LeBel J & Fish J
(delivered the judgment of the court)
Criminal responsibility does not fall solely upon direct perpetrators of crime. A murder conviction, for example, can attach equally to one who pulls the trigger and one who provides the gun. Complicity is a defining characteristic of crimes in the international context, where some of the world’s worst crimes are committed often at a distance, by a multitude of actors.
While principal perpetrators may be distinguished from secondary actors for sentencing, the distinction is irrelevant for the purposes of art. 1F(a) of the United Nations Convention Relating to the Status of Refugees, Can. T.S. 1969 No. 6 (“Refugee Convention”). Article 1F(a) excludes individuals from the definition of “refugee” if there are “serious reasons for considering that [they have] committed a crime against peace, a war crime, or a crime against humanity”. Those who commit these offences are not entitled to the humanitarian protection provided by the Refugee Convention. Where exclusion from refugee status is the only “sanction”, it is not necessary to distinguish between principals, aiders and abettors, or other criminal participants. Individuals may be excluded from refugee protection for international crimes through a variety of modes of commission.
Guilt by association, however, is not one of them.
This appeal homes in on the line between association and complicity. It asks whether senior public officials can be excluded from the definition of “refugee” by performing official duties for a government that commits international crimes. It is the task of this Court to determine what degree of knowledge and participation in a criminal activity justifies excluding secondary actors from refugee protection. In other words, for the purposes of art. 1F(a), when does mere association become culpable complicity?
In contrast to international crime, determining responsibility for domestic crime is often direct. While party liability plays a role, domestic criminal law, in its simplest form, asks whether one individual has committed one crime against one victim. In international criminal law, the focus often switches to the collective and to the links between individuals and collective action. International criminal law typically asks whether a group of individuals, an organization or a state has committed a series of crimes against a group of victims. In other words, party liability plays a much greater role in the commission of those crimes recognized as some of the most serious in the international legal order: R. Cryer, et al., An Introduction to International Criminal Law and Procedure (2nd ed. 2010), at p. 361.
Aware of the collective aspects of international crime, the Federal Court of Appeal correctly concluded that senior officials may be held criminally responsible for crimes committed by their government if they are aware of the crimes being committed yet remain in their position without protest and continue to defend the interests of the government.
However, this does not mean that high-ranking government officials are exposed to a form of complicity by association. Complicity arises by contribution. The collective nature of many international crimes does not erase the importance of holding an individual responsible only for his or her own culpable acts: G. Werle, “Individual Criminal Responsibility in Article 25 ICC Statute” (2007), 5 J.I.C.J. 953, at p. 953.
While individuals may be complicit in international crimes without a link to a particular crime, there must be a link between the individuals and the criminal purpose of the group – a matter to which we will later return. In the application of art. 1F(a), this link is established where there are serious reasons for considering that an individual has voluntarily made a significant and knowing contribution to a group’s crime or criminal purpose. As we shall see, a broad range of international authorities converge towards the adoption of a “significant contribution test”.
This contribution-based approach to complicity replaces the personal and knowing participation test developed by the Federal Court of Appeal in Ramirez v. Canada (Minister of Employment and Immigration),  2 F.C. 306. In our view, the personal and knowing participation test has, in some cases, been overextended to capture individuals on the basis of complicity by association. A change to the test is therefore necessary to bring Canadian law in line with international criminal law, the humanitarian purposes of the Refugee Convention, and fundamental criminal law principles.
We would therefore allow the appeal and send the matter back to a different panel of the Refugee Protection Division of the Immigration and Refugee Board (the “Board”) for redetermination in accordance with these reasons. The panel will decide whether there are serious reasons for considering that the appellant’s knowledge of, and participation in, the crimes or criminal purposes of his government meet the complicity by contribution test. As always, whether art. 1F(a) operates to exclude an individual from refugee protection will depend on the particular facts of the case.
A. Background Facts
The appellant, Rachidi Ekanza Ezokola, began his career with the government of the Democratic Republic of Congo (“DRC”) in January 1999. He was hired as a financial attaché at the Ministry of Finance and was assigned to the Ministry of Labour, Employment and Social Welfare in Kinshasa. He later worked as a financial adviser to the Ministry of Human Rights and the Ministry of Foreign Affairs and International Cooperation.
In 2004, the appellant was assigned to the Permanent Mission of the DRC to the United Nations (“UN”) in New York. In his role as second counsellor of embassy, the appellant represented the DRC at international meetings and UN entities including the UN Economic and Social Council. He also acted as a liaison between the Permanent Mission of the DRC and UN development agencies. In 2007, the appellant served as acting chargé d’affaires. In this capacity, he led the Permanent Mission of the DRC and spoke before the Security Council regarding natural resources and conflicts in the DRC.
The appellant worked at the Permanent Mission until January 2008 when he resigned and fled to Canada.
The appellant says he ultimately resigned because he refused to serve the government of President Kabila which he considered to be corrupt, antidemocratic and violent. He claims his resignation would be viewed as an act of treason by the DRC government. He claims that the DRC’s intelligence service harassed, intimidated, and threatened him because it suspected he had links to Jean-Pierre Bemba, President Kabila’s opponent. It was on this basis that he sought refugee protection for himself, his wife, and their eight children in Canada.
B. Judicial History
(1) Immigration and Refugee Board – Refugee Protection Division, 2009 CanLII 89027
The issue for the Board in determining the appellant’s application for refugee protection was whether the appellant should be excluded from Canada on the basis of s. 98 of the Immigration and Refugee Protection Act, S.C. 2001, c. 27 (“IRPA”). This provision directly incorporates art. 1F(a) of the Refugee Convention into Canadian law. The Board excluded the appellant from the definition of “refugee” under art. 1F(a). It held that, although the government of the DRC was not an organization with a limited and brutal purpose, it had committed crimes against humanity as defined by the Rome Statute of the International Criminal Court, U.N. Doc. A/CONF.183/9, July 17, 1998 (“Rome Statute”), and Canadian jurisprudence: paras. 31 and 43. The Board concluded that the government’s crimes continued even as the government itself changed: para. 33.
