Who Are the ‘Gatekeepers’?: In Continuation of the Debate on the Direct Applicability and the Direct Effect of EU International Agreements

Who Are the ‘Gatekeepers’?: In Continuation of the Debate on the Direct Applicability and the... Abstract The article revisits the debate on direct applicability and direct effect of EU international agreements by questioning the role of the so called gatekeepers. It considers the established role of the Court of Justice of the EU as the gatekeeper of the EU legal order through identifying the stages of gatekeeping and their implications. It further analyses the possibilities of sidelining the Court through various techniques, which include the agreement between the parties to the international agreement. A more controversial challenge to the Court’s position stems from a practice emerging from Council decisions concluding a number of international agreements. These decisions make a strong pronouncement on the exclusion of direct effect for the entire agreement. The status of such pronouncements is analysed with reference to CJEU’s jurisprudence as well as the relevant rules of international law. I. Introduction The reception and the status of international law in domestic legal orders is a problem which many modern states continue to grapple with. Similarly, in the European Union (EU), the last decade witnessed a renewed debate on the relationship between the EU legal order and international law, not least due to the Kadi saga and Opinion 2/13.1 The openness of the EU legal order to wider international law has been increasingly challenged.2 Certain recent cases, denying direct effect to such multilateral agreements as the United Nations Convention on the Law of the Sea (UNCLOS) and the Kyoto Protocol to the United Nations Framework Convention on Climate Change,3 also raised question marks overs the EU’s commitment to international law and emphasized the ‘gatekeeping’ exercised by the Court of Justice of the EU (CJEU).4 As far as international agreements are concerned, the gatekeeper role has been assumed by the Court in a legal vacuum. Despite the significant developments in international law propelling the role of the individual in terms of creation of rights and imposition of personal responsibility,5 there is no general rule in international law imposing an obligation on states or international organizations, such as the EU, to satisfy rights of individuals stemming from international treaties in domestic courts.6 Neither does international law establish rules on the incorporation of an international agreement into the domestic legal systems. These matters are decided internally on the basis of national constitutions or other foundational documents, or through institutional intervention. The original Rome Treaty and its successive revisions neither referred to the status of international agreements nor their effects merely stating that agreements were binding upon the EU institutions and Member States.7 The judicial intervention on the matter initially came through the seminal Haegeman judgment according to which the provisions of international agreements upon their ratification by the EU become part of its legal order.8 The Court broke further grounds by finding that international agreements can have direct effect by transposing its ‘internal’ concept of direct effect into its external relations.9 Established in Van Gend en Loos,10 direct effect recognized the justiciability of the provisions of the Rome Treaty and came to be viewed as a ‘defining characteristic of EU law’.11 Any provision of the Rome Treaty, which was clear, unconditional, and required no further implementing measures was capable of direct effect. The paradox of the extension of this doctrine to international agreements lies in the fact that the doctrine helped to ‘define’ the EU legal order ‘in opposition to international law’,12 largely due to reversing the presumption that international treaties were not capable of having direct effect.13 In simple terms, the extension of direct effect to international agreements renders their provisions justiciable without internal implementing measures.14 The externalization of direct effect led to a gradual moulding of its rather flexible criteria on the basis of which various bilateral and multilateral agreements were found to have justiciable provisions, with some notorious exceptions, namely the persistent finding of the lack of direct effect of agreements under the Gereral Agreement of Tariffs and Trade (GATT) or the World Trade Organization (WTO).15 Most significantly, the direct effect became a fixture of a vast amount of cases involving challenges to the actions of both the Member States and the EU institutions in light of the commitments undertaken through EU international agreements. The decades of jurisprudence appeared to settle the Court’s gatekeeper or ‘door opener’16 role. The latter, however, is not set in stone. While the Court’s findings on the incorporation of international agreements remain largely unchallenged by other institutions, the opposite seems to be true in relation to direct effect. A number of recently concluded agreements, in a rather unusual fashion, set out the effect of their provisions themselves, leaving no choice to the CJEU and signalling the intention of the treaty-making institutions to take back control over the matter. In addition, the Council of the EU in its decisions concluding certain agreements makes strong statements on the exclusion of direct effect for entire agreements.17 In the past, a similar statement made in the Council Decision adopting the WTO Agreement, did not receive any clarification as to its status by the Court of Justice.18 What weight should then be accorded to these decisions? This article aims to answer this and other questions by revisiting the issue of who the gatekeepers are. The discussion is restricted to international agreements (and decisions of bodies established under the latter), and does not include customary international law as the trends described above are particular to international agreements (and by implication to the bodies mentioned above). The role of the Court of Justice as the main gatekeeper to international law is considered first. For this purpose, the Court’s key jurisprudence on the issue of the incorporation of international agreements and direct effect is analysed. Next, the sidelining of the Court as a gatekeeper is questioned through two techniques: either through the agreement between the parties or through unilateral pronouncements by the Council of the EU. The article is concluded with a brief summary of findings. To start with, a clarification of the terms used in this article is in order. II. A brief terminological clarification The Court’s gatekeeper role evolved primarily through its findings on the issues of the incorporation of international agreements in the EU legal order and the justiciability of the provisions of international agreements. Various terms have been used in the past to deal with these two issues. While some scholars find the term ‘self-executing treaties’ helpful in analysing the effects of EU international agreements,19 the latter never became a fixture of EU law unlike in the US legal order.20 The concepts of ‘direct applicability’ and ‘direct effect’ are the ones that are used in EU law most commonly even though accompanied by lack of clarity inter alia due to their interchangeable use both in jurisprudence and scholarship. Early on, Winter cautioned against treating these concepts as equivalent.21 Many have debated the linkages and correlation between the two but their interchangeable use has continued.22 The reason for this might be in the very fact of interchangeable use of these terms by the Court to denote the same notion. This lead to many commentators distinguishing them within the parameters of one single concept. Winter himself viewed the concept of direct applicability as one relating to ‘the specific nature of the treaty contents’ which he advises should not be confused with the issue of incorporation, even though he does not object against using the term in the latter sense in the internal EU legal order.23 Cheyne also viewed both concepts of direct applicability and direct effect as relating to the effects of international agreements without necessarily separating the issue of their incorporation: direct applicability relates to such features of the specific provision, as clarity, precision, unconditionality and lack of further implementing measures, direct effect is a narrower concept denoting the possibility of individuals relying on a particular provision.24 These two issues, however, are both encompassed within the notion of direct effect: one is part of the concept itself, the other refers to its conditions. Most recently, Lenaerts, writing extrajudicially, also viewed direct applicability and direct effect as two distinct notions: direct applicability denotes whether an international agreement requires further implementing measures to be deduced from the parties’ intentions, most commonly by examining the nature and the logic of the agreement, while direct effect is simply a quest into the provision’s unconditional and precise characteristics.25 Such distinction is somewhat problematic if one turns to Demirel, considered to be the classic authority on what constitutes direct effect: A provision … must be directly applicable when, regard being had to its wording and the purpose and nature of the agreement itself, the provision contains a clear and precise obligation which is not subject, in its implementation or effects, to the adoption of any subsequent measures.26 The case sets the conditions for direct effect to include an inquiry into the nature and purpose of the agreement, and the provision itself to determine whether it is clear, precise and requires no further implementation in light of the agreement. Here, as an example of interchangeable use, direct applicability is applied to denote direct effect. While the Demirel definition is not without its critics,27 the latter and the subsequent jurisprudence of the Court does not necessarily make the distinction between direct applicability and direct effect in a manner suggested by Lenaerts. Instead, such delineation is somewhat problematic in view of the Court’s practice on applying the conditions noted above. Even though the application of these conditions is characterized by significant flexibility, discussed further below, such delineation fragments the concept of direct effect and is capable of adding further confusion. First, the analysis of the nature and the broad logic of the agreement focuses on a much wider range of factors than whether the agreement requires implementing measures or not.28 Besides, the need for further implementing measures is usually considered in relation to the second Demirel condition, that is whether the provision is clear, sufficiently precise and requires no further implementing measures, which indeed would be determined within the wider context of the agreement. A taxonomical clarification was also done in passing by the Advocate General Sharpston in Brown Bear according to which the term direct applicability—corresponding to the notion of ‘self-executing treaties’—denotes the instances where international agreements require no EU or national implementing legislation.29 It is not clear, however, whether the lack of the need of implementing measures is about the incorporation of the international agreement into the EU legal order or its effects in courts, which are two distinct matters.30 This terminological confusion even led to suggestions to dismiss the familiar terms in favour of new concepts such as ‘direct judicial enforceability’,31 ‘invocability’,32 or ‘direct invocability’ as ‘the capability of a legal subject to rely on (ie use or invoke) that norm in a particular context’.33 However, calls to replace the familiar terms remain unanswered, as direct applicability and direct effect are being continuously used. These terms, nonetheless, are well suited to denote two distinct but related issues: direct effect, pertaining to the justiciability of a particular provision (in its broader understanding),34 can be distinguished from direct applicability referring to a separate issue of incorporation or ‘automatic integration’ of the EU international agreements into the internal legal order of the EU and its Member States without the need for transposing measures.35 Distinguishing direct applicability from direct effect in this manner is preferable for a number of reasons. First, it is semantically more accurate. Second, it avoids fragmenting the concept of direct effect as applied by the CJEU. Finally, it is more compatible with the internal use of the term. The reference to regulations in Article 288 of the Treaty on the Functioning of the EU (TFEU) uses direct applicability not to denote its effects but rather the mechanism of its automatic transposition into the legal order of the Member States.36 Transposition measures are distinct from implementing measures, as individual provisions of regulations (as well as Treaties) might require implementing measures. The Court’s case law on the direct effect of directives is particularly instructive in the internal distinction made between direct applicability and direct effect.37 Making the distinction between these two terms in the fashion suggested above would be most faithful to the logic, the findings and the language of Van Gend en Loos. At the same time, we should keep a close sight of the relationship between the two concepts. Direct applicability is the precondition for direct effect as international agreements have to be incorporated into the internal legal order prior to the consideration of their effects. The Court’s has played a paramount role in developing both of these notions. III. The tales of Haegeman and direct applicability of EU international agreements as the primary stage of judicial gatekeeping The Court of Justice’s gatekeeper role was assumed first through its findings on direct applicability of international agreements. The latter is essentially about the binding nature of international agreements concluded by the EU and the EU and its Member States jointly. While in certain cases the EU can be bound by agreements it did not conclude or accede to,38 in the majority of the cases the question turns to the mechanism of incorporation of agreements concluded by the EU or the EU and its Member States jointly into the internal legal order of the EU. It is often with reference to the concepts of monism and dualism that the status and validity of international treaties in internal legal systems is determined. Many have been sceptical, however, about the ultimate utility of these concepts not only in relation to the EU, but more generally,39 since monism or dualism in their pure form are rather uncommon. Instead, alternative frameworks have been advanced to address the issue of coexistence of different legal systems.40 As the concepts of monism and dualism are not central to the discussion that follows, they are referred to when relevant for considering the pathway to the Court’s gatekeeper role. We start with the Haegeman story concerning the Court’s jurisprudence to interpret the provisions of an Association Agreement with Greece in a preliminary reference procedure. The Court in an apparent straightforward manner proclaimed that upon entering into force the provisions of EU international agreements formed part of EU legal order.41 An EU international agreement enters into force after its ratification according to the procedure set out in the Treaty,42 currently found in Article 218 TFEU. The Court’s reasoning was minimal and simplistic. It was based on the assertion that the Council decision on the conclusion of an international treaty is ‘an act of one of the institutions of the [Union]’,43 and therefore part of the EU legal order. This shortcut, however, left a glaring gap in the judgment, that is that the agreement itself is not an act of an EU institution, but rather an act of the EU as an organization.44 This approach might have been driven by the determination to comply with the wording of Article 267 whereby the Court’s interpretative function is limited to Treaties and ‘acts of the institutions, bodies, offices or agencies of the Union’.45 It, nonetheless, paved the way for conflicting interpretations. On the one hand, the reasoning was interpreted by many as an indication of the adoption of a