The Politics of Legal Facts: The Erasure of Pushback Evidence from the European Court of Human Rights

Abstract

This article analyzes how the European Court of Human Rights (ECtHR) handles evidence of pushback, where states violently force asylum seekers away from borders. An examination of how the experiences of pushback survivors get translated (or not) into judgments contributes to theoretical discussions about truth, epistemic practices, and law. The article asks why so little of what researchers, journalists, civil society actors, and international organizations have documented about European border violence is visible in the court’s judgments. Based on a mix of legal and anthropological research methods, the article traces how states and the ECtHR erase pushback evidence at borders and during litigation. Taking seriously on equal grounds the construction of facts outside and inside a court room, the article connects external perspectives on the production of evidence with an internal analysis of evidence in judgments. In doing so, the article highlights the political dimensions of seemingly merely technical and legal procedures. We argue for a clearer separation of courts’ and states’ versions of facts, contending that the presumption of the states’ good faith should no longer apply when there is evidence, including in case law, of misrecordings and false statements by respondent states.

Introduction

In the autumn of 2020, Cypriot authorities pushed back several boats to Lebanon with Syrians who had wanted to seek asylum (Alpes 2021). After months of work by Jill Alpes to establish contact with some of the roughly seven hundred pushback survivors, a Cypriot litigator sent twenty-six complaints against Cyprus to the European Court of Human Rights (ECtHR) in Strasbourg. The registry of the ECtHR decided to register none of these twenty-six applications. The adjudication of an application by the court with regard to both its admissibility and merits necessarily requires prior registration. While the court’s admissibility decisions are taken by judges, the decision to register or not an application is taken by the registry staff alone and considered to be a merely bureaucratic process.

The twenty-six rejection letters drafted by the same registry officer stated that “there [was] no official confirmation of the alleged government action having taken place.” The requested “official confirmation” did not exist. As the litigator had stated in her application to the ECtHR, the pushback survivors had “never [been] notified with an entry ban/deportation/expulsion decision in writing explaining the reasons in fact and in law,” nor had any “record [been] made of [their] wish to apply for asylum.” ¹ The ECtHR’s registry did not register the cases because documents that do not exist were not submitted.

This incident is interesting in at least three ways. First, the incident poses questions about the inherent logic and assumptions of how the ECtHR comes to establish the facts of a case. The supposedly merely formalistic check by the registry required official proof for unofficial state practice, presenting the litigator with a paradox. The paradox raises questions about the extent to which the court is taking into account factual realities as well as about how it weighs official documents, testimonies, and evidence from state sources.

Second, the non-registration of these twenty-six applications reminds us of what remains invisible when academics analyze the court from the perspective of case law. Judgments are only the tip of the iceberg. Most survivors of human rights violations are not able to get a lawyer willing to file an application pro bono for them. And most applications before the ECtHR are either not registered or rejected at the admissibility state. ²

Third, the incident illustrates the uncertainty and challenges that litigators and their allies have to navigate. In a desire to stem its overwhelming workload, the court’s registry staff generally do not provide justifications for non-registration and inadmissibility decisions. Hence, in the vast majority of cases, its reasoning and evidentiary requirements for non-adjudicated cases remain unknown. The lawyers of pushback survivors do not receive explanations for non-registration and inadmissibility decisions after they and frontline responders have overcome a series of challenges at borders, including in the respective countries to which people on the move are pushed back.

Since February 2022, challenges for pushback litigators and their partners have further increased as they now only have four months to submit complaints. Despite these time pressures, formal rules are stringent. The ECtHR, for example, does not in principle accept digital signatures. In the case of the earlier-discussed applications against Cyprus, this meant that Alpes and her Lebanese colleague took an entire week to physically drive and collect the twenty-six signatures from the different pushback survivors who had found refuge in different, sometimes remote, areas of Lebanon.

Against the backdrop of this case, the article analyzes the ECtHR’s evidentiary regime for pushback cases and the challenges it poses to litigators and their allies who fight for the rights and safety of people on the move. With respect to the evidentiary regime, we refer to a system of principles and standards governing the admissibility and evaluation of evidence during legal proceedings. Litigators at European borders have over the last decade mostly sought to challenge pushbacks in front of the regional human rights court of the Council of Europe—the ECtHR—which provides individual applicants the possibility to directly challenge member states for human rights violations in court. The system of the European Convention on Human Rights (ECHR) includes oversight mechanisms for the implementation of judgments by member states, which through its interplay between law and diplomacy paved the way for the emergence of European human rights (Madsen 2007). ³ As there are typically no avenues for domestic remedies in pushback cases, litigators have complemented regional human rights proceedings with requests for criminal investigations at the domestic level (Fehr and Alpes 2023).

Despite the importance of the ECtHR for pushback litigation at European borders, there are striking gaps between, on the one hand, evidence on pushbacks as produced by people on the move, civil society actors, international organizations, and journalists and, on the other hand, the acknowledgment of border violence as legal facts in judgments. Why is so little of what has been documented about border violence visible in the court’s judgments? In order to answer this question, the article examines different processes at borders and at the court that render invisible the perspectives and experiences of people seeking asylum at European borders. We consider the ECtHR to be constituted by its judges, its registry officers, and other staff, thus including both adjudicating and more bureaucratic practices. In studying actions and actors that mediate what becomes knowable about border violence to ECtHR judges sitting at a desk in Strasbourg, the article also pays attention to what is not translated.

