The ECtHR has been for a long time criticized for its approach to immigration detention that diverts from the generally applicable principles to deprivation of liberty in other contexts. As Cathryn Costello has observed in her article Immigration Detention: The Ground beneath our Feet, a major weakness in the Court’s approach has been the failure to scrutinize the necessity of immigration detention under Article 5(1)(f) of the ECHR. The Grand Chamber judgment in Ilias and Ahmed v Hungary delivered on 21 November 2019 has further eroded the protection extended to asylum-seekers under the Convention to the point that restrictions imposed upon asylum-seekers might not even be qualified as deprivation of liberty worthy of the protection of Article 5. The Grand Chamber overruled on this point the unanimously adopted Chamber judgment that found that the holding of asylum-seekers in the ‘transit zone’ between Hungary and Serbia actually amounts to deprivation of liberty.
In this blog, I will briefly describe the facts of the case, the findings of the Grand Chamber under Article 3 ECHR that was also invoked by the applicants and then I will focus on the reasoning as to the applicability of Article 5.
The case concerned two Bangladeshi nationals who transited through Greece, the Republic of Northern Macedonia (as it is now known) and Serbia before reaching Hungary, where they immediately applied for asylum. They found themselves in the transit zone on the land border between Hungary and Serbia, where they were held for 23 days pending the examination of their asylum applications. The applications were rejected on the same day on the ground that the applicants had transited through Serbia that, according to Hungary, was a safe third country. The rejections were confirmed on appeal, an order for their expulsion was issued, the applicants were escorted out of the transit zone and they crossed back into Serbia.
Procedural Breach of Article 3 ECHR
The Grand Chamber established that Hungary ‘failed to discharge its procedural obligation under Article 3 of the Convention to assess the risks of treatment contrary to that provision before removing the applicants from Hungary’ to Serbia (para 163). No finding was made on the issue as to whether Hungary was substantively in breach of the right not to be subjected to refoulement given the conditions in Serbia and the deficiencies in the Serbian asylum procedures that might lead to chain refoulement. This omission follows a trend in the Court’s reasoning that can be described as a procedural turn: focus on the quality of the national decision making processes rather than on the substantive accuracy of the decisions taken at national level. This omission, however, had important consequences for the application of Article 5 to the applicants’ case, the most controversial aspect in the Grand Chamber’s reasoning.
The Chamber’s reasoning under Article 5 ECHR
On this aspect, the Grand Chamber departed from the Chamber’s conclusion that the applicants were deprived of their liberty. The fundamental question here is whether ‘the stay’ (Hungary used the term ‘accommodation’) of asylum-seekers in the ‘transit zone’ with an exit door open to Serbia, but closed to Hungary, amounts to deprivation of liberty (i.e. detention) in the sense of Article 5 ECHR. Asylum seekers in the transit zone were denied access to the Hungarian territory, but they could leave to Serbia. This creates a complex intertwinement between deprivation of liberty (Article 5(1)(f)) normally understood as not allowing somebody to leave a place, on the one hand, and not allowing somebody to enter a place. Entering a State can be very relevant from the perspective of the obligation upon this State not to refoule, which necessitates a procedure for determining whether there is a risk of refoulement.
In its judgment from 14 March 2017 the Chamber unanimously answered in positive: by holding them in the transit zone, Hungary deprived the applicants from their liberty, which was in violation of Article 5(1)(f) since this measures had no legal basis in the national law. The Chamber clarified that‘[t]he mere fact that it was possible for them to leave voluntarily returning to Serbia which never consented to their readmission cannot rule out an infringement of the right to liberty.’ (para 55). In this way the Chamber reaffirmed the reasoning in Amuur v France where the Court observed ‘[…] this possibility [to leave voluntary the country] becomes theoretical if no other country offering protection comparable to the protection they expect to find in the country where they are seeking asylum is inclined or prepared to take them in.’ (para 48) It follows that although the transit zone at the French airport was, as France argued, “open to the outside”, the applicants were still considered as having been detained since this ‘outside’ did not offer a level of protection comparable to the one in France.
