Resolving Negusie: The Attorney General Should Recognize a Broad Duress Defense to Article 1(F)(a) and Grant Asylum

Chris Opila

Third-year student at the University of Michigan Law School

Article 1(F)(a) of the Refugee Convention excludes an individual from refugee protection if there are “serious reasons for considering that he has committed a crime against peace, a war crime, or a crime against humanity, as defined by the appropriate international instruments.”[1] There is a growing international consensus that this provision encompasses a duress defense,[2] despite some U.S. adjudicators’ contentions otherwise.[3] The specific elements of such a defense, however, remain in dispute.[4] Using the Board of Immigration Appeals’ (BIA) recent decision in Matter of Negusie[5] as a starting point, this Note argues that Article 1(F)(a) contains a duress defense and compares the tests U.S., British, and Canadian tribunals have developed for it. The Note then contends that none of these tests are correct since the proper duress test not only adopts the language of Rome Statute Article 31(1)(d), but also resolves unsettled questions of its interpretation in the applicant’s favor. Finally, the Note returns to Negusie and asserts that if the U.S. Attorney General were to apply this proposed test in his review of the BIA’s decision, he should find that Negusie acted under duress, and thus remains eligible for refugee protection.

I. Introduction: Negusie’s Facts & Holding

Before arguing for the existence of a duress defense to Article 1(F)(a) and proposing its proper form, it is useful to review Negusie’s facts and findings to lend the argument factual context. Daniel Girmai Negusie is an Eritrean and Ethiopian national.[6] He was forcibly conscripted into the Eritrean military but refused to fight and consequently was detained and tortured for two years.[7] After his release, the Eritrean military forced Negusie to work as a guard at a prison camp where his duties included guarding prisoners and preventing them from taking showers and obtaining fresh air.[8] At times, he also guarded prisoners whom his supervisors had punished by placing them in the hot sun; and he witnessed at least one man die from this maltreatment.[9] Negusie’s supervisors twice caught him disobeying orders and assisting prisoners.[10] They reprimanded him and threatened to kill him because of this but did not harm him.[11] Negusie’s supervisors also executed two of his friends after their failed escape attempt.[12]

In 2018, the BIA denied Negusie asylum under the persecutor bar.[13] Two members held that his participation in persecuting prisoners[14] was not under duress. They found that the death threats Negusie faced, “when viewed in context, did not constitute the [required] imminent threat of death or serious bodily injury,” and that Negusie had failed to seize “a reasonable opportunity to escape or otherwise avoid guarding prisoners who were subject to harm and torture.”[15] The third BIA member concurred in the result but dissented from the majority’s rationale, arguing that there is no duress defense to the U.S. persecutor bar or Article 1(F)(a).[16]

II. Article 1(F)(a) Contains a Duress Defense

The BIA is not the only tribunal that has acknowledged a duress defense to Article 1(F)(a);British and Canadian tribunals have done so as well.[17] Nonetheless, some adjudicators still maintain that the defense either does not exist,[18] or that its existence remains an open question.[19] These holdouts are mistaken. A good faith[20] interpretation of Article 1(F)(a) recognizes a duress defense.

The text of Article 1(F)(a) demands a duress defense finding. Article 1(F)(a) requires “a crime against peace, a war crime, or a crime against humanity, as defined by the appropriate international instruments drawn up to make provision in respect to such crimes.”[21] These terms incorporate criminal law principles, such as duress, and focus on individual criminal culpability, i.e., “criminals whose conduct subjected them to extradition or international criminal proceedings and whose unworthiness would weaken public support for the refugee regime.”[22] Furthermore, the plain language of Article 1(F)(a) ties the definitions of its terms to “appropriate international instruments,”[23] meaning that they are not fixed but rather evolve alongside “contemporary understandings of relevant crimes.”[24] The most contemporary understanding of international criminal law is codified in the Rome Statute, an international instrument that expressly recognizes a duress defense.[25]

