Categorie archief: Thomas Blogs

A less disastrous policy on the war in Yemen (2016)

From: Constant: 8 x la guerre (1951)

Since 2015 an armed conflict takes place in Yemen. One of the warring parties is supported by an international coalition headed by Saudi Arabia, which is provided with arms by the US and European states. Violations of international humanitarian law (including bombing hospitals) have become routine. Out of a 24 million population, 14 million people are now reliant on food aid, and 3 million people are internally displaced. Over 180.000 people have fled the country. The number of refugees that has left the country is low because Yemen is surrounded by other conflict zones (Somalia, Eritrea, Sudan) and by Saudi Arabia, which blocks the escape routes in the desert by military means. Just like with other refugee situations, the UN reports that only half of the funds needed for supporting displaced persons and refugees is raised by the international community.

There are hardly Yemenite refugees who succeed in reaching Europe – over the past five years 3.845 people from Yemen applied for asylum in the entire European Union, 150 of which in the Netherlands. This is a very limited group of people. When the armed conflict in Yemen broke out in 2015, the Dutch state secretary of justice decided to suspend decision making in Yemenite asylum cases, by declaring a so-called moratorium. European and Dutch law allows for this so as to enable the authorities to gather information and formulate policy in situations which are volatile and changing – as was the case in Yemen in 2015.

The Dutch moratorium for Yemenite asylum cases expired in August 2016. According to law, the authorities should now resume taking decisions in Yemenite asylum cases. However, in October the state secretary for justice reported to parliament that he is working on new policy. The idea of the moratorium is that the authorities will use the period of the moratorium to formulate policy. Instead, the moratorium has been used to wait with making new policy until after it has expired.

If the state secretary of justice would have taken decisions in Yemenite asylum cases, he would have granted asylum. The situation in Yemen leaves no other options. During a moratorium, asylum seekers have to remain in asylum reception centres. These are fit for sheltering people for a few months. Not for the 18 to 24 months that Yemenite asylum seekers have been forced to remain there, while the inaction of the Dutch authorities forced them to do nothing but wait.

This makes Yemen a miniature of failing European policies:
1. By supporting and arming Saudi Arabia, the conflict is made worse;
2. The resulting humanitarian emergency is addressed only partly by insufficient funding;
3. And even the few refugees who succeeded in reaching Europe are left in a dead-end situation for an indefinite period.

Because this is a relatively small problem, Yemen could serve as a try-out for a less disastrous policy on three fronts. The Dutch authorities themselves can grant asylum to Yemenite refugees, for the sound reason that they qualify for it. Together with other European countries, the funds could be collected for supporting internally displaced persons and refugees in the region. And by ending arms supplies to the warring parties Europe could stop making the war worse than it already is.

Dutch version published in Trouw, 21 November 2016

Fact Check: Did ‘Wir Schaffen Das’ Lead to Uncontrolled Mass Migration? (2016)

From: Constant: 8 x la guerre (1951)

From: Constant: 8 x la guerre (1951)

It is widely assumed that the German ‘decision to suspend Dublin and open the borders,’ epitomized by Angela Merkel’s ‘Wir schaffen das’, led to ‘uncontrolled mass migration’ in the summer and fall of 2015. This belief is not only held by media and politicians, but also by prominent academics like Ruud Koopmans. The empirical claim contained in this belief has two elements: one relating to the ‘suspension of Dublin’, and one relating to the ‘opening of the borders.’ Are these claims correct?

Read my blogpost on Border Criminologies.

Fact Check: Did the EU-Turkey Deal Bring Down the Number of Migrants and of Border Deaths? (2016)

From: Constant: 8 x la guerre (1951)

From: Constant: 8 x la guerre (1951)

On March 18th, 2016, the EU and Turkey agreed that all irregular migrants arriving on the Greek islands were to be returned to Turkey. It is generally believed that this agreement has led to a dramatic decrease of the number of irregular migrants, and hence of the number of border deaths. Exemplifying this idea is this quote from the second progress report of the European Commission on the implementation of the EU-Turkey Agreement: ‘The sharp decrease in the number of irregular migrants and asylum seekers crossing from Turkey to Greece is proof of the Statement’s effectiveness – and in particular, that the business model of smugglers can be broken. The clear message to migrants is that getting on a boat in Turkey, and endangering lives in the process, is not worth the risk given that there is a legal and safe pathway through resettlement.’ Are these claims correct?

