1. Introduction: The role of distance

The externalization of European border control can be defined as the range of processes whereby European actors and Member States complement policies to control migration across their territorial boundaries with initiatives that realize such control extra-territorially and through other countries and organs rather than their own. The phenomenon has multiple dimensions. The spatial dimension captures the remoteness of the geographical distance that is interposed between the locus of power and the locus of surveillance. But there is also a relational dimension, regarding the multiplicity of actors engaged in the venture through bilateral and multilateral interactions, usually through coercive dynamics of conditional reward, incentive, or penalization. And there are functional and instrumental dimensions too, concerning the cost-effectiveness of distance-creation (in both ethical and legal grounds) vis-à-vis the (unwanted) migrant, who, removed from sight, is no longer considered of concern to the supervising State,[1] and the range of externalizing policy devices at the service of externalising agents in terms of purpose, format, delivery, and ultimate control.[2] European borders thus (re-)emerge as ubiquitous, multi-modal and translational systems of coercion – as an interconnected network of ‘little Guantánamos’.[3] This, in turn, creates a distance, both physically and ethically, that is utilized to shift away concomitant responsibilities.[4]
Distance, as the next sections will demonstrate, plays a crucial role as a mechanism not only of dispersion of legal duties, blurring the lines of causation and making attribution of wrongful conduct a difficult task, but also as an artefact of oppression and displacement in itself. It does not prevent (unwanted) migration but rather makes it unviable through legally sanctioned, safe channels, diverting it through ever more perilous routes. The immediate effect of this distance that externalization engenders is at least threefold. First, it leads to the disempowerment of migrants, who are left with no options for safe and legal escape, being instead coerced into dangerous courses operated by smugglers. Second, it legitimizes the actors enforcing externalized control on behalf, and for the benefit, of the European Union and its Member States. Repressive forces in third countries gain standing as valid interlocutors for cooperation, as a result; their democratic and human rights credentials becoming secondary, if at all relevant, as the Libyan case illustrates below. Third, legal alternatives, like the relaxation of controls or the creation of safe and regular pathways, are rejected; perceived as an illogical concession to the failure of the externalization project.
The final outcome, and what constitutes the focus of this contribution, is the ‘border-induced displacement’ effect,[5] resulting from the combination of the processes of extraterritorialisation and externalization taken together. Border-induced displacement is not equivalent to the original reasons forcing people into exile, but rather functions as a second-order type of (re-)displacement, produced precisely via (the violence implicated in) border control. This then leads to forms of ‘engineered regionalism’, that is, politics re-producing displacement in certain areas closest to the origin of flows.[6] ‘Safe third country’ rules and practices are the main vehicle of this development, discernible also within the EU, where the Dublin System has ‘rulified’ an asymmetric allocation of responsibility for asylum claims to peripheral countries situated at the external common frontiers of the Union, like Spain, Italy and Greece.[7] In the case of externalization, border-induced displacement is then imposed upon already-displaced persons by non-European actors implementing the EU’s pre-emptive control agenda, reinforcing prevailing patterns of exploitation and existing hierarchies of exclusion and subordination.
The ethical and legal consequences of ‘distance-creation’ are what we turn to analyse in the remainder of this article. Section 2 pays attention to the assumptions and ethical and political-economic dimensions behind this strategy, discussing exit control, coercion, and the democratic legitimization of unelected actors enforcing the EU border within third countries. Section 3 investigates the legal impact of externalization and extraterritorialization, centring on the apparent accountability gaps that it generates, contesting the legality of responsibility dispersion mechanisms. The overall conclusion we reach is that the ‘rulification’ of externalization at EU level does not render it ethically and legally tenable under international law. The ‘lawification’ at EU level of practices inconsistent with human rights is insufficient to render them compatible with international legal standards.

2. Ethical distance-creation: Examining attempts to justify externalization and border-induced displacement

Although immigration ethics has thrived as a discipline since its late arrival in the 1980s, debates on border control between cosmopolitanism and liberal nationalism have often remained at an ideational level and generally based on liberal democratic foundations,[8] thus overlooking the composite ways through which border control is realized and experienced on the ground. This includes practices of externalization and extra-territorialization. Often, the assumptions guiding ethical debates on border control have reproduced a territorially trapped gaze, circumscribed by methodological nationalism,[9] which, through a set of idealized premises, reduces the complex and transnational dynamics of displacement and border control to a phenomenon of mis-placement between territorially bordered societies.[10] Such reduction is marred by what can be called reactive and regionalist postulations. These view border control, first, as a manifestation of State agency, and, second, as only a response to migration flows. Third, they naturalize the containment of displacement within certain regions, perceiving the phenomenon as geographically and morally distant from Europe.
But immigration ethics is far from alone in reproducing methodological nationalism and reactive and regionalist conjectures, as these mirror prevailing paradigms about the relationship between displacement and borders.[11] However, it is instructive, nonetheless, to examine European externalization by applying existing ethical debates about the democratic legitimacy, coercion, and rights of border control to the issue of externalization.[12]

2.1. The democratic legitimacy question

One fundamental debate has concerned the democratic legitimacy of border control as such. Assuming that freedom and democracy are instrumentally valuable for securing individual autonomy, a principled concern is that the coercive aspects of border control amount to violations of autonomy when they happen without the consent of those exposed to them. In order for border control to be legitimate from a liberal democratic perspective, it would have to be justifiable to non-members – however the demos may initially be defined – through a deliberative process.[13] Yet, proponents of border control might argue that access to asylum procedures can resolve this concern, if asylum applications are seen as granting such deliberative voice to them. Although this debate has only concerned an undifferentiated notion of border control, we can extend it to the politics of externalization, if we imagine proponents to argue that, if externalized control is able to respect individual autonomy, it might also be deemed democratically legitimate.[14] The strength of such an argument will then depend on the meaning and function of externalization.
European externalization processes occur when European Member States, through bi-, multi- or supranational venues, complement policies of controlling cross-border migration into their territories with pre-emptive initiatives realizing such control extra-territorially and/or through sub-contracting to actors and agencies other than their own.[15] Externalization has been discussed in terms of policy transfer, issue-linkages, and ripple effects,[16] but, crucially, its dynamics apply also to intra-European relations. For many years, the Dublin system has served to transfer the border control burdens of North-Western Member States to South-Eastern ones, causing heated discussions about lacking solidarity,[17] similar to those between European and non-European countries.[18]
Justifications offered for externalization oscillate between grammars of securitized control and humanitarian care.[19] For instance, the June 2018 proposal by the EU ministers about ‘controlled centres’ and ‘regional disembarkation platforms’, whereto ‘boat migrants’ can be deported, is framed as an innovative idea allowing Member States both to ‘stem illegal migration’ and simultaneously save vulnerable migrants by breaking the ‘business model’ of smugglers and traffickers purportedly in accordance with human rights and the rule of law.[20]
Yet, the 2018 externalization proposal is not as innovative as it may seem. Between the 1980s and mid-2000s, five very similar – and similarly controversial – externalization proposals were put forth by the British, Danish, Dutch, and German governments and by the European Commission. And they all revolved around externalized centres in Eastern Europe and North Africa whereto EU Member States would send asylum seekers or interdicted ‘boat migrants’. The terminologies varied from ‘regional protection areas’ by the British, ‘processing centres’ by the Danes, ‘reception centres’ by the Dutch, ‘EU reception centres’ by the German, and ‘Regional Protection Programmes’ (RPPs) by the European Commission.[21] All but the RPP proposal focused on administrative deportation from European territory, so that, as put by the Blair government, ‘refoulement should be possible and the notion of an asylum seeker in[land] should die’.[22] By 2005, the German proposal had dropped any talk of extraterritorial asylum processing and moved on to identifying Libya as a promising collaborator for pre-emptive containment.[23] In light of the concurrent dysfunctional intra-European dynamics of the Dublin system, the proposals between 1986 and 2018 illustrate how the externalization logic has long been invoked as a magic remedy to the Dublin ills, always couched in crisis-laden and emergency-driven rhetoric, while also holding out vague promises of protection.
Externalization can be criticized for co-opting protection in favour of methods of ‘consensual containment’ that re-produce displacement in regions neighbouring the EU.[24] For instance, especially since 2017, Italy and the EU have pursued a policy of transferring search and rescue to the so-called Libyan Coast Guard (LYCG), thereby effectively turning missions into operations of exit control. It is due to their material contribution and close involvement in the internal command-and-control structure of the Libyan forces that the LYCG performed 19,452 pull-backs in 2017.[25] Political discourses on externalization can, however, be seen as arguing that this kind of regionalist engineering creates ‘protection elsewhere’ based on three claims, popular in ethical discussions on border control within liberal national regimes. In the following, we analyse them through standing ethical debates about coercion and prevention, peoples’ rights to enter and exit territories, and democratic legitimacy.

