In a little-noticed decision in April, the Court of Justice of the European Union (CJEU) significantly revised its approach to the doctrine of mutual trust among the member states. Even though the decision was issued only as an interpretation of the European Arrest Warrant, it will have profound consequences for the Area of Freedom Security and Justice more generally, including ongoing controversies concerning refugees.
Mutual trust in the area of freedom, security, and justice has been at the center of judicial disagreement across Europe. Well known, for instance, is the longstanding skirmish between the CJEU and the European Court of Human Rights (ECtHR) in Strasbourg about the limits of the Dublin system’s requirement that a refugee be returned to the Member State of first entry for processing. The ECtHR has held that any court – even a court of an EU member state – must first conduct an individualized inquiry to ensure that a refugee not be returned to another state if such return would create a “real risk” that the individual will suffer inhuman and degrading treatment. The CJEU, by contrast, has insisted that the EU’s foundational principle of mutual trust constrains such inquiries by one member state into the affairs of another. The CJEU limited a refugee’s claim against return to the member state of first entry to cases in which there is proof of a “systemic deficiency” in the member state of first entry. The CJEU’s insistence on “systemic” risk thus potentially foreclosed certain individualized claims of harm, unless they were accompanied by proof of broader deficiencies in the recipient country that were system wide. Member State courts caught in the middle between these two doctrines have shown some resistance to Luxembourg, with the U.K. Supreme Court, for example, saying it would disregard Luxembourg’s interpretation and heed the U.K.’s obligations to Strasbourg instead.
On another front, the CJEU has been doing battle with Member State courts on the European arrest warrant. Recall that in the landmark Melloni decision, the CJEU shut down the Spanish Constitutional Court’s objection that surrender of Mr. Melloni to Italy could be refused because it might violate his particular right against trial in absentia under the Spanish Constitution. The CJEU held that under the circumstances of that case there was no EU fundamental rights violation, and that Spain could thus not invoke an individual right under its own constitution to refuse surrender. In response, the Spanish Constitutional Court gave in – albeit not by following the ECJ directly, but by adjusting its own domestic jurisprudence under the Spanish Constitution.
As if in response to Luxembourg having swatted down Spain, the German Bundesverfassungsgericht (BVG) last December took a more creative line of attack. The BVG refused to surrender an American to Italy under an EAW request on the grounds that he had been convicted in absentia, and without proper notice. The facts were indeed far worse for Italy than they had been in Melloni, but the BVG (in contrast to the Spanish Constitutional Tribunal) did not risk asking Luxembourg. Instead, the BVG cleverly held that surrender under those circumstances would violate the individual’s dignity under the German Constitution, and would therefore compromise the constitutional identity of Germany. (Recasting the rights challenge as identity review allowed the BVG to circumvent its own Solange jurisprudence that would have limited the German court to reviewing the EU’s fundamental rights record only on what I like to call a “wholesale” basis, i.e. asking only whether, en gros, the EU/CJEU are taking rights seriously.) After a long discourse on the limits that national constitutional identity places on the EU’s principle of mutual trust, the German Court explained that there was, fortunately, no need in this case to block EU law. Why? Because in the German Court’s view, the European Arrest Warrant fortunately incorporates the protection of fundamental rights, including human dignity. Hence, surrender was not warranted under a proper interpretation of EU law itself. In a final twist, the German court said all this was so clear that there was no need for a reference to the CJEU.
Luxembourg revisits Mutual Trust
These two strands of decisions on mutual trust came together in last April’s Aranyosi and Căldăraru judgment. Responding to a reference from a German regional high court (the Oberlandesgericht, Bremen), the CJEU confronted the question whether Germany’s obligation of mutual trust under the EU Treaties could be overcome by a claim that the individual would suffer inhuman and degrading treatment in the prisons of the requesting state, in this case Hungary. Even though the case arose under the European Arrest Warrant, its substance tracks the big debate over mutual trust and the CJEU’s requirement of “systemic” deficiencies to defeat a return to the member state of first entry under the Dublin rules.
Aranyosi and Caldararu confirms that “mutual trust” is “not absolute.” But in what looks like a notable retreat from its previous insistence that only “systemic deficiencies” could defeat an EU obligation to transfer, the Court categorically declares: “The consequence of the execution of [a European Arrest Warrant] must not be that that individual suffers inhuman or degrading treatment.” (para. 88) Here we have now, in unequivocal terms, an absolute requirement to prevent individual harm.
The Court then walks the reader through the various steps of the inquiry. Read superficially, it seems to require an initial showing of “systemic deficiencies,” followed by a showing that this systemic deficiency will also produce individualized harm for the claimant. A closer reading, however, shows that the Court is now placing the question of “systemic” deficiencies in a distinctly secondary role:
“[T]he executing judicial authority must, initially, . . . rely on information . . . that demonstrates that there are deficiencies, which may be systemic or generalised, or which may affect certain groups of people, or which may affect certain places of detention.” (para. 89)
Use of the word “may” (instead of, say, “must”), and the juxtaposition of “systemic” deficiencies with many other types of qualifying deficiencies, suggests a loosening of the mutual trust obligation as compared to previous cases. Read properly, there must be “deficiencies” of some sort. What specific form these deficiencies “may” take, seems to be a much more open matter now.
Only after emphasizing once more the importance of protecting the individual, does the Court say that a finding of “general” conditions alone cannot lead to refusal to execute a warrant. It is necessary, so the Court, to show that the individual will be exposed to that general risk. So, generalized deficiencies alone are not enough. At the same time, “systemic” deficiencies (whatever that might have really meant) are no longer required.
If we compare the different language versions of the opinion, we see the Court still struggling with this change. The German version of that same passage says:
“Dabei muss sich die vollstreckende Justizbehörde zunächst auf . . . Angaben . . . stützen, die das Vorliegen systemischer oder allgemeiner, bestimmte Personengruppen oder bestimmte Haftanstalten betreffender Mängel belegen.”
Here, the emphasis is on the finding of “systemischer oder allgemeiner”, i.e. “systemic or general,” deficiencies (“Mängel”), not on the idea that simply “deficiencies” of some sort must be found. So we have some minor discrepancy between the German and the English. The French version, in turn, is closer to the English:
“[L]’autorité judiciaire d’exécution doit, tout d’abord, se fonder sur des elements . . . démontrant la réalité de défaillances soit systémiques ou généralisées, soit touchant certains groupes de personnes, soit encore certains centres de detention.”
Here, too, the emphasis is on “deficiencies” (“défaillances”) be they (“soit”) of one kind or another, plus expressly allowing deficiencies that pertain to a particular prison as a separate qualifying event.
The language of the case is German, and so the somewhat more restrictive formulation might be considered more authoritative (for now). But the cracks in the doctrine of mutual trust, opened by the unqualified requirement to ensure against individual deprivations of fundamental rights in other states, are remarkable. And for the CJEU to dispense with the requirement that deficiencies causing such rights deprivations be “systemic,” with the strong suggestion (at least in the English and French versions) that deficiencies can come from whatever source, is nothing short of an about-face in Luxembourg’s case law.
To some, these may seem like minute changes in legal language. In the context of the foundational principle of “mutual trust,” however, and for the lives of real litigants the change is of great significance. What is more, we see the Court stepping back from what some have criticized as an attempt to ignore social reality, and to impose an absolute doctrine of mutual trust by judicial fiat. Finally, in terms of high theory of the Union that continues to be debated in contributions here and elsewhere, this case suggests the CJEU is not just speaking, but also listening, in the bracing exchange among high courts of Europe.