Abstract
This paper critically examines the evolving landscape of effective judicial protection within the European Union’s Common Foreign and Security Policy (CFSP) following the case of KS and KD. While the right to effective judicial protection is enshrined in Article 47 of the Charter of Fundamental Rights and Article 19(1) TEU, the CFSP remains a constitutional anomaly due to explicit jurisdictional carve-outs laid out in Articles 24 TEU and 275 TFEU, which restrict the jurisdiction of the Court of Justice of the European Union (CJE]U). The paper argues that, although the CJEU consistently claims to enhance individual remedies, its case law reveals a primary commitment to preserving the structural coherence and uniformity of EU law, sometimes paradoxically at the expense of legal certainty. From this vantage point, the case law of the Court, culminating in KS and KD, has failed to take seriously the important role that Member State courts could play in ensuring effective judicial protection in the CFSP. Meanwhile, KS and KD may just mark the beginning of a new line of case law, with the CJEU facing pressure to adapt EU non-contractual liability rules so they can function as human rights remedies. Whether the Court will achieve this remains unresolved.
Introduction
The right to effective judicial protection is a cornerstone of the European Union's legal order. Enshrined in Article 47 of the Charter of Fundamental Rights and reinforced by Article 19(1) of the Treaty on European Union (TEU), it guarantees individuals access to remedies before independent and impartial tribunals. This right is not merely procedural; it is foundational to the rule of law, ensuring that rights conferred by EU law are not merely theoretical but practically enforceable. 1 To ensure that rights are effectively protected, the EU Treaties put in place a system of judicial protection that, according to the Court of Justice of the EU (CJEU), is both complete and coherent. 2 It is complete in the sense that, as a leading textbook puts it, ‘sufficient legal remedies and procedures exist before the Union Courts and the national courts to allow for the enforcement of Union law rights and to ensure the judicial review of Union acts.’ 3 It is also coherent, because ‘there exist both direct and indirect routes by which to enforce rights based on Union law and to review the legality of Union acts.’ 4
Yet the architecture of EU judicial protection is not uniform across all areas of EU activity. Nowhere is this more evident than in the Common Foreign and Security Policy (CFSP), where the jurisdiction of the Court of Justice of the European Union (CJEU) is explicitly limited by Articles 24 TEU and 275 of the Treaty on the Functioning of the European Union (TFEU). These carve-outs create a constitutional anomaly – a lex imperfecta: a core area of EU external action that is, at least on paper, partially insulated from judicial scrutiny. 5
This paper critically examines how the CJEU has interpreted these jurisdictional limits. It argues that the Court has consistently sought to assert jurisdiction in CFSP matters – not primarily to ensure that individuals have access to remedies, but to preserve the structural coherence and unity of the EU legal order. While the Court frequently invokes the principle of effective judicial protection, expressing a concern for the ability of individuals to obtain relief for infringements of their rights by the EU, its reasoning reveals a deeper concern with maintaining the uniform interpretation and application of EU law. This structural preoccupation has led the Court to stretch the Treaty text, often at the expense of legal certainty.
The paper challenges the assumption that judicial protection in the CFSP must be guaranteed by the CJEU. It argues that a more normatively coherent approach would have been for the Court to respect the jurisdictional carve-outs and to recognize the role of national courts in providing effective remedies. Article 19(1) TEU imposes a positive obligation on Member States to ensure effective legal protection in all fields covered by Union law, including the CFSP. National courts, while not well placed to ensure the uniform interpretation of EU law, are capable of offering individual redress. Their role is not merely residual; it is constitutionally mandated. Following KS and KD, however, their role will likely remain limited. From the vantage point of individuals seeking redress, this is not necessarily a positive development as it remains to be seen how the CJEU will grapple with the complexities of applying – and where necessary adjusting – its non-contractual liability rules to cases involving human rights violations.
The paper proceeds in three parts. The first part (section 2) sets out the legal and conceptual foundations of effective judicial protection in EU law, distinguishing between its structural and individual rights dimensions. The second (section 3) examines the evolution of the CJEU's case law on the CFSP, with a focus on how the Court has interpreted its jurisdiction in light of the Treaty carve-outs. It shows that the Court's reasoning is driven more by structural concerns than by a commitment to individual remedies. The next part (section 4) shifts the perspective by exploring the implications of the Court's broad reading of its own jurisdiction in KS and KD for the ability of individuals to obtain effective relief for human rights violations.
Effective judicial protection in EU law: A primer
As mentioned, the principle of effective judicial protection is a cornerstone of the European Union's legal order. It is enshrined in Article 47 of the Charter of Fundamental Rights and reinforced by Article 19(1) TEU. These provisions establish that individuals whose rights under EU law are violated must have access to a remedy before an independent and impartial tribunal.