monist system,46 or automatic incorporation.47 Others pointed to the failure of the judgment to provide clarity on whether the treaty automatically becomes part of internal EU law or by virtue of the Council’s act.48 The phrasing used by the Court—it is the provisions of the international agreement, rather than the agreement itself, that become the integral part of the Union law—is viewed as pointing towards the possibility that it is due to the Council’s act and not by virtue of the agreement on its own.49 This would be indicative of dualist characteristics. Another omission in Haegeman might indicate otherwise, though. The Court’s disregard for the nature and the form of the Council’s act (decision in the case) can be interpreted as a testimony to the latter’s insignificance. In its subsequent case law, the Court similarly did not pursue the apparent importance the national courts attached to the form and nature of the act approving the agreement.50 Besides, the emphasis put on the fact that the provisions of an international treaty become an integral part of EU legal order from the moment the latter enters into force also suggests the lesser importance of the decision.51 Various agreements were concluded both by Council decisions and regulations,52 fuelling assumptions that regulations are relied upon for agreements capable of direct effect, while decisions are used for agreements with no such effect.53 Such attempted distinctions, however, conflate the issues of direct applicability and direct effect.54 Only as late as in 2010 did the Court confirm that the form of the act is of no consequence for the issue of direct effect.55 In any event, currently Article 218 TFEU does not leave much choice: the Council ‘shall adopt a decision concluding the agreement’. This would have no bearing on the Haegeman findings. Returning to the limited reasoning in Haegeman, why was such an important question given such a facile and shallow answer? Perhaps, because the case evolved around a related but a different question. It would be recalled that Haegeman involved a challenge to the Court’s jurisdiction to interpret the provisions of the EC–Greece Association Agreement. The main findings, therefore, do not intend to address primarily the issue of the status of international law in the EU legal system or its doctrinal underpinnings. Rather, it is a shortcut to establish the Court’s jurisdiction to interpret the agreement, later also extended to decisions of bodies established under international agreements.56 The finding on the Court’s interpretative role was paramount as ‘whoever controls the process of interpretation, therewith controls the truth, or at least the meaning to be given to the text subject to interpretation’.57Haegeman, therefore, was the stepping stone for the Court’s gatekeeper role laying the foundations for further stages of gatekeeping related to the effects of agreements. Most importantly, the case signalled a clear openness to international law, despite leaving much scope for speculation on how exactly international agreements became part of EU law. Further clarification was due; although instead of answering the question how, the Court opted to answer the question why. The subsequent judgment in Kupferberg has been viewed as offering a ‘sounder’ analysis of the issue of direct applicability.58 Building upon Haegeman, the Court adds a new rationale to the automatic incorporation of EU international agreements: it derives from the assumption of obligations by the EU and its Member States towards third countries, as well as the obligations assumed by the Member States towards the EU.59 An outwards- and inwards-looking rationale should be distinguished here. The outwards-looking rationale is tied to the principle of pacta sunt servanda entrenched in the current Article 216(2) TFEU.60 The inwards-looking rationale, even though it upholds the same principle of pacta sunt servanda, acknowledges the internal dynamics of the EU. Rather than being preoccupied with the concepts of monism and dualism, the Court’s basic concern is the implementation of the Union’s commitments which might depend on the Member States.61 The latter should not be ‘undermined by recalcitrant Member States’ and stems from ‘the perceived necessity of protecting the autonomy of the [Union] legal order’.62 This obligation reinforces the internal commitments required from the Member States based on Article 4(3) of the Treaty of European Union (TEU) on duty of cooperation enabling the EU to perform its obligations internationally.63 Such clarification of the nature of the obligations of the Member States also solidifies the Court’s ‘full control’ over the application of EU international agreements in line with Haegeman.64 Both Haegeman and Kupferberg, hence, can be seen as being concerned with the internal dynamics of the EU legal system. Kupferberg, nonetheless, failed to address the shortcomings of Haegeman related to the mechanism of incorporation of agreements. Even if one interprets the cases above in favour of a monist view, it has been suggested that the Haegeman formula does not lead to automatic incorporation as there are certain qualifications to be made, including the respect for the constitutional values of the EU, and the application of this formula only in the areas of exclusive competence in the cases of mixed agreements.65 However, rather than affecting the incorporation of the agreement into the EU legal order, these factors are relevant for the issue of the validity, rather than the incorporation of the agreement, and as such do not impact the Haegeman findings. While the jury is still out on whether the EU legal order displays monist or dualist features (some argue that both monist and dualist interpretations can be advanced),66 the cases have been commonly interpreted to signal the ‘openness’ of EU law to international law. Such openness had significant implications for other actors both at supranational and national levels. It has an exclusionary impact on other EU institutions as it allows only for ex-post assessment of the essence of the rules of international law.67 The ability of other institutions to adapt to international law is curtailed,68 albeit that at the time the impact would have been greater for the Council than the Parliament which had limited powers in EU foreign relations. The exclusionary effect, however, guarantees that the application of the agreement cannot be undermined by the refusal of political institutions to transpose it into internal law.69 On the other hand, even bigger losers in this process are the Member States, as such moulding of the EU legal order is at the expense of modifications to their constitutional rules on external relations.70 While direct applicability at its core is about ‘the transfer of a provision between [two legal] systems’,71 in this case it is instead about three legal orders. The findings in Haegemen and Kupferberg determined not only the relationship between international law and EU law, but also between international law and the national legal orders of the Member States.72 This has been branded as the ‘Europeanization’ of international law which introduces a European element to the ‘“classical” dual legal relationship international law/national law’ turning it into ‘a new triangular relationship, international law/EU law/national law’.73 Such ‘indirect’ reception of international law in the legal order of the Member State,74 in combination with the principle of supremacy in EU law, introduced a detectable ‘openness’ in the internal legal orders of the Member States irrespective of their monist or dualist traditions,75 ensuring a uniformity as far as the supremacy of EU international agreements was concerned.76 While for Member States with monist legal systems, such intervention at the supranational level might not signal any drastic changes at first sight, for those with clear dualist features, this demonstrates a radical departure in terms of guardedness towards international agreements, and introduces a bifurcation between agreements accepted via EU law and those concluded by the Member States through their usual procedures requiring transposing measures.77 Others, however, argue that the openness to international law is not settled with direct applicability alone. This is where the direct effect of international agreements makes an entrance and takes central stage in this debate by building upon the Haegeman findings which provided ‘a basis for recognizing’ the direct effect of international agreements within the EU legal order.78 Indeed, direct effect requires prior rules settling the matter of the incorporation of internal law into an internal legal order.79 Perhaps, because the case law was not sufficiently or substantively clear on the matter of the incorporation of international law into the EU legal order, it is the concept of direct effect that has come to be seen as adding another layer of understanding to the position international agreements occupy in the EU legal order. Thus, various commentators, even though distinguishing the effects of the agreements from the issue of their incorporation, nonetheless interpret direct effect as a qualifying factor for the issue of incorporation of international treaties,80 possibly introducing a dualist element to this exercise.81 These views can be justified if one considers monism and dualism to be notions determining the relationship between different legal systems.82 More commonly, however, monism and dualism denote the process of the incorporation of a treaty into a domestic legal order.83 From this perspective, caution is required against conflating the issues of direct effect or ‘invocability of a treaty provision’ with the classification of the system as monist or dualist.84 Indeed, in this latter sense, the rejection of direct effect does not reverse the finding that EU international agreements are an integral part of the EU legal order.85 Neither does it affect the manner in which the Court arrived at that conclusion. However, rejection of direct effect can certainly cast a shadow over the exposure of EU law to international law. In this respect, direct applicability can be seen as the initial stage of gatekeeping that is static in nature, as the outcomes of individual cases cannot vary: all agreements concluded by the EU are directly applicable. The Court’s stance here can be interpreted as signalling an unconditional openness to international law. It is the further stages of gatekeeping, revolving around the concept of direct effect, that are capable of closing the proverbial gates and rendering international agreements toothless as far as their enforcement in national courts is concerned. IV. Judicial gatekeeping through direct effect: implications and phases The extension of the internal doctrine of direct effect to EU international agreements (including those concluded jointly with Member States) as a category to a wide range of agreements was seen as its ‘second major expansion’ (alongside directives) after its extension to a vast number of treaty provisions.86 Early on, the case of International Fruit demonstrated that the Court was willing to open the EU legal order to international law by considering the possibility of recognizing direct effect, although denying the latter to the GATT.87 This possibility materialized in Bresciani, the first case positively acknowledging the direct effect of the Yaoundé Convention.88 This positive finding was also extended to a wide range of agreements through the course of the next decades.89 The recognition of direct effect in principle was a progressive development as the justiciability of norms of international treaties is the exception rather than the default,90 contrary to Pescatore’s well-known remark that direct effect was ‘the normal condition of any rule of law’.91 There has even been a suggestion that in view of the proliferation of international instruments setting rights or remedies for individuals, the development of international law might lead towards ‘a more general obligation to allow invocation of treaty norms by individuals in national courts in situations where the treaty contains provisions that are protective of individuals’, stimulated by the same considerations as those that motivated the Court of Justice in finding direct effect.92 The finding that international agreements are capable of direct effect was, thus, the next significant phase in signalling openness to international law. This, however, was a two-sided coin for this general finding was accompanied by an intricate and fluid jurisprudence allowing the Court to close the curtain at different sub-phases of this exercise. We shall start with the general finding of direct effect by the Court and its implications before turning to its sub-phases. A. The implications of granting direct effect to international agreements The extension of an internally developed legal technique to international agreements was bound to have political implications. While political considerations already transpire at the stage of direct applicability, the politics is even starker with direct effect as an ‘inherently political’ concept used for political purposes,93 and according the court with a political power.94 Even the arguments made to revisit the notion of direct effect are political in nature.95 Direct effect is said to be ‘about the separation of powers, and specifically about the extent of the judicial power to enforce the obligations of the state’.96 In the context of the EU, it is the judicial power to determine the EU’s obligations, namely those of its political institutions and Member States, that is at stake. While the positive finding of direct effect was not imperative,97 as noted above, the Court opted to grant direct effect to international agreements in principle. The earlier cases, referred to in the previous paragraph, did not shed much light on the reasons behind the Court’s assumption of responsibility on the matter of the effects of international agreements. In this respect, the already familiar to us later case of Kupferberg was more significant. Here, Article 21 of the EEC–Portugal Trade Agreement prohibiting discriminatory internal fiscal measures was found to be directly effective.98 The Court dismissed the argument against direct effect on the basis of the relationship between EU institutions: In conformity with the principles of public international law [Union] institutions which have power to negotiate and conclude an agreement with a non-member country are free to agree with that country what effect the provisions of the agreement are to have in the internal legal order of the contracting parties. Only if that question has not been settled by the agreement does it fall for decision by the courts having jurisdiction in the matter, and in particular by the Court of Justice within the framework of its jurisdiction under the Treaty, in the same manner as any question of interpretation relating to the application of the agreement in the [Union].99 At the outset, the Court takes a step back. It does not claim to have an inherent jurisdiction to decide on the matter: basing its analysis on the premise of the principles of international law, the Court’s role would be ‘residual’100 or secondary to the agreement between the parties. The acknowledgement that it is the prerogative of the parties to decide is seen as homage to Danzig guidance in international law, where the object of the agreement ‘according to the intention of the contracting Parties, may be the adoption by the Parties of some definite rules creating individual rights and obligations and enforceable by the national courts’.101 However, the Court’s argument has been criticized for being ‘naïve and inconsistent’ since issues related to the internal effects of international agreements do not have a significant part in the negotiations.102 This aspect of the Court’s finding is discussed further in the article. Next, the Court’s analysis moves to the domain of EU law to substantiate its competence: this is done with reference to its jurisdiction to interpret EU law.103 Having established that international agreements upon their conclusion form part of EU legal order, by implication the Court’s jurisdiction extends to their interpretation as well. Thus, the judgment in Haegeman laid ‘the legal foundation of the Court’s competence in all cases where international obligations have been accepted by an act of the Council under treaty-concluding powers’.104 The Court’s