2.2. Coercion: From ‘protection elsewhere’ to ‘protection nowhere’

First comes the claim that border control, and thus also its externalized manifestations, is not illegitimately coercive, because it is only preventive. Here, coercion has been referred to as when individuals are forced to do a specific thing, while prevention is taken to mean when they are forced not to do a specific thing.[26] Second comes the aforementioned argument that border control can be legitimate when agreed upon democratically.[27] Third follows the statement of an entry/exit-asymmetry signifying that people’s rights against one State not to prevent them from exiting its territory is held to be morally paramount, but that it does not entail an equally forceful obligation on any other State to let them enter their territory.[28]
Combining these claims, we then arrive at a ‘protection elsewhere’ argument maintaining that externalization is legitimate, since agreed to by all governments involved, and because it preserves displaced persons’ rights through extraterritorial asylum processing. Even if the policy may block their movement, this argument goes, it only prevents them from entering European territory, while still allowing them to find protection elsewhere, after having exited their own country. The zero-sum game effect that the generalisation of this policy would generate goes unaverted – if all countries did the same there would be ‘protection nowhere’.[29]
But this argument is categorically flawed. Its definitions of coercion and prevention are problematic and rest upon a disconnect between abstract assumptions about border control guiding liberal nationalistic immigration ethics and the actual reality of displacement and European border surveillance, discounting its concrete effects on the ground. EU externalization practices yield extremely coercive checks amounting to violent regimes of exit control, also contravening the legally-sanctioned right – assumed in debates on immigration ethics – to leave one’s own country.[30] That is, even if one, for the sake of argument, assumes the right to exit to hold more value than that of entry – since at international law one is universally applicable while the other is only opposable to one’s own country[31] – actual externalization practices still violate not just the latter, but also the former.[32] The containment of migrants in Libyan detention structures, for instance, reveals an abusive regime that bars access to asylum. Amnesty International has counted twenty reports from reliable monitors, including UN and EU sources, attesting to this reality.[33] The abject brutality facing displaced persons, contained and circulated through externalization, can only be labelled non-coercive prevention from a Eurocentric, and extremely abstract vantage point. In truth, they cause suffering on such a scale that they may amount to atrocity crimes, according to the ICC Prosecutor,[34] and, as the UN High Commissioner for Human Rights has put it, they constitute ‘an outrage to the conscience of humanity’ – at least as far as the situation in Libya is concerned.[35] Collaborative border infrastructures are endowed with the power to coerce at a distance, with externalization leading to practices of ‘remote control’ that extraterritorially negate access to the European asylum systems to those (theoretically) entitled to international protection,[36] literally ‘trapping’ migrants in a constant ‘cycle of abuse’.[37]
Nevertheless, even if the ethical ‘protection elsewhere’ argument must be rejected as an invalid justification for current European externalization policies the reasons for it are instructive. Seeing how externalization produces highly coercive collaborative regimes of exit control makes clear the problematic ramifications of the reactive and regionalist assumptions on which it rests. Conventional views on international relations and forced migration see the displacement to which borders respond as induced by conflicts or developmental or environmental factors.[38] Yet, while attention to the causes of displacement is important, this model embraces borders as only reactive to – rather than also constitutive of – displacement. But this is wrong. A range of border practices and infrastructures, performed at or beyond the physical frontiers of the EU, such as interdiction, detention, and deportation, do not just react to, but also in themselves cause displacement, by diverting flows towards increasingly dangerous routes and by multiplying death ratios at sea and at border zones.[39] This ‘border-induced displacement’, therefore, challenges the regionalist and reactive premise that the production of forced migration is primarily a problem created outside European territory and agency and contests the structural incorporation of (foreseeably lethal) coercion as a legitimate mechanism of border control.
EU-Libyan relations, since the 2000s, illustrate how externalization has built the infrastructures enabling this kind of coercive re-displacement. This problematizes prevailing assumptions still dominating immigration ethics and politics, namely that the agency of border control consists of States’ discretion over movement across their territorial borders. Externalization underscores the need to consider more composite notions of agency – and thus responsibility – decoupled from national territories, and spanning several governments, organisations as well as non-state actors.
The decades-long European-Libyan collaboration on border control is a case in point. After the European Commission decided to lift its arms embargo against Libya in 2004, two ‘technical missions’ followed. The first, in 2004, was meant to ‘identify concrete measures for possible balanced EU-Libyan cooperation particularly on illegal immigration’ and the second, in 2007, to develop ‘an operational and technical partnership’ for extraterritorial border control.[40] The case of Libya is but one example of how European externalization policies have facilitated the transformation of European border control into a flourishing market of violent deterrence and containment,[41] with little to do with a rights-based protection paradigm, and also how third countries’ control apparatuses have become a lucrative export venture for the arms-, security-, and IT-industries of the EU Member States.[42]

2.3. Trading in rights for border control

Companies like Spanish Indra, British BAE Systems, Italian Leonardo, French Thales and Ocea, Dutch Damen, German Rheinmetall and Airbus all compete for contracts to expand the capacity for surveillance and control of not just Libya, but also other Eastern European, North African and Middle Eastern countries collaborating on EU externalization. In 2012, an industrial consulting actor valued the global border industry at €25.8 billion, projecting an increase to €56 billion by 2022.[43] And European sales of patrol boats, jeeps, planes, drones, satellites, helicopters, radar systems and whole surveillance mechanisms for border control purposes were part of the EU export licenses worth €82 billion to the Middle East and North Africa between 2005–2014.[44] This political economy of externalization also applies to the industries of EU partner countries. For instance, in 2016, the EU channelled more than €83 million to contracts with Turkish Aselsan and Otokar to provide heavily armoured vehicles placed, respectively, at the Greek-Turkish border and the newly constructed 911 kilometre border-wall between Turkey and Syria.[45]
The dynamics reshaping third-country border infrastructures elucidate how borders can function as engines of, rather than just responses to, displacement. This means that arguments for externalization appealing to democratic legitimacy face more problems than merely the barring of access to asylum procedures: First, because when EU Member States use their political-economic leverage to make externalization deals with non-EU countries, they are effectively asking them to replace their own public interest with the EU preference of avoiding asylum seeker flows towards the Member States. Second, because several examples, like the EU collaboration with Libyan actors, including militias and former traffickers, as further discussed in the next section, illustrate how the EU’s externalization partners very often lack democratic legitimacy.[46] EU border externalization entrenches forms of undemocratic governance in third countries, empowering undemocratic actors, transforming their relative weight within domestic structures, and weakening democratic channels of scrutiny, accountability, and power control. Externalization thereby risks creating a vicious cycle, where the influx of arms and funds to those actors willing to enact the European containment agenda grants them political validity, which is then used to undermine not only migrant rights, but also to repress domestic opposition and dissidence and thus destabilize internal democratisation processes. The short-term European goal of preventing asylum seeker flows thereby risks compromising the stated long-term goal of tackling the root causes of displacement,[47] which is sacrificed in the altar of externalised ‘integrated border management’.[48]