The EU's system of legal remedies and procedures operationalizes the principle of effective judicial protection. It is structured around a division of responsibilities between the Court of Justice of the European Union (CJEU) and the national courts of the Member States. This multilevel architecture reflects two interrelated dimensions of the rule of law. 6 The first is a structural dimension, which concerns the uniform interpretation and application of EU law across the Union. It is focused on the preservation of legality; its normative object of concern is the protection of the legal order as a whole. The second is an individual rights dimension, which concerns the availability of effective remedies for individuals whose rights under EU law have been infringed. It is focused on the protection of the individual, be it vertically, against unfair or arbitrary treatment by public authorities, or horizontally, against actions or omissions by other individuals.
The CJEU plays a central role in safeguarding the structural integrity of EU law. 7 It ensures, as Article 19(1) first subparagraph TEU puts it, that in the interpretation and application of the Treaties ‘the law’ is observed. The CJEU does so through mechanisms such as the preliminary ruling procedure (Article 267 TFEU), the action for annulment (Article 263 TFEU) and the infringement procedure (Article 258 TFEU). These instruments are designed to maintain the coherence of the EU legal order and to ensure that EU law is applied consistently across all Member States.
At the same time, the responsibility for ensuring that individuals can obtain effective remedies lies primarily with national courts. 8 Article 19(1), second subparagraph, TEU requires Member States to ‘provide remedies sufficient to ensure effective legal protection in the fields covered by Union law’. This obligation is operationalized through the principle of national procedural autonomy: Member States are free to design their own procedural rules, provided that those rules comply with the principles of equivalence and effectiveness. 9 That is, remedies for breaches of EU law must be no less favourable than those available for breaches of domestic law, and they must not make it impossible or excessively difficult to exercise rights conferred by EU law. Article 19(1) TEU establishes a positive obligation for Member States to ensure that national courts can provide effective relief in accordance with Article 47 of the Charter. 10 This obligation highlights the role of national courts within the EU judicial system.
This decentralized model of enforcement means that national courts are the first line of defence for individuals seeking to vindicate their EU rights. 11 While the CJEU ensures that EU law is interpreted uniformly, it is primarily the national courts that apply that law in concrete cases and provide remedies where rights have been violated. The system thus relies on a functional division of labour: the CJEU safeguards the structural coherence of the legal order, while national courts ensure that individuals can obtain effective judicial protection. Only where the EU acts alone, and its actions directly affect individuals, does it fall on the CJEU to itself ensure that individuals whose rights are infringed obtain effective relief. This is relevant in areas of direct as opposed to indirect EU administration, such as competition law, where the Commission adopts decisions that target specific companies, and increasingly, migration law, where Frontex, an EU agency, plays a growing role in the patrolling the EU's external borders. 12
This division of responsibilities has important implications for areas of EU law where the CJEU's jurisdiction is limited, such as the Common Foreign and Security Policy (CFSP). As the next section will show, the Treaties restrict the CJEU's jurisdiction in CFSP matters, but they do not eliminate the obligation of Member States to provide effective remedies. Understanding the EU's system of judicial protection as a multilevel framework helps clarify how effective judicial protection could still be ensured even where the CJEU cannot act directly. As the next part will show, however, the CJEU's case law in CFSP matters has not always given due consideration to this division of responsibilities, raising important questions about the coherence of the Court's approach and the adequacy of the remedies available to individuals.
Looking back: Effective judicial protection in the CFSP from Opinion 2/13 to KS and KD
The principle of effective judicial protection, as embedded in Article 47 of the Charter and Article 19(1) TEU, applies across all fields governed by EU law. Yet its application within the CFSP has long been a source of constitutional tension. 13 The CFSP occupies a unique position in the EU's legal order: it is formally part of the EU legal order, and thus the constitutional principles governing that order, but it is subject to specific institutional arrangements and jurisdictional limitations. 14 Most notably, Articles 24(1) TEU and 275 TFEU carve out portions of CFSP activity from the jurisdiction of the Court of Justice of the European Union (CJEU), while allowing for limited exceptions.
This section examines how the CJEU has navigated the tension between the principle of effective judicial protection and the jurisdictional constraints imposed by the Treaties in the CFSP domain. It argues that the Court's case law reflects a persistent effort to preserve the structural integrity of the EU legal order – ensuring the uniform interpretation and application of EU law – even where this requires stretching the Treaty text. The Court has by contrast been inconsistent in protecting individual rights related to CFSP measures. While greater access to the CJEU can in some scenarios be beneficial for applicants, the case law reveals that this is not always the case. The Court's case law on its CFSP jurisdiction shows effective judicial protection as a secondary concern, with the structural dimension of the rule of law taking precedence.
Through a close reading of three key judgments – Rosneft, Bank Refah Kargaran and KS and KD – this section traces the evolution of the Court's approach. It shows how the Court has expanded its jurisdiction incrementally, often invoking the need for a ‘complete system of legal remedies,’ while leaving unaddressed the question of whether national courts can or should fill the remaining gaps. The analysis highlights the Court's prioritization of systemic coherence (the structural dimension of the rule of law) over individual redress (the individual rights dimension), and sets the stage for the next section, which considers the path not taken, that is, the role national courts could have played in ensuring effective judicial protection where the CJEU's jurisdiction is limited. Before moving to the case law, it is, however, important to sketch the Treaty framework.
