The Greek Council of State decision of 22 September 2017 is strongly politically charged. Greece is acting on behalf of the entire EU, so the Council of State may have been subjected to substantial pressure to somehow endorse the EU/Turkey deal, like the Asylum Appeals Board and Committees before it. It is noteworthy that the Council of State decided not to refer the case to the CJEU only by a narrow majority of 13 votes to 12.

Professor Steve Peers recently pointed to the fact that the CJEU has never ruled on the substance and legality of the safe-third-country concept and argues that, since the matter is less than clear, there is a duty to refer to CJEU. The CJEU itself in Gerhard Köbler v. Republik Österreich in 2003 ruled that the Member States might be liable in damages if final courts fail in their duty to refer. Also, the ECtHR in Schipani v. Italy in 2015 held that the failure by the final court to refer to CJEU may breach ECHR (EU Law Analysis), so, if an applicant is ordered to return to Turkey, s/he can ask for interim measures under Rule 39 of the ECtHR Rules of Court. If the person is returned and subjected to the violation of human rights, there may be the liability of the Greek state because, despite a low degree of clarity and precision of the law, the state decided to act. So, if the Council of State does not refer the case for a preliminary reference despite the lack of clarity of the law – even when there was an obligation – and takes a risk that this may result in an error, such error may not be excusable.


Forced deportations approved by the Greek Supreme Court

Putting the Case into Context

Another case on the same matter is now pending before the ECtHR Grand Chamber. The ECtHR’s Ilias and Ahmed case concerns the risk of chain refoulement of two Bangladeshi nationals who transited through Greece, the former Yugoslav Republic of Macedonia and Serbia before reaching Hungary. The case is destined to set an important precedent. It is quite ironic that, while the Council of State argues that Turkey may be considered a safe-third-country, in Ilias and Ahmed the ECtHR’s Fourth Section held that there is a risk of chain refoulement – from Serbia to Greece –  on account of the degrading reception conditions in Greece, which is not so different from the conditions of extreme poverty and lack of access to basic amenities or support in Turkey.

This takes us to the very issue at stake in both cases – namely, the fact that since the ECtHR M.S.S. and CJEU N.S./M.E. cases, concerning the reception conditions in Greece, a conclusive and not rebuttable presumption of safety of any country – be this an EU Member State or a third country –  is no more permissible under EU law. Especially if there is substantial evidence that, as in the Turkish case, there is a foreseeable risk of chain refoulement to Syria following the post-coup reform. So, before taking a decision, the Greek authorities and courts need to conduct a proprio motu assessment of the risk of possible violation of Article 4 of the Charter of Fundamental Rights in Turkey, or further down the chain of possible transfers to other countries. In the end, most likely, Greece will be forced to refer the case to the CJEU and the latter will need to take into account the outcome of Ilias and Ahmed, where the ECtHR – at least before the Fourth Section – already established that, as there is a risk of chain refoulement to Greece from Hungary, the transfer cannot take place.


Opening the Pandora Box

If the Greek case is referred to the CJEU, it may bring some surprises. One of the reasons is that, differently from the ECHR, the Charter of Fundamental Rights includes Article 18 on the right to seek asylum, so assessing the case against the background of such article would open the Pandora box. In fact, Article 18 must be interpreted with due respect for the rules of the Geneva Convention of 1951 and the Protocol of 31 January 1967 relating to the status of refugees and in accordance with the Treaty on the European Union and the Treaty on the Functioning of the EU. This means that the CJEU would need to address the case in the context of the STC but with due regard to its own precedent established in N.S./M.E., where the CJEU held that Member States exercising a discretionary power under the Dublin Regulation –  hence applying EU law (for the purposes of Article 51(1) TFEU) – must interpret the national law and exercise their powers in line with the EU law and fundamental rights, including Article 18 CFR. However, since the Dublin Regulation is strongly connected to the Asylum Procedure, the CJEU would also need to consider whether the Asylum Procedure Directive (an instrument of secondary legislation as well) and the STC do not breach fundamental rights and the EU law and international obligations.

If the Greek case is referred to the CJEU, it is very likely that in the end it will need also to be framed against the backdrop of the recent very contentious CJEU decision in the C-695/15 PPU Mirza case, where the CJEU failed to uphold the IRL by allowing an unconditional and not comparative inadmissibility procedure, which may de facto result in an unlawful penalty contrary to Article 31(1) 1951 Geneva Convention. Such decision is widely considered contrary to EU law, as it allows to override the Dublin Regulation and the proactive obligations under Article 7 CFR, which under IRL requires a comparative test to verify if the applicant has family members in the EU. This may also open the discussion on the current reform of the Dublin Regulation, under which it has been suggested to introduce an inadmissibility procedure which de facto violates Article 7 CFR.

