AILA Comments to Proposed Regulations on Gender-Based Asylum

January 17, 2001

Policy Directives and Instructions Branch
Immigration & Naturalization Service
425 I Street N.W., Room 4034
Washington, D.C. 20536

Re: 8 CFR Part 208, Asylum and Withholding Definitions
Proposed Rule, Federal Register Vol. 65, No. 236, at 76588, December 7, 2000
INS No. 2092-00

Dear Director:

The American Immigration Lawyers Association (AILA) is writing to comment on proposed regulations published in the Federal Register on December 7, 2000. See 65 FR 76588-76598 (Dec. 7, 2000). The American Immigration Lawyers Association (“AILA”) is a voluntary bar association of 7,000 attorneys and law professors practicing and teaching in the field of immigration and nationality law. AILA’s members represent and advocate on behalf of many thousand asylum seekers each year, and AILA remains committed to justice and fairness in the application of our nation’s asylum laws and regulations.

AILA believes that these regulations are a worthy effort to clarify principles for interpreting and applying some components of the statutory definition of “refugee, ” in the context of claims for asylum and withholding of removal. We commend the U.S. Attorney General and the Immigration and Naturalization Service for their efforts. We believe that these proposed rules are an important step in clarifying the criteria for assessing claims based on membership in a particular social group, especially in cases involving domestic violence, to ensure that those applicants who qualify may obtain asylum. The regulations make clear, as is well-established in long-standing precedent, that gender can define a particular social group.

We also believe these proposed rules are, in many respects, consistent with our international obligations under the UN Convention related to the Status of Refuges and Protocol and with evolving U.S. jurisprudence. They help correct clear errors and inconsistencies with established law found in the majority opinion of the Board of Immigration Appeals in Matter of R-A-, Interim Decision 3403 (1999).

Nonetheless, there are several areas where the proposed rules create confusion and potential confusion in interpretation. We believe the proposed language should be simplified and clarified. In some cases, proposals addressing certain areas of the law should be set aside at this time. Some of the language should be redrafted in specific ways as suggested below to maintain the focus on correcting errors in R-A-, including and most importantly clarifying the criteria for the particular social group ground. These comments are intended to identify several areas of concern as well as to propose alternative language consistent with the refugee definition, evolving jurisprudence and the U.S. statute.

208.13; Establishing asylum eligibility
(b) (1) “Past Persecution”

The regulation that has been in operation, at least until the recent changes made final on January 5, 2001, provided a workable framework for analyzing past persecution. The final regulations, among other changes, modified the presumption in past persecution cases, where the applicant’s fear of future persecution is unrelated to the past persecution suffered. These proposed regulations make an additional change, stating that the presumption of past persecution does not “relieve the applicant of the burden of producing testimonial evidence or where reasonably available to the applicant, documentary evidence relating to future persecution, including to a fundamental change in circumstances or the reasonableness of internal location.”

This change is not a necessary part of correcting the clear errors in R-A-, or creating a framework for analyzing the particular social group ground, which is the main stated goal of these proposed regulations. In addition, this language is inconsistent with the language of the statute; it not only deepens the tension between the regulations and the statute, but also between the regulations and international trends as well as the opinions of international authorities.

As has been held by the Board, by numerous courts and as is clear on the face of the statute, past persecution is a separate basis for asylum eligibility under U.S. law. (Law of Asylum; Cites). Even the current regulations, especially with the changes made in the recent new final regulations, are potentially problematic under the statute, since they do not put the past and future persecution standards on equal footings but only create a presumption in past persecution cases; that presumption is not even operative in all cases, but an exception is made in cases where the past and future persecution are unrelated. Under the law of many states parties to the UN Refugee Convention where the term refugee is exclusively defined in terms of a future, well-founded fear of persecution, past persecution still creates a strong evidentiary presumption of future persecution. (Hathaway).

The presumption of future persecution, based in U.S. law on a statutorily mandated distinct basis for asylum eligibility, should not be further changed, weakened or effectively eliminated. The proposed language seems to put the burden back on the applicant in past persecution cases, requiring the applicant to present proof and to establish a well-founded fear of future persecution. The presumption - again, grounded in a statutorily mandated distinct basis of eligibility - would be effectively eliminated, if the applicant had the burden to prove future persecution; this is what the proposed regulation on its face states. If the intent is only to make this the applicant’s burden in cases where the presumption does not arise (where past and future persecution claims are unrelated), at a minimum this limitation should be clearly stated.

