(PDF) Forensic Oceanography

Forensic Oceanography Tracing Violence Within and Against the Mediterranean Frontier’s Aesthetic Regime 1 Charles Heller and Lorenzo Pezzani While the emergence of the Mediterranean’s unequal mobility regime can be traced back to European imperial expansion towards the sea’s southern shores in the nine- teenth century, illegalized migration across the Mediterranean and fatalities at sea became structural and highly politicized phenomena only as of the end of the 1980s, in conjunction with the consolidation of the freedom of movement within the EU. With the Europeanization of migration policies, a truly European “color line” was institu- tionalized, as the populations who were excluded from accessing the European ter- ritory were marked out within a matrix of race and class. However, as a result of the perpetuation of the systemic conditions underpinning migrants’ movements towards Europe — in particular the need for migrant labor, global inequalities, and existing mi- grant networks — the illegalization of certain categories of migrants only resulted in their movement operating in an increasingly clandestine form, in particular by cross- ing the sea on overcrowded vessels.2 In the attempt to control the Mediterranean, now corresponding to the extremities of European space and transformed into a vast fron- tier zone, European states have deployed a vast array of bordering practices and tech- niques to contain and channel migrants’ movements. Crucially, since the early 2000s, the EU has increasingly outsourced border control to authoritarian regimes in North Africa to contain migrants on their shores. These policies have never more than temporarily succeeded in stemming migrants’ crossings, and for every route that was sealed off, several new ones — often longer and more dangerous — were opened. This dialectic between control and escape which re- sults from this mobility conf lict has had a harrowing human cost: more than 30,000 migrants have perished at sea since the end of 1980.3 Most migrants’ deaths across 1  T his article draws on conversations that have taken place in many different contexts, and we are in- debted to the friends and colleagues who have shared their thoughts with us in each one of them. In particular we would like to thank Thomas Keenan, Maria Iorio and Raphael Cuomo, Paolo Cuttitta, Maurice Stierl, Martina Tazzioli, Michael Neuman, Aurélie Ponthieu, and Julian Koberer for their in- sights. 2  De Genova, “Spectacles of Migrant ‘Illegality’.” 3  U NITED for Intercultural Action, “List of 35,597 documented deaths of refugees and migrants due to the restrictive policies of “Fortress Europe.” 96 Charles Heller and Lorenzo Pezzani the Mediterranean frontier have not only occurred at sea, but through the sea, which has been turned into a deadly liquid as a result of the EU’s exclusionary policies which precaritize their crossings. The sea’s “geopower”4 has become embedded in a form of killing operating without state actors directly touching migrants’ bodies, in which vio- lence is rather inf licted in a mediated way, through water: it is the liquid element that transmits the violence of state policies to the bodies and lives of migrants.5 The precari- ous travelers whose lives are taken during their attempt to cross the Mediterranean frontier are thus the victims of what we call “liquid violence,” the specific modalities of which are, as we will see, ever changing.6 To contest the EU’s liquid violence, we initiated the Forensic Oceanography project in 2011.7 In that period, with the toppling of the authoritarian regimes in North Africa that had served as the EU’s outsourced border guards, migrants were able to ‘re-open’ maritime routes to the European continent. In so doing, they ushered in a phase of increased turbulence in the Euro-Mediterranean border regime. Since then, we have sought to understand and document the shifting modalities of the liquid violence op- erating across the maritime space between Libya and Italy (the central Mediterranean) with the aim of contesting them in legal forums and beyond. In this article, we chart some of these momentous shifts, and ref lect upon the way we sought to respond to them, focusing in particular on three moments of change: The first is the moment of rupture in the border regime marked by the Arab uprisings, which led European states to adopt recurrent practices of non-assistance exemplified by the left-to-die boat. The second corresponds to the lethal policies of non-assistance implemented by European states in terminating the Italian Mare Nostrum “humanitarian and security” operation in the aim of deterring migrants from crossing. Finally, the third charts the combined process of criminalizing civilian rescue initiatives and (re-)outsourcing border control to the Libyan coast guard. Distinct forms of violence have emerged within each of the phases, which in turn posed new challenges to register their traces and translate them into legal violations for which states might be held accountable. In our attempt to contest the shifting modes of violence operating at the maritime frontier, it has been essential for us to understand two dimensions that are central to organizing the way the maritime frontier operates. First, the mobility conf lict — of 4  Grosz, “Geopower.” 5 Here, we draw on Sean Cubitt’s expanded understanding of mediation, which, beyond technologically mediated communication processes between humans, he defines as “the material processes connect- ing human and nonhuman events […]. Mediation is the primal connectivity shared by human and non- human worlds.” (Finite Media, 3). The way in which, in another text, he talks about sunlight as that which “mediates the sun and the earth” (“How to Connect Everyone with Everything”), further points to the understanding of mediation that inspires us here. 6  W hile the Mediterranean is according to the International Organization for Migration the deadliest crossing in the world, it is far from being the only place in which the very geo-physical characteristics of the environment have been enlisted and harnessed as crucial tools of border control. From the arid lands of the Sonoran and Sahara Desert to the rugged mountain passes in the Alps or between Iran and Turkey, or the oceans encircling Australia, the US, Europe, and the Arabian Peninsula, migrants have been forced to traverse more and more inhospitable and hazardous terrains in the hope that the risk of injury and death they will face might deter them from attempting the crossing. See Heller and Pezzani, “Hostile Environment(s).” 7  See the “Left-to-Die Boat” project: www.forensic-architecture.org/case/left-die-boat/. Forensic Oceanography 97 which the sea has become the liquid terrain — opposes migrants’ movements not only to the bordering policies and practices of states, but to a multiplicity of other actors. It sees international diplomatic disputes between states, conf licts between distinct agencies within a single state — such as the coast guard, the police and the mili- tary — and the confrontation of a multiplicity of non-state actors such as international and non-governmental organizations, fishermen, shipping companies, and their trade associations, researchers, journalists, artists — all of which contribute, in various ways, to enabling or limiting migrants’ movement in a more or less deliberate way. A crucial axis that organizes this wide spectrum of actors and their relations are the logics of security and humanitarianism, both of which have become fundamental rationales, discourses and practices that cut across governmental and nongovernmen- tal actors.8 The coexistence and circulation of these logics within each actor and within the border regime as a whole, is always fraught with tensions, and the balance be- tween them is in constant f lux, with migrants being constituted simultaneously as “a life to be protected and a security threat to protect against.” 9 While, as Paolo Cuttitta reminds us, these f luctuations leave fundamentally unchanged the “restrictive migra- tion and border regime” imposed by the EU,10 they do have an important impact on the conditions of migrants’ crossings and the changing modalities of liquid violence we chart below. Each of the shifts we analyze corresponds to a reorganization of the logics of security and humanitarianism. Second, while the violence of the maritime frontier is mediated by water, it is also mediated by images and a constantly shifting aesthetic regime. We use the term “aes- thetic” in the sense underlined by Jacques Rancière as what presents itself to sensory experience.11 Distinct conditions of (in)visibility and (in)audibility are imposed onto the maritime frontier by states’ restrictive policies, but also shaped, transformed, and contested by the multiple other actors mentioned above, including of course mi- grants themselves, and their various media. Images, surveillance technologies, and their use within states’ and migrants’ strategies of (in)visbility shape the modalities of liquid violence in decisive ways. In turn, the changing configurations of security and humanitarian logics have distinct aesthetic dimensions, with the (in)visibility of migrants’ deaths in particular oscillating between concealment within the security logic and spectacularization in the humanitarian one. Each of these shifts have forced us to reposition our own practice, and seek to exercise anew what we call a “disobedi- ent gaze”— revealing what state actors have sought to conceal, and not revealing that which they seek to shed light upon. In what follows then, we chart the changes at the nexus of the modalities of liquid violence, the articulation of humanitarian and secu- rity logics, and the reconfiguration of the aesthetic regime operating at the maritime frontier, and ref lect upon the way each of them has shaped our project. We first discuss the aesthetic regime within and against which our project sought to position itself. 8 Fassin, Humanitarian Reason. 9  Vaughan-Williams, Europe’s Border Crisis, 3. 10  Cuttitta, “Repoliticization Through Search and Rescue?”, 649. 11  Rancière, The Politics of Aesthetics, 13. 