Treaty framework: Jurisdictional limits and exceptions in the CFSP
The Treaties impose significant limitations on the jurisdiction of the CJEU in CFSP matters. These limitations are laid down in Article 24(1) TEU and Article 275 TFEU. Article 24(1) TEU provides that the CFSP ‘is subject to specific rules and procedures’ and that ‘the Court of Justice of the European Union shall not have jurisdiction with respect to these provisions,’ subject only to two exceptions. Article 275 TFEU echoes this carve-out, stating that the Court ‘shall not have jurisdiction with respect to the provisions relating to the [CFSP] nor with respect to acts adopted on the basis of those provisions’. However, it then introduces two exceptions: first, the Court retains jurisdiction to monitor compliance with Article 40 TEU, which safeguards the division of competences between the CFSP and other areas of Union law; second, it may review the legality of certain restrictive measures adopted against natural or legal persons in the context of ‘proceedings, brought in accordance with the conditions laid down in the fourth paragraph of Article 263 [TFEU]’.
This jurisdictional architecture has been described by Advocate General Wathelet as a ‘carve-out’ accompanied by two ‘claw-backs’. 15 The carve-out excludes the CJEU from reviewing most CFSP acts, while the claw-backs reintroduce limited judicial oversight in narrowly defined circumstances. The first claw-back ensures that CFSP measures do not encroach upon competences governed by the ordinary Union framework. The second allows for judicial review of restrictive measures – such as travel bans – when they directly affect individuals or entities. 16
If properly applied, the functional consequences of this arrangement are significant. Within the scope of the carve-out, the CJEU lacks jurisdiction to review the legality of CFSP decisions or explain their meaning, including those establishing or implementing Common Security and Defence Policy (CSDP) missions, unless they fall within the narrow confines of the claw-backs. This stands in contrast to the Court's jurisdiction in other areas of EU law, where the CJEU exercises broad powers of judicial review and interpretation – even if, there as within the CFSP, the powers of the Court are conferred powers, meaning that where powers are not conferred to the CJEU, Member State courts have jurisdiction. 17
At the same time, the Treaty is not entirely unambiguous. While the claw-backs are phrased in restrictive terms, the incorporation of the CFSP within a unified European Union governed by a single set of constitutional principles following the entry into force of the Lisbon Treaty pulls in the opposite direction towards a greater role for the CJEU within the CFSP. 18 This ambiguity has opened the door for the CJEU to adopt a more expansive reading of its jurisdiction in CFSP matters – particularly where the Court perceives a risk to the coherence of the EU legal order or a gap in the system of judicial protection. As the case law discussed below illustrates, the Court has seized upon this ambiguity to incrementally expand its jurisdiction, often invoking the principle of the rule of law and the need for a ‘complete system of remedies’ to justify its approach. The following analysis traces this evolution through a close reading of the Court's key judgments in Rosneft, Bank Refah Kargaran and KS and KD.
Judicial interpretation: From Rosneft to KS and KD
The Rosneft case concerned a request for a preliminary ruling made by the High Court of Justice of England and Wales. 19 Rosneft, a Russian oil company, had been the target of CFSP restrictive measures. Rosneft had challenged the implementation of those measures in the United Kingdom. Before the High Court, it had questioned the legality of the CFSP decisions on which the implementing measures were based, and had requested the High Court to ask preliminary questions on the matter to the CJEU. This led the High Court to ask the CJEU whether the latter has jurisdiction to rule on preliminary ruling requests on the validity of CFSP decisions imposing restrictive measures. The question was pertinent, considering that the second claw-back provision, Article 275 TFEU, seemingly restricts its scope to annulment actions aimed at contesting the validity of CFSP decisions that impose restrictive measures.
In response to the High Court, the CJEU held that it is possible for applicants to challenge the validity of restrictive measures adopted by a CFSP Council decision not only directly, by bringing an action for annulment on the basis of Article 263 TFEU, but also indirectly, by bringing a case before a national court, which can then issue a request for a preliminary ruling on the validity of the CFSP measure imposing the restrictive measures.
