‘Let them drown’: Rescuing migrants at sea and the non-refoulement obligation as a case study of international law’s relationship to ‘crisis’: Part I

This post is part of the ESIL Interest Group on International Human Rights Law blog symposium on ‘The Place of International Human Rights Law in Times of Crisis’.

“Approaching crises with criticism reminds us that crises are produced: they are negotiable narratives that can mask as well as reveal, a recognition that should be central when we respond to crises of human rights within international law.” Benjamin Authers and Hilary Charlesworth (‘The Crisis and the Quotidian’, p. 38)

The situation of the movement of certain migrants to and within Europe since 2015 has been described as a ‘crisis’.  The ‘crisis’ designation has been used because of the numbers involved—commonly depicted as the largest movement of people in Europe since the Second World War—and the consequent challenge of how the role of European states in assisting such people should be determined in a fair and equitable manner, in the face of sharp inequities in how things played out in practice.   A typical response from international lawyers has been to implore states to implement fully their relevant legal obligations, including in international human rights law.  Such a position is reflected, for example, in the open letter, signed by over 900 international lawyers, coming out of the 2015 ESIL conference in Oslo [I should declare I was responsible, with Başak Çali, Cathryn Costello, and Guy Goodwin Gill, in drafting and organizing the signatures for this letter].At the same time, others have drawn the opposite conclusion about the law, suggesting that legal rules were more part of the problem than the solution.  For example, in 2015 Germany partly suspended the operation of the Schengen border-free rules of EU law, on the basis that, absent a co-ordinated and equitable European approach to the situation, the cross-border free movement such rules permitted was objectionable (see here and here).

These responses epitomize the dual way international law can be and is invoked in relation to crisis: as part of the solution and as part of the problem.  In two posts I would like to explore this duality by considering the migration ‘crisis’ and the debates around one particular policy prescription relating to it: the ‘rescue’ of migrants at peril at sea performed by states acting extraterritorially, in the context of the operation of the non-refoulement obligation in human rights law.

Non-Refoulement in Human Rights Law

For some time, human rights law has been invoked to provide complementary protection for individuals from that provided in refugee law.  One important potential area of such protection concerns the obligation not to transfer individuals to circumstances where they face a risk of certain forms of human rights abuse, expressly provided for using the term non-refoulement, as in the 1951 Refugee Convention and the 1984 Torture Convention, and read into other human rights treaties, such as the 1950 European Convention on Human Rights.  The applicability of this obligation extraterritorially is of central significance to the forced migration context, given that many important forced-migration-related practices and policies by states are implemented extraterritorially, whether interception, rescue and ‘push back’ at sea, the operation of extraterritorial detention facilities to house migrants, or the posting of immigration officials at ports of exit by foreign states who wish to restrict entry to their territories.

Indeed, there is an emerging practice to invoke successfully the non-refoulement obligation in human rights law in the context of such extraterritorial forced-migration-related-policy activities, for example in the Hirsi case against Italy before the European Court of Human Rights concerning Italian maritime push-backs of migrants to Libya.

Jacques Hartmann and Irini Papanicolopulu have suggested that:

[t]he application of the principle of non-refoulement on the high seas has been claimed to create a perverse incentive for States not to conduct proactive search and rescue operations on the high seas.

The authors do not indicate who has made such a claim, but their observation invites us to consider the question of what is at stake in discussions about whether states should engage in sea-rescues of migrants, bearing in mind the operation of the non-refoulement obligation.  I propose to do this, in the light of a particular argument that has been made to resist involvement in sea-rescues (for a broader treatment of the general topic, see Itamar Mann’s Humanity at Sea, this article by Thomas Gammeltoft-Hansen and Tanja Alberts, and this forthcoming book chapter by Efthymios Papastavridis) In general, I am not going to address the ins and outs of when the non-refoulement obligation arises—what level of extraterritorial control is required—important though this is, although I will have a general observation to make about that matter.  Rather, I would like to look at some of the underlying policy implications, and to do so bearing in mind the aforementioned dual fashion in which international law can be invoked in relation to crisis.

These remarks part of a broader research project on the extraterritorial application of international human rights law, called ‘human rights beyond borders’, funded by the European Research Council, the academic research funding body of the EU.

Non-Refoulement and Rescue at Sea

The introduction of the non-refoulement obligation to sea rescues involves grafting onto the rescue objective an additional policy objective: an obligation borne by the rescuing state to ensure that the people under its control are not transferred out of this control in circumstances where they face human rights abuse.

