EU border violence at the European Court of Human Rights: Reflections on G.R.J. v Greece

On 7 January 2025, the European Court of Human Rights issued its (inadmissibility) decision in G.R.J. v Greece and its judgment in A.R.E. v Greece. G.R.J. and A.R.E., the first Greek ‘pushbacks’ cases to be heard by the court, were chosen from a group of 30 some similar cases that remain pending before it.

G.R.J. v Greece was filed in March 2021, when members of de:border were part of the Global Legal Action Network, in partnership with Prakken d’Oliveira and with support from the Dutch Refugee Council. As part of a holistic approach to lawyering, de:border also accompanied GRJ and provided him with different forms of support throughout the case. The case concerns GRJ’s abduction, secret deportation and life-endangering abandonment at sea by Greek state agents in September 2020, arguing that these actions amount to serious violations of the Convention by Greece, including of the right to life (Article 2) and of the prohibition of torture (Article 3), that could not be effectively remedied in Greece (Article 13). A detailed account of the facts of the case and the legal arguments is available here.

In its decision, the court found that, despite “serious indications that there was a systematic practice of ‘pushbacks’ from the Greek islands to Turkey at the time of the alleged events”, the applicant “has not adduced any prima facie evidence of his presence in Greece and of his deportation to Turkey from the island of Samos on the dates alleged, and that he cannot therefore claim to be a victim within the meaning of Article 34 of the Convention” (§ 225). The court thus held that the applicant had not made out his ‘victim’ status with regards to the alleged violations, and declared the case inadmissible ratione personae.

The court’s (in)admissibility decision rendered moot all substantive issues concerning the merits of the case, including the examination of the state’s violations, and, in so doing, bars the possibility for appeal. Still, the court’s analysis of the (legal) facts is substantively incisive and revealing of its (proceduralised) politics, which enable the normalisation of state violence.

The following are our initial reflections on the court’s reasoning and approach in GRJ’s case, and its implications for legal struggles against borders in Greece and at Europe’s borders elsewhere. We offer these counternarratives of the limits and impossibilities of legal struggles against borders pursued through justice systems that are statist and juridical, including human rights institutions, as part of a praxis of movement law/yering on migration justice. This praxis is rooted in our self-accountability towards communities in struggle against borders, acknowledges and resists the harms and violence of law (epistemic and otherwise), and reorients towards transformative and decolonial justice for the global immobility regime. We had the opportunity to share and hold some of these thoughts and feelings in community in meetings we co-convened in Strasbourg around the case’s hearing in June 2024, and at a migrant justice convening in İzmir in October 2024.

In what follows, we consider four areas of the court’s reasoning and analysis: the proceduralisation of substantive evidentiary matters through an (in)admissibility decision (section 1); the decoupling of GRJ’s experience and case from systematic patterns of border violence (section 2); the court’s ‘free’ assessment of the consistency and credibility of the evidence (section 3); and the raised evidentiary standard for a prima facie case (section 4). In a second part, we offer two sets of self-reflexive remarks that contextualise the court’s narrow and exceptionalising approach to remedies for systemic border violence (section 5), and that counternarrate the impossibilities of certain legal struggles against borders (section 6).

We offer these initial reflections primarily to communities of practice with some contextual knowledge of the realities of border violence in Greece and of the court’s judgments (the operative parts of which we reference and summarise as relevant). As we continue to reflect on the implications of these decisions in community with others (and to develop this text for publication elsewhere), we would be grateful to receive and exchange around any feedback and questions you might have (info@debordercollective.org).

1. Proceduralising substantive matters as (in)admissibility

The court’s approach to its (in)admissibility decision in G.R.J. is confusing and arguably misleading. The court appears to have strategically chosen to dismiss the case on an inadmissibility ground relating to the court’s jurisdiction (namely, incompatibility ratione personae), rather than the merits of the case (e.g., evidentiary substantiation of the complaint or ‘manifestly ill-founded’). In other words, and given the court’s reasoning for its dismissal decision, which focuses on substantive evidentiary matters, the court arguably should have dismissed the complaint on grounds of lack of evidence, as opposed to the perceived lack of jurisdiction. This choice allowed the court to sidestep the need to examine the implications of systemic patterns of border violence and lack of domestic remedies in Greece on the ability of (pending and future) applicants to substantiate their ‘pushback’ claims.