98 Charles Heller and Lorenzo Pezzani The Mediterranean Frontier’s Aesthetic Regime of (in)Visibility At the EU’s maritime frontier, we find at work a complex and ambivalent aesthetic re- gime of (in)visibility, inextricably bound to the way the border regime itself operates. As a result of their illegalization through the EU’s policies of exclusion, people who decide to migrate despite legal denials are forced to resort to an informal infrastruc- ture of mobility: transnational networks of migrants who exchange information and services, the smuggling networks they resort to for a portion of their journey, as well as actual means of transport such as overused and overcrowded boats. Migrants are illegalized — their illegality is a product of state laws — and therefore they must migrate clandestinely, a term with etymological connotations of hiddenness and secrecy, and seek to cross borders undetected. The EU’s migration regime thus imposes a particular “partition of the sensible” in the terms of Jacques Rancière: it creates particular condi- tions of (dis)appearance, (in)audibility, (in)visibility.12 As opposed to the logic of clan- destinity, what all security-oriented agencies aiming to control migration try to do is to shed light on migration and in particular on acts of unauthorized border crossing in order to make the phenomenon of migration more knowable, predictable and govern- able. To this effect, a vast dispositif of control has been deployed at the maritime fron- tier of the EU, one made of mobile patrol vessels but also of an assemblage of multiple surveillance technologies, through which border agents seek to achieve the most com- plete possible “integrated maritime picture” in the aim of detecting and intercepting migrants’ vessels. These technologies range from vessel tracking technologies, coastal and shipborne radars, to optical and synthetic aperture radar imagery. Together, they compose what Karin Knorr Cetina has called a “scopic system”: “an arrangement of hardware, software, and human feeds that together function like a scope: like a mech- anism of observation and projection.” 13 However, the partition of the sensible of the EU’s maritime borders is more am- bivalent than this binary opposition would let us believe. For their part, migrants in distress may do everything they can to be seen so as to be saved from drowning. In this, they are not only seeking to avert the possibility of their imminent death, but also seeking to use the humanitarian logic that has become embedded in the practices of all actors at sea — including those whose very aim is preventing illegalized migration — to forward their own objective of crossing borders. Conversely, border agents not only attempt to deliberately hide the structural violence inherent to practices of policing maritime migration — thus allowing these practices to perpetuate themselves in full impunity — they may also choose not to see migrants in certain instances, considering that rescuing them at sea entails the responsibility for disembarking them and pro- cessing their asylum claims and / or deporting them. This has led to repeated cases of 12  I n a synthetic article, Jacques Rancière argues that politics is not the exercise of power or struggle for power, but first resides in the “configuration of a space as political, the framing of a specific sphere of experience.” It is precisely “that distribution and re-distribution of times and spaces, places and iden- tities, that way of framing and re-framing the visible and the invisible, of telling speech from noise and so on” that Rancière calls “the partition of the sensible.” For Rancière then, “to the extent that it sets up such scenes of dissensus” politics can thus be characterized as an “aesthetic” activity (“The Politics of Aesthetics”). 13  Knorr Cetina, “The Synthetic Situation,” 64. Forensic Oceanography 99 migrants who have been left abandoned to drift at sea, as in the left-to-die boat case we will discuss below. In all these different cases, visibility and invisibility then do not designate two discrete and autonomous realms but, rather, an entangled topological continuum. We find the same ambivalence at work in photographic and video imagery of the maritime frontier. In addition to the different remote sensing means described above, patrol vessels are also equipped with cameras — those of border guards or of “embed- ded journalists”— which are used to document the moment of encounter between il- legalized migrants and those seeking to police their movement. This results in a highly controlled and ambivalent spectacularization of borders, incisively analyzed by Nicholas de Genova.14 In the countless images of intercepted / rescued boats that are circulated by state agencies and the press, the threat of illegalized migration and the securitiza- tion work of border control are simultaneously made visible and naturalized, follow- ing a circular logic. If migrants are being intercepted through militarized means, it is because they are a threat. If they are a threat, then they must be policed by all means. The sense of migration as threat is only exacerbated by the profusion of similar images, which suggest an invasion of European space by those who have been constructed as radically other — racialized and impoverished migrants from the global South. These racialized representations of migration at Europe’s external maritime borders, which produce “a dominant associative notion of irregular migration to non-whites bodies,” shape in turn racialized border control within Europe, as “any non-white body on the move” is perceived as a potential ‘illegal’ traveller that must be checked.15 What emerges then is a fundamental link between the three distinct dimensions of migrants’ exposure emphasized by Georges Didi-Huberman:16 the visual exposure of illegalized migrants, their being “ex-posed” (rendered outside and excluded from a given com- munity), and the exposure of their bodies to conditions of precarity and death. The ex- posure operated by state actors is however highly selective. While focusing on the scene of border enforcement, the conditions that lie before and after — the multiple forms of violence migrants sought to escape in the first place, the illegalization of their move- ment through policies of exclusion, the future exploitation of illegalized migrant labor in European economies — remain hidden as obscene supplements. However, as we will see in more detail in the course of this chapter, the regime of (in)visibility imposed by the border spectacle is constantly changing, and is in par- ticular shaped by the unstable equilibrium between the logics of security and humani- tarianism. If within a logic of security migrants are constituted as a threat to Europe, within the humanitarian perspective they are rather victims whose lives are threat- ened — either in their countries of origin or in the process of crossing borders.17 Illegal- ized migrants crossing the sea thus operate as an inherently unstable and “f loating” signifier,18 and their (in)visibilization is equally f luctuating. While the deaths of migrants may at times remain hidden, at others they are spec- tacularized by state actors to cover the violence of borders with a humanitarian var- 14  De Genova, “Spectacles of migrant ‘illegality’.” 15  Keshavarz and Snodgrass, “Orientations of Europe.” 16  Didi-Huberman, Peuples exposés, peuples figurants. 17  Chouliaraki and Stolic, “Rethinking Media Responsibility in the Refugee ‘Crisis’.” 18  Hall, “Race, the Floating Signifier.” 100 Charles Heller and Lorenzo Pezzani nish: border control becomes framed as an act of saving migrants, occluding the fact that state policies endanger their movement in the first place. The spectacularization of migrants’ distress and deaths at sea is also mobilized by non-governmental actors to contest the lethal effects of borders. As these examples show, the sensing practices operating within this aesthetic regime do not simply document the violence of borders, but actively participate in it. Whether through the logic of the spectacle or that of state secrecy, the very act of exclusion that underpins the EU’s politics of migration takes place as well within and through its various visualizations. Struggling for the rights of migrants, then, means also intervening in this regime of (in)visibility and challenging the very borders of what can be seen and heard. For these reasons, understanding the shifting and ambivalent configurations of the aesthetic regime operating at the Medi- terranean frontier has been fundamental at each stage of Forensic Oceanography’s successive projects, which have always needed to navigate and constantly re-negotiate a fine and unstable line between complicity, resistance and evasion. Exercizing a disobedient gaze: Reconstructing the Liquid Traces of the Left-to-Die Boat Case Our project began at a time when the border regime was highly securitized and both migrants’ deaths and the violent (in)action of states were largely kept in the shadows. As both migrants’ crossings and fatalities at sea increased again in 2011 in the wake of the Arab uprisings, and with indications of state actors’ responsibility for this loss of lives, we launched the Forensic Oceanography research project within the wider Fo- rensic Architecture agency. Historically, forensic science can be understood as “a disciplinary project that af- firms the power of states.”19 Since at least the beginning of the twentieth century, states have relied on experts deploying scientific methods to find traces of events under in- vestigation so as to reconstruct them to prove or disprove a crime — an (in)action that constitutes a violation of legal norms. In doing so, states have often also policed and silenced the victims of their own violence, pitting the alleged objectivity of technology and science against the fallibility of human testimony. Our aim has been to somehow reverse this process and reinvent forensics as a counter-hegemonic practice that could be used by non-governmental actors to hold state and other non-state actors account- able for their crimes, focusing on events that occur in zones outside conventional state jurisdiction and beyond established frames of criminal justice.20 However, if the traces considered by the inventors of forensic science since the times of Edmond Locard (1877–1966) could be stains, fingerprints, or gunpowder, etc., today events are potentially registered by an infinite amount of materials and media, from phone communication to payment data, from videos shot with mobile phones to satellite images and vehicle tracking data, from sound recordings to rubble analysis.21 Despite the limits and even ambivalences of strategic litigation which we discuss further on, we considered that forcing states to account for the practices and 19  Weizman, “Introduction: Forensis,” 10. 20  Weizman, “Introduction: Forensis.” 