Corps de l’article


The international refugee definition in the United Nations Convention Relating to the Status of Refugees does not enumerate “gender” as a recognized ground of persecution.[1] The definition requires claimants to demonstrate a well-founded fear of persecution against which their home state is unable or unwilling to protect, and which is linked to one of five enumerated grounds: “race, religion, nationality, membership of a particular social group or political opinion.”[2] Canada became the first state signatory to the Refugee Convention to address this omission when, in 1993, the Immigration and Refugee Board (IRB) released guidelines on Women Refugee Claimants Fearing Gender-Related Persecution.[3] While the Guidelines do not add “gender” to the definition, they outline a framework for analysis, complete with substantive and procedural directives, for interpreting the definition in a gender-sensitive manner. One of the Guidelines’ most profound contributions was the recognition that domestic violence perpetrated by non-state actors can amount to persecution and form the basis for a refugee claim. This recognition marked a sea change in Canada’s approach to refugee protection and extended refugee status to countless women previously denied protection. It has been hailed as one of the “most remarkable achievements in Canadian legal history.”[4]

In this article, in anticipation of the Guidelines’ twentieth anniversary, I examine how Canadian adjudicators have approached refugee cases involving domestic violence over the last two decades. My aim is to evaluate whether the Guidelines’ goal of encouraging a gender-sensitive refugee-determination process has been meaningfully realized in cases involving domestic violence. I survey the publicly available refugee-determination decisions involving domestic violence issued subsequent to the Guidelines’ release, and examine these through a comparison with cases involving forced sterilization and genital cutting. Examining 645 cases, I argue that over the last two decades, adjudicators have adopted distinctly different methods of analysis in cases involving domestic violence as compared with these other gender-based claims. Most notably, in cases involving forced sterilization and genital cutting, in keeping with the Guidelines’ directives, adjudicators consistently identified these practices as rights violations. In these cases, they generally recognized the claimant as a refugee because she was victim of a persecutory practice. In contrast, adjudicators rarely identified domestic violence as a rights violation in itself and demonstrated a general predisposition to finding persecution in cultural difference. That is, adjudicators generally recognized domestic violence claimants as refugees not because they were victims of persecutory practices but because they were victims of persecutory cultures.

While this approach has proven successful in securing protection for certain refugee women, it establishes incorrect criteria by which to evaluate domestic violence claims. I develop my analysis in four parts. In Part I, I chart an overview of the principles governing gender-related refugee claims, to provide context for discussion. In Part II, I detail the results of my study and explain the different patterns of analysis emerging from the domestic violence cases as compared with other gender claims. Based on these findings, in Part III, I examine the domestic violence cases in more detail. Through a close reading of select decisions, I identify two key problems with the adjudicative tendency to locate domestic violence persecution in cultural difference. First, I argue that by selectively blaming “culture” for refugee women’s persecution, the cases do not accord due weight to factors other than culture that make women vulnerable to domestic violence, such as material disparities, gender hierarchies, and existing power arrangements. Second, I argue that this approach constructs non-Western culture as a place where domestic violence occurs because of the so-called “inherent” vulnerability of the women located in that cultural milieu. As a result, women who cannot authentically narrate their experience of violence through the script of vulnerability face legal and conceptual barriers in proving their refugee claims. In these two key respects, the domestic violence cases promote an incorrect understanding of gender violence and gender difference. They establish unduly narrow criteria by which to evaluate women’s vulnerability to domestic violence persecution and thus risk excluding genuine refugees from protection.

Building on these critiques, in Part IV, I argue that the adjudicative tendency to locate domestic violence persecution in cultural difference both stems from and reflects a defensive anxiety. Unlike forced sterilization and genital cutting—“exotic” practices perceived to occur only in foreign countries—domestic violence is also common within Canada. Domestic violence claims thus cut along familiar lines of difference and cannot easily be identified as rights violations in the Canadian legal lexicon. Viewed in this light, the tendency to locate persecution in cultural difference can be seen as a protective device that distinguishes the violence suffered by refugee women from the violence suffered by Canadian women. The reluctance to inquire into broad power arrangements that make women vulnerable to domestic violence abroad makes it easier to ignore the prevalence of these same structures in Canada. In this key respect, the approach adopted in the domestic violence cases operates to replicate problematic assumptions about gender violence and gender difference, while also making it harder to challenge these assumptions in the Canadian legal discourse.

I conclude by arguing in favour of a discursive shift in the adjudicative approach to domestic violence cases. I suggest that a re-examination of the assumptions at play in these cases will allow Canadian legal actors to view domestic violence in a more accurate light and will create a more sophisticated body of law that can better address gender violence claims. Such an approach would better comply not only with the spirit of the Guidelines but also with Canadian standards of equality and rights protection.

I. The Legal Framework

A. The Refugee Definition: An Overview

To advance a gender-related refugee claim, a claimant must first demonstrate compliance with the refugee definition set out in article 1(A)(2) of the Refugee Convention, as amended by the 1967 Protocol Relating to the Status of Refugees.[5] Article 1(A)(2) defines a refugee as a person who

owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it.

This definition is incorporated into domestic law, almost verbatim, via section 96 of the Immigration and Refugee Protection Act (IRPA).[6] The definition involves a range of considerations but can be parsed into three central components. Simplified, it requires proof of (i) a well-founded fear of persecution (ii) demonstrative of a failure of state protection (iii) that is causally connected to one of the grounds enumerated in the definition: race, religion, nationality, political opinion, or membership of a particular social group.[7] As noted above, the ground “gender” remains conspicuously absent.

