, Identity Crisis: “Intersectionality,” “Multidimensionality,” and the Development of an Adequate Theory of Subordination, 6 Mich. J. Race & L. 285, 300 (2001).

      [156].     Windsor v. United States, 699 F.3d 169, 180–84 (2d Cir. 2012).

      [157].     Nat’l Ctr. for Transgender Equality & the Nat’l Gay and Lesbian Task Force, National Transgender Discrimination Survey 1 (2009), [].

      [158].     Grant, supra note 58, at 3.

      [159].     Id.

      [160].     Press Release, U.S. Census Bureau, Income, Poverty and Health Insurance Coverage in the United States: 2016 (Sept. 12, 2017), [].

      [161].     Id.

      [162].     Id.

      [163].     Id.

      [164].     Am. Psychol. Ass’n, Position Statement on Discrimination Against Transgender and Gender Variant Individuals (2012), [].

      [165].     Id.

      [166].     See Sandhya Somashekhar, Moriah Balingit & Emma Brown, Trump Administration Poised to Change Transgender Student Bathroom Guidelines, Wash. Post (Feb. 21, 2017), [].

      [167].     See Daniel Trotta, Transgender Military Personnel Sue Trump Over Service Ban, Reuters (Aug. 9, 2017), [] (internal quotation marks omitted).

      [168].     Martha Bebinger, Trump Swaps Complete Ban For ‘Qualified Ban’ On Transgender Military Service, NPR (Mar. 24, 2018), [].

      [169].     See id.

      [170].     See, e.g., Michael Gordon, Mark S. Price & Katie Peralta, Understanding HB2: North Carolina’s Newest Law Solidifies State’s Role in Defining Discrimination, Charlotte Observer (Mar. 26, 2016), []. The North Carolina legislature subsequently repealed this measure. See Daniella Silva, HB2 Repealed: North Carolina Overturns Controversial ‘Bathroom Bill’, NBC News (Mar. 30, 2017), [].

      [171].     Brocksmith v. United States, 99 A.3d 690, 698 n.8 (D.C. 2014).

      [172].     Adkins v. City of New York, 143 F. Supp. 3d 134, 139–40 (S.D.N.Y. 2015).

      [173].     City of Cleburne, Texas v. Cleburne Living Ctr., Inc., 473 U.S. 432, 442 (1985) (emphasis added).

      [174].     See Paisley Currah, Expecting Bodies: The Pregnant Man and Transgender Exclusion from the Employment Non-Discrimination Act, 36 Women’s Stud. Q. 331, 330 (2008).

      [175].     Goodmark, supra note 53, at 69.

      [176].     Windsor v. United States, 699 F.3d 169, 183 (2d Cir. 2012).

      [177].     Id. at 183–84.

      [178].     Adkins v. City of New York, 143 F. Supp. 3d 134, 139–40 (S.D.N.Y. 2015).

      [179].     See Valerie B, supra note 93, at 147; see also Mottet & Ohle, supra note 116.

      [180].     Unfortunately, it remains difficult to pinpoint how often this type of discrimination occurs among the transgender population specifically, because there exists little to no reliable empirical research on the experiences of transgender people outside of the medical context. See, e.g., Mottet & Ohle, supra note 116, at 6 (“No one has a clear idea of the true number of transgender individuals in need of services, because no one is accurately counting.”). However, surveys on the prevalence of discrimination against the larger LGBTQ population, which includes transgender people, suggest that discrimination is highly likely. See Nat’l Coal. of Anti-Violence Programs, Community Action Toolkit for Addressing Intimate Partner Violence Against Transgender People 2 (2014), [] (“More than 61% of LGBTQ survivors were turned away from domestic violence shelter[s] in 2012 . . . .”). In a survey of domestic violence shelters, “94% of the respondents said they were not serving LGBTQ survivors.” Id.

      [181].     See Marc R. Shapiro, Treading the Supreme Court’s Murky Immutability Waters, 38 Gonz. L. Rev. 409, 430 n.169 (2002).

      [182].     See id.

      [183].     See Kylar W. Broadus, The Evolution of Employment Discrimination Protections for Transgender People, in Transgender Rights 93, 98 (Paisley Currah, Richard M. Juang & Shannon Minter eds., 2006).