21  Ruffel and McKingley, Geoforensics; Schuppli, “Walk-Back Technology.” Forensic Oceanography 101 policies that lead to migrants’ deaths had the potential to block the forms of border violence that we were observing at the maritime frontier. We thus began to explore the ways in which we might take this approach to the sea. In the process, however, we further had to challenge a well-ingrained imaginary of the maritime space as an empty expanse without history, where all traces of past events seems to be constantly erased by winds and currents.22 Our project was sparked by a 2011 incident that came to be known as the ‘left-to-die boat’ case.23 At the height of the NATO-led military intervention in Libya, during which more than 38 warships were deployed off the coast, 72 migrants f leeing the warzone were left to drift in the central Mediterranean Sea for fourteen days. 63 human lives were lost, despite the survivors calling Father Zerai (an Eritrean priest based in Rome) via satellite phone, despite distress signals sent out to vessels navigating in this area, and despite several encounters with military aircrafts and a warship. As we demon- strated in our report, this was not an isolated incident, but rather the outcome of re- current conf licts between Italy and Malta with regard to rescue at sea, as well as the aim of the military actors to restrict their activities to military objectives and not be seized by humanitarian ones. This was a time, then, when a security logic dominated the practices of the differ- ent actors operating in the central Mediterranean, and when the humanitarian prac- tice of rescue was instead repeatedly avoided despite maritime law imposing on all actors the duty to assist passengers in distress at sea. While survivor testimonies in- dicated increasing instances of non-assistance, during this period the Mediterranean appeared as a “black box” for civilian actors in which the capacity to see and document the events occurring at sea was nearly entirely in the hands of state actors — with the exception of Father Zerai’s unique capacity to listen to the distress of migrants at sea and pressure states into complying with their obligations. The challenge we faced as we embarked on our investigation in support of the nine survivors and a coalition of NGOs, was precisely in wresting the capacity to sense the sea away from state actors, so as to make the violence of abandonment visible and breach the impunity in which it was being perpetrated. Images could be of only limited assistance in the process. While several photo- graphs were taken at different moments during these tragic events by military person- nel as well as the passengers themselves, only one of them — taken by a French surveil- lance aircraft during the first day of the migrants’ journey — was released in response to a parallel investigation by the Council of Europe 24 (Fig. 6). In the absence of rev- elatory images documenting these events, our investigation had to rely on the “weak signals” that underpin truth production practices in the field, which Thomas Keenan (after Allan Sekula) has called “counter-forensics.” 25 22  We further elaborate on this aspect in Heller and Pezzani, “A Disobedient Gaze.” 23  F or our reconstruction of these events, see our report: www.forensic-architecture.org/wp-content/ uploads/2014/05/FO-report.pdf. Our video animation Liquid Traces summarizes our findings: www. vimeo.com/128919244. 24  Parliamentary Assembly of the Council of Europe (PACE), “Lives lost in the Mediterranean Sea.” 25  Keenan, ‘Getting the Dead to Tell Me What Happened.” 102 Charles Heller and Lorenzo Pezzani 6. Reconnaissance picture of the lef t- to-die-boat taken by a French patrol aircraf t on 27 March 2011 By corroborating survivors’ testimonies with information provided by the vast appa- ratus of remote sensing technologies that have transformed the contemporary ocean into a digital archive, we assembled a composite image of the events. The expertise of an oceanographer allowed us to model and reconstruct the drifting boat’s trajectory, and satellite imagery analysis to detect the presence of a large number of vessels in the vicinity of the drifting migrant boat that did not heed their calls for help (Figure 7.1 and 7.2). While, as we discussed above, these technologies are often used for the purpose of policing and detecting illegalized migration as well as other ‘threats,’ we repurposed them to find evidence of the failure to render assistance. Through our work on the ‘left-to-die’ case, we sought to put into practice a disobedient gaze that used some of the same sensing technologies as border controllers, but redirected their ‘spotlight’ from unauthorized acts of border-crossing, to state and non-state practices violating mi- grants’ rights. We conceived of this gaze as “[aiming] not to disclose what the regime of migration management attempts to unveil — clandestine migration — but unveil that which it attempts to hide, the political violence it is founded on and the human rights violations that are its structural outcome.” 26 26  Heller and Pezzani, “A Disobedient Gaze,” 294. Forensic Oceanography 103 7.1 Id Length Confidence Envisat satellite image 29_6 29_9 29_0 75m 80% 29_1 75m 70% showing the modelled 29_5 29_2 75m 95% 29_0 29_3 75m 80% position of the “lef t-to-die 29_2 29_23 29_4 75m 70% 29_1 29_4 29_3 29_5 75m 90% boat” and the nearby 29_6 29_7 225m 75m 95% 80% 29_13 presence of several military 29_14 29_8 29_9 75m 75m 70% 80% vessels who did not intervene 29_15 29_10 75m 80% 29_11 150m 95% to rescue the migrants 29_12 29_13 150m 75m 95% 80% 29_7 29_8 29_14 300m 95% 29_15 150m 95% 29_16 75m 80% 29_17 29_17 75m 95% 29_16 29_18 29_18 300m 95% 29_19 29_19 75m 80% 29_21 29_20 29_20 75m 80% 29_21 75m 70% 29_22 29_22 150m 95% 29_23 225m 95% 29_10 29_11 0 25 50 100 Kilometers 29_12 0 10 20 40 Nautical Miles 7.2 ITA KILOMETERS LY S Chain of events 0 60 120 RR LAMPEDUSA in the “lef t-to-die boat” MALTA SRR NATO MARITIME SURVEILLANCE AREA AS OF MARCH 23, 2011 D C B NATO MARITIME SURVEILLANCE AREA AS OF APRIL 8, 2011 A LIB E YA SR R TRIPOLI ZLITAN © Forensic Oceanography and SITU Research,  Report on the Lef t-to-Die Boat Case (both) 104 Charles Heller and Lorenzo Pezzani In addition to reconstructing the traces of events at sea, crucial to our project was the task of spatializing the practices of actors and inscribing them within the politi- cal geography of the sea. At sea, the moment of border crossing is expanded into a process that can last several days and extend across an uneven and heterogeneous ter- ritory that sits outside the exclusive reach of any single polity. The spatial imaginary of the border as a line without thickness dividing isomorphic territorial states is here stretched into a deep zone “in which the gaps and discrepancies between legal borders become uncertain and contested.” 27 The maritime territory constitutes, then, a space of “unbundled sovereignty” in Saskia Sassen’s terms,28 one in which sovereign rights and obligations are disaggregated from each other and extended across complex and variegated jurisdictional spaces. The multiple jurisdictional regimes that crisscross the Mediterranean have allowed states to simultaneously extend their sovereign privileges through forms of mobile government and elude the responsibilities that come with them.29 Facing their mobile and f leeting practices of bordering and (non-)assistance, we have sought to inscribe as precisely as possible the lethal events within specific jurisdictional zones and bounda- ries (such as SAR zones, but also in the case of the left-to-die boat, NATO’s maritime surveillance area) so as to assign responsibility for them. While the fragmentation of juridical regimes at sea often allows for the evasion of responsibility, we have sought to mobilize this very fragmentation strategically toward the multiplication of poten- tially liable actors who could be held accountable for not abiding by the duty to rescue inscribed in maritime law. Not only did our reconstruction of the migrants’ drift allow us to demonstrate that the migrants had remained within NATO’s maritime surveil- lance area during their 14 days of deadly drift (see Figures 8.1 & 8.2), but by identifying many ships in the vicinity of the migrants’ boat our report allowed the NGO coalition we collaborated with to file several legal cases against the different states — including France, Spain, Italy and Belgium — whose assets had taken part in the NATO-led op- eration, and who shared a degree of responsibility for the death of the 63 passengers.30 While these different complaints did have an impact insofar as they put pressure on state actors to change their practices, eight years after the events the legal process is still ongoing and none of the actors involved has been condemned by a court for their practices of non-assistance. Aware of the structural limitations of “international jus- tice to isolate a few culpable individuals while leaving the social and economic hier- archies of a society intact,” 31 we have also attempted to multiply the forums where the violence against migrants at sea could be debated and challenged: not only established legal arenas but also emerging social, political and cultural spaces, both institutional and informal.32 27  Neilson, “Between Governance and Sovereignty,” 126. 28 Sassen, Territory, Authority, Rights. 29  Steinberg, The Social Construction of the Ocean; Gammeltoft-Hansen and Alberts, “Sovereignty at Sea.” 30  “63 migrants morts en Méditerranée : des survivants poursuivent leur quête de justice.” 31  Weizman, “Introduction: Forensis,” 29. 32  I n negotiating the tension between these different forms of intervention, we have been inspired by what critical race theorist Mari Matsuda has called a dualist approach. “There are times,” she wrote, “to stand outside the courtroom door and say ‘this procedure is a farce, the legal system is corrupt, justice will never prevail in this land as long as privilege rules in the courtroom.’ There are times to Forensic Oceanography 105 As the events of the left-to-die case demonstrate, the initiation of the Forensic Oceanography project was marked by a context in which a security logic dominat- ed the maritime frontier, and with it an aesthetic regime of invisibilizing migrants’ deaths while spectacularizing their interceptions. In response, Forensic Oceanogra- phy’s strategy to make up for the near absence of available images documenting the lethal events of the left-to-die boat case was to create a composite image of events, assembling the traces generated by multiple sensing devices. In so doing, we opened a fissure in the partition of the sensible imposed by states, and in the impunity that prevailed for migrants’ deaths. While the methods we developed in the process re- main important to this day, the profound shifts in the modalities of violence and the aesthetic regime that emerged as part of rapid processes of humanitarianization and de-humanitarianization of the maritime border we observed in the wake of the Octo- ber 2013 shipwrecks near Lampedusa forced us to reposition ourselves and adapt our research and aesthetic strategies accordingly. A Forensics of the EU’s policies of non-assistance: Mare Nostrum and its Demise On October 3, 2013, a boat carrying more than 500 migrants sank less than one kilom- eter from the coast of Lampedusa, causing the death of at least 366 people and a public outcry.33 Not only did this boat manage to cross the multiple layers of surveillance sur- rounding Lampedusa undetected, but survivors of this incident also claimed that, a few hours before the boat capsized, two or three fishermen’s ships ignored their calls for help (this has not been confirmed or disproven to date). On October 11, another boat carrying over 400 people sank after rescue deployment was delayed for over 5 hours due to the conf licting responsibility of the Italian and Maltese Coast Guards, and more than 200 people died.34 Since both of these tragedies involved practices of non-assis- tance, they initially appeared as a tragic repetition of the left-to-die boat, with an even more exorbitant death toll. In hindsight, however, we can see that these shipwrecks were indices of much deeper changes. In the wake of these two tragic shipwrecks, migrants’ deaths suddenly gained tremendous public visibility as the haunting underwater images of the Lampedusa wreck circulated in the international press, forcing policy-makers to articulate their positions. During his visit to Lampedusa, Jose Manuel Barroso’s, then President of the European Commission, posed in front of the coffins containing the bodies of the ship- wrecked in the hangar of Lampedusa airport, which was transformed into something in between an improvised media center and a mortuary. In his statement on October 8, 2013, he declared: ‘We in the European Commission, […] believe that the European Un- stand inside the courtroom and say ‘this is a nation of laws, laws recognizing fundamental values of rights, equality and personhood.’ Sometimes, as Angela Davis did, there is a need to make both speeches in one day.” (“When the First Quail Calls,” 298). 33  S ee Watch the Med, “At least 366 people dead in wreck 1 km from Lampedusa”: www.watchthemed. net/reports/view/31. 34  S ee Watch the Med, “Guardia Civil runs over refugee boat near Lanzarote”: www.watchthemed.net/ index.php/reports/view/33. 106 Charles Heller and Lorenzo Pezzani ion cannot accept that thousands of people die at its borders.’ Despite denouncing the deaths of migrants as unacceptable, Barroso’s conclusion was not to take responsibility for these deaths or to challenge the EU’s exclusionary border regime which has proven so deadly. In the same speech, Barroso announced instead an increase in Frontex’s budget and the launch of Eurosur, the European Border Surveillance System — that is, the continuation of a predominantly security-based approach to migration. This policy, however, was now less framed as a response to the “risk” that the movements of illegalized migrants was presumed to constitute for the EU than to the risk they faced in attempting the crossing. Barroso’s speech, and the mise-en-scène of his and other policy makers’ compas- sion, are exemplary of what we may refer to as the humanitarianization of the border. What William Walters refers to as the “humanitarian border” emerges “once it be- comes established that border crossing has become, for thousands of migrants seek- ing, for a variety of reasons, to access the territories of the global North, a matter of life and death. It crystallizes as a way of governing this novel and disturbing situation, and compensates for the social violence embodied in the regime of migration control.”35 While rescues at sea have long been the humanitarian counterpart of the illegalization of migrants, the October 2013 shipwrecks marked a turning point where border con- trol operations themselves became framed as acts of saving, all while European states perpetuated their exclusionary border regime. Days later, Italian authorities, faced with the impossibility of ignoring the public outcry caused by these shipwrecks, single-handedly launched what has been by far the largest “humanitarian and security” operation in the Mediterranean: Mare Nostrum.36 In the framework of this operation, an unprecedented number of Italian Navy ships proactively patrolled close to the Libyan shores to rescue migrants and disembark them on Italian territory, thus marking a clear shift away from the principled reluc- tance observed in previous years to initiate rescue operations. Humanitarianism be- came a central dimension of this mission, both at the discursive and operational level, transforming as well the conditions of (in)visibility imposed on migrants’ crossings and state activities. As Martina Tazzioli has importantly highlighted, with Mare Nostrum the border spectacle was temporarily transformed and humanitarianized, although always in ambiguous ways.37 Now, instead of foregrounding the securitized scene of neutral- izing the threat of migration through border control, Mare Nostrum activities focused public attention on the good “scene of rescue,” recasting the role of the state and the military as that of a merciful savior.38 However, the “humanitarian border spectacle”39 was just as selective in the (in)visibilization of the maritime frontier as the primarily security-oriented spectacle that had preceded it: the good scene of rescue ended with disembarkation, and the illegalization and future exploitation of migrants onshore continued to remain hidden, as did their previous identification and fingerprinting that (at times) began on military ships while still at sea. Because the policies that pre- 35  Walters, “Foucault and Frontiers,” 138. 36  “Italy launches Mare Nostrum, 400 more saved.” 37 Tazzioli, Spaces of Governmentality; Tazzioli, “The Desultory Politics Of Mobility.” 38  Chouliaraki and Musarò, “The Mediatized Border.” 39  Cuttitta, “‘Borderizing’ the Island.” Forensic Oceanography 107 caritize migrants’ crossing were perpetuated, the large-scale rescue activities did not prevent them from dying in even greater numbers as the scale of crossings increased with the exodus of Syrians. In this sense, just like the spectacularization of the Lampe- dusa deaths, the display of heroic rescue activities occluded political responsibility for migrant deaths. Furthermore, the control of the capacity to know, document and produce imagery of the maritime frontier remained firmly in state hands. Even though select cases of migrants’ deaths as well as military-led rescue operations were now spectacularized, lending a sense of profound change in the aesthetic regime of the maritime frontier, there were deeper continuities in what was occluded by the border spectacle in both its security and humanitarian variants. Importantly, this phase demonstrated that the mourning of migrants’ deaths, which until then the border regime had attempted to keep largely hidden, could be rendered hyper-visible and spectacularized, without this entailing any increased accountability for their deaths. This situation posed a challenge to Forensic Oceanography’s strategy of demanding accountability by making deaths visible, a challenge that emerged even more acutely as the Italian-led humanitarian turn came under attack from its European counter- parts, leading to a shift in the modality of liquid violence perpetrated by states from practices to policies of non-assistance. While the causal relation between policies of clo- sure and migrant deaths had been undermined by the humanitarian border spectacle, re-establishing this connection would become the central task of our “Death by Rescue” report.40 The break with practices of non-assistance and the strength of the humanitarian logic marked by Mare Nostrum proved short-lived, as the operation was soon criticized for allegedly constituting a “pull-factor” for people crossing the Mediterranean. The UK Foreign Office Minister, Lady Anelay, exemplified this position when she stated “We do not support planned search and rescue operations in the Mediterranean. We believe that they create an unintended ‘pull factor,’ encouraging more migrants to at- tempt the dangerous sea crossing and thereby leading to more tragic and unnecessary deaths.”41 Translated in more frank terms by François Crépeau, United Nations Rap- porteur on the rights of migrants, this statement amounted to saying “let them die because this is a good deterrence.”42 As the balance of the border regime tilted once again towards securitization, European member states refused to Europeanize Mare Nostrum as Italy requested, and Italy terminated the operation at the end of 2014. As of November 1, 2014, the Frontex-led Triton operation was launched instead, deploying fewer vessels in an area further from the Libyan coast, and prioritizing border control instead of rescue. Through this operational shift, the EU and its member states hoped to make migrant crossings more difficult, so as to deter migrants from crossing. Nev- ertheless, as the UK statement exemplifies, the process of de-humanitarianization of the border European policy-makers sought to impose at the operational level was still couched in a humanitarian discourse: migrants should not be rescued proactively so as to deter them and ultimately prevent their deaths. At work then, was an paradoxical anti-humanitarian humanitarianism. 40  See www.deathbyrescue.org/. 41  See publications.parliament.uk/pa/ld201415/ldhansrd/text/141015w0001.htm. 42  See www.ohchr.org/EN/NewsEvents/Pages/DisplayNews.aspx?NewsID=15239&LangID=E. 108 Charles Heller and Lorenzo Pezzani But this humanitarian varnish was fissured by the dramatic scale of human suffer- ing that would unfold in early 2015 as a result of this policy shift. The week commenc- ing April 12, 2015 saw what is believed to be the largest loss of life at sea in the recent history of the Mediterranean. On April 12, 400 people died when an overcrowded boat capsized due to its passengers’ excitement at the sight of platform supply vessels ap- proaching to rescue them. Less than a week later, on April 18, a similar incident took an even greater toll in human lives, leading to the deadliest single shipwreck recorded by the United Nations’ High Commissioner for Refugees (UNHCR) in the Mediterra- nean.43 Over 800 people are believed to have died when a migrants’ vessel sank after a mis-manoeuvre led it to collide with a cargo ship that had approached to rescue its passengers (see Fig. 8.1 & 8.2). More than 1,200 lives were thus lost in a single week. As Médecins Sans Frontières (MSF) commented at the time, these figures eerily resemble those of a war zone.44 Beyond the huge death toll, what was most striking about these events — which resulted in few recorded images and which we therefore had to recon- struct on the basis of survivor testimonies and AIS vessel tracking data — was that they were not the result of a reluctance to carry out rescue operations, which were a struc- tural cause of migrants’ deaths in the past. In these two cases, the actual loss of life has occurred during and partly through the rescue operation itself. While it might appear (as state actors were quick to argue) that only the ruthless smugglers who overcrowded the unseaworthy boats to the point of collapse were to blame, our report titled “Death by Rescue — The Lethal Effects of the EU’s Policies of Non-assistance” argued that the absence of any immediate violation perpetrated by vessels in vicinity to the boats in distress hid a form of policy violence operating at a dif- ferent scale and temporality than that of the migrants’ crossing.45 In order to account for this violence, we had to go beyond the reconstruction of specific cases of death at sea and resort to what we have called a forensics of policies. In particular, our report meticulously reconstructed the policy and operational decisions taken in the forums of the European Commission in Brussels or in meeting rooms in Rome through which Frontex and EU member-states created what we have called a “rescue gap” in full knowledge of its lethal effects. 43  U nited Nations High Commissioner for Refugees (UNHCR), “Mediterranean boat capsizing: deadliest incident on record.” 44  M édecins Sans Frontières, “MSF calls for large scale search and rescue operation in the Mediterra- nean.” 45  See: www.deathbyrescue.org/. Forensic Oceanography 109 8.1 21:01 Automatic Identification OB AC GJ System (AIS) vessel tracks in KIN !! ! the Mediterranean following ! 05:39 [19/04] 4] !! ! ! !!!! 05:58 [19/04] ! !!!!!! !!! ! a Mediterranean shipwreck ! !! ! !!!! !!!!! ! ! ! ! ! !!!!!!!! 08:35 [19/04] of April 18, 2015 !01:22 ! 