Since March 2020 asylum seekers arriving on Greek islands or in territorial waters have been dragged out to sea, forced by the Hellenic Coast Guard into inflatable orange life rafts, ‘shaped like a tent’ Footnote 1 and left to drift. Footnote 2 These incidents highlighted a pattern of violations that developed into a de facto government policy of summary and collective expulsions Footnote 3 – one which implicated the European Border and Coast Guard Agency, Frontex. Footnote 4 Survivors of the initial episodes recounted how the authorities they relied on for protection became agents of their persecution. The equipment designed to save their lives – objects designated for rescue and prescribed by the international law of the sea – facilitated their expulsion. The paradox of repurposing rescue equipment, demanded under the law of the sea, against those seeking protection represents the starting point of our inquiry.
These ‘drift backs’, as the practices have been called, are instances of direct and extreme violence by Greek state officials. Under international law, they engage Greek state responsibility and potentially individual criminal liability. Yet, beyond what international legal analysis suggests on its own terms, they were not made possible simply by a single state reneging on its duties, nor were they merely breaches of rogue individual actors. We argue that these violations reveal an important and still little-examined dark side of European ‘migration management’ as a process purportedly aimed to pacify and ‘civilize’ border violence. Footnote 5 In transforming life-saving materials into life-threatening ones, the incidents reveal important insights into the relationship between law, politics, and the materiality of physical objects. Footnote 6 Engaging with literature on ‘legal materiality’, we show how absent concrete political and moral commitments to international protection, rescue’s physical infrastructure has been weaponized. The repurposed rescue raft is emblematic of the way safety and humanitarianism are tied with security and deterrence in the EU’s migration management paradigm. As long as this paradigm persists, the dynamics the weaponized raft encapsulates are likely to reproduce themselves at the EU’s external borders.
What can the appropriation of lifesaving equipment for border violence teach us about the law pertaining to European migration policy specifically and the general aspiration to quell violence by appeals to law? Looking at the migration management policies at Europe’s external Aegean border, we examine how and why these infrastructures of protection come to function as technologies of border violence. Footnote 7 This allows for a critique of the policy framework of migration management through which human rights have been ‘mainstreamed’ in the context of European migration policy. Footnote 8 According to our critique, at the heart of this policy framework is a self-contradiction: an insistence on a technical and legal understanding of human safety, decoupled from a moral commitment to the real-life security of migrant individuals.
Migration management, as explained by lawyers and social scientists, is a problem-solving orientation to migration policy which seeks to be politically neutral. Footnote 9 By focusing on technical fixes such as equipment, training, standard operating procedures, and internal review, it seeks to eschew moral questions. ‘Management’ or ‘managerialism’ serve as an allusion to corporate governance. Across a large swath of policy questions, the EU has been considered, at least since the 1990s, as a quintessential example of managerialism, both by its proponents and by its detractors. EU managerialism extends to fiscal and monetary policy, environmental planning, and national security. Managerialism is arguably the dominant paradigm for migration policy throughout a wide political centre. Footnote 10 It may be advanced by both welfarist and neoliberal parties. This article seeks to join several (academic) exceptions to that general rule Footnote 11 by showing how puzzling the material results can be when human rights are set into a managerial framework, where they are a box to tick in the implementation of policy. An undercurrent impetus to circumvent human rights prohibitions leads to displacing violence in unexpected ways, but not necessarily to its reduction. Specifically, since March 2020, Footnote 12 border violence in the Aegean has been shoehorned into the legal commitment to ‘rescue lives at sea’. Footnote 13
We argue that this weaponization of rescue equipment is not incidental. The life raft-turned-weapon of torture functions as an interpretive key, demonstrating outcomes of a de-politicized, managerial approach to rescue. It embodies the ways objects reflect the social and political world in which they operate. Footnote 14 This analysis highlights the tension between the presiding paradigm of border management in the European Union and the legally binding human rights commitments of its member states. Footnote 15 In other words, the weaponized life raft challenges the assumptions underpinning European ‘migration management’: the idea that technocratic solutions can fix structural injustices, or that such ‘neutral’ solutions can ensure human rights compliance. Footnote 16 Our case study suggests that in the context of contemporary migration, managerialism and human rights protections may be incompatible.
The article consists of five sections. Section 2 offers a legal assessment of the ‘pushback’ practices employed by Greek authorities since March 2020, considering international and European law. Footnote 17 Section 3 contextualizes these maritime expulsions within a recent history in which rescue has served as a pretext for pushbacks. We attempt to explain this phenomenon by illustrating the relevant legal framework and policy context: the entanglement of interception and rescue, and international law’s delineation therein. The phenomenon of weaponized rescue is explored through Violeta Moreno-Lax’s ‘rescue-through-interdiction/rescue-without-protection’ paradigm. Footnote 18 This paradigm was crystalized in recent decades through the EU’s general approach to migration management, and the specific operations of the EU Border and Coast Guard Agency, FRONTEX. Footnote 19 The EU’s interventions in the Greek response to migration have often exemplified this managerial approach. Section 4 returns to the instruments of the current violence unfolding on the Aegean to examine what these objects reveal about their legal environment. We explore the approach and discourse through which Greece has pursued its duty to render assistance at sea under international and EU law. This reveals the lessons that the appropriation of lifesaving equipment for border violence can teach us about the relationship between law, objects, and migration management. Section 5 concludes.
2. Systematic ‘drift backs’ in the Aegean Sea
In the shadow of the coronavirus pandemic, Greek authorities established a novel deportation policy. Since March 2020, migrants have been found drifting in orange, tent-like inflatable life rafts without motors or propellers and which cannot be steered. Footnote 20 The Turkish Coast Guard reported these apparitions, but Greek authorities have provided neither explanation nor documentation. Apart from its bizarre and dramatic visual aspect, Turkish propaganda efforts surely contributed to the practice’s quick emergence as an icon of human rights violations in the Aegean. Footnote 21 Below, we further analyse the conditions that made this practice possible. But first, its basic violation of fundamental tenets of international law must be clarified.