See my blogpost on Border Criminologies

Advocating human rights as gorilla behaviour (2016)

media_xll_1212576Human rights are increasingly considered to be crucial for the legitimacy of European institutions. Since 1983, the Dutch Constitution opens with a chapter on fundamental rights, in the first place equality and non-discrimination. The 1998 Human Rights Act in the United Kingdom is another case in point. The European Union proclaimed its Charter on Fundamental Rights in 2000, and made it into binding law in 2009. These are formal phenomena, but they have effects in everyday life. People who want to make a point about Islam will easily refer to equal rights of women and LGBT people, which they are purported to lack and we are said to have. This happens in politics, at parties and in the pub.

Therefore, it is remarkable that human rights are being relativized more and more frequently as well. The Dutch prime minister, Mark Rutte, last Friday had to explain why he had not publicly raised the issue of freedom of expression when a Dutch-Turkish journalist had been arrested in Turkey after critical tweets about Erdogan. He defended himself by stating that behaving like a gorilla is not going to get us anywhere. This kind of statement is not an isolated thing. Over the past few weeks, human rights organizations Amnesty International, Human Rights Watch and the Syrian Observatory for Human Rights have published well researched reports about serious violations of the rights of Syrian refugees – from systematic refoulement to shooting them, including women and children. Prime minister Rutte and Dutch foreign secretary Koenders have systematically referred to these reports as “rumours”. They have made enquiries with the Turkish authorities, who have assured them that there is nothing whatsoever to worry about. There is “no reason at all to point a reproaching finger at Turkey.”

It might be objected that these statements are a bit awkward, but are related only to the refugee deal between Turkey and the EU. Because the Netherlands is President of the EU Council these months and the deal is very important, human rights have to be toned down. But this ignores that this is merely an escalation of a more long term process. Two years ago, Geert Wilders promised to make sure that there would be less Moroccans in the Netherlands. He now leads the polls, and faces criminal prosecution in court. When the Dutch legislation on family reunion was made stricter, there was wide support in Parliament for proposals which explicitly aim at reducing family reunion for people with a Turkish and Moroccan background. Why does Wilders face prosecution for something the Dutch legislator actually makes happen?

Human rights, like law in general, are always multi-interpretable. You can take them in different directions. The question, however, is whether they can be taken in different directions at the same time. That is what politicians of all shades are trying to do now, by turning up the volume on both points. They say ever more loudly how important human rights are. And they emphasize more and more that human rights are overdone – with the gorilla metaphor as a high point until now. The dissonance is increasing, and just as in music it will have to be resolved at one moment. Either Europe joins the likes of Putin and Erdogan, who find human rights silly nonsense. Or they join the ranks of people like Max van der Stoel, a former Dutch labour Party politician who, without ever raising his voice, unflinchingly advocated human rights.

Dutch version in NRC-Handelsblad 3 May 2016

Is the EU-Turkey refugee and migration deal a treaty? (with Maarten den Heijer, 2016)

Maarten den Heijer & Thomas Spijkerboer

In the European Parliament, questions were asked about the legal nature of the EU-Turkey Statement of 18 March, pursuant to which Greece has started to return asylum seekers to Turkey this week. Apparently, the EU’s procedure for negotiating and concluding treaties withIMG_0321 third countries, laid down in in Art. 218 TFEU, has not been followed. The European Parliament wants to know whether the Council nonetheless considers the Statement to be a treaty, and, if not, whether Turkey has been informed about the non-binding nature. Importantly, for treaties “covering fields to which the ordinary legislative procedure applies” (asylum and immigration is such a field), the Council may only conclude a treaty with a third country after obtaining consent of the European Parliament (Art. 218(6)(a)(v) TFEU).

The full blog post was published on EU Law Analysis.

Wij hebben gewonnen (2016)

tegelWij, een overdonderende meerderheid van de kiezers, hebben gisteren gebruik gemaakt van ons democratische recht om niet naar de stembus te gaan. Vrijwel altijd geven wij trouwhartig gehoor aan de oproep om te gaan stemmen. Tot die malle deelraadverkiezingen aan toe. Maar gisteren hebben wij het vertikt om onze schoenzolen nodeloos te verslijten. We hebben namelijk, als we dat ergens goed voor vonden, al gestemd voor het Europees Parlement. En de Tweede Kamer vertegenwoordigt ons prima. Soms vinden we bij nader inzien van niet, en dan stemmen we de volgende keer lekker anders. Maar een handelsakkoord met de Oekraine? Straks moeten we nog gaan stemmen over de dienstregeling van lijn tien! Begrijp me niet verkeerd: we vinden de vertrektijden van de tram enorm belangrijk, maar die kunnen beter worden vastgesteld in een samenhangende afweging met allerlei andere dingen. We willen gewoon met de tram mee, en verder moet er niet aan onze kop gezeurd worden.