      [184].     Paisley Currah, Gender Pluralisms Under the Transgender Umbrella, in Transgender Rights 3, 18 (Paisley Currah, Richard M. Juang & Shannon Price Minter eds., 2006).

      [185].     Id. (internal quotation marks omitted).

      [186].     Id.

      [187].     See Barry, supra note 128, at 560.

      [188].     See Kyla Bender-Baird, Transgender Employment Experiences: Gendered Perceptions and the Law 36 (2011).

      [189].     See San Antonio Indep. Sch. Distr. v. Rodriguez, 411 U.S. 1, 19 (1973).

      [190].     See William D. Araiza, Enforcing the Equal Protection Clause: Congressional Power, Judicial Doctrine, and Constitutional Law 67–68, 71, 115 (2015) (highlighting the way in which the Court regularly invokes the “floodgates” argument to deny suspect class status).

      [191].     See Ross, infra note 193, at 1814.

      [192].     See Russell K. Robinson, Unequal Protection, 68 Stan. L. Rev. 151, 154 (2016).

      [193].     Bertrall L. Ross, Administering Suspect Classes, 66 Duke L. J. 1807, 1814 (2017).

      [194].     Compare Windsor, 699 F.3d 169, 176 (2d. Cir 2012) (“Section 3 of DOMA is subject to intermediate scrutiny under the factors enumerated in City of Cleburne v. Cleburne Living Center.”) with Windsor v. United States (Windsor II), 133 S. Ct. 2675 (containing no analysis of suspect class status).

      [195].     See Brief for Haven Hills, Inc., supra note 105, at 29–30.

      [196].     See id.

      [197].     Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 303 (1978).

      [198].     United States v. Virginia, 518 U.S. 515, 533 (1996) (emphasis added).

      [199].     See id. at 536–39.

      [200].     See, e.g., Craig v. Boren, 429 U.S. 190, 201–02 (1976).

      [201].     See, e.g., Reed v. Reed, 404 U.S. 71, 76–77 (1971).

      [202].     See id.

      [203].     See Reed, 404 U.S. at 76–77.

      [204].     Frontiero v. Richardson, 411 U.S. 677, 680–81 (1973).

      [205].     Id. at 681.

      [206].     Id. at 684.

      [207].     Craig v. Boren, 429 U.S. 190, 191–92 (1976).

      [208].     Id. at 201–02.

      [209].     Keith Cunningham-Parmeter, (Un)equal Protection: Why Gender Equality Depends On Discrimination, 109 Nw. U. L. Rev. 1, 1, 19 (2015).

      [210].     United States v. Virginia, 518 U.S. 515 (1996) [hereinafter VMI]; Miss. Univ. for Women v. Hogan, 458 U.S. 718 (1982); Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721 (2003).

      [211].     VMI, 518 U.S. at 520, 536.

      [212].     Id. at 549.

      [213].     Id. at 522.

      [214].     Id. at 524–25.

      [215].     Id. at 535.

      [216].     Id. at 542–43.

      [217].     Id. at 550–51 n.19.

      [218].     Id. at 550 (internal quotations and emphasis omitted).

      [219].     458 U.S. 718, 725 (1982).

      [220].     Id. at 728.

      [221].     Id. at 729.

      [222].     Hogan, 458 U.S. at 729.

      [223].     Id. at 731.

      [224].     Nevada Dep’t of Human Res. v. Hibbs, 538 U.S. 721 (2003).

      [225].     Id. at 731–36

      [226].     United States v. Virginia, 518 U.S. 515, 533–34 (1996) (internal citations omitted).

      [227].     Califano v. Webster, 430 U.S. 313, 318 (1977) (quoting Kahn v. Shevin, 416 U.S. 351, 353 (1974)) (holding that a formula that gave more old-age insurance benefits to women than men was not discriminatory because the statute compensated women for past economic discrimination by taking into account the wage gap between men and women and in no way penalized women wage earners).

      [228].     Cunningham-Parmeter, supra note 209, at 7.

      [229].     Michael M. v. Superior Court of Sonoma Cty., 450 U.S. 464, 467 (1981).

      [230].     Id. at 467.

