Killing by Omission

Written by

On Monday, the Forensic Architecture team at Goldsmith College, London, published Death by Rescue. The report exposes a rather complex set of facts, but the basic argument is as simple as it is alarming.

Operation Triton, facilitated by Europe’s border security agency, Frontex, began on 1 November 2014 and is mandated to enforce Italy’s maritime border. Triton replaced an earlier and much wider Italian Navy operation, Mare Nostrum, which began in October 2013 and was mandated to save migrant lives beyond Italy’s territorial waters. When EU officials decided on the more limited scope of Triton, they knew their decision would result in the drowning of numerous migrants. As one Frontex official wryly noted, “the withdrawal of naval assets from the area, if not properly planned and announced well in advance, would likely result in a higher number of fatalities.” But the European Commission turned a blind eye – leading to a spike in migrant deaths, which the authors, Charles Heller and Lorenzo Pezzani meticulously document.

From a legal perspective, this set of circumstances raises the question whether the migrants’ rights were violated, and if so, whether EU actors can be held legally accountable. In my view, the report exposes no illegal activity by European agents, either at the operational or at the policymaking level. Perhaps more troubling, the report raises the specter of unaccountable violence ingrained in the very structure of international law. If international law is somehow to blame for circumstances that made these utterly preventable deaths possible, then perhaps it is law itself that should be indicted.

Law of the Land, Law of the Sea

To explain what I mean by that, several rather theoretical remarks are required.

In common law countries, one of the first things law students learn is that law imposes no duties of rescue upon individuals qua individuals.  The classical jurisprudence on this includes comically macabre examples. A characteristic hypothetical describes a bystander witnessing a drowning baby. Law professors often use the initially astonishing absence of a duty of rescue to illustrate a basic tenet of legal positivism: the distinction between legal and moral prescription (or “the separation thesis”). Students are expected to adopt this distinction as a second nature. Rescuing the drowning stranger, they are comforted, is morally required. Of course, there are important exceptions to the general absence of a duty of recue. The basic point nevertheless stands: law does not impose a duty of rescue. Law does not always follow moral prescription.

The absence of a duty of rescue is, famously, not uniform across the comparative law terrain. In civil law traditions, typically we find criminal provisions penalizing certain failures to rescue. Upon scratching the surface, however, one quickly discovers that duties of rescue in civil law countries too are not very robust. Indeed, they are but wrinkles overlaying the fundamental presumption: individuals are not directly responsible for each other’s security.

As legal theorist Arthur Ripstein has elegantly argued, the absence of a duty of rescue is intimately related to the very basis of sovereignty in western political thought, i.e., to the social contract. By granting the responsibility for security to a public entity, law somehow “relieves” individuals from direct mutual responsibility, opening a space for private life. More recently, David Singh Grewal explained the Hobbesian roots of this allocation of responsibility to sovereigns, labeling it “realist utopianism.” The responsibility for securing people within a state’s jurisdiction dovetails with its monopoly over legitimate violence. The dominant understanding of human rights law redoubles this structure. It relies on states as both the enforcers of law and the sources of law. States bare the duties that correspond to the rights of people within their respective jurisdictions.

More rarely acknowledged in such basic courses is the fact that travelers on the earth’s oceans and seas are bound by a duty of rescue. This is not the case in a state’s territorial waters. Here states are still required to provide assistance to vessels in distress. It is also not true in a state’s Search and Rescue (SAR) zone (which may exceed a state’s territorial waters, and is defined in the 1979 International Convention on Maritime Search and Rescue). However, the fundamental tenet of law according to which the responsibility for security is ultimately allocated to public authority doesn’t hold on the high seas. Here, private vessels are required to respond to each other’s signals of distress. The sources of this private duty of rescue at sea are in treaty law and – perhaps even more importantly – in customary international law.

Stepping out of sovereignty and into the high-sea commons, the construction of relations between private and public authority transforms. Law defines quite differently the ways in which persons are implicated by each other’s claim upon life. A ship’s flag carries significant legal consequences, seemingly anchoring the ship back to a terrestrial, state-centered, legal order. Yet by imposing mutual duties of security upon individuals, law is radically privatized at sea. This dual division of labor between public and private authorities, at land and at sea, is at the very basis of international law, and is constitutive of law more generally.

Against the backdrop of this conceptual architecture, the Death by Rescue report provides an invaluable case study. It reflects how this modern legal division of labor is put under pressure in the context of contemporary crises – most importantly the so-called “refugee crisis.” For example, what happens to the legal ordering of the sea when one sovereign disintegrates? Framed in these terms, Death by Rescue opens a question about the very foundations of international law.

