The Yale Law Journal

May 2014

Domestic Violence Asylum After Matter of L-R-

abstract. Women fleeing severe domestic violence have sought asylum in the United States for at least twenty years. Yet the legal system has been reluctant to understand domestic violence as occurring “on account of” gender or to see domestic violence victims as deserving asylum-seekers. Even after the Department of Homeland Security indicated its support for some domestic violence asylum claims in Matter of L-R-, there remains no binding legal precedent supporting such claims. This Note first argues that feminist theory and sociological evidence present a compelling case that domestic violence occurs “on account of” gender and that domestic violence victims may thus be eligible for asylum as members of a “particular social group” of women. The Note then critiques the weak legal framework that now exists for domestic violence asylum claims and proposes a novel regulatory reform to put such claims on firmer footing. A regulatory reform explicitly embracing domestic violence asylum would not only provide meaningful relief to many asylum-seekers but also be a significant symbolic statement of U.S. commitment to women’s human rights.

author. Yale Law School, J.D. expected 2014. I am deeply grateful to Jean Koh Peters for introducing me to asylum law and the discipline of client-centered lawyering, to Alexandra Lu and Jessica Vosburgh for countless hours of teamwork and camaraderie, and to “Angela” for putting her trust in us. Thanks also to Nicholas Parrillo for his guidance on the early drafts of this project, to Cristina Rodríguez for thoughtful comments at a workshop on this Note, and to Ravi Ramanathan and the editors of the Yale Law Journal for valuable suggestions throughout the editorial process.


If there is a popular conception of an asylum-seeker, it might be Victor Laszlo, the fictional fugitive leader of the anti-Nazi Resistance who escaped on a late-night flight to freedom in the movie Casablanca.1 A more modern example is Chen Guangcheng, the blind Chinese “barefoot lawyer” who was smuggled into the U.S. embassy in Beijing in a diplomatic car chase.2 But far more common are asylum-seekers like “Angela,” a client of the Immigration Legal Services Clinic at Yale.3 Pressured to marry when she was only sixteen, Angela endured more than ten years of marriage to a man who repeatedly raped and beat her. On at least two occasions, local police refused to intervene or protect her. When her husband began to threaten to kill her, Angela realized that she had to escape to save her life. She fled her home in Central America and traveled overland to the U.S.-Mexico border, where U.S. Border Patrol agents arrested her. Eventually, a government official at the detention center realized that she had a “credible fear” of persecution and might be eligible for asylum.

Angela took many risks when she decided to leave her home: crossing multiple borders without a passport; spending nights vulnerable to assault by male guides; and riding a raft across the Rio Grande. But she likely did not realize one of the biggest risks she took: applying for asylum as a domestic violence victim. The odds of this particular form of “refugee roulette” vary wildly from jurisdiction to jurisdiction, immigration judge to immigration judge, and asylum officer to asylum officer.4 As one practitioner describes it, “whether a woman fleeing domestic violence will receive protection in the United States seems to depend not on the consistent application of objective principles, but rather on the view of her individual judge, often untethered to any legal principles at all.”5 Perhaps because Angela’s story does not resemble that of Victor Laszlo or Chen Guangcheng—she fled a violent spouse, not a repressive government, and snuck across the border rather than being welcomed with open arms—asylum adjudicators have struggled to fit her experience into the “typical” asylum narrative.

While supportive of domestic violence asylum in principle, this Note critiques the legal framework currently relied on by adjudicators who grant asylum to victims of domestic violence. That framework has no legal support in Board of Immigration Appeals (BIA) or federal court opinions, and, in fact, is inconsistent with some of the binding decisions of those courts. Instead, I propose a regulatory reform to make clear that domestic violence occurs “on account of” gender and that severe domestic violence can be the basis for an asylum grant. Unlike another recent proposal regarding domestic violence asylum,6 I argue that creative arguments based on existing law will not overcome the innate hostility of some adjudicators to this type of claim. Instead, the law itself must be changed to clear away existing adverse precedent and put domestic violence asylum claims on solid legal ground.

In Part I of the Note, I offer a normative account of why domestic violence victims should be granted asylum as victims of human rights violations that occur “on account of” their gender. In Part II, I describe the current status of the law on domestic violence asylum and gender-based asylum claims generally. In Part III, I argue that the legal framework proposed by the Department of Homeland Security (DHS) in 2009 has failed to provide for consistent adjudications of gender-based claims and may be contrary to existing BIA and circuit court precedent on the requirements for “particular social group” asylum claims. Finally, in Part IV, I offer a proposal for regulatory reform to put asylum claims based on domestic violence on a firmer legal footing. I argue that such a regulation would provide for more consistent adjudication of such asylum claims without resulting in a “flood” of new asylees to the United States.

As an initial matter, I discuss gender-based violence in which a woman is persecuted on account of her gender, not necessarily in gender-specific ways. Gender-specific violence includes “rape, sexual violence, forced abortion, compulsory sterilization, and forced pregnancy.”7 These harms, while devastating, are not necessarily perpetrated because of the victim’s gender.8 For instance, rape may be used as “punishment” for a woman’s political beliefs that have nothing to do with her gender. Where gender-specific harm occurs on account of a protected ground, adjudicators since the 1990s have generally accepted it as the basis for a successful asylum claim.9 In contrast, gender-based violence, in particular domestic violence, has faced much greater resistance from both administrative agencies and courts.10

I. domestic violence is “asylum-worthy”

The key obstacle to asylum for domestic violence victims has been the perception that domestic violence is not the “type” of persecution whose victims the asylum system is intended to protect. But in this Part, I argue that the history of modern asylum law shows that its protections were meant to apply broadly, including to harms only recently recognized as human rights violations. I then note that both feminist theory and empirical studies suggest that domestic violence is best understood as a form of human rights violation that is perpetrated against women because they are women. Finally, I show that the hostility to domestic violence asylum claims stems from two fundamental misunderstandings of the nature of domestic violence: that domestic violence is “private” rather than “public” harm, and that it occurs for reasons other than the victim’s gender. Once we reject these misunderstandings, it becomes clear that our Refugee Convention obligations not only allow but in fact require that we extend asylum protections to qualifying victims of domestic violence.

A. The Refugee Convention and “Particular Social Group”

The underlying goals of the asylum system are twofold: (1) the protection of an individual’s right to core aspects of identity and human dignity; and (2) the rejection of states’ efforts to use violence to enforce status hierarchies drawn along those lines. The 1951 Refugee Convention and its 1967 Protocol name four specific aspects of identity that are protected: race, religion, nationality and political opinion.11 But the drafters of the Convention also recognized that no list could possibly encompass all of the reasons for which a deserving asylee might be persecuted. A fifth protected ground, “membership in a particular social group,” encompasses the many distinct groups that could be targeted, typically by a government, for persecution.12 The drafters may have been thinking specifically of the victims of persecution in newly socialist states, such as “landowners, capitalist class members, independent business people, the middle class and their families.”13 But the language of the Convention itself is broad, suggesting that the drafters did not intend to limit its protection to groups that it knew were subject to persecution in 1951.