It cannot be forgotten that an applicant for asylum in the EU may have a ‘good cause’ for requesting to be transferred to another Member State, rather than remaining stuck in Greece or Hungary or being returned to a safe-third-country such as Turkey. Among the good causes, there is the presence of family members in another Member State. As a result, an application in the EU cannot be declared inadmissible according to Article 33(1) APD before a comparative test is run within the Dublin Regulation rules. Due to the cross-referencing in Article 33(1) APD, the inadmissibility decision under the same article can only follow the responsibility determination under Article 3 of DRIII. In this regard, a recent landmark decision of a UK Court established that Article 17 DRIII is a justiciable right (approach endorsed indirectly by the same CJEU in Mengesteab, where the CJEU underlined that the Dublin Regulation not only regulates interstate relations but also generates subjective rights for asylum-seekers), so a Member State may be forced to send a Dublin request to another EU Member State, in order to bring together any family relations on humanitarian grounds based, in particular, on family or cultural considerations, even where that other Member State is not responsible.


Being Willing but Unable

Besides these purely legal and quite technical aspects, which the CJEU will need to address unless the Court decides to take a restrictive approach, the CJEU will need in the context of the EU/Turkey to assess the Turkish willingness to accept back an applicant premised on the existence of two key safeguards. Firstly, Turkey must be able to provide effective protection, which must exist in practice and not only in theory. For the UNHCR the effective protection is quality protection that should be regarded as sufficient if, at a minimum, the following is reliably guaranteed:


  • There is no likelihood of persecution, of refoulement or of torture or other cruel and degrading treatment;
  • There is no other real risk to the life of the person[s]concerned;
  • There is a genuine prospect of an accessible durable solution in or from the asylum country, within a reasonable timeframe;
  • Pending a durable solution, stay is permitted under conditions which protect against arbitrary expulsion and deprivation of liberty and which provide for adequate and dignified means of subsistence;
  • The unity and integrity of the family is ensured; and


The specific protection needs of the affected persons, including those deriving from age and gender, can be identified and respected.


A recent study conducted by the Vrije Universiteit Amsterdam demonstrated that Turkey clearly does not meet several of these criteria and its asylum system leads to serious human rights violations.


Secondly, when deciding on the nature and quality of protection available, the assessment must be individual and must take into account the current and actual circumstances prevailing in Turkey. Group decisions based on nationality are not possible and each person must be allowed to rebut the presumption of the safety of Turkey. The return decision must also include an evaluation of the eventual foreseeable risk, including chain refoulement. In this regard, legal changes introduced to Turkish law under the post-coup state of emergency have increased the risk of chain refoulement, so even if we do not consider the living conditions in Turkey, the risk of chain refoulement represents a great obstacle for any return to Turkey. Furthermore, another obstacle is that after the coup Turkey notified the Council of Europe that it may derogate from the ECHR, and Turkey is a signatory to the 1951 Refugee Convention but retains the geographic limitation to Europe, so it does not recognize any obligation towards the modern refugees and doesn’t reliably do so in practice.


Everyone Asks Now: Is Turkey a Safe-Third Country?


If the case is referred to the CJEU, the Court will need to take into account – among others – the above considerations. This may lead the CJEU to take a step back on its previous contentious decision in Mirza, where the CJEU failed to respect the IRL and EU law by allowing an unconditional and not comparative inadmissibility procedure. But it may also force the Court to address the right to seek asylum in relation to the STC concept and the Dublin Regulation construction. It shouldn’t be forgotten that STC principle may be applied outside the EU, but also within the EU going well beyond the Dublin Regulation allocation criteria. So a person with close or meaningful links to an EU Member State should be able to request to be transferred to such country rather than being pushed away to a country where it merely transited for few days.



As a way to draw some conclusions, contrary to the Amnesty International’s position and the recent Washington Post article, the Greek decision is very unlikely to pave the way for forcible returns to Turkey. The decision must be subject to the CJEU scrutiny and may become revocable. The ECtHR final decision in Ilias and Ahmed will also have an impact on possible returns to Turkey. Turkey does not meet most of the criteria for a being considered a safe-third-country. Because it is so close to Syria and because its legislation is being constantly reformed, an absolute presumption of safety is out of the question. It cannot be considered as a country able to offer effective protection, even if we stick to the core definition of this term.

Even if the CJEU will decide to somehow endorse the Greek Council of State decision and go against the ECtHR and its own precedent, that decision would still not pave the way for en masse forcible return to Turkey. The inadmissibility procedure needs to be an individual procedure that must take place with appropriate safeguards, including access to a fair asylum procedure. A conclusive presumption of Turkey as a safe-third-country for every person arriving in Greece, or making their way, is not acceptable and each case must be decided on its own merits.