17 February 2020

A Restrictionist Revolution?

A Counter-Intuitive Reading of the ECtHR’s N.D. & N.T.-Judgment on ‘Hot Expulsions’ at the Spanish-Moroccan Border

In times of twitter and social media, we get used to quick reactions and clear-cut opinions that lend themselves to intuitive approval or rejection. Not surprisingly, the immediate response to the Grand Chamber’s N.D. & N.T. judgment rectifying the Spanish policy of ‘hot expulsions’ of irregular migrants was met with ‘shock’ – a ‘slap in the face’ of human rights law that ‘refutes the raison d’être’ of the European Convention on Human Rights (ECHR). These first analyses are correct insofar as they express the utter disappointment of the authors at the immediate outcome of the case and the initial conclusion that judges backtracked from an earlier dynamic interpretation of the prohibition of collective expulsion. 

This blogpost presents a different reading. It will highlight that the ruling is defined by a series of inbuilt ambiguities that combine restrictionist tendencies with dynamic elements, which are bound to cause heated debates of both principle and practice in the coming years. These uncertainties concern the scope and the meaning of the novel exception for those entering illegally and the normative contours of the – potentially wide-ranging – judicial insistence on legal pathways for refugees and migrants. Moreover, greater emphasis should be put on the statutory guarantees of EU asylum law, which are better placed to assess the situation at the external borders. Such technical issues would gain further relevance if follow-up cases confirmed the intuitive impression that the N.D. & N.T. judgment heralds an endpoint to 25 years of migrant-friendly human rights case law.

Judges Reaffirmed a De Facto ‘Right to Asylum’ 

Despite the ECtHR’s rejection of the application, the court’s reasoning is not as straightforward as it at first seems. In particular, the Spanish government did not succeed in reversing the dynamic interpretation of the prohibition of collective expulsion in the seminal Hirsi judgment, which had brought the hitherto obscure provision of Article 4 of the Additional Protocol No. 4 to the forefront of legal debates about migration. 8 years ago, judges had boldly extended the concept of ‘expulsion’ to cases of non-admission at the border, even though the provision had originally been understood to protect those residing on a territory. Spain was supported by several states in challenging this dynamic interpretation. It did not succeed. The Grand Chamber invested considerable energy in reaffirming its earlier stance (paras 15, 173-87). That’s a victory for the applicants and NGOs. 

What sounds like a lawyeristic quarrel is of great practical and conceptual importance, since the prohibition of collective expulsion applies to all migrants. Article 4 of the Additional Protocol No. 4 is not limited to those with a prima facie claim to be persecuted in countries of origin or transit. The N.D. & N.T. ruling is a good example, since judges had rejected a violation of the principle of non-refoulement in Article 3 of the ECHR at an early stage of the procedure. If the Additional Protocol applies nonetheless, Article 4 turns into a freestanding procedural guarantee for people without a realistic opportunity to be recognised as a refugee. 

To be sure, the procedural standards under the Additional Protocol may be lower than in regular asylum procedures (here, paras 237-42), but states are nonetheless subject to a novel human rights standard when rejecting irregular migrants at the border. Such an obligation usually stems from an individual right to asylum for everybody (which the ECHR does not contain) that transcends negative non-refoulement obligations for those facing a real risk of persecution (for instance under Article 3 ECHR). That is why states had objected the dynamic interpretation of the prohibition of collective expulsion so vigorously. Notwithstanding the assertion of the Grand Chamber to the contrary (para 188), the N.D. & N.T. judgment reaffirmed a de facto right to asylum for those rejected at the border. 

The significance of these findings is mitigated, but not undone by the alternative suggestion of the ECHR to exclude those entering irregularly from Article 4 of the Additional Protocol ratione personae. It will be discussed below in how far this sanction is conditioned on the – decidedly vague – availability of legal pathways. Whenever that condition is not met, the procedural guarantees against collective expulsions will resurface. They also cover those applying at a regular border crossing point and, possibly, in consulates abroad. Our analytical acclaim of the N.D. & N.T. judgment should recognise an obscure combination of restrictions and dynamism.