The ‘drift backs’, whereby asylum seekers are left to drift towards Turkish waters in life rafts, can be classified into two types: expulsions which take place following interception at sea and those which occur when asylum seekers have already made landfall on Greek islands. Footnote 22 In cases of expulsions following interception at sea, physical force and violence are typically used to interdict dinghies carrying asylum-seekers. Various tactics are employed to damage the motors of dinghies and to render them immobile. Asylum seekers are subsequently transferred to Hellenic Coast Guard vessels, where survivors described being beaten. Footnote 23 Having their personal belongings, including mobile phones and identification documents confiscated, Footnote 24 they are then forcibly transferred onto inflatable, motorless life rafts, dragged out to sea and left to drift. Footnote 25
Expulsions of asylum-seekers following landfall on Greek islands differ more substantially in terms of the precise circumstances of each case. There is, nevertheless, a clearly identifiable fact pattern spanning a large geographical area throughout the Aegean islands and involving various state actors. On the Greek islands, newly arrived asylum-seekers have been arbitrarily detained in unofficial detention sites in inhuman and degrading conditions. Footnote 26 Reports of deprivation of adequate food, water, medical assistance, and denial of access to legal assistance or information, prior to expulsions, are routine. Footnote 27 Importantly, the covert nature of these practices means that no orders of detention or expulsion are provided. Footnote 28 Asylum seekers are prevented from challenging their detention or expulsion in court and denied access to an effective remedy. Footnote 29
From the perspectives of international as well as European refugee and human rights law, these expulsions by Greek officials represent violations of the principle of non-refoulement. Footnote 30 The underlying reasoning for this remains firm irrespective of whether asylum seekers land on Greek islands or are intercepted at sea. Footnote 31 European human rights law sets out specific guarantees and procedures to safeguard against refoulement. EU member states are obliged to provide asylum seekers with access to procedures for the determination of refugee status and other categories of international protection. Footnote 32 Deportations, or ‘returns’ as they are referred to in EU law, must be preceded by individualized assessments to ensure the person deported will not be exposed to the risk of direct or indirect refoulement. Footnote 33
These obligations also apply in the context of returns to ‘third countries’ to safeguard against chain refoulement, even where such countries are recognized as ‘safe third countries’ pursuant to the Asylum Procedures Directive. In such cases, the Grand Chamber of the European Court of Human Rights has affirmed, that it is incumbent on the domestic authorities to assess ‘whether or not the individual will have access to an adequate asylum procedure in the receiving third country’ as well as the risk ‘of being subjected to treatment contrary to Article 3’. Footnote 34 The Court stressed that an expelling state’s failure to assess the risks of treatment contrary to Article 3 results in a violation of the state’s procedural obligations. Footnote 35
Accordingly, where asylum seekers are intercepted in the territorial waters of a Member State, the responsibility of that state is engaged and there is an obligation under EU and domestic law to provide access to asylum procedures. Footnote 36 In such cases, the Asylum Procedures Directive envisages that asylum seekers ‘should be disembarked on land and have their applications examined’ in accordance with EU and international law. Footnote 37 The human rights obligations, including extraterritorial obligations, triggered by interception at sea, have been clearly defined by the ruling of the European Court of Human Rights in Hirsi Jamaa and others v. Italy. Footnote 38 These obligations continue to be applicable, even after the asylum seekers are abandoned in the rafts, as they remain dependent on Greek authorities for their survival. Footnote 39
Given their summary nature, ‘drift backs’ violate the prohibition of collective expulsions. Footnote 40 This prohibition flows from the procedural guarantees necessitated by the principle of non-refoulement. Footnote 41 This applies irrespective of ‘refugee status or intention or ability to claim asylum in the country concerned or in a transit country’. Footnote 42 Several attempts to challenge ‘pushback policies’ at Europe’s external borders, however, have demonstrated the erosion of this protection in the European legal imagination. Judgments including Khlaifia v. Italy and N.D and NT v. Spain, ‘[shift] the burden of proof to individual asylum-seekers to assert their rights to non-refoulement or other international protection’, Footnote 43 demonstrating a certain willingness to defer to states’ prerogatives and tolerate exceptions to this rule. Footnote 44
Additional exceptions were generated in N.D and N.T v. Spain, where the applicants challenged ‘a systematic policy of removing migrants without prior identification’. Footnote 45 The Grand Chamber found that the ‘pushback’ the applicants were subjected to did not violate Article 4, Protocol No. 4. This conclusion was based on the creation of ‘a new exception to the prohibition on collective expulsion’, Footnote 46 where the absence of an individual decision ‘can be attributed to the applicant’s own conduct’. Footnote 47 This was derived from some, albeit unsettled, case law of the Court, according to which the applicant’s ‘lack of cooperation with the available procedure for conducting an individual examination of the applicants “circumstances”, excludes State responsibility for the absence of such assessment’. Footnote 48 By extending this principle, the applicants were considered culpable because of their unauthorized entry into Spanish territory, despite the fact that legal routes identified by the Spanish government were neither accessible nor effective in practice. Footnote 49
In the 2021 judgment of M.H and Others v. Croatia, the Croatian government argued that the applicants, who had been forcibly returned without an individualized assessment, ‘had engaged in “culpable conduct” by circumventing legal procedures that existed for entry into Croatia’. Footnote 50 The argument was not upheld and instead, a violation of Article 4 of Protocol No. 4 was found. However, the Croatian government’s counterargument was not dismissed on account of principle, but rather due to the Court’s inability to determine whether ‘genuine and effective’ access to procedures for legal entry had been provided. Concepts provided by EU law have long been operationalized toward this same aim, paving the way for punitive policies that narrow access to asylum procedures and legitimize violent modes of deterrence. At Greece’s maritime border, where protection seekers transit through Turkey, a designated ‘Safe Third Country’, Footnote 51 irregular migration, even for the purposes of claiming international protection, has already been politically and discursively framed as a form of ‘culpable’ conduct. Footnote 52 Whether this will evolve to formally provide legal grounds to legitimize ‘pushbacks’, however, remains to be seen: eight cases regarding maritime pushbacks by Greek authorities were communicated by the European Court of Human Rights in December 2021. Footnote 53