Canada was the first state signatory to the Refugee Convention to formally recognize that gender can ground a claim for refugee status in 1993, when the Immigration and Refugee Board released the Guidelines.[8] The Guidelines do not incorporate “gender” into the refugee definition but instead instruct adjudicators that the definition “may properly be interpreted as providing protection for women who demonstrate a well-founded fear of gender-related persecution by reason of any one, or a combination of, the enumerated grounds.”[9] While the Guidelines lack legislative force, the IRB gives members bureaucratic incentive to apply them seriously by requiring that written reasons be provided for any decision that rejects the principles set out in the Guidelines.[10] The central claim—and operating rationale—that underpins the Guidelines is that the “existing bank of jurisprudence on the meaning of persecution is based, for the most part, on the experiences of male claimants.”[11] The Guidelines explain that since the “circumstances which give rise to women’s fear of persecution are often unique to women,” gender claims necessitate a different method of analysis.[12] They also explicitly recognize that women may be targets of persecution “at the hands of non-state agents of persecution, where the state is either unwilling or unable to protect.”[13]

Several months after the introduction of the Guidelines, the Supreme Court of Canada released its decision in Canada (Attorney General) v. Ward, in which it confirmed the possibility of naming “gender” as a ground of persecution.[14] While the case did not directly involve a gender claim, the Court held that “gender” can constitute a “particular social group” within the meaning of the refugee definition.[15] Importantly, the Court in Ward also recognized that persecution can be perpetrated either directly by the state or by non-state agents.[16] This finding has proven crucially important in substantiating many gender claims, particularly claims involving domestic violence.

The Court in Ward further explained that state protection need not be perfect to satisfy the refugee definition.[17] States that have functioning democratic governments, the Court held, can be presumed capable of protecting their citizens. In such cases, the claimant must demonstrate “clear and convincing” evidence of her home state’s inability or unwillingness to protect her from harm.[18] Recognizing that this evidentiary standard may establish difficult hurdles for women facing gender violence, the IRB revised the Guidelines after Ward’s release. In their current form, the Guidelines instruct that “where the claimant cannot rely on the more standard or typical forms of evidence as ‘clear and convincing proof’ of failure of state protection, reference may need to be made to alternative forms of evidence to meet the ‘clear and convincing’ test.”[19] Elsewhere, the Guidelines also encourage adjudicators to “consider, among other relevant factors, the social, cultural, religious, and economic context in which the claimant finds herself.”[20] Before examining the gender cases in detail, I first chart an overview of the basic principles governing the adjudication of these claims, following the refugee definition’s three central components.

B. Finding a Human Rights Violation

The persecution requirement lies at the heart of the refugee definition and obliges claimants to demonstrate a subjective fear of persecution related to an objective risk of harm.[21] In Ward, the Supreme Court of Canada defined persecution as the “sustained or systematic violation of basic human rights demonstrative of a failure of state protection.”[22] Ward thus requires adjudicators to make two distinct but interrelated findings: that the claimant faces a violation of basic human rights and that her state is unable or unwilling to protect her from this violation. The centrality of human rights to this analysis is notable and signals a clear decision on the part of the Court to establish human rights as the yardstick against which refugee claims will be measured.[23] This approach, the Court explained, both complies with and reflects the underlying purpose of the United Nations Refugee Convention, namely “the international community’s commitment to the assurance of basic human rights without discrimination.”[24] The requirement that these violations be sustained or systemic further signals that minor harms—or instances of “mere discrimination”, as these are dubbed in the case law—will not amount to persecution.[25]

Like Ward, the Guidelines interpret the persecution requirement by reference to human rights principles and instruct that claimants facing gender persecution must show evidence of “serious ... harm which detracts from the claimant’s fundamental human rights.”[26] Recognizing that violence is often experienced in uniquely gendered ways, the Guidelines identify several “female-specific experiences”[27] that amount to persecution, listing rape, infanticide, genital cutting, bride burning, forced marriage, domestic violence, forced abortion, and compulsory sterilization as examples. The Guidelines further pre-empt and counter the argument that the apparent universality of gender violence might preclude its recognition as persecution: “The fact that violence, including sexual and domestic violence, against women is universal is irrelevant when determining whether rape, and other gender-specific crimes[,] constitute forms of persecution.”[28] Instead, they explain, “The real issues are whether the violence -- experienced or feared -- is a serious violation of a fundamental human right for a Convention ground and in what circumstances can the risk of that violence be said to result from a failure of state protection.”[29]

The fact that both Ward and the Guidelines anchor the refugee analysis in human rights principles is significant. First and foremost, it expanded refugee law’s substantive eligibility grounds and extended refugee status to countless claimants previously denied protection, many of them women. The significance of these outcomes cannot be underestimated. Second, the emphasis on human rights signalled a shift from what Matthew Price has termed the “political model” of asylum toward the “humanitarian model of asylum.”[30] As originally conceived, international refugee law was devised as a political instrument, a vehicle for protecting persecuted people while also advancing Western political values.[31] While this approach was prominent throughout the Cold War era, as the global political climate changed, so did the international community’s approach to refugee law.[32] With the growing momentum of the international human rights movement, signatory states increasingly turned to human rights principles in their interpretation of refugee law. This shift sought to distance refugee law from its political roots and anchor it in a universally accepted and ostensibly “neutral” rights-based model of assessment.[33] Because human rights principles are thought to transcend the particular—or, in the words of the Supreme Court of Canada, “transcend subjective and parochial perspectives and extend beyond national boundaries”[34]—they are perceived to form the basis of universal protection standards not subject to the political preferences of receiving states.[35] Anchoring refugee law in the human rights paradigm allows state actors to refrain from expressing overt value judgments about the conduct of persecuting states.[36]

In the Canadian context, Ward has been credited with shifting refugee law away from politically motivated considerations and anchoring it firmly in the human rights paradigm.[37] Critics praised the decision for steering “a course away from the days when refugee law was used to condemn publicly enemy states for their misbehaviour.”[38] They further hailed the decision as a powerful affirmation that the “protection of those at risk of serious human rights violations is the lens through which refugee law must be focused.”[39] Since Ward, the primary goal of human rights protection has been formally inscribed in the IRPA, which provides that Canada’s refugee protection regime “is in the first instance about saving lives and offering protection to the displaced and the persecuted,” as an expression of “Canada’s respect for the human rights and fundamental freedoms of all human beings.”[40]

Indeed, both Ward and the Guidelines have been credited with shifting the refugee analysis toward a rights-based model of assessment, which serves to channel the language of international human rights and thereby sidestep accusations of cultural imperialism. The drafters’ decision to incorporate human rights principles into the Guidelines was devised to do just that.[41] As Nurjehan Mawani, former chair of the Immigration and Refugee Board, has explained, the Guidelines’ application “is not simply a matter of imposing western standards on other countries” but rather a “matter of respecting internationally accepted human rights standards.”[42] As Audrey Macklin has shown, this approach was intended to “bolster the proposition that characterizing certain forms of sex-based discrimination and oppression as persecution represents not only Canada’s view, but also that of the international community.”[43] To this end, the Guidelines incorporate eight international human rights instruments to “assist decision-makers in determining what kinds of treatment are considered persecution.”[44] They further assert that these instruments establish an “objective standard” for evaluating women’s persecution claims.[45]