001:2 22 [19/04] 0111:::2 :22 23:46 2 ! !! / 11:22 [19/04] !!! 01:06 [19/04]] !!!! !!! ! ! !!!! !! ! ! ! !! !!!!! !! ! CP 10:14 [19/04] 44]] !!! ! ! ! ! ! !!! ! ! ! ! 10:27 : [19/04] 04] 92 21:155 !! ! 23:477 !! 0- ! ! !! !!!! BR !!! 1 21:31 ! ! ! !!! 22:17 2 7 UN !2 2 ! 44]] 00:26 [19/04] 22 40 22:40 !!21:27 ! O !!! !! ! GR ! ! 00:47 00 [19/04] ! COLLISION EG x O RE TT WRECK I 21:49 MIGRANTS’ BOAT SIGHTED LUTECE SPEED (in Knots) 0-1 2-5 6-10 >11 BRUNO GREGORETTI 21:38 CITY OF KING JACOB CITY OF LUTECE 11:30 [19/04] 11:24 [19/04] 0 1 2 4 km 8.2 Video still of an interview with a survivor of the April 18, 2015 shipwreck, showing his drawing of the collision between the migrants’ boat and the cargo ship 8.3 Map comparing the operational zones of Italian Navy’s Mare Nostrum and Frontex’s Triton © Forensic Oceanography (all), Death by Rescue report. GIS anal- ysis: Rossana Padeletti. Design: Samaneh Moafi  (8.1 and 8.3) 110 Charles Heller and Lorenzo Pezzani 9.1 and 9.2 Minutes of a technical meeting held in Rome on September 24, 2014 between representatives of the European Commission, Frontex, the Italian Ministries of Interior, Defense and Foreign Af fairs as well as the Italian Coast Guard Border Police and the Navy. In the meeting, a foreseen increase in search and rescue operation as a result of the end of Mare Nostrum is discussed Forensic Oceanography 111 112 Charles Heller and Lorenzo Pezzani Through the spatial analysis of operational zones, interviews with state officials concerning their operations at sea, and statistical data referring to migrant arrivals, deaths and SAR operations, our report reconstructed the reality that began to unfold in early 2015: migrants’ crossings continued unabated, but instead of a f leet of state- operated vessels, a lethal search and rescue gap awaited them. Seeking to fill this gap, the Italian Coast Guard increasingly called upon large merchant ships transiting in the area to carry out rescue operations. But because the rescue of migrants’ overcrowded boats can easily lead to tragedies if not operated with the most adaptive means and standards, the large vessels of the shipping industry were unfit for the task.46 In this context, the April 2015 tragedies were waiting to happen. On April 29, 2015, the Presi- dent of the European Commission, Jean-Claude Juncker, admitted that “it was a seri- ous mistake to bring the Mare Nostrum operation to an end. It cost human lives.”47 The ending of Mare Nostrum and its (non-)replacement by Frontex’s Triton operation, how- ever, cannot adequately be described as a “mistake” since it was a carefully planned policy implemented with full knowledge of its outcomes. Our report demonstrates that EU agencies and policymakers deliberately implemented policies of non-assistance that created the conditions that made the April shipwrecks inevitable. In the wake of the October 2013 shipwrecks, the rapid process of partial humani- tarianization and de-humanitarianization of the border that we saw at work between Mare Nostrum’s launch and termination involved profound shifts both in the regime of (in)visibility operating at the maritime frontier and the modalities of liquid violence. These changes demanded that we reposition our strategies in the aim of effectively contesting the violence of borders. The spectacular display of migrants’ deaths as part of the humanitarianization of the border was mobilized to justify border control, but also to justify the temporary deployment of a new scale of state-led rescue operations to mitigate the structural violence of borders. The conditions in which these lethal events unfolded, as well as the responsibilities of policy makers for restrictive policies that precaritize migrants’ crossings in the first place, were however still largely kept in the shadows. Contesting the partition of the sensible operating at the maritime frontier by using forensic techniques to reconstruct lethal events remained important; it proved insuffi- cient, however, in relation to the changing modality of liquid violence. Moreover, while in the case of the “left-to-die” boat our investigation was mainly concerned with finding new ways to shed light on an episode of violence that had been deliberately kept hidden, in this case the two shipwrecks of April 2015 received immediate and quite extensive media attention. While our reconstructions still proved important, the unfolding of the immediate events themselves was never particularly controversial. What was at stake was less the “draw[ing] open [of] a theatrical curtain […] behind which violence is lurking,”48 and more the attempt to re-establish the link that exists between the policy decisions taken in meeting rooms in Rome and Brussels, and their deadly effects at sea — thus displacing the scene of violence onto a different scale. In other words, the challenge was to bridge the gap that exists between the “visibility” of certain violent 46  “ICS: Rescue of all persons at sea is a must.” 47  E uropean Commission, “Speech by President Jean-Claude Juncker at the debate in the European Par- liament on the conclusions of the Special European Council on 23 April.” 48  Winter, “Violence and Visibility,” 199. Forensic Oceanography 113 events — in this case the two shipwrecks and the deaths these caused — and their “say- ability” in terms of violence, the possibility of producing statements that would be able to link these deaths with the political decisions that made them inevitable so as to seek accountability for them.49 The shift from practices to policies of non-assistance as a dominant modality of liquid violence in the context of the humanitarianization of the border, demanded that we supplement a forensics of cases with a forensics of policies, the scale of which could no longer limited to a particular incident, or even the central Mediterranean as a whole, but had to extend to policy forums and meeting rooms from Rome to Brussels. The form of liquid violence we reconstructed, which operates in an even more indirect way than in the past, has however proven challenging so far for lawyers to translate into a violation that might be accounted for in the language and forums of the law. As a result, the impunity which prevailed for the implementation of this lethal policy has allowed it to be perpetuated. Navigating the Architectural-Image Complex and the Ambivalences of the Humanitarian Border Spectacle Like the twin October 2013 shipwrecks, the twin April 2015 shipwrecks signaled an- other wave of impressive shifts in the assemblage of security and humanitarian logics shaping rescue and bordering practices, as well as in the aesthetic regime operating at the maritime frontier. Despite Jean-Claude Juncker’s admission of guilt for the in- crease in migrants’ deaths, the EU has continued to refuse to launch a new proactive search and rescue operation. Instead, it strengthened its security-oriented operations. On the one hand, it increased Frontex’s budget as well as the scope of its operations; on the other, on June 22, 2015, it launched a European anti-smuggling operation named EUNAVFOR MED. In opposition to Mare Nostrum’s combined “humanitarian and mili- tary” dimensions, the EU anti-smuggling operation was entirely “a police operation with military means,” as Rear admiral Hervé Bléjean, the Deputy Operation Com- mander in the Mediterranean, described it,50 and the rescue of migrants was far from the mission’s operational priority. However, this security-oriented mission continued to be justified in the name of saving migrants from the dangerous crossings they were subjected to by ruthless Libyan “traffickers,” not smugglers,51 described in a New York 49  T his attempt is connected with our reflection on structural violence. As we have argued in Liquid Trac- es, at the core of this notion lies an aesthetic problem. For many of the authors that have employed it, in fact, what is at stake is a distinctive relationship that is instituted between violence and visibility, one by which violence is concealed in plain view. As Nevins notices, “structural violence is not hidden simply because it occurred ‘behind the scenes’ […] but because powerful actors reproduce its hidden nature and / or construct it as something legitimate or other than violence through various represen- tations.” (“A Beating Worse than Death,” 16). In this context, “it is not invisibility that allows violence to be repeated and reproduced but [rather] repetition and reproduction [that] make violence invisible.” (“Violence and Visibility,” 202). 50  Padovani, “Les Passeurs sont Souvent des Migrants.” 51  F or the conceptual and legal distinction between smuggling and trafficking, see Monzini, Aziz, and Pastore, The Changing Dynamics of Human Smuggling and Trafficking in the Mediterranean. Monzini also underlines how the actual practices of the “actors practicing the commerce of illegalized passage” 114 Charles Heller and Lorenzo Pezzani Times op-ed by Prime Minister of Italy Matteo Renzi as the “slave traders of the 21st century.”52 This discursive humanitarian spin to this security mission was clearly il- lustrated when, in September 2015, Federica Mogherini, the EU’s High Representative for Foreign Affairs and Security Policy, suggested the operation be re-named “Sophia” in honor of a Somali baby born on one of its warships following a rescue operation.53 In adopting this name, Mogherini declared, she wanted “to pass the message to the world that fighting the smugglers and the criminal networks is a way of protecting hu- man life.”54 While the operation, in its initial phases, has come to rescue a substantial number of migrants — 54,000 between 2015 and 2016 — it has also consistently sought to refrain from initiating rescue operations in order to prioritize its anti-smuggling activities, and has almost entirely pulled back from its rescue activities as of 2017. Faced with the continued refusal of states either to fundamentally change their exclusionary policies or to redeploy a proactive rescue mission, a growing number of NGOs courageously stepped in with their own vessels to fill the lethal gap in rescue capabilities left by the ending of Mare Nostrum, progressively constituting a veritable civilian rescue f leet. As the work of Stierl and Cuttitta have shown, rescue NGOs are far from homogenous, and can be positioned on a wide spectrum in terms of their (de)politicization. While organizations such as MOAS have framed their activities as putting themselves at the service of states, and others such as SOS Méditerranée have tended to restrict their activities and discourse to the urgency of saving lives at sea, critical humanitarian organizations such as MSF, and much smaller NGOs such as Sea Watch, have adopted a far more politicized stance, denouncing the retreat and inac- tion of states, and calling on them to redeploy a large-scale SAR operation.55 These lat- ter organizations have further underlined that, as Mare Nostrum had already demon- strated, as essential and urgent as saving migrants in distress at sea might be, it could not put an end to deaths as long as the exclusionary EU migration policy remained in place. As a result, these organizations have called for a fundamental reorientation of the EU’s policies to enable “safe passage.” 