Our recommendation is that this proposal be eliminated altogether. The recent new final regulations made a fundamental (and potentially problematic) change, specifically in cases of past persecution. Additional changes at this time are confusing, and these particular changes are substantively confusing and create additional inconsistencies with the statute.

208.13 (a) Definitions; “Persecution”

We commend DOJ/INS for recognizing that persecution does not require a subjective, malignant, or punitive intent to cause harm. Kasinga; Pitcherskaia v. INS, 118 F.3d 641 (9th Cir. 1997). However, the proposed rule contains a definition of persecution that departs from established case law. The proposed definition of persecution would be: “the infliction of objectively serious harm or suffering that is experienced as serious harm or suffering by the applicant, regardless of whether the persecutor intends to cause the harm.” Proposed 8 C.F.R. §208.15(a) (emphases added). Although we agree that persecution does not encompass trivial harms, we fear that the proposed addition of the word “serious” may be interpreted by adjudicators to heighten the level of harm necessary for a finding of persecution.

More helpful guidance for adjudicators would make reference to the international treaties and human rights standards, serious violation of which amounts to persecution. See, e.g., UK Gender Guidelines 2A.15 (“Whether particular treatment amounts to ‘serious harm’ should be decided on the basis of international human rights standards.”); UNHCR Handbook ¶ 51 (“serious violations of human rights...would constitute persecution.”). Accordingly, we request that the language be changed to read that “persecution is the infliction of suffering or harm upon persons who differ in a way regarded as offensive.” We further urge that adjudicators seeking to determine if persecution has taken place be referred to international human rights standards.

Under now accepted international understandings of the term “persecution,” persecution is harm, measured by international human rights standards, plus the absence of state protection. The proposed regulation takes an important interpretative step, consistent with this accepted doctrine, in emphasizing that “absence of state protection” is inherent in the term persecution, as well as in reinforcing the doctrine that the agent of persecution can be a non-state actor.

Intent: The proposed regulatory language, however, is confusing in some respects. First, we think it would be clearer and more direct to eliminate the word “infliction” from the description of persecution; persecution is the measure of the harm and the word infliction could imply an evaluation of intent, which is clearly not intended by the proposed regulation.

“Objective” persecution: Second, and perhaps most importantly, the use of “objective” and “subjective” in defining persecution itself causes confusion since these are terms which have specific meanings in the interpretation of well-founded fear. The use of the term “objective” in the definition of persecution itself could readily imply some heightened standard or measure of proof, which, as noted, could undermine the well-founded fear standard itself; that standard allows for various forms of proof, including testimonial evidence, in establishing the level of risk. The standard of proof and the substantive elements of the definition, in particular persecution, should be clearly distinguished.

Instead of “objectively” - an abstract and potentially confusing concept, leading to potentially ad hoc judgments - we think it is critical to spell out by what “objective” standards the harm which constitutes part of the dual meaning of persecution (i.e., harm plus absence of state protection), must be measured. Numerous cases, other U.S. authorities such as the INS Gender Guidelines and the INS Basic Law Manual, as well as an overwhelming international consensus, have established that the harm inherent in persecution must be measured by international human rights standards. As cited in the preface to the proposed regulations, this is what congress meant when it first described persecution as harm that is offensive under the rules and standards of “civilized nations.”

As substitute language we would suggest that this part of the proposed regulation state that “Persecution constitutes harm plus the absence of state protection. Harm must be measured by accepted international human rights standards. The subjective intent of the agent of persecution is not relevant to the term persecution, or any other criterion of the refugee definition.”

Absence of state protection: Third, we believe there should be modification in the proposed language describing the absence of state protection as involving the government’s unwillingness or inability to control the infliction of harm or suffering and the reasonableness of the government’s steps in this regard. The standard for protection must be contextualized in light of the particular applicant’s well-founded fear. As the New Zealand authorities have held, the state fails to protect when it does not bring the risk of harm from the non-state actor to a level below that of a well-founded fear.

