The Trump Administration, Asylum Law, and Private-Actor Persecution
Written by Rachael Stryer*
Over the past three and a half years, the Trump administration has sought to radically restrict access to asylum in the United States through administrative decision-making and rulemaking. Two recent circuit court decisions grapple with the administration’s attempts to limit the eligibility of asylum applicants fleeing persecution from non-state actors — while the Second Circuit upheld the U.S. Attorney General’s newly-stringent requirements, the D.C. Circuit found them arbitrary and capricious. The new system would, in practice, bar a woman seeking protection from domestic violence or murder witnesses fleeing gang retribution from gaining asylum, even where their states of origin struggled to stem or were blind to this violence. These changes affect the core tenants of asylum law and contradict tentative global trends towards the recognition of private-actor persecution.
The Global Asylum System
The contemporary global asylum regime was established by the 1951 Refugee Convention and the subsequent 1967 Refugee Protocol. The Convention sets out the rights of refugees and defines the responsibility of signatory states towards refugees. Article 1 defines refugees as individuals who have a “well-founded fear of persecution for reasons of race, religion, nationality, membership of a particular social group or political opinion” and who are “unable or, owing to such fear,  unwilling to avail [them]self of the protection of” their country of nationality. The Convention’s focus on government responsibility reflects a Euro- and state-centered worldview, responding to episodes of state renunciation of segments of its population, notable the Armenian Genocide and the Holocaust. With the emergence of the Cold War, the refugee system effectively adjusted to an individualized model of state persecution. However, the Convention excluded non-Europeans from refugee status until 1967, ignoring parallel mass refugee situations created by, for example, the Chinese Communist Revolution, the Korean War, the Arab-Israeli War, and the partition of India and Pakistan. Further, the asylum regime failed to address the status of victims felling private persecution. Rather, the Convention required applicants to prove that, where the state was not the persecutor, the state was unable or unwilling to provide protection.
Signatories have interpreted this state-centered requirement differently, creating two diverging systems. In the majority of Convention and Protocol signatories, including the United Kingdom, most of Western Europe, Canada, New Zealand, Australia, and the United States, judicial inquiry has focused on whether the home state could and would protect the asylum seeker. Applicants were eligible for asylum relief where a state was averse to providing real protection to the individual or where the state was incapable of doing so. In contrast, civil law states, in particular Germany, Switzerland, Italy, and France, retained a state-centered concept of persecution which focused on state responsibility for the persecution in question. Under the protection regime, asylum applicants needed to prove a failure of state protection separately from the persecution at hand; under the accountability regime applicants needed to show that the state failure was directly responsible for or that the state supported, encouraged, or tolerated the persecution by third parties.
Increasing Global Recognition of Non-State Persecution
This protection-accountability dichotomy has shifted over time in reaction to changes in international dynamics and law. The 1990s witnessed increasing internal conflict and state decline. As a result, “popular repression by an all-powerful state is no longer the primary context or metaphor for human rights abuses.” Two new paradigms, while by no means new, have become increasingly prominent. First, in areas where states lack control, non-state actors such as FARC in Colombia and Mara Salvatrucha in El Salvador have adopted state-like functions. Second, where states retain control, their unwillingness to use their authority to protect minorities and women, such as the cases of Rohingya refugees in Bangladesh or unpatrolled Palestinian refugee camps in Lebanon, have emerged as central features of asylum debates.
States’ implementation of asylum obligations have developed in response to paradigms of state incapacity and indifference. Protection-focused countries increasingly recognize gender-based claims when states are unwilling to provide protection against gender-based violence. A landmark 1999 British decision, Ex Parte Shah, recognized the petitions of two Pakistani women seeking asylum from their abusive husbands based on their particular social group “Pakistani women.” Although gang-based claims remain controversial, states have increasingly recognized these types of claims through a focus on the ineptness of states to counter the groups. For example, a 2009 Australian case granted asylum to a man claiming persecution as a bus driver targeted by MS-13 in El Salvador and a 2008 Canadian casegranted asylum to a Mexican journalist who had investigated gang violence.
Likewise, accountability-centered countries have shifted towards protection-based models. In response to a 2004 European Union directive, Germany, France, and Italy broadened their definitions of persecution to include non-state actors. Since then, Germany, which has traditionally had the most stringent accountability system, has recognized criminals, mafias, religious extremists, and terrorists as non-state agents of persecution and recognized gender-based attacks such as forced marriage as forms of persecution. These decisions increasingly consider the ability and willingness of the state to provide protections, recognizing state ineffectiveness and indifference beyond failed or incompetent states. For example, a 2006 Swiss decision granted asylum to a woman fleeing gender-based violence, finding Ethiopia unable to provide protection despite recent improvements in the Ethiopian legal system. Thus, while these systems continue to struggle to adapt to contemporary realities and to address historical blind spots, they have shown increasing flexibility with regards to the types of persecution they recognize.
Regression in American Asylum Law
American jurisprudence historically has developed in parallel with other protection-based regimes. Immigration courts require applicants seeking asylum in the United States to show past persecution or a well-founded fear of future persecution on the basis of their membership in a protected group “inflicted either by the government of a country or by persons or an organization that the government was unable or unwilling to control.”
