Unlocking CEDAW's Transformative Potential: Asylum Cases Before the Committee on the Elimination of Discrimination Against Women

Abstract

One of the most important developments in international law for the protection of displaced women and girls—the implied non-refoulement obligation in the Convention on the Elimination of Discrimination Against Women has received little scholarly or jurisprudential attention. This Article presents, for the first time, a doctrinal analysis of the full corpus of asylum complaints decided by the Committee on the Elimination of Discrimination Against Women (Committee). It analyzes the early development of a prohibition on return to real, personal, and foreseeable risks of serious forms of discrimination against women; highlights critical shortcomings in the Committee's asylum practice; identifies key areas for improvement; and reflects on the Committee's as-yet-unrealized potential to contribute to international law in this area.

I. Introduction

In 1991, this Journal published a seminal treatise on Feminist Approaches to International Law by Hilary Charlesworth, Christine Chinkin, and Shelley Wright, challenging the purported neutrality and objectivity of international law.¹ Unlike its domestic counterpart, which had already been subject to feminist critique, the authors observed the international legal order to be “virtually impervious to the voices of women.”² Women were silenced and subjugated by the organizational and normative structures of international law that privileged men. The authors identified as core issues the exclusion of women from international power structures, the dominance of (masculine) state interests over respect for human dignity, and the weaponization of the public/private dichotomy to assign “women's issues” a lesser or marginal status. To address these phenomena, they called for “further study of traditional areas of international law from a perspective that regards gender as important,” with the hope that such study might precipitate a challenge to the androcentric nature of modern international law.³

Within the field of international refugee law, this study had already begun. Throughout the 1980s, especially, feminist scholars and advocates had challenged the omission of sex or gender⁴ from the list of protected characteristics in the definition of “refugee” in Article 1A of the 1951 Convention Relating to the Status of Refugees (Refugee Convention).⁵ This definition was “heavily weighted towards the experience of public actors: those who participate in big ‘P’ political activities and who join group activities,” rendering refugee protection “a much better fit for men” than for women fleeing “private” harms such as domestic or sexual violence.⁶ Galvanized by the rise of feminist approaches to international law in the early 1990s, a critical discourse continued over the coming decades to explore how violence and exclusion—both in life and in law—oppressed displaced women and girls. This scholarship focused, disproportionately, on the challenges manifest in Global North refugee status determination (RSD) procedures. Thus, scholarship has critiqued the difficulties in “fitting” women's experiences within the “particular social group” category and satisfying the nexus requirement of the refugee definition,⁷ the persistent devaluing of gendered or “private” harms,⁸ the tendency to typecast all “refugee women” as a homogenous category of passive victims while overlooking how other forms of oppression intersect with gender,⁹ and other procedural challenges faced by women.¹⁰ New guidelines on gender-related persecution were published by the UN High Commissioner for Refugees (UNHCR) in 2002,¹¹ and legal and administrative frameworks governing the processing of women's protection claims evolved in some jurisdictions.¹²

Since that time, the highpoint of critical feminist engagement seems to have passed, and progress toward implementing the gender gains of the previous era has “stalled.”¹³ Perhaps as a result, a crucial normative development has gone largely unremarked. In July 2013, in its Views on an individual complaint, the UN Committee on the Elimination of Discrimination Against Women (Committee) read an implied non-refoulement obligation into the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW).¹⁴ It affirmed that Article 2(d) of CEDAW requires states parties “to protect women from being exposed to a real, personal and foreseeable risk of serious forms of gender-based violence, irrespective of whether such consequences would take place outside the territorial boundaries of the sending State party,”¹⁵ and upheld this finding the following year in General Recommendation No. 32 on the Gender-Related Dimensions of Refugee Status, Asylum, Nationality and Statelessness of Women (General Recommendation 32).¹⁶ In doing so, the Committee affirmed that serious forms of gender-based discrimination are an affront to human dignity warranting the same protective measures as other serious human rights violations.

Given that 189 of the 193 UN member states are states parties to CEDAW,¹⁷ this protection should have addressed many of the outstanding challenges identified in refugee scholarship. It is extraordinary, then, that this major development received almost no scholarly attention, either at the time or since. Catherine Briddick is one of the few scholars to consider the Committee's work in the migration context. In a recent article, she compared how the Committee deals with complaints within and outside of the migration context, citing a selection of the Committee's case law to evidence a disparity between how it deals with migration and other cases.¹⁸ However, she does not comprehensively review the Committee's asylum jurisprudence. Başak Çalı, Cathryn Costello, and Stewart Cunningham have undertaken a more rigorous review of asylum case law, but the Committee occupies only a small part of their study.¹⁹ Apart from these analyses, the Committee's work is largely absent from refugee law scholarship.²⁰ This marginalization of CEDAW within international refugee law is striking, and is replicated in international human rights law more broadly. As Loveday Hodson observed in 2014:

In light of concerns expressed by feminists about the silencing of women's voices in international law, one might well expect the jurisprudence and working methods of the Women's Committee to be of interest to a number and range of international legal scholars; in practice, however, its work has failed to generate a great deal of excitement or debate.²¹

Conversely, the relatively limited body of scholarship which does focus on CEDAW contains little analysis of the Committee's asylum work in particular. This lack of scholarly engagement is echoed in a dearth of judicial interpretation of CEDAW within national jurisdictions.²²

The result is a critical gap in our understanding of how CEDAW could and should affirm the rights of women fleeing gender-based harm. To address this gap, this Article undertakes a doctrinal analysis of the Committee's asylum jurisprudence “from a perspective that regards gender as important.”²³ A search of the jurisprudence database of the Office of the High Commissioner for Human Rights (OHCHR) produced 116 admissibility and merits decisions of the Committee between 2006 and 2022, of which thirty-six involved women seeking asylum.²⁴ For completeness, an important additional case decided in 2023 but not yet available through the OHCHR database was added to the analysis. A unique database of these thirty-seven asylum cases was compiled for close textual and doctrinal analysis.²⁵ This analysis produced, for the first time, a holistic view of the full corpus of the Committee's asylum jurisprudence. All thirty-seven communications involved women fearing return to gender-based violence (GBV).²⁶ Some also alleged that their rights under CEDAW had been violated during asylum proceedings in the respondent states, being Denmark (twenty-six), Switzerland (four), Canada (three), the Netherlands (two), and the United Kingdom (two). Twenty-nine of the thirty-seven cases were ruled inadmissible, while only three of the eight cases that proceeded to a decision on the merits were found to involve a violation of CEDAW.

Briddick's comparative analysis of the Committee's migration and other work concluded that its scrutiny of asylum cases was “much more cursory,”²⁷ and involved a “significantly wider” margin of appreciation to states which was “over-wide, characteristic not of appropriate (quasi) judicial restraint, but unprincipled deference.”²⁸ The present research generally supports Briddick's findings, but also identifies a shift in the Committee's approach to asylum cases depending on when they were decided. The reading of a non-refoulement obligation into CEDAW in 2013 was a significant and progressive normative development, and in the immediate aftermath the Committee's Views reflect an attempt to grapple with its new role in determining asylum cases (with mixed results). Then, from late 2016, a radical shift in approach becomes evident in the Committee's asylum work. With limited exceptions, most of the Views in this period meet Briddick's description of “unprincipled deference,” with the Committee deferring excessively to decisions made by national authorities without interrogating apparently discriminatory submissions and possible misinterpretations of CEDAW.

The aims of this Article are threefold: to draw attention to the implied non-refoulement obligation in CEDAW and its implications for international protection; to critique recent failures of the Committee to adequately monitor implementation of CEDAW in cases involving asylum seeker women; and to resuscitate the protective potential of the Committee's contribution to national and international decision making on gendered asylum claims. Part II provides some preliminary comments on the mandate and working methods of the Committee, before Part III introduces the CEDAW non-refoulement obligation and maps it against parallel obligations under international refugee and human rights law. Parts IV and V then set out a close doctrinal analysis of different periods of the Committee's asylum jurisprudence. Part VI looks ahead to the Committee's future asylum work, identifying four key areas for improvement: the quality of the Committee's decision making and Views, the extent of its deference to national decision making, its approach to credibility issues, and its response to the unique challenges of the asylum context. Part VII concludes by reflecting on how a more adventurous approach to the Committee's complaints mechanism might help it to challenge entrenched normative structures which perpetuate the subjugation of displaced women's rights and experiences to those of men.

II. The Committee and Its Working Methods

Upon adoption in 1979, CEDAW was heralded as “the international bill of rights for women.”²⁹ It initially focused on achieving equal opportunities and rights for women, before a milestone development in 1992 saw its remit expand to include protection against GBV, which was conceptualized as “a form of discrimination that seriously inhibits women's ability to enjoy rights and freedoms on a basis of equality with men.”³⁰ CEDAW also established the Committee as the third of now ten UN treaty bodies (UNTBs) within the human rights system. It was empowered to receive and review periodic reports from states parties and make general recommendations on the interpretation and implementation of CEDAW.³¹ However, it was not established on an equal basis as other treaty bodies and battled marginalization within the human rights system from its inception. Described as the “poor cousin”³² of other UNTBs, the Committee suffered from “a particularly disadvantaged position” due to inadequate resourcing and secretariat support, geographical and administrative separation from other treaty bodies, and, critically, the lack of an individual complaints mechanism.³³

This last deficiency was remedied in 1999 with the adoption of the Optional Protocol.³⁴ Its entry into force in December 2000 enlivened the first gender-specific complaints mechanism within the international human rights system, providing the Committee “an enhanced opportunity . . . to discover its voice.”³⁵ The Committee became empowered to receive and consider complaints from any individual or group of individuals under the jurisdiction of a state party who claimed to have suffered a violation of CEDAW.³⁶ While it has certain fact-finding powers, including the ability to request that states parties and authors submit additional written explanations or statements relevant to issues arising in a case,³⁷ the Committee does not hold hearings or receive oral testimony from the parties. Instead, it relies on written submissions to determine whether a violation of CEDAW has occurred, and issues findings in the form of “Views.”

The Committee's Views resemble those of other UNTBs. First, they summarize the submissions of the parties in relatively comprehensive detail. Since the original submissions are generally not made public, these summaries often provide the only insight into the allegations of fact, evidenced adduced, and legal arguments made in each case. The Committee then determines whether a communication is admissible (pursuant to the criteria in Articles 2 to 4 of the Optional Protocol) before proceeding to examine the merits. By contrast to the submission summaries, the actual findings on admissibility and merits tend to be brief. They can be superficial, opaque, inconsistent, and formulaic, and may fail to engage with all relevant submissions, clarify findings of fact and law, or provide full and transparent reasons for findings.³⁸

In light of these critiques, it is worth reflecting on how the Committee's Views are formulated. The individual complaints mechanisms for various treaties, including CEDAW, are administered by the UN Secretariat—specifically, the “petitions unit” within the Human Rights Treaties Division at OHCHR. Ibrahim Salama, the former director of this Division, describes it as “essential” to the work of the UNTBs and “acting as a built-in coherence vector, as well as the institutional memory of the jurisprudence under the different treaties.”³⁹ Yet publicly available information about specifically how the Secretariat acts as a “coherence vector” is sparse.⁴⁰ If the Secretariat prepares draft Views, a practice which has been critiqued in other contexts,⁴¹ this involvement is not reflected in the Views themselves or the Committee's published working methods.

What is known is that a Working Group on Communications comprised of five Committee members leads the drafting process for the Committee.⁴² One Working Group member is appointed as case rapporteur for each registered communication to “examine all information contained in the case file, carry out the necessary research and propose to the Working Group the course of action that she or he considers appropriate, including recommendations on admissibility and merits.”⁴³ The Working Group considers the case rapporteur's draft Views and tries to reach consensus, but if consensus is not possible the majority Views are sent to the Plenary for discussion.⁴⁴ The Committee holds “closed meetings” when examining communications,⁴⁵ and final Views are published under the name of the Committee as a whole, listing the names of all individual members who “participated in the examination of the present communication.” The Views themselves do not identify the case rapporteur or Working Group members for a given communication. Committee members who disagree with the majority have the option of appending individual opinions, but in practice this is rarely done.⁴⁶ Absent individual opinions, the published Views may be “purged . . . of controversial matter on which members differ,”⁴⁷ providing no indication of whether the Working Group reached consensus or the length, level and subject of any debate that informed the decision at the Plenary level.

The legal or normative status of these Views, like those of the other UNTBs, is contested.⁴⁸ To borrow from the context of the Human Rights Committee, they “exhibit some of the principal characteristics of a judicial decision” and are “arrived at in a judicial spirit.”⁴⁹ They represent an “authoritative”⁵⁰ interpretation of CEDAW and, according to their proponents, should be ascribed “great weight” in order to “achieve the necessary clarity and the essential consistency of international law, as well as legal security, to which both the individuals with guaranteed rights and the States obliged to comply with treaty obligations are entitled.”⁵¹ Yet the Committee is not a court, indeed its members may not even have legal qualifications, and its Views lack express binding force.⁵² Views are often described as “quasi-judicial,”⁵³ although the meaning of this term may be disputed.⁵⁴ Rather than seek to resolve these disputes, the subsequent analysis focuses on the overall quality and effectiveness of the Committee's Views, as determined by criteria such as their coherence, consistency, predictability, rigor of analysis, and persuasiveness or ability to shape normative developments at the national and international levels.⁵⁵

III. Reading an Implied Non-refoulement Obligation into CEDAW

The Committee began to consider the issue of returning asylum seeker women to gender-based harm in a substantive way in a series of cases communicated to it from 2010 onward.⁵⁶ The respondent states—Denmark, Canada, and the Netherlands—rejected what they described as attempts to apply CEDAW “extraterritorially.”⁵⁷ They argued that their obligations under CEDAW did not extend to discrimination occurring elsewhere, and that the treaty did not contain a positive non-refoulement obligation preventing removal to gender-based harm. Denmark was particularly staunch in its resistance to Committee interference in its asylum system, arguing that “wholly exceptional circumstances” were required to prevent removal.⁵⁸ It submitted that deportation to risks of arbitrary deprivation of life, torture and cruel, inhuman, or degrading treatment or punishment [hereinafter ill-treatment] might engage its responsibility, but that the UN Human Rights Committee had “never considered a complaint on its merits regarding deportation of a person who feared a lesser human rights violation in the receiving State (e.g., violation of a derogable right) by the receiving State.”⁵⁹ Thus, according to Denmark, “the returning of a woman who arrives in Denmark simply to escape from discriminatory treatment in her own country, however objectionable that treatment may be, cannot constitute a violation of the Convention.”⁶⁰ These submissions suggested that there was a “lesser” and “derogable” right to freedom from discrimination which was not “serious”⁶¹ enough to engage the responsibility of a state returning a woman to such treatment.