56 With their sudden presence at sea, rescue NGOs have both denied states the monopoly over intervention in — and the monitoring of — the seas. Rescue NGOs quickly demonstrated a fundamental impact at the opera- tional level, as they came to rescue a greater number and share of people — reaching a peak of 35 % in 2017. As EU states emphasized the security dimension of their operations, and NGOs deployed their own rescue missions, a new and surprising situation began to emerge as of the spring of 2015. It was as if the Janus face of Mare Nostrum, humanitarian and security-focused, had been split in two, its distinct logics now separated into two dis- as we would rather call them, often blur between practices of smuggling and trafficking. See also Guilfoyle, “On being there to help.” 52  Renzi, “Helping the Migrants is Everyone’s Duty.” 53  F or more on the re-naming of ENAVFOR MED “Operation Sophia,” see Krista Lynes’s “SOPHIA” in this volume. 54  S tatement by Federica Mogherini, EUNAVFOR MED operation Sophia Operational Headquarters, Rome, 24th September 2015. 55  A particularly sticking example of MSF’s communication is provided by its April 20, 2015 statement: www.msf.org/article/msf-calls-large-scale-search-and-rescue-operation-mediterranean. 56  Médecins Sans Frontières, “EU: your fences kill. Provide safe and legal passage.” Forensic Oceanography 115 tinct actors. Would they be better reconciled in this new configuration? By spring 2015, it could seem so: after the transgressive arrival of NGOs, a relative complementarity settled in in which NGOs ran rescue operations, allowing state agencies to focus on destroying migrant boats in their wake. In this sense, even if security and humanitar- ian logics were operated primarily by distinct actors, they remained bound together in a “secret solidarity,” to borrow Michel Agier’s term.57 This surprising division of la- bor in fact pointed to a recurrent ambivalence in humanitarian practice, which, again in Agier’s words, is always at risk of becoming the “left hand of Empire,” healing the wounds wrought by the violence of the right hand.58 While unable to end migrants’ deaths at sea, and proving more effective in replacing state-led rescue at sea than in forcing states to reinstate their operations, NGOs nevertheless rescued 75,000 people between 2014 and 2018 — many of whom might have died without their presence. They further profoundly transformed the aesthetic regime of the maritime frontier in ways that have yet to be fully accounted for.59 Through the presence of NGO vessels, non-governmental actors suddenly had an unprecedented capacity to claim a right to look at the EU’s maritime frontier, wresting from states the monopoly over the knowledge and imaging of migration and border- ing across the sea.60 The outcome has been as impressive as it has been ambivalent.61 In some ways, rescue NGOs have clearly reproduced and even heightened the humani- tarian border spectacle that had taken form during the Mare Nostrum operation. Just like the Italian operation, they substituted the securitized scene of border interception with the good scene of rescue. Taking on board journalist teams who f locked in droves to the Mediterranean frontier as the “migration crisis” intensified in 2015, but also de- ploying their own media teams and equipping their crews with GoPro cameras, the 57  Agier, “Humanity as an Identity.” 58  A gier, “Humanity as an Identity.” Rescue NGOs have been well aware of these ambivalences, and have led their own reflection as to how to respond to them. This has been conveyed to us in numerous in- terviews, and can be seen as well in the documentary film Iuventa, directed by Michele Cinque, 2018. See also Newman, “Médecins sans Frontières France.” 59  T his act of bearing witness to under-reported catastrophes around the world as well as the ambigui- ties it carries with it are of course not a novelty but sit at the very core of the history of humanitarian- ism and the international human rights movement at large. Humanitarian rescue ships have played a crucial role in this field at least since the Indochinese refugee crisis that started in the late 1970s, which “stands out as a key event in the history of this new media-savvy interventionist humanitarianism.” See Whyte, “Human Rights;” and Pezzani, Liquid Traces, in particular chapter 2. 60  I n this, they seem to instantiate one of the very definitions of politics provided by Rancière’s “Ten The- ses on Politics”: against the police injunction to “Move along! There is nothing to see here!, […] politics, in contrast” the French philosopher argues, “consists in […] refiguring the space, of what there is to do there, what is to be seen or named therein” (22). 61  T he presence of rescue NGOs at sea is not an absolute novelty but has an important precedent in the South China Sea refugee crisis of the late 1970s, in the context of which the very term “boat-people” was coined. This episode “stands out as a key event in the history of this new media-savvy interven- tionist humanitarianism” (“Human Rights”), insofar as the intervention of humanitarian rescue ships in Indochina was crucial in the emergence and consolidation of a new wave of humanitarian actors “without borders.” The importance these actors placed on the act of bearing witness and their reck- oning with the ambiguities that the latter brought about have been a defining feature of this new chapter in the history of humanitarianism and the international human rights movement at large. See Pezzani, Liquid Traces, in particular chapter 2. 116 Charles Heller and Lorenzo Pezzani images produced by rescue NGOs (or through them) brought distant viewers into an unprecedented proximity with the extreme situation of life and death that each rescue operation constitutes. Making visible their rescue activities may have been necessary to convey the continuing urgent reality of migrants’ distress at sea, and more prosaically to allow NGOs to justify their activities to their funders. In the process however, as had been noted by numerous observers,62 rescue NGOs reproduced the intertwined tropes of the white savior and the racialized, precarious subject fighting for survival.63 While many members of rescue NGOs we have spoken to are well aware of these problems, and have sought to address them as best they can — for example by alternating the type of imagery they have circulated between images of distress and empowerment, or conducting interviews with rescued migrants on the deck of their vessels to allow them to unfold their life stories64 — there is no easy way out of the radically unequal positions our postcolonial world generates, which permeate humanitarian practices in general.65 In fact, moments of radical precarity in the open sea bring these inequalities into even sharper relief.66 Furthermore, NGOs’ radicalized version of the humanitar- ian border spectacle also reproduced some of the configurations of the partition of the sensible imposed by the spectacle of border security: by spectacularizing the maritime crossing, the imagery produced by NGOs continued to echo the iconography of the Eu- ropean imaginary of invasion, as well as this imaginary’s occlusion of the conditions lying before and after the sea crossing. No matter how hard different rescue NGOs have tried to frame their images differently — and a more careful study than what we can offer here would be necessary — in effect many of the images they have produced have been nearly identical, and practically interchangeable, from one rescue organiza- tion to another. Despite these ambivalent outcomes, which should not be taken lightly, rescue NGOs have also re-organized the Mediterranean frontier’s partition of the sensible in other highly oppositional and decisive ways. By wresting the monopoly over the capac- ity to document events at sea away from states, they were also able to redirect part of the “light” shed by the humanitarian border spectacle towards the violent (in)actions of state actors. Without the civilian oversight at sea that NGOs have permitted, much of this violence would have remained in the shadows. This role would become even more crucial when, after an initial phase of relative complementarity between the se- curity-oriented operations of states and the rescue activities of NGOS discussed above, European states deployed drastic measures to seal off the Mediterranean frontier by criminalizing solidarity and (re-)outsourcing border control as of the summer of 2016. As migrants’ capacity to overcome the EU’s borders peaked in summer 2015, when the Aegean temporarily replaced the central Mediterranean as the main area of cross- 62  D anewid, “White Innocence in the Black Mediterranean;” Stierl, “A Fleet of Mediterranean Border Hu- manitarians;” Cuttitta, “Repoliticization Through Search and Rescue?” 63  T hese issues have also been critically reflected upon from within rescue NGOs. See for example New- man “Médecins sans Frontières France.” 64  S ee for example MSF’s dedicated Facebook page “Voices from the Road”: www.facebook.com/MSF. VoicesFromTheRoad/. 65 Fassin, Humanitarian Reason, 4. 66  Tazzioli, “When Rescue is Capture.” Forensic Oceanography 117 ing, plunging the EU into a deep crisis, European states sought to impose a violent roll-back of the border regime. Unable to manage migrants’ transgressive movements once they arrived on European territory, in the aim of re-imposing order on the bor- der regime, European states found only one solution: (re)outsourcing border control to whatever partner they could find along migrants’ entire trajectories. After success- fully sealing off the Aegean thanks to the EU-Turkey deal in March 2016,67 the atten- tion of policy makers returned to the central Mediterranean, where the only available partner facing Italian shores was the shattered Libyan state, and particularly its un- savory coast guard units. Outsourcing border control once again to these Libyan part- ners — as had been the case already in 2009 — demanded that NGOs be sidelined. After all, if the newly-equipped Libyan units were to intercept migrants leaving their shores effectively, the same migrants could not be rescued by NGOs which would bring them to European soil. Furthermore, to allow the Libyan coast guard to intercept migrants at gunpoint with impunity, the monopoly of states over the maritime frontier’s aes- thetic regime had to be restored. For all purposes then — inextricably operational and aesthetic — the Mediterranean had to be de-humanitarianized. Since the summer of 2016, then, Italy, with the full support of the EU, has stepped up its collaboration with the Libyan coast guard, and at the same time led a virulent campaign of delegitimization and criminalization of NGOs — a two-pronged policy we have called Mare Clausum.68 These two dimensions have been the focus of our work since 2016. Knowing the looming catastrophe that the attacks against rescue NGOs signaled, we first attempted to intervene in this debate through our report, “Blaming the Rescuers,” which demonstrated the fallacy of the “pull-factor” attributed to SAR NGOs, an argument that had already been mobilized against Mare Nostrum. In collabo- ration with Forensic Architecture, we have further offered a counter-reconstruction of events of alleged collusion with smugglers by the NGO Jugend Rettet, whose vessel was seized on August 2, 2017.69 Thanks to the documentation produced by this NGO — in particular through GoPro cameras mounted on its crew’s helmets and the images pro- duced by journalists on board its own and other NGO vessels — we have demonstrated that the accusations against NGOs have been spurious and amounted to “factual lies” (the use of factual elements to weave a narrative that is intentionally false).70 Our efforts — as that of many other actors — have however proven insufficient, and since the summer of 2017 the NGO f lotilla has been reduced to only a few remaining as- sets, leading as in the past to a greater risk for migrants of dying at sea.71 Furthermore, as the share of migrants rescued by NGOs fell, the share of interceptions operated by the Libyan coast guard increased in parallel. Between 2016 and 2018, 55,000 migrants were pulled-back to Libya, where they have faced detention, forced labor, torture, and rape.72 In this shift then, we can sea a repetition of the EU’s policy of non-assistance, enacted with the ending of Mare Nostrum, now supplementing the indirect violence 67  Heck and Hess, “European restabilization attempts of the external borders and their consequences.” 