The proposed language could be interpreted to permit an abstract assessment of whether the state has taken reasonable steps for protection. Reasonable steps in the abstract are not relevant to refugee status, which makes well-founded fear the centerpiece of the risk assessment. Refugee protection is not concerned with abstract judgements about a state’s conduct or abstract assessment of its system of protection for its citizens generally. A state may in some abstract sense have taken “reasonable” protection steps (whatever that means), but those reasonable steps are not relevant if in the concrete situation, the applicant, despite these reasonable steps and “reasonable” access, continues to have a well-founded fear of harm, or human rights abuse. Of course the state is not required to eliminate all risk. It is, however, required to reduce the risk for the particular applicant to below the level of a well-founded fear.

Although the proposed regulation notes that a state’s inability to protect may constitute an absence of state protection, the term reasonable and other language of the proposed regulation creates confusion and undermines this critical point of doctrine. The word “reasonable” is subject to differing interpretations and could be misunderstood as denying status when circumstances such as war, civil disturbance or natural disaster make a government unable to protect, although it is willing to do so. We would suggest, therefore, that the third sentence of the proposed 208.15 (a) be eliminated. In the alternative we would suggest that the word “effective” be substituted for “reasonable,” and that the regulation state that “effective means that the applicant’s fear be reduced to below the level of a well-founded fear, and that it includes circumstances where the government is willing but nonetheless unable to provide protection.”

The list of examples (“evidence” of unwillingness or inability to protect) is helpful, but it should be underscored that this list is not exclusive. In addition, all of the examples given involve a government which in some sense is unwilling (as opposed to unable) to provide protection. Other examples of a state or a government’s inability to protect should be incorporated. This would include: “a pattern of lack of protection;” “unavailability (not just denial) of services”; other evidence that the government is unable to provide protection.

208.13 (b) “On Account of” the applicant’s protected characteristic

Imputed Political Opinion. The proposed rule codifies the doctrine of imputed political opinion, long recognized by the U.S. Supreme Court, the federal courts, the Board of Immigration Appeals, and the INS. See Elias-Zacarias v. INS, 502 U.S. 478, 482 (1992)(“nor is there any indication (assuming, arguendo, it would suffice) that the guerrillas erroneously believed that Elias Zacarias’ refusal was politically based”); Canas-Segovia v. INS, 970 F.2d 599, 601 (9th Cir. 1992)(Canas II)(“Imputed political opinion is still a valid basis for relief after Elias-Zacarias”);Surinder Singh v. Ilchert, 69 F.3d 375, 379 (9th Cir. 1995); In re S-P-, Interim Dec. 3287, at 1, 5-6 (BIA 1996). At the same time it recognizes that the principle of “imputation” applies to the other grounds of protection. The proposed regulation provides that an applicant may satisfy the “on account of” requirement if the persecutor acts against the victim “on account of what the persecutor perceives to be the applicant’s race, religion, nationality, membership in a particular social group, or political opinion.” 8 C.F.R. Sec. 208.15(b).

We believe that the codification of the imputed political opinion doctrine and its extension to the other grounds of protection is important. We also feel that it is important to recognize explicitly that a victim of domestic violence may be persecuted or fear persecution not only because of her membership in a particular social group, but also because of her race, religion, nationality or, perhaps most importantly, her political opinion, see, e.g., In re S.A., Interim Decision 3433 (June 27, 2000), or because of what her abuser perceives to be her race, religion, nationality or political opinion.

Although the INS and the Board have been relatively receptive to recognizing claims based on domestic violence where the Respondent can show a nexus to race, religion or nationality, they have been less willing to recognizing claims grounded in the victim’s political opinion. The denial of women’s rights takes many forms, including, bride burning, FGM, discriminatory treatment in access to employment, education and public life, and violence in the home. When women raise objections to these practices, whether through public expression, organizational activity, defiance of the law, or, in the case of domestic violence, through acts of resistance to individual experiences of oppression, this opposition should be recognized as the expression of a political opinion. See Anker, Law of Asylum in the United States (Refugee Law Center, 1999) at pp. 370-371. See also Canadian Guidelines at 807.

