I. The Legal and Political Framework

The 2008 Treaty of Friendship, as developed in the Memorandum of Understanding (MoU) of February 2017,Footnote ⁴⁶ is the pivotal agreement, providing legal coverage to the Italian-Libyan cooperation in the field of irregular migration. It specifically buttresses the re-establishment of a Libyan Navy and Coast Guard (LN/LCG), with Italy assuming “a leading role.”Footnote ⁴⁷

The Treaty contains a provision, in Article 19, calling on both parties to intensify their collaboration in the establishment of an integrated system of frontier surveillance in Libya, for the Italian actors with the requisite technological competence to administer, committing Italy to pay half of the cost, with the EU bearing the other half.Footnote ⁴⁸ The provision also explicitly commits the parties to jointly define actions to “stem irregular migration flows”Footnote ⁴⁹—with no mention of human rights obligations. While the implementation of the Treaty led to the joint push-back campaign conducted in 2009, for which Italy was condemned in Hirsi,Footnote ⁵⁰ cooperation was halted during the civil war period.

The 2017 MoU has revived the Treaty of Friendship by expanding on its Article 19.Footnote ⁵¹ It sets up, on that basis, specific structures of collaboration, including a “Joint [Italy-Libya] Commission” charged with the definition of priorities, funding needs, implementation strategies, and monitoring actions.Footnote ⁵² The ultimate goal remains to “stem irregular migrant flows”Footnote ⁵³—again, with no reference to human rights. To that end, the division of labor foresees that Italy provide the financial, technical, technological and other means, specifically to the LYCG.Footnote ⁵⁴ The financing of detention centers, the training of its personnel, and overall support to return and readmission from Libya is also part of the agreement.Footnote ⁵⁵And Article 4 reiterates that it is for Italy, including via EU funding, to cover the expense.Footnote ⁵⁶

Regarding political support, Italy has not been alone in sustaining the LYCG and the plan for comprehensive containment of unwanted flows departing from Libya. The EU, besides providing significant financial and logistic assistance, has also celebrated the Italian-Libyan cooperation at the highest political level. Already in January 2017, the EU Commission and the EU High Representative for Foreign Affairs called for the enhancement of support to Libya and the LYCG.Footnote ⁵⁷ And, far from condemning the MoU, the Malta Declaration, adopted by all EU Heads of State and Government, “welcomes and … support[s] Italy in its implementation,” pledging funds and capacity building, with the explicit aim of “preventing departures and managing returns.”Footnote ⁵⁸ Despite the wealth of sources denouncing it, the situation facing migrants in Libya—known to former Italian Minister of Interior, MinnitiFootnote ⁵⁹ and his fellow ministers of the other Member StatesFootnote ⁶⁰—has been no impediment to the EU’s backing of this cooperation.

II. Funding and Equipment

Capacity-building initiatives within the framework of the Treaty of Friendship and the MoU intensified in the summer of 2017, with Italy creating a dedicated “Africa Fund” and allocating €2.5 million for the maintenance of Libyan boats and the training of their crews.Footnote ⁶¹ In parallel, Italy also secured EU funding in excess of €160 million for Libya. An EU project was awarded to the Italian Coast Guard, through which €46.3 million have been channeled to border management and migration control in Libya.Footnote ⁶² The project specifically aims at “[s]trengthening the operational capacities of the Libyan coastguards,” via “training, equipment … repair and maintenance of the existing fleet,” so as to “strengthen the authorities’ capacities in maritime surveillance and rescuing at sea.”Footnote ⁶³ The final goal is “to provide the Libyan coast guards with initial capacity [absent hitherto] to better organise their control operations” and “coordinate maritime interventions.”Footnote ⁶⁴ This, the EU Commission has noted, “will involve the full design of an Interagency National Coordination Centre … and a Maritime Rescue Coordination Centre,”Footnote ⁶⁵ which does not yet exist—its completion being “estimated in 2020”Footnote ⁶⁶—as well as “assistance to the authorities in defining and declaring a Libyan Search and Rescue Region [SRR]”Footnote ⁶⁷—which was only recorded by the International Maritime Organization (IMO) in June 2018.Footnote ⁶⁸

In terms of equipment, Italy has donated ten fast patrol boats to the LN/LCG,Footnote ⁶⁹ which seem to be “the most effective and reliable ships [in the LYCG inventory].” The best appears to be precisely the Ras Al Jadar, which performed “approximately half of all sorties” between October 2017 and January 2018,Footnote ⁷⁰ including the one of November 6, 2017. The vessels were gifted disregarding the widely publicized malpractices of the LYCG—also witnessed in the S.S. events—and the series of violent incidents occurred just a few days before the ceremony of award.Footnote ⁷¹ In one such incident the LYCG had interrupted a rescue, intercepted migrants at gunpoint, and pulled them back to Libya using perilous tactics.Footnote ⁷² Several actors, including the UN Secretary-General, have denounced similarly violent behavior by the LYCGFootnote ⁷³—of which the Italian Coastguard was awareFootnote ⁷⁴—including the firing of live shots,Footnote ⁷⁵ the intimidation of NGO rescue boats,Footnote ⁷⁶ and the use of force against migrants.Footnote ⁷⁷

III. Operational Involvement

For many years, and especially since the Arab Spring, “the only country that provide[d] SAR to the area sitting next to the territorial waters of Libya [was] Italy.”^{Footnote 78} After the termination of the Mare Nostrum operation in 2014, the Italian Government carried on “coordinat[ing] virtually all rescue operations” in that area^{Footnote 79}—a fact corroborated by EUNAVFOR MED, confirming that the “Italian MRCC … continued to coordinate rescue operations” throughout 2016 and 2017.^{Footnote 80} In fact, LYCG coordination capabilities peaked at a mere “54% of [all] SOLAS events” only in the second semester of 2018^{Footnote 81}—long after the S.S. events.