This standard has allowed applicants fleeing gender- and gang-based violence to gain asylum in the United States, albeit inconsistently. In Matter of Kasinga, the United States recognized gender-based persecution in the case of a woman fleeing forced marriage and female genital mutilation in Togo, finding that Togo was unwilling to protect her. And, a number of circuit court decisions have granted asylum to those facing threats from gangs. For example, in 2011, the Third Circuit in Garcia v. Attorney General granted protection to a woman who had assisted law enforcement against violent gangs in Guatemala, finding that the government was unable to protect the applicant even if it was willing and even had attempted to do so. In contrast, where these gang-based claims have failed, they have typically done so because the court found their group membership incognizable, rather than because of the nature of the persecuting actor.
In 2018, however, the Attorney General held in Matter of A-B- that the “unable and unwilling” standard requires asylum seekers to demonstrate the government “condoned the private actions or ‘at least demonstrated a complete helplessness to protect the victims.’ In particular, the Attorney General rejected the use of country condition reports, requiring applicants to prove these standards based on individual circumstances rather than broader social dynamics. This new standard would be nearly impossible for applicants to meet. While Matter of A-B- has received significant attention for rolling back broad recognition of the cognizability of gender-based particular social groups, its repercussions extend beyond asylees fleeing domestic violence to the viability of broad-reaching private-actor claims.
In the past year, two circuit courts have grappled with these repercussions, coming to diverging conclusions. In the Second Circuit, Scarlett v. Barr broke with circuit precedent to uphold the heightened “complete helplessness” standard, denying asylum to a Jamaican constable facing threats from a gang. Upon asking for protection, the applicant’s superiors transferred him to a station four miles away from his original station, where he continued to receive threats. The police did nothing further to protect the applicant. The court found that the superiors’ initial response was enough to show that the state was not tolerant of the applicant’s persecution.
The Second Circuit held that the applicant did not meet the “condoned” or “complete helplessness” standard. “Condoned” suggests active approval of persecution far beyond an unwillingness, while “completely helpless” connotes a state that is totally impotent, significantly more extreme than “unable.” Despite the clearly heightened standard, Scarlett deferred to the government’s position that the complete helplessness standard was not a change from the traditional “unable or unwilling test.”
In contrast, the D.C. Circuit ruled for twelve asylum applicants fleeing gender- and gang-based violence in Grace v. Barr, finding that the AG’s “condoned-or-completely-helpless” standard broke with established Board practice. While it rejected the authority of the United Nations High Commissioner for Refugees Handbook, the Circuit found this change arbitrary and capricious. The court establish that “[a] government that ‘condones’ or is ‘completely helpless’ in the face of persecution is obviously more culpable, or more incompetent, than one that is simply ‘unwilling or unable’ to protect its citizens,” drawing on a recent First Circuit decision which found the Mexican state unable to provide protection to an applicant whose son was murdered. Although police had responded to the murder of the applicant’s son by visiting the scene of the murder, taking statements, and performing an autopsy, the general corruption, low trust, and high homicide rates in the region were sufficient evidence to demonstrate that the state was incapable of adequate guarantees. Grace argued that, under the condoned or completely helpless standard, the claim would almost certainly fail (because the government had responded to the murder), thereby creating two differing standards.
As Grace v. Barr demonstrates, replacing a “unwilling or unable to protect” model with “condoned or complete helplessness” fundamentally shifts the American asylum system. While other signatories move to recognize non-state persecution (albeit unevenly and with gaping failures of their own), the United States recedes from a protection-based model towards the increasingly antiquated accountability-based model. This decision signals a regression in American asylum protection and moves the United States out of step with both global realities and international asylum developments.
*J.D. candidate, Yale Law School; Features Editor, Yale Journal of International Law
 UN General Assembly, Convention Relating to the Status of Refugees, 28 July 1951, 189 UNTS 137.
 UN General Assembly, Protocol Relating to the Status of Refugees, 31 January 1967, 606 UNTS 267.
 Lucy Mayblin, Asylum After Empire: Colonial Legacies in the Politics of Asylum Seekers 22-23 (2017).
 The United States is only party to the Protocol, which it ratified in 1968. Congress incorporated the definition into immigration law through the Refugee Act of 1980.
 Catherine Phuong, Persecution by Third Parties and European Harmonization of Asylum Policies, 16 Geo. Immigr. L.J. 81, 83 (2001).
 See Mayblin, supra note 3, for a critique of the narrative of “new” refugees as constructed through the systematic historic exclusion of non-European refugees.
 Jennifer Moore, From Nation State to Failed State: International Protection from Human Rights Abuses by Non-State Agents, 31 Columb. H.R. L. Rev. 81, 85 (1999).
 Id. at 104.
 Id. at 105.
 Elizabeth Keyes, Unconventional Refugees, 67 Am. U. L. Rev. 89 (2017); Shoenholtz, supra note 6.
 Matter of Acosta, 19 I&N Dec. 211, 222 (BIA 1985).
 27 I&N Dec. 316, 337 (AG 2018).
 Id. at 340.
 957 F.3d 316 (2d Cir. 2019).
 Id. at 323.
 Id. at 333.
 965 F.3d 883, 898 (D.C. Cir. 2020).
 Id. at. 897.
 Id. at 888-89.
 Id. at *10.
 See Steve Y. Koh, Nonacquiescence in Immigration Decisions of the U.S. Courts of Appeals, 9 Yale L. & Pol’y Rev. 430 (1991) for an explanation of why Matter of A-B- still applies despite an appellate court’s rejection of the standard.