The context, object, and purpose of Article 1(F)(A) also compel a duress defense finding. The Refugee Convention’s travaux préparatoires and connection to the post-WWII military tribunals manifest the drafters’ clear intent to limit the category of individuals excluded from refugee protection to those who bore individual responsibility for their actions, i.e., not those who acted under duress.[26] Delegates drafted the Convention against the backdrop of the humanitarian crisis and war criminal prosecutions that followed the Holocaust and World War II.[27] They added Article 1(F)(a) to the Refugee Convention because they felt that such war criminals were undeserving of refugee status and wished to render them ineligible for protection under the Convention.[28] The Article’s initial language specifically referenced the Charter of the International Military Tribunal, and while the drafters did later rephrase it to its present form, they only did so to placate Germany’s desire for more politically neutral language.[29] This drafting history thus demonstrates the drafters’ intent to link the ambit of Article 1(F)(a) to jurisprudence of the international criminal tribunals at the time,[30] both of which recognized a superior orders duress defense.[31]

III. The Proper Duress Test Adopts Rome Statute Article 31(1)(D) and Resolves Its Open Interpretative Questions in the Applicant’s Favor

Having established that a good faith reading of Article 1(F)(a) compels recognizing a duress defense, it is time to consider the concomitant, unsettled question of the contours of the defense itself. U.S., British, and Canadian tribunals have all cited Rome Statute Article 31(1)(d)[32] as a basis for their duress tests and adapted many of its elements.[33] However, each tribunal has articulated a different test because they have resolved the Article’s unsettled interpretative questions differently,[34] and/or allowed considerations of domestic criminal law to infect their analysis.[35] None of the tribunals’ tests are correct, however, since the proper test actually adopts Article 31(1)(d) verbatim and resolves its unsettled interpretative questions in the applicant’s favor.

A. The U.S., British, and Canadian Duress Tests Differ on Each of Article 31(1)(d)’s Elements

Before detailing the proper test, it is useful to contrast the elements of the U.S., British, and Canadian tests, as they inform it. In Negusie, the BIA held that the U.S. duress test contains five elements: (1) an “imminent threat of death or serious bodily injury to the applicant or another person;” (2) a “reasonab[le] belie[f] that the threatened harm would be carried out unless the applicant acted or refrained from acting;” (3) no “reasonable opportunity for the applicant to escape or otherwise frustrate the threat;” (4) reasonable unforeseeability, i.e., the applicant “did not place himself a situation in which he knew or reasonably should have known he would likely be forced to act or refrain from acting;” and (5) proportionality, i.e., the applicant “knew or reasonably should have known that the harm he inflicted was not greater than the threatened harm to him or another person.”[36]

In Ezokola v. Canada, the Canadian Supreme Court affirmed Rome Statute Article 31(1)(d) as the Canadian duress test, but did not elucidate its own articulation of the test’s elements.[37] The Ezokola test therefore is comprised of the three elements of the Article itself: (1) a “threat of imminent death or of continuing or imminent serious bodily harm to the applicant or another person;” (2) necessity, i.e., the applicant acted “necessarily and reasonably to avoid this threat” and; (3) proportionality, i.e., the applicant did “not intend to cause a greater harm than the one sought to be avoided.”[38] The Ezokola Court also interpreted the necessity element to extend to the applicant’s complicity in bringing about the threats he faced and specified the “method of recruitment” and “any opportunity to leave” as relevant factors in this analysis.[39]

In MT Zimbabwe v. The Secretary of State for the Home Department, the Immigration and Asylum Chapter of the Upper Tribunal of the United Kingdom (UKUT) held that the British duress test has four elements: (1) “threat [of] imminent death or of continuing or imminent serious bodily harm;” (2) causation, i.e., “the threat must result in duress causing the crime;” (3) objectivity, i.e., “a threat results in duress only if it is otherwise unavoidable (i.e.[,] if a reasonable person incomparable circumstances would have submitted and would have been driven to the relevant criminal conduct);” and (4) necessity, i.e., “the act directed at avoiding the threat must be necessary in terms of no other means being available and reasonable for reaching the desired effect.”[40] The UKUT further expressed that whether duress “is a complete defense […that] can apply in all types of cases and circumstances remains unsettled.”[41]