The ‘drift backs’ are more than a means of pushing asylum seekers back from the border of Europe. The practice of abandoning people at sea in inflatable life rafts violates the prohibition on torture and inhuman and degrading treatment and the right to life. Footnote 54 The right to life, as reflected in Article 6 of the International Convention on Civil and Political Rights (ICCPR) and Article 2 of the European Convention on Human Rights (ECHR), provides that states not only refrain from actions that pose risk to human life; it requires them to take active measures to prevent the loss of life. Footnote 55 In the context of border surveillance operations, such as Operation Poseidon in the Aegean Sea, these obligations may extend to the provision of rescue services. Footnote 56 Instead, life-threatening situations are actively generated as a result of Greek bordering practices, at the core of which is an overall sense of indifference as to whether the migrants being expelled live or die. Footnote 57 The Special Rapporteur on extrajudicial, summary or arbitrary execution emphasized that in the context of ‘pushbacks’ at sea, ‘[t]he State bears responsibility under international law for any deaths or injuries that may occur in these circumstances’, and that ‘State officers themselves may be potentially culpable for murder’. Footnote 58
The practices fit squarely within the torture definition under international and European law and contain the integral elements of torture derived from the jurisprudence of the ECtHR. Footnote 59 The European Commission of Human Rights, the predecessor of the ECtHR, held that ‘all torture must be inhuman and degrading treatment’. Footnote 60 From this perspective, inhuman and degrading treatment are both subsumed within torture. In a 1967 case, the Commission referred to the definition of torture, noting that ‘[t]he word “torture” is often used to describe inhuman treatment, which has a purpose, such as the obtaining of information or confessions, or the infliction of punishment, and it is generally an aggravated form of inhuman treatment’. Footnote 61
In the context of ‘drift backs’, the treatment is premeditated, as evidenced by the numerous reports documenting the same tactics employed by Greek officials across several Aegean islands and territorial waters. It is ‘applied for hours at a stretch’ in many recorded incidents and causes ‘actual bodily injury’ and ‘intense physical and mental suffering’; Footnote 62 falling clearly within the settled definition of ‘inhuman’ treatment established by the ECtHR. Respondents describe life-threatening situations during and resulting from their encounters with the Greek Coast Guard. The degrading techniques, therefore, arouse ‘feelings of fear, anguish and inferiority capable of humiliating and debasing [them] and possibly breaking their physical or moral resistance’. Footnote 63 Finally, the deliberate and intimidating nature of the treatment, as required to meet the threshold of the torture definition, is further evidenced by its connection with a systematic asylum seeker ‘deterrence’ policy. Footnote 64 This policy is employed to intimidate and coerce, including by means of physical force, for the specific purpose of expelling asylum seekers and demonstrating to future asylum seekers that they should not attempt to exercise their rights and seek international protection in Greece. This is reflective of the broader deterrence strategy pursued by Greece, particularly since March 2020. Footnote 65
This deterrence strategy is comprised of both a practical aspect, including the use of physical force to keep out asylum seekers, and a rhetorical component. Since the implementation of the 2016 EU-Turkey Statement, the Turkish Coast Guard has been tasked with containing refugees and migrants within its territory. Footnote 66 In late February 2016, however, Turkey announced it would no longer prevent migrants and refugees from leaving ‘irregularly’. Following a meeting of the National Security Council (KYSE), the Greek government reinforced its border forces ‘to prevent illegal entry into the country’ and asylum procedures were suspended for one month. Since then, collective expulsions – a longstanding feature of this border – became increasingly visible and systematic. The abandonment of asylum seekers in rafts is now an established modality in the Greek government’s de facto policy of ‘pushing back’ migrants from its territory. And yet, the reality of violence and illegality in the Aegean continues to be distorted by the Greek government.
The government’s strategy of denial is central to reinforcing the situation of de facto rightlessness generated by these practices, which attempt to ‘place [asylum seekers] outside the protection of the law’. Footnote 67 Indeed, if asylum seekers were not documenting their own presence, there would be no way of knowing that these systematic expulsions from Greek territory are taking place. Footnote 68 Arrivals are omitted from official records. Footnote 69 Against this backdrop, and particularly given the ‘abduction’ of asylum seekers from Greek islands and other forms of deprivation of liberty experienced by victims, ‘drift backs’ risk violating the prohibition on enforced disappearance. Footnote 70
This strategy of denial is reinforced by the construction of a public narrative propagated by the Greek government and Frontex, which focuses on an interpretation of ‘interception’ at sea that distorts the factual pattern of the Hellenic Coast Guard’s modus operandi, and Greece’s resulting obligations under international human rights law and refugee law. Footnote 71 This is consistent with a particular conception of human rights obligations in maritime border spaces advanced through the structure and implementation of EU migration policy for decades.
3. Rescue as interdiction in law and infrastructure
Yet, beyond human rights law, ‘drift backs’ also raise concerns relating to the international law of the sea. Examining these acts from an international law of the sea lens reflects how these violations of human rights law are also enabled and constructed by a certain managerial approach to the application of the law. When set in a managerial context, the law of the sea may be a conduit for human rights violations. Indeed, as will become clear below, the law of the sea has served the managerial legal vocabulary: it helps reduce the politically laden vocabulary of ‘asylum’ to ostensibly less political elements such as ‘innocent passage’ and ‘saving life’. Footnote 72 This contradictory relationship with the international legal environment and rhetoric is what makes ‘drift backs’ noteworthy from the perspective of the critique of managerialism.
The imperative of protecting life at sea is advanced through the duty to render assistance to vessels in distress. It is a recognized norm of customary international law and is codified in numerous modern international treaties, foremost of which are the United Nations Convention on the Law of the Sea [UNCLOS] Footnote 73 and the International Convention for the Safety of Life at Sea [SOLAS]. The international legal regime governing all maritime activities is centred primarily around the UNCLOS, which delineates different maritime zones and defines the types of enforcement (and other) measures states can take accordingly. SOLAS establishes minimum safety standards for the construction, equipment, and operation of ships. Together, the various international instruments seek to reinforce the legal obligation of rescue.