A Maritime Legal Black Hole

When Libya collapsed back in 2011, after a western coalition intervened, its own SAR zone became a kind of legal black hole. Private duties among seafarers failed to provide safety for the many migrants that Libya’s fall triggered. Starting from October 2013, the Italian government launched Mare Nostrum. This was an unprecedented rescue operation beyond Italy’s SAR zone. But Mare Nostrum provided no magic solution. Whether “pushed” to the sea by the increasingly violent conditions in Libya, or “pulled” by the hope of being saved, the operation saw a continued swell in migrant departures, and continued deaths at sea. Smugglers observed fewer safety precautions, and in some cases exposed their passengers to threats and violence.  A narrative emerged among some European policymakers that the rescue operation in fact reflected a danger for the migrants. By creating incentives for embarkation, it encouraged people to travel to their own deaths.

This reasoning solidified into a change in policy when Mare Nostrum was terminated, and On 1 November 2014 the more limited Operation Triton replaced it. Triton did not ordinarily patrol waters beyond Italy’s SAR zone. Any rescue operation on the high seas would have to be individually approved. The Death by Rescue report compellingly demonstrates the most chilling aspect of the decision-making process that led to this reduced mandate. High-ranking officials in Frontex knew well in advance that numerous deaths were imminent without the safety blanket of Mare Nostrum. They informed actors within the Commission of this likely consequence, yet the warning wasn’t heeded. Indeed, the realization of the likely consequences quickly became public. A day before Triton began The Guardian noted “expert” warnings that the end of Mare Nostrum “put thousands at risk.”

As one graph shows, the rate of migrant deaths increased more or less immediately. But the most dramatic part of the report addresses the so-called “black week” in April 2015: more than 1,200 adults and children drowned in the maritime space between Libya and Sicily. With no systematic high-seas rescue mission in place, and indeed no such activity in Libya’s SAR zone, the Italian authorities relied upon private seafarers. But these were untrained and ill equipped. As the report describes, private vessels initiated rescue operations on the high seas, but ended up contributing to the loss of life. The report leaves no doubt that the carnage had to be an utterly foreseeable result for several key figures, chiefly EU Commissioner Cecilia Malmström.

But how is the foundational structure of international law to blame for these results?

Even with the knowledge that Triton’s limited mandate will spell the death of numerous innocent people, it seems to me that neither Italy nor the European Union had a legal duty to continue Mare Nostrum. At the outset, launching Mare Nostrum was not the fulfillment of a legal duty. It was a discretionary response to the increasing perils that developed in the relevant maritime space. Mare Nostrum thus signaled the temporary displacement of the legal division of labor described above: the Italian state assumed an expanded, essentially public, role of security at sea. When Mare Nostrum ended, the sea was supposedly reinstituted as a space of private reciprocal duties of rescue among seafarers. As Death by Rescue explains, Italy voluntarily tried to help these private actors fulfill their roles, and indeed used them as the privatized arms of its own aborted rescue initiative. But the disaster that unfolded does not point to a human rights violation by any of the European actors involved.

Law does not recognize these casualties as victims. Under Article 2 to of the EU Charter of Fundamental Rights “Everyone has a right to life.” But the political urgency of Death by Rescue stems precisely from how it shows that this right can be detached from any corresponding duty. The insight reflects how utterly arbitrary the legal division of legal labor between public and private authorities is becoming in the gruesome spectacle of maritime migration.

Violence beyond Jurisdiction

Back in 23 February 2012, the European Court of Human Rights (ECtHR) extended human rights duties to migrants intercepted on the high seas. In Hirsi Jamaa and Others v Italy, the Court found that by pushing back a migrant vessel that left Libya, Italy had violated its obligations under Articles 3 and 4 of the European Convention on Human Rights.

The case includes an important discussion of jurisdiction. Only because the migrant boat had come under the control of Italian agents were duties under the Convention triggered. This control, the court explains, can be established either de-facto (for example by the presence of Italian coastguard officials) or de jure (for example by the flag on the intercepting boat). For personal jurisdiction to kick in, a point of contact between a state or its representatives and the individual making a human rights claim is necessary.

At the time, human rights advocates celebrated the judgment as an historic one, presumably not only because law was applied, but also because of a perceived moral triumph. For the Court to step so boldly beyond territorial jurisdiction was thought of as a step of advancement towards universal justice. But, crucially, when a court defines the limits of human rights jurisdiction – even expansive ones such as in Hirsi – it may invite states to reap the fruits of lawless violence beyond their jurisdiction.

In the last decade or so, surveillance technologies have allowed developed countries like Italy to easily transform parts of the high seas into bits of functionally sovereign territory. But because jurisdiction is still based either on territoriality or on effective control (as in Hirsi), such functional sovereignty may be bifurcated from the legal duties that would otherwise attach to it. The unaccountable violence human rights jurisdiction generates, often occurs just beyond its margins.