In July 2013, in groundbreaking Views in M.N.N., the Committee definitively rejected these submissions and reaffirmed GBV as a form of discrimination against women which impairs or nullifies the enjoyment of all other rights.⁶² It noted that both international human rights and refugee law prohibit return to serious violations of human rights or persecution, including “forms of persecution that are directed against a woman because she is a woman or that affect women disproportionately.”⁶³ In doing so, it affirmed that gender and gender-related harm are relevant considerations in the determination of claims for protection under the Refugee Convention, Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (Torture Convention)⁶⁴ and International Covenant on Civil and Political Rights (ICCPR).⁶⁵ Indeed, as the Committee reiterated the following year in General Recommendation 32, CEDAW is an important “complement” to the legal protections for displaced women and girls under other treaties because they lack explicit references to gender equality or gender-based persecution and “understanding the way in which women's rights are violated is critical to the identification of those forms of persecution.”⁶⁶

But CEDAW's normative value was not limited to promoting a gender-sensitive interpretation of the non-refoulement obligations under other treaties. The Committee also read it as the source of an independent prohibition on refoulement. Article 2(d) of CEDAW requires states parties “to refrain from engaging in any act or practice of discrimination against women and to ensure that public authorities and institutions shall act in conformity with this obligation.” According to the Committee:

This positive duty encompasses the obligation of States parties to protect women from being exposed to a real, personal and foreseeable risk of serious forms of gender-based violence, irrespective of whether such consequences would take place outside the territorial boundaries of the sending State party: if a State party takes a decision relating to a person within its jurisdiction, and the necessary and foreseeable consequence is that that person's rights under the Convention will be violated in another jurisdiction, the State party itself may be in violation of the Convention.⁶⁷

It affirmed that “[w]hat amounts to serious forms of gender-based violence will depend on the circumstances of each case.”⁶⁸

The Committee expanded on this implied obligation in General Recommendation 32. It affirmed that Article 2(d) obliges states parties to protect women from being exposed to a real, personal, and foreseeable risk of serious forms of discrimination against women, including (but not limited to) GBV.⁶⁹ Specifically, the Committee stated that:

States parties have an obligation to ensure that no woman will be expelled or returned to another State where her life, physical integrity, liberty and security of person would be threatened, or where she would risk suffering serious forms of discrimination, including serious forms of gender-based persecution or gender-based violence.⁷⁰

The Committee's use of the conjunction “or” suggested that the non-refoulement obligation under CEDAW had two limbs: a first, prohibiting return to a place where a woman's “life, physical integrity, liberty and security of person would be threatened”; and a second, prohibiting return to a “risk [of] suffering serious forms of discrimination, including serious forms of gender-based persecution or gender-based violence.” However, the Committee has not subsequently articulated an understanding of the obligation as having two limbs, and instead has focused simply on whether there is a real, personal, and foreseeable risk of serious forms of gender-based persecution or violence.⁷¹

Initially, after M.N.N., Denmark sought to limit the impact of the Committee's Views by requiring active state involvement in harm perpetrated by non-state actors. It proposed two different standards. First, with reference to the jurisprudence of the Committee Against Torture, Denmark argued that CEDAW did “not encompass an obligation for States parties to refrain from expelling a person who might risk pain or suffering inflicted by a private person, without the consent or acquiescence of the State authorities.”⁷² Second, with reference to the jurisprudence of the European Court of Human Rights (ECtHR), Denmark argued that a woman needed to show that authorities in her home state were “unable to obviate the risk by providing appropriate protection.”⁷³ There was a critical difference between these two standards. “[C]onsent or acquiescence” by the state is a significantly higher threshold which is specific to the torture context.⁷⁴ By contrast, inability (or willingness) to obviate risk is the more common standard in other contexts. For example, UNHCR's guidelines on gender-related persecution state that: “serious discriminatory or other offensive acts committed by the local populace, or by individuals, can . . . be considered persecution if such acts are knowingly tolerated by the authorities, or if the authorities refuse, or are unable, to offer effective protection.”⁷⁵

In General Recommendation 32, the Committee affirmed that harm perpetrated by non-state actors “is persecution where the State is unable or unwilling to prevent such harm or protect the claimant because of discriminatory governmental policies or practices.”⁷⁶ This standard definitively and correctly rejected the “consent or acquiescence” threshold. However, it also appeared more stringent than the traditional refugee law standard by requiring that the state's inability or unwillingness be attributed to discriminatory policies or practices. Since the Committee drew this test from UNHCR's guidelines, any difference between the two standards should be put down to error or imprecision rather than a deliberate attempt to establish a higher threshold. The language of General Recommendation 32 should have replicated the UNHCR standard exactly by providing that serious forms of discrimination against women can engage CEDAW's implied non-refoulement obligation if the authorities “knowingly tolerate” them, or “refuse or are unable” to offer effective protection against them, without the need for such refusal or inability to be linked to a discriminatory state practice. Clarification on this point is critical, since the extent of a state's tolerance of harm perpetrated against women by non-state actors is central to understanding the exact contours of the implied non-refoulement obligation.

The relationship between the non-refoulement obligation read into CEDAW and those found under other treaties also warrants further attention. For states that are parties to CEDAW but not the Refugee Convention, Torture Convention, and/or ICCPR, the reading of a non-refoulement obligation into CEDAW is particularly significant.⁷⁷ It either establishes or expands those states’ treaty-based obligations.⁷⁸ By contrast, for the states already bound by the prohibitions on refoulement under international refugee and human rights law, the relative scope of the CEDAW obligation requires clarification to determine whether it adds anything “new.” The Committee did not directly address this question at the time, nor has it done so since. This omission is unfortunate, since mapping the CEDAW non-refoulement obligation against those established under other treaties is essential to determining the scope of state obligations and potential value of CEDAW to displaced women and girls. In the absence of authoritative guidance from the Committee, this Article offers some preliminary reflections.⁷⁹

Turning first to the Refugee Convention, Article 33 prohibits return of a “refugee” to any place “where his [sic] life or freedom would be threatened on account of his [sic] race, religion, nationality, membership of a particular social group or political opinion.” This protection covers people who meet the definition of a “refugee” in Article 1(A) of the Refugee Convention as modified by the 1967 Protocol, irrespective of whether or not they have been formally recognized as such.⁸⁰ The question then becomes whether all women who fear serious forms of discrimination (such as to engage the implied non-refoulement obligation in CEDAW), also constitute “refugees” such as to engage the protection of Article 33 of the Refugee Convention?⁸¹ Or are there some women who fall outside the scope of the Refugee Convention but might nevertheless be entitled to protection against refoulement under CEDAW?

It is “widely accepted that [gender] can influence, or dictate, the type of persecution or harm suffered and the reasons for this treatment,” and therefore that “[t]he refugee definition, properly interpreted, . . . covers gender-related claims.”⁸² But this general position warrants deeper interrogation when comparing the scopes of protection against refoulement under CEDAW and the Refugee Convention respectively. Central to the refugee definition is the concept of “persecution.” Despite being a somewhat “indeterminate” term,⁸³ persecution is generally understood as encompassing acts which are sufficiently serious by their nature or repetition as to constitute severe violations of human rights, including but not limited to torture and threats to life or freedom.⁸⁴ Additionally, “a pattern of discrimination or less favourable treatment could, on cumulative grounds, amount to persecution and warrant international protection,” in particular where discriminatory measures lead to “consequences of a substantially prejudicial nature for the person concerned.”⁸⁵ Since serious forms of discrimination against women impair or nullify the enjoyment of all other rights, they are likely to reach the severity threshold for “persecution,” meaning CEDAW and the Refugee Convention offer protection against return to comparable harms.⁸⁶

However, even when reaching a sufficient level of severity to constitute “persecution,” such harm is insufficient on its own to ground a refugee claim. In addition to an absence of state protection, it is also necessary to establish a causal link between the relevant harm and at least one of the five grounds set out in Article 1(A). Women (or a subset of women) may constitute a “particular social group” in certain circumstances, and/or may base a refugee claim on any of the other four grounds (race, religion, nationality, and political opinion).⁸⁷ The relevant ground “need not be the sole, or even the dominant cause of the risk of being persecuted, but it must be a contributing cause to the risk.”⁸⁸ This causal requirement may pose procedural and substantive difficulties for women, especially those fleeing “non-exotic” harms such as domestic violence.⁸⁹ Even when the five grounds are given a gender-sensitive interpretation, they create an additional obstacle that is absent from the CEDAW formulation. For example, under CEDAW, a woman fleeing domestic violence need only prove that the risk she fears is serious, real, personal, and foreseeable, and that the state of origin is unable or unwilling to provide effective protection. She need not establish herself within a particular social group, or otherwise prove a perpetrator's motivations by reference to one of the other grounds for persecution in the Refugee Convention. This difference renders the CEDAW non-refoulement obligation broader in its application than that contained in Article 33 of the Refugee Convention. It should, in theory, mean that some women who do not meet the definition of a refugee such as to engage the full range of protections set out in the Refugee Convention may nevertheless benefit at least from protection against refoulement under CEDAW.⁹⁰

Turning to map the CEDAW prohibition on refoulement against comparable provisions of international human rights law, it is obviously broader than that contained in Article 3 of the Torture Convention, since it does not require the relevant harm to be committed “by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity.”⁹¹ However, its relationship with the implied obligation under the ICCPR is less clear. Throughout the 1990s, the Human Rights Committee began reading an implied non-refoulement obligation into Article 2 of the ICCPR and affirmed that:

if a State party takes a decision relating to a person within its jurisdiction, and the necessary and foreseeable consequence is that that person's rights under the Covenant will be violated in another jurisdiction, the State party itself may be in violation of the Covenant.⁹²

The Human Rights Committee later clarified that this obligation prevents removal “where there are substantial grounds for believing that there is a real risk of irreparable harm, such as that contemplated by Articles 6 and 7 of the Covenant”⁹³ (arbitrary deprivation of life and torture or cruel, inhuman or degrading treatment or punishment), and that the relevant risk must be both “real and personal.”⁹⁴

In reading a comparable non-refoulement obligation into CEDAW, the Committee's approach was broadly consistent with that of the Human Rights Committee, particularly with respect to the standard of proof and need for a “real, personal and foreseeable risk.”⁹⁵ But there is some ambiguity around the respective scopes of the harm to which return is prohibited. Specifically, is the CEDAW prohibition on return to serious forms of discrimination against women coterminous with, or broader than, the ICCPR prohibition on return to “irreparable harm”? Some cases will fall squarely within the scope of both, such as those involving serious GBV which also violates Articles 6 and/or 7 of the ICCPR.⁹⁶ However, for other cases involving forms of discrimination against women which have not traditionally been recognized as reaching the Article 6 and/or 7 threshold (such as general, non-violent measures of gender inequality), some greater clarity about the relationship between the prohibitions on refoulement under the ICCPR and CEDAW would be helpful.

Unfortunately, the Committee has provided no such clarity, and a comparison between the two is complicated by several factors. First, the scope of the “irreparable harm” concept in the ICCPR formulation remains open to debate. It is unclear whether, and if so in what circumstances, violation of a right other than Article 6 or 7 of the ICCPR might amount to irreparable harm. Scholars and international bodies have argued that “the list of rights giving rise to complementary protection is not closed,”⁹⁷ with the Human Rights Committee in particular adopting an “evolutive approach” to the concept (including through emerging jurisprudence on return to socioeconomic deprivation).⁹⁸ Nevertheless, to date the Human Rights Committee has only ever found a violation of Article 6 or 7 to reach the threshold of “irreparable harm.”⁹⁹ Whether this concept remains constrained in this way, or is expanded through further evolutions in the Human Rights Committee's work, will be key to determining whether the ICCPR non-refoulement obligation covers the same ground as CEDAW.¹⁰⁰

A second complicating factor is that while the CEDAW non-refoulement obligation is articulated as encompassing “serious forms of discrimination” broadly, all of the individual complaints to the Committee have involved fear of return to GBV specifically, and General Recommendation 32 does not articulate any examples of non-violent discrimination which might prevent return. Thus, it remains to be seen whether women can rely on CEDAW to challenge return to countries where they do not necessarily fear physical violence, but would face serious forms of discrimination with respect to, for example, education, employment, inheritance, property rights, health and reproductive rights, family law, or custody rights.¹⁰¹ If CEDAW is interpreted as preventing return to non-violent discrimination (as the Committee's language would suggest), further clarification is required as to the standard at which discrimination reaches the “serious” threshold such as to enliven the non-refoulement obligation. Until then, the extent to which the non-refoulement obligation implied into CEDAW (potentially) offers a broader protection than that which is implied into ICCPR remains unclear.

IV. Asylum Decision Making in the Immediate Post-M.N.N. Period (2013–2016)

A. Promoting Gender-Sensitive Asylum Processes

The Committee's Views in M.N.N. marked the beginning of a new phase in which the relevance of CEDAW to asylum decision making had been affirmed, but its application to specific cases remained in its infancy. Over the following years, the Committee sought to clarify this application—with mixed results. A decade earlier, UNHCR had already called on states parties to the Refugee Convention to adopt “gender-sensitive” approaches to women's asylum claims.¹⁰² “Gender-sensitivity,” in this sense, meant at a minimum ensuring that the refugee definition was not always interpreted through the framework of male experiences.¹⁰³ Other treaty bodies had also recognized that proper attention to gender was critical to ensuring respect for the fundamental human rights of all people, including women.¹⁰⁴ But a complete understanding of what gender-sensitivity entailed in the asylum context was yet to be elucidated when the Committee began to emerge as a body which might contribute to the normative understanding of this issue.