This situation of de facto Italian-led Libyan interventions was consolidated in 2017, on the basis of the MoU. Within that framework, not only the establishment of a capable coast guard, but also of a reliable Libyan MRCC became top priorities. The aforementioned EU project awarded to the Italian Coast Guard supported implementation.^{Footnote 82} Completion was planned in consecutive phases, including activities such as “organiz[ing] [LYCG] SAR units” and “develop[ing] SAR SOPs.”^{Footnote 83} But the actual creation of the Libyan MRCC only began in December 2018, when the project entered its second phase, with the “development of the MRCC Communication network along the coast.”^{Footnote 84}

Meanwhile, an incipient LYCG—still “far from being fully operational” by EUNAVFOR MED’s own admission^{Footnote 85}—started operating with the support of a Joint Operation Room (JOR), consisting of some “basic operational rooms in a joint building in Tripoli” set up in the first phase of the project,^{Footnote 86} but “with limited [space] and communication capabilities [and] relatively equipped to communicate with naval assets at sea.”^{Footnote 87} The JOR, involved in the events of November 6, 2017, was and still remains “in a critical infrastructural situation … [that] is further adversely conditioned by a limited presence of personnel with insufficient language (English) skills and limited software tools … knowledge.”^{Footnote 88} In fact, the JOR is incapable of operating at a “self-sustaining level,”^{Footnote 89} and its capacities “do[] not allow properly carrying out the institutional tasks as MRCC,”^{Footnote 90} so that, as per the EUNAVFOR MED’s assessment, they “still need further sustainment … also in operational terms.”^{Footnote 91}

Especially, the “lack of effective and reliable communication systems hampers Libyan capacity for the minimum level of execution of command and control [C2], including that necessary to coordinate SAR/SOLAS events,”^{Footnote 92} hence Italy has secured the necessary functions. To this effect, in August 2017, it launched Operation Nauras, an extension into Libyan territorial and internal waters of the military mission Mare Sicuro,^{Footnote 93} including “a factory vessel” sent to Tripoli with the task “to restore the efficiency of other Libyan naval units, and coordinate patrol and sea rescue operations.”^{Footnote 94}

Operation Nauras consists of four ships, four helicopters, and 600 servicemen, of which 70 per cent are deployed at sea, with the remaining 30 per cent staying in Tripoli harbour. Their key mission is, specifically, to “establish [the] operational condition[s] for LN/LNCG assets and develop C2 capabilities.”^{Footnote 95} In the interim, their “naval asset in Tripoli Harbour [is] acting as LNCC [i.e., Libyan Navy Communication Centre] and logistic assistance/support hub.”^{Footnote 96} This vessel is permanently “in contact with SAR assets and ITCG [i.e., Italian Coast Guard] and MRCC Centres,”^{Footnote 97} thus playing the role of a floating MRCC for Libya. Its function—also at the time of the S.S. events^{Footnote 98}—was explicitly “the cooperation and coordination of the joint activities of the Libyan Coast Guard and Navy, with a view to carrying out their Command and Control (C2) tasks and maintaining an adequate Maritime Situational Awareness to fight illegal migration.”^{Footnote 99}

It is, therefore, via the Italian authorities, within the MRCC Rome and aboard the Nauras warship in Tripoli, that the LYCG received distress calls. And, because, on receipt, it lacked the means to further communicate with, let alone coordinate, assets at sea, the LYCG systematically relied on Italian (and EUNAVFOR MED^{Footnote 100}) infrastructure to liaise with the relevant actors. A case of 2019, documented by the SAR NGO Mediterranea, discloses how, oftentimes, communication is even entirely done by Italian officials supposedly “on behalf of” their absent LYCG counterparts, creating the impression of autonomous Libyan action.^{Footnote 101} A usual mode of engagement—confirmed by the EU Commission—involves the early detection via “sightings” performed by Italian or EUNAVFOR MED aerial assets, transmission of the information to the LYCG through the Nauras warship in Tripoli acting “as a ‘communication relay,’”^{Footnote 102} and then further action coordinated by Italy “on behalf of” the LYCG.^{Footnote 103}

This pattern consolidated through sustained practice since August 2017,^{Footnote 104} and has been reinforced with Italy and the EUNAVFOR MED introducing a post-operation evaluation of the LYCG’s conduct, precisely as a consequence of the November 6, 2017, incident. The lack of “professional behaviour” of LYCG personnel was raised through this channel on this occasion and a “basic ‘lessons learnt’ process” introduced, with disciplinary measures taken “in one specific case.”^{Footnote 105} Apparently, the monitoring system in place entails an “advising role in order to strengthen accountability and follow up,”^{Footnote 106} including “feedback and recommendations” to which the LYCG has been “receptive” so far.^{Footnote 107}

Accordingly, what the next sections will substantiate is that, from the launch of Nauras, it has been Italy, both remotely through its MRCC and via direct military presence in Libya, which has assumed the overall coordination of the LYCG operational response in the Central Mediterranean in a way that amounts to an exercise of extraterritorial jurisdiction. Italy’s pervasive political, financial, and operative involvement equates “effective control.”