The UNCLOS Footnote 74 incorporates the duty to render assistance at sea in two, interconnected provisions: Article 98 and 19. Article 98 requires states to oblige masters of ships flying their flag to ‘render assistance’ to persons in danger at sea and to proceed as soon as possible to a vessel in need of assistance if informed of this need, insofar as may be reasonably expected of the master. Footnote 75 This obligation applies to ‘any person in danger of being lost’, ‘persons in distress’ as well as to the crew and passengers of ships in the context of a collision. Footnote 76
This norm, and the international law of the sea more broadly, predates the international instruments in which it is enshrined. It emerged through inter-state conduct and evolved into customary international law, primarily as a means of regulating and furthering trade. The maritime customs the law is based on are rooted in an understanding of the sea that has changed over time. Footnote 77 As a result of contemporary developments in ship manufacturing and standardized safety procedures, however, the duty to render assistance at sea no longer plays the role it once did. Numerous international legal regulations have since developed to make travel safer. Footnote 78 These legal regulations form one aspect of what Spijkerboer conceptualizes as the ‘Global Mobility Infrastructure’, that is, the ‘physical structures, services, laws that enable some people to move across the globe with high speed, low risk, and at low cost’. Footnote 79
The duty to render assistance at sea, however, is universal. Footnote 80 Unlike other aspects of international maritime law, ‘the international legal obligations concerning maritime search and rescue cover both regular and irregularised travellers’. Footnote 81 The obligations imposed by the duty to render assistance at sea are now, therefore, most relevant and potentially transformative for unauthorized maritime migrants, including those seeking international protection in the EU, populations who are otherwise denied access to the ‘global mobility infrastructure’. Footnote 82
Under UNCLOS, the maritime space extending 12 nautical miles from the coast is designated as the territorial sea (UNCLOS Articles 2 and 3). Articles 27 and 28 provide coastal states with prescriptive and enforcement jurisdiction over vessels and individuals within this zone. Footnote 83 UNCLOS distinguishes between ‘innocent’ and ‘non-innocent’ passage through territorial seas. Footnote 84 Foreign flagged vessels are permitted where such transit does not ‘prejudice the peace, good order or security of the state’. Footnote 85 The question of whether the ‘unloading’ of persons in violation of national migration laws is considered under Article 19 UNCLOS as ‘an activity in violation of innocent passage’ depends on the extent to which unauthorized migration is framed as a crime or security threat.
Within the current paradigm of border securitization, the ‘unloading’ of unauthorized migrants, people smuggling, is typically interpreted as precluding innocent passage. Footnote 86 This interpretive move is perhaps the first application of the international law of the sea which makes the latter useful for a managerial approach aiming to stem migration, including the passage of asylum seekers. Under such an interpretation, closing borders to certain asylum seekers may be viewed as an objective application of the law rather than a violent act. Hence the ‘civilized’ veneer that international legal vocabulary may grant to border violence. Accordingly, the vessel may be intercepted and subject to certain enforcement measures. Footnote 87 Article 25 UNCLOS, for example, provides the rights of protection of the coastal state and provides that ‘[t]he coastal State may take the necessary steps in its territorial sea to prevent passage which is not innocent’. Footnote 88 The Migrant Smuggling Protocol Footnote 89 provides further basis for the interception of stateless vessels. The Protocol entitles states to ‘board and search’ a stateless vessel if there is reasonable grounds to suspect people smuggling. Footnote 90 While the Protocol’s mandate to ‘prevent and suppress the smuggling of migrants by sea’ might appear as a green light to thwart unauthorized migration entirely, Article 8(7) expressly states that the measures must be in accordance with ‘relevant domestic and international law’. Footnote 91 And yet, this interpretive territory is obscure and has allowed the utilization of the law of the sea in order to counteract, in ‘objective’ and seemingly non-political terms, obligations towards non-citizens who may be in dire need; even such need that is recognized under other branches of international law – notably refugee law – requires a (temporary) permission of entry. Part of the underlying confusion stems from how international human rights law enshrines a right of exit, but not a right of entry. Footnote 92
This use of the international law of the sea to sanitize what might otherwise be perceived as a form of violence is echoed in certain aspects of EU law. Frontex Regulation 656, which governs ‘sea surveillance’, authorizes participating units to take certain measures to acquire information on the vessel and its passengers, ‘where there are reasonable grounds to suspect that a vessel may be carrying persons intending to circumvent checks at border crossing points or is engaged in the smuggling of migrants by sea’. Where ‘evidence confirming that suspicion is found’, several enforcement measures, such as seizing the vessel, apprehending those on board, and ordering the vessel to alter its course, may be carried out though they too must be in full compliance with international legal obligations. Footnote 93
While it is still unclear how a managerial orientation toward the international law of the sea relates to a predatory use of rescue equipment, a basic point might already be apparent; one that Moreno-Lax and others have illustrated: that when it comes to pushbacks, refugee law and the international law of the sea have militated towards different directions, ambivalently proscribing, and permitting, patterns of border violence. Footnote 94 The two bodies of law effectively transmit the debate on refoulement from a bright-line to an interpretive disagreement. The veneer of legality is no longer the rhetorical privilege only of one side of the debate. Footnote 95
Vessels in distress represent an important exception to rules distinguishing innocent and non-innocent passage under customary international law. Article 18(2) UNCLOS provides that coastal states ‘permit ships in distress or under force majeure access to its territorial sea and ports’. On the other hand, the status of vessels in distress can also work in the reverse, as authorities can intervene to rescue vessels in distress, regardless of whether they are located in their own territorial waters and may even be obliged to perform rescue outside of their legally defined Search and Rescue (SAR) zone. Footnote 96 This further demonstrates why rescue is significant in authorizing and facilitating operational flexibility for border and migration ‘management’ policies.