Maritime Migration Between Politics and Law

Whether moral blame is appropriate is a different question, the answer to which should be a resounding yes.

When it initiated Mare Nostrum, Italy demonstrated that it was able to save people outside its territorial waters (even if its efforts were not always successful). As noted, critics of Italy’s operation argued that it created a “pull factor” for migrants who sought to be saved. Behind this argument lurks the ugly notion of deterrence, which policymakers around the world have been raising for at least two decades now. The argument is that by exposing some migrants to extreme suffering, or indeed to death, western countries can dissuade others from arriving at their shores (or indeed, their borders). This idea is in direct contradiction with the most basic understanding of human dignity. But As James Hathaway and Thomas Gammeltoft-Hansen have shown, it has consistently been raised as a result of states’ attempts to enforce their borders legally. The report describes circumstances in which the international division of legal labor opened a maritime space in which a deterrent form of killing became legally available.

Under Triton, a pattern emerged of deaths in previously-patrolled waters, which were now far less accessible to assistance efforts and consequently unnecessarily deadly: “in some instances boats were left to drift for days before being detected and, once detected, it took many more hours for rescue vessels to reach the location of distress, due to their positioning much further away from the location of SAR events.” What is clear is that EU actors chose not to prevent the wanton loss of life, and in that sense, they participated in it.

That they could have done differently seems clear enough. A minimal demand would be to scale back Mare Nostrum only gradually, allowing would-be migrants to adjust to the change, and creating legal alternatives. Another option would simply be to leave Mare Nostrum in place, albeit its imperfections. A third option, advanced by the report’s authors, would be to provide a “safe passage” for migrants into Europe, presumably by escorting migrant boats and safeguarding them. Outrageously, none of these was pursued.

The proper forum for holding morally blameworthy leaders accountable is political. It remains to be seen whether such political accountability will be realized. But if the circumstances that led to these events are somehow inscribed in the structure of international law, simply “keeping things legal” will ensure that similar events continue to recur.

The author provided legal consultation to the authors of the Death by Rescue report. The opinions expressed in this post are the author’s own and are not to be attributed to any other source.

Print Friendly, PDF & Email

Leave a Comment

Comments for this post are closed


Melanie Fink says

April 20, 2016

Dear Itamar,
thank you for this thought provoking post. I am very much looking forward to reading your book.

Nick says

April 20, 2016

Interestingly,a similar obligation applies under Article 25 of the Chicago Convention on States parties with respect to aircraft.

Eliav Lieblich says

April 20, 2016

Thanks Itamar. Fascinating.

Gabriel Armas-Cardona says

April 20, 2016

I appreciate your radical approach to IL obligations. I have two questions. First, you talk about how surveillance has led to an extended area were a state is "functionally sovereign." Yet, you later quote saying that "in some instances boats were left to drift for days before being detected." The two seem to be in contradiction unless you're saying that since the technology for surveillance exists, a State has some duty (whether legal or not) to use it.

Second, I'm trying to sense the upper-bound of a state's obligations. Your general point is that it was wrong for Europe to scale down the Mare Nostrum program. I believe you think that that scale down is or should be a violation of international law. If so, then what's the rule for elaborating a state's obligation? What is the tool to show the cutoff, geographic distance? My best understanding of your work is that it would be a progressively realizable duty to *everyone*. Thus as technology improves and world crises expand, the duty would grow ever larger. This is a big jump from what we have now.

Itamar Mann says

April 20, 2016

Thanks for your questions!

First, I apologize if I've been unclear. I meant to say that while "functional sovereignty" was made possible by surveillance technologies, it existed during Mare Nostrum. The boats were left to drift after that, during Triton. At this time, it was possible, perhaps even easy, to apply such sovereignty, but that was not done.

Second, the post is premised on the proposition that scaling back Mare Notrum was indeed legal under extant law (though I do think it was wrong as a matter of moral and political judgment). That in essence is the outrage. Whether law should change is a very good question best left for another time. For now it is enough to say that we should not expect current law to solve this problem, and there are other and better political avenues for that.

Whether is “radical” or not is for others to say.

Jordan says

April 20, 2016

Xlnt post. I wonder whether estoppel theory would be operative at a certain point, when people learned of ongoing rescues and responded in reliance and to their detriment when rescues were cut back.

JR says

April 21, 2016

there is another option: the Australian way. Safe the boats and take them back to where they started. This way, you don't create perverse incentives (this is not a "narrative"; it's human beings reacting to the prospect of being brought to Europe once they reach a certain point) and you avoid the deaths of people endangering themselves.