I. Jurisdiction as Sovereign-authority Nexus

The definition of the concept and its specific role in international human rights law has long attracted doctrinal attention. But there is disagreement as to its utility and its centrality for the establishment of responsibility for human rights violations. Some authors, like Scheinin, argue that “jurisdiction” does not add anything to the key aspects of admissibility within the state responsibility framework and should, therefore, be considered an empty notion for the purposes of substantiating legal accountability. For him, there is apparently no distinction between the attribution of wrongful conduct to the state concerned and the determination of an exercise of its jurisdiction. The two are one and the same. Adding an extra step that functions as a threshold and precludes the establishment of responsibility is, therefore, seen as unhelpful.^{Footnote 108} Another strand of the literature questions the appropriateness of attempting a general synthesis of the concept, in light of the variety of human rights duties and their different manifestations, which would require a more tailored and nuanced approach. Only so can the complexities of (especially positive “facilitation” and “fulfillment”) obligations, entailed in particular by economic, social and cultural rights, be adequately reflected.^{Footnote 109}

By contrast, other writers, such as Besson, consider jurisdiction to be fundamental to the proper understanding of the relationship that unites human rights holders and duty bearers.^{Footnote 110} For her, without jurisdiction, the universality of human rights would imply that any state would owe human rights duties to any human rights holder, regardless of any specific political-legal nexus between them. This is why jurisdiction, in this relational sense,^{Footnote 111} has an essential role to play in arbitrating between duty, capability, and desirability of compliance by any specific state vis-à-vis any specific human rights holder. And this is also why jurisdiction should be understood as an “all-or-nothing” condition for the activation of human rights obligations, rather than as gradual or incremental.^{Footnote 112} Either there is a jurisdictional link between the state and the person concerned or there isn’t. What may, then, be “divided and tailored” in the specific case, and be proportionate to the level of control applied, are the ensuing obligations, but not jurisdiction per se.^{Footnote 113}

From this perspective, the term should best be understood as the “de facto political and legal authority” of the sovereign, amounting to more than mere coercion,^{Footnote 114} including a normative dimension that demands compliance. It is not “facticity [that] creates normativity.”^{Footnote 115} Normativity must precede and underpin the account of a factual basis qua jurisdiction. It is the normative aspect of an exercise of state power that makes its interaction with a particular individual human-rights relevant. In Besson’s view—which I espouse—jurisdiction refers to “some kind of normative power” that the sovereign exercises vis-à-vis an individual “with a claim to legitimacy,” and that serves to establish the human-rights relevant link between them. Whether the state concerned may have acted ultra vires in the specific situation constitutes a separate question. A priori, to be an expression of jurisdiction, state actions/omissions do not have to be lawful, but only stem from a “lawfully organized institutional and constitutional order.”^{Footnote 116} What matters to characterize state conduct as jurisdiction in the human rights sense is the underlying sovereign-authority nexus that connects the state to those within its might and the control it thereby purports to exercise, whether de jure or de facto, rather than the legality of its conduct. In this sense—which seems to be the one tacitly embraced by the Strasbourg Court—jurisdiction works as a trigger of human rights obligations.^{Footnote 117}

Without a (pre-existing) jurisdictional link between a State party and a certain individual, no human rights duties can be owed in specific circumstances. Potential or hypothetical connections are hence irrelevant. Also claimed connections, which are not effectuated in the real world, are immaterial.^{Footnote 118} Jurisdiction requires an “external manifestation of the power of the State”^{Footnote 119}—whether having a legal or factual dimension, or being constituted by a combination of both. So, for instance, simply having the capacity to counter famine in a remote land to which there is no prior public-power relation does not suffice to entail responsibility. Unless there is an underpinning basis of prescriptive, executive, and/or adjudicative authority—with or without legal title—through which actual state activity has taken place, the jurisdictional link will not be established. If, on the contrary, there is a piece of legislation enacted, a policy plan implemented, and/or a court decision enforcing the legislation or the policy plan in relation to said famine in said remote land, there should be no obstacle to consider such action as one demonstrative of state jurisdiction. Once the sovereign authority-nexus has been ascertained, there seems to be no principled reason justifying a distinction on the basis of the locus of such activity in deeming it a manifestation of jurisdiction, whether territorially or extraterritorially exercised. It would be “unconscionable” to create a double standard on that ground alone and, in consequence, “permit a State … to perpetrate violations … on the territory of another State, which violations it could not perpetrate on its own territory.”^{Footnote 120}

To my mind, the role that territoriality plays within this understanding of the concept—in line with the basic tenets of public international lawFootnote ¹²¹—is to generate a (rebuttable) presumption of the existence of such a link within the national domain, applying “throughout the State’s territory.”Footnote ¹²² What distinguishes extraterritorial settings is the absence of such a presumption, given the principles of territorial integrity and non-interference in domestic affairs. But that does not alter the fundamental premise on which the concept of jurisdiction rests. As soon as a concrete public-power relation has been established, a jurisdictional connection is activated, triggering the application of human rights obligations. This, however, does not mean that all human rights will be owed in all situations. For instance, a military surveillance mission over non-national territory will be irrelevant to the right to education of those concerned, but it may engage responsibility from the perspective of the right to privacy, if it entails the collection of personal data.^{Footnote 123}

This approach, therefore, unifies the premise underpinning all forms of jurisdiction qua normative power with a claim to legitimacy by a state that, if and when acted upon, establishes a sovereign-authority link with those concerned. It also “normalizes” the possibility of extraterritorial manifestation—just like the Strasbourg organs did before Bankovic.Footnote ¹²⁴ Indeed, the now-disappeared European Commission on Human Rights consistently held that the “High Contracting Parties are bound to secure the … rights and freedoms [in the Convention] to all persons under their actual authority and responsibility, not only when the authority is exercised within their own territory, but also when it is exercised abroad.”^{Footnote 125} The Convention was supposed to govern the actions and omissions of Contracting Parties wherever they exercised jurisdiction. And jurisdiction, under Article 1 ECHR, was not deemed “equivalent … to or limited to the national territory of the High Contracting Party concerned.” This was “clear from the language … and the object of this Article, and from the purpose of the Convention as a whole ….”^{Footnote 126} It has been in Bankovic that the Court “exceptionalized” extraterritorial jurisdiction and conceptually decoupled it from its territorial counterpart.