As a preliminary matter, the Committee affirmed that states need to be alert to the possibility of discrimination in asylum claims even if they are not explicitly framed in such terms. This position originated in a joint dissenting opinion in M.E.N., in July 2013. The author, a political dissident, had been gang-raped by three unknown men as she fled an attack on her town in Burundi, but had not gone to the police because “she was trying to hide from them.”¹⁰⁵ If returned to Burundi, she feared “being imprisoned and raped while in custody by prison guards and being unable to report such crimes to the police because of the prevailing impunity for rape perpetrators.”¹⁰⁶ Denmark argued that the author had failed to exhaust domestic remedies because she had not raised allegations of gender discrimination during her asylum proceedings.¹⁰⁷ A majority of the Committee agreed, but a strong joint dissenting opinion led by Committee member Dubravka Šimonović held otherwise. It argued that the pertinent question for admissibility was “whether the author raised sex-based discrimination, whether intersecting other grounds of persecution or alone, as the basis for her claim during the asylum procedure.”¹⁰⁸ Since the author had mentioned her rape during her asylum proceedings, that should have been sufficient to put Denmark on notice that her claim raised allegations of gender-related persecution.¹⁰⁹ According to the joint dissent, a woman must provide the relevant facts to substantiate her claim but “should not be required to make explicit reference to rape as a form of discrimination against women,”¹¹⁰ and “it should not be incumbent upon the asylum seeker to use in her asylum claim words such as ‘discrimination based on sex’ and/or ‘gender-related persecution.’”¹¹¹ This minority view proved influential. When the next case, N., was decided in February 2014, the Committee followed the joint dissent in M.E.N. in rejecting the Netherlands’ submission that the claim should be ruled inadmissible because the author had not explicitly mentioned gender-discrimination and therefore had failed to exhaust domestic remedies.¹¹²

The joint dissent influenced the Committee's approach in other ways, too. It modeled “gender-sensitivity” by insisting that women's asylum claims not be assessed through the lens of male experiences. After affirming that “women's political activity may not always look like male political activity or may not be equally valued in the male-dominated political environment,” the joint dissent concluded that Denmark had “failed to assess in a non-discriminatory manner the risk of the author's future political persecution or gender-related persecution or whether the author could benefit from State protection.”¹¹³ It reiterated that “[g]ender-related claims to asylum may intersect with other proscribed grounds of persecution, such as membership of a particular social group or political opinion,”¹¹⁴ and insisted that states parties “put in place asylum procedural safeguards to ensure that women's claims are properly heard and assessed, especially because women may not know the motivation or reasons for the acts perpetrated against them, or the identity of their perpetrators.”¹¹⁵

The joint dissent also took issue with the characterization of rape in the context of conflict and political dissent. Denmark had argued that there was no evidence that M.E.N.'s rape was related to her political activities and that “at no time during the domestic proceedings did the author assert that the rape was an instance of politically motivated persecution.”¹¹⁶ The majority of the Committee also noted that the author “did not know the identity of the perpetrators, nor why they had raped her in particular.”¹¹⁷ By contrast, the dissent privileged the author's account of why she believed her rape had been politically motivated.¹¹⁸ Describing rape as “the most notorious form of sexual violence directed against women because of their sex or gender” and “a gender-related form of persecution,”¹¹⁹ the dissent concluded that Denmark had failed to “adequately consider the environment surrounding the rape, including impunity for the crime” or to “recognize the rape of the author as a separate or intersecting form of gender-related persecution and sex-based discrimination.”¹²⁰

These aspects of the M.E.N. dissent informed the Committee's guidance in General Recommendation 32. Using almost identical language to the dissent, the Committee called on states to adopt “[a] gender-sensitive approach . . . at every stage of the asylum process” informed by “a thorough understanding of the particular forms of discrimination or persecution and human rights abuses that women experience on grounds of gender or sex.”¹²¹ The General Recommendation affirmed that “violence against women . . . is one of the major forms of persecution experienced by women in the context of refugee status and asylum,”¹²² and expressed concern that “many asylum systems continue to treat the claims of women through the lens of male experiences, which can result in their claims to refugee status not being properly assessed or being rejected.”¹²³ These statements had the potential to have a positive influence on how women's asylum claims were considered by both states and international bodies. However, as discussed in Part V, the Committee subsequently missed opportunities to model a gender-sensitive approach to asylum claims in a later stage of its work.

B. Clarifying the Need to Seek State Protection Against Private Harm

One shortcoming that began to emerge early in the Committee's post-M.N.N. asylum work was the lack of a consistent and principled approach to the question of whether a woman must seek the protection of national authorities before fleeing harm by private actors. This issue was exemplified in the case of N., decided in 2014.¹²⁴ N. was raped by her employer, L., and became pregnant. She reported her assault to the police who questioned L. but released him a few days later. He told N. that “she could not do anything to him because he had money and connections,” and because he was in possession of her passport, birth certificate, and diploma.¹²⁵ The author was locked in sexual slavery and regularly abused until she escaped and again went to the police. They took her statement and photographs of her injuries, but no further action. L. later told the author “that the police had contacted him and that he had bribed them, meaning that they would do nothing to protect her.”¹²⁶ N. eventually fled to the Netherlands, where her asylum claim was rejected on the basis that she “failed to demonstrate that she was unable to apply to the Mongolian authorities for effective protection from Mr. L.'s conduct.”¹²⁷

The Committee's Views in N. were problematic. It concluded that nothing on file indicated “that the Mongolian authorities had in fact acted in bad faith or had failed to react promptly to the author's complaints.”¹²⁸ The Committee claimed that “the author merely states that she fears that the Mongolian authorities would fail to protect her against Mr. L., but provides no further explanation,” and that she “has not explained how, in the past, the Mongolian authorities have failed to protect her in her personal circumstances.”¹²⁹ It took issue with the fact that N. did not “follow up on her complaints with the police” or “complain to the Mongolian prosecuting authorities or courts.”¹³⁰ The Committee appeared to accept the Netherlands’ submissions that the author should have adduced evidence that her employer had bribed the police,¹³¹ without discussing whether such an evidentiary burden was reasonable. It also appeared to accept, unquestioningly, the Netherlands’ conclusion that the mere fact of the police taking minimal steps to record N.'s complaint before releasing her employer to reoffend indicated that they were not unwilling to protect her.¹³²

The refusal to find any indication of a lack of police protection is difficult to reconcile with the author's account (which was deemed credible by the state party¹³³) and with the Committee's own Concluding Observations on Mongolia from the relevant period. In November 2008, contemporaneously with the start of N.'s abuse, the Committee had expressed “deep concern” with the high rates and prevalence of violence against women in Mongolia.¹³⁴ It expressed concern “that domestic violence continues to be seen as private matters, including among the law enforcement personnel” and that prosecutions for domestic violence remained “very low.”¹³⁵ N. cited these Concluding Observations, as well as corroborating evidence about the lack of protection against rape, domestic violence, and sexual harassment in Mongolia, in her complaint.¹³⁶ Yet the Committee made no reference to these sources in its findings. Instead, it noted that Mongolia was a state party to CEDAW, before deciding that “the facts as presented do not permit a conclusion to be reached that there is no effective legal system in Mongolia capable of establishing, prosecuting and sanctioning Mr. L..”¹³⁷

The Committee's findings in N. and another case with comparable facts, S.O.,¹³⁸ were out of step with its own guidance in General Recommendation 32, which was published around the same time. While acknowledging that states have primary responsibility for protecting their citizens, and that “it is only when such protection is not available that international protection is invoked to protect the basic human rights that are seriously at risk,” the Committee also noted that:

the fact that a woman asylum seeker has not sought the protection of the State or made a complaint to the authorities before her departure from her country of origin should not prejudice her asylum claim, especially where violence against women is tolerated or there is a pattern of failure in responding to women's complaints of abuse. It would not be realistic to require her to have sought protection in advance of her flight. She may also lack confidence in the justice system and access to justice or fear abuse, harassment or retaliation for making such complaints.¹³⁹

In line with this guidance, the asylum claims of N. and S.O. should not have been prejudiced by the fact that they did not continue to seek police protection after earlier attempts had failed.

C. The Case of “A.”

The Committee modeled a better approach to gender-related asylum claims in November 2015 in A., the first asylum case in which the Committee found a violation of CEDAW. A. had suffered three attacks in Pakistan. First, a group of men had broken into the beauty salon where she worked, committed vandalism, called the place a “sex clinic” and accused the employees of performing “dirty work.” A few days later, a group of men had broken into her home, accused her of being a prostitute and performing “dirty work,” then “beat her, kicked her, threw inflammable liquid at her and set fire to her clothes, causing severe burns to her torso and arms.”¹⁴⁰ Finally, she was shot at by unknown men on motorbikes while taking a taxi. A. never made any attempt to report her attacks to the police because she had been in hospital and:

did not dare to do so after she was released because she had been informed by her neighbours that, notwithstanding the fact that they had reported the severe attacks to the police, the police had refused to investigate because they considered her to be a prostitute.¹⁴¹

Like the author in M.E.N., A. could not prove who had committed the attacks or why, but cited various possibilities, including that she had married against the will of her husband's family, that she was a religious minority, and/or that she was “a woman who behaved in a way that was not accepted by some sectors of society.”¹⁴² She argued that the attacks “were aimed directly at her because she was a woman who behaved contrary to the established gender roles in Pakistan.”¹⁴³

Denmark accepted the author's accounts of her attacks but was not satisfied that she had proved that her husband's family were behind them or that they had been directed against her personally. It also argued that she had failed to sufficiently substantiate her claim that the Pakistani authorities would be unable or unwilling to provide her with the protection necessary to obviate the risk of GBV.¹⁴⁴ The Committee took a different approach, finding that the author's asylum application had been denied by the Danish authorities “simply” because they were not satisfied that the attacks had been instigated by her husband's family, and that they had given “no due consideration” to other relevant factors, including that the author was “an illiterate ethnic Punjabi of Christian faith with no family support, living in a village in Pakistan away from her husband and being treated as a ‘prostitute’ by society at large, including the police.”¹⁴⁵ The Committee recalled the three attacks against the author and confirmed that “the nature and circumstances of those attacks all indicate that they were targeted at the author and were therefore ‘personal.’”¹⁴⁶ It clarified that “the inability of the author to provide precise information on the exact identity of the persons responsible for the three attacks did not compromise her credibility” and that the denial of her asylum application was therefore “manifestly arbitrary.”¹⁴⁷

With respect to the author's failure to seek police protection, the Committee contextualized her individual circumstances—including the intersecting issues of her gender, ethnicity, religion, level of education, family situation, living arrangements, and perception by the community and police—within the broader setting of prevailing attitudes toward women in her position in Pakistan. On these bases, it rejected Denmark's submission that the author had failed to sufficiently substantiate her claim that the Pakistani authorities would be unable or unwilling to provide her with the protection necessary to obviate the risk of GBV. Instead, citing General Recommendation 32, the Committee took the view:

that the fact that the author did not seek the protection of the State or make a complaint to the authorities before her departure from Pakistan should not have prejudiced her asylum claim, especially taking into account the level of tolerance towards violence against women and the pattern of failure in responding to women's complaints of abuse, which are reflected in the information provided by the author and make it unrealistic to require the author to have sought protection in advance of her flight.¹⁴⁸

The Committee recalled that “gender-related asylum claims may intersect with other proscribed grounds of discrimination, including ethnicity and religion,” and relied on UNHCR guidance that “in many instances, the [Pakistani] authorities are reportedly unable or unwilling” to protect the lives of Christians or bring perpetrators of GBV against Christian women to justice.¹⁴⁹ The Committee also relied on its own Concluding Observations on Pakistan in which it too raised concerns about impunity for violence against women.¹⁵⁰

The Committee's reasoning in A. was set out in considerably more detail than most previous cases, making it possible to understand how it reached its final decision. Ultimately, the Committee concluded that:

the author has been subjected to gender-based violence in Pakistan . . . either because she was a woman living on her own and working at a beauty salon, which was perceived as “immoral” by her community, or because she married against the wishes of her husband's family and her family, or both.¹⁵¹

Satisfied that the author would not receive protection from the authorities in Pakistan, the Committee found that there were “substantial grounds for considering that the author would be exposed to a real, personal and foreseeable risk of serious forms of gender-based violence in case of return to Pakistan,” contrary to Articles 2(c) and (d) of CEDAW.¹⁵² It recommended, inter alia, that Denmark refrain from forcibly returning A. to Pakistan and take all measures necessary to prevent similar violations in the future.

V. A Turn to Passivity in the Committee's Asylum Work

From late 2016, a shift began to emerge in the Committee's asylum work. Whereas previously the Committee had showed some willingness to assert itself as an authority in asylum cases involving discrimination against women, this assertiveness progressively gave way to passivity. The Committee appeared to abandon its progressive agenda and became increasingly unwilling to interfere with national decision making, especially in relation to credibility.

A certain degree of deference to national authorities is an ordinary feature of treaty body jurisprudence,¹⁵³ and had previously been discussed by the Committee in a series of cases involving gender-based myths and misconceptions about women in rape and sexual harassment prosecutions in the Philippines.¹⁵⁴ It first arose in the asylum context in N.Q. and S.A. in February 2016. The United Kingdom argued that “although the Committee has certain fact-finding powers under its rules of procedure, there should be compelling reasons before it uses them.”¹⁵⁵ It invited the Committee to accept its findings of fact and credibility and argued there was “no proper basis on which the fairness of the British legal process [could] be impugned in the present case.”¹⁵⁶ On the merits, the Committee stated that:

it is generally for the courts of the States parties to the Convention to evaluate the facts and evidence or the application of the national law in a particular case, unless it can be established that the evaluation was biased or based on gender stereotypes that constitute discrimination against women, was clearly arbitrary or amounted to a denial of justice.¹⁵⁷

However, it also ruled the complaint admissible and considered in some depth the asylum process.¹⁵⁸ It reviewed the evaluation of facts and evidence by the national authorities and reached its own findings about the author's credibility in light of inconsistencies in her evidence.¹⁵⁹ Ultimately, it concluded that the removal of the author and her husband to Pakistan would not breach CEDAW.

This and the next two cases decided by the Committee in 2016 did not raise overt concerns about the deferential approach.¹⁶⁰ From November 2016 onward, however, the nature and extent of this deference became problematic. The Committee decided nineteen cases between November 2016 and February 2022: seventeen involving the attempted removal of women from Denmark (sixteen) or Switzerland (one), and two involving transfers from Switzerland to other European states pursuant to the Dublin Regulation (discussed separately below). Credibility was a central issue, with both Denmark and Switzerland making potentially gendered adverse credibility findings against the authors that should have been reviewed by the Committee. Some of these cases also raised questions about the correct interpretation of CEDAW and application of its implied non-refoulement obligation. Yet, despite these matters falling squarely within the Committee's mandate and competence, it adopted an unduly deferential approach to the findings of national asylum authorities and failed to provide vital normative guidance on how the issues in dispute should have been resolved consistently with CEDAW. In all but one anomalous case,¹⁶¹ the Committee afforded Denmark and Switzerland an excessively wide margin of appreciation, “characteristic not of appropriate (quasi) judicial restraint, but unprincipled deference.”¹⁶² Key issues received only cursory consideration or were omitted entirely from the Committee's Views, many of which read more as template judgments than carefully reasoned findings based on the individual facts of each case. In formulaic terms, the Committee affirmed that “significant weight” should be given to the asylum decisions of states parties, before concluding—often summarily—that “nothing on file” demonstrated bias, stereotyping, discrimination, arbitrariness, or a denial of justice.