Interpreting courts agree that the “particular social group” ground is not limited to those groups that were being persecuted in 1951. The seminal case in this area is the BIA decision in Matter of Acosta,14 which is widely cited by both U.S. and foreign courts. The BIA observed that the other four grounds for asylum each “describes persecution aimed at an immutable characteristic: a characteristic that either is beyond the power of an individual to change or is so fundamental to individual identity or conscience that it ought not be required to be changed.”15 Applying ejusdem generis, it concluded that a cognizable particular social group must also be defined by a shared “immutable” characteristic.16 This characteristic need not be literally immutable; if the characteristic is “fundamental to [the group members’] individual identities or consciences,” that will also be a sufficient basis for a particular social group.17This standard “preserve[s] the concept that refuge is restricted to individuals who are either unable by their own actions, or as a matter of conscience should not be required, to avoid persecution.”18

Other countries and the Office of the United Nations High Commissioner for Refugees (UNHCR) have also accepted the Acosta immutability standard for “particular social group.” In Canada v. Ward, the Supreme Court of Canada observed that “[t]he dominant view . . . is that refugee law ought to concern itself with actions which deny human dignity in any key way.”19 Accordingly, no matter which of the five protected grounds forms the legal basis for an asylum claim, the appropriate standard for granting refugee status is “the sustained or systemic denial of core human rights.”20 A particular social group will not be recognized where the group is “defined by a characteristic which is changeable or from which disassociation is possible, so long as neither option requires renunciation of basic human rights.”21 The UNHCR likewise endorses an interpretation of particular social group that is grounded in the norms of international human rights. In the absence of a “closed list” of groups that may constitute a particular social group, the term “should be read in an evolutionary manner, open to the diverse and changing nature of groups in various societies and evolving international human rights norms.”22

B. Domestic Violence as Human Rights Violation

As described in Acosta and Ward, the principles behind the asylum statute clearly support granting asylum to victims of domestic violence whose abuse reaches the level of persecution. Feminist theory offers a persuasive account of domestic violence as a tool of female subordination that differs from “traditional” asylum-worthy persecution in manner but not in kind. Cross-cultural studies of domestic violence lend support to this theory by demonstrating that rates of domestic violence vary across different cultures and are correlated with other markers of gender equality.

Feminist theorists, activists, and international human rights organizations agree that domestic violence, and the state’s failure to protect women from it, reflects and reinforces women’s lower social status relative to men. In her seminal early study of battered women, Lenore Walker observed the importance of socialization, both in causing men to believe they had a right to abuse their wives that grew out of their historical “rights to rule their women,” and in causing women to tolerate abuse.23 Writing fifteen years later, Rhonda Copelon described domestic violence against women as “systematic and structural, a mechanism of patriarchal control of women that is built upon male superiority and female inferiority, sex-stereotyped roles and expectations, and the economic, social and political predominance of men and dependency of women.”24 Celina Romany sees domestic violence as part of the “continuum of subordination which deeply affects women’s ability to develop as citizens.”25 In these accounts, domestic violence is not a private relationship gone awry but rather a self-reinforcing expression of widespread social norms. Society communicates to men that they have the right and the power to abuse female partners. The experience of violence then further subordinates women vis-à-vis their male partners and male society more generally.

The history of legal permission for men to abuse their wives also supports the conclusion that it has a different, and greater, social meaning than other forms of physical abuse, including abuse by women against male partners. Until the nineteenth century, Anglo-American common law protected the husband’s right to use corporal punishment against his wife so long as he did not inflict permanent injury.26 The husband’s right to beat his wife was part of a larger structure of marital rights and privileges that established the husband’s superiority over his wife. He possessed rights to her person, her labor, and most property that she possessed before the marriage.27 She was obligated to obey and serve him, and her legal identity was subsumed into his.28 Early women’s rights advocates criticized wife-beating as “a practical and symbolic embodiment of the husband’s authority over his wife.”29 As Reva Siegel documents, the right of marital chastisement has been formally repudiated by the American legal system, but many of the same behaviors continued to be tolerated under the guise of marital privacy.30 Similar patterns have been recognized outside of the Anglo-American legal world. A United Nations session on gender equality observed that women’s subordinate place in society makes them vulnerable to particular forms of violence, including intimate partner violence, and that “[d]omestic violence . . . is still treated as a private matter in some countries.”31 There is, by contrast, no comparable history of legal tolerance for violence committed by women against their husbands or male partners.

A 2005 World Health Organization (WHO) cross-country study of domestic violence supports the theorists’ connection between domestic violence and gender hierarchies. Researchers found that the prevalence of domestic violence varies widely across different cultures and social settings. WHO researchers surveyed more than 24,000 women from fifteen sites in ten countries and found that rates of domestic violence varied from fifteen percent to seventy-one percent in different sites.32 In the countries where the researchers sampled both rural and urban sites, overall levels of domestic violence were consistently higher in the rural settings.33 Researchers found a strong association between the prevalence of domestic violence and women’s belief that such violence is “normal.”34 This association was particularly marked in rural and more traditional societies, which “reinforces the hypothesis that the status of women within society is a key factor in the prevalence of violence against them.”35 Because of the connection between domestic violence and the status of women in the society, researchers’ first recommendation for reducing domestic violence is that policymakers take steps to promote gender equality: “Improving women’s legal and socioeconomic status is likely to be, in the long term, a key intervention in reducing women’s vulnerability to violence.”36

Individual country studies of domestic violence also confirm the connection between domestic violence and social status relationships. To take Angela’s home country as an example, Nicaragua’s rates of domestic violence are at the upper range of countries studied in the WHO report. The lifetime prevalence of spousal violence in Nicaragua is fifty-two percent among ever-married women;37 of the fifteen sites studied by the WHO, only five had higher reports of physical and/or sexual violence by ever-partnered women.38 The same researchers found that the high rate of domestic violence in the country was related to traditional Nicaraguan conceptions of gender identity, namely the paired concepts of machismo—which “emphasi[zes] male moral, economic and social superiority over women”—and marianismo—which emphasizes women’s role as mothers and submissive partners.39 The norm of marianismo means that “[r]ather than taking active steps to change her situation, a devout woman is expected to hold the family together at all costs, to endure abuse patiently and to pray to the Virgin for her husband’s conversion.”40 Another study found that “sexual violence by intimate partners is sometimes triggered . . . when men feel at risk of losing control of the relationship.”41 Domestic violence is thus used to suppress women’s autonomy in order to promote and preserve the relative power of men in society.

It is true that domestic violence also occurs in same-sex couples42 and is sometimes perpetrated by women against men.43 Certainly, this suggests that “tensions, stress, jealousies,” and other power dynamics may lead to violence even in relationships that are not shaped by historic status hierarchies.44 These same factors may even be the impetus for individual acts of violence by men against their female partners. But excessive focus on non-gender-related causes of domestic violence can obscure the gender dynamics that permit and inspire men to attack their partners.45 In the WHO study, an extraordinarily small percentage of women reported that they had ever initiated violence against a spouse or partner compared to the number of women who reported being victims of domestic violence themselves.46 Although the WHO methodology raises obvious self-reporting concerns, another survey of Cambodian men and women found broad agreement between the number of men who reported being abused by their female partners and the number of women who reported acts of violence against male partners.47 Furthermore, intimate partner violence committed by women is much more likely to occur when women are acting to defend themselves; “woman-initiated aggression is relatively rare.”48 There is also no history of legal tolerance of female-on-male domestic violence or chastisement, in contrast to the long history of legal permission that was granted to men to chastise their wives. Although female-on-male intimate partner violence does occur, it is less common and lacks the social and historical meaning that makes male-on-female intimate partner violence distinctive and deserving of special consideration in asylum law.

Because domestic violence subordinates women on account of their gender, it should be the basis for a viable asylum claim. Gender is an immutable characteristic that is sufficient to define a particular social group under the Acosta standard. If medical advances have rendered it not quite immutable, gender is precisely the type of characteristic that “ought not be required to be changed.”49 And although sex and gender were not included in the 1951 Refugee Convention, sex is included as a “fundamental” aspect of identity—alongside race, religion, and national origin—in state and federal antidiscrimination statutes.50 Those statutes share with the Refugee Convention the purpose of prohibiting the exclusion or marginalization of people based solely on their group identities. The recognition of gender as a protected characteristic is an acknowledgment that women have a history of “shared marginalization,”51 including the inability to vindicate their human right to be free from violence. As long as victims of domestic violence can satisfy the other requirements for asylum—fear of harm reaching the level of persecution, the inability to obtain state protection or to relocate away from the harm, and the absence of serious criminal history—they should be entitled to asylum’s protection.