The human rights approach to refugee law has successfully expanded its substantive eligibility grounds but has done little to move the refugee analysis away from politically motivated considerations and toward a politically neutral model of assessment. Human rights principles do not embody an objective morality, and nor does Canadian refugee law simply by incorporating these principles. In fact, it is the other way around. The Canadian national identity—our understanding of ourselves, our national values, and the ideals that we privilege in our laws—informs what we see as good or bad, prescribes our political choices, and determines which practices we designate as rights violations. Put another way, the Canadian national identity does not merely reflect some external or universal morality but in fact precedes it. The pretense that Canadian refugee law somehow ceases to be value laden because it references human rights principles obscures the extent to which refugee determinations, even when grounded in human rights principles, are always already informed by national ideologies and political choices. In fact, Canada’s stated commitment to rights protection itself operates as an expressive vehicle through which to assert and define the Canadian national identity. By asserting a commitment to human rights—which, as Peter Fitzpatrick argues, represent the “pervasive criteria” by which global standards of civilization and progress are judged—Canada claims its status as “universal exemplar” in refugee protection.[46] Viewed in this light, far from distancing the refugee analysis from Canadian political ideologies, the incorporation of human rights principles instead operates to reinscribe them. Throughout the refugee cases, the commitment to rights protection itself serves as a means through which to assert Canada’s claimed identity as human rights protector and refugee acceptor. This becomes clear in the context of the state-protection analysis, to which I now turn.

C. Finding a Failure of State Protection

In order to constitute persecution, a practice must be shown not only to violate human rights but also to do so in ways that demonstrate a failure of state protection. This aspect of the definition—described in Ward as the “lynch-pin of the analysis”[47]—requires claimants to demonstrate “clear and convincing confirmation of [the] ... state’s inability to protect.”[48] This requirement reflects the underlying premise of international refugee law. As stated in Ward, since refugee law “was formulated to serve as a back-up to the protection one expects from the state of which an individual is a national”[49] it comes “into play where no alternative remains to the claimant.”[50] This is because refugee claims “were never meant to allow a claimant to seek out better protection than that from which he or she benefits already.”[51] This “surrogate or substitute protection,” the Court explained, can be activated only upon demonstrated proof that the claimant’s home state is unable to protect her from harm.[52]

Such formulations, as Price explains, prescribe that when states fail to protect their nationals, sovereignty can no longer “serve as a shield of immunity behind which unjustified harm can be inflicted with impunity.”[53] In the Court’s analysis, a state’s inability to protect its nationals is reflective of a failure of state sovereignty. States “should be presumed capable of protecting their citizens [since s]ecurity of nationals is, after all, the essence of sovereignty.”[54] By this logic, Ward effectively asks adjudicators to counteract a finding of a foreign state’s inability to protect with a finding of Canada’s ability to protect. The ability to protect—the “essence” of sovereignty—thus becomes the test not only of a subject’s status as a refugee but also of Canada’s status as provider of refuge. It is mobilized as an expression of the Canadian identity, a reflection and embodiment of Canadian national values and rights-protecting norms. In this way, the Ward analysis operates to subtly reinforce Canada’s claimed identity as rights protector: with every refugee admission, the nation is repeatedly affirmed as a place that protects. In granting a claimant refugee status, Canada both guarantees her protection and offers her provisional membership in a national identity constructed, in part, by the ability to protect.

Proof of Canada’s ability to protect thus also operates to position Canada as “refugee acceptor” in the division between “refugee-accepting” and “refugee-producing” states. By now thoroughly considered in the scholarly literature, this division replicates the familiar divide between us and them, self and other, West and rest; it also depicts refugee-receiving states as civilized, superior, and law-abiding, in opposition to the backward, rights-violating, refugee-producing states of the global East and South. As Catherine Dauvergne and Jenni Millbank explain, refugee law is “erected on a foundation of ‘othering’,” which is “sustained by a recurrent division between ‘us’ and ‘them’.”[55] Within this framework, as Macklin argues, Western states like Canada assert their identities as refugee acceptors by distinguishing themselves from “what they are not, namely, the kind of governments that do the kinds of things to people that propel them to claim refugee status.”[56]

D. Grounds of Admission

To satisfy the refugee definition, after having demonstrated the state’s failure to protect from persecution, a claimant must next establish a nexus between the feared persecution and at least one of the definition’s enumerated grounds: race, religion, nationality, political opinion, or membership of a particular social group. While the Guidelines encourage claimants to advance gender claims under all five grounds, the most common (and, arguably, most effective) method shows the claimant’s membership in a gender-defined “particular social group”.[57] According to Ward, commonality and immutability are the two requirements necessary for defining a social group: group members must “share a common, immutable characteristic” that “either is beyond the power of the individual to change or that is so fundamental to his identity or conscience that it ought not be required to be changed.”[58] Further emphasizing the significance of human rights to the interpretation of the refugee definition, Ward instructs that particular social group designations should “take into account the general underlying themes of the defence of human rights and anti-discrimination that form the basis for the international refugee protection initiative.”[59]

Subsequently to Ward’s release, the “membership of a particular social group” ground has evolved into a catch-all category that encompasses a range of ascriptive characteristics not specifically enumerated in the definition, such as gender, sexual orientation, disability, or lineage, to list a few.[60] Canadian adjudicators have adopted two distinct methods by which to define gender-specific social groups. Some have recognized groups defined as “women” or by the shared attribute “gender”, following the Guidelines’ directive that gender “is an innate characteristic and, therefore, women may form a particular social group.”[61] Others have recognized more particularized formulations, defining the social group by reference to gender in addition to other criteria. The Guidelines also endorse this latter approach and instruct as follows:

Particular social groups comprised of sub-groups of women may also be an appropriate finding in a case involving gender-related persecution. These particular social groups can be identified by reference to factors, in addition to gender, which may also be innate or unchangeable characteristics. Examples of other such characteristics are age, race, marital status and economic status. Thus, for example, there may be sub-groups of women identified as old women, indigenous women, single women or poor women. In determining whether these factors are unchangeable, consideration should be given to the cultural and social context in which the woman lives, as well as to the perception of the agents of persecution and those responsible for providing state protection.[62]

It is worth noting that signatory states like the United Kingdom[63] and Australia[64] have rejected this approach, making these highly particularized formulations distinctly Canadian.[65]

II. The Domestic Violence Cases: An Overview

A. Methodology

Using the three central components as a framework for analysis, my study examines how adjudicators approach refugee-determination cases involving claims of domestic violence. The domestic violence cases comprise the largest category of reported cases, numbering 528 in total.[66] This figure includes claims of spousal or familial physical violence, sexual violence, and spousal or familial rape. It does not include claims involving rape or assault by state actors or non-spousal or non-familial third parties. Of the cases in the category, 107 resulted in positive outcomes (either refugee admissions or successful appeals) and 421 resulted in negative outcomes (denials of status requests or appeal requests). Most of the unsuccessful cases were rejected on the grounds that the claimant could not prove a failure of state protection, did not advance sufficiently credible evidence, or lacked credibility.[67] These figures show an acceptance rate of roughly 20 per cent. Since the IRB publishes only a small fraction of its decisions, these figures do not accurately reflect the acceptance rate in the domestic violence cases. Indeed, partial statistics obtained from the IRB report a different rate. Between 1 January 2002 and 31 December 2006, the IRB recorded an acceptance rate of 43.9 per cent in cases involving principal claimants who alleged domestic violence persecution.[68] Between 1 January 2008 and 30 June 2012, the IRB recorded an acceptance rate of 48.7 per cent in cases involving principal claimants who alleged domestic violence persecution.[69] Because the IRB has not electronically tracked data outside of these specific periods, it is difficult to estimate the acceptance rate for the twenty-year period here examined. With this in mind, my study does not purport to analyze the domestic cases quantitatively but rather reflects on broad patterns of adjudication based on the best available data.

My study analyzed the domestic violence cases through a comparison with the second- and third-largest categories of reported gender cases. Respectively, these categories involve cases of forced abortion or compulsory sterilization, and cases of genital cutting. I examined 60 reported cases involving forced abortion or compulsory sterilization, which I refer to below under the broad heading “forced sterilization”. Of these, 15 resulted in positive outcomes and 45 in negative outcomes, amounting to an acceptance rate of 25 per cent. I also examined 57 reported cases involving genital cutting, of which 23 resulted in positive outcomes and 34 in negative outcomes, amounting to an acceptance rate of 40 per cent. As noted above, since the IRB publishes only a small fraction of its decisions, it is difficult to ascertain whether this data set accurately reflects actual patterns and trends.

I thus examined a total of 645 decisions,[70] restricting my analysis to female claimants only, on the basis that the Guidelines do not apply to male claimants. I did not examine cases involving the remaining “gender-specific” forms of persecution listed in the Guidelines, as these cases were either too few in number or have been well canvassed elsewhere.[71] I also did not examine cases involving persecution on grounds of sexual orientation or sexual identity since the IRB assesses these cases differently and often without referencing the Guidelines.[72] I classified the decisions based on the nature of the claim alleged, following the central components of the refugee definition: the persecution analysis (divided into the human rights–violation assessment and the state-protection assessment) and the grounds-of-admission analysis.

My research suggests that cases involving forced sterilization and genital cutting were characterized by three general tendencies: adjudicators consistently identified forced sterilization and genital cutting as rights violations; they directed comparatively little attention to the state-protection analysis; and they generally admitted claimants on the ground of their membership in broad “particular social groups” defined as “women” or by the common attribute “gender”. In contrast, cases involving domestic violence were characterized by three different tendencies: adjudicators rarely identified domestic violence as a rights violation in itself; they based their determinations on the availability of state protection (and frequently located the source of persecution in cultural difference); and they generally admitted claimants on the ground of their membership in a highly specific “particular social group”. I chart an overview of these findings below, using the three central components of the refugee definition as a framework for analysis.

B. Finding a Human Rights Violation

1. Forced-Sterilization and Genital-Cutting Cases

In cases involving forced sterilization, adjudicators frequently referenced human rights principles in assessing whether the practice amounted to persecution. Most of the reported IRB cases followed the Federal Court of Appeal’s findings in Cheung v. Canada (Minister of Employment and Immigration) (C.A.) that forced sterilization “is a fundamental violation of basic human rights,” a “serious and totally unacceptable violation of ... security of the person” that subjects women to “cruel, inhuman and degrading treatment.”[73] In Zheng v. Canada (Minister of Citizenship and Immigration), the Federal Court similarly held that forced sterilization is a violation of basic rights “ranking high on our scale of values” and amounting to persecution.[74] Following these decisions, at the IRB level, adjudicators consistently held that forced sterilization is a violation of the human right[75] variously identified as the right to security of person,[76] the right to reproductive choice,[77] or the right to bear a child,[78] and that it thus amounts to persecution.[79] The emphasis on human rights was also prominent in a series of cases that excluded claimants from protection pursuant to section 98 of the IRPA because of their role in enforcing China’s one-child policy.[80] While these cases were comparatively few in number, their results were consistent: adjudicators not only recognized forced sterilization as a rights violation but also identified it as a crime against humanity.[81] It is also notable that, in many of these cases, adjudicators used strong language to condemn forced sterilization as an “inhuman” practice,[82] an act of “barbarous cruelty”,[83] and an act “contrary to human dignity.”[84]

Cases involving genital cutting largely followed a similar pattern. In its 1995 decision in Annan v. Canada (Minister of Citizenship and Immigration) (T.D.), the Federal Court identified genital cutting as a “cruel and barbaric practice” that constitutes persecution.[85] IRB adjudicators have arrived at similar determinations: adjudicators have held that it is “well established” that genital cutting “is a persecutory act,”[86] which “must be characterized as a form of persecution.”[87] IRB adjudicators have further held that genital cutting is a “sustained and systemic violation of several of the most fundamental [human] rights,”[88] including the right to life, liberty, and security of person; the right to health; the right against cruel or inhuman treatment; the right not to be married against one’s consent; and the right to “special protection for motherhood.”[89] The rationale underpinning these findings is that genital cutting is a practice “so severe” that its “characterization [as] ... persecution [is] beyond dispute.”[90] In keeping with the Guidelines’ instructions, adjudicators frequently cited human rights instruments in these decisions.[91] As with the forced-sterilization cases, many adjudicators depicted genital cutting as “cruel and barbaric”,[92] a “horrendous” custom,[93] a “torturous custom”,[94] a “horrific torture”,[95] and an “atrocious mutilation”.[96]