The article’s methodological approach to answer its research question is tightly interwoven with its theoretical approach to the production and assessment of evidence. Drawing on the methodological tool sets of both a legal scholar and an anthropologist, the article is based on open-ended semi-structured interviews with frontline responders, pushback survivors, and their legal representatives, observations with actors at European borders, as well as a forensic reading of silences in ECtHR case law and other types of documents that are part of adjudication, litigation, and border procedures. More precisely, Grażyna Baranowska, as a legal scholar, analyzed the ECtHR pushback case law. She collected the judgments through a HUDOC database search of Article 4, Protocol 4, cases (prohibition of collective expulsions) and additionally complemented them with relevant Article 3 (principle of non-refoulement) judgments under the ECHR. ⁴ As a legal anthropologist, Alpes interviewed thirty-five litigators, journalists, activist researchers, and members of civil society organizations between May 2021 and May 2022 and has also been actively involved in litigation efforts for the pushback corridor from Cyprus to Lebanon since November 2020. ⁵ In addition, Alpes and Baranowska co-analyzed decisions of non-registration and litigators submissions, including rebuttal letters written by litigators in reaction to state responses to the ECtHR’s decisions. Such documents are not publicly available on the court’s database and, as such, are not normally part of the corpus of texts analyzed by legal scholars.

Focusing on questions of visibility, the article does not unpack the voices of pushback survivors per se. Instead, it examines translation processes and the erasure they produce. By also shedding light on evidence that never makes it into a judgment, the article brings external perspectives on the production of evidence into dialogue with an internal analysis of evidence in human rights judgments. While external and internal are mediated constructs that in turn shape our vision of law itself (van Oorschot 2021, 36), we use this conceptual frame to highlight that our interdisciplinary work not only mixes different types of data but also confronts different views of how we can come to understand the world (epistemology).

Legal scholars typically offer critiques of adjudication from “within” the law, meaning their analysis is constructed within the constraints of legal concepts and procedures. Social scientists, on the other hand, typically offer critiques from “external” perspectives, meaning that the physical context of adjudication, rather than its normative outcomes, becomes the object of analysis. External perspectives on law creates space for the explanatory power of politics in legal decisions (Feldman 2005; Canfield 2023), while also allowing positivism in legal scholarship to be counterbalanced with more nuanced insights into power and knowledge (Janmyr 2022). The benefits of external critiques can come at the expense of audibility for legal practitioners who operate from within the logics of law. In this article, we confront and integrate both internal and external perspectives.

The article’s interdisciplinarity allows us to highlight the political dimensions of seemingly mere technical and legal procedures. Procedures during litigation and adjudication do not account for legal and practical limitations at borders, and, thus, the ECtHR’s version of the facts tends to follow the states’ version of facts. By exploring a series of reasons for this, the article demonstrates how the court’s evidentiary regime for border violence constructs what in other contexts has been called a “sphere of willful ignorance” (Weisberg 2014, xii). The article argues that the presumption of the states’ good faith should no longer apply when there is evidence, including in ECtHR case law, of misrecordings and false statements about facts by respondent states. This normative argument is the building block for human rights courts to draw stronger boundaries between states’ versions of facts and their own.

Inspired by theoretical discussions on epistemic injustice and the attribution of authority to different kinds of situated knowledge (Fricker 2007; Medina 2017), this article contributes to social legal scholarship on the materiality of epistemic practices in legal processes and investigations (Latour 2010; Allen 2021) as well as on how experiences of violence get documented and translated into human rights frameworks (Merry 2006; Merry and Coutin 2014). While counternarratives to knowledge claims by state authorities about border violence create epistemic friction (Davies, Isakjee, and Obradovic-Wochnik 2022), such friction fails to materialize in the corridors of the ECtHR. As examined for the case of environmental harm for racialized citizens, for example, “violence does not persist due to a lack of arresting stories … but because those stories do not count” (Davies 2022, 3). In a quest to examine the factors that make a story count, this article’s argument pays attention to the role of legal forms in translation processes between different epistemic orders (Riles 2011).

While the idea of a direct translation of violence, power, or interests into legal decisions is necessarily an illusion, the question remains why some narratives are translated into evidence and not others. On the one hand, marginalized groups need to overcome deep power differentials when they present facts and arguments in legal proceedings (Botha and Kok 2019; Fuchs 2020, 2024). On the other hand, case files also actively erase histories (van Oorschot 2021), including by producing the nexus itself between what is known in “the world of law” and “the world out there” (van Oorschot and Schinkel 2015). For the case of other human rights violations and adjudicating bodies, for example, anthropologists have examined how bureaucratic cultural practice allows “content” to fade away and “realities” to be obscured (Cowan and Billaud 2015; Fuchs 2022). In a quest to bring together both explanatory models, our article takes seriously on equal grounds the construction of facts outside and inside a court room.

The article opens by placing the erasure of pushback evidence at the ECtHR into its contexts, discussing respectively the literature on evidentiary regimes for migration cases at the court and the legal context for border violence at European borders. The core of the article’s argument about the politics of legal facts is based on two argumentative moves, which respectively foreground external and internal perspective. First, we discuss the erasure of pushback evidence at borders, inquiring into the challenges that people on the move, civil society actors, and journalists face when seeking to produce and submit evidence about border violence to the ECtHR. Second, we discuss the erasure of pushback evidence in the court room, analyzing in detail how the court’s registry staff and judges handle and assess evidence submitted by both pushback survivors and respondent states. In our conclusion, we consider broader implications of our findings for international human rights regimes and the legitimacy of state power.

Migration Evidence at the European Court of Human Rights

Despite migrants’ needs for human rights protection, the ECtHR does not have a very strong body of case law on migrants in general (Dembour 2015, 246–47) and certainly not on cases concerning the deaths of persons trying to reach Europe (Spijkerboer 2013; Mann 2018; Costello and Mann 2020; Moreno-Lax 2020). The imbalance in how the ECtHR protects citizens versus non-citizens is rooted in the history of the ECHR (Dembour 2015, 35–61) as well as in its geopolitical context (Madsen 2017). This article contributes an evidentiary angle to existing scholarly critiques of the ECtHR’s inability to uphold migrants’ rights.

The ECtHR’s evidentiary regimes is not codified by explicit guidelines (Gunn 2020; Dembour 2023). Principles that guide the assessments of facts, such as the burden and standard of proof, need to be deducted from case law analysis. For the case of migration cases at the ECtHR, migrants usually carry the burden of proof (Baranowska 2023) and face extraordinarily high standards of proof (Keller and Heri 2014; Speck 2020). As a result, the ECtHR judges’ default position in migration cases “is to follow the state—against the applicant” (Dembour 2015, 232). In addition, evidentiary practices can be inconsistent between different rights violations. Despite similarities between enforced disappearances and some pushbacks, for example, the court has so far failed to apply the same evidentiary regime (Baranowska 2023).