The court’s decision states that “the application is incompatible ratione personae with the provisions of the Convention as per Article 35 § 3 a) and must be dismissed in application of Article 35 § 4” (§ 226). Under Article 35 § 4 ECHR, the court “shall declare inadmissible any individual application submitted under Article 34 if it considers that: a) the application is incompatible with the provisions of the Convention or the Protocols thereto, manifestly ill-founded, or an abuse of the right of individual application”.

The court’s guidance on admissibility ratione personae concerns the applicant’s ability “to show that he or she is a victim of the alleged violation” (§ 281). The case-law referenced in the guidance concerns applicants who are unable to show in/direct victimhood due to issues of ‘standing’ before the court, or to the submission of ‘abstract’ complaints (e.g., issues of potential, future material prejudice). To be sure, neither of these instances are applicable in G.R.J., where incompatibility ratione personae arose, according to the court, from the applicant’s (in)ability to show prima facie evidence of his presence in Greece, let alone of a violation of the ECHR.

Given this, it appears that the court should have reasoned G.R.J.’s inadmissibility through the substantive grounds of it being ‘manifestly ill-founded’ (instead than incompatible ratione personae), not least to clarify the difference between the two grounds, as well as to clarify the difference with the approach it took in A.R.E.. In A.R.E., the Greek government also challenged all of the applicant’s factual allegations and raised the same objections with regards to the applicant’s (lack of) victim status and the “abusive nature of the application”. There, however, the court found these objections to be “closely linked to the establishment of the facts and, therefore, to the merits of the case” and therefore “must be joined to the examination of the merits of the complaints made in the application” (see, mutatis mutandis, D v. Bulgaria, […] § 93)” (§ 202; emphasis added).

The court’s approach in G.R.J. also contrasts with its recent judgment in B.Y. v Greece. The complaint was deemed admissible despite, as the court proceeded to find at merits stage, “the applicant […] has not provided prima facie evidence to support [his] version [of events]” and there was “no concrete and consistent evidence, in particular as to his presence on Greek territory, on the basis of which the burden of proof would lie with the Government” (§ 89).

The court’s guidance makes clear that an assessment of the evidence of victim status and a ‘well-founded claim’ requires, in some cases, a decision ‘on the merits’ (Siliadin v. France, 2005, § 63; Hirsi Jamaa and Others v. Italy [GC], 2012, § 111; and also A.R.E. v Greece, 2025, § 202, as seen above). This means that the court evaluates the application in its entirety before reaching an in/admissibility decision (though without necessarily progressing to the merits stage). It would find an application to be ‘manifestly ill-founded’ if the applicant “omits […] to produce documentary evidence in support of his allegations”—but not before examining whether “there are exceptional circumstances beyond [the applicant’s] control which prevent [them] from doing so” (§ 384 of the court’s guidance on ‘unsubstantiated complaints’, emphasis added).

The very nature of ‘pushbacks’ entails practices of concealment and destruction of evidence, including through the confiscation of cell phones, that are arguably exemplary of the kind of ‘exceptional circumstances’ intended to protect applicants like GRJ. By choosing incompatibility ratione personae, the court altogether sidesteps the need to consider the existence of such ‘exceptional circumstances’ and their implications for substantiating ‘pushback’ claims.

2. Decoupling the specific case from a systematic practice

A second facet of the court’s rationale for G.R.J.’s (in)admissibility concerns the link between the existence of a systematic practice of ‘pushbacks’ in Greece, including their non-remediation, and the applicant’s specific case.

In G.R.J., the court begins by recalling that an applicant is not normally required to establish that their case is part of a “systematic or widespread practice of refoulements” in order to “satisfy the burden of proof”, but proceeds to follow a “reverse approach” (§ 182) and first examines whether a systematic or widespread practice of ‘pushbacks’ existed at the time. It does so in line with the applicant’s submissions that such a determination “will assist the Court in taking into account, where appropriate, the general context prevailing at the relevant time in the Greek islands” (§ 182).