      [231].     Id. at 467, 473.

      [232].     Nguyen v. I.N.S., 533 U.S. 53, 59–60 (2001).

      [233].     Nguyen, 533 U.S. at 59–60.

      [234].     Id. at 64–65.

      [235].     Id. at 62.

      [236].     Id. at 65.

      [237].     Id. at 64.

      [238].     See Katherine M. Franke, The Central Mistake of Sex Discrimination Law: The Disaggregation of Sex from Gender, 144 U. Pa. L. Rev. 1, 3 (1995).

      [239].     See Michael M. v. Superior Court of Sonoma Cty., 450 U.S. 464, 467 (1981).

      [240].     See Nguyen, 533 U.S. at 65.

      [241].     See id.

      [242].     See Frontiero v. Richardson, 411 U.S. 677, 684 (1973).

      [243].     See Nguyen, 533 U.S. at 68.

      [244].     See Miss. Univ. for Women v. Hogan, 458 U.S. 718, 726 (1982).

      [245].     See Frontiero, 411 U.S. at 684.

      [246].     See Nguyen, 533 U.S. at 65.

      [247].     Robinson, supra note 192, at 154 n.14.

      [248].     Franke, supra note 238, at 3.

      [249].     See Regents of Univ. of California v. Bakke, 438 U.S. 265, 303 (1978).

      [250].     See Geduldig v. Aiello, 417 U.S. 484, 494 (1974).

      [251].     See id.

      [252].     See Geduldig, 417 U.S. at 496–97.

      [253].     See id. at 496.

      [254].     See id.

      [255].     See id. at 503 (Brennan, J., dissenting) (“[T]he Court appears willing to abandon that higher standard of review without satisfactorily explaining what differentiates the gender-based classification employed in this case from those found unconstitutional in Reed and Frontiero.”).

      [256].     See id. at 496–97.

      [257].     See Geduldig, 417 U.S. at 494 (emphasis added).

      [258].     Miss. Univ. for Women v. Hogan, 458 U.S. 718, 729 (1982).

      [259].     See supra Section II.B.

      [260].     See supra note 112 and accompanying text.

      [261].     See Woo, supra note 97, at 16.

      [262].     See Craig v. Boren, 429 U.S. 190, 202 (1976).

      [263].     See Hogan, 458 U.S. at 728.

      [264].     See United States v. Virginia, 518 U.S. 515, 533–34 (1996) (internal citations omitted).

      [265].     Califano v. Webster, 430 U.S. 313, 318 (1977).

      [266].     See Califano, 430 U.S. at 318.

      [267].     See Hogan, 458 U.S. at 728.

      [268].     See id.

      [269].     See Michaela Rogers, Challenging Cisgenderism Through Trans People’s Narratives of Domestic Violence and Abuse, 0 Sexualities 1, 1 (problematizing the “gender asymmetry debate” as prone to “cisgenderism and a heteronormative bias”).

      [270].     See Brown, supra note 51, at 3.

      [271].     See Hogan, 458 U.S. at 728–30.

      [272].     See supra notes 114–115.

      [273].     See Hogan, 458 U.S. at 729.

      [274].     See Hassouneh & Glass, supra note 44, at 312.

      [275].     See Nguyen v. I.N.S., 533 U.S. 53, 89 (2001) (O’Connor, J., dissenting).

      [276].     See Frontiero v. Richardson, 411 U.S. 677, 684 (1973).

      [277].     See Nevada Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003).

      [278].     See United States v. Virginia, 518 U.S. 515, 550 (1996).

      [279].     See Craig v. Boren, 429 U.S. 190, 201–02 (1976).

      [280].     See generally supra Part II (discussing the rise of the shelter movement in the 1970s).

      [281].     See Haaken & Yragui, supra note 67, at 53, 49.

      [282].     See Nicki, supra note 99, at 154–55.

      [283].     See Woo, supra note 97, at 15–16.

      [284].     See Nicki, supra note 99, at 157.

      [285].     See generally supra Part II (discussing shelters’ attempts to empower survivors through feminist education and the fostering of healthy female relationships).

      [286].     See United States v. Virginia, 518 U.S. 515, 535 (1996).