In its reasoning, the Court relied heavily on the rule of law in both its structural and its individual rights dimensions. The Court's reasoning has two, somewhat disjointed parts. 20 In a first part, the Court held that the option for national courts to ask preliminary questions on the validity of CFSP measures is crucial to ensure that individuals can enjoy effective judicial protection. The Court presented judicial review of such measures as part of the ‘complete system of legal remedies and procedures’, invoking Article 47 of the Charter, mentioned earlier. 21 Next, in a second part, the Court invoked the structural dimension of the rule of law, emphasizing how judicial review through the preliminary ruling procedure is needed to protect not only the coherence of the system of judicial protection, but also the ‘very unity of the European Union legal order’. 22 The Court referred to Foto-Frost, where it had held that, to protect the unity of the EU legal order, lower national courts had to refer validity questions to the CJEU when in doubt about the legality of a norm of EU law. In Rosneft, the Court extended the Foto-Frost logic to preliminary ruling requests on the validity of CFSP decisions imposing restrictive measures. 23
While constitutional principles, in particular the rule of law in both of its dimensions, played a central role in the Court's reasoning, the Court did try to square its conclusion with the Treaty text. Article 275 TFEU, mentioned earlier, makes clear that the CJEU has jurisdiction to rule on annulment actions. In particular, the above-mentioned claw-back covers ‘proceedings, brought in accordance with the conditions laid down in the fourth paragraph of Article 263 [TFEU], reviewing the legality of decisions providing for restrictive measures against natural or legal persons adopted by the Council on the basis of Chapter 2 of Title V [TEU].’ Article 263 TFEU covers the action for annulment. On its face, the provision suggests that the claw-back covers only annulment actions targeting CFSP decisions imposing restrictive measures. However, the Court interpreted the claw-back more broadly as covering not only the above-mentioned annulment actions, but all proceedings that allow for judicial review of such CFSP decisions.
The Court's reading of the claw-back has been criticized – and for good reason. The Court bent the meaning of the claw-back beyond what its text can bear. 24 The parallel with Foto-Frost is misleading, moreover, since the jurisdictional claw-back of Article 275 TFEU needs to be read against the backdrop of the carve-out, whereas in Foto-Frost there was no explicit Treaty language limiting the Court's jurisdiction. As a result, while in Foto-Frost the Court could be understood as filling a gap left by the Treaty framers, in Rosneft the Court moved beyond gap-filling to override the intention of the framers. 25
In Bank Refah Kargaran, the Court followed a similar argumentative path as it did in Rosneft to deal with a different question. 26 Bank Refah Kargaran is an Iranian bank. It had been the object of CFSP sanctions. In an earlier case, the General Court had ruled that the sanctions were illegal. 27 Following the General Court's judgment, Bank Refah Kargaran had brought an action for damages for the harm caused to it by the illegal CFSP sanctions. At issue in this case was whether the CJEU had jurisdiction on the basis of Article 268 TFEU to award damages for harm caused by CFSP restrictive measures that had previously been annulled by the Court.
In contrast to Rosneft, where Article 275 TFEU explicitly dealt with the issue of legality review, neither Article 24 TEU nor Article 275 TFEU explicitly mention the action for damages. The Court concluded that the CJEU did have jurisdiction to award damages. The Court highlighted that the action for damages is not mentioned in the claw-backs, and that the carve-out should be understood as an exception to the general rule, allegedly laid down in Article 19(1) TEU, that the CJEU has jurisdiction. As an exception to the general rule, the exception should be interpreted narrowly. 28 From this starting point, the Court built an argument around the need for the system of judicial protection in the EU to be complete. Without the possibility for plaintiffs to bring an action for damages for harm caused by CFSP sanction decisions, there would be a gap in the system, and the system would not be coherent. 29 Similar to the first part of the Court's reasoning in Rosneft, in Bank Refah Kargaran the Court's reasoning revolved around the rule of law's individual rights dimension.
The Court's reasoning was based on two premises. First, the Court suggested that the action for damages falls outside of the scope of the carve-out as the action for damages is not expressly mentioned in either Article 275 TFEU or 24 TEU. 30 Second, the Court suggested that, absent CJEU jurisdiction to award damages, there would be a gap in the system of judicial protection because Member States would not provide effective remedies. 31 Neither premise is convincing, however.