To this end, the court observes that there are “numerous official reports indicating a systematic practice by the Greek authorities of returning to Turkey, from the region of Evros and the Greek islands, foreign nationals who entered Greek territory irregularly in order to access asylum procedures” (§ 187) that, in the court’s assessment, “reveal a fairly uniform modus operandi […] on the part of the Greek authorities” (§ 187). It therefore proceeds to find that these constitute “serious evidence suggesting that there existed at the time of the alleged facts a systematic practice of pushbacks by the Greek authorities of third-country nationals from the Greek islands to Turkey” (§ 190; emphasis added).

Critically, the court clarifies that, in line with its case-law, the determination of the systematic nature of Greece’s practice of ‘pushbacks’ in the Aegean at the time of the case (i.e., September 2020) “does not relieve an applicant from the duty to provide a prima facie case in support of his allegations” (§ 182). The court insists that the applicant is required to “demonstrate that [he] entered Greece and then ended up in Turkey on the dates alleged, [and to] establish a link between these two facts in order to be able to satisfy itself that the alleged pushback took place” (§ 191). This is despite “the fact that […] demonstrating what happened in the meantime […] is extremely difficult to prove because of the inherently secret and unofficial nature of the actions in question” (§ 191), and that “applicants may find themselves in an inherently difficult position in terms of evidence and be unable to establish the truth of their account” (§ 183).

By insisting on specific evidence and claiming that “any third-country national could claim to be a victim of a violation of the Convention by shaping his account on the practice described in the reports from national and international institutions” (§ 183), the court is clearly resorting to the logic of the so-called ‘floodgates principle’ in seeking to (considerably) limit the possibility for future claims.

This approach is in contrast with its previous jurisprudence in M.H. v Croatia, where it used evidence of a systematic practice of ‘pushbacks’—i.e., “numerous reports by civil society organisations, national human rights structures and international organisations concerning summary returns of persons”—to support the “prima facie evidence in favour of the applicants’ version of events”, despite there being “no physical evidence to confirm that the applicants entered Croatia on 21 November 2017 and were returned to the border with Serbia by the Croatian police” (§ 269). Similarly, in El-Masri v The Former Yugoslav Republic of Macedonia, the court attached particular importance to material about “the ‘rendition programme’ run by the US authorities at the time” that “does not refer to the applicant’s case as such”, but that “sheds light on the methods employed in similar ‘rendition’ cases to those described by the applicant” (§ 160). By “draw[ing] inferences from” this evidence of a systematic practice, the court found the applicant’s account “sufficiently convincing and established beyond reasonable doubt” (§ 167).

What appears to materially distinguish the court’s reasoning in both M.H. and El-Masri from G.R.J. is the “applicant’s consistent and coherent description of events” (El-Masri, § 162) that was “specific and consistent throughout” (M.H., § 269). Given the witness testimonies and documentary evidence that support his account, GRJ’s claim is not remiss for an ‘absence of circumstantial evidence’. The crux of the court’s reasoning around his (in)ability to substantiate his case is rather based on its ‘free’ assessment of GRJ’s evidence, which we turn to next (section 3).

3. A ‘free’ assessment of the evidence

In the assessment of the coherence and consistency of the applicant’s description of the events, and the overall credibility of his account, the court is not bound by any “procedural obstacles to the admissibility of evidence or predefined formulas for its assessment” (G.R.J., §174): “[a]s master of its own procedure and rules, [it] assesses in complete freedom not only the admissibility and relevance, but also the probative value of each element of the case file” (§ 178).

In line with its jurisprudence, however, the court is required to exercise this freedom in a manner that accounts for the applicant’s “inherently vulnerable evidentiary position” (§ 45) and treats it with due care. The court’s problematic (over)reliance on ‘objective’ evidence to the detriment of human testimony—which was evident already during the hearing and was confirmed in the decision—is at odds with this requirement.