The details of the ECtHR’s evidentiary regime play out in a context where the court assumes the “good faith” of states party to the ECHR. In its proceedings and judgments, actors at the ECtHR assume that states fulfill their obligation under the ECHR in “good faith.” ⁶ The requirement of “good faith” “permeates the entirety of the Convention system” (Tsampi 2020, 146) and allows the court to recognize domestic institutions as co-appliers and co-interpreters of the convention (Çali 2022, 256). As human rights systems can only lean into state sovereignty (Madsen 2019), this move is a pragmatic one that allows the court to be operational by respecting national sovereignty and avoiding backlash against its own institution.

This default position is problematic, however, when ECtHR judges disregard evidence of states’ bad faith (Çalı 2010; Heri 2020, 61). In the case of enforced disappearances specifically, for example, states have created a “strategic wall of silence” by refusing to provide evidence about facts (Keller and Heri 2014; Baranowska 2021). And in migration cases more generally, states have subverted the ECtHR by learning from the limits of the law and adjusting state practices to escape accountability, notably by giving up legal sovereignty to ensure effective sovereignty (Greenberg 2021). Finally, the rise of de facto illiberal democracies in Europe renders the Court’s assumptions of states” good faith yet more problematic (Madsen 2021). The ECtHR’s case law on Turkey, for example, does not make visible its disrespect for the rule of law itself (Çali 2022, 252).

The Cypriot case, involving non-registered pushback complaints discussed at the outset of this article, highlights how legal procedures enable ECtHR judges to come to understand (or not) the empirical realities faced by applicants. While judges will inevitably restrict themselves to clarify only those facts that they consider to be relevant for a judicial decision (Speck 2022), there is politics to what is to be considered legally relevant—both at the bureaucratic registration and judicial adjudication stage. When studying pushback evidence at the ECtHR, it is thus important to consider both the legal technicalities and the dynamics at borders.

Pushbacks at European Borders

There is no internationally agreed legal definition of “pushbacks.” Nevertheless, the United Nations (UN) Special Rapporteur on the Human Rights of Migrants (2021, para. 34) offered the following working definition of pushbacks in 2021:

[V]arious measures taken by States, sometimes involving third countries or non-State actors, which result in migrants, including asylum seekers, being summarily forced back, without an individual assessment of their human rights protection needs, to the country or territory, or to sea, whether it be territorial waters or international waters, from where they attempted to cross or crossed an international border.

Under the ECHR, pushbacks are typically litigated as violations of the prohibition of collective expulsions (Article 4 and Protocol 4) or the principle of non-refoulement (derived from Article 3). ⁷ The international human rights principle of non-refoulement is defined as the prohibition of the return of anyone who has a well-founded fear of prosecution or where there are substantial grounds for believing that they would be in danger of torture and inhuman and degrading treatment (Goodwin-Gill and McAdam 2007). ⁸ To protect this principle, individuals arriving at both sea and land borders have a right to an individual assessment of their international protection needs.

Despite a clear strong international framework for human rights at borders, European governments have increasingly invested resources into closing borders in the aftermath of the governance crisis of refugee flows in 2015. In one of countless border violence cases along the Aegean Sea in 2020, for example, journalists and civil society organizations reported how the Greek coastguard took several hours to arrive after a rescue call and then twice failed to initiate a rescue procedure (Fallong and Malichudis 2020). ⁹ This reflects a global trend in migration governance, which has shifted toward deterrence (Pickering and Weber 2014) and “crimmigration,” which means increasingly keeping people on the move from accessing territory and dealing with migration from a security and crime perspective (van der Woude, van der Leun, and Nijland 2014; Rosenberg Rubins 2022). Deterrence and crimmigration are manifest in both policy (Greenberg 2021) and adjudication (March 2021), even if at times they are dressed up within an ethics of liberal humanitarianism (Macías-Rojas 2021).

Despite varied sources of documentation, states with external European Union (EU) borders have moved to outright denying that they are pushing back people, amongst others, by purposefully mis-recording or, as in the Cypriot case described earlier, simply not identifying, registering, or recording individuals who arrive at sea or land borders (Davies, Isakjee, and Obradovic-Wochnik 2022, 7–8). Attempting to eschew responsibilities and accountability, border guards also confiscate and destroy the phones of people who seek to cross borders, thus destroying or preventing alternative means of documenting state practices at European borders (Tazzioli 2020). Finally, states have also shifted to informalizing border guard practices, thus blurring boundaries of responsibility for pushbacks as well as lowering legal thresholds at the domestic level of what is supposedly the legitimate practice at borders (Baranowska 2022).

Pushback practices and irregularities in recording them have been widely reported on, including by Council of Europe institutions. The European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment (CPT), for example, organizes visits to places of detention to assess how persons deprived of their liberty are treated. After each visit, the CPT sends a detailed report to the concerned state. In their 2020 report for Greece, the CPT (2020, 25–26) went on record stating:

As the Hellenic Police did not keep any record of the persons who had been held at the Poros detention facility, it was not possible to trace the location to which these persons had been transferred. Records at Feres and Soufli did not state whether they had been held at Poros or not.

In its report, the CPT recognized not only that pushbacks from those facilities had taken place but also that state records did not match the realities on the ground. Prior to discussing whether and how ECtHR judges acknowledge these types of evidence, this article first examines from an external perspective how evidence comes (or not) to be produced and submitted to the court.