      [287].     See Epstein, supra note 66, at 345–46.

      [288].     See Haaken & Yragui, supra note 67, at 56.

      [289].     See Virginia, 518 U.S. at 550 (internal quotations and emphasis omitted).

      [290].     See Miss. Univ. for Women v. Hogan, 458 U.S. 718, 726 (1982); Michael M. v. Superior Court of Sonoma Cty., 450 U.S. 464, 467 (1981).

      [291].     See, e.g., Judith Butler, Gender Trouble: Feminism and the Subversion of Identity (1990) (criticizing the binary that gender is socially constructed and sex is biologically determined, and instead arguing that both sex and gender are both socially constructed).

      [292].     See Nguyen v. I.N.S., 533 U.S. 53, 68 (2001).

      [293].     See id. at 62.

      [294].     See Nicki, supra note 99, at 155; Nguyen, 533 U.S. at 62.

      [295].     See Woo, supra note 97, at 15.

      [296].     See Goodmark, supra note 53, at 69.

      [297].     See, e.g., Price Waterhouse v. Hopkins, 490 U.S. 228, 258 (1989) (holding that when stereotypes regarding what constitutes appropriate female behavior guide employment decisions in part, the employer has engaged in an unlawful employment practice).

      [298].     See supra Section II.A.

      [299].     See supra Section II.B.

      [300].     See Weiner, supra note 72, at 58.

      [301].     See Part of the Solution, supra note 112.

      [302].     See United States v. Virginia, 518 U.S. 515, 550 (1996).

      [303].     Id. at 543.

      [304].     See id. at 550 n.19.

      [305].     See Goodmark, supra note 53, at 69; Woo, supra note 97, at 16; Nicki, supra note 99, at 154–55.

      [306].     Katy Steinmetz, Why LGBT Advocates Say Bathroom ‘Predators’ Argument is Red Herring, Time (May 2, 2016), [] (collecting statements from police officers and school officers who say that the adoption of gender-neutral bathroom laws have not resulted in an uptick of sexual assaults in bathrooms).

      [307].     Marcie Bianco, Statistics Show Exactly How Many Trans People Have Attacked You in Bathrooms, Mic (Apr. 2, 2015), [] (stating that there have been zero incidents of transgender people attacking others in bathrooms); Jeff Brady, When A Transgender Person Uses A Public Bathroom, Who Is At Risk?, NPR (May 15, 2016), [] (stating that, in states that have allowed transgender people to use the bathroom of their choice, “[a]s far as we know there hasn’t been some sort of, you know, devolving into chaos in restrooms”); Emanuella Grinberg & Dani Stewart, 3 Myths that Shape the Transgender Bathroom Debate, CNN (Mar. 7, 2017), [] (noting that there has been only one case of a Seattle man undressing in a women’s locker room using gender-neutral policies as his motivation).

      [308].     See Mottet & Ohle, supra note 116, at 14.

      [309].     Id. at 13–14.

      [310].     See U.S. Dep’t of Housing and Urban Dev., Equal Access for Transgender People 15,

      [311].     See Ms. Wendy, Unity (Aug. 3, 2010), [] (describing the story of a domestic violence survivor who used UNITY services).

      [312].     See What We Do: Valley Oasis Shelter, Valley Oasis, [] (describing facilities offered by the shelter and their client base).

      [313].     See Miss. Univ. for Women v. Hogan, 458 U.S. 718, 731 (1982).

      [314].     See Craig v. Boren, 429 U.S. 190, 201–02 (1976).

      [315].     See supra note 308.

      [316].     See United States v. Virginia, 518 U.S. 515, 550 n.19 (1996).

      [317].     See Reed v. Reed, 404 U.S. 71, 76–77 (1971).

      [318].     Adkins v. City of New York, 143 F. Supp. 3d 134, 139 (S.D.N.Y. 2015).

      [319].     Robinson, supra note 192, at 193 (internal citations omitted).

      [320].     See Palmore v. Sidoti, 466 U.S. 429, 431–32 (1984).

      [321].     Id. at 433.

      [322].     See id.

      [323].     Bd. of Tr. of Univ. of Ala. v. Garrett, 531 U.S. 356, 381 (2001) (Breyer, J., dissenting) (quoting Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 448, 450 (1985)) (internal quotation marks omitted).