GRJ’s account is dismissed for “not provid[ing] a prima facie case in support of his allegations” (§ 224), despite the fact that his account “largely correspond[s] to the modus operandi” of Greece’s ‘pushbacks’ (§ 191) and despite the failure of the government to provide a satisfactory and convincing alternative explanation” (§ 190) of the chain of events leading to GRJ being ‘rescued’ by the Turkish Coast Guard.

To reach this conclusion, the court evaluated the three types of evidence submitted before it: 1) documentary evidence; 2) audiovisual evidence; and 3) testimonies and other evidence.

---

---

First, the court dismissed the documentary evidence—an extract from the register of events kept by the Turkish Coast Guard indicating the ‘rescue’ of two Afghan minors on 9 September 2020 and a detailed case log—by pointing out that, even if the documents were issued by the Turkish Coast Guard (a fact that was challenged by the Greek government), they did not indicate the specific identity of the ‘rescued’ individuals and thus could not “constitute prima facie evidence” (§ 198). The court does not, however, comment on the substantive content of the detailed case log, which states that “[it] has been determined from the clothing of the rescued irregular migrants that they are the same [people] as those in the photographs shared on Aegean Boat Report’s social media account on September 8 2020 after crossing to Samos” (machine translated from Turkish).

Then, with regards to the filed audiovisual evidence, the court places decisive emphasis on the implausibility of the “[a]bsence of a photograph” of GRJ on Samos, which it deems “is such as to cast serious doubt on the credibility of [the applicant’s] account” (§ 212).

What the court appears to ignore is that GRJ’s account consistently makes reference to pictures (included in the case file) that he had taken of H, another Afghan minor with whom he arrived on Samos and with whom he was later pushed back to Turkey. In addition, GRJ mentions a photo of H taken inside Samos’ RIC, which was shared with H’s relative, ZH, and with Aegean Boat Report (ABR). Despite the pictures matching GRJ’s account of the events, the court does not accept the possibility that the photo of GRJ on Samos, taken by H, was lost when the Greek authorities confiscated GRJ’s phone, and that the copy of that photo, shared with the eyewitness MNA in September 2020, was no longer accessible due to MNA having changed phones. The court goes as far as to dismiss the applicant’s account—that his photos were sent to MNA, and that H’s picture was recovered through his relative ZH, who forwarded it to ABR—as a misrepresentation.

In considering the video of his arrival on Samos, the court is more concerned with the legal representatives’ error in pointing out the applicant’s position (as being on the left during the hearing, and not on the right as indicated in the case file) than the value of this circumstantial evidence and its cumulative effect.

Then, while analysing the Turkish Coast Guard’s video footage showing GRJ, who identified himself in the video, and H, who is clearly identifiable in the photo in Samos’ RIC and is wearing the same clothes in the video, being ‘rescued’ and disembarked in Turkey, the court found that since the individuals’ “faces are covered by medical masks, and they are therefore not fully visible”, “it is highly doubtful” (§ 215) that the footage includes the applicant. The dismissal of the applicant’s self-identification in the video is at odds with the court’s approach in N.D. and N.T. (as discussed below in section 4), and, once again, does not take into account the cumulative value of the submitted evidence.

The court then analyses the testimonies and other evidence. It confirms that “different versions [of the events] confirm that two minors of Afghan nationality were returned from Samos to Turkey on the alleged dates” (§ 219), but dismisses their probative value on the basis that “the evidence in the case file does not enable [the court] to establish beyond reasonable doubt that the applicant was […] one of the two minors in question” (§ 223, emphasis added). This is despite witness testimonies placing GRJ on Samos in September 2020 and the video footage of the Turkish Coast Guard capturing GRJ and H being ‘rescued’ and disembarked in Turkey. As we discuss in section 4 below, the ‘beyond reasonable doubt’ test is the appropriate test for a violation, but not for establishing whether the applicant has substantiated a prima facie case.