The Board relied on various reports – including media, governmental, and non-governmental – to find international crimes were committed, on both sides of conflicts spanning several years. For instance, at para. 39, it cited a report by Human Rights Watch, dated January 18, 2006, which condemned the conduct of the DRC government (and others):
In 2005, combatants from armed groups as well as government soldiers deliberately killed, raped, and abducted civilians and destroyed or looted their property in repeated attacks, particularly in eastern Congo. A feeble justice system failed to prosecute these recent crimes and did nothing to end impunity for war crimes and crimes against humanity committed during the previous two wars. The September 2005 discovery of mass graves from 1996 in the eastern region of Rutshuru served as a reminder of the unpunished mass slaughter of civilians in Congo in the last decade.
The government failure to integrate troops of former belligerent groups into the national army and to properly train and pay its soldiers underlay some military abuses. Military abuses such as those that occurred in December 2004 in North Kivu where government soldiers and combatants refusing integration fought and killed at least one hundred civilians, many of them targeted on an ethnic basis, were repeated elsewhere in 2005. In Walungu, South Kivu, government soldiers raped civilians and looted property during operations against the FDLR in late 2004 and early 2005. In Equateur, poorly paid and undisciplined troops went on a rampage in July 2005, killing, raping, and stealing from civilians.
As government soldiers tried to take control of Ituri and parts of North and South Kivu, Maniema and Katanga in late 2004 and 2005, both they and the combatants fighting them committed grave violations of international humanitarian and human rights law.
(World Report 2006: Events of 2005 (2006), at pp. 90-92)
The Board also cited reports condemning the DRC authorities’ practice of recruiting child soldiers: paras. 36 and 40.
The Board concluded that “[t]he evidence clearly shows that the Congolese government represses human rights, carries out civilian massacres and engages in governmental corruption”: para. 43.
In the Board’s view, the appellant was complicit in these crimes. Based on the appellant’s official rank, he had “personal and knowing awareness” of the crimes committed by his government: para. 71. The Board pointed out that the appellant had joined the government voluntarily and continued to act in his official capacity until he feared for his own safety. In the Board’s view, the appellant’s functions and responsibilities helped to sustain the government of the DRC. The Board therefore had serious reasons for considering that the appellant was complicit in the crimes committed by the government.
(2) Federal Court – Mainville J., 2010 FC 662,  3 F.C.R. 377
The Federal Court allowed the appellant’s application for judicial review. The court determined that an individual cannot be excluded under art. 1F(a) merely because he had been an employee of a state whose government commits international crimes. Complicity requires a nexus between the claimant and the crimes committed by the government.
In order to determine whether this link existed, the court considered the modes of commission provided for in the Rome Statute and concluded that “criminal responsibility for crimes against humanity requires personal participation in the crime alleged or personal control over the events leading to the crime alleged”: para. 86.
In the court’s view, the required nexus between the claimant and the crimes “may be established by presumption if the claimant held a senior position in the public service, where there are serious reasons for considering that the position in question made it possible for the refugee claimant to commit, incite or conceal the crimes, or to participate or collaborate in the crimes”: para. 4. However, in this case there was “no evidence that tend[ed] to show direct or indirect personal participation by the applicant in the crimes alleged, and there [was] no evidence of incitement or active support by the applicant for those crimes”: para. 104. It was an error to assign responsibility to the appellant solely on the basis of his position within the government, absent a personal nexus between his role and the army or police of the DRC.
The court certified the following question [2011 FCA 224,  3 F.C.R. 417, at para. 28]:
|For the purposes of exclusion pursuant to paragraph 1F(a) of the United Nations Refugee Convention, is there complicity by association in crimes against humanity from the fact that the refugee claimant was a public servant in a government that committed such crimes, along with the fact that the refugee claimant was aware of these crimes and did not denounce them, when there is no proof of personal participation, whether direct or indirect, of the refugee claimant in these crimes?|
(3) Federal Court of Appeal – Noël J.A. (Nadon and Pelletier JJ.A.)
The Federal Court of Appeal answered the certified question in the affirmative, but not before reformulating it as follows [para. 44]:
For the purposes of exclusion pursuant to paragraph 1F(a) of the United Nations Refugee Convention, can complicity by association in crimes against humanity be established by the fact that the refugee claimant was a senior public servant in a government that committed such crimes, along with the fact that the refugee claimant was aware of these crimes and remained in his position without denouncing them?
The Federal Court of Appeal rejected the Federal Court’s approach to complicity, describing it as inconsistent with Canadian jurisprudence and too narrow: paras. 46 and 57. The Federal Court of Appeal concluded that [para. 72]
a senior official may, by remaining in his or her position without protest and continuing to defend the interests of his or her government while being aware of the crimes committed by this government demonstrate “personal and knowing participation” in these crimes and be complicit with the government in their commission.
The court added that “the final outcome will always depend on the facts particular to each case”: para. 72.
The Federal Court of Appeal decided it was unnecessary to determine whether the conclusion of the Board was reasonable because it had applied the wrong test for complicity. Instead of applying the “personal and knowing participation” test, the Board considered the appellant’s “personal and knowing awareness”: para. 75. The Federal Court of Appeal therefore remitted the matter to a different panel of the Board to apply the personal and knowing participation test to determine whether the appellant was an accomplice in the crimes committed by the DRC.
Whether or not the appellant should ultimately be excluded from refugee protection for having committed international crimes will be determined by the Board at a de novo hearing. The task for this Court is to determine what test for complicity will be applied by the art. 1F(a) decision maker. To answer this question, the Court must also address the evidentiary standard applicable to art. 1F(a) determinations.
For the reasons that follow, we conclude that an individual will be excluded from refugee protection under art. 1F(a) for complicity in international crimes if there are serious reasons for considering that he or she voluntarily made a knowing and significant contribution to the crime or criminal purpose of the group alleged to have committed the crime. The evidentiary burden falls on the Minister as the party seeking the applicant’s exclusion: Ramirez, at p. 314.
In rejecting a guilt-by-association approach to complicity, we have considered
the purpose of the Refugee Convention and art. 1F(a);
the role of the Board;
the international law to which art. 1F(a) expressly refers; and
the approach to complicity under art. 1F(a) taken by other state parties to the Refugee Convention.
Each of these demonstrates the need to rein in the Canadian approach to complicity under art. 1F(a) to ensure that individuals are not excluded from refugee protection for merely being associated with others who have perpetrated international crimes.