In short, rescue has emerged as an imperative in efforts to ‘manage’ migration. From the Global Approach to Migration  Footnote 97 to the ‘New Pact on Migration and Asylum’ , Footnote 98 EU migration policy has implicitly relied on the obligation to save lives at sea to legitimize closed borders. Moreno-Lax conceptualizes this dynamic as the ‘rescue-through-interdiction/rescue-without-protection’ paradigm. Footnote 99 By framing physical acts of border enforcement, such as interception at sea, as acts of rescue, they are interpreted as detached from the international legal obligations that flow from the exercise of sovereignty within territorial waters, or the effective control exercised in the case of those acting beyond their territorial seas. Footnote 100 This model shows how interdiction is ‘re-defined into a life-saving device’. Footnote 101 The EU and its member states have, thus, actively and strategically forged gaps in refugee and human rights protection. Footnote 102
Europe is not the only example, since European policies are arguably based on American and Australian precedents. Footnote 103 Policy-makers across the Global North rely on the rescue duty as a basis for intercepting migrants at sea where legal grounds for such action may otherwise be weak or absent. At the same time, they often attempt to erase the jurisdictional link from which obligations stemming from human rights law and refugee law follow. Footnote 104 These earlier precedents also embody the paradigm captured by Moreno-Lax: the goal is to perform ‘rescue’ without being bound to the full range of human rights obligations, or even to a core of non-refoulement obligations. Footnote 105 Importantly, the debate’s transformation into one about ‘legal fragmentation’ Footnote 106 helps cabin it in a realm of expertise; Footnote 107 ordinary people are thus invited to suspend their own moral or political judgement (whether it is anti- or pro-protective migration policies). This, of course, is illusory. What expertise offers here is nothing but interpretive disagreement; and a (real or forged) ‘regime collision’. Footnote 108
Specific factors relating to the structure of the international law of the sea, at times relating to the text of the law itself, enabled this ‘humanitarian’ duty to be interpreted for opposite ends. For one, neither interception nor distress is clearly defined in international law. The lack of universal definitions of both has been frequently exploited by states. In 2000, the UNHCR proposed defining interception as:
all measures applied by a State, outside its national territory, in order to prevent, interrupt, or stop the movement of persons without the required documentation crossing international borders by land, air or sea, and making their way to the country of prospective destination. Footnote 109
The same UNHCR ExCom conclusion further affirms in the preamble that ‘when vessels respond to persons in distress at sea, they are not engaged in interception’. Footnote 110 As one commentator observes, ‘this distinction recognizes the predominantly humanitarian character of rescue in contrast to the migration control policy objectives that underpin interception practices’. Footnote 111 In practice, however, this distinction is easily exploited and undermined. Since 2006, ‘the practice whereby several States were classifying some interception measures as rescue at sea operations, in order to use SAR operational capacity for such activities’ has been observed. Footnote 112 These are ‘cloaked interceptions’. Footnote 113
Frontex Regulation 656 seemingly provides more certainty around the concept of ‘distress’. Footnote 114 It outlines three phases of uncertainty, alert, and distress, and includes factors considered in the evaluation of whether a boat is in distress. For example, ‘the existence of a request for assistance; the seaworthiness of the vessel; overcrowding; the presence of supplies; the presence of qualified crew; weather conditions and the needs of those on board’. Footnote 115 These are all factors commonly present with asylum seeker dinghies. However, in addition to other factors, the definition’s different prongs allow the issue of ‘distress’ to be continually contested. The different phases arguably create a strategic ambiguity that allows Frontex to initiate operations framed as rescue in moments that are beneficial in an overall policy of closing borders and not of asylum. Footnote 116
The question of how a managerialist orientation towards the international law of the sea contributes to the use of rescue for violent deterrence is not one answered in directly causal terms. However, the International Convention on Maritime Search and Rescue (SAR Convention) (1979), which Greece has ratified, is instructive. Under Article 2.4, the SAR Convention requires parties to designate ‘rescue units’ tasked with implementing rescue obligations in their respected ‘search and rescue zones’. These rescue units must be ‘suitably equipped and located’ (Article 188.8.131.52) and ‘each rescue unit should be provided with facilities and equipment appropriate to the task’. The provisions are part of a larger legal environment, including the Safety of Life at Sea Convention (SOLAS) in which multilateral treaties set the technical minutiae of maritime safety.
In 2009, when Frontex launched its first ‘rapid deployment’ in Greece, the surrounding rhetoric was one of providing competencies: ‘technical assistance’ and operational equipment. This infrastructural aspect of the EU’s involvement has remained salient. In 2015, Greece experienced an upsurge of migrant entries, later to be labelled ‘the refugee crisis’. As many entered through maritime routes, the lack of equipment was seen by some humanitarian agencies as a significant obstacle to migrant safety. In a May 2015 report, the International Federation of Red Cross and Red Crescent Societies (IFRC) recommended the purchasing of search and rescue equipment such as helmets and spine boards. Footnote 117 In June 2015, the EU’s Internal Security Fund provided the Hellenic Coast Guard with €480,000 in ‘[e]mergency assistance for the procurement of Search and Rescue Equipment to avert losses of migrant’s life at sea’ and €2.2 million for ‘emergency assistance covering staff related costs in order to ensure a high-level domain awareness of the severely affected Eastern Aegean EU external borders and to minimize the losses of human lives at sea’. Footnote 118 The Hellenic Coast Guard received a further €2.89 million in October 2015. In December, Greece proceeded to request ‘items such as tents, generators, beds, sanitary equipment and emergency first aid kits’. As EU authorities explained, ‘[t]his voluntary delivery of aid is coordinated by the European Commission’s Emergency Response Coordination Centre (ERCC) which is working closely with the Greek authorities and the other participating states in the Mechanism for a swift response to the request’. Footnote 119 The December request for relief coincided with further Greek agreements and requests for enhanced border enforcement and Frontex presence in the Aegean islands. Humanitarian and policing assistance have since been constantly intertwined. Footnote 120
This moment of considerable danger caught Greek and EU authorities ill-prepared and in violation of the SAR convention’s requirement to be ‘suitably equipped’. In a 2020 film investigating a rescue incident carried out on the island of Lesvos on 28 October 2015, that resulted in 43 deaths, the investigation agency Forensic Architecture finds that the Hellenic Coast Guard had received CPR training from non-governmental volunteers, but not as part of the official EU training. More generally, the Greek authorities were ill-equipped: ‘the EU’s long-term policy of repelling migrants left the local coastguard and Frontex crews under prepared and under-equipped for rescue, leading to this tragic loss of life, only three kilometres from European shores’. Footnote 121
In an application of the SAR convention requirements and with EU support, the Greek government attempted to amend potential violations of the SAR convention. According to the government’s procurement records, the life-rafts that were later used for ‘drift backs’ were purchased for the Greek navy in 2017. Footnote 122 Government procurement records also show that the Ministry of National Defense, Ministry of Interior and Administrative Reconstruction and Ministry of Shipping and Island Policy have contracts with LALIZAS. Footnote 123 The life rafts themselves have likely been made available to the Hellenic Coast Guard for ‘suitably equipped’ rescue operations.
The Greek ‘drift backs’ should therefore be understood as a direct continuation of a history in which rescue has served as a pretext for pushbacks and where the purported need to control borders has served to obscure the right to seek asylum. The starting point can be traced to the United States’ interdiction of Haitian refugees during the 1980s and 1990s which involved the interception and systematic summary expulsion and offshore detention, of Haitians fleeing by boat. Footnote 124 While Moreno-Lax, Daniel Ghezelbash, Natalie Klein, and Daniel Opeskin have exposed and examined the global trajectory of this interpretive move surrounding maritime rescue authorities, Footnote 125 the analysis we provide is different. It shines a light not only on legality but also on its ‘material’ aspects.