2. Domestic Violence Cases

The cases examined in this data set confirm one very positive trend: adjudicators consistently recognized domestic violence as persecution for the purposes of the refugee analysis. This finding is consistent with those of Constance MacIntosh, who examined 135 domestic violence decisions released between 2004 and 2009.[97] Two distinct patterns further distinguish adjudicators’ approach to the first branch of the persecution analysis, the human rights–violation assessment. First, despite Ward’s instruction that persecution is defined as a sustained and systemic violation of human rights, adjudicators rarely identified domestic violence as a rights violation in itself. Second, despite the Guidelines’ instructions that adjudicators should consider human rights instruments and norms, adjudicators rarely referenced human rights in the domestic violence cases. My study identified only fifteen reported decisions over the twenty-year period examined in which adjudicators either explicitly identified domestic violence as a human rights violation or relied on human rights instruments.[98] All of these decision resulted in positive admissions. I did not identify a single negative decision in which adjudicators cited human rights principles. The lack of references to human rights instruments and principles in the domestic violence cases is notable.

C. Finding a Failure of State Protection

1. Forced-Sterilization and Genital-Cutting Cases

At the state-protection stage of the analysis, my study also identified key differences between the domestic violence cases and the forced-sterilization and genital-cutting cases. In the forced-sterilization and genital-cutting cases examined, adjudicators by and large engaged in the state-protection analysis only minimally, often because the state was directly responsible for the persecution at issue. In the cases of forced sterilization, adjudicators generally agreed that, since forced sterilization is a “state-enforced suppression of ... [a woman’s] reproductive capacity,”[99] it would be “objectively unreasonable to expect the claimant ... to enlist the protection of the state, which was acting as an agent of her persecution.”[100] In most of the genital-cutting cases, adjudicators either did not comment on the availability of state protection,[101] presumed state protection would be lacking,[102] or engaged in the state-protection analysis only minimally.[103] This pattern suggests that, in cases involving practices identified as rights violations, the state-protection analysis is less determinative to the refugee assessment.

2. Domestic Violence Cases

In contrast, in the domestic violence cases, adjudicators’ determinations generally hinged on the state-protection analysis. In cases resulting in negative decisions, adjudicators rarely engaged in a contextual analysis when evaluating the availability of state protection, a fact that suggests the standard for demonstrating meaningful enforcement is relatively low.[104] For example, many adjudicators identified the existence of anti–domestic violence legislation, women’s shelters, or other protective services as sufficient to show the availability of state protection, with few referring to documentary material on the adequacy or accessibility of these measures. Adjudicators usually found state protection was available in cases where the state made good-faith efforts to take the problem of violence against women seriously[105] by enacting legislation,[106] training specialized police units,[107] providing legal-aid services,[108] or establishing shelters or other forms of recourse[109] and support.[110] As MacIntosh argues, this approach casts doubt on whether adjudicators substantively engaged with the evidence before them or merely “conflate[d] the fact that states have enacted protective legislation with the finding that there is protection for the claimant.”[111]

In many of the cases resulting in positive admissions, adjudicators found state protection was unavailable to the claimant given evidence of cultural norms. For example, in Re L.T.D., the IRB reasoned that while “wife battering is a criminal offence” in Ghana, it was still “condoned in traditional society” and “culturally” accepted, with the result that state protection was not available to the claimant.[112] In Re L. (C.B.), the IRB concluded that state protection was unavailable in Argentina because of its “machismo” and because of the prevalence of gendered “misconceptions ... deeply rooted in the Argentinean culture.”[113] In Re G. (D.M.), the IRB cited evidence that “culture and tradition inhibit the achievement of full equality for women” and that, as a result, state protection was not available.[114] In Re H.T.O., the IRB found state protection was absent given the “cultural ethos” in Bangladesh.[115] Similarly, in Re O.E.X., the IRB found that state protection was lacking in Nigeria given evidence that “[c]ultural norms endorse wife assault among the Yoruba.”[116] There are numerous examples of similar findings in the reported decisions.[117] Such findings were comparatively absent in the forced-sterilization and genital-cutting cases.

My concern with these cases is not with the finding that the claimant’s culture was violent, patriarchal, or oppressive. Indeed, cultural norms and traditions can often be oppressive and can often give rise to or reinforce women’s persecution in both domestic and nondomestic settings. My concern is rather with the unintended consequences that stem from locating persecution so narrowly in cultural difference. I return to this point below, but first, I examine how adjudicators approached the grounds-of-admission stage of the refugee analysis.

D. Grounds of Admission

1. Forced-Sterilization and Genital-Cutting Cases

At the grounds-of-admission stage of the analysis, the cases involving forced sterilization or genital cutting also followed a distinctly different adjudicative pattern as compared with those involving domestic violence. As noted above, the Guidelines contemplate two ways through which to designate a gender-specific social group. The first is through the broad designation “women”, while the second is through more particularized formulations specific to a subgroup of women. In most of the reported decisions involving forced sterilization or genital cutting, adjudicators adopted the former approach. In the forced-sterilization cases, for example, most adjudicators either did not define the social group,[118] or defined it as “women”.[119] A small number of adjudicators identified more particularized formulations[120] or admitted claimants under the “political opinion” ground.[121] Similarly, in the genital-cutting cases, adjudicators generally recognized social groups defined either as “women”[122] or as “women facing genital mutilation.”[123] Only a few cases recognized more particularized formulations.[124]