Specifically, the court dismisses the testimonies and other evidence in the case file based on three (perceived) inconsistencies. First, during a videoconference exchange in April 2024 between MNA and the Greek National Transparency Authority, described in a confidential report by the NTA, MNA was shown two pictures of GRJ, sent by the Ministry of Migration and Asylum, and was unable to confirm with 100% certainty that the person in the photos was the same person he met briefly in the Samos’ RIC in September 2020. Whilst recognising that “MNA expressed hesitation due to the four years that had passed since the alleged incident”, the court remained of the opinion that MNA’s failure to identify GRJ “considerably weaken[ed] the credibility of the applicant’s account” (§ 220). Remarkably, this is in contrast to the court’s approach in A.R.E. v Greece, where it acknowledges the existence of discrepancies between the eyewitness’ testimonies (as well as between them and certain allegations made in the application), but that given the four-year time lapse, what matters is that they “converge on certain crucial points” and “corroborate the applicant’s account concerning her presence in Greece and her arrest by the Greek authorities” (§ 264). The same standard and approach was arguably warranted in G.R.J..

Second, the court placed emphasis on an inconsistency between UNHCR’s records of GRJ’s arrival on Samos and GRJ’s asylum card—the former reading “First name: R.; surname: J.; Father’s first name: R”, and the latter “First name: GR; Surname: J.; Father’s first name: MY” (§ 221). The court sets aside the clear convergence in ‘First name’ and ‘Surname’, stating that “these […] sets of identification details […] do not coincide [and] make it extremely doubtful that the applicant […] was […] concerned by the incident” (§ 221), without recognising that misregistration is a common practice.

Thirdly, the court notes that when the applicant finally managed to re-enter Greece and register his asylum claim in November 2021, he “made no reference to an earlier entry into Greece” (§ 222), though finding that “these assertions cannot be considered as conclusive” (§ 222). Again, this finding does not explicitly acknowledge the reality in which persons in the position of the applicant fear the real consequences of their making reference to prior, especially unregistered, arrivals, let alone ‘pushbacks’. for their protection claims.

Overall, the court is convinced that two Afghan minors were returned to Turkey by Greece on the day GRJ was pushed back, in a manner consistent with his account of events. However, because his “identity cannot be established with certainty in light of the documents produced before it” (§ 194), the court summarily declares the case inadmissible.

The court states that it can draw inferences unfavourable to the government if it fails “to disclose essential documents to enable the Court to establish the facts or […] to provide a satisfactory and convincing explanation of how the events in question occurred” (§§ 175-176). Yet, at no point does the court thoroughly scrutinise the government’s assertions, including its outright denial of the factual allegations in their entirety and failure to provide a satisfactory explanation as to how two minors found themselves at sea on a motorless raft (§ 190). Nor does the court require that the government disclose any relevant documentation in support of its version of events.

The court superficially takes note of “the apparent delays by the [Greek] national authorities in conducting a criminal investigation into the facts of the present case”, which it states “could have shed light on certain aspects of the case which divide the parties” (§ 224) but do not “affect the overall conclusion that the applicant has not provided a prima facie case” (§ 224). It does not, however, scrutinise either Greece’s systemic investigative failures (discussed in several TPIs, the court’s case-law, e.g., B.Y. v Greece; and A.R.E. § 197-201) or the Greek authorities’ practices of concealment and evasion of accountability, including its fabrication of (legal) facts by falsifying ‘prevention of departure’ records (and Frontex’s misuse of its JORA system to disguise ‘driftbacks’, exposed by Lighthouse Reports).

In sum, the court dismisses GRJ’s submissions as fabricated and spurious—short of abuse of application, as alleged by the Greek government (§ 145)—and disbelieves his experiences of violence at the EU’s borders in Greece. By decoupling GRJ’s case from the systemic patterns of violence, and taking a rigid approach in its analysis that prioritises ‘objective’ evidence over human testimony, the court effectively accepts the government’s position that the applicant’s allegations are ‘unreasonable and unconvincing’ (§ 144).

4. Raising the bar for a prima facie case

The court’s ‘free’ assessment of the consistency and credibility of the evidence in G.R.J. both diminishes the probative value of human testimony, and entrenches the court’s reticence to reverse the burden of proof onto the state in previous cases of covert border operations. Its approach in G.R.J. reinforces the court’s problematic track record of conflating and upscaling different evidentiary standards and tests: a) on admissibility, it conflates the requirements for the court’s jurisdiction and those for the ‘foundedness’ of a complaint (see above in section 1); and b) on the merits, it conflates the standard for a ‘prima facie case’, relevant for the reversal of the burden of proof onto the state, and the ‘beyond reasonable doubt’ test, applied in proving a violation.