The Erasure of Pushback Evidence at Borders

To counter state denials of pushback practices, people on the move, civil society actors, international organizations, and journalists have professionalized their approach to investigating and documenting policing practices at sea and land borders. The role of frontline responders is pivotal in the production of evidence. All individuals who initially encounter new arrivals at the EU external borders are categorized as “frontline responders,” regardless of whether they are part of grassroots movements, civic neighborhood groups, or international non-governmental organizations (NGOs). While the assistance provided by frontline responders has primarily been humanitarian, it can also take on a legal dimension. In fluid and evolving collaborations, frontline responders work alongside lawyers, who are at times situated further away from the border.

This section delves into the challenges faced by frontline responders in offering humanitarian support, documenting rights violations, and presenting cases to the ECtHR. We contend that border regimes actively expunge pushback evidence from the historical record through three distinct mechanisms. First, we explore the impact of the criminalization of frontline responders on access to justice and the creation of evidence about border violence. Second, we address disjunctions between time at the border and time at the court, even within the framework of its designated “urgent” measures. Third, we identify the active role played by state actors in eradicating pushback evidence at borders.

Criminalization of frontline responders

When borders become militarized, frontline responders are barred from directly accessing zones where the rights and needs of people on the move are violated. Domestic legal frameworks in Europe have in addition moved toward the criminalization of both people on the move and frontline responders (Rodrik 2021; Schack and Witcher 2021; Haddeland and Franko 2022). Due to a variety of factors, these processes diminish opportunities for evidence production, which in turn makes litigation both more demanding and essential.

First, the criminalization of frontline responders has made it increasingly challenging to establish contact with new arrived people. The founder and director of the Aegean Boat Report, Tommy, for instance, explained in an interview how the nature of his work had undergone a significant transformation between 2015 and 2022. When he began volunteering on Lesbos in 2015, he offered medical first aid to survivors of perilous boat journeys. Seven years after his initial trip to Lesbos, civil society organizations are no longer permitted to conduct search and rescue operations in the Aegean. Now, this Norwegian national is forced to limit himself to merely documenting pushbacks based on photo and video material sent to him by passengers on boats at risk of sinking. ¹⁰ Even so, as a frontline responder, he faces the risk of being accused of smuggling.

Second, the criminalization of asylum seekers at European borders generates new needs for legal aid and, thus, additional work pressures for legal aid providers. In Greece, for example, it is illegal to assist someone who has not yet been registered with the police. Asylum seekers who arrive by boat to the Greek islands, however, are afraid to go to the authorities as it is the Greek police that carries out the pushbacks into Turkish waters, even after people have arrived on Greek soil. In addition to the pushbacks, asylum seekers also risk life imprisonment if the Greek police accuses them of being smugglers. It is a regular practice for smugglers to leave people on the move in charge of navigating the boat journey. Nonetheless, the Italian and Greek authorities prosecute people on the move who have held the steering wheel of a boat as smugglers. In March 2020, for example, the Greek police arrested and charged the father of a boy who had drowned during the crossing with the murder of his child (Fallong and Malichudis 2020). ¹¹

In the light of these crimmigration logics, filing for interim measures with the ECtHR can be a strategic move for frontline responders to try and prevent pushbacks and oblige the police to register protection claims for asylum seekers.

Practical hurdles when filing interim measures (Rule 39) and full applications

Interim measures at the ECtHR are provisional measures issued by the ECtHR to prevent irreparable harm and to protect the rights of individuals in urgent situations while a case is pending before the court. These measures are intended to preserve the status quo until the court can make a final determination on the merits of a case within the context of a full application. In pushback cases, litigators typically ask the court to halt deportations, ensure the protection of individuals’ rights, and secure necessary interventions to safeguard the well-being of those affected.

Interviewed litigators who have filed interim measures applications for asylum seekers at European borders have explained that the court in practice often choses to suspend the examination of interim measures in order to request further information from the respondent states. In the Cypriot pushback case in 2020, for example, the court asked questions about the whereabouts of the applicants and the access possibilities to international protection at the Cypriot embassy in Lebanon. Given the fast pace at which people are pushed back, the applicants were already back in Lebanon by the time Cyprus answered the court’s questions. By asking questions to the respondent states, the court thus de facto annihilates the effectiveness of interim measures in border cases. Whether or not one assumes that the ECtHR’s registry staff is aware of the consequences of their actions depends on how knowledgeable one considers staff to be about the realities at the borders.

Additional challenges in submitting applications for interim measures stem from the inherent nature and operational schedule of the ECtHR. When discussing the court’s office hours, frontline responders regularly expressed their frustrations, sometimes outright laughing at the irony of non-aligning time frames. In his interview, the founder of the Aegean Boat Report shared:

The office hours don’t align with the needs of refugees. We can only file these applications because I have a network of lawyers willing to do this in the middle of the night, so that we can fax it through in the morning. If incidents occur on a Friday, getting to the Court becomes practically impossible since the ECtHR operates only during weekdays from 08:00 to 16:00.

The court’s website states that requests sent after 4:00 p.m. will not normally be dealt with on that day and additionally displays a list of public holidays. Even if, in practice, some high-profile interim requests might get treated outside of this rigid time frame, the temporalities of registry staff and judges do not align with the needs of asylum seekers, frontline responders, and litigators. This mundane aspect of working hours and weekends restricts the evidence that frontline responders can submit to the court.

Moreover, litigators and frontline responders encounter challenges intrinsic to the context of boat arrivals and land crossings. Another frontline responder who works for a legal aid organization on a Greek island explained the following:

In many cases, when the applicants manage to contact us, they are on the island for a few hours, so maybe their phones no longer have battery power. Also, the phone will not work if people had to jump into the water or swim in order to access the island. When their lives are at risk, they will not necessarily seek out a lawyer as a first step. If there is a person with a medical issue, the attention will go there and not to access a lawyer. Because of the emergency of the situation, often the people are splitting into groups. They are then not aware of where the other people are going. They don’t know the island. In the beginning, we might be in touch with a whole group and in the end maybe we only file for a small part of the initial group…. If they have a mobile phone, they have a sim card from another country. So, we need them to have internet data. Also, if it’s raining and the group is outside, then they have to find shelter and, the sim card doesn’t work as well. And, even if for us the applicants are a group, we have to be aware that in most of the cases, they do not know each other.