Regarding the first premise, it is true that the action for damages should be understood as an ‘autonomous form of action’, distinct from the avenues for legality review such as the annulment action or the preliminary ruling procedure on the validity of CFSP decisions. 32 It is not true, however, that the action for damages is not addressed in either Article 275 TFEU or 24 TEU. As mentioned, Article 24 TEU phrases the carve-out as follows: ‘The Court of Justice of the European Union shall not have jurisdiction with respect to these provisions’ – ‘these provisions’ being the TEU provisions concerning the CFSP. Article 275 TFEU phrases it as follows: ‘The Court of Justice of the European Union shall not have jurisdiction with respect to the provisions relating to the [CFSP] nor with respect to acts adopted on the basis of those provisions.’ An action for damages for harm caused by the adoption of restrictive measures on the basis of Article 28 TEU concerns the CJEU's jurisdiction ‘with respect to the provisions relating to the CFSP.’ As such, the carve-out does cover actions for damages, which means CJEU jurisdiction on the basis of Article 268 TFEU for harm caused by CFSP decisions ought to be excluded. 33
Regarding the second premise, the assumption that absent CJEU jurisdiction on the basis of Article 268 TFEU there would be a gap in the system of judicial protection is not persuasive either. The Court did not develop the point, instead referring to the Advocate General's opinion. Yet the AG had not explained the nature of the alleged gap either. 34 As will be argued in greater detail in section 4, the claim of there being a gap where the CJEU lacks jurisdiction is inaccurate. As already discussed in section 2, where the CJEU lacks jurisdiction, the obligation derived from Article 19(1), second subparagraph, TEU of Member States to provide for effective relief revives. Because of this obligation, there arguably is no gap in the system of judicial protection – at least not any more than there is a gap in other areas of EU action. 35
Taken together, the reasoning of the Court that is based on the requirement that the system of judicial protection in the EU be complete, fails to convince. The Court does not address the point that the action for damages is covered by the carve-out, and it does not engage with the point that Member State courts should step in where the CJEU lacks jurisdiction. The offered justifications falling short, it is worth speculating on why the Court nonetheless concluded that it had jurisdiction to award damages. In a multilevel judicial system consisting of the CJEU and Member State courts, the CJEU could simultaneously have the necessary jurisdiction to invalidate the CFSP decision imposing restrictive measures, and lack the jurisdiction to award damages for the harm caused by the illegal measures. From the vantage point of the individual rights dimension of the rule of law, this is not necessarily a problem. As long as the applicant obtains relief, the principle of effective judicial protection is respected. Which court offers the relief is immaterial. 36
By contrast, from the vantage point of the structural dimension of the rule of law, a conclusion that the CJEU lacked jurisdiction to award damages would have been problematic. It would have led national courts to apply tort rules, and possibly Article 340 TFEU governing the EU's non-contractual liability, to actions taken directly by the European Union. (To recall, the illegality of the measure was caused by the Council, which had adopted it.) In an arrangement where the CJEU lacks jurisdiction, there would be no guarantees that EU law be applied and interpreted uniformly. From the vantage point of the unity of the EU legal order, this would not be a desirable outcome.
The prospect of national courts applying EU liability rules may have led to unease within the CJEU. Such unease could explain the Court's reference to the ‘complete system of legal remedies and procedures’, first introduced in Les Verts. 37 As the Court held in that judgment, the complete system was designed specifically ‘to permit the Court of Justice to review the legality of measures adopted by the institutions.’ Les Verts thus established a connection between effective judicial protection and the central role of the CJEU in ensuring the uniform interpretation and application of EU law. The suggestion was and remains that, for judicial protection to be effective, it must be provided by the CJEU itself. Les Verts remains a crucial judgment today, as it is emblematic of the conceptual confusion between the structural and the individual rights dimensions of the rule of law in the CJEU's case law, including in the three judgments discussed in this paper. This confusion enables the Court to state in Bank Refah Kargaran that ‘the principle of effective judicial protection of persons or entities subject to restrictive measures requires, in order for such protection to be complete, that the [CJEU] be able to rule on an action for damages brought by such persons or entities seeking damages for the harm caused by the restrictive measures taken in CFSP Decisions.’ 38
KS and KD was a case brought by family members of residents of Kosovo who had gone missing during the hostilities that took place in Kosovo in the late 1990s. In the framework of its CSDP mission in Kosovo, the EU had committed to investigating the disappearances. The plaintiffs considered that the EU's efforts to do so had fallen short. The family members claimed damages before the General Court. The General Court had ruled it lacked jurisdiction because of the carve-out. 39 The plaintiffs appealed before the CJEU, requiring the CJEU to determine whether it has jurisdiction to award damages for harm caused by EU CSDP missions. As was the case with Bank Refah Kargaran, KS and KD thus concerned an action for damages for harm caused within the framework of the CFSP. However, while Bank Refah Kargaran concerned harm caused by CFSP restrictive measures, KS and KD concerned harm caused in the framework of a CSDP mission.
The Court held that it has jurisdiction to award damages for harm caused by CSDP missions, except if the harm was caused by actions or omissions of the EU that related to ‘political or strategic choices made by the institutions, bodies, offices and agencies of the Union in the context of the CFSP, and in particular the CSDP.’ 40
The Court's argumentative approach was different from the one pursued in Bank Refah Kargaran. While in Bank Refah Kargaran the Court relied heavily on structural principles – in particular the coherence of the system of judicial protection – the Court in KS and KD started by downplaying the significance of structural principles such as the rule of law and the requirement of effective judicial protection. Neither of these principles, the Court made clear, can extend the Court's jurisdiction beyond the limits set out in the Treaties. To disregard these limits, would violate the principle of conferral, the Court held. 41
Instead, the Court's interpretative approach was more minimalist, suggesting that its reading of the carve-out as being limited to issues of political and strategic significance is in keeping with the text of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU, their context and their aim. The Court here referred back to earlier case law, in particular the case of H v. Council, which was a combination of an action for annulment and an action for damages aimed at a staff management decision taken in the framework of a CSDP mission.