B. The Purpose of the Refugee Convention and Article 1F(a)
In our view, the purpose of the Refugee Convention, together with the purpose of art. 1F(a), sheds light on the proper approach for determining exclusions from refugee protection based on complicity in international crimes: Vienna Convention on the Law of Treaties, Can. T.S. 1980 No. 37, art. 31.
The preamble to the Refugee Convention highlights the international community’s “profound concern for refugees” and its commitment “to assure refugees the widest possible exercise of .... fundamental rights and freedoms”. Our approach to art. 1F(a) must reflect this “overarching and clear human rights object and purpose”: Pushpanathan v. Canada (Minister of Citizenship and Immigration),  1 S.C.R. 982, at para. 57.
That said, the Refugee Convention’s commitment to refugee protection is broad, but not unbounded. It does not protect international criminals. Incorporated directly into Canadian law by s. 98 of the IRPA, art. 1F(a) guards against abuses of the Refugee Convention by denying refugee protection
to any person with respect to whom there are serious reasons for considering that:
As the Federal Court of Appeal recognized in Sivakumar v. Canada (Minister of Employment and Immigration),  1 F.C. 433, at p. 445: “When the tables are turned on persecutors, who suddenly become the persecuted, they cannot claim refugee status. International criminals, on all sides of the conflicts, are rightly unable to claim refugee status.” In other words, those who create refugees are not refugees themselves: Pushpanathan, at para. 63; Zrig v. Canada (Minister of Citizenship and Immigration), 2003 FCA 178,  3 F.C. 761, at para. 118.
On the one hand then, if we approach art. 1F(a) too narrowly, we risk creating safe havens for perpetrators of international crimes – the very scenario the exclusion clause was designed to prevent. On the other hand, a strict reading of art. 1F(a) arguably best promotes the humanitarian aim of the Refugee Convention: United Nations High Commissioner for Refugees (“UNHCR”), “Guidelines on International Protection: Application of the Exclusion Clauses: Article 1F of the 1951 Convention relating to the Status of Refugees”, HCR/GIP/03/05, September 4, 2003 (online), at para. 2.
The foregoing demonstrates the need for a carefully crafted test for complicity – one that promotes the broad humanitarian goals of the Refugee Convention but also protects the integrity of international refugee protection by ensuring that the authors of crimes against peace, war crimes, and crimes against humanity do not exploit the system to their own advantage. As we will explain, these two aims are properly balanced by a contribution-based test for complicity – one that requires a voluntary, knowing, and significant contribution to the crime or criminal purpose of a group.
C. The Role of the Refugee Protection Division: Exclusion Determinations, not Findings of Guilt
In addition to the purposes of the Refugee Convention and art. 1F(a), the test for complicity must reflect the role of the Board and must work within the practical realities of refugee proceedings.
A refugee hearing is not a criminal trial before an international tribunal. International criminal tribunals render verdicts for some of the most serious crimes in the international legal order. In contrast, the Board makes exclusion determinations; it does not determine guilt or innocence. The purpose of art. 1F(a) “is to exclude ab initio those who are not bona fide refugees at the time of their claim for refugee status”: Pushpanathan, at para. 58.
To achieve this purpose, Board hearings tend to be less formal than criminal trials. The Board is not bound by traditional rules of evidence: IRPA, at s. 170(g) and (h); Thamotharem v. Canada (Minister of Citizenship and Immigration), 2007 FCA 198,  1 F.C.R. 385, at para. 41; Kumar v. Canada (Minister of Citizenship and Immigration), 2009 FC 643 (CanLII), at paras. 28-29. Section 162(2) of the IRPA instructs each division of the Board to “deal with all proceedings before it as informally and quickly as the circumstances and the considerations of fairness and natural justice permit”.
As we will discuss in more detail below, the differences between a criminal trial and a Board hearing are further reflected in – and accommodated by – the unique evidentiary burden applicable to art. 1F(a) determinations: a person is excluded from the definition of “refugee” on the basis of the “serious reasons for considering” standard.
In light of these features of refugee proceedings, it is unnecessary to craft a multitude of tests for each mode of commission through which a government official may be held complicit in the crimes committed by his or her government. Unique considerations may arise in cases where the individual is said to have control or responsibility over the alleged perpetrators, or where the individual allegedly made specific contributions to a specific crime (in the form of instigating, ordering, or inciting, for example). However, here we are concerned with general participation in a group’s criminal activity. We must determine when that participation becomes a culpable contribution.
D. The Board Must Rely on International Law to Interpret Article 1F(a)
Following the express direction in the text of art. 1F(a), we now turn to international law for guidance. As mentioned, art. 1F(a) excludes individuals when “there are serious reasons for considering that” they have “committed a crime against peace, a war crime, or a crime against humanity, as defined in the international instruments”. We must therefore consider international criminal law to determine whether an individual should be excluded from refugee protection for complicity in international crimes: Harb v. Canada (Minister of Citizenship and Immigration), 2003 FCA 39, 302 N.R. 178, at para. 8. We will also look to international jurisprudence for guidance: Mugesera v. Canada (Minister of Citizenship and Immigration), 2005 SCC 40,  2 S.C.R. 100, at paras. 82 and 126.
In our view, international law is relevant both for the elements of the offences and their potential modes of commission. As the appellant stated, art. 1F(a) is not concerned with simply identifying the substantive elements of the offence, but with whether there are serious reasons for considering that the individual has committed a crime as defined in international law. Since there is no dispute in this appeal that the elements of the crimes have been carried out by the government of the DRC, we are concerned here with modes of commission.
Whether an individual is complicit in an international crime cannot be considered in light of only one of the world’s legal systems: Ramirez, at p. 315; Moreno v. Canada (Minister of Employment and Immigration),  1 F.C. 298 (C.A.), at p. 323. This flows not only from the explicit instruction in art. 1F(a) to apply international law, but also from the extraordinary nature of international crimes. They simply transcend domestic norms. As Fannie Lafontaine explains in Prosecuting Genocide, Crimes Against Humanity and War Crimes in Canadian Courts (2012), at p. 95:
Genocide, crimes against humanity and war crimes, because of their very raison d’être, their particular magnitude and the context of their commission, cannot be assimilated to ordinary crimes, regardless of the latter crimes’ intrinsic gravity.