jurisdiction to decide on the effects of international agreements was tied to the exclusivity of this exercise. In the name of guaranteeing the uniformity of EU law, the Court effectively ‘monopolised’ this issue.105 One of the most forceful arguments in favour of this finding was put forward by Advocate General Mayras in International Fruit: The unity and, it can be said, the very existence of [Union] law require that the Court is alone empowered to say, with the force of law, whether an agreement binding the [Union] or all the Member States is or is not [directly effective] within the territory of the [Union] and, if it is, whether or not a measure emanating from a [Union] institution conforms to that external agreement.106 This can be linked to one of the two, at times rivalling considerations, which motivated the Court in its finding of direct effect, that is the inclination to advance international law and its desire to preserve the ‘unique character’ and the autonomous legal order of the EU.107 The first would suggest that the very finding of the possibility of direct effect would indicate advancement of international law and ultimately its efficiency.108 Thus, one rationale in the opening of the gates to EU international treaties by according them direct effect is linked to the arguments of the efficient enforcement of law, direct effect being one of the underlying notions of the principle of effectiveness in EU law,109inter alia to mitigate the limitations of public enforcement.110 Individual reliance on international treaties strengthens their enforcement by adding a further element of supervision.111 The enforcement of international agreements through granting them direct effect is seen as an example of the Court of Justice’s so called ‘maximalist’ approach to the implementation of international law.112 The Court, therefore, assumed the role of the guarantor of the application of international law in the EU legal order. From this perspective, the Court’s leading role was needed to guarantee not only the uniformity of interpretation, but also the efficiency of Union law of which the agreements formed an integral part.113 Returning to the second consideration of preserving the ‘unique character’ and the autonomous legal order of the EC, vertical and horizontal dimensions of the institutional or power balance can be identified, that is between the EU and its Member States vertically and between the EU institutions horizontally.114 The ‘door-opening’ through direct effect furthers the impact of direct applicability on the national legal orders of the Member States. The finding that provisions of agreements concluded by the EU and third countries can be invoked in national courts on the basis of centrally set criteria sidelines domestic constitutional arrangements pertaining to the effects of international treaties with varying impact depending on the constitutional traditions of the Member States.115 But even for monist states, it signals a significant limitation of the powers of national courts since the Union concept of direct effect limits their ability to determine how far the effects of international agreements can reach.116 In this respect, the ‘Europeanization’ of international law entails the central role of the CJEU not only in relation to the status, but also to the effects of ‘Europeanized’ agreements.117 In fact, it is the ‘centralization’ of the issue of direct effect that is seen at times as the ‘crucial contribution’ of the doctrine of direct effect.118 Besides, Kupferberg was decided at a time when the Court played a ‘dominant’ role in the EU, and in view of the relative weaknesses of the EU as an emerging external actor, it preferred to be ‘closely guarded’ by (international) law.119 Should the Court have ruled out a direct effect of EU international agreements, it would have clearly undermined them as a source of EU law. Thus, the extension of direct effect to international agreements should also be placed within the context of the Court’s wider findings on external relations in this period, including its paramount role in drawing up the external competences of the EU.120 At the same time, the Court was not the only institution whose profile was to be boosted by the finding of direct effect. By giving direct effect to international agreements, the Court in addition ‘elevated the [EU] institutions as a whole in their power struggle with the Member States’ through an emphasis on the role of the Union as an international actor ‘capable of concluding treaties with direct effect and supremacy in the domestic legal systems of the Member States’.121 The finding of direct effect, therefore, might have been motivated by ‘the desire to provide an effective way of enforcing agreements against Member States’.122 Particularly in the areas of emerging EU competences, direct effect is seen as a means of affirming the powers of the EU vis-à-vis the Member States as ‘competing internal actors’.123 Another development of the doctrine should be noted here. Significantly, the Court also stretched its finding that EU international agreements are capable of having direct effect to ‘mixed agreements’ which are concluded jointly by the EU and the Member States.124 This finding is, of course, relevant for the parts of the agreements falling solely under EU competences. On the downside, while mixed agreements permit avoiding strict delimitation of competences, the possible finding of direct effect invites the Court to do exactly that.125 This might result in an interpretation of competence unfavourable to the Member States. For instance, in Demirel, the Court interpreted the EC–Turkey Association Agreement to conclude that the provision on the free movement of workers fell under EU competences against the objections of the Member States.126 Notwithstanding the affirmation of the EU’s overall position vis-à-vis the Member States, through its finding of direct effect the Court also pitched itself horizontally against other EU institutions. While in some jurisdictions, the national constitutions might provide for direct effect (even though rarely),127 in all others the choice of the institution is about the separation of powers as noted earlier.128 The judicial finding of direct effect can have implications pre- and post-conclusion of the international agreement. It can present a potential ‘threat’ during the process of negotiations undermining the position and bargaining strength of other EU institutions.129 This observation is particularly made in relation to WTO agreements. For the majority of EU agreements, however, it is the post-conclusion finding of direct effect that has a restrictive impact on institutional choices. It ‘entails a decisive shift of responsibility from the executive to the judiciary’ whereby the Court determines issues which are usually part of the responsibility of the executive.130 The legislative choices are also impacted due to the principle of supremacy where international agreements rank below the Treaties, but above secondary legislation.131 Hence, the legislation adopted subsequent to the international agreement cannot override the latter,132 rendering the legislature unable to compensate for the shift of the responsibility from the executive to the judiciary.133 Despite the encroachment upon the executive and legislative powers, there are arguments in favour of the judicial lead on this matter. Since international agreements are part of EU law and it is the Court’s task to interpret EU law, ‘it would be contrary to the structure of the [EU] to leave the auto-interpretation of international obligations to an institution other than the Court’.134 The judiciary is seen as more objective in its interpretative task in comparison with the executive,135 and is more mindful of upholding the rule of law.136 Nonetheless, as noted above, by assuming this role the Court casts itself into the political realm. The politics of judicial gatekeeping comes across most prominently through the various sub-phases of what constitutes the direct effect exercise rooted in the setting and the application of the conditions on direct effect, as well as interpreting the directly effective provisions considered in turn below. B. The setting and application of the conditions for direct effect as the sub-phases of the direct effect exercise It was clear at the outset that any automatic transposition of an internally developed concept to EU international agreements would be problematic in terms of the establishing of the conditions for international agreements’ direct effect.137 Caution was required due to the difference in context and given the political nature of the negotiation and conclusion of international agreements.138 The conditions set in Van Gend en Loos, that of the clarity, precision, and unconditional nature of the provisions, would, therefore, have to be supplemented by additional conditions.139 These conditions did not crystallize immediately, however, and a string of early case law, including cases recognizing direct effect lacked clarity in this regard.140 Two conditions can be deduced from Demirel, cited above. The first relates to the entire agreement, its nature and purpose, whereas the second focuses on a specific provision to establish its ‘normative intensity’ through ‘a positive test’.141 The initial point of the Court’s flexibility is the uncertain relationship between the two conditions. Commentators even diverge on the sequence of the conditions,142 which is reflective of the schizophrenic case law of the court, potentially changing the nature of the exercise.143 The application of the first condition will be addressed prior to returning to the issue of the relationship between the two conditions. Even though at various times the Court deployed different terminology to denote the condition on the nature and logic of the agreement,144 it is essentially a ‘policy test’,145 aimed at determining the intentions of the parties in accordance with the principles of international law.146 While the level of scrutiny of the nature, logic, structure, scheme, the spirit, etc, varies from case to case, the Court is also at liberty to imply different factors within its analysis.147 The political underpinnings of this condition were particularly evident in the WTO law-related case law which has been extensively analysed and commented upon.148 Suffice it to refer here to the factors that led to the rejection of direct effect and the reasons behind it. (i) The GATT and WTO saga Early cases denying direct effect to the GATT agreement did so on the basis of certain of its features, such as the principle of negotiations, characterized by flexibility, including the possibility of derogations and the special dispute settlement mechanism.149 This was criticized for being an ‘unsatisfactory legal test’ due to the uncertainty embedded in the idea of ‘flexibility’ as many international agreements would include derogations, as well as procedures for reserving conflicts.150 It is significant that these early cases concerned challenges to EU law, whereby the denial of direct effect effectively meant ensuring the validity of the then Community acts.151 It should be noted here, that while the finding that international agreements are capable of direct effect signified an openness to international law, the Court simultaneously took a significant step in the opposite direction by linking the issue of validity of secondary EU law in light of obligations assumed under international agreements to the latter’s direct effect. The Court’s jurisprudence is seen here to be based on an ‘(unspoken) assumption’ that EU law should be presumed to be compatible with international law.152 Despite continuous criticism and judicial challenges against such an extension of the doctrine,153 this position is entrenched with no signs of a reversal in sight. It can, however, be defended on the ground that invalidating EU legislation in actions by individuals who themselves are incapable of relying on the agreement would be ‘a draconian step’ hardly envisaged by the Treaties.154 The linking of legality actions to the direct effect of international agreements is not restricted to challenges by individuals and includes those by Member States which is seen as justified due to the ‘broad construction’ of the principle of direct effect.155 Due to this wider function of direct effect, some have suggested a wider definition for the latter to highlight its function as a measure for legality review.156 In this context, direct effect assumed an additional function to those noted earlier becoming also a means of solving the problem of a ‘collision of norms’.157 Such a ‘collision of norms’ also took place beyond the GATT, where the direct effect of the WTO agreement was similarly ruled out despite the differences in GATT and WTO arrangements. The decisive factors for the rejection of any direct effect of the WTO agreement were the centrality of negotiations, the dispute settlement mechanism which might have an impact on the negotiation position of the EU legislative and executive institutions, as well as the so called reciprocity.158 Reciprocity raises the issue of whether the other parties to the agreement have granted direct effect.159 This particular feature is seen as introducing ‘a clear political element’ to the matter,160 as it allows a determination of the effect of international law vis-à-vis the position of other parties. Even though a stumbling block for direct effect, the Court ignored this argument as far as bilateral agreements are concerned.161 The Court itself acknowledges this inconsistency in Portugal v Council but justifies it by noting that a lack of reciprocity would lead to ‘disuniform application of the WTO rules’.162 Another justification for this dichotomy was proposed by Rosas whereby the lack of recognition of direct effect by other parties was not problematic per se, but the explicit exclusion of direct effect by them can be.163 Indeed, the direct effect of WTO agreement is excluded by most other members.164 Others, however, criticize this reasoning for belonging to the realm of economics than more law.165 As to the WTO dispute settlement mechanism, despite its mandatory nature, the Court linked it to the negotiating freedom of the legislative or executive organs.166 The presence of the dispute settlement mechanism does not necessarily bar direct effect in non-WTO law related case law,167 and at times does not even merit the Court’s attention.168 However, maintaining such distinct approaches has become unsustainable.169 Recently, a number of trade and other agreements include dispute settlement mechanisms which are modelled after the WTO and are capable of ruling on WTO related obligations. The Court, however, will not be required to reconsider its position as the matter has been decided by explicitly precluding direct effect as discussed further below.170 Most significantly, the features of the WTO rules chosen to negate direct effect demonstrate a political concern for the position of the EU and its institutions. The combined reasoning of the Court demonstrates a conscious limitation of its own role in recognition that in certain circumstances the obligation to comply with international agreements is a matter for the legislative and the executive institutions.171 In particular, it is the Council’s role as a legislator (later a co-legislator with the European Parliament) and the Commission’s role as a negotiator that is at stake.172 The interests of the political institutions of the EU are protected not only externally, but also internally. By linking the validity of internal EU legislation to the conditions of direct effect, the Court guarantees the latter’s intactness retaining the prerogatives of the legislative. Besides, secondary EU legislation often represents an, at times painstaking, institutional consensus which the Court would be unwilling to strike down. This is particularly the case when challenges to EU legislation are brought by the Member States: by denying direct effect the Member States are being directed ‘to the place where they are supposed to exert their influence, through the political institutions’.173 The Court’s approach attracted much criticism. Its ‘purposive interpretation’ has been seen as falling short of the principles of good faith interpretation in international law.174 Despite acknowledging the political sensitivity entailed by granting direct effect to WTO law, the Court’s position is also interpreted as lacking openness to