A. Credibility of Evidence About Violence, Abuse, and Torture

During this period of the Committee's asylum work, credibility emerged as a prominent and consistent basis for the rejection of women's asylum claims. Denmark based negative credibility findings on the ways in which women narrated their personal experiences of violence, abuse, and torture, concluding that their stories were “unlikely to have occurred,”¹⁶³ “fabricated” for the occasion,¹⁶⁴ and “not based on . . . personal experiences,”¹⁶⁵ noting areas where they were “vague,”¹⁶⁶ “evasive,”¹⁶⁷ “sketchy,”¹⁶⁸ “elaborative,”¹⁶⁹ “implausible,”¹⁷⁰ “incoherent,”¹⁷¹ “inconsistent,”¹⁷² or “lack[ing] inner logic.”¹⁷³ Switzerland, by contrast, did not critique the way the author in L.O. told her story, but rather privileged “credentialed knowledge”¹⁷⁴ from a legal report over her personal testimony. Both approaches should have alerted the Committee to the possibility that “rape mythology,” and “victim fabrication myths” in particular, were influencing national decision making.¹⁷⁵ They should have prompted a close examination of each credibility finding to ensure they had sound evidentiary bases, free from gender stereotypes or discriminatory assumptions about women. Had such an examination occurred, the Committee would have uncovered three areas of concern: credibility findings based on (1) the timing or manner of disclosure of GBV; (2) assumptions about women victims and male perpetrators; and (3) inconsistencies in evidence.

1. Timing or Manner of Disclosing Serious GBV

Denmark based a series of negative credibility findings on the timing or manner in which women disclosed GBV to asylum authorities. In E.W., Denmark cited the author's failure to disclose her alleged torture and sexual abuse at an initial interview as a factor undermining her credibility.¹⁷⁶ Similarly, in M.K.M., the Danish authorities concluded that the author's evidence was “unlikely and non-credible” because she had not raised her alleged torture and gang rapes by Chechnyan officials on her asylum request form or during her first interview.¹⁷⁷ Both women provided plausible explanations for only disclosing their abuse later in proceedings. E.W. argued that her accounts had been different but not contradictory, and that the differences could be explained by the fact that her first interview had been with a policeman (and she had been raped by policemen in China), whereas “she had been more open with regard to relating the abuses when she talked to the female case handler at the [Danish Refugee] Council.”¹⁷⁸ M.K.M. argued that she had found it difficult to discuss her abuse through a male interpreter, and attributed her delay in detailing the abuse to “the fact that, in Chechnya, culture and tradition dictate that speaking openly about sexual abuse often results in the stigmatization of the victim and exclusion from society.”¹⁷⁹ She argued that it was “understandable that she explained all the facts only at the hearing of the Refugee Appeals Board, at which she was accompanied by her own legal representative, who was a woman,”¹⁸⁰ and that “she could talk about having been raped only after she realized that in Denmark it was possible to talk about sexual abuse, whereas in her country doing so would have been shameful.”¹⁸¹

In General Recommendation 32, the Committee had explicitly called on states parties to ensure “[t]hat late disclosure by the claimant during the asylum procedure of sexual violence and other traumatic events does not automatically lead to an adverse judgement on her credibility,” noting that “[r]eluctance to identify the true extent of the persecution suffered or feared may stem from feelings of shame, stigma or trauma.”¹⁸² This recommendation is consistent with guidance from UNHCR that women submitting gender-related asylum applications “may feel unable or reluctant to disclose information for many reasons . . . [which] include, but are not limited to, the effects of trauma, other mental health problems, stigma and shame, lack of trust in authorities, fear of rejection or ostracism, and fear of serious harm as a reprisal.”¹⁸³ It is also consistent with guidance in the criminal law context that victim fabrication or “hue and cry” myths should not inform credibility decision making in cases involving sexual violence,¹⁸⁴ and that no adverse inferences should be drawn solely from any delay between alleged offenses and their reporting.¹⁸⁵ In E.W. and M.K.M., however, the Committee neither cited its own guidance nor addressed the issue of late disclosure in any substantive way. It was entirely silent on the point in E.W. and merely “recalled” the author's submissions in M.K.M.,¹⁸⁶ before concluding that nothing on file indicated any “irregularities” in the Danish process.

In other cases, Denmark focused more on women's manner when disclosing abuse. In N.M., the Danish authorities described the author's account of torture as “superficial and constructed” and “fabricated and not reflective of personal experience” because she had “replied evasively to the question of how the torture was carried out and she had also been unable to give any information as to how the torture affected her.”¹⁸⁷ The Views do not explain how, specifically, the author's answers were “evasive.” The author argued “that, being only 26 years old, it was very difficult to express how she was affected and also that she found it traumatizing to explain her account again and again.”¹⁸⁸ In concluding, again, that there was no irregularity in the Danish process, the Committee failed to provide any caution about the unreliability of demeanor as an indicator of credibility or guidance on how trauma can impact memory and behavior. It did not recall that “[t]he type and level of emotion displayed during the recounting of her experiences should not affect a woman's credibility.”¹⁸⁹ Nor did the Committee reflect on how the intersecting factors of age, gender, background, and trauma might cause a survivor of serious GBV to avoid thinking or talking about the details of distressing events.¹⁹⁰

In S.F.A. and H.H.M., the Danish authorities drew negative credibility findings from the fact that the author's statements were “vague.”¹⁹¹ The author, a young mother traveling alone with a baby, submitted that “she had her child on her lap at the time of her interviews . . . which confused her and, accordingly, had an impact on the statements that she gave regarding the grounds for her asylum application.”¹⁹² Denmark rejected this explanation, stating that “[a]t no point during those interviews did she express any confusion,” that she “had the option of requesting a babysitter during the interviews,” and that she had reviewed and signed the reports of her interviews.¹⁹³ In General Recommendation 32, the Committee had acknowledged the challenges women might face in presenting their claims in the presence of children and recommended that states parties ensure “that childcare is made available during the interviews so that the claimant does not have to present her claim, involving sensitive information, in front of her children.”¹⁹⁴ In S.F.A. and H.H.M., however, it made no reference to this guidance or the specific challenges that single mothers might face during asylum proceedings. Nor did it address the implication in Denmark's submissions that, having chosen not to leave her child with a babysitter, the author might have forfeited the right to cite the presence of her child as an explanation for “vagueness.” Instead, the Committee simply noted the parties’ submissions before preceding to rule the communication inadmissible.

2. Assumptions About Women and Male Perpetrators of GBV

In some cases, Denmark appeared to base negative credibility findings on general assumptions about how women victims and male perpetrators of GBV should act in certain circumstances. In M.K.M., for example, Denmark argued that it was “unlikely” the author would have continued to refuse to work as an informer for the Chechen authorities after they had tortured and gang raped her when she first refused to cooperate.¹⁹⁵ In A.R.I., the Danish authorities questioned why a mother would use money to help her daughter escape rather than pay a ransom for her son.¹⁹⁶ In F.H.A., they found it “implausible” that the author had maintained a relationship with her future husband against her uncle's will.¹⁹⁷ In S.A.O., they ruled it not “credible” that the author would have visited her boyfriend around the time of her forced marriage to an older man.¹⁹⁸ No evidentiary bases for these conclusions are apparent in the Committee's Views, leaving open the possibility that they were (wrongly) based on “unfounded assumptions, subjective speculation, conjecture, stereotyping, intuition, or gut feelings”¹⁹⁹ about how women would or should act in each situation.

Similarly, the Committee's views do not disclose any evidentiary bases for assumptions about the behavior of GBV perpetrators. In A.R.I., Denmark asserted that it was “unlikely” the author's brother would have believed rumors that she had been raped because her mother had denied them.²⁰⁰ In A.N.A., Denmark found it “unlikely” that Al-Shabaab would wait until six years after taking control of the author's town to challenge her marriage to a man who did not work for Al-Shabaab.²⁰¹ In A.M., it also ruled the author's claim that her ex-husband had returned ten years after their divorce “unlikely” and “fabricated for the occasion,” despite the author attributing the interval to him traveling abroad to train and fight as an Al-Shabaab militant.²⁰² In all three cases, the Committee appears not to have interrogated the evidentiary bases for these assumptions, and failed to provide guidance on the related issue of whether women bear the burden of explaining the allegedly implausible behavior of their persecutors.²⁰³

In E.W., one of the few cases disclosing an evidentiary basis for credibility findings, that basis also appeared tainted by assumptions about how women would or should act. The case contained complex facts, and it may have been open to Denmark and the Committee, looking at the case in its totality, to conclude that the author was not eligible for protection. However, it also revealed problematic assumptions that should have been addressed by the Committee. Denmark found it “improbable” that the author had been subjected to torture and sexual abuse by local public officials for more than ten years without complaining to higher authorities.²⁰⁴ The evidentiary bases for this conclusion included the facts that rape was criminalized in China, that “Chinese nationals whose rights or interests are infringed by public officials can lodge a complaint against such officials,” and that the author had managed to acquire a passport after subsequently reuniting with her husband.²⁰⁵ By contrast, there is no indication that the Danish credibility assessment factored in: (1) the author's intersecting vulnerabilities as “a single, illiterate woman” who “could not write to anyone outside the village” and was “fully dependent” on her alleged abuser for the necessities of life;²⁰⁶ (2) her claim that her circumstances left her feeling “too helpless” to leave and that “if she had left without [her abuser's] consent she would have been dragged back immediately”;²⁰⁷ (3) her claim that she did not complain to higher authorities in China “because she had been told that her husband had been killed for having tried to complain about his harassment by local authorities” and she “feared that the same could happen to her, as most of her rapists were employed by the local authorities”;²⁰⁸ or (4) the fact that there was a material change in her ability to escape gender-based harm in China after reuniting with her husband.

Without drawing conclusions on how Denmark or the Committee should ultimately have ruled in E.W.'s case, there are some matters on which the Committee should have provided guidance. First, it should have clarified that the criminalization of rape will not be sufficient on its own to establish that women in fact enjoy effective protection against sexual violence, or redress where it is committed by local authorities and police officers.²⁰⁹ Second, it should have recalled its guidance from General Recommendation 32 that “the fact that a woman asylum seeker has not sought the protection of the State or made a complaint to the authorities before her departure from her country of origin should not prejudice her asylum claim,”²¹⁰ and certainly should not be determinative of her credibility. Third, it should have affirmed that the credibility of the author's account, as extraordinary as it may have sounded to a Danish decision maker, should have been determined from the perspective of the author bearing in mind her individual circumstances. Instead, the Committee ruled the complaint inadmissible without any comment on credibility.

3. Inconsistencies

The credibility of narratives of GBV was also impugned by a focus on apparent inconsistencies, either within the authors’ evidence or between that evidence and other sources.²¹¹ While “internal” and “external” consistency are commonly used as indicators of credibility in RSD,²¹² a gender lens reveals issues in these cases specific to women fleeing GBV. When inconsistency evidence arose as a key issue in F.H.A., the author claimed that:

a lower priority seems to have been given in the proceedings to gender-based issues than to details surrounding those claims. The gender-based violence, forced marriage, domestic abuse and discriminatory, patriarchal societal norms were not properly taken into account and, instead, the proceedings were focused on details related to the flight of the author, with the minor divergent information apparently being sufficient grounds for rejecting her entire claim for protection.²¹³

Other authors made similar claims about a disproportionate focus on “minor divergent information” instead of evidence of gender-based harm. A.M. argued that Denmark was “deliberately seeking to find inconsistencies and elaboration in her statements, instead of taking into account her vulnerabilities as an illiterate person who has been subjected to severe psychological and physical abuse.”²¹⁴ K.I.A. alleged that Denmark had maintained a “central focus” on discrepancies in the evidence about her departure from Jordan, while “[t]he original honour-related sanctions for not obeying traditional norms, the forced marriage, the domestic violence and abuse, and the discriminatory legal system in Jordan were not duly taken into account.”²¹⁵ N.M. also claimed that “it appeared that the [Danish] Refugee Appeals Board sought to identify the smallest inconsistency upon which to base a finding of lack of credibility in order to allow it to dismiss the case.”²¹⁶ This line of argument was refuted by Denmark, which gave evidence about how its asylum authorities sought to clarify evidence characterized by inconsistencies, expansions or omissions, taking into account “the asylum seeker's particular situation, including cultural differences, age and health.”²¹⁷

While the Committee's role is not to replace national authorities in determining asylum claims, it does have a crucial role to play in ensuring that those authorities have considered all relevant gender-sensitive explanations for apparently inconsistent evidence and assigned such explanations proper weight. Unfortunately, the only case in which the Committee made substantive findings on inconsistency evidence was in F.F.M., and these were imperfect. Denmark argued, and the Committee agreed, that the author's credibility was undermined by inconsistencies in her evidence about secretly marrying her boyfriend. She claimed both that she had “destroyed the marriage certificate out of fear of being killed if the marriage were discovered” and that she had married her boyfriend to avoid being forced to marry a wealthy member of Al-Shabaab.²¹⁸ The Views contain no indication that either Denmark or the Committee considered the possibility that a young woman in the author's position might have had changing and conflicting motivations for the choices she made at a difficult time. Denmark also noted that the author had not disclosed until after the refusal of her asylum application an additional incident in which members of her boyfriend's family had come to her house, thrown boiling water on her, and stabbed her. Neither Denmark nor the Committee accepted as adequate the author's explanation that she had “initially focused on her immediate reasons for escaping,” being “increasing persecution from her husband's family, culminating in violent attacks at the family home,” and that she “thought that her case was clear and that the scars on her body were living evidence of her ordeal.”²¹⁹ Finally, Denmark and the Committee noted that the author had told a Danish doctor that she received the beating on the street rather than as a result of domestic violence within her home, and that “her account of the nature of the attack . . . changed” from a beating on both shoulders to “strikes to the heart.”²²⁰ The Views contain no consideration of the possible reasons why a woman might not disclose familial GBV immediately and consistently, nor do they give weight to the author's submissions that she was “still suffering from the psychological and physical effects” of her journey and that “while any lack of consistency can be attributed to her medical condition, her overall statements have been very consistent.”²²¹

Aside from in F.F.M., the Committee either did not mention at all, or only noted in passing, the parties’ submissions on inconsistency evidence and credibility. The Committee's failure to provide guidance on this issue was especially unfortunate in A.S., the only case involving a woman seeking asylum on the basis of her sexual orientation decided by the Committee to date. Denmark declared A.S. not credible in part because her evidence about when and how she had “discovered” that she was a lesbian, and about her sexual relationships with other women, had been “inconsistent, elaborative and sketchy.”²²² Denmark emphasized that the author had originally stated she “discovered that she was a lesbian” in 2007, when she commenced a relationship with another woman, but then added in later interviews that she had “known” she was a lesbian earlier in her life.²²³ A.S.'s failure to date her sexual orientation to a single and consistent point in time led the Danish authorities to reject that she “[was] in reality a lesbian.”²²⁴ In its Views, the Committee recalled in passing its guidance in General Recommendation 32 that “[g]ender-related claims to asylum may intersect with other proscribed grounds of discrimination, including . . . being lesbian, bisexual or transgender,”²²⁵ but provided no guidance on how this intersectionality should have been applied in the case before it. Nor did it provide guidance on how a person's understanding of their own sexual identity might emerge and evolve in a non-linear manner.²²⁶

Finally, in L.O., the Committee found no errors with Switzerland's reliance upon external inconsistencies—that is, differences between the author's testimony and other available evidence—to refuse her asylum claim. L.O. and her children had experienced sustained and severe domestic abuse and fled to Switzerland after her attempts to seek protection in Mongolia proved fruitless. The Swiss State Secretariat for Migration denied L.O.'s asylum claim because of apparent inconsistencies between her evidence and a “confidential” report prepared by “a trusted lawyer” in Mongolia.²²⁷ The author was not permitted to read the report, but was advised that it contradicted her evidence on various administrative matters, and found no record of her former partner being registered as a domestic violence perpetrator. In response, the author gave detailed evidence about naming protocols, registration records, and rental practices in Mongolia, as well as possible typographical errors or mistranslations that could plausibly explain why the lawyer had not found documentary evidence to substantiate those aspects of her evidence.²²⁸ As for the lack of police records, the author advised that her former partner had worked as a police officer so it was possible that certain files had been “lost.”²²⁹ She also argued that the relevant police stations would probably have records, but that it would be difficult for her to obtain them from Switzerland.