The Chamber followed this reasoning from Amuur v France in Ilias and Ahmed v Hungary, which led to the recognition that ‘[…] the applicants could not have left the transit zone in the direction of Serbia without unwanted and grave consequences, that is, without forfeiting their asylum claims and running the risk of refoulement’ (para 55). The Chamber also added that ‘To hold otherwise would void the protection afforded by Article 5 of the Convention by compelling the applicants to choose between liberty and the pursuit of a procedure ultimately aimed to shelter them from the risk of exposure to treatment in breach of Article 3 of the Convention.’ (para 56)
The ‘practical and realistic’ approach of the Grand Chamber under Article 5 ECHR
The Grand Chamber in its reasoning broke precisely this linkage between the applicability of Article 5 (the qualification of a treatment as deprivation of liberty) and Article 3 (protection from refoulement). The Grand Chamber performed the following important moves to achieve this. First, it stated that ‘its approach should be practical and realistic, having regard to the present-day conditions and challenges’, which implied that States were not only entitled to control their borders, but also ‘to take measures against foreigners circumventing restrictions on immigration.’ (para 213). With Ilias and Ahmed v Hungary the Court has thus added another nuance to its well-established point of departure in cases dealing with migrants. This point of departure has been that States are entitled, subject to their treaty obligations, to control their borders. The new addition introduced with Ilias and Ahmed v Hungary and also repeated in Z.A. and Others v Russia, a Grand Chamber judgment issued on the same day, concerns States’ right to prevent ‘foreigners circumventing restrictions on immigration’. This addition, however, does not seem appropriate given that the applicants themselves in Ilias and Ahmed v Hungary never circumvented any immigration control restrictions. They applied immediately for asylum.
This ‘practical and realistic approach’ also implied an endorsement of the representation of the situation as one of ‘crisis’: ‘the Court observes that the Hungarian authorities were in conditions of a mass influx of asylum-seekers and migrants at the border, which necessitated rapidly putting in place measures to deal with what was clearly a crisis situation.’ (para 228) In the same paragraph, the Grand Chamber went on to almost praise Hungary for having processed the applicants’ claims so fast event though it was ‘a crisis’: ‘Despite the ensuring very significant difficulties, the applicants’ asylum claims and their judicial appeals were examined within three weeks and two days.’ It appears as if the Grand Chamber at this stage had already forgotten its findings made earlier in the judgment under Article 3 that the national procedure for examining the applicants’ claims was deficient. This ultimately gave the basis for the Grand Chamber to find a violation of Article 3.
The distinction based on how asylum-seekers arrive and the type of border they find themselves at
The second move performed by the Grand Chamber implied the introduction of a distinction between ‘staying at airport transit zones’ (para 214) and at reception centers located on islands (para 216), on the one hand, and a transit zone located on the land border between two Council of Europe Member States (para 219). This meant, as the Court reasoned, that the applicants did not have to take a plane to leave the zone, they could simply walk out of the zone. In other words, it was practically possible for them to do it on their own and they did not need anybody’s help. As the Court continued to reason in para 236, ‘Indeed, unlike the case of Amuur, where the French courts described the applicants’ confinement as an “arbitrary deprivation of liberty”, in the present case the Hungarian authorities were apparently convinced that the applicants could realistically leave in the direction of Serbia [emphasis added].’ This quotation also begs the comment as to why what the national authorities were or were not convinced about actually mattered. In addition, the reference in Ilias and Ahmed v Hungary as to how the national authorities had qualified the situation is also bizarre given that ‘deprivation of liberty’ is an autonomous concept under the Convention. On this point, the two dissenting judges, Judge Bianku and Judge Vućinić criticized the majority by highlighting that ‘the Court has reiterated on many occasions that it does not consider itself bound by the domestic courts’ legal conclusions as to the existence of a deprivation of liberty.’