      [324].     Id. at 367.

      [325].     Id. at 374 (Kennedy, J., concurring).

      [326].     Romer v. Evans, 517 U.S. 620, 635 (1996).

      [327].     Id. at 634.

      [328].     Id.

      [329].     Robinson, supra note 192, at 185.

      [330].     Id. at 186.

      [331].     Susannah W. Pollvogt, Unconstitutional Animus, 81 Fordham L. Rev. 887, 891 (2012).

      [332].     Romer, 517 U.S. at 634.

      [333].     Lawrence v. Texas, 539 U.S. 558, 582 (2003) (O’Connor, J., concurring).

      [334].     Windsor v. United States (Windsor II), 133 S. Ct. 2675, 2694 (2013).

      [335].     Id.

      [336].     See Palmore v. Sidoti, 466 U.S. 429, 431–32 (1984).

      [337].     Id.

      [338].     Obergefell v. Hodges, 135 S. Ct. 2584, 2594 (2015).

      [339].     See Robinson, supra note 192, at 187–88.

      [340].     See id. at 188.

      [341].     Obergefell, 135 S. Ct. at 2602.

      [342].     See Robinson, supra note 192, at 162.

      [343].     See id. at 186.

      [344].     Obergefell, 135 S. Ct. at 2602.

      [345].     Pollvogt, supra note 331, at 899; see also Raphael Holoszyc-Pimentel, Reconciling Rational-Basis Review: When Does Rational Basis Bite?, 90 N.Y.U. L. Rev. 2070 (2015).

[346].    Susannah W. Pollvogt, Unconstitutional Animus, 81 Fordham L. Rev., 888, 927 (2012).

      [347].     Pollvogt, supra note 331, at 927 (internal quotation marks omitted).

      [348].     See supra Section III.B. (discussing the ways that the Court’s emphasis on “discreteness” and “immutability” harm the diversity of transgender identity).

      [349].     Romer v. Evans, 517 U.S. 620, 634 (1996)

      [350].     Raymond, supra note 62, at 10.

      [351].     See Nicki, supra note 99, at 154–55.

      [352].     Id. at 154.

      [353].     Id. at 155–57.

      [354].     See Robinson, supra note 192, at 188.

      [355].     See supra Section II.B (recounting the problems transgender people have accessing safe domestic violence shelters).


More in this Issue

A Theory of Corporate Joint Ventures

In a corporate joint venture, two corporations—often competitors—collaborate on a project. But how can corporations be partners and competitors at the same time? Though it sounds like a contradiction, such collaborations are commonplace. Many of the most familiar products come from corporate joint ventures, from high-technology like solid-state drives for laptops or rocket boosters for […]

Remedial Convergence and Collapse

This Article describes and interrogates a phenomenon of spillovers across remedies—how the legal standards governing the availability of remedies in cases regarding executive violations of individuals’ constitutional rights, particularly in the area of policing, have converged around similar ideas that narrow the availability of several different remedies. A similar set of limits restricts the availability […]

Legislating for Litigation: Delegation, Public Policy, and Democracy

When Congress enacts command-and-control regulation, it chooses between implementation through litigation and courts, through bureaucracy, or through a hybrid regime. Since the late 1960s, the frequency with which Congress has relied on civil litigation for frontline enforcement of statutes grew dramatically, and with it grew rates of federal statutory litigation and the role of courts […]

Redefining the Legality of Undocumented Work

Undocumented workers face a new harsh reality under the Trump administration. Federal law’s prohibition of undocumented work has facilitated exploitation because workers fear being brought to the attention of immigration authorities. The current administration’s aggressive stance towards worksite enforcement will only exacerbate abuses against undocumented workers, such as wage theft, dangerous working conditions, or human […]

All Disputes Must Be Brought Here: Atlantic Marine and the Future of Multidistrict Litigation

Multidistrict litigation (“MDL”) is an immensely powerful tool. In an MDL, cases that share a common question of fact are consolidated in a single district for pretrial proceedings. MDLs abide by the general principle that governs all transfers within the federal system: because transfer is no more than a “housekeeping measure,” an action retains the […]