3. Legal distance-creation: The juridical implications of externalization and border-induced displacement

Externalization has not only been encapsulated in political and policy arguments and practices, but has also been embedded in law through the ‘protection elsewhere’ model. The ‘protection elsewhere’ model ultimately rests on the assumption that refugees and migrants are best served ‘at home’, whether it be in their countries of origin or in the neighbouring region (but away from the EU at any rate). ‘Onward movements’ defy this logic and are thus seriously penalized. Responsibility for reception and asylum has accordingly been delegated (or redirected) to countries proximate to the source of flows, via targeted rules on ‘safe third countries’ and readmission agreements that legalise the practice. But, as stated above, this (re-)allocation of protection duties to peripheral States is also part and parcel of the Common European Asylum System within the EU. The Dublin Regulation enshrines and ‘rulifies’ this vision for the Member States, allowing non-external border countries to deflect responsibility in a legal manner.
Against this background, EU countries feel legitimized to claim their own irresponsibility vis-à-vis non-Member States,[49] projecting the model onto their external relations and imposing compliance with EU control rules as a matter of course. Fatalities at sea and elsewhere are then presented as the result of disorder and illegality; something avoidable if only (EU) rules were observed and effectively enforced by non-EU partners. The structural conditions imposed by the externalization apparatus, and the injustice that ensues, are usually disregarded or downplayed as unintended collateral damage. The fact that illegality is the only way out of a situation of want or persecution, and that smuggling is the only remaining vehicle to reach safety, is routinely silenced. It is the smugglers who profit of the precarious situation of ‘boat migrants’ – the argument goes. So, the eradication of smuggling and a return to (EU) law and order is portrayed as the solution. The option to relax border control rules and adapt them to the imperatives of human dignity, decriminalising the irregular movement of forced migrants, is not even contemplated. That would be perceived as an illogical concession; a descent into chaos and the negation of the rule of (EU) law. This EU-centric conception of the law is what sustains the externalization edifice and nurtures the collaboration with third countries.
At the legal-strategic level, externalization politics are accompanied by at least two degrees of ‘irresponsibilitization’, enshrined in, and sanctioned by, EU law: responsibility diffusion and responsibility denial. ‘Diffusion’ refers to the relational dimension of externalization, to situations of multi-actor alliance where the causation chain and attribution operation become unclear, with different agents and organs of different States contributing to a particular (unlawful) result. By contrast, ‘denial’ captures scenarios of outright disclaiming of responsibility, where this is said to belong to a different actor altogether, according to the (usually EU-based) rules in place (or their self-serving interpretation).

3.1. Responsibility diffusion

The creation of physical distance, via exit control, disembarkation platforms, holding sites, or reception camps abroad, contributes to ‘irresponsibilitization’ through diffusion. None of the proposals put forth so far clarifies exactly who should be considered responsible for those intercepted in, and repatriated to, Libya or any alternative location hosting the centres. The overall supposition appears to be that EU Member States would ultimately escape the task.[50] But there is some residual notion that European countries could not completely ‘circumvent’ their obligations[51] – albeit without elaboration, even the Legal Service of the European Parliament concedes that migrants sent to disembarkation platforms located outside the territory of the Member States ‘should benefit from the guarantees provided for in the 1951 Geneva Convention […] and in the European Convention of Human Rights’, including the principle of non-refoulement.[52]
Actually, under international law, ‘no State can avoid responsibility by outsourcing or contracting out its obligations’.[53] Cooperation with third countries does not exonerate EU Member States from their non-refoulement and related duties – both under general customary law and as per the relevant international Conventions.[54] According to the Strasbourg Court, ‘[w]here States establish […] international agreements to pursue cooperation in certain fields of activity’, whatever their legal nature, validity, and intent,[55] ‘there may be implications for the protection of fundamental rights’. With this in mind, it would be ‘incompatible with the purpose and object of the [European Convention of Human Rights][56] if Contracting States were thereby absolved from their responsibility under the Convention in relation to the field of activity covered by such [agreements]’.[57] As a result, ‘[i]n so far as any liability under the Convention is or may be incurred, it is liability incurred by the Contracting State […]’.[58] Despite its cooperation with Libya or any other third country, the independent responsibility of each EU Member State participating in the scheme of externalized migration controls subsists, ‘where the person[s] in question […] risk suffering a flagrant denial of the guarantees and rights secured to [them] under the Convention’.[59]
Nor would Member States be able to evade responsibility by transferring functions to the UNHCR or the IOM – whatever their support and potential separate liability.[60] ‘Absolving Contracting States completely from their Convention responsibility in the areas covered by such a transfer would [again] be incompatible with the purpose and object of the Convention’, as Strasbourg clarifies. The final effect would be for ‘the guarantees of the Convention [to] be limited or excluded at will thereby depriving it of its peremptory character and undermining the practical and effective nature of its safeguards’,[61] negating the basic premise of the pacta sunt servanda principle.[62] And the same is true in regard to other instruments of international human rights law.
Even though several actors combine to produce re-displacement, individual responsibility for its effects cannot be deflected. The principle is well established in international law. Article 47 of the ILC Articles on Responsibility of States for International Wrongful Acts (ARSIWA) contemplates precisely the scenario where several States participate in the same internationally wrongful act, stipulating that in such cases ‘the responsibility of each State may be invoked in relation to that act’.[63] Each State retains responsibility and, according to the ILC Commentary, ‘is separately responsible for the conduct attributable to it’. The fact that one or more additional States also contribute to the same act in no way reduces the responsibility of each single country.[64] So, any orders or transfers performed, or orchestrated by, EU Member States will engage their responsibility for any resulting breaches of their international commitments.
Neither the ‘disembarkation platforms’ proposal, nor any other of the similar initiatives emerged since the 1980s explored above specifies where exactly those repatriated or ‘pulled back’, whether to Libya or other third countries, would be accommodated.[65] It is conceivable that proponents envisage offshore reception centres to be closed, since the ultimate aim is to contain and deter irregular movement.[66] This then entails large-scale, and potentially long-term, detention, in breach of Article 5 ECHR guarantees,[67] which have been recognised to apply extraterritorially, extending to cases of deprivation of liberty abroad.[68] Yet, the border-induced displacement effects of externalization practices, like involuntary retention in international waters, forcible transfer to warships, coercive escorting or imposing of a certain course, constitute restrictions of physical freedom and need to accommodate the legal safeguards of the Convention.[69]
It is not known whether the ‘disembarkation platforms’ proposal foresees transfers to the country concerned to be automatic. Should that be the case, EU Member States risk incurring direct and indirect violations of the prohibition of collective expulsion and the (non-derogable/non-limitable) protection against refoulement. Regarding the latter, the Strasbourg Court attaches paramount importance to country information contained in reports from independent sources,[70] so that when reliable accounts of the circumstances prevailing in the receiving State make it ‘sufficiently real and probable’ that the general situation entails a ‘real risk’ of ill treatment in the sense of Article 3 ECHR, a refoulement presumption is activated and removal cannot be performed.[71] What is more, on account of the absolute character of Article 3, Contracting Parties must undertake the relevant investigation proprio motu and abstain from actions/omissions that put individuals at risk. As the Court asserted in Hirsi, ‘it [is] for the national authorities, faced with a situation in which human rights [are] systematically violated […] to find out about the treatment to which the applicants would be exposed after their return’.[72] So, the Member States concerned are to comply with their non-refoulement obligations proactively, regardless of whether the persons in question seek protection or specifically alert of the dangers faced upon return. The fact that potential applicants fail to request asylum or to formally oppose their removal does not absolve Contracting Parties of their Convention duties,[73] and especially their positive due diligence obligations.
This includes the requirement to provide access to adequate procedures.[74] Member States must offer a real opportunity for individuals to submit and defend their claims,[75] including an ‘effective remedy’.[76] This requires that the remedy in question be able to ‘prevent the execution of measures that are contrary to the Convention and whose effects are potentially irreversible’. Therefore, ‘it is inconsistent with Article 13 [ECHR] for such measures to be executed before the national authorities [of the Member State concerned] have examined whether they are compatible with the Convention’.[77] In these cases, appeals must have ‘automatic suspensive effect’.[78] And screening on board interdicting vessels or somewhere else offshore cannot satisfy these requirements.[79] Procedural responsibilities, just like substantive guarantees, cannot be deflected, postponed, or negated. The ultimate guarantors of ECHR safeguards are the Contracting Parties, which must ‘secure to everyone within their jurisdiction the rights and freedoms defined in [the] Convention’.[80]
Due diligence commands the dual duty to refrain from any conduct that may result in arbitrary violations as well as the obligation to enact laws and policies that effectively protect individuals against abuse. Following the Human Rights Committee’s recent General Comment on the Right to Life, by analogy, State Parties are required to ‘organise all State organs and governance structures through which public authority is exercised in a manner consistent with the need to respect and ensure [human rights]’. This includes a duty of ‘continuous supervision’ in order to ‘prevent, investigate, punish and remedy’ any harm.[81] As a result, actions such as the ‘sale […] of […] weapons’, and presumably other similar law enforcement and border control equipment, must be preceded by a conscientious examination of its foreseeable impact on human rights.[82] As members of the international community and as subjects of customary law, States must take into account

‘their responsibility […] to protect lives and to oppose widespread or systematic attacks on [human rights]’[83] – like those sustained by migrants in Libya.[84] And, in particular, States have an obligation under general international law ‘not to aid or assist activities undertaken by other States and non-State actors that violate [human rights]’.[85]

