Mirza and the application of the Safe Third Country concept under the Dublin Regulation: a foreboding of things to come?

Printer-friendly versionPrinter-friendly versionPDF version of SummaryPDF version of Summary
Wednesday, October 12, 2016


The judgment by the Court of Justice of the European Union (CJEU) in Mirza is a tightly constructed decision in response to a series of highly circumscribed questions posed by the Debrecen Administrative and Labour Court in Hungary. Given the very precise answers one may doubt the judgment’s utility, especially since the Court’s decision is completely detached from the context of the case, namely that Hungary deems Serbia to be a Safe Third Country (STC), and does, therefore, not broach the interpretation to be given to the concept of safety under Article 38 of the recast Asylum Procedures Directive (rAPD). Whilst this omission is inevitable given the CJEU is limited to the parameters of the questions referred, examining the requirements of safety under Article 38 would have been a particularly fruitful exercise, especially since the applicant had received an inadmissiblity decision on the basis that the Hungarian administrative authority did not consider the applicant to have substantiated why Serbia was not a Safe Third Country for him. However, the line of reasoning which the CJEU follows, as well as the interpretation and application of the provisions it discusses and its effects, touch on the fundamentals of the Dublin procedure and places a renewed question mark over the obligations of transferring States under Dublin. This case commentary will assess two of the three questions referred by the Hungarian Court to the Luxembourg Court whereby the latter was asked to clarify the situations in which the STC concept could be applied and more particularly the existing links between the APD and the Dublin Regulation III.

An opening for inadmissibility before allocation of responsibility for a claim?  

The judgment acknowledges that a Member State may rule on the inadmissibility of a claim under Article 33 of the recast APD regardless of said Member State taking responsibility for the applicant’s claim under the Dublin Regulation. By virtue of Article 3(3) of the Dublin Regulation the Fourth Chamber finds that any Member State, whether it be responsible under the Regulation or not, can send an applicant for international protection back to a Safe Third Country. Additionally, the Court specifies that there is nothing in the Dublin Regulation or the APD which states that an inadmissibility decision can no longer be applied where a State has accepted responsibility for a claim, specifically in the context of a take back procedure.

The first finding of the Court merits specific attention since it provides that subject to the rules of the recast APD, optional Article 33 APD may be used by any Member State, even a State who is in no way responsible under Dublin. The leniency of the court’s approach to the application of Article 3(3) and its possible use outside the Dublin responsibility mechanism is problematic given that Article 3 is set out to firstly identify a responsible state according to the hierarchy criteria in Chapter III and Article 4 of the Charter of Fundamental Rights before any consideration of the Safe Third Country concept. The inverse application would lead to a great deal of arbitrariness given that Article 33 APD is optional and not all Member States have provided for inadmissibility procedures in their domestic legislation, moreover it would directly counter the wording of Article 20(1) of Dublin which specifies that the determination of responsibility must be started as soon as the application is first lodged with a MS. Additionally and whilst not explicitly said by the Court the implication of allowing Member States to rule on admissibility without being responsible under Dublin means that the hierarchy criteria under Chapter III could be ignored as well as the rights within Article 4 and 5 of the Regulation. Such an approach seems to be in opposition with the court’s logic in Ghezelbash in which the CJEU confirmed that the incorrect application of Dublin criteria can give rise to a justiciable claim for the applicant under Article 27(1) and that Dublin III has been characterised by a re-fortification of rights specifically to involve the asylum applicant. Accepting that Member States could invoke the Safe Third Country concept outside Dublin responsibility does away with a considerable part of the judicial protection and effective procedural rights under the Regulation which the Court so fervently supported in Ghezelbash and Karim as well as the Court’s previous case law in Petrosian and MA and others whereby EU law provisions must be interpreted in light of the context and objectives of the framework of rules of which it takes part.

By focusing far more on the preferential treatment and incentives to move onwards if the option of applying the Safe Third Country concept was restricted temporally the Court leaves some troublesome omissions in its wake. The most noteworthy of which is if a Member State has the option of applying the Safe Third Country concept and its application is in conformity with Article 38(1) but the applicant has a family member elsewhere which should be prioritised over which? As said before, on the basis of the above Article 3(1) in conjunction with Article 20(1) the application of Dublin hierarchy should legally come first, however if the Commission’s proposal on Dublin IV goes ahead in its current form, ie mandatory admissibility procedures before Dublin responsibility, one could argue that any prioritisation of admissibility would be in contradiction with the interpretation given to Article 8 of the European Convention on Human Rights (and by virtue of Article 52 of the Charter the interpretation given to Article 7 of the Charter). Strasbourg has affirmed the pro-active obligations under Article 8 (Senigo Longue and Tanda- Muzinga) read in conjunction with the primacy of the best interests of the child under the UN Convention on Rights of the Child to allow entry for asylum applicants to reunite with their family members in other Member States and domestic courts have reiterated the positive duties incumbent on States to comply expeditiously with the procedural and substantive elements of Article 8 as both an autonomous right and as part of Dublin III. The prioritisation of an admissibility assessment and its application over the right to family unity could easily breach the proportionality examination under Article 8, especially where a child is involved, given that the child’s best interests must be the starting point and only then should other factors be weighed against it. Therefore, such an admissibility prioritisation not only weakens Article 8 ECHR and Article 7 Charter’s stand-alone application but it may also mean that the only open route to family unity will be through the Family Reunification Directive, the use of which is consistently hampered by administrative bureaucracy and stringent evidentiary obligations.