International criminal law, while built upon domestic principles, has adapted the concept of individual responsibility to this setting of collective and large-scale criminality, where crimes are often committed indirectly and at a distance. As Gerhard Werle puts it, at p. 954:
When allocating individual responsibility within networks of collective action, it must be kept in mind that the degree of criminal responsibility does not diminish as distance from the actual act increases; in fact, it often grows. Adolf Hitler, for example, sent millions of people to their deaths without ever laying a hand on a victim himself. And mass killer Adolf Eichmann organized the extermination of European Jews from his office in the Berlin headquarters of the “Reichssicherheitshauptamt” of the SS.
We are therefore required by both the text of art. 1F(a) and the realities of international crime to look beyond the bounds of Canadian criminal law. We must refrain from interpreting and applying international criminal law as if it were simply the mirror of our domestic criminal law: Cassese’s International Criminal Law (3rd ed. 2013), revised by A. Cassese, et al., at pp. 6-7.
The question is – what are the relevant sources of international criminal law?
In our view, the best place to start is the Rome Statute. As Lord Brown of Eaton-under-Heywood J.S.C. recognized in R. (J.S. (Sri Lanka)) v. Secretary of State for the Home Department,  UKSC 15,  A.C. 184 (“J.S.”), at para. 9:
It is convenient to go at once to the [Rome] Statute, ratified as it now is by more than 100 states and standing as now surely it does as the most comprehensive and authoritative statement of international thinking on the principles that govern liability for the most serious international crimes (which alone could justify the denial of asylum to those otherwise in need of it).
Canada’s acceptance of the Rome Statute as authority on international criminal principles is beyond dispute. Canada is not only party to the Rome Statute, Parliament has implemented the treaty into domestic law through the Crimes Against Humanity and War Crimes Act, S.C. 2000, c. 24.
Article 25 of the Rome Statute provides extensive descriptions of modes of commission. These enumerated modes of liability have been described as the culmination of the international community’s efforts to codify individual criminal responsibility under international law: A. Cassese, “From Nuremberg to Rome: International Military Tribunals to the International Criminal Court”, in A. Cassese, P. Gaeta and J. R. W. D. Jones, eds., The Rome Statute of the International Criminal Court: A Commentary, vol. I (2002), 3, at pp. 3-4); and E. van Sliedregt, Individual Criminal Responsibility in International Law (2012), at pp. 74-75.
That said, we may not rely exclusively on the approach of the International Criminal Court (“ICC”) to complicity. Despite its importance, the Rome Statute cannot be considered as a complete codification of international criminal law. International criminal law derives from a diversity of sources which include the growing body of jurisprudence of international criminal courts: Cassese’s International Criminal Law, at pp. 9-21. Article 1F(a) of the Refugee Convention refers generally to international instruments and the ICC itself has relied on the jurisprudence of ad hoc tribunals to interpret its own statute: Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10-465-Red, Decision on the Confirmation of Charges, 16 December 2011 (ICC, Pre-Trial Chamber I), at para. 280. See also B. Goy, “Individual Criminal Responsibility before the International Criminal Court: A Comparison with the Ad Hoc Tribunals” (2012), 12 Int’l Crim. L. Rev. 1, at p. 4. In Mugesera, at paras. 82 and 126, this Court highlighted the international law expertise of the ad hoc tribunals and explained that the decisions of the International Criminal Tribunal for the Former Yugoslavia (“ICTY”) and the International Criminal Tribunal for Rwanda “should not be disregarded lightly by Canadian courts applying domestic legislative provisions .... which expressly incorporate customary international law”: para. 126. Accordingly, while our focus will remain on the most recent codification of international criminal law in the Rome Statute, we will also consider other sources, more particularly the jurisprudence of the ad hoc tribunals.
As explained above, we are concerned here with the dividing line between mere association and culpable complicity. While further distinctions between modes of commission may be important for sentencing purposes, exclusion from refugee protection applies when there are serious reasons for considering that an individual has committed an international crime, whatever the mode of commission happens to be. Our task then is to identify threshold criteria for the application of the exclusionary clause, art. 1F(a) of the Refugee Convention. Accordingly, the broadest modes of commission recognized under current international criminal law are most relevant to our complicity analysis, namely, common purpose liability under art. 25(3)(d) of the Rome Statute and joint criminal enterprise developed in the ad hoc jurisprudence.
These two related modes have adapted the concept of individual criminal responsibility to the collective aspects of international crime. However, as the following analysis will show, individual criminal responsibility has not been stretched so far as to capture complicity by mere association or passive acquiescence. In other words, when we look to international criminal law for guidance, even the broadest modes of commission require a link between the individual and the crime or criminal purpose of a group. Therefore, to the extent that the Federal Court of Appeal’s reasons expand complicity under art. 1F(a) such a way that it includes mere complicity by association or passive acquiescence, they should not be followed. We shall return below, in greater detail, to this issue.
E. Common Purpose Under Article 25(3)(d) of the Rome Statute
Article 25(3)(d) of the Rome Statute recognizes a broad residual mode of commission, by capturing conduct that “[i]n any other way contributes” to a crime committed or attempted by a group acting with a common purpose:
See Cassese’s International Criminal Law, pp. 175-76.
In other words, art. 25(3)(d) captures contributions to a crime where an individual did not have control over the crime and did not make an essential contribution as required for co-perpetration under art. 25(3)(a), did not incite, solicit or induce the crime under art. 25(3)(b), or did not intend to aid or abet a certain specific crime under art. 25(3)(c): Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06-2842, Judgment pursuant to Article 74 of the Statute, 14 March 2012 (ICC, Trial Chamber I), at para. 999; Prosecutor v. William Samoei Ruto, ICC-01/09-01/11-373, Decision on the Confirmation of Charges, 23 January 2012 (ICC, Pre-Trial Chamber II), at para. 354; Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10-514, Judgment on the Prosecutor’s Appeal against the Decision on the Confirmation of Charges, 30 May 2012 (ICC, Appeals Chamber), at para. 8, per Judge Silvia Fernández de Gurmendi, concurring; Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06-803-tEN, Decision on the Confirmation of Charges, 29 January 2007 (ICC, Pre-Trial Chamber I), at para. 337; and Lafontaine, at pp. 237-38.