42
The Court had held in that case that the carve-out cannot be considered to be so extensive as to exclude the jurisdiction of the EU judicature to review acts of staff management relating to staff members seconded by the Member States the purpose of which is to meet the needs of that mission at theatre level, when the EU judicature has, in any event, jurisdiction to review such acts where they concern staff members seconded by the EU institutions…
43
The Court claimed this reading aligns with the Treaty text. Yet the Treaty text does not draw a distinction between political or strategic decisions on the one hand, and other, perhaps more mundane decisions on the other. 44 The text did not bring the Court where it wanted to land, namely at a place where it does have jurisdiction to rule on the merits of KS and KD's claims for damages. To further substantiate its decision, the Court referenced the context provided by Articles 24 TEU and 275 TFEU. In para. 119 of the judgment, it held that the above distinction is ‘supported by the context of those provisions, since it enables the effectiveness of the provisions to be preserved, without, however, unduly prejudicing the right to an effective remedy…’
The Court's contextual reading brought in the individual rights dimension of the rule of law by referencing the right to an effective remedy. The reference to the right to an effective remedy is reminiscent of the Court's reasoning in Bank Refah Kargaran. In that case as well, the Court held that it had jurisdiction to award damages to ensure that effectiveness of the applicant's right to an effective remedy, since absent CJEU jurisdiction Bank Refah Kargaran would, according to the CJEU, not have access to any court. Such an outcome would have revealed a gap in the judicial protection system. To fill the gap, the CJEU should have jurisdiction. Yet, as discussed in the above, it is not obvious that the gap actually exists, as Member States are under an obligation derived from Article 19(1) second para. TEU to provide sufficient remedies. In KS and KD, as in Bank Refah Kargaran, the Court did not develop the claim that absent CJEU jurisdiction there would indeed be a gap. Moreover, if such a gap existed in practice, it is legally irrelevant as Member States, not the EU, would be under an obligation based on Article 19(1) second subparagraph TEU to close the gap.
As a third element, the Court referred to the purpose of the carve-out, but there is no historical evidence that the Treaty framers intended to limit it to strategic or political decisions. The carve-out originated during the Constitutional Convention at the start of the millennium, leading to the draft European Constitution. The records show delegates could not agree on its exact limits. 45 This suggests caution in relying on the framers’ original intent when interpreting the Treaty, as invoking an alleged intention without documentary evidence is flawed legal reasoning. More compelling, in this regard, was the suggestion by Advocate General Wahl in the above-mentioned H v. Council case that ‘had the drafters of the Treaties had the intention of introducing an exception of a more limited scope to the CJEU's jurisdiction – essentially confining it to the acts of sovereign foreign policy – they would have drafted Articles 24(1) TEI and 275 TFEU differently.’ 46 They did not do so, which suggests they did not intent to limit the scope of the carve-out to issues of ‘high politics’.
With the text, context and aim not offering solid support for the Court's conclusion, the structural dimension of the rule of law reemerges as an explanatory factor. The structural dimension of the rule of law is not visible in the Court's reasoning in KS and KD. Yet between the lines one can discern the same unease as in Bank Refah Kargaran with the possibility of Member State courts being asked to rule on the merits of non-contractual liability claims against the EU. In para. 90 of KS and KD, the Court stated that ‘it is settled case-law that the Court of Justice of the European Union has exclusive jurisdiction in disputes involving the non-contractual liability of the European Union, to the exclusion of national courts and tribunals…’ 47 However, the exclusive nature of the Court's jurisdiction in this space is not self-evident. The justification for such exclusivity would be functional: absent exclusive CJEU jurisdiction, national courts could rule differently on similar claims brought by applicants in different Member States. This concern is real, but it has little to do with the right to effective judicial protection. It is instead a concern that a lack of exclusive CJEU jurisdiction – and a fortiori a complete lack of CJEU jurisdiction – would represent a threat to the unity of the EU legal order, as it would lead national courts to apply Article 340 TFEU without the CJEU having control over how it is interpreted in the framework of actions that concern the CSDP.
In the final analysis, KS and KD can be understood as another instance of the CJEU conflating the two dimensions of the rule of law: the individual rights dimension on the one hand, and the structural dimension on the other. The CJEU extended its jurisdiction to award damages as it found it difficult to accept the risk of interpretative fragmentation that comes with a lack of CJEU jurisdiction.