The Swiss authorities attached little probative weight to L.O.'s explanations for the apparent contradictions between her evidence and the lawyer's confidential report. Instead, the State Secretariat concluded that “the inconsistencies could not be ignored” and that the author's allegations “did not meet the credibility requirements in an asylum case.”²³⁰ On appeal, the Federal Administrative Court ruled that “[t]he lack of evidence concerning the violence suffered shows the lack of credibility of the alleged facts, all the more so since the author had stated that she had gone to the police on several occasions,” and thus that “the author's allegations of domestic violence and related problems were not plausible.”²³¹ Notably, Switzerland did not appear to have placed significant weight on the fact that the author's descriptions of the violence against herself and her children had remained consistent, or on other evidence corroborating the author's experiences of abuse. Even when the author did later obtain police and medical records of her abuse from Mongolia, Switzerland argued that those records were “not of a nature that would modify the State party's assessment of the author's request.”²³²

Had the Committee reviewed the Swiss approach through a gender lens, it might have made several findings. First, it might have clarified Switzerland's evidentiary basis for assuming that a Kazakh woman in Mongolia would know the second surname of her partner if they were unmarried.²³³ Second, it might have interrogated the respective weight given by Switzerland to the lack of documentary administrative records on the one hand, and the consistent account of abuse on the other, to ensure the author's account was not deemed less important. Third, it might have questioned Switzerland's submission to the Committee that the police and medical records would not have changed the decision made by its asylum authorities, given that the lack of records had been explicitly identified as key to their negative credibility findings. Instead, however, the Committee chose to exclude those records from its consideration because they had not formed part of the Swiss asylum process.²³⁴ It did so despite the gravity of the alleged risk to L.O. and her children, L.O.'s explanation that she had been unable to obtain them during the asylum proceedings “as she did not dare to contact the Mongolian authorities out of fear that her partner would be informed of her whereabouts,”²³⁵ and Switzerland's submission that even if she had produced them they would not have changed the outcome of her case.²³⁶ Moreover, in ruling the communication inadmissible, the Committee made no reference to its own longstanding concerns about violence against women in Mongolia,²³⁷ or its findings about inadequate implementation of CEDAW in Switzerland.²³⁸

B. Interpreting CEDAW

While credibility issues dominated this period of the Committee's asylum work, it also appeared unwilling to assert its expertise or challenge other aspects of national asylum decision making involving possible misapplication of CEDAW. Earlier, the Committee had cautioned states not to view women's asylum claims through the lens of male experiences, yet it did little to model this approach after 2016. One missed opportunity arose in A.R.I., where Denmark argued that Chechen authorities rarely paid attention to “distant relatives” of insurgents, whereas the author argued that the situation was “particularly difficult for female supporters or alleged supporters of insurgents, given that women risk violence, sexual harassment and sexual assault by the Chechen authorities.”²³⁹ Neither Denmark nor the Committee appears to have considered the specific risk profile of a female relative providing medical treatment to an insurgent, or the fact that “women's political activity may not always look like male political activity.”²⁴⁰

Second, the Committee failed to comment when it appeared that Danish asylum law and processes had not been updated to give effect to its earlier findings. In K.S., Denmark submitted that, under its Aliens Act, a residence permit would only be granted to an asylum seeker who was “at risk of the death penalty or of being subjected to torture or inhuman or degrading treatment or punishment should he or she return to his or her country of origin,” without reference to serious forms of gender-based discrimination.²⁴¹ Similarly, in E.W., Denmark submitted that even if the author had been repeatedly raped by local officials, such harm would “not fall within the concept of asylum and protection” under Danish asylum law.²⁴² Denmark's submissions in these cases should have alerted the Committee to the possibility that Denmark had not fully accepted its findings in M.N.N., and was wrongly comparing sexual violence with other forms of “mere” private criminality.²⁴³ Yet the Committee disposed of these cases without clarifying whether the Danish asylum system properly conceived of sexual violence as a form of gender-based persecution.

Third, the question of whether women bear an evidentiary burden to prove the identities and motivations of their abusers remained unanswered. In A.M., Denmark drew adverse credibility findings against the author because she was unsure whether Al-Shabaab or her ex-husband had attacked her and her spouse.²⁴⁴ Moreover, it claimed that she had:

failed to give a reasonable explanation as to why she was able to stay in . . . her home town, for the first 10 years of her marriage to her second spouse and subsequently for two years after his murder, without being contacted by members of Al-Shabaab, if her first spouse wanted revenge.²⁴⁵

For her part, the author argued that “she had only gradually realized that her first husband was behind the attack” and that “[b]lame should not be attributed to a victim for not knowing precisely who her attackers were.”²⁴⁶ As possible explanations for the periods of safety, she suggested that “her first husband had been abroad training with Al-Shabaab,” and that “the attackers, seeing her lying on the floor after having been shot, must have thought that she was dead.”²⁴⁷ In concluding that the author's removal to Somalia would not violate CEDAW, the Committee provided no guidance on victim-blaming, and merely noted some of these submissions but made no substantive findings.

Fourth, in L.O., the Committee failed to clarify when it is reasonable to conclude that protection against GBV is available in another country.²⁴⁸ On the issue of national protection, L.O. claimed that she:

contacted the police several times to submit complaints about the abuse that she was suffering. However, the police were unwilling to help, partly because the author and B.Y. were not married, and her complaints were not considered “valid.” The police would occasionally detain B.Y. for a few hours and then release him. After he was released, the abuse towards the author and the children intensified. As a result, the author stopped notifying the authorities.²⁴⁹

L.O. tried to seek help from a women's shelter but was unsuccessful and fled to various safe houses but each time her partner found her.

Despite these claims, the Swiss authorities concluded that the Mongolian authorities could provide L.O. and her children with effective protection, because:

Given the practice of the Swiss authorities, there was reason to believe, in general, that the Mongolian authorities had the will and the capacity to take protection measures, and that the infrastructure available for such measures was sufficient. . . . Mongolia was considered to be a secure State in terms of the right to asylum. That status presupposed a functional police and legal system.²⁵⁰

They noted that Mongolia had a law against domestic violence and concluded that the fact the author went to the police on several occasions demonstrated their “willingness to offer protection.”²⁵¹ On appeal, the Federal Administrative Court held that there was a “presumption that Mongolia was in a position to offer protection from violence perpetrated by a third person” and that the author could not “reverse” that presumption because her former partner had been detained on several occasions.²⁵² The Committee endorsed the approach of the Swiss authorities, finding that they “took sufficiently into consideration in their assessment the extent of domestic violence in Mongolia, the existing legal framework and the availability of protection from the authorities.”²⁵³

The Committee failed to address various aspects of the Swiss approach warranting scrutiny. Was there, as the Federal Administrative Court suggested, a “presumption” that national authorities offer protection from GBV committed by non-state actors which women must “rebut?” Was it appropriate to rely on the unspecified “practice of the Swiss authorities” which indicated that “in general” Mongolia had “a functional police and legal system,” rather than the specific facts of the author's individual case, which included being an unmarried woman and the perpetrator being a former police officer? Was there any evidence that the “presupposed . . . functional police and legal system” in Mongolia protected women against domestic violence in particular, as opposed to crime generally? Was it sufficient for Switzerland merely to note that Mongolia criminalized domestic violence, or did it need to examine whether that law was, in fact, implemented such as to provide effective protection to women? And, given the severity and consistency of the author's alleged abuse, was it appropriate for Switzerland to conclude that the police occasionally detaining a perpetrator amounted to a “willingness to offer protection?” Not only did the Committee fail to address, or even identify, any of these issues, it noted that the author had “not explained why she did not submit her complaints against her ex-partner to the Mongolian prosecuting authorities or courts,”²⁵⁴ without first establishing an evidentiary basis for believing that it was even open to L.O. to bypass the police and receive effective protection from prosecutors or courts directly. The deficiencies in these Views were exacerbated by the Committee's failure to reference its own relevant Concluding Observations, which were highly critical both of “the high prevalence of violence against women, in particular domestic and sexual violence” and the lack of sufficient support for victims of violence in Mongolia,²⁵⁵ and wide-ranging deficiencies in Switzerland's implementation of CEDAW.²⁵⁶

Fifth, the Committee failed to establish sound normative guidance on the issue of Somali women enjoying protection from “male networks” against GBV. This issue became prominent after the ECtHR ruled in 2015 that “a single woman returning to Mogadishu without access to protection from a male network would face a real risk of living in conditions constituting inhuman or degrading treatment.”²⁵⁷ Following this judgment, some women began to challenge the rejection of their asylum claims. These challenges raised the critical questions of whether expecting a woman to rely on male family members for safety was itself a form of discrimination against women, and whether protective family members were sufficient to alleviate a woman's risk of serious forms of GBV in the absence of state protection. Unfortunately, neither question was resolved in the Committee's Views. Instead, both Denmark and the Committee focused on identifying male family members in Somalia, without assessing whether those men comprised a network that would in fact protect the relevant authors against harm.²⁵⁸

This approach was particularly problematic in cases where the purported male network included the men accused of perpetrating GBV. In S.F.A. and H.H.M., for example, the author challenged Denmark's finding that she had a male network and could safely be returned to Somalia, arguing that “the male network referred to by the State party comprises the very same persons, her father and brothers, whom she fears will kill her upon her return to Somalia.”²⁵⁹ Denmark took a similar approach in S.A.O. and F.H.A. In all three cases, Denmark rejected the authors’ evidence about conflict with male family members as not credible, and on that basis concluded that they would not be single women without male networks in Somalia.²⁶⁰ Additionally, in A.N.A. v. Denmark, the Committee appeared not to detect critical contradictions in Denmark's submissions on the author's male network. During her asylum proceedings, the Danish authorities had taken into account the fact that the author was “a young woman with no social network in Somalia.”²⁶¹ Then, before the Committee, Denmark argued that she had “failed to substantiate that she would find herself in a situation with no male support network.”²⁶² For its part, the Committee found no evidence of irregularities in the Danish process. Similarly, in D.N.S., the Committee failed to detect any error in Denmark basing its finding that the author had a male network on: (1) the fact that D.N.S. had a brother in a Kenyan refugee camp (not Somalia); and (2) its assertion that the author was likely to have “substantial” (but unidentified) connections in Somalia because she had been able to leave the country so quickly after escaping from detention.²⁶³

C. The Unexplained Anomaly: R.S.A.A. v. Denmark

Within the Committee's asylum work at this time, R.S.A.A. stands out as a striking anomaly. Decided in July 2019, it was the second asylum case in which the Committee found that removing a woman to her country of origin would violate CEDAW. To highlight the incongruity between R.S.A.A. and other cases decided around the same time, it is helpful to analyze it alongside K.I.A., a comparable case decided just four months later in November 2019. K.I.A. and R.S.A.A. were both Palestinian refugees and Jordanian nationals²⁶⁴ who claimed to have fled to Denmark to escape domestic violence. K.I.A. had been forced as a child into a marriage with a forty-seven-year-old man “as an honour-related sanction for not following traditional family norms.”²⁶⁵ Her marriage was “characterized by repeated domestic violence, rape and controlling and degrading treatment,” and her husband was also violent toward their children.²⁶⁶ R.S.A.A. claimed that she and her daughters had been exposed to threats, violence, abuse, and degrading treatment by her husband, which had escalated after she opposed the forced marriage of one of her daughters to an older man, to the point where she was “beaten up and tortured” by her husband.²⁶⁷ Both women lied to their husbands about their reasons for traveling to Denmark and, once there, sought asylum. Both women emphasized their particular vulnerability as Palestinian women in Jordan, and challenged Denmark's claims that they were “entitled . . . to the same protection as all Jordanians” and could “rely on the same rights as other Jordanian nationals” with respect to protection against domestic violence by the Jordanian authorities.²⁶⁸

Denmark made adverse credibility findings against both women and rejected their asylum claims. It ruled that K.I.A.'s “statements on the intensity and scope of the violence and sexual abuse in her marriage could not be considered as facts” because they “appeared to be inconsistent and increasingly elaborate on a number of essential points.”²⁶⁹ For her part, K.I.A. argued that there was nothing contradictory about the additional information she provided about her abuse, and that:

it is reasonable to assume that an abusive relationship of the nature described, over such a long period, can include other types of abuse . . . and that it cannot be expected that such details will automatically be given up during an interview in which the author has already described severe domestic abuse.²⁷⁰

Denmark also ruled R.S.A.A.'s domestic violence testimony “inconsistent,” “non-credible,” and “fabricated for the occasion,” noting that she had omitted important information at various stages of the proceedings both in Denmark and before the Committee.²⁷¹ Having determined that the authors’ domestic violence testimonies were not credible, Denmark found it unnecessary to assess whether the Jordanian authorities would offer the authors or their children protection upon return.²⁷²

Despite the similarities between these two cases, the Committee's findings in each were conspicuously different. The Committee ruled K.I.A. inadmissible without providing any substantive reasons.²⁷³ By contrast, the Committee ruled R.S.A.A. admissible and then proceeded to find that the deportation of the author and her daughters would amount to a breach of CEDAW. In doing so, it reiterated that it is generally for states parties to evaluate facts, evidence and the application of national law, but added that:

it was incumbent upon the State party to undertake an individualized assessment of the real, personal and foreseeable risk that the author would face, as a woman who has knowingly abandoned her violent husband and fled Jordan with their two minor daughters who were at risk of forced marriage there, rather than relying exclusively on a number of inconsistent statements and the inferred non-credibility of the author.²⁷⁴

In other cases, the Committee had insisted that authors identify, with a high degree of specificity and proof, irregularities in national decision making which amounted to bias, gender stereotyping, discrimination against women, arbitrariness, or a denial of justice. By contrast, in R.S.A.A., it did not articulate which of these grounds (if any) had been established. Instead, it simply concluded in general terms that Denmark had “failed to give sufficient consideration to the real, personal and foreseeable risk of serious forms of gender-based violence faced by the author and her daughters should they be returned to Jordan,”²⁷⁵ and thus that return to Jordan would violate CEDAW.²⁷⁶

Five key factors appear to have informed the Committee's findings in R.S.A.A.: (1) recent Concluding Observations in which the Committee had expressed concern about the treatment of women and girls in Jordan; (2) Denmark's failure to assess whether the Jordanian authorities were able to ensure adequate protection for the author and her daughters upon their return; (3) Denmark according “no due weight to the author's vulnerable status as a Palestinian refugee” and her risk of statelessness should her Jordanian nationality be withdrawn; (4) Denmark's decision not to request verification of the authenticity of an arrest warrant issued against the author and assess the risks that she would face if prosecuted for the alleged abduction of her daughters to Denmark; and (5) Denmark's failure to give “any specific consideration” to the claim that the author's daughters would face risks of forced marriage if returned to Jordan. While the Committee's reliance on these grounds in R.S.A.A. was sound, it is unclear why they were completely absent from its Views in K.I.A., given they could equally have applied there.