The actus reus under para. (d) is distinguishable from the preceding paragraphs under art. 25(3) primarily by the magnitude of contribution required. While the jurisprudence is not completely settled, a pre-trial chamber of the ICC has said that the level of contribution required by art. 25(3)(d) is lower than the forms of commission under paras. (a) to (c). Where commission under para. (a) requires an essential contribution, and para. (c) a substantial one, Pre-Trial Chamber I has concluded that art. 25(3)(d) requires only a significant contribution: Mbarushimana, at paras. 279-85.
While the phrase “any other way” captures every imaginable contribution in a qualitative sense, it does not necessarily apply as broadly in a quantitative sense. Not every contribution, no matter how minor, will be caught by art. 25(3)(d). Setting the threshold at significant contribution is critical. As Pre-Trial Chamber I of the ICC said in Mbarushimana, at para. 277:
.... such a threshold is necessary to exclude contributions which, because of their level or nature, were clearly not intended by the drafters of the Statute to give rise to individual criminal responsibility. For instance, many members of a community may provide contributions to a criminal organisation in the knowledge of the group’s criminality, especially where such criminality is public knowledge. Without some threshold level of assistance, every landlord, every grocer, every utility provider, every secretary, every janitor or even every taxpayer who does anything which contributes to a group committing international crimes could satisfy the elements of 25(3)(d) liability for their infinitesimal contribution to the crimes committed.
The pre-trial chamber went on to explain that the significance of a contribution will depend on the facts of each case, “as it is only by examining a person’s conduct in proper context that a determination can be made as to whether a given contribution has a larger or smaller effect on the crimes committed”: Mbarushimana, at para. 284. On an appeal by the Prosecutor, the majority of the Appeals Chamber declined to determine the degree of contribution required under art. 25(3)(d): Mbarushimana, at paras. 65-68.
As for the mens rea requirement, the text of art. 25(3)(d) states that a contribution must be intentional, “made with the aim of furthering the criminal activity or purpose of the group” or “in the knowledge of the intention of the group to commit the crime”. The Pre-Trial Chamber I explained in Mbarushimana, at para. 289, that individuals may be complicit in crimes without possessing the mens rea required by the crime itself:
Differently from aiding and abetting under article 25(3)(c) of the Statute, for which intent is always required, knowledge is sufficient to incur liability for contributing to a group of persons acting with a common purpose, under article 25(3)(d) of the Statute. Since knowledge of the group’s criminal intentions is sufficient for criminal responsibility, it is therefore not required for the contributor to have the intent to commit any specific crime and not necessary for him or her to satisfy the mental element of the crimes charged.
While the subjective element under art. 25(3)(d) can take the form of intent (accused intends to contribute to a group’s criminal purpose) or knowledge (accused is aware of the group’s intention to commit crimes), recklessness is likely insufficient. The text of art. 25(3)(d) itself does not refer to conduct that might contribute to a crime or criminal purpose, and the mental element codified by art. 30 has been held to exclude dolus eventualis, that is, the awareness of a mere risk of prohibited consequences: Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08-424, Decision on the Confirmation of Charges, 15 June 2009 (ICC, Pre-Trial Chamber II), at para. 360. We note that Pre-Trial Chamber I took a different view of art. 30 in Lubanga, at paras. 351-55.
As the foregoing demonstrates, complicity under art. 25(3)(d) is not based on rank within or association with a group, but on intentionally or knowingly contributing to a group’s crime or criminal purpose.
F. Joint Criminal Enterprise
Having considered the broadest form of accessory liability under the Rome Statute, we now turn to what is perhaps the broadest and most controversial mode of liability recognized by the ad hoc tribunals: joint criminal enterprise. See Cassese’s International Criminal Law, at pp. 163-75; Cryer, at p. 372.
Even though joint criminal enterprise is considered to be a form of principal liability, it is relevant to our task of setting threshold criteria for art. 1F(a) of the Refugee Convention. The line between principal and accessory is not necessarily drawn consistently across international and domestic criminal law. Joint criminal enterprise, like common purpose liability under art. 25(3)(d), captures “lesser” contributions to a crime than aiding and abetting. While aiding and abetting likely requires a substantial contribution to a certain specific crime, joint criminal enterprise and common purpose liability can arise from a significant contribution to a criminal purpose. To the extent that the ICTY Trial Chamber may be seen to have applied a more exacting standard in Prosecutor v. Jovica Stanišić, IT-03-69-T, Judgment, 30 May 2013 (ICTY, Trial Chamber I), it is not in accordance with prevailing appellate authority: Prosecutor v. Duško Tadić, IT-94-1-A, Judgment, 15 July 1999 (ICTY, Appeals Chamber), at para. 229, cited in Lafontaine, at p. 237; Prosecutor v. Radoslav Brđanin, IT-99-36-A, Judgment, 3 April 2007 (ICTY, Appeals Chamber), at paras. 427-28 and 430. Joint criminal enterprise therefore captures individuals who could easily be considered as secondary actors complicit in the crimes of others: Cryer, at p. 372; S. Manacorda and C. Meloni, “Indirect Perpetration versus Joint Criminal Enterprise: Concurring Approaches in the Practice of International Criminal Law?” (2011), 9 J.I.C.J. 159, at p. 166.
In Tadić, the ICTY articulated three forms of joint criminal enterprise: paras. 196-206. For all three, the actus reus is a “significant” contribution to the criminal enterprise: Brđanin, at para. 430.
However, the mens rea varies for each form. The first form, JCE I, requires shared intent to perpetrate a certain crime. The second, JCE II, requires knowledge of a system of ill treatment and intent to further this system. The third, JCE III, requires intention to participate in and further the criminal activity or purpose of the group, and intent to contribute to the joint criminal enterprise or the commission of a crime by the group. Under JCE III, liability can extend to a crime other than one agreed to in the common plan if the accused intended to participate in and further the criminal activity of the group and (i) it was foreseeable that such a crime might be perpetrated by members of the group and (ii) the accused willingly took that risk. In other words, where an accused intends to contribute to the common purpose, JCE III captures not only knowing contributions but reckless contributions: see Tadić, at para. 228.