Unlike other human rights bodies, for whom the systemic nature of the violations is, in and of itself, a basis for the reversal of the burden of proof onto the state, the court continues to require case-specific prima facie evidence to proceed with such reversal. In line with its jurisprudence, in G.R.J. the court notes that “it is essential to determine whether the applicant has provided a prima facie case”, that is “a detailed, specific and consistent account of the events in question” (§ 179), to reverse the burden of proof onto the state.

However, in both G.R.J. and A.R.E., the court proceeds to wrongly apply the higher standard of the ‘beyond reasonable doubt’ test as part of the requirements for establishing a prima facie case, and thus the reversal of the burden of proof onto the state.

In A.R.E., the court concludes that the burden of proof is reversed onto the state—meaning that the Greek authorities must “prove that the applicant had not entered Greece and had not been returned to Turkey on the dates alleged” (§ 265)—by finding that the “applicant’s allegations are sufficiently convincing and established beyond reasonable doubt” (§ 267, emphasis added). Whereas in G.R.J., the court applies the ‘beyond reasonable doubt’ test, which is usually applied to the substantive assessment of a violation, in an inadmissibility decision, and finds that “the evidence in the case file does not enable it to establish beyond reasonable doubt that the applicant was [returned to Turkey]” (§ 223, emphasis added).

The court explains its broad discretion to choose this approach based on the absence of “procedural obstacles to the admissibility of evidence or predefined formulas for its assessment” (§ 174), and on “[t]he distribution of the burden of proof [being] intrinsically linked to the specificity of the facts, the nature of the allegation made and the Convention right at stake” (§ 177; see also, para 27 of the court’s guidance on admissibility‘s reference to N.D. and N.T. v. Spain [GC], 2020, §§ 83-88). As a human rights court, the ECtHR should have arguably proceeded to the examination of the merits of the case, rather than subsuming its substantive analysis of the evidence (see Casarini v Italy) in a (seemingly) procedural assessment of admissibility, in a manner that undoes the purpose of this stage of the proceedings.

The approach in G.R.J. significantly differs from N.D. and N.T. v Spain, where the court was satisfied that the “video-footage showing the storming of the fences as described by [the applicants], and on which they claimed to recognise themselves” (§ 86) amounted to “prima facie evidence of their participation in the storming of the border fences in Melilla on 13 August 2014 which has not been convincingly refuted by the Government” (§ 88). The court dismissed the government’s preliminary objections to the applicants’ victim status (denying they were part of the expulsions on the day) and was satisfied to “presume the account of the events presented by the applicants to be truthful” (§ 88)—without any reference to the ‘beyond reasonable doubt’ test as the appropriate standard of proof and despite finding that the evidence adduced by the applicants did “not appear conclusive” (§ 84-85).

The court departs from this approach in G.R.J. without explanation. In a manner that conflates the prima facie and ‘beyond reasonable doubt’ standards, it raises the standard for proving undocumented treatment to an “account, which must […] be detailed, specific and coherent, that is to say free from contradictions”, supported “with concrete, circumstantial and consistent evidence on the basis of which the burden of proof will be reversed to the respondent Government” (§ 182). Not only is the court creating an impossible standard by ‘upscaling’ the prima facie test in ‘pushback’ cases (Dembour; Baranowska), but it also renders meaningless the very purpose of such reversal of the burden of proof onto the state by applying the ‘beyond reasonable doubt’ test (used to prove violations).

5. Narrowing and exceptionalising remedies for systemic border violence

As the first two judgments in the group of cases pending before the ECtHR concerning ‘pushbacks’ at the EU’s external borders in Greece, the court’s inadmissibility decision in G.R.J. and its determination of violations in A.R.E. crucially shape the court’s approach to remedies for ‘pushback’ practices in Greece and beyond.