international law.175 On the other hand, the denial of direct effect to WTO law does necessarily indicate that the Court’s case law is at odds with ‘the structural principles of the world trade system’.176 Moreover, the alternative could have been counterproductive as not only it would have failed in reforming the WTO law, but it could also have undermined the interests of EU producers facing increased challenges by individuals from other WTO members.177 The harshness of the exclusion of direct effect for WTO law was somewhat mitigated through he Nakajima and Fediol exceptions, which allow for challenges against the legality of EU secondary legislation against WTO law in case of either a clear reference or transposition.178 In addition, the duty of consistent interpretation, the so called indirect effect of international agreements, was developed to oblige the European courts to interpret EU secondary legislation in light of the relevant international agreements.179 The indirect effect is enabled precisely because of the direct applicability of international agreements and their ranking above the secondary legislation as mentioned earlier. Neither the technique of indirect effect, nor the limited application of the exceptions, however, can fully compensate for the exclusion of direct effect. Until recently, the lack of openness characterizing the GATT/WTO line of case law was seen as exceptional, even though it qualified for ‘a substantial part of the empirical material’ on the subject.180 Currently, it is safe to say that WTO law is not exceptional in its rejection of direct effect on the basis of the first conditions as it had been denied to two other—notably multilateral—agreements. Even though in all of these cases direct effect was denied through the application of the first condition, the latter is applied differently in all three cases.181 (ii) Intertanko and lack of direct effect of the UNCLOS Intertanko raised inter alia the issue of the validity of secondary EU legislation in light of the MARPOL Convention (the International Convention for the Prevention of Pollution from Ships) and the UNCLOS.182 While the challenge against the MARPOL Convention failed on a different ground, the UNCLOS was found to lack direct effect precluding the legality review of EU legislation. Even though direct effect is not mentioned in the case, the Court focused on the nature and broad logic of the agreement,183 and it is here that the judgment ‘innovated’ by focusing on the issue of conferral of rights.184 Accordingly, in setting a wide regulatory regime the UNCLOS aims to achieve a balance between the interests of various states and does not grant any individual rights or freedoms. While the conferral of rights has been largely dormant within the case law on WTO and bilateral agreements,185 it is nonetheless not a complete novelty, as the issue of conferral of rights originally featured within International Fruit.186 While some have suggested incorporating the conferral of rights into the analysis of direct effect,187 others view this with caution since this would most probably preclude the direct effect of the majority of international agreements.188 It would also entail the narrowing of the concept of direct effect from its wider understanding as the justiciability of the norm. While some praised the ‘correctness’ of Intertanko,189 others noted the lack of effort on the part of the Court to address any of the aspects of its reasoning which were the stumbling blocks for the direct effect of WTO law.190 Even if one views the WTO law as ‘a case apart’ deserving exceptional treatment, Intertanko unnecessarily relies on the issue of conferral rights injecting further inconsistency in the case law.191 The particularly narrow view of conferral of rights adopted in the case has also be criticized by the proponents for incorporating this factor into the relevant analysis.192 Some, however, see parallels with the case law on WTO law. Apart from the multilateral nature of the agreement,193 it has been suggested that, similar to WTO law, concerns about binding the hands of the EU executive and legislative would have played a part,194 even though the judgment itself avoids such reasoning. Intertanko is also seen as shifting the emphasis to the Member States to ensure compliance with international law when drafting legislation.195 In both Portugal v Council and Intertanko, the Court’s position is seen as cautionary taking stock of ‘the structures and processes established by the agreement, the role played in them by the EU and its Member States, and the need to avoid fragmentation in the presentation of the Union interest in such international regulatory regimes’.196 Furthermore, some have interpreted Intertanko as suggesting that the Court was protecting its own interests to avoid following the rulings of such a powerful international court as the International Tribunal on the Law of the Sea.197 If such considerations indeed played a part in the Court’s reasoning, then they are masked by its reliance on the conferral of rights. By denying direct effect through the latter as part of the first condition results in a ‘general immunisation of EU norms from review vis-à-vis UNCLOS’,198 which would also include any subsequent norms. (iii) ATTA and the Kyoto Protocol The third instance of denying direct effect on the basis of the first conditions in Air Transport Association of America and Others (ATAA) concerned the issue of the validity of EU Directive 2008/101/EC including aviation in the EU emission trading scheme in light of the Chicago Convention, the Kyoto Protocol, and the Open Skies Agreement between the European Union and the United States of America.199 For the purpose of this discussion it is the findings on the Kyoto Protocol to the United Nations Framework Convention on Climate Change that are most relevant. Even though the case is not viewed as ‘groundbreaking’ in setting any new rules,200 it nevertheless introduced certain new features as far as the first condition of direct effect is concerned. Instead of relying on its previous case law on multilateral treaties to draw on the factors decisive in the ruling out of direct effect, here the Court focuses on a different issue—that of the flexibility of the implementation in the obligations of the parties.201 The flexibility relates to the manner and the speed of fulfilling the relevant obligations depending on the parties’ agreement. In addition to such novel element, another peculiarity in ATAA relates to the way the Court appears to incorporate the analysis of the second condition into the first one.202 In support of its reasoning on the flexibility available to the parties in terms of the manner and speed of meeting their obligations, the Court cites Article 2(2) of the Kyoto Protocol which ‘cannot in any event be considered to be unconditional and sufficiently precise so as to confer on individuals the right to rely on it in legal proceedings’ contesting internal EU law.203 While the emphasis on conferral of rights appears to be in accord with Intertanko, it nonetheless appears to blur the lines between the two conditions. The case, therefore, does not in any way help to resolve the confusion as to whether direct effect requires conferral of individual rights.204 It can be argued that what the case has in common with WTO law is deference to the political institutions of the EU which will be in charge of implementing the provisions of the Kyoto Protocol, including via the International Civil Aviation Organization (ICAO). Another possible explanation for the judgment is that the Court attempted, in this way, to protect a more developed internal EU regulation in comparison with international norms.205 While, on the one hand, a clear inconsistency emerges in the application of the first condition of direct effect, the position described above demonstrates the flexibility injected into its application. Although this might be justified in view of the necessity to make ‘various adjustments in the light of factors subject to constant evolution’,206 the lack of convincing reasoning as to the introduction of new factors undermines the coherence and the continuity of the case law.207 The flexibility in the application of conditions of direct effect is not solely restricted to the first condition. (iv) The flexibility embedded in the application of the second condition and the reversal of conditions The application of the second condition on clarity and unconditionality of specific provisions is also characterized by a high level of flexibility and often leniency,208 as well as a certain subjective elements present in its assessment.209 The leniency can manifest itself in various forms. The Court can declare a provision in an agreement to be unconditional and requiring no further implementation even though the latter is subject to recommendations by an institution established under the agreement.210 The Court’s lenient approach at times also surfaces in a scarce and fleeting analysis.211 Leniency is also apparent where the Court supports its positive findings on the second condition with past precedents on other types of agreements without having analysed the nature and logic of the agreement first.212 Furthermore, as noted above, there is also a certain flexibility in terms of the choice of which condition to apply first, which will potentially impact the nature of the exercise. In many cases, the features of the specific provision are analysed first whereas the first condition on the nature and the purpose of the agreement is given scarce attention post factum to determine that it does not preclude direct effect.213 This approach has been criticized as being potentially counter-productive, bacause ruling on the specific provision prior to establishing the purpose of the agreement would be useless if the purpose of the agreement were not such as to allow for direct effect.214 On the other hand, the reversal of the conditions’ order might suggest that the Court is predisposed to granting direct effect. Indeed, cases of this kind characterize the Court’s approach at its most permeable with a high rate of success.215 In some cases, where a specific type of agreement is concerned, this approach might not be as controversial. For instance, due to the extensive case law on association agreements, it is accepted that their nature is such as to afford direct effect. Excluding direct effect for such agreements would be equal to ‘ignoring [their] raison d’être as the foundation for gradual integration between parties’.216 However, the practice of the reversal of the conditions is not restricted to the types of agreements which were perviously found to be directly effective. A useful example in terms of the flexibility embedded in the application of the second condition as well as the order of the conditions is Simutenkov involving a non-discrimination clause in the EU–Russia Partnership and Cooperation Agreement (PCA).217 According to Simutenkov, as long as a ‘clear and precise obligation’ is found which is not subject to any implementing measures, then the provision can be directly effective if the nature of the agreement does not preclude it.218 Jacobs notes that the essential issue here was whether the agreement includes provisions which can ‘directly govern the position of individuals’ in which case the analysis should turn to the specific features of the relevant provision.219 In this instance, it appears to be manifested through the imposition of an obligation despite the lack of clarity in the phrasing of the provision.220 The reversal of the order of the conditions, and the limited emphasis on the first condition meant the latter played only ‘a marginal role’ in the Court’s findings, which is strikingly at odds with the case law on WTO law.221 In addition to the reversal of the order of the conditions, Simutenkov demonstrates the flexibility the Court injects into the application of the second condition. The principle of non-discrimination was to be implemented on the basis of the recommendations of the PCA Cooperation Council. In earlier Demirel, the Court ruled out direct effect for a provision of the EEC–Turkey Association Agreement due to the measures to be adopted by the Association Council.222 In a departure from Demirel, the Court found that the Cooperation Council recommendations were intended solely to ‘facilitate’ the compliance with the principle of non-discrimination, rather than limit the immediate application of the prohibition on discrimination.223 What is remarkable is the Court’s reliance on case law related to the EC–Poland Association Agreement where a positive finding on direct effect was made in relation to the principle of non-discrimination.224 This is significant because the Court at that stage did not consider the nature and the purpose of the agreement, and therefore made no distinction between the two agreements. Only after finding that the provision is clear, unconditional, and does not require further measures, did the Court move to the first condition. Despite the PCA falling short of promising association, offering less advantageous cooperation and less ambitious objectives, the Court confirmed that the lack of close links similar to association does not preclude direct effect.225 Furthermore, in a circular motion, the Court relied on the agreement’s general ability to govern the position of individuals.226 Thus, the apparent differences in the status of the partners and their different prospects in forging close relations with the EU did not have much impact on the CJEU’s findings.227 Indeed, the early case law on other types of agreements had already demonstrated that looser integration prospects or weaker links with the EU did not necessarily have a bearing on the finding of direct effect.228 Even in bilateral agreements, merely providing for cooperation,229 the Court undertook a similarly structured analysis focusing on the second condition.230 On the one hand, this would suggest the lesser significance accorded to the first condition, whereby the gatekeeping takes place predominantly through the application of the second condition and is of a less intense character. The reversal of the conditions is not restricted to bilateral agreements only. Indeed, there is a suggestion that if the Court commenced its analysis with the second condition in the WTO line of cases, the outcome might have been different.231 Nonetheless, as discussed above, other fundamental considerations played their part in the rejection of the direct effect of the WTO law. In addition, the preference for which condition to start with is itself indicative of the nature of the gatekeeping exercise the Court is about to undertake. The Court exercises an assessment as to whether the agreement in question might require a legislative and executive intervention: ‘the [Court] operates a sort of ex post control … closely related to the mechanism of automatic incorporation of international treaty into the EU legal system’.232 Following this approach in the case of a multilateral agreement will not necessarily lead to a positive outcome. In Brown Bear, the Court commenced its analysis of the direct effect of the Aarhus Convention with the second condition but eventually rejected direct effect.233 By deciding initially that the relevant provision was subject to further implementing measures, the Court dismissed the need for evaluating the nature and the purpose of the agreement. This can be seen as an open-minded form of gatekeeping, leaving open the possibility of finding other provisions of the Aarhus Convention directly effective. While this approach might raise the issue of the continuous relevance of the first condition, the judicial choice as to which condition to consider first will depend on the nature of the agreement.234 As seen in Intertanko and ATAA, the Court can opts to commence its analysis with the first condition depending on the agreement in question. Thus, the first condition is still part of the legal test, albeit a very flexible one. To sum up the setting and the application of the conditions of direct effect, this stage of gatekeeping demonstrates that the Court has not bound itself to being either open or closed, rather it has created a wide margin of discretion for ad hoc solutions. The Court’s flexible application of the conditions of direct effect demonstrates that the EU legal order can be successfully closed to international agreements which are capable of impacting the interests of the EU as a political