En onze kiezerspas? Laten we die opsturen aan het partijbureau van D66, die ons met dit gedoe heeft opgezadeld. Dan kunnen D66’ers op het kleinste kamertje wat antiparlementaire viezigheid van hun achterwerk vegen.

7 april 2016

The systemic failure of the Common European Asylum System, as exemplified by the EU-Turkey deal

Yoko Ono: Ex It, 1997-2000, Istanbul Museum of Modern Art Collection

Yoko Ono: Ex It, 1997-2000, Istanbul Museum of Modern Art Collection

Maarten den Heijer, Jorrit Rijpma & Thomas Spijkerboer

18 March 2016

European politicians are doing lots of things to get the number of refugees under control – during the last few weeks mainly by wheeling and dealing with Turkey. The EU itself is cracking at the seams, Schengen may not survive, and a number of states is no longer prepared to abide by its international legal obligations. Yet the number of refugees in itself cannot be the problem. They constitute some 0,3% of the European population. Europe wants to keep refugees in Turkey, Jordan and Lebanon, yet these countries host 10 to 100 times as many refugees as Europe does. How can there be a crisis nevertheless?

Four system errors

In the past months, four system errors in the European asylum system have come to light. The first one is internal. The Dublin system distributes the responsibility for asylum applicants among EU member states, but it does so unfairly. If asylum seekers would be distributed in conformity with the Dublin system, Greece and Italy would be saddled with the vast majority of them. For a long time, these countries have tried to change the rules, but northern member states were not prepared to do so because the rules suited them just fine. The impotence – and at times unwillingness – of southern member states to maintain the system became evident. The Greek asylum procedure eventually turned out to be so substandard that it would be a violation of basic human rights to return asylum seekers there. Italy did little to prevent asylum seekers from travelling onwards. Because of the successful sabotage of countries along the Balkan route, countries like Germany and Sweden now discover how unfair the system can work out. Suddenly they do see the point of changing the rules, but now they face unwilling eastern member states.

The system is unfair for asylum seekers, too. There is European legislation in order to harmonise asylum law. But the chances of being recognised as a refugee can be twice as high in one member state compared to another one; and reception may function quite well in one country while in another people are sleeping in the streets. Refugees not only have a formal entitlement to international protection. They are also dearly in need of it. Just like the member states, they often have a substantial and reasonable interests in frustrating the system.

That is not allowed. The law is the law. So if member states and asylum seekers do not obey the rules, they have to be forced to do so. That is easier said than done. Rules that can only work through enforcement because key players find them unreasonable – this is a mission impossible in any case. This is true in particular between the EU and member states. The EU can only work via the member states, and therefore is rather defenceless in the face of obstruction.

The second system error consists of the great expectations of what borders can do – both as a matter of fact and as a matter of law. Borders cannot stop each and every refugee. The border between North- and South-Korea is pretty watertight, but does not only stop human beings but any form of economic traffic. Europe has enormous interests in smooth circulation of capital, goods and people. Systematically checking everyone and everything is possible in theory, but has enormous costs in terms of both finance and principles. People who propose to do this should mention the 10 or 20% of net income people would have to part with. It would require minefields. In addition, it would have a waterbed effect: when one route is closed, another will open. Usually the new one is longer and more dangerous and deadly. Also, the more borders are controlled, the more demand for smuggling is created.

From a legal perspective, it is also hard to make borders impermeable. People can only be returned to the country they came from if (1) that country is safe and (2) there is an opportunity to ask for asylum in individual (appeal) procedures. In the past, efforts have been made to sidestep these legal requirements, for example by picking up people in international waters and returning them (‘pushbacks’). These tricks are of the same calibre as Guantanamo Bay or the Russian secret service which thinks it actually is allowed to kill opponents with polonium provided they do so in London. The Grand Chamber of the European Court of Human Rights unanimously rejected such tricks in its Hirsi Jamaa judgement. Outside its territory, a state is not suddenly allowed to do things which it is definitely not allowed to do on its own territory.