C. Misconceiving Domestic Violence as “Private” Punishment

Notwithstanding the well-established link between gender and domestic violence, domestic violence-based asylum claims are routinely viewed with suspicion by adjudicators. This stems from two fundamental misunderstandings of the nature of domestic violence: first, that intimate-partner violence is “private” violence not within the ambit of asylum’s protections, and second, that intimate-partner violence occurs for reasons other than gender—as punishment, an expression of jealousy, or something else.

It is true that domestic violence is different from the forms of violence typically associated with asylum: concentration camps, ethnic cleansing, and religious oppression. If we think of Victor Laszlo as the asylum system’s ideal, a Central American domestic violence victim arrested at the border after paying a coyote for her passage is certainly far removed from that prototype. But it would be wrong to infer that because women are persecuted differently from other groups, they are not persecuted because of their gender or at all. The persecution of women simply takes different forms:

Unlike the ways in which men systematically enslave, violate, dehumanize, and exterminate other men, expressing political inequalities among men, men’s forms of dominance over women have been accomplished socially as well as economically, prior to the operation of law, without express state acts, often in intimate contexts, as everyday life.52

Violent homes become, in essence, one-woman prisons or concentration camps scattered across the country.53 Women should not lose asylum’s protections merely because they are persecuted in isolation rather than in a group; indeed, by systematically isolating women and making abuse a source of private shame rather than public indignation, domestic violence intentionally deprives abused women of social support and makes it harder for them to escape.

Because domestic violence is rooted in widespread social norms, the distinction between public and private harm cannot justify the rejection of domestic violence-based asylum claims. The concept of privacy has been used in both the domestic criminal context and the asylum context as a reason to avoid state action to prevent domestic violence or protect its victims.54 But feminist theorists challenge the very value of privacy, suggesting that it developed in part to entrench and protect male power over women.55 By excluding the state from the home, doctrines of privacy immunize men from legal responsibility for acts of violence against women. In the criminal context, feminists have successfully dismantled the public/private distinction in order to enhance the police’s ability to protect domestic violence victims.56 Similarly, asylum law should not privilege men’s abuse of women merely because it occurs in the home rather than in a prison, a concentration camp, or another more “public” place.

Opponents of domestic violence-based asylum claims have also focused on what they claim to be the “real” reasons for domestic violence: punishment, jealousy, or the idiosyncratic pathology of the individual abuser.57 This obscures, without disproving, the connection between domestic violence and gender hierarchies.58 The belief that domestic violence occurs because of jealousy or as a form of “punishment,” as opposed to on account of gender, actually reinforces rather than undercuts the relationship between partner violence and gender hierarchies. Women in many societies conceptualize domestic violence as “a form of chastisement for female behaviour that transgresses certain expectations,” particularly infidelity or “circumstances where women ‘disobey’ a husband or partner.”59 A man who believes he has a right to punish his wife believes that he can control her. If he believes she has violated rules that apply to her because she is a woman, his violent response is thus “on account of” her gender. This connection is sufficient to establish the nexus between immutable characteristic and persecution that is required to support an asylum claim.

II. today’s unstable foundation for domestic violence-based asylum

Even if courts consistently recognized the close ties between gender and domestic violence, existing doctrine would nonetheless hamper their efforts to grant asylum claims on those grounds. In this Part, I first describe courts’ unwillingness to recognize that persecution “on account of” gender gives rise to a cognizable asylum claim. I then analyze the efforts of asylum advocates to win asylum for domestic violence victims on the basis of social group definitions that combine gender with a variety of other characteristics. That type of social group definition was eventually endorsed by DHS, whose briefs are now the best available legal support for domestic violence asylum claims.

A. Rejecting “Women” as a Particular Social Group

Notwithstanding the strong normative reasons to construe gender as a core aspect of identity that should be protected by the asylum statute, courts have failed to follow their own promises of protection for women who are persecuted on the grounds of their gender alone. The original Acosta decision held that particular social groups will typically be defined by innate characteristics “such as sex.”60 Numerous U.S. courts have stated in dicta that gender is a permissible basis for an asylum claim.61 Yet in practice, courts have been reluctant to grant asylum on the basis of a particular social group defined solely or primarily by gender. Courts routinely mistake the size of a gender-based social group—which admittedly could be quite large—for the idea that all members of that group would automatically be eligible for asylum. In one example, the Eighth Circuit found that “Iranian women” was overbroad because “no factfinder could reasonably conclude that all Iranian women had a well-founded fear of persecution based solely on their gender.”62 But asylum law does not actually require that all members of a particular social group face the same threat of persecution.63 Nor does a finding that a group is cognizable mean that all group members are therefore eligible for asylum. To warrant asylum, a group member must also demonstrate that she has a “well-founded fear” of persecution, that the persecution she fears would occur “on account of” her membership of the particular social group, and that she is not subject to any of the bars to asylum.64

Because courts have treated particular social groups that hinge only on an applicant’s gender with skepticism, gender-based asylum claims often define the particular social group more narrowly to make the group appear smaller.65 In Gao v. Gonzales, for instance, the Second Circuit found that a particular social group could be defined as “women who have been sold into marriage (whether or not that marriage has yet taken place) and who live in a part of China where forced marriages are considered valid and enforceable.”66 This was the route taken by early domestic violence asylum claimants, as follows.

B. A Brief History of Domestic Violence Asylum

Domestic violence asylum-seekers began to use complex social group definitions, sometimes successfully, in the 1990s.67 However, those developments were cut short by the BIA’s precedential decision in Matter of R-A-.68 That decision overturned an immigration judge’s asylum grant to Rodi Alvarado, a Guatemalan woman who was harshly abused by her partner. Ms. Alvarado had initially been granted asylum on the basis of her membership in the social group of “Guatemalan women who have been involved intimately with Guatemalan male companions, who believe that women are to live under male domination.”69 The BIA held that while the abuse was “deplorable,” Ms. Alvarado’s proposed social group was not “recognized and understood to be a societal faction.”70 It was merely “a legally crafted description of some attributes of her tragic personal circumstances.”71 The Board further held that even if Ms. Alvarado’s social group were cognizable, she had failed to prove nexus to a protected ground—that her husband had persecuted her because of her membership in that group.72 Instead, her husband

harmed her, when he was drunk and when he was sober, for not getting an abortion, for his belief that she was seeing other men, for not having her family get money for him, for not being able to find something in the house, for leaving a cantina before him, for leaving him, for reasons related to his mistreatment in the army, and “for no reason at all.”73

The BIA held that none of these reasons was “‘on account of’ a protected ground.”74

The R-A- decision was immediately criticized, as many perceived that the decision threatened not only domestic violence asylum claims, but also a whole range of gender asylum cases.75 In response to this criticism, about one year after the BIA issued its decision in R-A-, Attorney General Janet Reno proposed regulations to address cases like Ms. Alvarado’s.76 Pending the promulgation of final regulations—which were abandoned after President George W. Bush took office in 2001—Ms. Alvarado’s case was stayed.77 When the case was reopened in 2003, DHS filed a brief rejecting the prior BIA position and arguing that Ms. Alvarado had stated an acceptable asylum claim.78 In its brief, DHS argued that “married women in Guatemala who are unable to leave the relationship” was a proper social group formulation for a domestic violence claim because “a primary animus for violence arises from the abuser’s perception of the subordinate status his wife occupies within the domestic relationship.”79 In addition, her inability to leave the relationship and social expectations regarding domestic violence in Guatemala “bolster[] his belief that he has the right to abuse” his wife.80 After another stay, Ms. Alvarado’s case was finally remanded to an immigration judge for resolution in 2008.81