II. The “Exceptionalization” of Extraterritorial Jurisdiction

In Bankovic the Court likened the term “jurisdiction” to the concept of legal title under international law, thus affirming that “the jurisdictional competence of a State is primarily territorial.”^{Footnote 127} In fact, “a State may not actually exercise jurisdiction on the territory of another without the latter’s consent, invitation or acquiescence, unless the former is an occupying State ….”^{Footnote 128} There must, otherwise, be a legal basis allowing the state to exercise its power extraterritorially, whether “nationality, flag, diplomatic and consular relations, effect, protection, passive personality [or] universality.”^{Footnote 129} This understanding, however, conflates jurisdiction under Article 1 ECHR with the existence of a right or prerogative of the state to act, which a contrario leads to the absurdity that states operating unlawfully abroad, without legal title conferred by international law, can additionally be human rights exempt.

Even in Bankovic did the Court avoid this conclusion and decided, instead, that the implication of “the ‘ordinary meaning’ of the relevant term in Article 1 of the Convention” was that jurisdiction should be understood as “primarily territorial,”^{Footnote 130} other bases “being exceptional and requiring special justification in the particular circumstances of each case.”^{Footnote 131} While it delivered other controversial findings regarding the effect of the so-called “colonial clause” in Article 56 ECHR and the “espace juridique européen,”^{Footnote 132} these have been subsequently overturned in Al-Skeini.^{Footnote 133}

What Al-Skeini has retained is the notion that extraterritorial jurisdiction is exceptional and, as such, must be demonstrated in the specific instance^{Footnote 134}—an assertion I only partly share: While I accept that jurisdiction should be “presumed to be exercised normally throughout the State’s territory,”^{Footnote 135} over which the state is sovereign, that alone does not render extraterritorial jurisdiction exceptional in the material sense, it only requires that proof of an actual sovereign-authority link be produced in the individual situation. The presumption allocates the burden of that proof, but should have no bearing on the substantive finding of whether jurisdiction has indeed been exercised. It is also unclear what “exceptional” refers to in the eyes of the Court: Does it concern frequency or justifiability? The elimination of the presumption does not make the occurrence of extraterritorial exercises of jurisdiction any less frequent, or any less legitimate, per se. Questions on the lawfulness of jurisdictional action are separate from whether such jurisdictional action obtains in a particular case.

In any event, this “exceptionalization” has led to a narrow understanding of the material circumstances that can count as an exercise of extraterritorial jurisdiction. Only two models have been accepted: The “State agent authority” or personal model and the “control over an area” or territorial model.^{Footnote 136} In both cases the accent is put on the factual dimension of jurisdiction, understood as equivalent to “effective control,” but without defining the term or clarifying what “effective” means in this framework.

The territorial model refers to situations in which jurisdiction arises as a consequence of state military action outside national territory, whether lawfully or unlawfully engaged.^{Footnote 137} The obligation to secure Convention rights derives from “the fact of such control,” whether exercised directly, by the state’s own army, or through a subordinate local administration.^{Footnote 138} In the latter case, if the existence of “overall control” can be established, then it becomes unnecessary to demonstrate that the state exercises detailed control over each and every of the policies and actions of the subordinate local administration.^{Footnote 139} And, again, determining whether effective control exists in such a situation is deemed a “question of fact,” which, according to the Court, must be resolved by reference to the strength of the military deployment in the area or the degree to which military, economic, and political support to the local administration is “decisive” to influence its behavior.^{Footnote 140}

“Overall control” is considered to involve a measure of constant dominium over the foreign area at hand, to a point comparable to state sovereignty. In this sense, “overall control” is the de facto counterpart of the de jure title entailed by state sovereignty, thus justifying the (re-)emergence of the presumption of jurisdictional authority throughout the area concerned and its transposition to the extraterritorial context. “Overall control” liability becomes equivalent to that of the de jure sovereign. Therefore, within the area under its overall control, the controlling state has the responsibility to secure “the entire range of substantive rights set out in the Convention.”^{Footnote 141} Otherwise, discrete forms of geographical control give rise to a duty to ensure only the rights that are relevant in the circumstances.^{Footnote 142}

This is also what happens under the personal model, where effective control over an individual also entails a duty to secure only the relevant protections—presumably on consideration that, unlike in situations of overall territorial control, there has not been a replacement of the territorial sovereign. Under this model, the Court operates under the general rule that jurisdiction may extend to acts of state authorities “which produce effects outside its own territory”^{Footnote 143} and distinguishes three cases.

First, the acts of diplomatic and consular agents, “present on foreign territory in accordance with provisions of international law,” may count as an exercise of jurisdiction whenever they “exert authority and control over others.”^{Footnote 144} Second, state acts that amount to an exercise of “public powers normally to be exercised by [a national] Government” may also reach the threshold, if underpinned by “the consent, invitation or acquiescence” of the territorial sovereign. If such is the case, responsibility may be incurred by the ECHR party “as long as the acts in question are attributable to it rather than to the territorial State.”^{Footnote 145}

These first two categories thus appear to attach importance to elements of de jure jurisdiction, but the Court has failed to provide a detailed elaboration. In Hirsi, it did suggest that legal bases under customary international law, and in particular “the relevant provisions of the law of the sea,” are significant, so that “acts carried out on board vessels flying a State’s flag” shall be considered “cases of extraterritorial exercise of … jurisdiction.”^{Footnote 146} But it did not dwell on whether on that ground alone—without additional elements of de facto control—Article 1 ECHR could have been engaged.^{Footnote 147}