These practical challenges hold true for both interim and full applications. In addition, frontline responders need to navigate gaps between the life worlds of new arrivals and the information needs that are necessary for legal proceedings. The same frontline responder shared:

It’s clear to us what information we need and why we get in touch with them. It’s not so clear for them. In general, the people are lost and don’t understand what is going on. Why do they need to be on the phone with a lawyer who is asking questions. They are not aware of many things when they are coming. They are sometimes not even clear where they are and what their rights are. In other cases, they know because they had already been pushed back before.

All of these factors make it difficult for frontline responders and litigators to file interim applications to prevent pushbacks as well as produce evidence on border violence for full applications. In addition, states also actively obstruct the production and submission of evidence at border crossings.

States are preventing and failing to produce evidence at borders

Many state actors fail to fully and correctly record events at borders (Barker and Zajović 2020; CPT 2020, 25–26). Such non-recording is not a symptom of state failure (Kalir and Van Schendel 2017; Rozakou 2017). Instead, ignorance can be strategically and actively produced (Borrelli 2018; McGoey 2019). A close reading of the judgment M. H. and Others v. Croatia in 2021, for example, reveals that Croatian border guards intentionally turned off their radio stations to prevent the recording of evidence regarding the practices of Croatian state authorities. ¹²

State failure to comply with positive obligations to record events at borders makes it extremely difficult for litigators to rebuttal the states’ versions of facts for two reasons. First, when the required information was not produced by states in the first place, litigators and migrants cannot acquire them, for example, through freedom of information requests. Second, the absence of state documents—just like in the Cyprus case—does not seem to sway the approach of the ECtHR’s judges to evidence. In D. A. and Others v. Poland in 2021, for example, the ECtHR judges noted that the applicants were not allowed to take photos or make video recordings while at the border, but they nonetheless did not give greater weight to their oral testimonies. ¹³

In practice, most pushbacks at European borders occur without state actors issuing administrative decisions, resulting in what one could call non-paper pushbacks. ¹⁴ Here, state officials do not register individuals prior to pushing them back or while potentially detaining them prior to a pushback. The non-recording of border incidents is in most instances in violation of domestic law and Council of Europe standards. As a result, despite being under the direct control of state officials, people on the move lack registration records, expulsion orders, or detention papers that could substantiate their presence in the territory. When states issue, but do not hand over, documents, pushback survivors are not able to provide official evidence about their presence on state territory either. In Croatia, for example, the Ombudsman has documented how police officers carry out identification procedures and issue return decisions but do not hand these to the migrants at the borders (Croation Omudsman 2018).

Although most pushbacks are non-paper pushbacks, most ECtHR judgments concern paper pushbacks. It is much more difficult—although not impossible—for litigators and applicants to evidence non-paper pushbacks as judges and registry staff require evidence produced first and foremost by the state. The few non-paper pushback complaints that nevertheless make it to the ECtHR show us what exceptional circumstances need to have occurred to overcome the court’s evidentiary regime. In M. H. and Others v. Croatia, for example, pushback evidence became visible only due to the domestic criminal proceedings that had documented how a child had died at the border (Fehr and Alpes 2023). And in N. D. and N. T. v. Spain in 2020, a journalist had been able to film the pushback. ¹⁵ While judges had accepted the pushback video as prima facie evidence in this case, ECtHR judges discarded a similar case as entirely inadmissible by doubting whether the applicants were truly on the video. ¹⁶

In the following section, the article examines how judges and registry officials handle and assess the pushback evidence that people on the move, frontline responders, and lawyers are able to produce and submit regardless of the circumstances and state actions. In light of the above challenges, the question emerges whether and, if yes, how the ECtHR’s evidentiary regime accounts for the erasure of pushback evidence at borders.

The Erasure of Pushback Evidence in the Court Room

The ECtHR lacks its own operational fact-finding capabilities, preventing it from conducting on-the-ground investigations. Typically, ECtHR judges base their normative assessments on facts that have been established in domestic proceedings. In pushback cases, however, domestic remedies are often unavailable, and litigators go directly to the regional human rights court in Strasbourg. Consequently, respondent states and applicants frequently contest factors surrounding border incidents. In their judgments, however, ECtHR judges often fail to acknowledge this factual uncertainty. Instead, they narrate situations in alignment with state lawyers, without critically reflecting on the methods employed to ascertain facts.

In this section, we contend that the ECtHR’s evidentiary regime both perpetuates and intensifies the erasure of pushback evidence at borders through several legal maneuvers. First, ECtHR judges willfully disregard evidence presented by non-state actors. These non-state actors include the EU’s Fundamental Rights Agency, the Council of Europe’s Commissioner for Human Rights, the Council of Europe’s Special Representative of the Secretary General on Migration and Refugees, the fact-finding mission of the Council of Europe’s Parliamentary Assembly Committee on Migration, Refugees and Displaced Persons, the UN Special Rapporteur on the Human Rights of Migrants, the UN High Commissioner for Refugees, and international NGOs such as Amnesty International.

Second, the ECtHR’s evidentiary framework justifies assigning significantly greater weight to state-produced evidence than to evidence presented by applicants and their litigators by characterizing state-produced evidence as “direct” evidence. Third, ECtHR judges dismiss instances where states misrecord events or fail to provide evidence. Fourth, when confronted with conflicting evidence from two states, judges accord greater weight to evidence from the state whose actions at borders are under scrutiny. Consequently, the ECtHR’s evidentiary regime lacks internal consistency and logic, even if one accepts that state-produced sources should be deemed more trustworthy in principle.

Neglecting evidence from non-governmental and international organizations

While law and society scholars have extensively analyzed assumptions inherent to legal categories in different domains (Bakewell 2008; Bonjour and de Hart 2013; Wray 2015; Alpes 2017; FitzGerald and Arar 2018; Moret, Andrikopoulos, and Dahinden 2021; Welfens and Bonjour 2021), less attention has been paid to normative assumptions in legal procedures and, more precisely, assumptions about truth and the construction of facts (Lynch and Cole 2005; van Oorschot 2021).