Thus the situation is transformed from one of simply the immediate preservation of life, to a longer-term commitment to protect the individuals involved, if they cannot be sent back to where they came from due to the risk of human rights abuse there.  Indeed, such an obligation of protection covers both being sent back to this location directly, and being transferred to another location which is itself not safe, either directly, or in terms of a risk of being sent on from there to the original site of risk (or another such site). The rescue is, therefore, just the start: a gateway to an entirely new and, for the individuals involved, existentially-determinative relationship between them and the rescuing state or states.  Because of the non-refoulement obligation, the state cannot simply rescue such individuals and then transfer them elsewhere, including back to where they came from, if the risk test triggering the obligation is met as far as the location to which they would be transferred is concerned.

Incentivizing Dangerous Crossings and Smuggling Activities

It might be questioned whether this process—rescue leading to an ongoing protection requirement via the non-refoulement obligation—creates an incentive for people to put their lives and the lives of their families in danger through perilous sea crossings.  In other words, that part of the reason people are willing to take this risk is in the hope that if things go wrong, they might be rescued, and the rescue might itself realize more broadly the purpose for the crossing, in terms of obtaining long term protection, because the rescue will immediately trigger a protection relationship.  Thus sea-rescues twinned with the extraterritorial non-refoulement obligation make the migration ‘crisis’ worse, by encouraging more people to put their lives at risk.  The existence of the non-refoulement obligation is crucial here, since were it not to apply, individuals would not gain anything, in terms of their long-term protection aspirations, from the rescue, as far as a legal entitlement is concerned.  The risk calculus for making the dangerous crossings is thus radically different because of the operation of the legal regime.

For an example of the suggestion that a policy of rescues encourages dangerous crossings, in 2014, a junior minister in the UK Foreign and Commonwealth Office (the foreign ministry), Joyce Anelay, was asked in a written question in Parliament what contribution the UK would make to rescues of migrants at sea in the Mediterranean.   Joyce Anelay replied as follows:

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 attempt the dangerous sea crossing and thereby leading to more tragic and unnecessary deaths.

An aspect of arguments such as this one is that rescues twinned with non-refoulement do the smugglers job for them, thereby incentivising and enabling the smuggling industry itself.  People get from the rescuing state, in the form of long-term protection, what they had been promised as the eventual outcome of the journey by the smuggler.  Equally, the costs of risk-taking by smugglers are reduced—if things go wrong, states will step in.  (It is notable here that, as Efthymios Papastavridis reports, some Italian courts have used the doctrine of intermediary proprietor to criminalise smugglers who ‘use’ Italian rescue authorities to bring people to Italy.) So again human rights law supposedly exacerbates the ‘crisis’ by supporting the activities of the smugglers who are viewed as the key agents driving it.

The first thing to say about these arguments is that their general logic would apply not only to extraterritorial rescues twinned with the non-refoulement obligation, but also to the operation of the non-refoulement obligation in the territorial context, i.e. when individuals have managed to reach or cross the state’s territorial borders.  Smuggler-enabled perilous journeys to seek protection are of course aimed ultimately at entry into the territory of the state against whom protection will be sought on the basis of invoking the territorial application of the non-refoulement obligation binding on that state.  To charge the existence of such a non-refoulement obligation with incentivizing dangerous journeys and rewarding smugglers is to make a point about this obligation in general, whether in its territorial or its extraterritorial operation.  If, then, human rights law needs to be modified in this way to prevent its supposed causal effect on dangerous, smuggler-enabled journeys, then, actually, the non-refoulement obligation needs to be scrapped in its entirety, both territorially and extraterritorially.

That said, the extension of the non-refoulement obligation to the extraterritorial context, when allied to the policy of engaging in rescues at sea, potentially attenuates the risk people face, and the efforts smugglers have to make. Smuggler-enabled dangerous journeys can be shorter than they would be otherwise if the individuals enter a zone of protection before they reach the territory of the destination state.

Insofar, then, as it is helpful to understand the non-refoulement obligation in terms of incentivizing dangerous journeys and rewarding smugglers, the extension of this obligation to the extraterritorial arena potentially renders this effect more potent.

In her seminal piece on international lawyers and crisis, Hilary Charlesworth reminds us that the discipline of international law ‘does not encourage the weighing up of competing versions of events…What we [international lawyers] glean…as ‘facts’ may be inaccurate or partial and the way we report and emphasize them is an act of political interpretation’ (p.384).  She also observes that the ‘crisis’ approach can also involve concentrating on a single event or series of events and missing the ‘larger picture’ (id.).

Bearing these cautionary observations in mind, in the second part of this post I will appraise the merits of the supposed insight regarding the incentives behind dangerous crossings by placing the supposed insight in a broader context and considering other causal factors.