All these reasons should lead to the rejection of ‘disembarkation platforms’ and similar initiatives as ‘externalization fantasyland’.[86] EU Member States should not invest in a formula that promotes cooperation with human rights perpetrators and impedes the fulfilment of their pre-contracted obligations – such a course would hardly qualify as a good faith implementation of their binding commitments.[87] Instead, domestic systems of territorial protection should be reinforced, including the necessary intra-EU solidarity and responsibility-sharing mechanisms to make them effective.[88] Physical distance-creation, through off-shoring and outsourcing, does not translate into an erasure or diminution of legal duties. EU rules on ‘safe third countries’ and readmission cannot (unilaterally) undo international standards.[89]

3.2. Responsibility denial

Besides tools of responsibility deflection, mechanisms of outright denial of obligations are equally challenging. Usually, the capacitation of third countries’ control infrastructures, mimicking the Schengen ‘integrated border management’ system,[90] is framed as unproblematic. The transfer of funds, know-how, and equipment, as in the cases referred to in the previous section, are considered to emanate from a spirit of solidarity with non-EU partners and to be fully in line with the relevant criteria. The ethical distance between the EU or Member State gifting assets, ceding resources, or providing training and any potential human rights violations that may ensue is taken to preclude liability. There is no intent – no dolus specialis – intervening in the operation. Thus, the denial of responsibility on the European side for the atrocities in Libya, the abuses in Turkey, or the fatalities at sea associated with border-induced displacement, commonly recurs.[91]
Yet, international law paints a more complex picture.[92] If one considers that it is ‘thanks’[93] to Italy, for instance, that the LYCG continues to exist in any functional form in the post-Kaddafi period,[94] an outright denial of responsibility becomes difficult.[95]
Especially since the signature of the Memorandum of Understanding between Italy and the Libyan Government of National Accord in February 2017,[96] the delivery of training, equipment, and assets (including the four main patrol vessels employed by the LYCG) has intensified. Italy has created a dedicated ‘Africa Fund’, € 2.5 million of which has been allocated to the maintenance of LYCG boats and the training of their crews.[97] The EU, too, has committed € 46 million to prop up Libyan interdiction capacity.[98] It has been calculated that the total combined investment by Italy and the EU will be € 285 million by 2023,[99] with the EU alone providing € 282 million – most of which via programmes administered, coordinated, or supervised by Italy.[100] In addition, an extension of the Mare Sicuro Operation, named NAURAS,[101] was approved by the Italian Parliament in August 2017, consisting of four ships, four helicopters, and 600 servicemen, of which 70 per cent are deployed at sea, with the other 30 per cent stationed in Tripoli harbour. Their key mission, as declared by the Italian Navy itself, is to ‘establish [the] operational condition[s] for LN/LNCG [i.e. Libyan Navy and LYCG] assets and develop C2 [ie command-and-control] capabilities’. Meanwhile, an ‘ITN [ie Italian Navy] naval asset in Tripoli Harbour [is] acting as LNCC [ie Libyan Navy Communication Centre] and logistic assistance/support hub’, thus assuming the function of a floating maritime rescue coordination centre.[102]
The nature of the LYCG as a proxy for Italian interdiction has furthermore been confirmed by the judge of Catania adjudicating on the 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 Italy in leading LYCG operations. The judge goes so far as to affirm that the interventions of Libyan patrol vessels happen ‘under the aegis of the Italian Navy’ and that the coordination of rescue missions is ‘essentially entrusted to the Italian Navy, with its own naval assets and with those provided to the Libyans’.[103] This corroborates the ‘high degree of integration’ between the two,[104] and the ‘effective control’ exercised by Italy over LYCG operations, making ensuing violations attributable to it.[105]
The subsequent abuse of those pulled back to Tripoli happens despite Italy’s knowledge of the desperate situation facing migrants in Libya, including widespread and systematic torture, rape, inhuman and degrading treatment, and enslavement. The Deputy Minister for Foreign Affairs himself admitted that ‘taking [migrants] back to Libya, at this moment, means taking them back to hell’.[106] Nonetheless, the interdiction by proxy policy of Italy continues.[107] Amnesty International estimates that there are over 10,000 persons currently held in official detention centres in Libya – all of which funded through EU/Italian money. And, virtually all of them have been brought there as a result of their interdiction at sea by the EU/Italian-equipped and -trained LYCG.[108] Consequently, the combination of control exercised – though ‘contactless’[109] – and the knowledge of the circumstances migrants face should be understood to render Italy answerable for the resulting human rights violations, even if the LYCG is used as a surrogate.
As per Article 8 ARSIWA, ‘[t]he conduct of a person or group of persons [such as the LYCG] shall be considered an act of a State [i.e. Italy in this case]’, when the group in question ‘is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct’. Taking the Italian Navy and the Judge of Catania’s assertions at face value, the LYCG are to be considered ‘auxiliaries’ of the Italian border machinery deployed extraterritorially, ‘instructed to carry out particular [interdiction] missions abroad’. The Italian Navy conducts the specific operations through its NAURAS effectives exercising coordination as well as command-and-control functions, meaning that the (wrongful) conduct of the LYCG shall be considered ‘an integral part of the operations’ aimed at impeding departures across the Central Mediterranean and thus be attributed to Italy.[110] It is the Italian authorities that locate targets, relay maritime coordinates, and equip and mandate the LYCG to proceed to the interdiction of migrant boats.[111] It is Italy that ‘directs’ the operations in a way that ‘does not encompass mere incitement or suggestion but rather connotes actual direction of an operative kind’.[112] Italian intervention is a sine qua non for the ‘pull-backs’ at sea to materialise, which could not be carried out autonomously by the LYCG.[113] Italy exercises ‘such a degree of control […] as to justify treating the [LYCG] as acting on its behalf’.[114]
Italy’s involvement in Libyan search and rescue (or rather, interdiction) operations, in different ways and throughout time, rather than just an instance of complicity,[115] engaging indirect responsibility, can thus be characterised as a breach entailing direct responsibility, consisting of a ‘composite act’. Article 15 ARSIWA establishes that an international obligation (of non-refoulement, for instance, and of non-arbitrary interference with the right to leave) may indeed be violated via ‘a series of actions or omissions defined in aggregate as wrongful’. The financing or training of the LYCG alone may be harmless and perfectly licit, but, when taken together and alongside the infiltration of the command-and-control chain of the LYCG by the Italian Navy, the whole, in light of the final outcome of pull-backs, becomes an illicit under international law.
Italian jurisdiction may indeed be engaged not only in relation to action occurring within its territory and in other areas subject to its ‘effective control’, but, as the Human Rights Committee has stated, also regarding conduct ‘having a direct and reasonably foreseeable impact on the right[s] […] of individuals [abroad]’.[116] The obligation to respect and protect human rights extends beyond territorial domain to all persons subject to its jurisdiction, that is, to ‘all persons over whose enjoyment of the right[s] [concerned] it exercises power’, including ‘persons located outside any territory effectively controlled by the State, whose [rights are] nonetheless impacted by its military and other activities’ – the transfer of money, equipment and enforcement capacity thus acquiring a significance of its own as a possible trigger of independent responsibility for wrongful conduct.[117] Not only the aiding and abetting of human rights violations is of relevance, whatever the form the assistance provided to the LYCG may take (whether commercial, financial, political, or logistical), but also actions (or omissions) that impede the effective enjoyment of human rights – counting the right to leave any country, to seek protection from harm, and to non-refoulement – matter too, from a legal perspective.[118] Following the Legal Service of the European Parliament in the context of its viability analysis of ‘disembarkation platforms’, engagement in any formal or informal arrangement with third countries – including Libya – to finance or contribute to the functioning of externalized structures of migration control ‘have to respect the prescriptions of the relevant provisions of international law’[119] – presumably including those under the ECHR, the ICCPR and general customary norms.[120] Failure to do so flouts the obligations concerned. Direct perpetration of an international wrong is not a pre-requisite for legal responsibility. Indirect contraventions – including via proxy – incur liability as well.[121]
Distance-creation, through the ‘rulification’ of ‘irresponsibility’ in legal texts or self-seeking effectuations, does not do away with international obligations, nor does it legitimize the suffering it provokes. The EU and its Member States must come to recognise the predictable effect and implications of their externalization agenda. And, alongside the UN Special Rapporteur on Torture, acknowledge that, as currently designed, their ‘migration policies can amount to ill-treatment’.[122] Actually, ‘[t]he primary cause for the massive abuse suffered by migrants […] is neither migration itself, nor organised crime […] but the growing tendency of States to base their official migration policies and practices on deterrence, criminalisation and discrimination’.[123] It is this distinct strategy that causes border-induced displacement, breaches human rights obligations and triggers international legal responsibility.[124]

