S.S. and Others v. Italy: Killing by Omission, Confirmed by Design

On 12 June the European Court of Human Rights issued its judgment in S.S. and Others v. Italy (App. No. 21660/18). The Court declared inadmissible the claims of seventeen applicants intercepted at sea and subjected to grave harm through a Libyan-led operation underwritten by the Italian state. At the heart of the judgement is the finding that the Court lacked jurisdiction under Article 1 of the European Convention on Human Rights.

Alongside Violeta Moreno-Lax and Loredana Leo (ASGI), I was one of the lawyers who brought this case before the court back in 2017. Moreno-Lax brilliantly developed the conceptual foundation of the case. Both of us worked at the time, in our capacity members of the Global Legal Action Network (GLAN). Apart from the important partnership with ASGI, the Sea Watch activists and Forensic Oceanography (now Border Forensics) provided important support. And yet, I think it is fair to say that everyone involved understood that the case would be an uphill legal struggle. It was an attempt—a fragile one from the outset—to test whether the Court’s doctrine could rise to meet the evolving structure of harm in transnational migration governance.

As a human rights lawyer, one’s ethical responsibility is towards the applicant(s), and, on a secondary level, towards the human rights cause more broadly. And yet, personally, I thought of the case also as an opportunity to challenge, and possibly falsify, hypotheses I had developed in earlier work, most directly in a 2013 piece titled “Dialectic of Transnationalism”. Following the much-celebrated judgement of Hirsi Jamaa and others v. Italy (2012), I argued a dozen years ago that the seemingly progressive jurisprudence from the Court would serve, in reality, as blueprint for future harms. The prediction, in other words, was that Strasbourg jurisprudence would function not as a mechanism for individual protection but as one foundation of a juridical architecture that enables state violence at sea.

I remember very well the conversation with a friend and long-time collaborator challenging me: why such grim an outlook based on theory? Why not bring a case and test whether the Court can change its jurisprudence and live up to its promise of defending human rights?  In S.S., the opportunity arose to test that thesis, in conditions that come as close as we ever we do — in legal scholarship and practice — to an experimental context. Within this “lab”, the hypothesis regrettably held.

Blueprint, Reaffirmed

In November 2017, a group of migrants launched from Libya in a rubber dinghy. Upon issuing a distress call, the Italian Maritime Rescue Coordination Centre received and relayed it to the Libyan Joint Rescue Coordination Centre, treating Libya as the relevant Search and Rescue (SAR) authority. The Libyan coastguard vessel Ras Jadir, which would not have existed save for Italy establishing, training, and financing the Libyan Coast Guard, responded. Tens of people drowned (the Court says, “at least 20”, but the exact number is disputed). Two applicants lost children. Many of those intercepted were forcibly returned to Libya, where they faced arbitrary detention and systematic abuse.

This chain of events is not accidental. It reflects the operational logic of extraterritorialised border enforcement: displace formal responsibility, delegate coercive capacity, and structure rescue in a manner that forecloses access to asylum. As I argued in the 2013 article, such arrangements are produced by a dialectical effect that the court helped trigger: the extra-territorialized executive acts, and extra-territorial human rights jurisdiction follows. The “trick”, however, is that the latter falls, structurally, ever-so-slightly behind. As Thomas Gammeltoft-Hansen and James Hathaway memorably put it, a “cat-and-mouse game” follows, “in which border control must be constantly reinvented to respond to the schemes hatched by imaginative smugglers motivated by extraordinary profits.” Of course, the game persists between governmental enforcement agencies on the one hand, and protection capacities on the other. This image is appropriate because it highlights that ultimately, the former ends up eating the latter. Courts, at best, draw the outlines of the space of a maritime legal black hole, in which executive can perform border violence with impunity.