I readily admit that my personal preference would have been to disentangle the prohibition of collective expulsion from the principle of non-refoulement under Article 3 ECHR. Individualised procedures would have been necessary when there are – unlike in N.D. & N.T. – indications of a real risk of inhuman or degrading treatment, in line with the commendable separate opinion of the Finnish judge Pauliine Koskelo. Further guarantees may exist under EU law (and will be discussed below). Many fellow academics will be unhappy with that suggestion, since it would have returned to the interpretative status quo of a decade ago. They should recognise nevertheless that such differentiation might have been an attractive alternative to the court’s baffling proposition that those entering irregularly cannot rely on the Convention, which provokes two follow-up questions: which rights does the exclusion cover? When are legal pathways realistically available?

Interaction with the Prohibition of Refoulement – at Land Borders and Elsewhere

It is the central novelty of the N.D. & N.T. judgment to have made recourse to the generous interpretation of the Additional Protocol conditional upon regular entry. Those crossing a land border in an unauthorised manner are effectively excluded from the protective scope of the prohibition of collective expulsion, since judges consider the absence of an individualised return decision, which would normally amount to a collective expulsion in light of the comments above, to be attributable to the applicants’ own conduct (paras 200-1, 231). This judicial argument, which can hardly be said to emanate from ‘settled case law’ (para 231), amounts to an argument about ‘estoppel’ or ‘abuse’, thereby effectively barring the invocation of a human right that would normally apply. It is quite unclear what consequences this judicial innovation will have in follow-up cases.

The most important uncertainty concerns the range of the exclusionary effects with regard to non-refoulement obligations, in particular under Article 3 ECHR. Most initial commentators presented the judgment – in line with the general language of the press release – as a statement of principle vindicating state practices of ‘hot expulsion’ or ‘pushback’ across the board. Such a broad reading would imply that anyone crossing irregularly can be returned without an individualised procedure (provided that there were legal pathways as discussed below). This would imply that NGO accusations against Croatia, Greece (in relation to the land border at the Evros River) and Spain were unfounded on human rights grounds.

I am not certain that this is correct. Remember that the N.D. & N.T. ruling was based on the previous finding that Morocco was to be considered safe for the applicants and that there was no danger of chain refoulement. In doctrinal terms, the Grand Chamber grounded the exclusion on the ‘own conduct’ of the applicants, in line with the general criteria guiding the application of Article 4 of the Additional Protocol. By contrast, the Strasbourg court had repeatedly resisted efforts by states to limit the absolute character of the prohibition of inhumane or degrading treatment under Article 3 ECHR. Terrorists, for instance, are protected against expulsion or extradition if there is a threat of torture (here, para 79-82; and here, Rn. 124-7). This absolute characters is so deeply entrenched in the case law – ‘irrespective of the (applicants’) conduct’ (here, para 127) – that the N.D. & N.T. judgment should not be read to apply to Article 3 ECHR and my prediction is that the ECtHR will confirm this position in the future.

Nevertheless, the practical consequences of this interpretation might be limited given that the ECHR has recently emphasised that important third states adjacent to the external borders the European Union are to be considered safe. Before Christmas, the Grand Chamber prominently underlined that conclusion for Serbia and it can similarly be defended for Turkey and possibly even Bosnia-Herzegovina (even though such generalised verdicts may have to be distinguished for vulnerable groups or depending on the circumstances of the individual case). The obvious example to the contrary is Libya, which continues to be unsafe at this juncture. Irrespective of how we assess the situation in specific countries, the argument presented above shows that NGOs can try to overcome the logic of the N.D. & N.T. judgment by questioning the underlying safety requirement.

That is not the only doctrinal condition upon which the generalised vindication of the Spanish policy of ‘hot expulsion’ is built. Note that the ECtHR’s description of the ‘own conduct’ fluctuates between the generic emphasis on irregular entry and additional qualitative criteria like the use of force or the large numbers of people involved (paras 201, 231). Does that logic extend to scenarios of nonviolent clandestine entry, for instance via the Evros river or the green border between Croatia and Bosnia? 