While Ms. Alvarado’s case was pending before the immigration judge, DHS filed another brief in a different domestic violence asylum case. L.R. was a Mexican woman who came to the United States seeking asylum on account of abuse by her male partner.82 In its 2009brief, DHS argued that a cognizable social group for domestic violence is defined by the nexus of (1) the asylum-seeker’s gender, (2) her relationship status, and (3) her society’s perception of that status. The particular social group could be described as “Mexican women in domestic relationships who are unable to leave” or “Mexican women who are viewed as property by virtue of their positions within a domestic relationship.”83

Both Rodi Alvarado and L.R. were eventually granted asylum by immigration judges, but neither of these cases produced a precedential decision on domestic violence asylum claims. There is anecdotal evidence, however, that the DHS briefs filed in these two cases have had an impact on asylum decision-making: a study of asylum decisions tracked by the Center for Gender & Refugee Studies revealed that immigration judges have increasingly “recognized groups that include some combination of the L-R- characteristics of gender, nationality, and status in the relationship.”84 Notably, 2010 saw what is believed to be the first grant of a domestic violence-based asylum claim in the Eloy, Arizona Immigration Court, a jurisdiction that had previously been particularly resistant to such claims.85

However, there is still no binding rule on the acceptability of domestic violence asylum claims. As a result, DHS attorneys still sometimes take positions that are inconsistent with the R-A- and L-R- briefs.86 Some DHS attorneys have even objected to applicants’ inclusion of the R-A- and L-R- briefs in their filings with the immigration court, ostensibly because of confidentiality concerns.87 Furthermore, immigration judges are not required to accept social groups that are consistent with those briefs. This is particularly damaging because, as I will discuss in the next Part, DHS’s legal framework is arguably inconsistent with recent BIA precedent that is binding on those judges.

III. the 2009 dhs framework

DHS strove mightily to fit its analysis of domestic violence asylum claims to the doctrinal requirements of asylum law. But, as this Part will show, domestic violence social groups are nonetheless inconsistent with several of the requirements for a doctrinally cognizable “particular social group.” First, domestic violence social groups are (1) insufficiently visible or distinct, (2) insufficiently particular, and (3) too circular under recent BIA and circuit court precedent. Second, the doctrinal flaws in the DHS social groups have contributed to the inconsistent adjudication of domestic violence asylum claims. Finally, the lack of a permanent legal footing for domestic violence asylum will jeopardize future asylum-seekers if the political climate for domestic violence asylum worsens.

A. Doctrinal Requirements

Three doctrinal requirements for “particular social groups” pose special challenges to domestic violence asylum claims: (1) social visibility/social distinction, (2) particularity, and (3) non-circularity. The first two requirements were developed in Board of Immigration Appeals decisions based on interpretations of the underlying asylum statute. The BIA, which hears appeals from immigration judges, is authorized to issue precedential decisions that are binding on all immigration judges and asylum officers across the country.88 However, the national uniformity of the system is disrupted by the system for appeals from BIA decisions, which are taken to the appropriate regional court of appeals.89 The BIA developed the social visibility and particularity requirements in 2006 and 2008 respectively.90 Different courts of appeals have reached contrary conclusions on the validity of these BIA interpretations of the asylum statute.91

The third requirement, non-circularity, is much less contentious and arises out of the inherent logic of the Refugee Convention. It has been endorsed by circuit courts,92 the Department of Homeland Security,93 and the Board of Immigration Appeals,94 as well as courts in the United Kingdom.95 The reasoning is simple: If sharing a “well-founded fear” could be the basis for a particular social group, then there would be no need for an applicant to show a connection to any protected ground. Therefore, the fear or fact of persecution cannot itself define a social group. I will consider the applicability of these three requirements to domestic violence asylum claims in turn.

1. Social Visibility/Social Distinction

Since 2006, the BIA has required that social groups be “visible” or “recognizable” to others in the applicant’s home country.96 In response to criticism from the circuit courts,97 the BIA recently clarified that “visibility” does not mean “‘ocular’ visibility” but rather that there must be “evidence showing that society in general perceives, considers, or recognizes persons sharing the particular characteristic to be a group.”98 They also renamed the requirement “social distinction.”99 This interpretation of “particular social group” has not yet been reviewed by any court of appeals.

Understood as “social distinction” or “social visibility,” this requirement is not clearly satisfied by the social group definitions offered by DHS in Matter of R-A- and Matter of L-R-. In countries with a strong norm that spousal violence is “normal,” society may not distinguish victims of abuse from the general population of married women. In many countries with high rates of domestic violence, advocates are only beginning to raise awareness of domestic violence as a social problem rather than a normal feature of intimate relationships. In Bangladesh, for instance, where women reported some of the highest rates of domestic violence in the WHO Multi-Country Study,100 more than 63.8% of urban women and 86.1% of rural women who had experienced violence believed that there were at least some circumstances in which a man was justified in beating his wife.101 Furthermore, victims of domestic violence may themselves conceal the abuse because of fear of greater violence or fear of social stigma.102 Again taking Bangladesh as an example, more than sixty-six percent of ever-battered women surveyed by the WHO reported not having told anyone about their partner’s violence.103 Finally, in countries that have legal tools for women to leave their relationships, society may not recognize that some spouses are not actually able to avail themselves of those means. Disturbingly, the social visibility analysis implies that women from countries that do not protect domestic violence victims at all will have a harder time proving asylum eligibility than women from countries that make some effort to protect victims.104 This counterintuitive result would be a clear failure to satisfy our obligations to such women.

Some immigration judges have in fact denied asylum to domestic violence victims because they have found the particular social groups to be insufficiently visible, even though the proposed social groups were virtually identical to one of those endorsed by DHS in Matter of L-R-.105 One judge rejected the social group of “Kenyan women in a domestic relationship who are unable to leave” because the problem of domestic violence is generally ignored in Kenya, and thus the group is not socially visible.106 In another case, an immigration judge found that an L-R-type social group was not visible in India because “only thirty percent of domestic violence victims there seek assistance.”107 These denials underscore the need to clarify the legal status of domestic violence asylum claims.

2. Particularity

Since 2008, the BIA has applied the “particularity” requirement to ensure that “the proposed group can accurately be described in a manner sufficiently distinct that the group would be recognized, in the society in question, as a discrete class of persons.”108 Essentially, the BIA and circuit courts that accept the requirement ask how easy it is to know who is inside the group and who is outside the group.109 What is needed is “a clear benchmark for determining who falls within the group.”110 Under this standard, groups such as “wealthy Guatemalans” have been deemed to be “too subjective, inchoate, and variable” to constitute a particular social group.111

As with social visibility, the particularity requirement poses a challenge to domestic violence social groups. In comparison to wealth, the characteristics of being “unable to leave” and “viewed as property by virtue of their positions within a domestic relationship”112 seem to be equally if not more indeterminate. DHS asserts that “assessments of a victim’s ability to leave a domestic relationship would involve case-by-case, fact-specific examinations of whether it would be reasonable to expect the victim to do so under all the circumstances.”113 But this misses the point of the BIA’s particularity requirement, which requires that each criterion not be subject to variable interpretations. Absent further guidance from the BIA, different adjudicators will have different expectations of when it would be “reasonable to expect the victim to [leave the relationship] under all the circumstances,” and therefore the group should be found to be insufficiently particular.