The ECtHR’s preference for state-produced evidence becomes apparent in the judges’ implicit hierarchy of sources. There is no single written record that spells out what the court’s hierarchy of sources consists of, and judges mostly do not render explicit in their judgments how they weigh and frame evidence from different sources. Litigators are thus left to deduct preferences and past framings of different sources through case law analysis. In practical terms, this implies that litigators need to set aside what they themselves know about the research methods and familiarity of organizations with border practices and engage in informed guess work when anticipating how ECtHR judges might weight different sources.

ECtHR judges spelled out the hierarchy of sources for assessing reports by non-governmental and international organizations somewhat more explicitly only in a 2014 judgment, Georgia v. Russia (I). ¹⁷ Articulating criteria, the judgment states that the respective weight attributed to evidence depends on the reputation of the authors, the rigor of the investigation, the coherence of the conclusions, and corroboration with other sources.

In their judgments, ECtHR judges have also put aside evidence from reputable international human rights organizations and international organizations, including UN and Council of Europe bodies. In N. D. and N. T., for example, reports provided by Amnesty International and Human Rights Watch clearly documented the racialization of migrants at borders. According to their evidence, N. D. and N. T. were not able to access regular border crossings in Morocco because of their skin color. ¹⁸ Disregarding this evidence, ECtHR judges refused to engage with the factual nonexistence of access to legal entry at Europe’s borders (Hakiki and Rodrik 2023). As a consequence, and despite the available empirical evidence about realities at borders, ECtHR judges have not only failed to find violations of the ECHR, but they have also produced judgments that reveal a distorted view of border realities (Wriedt 2019).

The Court’s neglect of non-state sources is crucial for understanding the politics that play into the construction of legal facts. ECtHR judges are driven by a worldview that attributes greater trust to states than to non-state actors and consequently assumes states to be more trustworthy sources of facts. The disregard for non-state sources of evidence raises the question of how ECtHR judges justify their choices.

Framing evidence produced by states as “direct” evidence:

ECtHR judges justify giving significantly greater weight to state-produced evidence by framing state-produced evidence as “direct” evidence. The notion of “directness” has permeated the evidentiary regimes of courts in other places and domains, too. American judges, for example, have granted greater weight to evidence that was produced closer to the event because they equate temporal proximity with greater credibility. In doing so, the judges failed to consider amongst others psychological dynamics for the recollection and reconstitution of potentially traumatic events (Scheppele 1998, 323).

The court’s framing of state-produced evidence as direct evidence can be exemplified by its judgment of M. A. and Others v Lithuania in 2019. ¹⁹ In this instance, the disputed facts revolved around whether the applicants had requested asylum or not. Lithuania upheld that the applicants had not applied for asylum. As evidence to the contrary, the applicants’ lawyer submitted a photograph of fully completed and signed asylum application forms together with their train tickets from that same day. State officials at border posts control whether or not people can take photos. Hence, the applicants could not have photographed their attempts to submit the forms to the state officals at the border post.

While the ECtHR judgment eventually accepted that the applicants had applied for asylum, the court’s diversion from the state’s version of facts was achieved only by a very narrow majority. Three out of the seven judges rejected the assertion that the applicants had filed an asylum application on that day at the border, contending that no “direct evidence” had been presented. These dissenting judges alligned themselves with the state’s version of facts, arguing that the photographs “cannot be regarded as evidence that the request was actually remitted to the authorities.” According to their perspective, “the photographs show nothing conclusive apart from the train tickets and the application.” ²⁰

In their dissenting opinions, the judges did not clarify what other reasons a family might have for traveling to the border with a completed asylum application, aside from intending to file a claim for international protection. They also did not consider the practical reality of not being allowed to take photos of the state officials and border posts. Lastly, the insistence of the dissenting judges on accepting evidence solely from state authorities as the exclusive source of “direct evidence” overlooks the possibility that state authorities may, whether due to error or ill intention, fail to fully or accurately record all incidents in their offices.

The case of M. A. and Others v Lithuania illustrates how ECtHR judges assume that states produce “direct” evidence, which is superior to supposedly “indirect” evidence produced by people on the move who experience, and frontline responders who observe, border practices on the ground. From an external perspective, this assumption is problematic for several reasons. First, state-produced documents can misrepresent realities on the ground. If law enforcement agents are responsible for interviewing individuals newly arrived at borders, for example, their inquiries will primarily address topics such as smuggling routes, smuggler practices, and other factors essential for criminal investigations. Due to the questions’ emphasis on criminal law, a lack of responses regarding human rights matters does not eliminate the possibility that the interviewed person may genuinely have international protection needs. State officials charged with documenting and reporting duties might furthermore also be complicit with officials involved in the respective pushback incidents.

Second, states and their national and regional agencies are also adapting how they record events at borders. State officials can mechanically produce documents that repeat the same motive of rejection for every individual, merely rubber-stamping decisions rather than actually evidencing individualized assessments and decisions (Alpes, van Liempt, and Tunaboylu 2017). It is also possible for states or state agencies to comply with formal requirements for state recordings in border zones but, simultaneously, to erase human rights from the picture. So-called “serious incidents reports” by the EU border agency Frontex, for example, frame pushbacks as the “prevention of entry” and as “standard border practices.” While framed as a human rights accountability mechanisms, the technical frames that shape how Frontex reports on “serious incidents” actually erases the risk to people’s lives and the violation of their human rights (Bachiller López and Keady-Tabbal 2021).

Finally, state-produced paper truths risk omitting the role of non-state actors in de facto carrying out pushbacks at EU external borders. At the Greece-Turkey border, for example, journalistic collectives, such as Lighthouse Report and Bellingcat, have gathered photo and video evidence documenting the actions of masked men pushing back people across the river, sometimes even relying on asylum seekers and other migrants as drivers of boats across the river (Hadavi and Deeb 2020; Schmitz et al. 2020). Such practices are not captured by standard recording practices by states as they occur outside of its official framework. Consequently, framing only state-produced evidence as “direct evidence” erases evidence from the court’s judgments.

