First Steps

On June 16, Attorney General Merrick Garland finally, mercifully vacated three decisions that formed a key part of the Trump administration’s unrelenting attack on the law of asylum.1  Matter of A-B-,  issued by Jeff Sessions in June 2018, took aim in particular at victims of domestic violence.2  Matter of L-E-A-, issued the following year by William Barr, sought to undermine protection for those targeted by gangs due to their familial ties.3  And on January 14, 2021, six days from the end of the Trump Administration, acting A.G. Jeffrey Rosen issued a second decision in A-B-, gratuitously criticizing the method for determining nexus in asylum claims employed by the U.S. Court of Appeals for the Fourth Circuit, while conveniently evading that court’s review of the original decision in the case through remand.4

Garland’s action restores the law to where it stood prior to June 11, 2018, but only for the time being.  Proposed rules on the subject (which Garland referenced) are due by October 30, when they will first be subjected to a period of public comment.  If final rules are eventually published, it will occur well into next year.

As we sigh in collective relief and celebrate the first steps towards correcting our asylum laws, let’s also take note of the imperfect place in which the case law stands at present.

As to domestic violence claims, the BIA’s 2014 decision in Matter of A-R-C-G- (which Matter of A-B- had vacated) has been restored as binding precedent.5  That decision was issued at a time when (as now) regulations addressing particular social groups were being contemplated by DHS and EOIR.6  While A-R-C-G- was an extremely welcome development, the Board used it to recognize a rather narrowly-defined group: “married women in Guatemala who are unable to leave their relationship.”  In a footnote to the decision, the Board declined to address the argument of several amici (including UNHCR) that a particular social group may be defined by gender alone.  Although A-R-C-G- led to many grants of asylum, some immigration judges relied on the limited scope of the group’s definition to deny claims involving slightly broader variations, in particular, where the victim was not legally married, but nevertheless in a domestic relationship that she was unable to leave.  While the BIA reversed some of those denials in unpublished decisions, it declined to speak to the issue through binding precedent.

As to Matter of L-E-A-, Garland’s recent action returns us to the BIA’s original opinion in that case.7  While the decision acknowledged that families constitute particular social groups (a point that was not in dispute, having been universally recognized for some 35 years and stipulated to by DHS), the BIA still denied asylum by invoking a legally incorrect standard for establishing nexus that it has continued to apply in all family-based asylum claims.

For these reasons, the content of the forthcoming regulations will be extremely important in determining the future of asylum in this country.  While a return to the test for social group cognizability expressed in the BIA’s 1985 precedent in Matter of Acosta tops most regulation wish lists, I will focus the discussion here on a couple of more specific items necessary to correct the shortcomings of Matter of A-R-C-G- and Matter of L-E-A-. 

First, the regulations need to explicitly recognize that a particular social group may be defined by gender alone.  In its 2002 Gender Guidelines, UNHCR identified women “as a clear example of a social subset defined by innate and immutable characteristics, and who are frequently treated differently than men,” and whose “characteristics also identify them as a group in society, subjecting them to different treatment and standards in some countries.”8  However, over the nineteen years since those guidelines were issued, the BIA has consistently avoided considering the issue.

The peril of defining gender-based groups in the more narrow manner employed by the BIA has been addressed by two distinguished commentators, who explain that such practice results in “constant re-litigating of such claims,” sometimes creating “an obstacle course in which the postulated group undergoes constant redefinition.”9  And of course, that is exactly what has happened here, as A-R-C-G- gave way to A-B-, which led to differing interpretations among different courts until Garland’s recent reset.  The above-mentioned commentators further decried the “nitpicking around the margins of the definition” resulting from the narrow approach when the true reason for the risk of persecution to the applicant “is simply her membership in the social group of ‘women.’”10  Regulations recognizing gender alone as a particular social group would thus provide clarity to judges and asylum officers, eliminate the wastefulness of drawn out litigation involving “nitpicking around the margins,” and bring our laws into line with international standards. 

But as L-E-A- demonstrates, recognition of a group alone does not guarantee asylum protection.  In order for a group’s recognition to be meaningful, the regs must also address an ongoing problem with the BIA’s method for determining nexus, or whether persecution is “on account of” the group membership.