organization. Such interests would pertain to the external and internal roles of EU institutions, as well as the intactness of secondary EU law. The more permissive approach towards bilateral agreements, however, does not necessarily mean ultimate openness to their ability to have an impact on individuals. This openness can be rebutted by the application of the second condition, where a negative answer leaves the international agreement toothless as far as the individual is concerned. In addition, even a positive finding in favour of the second condition does not guarantee a change in the legal position of the individual. This is due to the Court’s interpretative freedom in relation to the scope of the provision considered next. C. The interpretation of directly effective provisions as the final stage of gatekeeping Having found a provision to be directly effective, the Court’s gatekeeping culminates with the interpretation of the scope of a particular provision,235 save for cases where the Court directly turns to interpretation without establishing direct effect first.236 At this final stage, the Court ultimately secures or denies the impact of an international agreement on a given individual’s legal position as the finding of direct effect is fruitless unless the relevant provision receives a ‘favourable interpretation’ by the Court.237 The gatekeeping at this stage also ranges from rather sparing to strict attitudes. The Court relies on international law, in particular Article 31 of the Vienna Convention, to interpret the provisions of the agreement ‘in their context and in the light of its object and purpose’.238 One of the main issues pertinent to the gatekeeping through interpretation is whether the provisions of international agreements resembling those found in the TFEU (previously EC Treaty) should be granted a similar interpretation. Despite establishing that provisions resembling those found in the EC Treaty should not necessarily be accorded with the same meaning,239 the Court, nonetheless, accorded similar interpretations to international agreements in a range of cases.240 The comparative analysis of the context of the TFEU and other agreements had ‘considerable importance’ in the Court’s findings.241 On the one hand, comparing international agreements to the EC Treaty/TFEU might appear to be devoid of controversy as it merely represents the outcome of considering the objectives of each Treaty.242 On the other hand, Article 31 of the Vienna Convention ‘itself does not permit … comparative analysis’ between different agreements, and it is the context and the object and purpose of each treaty that should be separately analysed.243 It can be argued, that the comparison with the TFEU can indeed cast a shadow over the Court’s perception of the international agreement. The most straightforward transposition of internal interpretations to provisions of international agreements were found in instances of direct references to the EC Treaty. For instance, the Yaoundé Convention, in its Article 2, directly referenced then Article 13 EC on the abolition of charges having equivalent effect leading to a uniform interpretation.244 Opting for a similar interpretation entails an ‘extension’ of the EU legal order which can be justified for agreements with an element of EU acquis transposition.245 However, having an element of EU acquis transposition itself does not guarantee similar interpretation. Neither does a promise of association always secure a homogenous interpretation. It is, therefore, difficult to deduce a clear and consistent pattern in the Court’s case law.246 Some suggest that a successful outcome would often depend on the ‘amenability’ of the ‘scope, historical and legal context’ of the relevant agreement, while the unsuccessful outcome is tied to the provisions themselves, commonly on the rights of third country nationals to reside on the territory of the EU.247 For the seeming inconsistency between the cases with transposition of internal interpretation and cases denying the latter, a justification is found in the context of each agreement.248 The focus of the interpretation, however, can hover from the context of the agreement to the provision itself, or the Court could put more weight on one than the other. Lack of consistency is seen most acutely in relation to association agreements with candidate countries: in some cases, the accession prospects were factors favouring uniform interpretation, while in others, the accession factor was not given sufficient weight to influence the outcome of the interpretation.249 In the latter category of cases, the directly effective provisions on freedom of establishment and free movement of workers in certain Europe Agreements were not accorded a uniform interpretation with the the equivalent TFEU provisions, as a result of which the Member States could impose their own rules on entry, stay, and establishment, including their immigration rules.250 Here, the Court chose to limit ‘its own creative reading of the association’ by granting less significance to the political and historical context of the relevant agreements and the parties’ accession intentions.251 Some viewed the Court’s approach as merely respecting the differences between the Europe Agreements and the EC Treaty and avoiding a potential backlash against the accession strategy.252 This in itself manifests the making of a political choice. It is expressed in particular in the refusal to advance the enlargement agenda and in the tribute to the Member States’ concerns about migration.253 Ultimately, the finding of direct effect here turned into ‘a hollow victory’.254 The comparative approach is not restricted to the TFEU. In El-Yassini, the EC–Morocco Agreement was compared to the EC–Turkey Association Agreement: since the former did not provide for a prospect of association, the scope of its non-discrimination provision was narrowly interpreted.255 Such comparison might be appropriate for agreements which were concluded as part of a regional approach, as seen in Kolpak, where the Europe Agreement with Poland was compared to the Europe Agreement with Slovakia in order to draw out their similarities ‘to their objectives or the context in which they were adopted’.256 There was, however, no such regional context in Simutenkov, where the Court transferred its interpretation of the non-discrimination provision from the Europe Agreement with Poland to the EU–Russia PCA without unpicking the distinctions between the context of the two agreements,257 thus diluting the ‘differentiation’ the conclusion of distinct agreements was meant to signify.258 Moreover, despite the lack of comparison between the PCA and the EC Treaty,259 the Court also relied on Bosman to transpose its internal interpretation of the principle of non-discrimination to the PCA, without clearly demarcating the limits of such transposition.260 Ultimately, the Court is not an unequivocal gate-opener: despite the far-reaching implications of direct applicability and direct effect, the application of the conditions of direct effect provides a wide margin for limiting the permeability of EU law towards international agreements. Each case of denying direct effect or interpreting the scope of a directly effective provision to exclude a particular right is an example of shutting the proverbial gates. Direct effect is used, in particular, as a ‘shield’ to protect EU institutions and legislation.261 Indeed, the application of direct effect to international agreements has at times been assessed as ‘less generous’262 or ‘more reserved’263 in comparison with internal EU legal acts. The Court’s generosity or the lack thereof has been linked to the nature of the action whereby challenges against Member States’s actions have a higher likelihood of success than those against the EU.264 As far as individual reliance on provisions of international agreements is concerned, it has been noted that the successful cases even though constituting ‘a broad category’, inclusive of association agreements and trade agreements, are nonetheless the exception.265 In any case, it appears that other EU institutions do perceive the Court’s approach to be generous as they are keen to challenge the status quo. V. Sidelining the court as the gatekeeper?: Alternative gatekeeping techniques As noted earlier, the Court carved out its role on determining the effects of international agreements in the legal vacuum created by the lack of an agreement between the parties: only if the effect of the provisions is not settled by the parties would it ‘fall’ to the Court’s jurisdiction, as set out in Kupferberg.266 If the parties have come to a consensus on the effects of the agreement, the Court will be required to give it full effect by enforcing the intentions of the parties. At this stage the inquiry is within the domain of international law.267 If such a consensus is absent, the inquiry shifts to the level of European law as discussed above. The main question to ask here is whether the treaty-making institutions can pre-empt a judicial inquiry into direct effect. Both scenarios are considered in turn below. A. The agreement between the parties The agreement between the parties is viewed at times as one of the conditions for direct effect.268 Some qualifications are in order here. If the agreement is non-existent, then there can be no talk of it being a condition. If there is an agreement, then it can take two forms—positive or negative, both creating implications for the Court’s role. If the agreement positively sets out direct effect, the Court is precluded from finding otherwise. Alternatively, if the agreement excludes direct effect, the Court cannot find to the contrary. Cheyne refers to this as ‘the pre-emptive rights of the executive institutions to determine whether [the agreement’s] provisions should be given [direct effect]’.269 As noted earlier, most commonly the issue of effects does not occupy an important role in international negotiations. Furthermore, in some cases open opposition to any such prescriptions has been recorded. For instance any such possibility in relation to WTO law was firmly rejected by its members.270 Positive prescriptions of direct effect are extremely rare in practice. An example of such practice can be found in the agreement establishing the European Common Aviation Area which, in its Article 15, obliges its parties to ‘ensure that the rights which devolve from this agreement may be invoked before national courts’.271 This positive setting of direct effect can be linked, perhaps, to the purpose of the agreement aiming to create a single aviation market between the EU and certain European states, non-members of the EU. Providing for direct effect in the agreement itself enables challenges by individual travellers and members of the aviation industry which could speed up the removal of barriers to movement. As for the negative exclusion of direct effect, in departure from the established view that the effects of the agreements are not usually part of the negotiations, examples of negative preclusion can be found. Until recently, they were rare. An isolated example can be found in the 1990s in the Agreement on international humane trappings standards between the EU, Canada, and Russia.272 In particular, it established that ‘[the] Agreement is not self-executing’ and requires implementation by each party.273 It is not clear whether the term ‘self-executing’ is used here to refer to direct applicability or direct effect, however, the reference to the need for implementing measures ensures that the conditions for direct effect will not be satisfied.274 More recently, a trend of a more systematic exclusion of direct effect appears to be emerging. This negative preclusion can take two forms either by denying direct effect to the entire agreement or to its specific provisions. The first form is more extreme, examples of which, with a variety of formulations, can be found in recent trade agreements with Colombia and Peru,275 the somewhat ill-fated free trade agreement with Singapore,276 an association agreement with Central America,277 and the Comprehensive Economic and Trade Agreement with Canada (CETA).278 None of these agreements has an ‘integrationist potential’.279 Nonetheless, as such agreements have been found to be directly effective in the past,280 the effect of such provisions is to remove any possibility of making such a finding by the Court of Justice. Excluding direct effect for an entire agreement would by implication extend also to the decision of the bodies established thereunder. It can be argued that in bilateral agreements denying direct effect in the agreement itself is more warranted by the EU, in particular its treaty making institutions and its Member States, since in the case of the other party its ordinary constitutional arrangements would apply in any case. The same conclusion, however, does not stand for multilateral agreements: the other parties might also be interested in restricting the obligations to intergovernmental level without creating judicially enforceable rights for private parties. In terms of the variations in formulations, it should be noted that the agreement with Colombia and Peru, as well as the EU–Singapore trade agreement and CETA clarify that no rights or obligations are created for individuals beyond those created between parties under international law. This secures an outcome whereby the Court of Justice can no longer rule on the possible direct effect of such agreements. However, this does not mean that the Court’s interpretative function is surrendered. Presumably, the Court can still rely on indirect effect to interpret any relevant legislation in light of the international commitments. Unlike the agreements noted above, the Association Agreement with the Central American countries goes further to bar the justiciability of its provisions by private parties ‘unless otherwise provided in that party’s domestic legislation’. Such a formulation, besides sidelining the Court, ensures that control over the matter is firmly with the legislative institutions of the EU should they opt for conferring certain rights to individuals. Examples of the second type of exclusionary practice limited to specific provisions are found in the Association Agreements with Ukraine, Georgia, and Moldova respectively.281 The relevant provisions relate to the dispute settlement mechanisms, as well as the schedules of commitments in service laid down in annexes to the agreement. The exclusion of direct effect here is linked to WTO law-related content: this includes the dispute settlement mechanism based on the WTO model (with some adjustments) with a jurisdiction to rule on WTO-related obligations, and the schedules for commitments in service areas based on The General Agreement on Trade in Services (GATS).282 Precluding direct effect for these provisions is due to substantive incorporation of WTO law guaranteeing that the lack of direct effect of WTO law cannot be bypassed.283 As such, this practice does not aim to sideline the Court. On the contrary, it can be suggested that it confirms the Court’s jurisprudence. At the same time, while the EU and the relevant third country have agreed to exclude the direct effect of specific provisions, by implication this would entail that other provisions of the agreement—those meeting the relevant criteria—are capable of having direct effect. As noted earlier, at this stage a transition is made from the domain of international law to EU law where the Court steps in with its ‘monopolized’ role to determine the effects of the other provisions. But what if the Council decision adopting such agreements explicitly excludes the direct effect for the entire agreement? B. Internal institutional challenges to judicial gatekeeping Over a decade ago, doubts have been expressed over the tenability of the status quo predicting a certain tension over the Court’s power, particularly on the part of the legislature.284 The current challenges against the Court’s lead role is manifested in the Council’s practice precluding direct effect in its decisions on the conclusion of certain international treaties. Such decisions have been viewed as one ‘form’ or ‘way’ of excluding direct effect.285 Examples include the decisions concluding the Association Agreements between the EU and Ukraine, Georgia, and Moldova respectively which rule out direct effect for the entire agreement.286 This section aims to enquire into whether such decisions can indeed be seen as a