The practices described through the cases are emblematic of the ongoing, systemic, and increasingly violent border control policies and actions of the Greek authorities—enabled and supported by EU bodies, including Frontex—in the Aegean since at least 2020 and at the Evros land border since 2018. In this context, impunity for state violence is maintained through a concurrent set of  systemic institutional practices, namely the Greek law enforcement and judicial authorities’ denial, concealment and fabrication of (legal) facts—from the confiscation of mobile phones to the creation of false entry records—and the initiation of disingenuous and superficial domestic remediation procedures.

The extreme and secretive nature of this state violence has led the UN Committee on Enforced Disappearances to elaborate and adopt its General Comment No. 1 (September 2023) as a blueprint for the application of the International Convention for the Protection of All Persons from Enforced Disappearances (2010) in the migration context. It, critically, maintains that acts of ‘pushbacks’ are forms of enforced disappearance and that the abandonment, endangerment and unprotection policies in operation in border zones produce (enforced) disappearances. States and international organisations’ have concrete and detailed obligations to search for, trace, identify and remedy such cases and the underlying causes for such harms (see also F.A.A. v Greece).

The court’s approach in G.R.J. and A.R.E. regrettably renders those subject to these extreme forms of violence and repression even more vulnerable. These cases, which pave the way for the court’s approach to the other cases of ‘pushbacks’ in Greece pending before it, appear to have been chosen by the court strategically—not only to represent both land and sea ‘pushbacks’ supported by ‘digital’ evidence (including a forensic reconstruction), but also as cases that could have been exceptionalised based on the court’s jurisprudence on vulnerability (an unaccompanied minor, and a woman returned to and persecuted by Turkey).

The court’s arguably rigid and dismissive approach in G.R.J. is out of step with its stance on “the special situation in which asylum seekers often find themselves”, having previously held that “it is frequently necessary to give [asylum seekers] the benefit of the doubt when it comes to assessing the credibility of their statements and the documents submitted in support thereof”, and thus provide them with ample opportunity to explain any discrepancies (D.N.W v Sweden, citing, among others, Collins and Akasiebie v. Sweden (dec.), no. 23944/05, 8 March 2007; and Hakizimana v. Sweden (dec.), no. 37913/05, 27 March 2008). It is thus within the court’s responsibility to take a care-full approach in such cases by engaging in further probing and analysis of the facts on the merits, and giving proper weight to human testimony. Despite its limited fact-finding powers and resources, the court could and arguably should have required disclosure and scrutinised relevant documentation by the respondent state and engaged in a good faith assessment of the applicant’s evidence.

Furthermore, the discrepancies between the court’s evaluation of evidence in G.R.J. and A.R.E., respectively, are remarkable. This is perhaps unsurprising, given the court’s apparent construction of a ‘perfect victim’ in A.R.E.—one that has exhausted domestic remedies in Greece (being rejected at the level of the Supreme Court), was informed enough to gather their own (metadata-supported) digital evidence that is then analysed by experts, has credible witnesses (a Greek lawyer), was returned directly to the country where she faced persecution, and has a Turkish court judgment evidencing her return and persecution in Turkey, amongst other things. The court’s approach in A.R.E. is understanding of the fact that the lawyer-eyewitness statements did not always confirm the applicant’s claims, thus creating some discrepancies between the different accounts (which the court explains as resulting from a four-year time lapse). In contrast to G.R.J., in A.R.E. the court does accept that the applicant’s undocumented return was impossible to prove in the ‘particular circumstances of the case’ (see the discussion of the ‘manifestly ill-founded’ exception in section 1 above), because the applicant was not in possession of her mobile phone and because her return took place at night.

Despite its (partial) acknowledgement of the systematic nature of ‘pushbacks’ in its inadmissibility decision in G.R.J. (as well as in its judgment in A.R.E.), the court effectively discredits and dismisses GRJ’s account, in favour of that of the government, by raising the standards of proof and insisting on ‘objective’ evidence. This use of evidentiary requirements reinforces the court’s established approach in migration cases, which gives primacy to the state’s right to enforce borders over and above migrants’ (individual and collective) rights (see e.g., N.D. and N.T. v Spain). The court’s previously deployed ‘methods of deference’ to state sovereignty include the rejection of cases on grounds of ‘non-exhaustion of domestic remedies’, the refusal to acknowledge the lack of safe and legal pathways to enter Europe, and the delay of cases concerning a widespread policy (e.g., S.S. and Ors v Italy, pending since 2018).

