M.H. v. Croatia: Shedding Light on the Pushback Blind Spot
In the recent judgment of M.H. v Croatia, the European Court of Human Rights (ECtHR) has recognized for the first time the pushback of migrants and asylum-seekers from Croatia. The case centers around the death of six-year-old Afghan Madina Hussiny, who was struck by a train after she and her family members were denied access to asylum, instructed by police officers to follow train tracks towards Serbia and pushed back from Croatian territory without individualized examinations of their circumstances. Her tragic death in Croatia — like that of Alan Kurdi in Greece or of Mawda in Belgium — put European audiences face-to-face with the unjust and deadly impact of Europe’s immigration policies. In addition to a violation of the prohibition of collective expulsions set out in article 4, protocol 4 of the European Convention on Human Rights (ECHR), the Court unanimously found violations of article 2 (right to life) as to the authorities’ treatment of the child’s death, articles 3 and 5 (freedom from torture and right to liberty and security) as to the family’ subsequent detention in Croatia, and article 34 (right to individual petition) as to the criminalization of the family’s lawyer.
While this case has uniquely devastating features, it is emblematic of much larger systematic practices of pushbacks by Croatia — and of several other European Member States. In response to the proposal of 12 Member States last month to adapt the relevant EU legal framework “to counter the instrumentalization of illegal migration and other hybrid threats,” the Council of Europe’s Commissioner for Human Rights warned that human rights violations against migrants and asylum-seekers at European borders have “proliferated at an alarming rate, especially the denial of access to asylum and returns without individual safeguards, frequently accompanied by brutal violence or even the endangerment of human life.” Yet pushbacks have been systematically denied by Croatian authorities and largely ignored by the European Commission, despite consistent documentation and reporting, including from the EU Fundamental Rights Agency, on their widespread use as a means of border control in an EU state vying to join the Schengen Zone.
Against this backdrop, the M.H. v Croatia ruling holds important implications for how the Court might address highly contested, and continuing, pushbacks from EU States: first, via its interpretive application of the N.D. and N.T. v Spain Grand Chamber judgment of February 2020; second, via subsidiarity and a delicate handling of the legal proceedings available domestically for pushback victims; and third, through its acknowledgment that facts presented by consistent and credible reporting outweigh governments’ denials that pushbacks occur.
Application of N.D. and N.T.: Welcome findings on prima facie evidence and the lack of genuine and effective legal pathways
The Grand Chamber’s February 2020 ruling in N.D. and N.T. v Spain, concerning the summary expulsion of two men from Mali and Cote d’Ivoire from Spain to Morocco, was widely critiqued for inventing a new and dangerous exception to the prohibition of collective expulsions. Its application in M.H. v Croatia highlights the case’s legacy on both questions of evidence and on its analysis of alleged collective expulsions.
On evidentiary questions, M.H. illustrates the difficulties pushback victims face in accessing justice and challenging their treatment: Following their expulsions, victims likely have little material evidence other than their testimonies and are generally unable to connect with legal assistance necessary to navigate domestic proceedings. Although this was not the case in M.H., Croatian authorities systematically break the mobile phones of the individuals they expel, thereby destroying potentially valuable evidence. The very nature of covertly expelling individuals into dangerous conditions in third countries from which they cannot access recourse makes the full toll of these pushbacks’ impact — injuries, deaths, and disappearances — difficult if not impossible to fully record and measure: Madina’s death represents one especially tragic instance out of many unknown stories. In this case, the several witnesses to her death, including a doctor on Croatian territory, may have been key factors challenging Croatia’s denials of such practices altogether. Yet the Croatian government still denied that the family had entered Croatia and been pushed back before the child’s death (§266).