The third system error concerns the schizophrenia in the external dimension of European asylum law. Refugees are granted asylum if they succeed in entering Europe, but European states use all means available to prevent from them getting in: visa requirements, airlines are forced to check documents before embarkation, transit countries are pressured to introduce visa requirements. By now, even Jordan, Lebanon and Turkey require visa from Syrian refugees. This did not lead to less refugees, but to more smuggling, because Syrians did not wait until their home was hit by a bomb too. Smuggling was so successful that, for example, Eritreans who had remained in Sudan for a decade thought: now or never. So: refugees were prohibited from leaving their country. This led to a more chaotic and probably a bigger migration flow than would have been the case if Syrians had been allowed to seek refuge. In addition, reception in the region was badly overburdened and absolutely underfinanced – in the past few years, only half of the money needed was available for the UN.

These first three system errors concern the content of policy. The last one concerns the form of cooperation, which is somewhere between complete Europeanization and classic intergovernmental cooperation. Asylum policy is based on minimum harmonization and mutual recognition of (negative) decisions. But implementing the legislation seeking to achieve that is up to the member states. The EU has no executive competences. Such a form of cooperation can only work if member states really want to work together because they have a common vision. That political will and common vision are nowhere to be seen.

The EU response

The Syrian refugees form the first stress test for the European asylum system. And what does the European response consist of? The first, internal system error is not addressed. Redistributing asylum seekers from Italy and Greece to other member states did not get beyond a few hundred people. The Dublin system will be revised, but without addressing the fundamental unfairness. A few extra rules of marginal importance may make the system a tiny bit less unfair, but will most definitely make it more complicated. Member states will still have an interest in sabotaging the system. The same is true for asylum seekers, because the important differences between national asylum systems remain.

The deal with Turkey is typical for the second and third system error. Boat refugees are to be returned straight away, in return for which some (which?) European countries would be prepared to accept some (which?) Syrians from Turkey. As long as the EU does not lift the visa requirement for Turkish nationals, it is entirely unclear why Turkey would have an interest in actually doing this. In addition, Turkey is absolutely not a safe country. The last time Turkey was condemned by the European Court of Human rights for inhuman treatment of an asylum seeker was 15 December last year. So refugees will still have a legitimate interest in continuing their trip to other countries. The main problem in Turkey is not a lack of money, but the lack of a proper functioning system of rule of law and the connectedness of refugees with vital political issues such as the Kurdish question. And why would redistributing refugees from Turkey to European countries work, while for Italy and Greece they did not get beyond a few hundred? Also, of course refugees will try other routes. In sum: the reaction to the Syrian refugee inflow consists of a continuation of the system errors which created the problem to start with. Externally, prohibition remains the dominant approach towards refugees; there are unrealistic expectations of wat borders are able and should be able to achieve; and the unfairness for states and refugees remain. The fourth system error – an unclear form of cooperation which only works if states really want it to work – is not addressed either.

What is the alternative?

Just like the banking 2008 crisis could not be solved in a few weeks or months, the number of refugees cannot be brought under control as long as the conflict in Syria has not been resolved. For the short term, Europe will have to clean up the mess it has created by 25 years of ill conceived policies. Two things are important for this. First, states have to be prepared for high numbers of refugees, not because we are Gutmenschen but because it is better to be prepared for things that are bound to happen. Secondly, an effort has to be made to undermine the lively smuggling economy by unconditional resettlement of large numbers of Syrian refugees from (in order of urgency) Lebanon, Jordan and Turkey. This will reduce the demand for smuggling, and it will enable those countries to host a considerable number of refugees in an acceptable manner.

For the longer term, Europe has to engage in a fundamental reconsideration of its asylum policy. The internal unfairness, the excessive expectations of the border, and the prohibition of flight have to be abandoned. Which topics will be subject to which forms of cooperation has to be assessed in a comprehensive manner, and not in impromptu deals on details. An encompassing refugee policy has to be based on the following principles:

  • Refugees do not only have the right to flee violence and oppression; they will do so as a matter of fact, whether we like it or not;
  • Neighbouring countries offer great opportunities for protection, but have to be generously funded for that; large scale, unconditional resettlement is necessary if countries are overburdened like Jordan and Lebanon;
  • Borders can be used for checking who gets in, and for assessing who is to be kept out; if they are expected to do more, this results in irregular migration and death;
  • The European asylum system has to be fair for member states and refugees.

A policy that is based on these principles would make an end to the unrealistic governance fantasies on which present day policies are based. And they would ensure that the principles on which Europe is based are not treated like garbage.


