- Program in Refugee and Asylum Law
- Additional Resources
If you have no capacity for violence then you are a healthy productive citizen: a sheep. If you have a capacity for violence and no empathy for your fellow citizens, then you have defined an aggressive sociopath—a wolf. But what if you have a capacity for violence, and a deep love for your fellow citizens? Then you are a sheepdog, a warrior, someone who is walking the hero’s path.
I am the good shepherd. The good shepherd lays down his life for the sheep.
It is not without irony that the Court of Justice of the European Union (CJEU) decided an asylum case involving a former U.S. army soldier with the surname “Shepherd” at the same time U.S. pop culture celebrates a film built around a metaphor of protecting sheep. In Clint Eastwood’s American Sniper, we learn that the U.S. Army’s “most lethal sniper,” Chris Kyle, grew up listening to his father preach a story about “three types of people in this world: sheep, wolves, and sheepdogs.” The father’s story, modeled after a book by U.S. Army Veteran Dave Grossmann, taught Kyle that sheepdogs are “blessed with aggression, and the overpowering need to protect the flock” from the wolves. Kyle was part of the rare breed of the sheepdog. During his deployment in the most recent Iraq war, he killed 160 “wolves.”
Mr. Andre Lawrence Shepherd cannot expect the same admiration as his U.S. Army comrade Kyle. Shepherd was first deployed in Iraq from September 2004 to February 2005. Trained as a helicopter maintenance mechanic, he did not directly participate in combat, but repaired and maintained helicopters, in particular Apache combat helicopters. In April 2007, after a two-year deployment in Germany as part of an air support battalion, Shepherd was ordered back to Iraq. Instead of returning to Iraq, however, he left the army. He considered the war in Iraq illegal and alleged that the U.S. Army committed war crimes. Shepherd requested international protection from the German immigration authorities; they refused his claim. On appeal before the Administrative Court of Munich, it was undisputed that Shepherd, if returned to the U.S., would face prosecution and punishment for his refusal to serve in Iraq. Nor was it contested that, if returned to U.S., Shepherd would face social ostracism. Shepherd did have a real chance of an objective risk – a well-founded fear, as described by the Refugee Convention. The disputed issue was whether the harm he faced was relevant harm for the notion of “being persecuted,” a term EU Asylum law borrows from Article 1A(2) of the Refugee Convention.
The EU Qualification Directive (QD) adopts the refugee definition in Article 1(A)(2) of the Refugee Convention. It defines persecution with respect to human rights. Persecutory acts are either “sufficiently serious by . . . nature or repetition” so that they would be “a severe violation of basic human rights” (such as those rights of which no derogation can be made under the European Convention of Human Rights), or are “an accumulation of various measures, including violations of human rights” that are “sufficiently severe as to affect an individual in a similar manner.” The QD then gives examples of what could be regarded as an “act of persecution.” According to one example in Article 9(2)(e), an act of persecution is the “. . . prosecution or punishment for refusal to perform military service in a conflict, where performing military service would include crimes or acts falling under the exclusion clauses as set out in Article 12(2).” Those exclusion clauses cover both war crimes and crimes against peace.
While the provision of Article 9(2)(e) is plain on first blush, it raises more questions than it answers: Was the involvement necessary in order for military service to “include” such acts or crimes? What is the standard of probability? What circumstances are relevant to assess whether such acts are included in military service? Why does the provision include acts and crimes in the plural form? What if an act is an infringement of international humanitarian law but there exists a plausible defense? Finally, what if military service does not include an act falling under the exclusion clauses but would be irreconcilable with a person’s conscience? What are other forms of relevant harms for conscientious objection? These are, in essence, the questions the German referring court sought the CJEU to clarify.