This material incarnation of the weaponization of rescue is not exclusively Greek. Indeed, the Greek use of life rafts harks back to an Australian precedent. In 2013, Australia introduced a dozen bright-orange, windowless life vessels shaped like oblong dirigibles. Footnote 126 The vessels were geared with lifejackets and touted as ‘unsinkable’, seemingly highlighting concern for preserving (if not exactly saving) life. The vessels were equipped with navigational systems, air conditioning, an engine, and with ‘just enough fuel’ to reach Indonesia. Footnote 127 This equipment was subsequently replaced by Vietnamese-style fishing vessels, also tasked with pushing back those seeking international protection. Footnote 128 But like the Greek rafts, the orange vessel episode embodied the often-hypocritical moral stance of liberal democracies regarding strangers in need: a willingness to engage in extreme measures of violence to enforce borders, coupled with an emphasis on efficiency.
The Australian lifeboat most closely resembles a missile and its image conveys the omnipotence of a regional superpower. Compared to the Australian raft’s mechanical cruelty, the Greek tent-like raft is a poignant symbol of inhumanity. No fuel is rationed to reach a destination and the expectation is that the life raft will simply drift across the relatively narrow waterway.
Each country’s trajectory of managerial thinking is apparent in the public framing of its infrastructure. On 20 November 2020, Greek Member of Parliament Kyriakos Velopoulos, leader of the right-wing Greek Solution party, appeared on the Greek state-owned public broadcaster, ERT. Holding photos of Australia’s oblong orange vessels, he explained: ‘This here … is a raft made by the Australian government … with food, actual food, and it never sinks.’ An interviewer gasped: ‘There’s a humanitarian aspect to it!’. In the Greek and Australian examples, the tenets of rescue are weaponized.
Life rafts are thus one piece of a history in which humanitarian infrastructures degrade and dominate refugee and migrant populations. Another example is the involuntary use of painkillers to sedate migrant children in the US, which illustrates another protective material repurposed for an attack upon migrants and upon basic human rights principles. Footnote 129 The material and infrastructural aspect of ‘drift backs’, the violent transformation of equipment designed to save life at sea, further reveals the dire failure of managerial fixes for a deep moral and political dilemma. These examples reflect a wider phenomenon, characteristic of contemporary violence, that others have highlighted: when protective materials are co-opted as technologies of (migration) governance, they no longer serve to mitigate violence but become the means by which it is exercised. Footnote 130
4. Against migration managerialism
For its proponents, migration management denotes a civilized, rational way to deal with the policy dilemmas caused by unauthorized migration. In this vein, migration policy is about transforming migration from a crisis, whether political or humanitarian, into something ‘knowable’ and therefore ‘governable’. Footnote 131 At the core of the view is the discursive framing of migration as ‘something normal, something that has always been there, always will be here. It’s something that we can manage together’. Footnote 132 Importantly, migration governance is often couched in a civilizing discourse directed not only at migrants but also at transit and potentially host states. Greece is a long-time recipient of ‘support’ and ‘assistance’ to enable it to engage in migration and border management in line with EU standards. Footnote 133 This not only consists of money allocated to the majority of EU member states through funds such as the Asylum Migration and Integration Fund (AMIF) or the Internal Security Fund (ISF). It also includes emergency support, often in the form of financial or in-kind donations. Footnote 134 But within the migration governance discourse, such assistance is not only designed to help in migration policy per se, Footnote 135 but it also promises to nudge its subjects towards economic development and moral enlightenment. Footnote 136
As former EU Commissioner Jean-Claude Juncker explained at the height of the ‘refugee crisis’ in 2015, ‘migration must change from a problem to be tackled to a well-managed resource’. Footnote 137 He further contrasted ‘migration management’ with instances of refoulement and of private xenophobic violence, particularly ‘pushing back boats from piers’ and ‘setting fire to refugee camps’. Footnote 138 This framing suggests an underlying dichotomy between management (which comes from Brussels) and violence (in the Greek-European periphery); between top-down rational planning and bottom-up racism.
Migration management relies on the depoliticization of migration. To make migration beneficial at an EU level, ‘a well-managed system’ is required. This, of course, relies on political co-operation among EU member states and local communities in places like the Aegean islands. For such co-operation to be possible, migration is depersonified. Footnote 139 In the words of Martin Geiger and Antoine Pécoud, ‘[p]olitical issues are reduced to technocratic ones’. Footnote 140 The language of management presents itself as an objective language. Recently, Ursula von der Leyen, President of the EU Commission, repeated the familiar point: ‘migration has always been a fact for Europe, and it always will be’. Footnote 141 With the right systems and incentives in place, it suggests migration offers potential benefits that extend to receiving states and their citizens (and potentially also to at least some migrants).
Yet for some scholars, the notion of migration management is the subject of trenchant critique. The rhetoric of expertise in international law has in recent years been subjected to important criticism. The argument, generally, is that expertise disguises political preferences and moral judgements by couching them in technical vocabularies inaccessible to many relevant parties and democratic constituencies. Footnote 142 More specifically, scholars seeking to chart the global migration management project have drawn attention to migration management as a self-legitimizing discourse, ‘a notion that is mobilized by actors to conceptualize and justify their increasing interventions in the migration field’. Footnote 143 What may at times be drastically divergent policy measures, such as migrant rescue and detention, are cast under one label: ‘actors develop discourses to justify their existence and legitimize their practices; yet their actual activities and policy interventions often differ substantially from the rhetoric underpinning them’. Footnote 144
This is not to suggest that acts of refoulement committed by Greek border guards, including ‘drift backs’, have been ‘managerial’. As reflected in Juncker’s words above, managerialism and migration management, indeed, define themselves against such acts. However, ‘drift backs’ and the displacement of violence onto the infrastructure of rescue, is an indicative by-product of European managerialism. To understand this, it is necessary to move away from the dichotomy between rational-bureaucratic ordering and violence. ‘Drift backs’ have not been made possible merely due to the presence of ‘bad actors’ among the Greek personnel. The key to understanding such developments is, rather, a certain relationship between managerialism and the material conditions in which it is exercised.