The Negusie, Ezokola, and Zimbabwe tests differ with respect to each of Article 31(1)(d)’s three elements for duress:

1. Threat Element: An imminent threat of death or bodily harm satisfies each test.[42] But, the Ezokola and Zimbabwe tests also allow a continuing threat of serious bodily harm to satisfy this element as well.[43] The Negusie and Zimbabwe tests further specify that the threat must be objectively reasonable,[44] taking a stance on an open question in international law[45] that the Ezokola test leaves unanswered.[46] 

2. Necessity Element: Each test requires the applicant’s conduct to be necessary and reasonable after he perceives a threat.[47] But, they take different positions on whether the necessity inquiry extends before that,[48] another unresolved question of international law.[49] The Ezokola and Negusie tests extend the inquiry to actions an applicant took before perceiving the threat,[50] while the Zimbabwe test does not.[51] Moreover, the Negusie test also cabins its extended inquiry to actions taken once the threat was reasonably foreseeable, while the Ezokola test does not.[52]

3. Proportionality Element: Negusie and Ezokola incorporate proportionality into their tests, but Zimbabwe does not.[53] However, Negusie assesses the balance of harms objectively from the perspective of a reasonable person, while Ezokola does so subjectively from the perspective of the applicant himself.[54]

B. The Proper Duress Test Is One that Adopts Article 31(1)(d) and Resolves Its Unsettled Interpretative Questions in the Applicant’s Favor

The proper duress test adopts Rome Statute Article 31(1)(d) verbatim and resolves its open interpretative questions in the applicant’s favor.[55] The test’s elements are therefore: (1) a “threat of imminent death or of continuing or imminent serious bodily harm to the applicant or another person;” (2) necessity, i.e., the applicant acted “necessarily and reasonably to avoid this threat;” and (3) proportionality, i.e., the applicant did “not intend to cause a greater harm than the one sought to be avoided.”[56] The test’s threat element is analyzed subjectively from the applicant’s perspective[57] and may be satisfied by a threat of continuing serious bodily harm.[58] Likewise, the scope of the test’s necessity inquiry is limited to when the applicant has subjectively perceived the threat,[59] but gives consideration to his reasonable opportunity to escape at or after this moment.[60] Finally, the test’s proportionality element is assessed subjectively based on the applicant’s intent.[61]

The proposed duress test interprets Article 1(F)(a) of the Refugee Convention in accordance with its terms’ ordinary meaning in their context, and in light of their object and purpose, because it directly incorporates the duress test in Article 31(1)(d) of the Rome Statute and resolves this Article’s ambiguities in the applicant’s favor. Textually, Article 1(F)(a) subjects its terms to definition by “the appropriate international instruments.”[62] The Rome Statute is the most contemporary, authoritative international criminal law instrument; and the proposed test adopts its duress defense verbatim. Contextually and purposively, Article 1(F)(a) elucidates a clear intent for refugee law to “follow, not lead, on the substance of relevant international criminal law,” because the consequences of unlawful exclusion are so severe.[63] Exclusion is therefore improper “if a defense is plausible under an authoritative international criminal law instrument.”[64] The proposed test respects this intent by resolving all unsettled interpretative questions about the Rome Statute’s duress test in the applicant’s favor, thereby allowing him to benefit from the broadest plausible duress defense the Statute allows.

Conclusion: Resolving Negusie

Returning to the facts and holding of Negusie, if the Attorney General applies the proposed test on review, he cannot affirm the BIA’s decision since its interpretations of the threat and necessity elements are incompatible with the proposed test. The BIA found that Negusie did not act under duress for two reasons: (1) “the threats of death [Negusie] received should he disobey orders, when viewed in context, did not constitute [an] imminent threat of death or serious harm,” and (2) Negusie failed to seize “a reasonable opportunity to escape or otherwise avoid guarding the prisoners” because “he eventually escaped through a weak spot.”[65] The first justification is void because it assesses the threat element objectively rather than subjectively.[66] The second justification is similarly void since it analyzes the factual context of the reasonable escape inquiry objectively, starts the reasonable escape inquiry at the point when the threat was reasonably foreseeable rather than subjectively perceived, and does not give appropriate weight to the consequences of a failed escape.[67]