Thomas Spijkerboer says

April 25, 2016

Thanks for this post! Two observations.
(1) You refer to Mare Nostrum as a rescue operation. However, it was something much more complex than that, as Paolo Cuttitta & others have described - it was a mixed border control/prosecution/rescue operation, and some of the ambiguities you identify are related to this mixed nature of the operation.
(2) The central metaphor you use for the people whose lives are lost is that of the drowning baby. Thereby, you represent migrants and refugees being in smuggler boats as a given fact. However, migrants and refugees are in smuggler boats not as a given fact, but AS A DIRECT CONSEQUENCE OF EUROPEAN MIGRATION POLICIES since 1990. As the available data show, border deaths have risen in synch with innovations in European migration policies. See for an overview So the central metaphor should not be a baby happening to be in the water. A more appropriate metaphor would something like this: someone is creating a situation in which an increasing number of babies ends up in the water (is throwing them in, or puts strollers on a tilted platform so they roll in - or something like that). Does this person have a legal obligation to rescue the babies who risk drowning as a consequence of his/her behaviour? And does s/he have an obligation to change his/her behaviour so as to make less babies end up in the water (or dare I suggest: none at all)? What I am trying to say is this. The direct involvement of European states in creating the risk of people drowning affects the analysis of their obligation to rescue. And it is reason to expand the analysis so as to include an obligation to investigate the connection between their policies and the increased risk; and to adapt their policies so as to make them less lethal if the connection is established.

Itamar Mann says

April 25, 2016

Thanks for these very thoughtful, excellent, comments.

JR: The Australian option is, in my view, in direct violation of international law (the principle of non-refoulement). So inasmuch as legality is important, that is not a real option. Morally, it is surely not a better one. An interesting comparison is the torture / targeting shift in US policy. There too, international law and moral imperative do not necessarily track each other.

Thomas Spijkerboer:

(1) Absolutely right. I allude to these complexities, too much to describe in this context, when I write about MR's "imperfections." Rescue was nevertheless one of the goals, which is the one I address directly.

(2) I completely agree with your underlying normative argument, but I do not think it is reflected in positive law, as it stands - even not in the most charitable reading of it. In other words, as a lawyer, I don't feel I can in good faith say that your argument is a fair interpretation of existing law. But I would love to read a detailed legal argument showing otherwise - that would be a conclusion I would in fact be much more comfortable with.

Thomas Spijkerboer says

April 26, 2016

I don't know what you teach undergrads in American law schools about babies in water. And I am perfectly happy to be relegated to the position of a non-lawyer in good faith, or that of a lawyer in bad faith (these are the two options you allow for people taking my position).

However, if you will allow me to impersonate the role of a lawyer in good faith for just a minute, I would find it implausible that the fact that an actor had an active role in the creation of a risk does not impact in any way this actor's legal responsibility for addressing the consequences, or for mitigating the risk.

Detailed legal arguments to this effect have been made, and more is to follow. Relegating them to (non-legal or bad faith legal, that remains unclear) "normative argument" requires a bit more substance than you offer in your blog post. So I am eager to see the book!

Dariusz Gozdzik says

May 15, 2016

An interesting article but most seafarers would be surprised to read that the duty of rescue does not bind them in a state's territorial waters or SAR zone. Because it most certainly does. See the International Convention for the Safety of Life at Sea, 1974 (SOLAS), Chapter V, Regulation 33 which requires "the master of a ship at sea which is in a position to be able to provide assistance, on receiving information from any source that persons are in distress at sea, is bound to proceed with all speed to their assistance......."
No distinction is made between territorial waters and high sea.

Itamar Mann says

May 16, 2016

Dear Dariusz (if I may),

Thanks for the note. The essay does not make the argument that there is no duty of rescue in territorial waters / SAR zone. Where did you see that? I'm not sure I understand.

Dariusz Gozdzik says

May 16, 2016

Dear Itamar,
As a simple sailor who has merely dabbled in the law that's how I read the passage " on the earth's oceans and seas are bound by a duty of rescue. This is not the case in a state's territorial waters....".

Itamar Mann says

May 16, 2016

Right. Mine is a misleading phrasing, and if I ever do anything further with these ideas (as I plan), I will have to clarify. What I meant to say - and thought that was clear in the context - was that in SAR areas / territorial waters the state has a direct duty of protection that it doesn't have on the high seas (see e.g. Art 15 SOLAS). Thanks for the comment - it's very useful, and important to fix that. As I see it, this does not raise questions with regard to the validity of the analysis provided. Please let me know if I'm missing something.


Dariusz Gozdzik says

May 16, 2016

Dear Itamar,
Thank you for clarifying. I look forward to reading your future posts on this subject and your book when it is published.
Kind regards,

Itamar Mann says

May 16, 2016

Thanks. And thanks again for your important comment.

Best regards,