The CJEU delivered its answer to some of these questions in a judgment on February 26. Some answers lie, according to the CJEU, in the text of the QD, or rather in the lack of text thereof: According to the court, there is nothing in Article 9(2)(e) that suggests logistical and support staff is not covered under the provision, or that acts falling under the exclusion clauses must have taken place in the past. Furthermore, the CJEU points out that the text names military service in a “conflict” only and that Article 9(2)(e) “does not refer solely to the situation in which the applicant would be led to commit such crimes personally.” Thus, “situations in which the applicant would participate only indirectly in the commission of such crimes (because, inter alia, he is not a member of the combat troops but rather, for example, serves in a unit providing logistical or technical support) are not, as a matter of principle, excluded.” For that very reason, an action in an armed conflict does not have to be of such nature to give rise to a criminal conviction, if it ever were to be prosecuted.
Whoever had hoped for an inclusive judgment will see his or her hope end here. The CJEU proceeds to clarify what level of involvement is required for Article 9(2)(e). It declares that persons falling under the Article must perform tasks that “sufficiently directly and reasonably plausibly . . . lead them to participate in . . . [acts falling under the exclusion clause].” The Court points out that that past war crimes are not sufficient to make future war crimes credible, and creates, in effect, a presumption against the commission of war crimes in Security Council-mandated operations and in cases where there are domestic criminal provisions against such crimes.
[…] an armed intervention engaged upon on the basis of a resolution adopted by that Security Council offers, in principle, every guarantee that no war crimes will be committed and that the same applies, in principle, to an operation which gives rise to an international consensus. Accordingly, although the possibility can never be excluded that acts contrary to the very principles of the Charter of the United Nations will be committed in war operations, the fact that the armed intervention takes place in such a context must be taken into account. […] importance must also be attached to the possibility that the State or States which conduct the operations prosecute war crimes. The existence, in the legal system of those States, of legislation penalising war crimes and of courts which ensure the effective punishment of those who commit such crimes is liable to render implausible the hypothesis that a soldier of one of those States could be led to commit such crimes and, accordingly, may in no case be disregarded.
The Court thus places confidence in the deterrent effect of criminal law within individual states. In reality, this confidence is ill placed. For example, U.S. violations of international humanitarian law through the use of torture in Iraq are well documented. Also, the court mixes up the basic distinction between the ius ad bellum and the ius in bello. It is correct that not every soldier participating in a military operation that is unlawful under the ius ad bellum – the law on the use of force – commits a crime against peace (which counsel for the applicant appeared to have argued). Contrary to what has been suggested elsewhere, the CJEU therefore never had to seize on the legality of the Iraq war. Equally correct is that no act will amount to a war crime (presupposing a violation of norms of international humanitarian law) simply because the military operation in which the act happens is unlawful under the ius ad bellum. But neither is the reverse true: Whether or not violations of international humanitarian law occur, and thus prima facie war crimes, is simply independent of the lawfulness of a military campaign under the ius ad bellum.
The confusion continues. The CJEU concludes that the applicant must establish that the conditions under which tasks are performed in the military must make it “highly likely” that acts falling under the exclusion clauses will be committed – which both disrespects the shared burden of fact finding under the Refugee Convention and the QD, and creates an almost insurmountable burden for an applicant.
it is for the person seeking refugee status under Article 9(2)(e) of Directive 2004/83 to establish with sufficient plausibility that his unit carries out operations assigned to it, or has carried them out in the past, in such conditions that it is highly likely that acts such as those referred to in that provision will be committed.
Yet, the Court also holds that the factual assessment “is for the national authorities alone to carry out” and that it must, “be based on a body of evidence capable of establishing, in view of all the circumstances of the case . . . that the situation in question makes it credible that the alleged war crimes would be committed.” In effect, the probability required for Article 9(2)(e) remains unclear.
Once it is suggested that it is “highly likely” that an act falling under the exclusion clauses will be committed, it must be established that the applicant had no other choice but to request international protection. Therefore, it is particularly relevant whether or not he availed himself to a procedure for obtaining conscientious objector status, failure to do so “excludes any protection” under the QD, according to the CJEU. Finally, even if military service would likely include the commission of acts falling under the exclusion clauses and the applicant has made use of domestic conscientious objector procedures, prosecution and penalties must “go beyond what is necessary for the State concerned in order to exercise its legitimate right to maintain an armed force.”