Applying the proposed test, the Attorney General should find that Negusie performed his activities as a prison guard under duress. First, Negusie satisfies the test’s threat element since he perceived an imminent threat of death for disobeying orders and attempting to escape,[68] and subjective perception is sufficient to satisfy this element. Second, Negusie satisfies the test’s necessity element since the actions he took to avoid the threat were necessary and reasonable. He attempted to avoid participating in the torture of prisoners by disobeying orders until his supervisors’ reprimands caused him to fear for his life and made additional disobedience unreasonable.[69] Negusie then extracted himself from this situation by escaping, and there are no facts in the record that indicate a reasonable person in his position would have escaped earlier, especially when due weight is given to his fear of being executed should his escape attempt fail.[70] Finally, Negusie satisfies the test’s proportionality element as he did not intend to cause a greater harm than the one he sought to avoid by obeying orders. Negusie did not intend for an inmate whose torture he aided to die, and the torture he did intend for his actions to abet is not a greater harm than the death he sought to avoid by obeying orders.[71]

The United States adopted legislation to comply with its obligations under the Refugee Convention almost thirty years ago,[72] and Negusie applied for asylum nearly fifteen years ago.[73] The time has come for the United States to uphold these obligations with respect to refugee applicants who committed an Article 1(F)(a) crime under duress and stop excluding Negusie from asylum under the persecutor bar.


[1] Convention Relating to the Status of Refugeesart. 1(F)(a), July 28, 1951, 19 U.S.T. 6259, 189 U.N.T.S. 150 [hereinafter Refugee Convention].

[2] See Negusie, 27 I&N Dec. 347, 362 (BIA) (adopting “a limited and strictly construed duress defense”) stayed by 27 I&N Dec. 481, 481 (A.G. 2018); Ezokola v. Canada, 2013 SCC 40, [2013] 2 S.C.R. 678, para. 86 (Can.) (finding that the voluntariness requirement for contribution to a crime captures the defense of duress); MT Zimbabwe v. Sec’y of State for the Home Dep’t, [2012] UKUT 15, [106] (U.K.) (recognizing a duress defense in “situations where the defendant’s freedom or will and decision is so severely limited that there is eventually no moral choice of counter activity available”).

[3] Negusie, 27 I&N Dec. at 371 (Malphrus, G. concurring and dissenting) (“The text of Article 1F does not provide a duress exception for the prohibited acts.”); Id. at 481 (“Whether coercion and duress are relevant to the […] persecutor bar” is an unresolved question).

[4] Compare Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (endorsing the Rome Statue test whose elements are: threat, necessity, and proportionality) with Negusie, 27 I&N Dec. at 363 (articulating a test whose elements are threat, reasonable belief, absence of reasonable escape, reasonable unforeseeability, and proportionality) with MT Zimbabwe, [2012] UKUT 15, at [106] (positing a test whose elements are: threat, causation, objectivity, and necessity).  