The Court’s attention has rather focused on the third tier of the personal model, concerning the use of force, under which it has concluded that what tends to be “decisive” in this context is “the exercise of physical power” over persons abroad.^{Footnote 148} The circumstances that have been considered to reach the threshold, and that the Court invokes to illustrate its findings, are instances of arrest, detention, abduction, and extradition,^{Footnote 149} thus highlighting forms of de facto control. And the same is true on the high seas, where in most cases the Court has ascertained the existence of jurisdiction on account of the “full and exclusive control” exercised “in a continuous and uninterrupted manner” over a foreign vessel or persons apprehended aboard.^{Footnote 150} This was the test applied in Hirsi, in the context of the push-back operation of migrants to Libya carried out by Italy, where the Court concluded that, “in the period between boarding the ships of the Italian armed forces” after rescue “and being handed over to the Libyan authorities,” the applicants had been subjected to “the continuous and exclusive de jure and de facto control of the Italian authorities.”^{Footnote 151}

However, the Court has also made clear that direct physical contact is not always necessary as long as the control thereby exerted is indeed effective. So, in a case involving the maritime blockade of a Dutch vessel by the Portuguese authorities impeding access to Portugal’s territorial waters, the jurisdictional link was not contested.^{Footnote 152} In parallel, the rerouting of a foreign ship in Medvedyev, imposing a specific course, but without boarding it, was also deemed to meet the jurisdictional test. Jurisdiction was exercised “from the stopping” of the boat, throughout the period of enforced navigation.^{Footnote 153} This, as the next part elaborates, opens up a range of possible configurations in which instances of “contactless control” may be seen as an expression of jurisdiction—particularly when exercised against a background of existing legal competence in the relevant domain, lending a de jure basis for action.^{Footnote 154}

I. The Impact Element

Very much in the line of the Norstar case,^{Footnote 182} the impact element refers to the “sufficiently proximate repercussions” of state action “on rights guaranteed by the Convention” that the Strasbourg Court has deemed pertinent to the establishment of jurisdiction, “even if those repercussions occur outside” national territory.^{Footnote 183} What is of relevance is their origin in an exercise of public powers by the authorities of the state concerned. Sovereign activity—arguably of whatever nature: legislative, executive, or judicial^{Footnote 184}—with direct and predictable consequences beyond territorial boundaries can thus engage Article 1 ECHR. So, for instance, in Andreou, the opening of fire from within state territory on a crowd from close range was deemed to amount to jurisdiction, “even though the applicant sustained her injuries in territory over which Turkey exercised no control,” since the shooting by state officials was “the direct and immediate cause of those injuries.”^{Footnote 185}

The Inter-American Commission, in a very similar case, concluded the same. In Brothers to the Rescue, Cuba was considered to have exerted sufficient control through the shooting down of two aircrafts outside its aerial space, because “the victims died as a consequence of direct actions of agents of the Cuban State” operating within Cuban territory.^{Footnote 186} The Inter-American Court has followed suit and declared that “a person is under the jurisdiction of the State … if there is a causal link between the action that occurred within its territory and the negative impact on the human rights of persons outside its territory.”^{Footnote 187} So, the mere fact that the impacted individuals are situated outside national territory does not preclude the engagement of extraterritorial responsibilities. The jurisdictional link is established through the effects of state conduct that is initiated within territorial domain.

However, the significance of the presence of the state authorities exercising jurisdiction within national territory has, subsequently, been downplayed. The Human Rights Committee has inferred that the extraterritorial “impact,” which is the “direct and reasonably foreseeable” result of state action, is relevant also vis-à-vis “individuals who find themselves in a situation of distress at sea.”^{Footnote 188} Actually, the Committee had already previously held that a State party could be considered responsible for extraterritorial violations of the ICCPR,Footnote ¹⁸⁹ where there was a “link in the causal chain” that would make possible violations on the territory of another state—wherever the location of state organs.^{Footnote 190} In such situations, the risk of an extraterritorial violation must be a “necessary and foreseeable consequence,” judged on the knowledge the state had at the time of events.^{Footnote 191} So, knowledge of the probable result becomes a factor in the jurisdictional analysis, whereas the locus of the action is immaterial. In Munaf, for instance, the Committee evaluated the conduct of diplomatic staff in the Romanian Embassy in Baghdad applying this paradigm, and implying that only remote and unforeseeable consequences fail the jurisdictional test.^{Footnote 192}

The Strasbourg Court has also endorsed this understanding. In Loizidou, it declared that “the responsibility of Contracting Parties can be involved because of acts of their authorities, whether performed within or outside national boundaries, which produce effects outside their own territory.”Footnote ¹⁹³ And more recently, in Al-Saadoon, it applied the so-called Soering reasoning to an extraterritorial extradition by UK agents of a terrorist suspect in Iraq.Footnote ¹⁹⁴ Therefore, while pure causation is insufficient to establish jurisdiction in relation to utterly accidental and unpredictable outcomes,Footnote ¹⁹⁵ the proximate and predictable results must be taken into account when planning and executing state action, whatever the location of its agents and of the action itself.