At least since Norbert Elias’s classical work The Civilizing Process (1939), social theorists have observed the entanglement of political organization with violence and instruments we use in our everyday. Footnote 145 Elias observed the transformation of the knife from a weapon to an object of social decorum around the table. Footnote 146 The fundamental insight is that things, and indeed entire material infrastructures, have the tendency to follow and reflect even latent political tenets of the societies that employ them. Moreover, Elias believes the civilizing process is one in which overt violence is continuously concealed. In our own time, Sheila Jasanoff highlights the relationship between objects and their environment in power politics. She writes:
many nonfictional accounts of how technology develops still treat the material apart from the social, as if the design of tools and machines, cars and computers, pharmaceutical drugs and nuclear weapons were not in constant interplay with the social arrangements that inspire and sustain their production. Footnote 147
In her terms, the repurposed rescue raft can be considered as a kind of technology.
In recent years, international law scholars have taken up the challenge and rethought the material aspects of their studies. Footnote 148 Spijkerboer’s conceptualization of the global mobility infrastructure, discussed above, has led the way. Footnote 149 Moria Paz’s contributions on border walls revealed how human rights law’s transnational effect and universal aspiration have come alongside the erection of multiple cement and barbed wire barriers. Footnote 150 Like the repurposed raft, the wall is also a technology of border management and both represent low-tech solutions of migration management. Such solutions are uniquely tactile, giving us the dubious benefit of immediately visualizing and imagining the texture of the relevant materials. That is perhaps central to their role in the project of migration deterrence. Footnote 151
On the other end of the spectrum, legal scholars interested in materiality documented the de-materialization of borders with a rich literature on how surveillance technologies often replace physical barriers. Here, the border becomes as ephemeral as it is all-pervasive. As an example, Ayelet Shachar discussed how ‘government surveillance of movement and mobility – traditionally restricted to the actual location of the border crossing – is now seeping into the interior’. Footnote 152 Focusing specifically on the discriminatory aspect of ‘digital racial borders’, Tendayi Achiume exemplifies how border technologies mask racial distinctions ‘in the cloak of presumed neutrality that attaches so strongly to technology in the popular and policy imaginary’. Footnote 153 Fleur Johns documented how detection and tracking equipment gathering biometric data, such as iris recognition, has become a condition for the cross-border movement of returning Afghan refugees. As she discusses, the technology became enshrined in a treaty through a tripartite agreement between Pakistan, Afghanistan, and the UNHCR. Footnote 154 Petra Molnar argued that a managerial orientation to migration has advanced experimentation with AI as a policy tool on migrant populations. Footnote 155 And finally, Dimitri Van Der Meerssche has conceptualized the ‘virtual border’, a system:
scattered across digital systems without fixed territorial coordinates … [that] operates as a central site of data extraction and social sorting … a system of discrimination and division where the standards of hierarchy or inclusion … are continuously kept in play. Footnote 156
Through this prism, emphasizing the co-production of infrastructures and their political environments, repurposed rescue rafts illuminate the fundamental confusion managerialism generates between persecution and protection. Footnote 157 Yet our case study reveals a different dynamic from what is described in parts of the literature, where infrastructures function as constitutive foundations for political participation or its prevention. Footnote 158 Taken as a whole, one might say following Kingsbury and Maisley, that harsh border systems are constitutive infrastructures for preventing the emergence of transnational solidarities and ‘publics’. Footnote 159 Yet focusing on the rescue vessel, we see that the demand for rescue infrastructure was granted only for the infrastructure to be repurposed for violent exclusion. In other words, the rescue element of the infrastructure did not constitute new social conditions but was subsumed under a stronger social and historical force.
In this view, the repurposing of the rescue rafts amounts to an unintended admission of the telos of the political environment they are part of; beyond the manifest (benevolent) purposes that led to installing them in the first place. The use of a life-raft shaped like a tent could not be made up, and it reveals a fantasy element of a collective unconscious, beyond its confusion between border enforcement, maritime safety, and humanitarian relief (this is also reflected in the earlier Australian missile-shaped raft). Set adrift on the Aegean, a disquieting quality emerges from the fact that the tent-raft becomes a metaphor for the refugee’s condition. Footnote 160
Looking at the spectrum of legal studies on the materiality of migration, the repurposed raft stands out as a stark example of disjunction between design (rescue) and use (violence). From this perspective, it is precisely at the opposite end of its companion low-tech artefact of border control, that is, Paz’s border wall. The wall expresses the resolute and steadfast message of building for the purpose of blocking. Unlike the rescue raft, its form follows its function.
It may not be immediately clear what difference this disjunction between design and use makes: the violence of this practice would not be reduced if migrants would be forced to drift away in equipment specifically made for that purpose. And yet the fact that in our case the raft is pre-made for a benevolent purpose may allow it to be less visible to transnational audiences. It is easier for the Greek government to mask it in the cloak of maritime safety, an option that could not have been pursued were asylum seekers shot at the border. In that regard, the disjunction between design and use may offer the Greek Coast Guard some deniability already harnessed by the Greek government, namely by recording pushbacks as ‘rescue’ operations. Footnote 161
Under the Dublin-II Regulation, Greece became part of a buffer zone for asylum seekers who are prohibited to travel from their country of entry to other European destinations. The Dublin-II Regulation is indeed the foundation of ‘fortress Europe’, a system that has placed the task of guarding Europe’s borders on relatively poorer countries at the EU’s external borders. Footnote 162 Both enforcement and humanitarian contributions to Greece are part and parcel of the system established by the regulation. While Greece received rescue equipment to support lifesaving operations amidst a dramatic increase in maritime border crossers in 2015, it was also coaxed to show how asylum applications can be deterred and ultimately avoided. Footnote 163 The equipment Greece received included elements fit for the task of deterrence, such as new detention facilities and surveillance equipment partly made available by Frontex and partly given directly to the Greek government. But it also included rescue equipment, the physical embodiment of an old cosmopolitan tradition of maritime law. The latter could readily be enlisted to perform a new kind of torture. The phenomena of the ‘drift back’ can thus demonstrate what can happen to life saving equipment when it exists as part of the ‘material economy’ of deterrence. Footnote 164 As emphasized above, this material economy is also a certain orientation to governance, a social and cultural context, and a mental condition, but demonstrating these odd results undermines the viability of managerialism’s technical solutions.