[5] This case has a lengthy, complicated, and ongoing procedural history. In 2005, a U.S. immigration judge denied Negusie’s asylum application, finding that his activities as a prison guard triggered the “persecutor bar.” See Negusie, 27 I&N Dec. at 347. (The “persecutor bar” incorporates Article 1(F)(a) into the U.S. Immigration and Nationality Act by exluding individuals who “ordered, incited, assisted, or otherwise participated in the persecution of any person on account of race, religion, nationality, membership in a particular social group, or political opinion” from receiving asylum. 8 U.S.C. § 1158(b)(2)(A)(i); J.M. Alvarado, 27 I&N Dec. 27, 30 n.3 (BIA 2017) (finding that the U.S. Congress considered the persecutor bar to be consistent with Article 1(F)(a).) The BIA and Fifth Circuit Court of Appeals affirmed the immigration judge’s decision, holding that Fedorenko v. United States foreclosed finding a voluntariness element to the persecutor bar. Negusie v. Gonzales, 231 F. App’x 325, 327 (7thCir. 2007) (citing 449 U.S. 490 (1981)), rev’d,Negusie v. Holder, 555 U.S. 511 (2009). In 2009, the Supreme Court reversed, holding that the lower tribunals had misapplied Fedorenko because the statutory framework that Fedorenko interpreted differed from that of the persecutor bar. Negusie, 555 U.S. at 520–521. The Supreme Court remanded the case to the BIA to consider in the first instance whether a duress defense existed to the persecutor bar. Id. at 521. In June 2018, the BIA reaffirmed the immigration judge’s decision to deny Negusie asylum, holding that although the persecutor bar has a duress defense, Negusie’s activities as a prison guard exceed its scope. Negusie, 27 I&N Dec. at 367–368. In October 2018, the Attorney General stayed the BIA’s decision, certified the case to himself for review, and invited amici to brief the issue of whether there is a duress defense to the persecutor bar. Negusie, 27 I&N Dec. at 481. At the time of this Note’s publication, the Attorney General has yet to issue his review.

[6] Negusie, 27 I&N Dec. at 348.

[7] Id.

[8] Id.

[9] Id.

[10] Id.

[11] Id. at 348-49, 368.

[12] Id. at 352.

[13] Id. at 348,

[14] Negusie does not contest that his activities as a prison guard constitute persecution under U.S. law. For the purposes of analyzing the duress defense contained in Article 1(F)(a), this paper assumes that Negusie’s activities fall within its ambit, i.e., constitute a crime against humanity.

[15] Negusie, 27 I&N Dec. at 368.

[16] Id. at 369, 370.

[17] See id. at 363; Ezokola v. Canada, 2013 SCC 40, [2013] 2 S.C.R. 678, para. 86 (Can.); MT Zimbabwe v. Sec’y of State for the Home Dep’t, [2012] UKUT 15, [106] (U.K.).

[18] Negusie, 27 I&N Dec. at 371 (Malphrus, G. concurring and dissenting) (“The text of Article 1F does not provide a duress exception for the prohibited acts.”).

[19] Negusie, 27 I&N Dec. 481, 481 (A.G. 2018) (inviting amici to submit briefs on “whether coercion and duress are relevant to the […] persecutor bar”).

[20] Vienna Convention Article 31(1) requires treaties like the Refugee Convention to “be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in light of its object and purpose.” See May 23, 1969, 1155 U.N.T.S 331. See also James Hathaway & Michelle Foster, The Law of Refugee Status 5 n.25 (2d ed. 2014).

[21] Refugee Convention, supra note 1, at art. 1(F)(a) (emphasis added).

[22] Amended Brief for Scholars of International Refugee Law as Amici Curiae Supporting Petitioner Matter of Negusie, 27 I&N Dec. 347 (BIA 2018), at 7 [hereinafter Refugee Scholars].

[23] Refugee Convention, supra note 1, at Art. 1(F)(a).

[24] Hathaway & Foster, supra note 19, at 569.

[25] See Rome Statute of the International Criminal Court art. 31(1)(d), July 1, 2002, 2187 U.N.T.S. 90 [hereinafter Rome Statute].

[26] Refugee Scholars, supra note 21, at 3.

[27] See Negusie, 27 I&N Dec. 347, 357–358 (BIA) stayed by 27 I&N Dec. 481, 481 (A.G. 2018); Refugee Law Scholars, supra note 21, at 3–4.

[28] Hathaway & Foster, supra note 19, at 567–68.

[29] Id.

[30] Refugee Scholars, supra note 21, at 7-8 (citing Guy S. Goodwin-Gill & Jane McAdam, The Refugee in International Law163–165 (3d ed. 2007)).