The INS Gender Guidelines describe Fatin as “mak[ing] clear that an applicant who could demonstrate a well-founded fear of persecution on account of her (or his) beliefs about the role and status of women in society could be eligible for refugee status on account of political opinion.” With regard to claims based on domestic violence and in order to avoid misinterpretations of the Board’s decision in In re R-A-, Interim Decision 3404 (BIA 1999), the final regulations should incorporate language recognizing that a belief in gender roles can constitute the expression of a political opinion. This addition is consistent with case law regarding political opinion. See Fatin v. INS, 12 F. 3d 1233, 1242 (3d Cir. 1993)(“[W]e have little doubt that feminism qualifies as a political opinion within the meaning of the relevant statutes.”).We would propose that the final regulations add the sentence “When a persecutor is motivated to act based on what he perceives to be the victim's beliefs concerning gender, sexual orientation or gender roles, that this can constitute persecution on account of political opinion” after the sentence “An asylum applicant must establish that the persecutor acted …on account of what the persecutor perceived to be the applicant’s race, religion, nationality, membership in a particular social group, or political opinion.” We would also add the sentence “a belief in women's rights, whether expressed in word or action, in public fora or in the home, constitutes a political opinion.”

Mixed Motives. We believe it is important that the proposed regulations codify the “mixed motives” principle well-established in case law that as long as the persecutor acts “at least in part” because of a protected characteristic, the “on account of” requirement has been satisfied. Nonetheless, we believe that the language in the proposed rule, while recognizing this standard would, in fact, make it more difficult than it already is to show the motive of the persecutor. The proposed standard would require that the asylum seeker establish that “the applicant’s protected characteristic [be] central to the persecutor’s motivation to act against the applicant”. [Emphasis added.] We believe that this language is not consistent with existing case law and contrary to the principle that decisions should be made keeping in mind “the fundamental humanitarian concerns of asylum law.” Matter of S-P-, 21 I&N Dec. 486 (BIA 1996).

It would be a major step backwards in evolving jurisprudence and would have far broader implications than for the asylum seekers targeted by these proposed regulations. The proposed standard would affect any claim with mixed motives, including extortion and bribery cases, domestic violence cases, and cases where the government interrogates political opponents through torture to gain information. It would place an impossibly high evidentiary burden on asylum seekers to prove their attackers’ motivations, and result in judges and asylum officers denying asylum to individuals who could not prove their attackers’ central motivations. These are precisely the types of situations the mixed motives case law was designed to remedy.

Alternatively, we would propose compromise language that the proposed rule read that “In cases involving a persecutor with mixed motives, the ‘on account of’ requirement is satisfied if the persecutor acts ‘at least in part’ because of a protected characteristic.” This standard is consistent with Matter of S-P-, infra, Matter of T-M-B-, 21 I&N Dec. 775 (BIA 1997),overruled on other grounds sub nom. Borja v. INS, 175 F.3d 732 (9th Cir. 1999)(en banc) and appellate court decisions. It is worth noting that in Matter of T-M-B-, the Board applied the standard requiring that the persecutor act “at least in part” because of a protected characteristic, and yet the majority still found that the Respondent, despite presenting evidence of past political activities, had not demonstrated that he was persecuted for any thing other than pecuniary motives.

Evidence of Persecutor’s Intent to Act Against Others. One of the most disturbing aspects of Matter of R-A- was the majority’s statement that the violence against the applicant was not “on account of” a particular social group because there was no evidence that the applicant’s husband would harm any other member of the asserted social group. See Matter of R-A- at 19-20. The Board’s reasoning demonstrated that it did not understand the nature of domestic violence or the very public nature of this private harm in societies. Nor was it willing to consider the social context and cultural meanings of the harm inflicted, as it had in Matter of Kasinga, Int. Dec. 3278 (1996). It constituted a setback for the principle, established in the INS gender guidelines, that the claims to women to refugee protection must be considered equally and that women’s rights, including women’s rights within the family, are human rights, entitled to protection. We believe that the language in the proposed rules that such evidence, while relevant, is not required, is extremely important in establishing a workable framework for deciding asylum claims based on domestic violence. The analogy in the preamble of the slave owner who only beats his own slave, but not his neighbor’s is a powerful one and helpful in dealing with this thorny analytical issue. This change is a positive one and will clarify the standard and assist adjudicators in reaching a decision.