N.D. and N.T. recognized common evidentiary challenges faced by individuals alleging pushbacks by applying the evidentiary standards from extraordinary rendition cases ( El-Masri v Macedonia) to expulsion cases “where the absence of identification and personalized treatment of the respondent State [is] at the very core of an applicant’s complaint.” The Grand Chamber acknowledged the necessity of shifting the burden of proof to the Government provided applicants produced prima facie evidence of their expulsion (N.D. and N.T. §85). This dictum was drawn on in Shahzad v Hungary (§35) and confirmed by the court in M.H (§268), where the Court noted it had “no material evidence” of the applicants’ entry into Croatia and return to Serbia by Croatian authorities (§269), but found the consistency and specificity of the applicants’ descriptions of the events, paired with the “large number of reports” of pushbacks from Croatia (§270), to be prima facie evidence.
Until now, the Court’s findings of prima facie evidence had relied on the fact that Spain (in N.D. and N.T.) and Hungary (in Shahzad) entrenched their pushback practices in national law — and thus acknowledged them as such. In M.H., the Court applied this principle to a context in which the Government flatly denies these pushback practices. By doing so, the Court confirmed that the credibility of the applicants’ accounts plays a central role in assessing prima facie evidence (N.D. and N.T. §86; Shahzad §36; M.H. §269). Finding that the Government had “not submitted a single argument capable of refuting” the applicants’ prima facie evidence, the Court accepted their account as true (§§273-274).
On the N.D. and N.T. exception, the Grand Chamber’s criticized test had established that a collective expulsion does not violate article 4, protocol 4 where (1) there is an irregular land border crossing that deliberately employs large numbers and the use of force, “such as to create a clearly disruptive situation which is difficult to control and endangers public safety”; (2) the State provides “genuine and effective access to legal entry”; and (3) the applicants provide no cogent reasons they have not used them (§201; see also Shahzad §59). The first prong of this test makes clear that this exception should only be applied in those restrictive circumstances — as confirmed by the court in M.K. and Poland (§207) and Shahzad (§61). Yet in M.H. v Croatia, without evaluating whether these limited circumstances apply, the Court jumps into an analysis of whether Croatia has in fact provided genuine and effective access to legal entry (§295), ruling that Croatia has not established that it does so (§303). With this move, the Court appears to either presume the applicability of the N.D. and N.T. exception to this case, or to use it as an opportunity to address the lack of legal pathways into Croatia. It ultimately finds a violation of article 4, protocol 4.
Pushbacks and national authorities: A fine balance struck by Strasbourg
The use of pushbacks by European states as the main tool for land border control poses a dilemma to Strasbourg: it is stuck between its current focus on subsidiarity — especially for issues of migration and asylum — and domestic courts’ failure to address these systematic human rights violations. In this case, Strasbourg affirms the need to address “important issues in terms of immigration control” in Croatia (§123). Yet when examining assaults to the life or physical integrity of pushback victims — and the potential criminal conduct of state agents — it skillfully struck a balance between firmly condemning state violations and giving national courts a chance to address the issue themselves.
On the requirement to exhaust domestic remedies in relation to the right to life under article 2, the Court evaluated the M.H. case via its framework for deaths caused by reckless conduct of state agents or occurring in suspicious circumstances, or in circumstances in which the state agents disregarded their legal duties. In doing so, the Court drew on the officers’ disregard for procedural requirements under both asylum law and the Readmission Agreement with Serbia and concluded that a civil claim was not an effective remedy to be exhausted. Criminal proceedings were necessary, which correspondingly triggered Croatia’s obligation under article 2 (§§135-137).
On article 2 and the investigation carried out by national authorities into Madina’s death, the Court found a violation of its procedural limb. The Court noted that the national authorities had mainly relied on the statements of the officers incriminated, failed to question their credibility, unduly disregarded the family’s statements as non-credible, and failed to take crucial investigative steps. These included (1) verifying the officers’ affirmation that the material used for border control could not have recorded the events and (2) taking on the family’s proposal of GPS verification of their (disputed) presence on Croatian soil. The Court also took issue with the exclusion of the family and their lawyer from the investigative process (§§152-163). These failures are all too familiar to victims, lawyers, and NGO workers who routinely file criminal complaints regarding pushbacks in Croatia, Greece, and elsewhere.