Narrowing down the importance of Amuur v France
The third move performed by the Court is playing down the importance of and narrowing the relevance of Amuur v France. In Ilias and Ahmed v Hungary the Grand Chamber reiterated (para 239) the most significant pronouncement from Amuur: the possibility to leave the zone ‘becomes theoretical if no other country offering protection comparable to the protection they expect to find in the country where they are seeking asylum is included to take them in.’ It then noted that this reasoning ‘must be read in close relation to the factual and legal context in that case.’ This meant that in contrast to the situation in Ilias and Ahmed v Hungary, in Amuur the applicants could not leave ‘without authorization to board an airplane and without diplomatic assurance concerning their only possible destination, Syria, a country “not bound by the Geneva Convention Relating to the Status of Refugees.’ (para 240) On this point Ilias and Ahmed v Hungary can be also distinguished from Z.A. and Others v Russia, where the Grand Chamber observed that ‘[…] unlike in land border transit zones, in this particular case leaving the Sheremetyevo airport transit zone would have required planning, contacting aviation companies, purchasing tickets and possibly applying for a visa depending on the destination.’ (para 154) For the applicants in Ilias and Ahmed ‘it was practically possible […] to walk to the border and cross into Serbia, a country bound by the Geneva Convention.’ (para 241). The Grand Chamber acknowledged that the applicants feared of the deficiencies in the Serbian asylum procedure and the related risk of removal to the Republic of North Macedonia or Greece. (para 242) However, what seems to be crucial is that their fears were not related to ‘direct threat to their life or health’ (para 242). It follows that the possibility to leave for a place will not preclude the qualification of the situation as one of detention, only if this place poses a direct threat to life or health.
As noted by the two dissenting judges, it did not seem to matter for the majority that the applicants could not enter Serbia lawfully. In this way, the majority’s reasoning under Article 5 appears to endorse a situation where people are just pushed out of the border without some formal procedures with elementary guarantees.
Read as a whole the Grand Chamber judgment in Ilias and Ahmed v Hungary is inconsistent: it contains two findings that are difficult to square together. The Court concluded that since the applicants would not be exposed to a direct risk in Serbia, they were not detained in Hungary. At the same time, Hungary violated Article 3 of the Convention since it did not conduct a proper assessment of the risks that the applicants could face if they were to return to Serbia.
Overall weakening of the protection of Article 5 ECHR
One final comment is due. In Ilias and Ahmed v Hungary, the Grand Chamber summarized the following factors for determining whether ‘confinement of foreigners in airport transit zones and reception centers’ can be defined as deprivation of liberty: ‘i) the applicants’ individual situation and their choices, ii) the applicable legal regime of the respective country and its purpose, iii) the relevant duration, especially in the light of the purpose and the procedural protection enjoyed by applicants pending the events, and iv) the nature and degree of the actual restrictions imposed on or experienced by the applicants.’ (para 217) (see also Z.A. and Others v Russia, para 145) Among these criteria particular attention needs to be directed to the applicable legal regime and the availability of procedural protection. In principle, Article 5, if found applicable, offers certain guarantees (e.g. statutory basis for the deprivation of liberty, access to proceedings for challenging the lawfulness of the detention). The Court seems to have inserted such considerations at the definitional stage of its analysis. For example, in Z.A. and Others v Russia, the Grand Chamber when it examined whether the confinement of the applicants in the airport transit zone amounted to deprivation of liberty, noted that they were left ‘in a legal limbo without any possibility of challenging the measure restricting their liberty’ (para 146). This played a role for the Grand Chamber to conclude that the applicants in Z.A. and Others v Russia were indeed deprived of liberty and Article 5 was thus found applicable. In contrast, the Grand Chamber in Ilias and Ahmed v Hungary observed that certain procedural guarantees applied to the applicants’ case (para 226), which also played a role for the final conclusion that Article 5 was not applicable. In sum, instead of scrutinizing the national legal regime and the access to procedural guarantees as part of the substantive analysis under Article 5, where a single deficiency leads to a finding of a violation (i.e. it is sufficient to find a violation of Article 5 if there is no strictly defined statutory basis for the applicants’ detention), the Court has muddled these criteria together with other factors and made them pertinent for the definitional analysis. This ultimately weakens the roles of these criteria and creates uncertainty.
 See V Stoyanova, ‘How Exception must “Very Exceptional” Be? Non-refoulement, Socio-Economic Deprivation and Paposhvili v Belgium’ (2017) International Journal of Refugee Law 29(4) 580.
 See B Nagy, ‘From Reluctance to Total Denial: Asylum Policy in Hungary 2015-2018’ in V Stoyanova and E Karageorgiou (eds) The New Asylum and Transit Countries in Europe during and in the Aftermath of the 2015/2016 Crisis (Brill 2019) 17.
 Boldizsar Nagy has argued that this representation made by the Hungarian government is a lie. See B Nagy, Restricting access to asylum and contempt of courts: illiberals at work in Hungary, ▻https://eumigrationlawblog.eu/restricting-access-to-asylum-and-contempt-of-courts-illiberals-at