The BIA is accorded deference by Article III courts when it reasonably interprets immigration laws, provided that the meaning of the language in question is ambiguous.  However, the “on account of” standard included by Congress in defining the term “refugee” is quite clear; its meaning is long established, and in fact, is not particular to immigration law.

The Supreme Court referenced this standard last year in a non-immigration case, Bostock v. Clayton County.  The Court explained that the test

incorporates the “‘simple’” and “traditional” standard of but-for causation.... That form of causation is established whenever a particular outcome would not have happened “but for” the purported cause....In other words, a but-for test directs us to change one thing at a time and see if the outcome changes. If it does, we have found a but-for cause.11

In a 2015 decision, the U.S. Court of Appeals for the Fourth Circuit applied this exact test in the asylum context to conclude that persecution was on account of family, determining that the petitioner’s “relationship to her son is why she, and not another person, was threatened with death if she did not allow him to join Mara 18.”12  But for some reason, the BIA has felt entitled to reject this established standard outside of the Fourth Circuit in favor of its own excessively restrictive one.

Had the proper test for nexus been employed in L-E-A-, asylum would have been granted.  Under the facts of that case, once the familial relationship is removed from the equation, the asylum-seeker’s risk ceases to exist.  However, the BIA instead imposed an incorrect test for nexus requiring evidence of an “animus against the family or the respondent based on their biological ties, historical status, or other features unique to that family unit.”13

As a former circuit court judge, Garland is particularly qualified to recognize the error in the Board’s approach, as well as the need to correct its course.  The problem is compounded by the particular composition of the BIA at present.  For example, of the ten immigration judges who were promoted to the BIA during the Trump administration, nine denied asylum more than 90 percent of the time (with the tenth denying 85 percent of such claims).  Three had an asylum denial rate in excess of 98 percent.14

This matters, as those high denial rates were achieved in part by using faulty nexus determinations to deny asylum in domestic violence claims, even before the issuance of Matter of A-B-.  This was often accomplished by mischaracterizing the abuse as merely personal in nature, referencing only the persecutor’s generally violent nature or inebriated state.  The analysis in those decisions did not further examine whether gender might also have been one central reason that the asylum seeker, and not someone else, was targeted.

One BIA Member appointed under Trump recently found no nexus in a domestic violence claim by concluding that the persecutor had not targeted the asylum seeker because of her membership in the group consisting of “women,” but rather because she was his woman. There is no indication in the decision that the Board Member considered why the persecutor might view another human being as belonging to him and lacking the same rights he seems to enjoy.  Might it have been because of her gender?

Without a correction through published regulations, there is little reason to expect different treatment of these claims moving forward.  Let’s hope that the Attorney General views his recent action as only the first steps on a longer path to a correct application of the law.

Copyright 2021, Jeffrey S. Chase.  All rights reserved.

Notes:

1. Matter of A-B-, 28 I&N Dec. 307 (A.G. 2021) (“A-B- III”); Matter of L-E-A-, 28 I&N Dec. 304 (A.G. 2021) (“L-E-A- III”).

2. 27 I&N Dec. 316 (A.G. 2018) (“A-B- I”).

3. 27 I&N Dec. 581 (A.G. 2019) (“L-E-A- II”).

4. 28 I&N Dec. 199 (A.G. 2021) (“A-B- II”).

5. 26 I&N Dec. 388 (BIA 2014).

6. The regulations under consideration at that time were never issued.

7. 27 I&N Dec. 40 (BIA 2017) (“L-E-A- I”).

8. UNHCR, Guidelines on International Protection: Gender-Related Persecution within the context of Article 1A(2) of the 1951 Convention and/or its 1967 Protocol Relating to the Status of Refugees (May 2002) at para. 30.

9. James C. Hathaway and Michelle Foster, The Law of Refugee Status, Second Edition (Cambridge University Press, 2014) at 442.

10. Hathaway and Foster, supra.

11. Bostock v. Clayton County, 140 S.Ct. 1731, 1739 (2020).

12. Hernandez-Avalos v. Lynch, 784 F.3d 944, 950 (4th Cir. 2015).

13.  L-E-A- I, supra at 47.

14. See TRAC (Transactional Records Access Clearinghouse) Immigration Judge Reports https://trac.syr.edu/immigration/reports/judgereports/.