6. Counternarratives of legal struggles against borders

The ECtHR is the main judicial authority on human rights within the European context and, in the absence of the possibility for obtaining genuine and effective domestic remedies in Greece, the only pathway for seeking human rights-based remedies for Greece’s ‘pushback’ policies. In the EU institutional framework, the European Commission, which has the means and obligations to take enforcement action against Greek and EU institutions and agencies involved in this violence, has continuously refused to recognise the reality of systemic breaches of EU law, including fundamental rights, and its significant implications for the integrity of the rule of law and budget of the EU (see joint complaint against the Commission and ongoing inquiry by the European Ombudsman). This is despite abundant authoritative documentation and the significant determination by Frontex’s Fundamental Rights Officer in 2023 that the Greek authorities’ engagement in and non-remediation of ‘pushbacks’ is systemic policy (an opinion obtained through a document request by Lena Karamanidou).

The ECHR framework, which requires states to opt in to the prohibition on collective expulsion through Protocol 4 (which Greece has not done), is structurally limited to individualised claims against states and individuated forms of remedies that often decontextualise cases from broader state practice. Human rights institutions are failing to secure migrant rights’ as even the most favourable migration-related decisions, such as Hirsi Jamaa and Ors v Italy or M.S.S. v Belgium and Greece, have had limited or harmful impact.

The impossibilities of systemic justice for border violence, let alone the global immobility regime, which most applicants seek, makes it difficult to narrate even a favourable determination of a violation as a ‘success’. Indeed, the court’s relatively favourable determination in A.R.E. (however, not on Articles 2 and 3 for the violence of the ‘pushback’) should be understood in the context of its exceptionalising and narrowing approach to remedies in migration cases—an ominous outcome that foreshadows the likely dismissal (through inadmissibility or otherwise) of many other ‘pushback’ cases, in Greece and elsewhere.

Critically, the court has always, at least implicitly, given primacy to states’ sovereign rights to prevent illegalised migration, and maintained the state’s underlying rights to (forcefully, and even violently) ‘protect’ its borders and enforce crimmigration (asylum and criminal) laws. Being part of colonial legacies, the European human rights framework not only prevents scrutiny of the global socio-economic inequalities maintained by borders, but has often ‘authorised’ certain forms of state violence by reproducing the epistemic violences of erasure and invisibility.

The court is therefore not a forum for hearing, let alone care-fully holding, the lived experiences of migrants—who all too often lack (physical and other) access to, and a meaningful participatory role in, proceedings. There was no exception in the packed courtroom in Strasbourg in June 2024, where the most significant absence was that of the applicants, which occurred unremarked. While ARE was subject to security measures in Turkey, GRJ chose not to attend having become understandably discouraged by the length of the proceedings and their apparent irrelevance to his current reality. After spending the last years living in precarious conditions and being exposed to further life endangerment (undocumented in Istanbul and subject to land ‘pushbacks’ by Greece) before finally finding (relative) safety in Germany, the justice he seeks is healing for himself, and transformation of the harmful systems he experienced.

Movement law/yering invites us to self-reflexively and self-accountably situate our legal struggles in a broader context of political struggles against borders, their violences and the (legal) systems that enable them, and to acknowledge the impossibilities, failures, and harms of statist justice systems. How do we account for and align around the harms reproduced by our legal struggles through systems that uphold the violence of borders? As communities in struggle, we are called upon to hold the embodied knowledges of experiences of injustice before human rights institutions and to counternarrate the impossibilities of migration justice through individualistic and punitive ‘justice’ frameworks that reproduce carcerality and borders (described by Silvana Tapia Tapia as ‘human rights penality’). We are called upon to recentre and reorient towards no-borders visions of justice: where and how can we demand and pursue decolonial and transformative justice and reparations for border and immobility regimes?