The Good, the Bad and the Ugly

In its recent decision in ST v Frontex the CJEU missed once again an opportunity to review Frontex’s conduct in light of human rights standards. The decision is the latest in a series of key decisions concerning EU human rights responsibility over the course of the past year, including WS and Others v. Frontex, Hamoudi v. Frontex, Sea Watch v Frontex, as well as Kočner and KS and KD. While these cases concern different legal procedures, different actors and touch on diverse domains of EU law, such as integrated border management (IBM), data protection, and the EU’s Common Foreign and Security Policy (CFSP), a common thread runs through them: all arose from situations in which Member States collaborate inextricably with EU actors, and all concern the ensuing question of the EU’s legal responsibility for its complicity in human rights violations. This contribution explains how some of these cases perpetuate the shortcomings of the EU’s human rights responsibility regime, while others show the Court’s willingness and ability to redress these.

The Ugly: The EU’s Human Rights Responsibility Gap

The EU’s supposedly ‘complete system of remedies’ has faced growing scrutiny due to what is increasingly perceived as the EU’s human rights responsibility gap (see also here and here). This gap operates on two levels: first, the difficulty in holding EU actors legally responsible for (contributions to) human rights violations, and second, the challenges in attributing responsibility in cases of joint conduct between the EU and its Member States.When investigating the EU’s human rights responsibility regime – comprised of two sets of rules (liability rules and human rights obligations) across two different legal regimes (international law and EU law) – two overarching conclusions can be drawn.

Firstly, the liability rules outlined in the Articles on the Responsibility of International Organizations (ARIO) and the EU’s action for damages require, with some terminological variations, that for (human rights) responsibility to be established against an international organization (e.g., the EU), an individual must demonstrate harmful conduct attributable to the organization, causally linked to the damage. This necessitates proving four elements: attribution, causation, unlawful conduct, and damage.

The condition of attribution is particularly problematic. At least six attribution tests can be identified under international and EU law, all of which are applied inconsistently—and sometimes not at all—by courts, including the CJEU. This ambiguity in attribution tests creates a scenario where different courts—domestic and EU—apply different standards, through distinct procedures, to different actors (Member States and EU entities), for the same joint conduct and harmful outcome (see here). This disjointed approach allows both Member States and EU actors to evade legal responsibility, a risk explicitly acknowledged in Kočner (¶ 78).

Secondly, there has been insufficient theorizing, conceptualizing and concretizing about what it means for EU actors to be bound by fundamental rights, and how fundamental rights obligations embedded in the Charter translate to actionable and concrete negative, positive, procedural and substantive legal obligations that are justiciable in court (see here and here). Again, the CJEU in Kočner acknowledges as much and recalls that liability arises not only when this is provided by explicit provisions of EU law “but also by reason of positive or negative obligations which those provisions impose in a clearly defined manner, whether on individuals, on the Member States or on the EU institutions, bodies, and agencies” (¶ 119). In the relative absence of actionable human rights obligations, individuals cannot establish the condition of unlawful conduct, again impeding access to an effective remedy.

These two shortcomings reveal a flawed – dare I say, ugly – responsibility regime: poorly designed in holding EU actors legally responsible and politically exploited to let EU actors and Member States evade human rights obligations, thereby undermining the right to an effective remedy.

How do recent cases by the CJEU exacerbate or remedy these flaws?

The Bad: Evading the Question of EU Human Rights Responsibility

The EU’s IBM, and Frontex face widespread criticism for their perceived contributions to human rights violations. This has prompted several largely unsuccessful applications before EU Courts, which have generally neglected to engage in a substantive appraisal of Frontex conduct in light of fundamental rights standards.

Action for Damages

In both WS and others and Hamoudi the General Court concluded that the conduct in question (contributions to pushbacks) could not be attributed to Frontex, suggesting instead that claims of human rights violations should be directed at implicated Member States. In doing so, the General Court neglected to articulate a clearly defined standard and method of proof, and did not consider the sharing or reversal of the burden of proof as is typically done where a public authority has exclusive access to evidence of the alleged harm. Instead, it maintained that the burden of proof rested entirely with the Applicant and merely refuted the evidence offered. These rulings appear to corroborate the aforementioned conclusions inasmuch as the absence of a clearly defined attribution test allowed the General Court to skirt the responsibility question in its entirety, including a substantive assessment of whether Frontex was complicit in human rights harms (see here and here).

Action for Failure to Act and Action for Annulment

The recent order in ST v Frontex appeal adds a new dimension to the question of EU human rights responsibility, as the case concerns Frontex responsibility for complicity in human rights harms not within the context of the EU’s action for damages. Instead, the claim is brought within the EU’s failure to act procedure (Article 265 TFEU), and alternatively, within the EU’s action for annulment procedure (Article 263 TFEU).

The case centered on a denied request to suspend or terminate Frontex activities in the Aegean Sea under Article 46(4) Frontex Regulation due to serious fundamental rights concerns (e.g., non-refoulement violations). The Applicant argued before the General Court that Frontex had failed to act by not suspending or ending its operations despite widespread human rights violations. Alternatively, the Applicant sought the annulment of Frontex’s decision to continue operations, citing non-compliance with human rights obligations.

Aligning with Frontex’s submissions, the General Court dismissed the failure-to-act claim as inadmissible, holding that the Applicant had not proven they were the same individual who requested the suspension or termination. Furthermore, Frontex had effectively taken a position by refusing to suspend or terminate operations and thus could not be considered to have failed to act.