Back-tracking on investigative duties

The potential deleterious consequences of the Court’s interpretation of the Dublin Regulation extends to its second submission in which it finds that national legislation prescribing third countries as safe does not affect the determination of a responsible MS. Indeed, the Court highlights that there is no obligation in the Regulation for a responsible MS to inform the sending MS in a take-back procedure of its national legislation or administrative practice with regards to STC. With all due respect the Court’s argumentation fails entirely to abide by Article 4 of the Charter and its corollary article under the Convention which requires transferring States under Dublin to verify that the receiving State’s asylum procedure affords sufficient guarantees against direct and indirect refoulement. Indeed, in order for the individualised assessment of the risks posed to the applicant under Article 3 ECHR to be adhered to, the ECtHR in MSS v Belgium and Greece required the sending state to firstly assess how the Greek authorities applied their legislation on asylum in practice. Therefore, whilst there may not be any explicit obligation under Dublin for the receiving State to inform the sending state of their legislation and practice, it is for the latter to protect individuals from harm to their life or physical integrity of which it knew or ought to have known and to undertake a fulsome assessment of all evidence available in order to examine the risk posed to the applicant upon return to the receiving State. Similarly, in recent domestic jurisprudence the District Court of The Hague has stated in a proposed Dublin transfer to Lithuania that it was incumbent upon the Secretary of State, and not the applicant, to further investigate the application of Safe Third Countries and Safe Countries of Origin in Lithuania in order to comply with the requirements of Article 3.

The Court’s stance is one which approves passiveness on the part of the sending State, placing the burden of proof entirely upon the applicant and leaving it up to them to raise their contentions against transfer at appeal. The reasoning is highly reminiscent of the now overturned KRS v UK judgment before the ECtHR in which no individualised assessment of the risks posed to the applicant upon return to Greece, namely the Greek asylum system and reputable empirical evidence, was carried out. Moreover, it goes against the recent F.G v Sweden, which should be taken into consideration pursuant to Article 52(3) of the Charter, whereby States are under an obligation to carry out an ex nunc assessment of an Article 3 risk of their own motion using all means at their disposal where the State is made aware of facts, i.e. in the Dublin interview, which could expose the asylum seeker to a risk of ill-treatment.  The consequences of bypassing this pro-active obligation allows States to ignore factual evidence and argumentation and in fact flips NS and ME on its head, arguably allowing States to be (purposefully) unaware of procedures in the receiving State which give rise to a real risk of an Article 4 Charter violation. Indeed such a reading of Mirza has been capitalised upon in Sweden by the Migration Agency who has argued on the back of the judgment that legislation and practice on STC lack significance in deciding the responsible MS under Dublin and consequently that there would be no breach of Article 3 ECHR if the applicant were transferred to Hungary. This is despite the numerous reputable reports submitted to the Agency all showing that there are substantial grounds for believing that by virtue of the Hungarian asylum admissibility procedure, viewing Serbia as a Safe Third Country, there is a real risk that an asylum application will be automatically rejected and the applicant will face (in)direct refoulement to Serbia and from there his or her country of origin.


In the commented judgment, the Court of Justice determines that there is no temporal restriction to the STC concept and it may be applied outside the Dublin responsibility mechanism. Such an approach counters the logic of Dublin, the legal structure of Article 3 and could well pose problems for the well codified right to family unity. The focus on incentives to secondary movements (arguably spurred on by Dublin itself) in the judgment distracts from the actual workability of the Court’s conclusion as well as its compatibility with its own jurisprudence and that of Strasbourg. Moreover, the gaping omissions of the judgment with regard to admissibility v Dublin hierarchy may add fuel to the Commission’s fire in their design to externalise the responsibility for asylum processing and refugee protection.


Jean-Baptiste Farcy, PhD candidate Université catholique de Louvain (UCL).

Amanda Taylor, EDAL Coordinator.

(This journal entry is an expression of the authors' own views, and not those of EDAL or ECRE).

Safe third country
Effective access to procedures
Procedural guarantees
Procedural rights and safeguards
Dublin Transfer
Family unity (right to)