68  S ee content.forensic-architecture.org/wp-content/uploads/2019/05/2018-05-07-FO-Mare-Clausum- full-EN.pdf. 69  See our “Blaming the Rescuers” report, June 2017. www.blamingtherescuers.org/. 70  w ww.forensic-architecture.org/case/iuventa/. 71  See UNHCR, “Desperate Journeys.” 72  See our “Mare Clausum” report: www.forensic-architecture.org/case/sea-watch/. 118 Charles Heller and Lorenzo Pezzani 10.1 and 10.2 Video still from The Crime of Rescue — The Iuventa Case (2018) ©©Forensic Oceanography and Forensic Architecture (both) Forensic Oceanography 119 exercised at and through the sea with the all too direct forms of violence perpetrated by proxy on firm land. Just as the presence of rescue NGOs was being drastically reduced over the sum- mer of 2017, and with it the rift NGOs had opened in the state-imposed partition of the sensible, the German NGO Sea Watch radicalized its visual politics. In the face of the accusations of collusion with smugglers and mounting Libyan coast guard attacks on rescuers and migrants alike, Sea Watch supplemented the GoPro cameras of its already well-equipped crew by mounting six cameras on the deck and mast of its ship, as well as devices to record its radio communications with state actors. The Sea Watch ship be- came a kind of moving audio-visual recording apparatus, continuously documenting everything that occurred on the horizon 360 degrees around it. The footage Sea Watch would gather in early November through these unprecedented means would prove central to our attempt to counter Italy and the EU’s outsourcing of border control to the Libyan coast guard.73 On the morning of November 6, 2017, a conf lictual rescue / interception event oc- curred, after both Sea Watch and the Libyan coast guard were directed by the Italian coast guard to the position of a boat in distress carrying more than 130 migrants. With its recently repaired vessel (that had been handed to Libyan authorities by Italy in May of that year), the Libyans were able to arrive on scene first, where many of the pas- sengers had already fallen into the water after one of the boat’s tubes def lated. The Libyans captured those who were still clinging to the boat using very dangerous ma- noeuvres — despite 8 of the 13 crew members onboard having been trained by the EU’s anti-smuggling operation “Sophia.” As the Sea Watch vessel approached minutes later, its crew deployed small fast boats to rescue the passengers set adrift. Soon, the an- tagonistic logics of interception and rescue came to a head. As Sea Watch’s crew ap- proached the migrants’ boat to rescue the passengers struggling in the water around it and not assisted by the Libyans, the Libyan Coast Guard threatened them and threw hard objects at them to keep them away. For the migrants caught between the Libyan coast guard and the rescue NGO, the few meters between their vessels is what sepa- rated the prospect of a new life in Europe and the certainty of violence at the hand of Libyan captors, which many of the migrants had already experienced. Some of the migrants the Libyan coast guard had managed to bring on board — and were beat- ing — managed to jump overboard and reach the NGO boat instead. Losing control, the Libyan coast guard set off at high speed with one passenger still hanging on the boat’s side ladder. In total, at least twenty people died before or during these events and as consequence of the conf lictual and chaotic operation. Sea Watch was able to rescue 58 migrants and bring them to safety in Italy, while 47 were captured by the Libyan coast guard and brought back to Libya where, as several survivors later recounted, they suf- fered grave human rights violations — including arbitrary detention, brutal beatings, rape, and starvation. In some ways, this incident was far from exceptional: the Libyan coast guard inter- cepted and returned to Libya more that 20,000 people in 2017 alone. What was excep- tional was that this pull-back event was only partly successful, so that several survivors could be retrieved and interviewed in Italy, allowing in turn that contact be estab- 73  S ee our “Mare Clausum” report and associated video investigation: www.forensic-architecture.org/ case/sea-watch/. 120 Charles Heller and Lorenzo Pezzani lished with those brought back to Libya. Moreover, by turning its vessel and crew into a complex apparatus of audio-visual recording, Sea Watch was able to record the entire violent event with unprecedented precision. To measure the change in the aesthetic regime of the Mediterranean frontier that had occured, we need only compare the sin- gle photograph we could access for the left-to-die boat case in 2011 with the hours and hours of video recordings of the events from multiple angles that Sea Watch handed over to us in order to reconstruct these events. To respond to this overabundance of images, which was novel for events occurring at sea, we had to call upon the unique method the Forensic Architecture agency has developed in a number of contexts on firm land — what it calls “the architectural-image complex.”74 By locating each cam- era within a dynamic three-dimensional model of the scene, Forensic Architecture’s method enabled us to navigate within an image-space and offer a minute-by-minute reconstruction of the facts from multiple perspectives. The audiovisual material produced by Sea Watch, and the way we have been able to assemble it with Forensic Architecture, provided fundamental evidence of the violence of the Libyan coast guard and its collaboration with its EU counterparts. To recon- struct this collaboration at a policy level, we resorted to the method of policy forensics developed in our previous reports.75 Finally, the testimonies of several of the survivors who were brought back to Libya give account of the brutal violence to which they were subjected, which lies far beyond the frame of Sea Watch’s cameras, and beyond the sea itself. Together, these testimonies, the video reconstruction, and our written report served as the factual basis for a legal case before the European Court of Human Rights filed by seventeen survivors in May 2018 against Italy. The argument put forward by the legal team constituted by Global Legal Action Network and the Association for Juridi- cal Studies on Immigration is that — because of the multiform collaboration between It- aly, the EU and the Libyan coast guard — Italy was engaging in “refoulement by proxy,” and was responsible for the passengers’ fate at sea and in Libya.76 As important as the audiovisual material produced by Sea Watch has been for our re- construction and the ongoing demands for accountability that aim, through litigation, to force Italy and the EU to interrupt their policy of outsourced border control, it has also confronted us with new difficulties and questions that we still ponder. Beyond the abundance of available images of this violent incident — an outcome of the profound reorganization of the partition of the sensible at sea operated by rescue NGOs — there is another striking difference in relation to our previous investigations that were char- acterized by a paucity of accessible imagery: while the lack of images in the report on the left-to die boat and on the April 2015 shipwrecks forced us to assemble a composite image of events by combining survivor testimonies with georeferenced data and re- mote sensing technologies so as to map the unfolding of events with a certain distance, here the images produced by Sea Watch bring us to the heart of events, in an intimate and disturbing proximity with bodies struggling for their lives and their cries of dis- tress. We are faced with scenes of radical precarity, and even death (the video includes images of two people drowning). From the perspective of the continuously recording 74  “Interview with Eyal Weizman.” 75  See our “Mare Clausum” report: www.forensic-architecture.org/case/sea-watch/. 76  “ Legal action against Italy over its coordination of Libyan Coast Guard pull-backs resulting in migrant deaths and abuse.” Forensic Oceanography 121 11.1 Video still from Mare Clausum — The Sea Watch vs Libyan Coast Guard Case (2018) 11.2 Video still of Interview with A., survivor of the confrontation between Sea Watch and the Libyan Coast Guard (2019) ©©Forensic Oceanography and Forensic Architecture (both) 122 Charles Heller and Lorenzo Pezzani camera fixed on Sea Watch’s mast, we see a man’s body slowly swallowed by the sea’s liquid mass after Sea Watch’s crew has been prevented from rescuing him by the Lib- yan coast guard. These are horrific images of a horrific event, and their presentation as documents to a court entails that this violence must remain in the brutal form in which it was captured. Even if the survivors we have worked with wanted these im- ages to be seen as part of their demand for justice, and even if this footage constitutes essential evidence to incriminate both the Libyan coast guard and European states, working with and displaying such imagery conjures difficult — and perhaps unresolv- able — questions: How to contest the violence of the EU’s outsourced policies of border control without reproducing other forms of symbolic violence in the process? Our use of these images is first of all haunted by the ghost of the dead from whom we will never be able to request consent to use this footage, whose lives were only made visible to us “in the moment of their disappearance.” 