At the same time, the Court did not display a principled commitment to protecting the right to an effective remedy of the applicants. Paradoxically, the Court's solution in KS and KD may make it more challenging for applicants to obtain effective relief. Contrary to the suggestion of Advocate General Ćapeta, who had advised the Court to rule that it always has jurisdiction to award damages for fundamental rights infringements committed in the framework of the CFSP (including the CSDP), 48 the Court drew a distinction that will require applicants to litigate before the CJEU whether or not a decision or omission was sufficiently mundane (and thus not political or strategic) for it to fall within the scope of the CJEU's jurisdiction. Where the CJEU declares an action or omission of the EU to constitute a strategic or political decision, the matter would in theory revert to the national courts. However, national courts are unlikely to accept jurisdiction to assess a claim for damages for harm caused by an action or omission of the EU that the CJEU itself has declared to be an issue of ‘high politics’. Faced with such a label of ‘strategic’, Member State courts are likely to decline to entertain CFSP challenges, invoking comity or a lack of expertise. As a result, applicants may find it harder to obtain relief. If this indeed turns out to be the case, the approach taken by the Court in KS and KD will have created rather than closed a gap in the system of judicial protection of the EU as there is no reason to believe that strategic or political actions or omissions never cause harm to individuals.
This outcome demonstrates how concerns for the effective judicial protection of individual rights have not been the CJEU's main object of concern. Effective judicial protection and, more generally, the individual rights dimension of the rule of law, are relied upon to justify expanding the CJEU's jurisdiction within the CFSP. This may be beneficial for applicants, as arguably it was in Bank Refah Kergaran, but it may also be bad news for them, making it harder to obtain relief.
Looking ahead: Effective judicial protection in the CFSP after KS and KD
Unless KS and KD are overturned, an important result of the Court's decision will be that the CJEU is likely to become the main venue for cases about non-contractual liability related to harm occurring during CSDP missions. Looking ahead, this implication raises new, more substantive questions for victims of human rights abuses seeking relief. In the years ahead the CJEU will likely have to grapple with a series of substantive questions that arise as applicants start to rely on Article 340 TFEU governing the EU's non-contractual liability in search of relief for harm caused by actions or omissions of the EU institutions – or Member States – in the context of CSDP missions. This context may require the Court to revise important tenets of its existing case law, which has developed in a very different context of companies – often large companies with significant financial resources – seeking compensation for harm caused by EU regulatory activity. 49
Such questions will likely concern the attribution of the conduct that caused the harm as well as the appropriate burden of proof. Regarding attribution, questions may arise on whether the actions or omissions should be attributed to the EU or to the Member States. Where individuals are harmed by actions or omissions related to a CSDP mission, that action or omission can be attributable to either the EU or one or several Member States. To be sure, in many instances, the action or omission will be attributable to the EU. This will be the case in particular where the harm is done by an EU staff member who was acting ‘in the performance of its functions’. 50 By contrast, where the harm is done by a Member State official seconded to an EU mission, attribution would become more complicated to assess as the conduct of the official can only be attributed to the EU if the EU was exercising ‘effective control’ over the Member State official. 51 As the earlier case of H v. Council concerning a decision to relocate an Italian official working for the EU's rule of law mission in Bosnia made clear, it is not always straightforward to determine whether or not the EU was indeed exercising ‘effective control’ over the official. 52
Attribution questions may also arise horizontally, between EU institutions and bodies. As a matter of international law, such questions are moot: the international organization, which has legal personality, is responsible for harm caused by its bodies. Yet as a matter of EU law, the question becomes more complicated in so far as it intersects with rules of EU procedural law that require applicants to bring proceedings not against the EU qua international organization, but against the specific institution or body allegedly responsible for the conduct. This requirement may derail cases as the choice of which institutions or bodies to target has to be made at the very start of the proceedings. Unlike in some national contexts, applicants cannot compel parties that were not summoned at the start of the proceedings to intervene at a later stage. 53
The lack of a procedural mechanism to compel third parties to intervene in ongoing proceedings may pose challenges to applicants. Specifically in KS and KD, it led the General Court, on remand from the Court's judgment discussed above, to dismiss the action by means of an order rather than a judgment, because in its view the conduct that had caused harm to KS and KD could not be attributed to any of the EU institutions KS and KD had targeted in their action for damages. In particular, while KS and KD had brought their action against the Council, the Commission and the European External Action Service (EEAS), they should have targeted Eulex Kosovo – the CSDP mission – directly. 54 As a result, the entire action was dismissed as inadmissible. 55
Questions may also concern the burden of proof: human rights victims may be vulnerable individuals with limited financial resources who may not be able to provide the type of proof required by the CJEU for the establishment of the EU's non-contractual liability in litigation brought by companies seeking relief for harm caused by EU regulatory actions. In this respect, a case such as KS and KD shares similarities with non-contractual liability cases that have been brought before the Court in recent years in the migration context. Drawing on ECtHR case law and EU consumer protection legislation, Advocate General Norkus proposed in the case of Hamoudi that the CJEU relax its burden of proof to offset the power asymmetries between applicants (asylum seekers) and defendants (EU institutions and agencies and Member States). 56 Given the presence of similar power asymmetries between the parties, such a relaxing of the burden of proof may be needed in the CSDP context as well.