[31] When the Refugee Convention was drafted, the international criminal tribunals in existence included the International Military Tribunal [IML], established by the London Charter to try major war crimes, and the Nuremberg Military Tribunals, created by Control Council Law No. 10 “for the prosecution of war criminals…, other than those dealt with by the [IML].” Negusie, 27 I&N Dec. at 357–358 (quoting Control Council Law No. 10 in I Trial of War Criminals Before the Nuremberg Military Tribunals Under Control Council Law No. 10, XVI (Oct. 1946–Apr. 1949), http://www.loc.gov/rr/frd/Military_law/NTs_war-criminals.html) [hereinafter Trials of War Criminals]). At this time, the IML had recognized a superior orders defense under which individuals were not culpable for war crimes or crimes against humanity if they had executed an order absent the ability to make a moral choice. Id. (citing I Trial of the Major War Criminals Before the International Military Tribunal 224 (Nuremberg 1947), https://www.loc.gov/rr/frd/Military_Law/pdf/NT_Vol-I.pdf). Likewise, the Nuremberg Military Tribunals had also recognized a superior orders defense when duress was involved. Id. (citing IV Trials of War Criminals, supra, at 480).

[32] “A person shall not be criminally responsible if, at the time of that person’s conduct, the conduct […] has been caused by duress resulting from a threat of imminent death or of continuing or imminent serious bodily harm against that person or another person, and the person acts necessarily and reasonability to avoid this threat, provided that the person does not intend to cause a greater harm than the one sought to be avoided.” Rome Statute, supra note 24, at art. 31(1)(d).

[33] See Negusie, 27 I&N Dec. at 362; Ezokola v. Canada, 2013 SCC 40, [2013] 2 S.C.R. 678, para. 86 (Can.); MT Zimbabwe v. Sec’y of State for the Home Dep’t, [2012] UKUT 15, [106] (U.K.).

[34] Compare Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (including a proportionality requirement in its duress defense test) with MT Zimbabwe, [2012] UKUT 15, at [106] (excluding a proportionality requirement from its duress defense test because the requirement remains an unsettled question of international law).

[35] See Negusie, 27 I&N Dec. at 363 (“The minimum threshold requirements of the duress defense […] are based on the formulations found in United States’ criminal law and international law.”). The BIA’s reliance on U.S. domestic criminal law in its analysis is inappropriate as the defenses to Article 1(F)(a) “must be based on the standards of an ‘international instrument,’ not domestic norms.” See Hathaway & Foster, supra note 19, at 568.

[36] Negusie, 27 I&N Dec. at 363.

[37] Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86.

[38] Rome Statute, supra note 24, at art. 31(1)(d).

[39] See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (“To assess the voluntariness of a contribution, decision makers should, for example, consider the method of recruitment by the organization and any opportunity to leave the organization.”).

[40] MT Zimbabwe v. Sec’y of State for the Home Dep’t, [2012] UKUT 15, [106] (U.K.).

[41] Id. (citing Prosecutor v. Erdemović, Case No. IT-96-22, Judgement, (Int’l Crim Trib. for the Former Yugoslavia Oct. 7, 1997)).

[42] See Negusie, 27 I&N Dec. 347, 363 (2018) stayed by 27 I&N Dec. 481, 481 (A.G. 2018); Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86; MT Zimbabwe, [2012] UKUT 15, at [106].

[43] See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, para. 86 (endorsing the Rome Statue test, which allows the threat to be one of continuing serious bodily harm); MT Zimbabwe, [2012] UKUT 15, at [106] (allowing the threat to be of continuing serious bodily harm). See also Hathaway & Foster, supra note 19, at 585 (“While the availability of duress as a defense was traditionally limited by reliance on […] strict imminence [this] no longer figure[s] in the Rome Statute’s contemporary codification of the case.”). Negusie likely does not allow a non-imminent threat to satisfy the threat element since the BIA incorporated domestic law into its duress analysis and the Supreme Court has held that duress requires an imminent threat. 27 I&N Dec. at 363 n.17 (citing Dixon v. United States, 548 U.S. 1, 4 n.2 (2006)).

[44] See Negusie, 27 I&N Dec. at 363; MT Zimbabwe, [2012] UKUT 15, at [106].