Despite the overlap between joint criminal enterprise and art. 25(3)(d), ICC jurisprudence has kept the two modes distinct. Commentators suggest that JCE III will not play a role at the ICC, largely because of the recklessness component: van Sliedregt, at p. 101; Lafontaine, at p. 238; A. Cassese, “The Proper Limits of Individual Responsibility under the Doctrine of Joint Criminal Enterprise”, (2007), 5 J.I.C.J. 109, at p. 132; Manacorda and Meloni, at p. 176.
For our purposes, we simply note that joint criminal enterprise, even in its broadest form, does not capture individuals merely based on rank or association within an organization or an institution: Cassese’s International Criminal Law, at p. 163. It requires that the accused have made, at a minimum, a significant contribution to the group’s crime or criminal purpose, made with some form of subjective awareness (whether it be intent, knowledge, or recklessness) of the crime or criminal purpose. In other words, this form of liability, while broad, requires more than a nexus between the accused and the group that committed the crimes. There must be a link between the accused’s conduct and the criminal conduct of the group: Brđjanin, at paras. 427-28; Lafontaine, at p. 234; Cryer, at p. 369.
G. Summary of Complicity under International Law
In sum, while the various modes of commission recognized in international criminal law articulate a broad concept of complicity, individuals will not be held liable for crimes committed by a group simply because they are associated with that group, or because they passively acquiesced to the group’s criminal purpose. At a minimum, complicity under international criminal law requires an individual to knowingly (or, at the very least, recklessly) contribute in a significant way to the crime or criminal purpose of a group.
H. Comparative Law and Decisions of Other National Courts
Other state parties to the Refugee Convention have approached art. 1F(a) in a manner that adheres to the minimum requirements for complicity set by the international law principles discussed above.
In J.S., the U.K. Supreme Court rejected the presumption that an individual is complicit in war crimes if he joins an organization, even where that organization has a limited and brutal purpose. Lord Hope of Craighead D.P.S.C., concurring, stated that “mere membership of an organisation that is committed to the use of violence for political ends is not enough to bring an appellant within the exclusion clauses”: para. 43; see also paras. 31 and 44. Rather, as Lord Kerr of Tonaghmore, J.S.C., also concurring, wrote, decision makers must “concentrate on the actual role played by the particular person, taking all material aspects of that role into account so as to decide whether the required degree of participation is established”: para. 55. In his view, this approach “accord[s] more closely .... with the spirit of articles 25 and 30 of the ICC Rome Statute”: at para. 57.
According to J.S., an individual would only be excluded under art. 1F(a) “if there are serious reasons for considering him voluntarily to have contributed in a significant way to the organisation’s ability to pursue its purpose of committing war crimes, aware that his assistance will in fact further that purpose”: para. 38.
To assess the accused’s mental state and degree of participation, J.S. provides factors that are remarkably similar to those used by Canadian courts in art. 1F(a) cases [para. 30]:
These factors are largely subsumed by the six “non-exhaustive” factors set out in Ryivuze v. Canada (Minister of Citizenship and Immigration), 2007 FC 134, 325 F.T.R. 30, at para. 38:
the nature of the organization;
the method of recruitment;
position/rank in the organization;
knowledge of the organization’s atrocities;
the length of time in the organization; and
the opportunity to leave the organization.
The factors recognized in U.K. and Canadian jurisprudence help guard against a complicity analysis that would exclude individuals from refugee protection on the basis of mere membership or failure to dissociate from a multi-faceted organization which is committing war crimes.
Similarly, United States appellate jurisprudence on refugee exclusions does not recognize complicity based on passive acquiescence or “tangential” contributions. While the U.S. “persecutor bar” does not directly incorporate art. 1F(a), it nevertheless represents an approach to exclusion that would only capture those who have committed international crimes as recognized by international criminal law: A.F., at para. 167. A recent decision by the Second Circuit Court of Appeals said:
.... the mere fact that [a person] may be associated with an enterprise that engages in persecution is insufficient by itself to trigger the effects of the persecutor bar. As the Supreme court’s oft-quoted dicta in Fedorenko v. United States, 449 U.S. 490, 101 S.Ct. 737, 66 L.Ed.2d 686 (1981), illustrates, a “guilt by association” approach to the persecutor bar is improper ....
.... Before [a claimant] may be held personally accountable for assisting in acts of persecution, there must be some evidence that he himself engaged in conduct that assisted in the persecution of another.
(Xu Sheng Gao v. United States Attorney General, 500 F.3d 93 (2007), at paras. 5-6, cited in P. Zambelli, “Problematic Trends in the Analysis of State Protection and Article 1F(a) Exclusion in Canadian Refugee Law” (2011), 23 Int’l. J. Refugee L. 252, at pp. 284-85.)
In our view, the approach to complicity adopted by these state parties adheres to the UNHCR’s recommendation in its Guidelines, at para. 18, although it would ask for a “substantial” contribution:
For exclusion to be justified, individual responsibility must be established in relation to a crime covered by Article 1F .... In general, individual responsibility flows from the person having committed, or made a substantial contribution to the commission of the criminal act, in the knowledge that his or her act or omission would facilitate the criminal conduct. The individual need not physically have committed the criminal act in question. Instigating, aiding and abetting and participating in a joint criminal enterprise can suffice.
In sum, the foregoing approaches to complicity all require a nexus between the individual and the group’s crime or criminal purpose. An individual can be complicit without being present at the crime and without physically contributing to the crime. However, the UNHCR has explained, and other states parties have recognized, that to be excluded from the definition of refugee protection, there must be evidence that the individual knowingly made at least a significant contribution to the group’s crime or criminal purpose. Passive membership would not be enough, as indicated above in paras. 70-76.
I. The Canadian Approach to Criminal Participation Has Been Overextended
Before being overturned by the Federal Court of Appeal, the Federal Court’s decision in this case was viewed as a potential signal of “a clearer jurisprudence, more closely tied to international standards and to the original wording of the Convention”: A. Kaushal and C. Dauvergne, “The Growing Culture of Exclusion: Trends in Canadian Refugee Exclusions” (2011), 23 Int’l. J. of Refugee L. 54, at p. 85. The Federal Court rightly concluded that neither mere membership in a government that had committed international crimes nor knowledge of those crimes is enough to establish complicity: para. 4.