4. Conclusion: ‘Rulification’ as the co-option of protection

‘Rulification’ does not represent a paradigm shift in European politics, but rather an up-scaling of the logic observable also in proposals pursued from the 1980s and onwards and which have led to the integration of the concepts of ‘first country of arrival’, ‘safe third country’ and maritime interdiction within the legal architecture of the common borders and asylum acquis, the primary purpose of which has been the avoidance of asylum seekers on EU territory. It is the abuse and exploitation entrenched within externalization strategies that engenders border-induced displacement in Europe’s border-region. With EU Member States viewing the opening up of legal escape routes as an irrational concession, the side-effects of externalization are exacerbated as the systemic logic of asymmetric, diffused, and denied responsibility for displaced persons is reproduced further and further away from Europe, and closer and closer to the repressive regimes people attempt to escape from.
The reactionary and regionalist assumptions underpinning externalization arguments and practices tell a securitized tale of displacements constantly generated and managed far removed from European territory and agency. However, distance-creation strategies, whether ethical, spatial, or legal, belong to the category of ‘policies based on deterrence, militarization and extraterritoriality’, denounced by UN Special Rapporteurs and others, ‘which implicitly or explicitly tolerate [and perpetuate] the risk of migrant deaths as part of an effective control of entry’.[125] As the previous sections demonstrate, the structural nature of externalization problematizes traditional assumptions and debates in immigration ethics and politics. It traps migrants in a ‘vicious circle’ of more control, more danger, and more displacement, where they must rely on facilitators to escape life-threatening perils.[126]
But smuggling and trafficking is the consequence, rather than the cause, of suffering. Suffering is embedded in the externalization system by design through the vehicle of ‘rulification’, which serves to launder the pernicious (and perfectly foreseeable) impact of extra-territorialised/externalised coercion into ‘law-ified’ (and purportedly unintended) side effects. At the same time, the European transfer of equipment and capacity for control outwards also risks undermining processes of accountability and democratic legitimacy in regions bordering Europe. And the ‘rulification’ of border-induced displacement does not make these implications any more palatable. In the words of UN Special Rapporteur Agnès Callamard, it is simply ‘not acceptable’ to deter entry by endangering life.[127] The fallacy of coercion-based protection needs to give way to an ethically grounded and legally sustainable rights-honouring paradigm. This is not to contest the legal existence of borders or their enforcement, but to challenge the legitimacy of mechanisms through which they are presently enacted in a manner incompatible with the most basic requirements of international law.



* Senior Lecturer in Law, School of Law, Queen Mary University of London, and Visiting Professor, College of Europe.

** Assistant Professor in Global Refugee Studies, Department of Culture and Global Studies, University of Aalborg.

[1] The ‘where’ of borders matters and determines access to protection. See eg M Dikeç, ‘The “Where” of Asylum’ (2009) 27 Environment and Planning D: Society and Space 183; P Novak, ‘The Neoliberal Location of Asylum’ (2019) 70 Political Geography 1.

[2] On this multi-dimensional taxonomy of externalization, see R Zaiotti, ‘Mapping Remote Control: The Externalization of Migration Management in the 21st Century’, in Zaiotti (ed.), Externalizing Migration Management (Routledge 2016) ch 1.

[3] M Gibney, ‘A Thousand Little Guantánamos: Western States and Measures to Prevent the Arrival of Refugees’, in K Tunstall (ed), Displacement, Asylum and Migration: The Oxford Amnesty Lectures 2004 (OUP 2006) 139. See also P Novak, ‘The Flexible Territoriality of Borders’ (2011) 16 Geopolitics 741.

[4] For a legal elaboration, see V Moreno-Lax, Accessing Asylum in Europe (OUP 2017) ch 2.

[5] M Lemberg-Pedersen, ‘Forcing Flows of Migrants: European Externalization and Border-Induced Displacement’, in DJ Andersen, M Klatt, M Sandberg (eds), The Border Multiple (Ashgate 2012) 35.

[6] M Gibney, ‘Forced Migration, Engineered Regionalism, and Justice between States’, in S Kneebone and F Rawlings-Sanaei (eds), New Regionalism and Asylum Seekers (Berghahn Books, 2007) 59.

[7] Regulation (EU) No 604/2013 ‘Dublin III’.

[8] J Carens, ‘Aliens and Citizens: The Case for Open Borders’ (1987) 49 The Review of Politics 251; B Barry and RE Goodin (eds) Free Movement: Ethical Issues in the Transnational Migration of People and Money (Harvester Wheatsheaf 1992); M Blake, ‘Distributive Justice, State Coercion, and Autonomy’ (2002) 30 Philosophy & Public Affairs 257; M Gibney, The Ethics and Politics of Asylum. Liberal Democracy and the Response to Refugees (CUP 2005).

[9] J Agnew, ‘The Territorial Trap: The Geographical Assumptions of International Relations Theory’ (1994) 1 Rev Intl Political Economy 51; A Wimmer, N Glik Schiller, ‘Methodological Nationalism and Beyond’ (2002) 2 Global Networks 301.

[10] The dominant grounding of ethical debates about border control on a fusion of liberal democratic values and methodological nationalism also means that critical engagement has been neglected concerning the politically popular perspectives, which refuse universal, liberal norms, and question or reject rights-regimes and the associated state obligations. This is beyond the scope of the current analysis, see though M Lemberg-Pedersen, ‘The “Imaginary World” of Nationalistic Ethics: Feasibility constraints on Nordic deportation corridors targeting unaccompanied Afghan minors’ (2018) 12 Nordic J of Applied Ethics 47.

[11] Generally on the different epistemological approaches to the relationship between migration and borders and their contradictions, see P Novak, ‘Back to Borders’ (2017) 43 Critical Sociology 847.

[12] Cf A Abizadeh ‘Democratic Theory and Border Coercion’ (2008) 36 Political Theory 37; D Miller, ‘Why Immigration Controls Are Not Coercive’ (2010) 38 Political Theory 111.

[13] See also P Cole, Philosophies of Exclusion: Liberal Political Theory and Immigration (Edinburgh UP 2000).

[14] This argument faces further myriad problems that we will not be able to address for reasons of space. These concern how to determine the proportional and legitimate levels of coercion that the access to asylum would allow, whether asylum procedures are deliberative in the manner required to fulfill the conditions of autonomy, etc.

[15] M Lemberg-Pedersen (n 5); V Moreno-Lax (n 4) chs 4, 5 and 6; T Balzacq, ‘The Frontiers of Governance: Understanding the External Dimension of EU Justice and Home Affairs’, in T Balzacq (ed), The External Dimension of EU Justice and Home Affairs (Palgrave Macmillan 2009) 1.

[16] M Lemberg-Pedersen, ‘Effective Protection or Effective Combat? EU Border Control and North Africa’, in P Gaibazzi, S Dünnwald and A Bellagamba (eds), EurAfrican Borders and Migration Management (Palgrave Macmillan 2017) 36; S Lavenex and EM Uçarer, ‘The External Dimension of Europeanization: The Case of Immigration Policies’ (2004) 39 Cooperation and Conflict 417.

[17] For a recent critique, see E Küçük, ‘The Principle of Solidarity and Fairness in Sharing Responsibility: More than Window Dressing?’ (2016) 22 ELJ 448.

[18] Cf D Mara and C Nathler, ‘Tunisian Migrant Issue Dominates EU Interior Minister Talks’, Deutsche Welle (11 April 2011) <www.dw.com/en/tunisian-migrant-issue-dominates-eu-interior-minister-talks/a-14981095>. For a legal perspective, see V Moreno-Lax, ‘Solidarity’s Reach: Meaning, Dimensions and Implications for EU (External) Asylum Policy’ (2017) 24 Maastricht J Eur and Comparative L 740.