[45] Jennifer Bond & Meghan Fougere, Omnipresent Threats: A Comment on the Defence of Duress in International Criminal Law, 14 Int’l Crim. L. Rev. 471, 487–488 (2014) (“Article 31(1)(d) does not specify the standard on which the actual presence of the threat needs to be evaluated. This silence occurs despite the fact that the drafting history of the Rome Statute reveals evidence of discussion as to whether the existence of the threat should be evaluated on an objective, subjective or combined standard.”).

[46] See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86.

[47] See Negusie, 27 I&N Dec. at 363; Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86; MT Zimbabwe, [2012] UKUT 15, at [106].

[48] Compare Negusie, 27 I&N Dec. at 363 (An applicant cannot “place himself a situation in which he knew or reasonably should have known he would likely be forced to act or refrain from acting.”) and Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (“Decision makers should, for example, consider the method of recruitment by the organization and any opportunity to leave the organization.”) with MT Zimbabwe, [2012] UKUT 15, at [106] (“The act directed at avoiding the threat must be necessary in terms of no other means being available and reasonable for reaching the desired effect.”).

[49] See Brian Myers, Right to Kill or the Obligation to Die: The Status of the Defence of Duress following New Zealand’s Implementation of the Rome Statute of the International Criminal Court, 2 N.Z. Y.B. Int’l L. 127, 148 (2005) (“The precise legal test contemplated by [Article 31(d)(1) is uncertain [and while] New Zealand courts will also engage in a secondary inquiry as to whether the threat was reasonably foreseeable and avoidance was possible at some earlier time[, it] is not clear whether such an inquiry is contemplated by Rome Statute article 3l(1)(d).”).

[50] See Negusie, 27 I&N Dec. at 363; Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86.

[51] See MT Zimbabwe, [2012] UTUK 15, at [106]. The test’s plain language – “the act directed at avoiding the threat must be necessary […] for reaching the desired effect” – cannot be read as consistent with either Negusie or Ezokola because an applicant must first subjectively perceive a threat in order to have a desire to avoid it and then take directed action to do so. Compare id. with Negusie, 27 I&N Dec. at 363 (requiring the necessity inquiry to consider the threat’s objective foreseeability) and Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (requiring decisionmakers to consider the method of recruitment and any opportunity to leave the organization).

[52] Compare Negusie, 27 I&N Dec. at 363 (requiring the applicant not to place himself in a situation where duress was reasonably foreseeable) with Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (suggesting that decisionmakers should consider any opportunity to leave during its inquiry). While Ezokola may be read narrowly to be consistent with Negusie, its call for decisionmakers to consider “the method of recruitment by the organization and any opportunity to leave the organization,” is unbounded by a reference to the threat and so invites a broader reading that includes actions the applicant took when the threat was not yet foreseeable. See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (emphasis added).

[53] Compare Negusie, 27 I&N Dec., at 363 and Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 with MT Zimbabwe, [2012] UKUT 15, at [106].

[54] Compare Negusie, 27 I&N Dec. at 363 with Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (adopting Rome Statute Art. 33(1)(d) as the duress defense test); Rome Statute, supra note 24, at art. 33(1)(d) (subjecting proportionality to an individual’s intent) and Hathaway & Foster, supra note 19, at 585 n.396 (“The drafters [of the Rome Statute] expressly rejected the notions that proportionality analysis should be objectively framed.”).

[55] See Jennifer Bond, Principled Exclusion: A Revised Approach to Article 1(F)(a) of the Refugee Convention, 35 Mich. J. Int’l L. 15,32 (2013) (arguing that “where there are unresolved doctrinal issues in international criminal law, decisionmakers should adopt the approach that most restricts individual liability”). Ezokola may be read narrowly to be consistent with the proposed test if the scope of its necessity element is restricted to situations in which the threat is subjectively perceived, but its mandate to consider “the method of recruitment by the organization and any opportunity to leave the organization,” invites a broader reading that is inconsistent with this test. See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86 (emphasis added).

[56] Rome Statute, supra note 24, at art. 31(1)(d).