Notably, the Court decided not to examine the case with respect to article 2’s substantive state obligations, highlighting that domestic authorities could and should resume their investigation (§165) and putting national authorities before their responsibility — and the subsidiarity principle.
The sub-text of the judgment: The Croatian practice of pushbacks is verified
One of the central but not explicitly acknowledged questions in the case was whether the numerous, long-standing reports on widespread, systematic pushback practices by Croatian authorities were verified. The Croatian government, like Greece, has forcefully denied or attempted to diminish these reports in court (§292) and in public discourse, calling migrants and refugees liars and stating that journalists only aim to tarnish the country’s reputation.
The reasoning of the Court in M.H. is an acknowledgement of the truthfulness — and credibility — of this body of reports but also of the Croatian state’s cover-up for such practices. The Court extensively refers to extracts of reports from 2018 to 2020 on the existence of a state pushback practice (§§103-115). In finding in favor of the applicants’ account of their pushback by Croatian authorities, the judgment enumerates the variety and authoritativeness of the reports (§270), drawing a comparison to its findings in M.K. v. Poland. This is significant because in the referenced paragraph of M.K. (§174), the Court had expressly concluded that the existing reports “indicate[d] the existence of a systemic practice.” Finally, in M.H. the Court draws on the authorities’ investigative failures into Madina’s death in its assessment of evidence (§§272-273), thus possibly pointing to an intentionally inadequate investigation.
Second, in its factual section, the judgment refers to Croatia’s obstruction of reporting on human rights violations arising in the context of pushbacks. The Court strongly sanctions the criminalization of the family’s lawyer by finding a violation of article 34 (right to individual petition). In doing so, the Court pointed out the lawyer’s exceptional commitment to the case (§334). Moreover, the judgment — and Judge Turković’s concurring opinion — emphasizes the important role played by NGOs such as the Center for Peace Studies and Are You Syrious, and the State’s obstruction of their work (§§103, 111, 327, 332). Finally, the judgment refers to how the Croatian Ombudsperson has been impeded from carrying out her mandate in relation to officers’ actions at Croatian borders (§§103, 111), despite being the National Prevention Mechanism (NPM) under the UN Convention Against Torture (Optional Protocol), the only independent police complaints authority and the only National Human Rights Institution in Croatia.
Conclusion
In nodding to the systematic nature of pushbacks in Croatia without explicitly saying as much, M.H. v Croatia remains a significant ruling for the ECtHR and future pushback cases before it. The judgment strongly undermines the EU Commission’s current position that pushback reports should be verified through national criminal investigations — as these do not seem to be effective — and that the solution rests on the establishment of an independent border monitoring mechanism by the Croatian State. As pointed out by the European Network of National Human Rights Institutions (ENNHRI), creating a new mechanism — instead of strengthening existing ones — is counter-productive.
In M.H. v Croatia, the ECtHR has confirmed, for the first time, the brutal practices that victims like Madina’s family and civil society have been exposing for years. The European Commission must wait no longer to finally acknowledge and act on the fact that pushbacks in Croatia exist, reflect widespread state practice, and are inherently unlawful. Pending cases before Strasbourg, notably S.B. and Others v. Croatia, present more evidence of the systematic nature of such pushbacks, as well as broader and interlinked tactics of destruction of evidence, targeted violence, and criminalization of assistance that they entail.
Yet despite the generally positive ruling, the judgment still struggles with fully acknowledging the severity and multiplicity of human rights violations committed during pushbacks. The mere 40,000 EUR in non-pecuniary damages offered to the applicants, a family of 14 who suffered numerous violations in Croatia, including the death of their child and sibling, is insufficient if intended to take seriously such egregious violations.
One question, how did her death take place in Croatia if the family was pushed back to Serbia? On that note, how efficient could have the investigation been if the death took place in a different country (non EU, so presumably the cooperation is diplomatic?) and, therefore, did not fall under Croatian jurisdiction?