D. Dublin Transfer Cases

Two final cases warrant separate consideration due to the distinct legal issues they raise. Abdi-Osman and M.A., decided in July 2020 and November 2021 respectively, involved women who were not facing return to their countries of origin following review of their asylum claims, but rather transfer from Switzerland to Italy and France pursuant to the Dublin III Regulation.²⁷⁷ Within the European Union (EU) and associated states, including Switzerland, this regulation establishes the “Dublin system,” a special regime based on a hierarchy of criteria for determining a single state responsible for examining a given asylum seeker's claim. The Dublin system is a central pillar of the Common European Asylum System (CEAS), which is underpinned by the principle of mutual trust.²⁷⁸ According to this principle, all states participating in the CEAS are presumed to observe EU law and respect fundamental rights, and therefore to be “safe” countries to which asylum seekers can be transferred in accordance with the Dublin criteria. This presumption, and the way it informs Dublin transfer procedures and decisions, render them fundamentally different from decisions to remove asylum seekers to their countries of origin following full and fair RSD.

Under an earlier version of the Dublin system, the principle of mutual trust was applied almost “blindly,”²⁷⁹ with some states establishing irrefutable presumptions that it was safe to return asylum seekers to other Dublin states.²⁸⁰ Transfer decisions could be made without any substantial consideration of an individual's asylum claim or the risks they might face in the Dublin state to which they were returned. This position later began to shift, with both the ECtHR and the Court of Justice of the EU (CJEU) affirming in 2011 that the presumption of compliance with fundamental rights in Dublin states may be rebutted in certain circumstances.²⁸¹ But the ability of national authorities to inquire into whether another Dublin state in fact respects fundamental rights remains “considerably circumscribed by the principle of mutual trust.”²⁸² Moreover, the threshold for rebutting the presumption of safety is contested and evolving.²⁸³

The presumption of safety may be rebutted where there are “substantial grounds for believing that there are systemic flaws in the asylum procedure and in the reception conditions” for asylum seekers in the relevant Dublin state, resulting in a risk of inhuman or degrading treatment contrary to Article 3 of the European Convention on Human Rights (ECHR) and Article 4 of the EU Charter of Fundamental Rights.²⁸⁴ This is a high threshold, requiring something akin to “the collapse of a national protection system.”²⁸⁵ The European courts have traditionally differed on the question of whether individual risks of inhuman or degrading treatment, absent “systemic flaws” in the broader system, can also preclude Dublin transfers.²⁸⁶ In a case of direct relevance to Abdi-Osman, the ECtHR held in Tarakhel v. Switzerland, in 2014, that conditions in Italy did not rise to the level of “systemic flaws” such as to bar all removals to that country, but that the presumption of safety could nevertheless be rebutted because of the individual circumstances of the applicants, a family with six minor children.²⁸⁷ The ECtHR held that the family's transfer to Italy would violate Article 3 of the ECHR unless Switzerland “first obtained individual guarantees from the Italian authorities that the applicants would be taken charge of in a manner adapted to the age of the children and that the family would be kept together.”²⁸⁸ The CJEU, by contrast, has been more resolute in its defense of mutual trust as a fundamental principle essential to the proper functioning of the EU. In an Opinion in 2014, it insisted that member states are “required to presume that fundamental rights have been observed by the other Member States” in all but “exceptional” cases.²⁸⁹ Exceptional cases were previously limited to those in which there were “systematic flaws” in the asylum procedure and reception conditions of a Dublin state, although there are some recent signs of CJEU convergence with the ECtHR in factoring individual circumstances into the assessment of risk of inhuman or degrading treatment.²⁹⁰

In addition to these debates about when the risk of inhuman or degrading treatment will displace mutual trust, it remains unclear whether there is a conclusive presumption of safety as regards other fundamental rights?²⁹¹ Specifically, can an asylum seeker challenge a Dublin transfer on the basis that she would face serious forms of discrimination against women (including prostitution or trafficking²⁹²) in the country of return? Are national authorities required or even permitted under EU law to inquire into the level of actual compliance with CEDAW in other Dublin states? These questions remain unresolved, and while it is beyond the scope of the Committee's mandate to determine them for the states participating in the CEAS, a more assertive Committee might at least have acknowledged that it should take a different approach to such cases.

At a minimum, the Committee should have interrogated whether the Swiss authorities ever actually considered whether the authors would face real, personal, and foreseeable risks of serious forms of discrimination against women in Italy and France, or whether they simply presumed that the rights contained in CEDAW would be respected in those states. The Views in Abdi-Osman and M.A. suggest the latter. In Abdi-Osman, the Swiss authorities refused to consider the author's asylum application because Italy was considered a “safe” country. They argued that if Italy was not fulfilling its obligations “the author would have to assert her rights in Italy.”²⁹³ Switzerland subsequently concluded that “there was no evidence that the author would be at risk of being subjected to serious gender-based violence in Italy or that the Italian authorities would not afford her effective protection against gender-based violence,”²⁹⁴ but the Views contain no indication of how Switzerland reached this conclusion. In M.A., Switzerland considered whether there were “systemic flaws” in the French asylum procedure and reception conditions,²⁹⁵ but does not appear to have considered the separate and more relevant question of whether, in light of her individual circumstances, the author's return to France would violate the implied non-refoulement obligation in CEDAW. In fact, Switzerland not only argued that article 2(d) of CEDAW did not “apply” to the author's claims, but indeed that it was not “directly applicable” in Switzerland because its wording was “not sufficiently clear or precise.”²⁹⁶

Thus, in both cases, there appeared to be insufficient information to conclude that Switzerland had properly assessed whether the authors would face relevant risks of harm as a result of Dublin transfers. Yet the Committee identified no issues with the Swiss process in either case. In Abdi-Osman it concluded, in familiar formulaic terms, that “nothing on file” demonstrated that Switzerland's examination of the claim was tainted by bias, discriminatory gender-based stereotypes, arbitrariness or a denial of justice.²⁹⁷ The Committee ruled M.A. inadmissible in even briefer terms, concluding summarily that “in the light of the information before it” the author had not sufficiently substantiated her claim.²⁹⁸ These summary dismissals are disappointing given the Committee's knowledge of extensive deficiencies in Switzerland's implementation of CEDAW.²⁹⁹ Its failure to address how CEDAW adds to or changes the obligations of states seeking to effect Dublin transfers leaves critical questions unresolved and will have repercussions for the Committee's pending case list, which includes a number of Dublin matters.

VI. Recommendations for More Robust Monitoring of Asylum Cases

A. Improving the Quality of Asylum Decision Making

In May 2023, the Committee published its Views in Bandboni, a case involving an Iranian woman who fled GBV at the hands of her father and brothers and sought asylum in Switzerland. Compared with its Views in asylum cases in preceding years, these provided relatively clear reasons for the Committee's findings. Credibility was not an issue. Instead, the Committee was required to review whether Switzerland had erred in finding that the author and her family would be able to obtain adequate protection from the Iranian authorities against future risks of harm. The Committee ruled the case admissible and, on the merits, clearly identified shortcomings in the Swiss assessment of the author's case. Specifically:

the persistent institutionalized discrimination against women and girls in public and private life enshrined within civil and penal law and practice in the Islamic Republic of Iran, the patriarchal values and misogynist behaviours that permeate many segments of Iranian family life, and the law enforcement agencies’ reluctance to intervene in domestic violence and honour crime cases were not sufficiently addressed in the context of the case at stake.³⁰⁰

The Committee critiqued the Swiss authorities for dismissing the author's fear of not being able to obtain protection from the Iranian authorities as “pure guesswork,” and for failing to accord any weight to the fact her father had been accompanied by a policeman on a previous occasion when he had sought her whereabouts. It concluded that “a more thorough risk assessment in connection with the capacity of Iranian authorities, including law enforcement agencies, to protect women and girls would have been required by the exigencies of the case.”³⁰¹ Like in R.S.A.A., the Committee did not identify specifically the irregularity in the Swiss decision-making process (arbitrariness, denial of justice, or gender stereotypes/bias), but did conclude in general terms that Switzerland had “failed to give sufficient consideration to the real, personal and foreseeable risk of serious forms of gender-based violence faced by the author should she be returned to the Islamic Republic of Iran.”³⁰²

The Committee's pending case list indicates that asylum issues will continue to comprise a significant amount of its work. With a view to these future cases, it would be premature to conclude that Bandboni indicates a shift in the Committee's approach to asylum cases. It does, however, provide a model for how the Committee might improve the quality of its Views and function as a more effective advocate of women's rights.

Any discussion about improving the quality and effectiveness of the Committee's decision making must be situated within the broader context of longstanding international efforts to reform and strengthen the UNTB system as a whole.³⁰³ Many of the deficiencies identified in the preceding analysis are not unique to the Committee, and derive from normative and structural factors outside its control, including the “watered down” non-binding legal status of Views,³⁰⁴ reliance on unpaid, part-time members elected by states,³⁰⁵ and inadequate annual meeting time and secretarial and other support.³⁰⁶ But the Committee need not, and indeed should not, wait until these broader issues are resolved to continue improving its own decision making. The asylum cases reviewed above indicate various ways in which the quality and effectiveness of Views may be improved in the short to medium term, to the benefit of its pending case list and future communications.

A substantial body of scholarship about the UNTBs and their complaints mechanisms assists to clarify the criteria of high-quality and effective Views. The present analysis adopts Egan's definition of “effectiveness” as the extent to which UNTB procedures are successful at producing desired results.³⁰⁷ Byrnes identifies the three main results or functions of the complaints mechanisms as:

• • providing an effective and timely remedy to the person whose rights have been violated;

• • bringing about changes to law and practice in the respondent state which will benefit others in a similar position to the complainant, now and in the future; and

• • through the elaboration of a jurisprudence of the relevant treaty, providing guidance to states parties and others on the meaning of the guarantees contained in the treaty and the measures that are needed to protect those rights.³⁰⁸

Some scholars express a degree of pessimism about whether UNTBs can actually achieve the first of these functions.³⁰⁹ For at least the other two, though, it would appear uncontroversial that the ability of UNTBs to produce these results depends on their persuasiveness.³¹⁰ Since Views do not have formally binding legal force, their normative value depends largely on their “persuasive authority,” which in turn stems from the legitimacy of the Committee which makes them and the quality or “merit” of the findings themselves.³¹¹

Analytical rigor is a key indicium of high-quality, and thus persuasive, decision making.³¹² On this point, Mechlem stresses the importance of UNTBs adopting appropriate legal methods of interpretation to produce decisions which are “comprehensible, rational, predictable, legitimate, and reproducible, and . . . do justice to the principles of legal certainty and . . . the rule of law.”³¹³ She argues that:

[M]ethodological weaknesses . . . call into question a treaty body's rigor of analysis and thereby weaken legal certainty. Lack of clearly discernible method means how a committee has reached its conclusions cannot be reconstructed, and its future approaches cannot be predicted. This result weakens both the work of the committees and the authority of the treaties they interpret.³¹⁴

In contrast to poor-quality Views which comprise “little more than a formulaic incantation of a justificatory mantra laid down in an earlier case,” a well-reasoned decision should: explain and justify the conclusion the UNTB has reached in a persuasive manner; indicate what remedy, if any, needs to be provided and the measures necessary to avoid future violations; and “give those faced with similar or analogous issues guidance as to what a particular human rights guarantee requires a government to do in practice, in order to comply with its obligations.”³¹⁵

With these criteria in mind, it becomes possible to begin assessing how the Committee's decision making in asylum cases could be improved, starting with the admissibility phase. Like other UNTBs, the Committee tends to consider admissibility and merits together. However, it is not always clear how consideration of a complaint is procedurally or substantively different between the two stages. In some cases, the Committee makes findings of fact and law at the admissibility stage. How would these findings have been different, if at all, had the complaints proceeded to consideration on their merits? And why have seemingly comparable cases resulted in different outcomes? The “somewhat erratic”³¹⁶ nature of the Committee's admissibility decisions undermines the transparency, predictability, and effectiveness of its complaint mechanism as a whole. To address these issues, the Committee might consider amending its Working Methods to provide more detailed guidance on how to resolve admissibility issues in a principled and consistent way, and ensure this guidance is publicly available.

The Committee should exercise particular care when dismissing complaints as manifestly ill-founded or not sufficiently substantiated under Article 4(2)(c) of the Optional Protocol. This is an opaque ground for inadmissibility, and an underexplored area of UNTB practice generally which warrants further scholarly and institutional attention.³¹⁷ The OHCHR indicates that a communication will be ruled inadmissible on this basis if an author has “not sufficiently developed the facts of [her] complaint or the arguments for a violation of the Covenant,” drawing parallels between this criterion and the rejection of cases as “manifestly ill-founded” by other courts.³¹⁸ But it is a vague standard, the content of which is difficult to discern from the asylum jurisprudence of the Committee itself. In the absence of clarity, the Committee leaves itself vulnerable to the suggestion that it relies on this ground to avoid making direct findings on more controversial matters, such as credibility.³¹⁹

The Committee itself has noted—on occasion—that issues pertaining to admissibility may be intimately linked to the merits of a case,³²⁰ and that “the threshold for admissibility should not be set too high” in view of the challenges women might face in obtaining documentary proof of GBV.³²¹ Given the gravity of the alleged harm to which women may be returned, the Committee may decide that Article 4(2)(c) should only be used sparingly, in cases where the lack of substantiation is flagrantly manifest. Even in such cases, the Committee should articulate with clarity and specificity the deficiencies in the complaint which led it to be ruled inadmissible. Overall, the Committee's admissibility findings should permit both the parties and external observers to understand exactly why the complaint failed, and how future complaints might be formulated to improve their chances of clearing the hurdle of admissibility.