In the S.S. case, the coordination of the rescue/interdiction operation was undertaken by MRCC Rome through a combination of prescriptive and executive action—with knowledge of the likely outcome. The Italian Coast Guard acted territorially, within its Headquarters, taking the decisions of launching the SAR response and delivering instructions to all assets in the SAR theater on the high seas. This alone, amounting to the “institution of … proceedings” extraterritorially by the authorities of an ECHR party, has, in comparable cases, been considered to be “sufficient to establish a jurisdictional link” by the Strasbourg Court.^{Footnote 196} Here, such action “produced effects outside its own territory” with very significant consequences for those concerned,^{Footnote 197} which Italy could and should have taken into account when planning and deploying its intervention. The fact that Italy’s conduct “facilitated the whole process” that led to the involvement of the LYCG and “created the conditions” for the several violations complained of to materialize,^{Footnote 198} is a further indication of the existence of jurisdiction under Article 1 ECHR.^{Footnote 199}

This factual dimension of the jurisdictional constellation present in the S.S. case is complemented by a de jure basis in international law. Indeed, the coordinating role assumed by MRCC Rome could not have been ignored or avoided. It was legally predetermined by the maritime conventions, which, rather than creating any new sovereign entitlements in favor of coastal states, instead produce “area[s] of responsibility” to be overseen (in good faith) in order to preserve the safety of human life at sea.Footnote ²⁰⁰ These conventions stipulate that upon receipt of a distress call, the first MRCC contacted becomes and remains responsible for the coordination of rescue procedures until the MRCC in charge of the SAR region (SRR) within which the incident occurs assumes responsibility.^{Footnote 201} Like Papastavridis argues in this Special Issue, it is the knowledge of the situation of distress that triggers the obligation under the law of the sea, in line with the object and purpose of the maritime conventions. Their objective is to ensure cooperation in completing the rescue and disembarking survivorsFootnote ²⁰²—a duty that would normally fall on to the MRCC in whose SRR the incident takes place.Footnote ²⁰³

However, in the absence of an officially declared SRR and a fully functioning Libyan MRCC, that responsibility could not be validly transferred to the LYCG, and the first MRCC receiving the distress call—and thus with knowledge of the event—remained bound to proceed with the effective coordination of the operation. This responsibility includes making sure that the rescue is conducted safely and in compliance with the relevant rules, bringing survivors to landfall in a place of safetyFootnote ²⁰⁴—which Libya is not.Footnote ²⁰⁵

Any information, instructions, and guidance delivered by MRCC Rome must take into account their likely repercussions—bearing in mind that reliance on law of the sea norms does not release from parallel human rights obligations concurrently applying in situations of distress.Footnote ²⁰⁶ In particular, an MRCC that coordinates a SAR operation outside its own SRR “should refrain from giving directions or advice which it knows or ought reasonably to know would have negative human rights implications for those requiring assistance.”^{Footnote 207} This arguably includes the requisitioning of vessels from actors, like the LYCG, which are known for their unsafe, threatening, and abusive conduct towards survivors, invariably leading to their refoulement.^{Footnote 208} While “the search and rescue service concerned … has the right to requisition ships [so that they] render assistance,”Footnote ²⁰⁹ it has also the duty to exercise this power in line with “other rules of international law.”^{Footnote 210} Arguably, this includes the prerogative to release masters of ships that could potentially be requisitioned from their obligation to render assistance, when they are unsuitable.^{Footnote 211} A shipmaster should only be asked to proceed to the rescue “in so far as such action may be reasonably be expected of him.”^{Footnote 212}

The Italian authorities knew or ought to have known that the LYCG was inadequate. They knew or ought to have known that calling upon it to intervene would mean for the survivors to be taken back to Libya,Footnote ²¹³ to face “dismal circumstances” amounting to “crimes against humanity,” as described in EUNAVFOR MED documentation.^{Footnote 214} And this foreseeability of the likely result of their actions was relevant to the establishment of a jurisdictional link with the S.S. applicants.

Acting in the knowledge that the life and integrity of the persons in distress will be threatened when delivered to the authorities of an unsafe country^{Footnote 215} amounts to an exercise of jurisdiction under the impact model, which thus suffices to activate the positive, due diligence obligations attaching to the rights of the persons directly affected by the action concerned.^{Footnote 216} In the S.S. case, SAR duties intersect with human rights responsibilities, which constrain state discretion and limit the options left for choice of action.Footnote ²¹⁷ Italy could, therefore, not legitimately indicate a transfer of responsibility for the survivors to the LYCG, whether directly or indirectly, including through the provisions regulating OSC, without thereby engaging its (functional) jurisdiction and violating its international obligations.^{Footnote 218} MRCC Rome should, instead, have avoided the intervention of the LYCG, by not calling on the Ras Al Jadar, as a measure “within the scope of [its] powers which, judged reasonably, might have been expected to avoid [the] risk.”^{Footnote 219} Alternatively, at the very least, it should have refrained from asking it to assume OSC, a task that MRCCs must allocate “taking into account the apparent capabilities of the on-scene co-ordinator and operational requirements.”^{Footnote 220} Rather, it should have preferred the better alternatives offered by the SW3 and the multiple units readily available within the Mare Sicuro and EUNAVFOR MED missions present in proximity, which could have completed the rescue safely.

II. The Decisive Influence Element

Besides the impact element, the decisive influence element regards the exercise of functional jurisdiction through indirect means. “Public powers,” in this instance, rather than being carried out by the authorities of the state concerned, are deployed through the medium of a local administration in a third country—whether with its legal consent, de facto connivance, or none of them, as the situation was in Ilaşcu and subsequent line of cases.Footnote ²²¹