[19] Cf J Carling, M Hernández-Carretero, ‘Protecting Europe and Protecting Migrants? Strategies for Managing Unauthorised Migration from Africa’ (2011) 13 The British J of Politics and Intl Relations 43; P Pallister-Wilkins, ‘The Humanitarian Politics of European Border Policing’ (2015) 9 Intl Political Sociology 53; P Cuttitta, ‘Delocalization, Humanitarianism, and Human Rights: The Mediterranean Border Between Exclusion and Inclusion’ (2018) 50 Antipode 783; and V Moreno-Lax, ‘The EU Humanitarian Border and the Securitization of Human Rights: The “Rescue-through-Interdiction/Rescue-without-Protection” Paradigm’ (2018) 56 J Common Market Studies 119.

[20] European Council Conclusions, 28 June 2018, Press Release 421/18. See also European Commission, Migration: Regional Disembarkation Arrangements (24 July 2018) <ec.europa.eu/home-affairs/sites/homeaffairs/files/what-we-do/policies/european-agenda-migration/20180724_factsheet-regional-disembarkation-arrangements_en.pdf>.

[21] For analysis, see M Lemberg-Pedersen (n 10); G Noll, ‘Visions of the Exceptional: Legal and Theoretical Issues Raised by Transit Processing Centres and Protection Zones‘ (2003) 5 European Journal of Migration and Law 303; V Moreno-Lax, ‘External Dimension’, in S Peers, V Moreno-Lax, M Garlick and E Guild (eds), EU Immigration and Asylum Law (Brill, 2nd edn, 2015) vol 3 ch 10.

[22] UK Government New Vision for Refugees (White Paper, 2003) 9.

[23] Bundesministers des Innern Effektiver Schutz für Flchtlinge, wirkungsvölle Bekämpfung illegaler Migration – berlegungen des Bundesministers des Innern zur Errichtung einer EU-Aufnahmeeinrichtung in Nordafrika (Pressemitteilung 2005) 3-6.

[24] V Moreno-Lax, MG Giuffré, ‘The Raise of Consensual Containment: From “Contactless Control” to “Contactless Responsibility” for Forced Migration Flows’, in S Juss (ed), Research Handbook on International Refugee Law (Edward Elgar forthcoming) <www.unhcr.org/5a056ca07.pdf>.

[25] IOM, Maritime Update Libyan Coast, 25 October-28 November 2017, <www.iom.int/sites/default/files/situation_reports/file/IOM-Libya-Maritime-Update-Libyan-25Oct-28Nov.pdf>.

[26] These definitions of coercion and prevention are offered by D Miller (n 12) 114.

[27] A Abizadeh (n 12) 47. Miller and Abizadeh disagree on the boundaries of the demos involved, but this is not crucial to the stated version of the externalization argument.

[28] D Miller, National Responsibility and Global Justice (OUP 2007) 207-209. Arguably, the assumption of such asymmetry reflects the norms of Cold War migration geopolitics, epitomized by the condemnation of the Berlin Wall, and emphasizing the ethical value of allowing individuals persecuted by totalitarian regimes to exit territories. But today, the border control systems of European countries, purportedly liberal democratic, also involve practices of walling, fencing, and exit control. Moreover, a principled ethical critique of the entry/exit-asymmetry has also been that the very reasons why emigration might be crucial for securing individuals’ fundamental rights, and thus autonomy, may also apply equally strongly to the reasons why people try to immigrate. Cf N Holtug, ‘The Ethics of Immigration Policy’ (2011) 1 Nordic J Migration Research 4.

[29] For a legal elaboration see V Moreno-Lax, ‘The Legality of the “Safe Third Country” Notion Contested: Insights from the Law of Treaties’, in Goodwin-Gill and Weckel (eds), Migration & Refugee Protection in the 21st Century (Nijhoff 2015) 665.

[30] Report of the Special Rapporteur on the Human Rights of Migrants, François Crépeau, UN Doc A/HRC/29/36 (2015).

[31] For a full-length analysis of the right to leave, see V Moreno-Lax (n 4) ch 9.

[32] ibid. Cf. N Markard, ‘The Right to Leave by Sea: Legal Limits on EU Migration Control by Third Countries’ (2106) 27 Eur J Intl L 591.

[33] Amnesty International, Libya’s Dark Web of Collusion (December 2017) 56-58 <www.amnesty.org/en/documents/mde19/7561/2017/en>. See also Human Rights Watch, No Escape from Hell (January 2019) <www.hrw.org/report/2019/01/21/no-escape-hell/eu-policies-contribute-abuse-migrants-libya>.

[34] ICC Prosecutor, Statement to the United Nations Security Council on the Situation in Libya, pursuant to UNSCR 1970 (2011) 8 November 2017 <www.icc-cpi.int/Pages/item.aspx?name=otp_lib_unsc>.

[35] ‘UN Human Rights Chief: Suffering of Migrants in Libya Outrage to Conscience of Humanity’ Press Release (14 November 2017) <www.ohchr.org/EN/NewsEvents/ Pages/DisplayNews.aspx?NewsID=22393&LangID=E>.

[36] L Bialasiewicz, ‘Off-Shoring and Out-Sourcing the Borders of Europe: Libya and EU Border Work in the Mediterranean’ (2012) 17 Geopolitics 843.

[37] Amnesty International, Between the Devil and the Deep Blue Sea (August 2018) 18 <www.amnesty.org/download/Documents/EUR3089062018ENGLISH.pdf>.

[38] A Betts, Forced migration and global politics (Wiley-Blackwell 2009) 4-10.

[39] M Lemberg-Pedersen (n 16) 54-6; V Moreno-Lax (n 4) chs 3 to 6. Among others, The Human Cost of Border Control project provides primary source information and statistical evidence of these correlations <www.borderdeaths.org/>.

[40] Cf Technical Mission to Libya on Illegal Migration, 27 November–6 December 2004, Council Doc 7753/05 <www.statewatch.org/news/2005/may/eu-report-libya-ill-imm.pdf>; Frontex-led EU Illegal Immigration Technical Mission to Libya, 28 May–5 June 2007 <www.statewatch.org/news/2007/oct/eu-libya-frontex-report.pdf>.

[41] M Lemberg-Pedersen, ‘Private Security Companies and the European Borderscapes’, in N Nyberg Sørensen and T Gammeltoft-Hansen (eds), The Migration Industry and the Commercialization of International Migration (Routledge 2013) 152; R Andersson, Illegality Inc. Clandestine Migration and the Business of Bordering Europe (U California Press 2014); T Baird, ‘Interest Groups and Strategic Constructivism: Business Actors and Border Security in the European Union’ (2017) 44 J Ethnic and Migration Studies 118.

[42] M Lemberg-Pedersen, ‘Security, Industry and Migration in European Border Control’, in A Weinar, S Bonjour and L Zhyznomirska (eds), The Routledge Handbook of the Politics of Migration in Europe (Routledge 2018) 239.

[43] Frost & Sullivan, Global Border and Maritime Security Market Assessment Human-Intensive Security on the Border and Maritime Domain Will Increase the Demand for New Technology (February 2014) 965 ff <www.frost.com/sublib/display-report.do? id=M965-01-00-00-00>.

[44] Seventeenth annual report according to Art 8(2) of Council Common Position 2008/944/CFSP defining common rules governing control of exports of military technology and equipment [2016] OJ C 163/1.

[45] Z Şentek, S Arsu, ‘No Way Out: The European Union is Funding Military Equipment used by Turkey to Stop Refugees from Fleeing the Syrian Civil War and Entering the EU’ The Black Sea (23 March 2018) <theblacksea.eu/billions-for-borders/article/en/no-way-out#>.

[46] Eg ‘Libyan Militia Cash in on EU’s Anti-Smuggling Strategy’ EU Observer (5 October 2018) <euobserver.com/migration/143003>.

[47] M Lemberg-Pedersen, ‘Externalizing Brutality to Libya Is Not an Answer to Displacement’, ECRE Op-Ed (1 December 2017) <www.ecre.org/op-ed-externalizing-brutality-to-libya-is-not-an-answer-to-displacement>.

[48] Art 77(1)(c) TFEU.

[49] Cf S Fine, T Lindemann, ‘European Democracies and the Responsibility to Not Protect’, Open Democracy, 20 September 2018 <www.opendemocracy.net/can-europe-make-it/shoshana-fine-thomas-lindemann/european-democracies-and-responsibility-to-not-pro>.

[50] For a similar analysis, see F Maiani, ‘“Regional Disembarkation Platforms” and “Controlled Centres”: Lifting the Drawbridge, Reaching out Across the Mediterranean, or Going Nowhere?’ EU Migration Law Blog (18 September 2018) <eumigrationlawblog.eu/regional-disembarkation-platforms-and-controlled-centres-lifting-the-drawbridge-reaching-out-across-the-mediterranean-or-going-nowhere>.