Similarly, it is questionable whether the argument can be applied to maritime borders. In N.D. & N.T., the Grand Chamber seems to suggest a differentiation between land and sea borders (para 167), mirroring the earlier distinction between different types of border procedures (here, paras 214-6). For the purposes of the ‘own conduct’ criterion, however, that distinction need not be categoric. Those entering a boat that is unfit for travel knowingly bring about a situation, in which they have to be rescued. That element of own conduct does in no way question the qualification of the maritime emergency as a situation of ‘distress’ and the corresponding unconditional obligation of maritime rescue. It may affect, however, the extent of procedural obligations after rescue. The prohibition of collective expulsion could similarly be considered to be inapplicable provided that neighbouring states are – unlike Libya – to be considered safe. 

Legal Pathways: An Illusory Judicial Revolution?

It is a general feature of the N.D. & N.T. judgment that it raises complex questions of practical and conceptual relevance, once you scratch the surface of a seemingly clear-cut restrictive ruling. One of the most dazzling elements is the astonishingly vague insistence upon legal channels for entry, upon which the exclusionary effects of the ‘own conduct’ criterion hinges. The argument is defined by a confusing mixture of generous and almost revolutionary language requiring states to ‘make available genuine and effective access to means of legal entry’, which ‘allow all persons who face persecution to submit an application for protection’ (para 209), and a decidedly nebulous assessment of the Spanish practice (paras 211-7) that leaves the reader with the impression that the insistence on legal pathways is nothing more than a humanitarian fig leaf for border closures. Many initial commentators rightly emphasised that point.

A threefold ambiguity of the judicial argument draws our attention. Firstly, the doctrinal implications of the plea for legal pathways remains uncertain. Is it a freestanding positive obligation, which flows in particular from Article 3 ECHR and which people in third countries can rely upon? Or is it, simply, an ancillary consideration guiding the application of the ‘own conduct’ requirement? The internal structure of the judgment supports the second reading, since the paragraphs on legal pathways do not form part of the ‘general principles’, but define the court’s assessment of the facts before it (paras 202 et seq.). Moreover, judges explicitly left undecided whether the Convention applies in consulates abroad (para 222) and emphasised that legal channels concerned ‘in particular border procedures’ (para 209). All the same, the principled insistence on legal pathways lends itself to be generalised at a later point, also considering the warning in the dissenting opinion of Judge Koskelo (paras 17-21). An opportunity to do so will be future judgments on humanitarian visas and harbour closures.

Secondly, the Grand Chamber’s petition for legal pathways is defined by an inbuilt weakness: the identification of those who can claim access. A simple jurisdictional threshold has the advantage of being comparatively straightforward to apply in practice, even though it can (unfairly) privilege irregular entries. If, alternatively, the Convention was found to contain positive obligations to establish legal pathways for those who are not yet physically present, judges would have to decide who benefits. The Grand Chamber recognised explicitly that the Convention ‘does not, however, imply a general duty … to bring persons who are under the jurisdiction of another State within its own jurisdiction’ (para 221). Does that mean that the positive duty to facilitate legal entry remains an abstract obligation individuals cannot invoke before courts, a sort of third generation right? 

Thirdly, the precise meaning of the legal pathways standard in the context of Article 4 remains imprecise. To be sure, the Grand Chamber insisted that legal pathways had to be ‘genuinely and effectively accessible to the applicants’ (para 211). In spite of this, the assessment of the Spanish law and practice remained decidedly abstract (paras 212-7, 223-8). It did not engage with a serious discussion whether the applicants had a realistic individual opportunity to avail themselves of these options, also considering that Spain cannot – rightly, in my view – be held responsible for the behaviour of the Moroccan authorities (para 218). The factual situation of the N.D. & N.T. case discloses a noticeable mismatch between potential aspirations and factual entry options: humanitarian visas concern refugees only; the number of asylum requests at border crossing points and corresponding countries of origin reveal their illusory character for West-Africans; similarly, the number of working visas for these states was minimal. 

One is left with the overall impression that the Grand Chamber embarked on a journey without having a realistic idea where to go. A clear judgment with a sense of direction looks different. That is regrettable and must be criticised. Judges lose credibility if their generous language on legal pathways degenerates into humanitarian window-dressing. It would have been honest not to insist on their availability in the first place. Moreover, implications for follow-up cases are wholly uncertain. The genie of legal pathways has been let out of the bottle and it will cause considerable legal controversies in the years to come. Instead of clarifying the legal picture, the Grand Chamber has created much uncertainty.