means of excluding direct effect by aborting the Court’s jurisdiction. As noted earlier, international law is characterized by it permissiveness to the domestic legal effects of international agreements; it is the prerogative of each state to fulfil its obligations according to its interests.287 This implies that a state, or in this case the EU, would establish its own rules on the implementation and the effects of the agreement. The Council decisions concluding the agreement cannot be equated with constitutional rules, however. Neither are they measures implementing the agreement.288 What, then, is their effect? Can an argument be made under EU law or international law for giving a decisive weight to the Council’s decision? (i) The Council decisions in view of the Kupferberg formula? While the Kupferberg formula cited above made no mention of the Council’s decision adopting an international treaty, on one occasion the matter did come to the Court’s attention. In 1994, the Council’s Decision adopting the WTO Agreement included the following statement in its preamble: the agreement is ‘not susceptible to being directly invoked in [EU] or Member States’.289 Initial assessments viewed it as indecisive in setting the effects of the agreement on individuals or EU institutions and Member States, which was a matter for the Court to decide.290 There were also predictions as to the decision’s potential to lead to a more cautious approach by the Court.291 The issue soon came to the Court’s attention in Portugal v Council.292 Advocate General Saggio in his opinion was unequivocal against the Council’s decision having an effect on the Court’s competence to rule on direct effect: it was merely a ‘policy statement’.293 His argument was twofold. The first stems from international law. Invoking Articles 31–33 on the rules of interpretation of the 1969 Vienna Convention on the Law of Treaties, the Advocate General ruled against a unilateral institutional declaration being seen as decisive at the level of international law in terms of limiting the direct effect of the agreements.294 Rather than according primacy to such declaration, the issue of direct effect should be resolved through the interpretation of ‘the objective content of the textual provisions of the agreement’ in accordance with the rules of customary international law,295 including Article 31 of the Vienna Convention binding the EU.296 The latter sets the general rule of interpretation according to which treaties should be interpreted ‘in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. The second argument derives from the EU legal order. The Council cannot by secondary legislation limit the Court’s (or national courts’) competence to establish the effects of international agreements: ‘a unilateral interpretation of the agreement made in the context of an internal adoption procedure cannot—outside the system of reservations—limit the effects of the agreement itself’.297 The Court in turn acknowledged that it was the parties’ prerogative to establish the means of implementation of the agreement, however in the absence of such an accord the parties are at liberty to choose the means of implementation.298 Instead of clarifying the status of the Council’s decision, the Court then proceeded with setting its arguments for denying direct effect to the WTO agreement and, only having established the latter, did it refer to the Council’s decision noting that its own finding ‘corresponded’ to the decision.299 First, it is clear that the Court finds the decision of relevance.300 Secondly, the Court only refers to it post factum, and not as a starting premise for its findings. Thirdly, there are no suggestions that the Court views the Decision as a measure implementing the agreement. In fact, the CJEU does not specify the capacity in which the Council’s decision was taken into account. The Court’s approach has been interpreted differently. According to Fabri, the Council’s decision did not play a significant part in the outcome of the case: the negative preferences of other institutions cannot bind the Court in a manner that primary law would.301 Others saw the decision as having impacted the Court’s approach as evidence of the intention of the EU as a party.302 Can a negative pronouncement in a Council decision be equated with the expression of the intention of the EU as a party, however? It is safe to say that the preamble of the decision demonstrates the views of both the Council and the Commission,303 ‘the treaty-making institutions’.304 It might be suggested that in dualist countries the requirement to implement the agreement eliminates the need for any such proclamations. In monist countries, the stance of the executive can play a part in the judicial finding of direct effect. Paying homage to the views of the executive, in deciding on the matter of direct effect, forms part of the judicial inquiry into the intentions of parties in the Netherlands, for instance.305 The Council decisions can also be compared to the declarations by the US President and two-thirds of the Senate on the non-self-executing nature of an international agreement. This practice has been criticized for its ‘neo-isolationist preferences’ depriving the judiciary of the opportunity of ‘contributing to the evolution of international practice’.306 Not only the status of such declarations in the US legal order is uncertain, both in courts and in scholarship,307 the comparison itself is somewhat problematic. First, the US declarations as to non-self-execution are about the need for further implementing measures in the US domestic legal order: individuals can derive rights stemming from the international treaty only if the Congress passes a legislative measure.308 In contradistinction, the Council’s decision merely excludes direct effect which does not necessarily imply that an implementing measure is required. Another consideration should also be born in mind when embarking upon such comparison. The US President and the Senate are the actors behind the treaty ratification: the US President signs it after the consent of two-thirds of the Senate. The other legislative chamber of the Congress, the House of Representatives, does not take part in this process. In the context of the EU, a principled objection can emerge towards treating the Council’s decision as reflective of the EU’s intention as a party for the reason of the involvement of the European Parliament. The latter, as a co-legislator with the Council, has seen an expansion of its involvement in the process of conclusion of international agreements to the extent of influencing their content or even rejecting them.309 As a party, the EU concludes agreements through the process established in Article 218 TFEU requiring the consent or consultation of the European Parliament.310 When an agreement within its provisions excludes direct effect for its entirety or for few provisions, the Parliament’s participation would thus make the intention of the EU as a party complete. Would a similar conclusion be applicable in relation to those cases where direct effect is excluded in a Council decision only? What is the evidence to suggest that the Parliament would support such restrictive practices? Support of this nature could have been found in case law where the Parliament became involved to argue against direct effect. However, only a handful of such examples can be found,311 and the case law offers no systematic conclusion on the Parliament’s position on the issue of direct effect more generally. On the other hand, a compelling argument can be made to suggest that the Parliament, in effect, acquiesces in this type of exclusionary practice. The preambles of parliamentary resolutions expressing consent to the conclusion of international agreements, refer first and foremost to the respective draft Council decisions. The Parliament, therefore, tacitly endorses the preferences of the Commission and the Council. This conformism, however, might be rooted in the Parliament’s own disapproval of the openness to international law expressed on occasion. In its 1997 resolution on the relationship between international law, Union law and the constitutional law of the Member States, the Parliament called for a provision in the EC Treaty setting out the process of the transposition of international law into the EC legal order.312 The resolution, thus, focused on the issue of direct applicability. Even though the Parliament appears not to have challenged the Court’s leading role, it nevertheless is discontent with the Court’s ‘solutions’: EU law is more ‘permeable’ to international law than domestic legal orders of the Member States.313 The Parliament evidently favoured a non-automatic transposition of agreements which should be directly applicable only if it ‘has been declared applicable by an internal legal act of the [EU] or after its substance has been transposed into [EU] legislation’.314 Even though these objections were aimed at direct applicability, clearly the Parliament would be interested in moulding the provisions which could potentially be invoked by individuals. This can be linked to the ‘democratic’ argument where such issues should be decided by representative institutions instead of leaving them to the fate of sporadic developments through individual claims.315 The Council decision, however, does not suggest any meaningful return of control over this issue of the conferral of rights to the Parliament as it does not necessarily imply any further legislative measures. This can be juxtaposed with the US practice, where the relevant declaration passes control to Congress.316 In the EU, if the Council decision is to be upheld on its effect of barring direct effect, it should rather be viewed as returning control to Member States. If one is to view the Council’s decision as representative of the EU’s intentions as a party, it does not necessarily lead to an argument that it should prevent the Court from ruling on the matter. As prescribed by the Kupferberg formula, the Court’s jurisdiction to determine the effects of the agreement is secondary only to the agreement between the parties. The Court made no such reservations for the acts of the Council concluding the agreement, neither did it consider it necessary to give a clear weight to the Council decision in Portugal v Council, as noted above. Any suggestions of treating the Council’s decision as definitive on the matter of direct effect will require the Court to revisit its settled jurisprudence and to make a new constitutional argument to determine its own role and potential limitations to it. Portugal v Council, however, would suggest that by mentioning the Council decision within its analysis of the first condition of direct effect the Court indeed viewed it as relevant for establishing the intentions of the parties. Crucially, it was merely used as one of the factors relevant for establishing the intentions of the parties. It can also be questioned whether any further weight could be granted to the relevant decisions under the provisions of the Vienna Convention on the Law of Treaties. (ii) Council decisions as an interpretative tool under the Vienna Convention on the Law of Treaties? At the outset, it should be noted that Council decisions excluding direct effect would not qualify for a so called ‘reservation’ under the Vienna Convention on the Law of Treaties.317 The latter defines reservations as ‘a unilateral statement, however phrased or named, made by a State, when signing, ratifying, accepting, approving, or acceding to a treaty, whereby it purports to exclude or to modify the legal effect of certain provisions of the treaty in their application to that State’.318 The stipulations on preclusion of direct effect in the Council decisions do not qualify for this definition as they are not aimed at excluding or otherwise modifying the legal effect of a particular agreement. Rather, they make a comprehensive pronouncement on the entire agreement in relation to domestic legal effects. So, what capacity can be accorded to such decisions? According to Verwey, even though non-decisive per se, the Council’s decision can be taken as an ‘additional source’ for the purposes of interpreting a provision in the agreement.319 But what type of ‘additional source’ would it be? Returning to Article 31 of the 1969 Vienna Convention, this allows the context of the agreement to be taken into account when interpreting the treaty. According to Article 31(2)(b), the context, in addition to the text, the preamble, and annexes, can include inter alia ‘any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty’. While the Vienna Convention itself does not clarify what the ‘conclusion’ of the agreement precisely refers to, it is accepted that it would include instruments ratifying or expressing the parties’ ‘consent to be bound’.320 The Council’s decision is indeed an instrument expressing the EU’s will to be bound by a particular agreement. That leaves the issue of ‘acceptance’. It is suggested that ‘the need for acceptance’ would distinguish the instrument envisaged in the provision above from ‘unilateral interpretative declarations made by a state when signing or ratifying’ the agreement.321 The acceptance by the other party is a condition,322 requiring the opposite parties at least to have ‘acquiesced in the instrument’.323 Even though the acceptance can be ‘informal’ or ‘tacit’, any party suggesting a particular interpretation on the basis of such extrinsic instrument will be required to demonstrate that other parties have accepted the position declared in the instrument.324 Verwey suggests that an explicit acceptance or rejection of the Council’s statements by the other party is not necessarily referring to Bresciani, where reciprocity, that is reciprocal recognition of direct effect, had no bearing on the Court’s finding of direct effect.325 However, these are two distinct matters. In assessing reciprocity as a factor indicative of the nature of the agreement, the Court is not affected by denial or recognition of direct effect by other parties (with the exception of the WTO case law). But when evaluating whether the Council’s decision precluding direct effect for an entire agreement is part of the context of the agreement under Article 31 of the Vienna Convention, the Court should require some form of acceptance of this exclusionary effect of the Council’s decision. If the Court is to view the Council’s decision as ‘any instrument’ comprising the context of the agreement under Article 31(2), there should be some evidence of some form of acceptance by other parties. Furthermore, there is no hierarchy among the sources used for interpretation: the order ‘appears to be that of logic, proceeding from the intrinsic to the extrinsic, from the immediate to the remote’.326 ‘Any instrument’ referred to above is ‘extrinsic to the treaty’,327 suggesting that it cannot per se set aside the text of the Treaty which may well indicate in favour of direct effect. Ultimately, under Article 31(1) of the Vienna Convention the interpretation of the agreement is undertaken on the basis of textual and teleological approaches in addition to the subjective quest for the intentions of the parties.328 (iii) Taking stock The discussion above is not a matter of historical debate, but it is pertinent due to a number of Council decisions excluding direct effect in relation to a range of agreements, including free trade agreements and association agreements with neighbouring countries, as noted above. For those agreements which incorporate WTO law to a certain extent, precluding direct effect in the Council decision can similarly be interpreted as an insurance policy for not bypassing the established practice of excluding direct effect for WTO law. This would not appear to create any issues for those agreements which themselves preclude direct effect in their entirety. However, where direct effect is precluded for specific provisions only, or which contain no provisions about direct effect, the Council’s decision can be viewed as problematic. As discussed earlier, the decisions concluding the Association Agreements between the EU and Ukraine, Georgia, and Moldova, respectively, rule out direct effect for the entire agreement, while the agreements themselves rule out direct effect for specific provisions only. Unlike the Council Decision on the WTO Agreement where the relevant limitation was included in the preamble, in the above cases it is found among the main provisions of the