As Jessie Hohmann writes,
taking the object as the primary frame of reference compels us to bring our abstract thinking, our focus on rules, doctrines and principles, down to the material level where their impacts are felt on and in communities, homes, and bodies. Footnote 165
On the level of rules, the orange life rafts simply correspond to several provisions of UNCLOS and the SAR convention, and more specifically the SOLAS convention. On the level of principles and doctrines, it refers to ideas such as ensuring a measure of state protection even for seafarers far away from home and of lending assistance to the unknown seafarer who has experienced a shipwreck. The life raft points to migration management as a process in which states and international bodies contribute equipment to other states in need. Footnote 166 However, when we step closer to the material level to follow the life raft through its real trajectory and look at its impact on bodies, a different story is revealed: one that illustrates how the international law of the sea interacts with refugee law and human rights law to demobilize their political and moral engines.
Records of pushback practices in Greece are traced at least to the 1990s. Footnote 167 Bolstering infrastructure cannot be expected to change such a course, especially when it enables Greece to fulfil its role in an EU-wide process of border externalization. Footnote 168 The New Pact on Migration and Asylum of July 2021, while framed as a fresh start, is just another iteration in a managerial cycle. The same argument applies to the construction of new camps as we write across the islands of Samos, Leros, Kos, Chios, and Lesbos. With the unfolding climate crisis affecting many regions in the Middle East and North Africa, it is difficult to see how the relationship between law and the physical infrastructure of migration management could escape this vicious circle. Increased migration leads to increased spending on management equipment which in turn comes to reflect the non-managerial underbelly: deep fears that people on the move upend the lives of those who have the privilege to remain sedentary.
Breaking the circle may require new modes of thinking about human movement, the outlines of which are still unclear. For now, we observe a negative dynamic clearly opposed to EU managerialism’s enlightened self-understanding. The latter rests on the hope that managerial support will incrementally harness forces of xenophobic violence into a rule of law framework. It thus assumes a positive dialectic in which human activity becomes cleaner and more rational. What we have seen, rather, is that managerial support ends up generating conditions for xenophobic and racist violence and is ultimately marred by such violence as well. Reversing this cycle would require much more radical thinking, in which Greece would also no longer be conceived as a subject to civilizing forces coming from Brussels or other more economically powerful centres of European policymaking. It would necessitate, at a minimum, the dramatic restructuring of the architecture of EU asylum law, which in its current form not only incentivizes but prescribes deterrence through its unequal distribution of responsibility to member states at the ‘external’ borders of the EU. This would be difficult to imagine within an international legal order without an unqualified right to mobility, and where ‘the logic of even progressive international legal theory remains that sovereign states have a right to exclude, but that certain political strangers warrant discretionary admission and inclusion’. Footnote 169
Our critique aims to discard the managerialist idea that deep moral and political problems can be solved by a few more camps or a few more rescue vessels. Migration management grows from a fear that if we open the questions of asylum and movement rights, including their associations with racial, distributive, and climate justice, the far-right will necessarily win. Footnote 170 We will remain with nothing but the violent component of closure and will choose to ignore, if not straightforwardly eliminate, the rights claims of racialized asylum seekers who are making their demands upon Europe.
In its embrace of the rule of law and a corresponding dual commitment to deterrence and protection, migration management presents itself as the ‘lesser evil’ of border restriction. Footnote 171 Mobilizing the logics of securitization and humanitarianism ‘helps to enhance the legitimacy and reputation of securitising forces’ Footnote 172 and serves to eclipse the moral and political realities of border violence.
Meanwhile, violence is simply funneled into unexpected canals such as rescue at sea. In this way, the managerialist orientation to governance has led to the ‘normalisation of abandonment’ at the external borders of the EU. Footnote 173 In Achille Mbembe’s words:
the dominant power practically abdicates governance. It abdicates any responsibility for the lives and welfare of a specific population group and yet, abandonment does not signify an end to domination. It rather has to do with the emergence of a paralysis whose function … is to ensure a systematic withdrawal of care and renunciation of obligations towards a designated population. Footnote 174
The duty to render assistance at sea has historically been one of the most concrete reflections of solidarity across the ‘international community’. It has embodied the notion of horizontal international comity in which all states contribute so that seafarers can enjoy global freedom of movement through the maritime commons. Together, they facilitate international trade and commerce. Like certain rules of humanitarian law, the duty to render assistance at sea is blind to categories such as nationality, race, and religion; it thus seems to extend universally to all members of humanity. Under international legal instruments such as the SOLAS Convention, this duty gains a material and infrastructural aspect: states are obliged to obtain certain kinds of equipment that make their SAR obligations more reliable. The duty to render assistance thus requires not only responding to emergencies but also preparing for them and allocating appropriate resources.
At the external borders of Europe, the managerialist orientation turned to law and infrastructure to elude a more open debate about moral and political commitments. But without a solid political basis, the laws and infrastructures of mutual assistance at sea do not only crumble but are also turned on their heads. The process is not only the product of the devious intentions of local actors. Material infrastructures may have the tendency to follow and reflect the underlying political commitments of the systems that bring them into being.
The EU contributes aid to Greece for asylum seeker and refugee assistance: tents, medicine, and maritime rescue equipment. It also contributes to the financing of enforcement tools including detention centres, vehicles, and state-of-the-art surveillance equipment. The message is dual: we stop refugees from entering our common space while adhering to fundamental human rights protections. But the dual task turns out to be quite difficult and the two objectives appear to be competing in a zero-sum game. Footnote 175 Despite the legal arrangements designed to deter movement, humanitarian assistance creates both legal and material conditions for mobility that bring unauthorized migrants under the protection of Greek and EU law.
EU policymakers and member states may wish to remain wedded, rhetorically, to the possibility of a ‘win-win’, but the weaponization of rescue illustrates that Brussels and Athens cannot embrace both objectives. When both sides of the hierarchical relationship choose restriction, it is but a small step until all the available material resources are recruited for the task. The role of migration management’s infrastructures of protection in generating conditions of de facto rightlessness for migrants at the external borders of the EU testifies to how violence at sea cannot be fixed merely by managerial solutions. Enforcing the international law of the sea or adding budgets for rescue equipment is not fit for the task. The real conversation lies in structural questions around what the future of borders should be and how, if at all, they might be democratized.