The Elephant in the Room: EU Asylum Law

An explanation for the lack of clarity may be the unsuitability of the normative standard for the complex situation at the external borders. It is difficult to balance multifaceted normative and factual arguments with an abstract insistence legal pathways, including for economic purposes, on the basis of a human rights instrument, which, moreover, does not contain an individual right to asylum. The Hirsi judgment may have left NGOs with the wrong impression that Strasbourg is the primary judicial forum to seek redress. Indeed, there is an alternative legal benchmark, which should be put into the limelight: EU asylum law, including the Asylum Procedures Directive 2013/32/EU, which stipulates in Article 3 that asylum can be requested ‘at the border’ and which has to be interpreted in light of the right to asylum in Article 18 of the Charter of Fundamental Rights.

I do not claim that the statutory rules of the EU asylum acquis provide easy answers to all questions. Thus, the arguments put forward in the N.D. & N.T. judgment demonstrate that the interpretation of Article 3 of Directive 2013/32/EU may not be as straightforward as many of us (including myself) would intuitively have thought. It can be argued, in particular, that Article 3 of the said Regulation read in conjunction with Article 5 of the Schengen Borders Code Regulation (EU) 2016/399 excludes asylum requests by those who crossed the external Schengen borders irregularly. The positions of the French government and the Grand Chamber in the N.D. & N.T. judgment point in that direction (paras 149 and 209 regarding Article 4 of the former Borders Code). At the same time, a different interpretation can be defended, thus effectively giving irregular entrants a statutory guarantee to claim asylum under EU law, even though the Grand Chamber of the ECtHR not find a violation of the Convention.

The correct judicial forum to resolve these matters is the Court of Justice of the European Union (ECJ) in Luxembourg. Spanish courts should assume their responsibility to apply EU law in these circumstances, including – within the confines of its jurisdiction – the Spanish constitutional court in the pending case on the ‘hot expulsion’ policy. In cases of doubt, judges should make a preliminary reference so that the Luxembourg court can assume its responsibility as the ultimate arbiter of EU law. If necessary, the ECJ will have to decide in how far Article 18 of the Charter contains procedural guarantees beyond negative non-refoulement obligations. The last word on the situation in Ceuta and Melilla remains to be spoken.

Political Salience, or: the End of Dynamic Interpretation?

Like many academics in the field of EU asylum law, I belong to the ‘end of history’ generation. I had just turned 16 when the Berlin wall fell and it seemed self-evident that the liberal democratic project would keep expanding. The ECHR case law on migration is a potent expression of this cosmopolitan honeymoon. Who would have thought thirty years ago that the prohibition of inhuman and degrading treatment would turn into a strong human rights guarantee against refoulement? And that the hitherto opaque prohibition of collective expulsion would be transformed into a de facto right to asylum? Or that the ECtHR might one day oblige states to establish meaningful legal pathways for refugees and migrants? 

The initial reactions to the N.D. & N.T. judgment show that many of us take the human right dynamism for granted. The widespread sense of shock builds on an entrenched expectation that the Strasbourg court would proceed steadily further on the dynamic jurisprudential path. That did not happen, even though the analysis above demonstrated that the seemingly state-friendly outcome conceals a complex (and ambiguous) normative combination of restrictive tendencies with dynamic promises. It seems to me that this restrictionist turn is no coincidence and I predict that judges will not embark on a wholehearted legal pathways revolution in follow-up cases.

As I have explained elsewhere, both in English and in German, courts can be powerful actors in a technocratic world of supranational policy development, while they are badly placed to embark on far-reaching changes in a policy environment, in which migration is politically salient. Courts will continue to play an important part in interpreting existing laws, including the statutory guarantees in the Asylum Procedures Directive. Their judgments will usually uphold the status quo, thereby preventing a potential downward spiral which tries to undo existing safeguards. Courts will not, however, turn European asylum policy or the international refugee regime on its head. The N.D. & N.T. judgment exemplifies that trend towards judicial standstill after more than two decades full of revolutionary judgments.