At this point, the Court has misread the approach of the QD – which has to be in accordance with the Refugee Convention. The QD, in agreement with practice on the Refugee Convention around the world, requires a human rights approach to interpret the notion of “being persecuted.” It is a separate debate to what extent human rights law, for matters of conscience or political or religious conviction, includes a right to conscientious objection even if military service would not include acts that fall under the exclusion clauses of the Refugee Convention. That was one of, but not the primary, question in Shepherd v. Bundesrepublik Germany. The QD states that prosecution or punishment are “acts of persecution” if directed against an applicant who escapes involvement in war crimes or crimes against peace, by refusing to perform military service. A closer look at the list in Article 9(2) reveals that the examples listed are examples that would be unlawful acts under human rights law and thus, do not require an additional assessment of whether they are justified or not. Consider these examples:
(a) acts of physical or mental violence, including acts of sexual violence;
(b) legal, administrative, police, and/or judicial measures which are in themselves discriminatory or which are implemented in a discriminatory manner;
(c) prosecution or punishment which is disproportionate or discriminatory;
(d) denial of judicial redress resulting in a disproportionate or discriminatory punishment
Under a human rights approach it appears that requiring prosecution or punishment of those who refused participation in war crimes is already relevant harm for the notion of “being persecuted.” Granted, domestic conscientious objector procedures, under such approach, appear relevant, but their effectiveness, as well as the failure to use these procedures should be assessed.
Back in Germany, Shepherd’s appeal is likely to be rejected. Here is why: Almost as if to make sure that Shepherd is not recognized as a refugee, the Court overstepped its ability to interpret EU law, rather than pronounce on the facts:
Although it can be seen from the information provided by the referring court that the applicant in the main proceedings runs the risk of a custodial sentence for desertion of 100 days to 15 months, or even of five years, nothing in the file submitted to the Court suggests that such measures clearly go beyond which is necessary for the State concerned to exercise its legitimate right to maintain an armed force.
Andre Lawrence Shepherd may or may not be a refugee. But sheep and non-violent shepherds can be refugees. Although hardly so in the world of the CJEU.
 Dave Grossmann, On Combat – The Psychology and Physiology of Deadly Conflict in War and Peace (2008)
 The Bible, John 10:11
 Grossmann, On Combat [n. 1].
 Beschluss vom 20.08.2013 – M 25 K 11.30288
 Directive 2011/95/EU of the European Parliament and of the Council of 13 December 2011 on standards for the qualification of third-country nationals or stateless persons as beneficiaries of international protection, for a uniform status for refugees or for persons eligible for subsidiary protection, and for the content of the protection granted (recast), OJ L337/9
 Beschluss vom 20.08.2013 – M 25 K 11.30288.
 Andre Lawrence Shepherd v Bundesrepublik Deutschland (C-472/13, Cout of Justice of the European Union, 26 February 2015)
 Ibid para 37.
 Ibid para 39.
 Ibid para 33.
 Ibid para 36.
 Ibid para. 37
 Ibid para 38.
 Ibid para 42.
 James C. Hathaway and Michelle Foster, The Law of Refugee Status (2 edn, Cambridge University Press 2014) Section 2.4.2.
 According to Article 4(1) of the Qualification Directive, “Member States may consider it the duty of the applicant
to submit as soon as possible all the elements needed to substantiate the application for international protection. In cooperation with the applicant, it is the duty of the Member State to assess the relevant elements of the application.”
 Andre Lawrence Shepherd v Bundesrepublik Deutschland, para 43. Emphasis added.
 Ibid, para 46.
 Ibid para 44-45.
 Hathaway and Foster (n. 17) section 3.2.
 Article 9(2) of the Directive 2011/95/EU, emphasis added.