[51] ‘Theo Francken Accuses EU of Hypocrisy’, The Brussels Times, 7 June 2018 <www.brusselstimes.com/eu-affairs/11558/theo-francken-accuses-eu-of-hypocrisy>.

[52] Opinion of the Legal Service of the European Parliament, Re: LIBE – European Union asylum and immigration policies – Regional disembarkation platforms and controlled centres – European Council Conclusions of 28th of June 2018 (on file with the authors) paras 44, 45, 53 and 58.

[53] G Goodwin-Gill, ‘The Extraterritorial Processing of Claims to Asylum or Protection: The Legal Responsibilities of States and International Organisations’ (2007) 9 UTS L Rev 26, 34. See also M Den Heijer, Europe and Extraterritorial Asylum (Hart 2012) ch 7.

[54] See further V Moreno-Lax, ‘Seeking Asylum in the Mediterranean: Against a Fragmentary Reading of EU Member States’ Obligations Accruing at Sea’ (2011) 23 Intl J Refugee L 174. See also V Moreno-Lax (n 4) ch 8 on the extraterritorial application of non-refoulement and its legal nature in international law.

[55] These agreements may indeed be null and void ipso jure if they conflict with jus cogens norms, pursuant to Art 53 VCLT, which some authors claim is precisely the rank of the prohibition of refoulement. See eg J Allain, ‘The Jus Cogens Nature of Non-refoulement’ (2001) 13 International Journal of Refugee Law 533. Cf C Costello and M Foster, ‘Non-refoulement as Custom and Jus Cogens? Putting the Prohibition to the Test’ (2015) 46 Netherlands YB Intl L 273.

[56] Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights) (adopted 4 November 1950, entered into force 3 September 1953) ETS 5 (ECHR).

[57] T.I. v United Kingdom App no 43844/98 (ECtHR, 7 March 2000) 15; K.R.S. v. United Kingdom App no 32733/08 (ECtHR, 2 December 2008) 15.

[58] Saadi v United Kingdom App no 37201/06 (ECtHR, 28 February 2008) para 126.

[59] W.M. v Denmark App no 17392/90 (EComHR, 14 October 1992).

[60] ‘IOM, UNHCR Appeal for Region-Wide Action by EU Countries over Mediterranean Tragedies’ UNHCR Press Release (27 June 2018) <www.unhcr.org/news/press/ 2018/6/5b33d8bf4/iom-unhcr-appeal-region-wide-action-eu-Countries-mediterranean-tragedies.html>.

[61] Bosphorus Hava Yolları Turizm ve Ticaret Anonim Şirketi v Ireland App no 45036/98 (ECtHR, 30 June 2005) para 154.

[62] Vienna Convention on the Law of Treaties (adopted 23 May 1969, entered into force 27 January 1980) 1155 UNTS 331 art 26 (heading).

[63] ILC Articles on Responsibility of States for Internationally Wrongful Acts, Annex UNGA Res 56/83 (2001), A/56/49(Vol. I)/Corr.4 (ARSIWA) (emphasis added).

[64] ILC Commentary on Articles on Responsibility of States for Internationally Wrongful Acts, fifty-third Sess, A/56/10 (2001) 124 (ARSIWA Commentary) (emphasis added). See also J Crawford, State Responsibility (CUP 2013) 355.

[65] Note that so far no African country has agreed to the strategy. See ‘EU Admits No African Country Has Agreed to Host Migration Centre’ The Guardian (21 June 2018) <www.theguardian.com/world/2018/jun/21/eu-admits-no-african-Country-has-agreed-to-host-migration-centre>.

[66] Some EU leaders have advocated for ‘floating centres’. See A Dastyari, D Ghezelbash, ‘Opinion: European Plans for Refugee “Floating Reception Centres” Will Endanger Lives’ ABC News (20 September 2018) <www.abc.net.au/news/2018-09-20/europe-asylum-seekers-australia-mediterranean-austria-italy/10265238>.

[67] Cf Saadi v United Kingdom (n 58). For a critique, see V Moreno-Lax, ‘Beyond Saadi v. UK: Why the “Unnecessary” Detention of Asylum Seekers is Inadmissible under EU Law’ (2011) 5 Human Rights & Intl L Discourse 166.

[68] Al-Saadoon and Mufdhi v United Kingdom App no 61498/08 (ECtHR, 2 March 2010). See also Al-Skeini v United Kingdom App no 55721/07 (ECtHR, 7 July 2011); Hirsi Jamaa and Others v Italy App no 27765/09 (ECtHR, 23 February 2012). See further, V Moreno-Lax, ‘Hirsi v Italy or the Strasbourg Court v Extraterritorial Migration Control?’ (2012) 12 Human Rights L Rev 574.

[69] Rigopoulous v Spain App no 37388/97 (ECtHR, 12 January 1999); Medvedyev v France App no 3394/03 (ECtHR, 10 July 2008).

[70] Hirsi Jamaa and Others v Italy (n 68) paras 118 and 123.

[71] ibid para 136.

[72] ibid para 133.

[73] ibid paras 133 and 157. See also G Goodwin-Gill, ‘The Right to Seek Asylum: Interception at Sea and the Principle of Non-Refoulement’ (2011) 23 International J Refugee L 443.

[74] Hirsi Jamaa and Others v Italy (n 68) paras 203-204.

[75] M.S.S. v Belgium and Greece App no 30696/09 (ECtHR, 21 January 2011) paras 301 and 319. See further, E Guild, C Costello, M Garlick, V Moreno-Lax, M Mouzourakis and S Carrera, New Approaches, Alternative Avenues and Means of Access to Asylum Procedures for Persons Seeking International Protection, PE509.989 (European Parliament 2014) ch 4 <www.europarl.europa.eu/RegData/etudes/STUD/2014/509989/ IPOL_STU(2014)509989_EN.pdf>.

[76] Jabari v Turkey App no 40035/98 (ECtHR, 11 July 2000) para 48.

[77] Conka v Belgium App no 51564/99 (ECtHR, 5 February 2002) para 79 (emphasis added).

[78] Gebremedhin v France App no 25389/05 (ECtHR, 26 April 2007) para 66; I.M. v France App no 9152/09 (ECtHR, 2 May 2012) paras 132, 134-135; A.C. v Spain App no 6528/11 (ECtHR, 22 April 2014) para 95.

[79] Cf ‘Austria, Italy Propose Processing Refugees on Ships’ Deutsche Welle (15 September 2018) <www.dw.com/en/austria-italy-propose-processing-refugees-on-ships/a-45496615>.

[80] Art 1 ECHR. For an elaboration and further references, see V Moreno-Lax (n 4) chs 8 and 10.

[81] HRC, ‘General Comment No 36’ (2018) on article 6 of the International Covenant on Civil and Political Rights (ICCPR) on the right to life UN Doc CCPR/C/GC/36, 30 October 2018, paras 7, 18, 19 and 21 (emphasis added). On the requirements of investigations, see paras 27-28. The ECHR imposes similar positive obligations under Arts 2 and 3. See eg Al-Saadoon and Mufdhi (n 68); M.S.S. (n 75); Othman (Abu Qatada) v UK App no 8139/09 (ECtHR, 9 May 2012).

[82] HRC (n 81) para 65.

[83] ibid para 70.

[84] Amnesty International (n 33); Human Rights Watch (n 33).

[85] HRC (n 81) para 63.

[86] C Woollard, ‘Editorial: Lost in Externalization Fantasyland’ ECRE Bulletin (22 June 2018) <perma.cc/NJL5-T4U5>.

[87] Art 26 VCLT: ‘Every treaty in force is binding upon the parties to it and must be performed by them in good faith’ (emphasis added).

[88] Art 80 TFEU. On the workings of the principle of solidarity and fair sharing of responsibility, see V Moreno-Lax (n 18).

[89] For a detailed discussion, see V Moreno-Lax (n 29).

[90] Art 77(1)(c) TFEU.