For complaints that proceed to merits, the Committee should clarify what is required to establish a violation of CEDAW in an asylum case. From the Committee's jurisprudence to date, including the recent Bandboni case, it appears that an author must prove both that she would be exposed to a real, personal, and foreseeable risk of serious forms of gender-based harm if returned to her country of origin and that the asylum process of the respondent state was tainted by bias, gender-based stereotypes, arbitrariness, or a denial of justice. If this is the case, the Committee should state so clearly, perhaps in a General Recommendation. It should provide principled guidance on the content, standard, and burden of proof for establishing deficiencies in a respondent state's asylum process. It might also consider updating the template communication form and guidance note on its website to make it clearer to potential complainants what submissions are required from them.³²² As discussed above, it would also be helpful for the Committee to provide further guidance on the contours of the CEDAW non-refoulement obligation and how it compares with those established under other treaties.

The following sections detail further specific ways in which the quality of Committee decision making could be improved, including by rethinking its deferential standard, adopting a more proactive approach to credibility issues, and navigating the unique challenges of the refugee context. More generally, the Committee should approach future asylum cases with an eye to the broader normative impact of its Views beyond the outcome of the matter at hand. While the Committee is not bound by any rule of precedent, any (unexplained) lack of coherence or consistency—whether between its Views in comparable cases, or between its Views and other work such as General Recommendations and Concluding Observations—undermines its persuasiveness.³²³ Limitations of time and resources may put significant pressure on Committee members to produce Views quickly, but there is a balance to be struck between speed and quality. Regardless of whether a final decision is correct, formulaic, brief, and poorly explained reasons threaten the legitimacy and persuasiveness of the Committee's complaints work and prevent external parties from commenting on the correctness of its approach to individual issues or cases.³²⁴ Even if it is possible for the Committee to dispose of a complaint in relatively brief and straightforward terms, there is value in it identifying any other aspects of the state's asylum practices or submissions which involve potential discrimination against women. Finally, Committee members should be encouraged to attach separate or dissenting opinions wherever they might assist to understand better the majority position or challenge entrenched modes of thinking and jurisprudence that perpetuate the subjugation of displaced women's rights and experiences to those of men.³²⁵

B. Rethinking the Deferential Standard

In order to improve the quality of its decision making and persuasiveness of its Views, the Committee must grapple with its recent tendency to grant respondent states an overwide margin of appreciation in asylum matters.³²⁶ As a preliminary step, the Committee will need to undertake an internal reckoning to determine how and why an excessively deferential standard came to assume such a decisive role in its asylum work. As noted above, this deference was largely absent from the Committee's asylum work until 2016, after which it became a dominant feature of all but two subsequent cases (R.S.A.A. and Bandboni). The reasons for this shift are not readily apparent from publicly available information. It is not possible to draw a clear causal link between changes in membership of the Committee and the rise in deference, as it is not known which members led the drafting of which Views, or even when they were written.³²⁷ It is striking, however, that most of the communications in the later period involved Denmark as the respondent state (a trend common to other UNTBs at the time),³²⁸ and possible that this fact had some influence. Sarah Scott Ford documents “contestation” by Nordic states against an increase in critical Views by UNTBs generally, which corresponds with the Committee's turn to passivity.³²⁹ She explains why features of the Danish asylum system render that state particularly averse to criticism that challenges the institutional authority of its decision makers, and contextualizes this “contestation” against broader “backlash politics towards the involvement of human rights institutions in migration matters,” both within and beyond Denmark.³³⁰ Ford does not claim that this backlash swayed UNTB decision making, but her analysis does offer a possible explanation for the change in the Committee's approach which warrants further scholarly investigation and institutional reflection by the Committee itself.

If the Committee is to be a credible and persuasive authority on CEDAW, its decision making in the context of individual complaints must be informed by appropriate legal methods and transparent analysis.³³¹ Changes in membership and responsibility for drafting Views, and external political pressures, should not influence the outcome of individual complaints. By contrast, scholarship and practice suggest that it is acceptable for the Committee's work to be informed by the “fourth instance” doctrine.³³² According to this “general rule,”³³³ well-established across the jurisprudence of the UNTBs³³⁴ and other international bodies,³³⁵ the tasks of assessing and making findings of fact and law fall primarily to the national authorities of states parties. UNTBs do not conduct a de novo review of matters referred to them. Instead, the standard of review conducted by UNTBs is generally limited to reviewing national processes and decision making to ensure that they are not tainted by clear arbitrariness or a denial of justice.³³⁶ Due to its specific mandate, the Committee also reviews whether national processes are “biased or based on gender harmful stereotypes that constitute discrimination against women.”³³⁷ If a relevant error is detected, the UNTB will usually refer the decision back to the state party to be made in accordance with the relevant treaty, rather than replace the state's findings with its own.

In applying this doctrine, a human rights body's deference to national authorities should not be so extreme as to preclude it from rigorous performance of its oversight responsibilities. In the context of a specific case there may be a fine line between sufficiently interrogating and according due (“considerable”³³⁸) weight to national decision making, but an appropriate balance between the two must be struck. Other supranational human rights bodies provide examples of how this might be done. For example, in R.C. v Sweden in 2010, the ECtHR affirmed that:

as a general principle, the national authorities are best placed to assess not just the facts but, more particularly, the credibility of witnesses since it is they who have had an opportunity to see, hear and assess the demeanour of the individual concerned.³³⁹

However, it then proceeded to conduct its own analysis of how the Swedish authorities dealt with evidence and reached conclusions about credibility, before making different findings and ruling that the applicant's deportation would violate the ECHR.³⁴⁰ The Human Rights Committee, Committee against Torture, Committee on the Rights of the Child, and Committee on the Rights of Persons with Disabilities have similarly interrogated and departed from states parties’ findings of fact, law and credibility in asylum cases, including in cases involving women who fled to Denmark to escape GBV.³⁴¹ Like the Committee, these bodies are liable to criticism due to inconsistency in the application of this deferential standard.³⁴² Nevertheless, their approaches to comparable issues may serve as useful guidance for the Committee.

In determining when and to what extent deference is appropriate, it is important to recall that the Committee is a specialized body of experts overseeing the only gender-specific complaints mechanism within the international human rights system. While it might be appropriate to defer to national authorities on general fact-finding, the Committee should be robust in its interrogation of any findings of fact or credibility relevant to an author's status as a woman, and alert to any aspect of an asylum process that fails to meet the appropriate standards of gender-sensitivity. Where return is proposed to a CEDAW state party, the Committee's own detailed and specific understanding of the situation in that country, as obtained through its reporting procedure and reflected in Concluding Observations, should be given due weight. The Committee has previously appeared reluctant to assume such an assertive role,³⁴³ but a failure to reengage critically with national asylum systems may jeopardize its credibility as the oversight mechanism for CEDAW.

Çalı et al. note that while gender-bias and stereotyping may be “pervasive” in national asylum decision making, the Committee has yet to find any examples of it, and suggest that it “could use this tool as a basis for deeper scrutiny into national systems, should it wish to develop a more proactive role.”³⁴⁴ The preceding analysis of the Committee's asylum work supports this recommendation. While the Committee does not replace national authorities in determining asylum claims, it does need to do more to satisfy itself that the decisions of those authorities are not tainted by discrimination or bias. It should be alert to discrimination that may be hidden, implicit, or disguised behind apparently “gender-neutral” or “gender-blind” processes. Discrimination will not always be as flagrant as in the Philippine criminal cases, which involved remarkably blatant stereotypes about how a “rational,” “ideal,” or “ordinary” Filipina rape victim should act.³⁴⁵ National asylum authorities do not always publish detailed reasons for their findings, and where they do those reasons may not contain explicit discrimination. Cognitive biases and gender-based stereotypes and assumptions may be less obvious from the face of an asylum decision and “indirect” discrimination can occur in more nuanced ways.³⁴⁶

Finally, there should be no presumption that a state party's asylum system is free of discrimination, nor should the burden of establishing deficiencies in that system fall solely on a complainant. Instead, the Committee should be rigorous in its scrutiny of national laws and procedures, and there should be some onus on the state party to prove (not just claim) that it adopts a gender-sensitive approach at all stages of the asylum process. In Y.W., in 2015, the author argued that there “appears to be a gender-biased practice to regard the problems of single female asylum seekers as unsubstantiated and manifestly unfounded” in Denmark, and that “the Danish legal system is lacking gender sensitivity, as demonstrated by the fact that relatively more single female asylum seekers are denied asylum under the manifestly unfounded procedure and deported, without the right to appeal, than male asylum seekers.”³⁴⁷ To support these claims, she referred to a number of other cases involving women whose asylum claims had been treated as manifestly unfounded, and argued that they “indicat[ed] that gender-specific issues are treated less seriously than ‘male issues’ such as political repression.”³⁴⁸ She also claimed that male asylum seekers with comparable claims to her had received more favorable treatment.³⁴⁹ These allegations were never addressed or resolved by the Committee, but Denmark should have been required to address them in light of the prevalence of complaints against it and the Committee's own concerns about inadequate awareness and implementation of CEDAW in Denmark.³⁵⁰

C. A More Proactive Approach to Credibility Issues

Another issue in urgent need of attention from the Committee is how the credibility of women fleeing gender-based harm should be assessed. Credibility is one of the most fraught aspects of the adjudication of asylum claims, yet can be central to determining cases that are not sufficiently clear from the facts on record or where “objective” corroborating evidence is unavailable.³⁵¹ Longstanding UNHCR guidance states that “[c]redibility is established where the applicant has presented a claim that is coherent and plausible, not contradicting generally known facts, and therefore is, on balance, capable of being believed.”³⁵² In determining whether this standard is met, decision makers should take into account factors such as:

the reasonableness of the facts alleged, the overall consistency and coherence of the applicant's story, corroborative evidence adduced by the applicant in support of his/her statements, consistency with common knowledge or generally known facts, and the known situation in the country of origin.³⁵³

Crucially, UNHCR advises that credibility findings must have a foundation in evidence, and “should not be based on unfounded assumptions, subjective speculation, conjecture, stereotyping, intuition, or gut feelings.”³⁵⁴ Despite this advice, various studies have shown how assumptions do often inform national asylum decision making, including in cases where they are not supported by empirical evidence about human behavior.³⁵⁵ UNHCR has also stressed that an applicant's demeanor is “not a reliable indicator of credibility” and should not inform credibility findings because “[a] determination of credibility by reference to demeanour has a subjective basis that will inevitably reflect the values, views, experience, prejudices, and cultural norms of the decision-maker and is, therefore, at odds with the requirement of objectivity and impartiality.”³⁵⁶

The challenges of determining credibility in the asylum context are well-recognized,³⁵⁷ and often amplified for women. According to UNHCR, “brief, vague or apparently inconsistent responses” may be attributable to the fact that:

A woman . . . may lack experience of and confidence in communicating with figures of authority. A woman . . . may be unaccustomed to communicating with strangers and/or persons in public positions due to a background of social seclusion and/or social mores dictating that, for example, a male relative speaks on her behalf in public situations. In addition, it may be common for a female applicant to be deferential in her country of origin or place of habitual residence.³⁵⁸

While these reasons may underlie negative credibility findings in some cases, care is required not to essentialize refugee women as passive, “deferential” victims. Women face obstacles to establishing their credibility regardless of their level of confidence in dealing with national authorities. Indeed, in many cases, the issue lies not (just) with how the applicant presents her case, but with decision makers and asylum laws and structures. Singer puts it bluntly: “The key reason why women are refused asylum is because they are not believed.”³⁵⁹

Within the literature critiquing asylum credibility assessments generally, the “gendered nature of credibility assessment”³⁶⁰ and the gender credibility gap have received particular attention.³⁶¹ Credibility has a “disproportionate impact” on the asylum claims of women, and there are various reasons why, “[c]ompared to men, women encounter additional hurdles in showing that their asylum claim is credible.”³⁶² Singer observes that:

Women are more likely than men to have claims based on persecution suffered in the private sphere. Thus, due to the nature of the harm they have suffered, it may be more difficult for women to obtain documentary evidence of the agent of persecution and of their activities and place in society.³⁶³

Documentary evidence of sexual and other violence will be particularly difficult to produce. Stigma, fear, and shame may have prevented women from reporting abuse in their countries of origin, and medical examinations may be inconclusive or retraumatizing. Moreover, since women fleeing serious harm by non-state actors must prove that the state was unable or unwilling to protect them, official records are less likely to be available to present with asylum claims. Women may also struggle to produce documentary evidence relating to their actual or imputed political affiliations and administrative matters such as their (and their children's) identities and former residences.³⁶⁴

In these circumstances, a woman's personal testimony of abuse assumes greater significance. A high degree of caution is warranted when a state sits in judgment of a woman's account of sexual, domestic, or other violence, especially when this judgment occurs in “the absence of a shared world between the applicant and decision-maker.”³⁶⁵ Demeanor evidence is fundamentally unsound as a basis for detecting honesty or deceit, and should never be relied upon to ground a negative credibility finding.³⁶⁶ Other facially neutral indicators of credibility such as “coherence,” “plausibility,” “reasonableness,” and even “consistency” also require care. Drawing on scientific research, Noll argues that it is “outright inadvisable” to employ criteria such as coherence and consistency, and that “plausibility” is a similarly “useless” criterion.³⁶⁷ In the context of gendered harms, these criteria may mask a discriminatory evidentiary hierarchy which privileges male experiences and “credentialed knowledge.”³⁶⁸ Reasonableness and plausibility are inherently subjective tests, vulnerable to being shaped by the perspectives and experiences of individual decision makers and undermined by gender-based myths and assumptions. For example, decision makers might rely on assumptions about how women in violent domestic relationships “should” or “would” act.³⁶⁹ The expectation that credible testimony be “coherent” and “consistent” is likewise gendered. There are many reasons other than deceit why a woman might provide apparently incoherent or inconsistent evidence. The Committee has already cautioned states not to draw negative credibility inferences from late disclosures of sexual violence and trauma,³⁷⁰ but similar care is required when attaching meaning to other apparent discrepancies.