Maarten den Heijer is assistant professor of international law at the Universiteit van Amsterdam; Jorrit Rijpma is associate professor of European Law at the Universiteit Leiden; Thomas Spijkerboer is professor of migration law at the Vrije Universiteit Amsterdam.

A Dutch version of this text was published in NRC-Handelsblad 19 March 2016

The NATO pushbacks in the Aegean and international law (2016)

On Thursday 11 February 2016, NATO announced that its ships would immediately be deployed in the Aegean in order to combat irregular migration, in cooperation with the relevant authorities and with Frontex. On 23 February, NATO Secretary-General Stoltenberg stated in the European parliament: “When we rescue those people, what we agreed with Turkey at a ministerial level, we agreed that if those people came from Turkey then we can return them to Turkey.” Stoltenberg repeated this on 24 February. Is this compatible with international law? reported the following on the NATO plans. A group of five vessels (from Germany, Italy, Canada, Turkey and Greece) already are present in the Eastern Mediterranean. Denmark

From: Constant: 8 x la guerre (1951)

From: Constant: 8 x la guerre (1951)

and the Netherlands are said to have promised vessels too. Stoltenberg said that Turkey and Greece will not operate in each other’s territorial waters, thereby addressing a political sensitivity. The mandate of the mission will not be to intercept or to return boats, but to engage in search and rescue (which however, as Stoltenberg made clear on 23 and 24 February, may consist of interception and return). Activities were to take place in Turkish territorial waters. The Guardian reported that the action was to start on 12 February. Operational reports on NATO activity in the Aegean is lacking so far.

A number of questions is relevant in order to assess the legitimacy of this in light of international law: are NATO member states exercising jurisdiction; are there international law objections; and are there ways to evade jurisdiction and international law obligations?

Do NATO member states exercise jurisdiction?
States are bound to international law when they exercise jurisdiction. If – for example a German – vessel picks up people while it is in Turkish territorial waters and brings them to the Turkish shore? This is not an irrelevant question. If such a German vessel exercises jurisdiction, Germany has a number of international law obligations, relating inter alia to asylum. The issue of jurisdiction has been the subject of a number of cases.

One of the first cases on the issue is a decision of the UN Human Rights Committee from the early 1980s. The case of Burgos v Uruguay concerned a Uruguayan refugee who enjoyed asylum in Argentina. After the military coup in Argentina, he was kidnapped by the Uruguayan secret service, detained in Argentina for two weeks, and transferred to Uruguay where he was tortured. The question arose whether Burgos was under Uruguayan jurisdiction during his initial arrest. The UN Human Rights Committee held that it would be unconscionable to interpret the International Covenant on Civil and Political Rights in such a manner that a State would be allowed to perpetrate acts on the territory of another State which it would not be allowed to perpetrate on its own territory. The Human Rights Committee formulated a fundamental rule: what a state is prohibited from doing on its own territory, it is not allowed to do somewhere else.

A case of the Committee Against Torture displays more similarities with the NATO plan. The Marine I-case concerned a cargo vessel carrying 369 migrants, which in 2007 issued a rescue call in international waters. A Spanish rescue vessel approached the Marine I and towed it to the Mauritanian coast. After a week and a half of negotiations , the Mauritanian authorities gave permission to tow the vessel into a Mauritanian harbour. The migrants were detained under supervision of Spanish personnel. In groups most of them were returned to their country of origin; a few received a humanitarian residence permit. The complaints concerned detention conditions and removal to the country of origin. One of the arguments brought forward on behalf of the Spanish authorities was that all this occurred outside Spanish territory. The Committee Against Torture held that a state exercises jurisdiction when it has, directly or indirectly, in whole or in part, de jure or de facto effective control. The Committee ruled that Spain exercised jurisdiction from the moment is came to the rescue of the Marine I.

The most directly applicable case is the 2012 Hirsi Jamaa judgment, in which the Grand Chamber of the European Court of Human Rights passed judgment on the Italian pushbacks, which consisted of transferring migrants from vessels onto Italian navy vessels and returning to Libya without any procedure. The Court held that a state exercises de jure jurisdiction over vessels flying its flag, and therefore the migrants were under Italian jurisdiction. It added that Italy could not evade the exercise of jurisdiction by arguing that its activities constitute a search and rescue action – just like NATO is doing at present.