decision. This might perhaps lead to a suggestion that a more significant weight should be attached to the latter as opposed to a preambular statement.329 On the other hand, in his forceful position in Portugal v Council, Advocate General Saggio referred to the inability of the secondary legislation to exclude the Court’s jurisdiction generally, without making a distinction between the relevant pronouncement being made in a preamble or in the text of the act. Furthermore, irrespective of the location of such limitations, the decisions are problematic for the following reasons. First, as noted above, these agreements themselves preclude direct effect for specific provisions implying that other provisions are capable of direct effect. In this context, it can even be suggested that the Council decisions are against the intentions of ‘the parties’. Secondly, such unilateral declarations by the Council, outside of WTO law, are at odds with past jurisprudence, especially if the agreement contains provisions identical to those which were found to be directly effective in the past in cases of comparable bilateral agreements.330 Van der Loo and others note in this connection Article 17 in the EU–Ukraine Association Agreement on the principle of non-discrimination in relation to workers.331 Similar provisions in Association Agreements with EU neighbouring countries in the south were found to be capable of direct effect in the past.332 In addition, this provision is identical to the non-discrimination provision in the EU–Russia PCA in Simutenkov.333 As discussed earlier, the Court in its analysis of the condition on the nature and the logic of the agreement embraces a comparative approach. The PCA with Russia offers much narrower integration prospects vis-à-vis the Association Agreement with Ukraine.334 The Ukrainian agreement is ‘the most advanced agreement of its kind ever negotiated by the [EU]’,335 offering the closest possible links to the EU falling short of membership, which has also served as a template for the respective agreements with Georgia and Moldova. In this light, the Council’s decision indeed appears to signal a clear preference for a departure from previous case law. It might be suggested that it is perhaps due to the integrationist agenda of such agreements that the Council felt compelled to restrict their direct effect. Ukraine, Georgia, and Moldova entertain European aspirations and view these agreements as a stepping stone on their future membership path. By denying direct effect to these Agreements, the Council sends a clear signal that they remain at an intergovernmental level falling short of creating a union of people. But this cannot be the sole logic driving the Council to exclude direct effect for EU bilateral agreements. A similar decision precluding direct effect was adopted for the EU–Korea free trade agreement.336 The latter cannot be compared to the Association Agreements noted above and contains no similar integrationist agenda. It might, therefore, indicate a general trend of restricting direct effect for agreements which were found to be directly effective in the past. The EU–Korea agreement, for instance, contains standstill provisions which could be directly effective as seen in earlier jurisprudence.337 The Council decisions, therefore, signal a manifest preference for a departure from the Court’s practice. It might only be a matter of time before the CJEU is called to clarify the status of such pronouncements by the Council. In particular, a clarification will need to be made whether excluding direct effect in a provision of the decision gives the latter more legal weight than a preambular statement, as in the WTO Decision, so as to exclude the Court’s jurisdiction. Alternatively, if the Court maintains the Kupferberg formula, it should interpret international treaties by giving ordinary meaning to their terms which might indicate in favour of direct effect according to the conditions set by the Court. If the Council is to insist that its decision should be viewed as part of the context of the agreement, then some form of ‘acceptance’ on behalf of other parties should be demonstrated. VI. Conclusion The Court of Justice has meticulously carved outs its role of gatekeeper, which is not one of an unequivocal ‘door opener’. The Haegeman ruling establishing the direct applicability of international treaties signalled an almost automatic openness to international law. It laid the foundations for the next opening act which is the general finding that intentional agreements are capable of having direct effect. Paradoxically, these developments were affected more by internal than external considerations. Apart from affirming the crucial role of the Court of Justice in EU international relations vis-à-vis other institutions, they also propelled the EU’s external actorness against the Member States. The Court, having placed itself at the core of the direct effect exercise, became a flexible gatekeeper. The manner in which the conditions of direct effect have been set and applied allowed for ample scope for both generous and ungenerous findings. Ridden with inconsistency, the setting and the application of the conditions of direct effect permitted the Court to limit the extent to which international agreements could empower individuals or affect the legality of secondary EU legislation, in particular when the external or internal interests of the EU institutions were at stake. The CJEU’s flexibility has been carried forward to the last phase of gatekeeping through the interpretation of the provisions of the agreement ultimately determining whether the individual would find the relief sought. Here, additional concerns—those of Member States—can also be taken into account. The reign of the Court is being challenged though. While challenges to direct applicability expressed by the European Parliament in its 1997 resolution did not lead to any consequences, the Council and the Commission are willing to mitigate the openness created by direct applicability through limitations to direct effect. The dual practice of excluding direct effect either for the entire agreement or for specific provisions reveal the ease with which the Court can be sidelined. The situation is less clear with the Council’s attempt to exclude direct effect for entire agreements in its decisions on their adoption. Despite mentioning the exclusionary preambular pronouncement in the Council’s 1994 decision in Portugal v Council in support of its rejection of the direct effect of the WTO agreement, the Court failed to clarify the weight it attached to the latter. In the recent decisions, the Council has made a stronger statement by relocating the exclusion of direct effect to the text of the decision from the preamble. If the issue of direct effect of any of these agreements is ever raised, the Court will be called upon to revisit its gatekeeper role and the possible concessions to it. The Court will have to clarify whether the decisions can be viewed as expressing the intention of the EU as a party, and, if so, whether it would be capable of setting the Court’s jurisdiction aside. Unless the Kupferberg formula is revised, Portugal v Council suggests that the Court would interpret the international agreement, ie would preserve its jurisdiction over the matter of direct effect, but would, nonetheless, take into account the Council’s decision as part of its analysis of the first condition of direct effect. The status of the pronouncements by the Council can also be clarified under international law. It is only a matter of time before this issue is raised in front of the Court. I would like to thank Jeffrey Kenner, Dominic McGoldrick, Daria Davitti, Marko Milanovic, Mary Footer and Or Bassok for their helpful comments on earlier versions of this article. Footnotes 1 Case C-402/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-06351; Opinion 2/13, ECLI:EU:C:2014:2454. 2 See for instance Jan Klabbers, ‘Völkerrechtsfreundlich? International Law and the Union Legal Order’, in Panos Koutrakos (ed.), European Foreign Policy: Legal and Political Perspectives (Cheltenham: Edward Elgar 2011) 95–114; Jan Klabbers, ‘Straddling the Fence: The EU and International Law’ in Anthony Arnull and Damian Chalmers (eds), The Oxford Handbook of European Union Law (Oxford: Oxford University Press, 2017) 52–71, 62–3; Gráinne de Búrca, ‘The European Court of Justice and the International Legal Order after Kadi’ (2010) 51 Harvard International Law Journal, 1. 3 Case C-308/06 Intertanko [2008] ECR I-4057; Case C-366/10 Air Transport Association of America and Others (ATAA) [ECR] 2011 I-13755. 4 The term ‘gatekeeper’ is borrowed form literature; Jan Klabbers, ‘International Law in Community Law: the Law and Politics of Direct Effect’ (2001) 21 Yearbook of European Law, 263, 296; Francis Snyder, ‘The Gatekeepers: The European Courts and WTO Law’ (2003) 40 CML Rev, 313; Marise Cremona, ‘External Relations and External Competence of the European Union: The Emergence of an Integrated Policy’ in Paul Craig and Gráinne de Búrca (eds), The Evolution of EU Law (2nd edn, Oxford: Oxford University Press, 2011) 215–68, 234. 5 Robert McCorquodale, ‘The Individual and The International Legal System’ in Malcolm D Evans (ed.), International Law (4th edn, Oxford: Oxford University Press, 2014) 280–305. 6 Sean D Murphy, ‘Does International Law Obligate States to Open Their National Courts to Persons for the Invocation of Treaty Norms That Protect or Benefit Persons?’ in David Sloss (ed.), The Role of Domestic Courts in Treaty Enforcement: A Comparative Study (Cambridge: Cambridge University Press, 2009) 61–119, 118; André Nollkaemper, ‘The Effects of Treaties in Domestic Law’ in Chrtistian J Tams and others (eds), Research Handbook on the Law of Treaties (Cheltenham: Edward Elgar, 2014) 123–50, 145. 7 Current Art. 216(2) TFEU [2012] OJ C326/47. 8 Case 181/73 Haegeman (Haegeman II) [1974] ECR 449. 9 Eeckhout distinguishes between ‘internal’ and ‘external’ direct effect; Piet Eeckhout, EU External Relations Law (2nd edn, Oxford: Oxford University Press, 2011) 229–330. 10 Case C-26/62 Van Gend en Loos [1963] ECR 6. 11 Bruno De Witte, ‘Direct Effect, Primacy, and the Nature of the Legal Order’ in Paul Craig and Gráinne de Búrca (eds), The Evolution of EU Law (2nd edn, Oxford: Oxford University Press, 2011) 323–62, 324. 12 Eeckhout (n 9), 324. 13 Bruno de Witte, ‘The Continuous Significance of Van Gend en Loos’ in Miguel P. Maduro and Loïc Azoulai (eds), The Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford: Hart, 2010) 9–15, 10. 14 If the international agreement has been implemented through regulations or directives, then the internal rules on direct effect will be relevant; Dominic McGoldrick, International Relations Law of the European Union (Harlow: Longman, 1997) 125. 15 Cases 21-24/72 International Fruit [1972] ECR 1219; Case C-280/93 Germany v Council [1994] ECR I-4973; Case C-149/96 Portugal v Council [1999] ECR I-8395; Case C-377/02 Van Parys v BIRB [2005] ECR I-1465; Joined Cases C-300/98 and C-392/98 Dior [2000] ECR I-11307. 16 Mayer as cited in Bart Van Vooren and Ramses A. Wessel, EU External Relations Law (Cambridge: Cambridge University Press, 2014) 218. 17 See nn 287, 337 below. 18Portugal v Council (n 15). 19 See for instance, Francesca Martines, ‘Direct Effect of International Agreements of the European Union’ (2014) 25 The European Journal of International Law, 129; Stefan A Riesenfeld, ‘The Doctrine of Self-executing Treaties and Community Law’ (1973) 67 American Journal of International Law, 504. 20 Carlos M. Vázquez, ‘The Four Doctrines of Self-Executing Treaties’ (1995) 89 American Journal of International Law 695; Carlos M. Vázquez, ‘Treaties as Law of the Land: the Supremacy Clause and the Judicial Enforcement of Treaties’ (2008) 122 Harvard Law Review 600; David Sloss, ‘United States’ in Sloss, (n 6), 504–54. 21 J A Winter, ‘Direct Applicability and Direct Effect: Two Distinct and Different Concepts in Community Law’ (1972) 9 CML Rev 425, 425. 22 See for instance Winter (n 21); Pierre Pescatore, ‘The Doctrine of “Direct Effect”: An Infant Disease of Community Law’ (1983) 8 EL Rev, 135; David Edward, ‘Direct Effect, The Separation of Powers and the Judicial Enforcement of Obligations’ in Scritti in onore di Guiseppe Federico Mancini (Essays in honour of Guiseppe Federico Mancini) (Milan: Dott. A. Giuffrè Editore, 1998), Volume II, 423–443; Ilona Cheyne, ‘International Agreements and the European Community legal system’ (1994) EL Rev, 581. 23 Winter (n 21), 427, 438. 24 Cheyne (n 22), 585–8. Elsewhere Cheyne analyses the Haegeman formula in relation to the effects of international agreements; Ilona Cheyne, ‘Haegeman, Demirel and their Progeny’ in Alan Dashwood and Cristophe Hillion (eds), The General Law of EC External Relations (London: Sweet and Maxwell, 2000) 20–41, 26. 25 Koen Lenaerts, ‘Direct Applicability and Direct Effect of International Law in the EU Legal Order’ in Inge Govaere and others (eds), The European Union in the World: Essays in Honour of Marc Marescegau (Leiden: Brill; Nijhoff 2013) 45–64, 46. 26 Case 12/86 Demirel [1987] ECR 3719, para. 14. 27 According to Eeckhout, the definition ‘does not put in much relief the distinction between the structure and nature of the agreement as such, and the conditions for direct effect of specific provisions’, Eeckhout (n 9), 337. 28 See for instance International Fruit (n 15); Portugal v Council (n 15); Intertanko (n 3); ATAA (n 3). 29 Opinion of AG Sharpston, Case C-240/09 Lesoochranárske zoskupenie VLK v Ministerstvo životného prostredia Slovenskej republiky (Brown Bear) [2011] ECR I–1255, para. 38, footnote 19. 30 Edward (n 22), 426. 31 Szilárd Gáspár-Szilágyi, ‘EU International Agreements through a US Lens: Different Methods of Interpretations, Tests and the Issue of Rights’ (2014) 39 EL Rev, 601, 602, 607. 32 Jean Groux and Philippe Manin, The European Communities in the International Order (European Perspectives) (European Commission, 1985) 118. 33 Holdgaard uses ‘direct invocability’ in a wider sense to incorporate not only direct effect as the ability of the individual to rely on a particular provision, but also other legal effects of international agreements; Rass Holdgaard, External Relations of the European Community: Legal Reasoning and Legal Discourses (Alphen aan den Rijn: Kluwer Law International, 2008) 244. 34 The justiciability of the norm—‘the capacity of a norm of Union law to be applied in domestic court proceedings’, should be distinguished from a narrower understanding of conferral of individual rights; De Witte (n 11), 323; Sasha Prechal, ‘Does Direct Effect Still Matter?’ (2000) 37 CML Rev, 1047, 1050; Sasha Prechal, ‘Direct Effect, Indirect Effect, Supremacy and the Evolving Constitution of the European Union’ in Catherine Barnard (ed), The Fundamentals of EU Law Revisited: Assessing the Impact of the Constitutional Debate (Oxford: Oxford University Press, 2007) 35–69, 37. 35 Edward (n 22), 426; Van Vooren and Wessel (n 16), 229; by analogy ‘direct application’ in Kees J. Kuilwijk, The European Court of Justice and the GATT Dilemma: Public Interest versus Individual Rights? (Nexed Editions, 1996) 82–103. The term ‘direct applicability’ is also used in scholarship on international law to denote the same concept: James Crawford, Brownlie’s Principles of Public International Law (8th edn, Oxford: Oxford University Press, 2012), 58; Nollkaemper (n 6), 138–42. 36 Art. 288(1) TFEU; Paul Craig and Gráinne de Búrca, EU Law: Text, Cases and Materials (6th edn, Oxford: Oxford University Press, 2015) 107; Robert Schütze, European Union Law (Cambridge: Cambridge University Press, 2015) 91; Damian Chalmers, Gareth Davies and Giorgio Monti, European Union Law (3rd edn, Cambridge: Cambridge University Press, 2014) 112. 37 Case C-41/74 Van Duyn [1974] ECR 01337, para. 12; Case C-51/76 Nederlandse Ondernemingen [1977] ECR 00113, para. 21; Case C-148/78 Ratti [1979] ECR 01629, para. 19. 38 For instance in International Fruit, the Court found that the EC was bound by the GATT even without being a party to the latter; International Fruit (n 15), paras 10–18. 39 Klabbers (n 4), 294–5; Armin von Bogdandy, ‘Pluralism, Direct Effect and the Ultimate Say: On the Relationship between International and Domestic Constitutional Law’ (2008) International Journal of Constitutional Law 397, 400; Van Vooren and Wessel (n 16), 222; Panos Koutrakos, EU International Relations Law (2nd edn, Oxford: Hart, 2015) 257. 