The false panacea of offshore deterrence

Governments take often shockingly blunt action to deter refugees and other migrants found on the high seas, in their island territories and in overseas enclaves.  There is a pervasive belief that when deterrence is conducted at arms-length from the homeland it is either legitimate or, at the very least, immune from legal accountability.

For example, the US maintains that it has no legal obligation to intercepted refugees, even if they manage to reach its territorial sea. Indeed, the US recently argued that a Cuban asylum seeker – traditionally a highly favoured group under its domestic law – could not assert a right to protection because the bridge where her tiny boat landed had been disconnected by storms from the American mainland.

When some 10,000 persons managed to reach the Italian island of Lampedusa this year, Italy responded by discontinuing its traditional practice of sending them to Sicily for processing of protection claims. Instead, the BBC reports that the “migrants were dispatched back handcuffed in military planes from Lampedusa direct to Libya. No questions asked.”

Spain erected dual razor-wire fences around its North African enclaves of Ceuta and Melilla to deter groups of largely sub-Saharan migrants anxious to enter the European Union. Even those who successfully scaled the barriers were often summarily sent back to Morocco, which is reported simply to have dumped them in desert border zones. The ‘success’ of this deterrent programme put renewed pressure on the Spanish Canary Islands, a favoured destination until 2002 when radar and sea patrols were instituted to deter travel from Morocco to the Canarian islands of Fuerteventura and Lanzarote, some 100 kilometres away. The most recent flows have thus been forced to take a much longer and more perilous route from northern Mauritania to Tenerife. The Spanish government has responded to the upsurge in arrivals by offering Mauritania patrol boats to stop departures and to set up refugee camps in Mauritania.

Are such practices legal?

The 1951 Refugee Convention and its 1967 Protocol do not allow states to refuse protection to refugees just because they have not yet entered the core of its territory. Simply put, the most basic duties – including the critical duty of non-refoulement, requiring states not directly or indirectly to return refugees to the risk of persecution – apply wherever a state exercises jurisdiction. Whether protection is sought on Lampedusa or in Rome, the refugee law implications are identical. It makes no difference whatever if asylum is claimed by a refugee clinging to the outermost razor-wire fence at Ceuta or at a police station in Madrid. Nor may there be any peremptory refoulement of refugees encountered by vessels patrolling a state’s territorial waters, or even of those intercepted on the high seas. Because jurisdiction is the lynchpin to responsibility, state parties to the Refugee Convention must provisionally honour the rights of persons under their authority who claim refugee status until and unless they are fairly determined not to qualify for protection.

Despite the clarity of these legal rules, two kinds of argument are made in support of deterrent measures.

The first is that insistence on rigorous respect for the rules of refugee law amounts to allowing the proverbial tail to wag the dog. Because in any given flow towards the developed world today refugees are significantly outnumbered by economic migrants – whose entry can normally be lawfully resisted – it is argued that governments must be free to respond effectively to the dominant (non-refugee) character of the arrivals.

As a matter of law, though, non-selective deterrent measures cannot be justified where genuine refugees are part of a mixed flow. There is no exception to the duty of non-refoulement for situations in which the cost or inconvenience of processing claims is great, or where only one in ten entrants is actually a refugee. Nor can states lawfully avoid refugee protection obligations by deciding simply not to assess claims made to them. As UNHCR rightly insists, a refugee does not become a refugee because of recognition, but is recognised because s/he is a refugee. In practice, this means that a person who may be a refugee must be provisionally treated as such until and unless he or she is fairly determined not to qualify for refugee status. Measures which deter refugee claimants from arriving in an asylum state are therefore no less in breach of refugee law than is the removal of a recognised refugee already present in a state’s territory.

A second and more complex argument for deterrence is sometimes made on humanitarian grounds. Particularly where refugees and others arrive by sea, often in rickety or grossly overcrowded vessels, it has been said that departures must be stopped in order to avoid risk to life or limb.

There is, however, a critical legal distinction between sensible efforts to provide information and to make it difficult for traffickers to exploit people on the one hand, and more aggressive efforts actually to stop departures on the other. Whatever the risks, every person has the legal right to make the decision about departure for him or herself. The relevant rule in such cases is not rooted in refugee law but in the requirement in the International Covenant on Civil and Political Rights[1] that all persons be allowed to leave any country, including their own. Allegedly humanitarian steps taken to shut down escape routes – such as the formal agreement between the US and Cuba in 1994 requiring Cuba to “... take effective measures in every way it possibly can to prevent unsafe departures using mainly persuasive methods” – are unlawful and paternalistic. It is the refugee’s right – not the prerogative of any state or humanitarian agency – to decide when the risks of staying put are greater than the risks of setting sail.

Until and unless the abuse that causes refugees to flee in the first place is ended, the only real answer is to provide safe alternatives to unsafe routes of escape. While blunt deterrence of refugee or mixed flows is unlawful, states are perfectly free to conceive creative protection alternatives. Most sensibly, the focus should be on the establishment of genuine protection options within regions of origin. Where intra-regional alternatives are truly safe and accessible and deliver rights-based protection, it is likely that most refugees will feel no need to undertake perilous voyages. Indeed, where protection options that meet international legal standards are declined for economic, social or other reasons not related to protection, refugees who travel farther afield may lawfully be returned to their own region. For this reason, a re-emphasis on making real protection available closer to home should be attractive to developed states: while less ‘efficient’ than (unlawful) deterrence, it is, nonetheless, consistent with their more general migration control objectives. It is also of real value to states in regions of origin, which desperately need binding guarantees of substantial resources to cope with endemic refugee flows. Most critically, it would enhance the welfare of the overwhelming majority of refugees not able or willing to flee beyond their own region.

Discussions along these lines are, of course, already occurring. There is clear interest in exploring both the operational flexibility which refugee law affords, and the value of systems to share out both the responsibilities and burdens inherent in refugee protection. It is not at all clear, however, that present initiatives are based on finding practical ways by which to respond to involuntary migration from within a rights-based framework. Potentially lost in the discussions as they have evolved to date is the imperative to reform the mechanisms of refugee law not simply to avert perceived hardships for states but also in ways that really improve the lot of refugees themselves. If the net result of reform is only to lighten the load of governments, or to renew the capacity of international agencies to meet the priorities of states, then an extraordinary opportunity to advance the human dignity of refugees themselves will have been lost.

The challenge, then, is twofold. Most obviously, we must flatly reject the legitimacy of generalised deterrence which can block refugee flight, including even deterrent measures prompted by genuine humanitarian concern. Second, we should embrace the opportunities which reform of the mechanisms of refugee law affords both to save lives now risked in the flight to asylum and to improve the quality of protection for all refugees in the world, wherever located.


James Hathaway is James E. and Sarah A. Degan Professor of Law, and Director of the University of Michigan’s Program in Refugee and Asylum Law ( His most recent book is The Rights of Refugees under International Law, Cambridge, 2005[2]). Email:



Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at