The above questions give a sense of how the centralization of the human rights responsibility contentieux involving CSDP missions in the hands of the CJEU may complicate efforts of victims to obtain relief. While such centralization may take away concerns about forum shopping and safeguard the unity in the interpretation and application of EU non-contractual liability rules, it also requires victims to make their way to Luxembourg and to participate in efforts to construct an EU-specific non-contractual liability regime that meets the specific needs of victims of human rights violations. As the above-mentioned order of the General Court in KS and KD demonstrates, even when the preliminary question of jurisdiction is settled, hurdles remain, and it remains to be seen if the action for damages can indeed be transformed into an effective fundamental rights remedy.
That human rights victims will need to bring proceedings before the CJEU is all the more unfortunate since Member State courts could very well have offered the type of financial or declaratory relief victims are seeking. In an alternative universe in which the Court of Justice would have concluded that the CJEU lacks jurisdiction to award damages for harm caused by actions or omissions of CSDP mission, Member States could have taken action to make it easier for individuals to obtain remedies before national courts. The Council decision establishing a CSDP mission could contain a clause granting jurisdiction to the courts of one or several specific Member States. 57 EU institutions other than the CJEU could have offered guidance to national courts on the applicable liability principles, thereby partially compensating for the guidance ordinarily offered by the CJEU. Member States could have facilitate judicial cooperation by agreeing on rules to recognize rulings by courts in other Member States or on rules that prevent parallel proceedings in different Member States by requiring courts to stay proceedings if another court is already seized of the same or related matter (that is, so-called lis pendens rules). 58 With regard to enforcement, the EU could have committed to complying with national court rulings and waive the immunity of enforcement it ordinarily enjoys. 59 More straightforward still, the EU could have committed in tempore non suspecto to complying with Member State court rulings awarding damages for harm caused in the framework of the CFSP. Taken together, such measures could have allowed the Member States to offer remedies that meet the conditions of Article 47 of the Charter.
Following KS and KD, whether such alternative strategies involving Member State courts may have left individuals better or worse off is likely to become a counterfactual question. Nevertheless, the Court's consistent emphasis on the structural aspects of the rule of law, rather than individual rights, raises ongoing questions about its capacity to provide the necessary support and protection for ensuring effective judicial safeguards for vulnerable human rights victims.
Conclusion
This paper has examined the evolving contours of effective judicial protection in CFSP, focusing on the CJEU's case law and the potential role of national courts. It has argued that key judgments of the Court – particularly Rosneft, Bank Refah Kargaran and KS and KD – reflect an eagerness to assert jurisdiction in CFSP matters. However, this eagerness is not primarily driven by a concern for the individual rights dimension of the rule of law. Rather, it is motivated by the Court's structural preoccupation with preserving the uniform interpretation and application of EU law and, ultimately, its own position as the EU institution responsible for ensuring that in the interpretation and application of the Treaties the law is observed.
This structural concern has led the Court to stretch the Treaty text, often at the expense of legal certainty. While the Court invokes the principle of effective judicial protection, both its reasoning and the practical implications of its judgments reveals that the individual applicant is not the central object of concern. In some cases, such as Bank Refah Kargaran, the Court's assumption of jurisdiction may indeed benefit the applicant. In others, such as KS and KD, it may make access to relief more difficult. What unites these cases is not a steady commitment to individual redress, but a consistent effort to preserve the unity of the EU legal order.
Looking ahead, the paper highlighted a number of questions that may come up in future claims for damages related to human rights violations during CSDP missions. It remains uncertain whether the CJEU will choose to modify key aspects of the EU's non-contractual liability framework to better protect vulnerable human rights victims. The Court's focus on the structural aspects of the rule of law – sometimes at odds with individual rights – makes it hard to foresee if such changes are likely.
Finally, a remark about the EU's possible accession to the ECHR. Before Rosneft came Opinion 2/13 in which the CJEU had identified the limits on its jurisdiction within the CFSP as one of the main impediments towards the EU's accession under the terms of the then draft accession agreement. 60 In her opinion in KS and KD, Advocate General Ćapeta had explicitly referred to the ongoing negotiations towards a revised accession agreement as a relevant piece of background information. 61 Member States had argued in the same case that the Treaty obligation in Article 6(2) TEU for the EU to accede to the CJEU should lead the Court to interpret its jurisdiction as widely as possible.
It is difficult to detach the legal questions regarding the CJEU's role within the CFSP from the broader political context in which they arise. However, there are sound constitutional reasons not to attach any legal significance to the negotiations that were taking place while KS and KD was being litigated. Where the CJEU disregards the Treaty text, as it has done most clearly in Rosneft but, as has been argued in this paper, also in Bank Refah Kergaran and KS and KD, the Court is implicitly amending rather than interpreting the Treaties. In so doing, the Court undermines its own legitimacy as the institution charged with the latter, but not the former. Treaty amendment requires unanimity among the Member States. At present, there is no such unanimity to expand the CJEU's jurisdiction within the CFSP, leaving the EU, the Member States and applicants in search of relief only with second-best options.
Footnotes
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author received no financial support for the research, authorship and/or publication of this article.