2. Domestic Violence Cases

In contrast, in the domestic violence cases adjudicators frequently identified highly particularized social groups defined by a combination of gender, nationality, and personal circumstance. Examples of such particularized formulations include: “women in Hungary who are subject to domestic violence,”[125] “women in China subject to spousal abuse,”[126] “Trinidadian women subject to domestic violence,”[127] “Mexican women subject to domestic violence,”[128] and “Ghanaian women subject to spousal and family abuse,”[129] to list just a few.[130] Adjudicators also identified more particularized social groups, such as “women in Turkey who are victims in abusive relationships or victims of abuse at the hand of rejected suitors, in circumstances where agents of the state, including the police, are unable or unwilling to provide protection,”[131] “Bulgarian women vulnerable to wife abuse by men with government influence,”[132] “Westernized Tajik woman in a society moving towards Islamic orthodoxy, with no male protection,”[133] “[wives] of ... ‘Machismo’ [men] ... targeted for domestic violence,”[134] and “elderly single women in Pakistan subject to abuse due to kinship.”[135] Perhaps most explicitly, one Federal Court case defined the social group as

new citizens of Israel who are women recently arrived from elements of the former Soviet Union and who are not yet well integrated into Israeli society, despite the generous support offered by the Israeli government, who are lured into prostitution and threatened and exploited by individuals not connected to government, and who can demonstrate indifference to their plight by front-line authorities to whom they would normally be expected to turn for protection.[136]

While these formulations seem attractive in their capacity to recognize that gender persecution is experienced in highly particularized ways, this potential is never realized in the cases.

III. The Limits of Protection

The above analysis suggests adjudicators approach domestic violence claims in distinctly different ways than they do other gender-persecution claims. In cases involving forced sterilization and genital cutting, adjudicators identified these practices both as acts of persecution and as fundamental rights violations. Adjudicators frequently cited human rights instruments and principles in these decisions and also deployed descriptors such as “morally shocking”, “inhuman”, “degrading”, and “barbaric”. The relative paucity of the state-protection analysis, combined with the designation of broad social groups defined as “women”, further suggests that adjudicators view the practices concerned as universal rights violations that target women as women. In contrast, in the domestic violence cases, adjudicators identified domestic violence as a problem of “culture” rather than as a human rights violation per se and adopted highly particularized social-group formulations. The effects of these formulations is clear: when legal actors portray domestic violence as the product of a foreign culture to which only Bulgarian, Pakistani, or Westernized Tajik women are subjected—as opposed to “women” writ large—they construct domestic violence as an “othered” harm perpetrated against only certain subcategories of (non-Western) women.

Viewed together, these cases generate what Kwame Anthony Appiah has identified as a “script”, a dominant narrative according to which the refugee woman’s experience is evaluated and understood.[137] As Appiah explains, demanding rights for people as members of identity groups—in Appiah’s analysis, African Americans or homosexuals—demands that there be a “script” for what it means to be black or gay. The problem with this approach is that it ascribes various attributes and characteristics to members of identity groups that often have little to do with their lived experiences or, in the case of refugee claimants, with their protection needs. In the domestic violence cases, the deployment of this script effectively requires claimants to perform the refugee identity by presenting as “victims of culture”.[138]

In the analysis that follows, I focus on two reasons why I view this approach to be problematic. First, I argue that, by depicting domestic violence so narrowly as a product of culture, the cases fail to recognize factors besides culture that make women vulnerable to domestic violence. This approach to domestic violence establishes unduly restrictive standards by which to evaluate women’s protection needs. Second, I argue that this approach constructs non-Western culture as a place where domestic violence occurs because of the so-called “inherent” vulnerability of women located in that cultural milieu. In so doing, it erects both legal and conceptual barriers for women who cannot authentically narrate their experience of violence through the script of cultural vulnerability. In these key respects, the adjudicators of domestic violence cases remain insufficiently attentive to the complex factors that make women vulnerable to persecution in domestic settings and risk denying protection to genuine refugees facing real threats of persecution. I address each of these points in turn.

A. Obscuring Factors Besides “Culture”

In many of the cases surveyed, adjudicators not only located persecution in cultural difference but also depicted the claimant’s culture as made up almost entirely of gender-subordinating values and thus as fundamentally oppressive of women. For example, in Re Z.D.M., the IRB held that state protection was lacking in Egypt because “Egyptian patriarchal culture and traditions” were fundamentally repressive and hostile toward women and “[w]omen’s experiences of beating, battering and sexual or psychological abuse have been accepted and normalized in Egyptian traditions.”[139] Similarly, in Re R.Z.B., the IRB held that state protection was lacking in Barbados because “Barbados, as other Anglophone Caribbean societies[,] has a strong patriarchal culture and prevailing gender ideologies place women’s interest as secondary to that of men.”[140] In Re C. (X.N.), the IRB held that state protection was lacking in Ecuador since Ecuador’s patriarchal “culture and tradition” rendered the state “unwilling to protect the rights of women who are subject to domestic abuse.”[141] Perhaps most explicitly, in Re B. (P.V.), the IRB held that state protection was lacking in Somaliland, given evidence presented at the hearing that described Somali culture as so “overwhelmingly restrictive and patriarchal” that Somali women had no protection from domestic violence.[142] Based on this evidence, the board concluded that “Somaliland means oppressed women.”[143] In this totalizing formulation, the board presented Somaliland as being synonymous with the oppression of women, suggesting that to be a woman from Somaliland is to be a victim of oppression.

These formulations are striking because they incorrectly present foreign cultures as internally homogenous and comprised primarily (at times, exclusively) of gender-subordinating values. By presenting foreign cultures as fundamentally oppositional to women’s rights, the cases convey the impression that culture is the root cause of the claimants’ persecution. They further convey the impression that the oppression of non-Western women (and the violence they suffer as a result) is integral to these cultures and produced by quintessentially foreign, barbaric, and misogynistic cultural norms. This approach fails to recognize that domestic violence is never created solely by culture, and it leaves unexamined the broader societal arrangements that make women vulnerable to persecution in domestic settings. It also risks obscuring, in potentially harmful ways, the factors besides culture that make women vulnerable to persecution. The sheer number of cases that come before the IRB, advanced by diversely situated women from various countries, should itself demonstrate that the problem of domestic violence cannot be attributed to culture alone. Narrowly depicting domestic violence as a product of culture incorrectly attributes responsibility to a static, insular, and fundamentally foreign constructed entity, and not, for example, to systemic gender hierarchies, uneven power distributions, and economic factors that transcend culture and permeate other aspects of women’s lives.