The Court referred to another case, in which it held that French agents who took over a vessel (suspected of engaging in drugs trafficking) flying a Cambodian flag exercised de facto jurisdiction. In conclusion, the Italian authorities had exercised jurisdiction during the pushbacks because the migrants were put on Italian vessels (de jure jurisdiction) and because the Italian navy had factual control over them (de facto jurisdiction).

Legal doctrine holds that the same applies to the 1951 Refugee Convention. It has to be noted that the US Supreme Court ruled that the Refugee Convention does not apply outside the territory of the United States. However, this interpretation of the Supreme Court is highly contested, and has more to do with US constitutional law (in particular with the plenary powers doctrine) than with international law. The Supreme Court projected a piece of domestic constitutional law onto the international convention on refugees. It is not to be expected that European courts will change their position in order to adopt this American interpretation, although that cannot be strictly excluded either.

In sum: it is evident that, when they return migrants to the Turkish shore, NATO vessels exercise jurisdiction in the sense of the European Convention on Human Rights, the International Covenant on Civil and Political Rights, and the Convention Against Torture – even when the entire operation takes place within Turkish territorial waters. That this is evident is clear from the fact that the judgment of the Grand Chamber of the European Court of Human Rights was unanimous on all major points. This underlines that the Court’s interpretation is not far-fetched or activist, but reflects a broad consensus. The Court did nothing else than restate the obvious and basic rule formulated in Burgos v Uruguay: what a state is not permitted to do on its own territory, it cannot do somewhere else. This is a fundamental rule. If the US is not permitted to waterboard people, it is not permitted to do so on Guantanamo Bay either. If the Russian secret service is not allowed to poison an opponent with polonium, it is not allowed to do so in London.

Can states evade jurisdiction?
It would be conceivable to use a construction with Spain is said to apply in its cooperation with some West-African countries. Image that on board of all participating NATO vessels a Turkish representative is posted who – even when s/he is taking a nap – is supposed to command the vessel. This would clearly be a mere construction to hide the jurisdiction the NATO state concerned is exercising under a formalist veil. But even for those who wish to go along with that, it would merely mean that the de jure jurisdiction of that NATO state would disappear. It would not do away with its de facto exercise of jurisdiction. The presence of a Turkish functionary could possible mean that Turkey would exercise de jure jurisdiction in addition to the exercise of jurisdiction of the NATO state concerned. The fact is that it is hard to think of a clearer example of the exercise of jurisdiction that a state has over its own navy vessels.

What are the relevant international law obligations?
A has been explained elsewhere, migrants who may want to ask for asylum can be returned to a third country, on the condition that
• The third country is safe;
• It has been established in an individual decision that the country is safe for this person as well;
• And the migrant has had the possibility to appeal this decision in a court of law.
As has been explained elsewhere, it is highly questionable whether Turkey is to be considered as a safe third country. The last time the European Court of Human Rights concluded that Turkey violation the human rights of an asylum seeker dates from 15 December last year. In addition to this, the NATO action excludes the possibility of individual decisions and appeal to courts. Therefore, the NATO action would be contrary to the prohibition of refoulement. The prohibition of refoulement contained in the ECHR, ICCPR and CAT is non-derogable.

Can NATO member states get rid of their obligations?
What NATO plans to do is contrary to the European Court of Human Rights (Hirsi Jamaa), the Convention Against Torture (Marine I) and the International Covenant of Civil and Political Rights (Burgos). States may denounce (i.e. stop being a party to) most conventions (the ICCPR does not foresee denunciation), but that is not an easy thing. EU law requires that its member states are party to these conventions. Therefore, denouncing them would require amendment of EU treaties (as well as of all secondary legislation referring to the prohibition of refoulement).

An additional problem would be that the prohibition of refoulement which the NATO member states would violate is part of international customary law, according to most authors. International customary law binds states even if they have not ratified any international treaty. So maybe it would not help to denounce all these conventions. On the other hand, one might argue that that the idea that the prohibition of refoulement is customary law is based primarily in the fact that it has been incorporated into a number of treaties (in addition to the ones already referred to also in regional treaties in Africa and the Americas). So s all NATO-states withdraw from these treaties, one might argue that customary law has changed.

The conclusion has to be that the NATO actions are in violation of international law; and that the relevant parts of international law are binding on NATO states because they exercise jurisdiction over migrants. Returning migrants to Turkey as envisioned violates the prohibition of refoulement, also when it happens in the form of search and rescue.

This blog was amended on 26 February 2016 in response to a tweet by Martin Scheinin. A previous version of this blog in Dutch was posted on Verblijfblog on 12 February 2016.