Having critiqued the notion of “direct evidence” from outside of the law, we will now look at the notion from an internal perspective, meaning that we will accept the supposed directness of state-produced evidence and examine how ECtHR judges mobilize the notion in different contexts.

Neglecting misrecordings by states

ECtHR judges give greater weight to evidence put forth by the state accused of pushbacks, even when there is evidence casting doubt on the state’s good faith in recording border practices. To enable this approach, ECtHR judges often disregard misrecordings by states. Given the significance of state records as crucial evidence in pushback cases, we contend that any errors made by states in their records should be thoroughly scrutinized by ECtHR judges. Such mistakes should influence the assessment of factual disagreements between the involved parties.

In Asady and Others v. Slovakia in 2020, for instance, ECtHR judges omitted an assessment of errors in state records and endorsed the state’s rendition of facts, notwithstanding evidence pointing to inconsistencies in the domestic recordings. ²¹ In this particular case, Asady, an Afghan national, along with eighteen other Afghans, filed a complaint against Slovakia regarding an alleged paper pushback to Ukraine. The disagreement between Asady’s lawyer and Slovakia centered on whether the applicants had sought asylum and whether their individual situations had been examined during the twenty-four hours they had spent on Slovak territory before the pushback.

Upon careful examination, discrepancies emerged between the claims made by the Slovakian government and its own evidence. Despite the government’s assertion of interpreters being present throughout the day, official records only confirmed the provision of interpretation services for a mere three hours. Given that there were thirty foreigners in state facilities at that time, this duration would only allow for individual assessments lasting approximately six minutes per person. In line with this hypothesis, the official records indicated that three different individuals were interviewed between 09:20 and 09:30 a.m., thus a mere ten minutes. ²² The Slovakian government dismissed both instances as mere “written errors,” a stance that the ECtHR judges accepted. Had the judges been willing to question the good faith of the Slovakian state, they would have had to conclude that no effective individual assessments had taken place, thus dismissing the government’s argument.

In Khlaifia v. Italy in 2016, ECtHR judges acknowledged the individual assessment of the applicants’ situations, despite the absence of official recordings. ²³ In this case of a paper pushback, Italy argued that individual records had initially existed but were later destroyed in a fire during riots. Here, the ECtHR judges failed to raise the additional question of how, in the absence of records, Italian authorities could ascertain that the outcomes of these assessments had been negative, justifying an expulsion. Given the lack of records, the ECtHR judges could have reasonably requested the state to provide evidence of the supposed individual assessments and their outcomes after the fire. This viewpoint was emphasized in one dissenting opinion by Judge Serghides:

Even assuming that the Government’s submission that the documents in question were destroyed in the fire on 20 September 2011 was true, since the applicants were in Italy for at least a further week the Italian authorities should have conducted another interview and should have made a fresh record—an obligation which they signally failed to fulfil. … The Government did not give any explanation at all as to why their authorities had not proceeded with a second interview, since the records of the first interview had been destroyed by fire. Even assuming that the authorities had been facing some administrative difficulties at the material time on account of the revolt, they should have abstained from proceeding with the expulsions until they were able to repeat the personal interviews. ²⁴

The dissenting judge furthermore argued that the lack of references to the personal interview in the refusal-of-entry orders is “strong indication, or even proof, that no such interview was conducted.” ²⁵ The dissenting judge’s opinion underscores that the legal facts of the case could have been assessed differently. Even though the state itself was unable to provide evidence, the majority of judges chose to accept its version of the facts.

Disregard for state misrecordings is not specific to pushback cases. In a readmission case in 2009 in Sivanathan v. United Kingdom, the government argued that the applicant had signed a letter expressing his desire to return to Sri Lanka voluntarily but claimed that the letter had been destroyed. ²⁶ In this instance, too, the ECtHR accepted the state’s version of facts despite the absence of the purported evidence (Dembour 2015, 246).

In giving greater weight to states’ versions of facts, even in the presence of evidence that raises doubts about its good faith, the court’s evidentiary regime applies different standards when assessing the credibility of evidence from state and non-state sources. This approach lacks internal consistency and logic, even if one accepts that state-produced sources should, in principle, be deemed more trustworthy. Having examined how judges handle misrecordings by states, we will now examine how judges assess conflicting evidence from two states.

Giving greater weight to evidence from states whose actions are under scrutiny

States on either side of the border involved in a pushback can diverge in their accounts of facts. Our analysis of case law reveals that ECtHR judges typically categorize only evidence submitted by the state against which the application is filed as “direct evidence.” This approach to conflicting evidence from different states assigns greater credibility to evidence presented by states accused of human rights violations and less credibility to evidence from states not directly implicated.

In M. H. and Others v. Croatia, for example, an applicant alleged a non-paper pushback from Croatia to Serbia. Croatia denied the pushback, contending that the absence of records regarding the incident constituted evidence that no pushback had occurred. On the contrary, the Serbian border police found that the pushback had indeed taken place, providing evidence on the violation of their bilateral readmission agreement, as recalled by dissenting judges in the Croatian Constitutional Court’s decision. ²⁷ Although the ECtHR judgment references this domestic judgment in its review of criminal proceedings, the judges refrained from questioning the Croatian version of the facts. In doing so, the judges disregarded state-produced evidence from Serbia about the pushback.

Not specific to only pushbacks, a case on secret Central Intelligence Agency prisons further highlights the counterintuitive nature of the ECtHR’s approach to direct evidence when there are documents provided by more than one state. In El-Masri v. Former Yugoslav Republic of Macedonia in 2012, the disagreement evolved around the question of whether the applicant had been subjected to secret detention in Macedonia. ²⁸ While Macedonia denied that it had happened, the German authorities had conducted international investigations, which had produced evidence to the contrary. In their judgment, the ECtHR considered the written statement by a Macedonian politician who served as the minister of the interior at the time of the detention as “the only direct evidence about the evidence complained of before the Court.” ²⁹ In doing so, the judges failed to consider the documents produced by the German state as “direct evidence.” The judgment literally spells this inconsistency out, stating: “The applicant’s account was supported by a large amount of indirect evidence obtained during the international inquiries and the investigation by the German authorities.” ³⁰ Here, the judges thus framed evidence as direct only when it was produced by representatives of the state responsible for the (alleged) human rights violations.