No less significantly, this approach fails to recognize that culture is always contested within communities and constantly subjected to change and transformation. As Bonnie Honig writes, “culture is something rather more complicated than patriarchal permission for powerful men to subordinate vulnerable women.”[144] She further explains:

[It] is a way of life, a rich and timeworn grammar of human activity, a set of diverse and often conflicting narratives whereby communal (mis)understandings, roles, and responsibilities are negotiated. As such, “culture” is a living, breathing system for the distribution and enactment of agency, power, and privilege among its members and beyond. Rarely are those privileges distributed along a single axis of difference such that, for example, all men are more powerful than all women.[145]

Homi Bhabha similarly cautions against totalizing formulations. The constructed opposition between (foreign) cultures and women’s rights, he argues, mistakenly constructs non-Western societies as patriarchal and oppressive, and identifies non-Western cultures as the root of these problems.[146] These formulations fail to recognize that cultural patriarchy and oppression are always constructed through, and informed by, a myriad of other factors, such as history, race, or socio-economic status. They erroneously depict minority women as “abject ‘subjects’” of their cultures, “huddled in the gazebo of group rights, preserving the orthodoxy of their distinctive cultures in the midst of the great storm of Western progress.”[147] In so doing, they also construct non-Western women as being deprived of meaningful choice and forced to endure the violence of their cultures.

Leti Volpp’s analysis of the role played by discourses of culture in gender-violence claims in the United States aptly summarizes the problems with such formulations. Analyzing disparities in US approaches to domestic violence in immigrant and “white” communities, Volpp explains:

Part of the reason many believe the cultures of the Third World or immigrant communities are so much more sexist than Western ones is that incidents of sexual violence in the West are frequently thought to reflect the behavior of a few deviants—rather than as part of our culture. In contrast, incidents of violence in the Third World or immigrant communities are thought to characterize the cultures of entire nations.

… The philosopher Uma Narayan has calculated that death by domestic violence in the United States is numerically as significant a social problem as dowry murders in India. But only one is used as a signifier of cultural backwardness: “They burn their women there.” As opposed to: “We shoot our women here.” Yet domestic violence murders in the U.S. are just as much a part of American culture as dowry death is a part of Indian culture. In the words of Narayan, when “cultural explanations” are given for fatal forms of violence only in the Third World, the effect is to suggest that Third World women suffer “death by culture.”[148]

Such an approach, Volpp argues, suggests that “only minority cultures are ... traditional, and made up of unchanging and longstanding practices that warrant submission to cultural dictates.”[149] “In hasty expressions of distaste for other cultures perceived as primitive and backward,” she cautions, “we miss the complex ways in which power actually functions in particular communities.”[150] This discursive practice overlooks the various relations of power that inform violence, both in “other” cultures and in our own cultures. It mistakenly conveys the impression that incidents of domestic violence that occur in Western settings are products of individual aberrations or deviances rather than cultural suppositions, entrenched inequalities, material disparities, or systemic failures.

B. Constructing Refugee Women as “Victims of Culture”

Indeed, by depicting foreign cultures as oppositional to women’s rights, the domestic violence cases construct non-Western culture as a place where domestic violence occurs because of the perceived vulnerability of women situated in that cultural milieu. These formulations seem to presume that women brought up in non-Western societies are less likely to be able to resist or struggle against domestic violence because of their culture. For example, in a case involving a claimant from Trinidad, the IRB concluded that state protection would be lacking because the “cultural tradition of the claimant’s nationality ... produced the dependency described.”[151] The board held that the violence suffered by the claimant “created a paralysing dependency which she has been unable to overcome,” which was “reinforced by her cultural tradition” and prevented her from seeking state protection.[152] Similarly, in Re Y. (O.E.), the board allowed the claim, in part, on grounds that the claimant “was under constant pressure to perform the role of an obedient, submissive, good wife, a role which prevails in the traditions of Indian culture.”[153] In a case involving a claimant from Venezuela, the board relied on evidence offering a “cultural interpretation” of the gendered attitudes of Venezuelan society to conclude that “Latin women typically feel helpless to confront their spouses directly,” and that “[f]ear of eliciting violence, including the threat of death, frequently inhibits women from proactively defending themselves.”[154]

As above, it is worth emphasizing that my concern with these cases is not with the finding that cultural norms and traditions can be paralyzing or oppressive, and thus trap certain women in cycles of violence. My concern with these cases is rather with the unintended consequences that stem from locating persecution so narrowly in cultural difference. My study pointed to a small number of cases in which claimants were denied refugee status because they did not present as “victims of culture”. In a case involving a Brazilian claimant fearing domestic violence, for example, the IRB cited the claimant’s perceived capacity for independence as a reason for denying her claim.[155] It held the claimant was unable to satisfy the refugee definition because “her adjustment in Canada demonstrate[d] her capabilities for independence” and suggested she would be able to live safely in Brazil.[156] In another case involving a claimant from St. Vincent, the board similarly determined that the claimant’s perceived capacity for independence militated against a finding of persecution.[157] While the board found that the “claimant’s story clearly show[ed] that she was a victim of domestic assault and death threats by her former boyfriend,”[158] it held that the claimant “did not exhaust all the avenues of protection available to her before leaving St. Vincent”[159] and thus could not demonstrate a failure of state protection. Noting that the claimant was “an astute, intelligent, young woman who ha[d] demonstrated that she [was] ... capable of seeking out information and support, if need be,”[160] the board concluded that

the claimant will be able to seek out the protection if and when she needs and that the protection available to her will be adequate, if not perfect. She will not be one of those victims of domestic violence who will be either reluctant to press charges or not seek out support from the existing facilities like the National Council of Women. The panel is satisfied that the rationale for the claimant not to have utilised all the resources available to her in seeking protection--namely, her lack of knowledge due to her age and inexperience--is no longer valid now.[161]

The board denied the claim based on the additional findings that the claimant’s former boyfriend did not have sufficient influence to interfere with law authorities as she had alleged, that his abusive behaviour stemmed from his drinking habits and was thus not necessarily directed only at her, and that she could procure financial assistance from relatives abroad to secure legal help.