3. Circularity

Above and beyond the core requirements for a particular social group, DHS’s domestic violence social groups run afoul of a core doctrinal requirement for particular social groups: that they may not be defined by the fact of the persecution that the applicant suffered.114 DHS argues that a woman may be “unable to leave” because of “economic, social, physical or other constraints . . . during the period when the persecution was inflicted” or because “the abuser would not recognize a divorce or separation as ending the abuser’s right to abuse the victim.”115 Insofar as it is the threat of physical abuse that causes women to be unable to leave their relationships, the group is defined by its persecution. DHS’s position is essentially that the applicant is persecuted on account of a relationship that she cannot leave without being persecuted; this clearly falls afoul of the circularity principle. At least one immigration judge has rejected a social group similar to that advocated by the DHS in L-R- on these grounds.116

B. Inconsistent Adjudications

In part because of the doctrinal flaws in the DHS social groups, reliance on the relevant DHS briefs as the sole legal authority for domestic violence asylum claims contributes to widespread inconsistency in the outcomes of those claims. Immigration judges vary widely in their willingness to accept social groups constructed to exactly match those proposed by DHS.117 A recent analysis of years of domestic violence-based claims suggests that the outcomes of such cases depend primarily on the identity of the presiding immigration judge.118 The five immigration judges at the Eloy Detention Center in Arizona exemplify this trend: in Eloy, it is “extremely rare” for any of the judges to grant a domestic violence claim, even when the social group proposed by the applicant is identical to that proposed by DHS in Matter of L-R-.119 In addition, DHS does not file an L-R-type brief in every domestic-violence based claim, so the burden lies on the applicant to bring the argument to the attention of the relevant adjudicator. In doing so, an applicant cannot rely on a binding regulation, a precedential decision, or even interpretative guidance from the agency; the best that she can do is to cite the DHS briefs as persuasive authority.

The structure of the immigration court system facilitates the profusion of inconsistent decisions by immigration judges. Immigration judges are technically required to base their decisions on existing precedent issued by the Board of Immigration Appeals and the Attorney General.120 But only a small number of BIA decisions are designated as precedential decisions; the vast majority go unpublished.121 This leaves immigration judges with a relatively thin body of law on which to base their decisions. In the absence of on-point BIA authority, judges may “exercise their independent judgment and discretion” and “take any action . . . that is appropriate and necessary for the disposition of such cases.”122 This expansive grant of authority to immigration judges, who need not issue written opinions,123 means that most enforcement of asylum law—including the law of domestic violence asylum—is shielded from public view. In fact, without the advocates’ own recordkeeping, there would be no way to know how such claims have fared. Without any published decisions on the subject, the most conscientious immigration judge has only the broad principles of asylum law to guide his or her decision, and it is therefore unsurprising that judges are reaching such contrary results.

Apart from being intuitively unfair, inconsistency in asylum adjudication is particularly risky for domestic violence victims. Attempts to leave an abusive relationship are closely correlated with an increase in the severity of the abuse.124 The possibility of safe harbor in the United States may give women the courage to leave severely abusive relationships. But if asylum is granted inconsistently, then a woman who through no fault of her own ends up before an adjudicator hostile to this type of claim will be deported to a more dangerous situation than the one that she left behind.125 Inconsistency also breeds inefficiency.126 Each domestic violence asylum applicant must re-litigate arguments about social group and nexus because there is no controlling precedent or rule to rely on for the proposition that severe domestic violence may give rise to a cognizable asylum claim.

Any satisfactory resolution to the problem of domestic violence asylum claims must provide for more consistent outcomes for similar claims. It is true that absolute consistency may be an impossible and even an undesirable goal for the asylum system. But more consistency is not only possible, but morally required to satisfy our human rights obligations and offer safety to women seeking refuge in this country.

C. No Permanent Protection

Perhaps the most significant objection to the DHS framework for domestic violence social groups is that it is a decidedly temporary solution to the problem of adjudicating gender-based asylum claims. Neither the BIA nor any court has adopted DHS’s reasoning in a precedential decision. United States Citizenship and Immigration Services (USCIS), a branch of DHS, currently instructs asylum officers that “in a domestic violence claim, an adjudicator would consider evidence that the abuser uses violence to enforce power and control over the applicant because of the social status that a woman may acquire when she is in a domestic relationship”—a formulation closely mirroring the DHS argument in L-R-.127 But these guidelines do not bind USCIS, are not enforceable by courts, and are open to revision at any time by the agency.

Given the current political salience of immigration policy and the recurrent fear of “opening the floodgates” to a vast number of asylum claims by women, it is not far-fetched to worry that a future administration would roll back DHS’s endorsement of gender-based asylum claims. Conservative immigration policy analysts have already taken aim at the prospect of what they term “spousal abuse asylum.”128 Their criticisms include the fear of a flood of asylees and, relatedly, a risk of fraud given what they believe to be the low evidentiary burdens of asylum law.129 The Center for Immigration Studies, a conservative immigration think tank, has described domestic violence asylum claims as part of “a larger quest to remake American legal norms, establish victim status for a number of officially recognized groups, and overhaul American society more generally.”130

In this political climate, it is imperative that asylum eligibility for victims of domestic violence and other gender-based crimes be placed on a more permanent legal footing. As domestic violence victims abroad begin to hear that women are able to gain protection in the United States, it is possible that some women who might otherwise have stayed in abusive relationships will find the courage to leave and come to the United States. But if domestic violence asylum were suddenly ended—or if a woman had the bad luck to end up before one of the immigration judges who reject domestic violence asylum claims in the status quo—she would be sent back to a situation that is likely to be more dangerous than the one she left.131 It is intolerable that the United States’ uncertain position on domestic violence asylum could make women’s situationsmore dangerous than they already are; immediate action is needed to clarify the status of domestic violence asylum claims.

IV. proposal for regulatory reform

The best way to solidify the legal foundation for domestic violence asylum, reduce inconsistency in asylum adjudications, and make domestic violence asylum permanent is through the regulatory process. In this Part, I first argue that rulemaking is the appropriate mechanism for legal reform because it can override BIA precedent and will bind both asylum officers and immigration judges. Second, I propose a regulation that defines gender as a sufficiently particular group to support an asylum grant and establishes that intimate-partner violence against women occurs “on account of gender” as a matter of law. Third, I discuss the legal obstacles to promulgating such a regulation in such a way that it would bind both the Department of Justice (DOJ) and DHS. Finally, I defend the regulation’s ability to resolve current legal ambiguities around domestic violence asylum without opening a “floodgate” to many more asylees.

A. Why Use Rulemaking?

As discussed in Section III.B, asylum adjudication in general is rife with inconsistencies among individual adjudicators, among different regional offices and different immigration courts, and among different levels of adjudication. Steps that would significantly increase consistency across the board, such as quotas or disciplinary actions for noncompliant adjudicators, are rightly rejected as incompatible with the highly individual, fact-dependent nature of the asylum inquiry.132

But to the extent that inconsistency arises because of misunderstanding or disagreement about the meanings of terms used in the asylum statute—for instance, “particular social group”—agencies may use rulemaking to clarify the definition.133 The question of how to handle domestic violence-based asylum claims is one of the “few specific key issues” where policy guidance in the form of regulations could meaningfully increase consistency.134 Rulemaking may not be able to resolve all inconsistency in asylum adjudications: for example, there is evidence that different personal backgrounds of adjudicators also contribute to disparities.135 But ongoing confusion over domestic violence asylum claims suggests that at least in the area of gender-based asylum claims, there continue to be real disagreements about substantive law that could be resolved via rulemaking.