In our view, the Federal Court’s approach in this case brings appropriate restraint to the test for complicity that had, in some cases, inappropriately shifted its focus towards the criminal activities of the group and away from the individual’s contribution to that criminal activity: see, for example, Osagie v. Canada (Minister of Citizenship and Immigration) (2000), 186 F.T.R. 143; Mpia-Mena-Zambili v. Canada (Minister of Citizenship and Immigration), 2005 FC 1349, 281 F.T.R. 54, at paras. 45-47; Fabela v. Canada (Minister of Citizenship and Immigration), 2005 FC 1028, 277 F.T.R. 20, at paras. 14-19. By answering “yes” to the certified question, the Federal Court of Appeal’s reasons could be seen as having endorsed an overextended approach to complicity, one that captures complicity by association or passive acquiescence.
As Noël J.A. noted in this case, a senior official may be complicit in the government’s crimes “by remaining in his or her position without protest and continuing to defend the interests of his or her government while being aware of the crimes”. Nonetheless, the Federal Court of Appeal reasons should not be improperly relied on to find complicity even where the individual has committed no guilty act and has no criminal knowledge or intent, beyond a mere awareness that other members of the government have committed illegal acts.
In our view, it is necessary to rearticulate the Canadian approach to art. 1F(a) to firmly foreclose exclusions based on such broad forms of complicity. Otherwise, high ranking officials might be forced to abandon their legitimate duties during times of conflict and national instability in order to maintain their ability to claim asylum. Furthermore, a concept of complicity that leaves any room for guilt by association or passive acquiescence violates two fundamental criminal law principles.
It is well established in international criminal law that criminal liability does not attach to omissions unless an individual is under a duty to act: Cassese’s International Criminal Law, pp. 180-82. Accordingly, unless an individual has control or responsibility over the individuals committing international crimes, he or she cannot be complicit by simply remaining in his or her position without protest: Ramirez, at pp. 319-20. Likewise, guilt by association violates the principle of individual criminal responsibility. Individuals can only be liable for their own culpable conduct: van Sliedregt, at p. 17.
Accordingly, the decision of the Federal Court of Appeal should not be taken to leave room for rank-based complicity by association or passive acquiescence. Such a reading would perpetuate a departure from international criminal law and fundamental criminal law principles.
J. The Canadian Test for Complicity Refined
In light of the foregoing reasons, it has become necessary to clarify the test for complicity under art. 1F(a). To exclude a claimant from the definition of “refugee” by virtue of art. 1F(a), there must be serious reasons for considering that the claimant has voluntarily made a significant and knowing contribution to the organization’s crime or criminal purpose.
We will address these key components of the contribution-based test for complicity in turn. In our view, they ensure that decision makers do not overextend the concept of complicity to capture individuals based on mere association or passive acquiescence.
(1) Voluntary Contribution to the Crime or Criminal Purpose
It goes without saying that the contribution to the crime or criminal purpose must be voluntarily made. While this element is not in issue in this case, it is easy to foresee cases where an individual would otherwise be complicit in war crimes but had no realistic choice but to participate in the crime. 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. The voluntariness requirement captures the defence of duress which is well recognized in customary international criminal law, as well as in art. 31(1)(d) of the Rome Statute: Cassese’s International Criminal Law, pp. 215-16.
(2) Significant Contribution to the Group’s Crime or Criminal Purpose
In our view, mere association becomes culpable complicity for the purposes of art. 1F(a) when an individual makes a significant contribution to the crime or criminal purpose of a group. As Lord Brown J.S.C. said in J.S., to establish the requisite link between the individual and the group’s criminal conduct, the accused’s contribution does not have to be “directed to specific identifiable crimes” but can be directed to “wider concepts of common design, such as the accomplishment of an organisation’s purpose by whatever means are necessary including the commission of war crimes”: para. 38. This approach to art. 1F(a) is consistent with international criminal law’s recognition of collective and indirect participation in crimes discussed above, as well as s. 21(2) of the Canadian Criminal Code, R.S.C. 1985, c. C-46, which attaches criminal liability based on assistance in carrying out a common unlawful purpose.
Given that contributions of almost every nature to a group could be characterized as furthering its criminal purpose, the degree of the contribution must be carefully assessed. The requirement of a significant contribution is critical to prevent an unreasonable extension of the notion of criminal participation in international criminal law.
(3) Knowing Contribution to the Crime or Criminal Purpose
To be complicit in crimes committed by the government, the official must be aware of the government’s crime or criminal purpose and aware that his or her conduct will assist in the furtherance of the crime or criminal purpose.
In our view, this approach is consistent with the mens rea requirement under art. 30 of the Rome Statute. Article 30(1) explains that “a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court only if the material elements are committed with intent and knowledge”. Article 30(2)(a) explains that a person has intent where he “means to engage in the conduct”. With respect to consequences, art. 30(2)(b) requires that the individual “means to cause that consequence or is aware that it will occur in the ordinary course of events”. Knowledge is defined in art. 30(3) as “awareness that a circumstance exists or a consequence will occur in the ordinary course of events”.
(4) Applying the Test
Whether there are serious reasons for considering that an individual has committed international crimes will depend on the facts of each case. Accordingly, to determine whether an individual’s conduct meets the actus reus and mens rea for complicity, several factors may be of assistance. The following list combines the factors considered by courts in Canada and the U.K., as well as by the ICC. It should serve as a guide in assessing whether an individual has voluntarily made a significant and knowing contribution to a crime or criminal purpose:
the size and nature of the organization;
the part of the organization with which the refugee claimant was most directly concerned;
the refugee claimant’s duties and activities within the organization;
the refugee claimant’s position or rank in the organization;
the length of time the refugee claimant was in the organization, particularly after acquiring knowledge of the group’s crime or criminal purpose; and
the method by which the refugee claimant was recruited and the refugee claimant’s opportunity to leave the organization.
See Ryivuze, at para. 38; J.S., at para. 30; and Mbarushimana, Decision on the Confirmation of Charges, at para. 284.