Given these inherent challenges in determining credibility in the asylum context, how should supervisory mechanisms such as the Committee approach the task of reviewing decision making at the state level? The interrogation of credibility findings is a challenge for all UNTBs,³⁷¹ but it is especially disappointing that the Committee has been reluctant to do so given its mandate and the specifically gendered nature of certain adverse assessments.³⁷² A less deferential approach would see the Committee draw on its specialized expertise to provide guidance on how gender-based harm may impact women's decision making, memories, testimonies, and engagement with asylum processes, and compel respondent states to reflect on their own possible biases. Given state resistance to Views that encroach upon “the discretionary domain of credibility,”³⁷³ these issues may be ripe for in-depth coverage in a new General Recommendation. Additionally, in the context of its Views in individual cases, the Committee should identify and interrogate the evidentiary bases (if any) for all negative credibility findings relevant to a woman's gender, and probe deeper into those findings to ensure they are not tainted—implicitly or explicitly—by discrimination, subjective assumptions about women or gender-based stereotypes. It should examine whether respondent states considered alternative explanations for apparently inconsistent or incoherent testimony, the respective weight they placed on evidence of violence as opposed to minor inconsistencies, and whether they adopted an intersectional approach to credibility findings.³⁷⁴

D. Navigating the Unique Challenges of the Refugee Context

Finally, if the Committee is to carve out a space for itself as an institution capable of upholding the rights of asylum-seeking women and girls, it will need to acknowledge and navigate the unique challenges of the refugee context which distinguish its work in this area from other cases. Three challenges are particularly noteworthy. First, as legal protections for displaced people have expanded, including through the reading of implied non-refoulement obligations into treaties such as CEDAW, some scholars have identified a “counternarrative” or “backlash trend” whereby economically advantaged states expand the use of non-entrée measures.³⁷⁵ Backlash may occur whenever “domestic political interests collide with developments at international institutions,”³⁷⁶ so it is unsurprising that it is observed in the context of migration control, an area which states have traditionally sought to guard within their domestic sovereign domain. As Ralph Wilde observes, “[j]ust as the scope of human rights legal protection in general, and the legal protection accorded to certain migrants in particular, has expanded, so too states have become less willing to provide such protection.”³⁷⁷ Backlash can also manifest as state rejection of the legitimacy of international bodies and their findings.³⁷⁸ While this is an issue common to all UNTBs, it may be particularly relevant to the Committee due to the relative novelty of the non-refoulement obligation read into CEDAW, the precise contours of which remain undefined.

A second and related issue, specific to the EU state parties to CEDAW, is the relationship between the asylum work of the ECtHR and Committee respectively. Ford identifies several relevant features of this relationship. First, she traces an increase in the ECtHR's use of the margin of appreciation (and thus a narrowing of access to remedies from that court) with an increase in the number of asylum seekers turning to other accountability bodies, such as the UNTBs.³⁷⁹ Second, she notes that some states might refer to the ECtHR's asylum jurisprudence to reject findings of UNTBs, particularly where the international human rights framework is more willing to scrutinize state action or provides greater protection for displaced people than its European counterpart.³⁸⁰ These observations highlight the potentially fraught international political environment in which the UNTBs undertake asylum monitoring with respect to states bound by EU law. For the Committee, a relative newcomer to international asylum adjudication, they emphasize the importance of clarifying the application of the CEDAW non-refoulement obligation, particularly in Dublin cases, and providing comprehensive, well-reasoned findings evidencing analytical rigor and an appropriate legal method. While quality decision-making alone may not be sufficient to ensure that any progressive developments spearheaded by the Committee are influential to, or even accepted by, states, without it there is little realistic chance of the Committee achieving its transformative potential.

Third, in reviewing state conduct for CEDAW compliance, the Committee must recognize the ways in which national asylum decision making differs from state action in other contexts. Scholars have critiqued refugee law's failure to “keep pace” with human rights law in recognizing women's rights, particularly in relation to domestic violence.³⁸¹ There are several possible explanations for this normative lag. Notably, migration control goes to the core of sovereign power, and states’ concerns about the movement of people into their territories may influence asylum decision making in women's cases in invidious ways. Of particular note is the fear of opening the proverbial “floodgates” to all women facing any form of GBV, which in turn may explain the privileging of “exotic” gender-based harm over more endemic or “everyday” harm such as domestic violence and rape.³⁸² As Siobhán Mullaly explains:

The widespread impunity of State and non-state actors for crimes of domestic violence brings into question the exceptional claim of the asylum seeker and raises the spectre of opening floodgates in response to a human rights violation that is both familiar and endemic.³⁸³

The “familiarity” of certain forms of GBV may prevent decision makers from recognizing them as persecution, since they also occur in the host state.³⁸⁴ National decision makers and courts may approach these asylum cases in a particular way, informed by sensitive local politics about domestic violence and gender equality within the state party itself.³⁸⁵ Indeed, drawing on the “racialised politics of protection,” Heaven Crawley argues that host states in the Global North prefer cases which allow them to “rescue” the “exotic other,” noting that “cases involving ‘everyday’ violence against women . . . threaten the self-identify of host states as progressive and as protectors of women's human rights.”³⁸⁶

The “floodgates” fear and preoccupation with “exotic” gendered harms contribute to a very specific form of structural violence against women seeking asylum. Stereotyping is prevalent in the asylum context, with the “Refugee Woman” expected to reproduce certain gendered “hegemonic narratives” or “act to a particular ‘script’” in order to establish her credibility, unlock protection, and be “saved.”³⁸⁷ Women are expected to present as “vulnerable victims” of male violence rather than “holders of rights for whom access is negated by patriarchal institutions and structures.”³⁸⁸ As Gerken observes:

successful asylum claims often depict women as passive and powerless victims who are seeking protection from a supposedly more advanced and egalitarian country… To ensure that they will get political asylum, women sometimes feel pressured to participate in a discursive nationalism which portrays Third World women as victims of their own culture.³⁸⁹

Indeed, some research suggests that “women are much more likely to secure protection where they experience ‘exotic’ forms of harm” and, conversely, they are “less likely to be successful where the risk is ‘only’ domestic violence.”³⁹⁰

Cases involving asylum seeker women also raise other complexities requiring the Committee's attention. The safe environment necessary for disclosure of violence “can be difficult to secure in the asylum context, where adjudicators may be preoccupied with factual details concerning country of origin information, travel routes, or alternative protections sought by the applicant.”³⁹¹ Some asylum-seeker women might need to overcome presumptions of non-credibility arising from their imputed “criminality” for having traveled or entered a country irregularly. And while much progress has been made, still more is required to shake off refugee law's androcentric origins and the “vision of a refugee as a courageous male dissident.”³⁹² Whether or not the Committee can reshape international decision making on asylum claims involving discrimination against women depends on its ability to engage directly with these peculiar features of the refugee context.

VII. Conclusion

After more than a decade of interpreting and applying CEDAW to asylum cases through its individual complaints mechanism, the time is ripe to assess the progress and achievements of the Committee. The expansion of CEDAW to include GBV, the reading of an implied non-refoulement obligation into CEDAW in M.N.N., General Recommendation 32 and the Views in three cases involving a violation of CEDAW (A., R.S.A.A., and Bandboni) provide a small but significant normative basis for the protection of rights of asylum seekers fleeing serious forms of discrimination against women. Yet the path to the Committee becoming a catalyst of radical reform for displaced women remains obstructed, both by the same normative and structural deficiencies which plague the UNTB system generally, and certain additional weaknesses in the Committee's own approach to asylum cases. Since 2016, in particular, its passivity in the face of potentially discriminatory national asylum practices has undermined the impact of some of its earlier achievements. Its deference to national decision making has been excessive and unprincipled. With limited exceptions, it has produced Views of relatively poor quality which leave key issues unaddressed, including in relation to credibility and interpretation of CEDAW. Overall, it does not appear to be fulfilling its mandate under the Optional Protocol in a way which is likely to influence states and other international bodies in a transformative way, or effectively address the protection concerns of displaced women.

In light of these criticisms, it is pertinent to ask whether the Committee is indeed the right body at the international level to pioneer improvements in how the asylum claims of women fleeing gender-based harm are processed? Despite the preceding analysis, there remain strong grounds to conclude that it is, albeit with some tempered expectations. Crucially, amongst international tribunals and bodies dominated by men,³⁹³ the Committee stands alone as “a space . . . that is headed by women decision-makers, whose remit is specifically gendered and whose task is to uphold the rights of women.”³⁹⁴ Deficiencies in the Committee's decision making to date reaffirm why the “add women and stir” approach is insufficient on its own to effect radical change,³⁹⁵ and why decision makers cannot be essentialized by their gender.³⁹⁶ Yet the value of an international body of predominantly female “experts of high moral standing and competence in the field covered by the Convention,”³⁹⁷ who can be expected to have greater expertise about gender biases, stereotypes, and discrimination against women than the average asylum decision maker, should not be diminished.

Moreover, the Committee boasts a level of geographically diverse female representation that is unparalleled in most other international bodies tasked with upholding the rights of displaced women.³⁹⁸ Members are elected with “consideration being given to equitable geographical distribution and to the representation of the different forms of civilization as well as the principal legal systems.”³⁹⁹ Again, Committee members cannot be essentialized by their national background, and in other contexts have been criticized for perpetuating artificial binaries between “Global North” and “Global South” women, and typecasting the latter as “passive victims of their own ‘culture.’”⁴⁰⁰ Nevertheless, while national diversity alone will not guarantee that intersectional identities such as “race,” class, ethnicity, etc. are fully incorporated into decision making, it does lend itself to “a perspective that recognizes that different women experience gender-based disadvantage or oppression differently.”⁴⁰¹ This composition may assist to challenge the historical reduction of refugee women to a “singular monolithic subject”⁴⁰² and mitigate the impact on asylum processing of “entrenched assumptions and behaviours that centre whiteness and see Western values as superior.”⁴⁰³

Of even greater significance than the composition of the Committee itself is the treaty underpinning it. As it approaches its half-century anniversary, CEDAW continues to be credited for advancing awareness of and respect for women's rights at both the national and international levels.⁴⁰⁴ As originally drafted, its provisions could be criticized for conceiving of women's rights in relation to those of men and excluding violence against women. Yet CEDAW has proved remarkably flexible in its ability to move beyond the strictures of formal gender equality and incorporate gendered harms such as GBV. Further evolutions are required,⁴⁰⁵ but if the Committee is willing to continue to interpret it with bold progressiveness, CEDAW has “a particularly potent transformative potential.”⁴⁰⁶ This potential could, under the right circumstances, extend to the asylum context, too.

At the normative level, CEDAW establishes an obligation of non-refoulement that has the (as-yet-unrealized) potential to operate as a powerful complementary source of protection for displaced women. Some ambiguity remains about the precise contours of this obligation. Does it merely require states to interpret their existing human rights obligations with a sufficiently attuned eye to gender issues, or does it go further than that in creating new and additional limits on states’ ability to remove women fearing serious gender-based discrimination in their countries of origin? Either way, it requires at a minimum that states parties review their national asylum legislation and procedures to ensure that serious forms of gender-based discrimination are expressly included as grounds for complementary protection, and that decision makers properly consider whether women would face real, personal, and foreseeable risks of serious forms of gender-based discrimination or violence if returned. Such consideration must be grounded in and informed by an evidence-based understanding of how women from different backgrounds experience discrimination and violence. Indeed, under a more radical normative agenda, the implied non-refoulement obligation could even challenge the gendered hierarchy of rights by extending protection against return to other serious or irreparable harms which disproportionately affect women, such as those relating to reproductive and women's health, motherhood, education, livelihoods, etc.

Of equal importance are the contributions the Committee can make toward understanding procedurally how asylum claims involving gender-based harm should be determined. It has acknowledged intersectionality in passing in General Recommendation 32, and sporadically modeled an intersectional approach to the cases of A. and R.S.A.A. But it could do much more to demonstrate what it means to “embrace the complexities of compoundedness” in assessing the asylum claims of people who are “multiply-burdened” by various grounds of discrimination (gender, race, etc.).⁴⁰⁷ Similarly, what specifically is required to adopt a “gender-sensitive” approach to every stage of an asylum process? What are common examples of indirect discrimination in national asylum procedures and how can they be remedied? How do women's experiences of persecution and serious harm differ from those of men? How should states determine the credibility of women's narratives of abuse and violence in a way that appropriately accommodates those experiences within a principled administrative or judicial framework? In cases involving GBV from non-state actors, what is required of a woman to establish that effective protection is not available in and from her home state? To what extent must she have sought the benefit of state protection before fleeing, and in what circumstances can it be deemed reasonable for her to have fled without addressing her concerns to national authorities?

Crucially, these substantive and procedural questions are not and should not be limited to the framework of CEDAW. They have direct relevance to how asylum claims involving gender-based harm are approached and determined under complementary legal frameworks, particularly under the Refugee Convention, the ICCPR, and the Torture Convention. If the Committee is to shake off its origins as the “poor cousin” of other UNTBs, it must not allow itself to be relegated to the margins as a niche or nascent review body. It must assert itself as the leading international institution with respect to the interpretation and implementation of the human rights of women. Reclamation of this role begins with improving the transparency, coherence, consistency, and predictability of its reasoning and Views. It needs to be, and be seen to be, producing quality decisions of high normative significance that are influential both amongst states parties to CEDAW and in their contributions to international law processes and institutions more generally. In this regard, the Committee should not shy away from a more “adventurous” use of its complaints mechanism to establish case law and norms which are, to a certain extent, either incompatible with or more progressive than existing standards, particularly where doing so challenges jurisprudence that perpetuates the marginalization of women's experiences.⁴⁰⁸ It should recognize that its contributions to the broader corpus of human rights “soft law” may “in turn influence the determinations of other treaty bodies and . . . help ‘mainstream’ women's rights.”⁴⁰⁹

Of course, the Committee cannot achieve these results simply by pronouncing them. Like all UNTBs, it must navigate complex relationships with states parties and other key actors, and the more radical or controversial its pronouncements the more difficult this navigation may be.⁴¹⁰ States which are genuinely committed to achieving equality for women and eradicating GBV should ensure that they are parties to CEDAW⁴¹¹ and paying close attention to the Committee's guidance. They should also ensure that the Committee receives the resources and support necessary for it to fulfill its mandate. There is a role for the broader international community of actors, too. As Lutz Oette argues, “legal recognition and/or strong institutional mechanisms at the national level are pivotal in enhancing the prospect of implementation,” and:

In the absence of a strong legal mandate and national mechanisms of this kind, treaty bodies will have to rely on their legitimacy, persuasion, and often the intervention of other actors, be it States, regional or international organizations, or civil society, including transnational networks, to add weight to their interpretations, findings, and recommendations.⁴¹²

The lack of scholarly or judicial engagement with CEDAW's non-refoulement obligation, and the Committee's broader work, contributes to its relegation to the margins of normative development within the international system. Moving forward, the Committee's jurisprudence and other work should be subject to greater scholarly attention and critique. When listing the international law sources of the prohibition on refoulement, scholars, advocates, and jurists should include CEDAW alongside more traditional sources such as the Refugee Convention, Torture Convention, and ICCPR. Finally, if the Committee's recent turn to passivity in asylum cases can be reversed, advocates may consider citing its work when raising gender-related claims before national courts and international bodies, both to raise awareness of it and to promote judicial and inter-institutional dialogue on key concepts and norms. Such a focus on the transformative potential of CEDAW would support the Committee in promoting tangibly better outcomes for displaced women and girls.