These disagreements—over whether women constitute a particular social group and whether domestic violence can be said to occur “on account of” that group—are different from other uncertainties in asylum law, such as the absence of a comprehensive definition of “persecution.” Rather than attempting to list all of the types of harm that might constitute persecution, the law treats that as a question of fact to be determined by an individual adjudicator.136 It would be impossible to write a law that anticipated every form of persecution that an abuser might use, so the assessment of whether persecution occurred must turn heavily on the unique set of facts presented by the applicant. Even individual acts that themselves are not persecution may reach the level of persecution when they are repeated or combined with other harmful acts.137 Rather than relying on a narrow statutory definition of persecution, common-law guidance as to its meaning has developed through decades of circuit court and BIA decisions. In contrast, whether “women” can be a particular social group is nearly a purely legal question, about which no judicial consensus has been reached. The concerns raised by judges who reject gender-only social groups are the breadth and size of the groups and their internal diversity. Since the specific facts of an individual applicant’s story have little bearing on this question, it is ripe for resolution by a binding regulation.138

Recent scholarly proposals to restyle domestic violence asylum claims within the existing legal framework do not go far enough to solve the problems of inconsistency and impermanence that were described in Part III. Marisa Silenzi Cianciarulo proposes that domestic violence claims should be treated as political opinion claims rather than particular social group claims.139 Barbara Barreno and Elsa M. Bullard both argue that the analysis should be shifted to focus on the government’s failure to act rather than the motives of the persecutor himself.140 However, requiring adjudicators to assess the motives behind the government’s failure to act will not resolve the inconsistencies that we now see in the outcomes of domestic violence asylum claims, which stem from a more fundamental disbelief by some adjudicators that asylum covers domestic violence claims at all. Natalie Rodriguez does advocate for regulations to refine the meaning of “particular social group.”141 While I agree with her that regulation is the right approach to solving the current problem, her proposal does not go far enough in making the law more favorable to this type of asylum claim. Among other differences, she would continue to allow adjudicators to determine that persecution occurred on the basis of gender as a matter of fact. In contrast, I argue that the historical and sociological evidence tying domestic violence to gender warrants drawing that connection as a matter of law.

There is historical precedent for resolving elements of the asylum inquiry as a matter of law. After China imposed the one-child policy in the late 1970s, Chinese immigrants to the United States and other Western countries began to seek asylum on the basis that they were persecuted because of their unwillingness to comply with the policy. A 1989 Board of Immigration Appeals decision held that enforcement of the one-child policy is not “on its face” persecution.142 In order for sterilization or forced abortions under the policy to qualify as persecution, the BIA held, the government must be motivated by a discriminatory reason other than general population control.143 Seven years later, however, Congress rejected this interpretation of the asylum statute. It amended the law to provide that forced abortions and sterilizations are persecution on account of membership in a political group as a matter of law.144 This statutory change does not mean that all applicants seeking asylum on the basis of the one-child policy are now automatically granted asylum. Asylum-seekers still must prove that they actually have faced or will face forced sterilization or abortion. However, the legal change guaranteed that the legal questions of whether forced sterilization constituted persecution and whether such persecution was on account of political opinion would be resolved consistently whether they were posed in an asylum office, immigration court or on appeal to a federal court of appeals.

The legislative response to the “one-child policy” asylum claims is a model for a resolution of the outstanding questions about domestic violence asylum claims. But there is no need to wait for Congress to act; the same effect can be obtained through regulatory reform. Indeed, waiting for Congress is likely to be futile. The expansion of protection for parents affected by the one-child policy had much greater political salience than enhancing protection for domestic violence victims is ever likely to have. First, initial U.S. government support for one-child policy asylum claims came just after the Tiananmen Square massacres of 1989.145 Since domestic violence asylum-seekers could come from any number of countries that fail to protect victims, increasing protection for them is unlikely to align so neatly with U.S. foreign policy goals. Second, protection for victims of the one-child policy—particularly those who have been or might be subject to forced abortions—aligns with the domestic agenda of the anti-abortion movement. Many of the politicians advocating for asylum protection for victims of the one-child policy explicitly did so as part of an anti-abortion agenda.146

In contrast, domestic violence is not the type of political issue that is likely to find a natural constituency in Congress. Passing any law that would expand access to legal immigration is increasingly difficult in the divided political climate. At the time of this writing, the Senate has passed a comprehensive immigration reform bill, but the bill remains stalled in the House of Representatives.147 Furthermore, the specific needs of women immigrants have not been a focus in this round of immigration reform, despite evidence that the immigration system as a whole is systematically biased against women.148 The Senate never considered an amendment to its immigration bill that sought to address some of these inequalities.149 In light of the low likelihood of congressional action, a regulatory strategy is more likely to lead to the legal changes necessary to protect domestic violence asylum-seekers.

Finally, a regulatory interpretation of the asylum statute approving of domestic violence asylum claims is consistent with the purpose of the Refugee Convention and the meaning of the U.S. asylum statute. As discussed in Section I.A, the drafters of the Refugee Convention likely intended for the “particular social group” to evolve according to changing human rights norms. Indeed, the Department of Homeland Security has already taken the position in some cases that the asylum statute encompasses domestic violence-based claims.150 USCIS trains new asylum officers that domestic violence “is often related to the historically more powerful position of men in the family and in society, the perceived inferiority of women and unequal status granted by laws and societal norms,”151 and contemplate that it could be the basis for a successful asylum claim.152 The regulatory reform that I propose merely ensures that this insight is applied equally to all asylum applicants, whether they apply initially in an asylum office or before an immigration judge.

B. The Proposed Regulation

A regulatory solution must resolve two primary difficulties with the current DHS framework for domestic violence-based claims: (1) how to define a particular social group for domestic violence-based claimants, and (2) whether intimate partner violence can be understood to occur “on account of” gender. Resolving these legal ambiguities will allow adjudicators to focus on the facts of the case in front of them, while removing the burden on applicants to make complex legal arguments to defend their claims. The latter point is particularly important, as many applicants are likely to be applying for asylum or appearing in removal proceedings pro se. The following regulation could be inserted into DHS regulations at 8 C.F.R. § 208.13(b)(1), which defines eligibility for asylum, or elsewhere in that chapter as appropriate:

For the purposes of assessing asylum eligibility,

(a) a social group defined solely by the gender of its members iscognizable as a particular social group; and

(b) where a woman has experienced intimate-partner violence that otherwise meets the standard for persecution, the victim’s gender shall be deemed to be one central reason for the persecution.

Like the one-child policy amendment of 1996, this proposed regulation leaves intact the structure of the asylum inquiry for domestic violence asylum claims even as it resolves some elements of the inquiry as a matter of law. The one-child policy amendment fixed the interpretation of two elements of an asylum claim: it (1) clarified that forced abortions and sterilizations qualify as persecution, and (2) established that all forced abortions and sterilizations will be deemed for asylum purposes to have occurred “on account of” a political opinion.153 Similarly, the proposed domestic violence regulation (1) clarifies that gender alone can constitute a particular social group, and (2) establishes that intimate-partner domestic violence against women will be deemed to have occurred “on account of” gender.

The first part of the regulation affirms that a social group based on gender alone is cognizable under current asylum law. As discussed in Section II.A, judges have been reluctant to accept gender-only particular social groups, even though the Acosta decision explicitly contemplated a social group defined by gender. The proposed regulation resolves this legal question; if it were enacted, all immigration judges and asylum adjudicators would be required to accept gender-only particular social groups.This would shift the analysis of domestic violence asylum claims away from the largely legal question of what qualifies as a social group and towards a factual inquiry into the applicant’s own circumstances.