[91] For a recent example, see the EU’s reaction to the claim by 17 Nigerian migrants rescued on 6 November 2017 by Sea Watch filed against Italy for their ‘pull-back’ operations conducted via the LYCG, ‘Migranti: UE “L’Italia rispetta le leggi”’ ANSA (8 May 2018) <www.ansa.it/sito/notizie/topnews/2018/05/08/migranti-ue-italia-rispetta-le-leggi_17faec35-69dc-445e-9ead-491706e595dc.html>. For human rights violations occurring at sea since the outbreak of the 2015 ‘refugee crisis’, consult the Search and Rescue Observatory for the Mediterranean (SAROBMED) <https://sarobmed.org/>.

[92] On the specific matter of intent, see ARSIWA Commentary (n 64) 36: ‘it is only the act of a State that matters, independently of any intention’. In this line, see also V Lanovoy, ‘Complicity in an Internationally Wrongful Act: An Appraisal of the State of the Art’, in A Knollkaemper, I Plakokefalos (eds), Principles of Shared Responsibility in International Law (CUP 2014) 134, 152 ff.

[93] Chiragov and Others v Armenia App no 13216/05 (ECtHR, 16 June 2015) para 178.

[94] Cf Catan and Others v Moldova and Russia App no 43370/04, no 8252/05, no 18454/06 (ECtHR, 19 October 2012) para 106.

[95] For a detailed review of Italy’s entanglement with the Libyan Coast Guard, see C Heller and L Pezzani, ‘Mare Clausum: Italy and the EU’s Undeclared Operation to Stem Migration across the Mediterranean’, Forensic Oceanography, May 2018, Forensic Architecture <www.forensic-architecture.org/wp-content/uploads/2018/05/2018-05-07-FO MareClausum-full-EN.pdf>.

[96] Italy-Libya Memorandum of Understanding (2 February 2017) English translation <https://eumigrationlawblog.eu/wpcontent/uploads/2017/10/MEMORANDUM_translation_finalversion.doc.pdf>.

[97] The practice has been contested and a legal challenge is currently pending in this regard. See ‘NGO Takes Italy to Court over Misappropriation of Development Funds Directed to Libyan Coastguards’ ECRE News (24 November 2017) <www.ecre.org/ngo-takes-italy-to-court-over-misappropriation-of-development-funds-directed-to-libyan-coastguards>.

[98] European Commission, EU Cooperation on Migration in Libya (8 May 2018) <https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/eutf-noa-ly-08052018.pdf>.

[99] ‘EU and Italy put Aside €285m to Boost Libyan Coast Guard’ EU Observer (29 November 2017) <https://euobserver.com/migration/140067>.

[100] EU Emergency Trust Fund for Africa: Libya (undated) <https://ec.europa.eu/trustfundforafrica/region/north-africa/libya>.

[101] Maritime Rescue Coordination Centre Rome, Guardia costiera italiana, Annual Report 2017 <www.guardiacostiera.gov.it/attivita/Documents/attivita-sar-immigrazione-2017/Rapporto_annuale_2017_ENG.pdf>.

[102] Italian Navy, Operation Mare Sicuro Presentation, SHADE MED Briefing (23-24 November 2017) (on file with the authors).

[103] Tribunale di Catania, no 3476/18 R.G.N.R. and no 2474/18 R.G.GIP (27 March 2018) 22 <www.dirittoimmigrazionecittadinanza.it/allegati/fascicolo-n-2-2018/256-trib-catania-27-3-2018/file>.

[104] Chiragov and Others v Armenia (n 93) paras 176 and 186.

[105] Al-Skeini v United Kingdom (n 68) para 138 ff.

[106] ‘Giro: Fare rientrare quelle persone vuol dire condannarle all’inferno’, La Stampa (6 August 2017) <www.lastampa.it/2017/08/06/italia/cronache/giro-fare-rientrare-quelle-persone-vuol-dire-condannarle-allinferno-SXnGzVlzftFl7fNGFCMADN/pagina.html>.

[107] ‘Nigerian migrants sue Italy for aiding Libyan coast guard’ Reuters (8 May 2018) <www.reuters.com/article/us-europe-migrants-italy/nigerian-migrants-sue-italy-for-aiding-libyan-coast-guard-idUSKBN1I9206>.

[108] Amnesty International, Between the Devil and the Deep Blue Sea (n 37) 17-18.

[109] V Moreno-Lax and MG Giuffré (n 24).

[110] ARSIWA Commentary (n 64) 47.

[111] A Palladino, ‘Cercate i guardacoste libici? Telefonate a Roma: 06/…’, Il Fatto Quotidiano (18 April 2018) <www.ilfattoquotidiano.it/in-edicola/articoli/ 2018/04/18/cercate-i-guardacoste-libici-telefonate-a-roma-06/4299475/>. For a visual reconstruction of events occurring on 6 November 2017 between Sea Watch and the Libyan Coast Guard in a rescue operation coordinated by Italy, see ‘“It’s an Act of Murder”: How Europe Outsources Suffering as Migrants Drown’ New York Times (26 December 2018) <https://www.nytimes.com/interactive/2018/12/26/opinion/europe-migrant-crisis-mediterranean-libya.html>. For the general trend, see Mare Clausum (n 95).

[112] ARSIWA Commentary (n 64) 69 (emphasis added) regarding art 17 ARSIWA on ‘direction and control exercised over the commission of an internationally wrongful act’ (heading).

[113] Confirming this finding, see IMO, Libyan Maritime Rescue Coordination Centre Project submitted by Italy, NCSR 5/INF.17 (15 December 2017), Executive Summary: ‘This document provides information on the Libyan Maritime Rescue Coordination Centre (LMRCC) Project, an initiative run by the Italian Coast Guard and funded by the European Commission, with the aim of conducting a feasibility study in order to establish a Libyan MRCC […]’, which, for the time being, is lacking (emphasis added).

[114] Military and Paramilitary Activities in and against Nicaragua (Merits) [1986] ICJ Rep. 14, 62 and 64–65, paras 109 and 115.

[115] Making this case, see G Pascale, ‘“Esternalizzazione” delle frontiere in chiave antimigratoria e responsabilità internazionale dell’Italia e dell’UE per complicità nelle gross violations dei diritti umani commesse in Libia’ (2018) 13 Studi sull’integrazione europea 413. See also R Mackenzie-Gray Scott, ‘Torture in Libya and Questions of EU Member States Complicity’, EJILTalk! (11 January 2018) <https://www.ejiltalk.org/ torture-in-libya-and-questions-of-eu-member-state-complicity/>.

[116] Mutatis mutandis HRC General Comment (n 81) para 22.

[117] ibid. para 63 (emphasis added). For a similar reasoning by the Strasbourg Court, see Al-Skeini (n 68); Jaloud v The Netherlands, App no 47708/08 (ECtHR 20 November 2014); Chiragov (n 93); Catan (n 94); Ilascu and Others v Moldova and Russia, App no 48787/99 (ECtHR, 8 July 2004).

[118] Cf A Brown, ‘To Complicity… and Beyond! Passive Assistance and Positive Obligations in International law’ (2016) 27 Hague YB Intl L 133.

[119] Opinion of the Legal Service of the European Parliament (n 52) para 42.

[120] International Covenant on Civil and Political Rights [1966] 999 UNTS 171 (ICCPR). See also HRC (n 81).

[121] For an elaboration, see V Moreno-Lax, ‘Responsibility by Proxy and the Functional Approach to Jurisdiction: Closing Accountability Gaps in Multi-actor Constellations of Extraterritorial Cooperation’, Conference Paper, REF-MIG Project, Oxford (10 November 2018).

[122] ‘Migration Policies Can Amount to Ill-Treatment and Torture, UN Rights Expert Warns’, OHCHR News, 1 March 2018, <www.ohchr.org/EN/News Events/Pages/DisplayNews.aspx?NewsID=22739&LangID=E>.

[123] Report of the Special Rapporteur on Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, Nils Melzer, UN Doc A/HRC/37/50 (2018) para 64(d).

[124] In this line, see I Mann, V Moreno-Lax and O Schatz, ‘Time to Investigate European Agents for Crimes against Migrants in Libya’, EJILTalk! (29 March 2018) <www.ejiltalk.org/time-to-investigate-european-agents-for-crimes-against-migrants-in-libya/>.

[125] Report of the Special Rapporteur of the Human Rights Council on Extrajudicial, Summary or Arbitrary Executions, Agnès Callamard, UN Doc A/72/335 (2018) para 10.

[126] H de Haas, ‘Borders beyond Control’, International Migration Institute Network (January 2015) <www.imi-n.org/imi-archive/news/borders-beyond-control-blog-by-hein-de-haas>.

[127] A Callamard (n 125) para 59.