77 While the initial use of these images within our video has been geared to a judicial context, as a consequence of their further circulation in different cultural, institutional and political forums, no ge- neric public can be assumed as viewer of those images. As such, echoing other debates on the consumption and circulation of images of Black suffering we must ask: who is meant to see those images and what do those images mean to her/him? Can they be effectively mobilized to counter the fungibility of Black life? If yes, by whom and un- der what conditions? Might their circulation in certain contexts undermine that aim? 78 Clearly we see a risk that, in recirculating these images of Black subjects in conditions of radical precarity, struggling for survival and being rescued by European activists, we reproduce racialized tropes of the rescued / rescuer. Our own work of reconstruc- tion is affected by the inherent ambivalence of the images produced by NGOs as part of the humanitarian border spectacle we have discussed above. Furthermore, as George Didi-Huberman has recently reminded us in a scathing critique of Ai WeiWei’s Human Flow (2018), if for Walter Benjamin criticism is “a matter of correct distancing,” we may ask if the images of NGOs haven’t brought us too close? How to re-introduce distance from within this extreme proximity? Moreover, in a context of over-abundance of im- ages, could the subtraction of visual imagery become an additional — and in certain situations not only more ethically sustainable but also more effective — tool in our rep- ertoire? These are among the questions and tensions that continue to inhabit us, and which, as Saidiya Hartman underlines, are “unavoidable in narrating the lives of the subaltern, the dispossessed, and the enslaved.” 79 After having submitted our factual video reconstruction based on Sea Watch’s foot- age to the European court, one of the ways we have sought to respond to these ques- tions is by accompanying the display of our own reconstruction in artistic contexts with a video interview in which a survivor looks back at and re-subjectivizes these im- ages. This is, however, far from exhausting them, and risks to continue reproducing the unequal relationship that the trope of “giving voice to” a under-represented group always institutes. What these difficulties point to is that relying on some of the tools 77  Hartman, “Venus in Two Acts,” 12. 78  W e are referring for instance to the debate that developed in the wake of the exhibition of Dana Schutz’ work at the Whitney Biennale in 2017. See Mitter, “What Does It Mean to Be Black and Look at This?” 79  Hartman, “Venus in Two Acts,” 12. Forensic Oceanography 123 mobilized and images produced as part of the border spectacle, whether in their se- curitized or humanitarian variants, demands careful positioning, and charting one’s course through a field fraught with ambivalences in which there is at times no unprob- lematic position. This, we would argue, is the risk that comes from intervening within and against the aesthetic regime of the maritime frontier. Conclusion The trajectory we have charted of the changing modalities of liquid violence, constant rearticulations of humanitarian and security logics, and reconfigurations of the aes- thetic regime operating at the maritime frontier, have all shaped the twists and turns of our own practice as researchers, aesthetic practitioners and activists striving to contest the violence of borders. The complex and ambivalent shifts at the maritime frontier we have described trace a sequence of openings and closures in relations to migrants’ movements, the presence of non-governmental actors and the aesthetic re- gime. While in 2011, migrants had pried open the Mediterranean frontier in the wake of and as an integral part of the uprisings in North Africa and the Middle East, with each group of illegalized migrants intercepted and pulled-back to Libya, or left strand- ed at sea for days, the sea is once again being closed down. The sea is also closing down on activists, who, in response to the demands for freedom and equality that emerged out of the so-called Arab uprisings, have in the last few years transformed the sea into a central space of political struggle, inventing new strategies and tactics to contest the violence of borders and support migrants’ movements. Contesting the partition of the sensible imposed by states on the liquid frontier has been a central dimension of each of these initiatives — even if their intervention conjured new problems in the pro- cess — and the attempt to criminalize them and expel them from the maritime frontier has also been an attempt by states to re-impose their own regime of (in)visibility. The closing down of the sea in these different respects has continued apace since the autumn 2017 Sea Watch vs Libyan Coast Guard incident described above. While our reconstruction was the basis for a legal challenge against Italy in front of the Eu- ropean Court of Human Rights filed on May 8, 2018, the deterrent effect we hoped our complaint might have on state practices even as the court examined the complaint, did not materialize. After far-right politician Matteo Salvini became Interior Minister on June 1, 2018, the de-humanitarianization of the border pursued through the Mare Clau- sum policy was only further radicalized: more NGOs have been prevented from operat- ing, rescued migrants have been prevented from disembarking, and more migrants have been pulled back to Libya.80 The rift in the monopoly of states over imagining and documenting the maritime frontier (which rescue NGOs had widened) has been nar- rowed. While we continue to consider strategic litigation as an essential political tool against the lowering of the threshold of acceptable violence (particularly in regards to push-backs), as we have repeatedly hinted at above it is also an inherently limited one. 80  I n this phase, the lives of migrants appear to have increasingly lost even their discursive value, as the absence of any representative of the current Italian government at the ceremony commemorating the Lampedusa shipwreck deaths in October 2018 indicates. “March in Lampedusa to honour victims of 2013 shipwreck.” 124 Charles Heller and Lorenzo Pezzani Despite the tendency of closure, the sealing off of the Mediterranean frontier remains an unstable, contested and open-ended process. The recent shifts in the Mediterranean frontier’s partition of the sensible have demanded new responses by non-governmental actors seeking to contest the violence of borders. With the vessels of rescue NGOs prevented from operating, the role of an activist hotline supporting migrants in distress in the Mediterranean we have contributed to forge, the Watch- TheMed Alarm Phone, has proven more important than ever over the last months.81 Launched in October 2014 by a coalition of freedom of movement, human rights, and migrant activist groups, the Alarm Phone connects more than 150 activists located in about 12 countries, who take shifts to respond to distress calls from migrants cross- ing the sea. The Alarm Phone activists enact what we might call a topological solidarity: while the activists that respond to calls are dispersed throughout Europe and North Africa, through the phone line, they can hear the precarious passengers as if they were beside them. The Alarm Phone has intervened in a crucial way in the Mediterranean frontier’s aesthetic regime. By providing support to and amplifying the voices of mi- grants to denounce state violence,82 the Alarm Phone has exercised a form of disobedi- ent listening, which is all the more crucial today when the visual means of monitoring the maritime frontier are challenged. The moment of closure we are facing is thus far from being the end of the story. Illegalized migrants from the Global South will continue to refuse their banishment from spaces of (relative) opulence, and non-governmental actors will strive to find new ways to contest the violence of borders, which, as we have shown, also involve contest- ing the boundaries of what can be seen and heard. But this open-ended sequence also underlines the limits of the hand-to-hand struggle of migrants and those who sup- port them with the border regime, since we have collectively proven unable to durably end border violence. Migrants’ claims to freedom of movement — which they express through their voices and practices of unauthorized mobility — have not been heeded. While they have pushed the European border regime into a phase of turbulence, lead- ing to the rapid changes we have outlined in this chapter, the fundamentals of the EU’s policy of migrant illegalization have remained unchanged. In this context, strategic litigation may have blocked certain violent practices temporarily, but it has not chal- lenged the foundational violence of the EU’s policies of exclusion. Likewise, state and NGO vessels deployed to rescue migrants have succeeded in mitigating the lethal ef- fects of the violence of the liquid frontier, but proven unable to put an end to the very necessity of rescue, and with it to the “asymmetry between rescuer and rescued,”83 that is intrinsic to this practice and that haunts the images produced by state and non-state actors alike. As important as these practices have been, they have proven to be “not enough.”84 These limitations are certainly not the ground to refrain from any form of intervention, but acknowledging them is necessary to navigate a fine and unstable line separating resistance, refusal and complicity. 81  S ee: www.alarmphone.org / . For an extended discussion see Heller, Pezzani and Stierl, “Disobedient Sensing and Border Struggles.” 82  Schwartz and Stierl, “Amplifying Migrant Voices.” 83  Tazzioli, “When Rescue is Capture.” 84  Moyn, Not Enough. Forensic Oceanography 125 While activist energies, and much of our own research, focus on countering the shifting modes of violence at the border, this risks leaving the systemic conditions within which borders are embedded, and which reproduce the mobility conf lict of which the Mediterranean is a major frontline, unaddressed. The problems arising from an excessive focus on the border as the site of political contestation is also partly an aesthetic question. As we have shown in relation to the humanitarian border spec- tacle, there is a risk that, even as we contest the securitized border spectacle and its lethal policies, one partly reproduces its selective (in)visibilisation — for example by contributing to the over-representation of racialized migration across the Mediterra- nean frontier that fuels Europe’s imaginary of invasion, or by occluding the forms of systemic violence that extend far beyond the sea’s coastline to shape the social, politi- cal and economic conditions that migrants escape in the first place. Our ongoing inter- est in migrants’ trajectories as an epistemic device85 stems precisely from the way they connect multiple locations and struggles, bringing into view an expansive field of rela- tions.86 While this may be part of a response to the challenges of contesting the border spectacle, we are well aware that each practice and perspective reveals as much as it conceals. That partiality cannot be undone, but it demands to be constantly ref lected upon and the configuration of the (in)visible it produces questioned anew. At best, its recognition can produce a sense of humility, and the acknowledgement of the need for practices and perspectives other than one’s own. 85  Mezzadra and Neilson, Border as Method. 86  Heller and Pezzani, “Contentious Crossings.”