The fact that gender is not expressly mentioned in the statute does not bar this regulation. As discussed in Section I.A, the drafters of the Refugee Convention had imagined that “particular social group” would be an expansive catch-all category to encompass other immutable or fundamental characteristics, including characteristics that have only been recognized as fundamental since the drafting of the Convention. UNHCR has made clear that protection from gender-related harms falls within the scope of the refugee definition, even though it is not an enumerated ground for relief.154 There is no need to add an additional ground to the 1951 Convention definition—or the U.S. codification of that definition—in order to bring gender-related claims within the ambit of asylum law.155 The technical problem addressed by this regulation is the unwillingness of courts to allow gender to define a “particular social group,” which is an enumerated ground in both the Refugee Convention and the asylum statute. This forces claimants to offer convoluted social group definitions that are susceptible to the types of attack described in Part III.

The second part of the regulation fixes the connection between intimate-partner violence and gender as a matter of law. The rule borrows the language of the REAL ID Act of 2005, which defined “on account of” to mean that a protected ground must be “one central reason” for an asylum applicant’s abuse.156 There may be other reasons for any given incident of abuse, but the historical and sociological connections between domestic violence and female subordination demonstrate that gender relations are deeply implicated in any act of intimate-partner violence by men against women.157

This reform would not mean that all domestic violence asylum claimants would actually receive asylum. Applicants would still have to prove that they satisfy other requirements for asylum, including a well-founded fear of persecution, their home government’s unwillingness to protect them, and the absence of other bars to asylum.158 The standard for what constitutes persecution is high. In the First Circuit, for instance, “[t]o qualify as persecution, a person’s experience must rise above unpleasantness, harassment, and even basic suffering.”159 Rape,160 torture,161 and death threats162 are all persecution, but discrimination and harassment—even minor beatings—may not be persecution.163 Applicants will also have to prove that their home government is unwilling or unable to protect them from persecution. While persecution by private actors can qualify someone for asylum, applicants in those circumstances must demonstrate that their home government is unable or unwilling to control the persecutors.164 This can be shown by evidence that government actors instigated or condoned the persecution or that the government was unable to prevent it.165 The applicant must also show that she sought the protection of the government and was denied it, or that she had a reasonable explanation for not seeking protection (which may include evidence that the government has been unwilling to act in similar situations).166 Evidence of prevailing conditions in the applicant’s home country can be enough to satisfy this requirement.167 This requirement ensures that the victims of domestic violence who are most in need of asylum—namely, women whose own governments do nothing to protect them—will be eligible for it. This is in contrast to the current doctrinal regime, in which women have been denied asylum because their government’s failure to protect them is seen as evidence that their social group is not sufficiently “visible” to justify protection.168

C. The Rulemaking Mechanism

The process of implementing this regulation would be complicated by the division of authority over asylum between the Departments of Homeland Security and Justice. Responsibility for asylum administration was divided between the two agencies when DHS was created in 2002.169 The effect of this reorganization is that DHS now has jurisdiction over affirmative asylum applications in the first instance, but the Executive Office for Immigration Review, a DOJ component, adjudicates the de novo review of applications denied by the Asylum Office, as well as asylum claims raised defensively in removal proceedings.170 As a result, both agencies have a claim on promulgating regulations related to asylum. Indeed, the entirety of the asylum regulations is duplicated at 8 C.F.R. § 208 (DHS) and 8 C.F.R. § 1208 (DOJ).171 Ten years after the creation of DHS, none of the duplicative asylum regulations have been revised or eliminated,172 even though they were supposed to be removed after “a further and more detailed division” of responsibility between the agencies.173

The agency regulations contemplate joint rulemaking efforts by DHS and EOIR over their common jurisdictions, but instances of coordination have been relatively rare.174 Where the regulations have been harmonized, DOJ has adopted regulatory changes years after they were implemented by DHS.175 Even nonsubstantive changes to the immigration regulations are subject to the same type of delay.176 More troubling is that inconsistencies have sprung up when DHS has revised its own regulations without coordinating with DOJ to change the duplicate EOIR regulation.177 In its 2012 Retrospective Regulatory Review, which examined only a subsection of the EOIR regulations, DOJ found that numerous EOIR regulations were inconsistent with the equivalent USCIS regulations because DOJ had not kept up with DHS’s amendments.178

Asylum advocates recognized the risks posed by this dual system of rulemaking almost as soon as it was announced.179 Of particular interest to this Note, the American Immigration Lawyers Association (AILA) was specifically concerned that DOJ would issue regulations “proscribing gender-related persecution claims” even as DHS was contemplating “a conflicting regulation recognizing” such claims.180 In fact, neither DHS nor DOJ has taken steps to promulgate a new rule pertaining to gender-based persecution. But AILA’s concern over the possibility of conflict underscores the importance of rulemaking by both agencies. If the domestic violence asylum regulation is promulgated by only a single agency, the rules governing an asylum claim would vary significantly depending on whether the claim was heard by an asylum officer or an immigration judge. If DHS’s rules are more permissive than DOJ’s rules, the fate of an applicant’s claim could vary depending on whether she first raises her asylum claim at the Asylum Office or in removal proceedings. Stricter rules at the DHS level would have a less significant effect on overall outcomes because every Asylum Office denial is entitled to de novo review by an immigration judge,181 but could nonetheless cause significant delays in securing immigration benefits. In contrast, a joint regulation would be binding on asylum officers and immigration judges and would ensure that the same standard is applied at every level of asylum review. In addition, the regulation would be entitled to deference by courts of appeals reviewing administrative decisions.182 This would avoid the problems of horizontal equity that exist now and that would be exacerbated by single-agency rulemaking.

D. Effects on Domestic Violence Asylum Claims

By resolving doctrinal inconsistencies and sending a clear statement that asylum should protect domestic violence victims, the proposed regulation would significantly improve the adjudication process for asylum claims. The proposal addresses or renders moot each of the doctrinal problems previously identified in the L-R- framework. The first prong of the regulation, approving gender-only social groups, avoids an individualized inquiry into the immutability, visibility, and particularity of an applicant’s social group. A domestic violence victim will be able to seek persecution as part of the particular social group of “women,” or perhaps “women in [her country].” The regulation also establishes as a matter of law the nexus between domestic violence and gender, which has been questioned by some immigration judges.183 This will narrow the scope of the adjudicator’s inquiry to focus on whether an applicant has proved that persecution occurred and whether her government was unwilling or unable to protect her. While it is still possible that different adjudicators would decide these questions differently, the number of judgments that an individual adjudicator has to make will be smaller, and thus the scope for inconsistencies narrower.

The regulation would also officially repudiate the view that domestic violence is simply not the type of persecution that asylum is meant to address. In the years since DHS first took the position that domestic violence victims may be eligible for asylum,some immigration judges have continued to deny women asylum because they do not believe domestic violence can be the basis for asylum under any circumstances.184 One judge denied immigration relief because while other countries are “‘not as good’ as the United States on women’s rights, . . . ‘that doesn’t mean that the United States should grant asylum to all women of the world.’”185 This regulation would clarify that this is not a valid reason to deny asylum to domestic violence victims; if some judges continued to deny asylum on this basis, it would be reversible error.

In addition, the regulation would address the concerns of adjudicators who want to protect women but feel that the law does not currently allow them to do so. Several immigration judges have denied domestic violence asylum claims even while stating on the record that they would like to be able to grant asylum in such cases.186 Some of these judges have recognized the doctrinal problems with visibility and particularity discussed earlier in this Note; others have said that they do not feel free to grant such innovative claims without guidance from the BIA.187 A joint DHS-DOJ regulation would have the same binding effect on immigration judges as a precedential BIA decision, and thus would free friendly immigration judges to begin granting asylum to deserving domestic violence victims.