The Strasbourg Court has maintained in this constant jurisprudence, regarding Russian and Moldovan (co-)responsibility for the violations perpetrated by the separatist government of Transdniestria, that an ECHR party engages its jurisdiction for the actions and (crucially also for the) omissions of a third actor, when the latter comes under its “decisive influence.”Footnote ²²² Such “decisive influence” can lead to the establishment of functional jurisdiction on account of the degree of dependency of the third actor in question on the support received by the ECHR party. Where the third actor survives “by virtue of the military, economic, financial and political support given to it” by the ECHR party,Footnote ²²³ this entails “that [same ECHR party’s] responsibility for its policies and actions.”Footnote ²²⁴ The reason is that this kind of critical support engenders a “continuous and uninterrupted link of responsibility … for the applicants’ fate.”Footnote ²²⁵ And this is true even when there may not be any “direct involvement” of the influencing ECHR party in the specific human rights violations alleged.Footnote ²²⁶ What is more, such a “continuous and uninterrupted link of responsibility” is considered to give rise to positive obligations to prevent human rights violations in the area controlled by the dependent third actor over which the ECHR party exercises “decisive influence.”Footnote ²²⁷

Although the Court designed this paradigm with a geographical rather than a functional area of control in mind, the parallels with S.S. are paramount, considering the multiple ways in which Italy has influenced Libya’s policy and practice in the Central Mediterranean, entailing control over a wide range of interdependent stakes, as Part C demonstrates. In November 2017, Libya lacked an SRR, an MRCC, and a coastguard function capable of receiving and responding to distress calls autonomously, which is why Italy’s input was essential.Footnote ²²⁸

In 2016, the LYCG was barely functional, due to vital assets and equipment having been destroyed by the NATO’s offensive during 2011–12.Footnote ²²⁹ For the former Italian Minister of Interior, Minniti, prior to 2017, “when we said we had to re-launch the Libyan coastguard, it seemed like a daydream.”Footnote ²³⁰ Plans to develop a system of border surveillance in Libya, in general, and a functioning LN/LCG, in particular, as Part C has shown, were entirely “dependent” on Italy’s (and the EU’s) assistance.Footnote ²³¹ It was only after the MoU, and the related financial, logistic, and operative support provided by Italy, that the LYCG performed 19,452 pullbacks in 2017,Footnote ²³² up from 800 in 2015.Footnote ²³³

However, rather than contributing to diminishing the “horrific abuses” faced by migrants,Footnote ²³⁴ in accordance with the due diligence obligations attached to (an exercise of functional jurisdiction taking the form of) decisive influence,Footnote ²³⁵ the Italian plan deliberately led to their containment in Libya. Its interventions so far “have done nothing … to reduce the level of ill-treatment suffered by migrants” in the country. On the contrary, UN monitoring “shows a fast deterioration of their situation,”Footnote ²³⁶ including at the hands of the LYCG and after being pulled back.Footnote ²³⁷

What is clear, and the European authorities have recognized, is that the “increased performance of the Libyan Coast Guard [is a] direct consequence of the support … provided.”Footnote ²³⁸ “[T]here could not be a sufficient operational capability [of the LYCG] without … [the] training [and] equipment” delivered.Footnote ²³⁹ As the Italian Ministry of Foreign Affairs acknowledged in a public report, it is their “partnership with Tripoli which … has … produced [these] important results.”Footnote ²⁴⁰ It is “thanks” to Italy,Footnote ²⁴¹ rather than to Libya’s independent efforts,Footnote ²⁴² that there has been a near 90 per cent decrease in the number of arrivals at Italian shores by mid-2018.Footnote ²⁴³

These results are not accidental, unforeseen, or unintended. They are planned and expected. They stem from the direct application of the Treaty of Friendship and the 2017 MoU. They constitute the concrete realization of their object and purpose. Indeed, Italy’s support has specifically been targeted at “reinforcing the autonomy of [Libyan] operational capacities,”Footnote ²⁴⁴ with a view to transferring coordination responsibilities for rescue and interdiction in what was to become the Libyan SRR. And that investment in capacity building of the LYCG is not unconditional. In the words of the EUNAVFOR MED command, it is provided “in exchange for [Libyan] cooperation in tackling the irregular migration issue.”Footnote ²⁴⁵ So, the support lent to the LYCG has explicitly been understood as a quid pro quo, in a bid to exert influence over the manner in which Libyan constabulary functions are implemented at sea, in order to achieve the desired outcome of foreclosing maritime crossings towards Italy. Accordingly, it has only been “under pressure” from Italy (and the EU) that “Libyan authorities [have] increased their efforts to address the irregular flow of migrants.”Footnote ²⁴⁶

The pressure has come from different directions, not only from the political and operational spheres, but also from the dedicated Italian-Libyan Joint Commission created by the MoU.Footnote ²⁴⁷ In accordance with its mandate, the Joint Commission has formulated the “strategic priorities” of the Italian-Libyan collaboration pursuant to which Italy has delivered funding, training, equipment, and the main patrol vessels in the Libyan fleet. So, the definition of such “strategic priorities” and their practical implementation are key towards the establishment and full capacitation of the LYCG. They are, arguably, tantamount to “the formulation of essential policy,” as defined by the Strasbourg Court in Jaloud,Footnote ²⁴⁸ further supporting the conclusion that Italy, although not directly involved in each and every individual action of the LYCG, did not merely exert pressure, but “decisive influence” in the overall implementation of the plan to stem irregular migration across the Central Mediterranean. It is Italy’s comprehensive investment that made pull-backs a reality in the course of 2017, thus providing “a strong indication” that it exercised decisive influence over the LYCG in a way such as to trigger Article 1 ECHR.Footnote ²⁴⁹

III. The Operative Involvement Element

Beyond its implication from a distance, through the “impact” and “decisive influence” elements identified in the previous parts, Italy’s involvement in the operative capacities of the LYCG, especially in the course of 2017, has been very direct too—so much so that it fits the “public powers” doctrine to the letter, as formulated in Al-Skeini. To be sure, not only did Italy assume state functions of those normally pertaining to the territorial sovereign, but it did so on the grounds of the MoU and related decisions of the Joint Commission established by it—therefore, with “the consent, invitation or acquiescence of the state concerned.”^{Footnote 250}