[57] Whether the presence of the threat is evaluated subjectively or objectively is an open question that neither Article 31(1)(d) nor its drafting history answer. See Bond & Fougere, supra note 44, at 487–488. By adopting the applicant’s subjective perception of whether there is a threat, the proposed test properly resolves this ambiguity in the applicant’s favor so that he may benefit from the broadest plausible defense the Rome Statute allows. See Hathway & Foster, supra note 19, at 572; Bond, supra note 55, at 32.

[58] Like Ezokola and Zimbabwe, the proposed duress test directly incorporates the threat element of Article 31(1)(d) without modifications. See Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86; MT Zimbabwe, [2012] UKUT 15, at [106].

[59] Both the starting point for the necessity inquiry and whether this inquiry is governed by a subjective or objective assessment of the circumstances remain unsettled questions. See Myers, supra note 49, at 148. By starting the inquiry when applicant subjectively perceives the threat and assessing the reasonableness and appropriateness of the applicant’s actions based on his perception of the facts, the proposed test properly resolves these ambiguities in the applicant’s favor so that he may benefit from the broadest plausible defense the Rome Statute allows. See Hathaway & Foster, supra note 19, at 572; Bond, supra note 55, at 32.

[60] The test considers the reasonable opportunity for escape to be a dispositive factor in the necessity inquiry since there is broad judicial and scholarly consensus on this issue. See Negusie, 27 I&N Dec. 347, 363 (BIA) stayed by 27 I&N Dec. 481, 481 (A.G. 2018); Ezokola, 2013 SCC 40, [2013] 2 S.C.R. 678, at para. 86; Hathaway & Foster, supra note 19, at 385 n.397; Refugee Scholars, supra note 21, at 23; UNHCR, Background Note on the Application of the Exclusion Clauses: Article 1F of the 1951 Convention Relating to the Status of Refugees, ¶70 (Sep. 4, 2003) [hereinafter UNHCR]. However, the escape assessment must be conducted based on the applicant’s perception of the facts once he has perceived the threat. It is stressed, moreover, that each case should be considered on its own facts and that weight should be given to the consequences of a failed escape. UNHCR, at ¶70. The test also rejects the first reasonable escape opportunity requirement that UNHCR recommends since objective foreseeability is a factor in its analysis, see id., and this test rejects a necessity element whose scope is bounded by objective foreseeability.

[61] The MT Zimbabwe court cites Prosecutor v. Erdemović, Case No. IT-96-22, Judgement, (Int’l Crim. Trib. For the Former Yugoslavia Oct. 7, 1997) to support its contention that the existence of a proportionality element is an open question. See [2012] UKUT 15, at [106]. But, since the Rome Statute was finalized after Erdemović, its drafters considered the arguments of both Erdemović’s majority and dissent. Their decision to codify the dissent’s position into Article 31(1)(d) is thus a deliberate resolution of this debate.

[62] Refugee Convention, supra note 1, at art. 1(F)(a).

[63] See Hathaway & Foster, supra note 19, at 572.

[64] See id.

[65] Negusie, 27 I&N Dec. at 368.

[66] See id. The proposed test assesses the existence of a threat subjectively from the applicant’s perspective, rather than objectively.

[67] See id. The proposed test measures the reasonableness of escape based on the applicant’s perception of the facts, starts the reasonable escape inquiry once the applicant has perceived the threat, and gives due consideration to the consequences of a failed escape.

[68] See id. at 348, 352.

[69] See id. at 348.

[70] See id. at 349.

[71] See id. at 352.

[72] Congress passed the Refugee Act of 1980 to implement the obligations that the United States assumed when it signed the 1967 Protocol Relating to the Status of Refugees. See Negusie v. Holder, 555. U.S. 511, 520 (2009).

[73] Negusie, 27 I&N Dec. at 348.

Suggested Citation: Chris Opila, Resolving Negusie: The Attorney General Should Recognize a Broad Duress Defense to Article 1(F)(a) and Grant Asylum, RefLaw (June 6, 2019), http://www.reflaw.org/resolving-negusie-the-attorney-general-should-recognize-a-broad-duress-defense-to-article-1fa-and-grant-asylum.

0 Comments

Submit A Comment

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>