Section 208.15(c) Membership in a Particular Social Group:

We applaud the clarification within the introductory comments and the proposed regulations addressing the application of the “particular social group” ground of the refugee definition, and the recognition of the Board’s decision in Matter of Acosta as the key decision on the issue. See Section 208.15(c)(1). The proposed regulations effectively address long‑standing confusion brought about by the Ninth Circuit’s decision in Sanchez‑Trujillo v. INS, 801 F.2d 1571 (9th Circuit), by adopting the clarification set out by that court in Hernandez‑Montiel v. INS, 225 F.3d 1084 (9th Cir. 2000) and reconciling the Ninth Circuit’s interpretation with Acosta. This interpretation is consistent with U.S. case law and with international understanding of the particular social group language of the Refugee Convention. See e.g. Fatin v. INS, 12 F3d 1233, 1242 (3d Cir. 1993); Canada (Attorney General) v. Ward (1993) 2 S.C.R. 689; Islam v. Secretary of State for the Home Department, 2 App. Cas. 629 (H.L. 1999)(United Kingdom); Re GJ, Refugee Appeal No. 1312/93, Refugee Status Appeals Authority (Aug. 30, 1999)(New Zealand). The introductory comments provide extremely helpful guidance for the application of the particular social group ground, as does the non‑exclusive list of factors to be considered in determining the existence of a particular social group contained in Section 208.15(c)(3) of the proposed regulations.

However, Section 208.15(c)(2), addressing the question of when past experience defines a particular social group, imposes an improper standard which is not in conformance with U.S. case law or international interpretations of the particular social group definition. This section, which provides that, to form the basis of a particular social group, a past experience must have been fundamental to the individual’s identity at the time it occurred, confuses the characteristics which define present association for purposes of particular social group membership and those defining past association. The key factor in determining whether a characteristic is one which defines a particular social group under the refugee definition is whether that characteristics is immutable ‑ either beyond the power of the individual to change or so fundamental to the individual’s identity or conscience that he or she should not be required to change it. A past experience, by virtue of its historical nature, is unchangeable. It is, without more, an immutable characteristic. The Board in Matter of Acosta addressed this issue by offering a past shared experience as one example of an immutable characteristic:

Applying the doctrine of ejusdem generis, we interpret the phrase “persecution on account of membership in a particular social group” to mean persecution that is directed toward an individual who is a member of a group of persons all of whom share a common immutable characteristic. The shared characteristic might be an innate one such as sex, color, or kinship ties, or in some circumstances it might be a shared past experience such as former military leadership or land ownership. Acosta at 233.

The Supreme Court of Canada, in Canada (Attorney General) v. Ward (1993) 2 S.C.R. 689, 739, adopted a similar standard, providing three general categories of particular social groups:

1) groups defined by an innate or unchangeable characteristic;

2) groups whose members voluntarily associate for reasons so fundamental to their human dignity that they should not be forced to forsake the association; and

3) groups associated by a former voluntary status, unalterable due to its historic permanence.

See also Fatin v. INS, 12 F3d 1233, 1242 (3d Cir. 1993); Islam v. Secretary of State for the Home Department, 2 App. Cas. 629 (H.L. 1999)(United Kingdom); Re GJ, Refugee Appeal No. 1312/93, Refugee Status Appeals Authority (Aug. 30, 1999)(New Zealand).

As an explanation for the imposition of the additional requirement on past association, the introductory comments to the proposed regulations offer past membership in a violent gang as an association which should not define a part social group for asylum purposes and state that “it is reasonable for any society to require its members to refrain from certain forms of illegal activity.” These legitimate concerns are amply addressed through other provisions in the Immigration Act. For example, an individual sought by his or her government because of prior illegal activity would likely be found to face prosecution rather than persecution upon return to his or her country and would be denied asylum on that basis. See e.g. Abedini v. INS, 971 F.2d 188, 191 (9th Cir. 1992). In addition, an individual who has engaged in nonpolitical criminal activity would be denied asylum under Section 208(b)(2).

Because proposed Section 208.15(c)(2) is not consistent with U.S. or international case law, and because the concerns it is intended to address are dealt with elsewhere, this section of the proposed regulations should be struck.


We hope that these comments assist the Service in examining the important issues involved in the proposed asylum regulation. AILA remains available to discuss with the Service the issues contained in these comments or in any aspect of the asylum program.

Respectfully submitted,

Jeanne A. Butterfield, Executive Director


Cite as AILA Doc. No. 01011801.