The Court’s response to the November 2017 tragedy is, of course, predictably too late to matter for the victims. But it is also legally coherent within the terms of border externalization. The applicants in S.S. v Italy were not intercepted directly by Italian agents. They were not brought aboard Italian vessels (as was the case in Hirsi). The Italian government argued that the operation took place within Libya’s SAR zone, an issue we disputed, but which the court found non-decisive. Italy’s material support to the Libyan coastguard — acknowledged as extensive — did not suffice to establish “effective control” or “authority” over the individuals concerned. On this basis, the case was deemed inadmissible ratione personae.

A Lethal Formalism

The Court applied primarily its familiar case law on jurisdiction — cases like Banković, Al-Skeini, Medvedyev, in a way that confirms the predictive arc. Most important for S.S.’s inadmissibility, however, was of course Hirsi. With a clear message that the state’s influence may surpass its Article 1 jurisdiction (and accountability), the real function of the Court’s jurisdictional doctrine in migration cases is not to constrain harm. It is to distribute it. The Court is thus helpful in allocating violent enforcement functions across actors, zones, and legal regimes, while keeping a protective apparatus formally intact. The main idea, as copious “externalization” scholarship now shows, is to create a veneer of human rights protection, while continuing to outsource violations.

This structural reading finds further support in the Court’s treatment of substantive obligations under Articles 2 and 3 ECHR. The applicants alleged a duty to prevent foreseeable death at sea, and to avoid refoulement or otherwise return into inhuman conditions. These claims were substantiated by overwhelming evidence, including findings from UN bodies. Yet the Court declined to examine them in any level or granularity. Without jurisdiction, it reasoned, it lacked competence. In short: the risk of drowning, torture, and the risk of refoulement, remain invisible if they are effectively delegated.

The Role of the Court

One could argue that the Court is constrained by its interpretive traditions. That it cannot step outside of its own structure. That it must defer to principles of sovereignty and territoriality. Above all, one could argue that it is naive to expect the court to expand its jurisprudence, in a time when it is under political pressures to protract (such as has been clear most recently form a letter by nine governments, led by Italy and Denmark).

I understand all these considerations. But it is equally important to insist that these are not neutral postures. They are normative choices that end up not only reflecting but also entrenching a specific scheme of border violence. Surely, in this instance, the choice was to reaffirm a jurisprudential architecture that legitimises harm offshore. From a consequentialist point of view, today it is no longer clear that the court provides any meaningful defence to migrants, beyond the “core” of refugee law. It is not enough to observe that the Court treats foreseeable violence—drowning, beatings, arbitrary detention—as beyond the reach of law, if it occurs beyond jurisdiction. That is of course true, but only because the Court itself draws the limits on jurisdiction. So, the objection is of course begging the question.

The applicants in S.S. also raised arguments under general international law. Article 16 of the ILC Articles on State Responsibility prohibits a state from aiding or assisting another state in committing internationally wrongful acts, when done with knowledge of the circumstances. The Court acknowledged the legal basis but declined to engage, citing the absence of Article 1 jurisdiction. It refused, in effect, to connect human rights responsibility with the design of foreseeable harm.

Conclusion

That refusal leaves open a question: if Strasbourg cannot account for these patterns of structural violence, where will accountability be found? Some have offered the human rights committees. Others have advocated for a greater role for Courts in the global south. Yet a third option — one that signals an exist from legal process — has been the formation of people’s courts. But frankly, each of the options has its own problems. In such conditions, it is inevitable to shift attention from the strategic or instrumentalist question of how to achieve certain predetermined ends, to the more fundamental question of what the ends should be. I hope to address that question elsewhere. Suffice it to say here, that not all strategic litigation is created equal.

Be that as it may, in S.S. the Court has revealed what we have to some extent already known, of course since ND and NT v. Spain, that we should not turn to it as a defender of migrant rights at this time. As far as externalization goes, it has confirmed the doctrine, confirmed the delegation, and confirmed the exclusion from rights. The blueprint, already drawn, has now been judicially underlined.

S.S. v. Italy offered a rare chance to test a critical hypothesis: that the Court’s jurisprudence, far from curbing violence, helps script it. The result is not just confirmation, but demonstration. What appeared as legal process was, in truth, the controlled conditions of a failed experiment — one that reveals the law’s complicity, not its constraint.