Overall, this proposal would do a better job addressing the ambiguities in the handling of domestic violence asylum claims than prior efforts at reform would have done. As noted in Section II.B, DOJ under Attorney General Reno tried once before to promulgate regulations governing domestic violence asylum claims, but the rules were never finalized.188 The DOJ Proposed Rule purported to alter the “particular social group” requirement in a way that would benefit domestic violence asylum claimants. But the definition of “particular social group” merely restated the Acosta “common, immutable characteristic” language, which would do nothing to alter the court and BIA precedents disapproving of particular social groups defined by gender alone.189 The DOJ Proposed Rule also suggested six additional factors for courts to consider in deciding whether something is a particular social group.190 None of these factors is determinative, and the rule does not make clear how many criteria a social group must meet or how adjudicators should prioritize the different factors.191 A social group of “women” would satisfy some of these factors and not others. In the absence of a clearer statement that social groups can be defined by gender alone, asylum applicants would likely continue to rely on complicated social group definitions that are more likely to be circular, to fall afoul of the BIA’s social visibility and particularity requirements, and to inaccurately reflect the true nature of the persecution suffered. In contrast, the rule proposed here resolves any ambiguity by making it clear that gender alone can define a particular social group.

The DOJ Proposed Rule explicitly noted that it was not announcing “a categorical rule that a victim of domestic violence is or can be a refugee on account of that experience or fear, or that persons presenting such claims may be found eligible for relief or granted relief as a matter of discretion in certain specified circumstances.”192 The Department “tentatively concluded” that a case-by-case approach would be better than a categorical rule.193 However, with the benefit of an additional thirteen years of case-by-case adjudication of domestic violence asylum claims, it is now evident that only a categorical rule will provide even-handed protection to all meritorious domestic violence asylum claims.

E. A Flood of Refugees?

One criticism that has been leveled at proposals to make it easier for women to seek asylum on the basis of domestic violence is that doing so will open a “floodgate” of female asylum-seekers to the United States.194 Even if a flood of immigrants werelikely, the argument that we should not extend protection to domestic violence victims because there are too many of them is simply unprincipled. Such a fear has not stopped the United States from extending asylum to other victims of persecution. When the asylum statute was amended to cover Chinese citizens fleeing the one-child policy, for instance, the potential beneficiaries included almost the entire adult female population of China, all of whom could be subject to forced abortion or sterilization if they violated the policy.195 To the extent that a fear of floodgates is really a concern about fraudulent asylum applications, then it is appropriate to take measures to ensure a rigorous review process for asylum claims.196 But excluding an entire class of asylees simply because there may be too many of them is unprincipled and unjust.

But for legal and practical reasons, a flood of asylum-seekers is unlikely to materialize, even if the reforms proposed by this Note are adopted. The critics are correct that the total number of abused women worldwide is likely incredibly high. While it is impossible to calculate a precise figure,197 the total number of female domestic violence victims is likely to be in the hundreds of millions. However, the regulatory reform proposed by this Note would not make all of those women eligible for asylum. Asylum-seekers would still have to show that the violence they suffered and would be likely to suffer in the future if they returned to their home country would reach the level of persecution, which is a high standard of harm.198 A single or occasional instance of physical abuse would not satisfy that standard. Abuse that was limited to controlling behavior, even severely controlling behavior, would also likely not be persecution if it was not accompanied by violence.199 Applicants would also have to show that their home country’s government would be unwilling or unable to protect them.200 In most cases, a domestic violence victim who had not actually gone to the authorities to seek protection and been refused would not be eligible for asylum.201

Furthermore, as a practical matter, it is unlikely that all of the potentially eligible asylees would actually take the drastic step of coming to the United States. Past experience supports the conclusion that no “flood” of immigrants will materialize when a new type of asylum claim is recognized. After the BIA’s decision in In re Kasinga allowed asylum claims based on female genital mutilation (FGM)202—opening the door to asylum applications by millions of women and girls who are required to undergo FGM—the Immigration and Naturalization Service did not see an appreciable increase in FGM-based asylum claims.203 Perhaps even more than other types of asylees, prospective domestic violence asylum-seekers are likely to be reluctant to come to the United States except as a last resort. As for all asylees, making the trip to the United States is costly and, in some cases, dangerous. Additionally, while all asylum-seekers lose crucial social and family ties when they leave their home countries, domestic violence asylum-seekers are likely to have an even more difficult time reestablishing those ties after they receive asylum. Successful asylees are entitled to bring their spouses and children to the United States on “following-to-join” petitions.204 But many countries require the permission of both parents for a child to leave the country.205 If an asylee’s abuser is also the parent of her children, he then has a veto power over her efforts to reunify her family. Leaving her home country to seek asylum may also reveal the abuse she suffered to friends and family for the first time. In light of the stigma that attaches to domestic violence in many cultures, it is likely that some women will continue to tolerate abuse rather than take the drastic step of coming to the United States.

But the fact that many qualifying domestic violence victims will not actually be able to take advantage of the protection of asylum underscores another problem—not that there are too many asylum-seekers, but that there are too few. This problem is endemic to the asylum system as a whole, not limited to domestic violence asylum claims. Unlike the refugee system, which places quotas on the number of refugees who will be admitted each year,206 asylum relies on the difficulty of getting to the United States as a natural cap on the number of asylees. Viewed cynically, it may appear that the United States has made an ostensible commitment to protect people all over the world from human rights violations while relying on the fact that it will never be required to deliver fully on its promise.

While it is important to recognize the limitations of asylum as a practical solution for large-scale human rights violations, it would nonetheless be a mistake to overlook either the real benefit it provides to those asylum-seekers who do make it to the United States or the symbolic value of asylum as an expression of our commitment to human rights. As a practical matter, even though expanded access to asylum will not protect all victims of domestic violence, it is a way to offer protection to at least some of the women facing dire abuse at home. Symbolically, official recognition for domestic violence-based asylum claims also reaffirms the country’s commitment to stopping human rights violations against women, including domestic violence. Congress understands asylum’s symbolic value: its decision to amend the asylum statute to include victims of forced abortions and sterilizations was a statement to the Chinese government of the U.S. position on acceptable forms of population control. Similarly, official recognition of domestic violence asylum claims will demonstrate that the United States does not tolerate domestic violence, even when it is sanctioned by traditional gender norms.207 Finally, a symbolic statement about domestic violence may help bring about policy changes in other countries. International recognition of FGM as a violation of women’s human rights—including the U.S. recognition of FGM as the basis for an asylum claim—may have contributed to the recent reduction in the practice in many countries.208 Countries that today do little to protect domestic violence victims may be prompted to greater action if they see that the United States is taking in many of their citizens as domestic violence asylees.


After nearly twenty years of legal limbo, it is past time to put domestic violence asylum claims on a firmer legal footing. Women like Rodi Alvarado, L.R., and Angela face enormous dangers when they choose to escape violent relationships and seek refuge in the United States. It is unjust that their fates often hinge on the random chances that decide whether their asylum claims will be heard first by asylum officers or immigration judges, in Hartford, Connecticut, or Eloy, Arizona. The regulatory reform proposed in this Note would resolve ambiguity over the legal foundation for domestic violence asylum and contribute to greater consistency in the application of asylum law nationwide.

But as much as it is needed for its practical effects, the proposed regulatory reform is also needed as a symbol of the United States’s commitment to women’s human rights. The Refugee Convention promises to protect individuals from persecution on account of fundamental aspects of their identity. Adjudicators’ reluctance to grant asylum to domestic violence victims often reflects a deep misunderstanding of the nature of domestic violence. It is not just a conflict between two individuals or an overreaction to jealousy, anger or frustration. Domestic violence subjugates women because they are women. As the United Nations has said, it is just as much persecution as if women were put into concentration camps or otherwise persecuted in more “traditional” ways. New regulations on domestic violence asylum would simply reaffirm our commitment—and our international law obligation—to protect women who seek asylum on our shores from being returned to face further violations of their basic human rights.