When relying on these factors for guidance, the focus must always remain on the individual’s contribution to the crime or criminal purpose. Not only are the factors listed above diverse, they will also have to be applied to diverse circumstances encompassing different social and historical contexts. Refugee claimants come from many countries and appear before the Board with their own life experiences and backgrounds in their respective countries of origin. Thus, the assessment of the factors developed in our jurisprudence, the decisions of the courts of other countries, and the international community will necessarily be highly contextual. Depending on the facts of a particular case, certain factors will go “a long way” in establishing the requisite elements of complicity. Ultimately, however, the factors will be weighed with one key purpose in mind: to determine whether there was a voluntary, significant, and knowing contribution to a crime or criminal purpose.
In the present case, it will be for the Board to determine which factors are significant, based on the application before it. To provide guidance to the Board in making this determination, it may be of assistance to briefly elaborate on each of the factors listed above.
The size and nature of the organization. The size of an organization could help determine the likelihood that the claimant would have known of and participated in the crime or criminal purpose. A smaller organization could increase that likelihood. That likelihood could also be impacted by the nature of the organization. If the organization is multifaceted or heterogeneous, i.e. one that performs both legitimate and criminal acts, the link between the contribution and the criminal purpose will be more tenuous. In contrast, where the group is identified as one with a limited and brutal purpose, the link between the contribution and the criminal purpose will be easier to establish. In such circumstances, a decision maker may more readily infer that the accused had knowledge of the group’s criminal purpose and that his conduct contributed to that purpose. That said, even for groups with a limited and brutal purpose, the individual’s conduct and role within the organization must still be carefully assessed, on an individualized basis, to determine whether the contribution was voluntarily made and had a significant impact on the crime or criminal purpose of the group.
The part of the organization with which the refugee claimant was most directly concerned. This factor may be relevant if particular parts of the organization were known to be involved with the crime or criminal purpose. For example, where only one part of the organization in question was involved in the crime or criminal purpose, a claimant’s exclusive affiliation with another part(s) of the organization may serve to exonerate him or her for the purpose of art. 1F(a).
The refugee claimant’s duties and activities within the organization. This factor is likely to be significant in any analysis of complicity, because it goes to the heart of a claimant’s day to day participation in the activities of the organization. The Board should consider the link between the duties and activities of a claimant, and the crimes and criminal purposes of the organization.
The refugee claimant’s position or rank in the organization. A high ranking individual in an organization may be more likely to have knowledge of that organization’s crime or criminal purpose. In some cases, a high rank or rapid ascent through the ranks of an organization could evidence strong support of the organization’s criminal purpose. Moreover, by virtue of their position or rank, individuals may have effective control over those directly responsible for criminal acts, possibly engaging art. 28 of the Rome Statute.
The length of time the refugee claimant was in the organization, particularly after acquiring knowledge of the group’s crime or criminal purpose. It may be easier to establish complicity where an individual has been involved with the organization for a longer period of time. This would increase the chance that the individual had knowledge of the organization’s crime or criminal purpose. A lengthy period of involvement may also increase the significance of an individual’s contribution to the organization’s crime or criminal purpose.
The method by which the refugee claimant was recruited and the refugee claimant’s opportunity to leave the organization. As mentioned, these two factors directly impact the voluntariness requirement. This requirement may not be satisfied if an individual was coerced into joining, supporting, or remaining in the organization. Similarly, an individual’s involvement with an organization may not be voluntary if he or she did not have the opportunity to leave, especially after acquiring knowledge of its crime or criminal purpose. The Board may wish to consider whether the individual’s specific circumstances (i.e. location, financial resources, and social networks) would have eased or impeded exit.
We reiterate that the factors discussed above should be relied on only for guidance. We agree with Lord Kerr J.S.C.’s statement in J.S., at para. 55:
.... they are not necessarily exhaustive of the matters to be taken into account, nor will each of the factors be inevitably significant in every case. One needs, I believe, to concentrate on the actual role played by the particular person, taking all material aspects of that role into account so as to decide whether the required degree of participation is established.
A full contextual analysis would necessarily include any viable defences, including, but certainly not limited to, the defence of duress, discussed above.
K. Evidentiary Standard: Serious Reasons for Considering
Ultimately, the above contribution-based test for complicity is subject to the unique evidentiary standard contained in art. 1F(a) of the Refugee Convention. To recall, the Board does not make determinations of guilt. Its exclusion decisions are therefore not based on proof beyond a reasonable doubt nor on the general civil standard of the balance of probabilities. Rather, art. 1F(a) directs it to decide whether there are “serious reasons for considering” that an individual has committed war crimes, crimes against humanity or crimes against peace. For guidance on applying the evidentiary standard, we agree with the Lord Brown J.S.C.’s reasons in J.S., at para. 39:
It would not, I think, be helpful to expatiate upon article 1F’s reference to there being “serious reasons for considering” the asylum seeker to have committed a war crime. Clearly the tribunal in Gurung’s case  Imm AR 115 (at the end of para. 109) was right to highlight “the lower standard of proof applicable in exclusion clause cases” – lower than that applicable in actual war crimes trials. That said, “serious reasons for considering” obviously imports a higher test for exclusion than would, say, an expression like “reasonable grounds for suspecting”. “Considering” approximates rather to “believing” than to “suspecting”. I am inclined to agree with what Sedley LJ said in Al-Sirri v Secretary of State for the Home Department  Imm AR 624, para. 33:
In our view, this unique evidentiary standard is appropriate to the role of the Board and the realities of an exclusion decision addressed above. The unique evidentiary standard does not, however, justify a relaxed application of fundamental criminal law principles in order to make room for complicity by association.
For the foregoing reasons, we would allow the appeal, with costs throughout, and remit the matter to the Refugee Protection Division of the Immigration and Refugee Board for redetermination in accordance with these reasons. A new panel shall apply the contribution-based test for complicity outlined above. A detailed assessment is required to determine whether the particular facts of this case establish serious reasons for considering that the actus reus and mens rea for complicity are present and therefore justify excluding the appellant from the definition of refugee by operation of art. 1F(a) of the Refugee Convention.
Jared Will, Annick Legault and Peter Shams (instructed by Jared Will, Montréal; Boisclair & Legault, Montréal; Peter Shams, Montréal), for the appellant.
François Joyal and Ginette Gobeil (instructed by Attorney General of Canada, Montréal), for the respondent.
all rights reserved