In sum, the ECtHR’s evidentiary regime employs varying standards to assess evidence from respondent and non-respondent states. Moreover, our analysis sheds light on the limitations of state-produced evidence. From the internal perspective of the law and its institutions, we demonstrate instances where state-produced evidence is inconsistent and lacks corroboration from other sources. By failing to acknowledge and address these limits, the court’s evidentiary regime further undermines its own claims to logic and consistency.

Conclusion

We opened this article with the ECtHR’s registry decision to not register twenty-six applications by pushback survivors from Cyprus. This incident raises questions about both the logic, fairness, and transparency of ECtHR decisions as well as about the erasure of pushback cases from the court’s record. ECtHR applications represent only the tip of the iceberg. Most pushbacks survivors are not able (or do not want) to file complaints against states. And most complaints are deemed inadmissible or, as in the Cyprus case, not even registered to be examined for admissibility. Moreover, pushback cases that reach the desks of ECtHR judges are not representative of the actual incidents at borders due to the the evidentiary dynamics within the court. Because of bias in favor of state-produced evidence, ECtHR judges predominantly adjudicate paper pushbacks, meaning the border incidents where states have fulfilled at least to some degree their documentary obligations. Most pushbacks at European borders, however, occur in an informal manner with states not carrying out any paperwork at all. While, by far more common, such so-called non-paper pushbacks are largely absent from the ECtHR’s case law.

In order to answer why so little available knowledge about violence at European borders becomes visible and validated in ECtHR judgments, this article has examined the court’s evidentiary regime and its implications for people on the move and their lawyers from both an external and an internal perspective. From an external perspective, we have examined the production and submission of evidence at borders, based on interviews with litigators, pushback survivors, and frontline responders. From an internal perspective, we have examined the legal assessment of submitted evidence by ECtHR judges and, to a lesser extent, registry officials. This mixing of perspectives has allowed us to examine the construction of legal facts both in contrast to epistemic orders grounded in empirical social sciences as well as in relation to the law’s own internal logics.

At borders, police officers and other state authorities erase pushback evidence by not fulfilling obligations to register individuals, by not recording all deprivations of liberty, and by controlling access of non-state actors to prevent them from recording what happens at borders. When people on the move and frontline responders nevertheless manage to produce evidence, state actors have actively destroyed it. In doing so, states have significantly reduced the pool of evidence that litigators can submit to the Strasbourg judges.

In the corridors of the ECtHR, judges frame state-produced evidence as “direct” evidence. The notion of “direct evidence” reveals a bias toward the credibility of states rather than to people on the move and civil society actors. By mobilizing this notion, ECtHR judges build the construction of legal facts for pushback cases on the assumption that people on the move and civil society members present at borders are only able to acquire and share “indirect” knowledge. Even if one were to accept the characterization of state evidence as “direct evidence,” however, our analysis exposes that the fact that ECtHR judges have deemed state evidence as such only when presented by the state against which the application is directed. Judges have disregarded state evidence presented by a country that received pushback survivors. In doing so, ECtHR judges assume states’ good faith both at borders and during the litigation process. This assumption drives the judges’ assessment of facts even when states have been internally inconsistent in their version of the facts.

Together, these framing devises and logical inconsistencies blur the boundaries between the state’s versions of facts and those presented by the ECtHR. This article hence concludes that the court has proven blind to evidence, including in the case law itself, that raises questions about states’ good faith both at borders and during litigation. Judges write up facts as well establish the standard and burden of proof with a belief system marked by deference to states. The legal procedures of the court do not hold judges accountable for how they deal with evidence and different sources for establishing the factual circumstances of cases.

Theoretically, this article has extended scholarship on the translation of experiences of violence into human rights violations by taking seriously on equal grounds the construction of facts outside and inside a court room. In doing so, the article has analyzed how boundaries between knowledge practices in the “world out there” and “the world of law” are actively produced and maintained. Studying the construction of these boundaries allows for new insights into existing discussions on roles and relationships between social scientists and legal professionals to work across and potentially shift these boundaries (Foblets, Sapignoli, and Donahoe 2024). In doing so, the article highlights possible contributions from anthropologists beyond the domain of cultural expertise (Vetters and Foblets 2016). The article also contributed to scholarship on power and non-knowledge in global politics (Aradau, Canzutti, and Perret 2023) as well as ignorance studies, notably by examining how power relations are reproduced in the legal domain (McGoey 2019). There is politics to legal procedures for the registration of cases and the factual assessment of cases by judges. Looking at willful ignorance and the politics of legal facts opens up new doors for inquiring into the respective independence of judges from government officials as well as about the construction of public truths and the role of court and state actors therein.

For litigators and civil society actors in quest of border justice, the conclusions of our article means that more evidence will not equate more justice in the legal realm. Epistemic justice requires reflexivity and transparency about why and how sources of evidence are accepted or rejected for analysis and then subsequently analyzed. If people on the move and their lawyers are to trust the epistemic practices and discretional powers of ECtHR judges, for example, then the presumption of the states’ good faith should no longer apply when there is evidence—including in ECtHR case law—of misrecordings and false statements about facts by respondent states. Whilst doing so could limit the ECtHR’s standing with states, it is possible for the court to harness internal language and existing tools when navigating such potential tensions (Baranowska, Alpes, and Kienzle, n.d.). In the bigger picture, clearer separations between the court’s and states’ interpretations of facts are essential not only for border justice but also for upholding the legitimacy of a regional human rights court. If human rights are to be a corrective to illegitimate state violence, the judicial records of courts need to accurately depict the realities of state practices.