The case of Re T.D.D. offers another example in which refugee status was denied to a claimant who did not fit the mould of a “victim of culture”.[162] It involved a claimant from India who alleged a well-founded fear of persecution on the grounds of her political opinion as well as her membership in the particular social group “women without male support.”[163] As in the cases above, the claimant was denied refugee status because her perceived capacity for independence militated against a finding of persecution.[164] The IRB reasoned that, as a “well-educated career woman, a witness who appeared articulate and self-possessed at her hearing ... [and] a wife who was unapologetically outspoken in her objections to her husband’s smoking and drinking,” it was unlikely that the claimant would be “completely passive and subordinate” as she had alleged.[165] It further found that the Guidelines were “of no assistance” in the case since the claimant, having a master’s degree, teaching experience, and supportive male family members, was unlike the traditional stereotype of an Indian woman and was well qualified to live without her husband and support her children.[166] Here, the claimant’s educational background prevented her from satisfying acknowledged cultural stereotypes and precluded her recognition as a refugee.[167]

These cases show the problems of linking a finding of persecution to a script of inherent cultural vulnerability. Such an approach effectively requires women to narrate their experience of persecution in terms of that vulnerability; but for some women, this is not their story. This approach has consequences that are potentially quite damaging: it perpetuates the faulty assumption that educated or self-sufficient women who do not appear to be “victims of culture” are less likely to be persecuted by domestic violence. By adopting this approach, the domestic violence cases establish incorrect criteria by which to assess women’s refugee claims. They risk excluding from protection women who face real threats of persecution but cannot authentically narrate their experience of violence through the script of cultural vulnerability.[168] The above analysis thus shows that while women persecuted by domestic violence are still being admitted as refugees, the criteria established for their admission are more limiting than they should be. The domestic violence cases remain insufficiently attentive to the conditions that make women vulnerable to persecution in domestic settings.[169]


The fact is that, over almost two decades and in a large numbers, adjudicators adopted such different methods of analysis in domestic violence cases as compared with other gender cases. These differences can be attributed to a range of factors, such as evidentiary hurdles and varying grant rates between IRB adjudicators,[170] to list just a few. But in addition to these factors, the differences can also be attributed to the perceived distinction between the types of claims at issue. Forced sterilization and genital cutting are generally regarded as exotic harms that can easily be labelled as rights violations in the Canadian legal lexicon and are thus broadly imagined as the proper subjects of refugee law. Because forced sterilization and genital cutting are perceived to occur only in foreign countries, Canada can assert its ability to protect women from these harms with unambiguous certainty. In contrast, domestic violence, rape, and sexual assault are familiar harms that are also clearly prevalent within Canada. These claims cut along familiar lines of difference and cannot be labelled as persecutory rights violations with quite the same rhetorical ease. Since protecting women from domestic violence poses a challenge to even the most well-intentioned of states, as Melanie Randall argues, domestic violence cases create “an unacknowledged dilemma for refugee-receiving states like Canada.”[171] Indeed, despite the availability of state protection in Canada, many Canadian women—particularly Aboriginal women—suffer severe violence in domestic settings.[172] Given the prevalence of domestic violence within Canada, can Canada conclusively assert its ability to protect victims of domestic violence from harm? If the ability to protect is indeed the test not only of a subject’s status as a refugee but also of Canada’s status as provider of refuge, what happens in cases where Canada’s ability to protect may not be clear? Can Canada still assert its value—and “essence”—as a place that protects if it cannot fully protect women from domestic violence?

Audrey Macklin’s analysis of this dynamic correctly summarizes its complexity. As Macklin argues, cases involving familiar harms and familiar scenarios require adjudicators to veer “perilously close to confronting the fact that the same country that has won deserved praise for enacting the Guidelines is also implicated in practices that amount to gender persecution.”[173] The adjudicative response to this is a defensive one: the careful exercise of characterizing the violence suffered by refugee women as a product of culture operates as a protective device that distinguishes it from the violence suffered by Canadian women. By locating persecution in cultural difference, adjudicators subtly sidestep the possibility that, despite Canada’s clear commitment to gender equality and rights protection, women in Canada still suffer persecution. Locating persecution in cultural difference allows Canadian legal actors to avoid the uncomfortable conclusion that the violence suffered by refugee women echoes, in Martha Minow’s words, “something familiar, in reality or metaphor, in the practices of the dominant Western nations” and in “gender hierarchies only too familiar in their own world.”[174] The assumption that domestic violence is a product of foreign cultures—rather than a combination of power structures, material disparities, and cultural arrangements—subtly colours how Canadian legal actors think about, adjudicate, and evaluate domestic violence claims. This assumption makes it harder to recognize domestic violence as a rights violation in the absence of “culture”, not just within but also beyond the confines of refugee law. In this key respect, the adjudicative tendency to locate domestic violence persecution in cultural difference operates to suppress the commonality of domestic violence across cultures and to elide its domestic prevalence.[175]

Beyond establishing incorrect criteria by which to understand domestic violence, the IRB’s domestic violence cases also suggest that the decisions made in this field of law do not merely reflect universally accepted human rights standards. Adjudicators do not assess refugee claims based solely on rights criteria or universally accepted norms. Rather, they filter these criteria through national ideologies, ideals, and anxieties. This pattern calls into question the claim that, by incorporating human rights principles into the refugee analysis, Canadian refugee law has moved away from politically motivated considerations. Instead, it implies that the process is deeply intertwined with Canadian national values and Canada’s national self-understanding.

Recognizing the choices made in the domestic violence cases as informed by a myriad of unstated assumptions, political ideologies, and defensive anxieties offers a more accurate account of the complex dynamics of refugee protection. It makes clear that the findings in these cases do not always reflect universally accepted or “objective” criteria. This realization makes it easier to deconstruct and reconstruct these findings to reflect more accurate understandings of gender violence and gender difference. It allows us to recognize “persecution” as an open-ended category that acquires content and meaning not through some external morality but through our own perceptions, positionings, and beliefs. It allows us to recognize that “culture” is not static, innate, or immutable, but fluid and discursively produced. It allows us to recognize domestic violence as a complex and globally pervasive phenomenon that stems from cultural dynamics—as well as extant power dynamics, material disparities, and gender imbalances that challenge even rights-protecting states like Canada. A more accurate understanding of the dynamics of domestic violence persecution makes it possible to advocate for a shift in the discourse, for an approach that moves beyond its current limits.