Next, the General Court held that annulling Frontex’s decision not to suspend or terminate its Aegean operations would provide no discernible benefit to the Applicant. Such annulment would neither suspend the operations nor facilitate the Applicant’s entry into Greece and would merely trigger a re-examination of the conditions to adopt a decision under Article 46 Frontex Regulation. Consequently, the Applicant failed to meet the annulment action requirement of demonstrating a vested and current interest.

The CJEU upheld the General Court’s reasoning. Regarding the annulment action, it underscored the discretion of the Frontex Executive Director in deciding whether to suspend or terminate operations due to serious human rights concerns. The Court concluded that the appeal was thus partly inadmissible and partly unfounded.

A (Not So) Complete System of Remedies?

Three types of direct actions can be brought before the CJEU to challenge human rights violations by EU actors: actions for damages, annulment, and failure to act. However, none of these avenues have led to a substantive investigation of Frontex’s conduct. The General Court has avoided any substantive appraisals of EU human rights conduct by relying on ambiguous attribution tests, undefined standards and methods of proof, and by placing the burden of proof exclusively on Applicants, who lack access to the necessary evidence.

ST v Frontex suggests that Frontex’s operational conduct (e.g., joint operations at sea) cannot effectively be challenged under the failure-to-act procedure, as long as the Executive Director responds to an Article 46(4) Frontex Regulation request, regardless of the substance of the response. EU courts place significant weight on the Executive Director’s discretion, making the failure-to-act procedure an implausible tool for scrutinizing alleged human rights violations by Frontex.

Similarly, ST v Frontex suggests that actions for annulment are not a viable pathway to contest Frontex conduct on human rights grounds. Frontex’s involvement in joint operations at sea, returns, or in third countries often creates a power imbalance between the Agency and individual applicants, as evidenced in ST, Hamoudi, Sea-Watch and WS and others, whereby Frontex retains near-exclusive control over critical evidence. This leaves individuals – already in a vulnerable situation – unable to prove their exposure to human rights violations, making it highly unlikely they can demonstrate the individual vested and current interest required for the annulment of a decision pertaining to ongoing Frontex operations. ST entrenches the evidentiary challenges faced by individuals in previous actions for damages also within the context of an action for annulment.

This leaves the action for damages as the sole potential avenue for challenging Frontex’s human rights violations. Yet here we run into the aforementioned human rights responsibility gap: Frontex’s legal responsibility will be difficult to establish, and the interplay between Frontex and Member States allows both to evade responsibility.

By eschewing substantive appraisals of EU conduct in light of human rights, these cases appear to have helped entrench the EU’s human rights responsibility gap.

The Good: EU Human Rights Responsibility Beyond Border Management

And yet, when stepping out of the Frontex silo, two significant Grand Chamber judgments by the CJEU this year, have the potential to shift the discussion on EU human rights responsibility in a more constructive – dare I say, good – direction.

Kočner and KS and KD both tackle EU actors’ human rights responsibility when collaborating with Member States. Kočner focused on joint and several liability for data protection breaches, questioning whether data protection violations were attributable to Slovakia or Europol. KS and KD examined the CJEU’s jurisdiction over CFSP mission EULEX Kosovo and its failure to effectively investigate the disappearance of the Applicants’ husbands.

Kočner is notable because it foregrounds the individual’s right to an effective remedy by relying on the joint and several liability mechanism in the 2016 Europol Regulation to ensure individuals can seek redress when harmful conduct is difficult to attribute to either a Member State or Europol.

In KS and KD, the CJEU underscores the right to an effective remedy by interpreting the CFSP exception to its jurisdiction narrowly. It affirms that the Court can review alleged human rights violations arising from CFSP missions in damage claims, provided they are not directly linked to political or strategic CFSP decisions.

Pending deeper analyses, Kočner and KS and KD are notable for several reasons. In both cases, the CJEU’s Grand Chamber addresses key issues surrounding EU human rights responsibility, particularly in cases of shared conduct between EU actors and Member States. However, the Court proceeds with caution. In Kočner the Court was careful to point to pre-existing provisions in the Europol Regulation concerning shared responsibility for data protection violations as the basis to establish Europol’s responsibility. In KS and KD the Court was careful to limit its assessment to the question of the Court’s jurisdiction for human rights harms in CFSP, which – although critiqued (Verellen) – remains rather limited (Schubert), while referring the case back to the General Court for a substantive review of the claims advanced by KS and KD.

An optimistic reading of these cases suggests that the CJEU’s Grand Chamber has recognized the urgency of addressing EU (shared) human rights responsibility (see also Krommendijk). Acknowledging the EU legislator’s role in shaping this responsibility, the Grand Chamber has relied on secondary law provisions and purposive interpretations to ensure the EU’s legal framework offers a ‘complete set of remedies’. In the Frontex saga, Article 7(4) Frontex Regulation similarly provides a legal basis for CJEU interpretation of shared human rights responsibility, which could be addressed in pending appeals and in tandem with transparency cases implicating Frontex such as the recent Sea-Watch case. But more importantly, these cases underscore that questions about EU human rights responsibility transcend the Frontex saga. Instead, they prompt more fundamental and constitutional questions about division of powers between EU Member States and EU actors and concomitant responsibilities, as well as about what type of human rights actor the EU was, is and could be.