As elaborated upon in Part C, November 6, 2017, was not an isolated occurrence, in terms of the overall functional authority undertaken by Italy in the coordination of SAR in the waters off Libya. Although Libya had ratified the SAR Convention, it had not officially declared an SRR according to the applicable formalities at the time of the S.S. events. An information document submitted by Italy (not Libya) to the IMO in December 2017 reveals that the process of “assist[ing] the relevant Libyan authorities in identifying and declaring their SRR” was still ongoing.^{Footnote 251}

Actually, for the declaration of an SRR to be valid, the SAR Convention foresees that there be an agreement among the Parties concerned (usually including all neighboring coastal states) to be notified to the IMO for dissemination,^{Footnote 252} and that SAR services be fully operational within the SRR being declared, so that they “are able to give prompt response to distress calls.”^{Footnote 253} That the existence of a functioning MRCC is “a prerequisite for efficiently coordinate [sic] search and rescue within the Libyan search and rescue zone, in line with international legislation,” has been jointly declared by the EU Commission and the EU High Representative for Foreign Affairs.^{Footnote 254}

The obligation on coastal states is to run “an adequate and effective” SAR service.^{Footnote 255} To that end, parties responsible for an SRR normally undertake “overall coordination of SAR operations,”^{Footnote 256} for which purpose they “shall make provision for the coordination facilities required to provide SAR services round their coasts” and “shall establish a national machinery for the overall coordination of SAR services,”^{Footnote 257} in the form of rescue coordination centers.^{Footnote 258} Above all, MRCCs “shall have adequate means for the receipt of distress communications” and “adequate means for communication with its rescue units and with MRCCs in adjacent areas.”^{Footnote 259} And rescue units attached to them must, in turn, be “suitably … equipped,” staffed, and managed, with appropriate “facilities and equipment” that allow for an effective response^{Footnote 260}—all of which was, and still is, lacking in the Libyan case.

As shown in Part C, Libyan MRCC functions have, instead, been secured by Italy, arranging for the dispatch and coordination of resources within SAR missions, ascertaining the movement and location of vessels in distress, developing rescue plans, designating OSC, communicating with rescue assets at sea, coordinating their action, and even arranging for briefing and debriefing of LYCG personnel.^{Footnote 261} Italy should, therefore, be considered to have assumed “overall control,” in the functional sense, of this Libyan competence,^{Footnote 262} which it exercises both “directly, through its [own naval] forces”—deployed in Libya and at sea, within Operation Nauras, and within its own Coastguard and MRCC—as well as “through a subordinate local administration” embodied in the LYCG.^{Footnote 263} It is Italy (also with the EU’s input) that has put in place the whole technical and material infrastructure (not only the ships and the equipment, but also the whole detection and communication apparatus) that enables the interception and return of migrants back to Libya. And it is Italy that has assumed “effective authority” over individual SAR operations,^{Footnote 264} including the one it deployed in S.S. As a result, Italy should be considered responsible to “secure, within the [policy] area under its control, the entire range of substantive rights set out in the Convention” that arise in SAR and interdiction situations.^{Footnote 265}

The nature of the LYCG as a subrogate Italian proxy for interdiction and pull-back at sea has been confirmed by the Tribunal of Catania adjudicating on a related case concerning the rescue ship Open Arms of the NGO Proactiva. In his decision, the judge takes as proven the crucial role played by the Italian Nauras assets in detecting migrant boats off the Libyan coast and in leading LYCG operations.^{Footnote 266} The judge goes as far as to affirm that the interventions of Libyan patrol vessels happen “under the aegis of the Italian navy” and that the coordination of SAR missions is “essentially entrusted to the Italian Navy, with its own naval assets and with those provided to the Libyans.”^{Footnote 267} The phone number of the LYCG, as provided in their official headed paper, at least until the spring of 2018, corresponded to the phone number of the Italian Nauras vessel docked in Tripoli,^{Footnote 268} which further corroborates the “high degree of integration” between the two.^{Footnote 269} Ayoub Qassem, a spokesperson for the LYCG Tripoli sector, back in November 2017, had already confirmed this modus operandi. He explained how the LYCG uses “the information [delivered by Italy] to intercept people and return them to Libya, even if they are apprehended [rather than rescued] in international waters.”^{Footnote 270}

Italy de facto commands the SAR and interdiction response of the LYCG. In these circumstances, it should not be able to “evade its own responsibility by relying on its obligations arising out of bilateral agreements with Libya.”^{Footnote 271} It should, instead, be considered that the practice it promotes of refoulement by proxy, employing the LYCG to that end, amounts to an “exercise of [Italy’s] sovereign authority, the effect of which is to prevent migrants from reaching [its] borders,” thus engaging ECHR responsibility.^{Footnote 272}

On November 6, 2017, the measure of comprehensive dominium that Italy exercised over Libya’s SAR and interdiction functions was similar to that recognized by the Strasbourg Court in relation to occupied areas of territory of a foreign country in its case law.^{Footnote 273} Against this background, it should not be necessary to determine whether Italy exercised “detailed control” over every individual action of the LYCG.^{Footnote 274} Italy’s significant naval presence, through its Nauras and Mare Sicuro missions, as well as its all-encompassing provision to the LYCG—which only “survives as a result of [that] support”^{Footnote 275}—determine that it exercised “effective control” over the S.S. applicants throughout the chain of events of November 6, 2017. This includes those who drowned or were injured at sea, alongside those who were maltreated by LYCG officers and/or pulled back to Libya, “during the course of or contiguous to [SAR/interdiction] operations” carried out under Italy’s direction.^{Footnote 276}