40 See for instance constitutionalism and pluralism in Ramses A. Wessel, ‘Reconsidering the Relationship Between International and EU Law: Towards a Content-Based Approach?’ in Enzo Cannizzaro, Paolo Palchetti, and Ramses A. Wessel (eds), International Law as Law of the European Union (Leiden: Martinus Nijhoff Publishers 2012) 7–33; Von Bogdandy (n 39); pluralism in Francis Snyder, The EU, the WTO and China: Legal Pluralism and International Trade Regulation (Oxford: Hart, 2010); neo-monism in Enzo Cannizzaro, ‘The Neo-Monism of the European Legal Order’ in Cannizzaro and others, ibid, 35–58; communitarisation in Anne Peters, ‘The Position of International Law Within the European Community Legal Order’ (1997) 40 German Yearbook of International Law, 9, 34–5; and Christian Tietje, ‘The Status of International Law in the European Legal Order: The Case of International Treaties and Non-Binding International Instruments’ in Jan Wouters, André Nollkaemper, and Erika de Wet (eds), The Europeanisation of International Law: The Status of International Law in the EU and its Member States (The Hague: TMC Asser Press, 2008) 55–69, 58–9. 41Haegeman (n 8), para. 5. 42Haegeman (n 8), para. 5; Case C-301/08 Bogiatzi [2009] ECR I-10185. 43Haegeman (n 8), para. 3–4. 44 Trevor C Hartley, ‘International Agreements and the Community Legal System: Some Recent Developments’ (1983) 8 EL Rev, 383, 391; Trevor C Hartley, The Foundations of European Community Law (4th edn, Oxford: Oxford University Press, 1998) 263; Mario Mendez, The Legal Effects of EU International Agreements (Oxford: Oxford University Press, 2013) 64. 45 Giorgio Gaja, ‘Trends in Judicial Activism and Judicial Self-Restraint Relating to Community Agreements’ in Enzo Cannizzaro (ed.), The European Union as an Actor in International Relations (Alphen aan den Rijn: Kluwer Law International, 2002) 117–34, 119. 46 Cheyne (n 22), 586–7; De Witte (n 11), 336; Allan Rosas, ‘The European Court of Justice and Public International Law’ in Wouters and others, (n 40), 71–85, 75; Joost Pauwelyn, ‘Europe, America and the “Unity” of International Law’ in Wouters and others (n 40), 205–25, 222, Kuilwijk (n 35), 101. 47 Mendez, (n 44), 63. 48 Klabbers (n 4), 264. 49 Klabbers ( n 4), 276. 50 For instance, in Polydor the Court of Appeal of England and Wales appeared to link the issue of direct enforceability by individuals to the fact that the agreement was adopted by regulation; Case 270/80 Polydor [1982] ECR 329, para. 10; Bundesfinanzhov, judgment of 5 August 1980 (1980) RIW 786 as cited in Geert A Zonnekeyn, ‘The Direct Effect of GATT in Community Law: from International Fruit Company to the Banana Cases’ (1996) 2 International Trade Law and Regulation 63, 64; Peters (n 40), 22. 51 Klabbers (n 4), 275. 52 Gerhard Bebr, ‘Agreements Concluded by the Community and Their Possible Direct Effect: From International Fruit Company to Kupferberg’ (1983) 20 CML Rev, 35, 38–9; Riesenfeld (n 19), 506. 53 Iain Mcleod, Ian D. Hendry, and Stephen Hyett, The External Relations of the European Communities: A Manual of Law and Practice (Oxford: Clarendon Press, 1996) 81; Jacques H J Bourgeois, ‘Effects of International Agreements in European Community Law: Are the Dice Cast?’ (1984) 82 Michigan Law Review, 1250, 1257. 54 Groux and Manin (n 32), 115–16. 55 Case C-160/09 Katsivardas [2010] ECR I-4591, para. 34. 56 Case C-192/89 Sevince [1990] ECR I-3461, paras 8–11; Case 30/88 Greece v Commission [1989] ECR 3711, para. 13. 57 Jan Klabbers, ‘Virtuous Interpretation’ in Malgosia Fitzmaurice, Olufemi Elias, and Panos Merkouris (eds), Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years On (Leiden: Martinus Nijhoff Publishers 2010) 17–37, 20. 58 Edwards (n 22), 435. 59 Case 104/81 Hauptzollamt Mainz v Kupferberg [1982 ] ECR 364, paras 11–13. 60 According to this maxim, treaties are binding upon parties and must be performed in good faith; Art. 26, Vienna Convention on the Law of Treaties, 23 May 1969, United Nations, Treaty Series, vol. 1155, p. 331. 61 Klabbers (n 4), 281; Hartley, The Foundations of Community Law (n 44), 217. 62 Klabbers, Völkerrechtsfreundlich (n 2), 100. 63 Mcleod and others, (n 53), 126–7. 64 Klabbers, Völkerrechtsfreundlich (n 2), 100. 65 Gáspár-Szilágyi (n 31), 605–6; Lenaerts (n 25), 55. 66 Pieter Verloren van Themaat, ‘The Impact of the Case Law of the Court of Justice of the European Communities on the Economic World Order’ (1984) 82 Michigan Law Review, 1422, 1435–6. 67 Martines (n 19), 134. 68 Pieter J. Kuijper, ‘The Case Law of the Court of the Court of Justice of the EU and the Allocation of External Relations Powers: Wither the Traditional Role of the Executive in EU Foreign Relations?’ in Marise Cremona and Anne Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Oxford: Hart, 2013) 95–114, 103. 69 John H. Jackson, ‘Status of Treaties in Domestic Legal Systems: A Policy Analysis’ (1992) 86 American Journal of International Law, 310, 322. 70 Mendez, ‘The Legal Effect of Community Agreements: Maximalist Treaty Enforcement and Judicial Avoidance Techniques’ (2010) European Journal of International Law, 83, 88; Mendez, The Legal Effects of EU International Agreements (n 44), 73–6. 71 Hartley refers to ‘direct effect’ as corresponding to the concept of direct applicability as used in this article; Trevor C Hartley, ‘The Constitutional Foundations of the European Union’ (2001) 117 Law Quarterly Review, 225, 237. 72 Cremona (n 4), 234. 73 Jan Wouters, André Nollkaemper, and Erika de Wet, ‘Introduction: The “Europeanisation” of International Law’ in Wouters and others (n 40), 1–3, 8–9. 74 André Nollkaemper, ‘The Netherlands’ in Sloss (n 6), 366. 75 Wessel (n 40), 12. 76 Peters (n 40), 29. 77 Mario Mendez, ‘The Enforcement of EU Agreements: Bolstering the Effectiveness of Treaty Law’ (2010) 47 CML Rev, 1719, 1724. 78 Emphasis added; Pescatore (n 22), 173. 79 Gerrit Betlem and André Nollkaemper, ‘Giving Effect to Public International Law and European Community Law before Domestic Courts: A Comparative Analysis of the Practice of Consistent Interpretation’ (2003) 14 European Journal of International Law, 569, 573. 80 Per Bourgeois, direct effect is part of ‘a broader problem of the status of international agreements’. Klabbers notes that in relation to at least the early case law ‘the idea of direct effect is inescapable when thinking about the reception of international law in the Community legal order’. According to Eeckhout, direct effect can serve as a ‘limitation to integration of international law’. Van Vooren and Wessel, while noting that the issue of the status and validity (matters for monism/dualism) of international law should at least ‘formally’ be distinguished from the issue of direct effect, also perceive that the concept of direct effect as applied in the case law can cast a shadow over the monist nature of the EU system; Bourgeois (n 53), 1255; Klabbers (n 4), 282–83; Piet Eeckhout, ‘The Integration of Public International Law in EU Law: Analytical and Normative Questions’ in Piet Eeckhout and Manuel Lopez-Escudero (eds), The European Union’s External Action in Times of Crisis (Oxford: Hart, 2016) 190–204, 204; Van Vooren and Wessel (n 16), 218–20, 231; Ernst-Ulrich Petersmann, ‘Application of GATT by the Court of Justice of the European Communities’ (1983) 20 CML Rev, 397, 402. 81 See for instance Klabbers (n 4), 292–4, 296–7; Pieter J. Kuijper and Marco Bronckers, ‘The WTO in the European Court of Justice’ (2005) 42 CML Rev, 1316, 1354; Christina Eckes, ‘International Law as Law of the EU: The Role of the ECJ’ in Cannizzaro, Palchetti, and Wessel (n 40), 353–77; Thomas Cottier, ‘International Trade Law: The Impact of Justiciability and Separation of Powers in EC Law’ NCCR Trade Working Paper No 2009/18, 8; Antonis Antoniadis, ‘The European Union and WTO Law: A Nexus of Reactive, Coactive and Proactive Approaches’ (2007) 6 World Trade Review, 45, 83. 82 Crawford (n 35), 48–50; David Sloss, ‘Treaty Enforcement in Domestic Courts: A Comparative Analysis’ in Sloss (n 6), 1–60, 6. 83 Anthony Aust, Modern Treaty Law and Practice (3rd edn, Cambridge: Cambridge University Press, 2013) 163, 167. 84 Groux and Manin (n 32), 119; Wessel (n 40), 13. 85 AG Maduro remarked in FIAMM that ‘the fact that WTO law cannot be relied upon before a court does not mean that it does not form pat of the Community legal system’; Opinion of Advocate General Maduro, Case C-120/06 FIAMM [2008] ECR I-06513, para. 37; Bourgeois (n 53), 1260. 86 de Witte (n 13), 11. 87International Fruit (n 15). 88 Case 87/75 Bresciani [1976] ECR 129. 89 For an overview see Holdgaard (n 33), 288–98. 90 Martines (n 19), 132. 91 Pescatore (n 22), 155. 92 Murphy (n 6), 109. 93 Klabbers (n 4), 264; Joel Trachtman, ‘Bananas, Direct Effect and Compliance’ (1999) 10 European Journal of International Law, 655, 664. 94 André Nollkaemper, ‘The Duality of Direct Effect of International Law’ (2014) 25 The European Journal of International Law, 105, 124. 95 Hélène R. Fabri, ‘Is There a Case—Legally and Politically—For Direct Effect of WTO Obligations?’ (2014) 25 European Journal of International Law, 151, 151. 96 Edwards (n 22), 425. 97 Hartley, ‘International Agreements and the Community Legal System’ (n 44), 386–7; Alan Dashwood and others, Wyatt and Dashwood’s European Union Law (6th edn, Oxford: Oxford University Press, 2011) 954–5. 98Kupferberg (n 59), para 26. 99Kupferberg (n 59), paras 16–17. 100 Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 90. 101 Jurisdiction of the Courts of Danzig, 1928 PCIJ Series B, No 15, at 17–18. 102 Kuijper and Bronckers (n 81), 1320; Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 90; Arnaud Van Wayenerge and Peter Pecho, ‘Free Trade Agreements after the Treaty of Lisbon in the Light of the Case Law of the Court of Justice of the European Union’ (2014) 20 European Law Journal, 749, 753. 103 Art. 19(1) TEU [2012] OJ C 326. 104 van Themaat (n 65), 1428–9. 105 Robert Schütze, Foreign Affairs and the EU Constitution (Cambridge: Cambridge University Press, 2014) 51. 106 Opinion of Advocate General Mayras, International Fruit (n 15), 1234. 107 Klabbers (n 4), 271. 108 Nollkaemper (n 94), 118. 109 Armin Von Bogdandy, ‘Founding Principles’ in Armin von Bogdandy and Jürgen Bast (eds), Principles of European Constitutional Law (Oxford: Hart, 2006) 11–54, 29–30; Prechal, ‘Direct Effect, Indirect Effect’ (n 34), 38–9. 110 Paul Craig, ‘Once upon a Time in the West: Direct Effect and the Federalisation of EEC Law’ (1992) 12 OJLS 454, 458–63. 111 Henry G Schermers, ‘The Direct Application of Treaties with Third States: Note Concerning the Polydor and Pabst Cases’ (1982) 19 CML Rev, 563, 564–5. 112 Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 89–93; Mendez (n 44), 107–3. 113 Bebr (n 52), 41. 114 Fabri (n 95), 161. 115 Nollkaemper (n 74), 367. 116 Mendez (n 77), 1725; Nikos Lavranos, Decisions of International Organisations in the European and Domestic Legal Orders of Selected EU Member States (Amsterdam: Europa Law Publishing, 2004) 45. 117 Wouters, and others (n 73), 9. 118 De Witte (n 11), 327. 119 Kuijper and Bronckers (n 81), 1322. 120 Case 22/70 ERTA [1971] ECR 00263; C-3/76 Kramer [1976] ECR 1279; Opinion 1/76 [1977] ECR 741. 121 Pauwelyn (n 46), 219. 122 Emphasis added; Hartley, ‘International Agreements’ (n 44), 386–7; see also Kuijper and Bronckers (n 81), 1317–8. 123 Pauwelyn (n 46), 222. 124Demirel (n 26); See further Panos Koutrakos, ‘Interpretation of Mixed Agreements’ in Christophe Hillion and Panos Koutrakos (eds), Mixed Agreements Revisited: The EU and its Member States in the World (Oxford: Hart, 2010) 116–37, 118; Andrea F Gagliardi, ‘The Right of Individuals to Invoke the Provisions of Mixed Agreements before the National Courts: A New Message from Luxembourg?’ (1999) 24 EL Rev, 276. 125 Nanette Neuwahl, ‘Freedom of Movement for Workers under the EEC Treaty Association Agreement’ (1988) 13 EL Rev, 360, 365. 126Demirel (n 26), para. 9. 127 The Dutch constitution is the example often cited in scholarship; Constitution of the Kingdom of the Netherlands, June 2002. 128 Pauwelyn (n 46), 222; Snyder (n 4), 326–35, 362; Prechal, ‘Does Direct Effect Still Matter’ (n 34), 1065; Geert de Baere, Constitutional Principles of EU External Relations (Oxford: Oxford University Press, 2012) 35. 129 Kuijper and Bronckers (n 81), 1317–23. 130 Claus-Diete Ehlermann, ‘Application of GATT Rules in the European Community’ in Meinhard Hilf, Francis G Jacobs, and Ernst-Ulrich Petersmann (eds), The European Community and Gatt (Alphen aan den Rijn: Kluwer 1986) 127–40, 137. 131 Case C-61/94 Commission v Germany [1996] ECR I-3989, para. 52; Joined Cases C-402/05P and C-415/05 P Yassin Abdullah Kadi v Council [2008] ECR I-06351, para. 307; Simon Marsden, ‘Invoking Direct Application and Effect of International Treaties by the European Court of Justice: Implications for International Environmental Law in the European Union’ (2011) 60 ICLQ 737, 742. 132 Marco Bronckers, ‘From “Direct Effect” to “Muted Dialogue”: Recent Developments in the European Courts’ Case Law on the WTO and Beyond’ (2008) 11 Journal of International Economic Law 885, 896; Kuijper (n 68), 103. 133 Ehlermann (n 130), 137. 134 Schermers (n 111), 566–7. 135 Schermers (n 111), 564. 136 Martines (n 19), 147. 137 Bebr (n 52), 36–7. 138 Gáspár-Szilágyi (n 31), 606–7. 139 AG Mayras suggested a more elaborate test in comparison with the direct effect of internal EU law to include such conditions as the analysis of the provisions, evaluation of the context and general scheme of the treaty, the aim of the provision in question, having regard to overall objective of the measure, and lastly the circumstances in which the contracting parties have decided to apply the treaty; International Fruit (n 15), 1235. 140International Fruit (n 15); Bresciani (n 88); Kupferberg (n 59); Holdgaard (n 33), 246; Cheyne, ‘Haegeman, Demirel’, (n 24), 24. 141 Van Wayenerge and Pecho (n 102), 754–5. 142 For instance, Martines and Zonnekeyn consider that the condition related to the nature and the broad logic of the agreement comes first followed by the attributes of the provision, whereas Holdgaard views the attributes of the provision to be the first condition followed by the purpose and the nature of the agreement; Martines (n 19), 138; Zonnekeyn (n 50), 66; Holdgaard (n 33), 251–71. 143 For criticism of this approach see Christine Kaddous, ‘Effects of International Agreements in the EU Legal Order’ in Marise Cremona and Bruno de Witte (eds), EU Foreign Relations Law: Constitutional Fundamentals (Oxford: Hart, 2008) 291–312. 144 For instance, ‘the purpose, the spirit, and the general scheme’ in International Fruit, ‘the spirit, the general scheme and the wording’ in Bresciani; the ‘nature and structure’ of the agreement in Portugal v Council, ‘the nature and broad logic’ in IATA; International Fruit (n 15), para. 8, 19–20, 27; Bresciani (n 88), para. 16; Portugal v Council (n 15), para. 47; Case C-344/04 IATA [2006] ECR I-403, para. 35. 145 Schütze (n 105), 52. 146 Cheyne (n 22), 594; Marise Cremona, ‘Who Can Make Treaties? The European Union’ in Duncan B. Hollis (ed.), The Oxford Guide to Treaties (Oxford: Oxford University Press, 2012) 93–124, 112; Klabbers (n 4), 279–80. 147 For observations on earlier case law see Bebr (n 52), 58; for a more general overview of the application of the first condition see Koutrakos (n 39), 259–67. 148 See further Zonnekeyn (n 50); Hilf and others (n 130); Piet Eeckhout, ‘Judicial Enforcement of WTO Law in the European Union: Some Further Reflections’ (2002) 5 Journal of International Economic Law, 91; Snyder (n 4); Nicholas Emiliou and David O’Keeffe (eds), The European Union and World Trade Law After the GATT Uruguay Round (Oxford: Wiley 1996); Kuijper and Bronckers (n 81); Fabri (n 95); Bronckers (n 132); Koutrakos (n 39), 280–301. 149 Even though the EC did not accede to the GATT, the Court established its binding effect by relying on succession theory; International Fruit (n 15), paras 10–18, 21; Case 9/73 Schlüter [1973] ECR 1135. 150 Cheyne, ‘Haegeman, Demirel’, (n 24), 22–3. 151 Bebr (n 52), 46. 152 Jan Klabbers, ‘The Validity of EU Norms Conflicting with International Obligations’ in Cannizzaro and others (n 40), 111–31, 125. 153 In Germany v Council, Germany argued that the issue of direct effect is irrelevant for judging the validity of EU law in light of GATT 1947; Case C-280/93 Germany v Council (Bananas) [1994] ECR I-4973, para. 103; AG Saggio in Portugal v Council argued against using direct effect for legality challenges (n 15); Geert Zonnekeyn, ‘The Status of the WTO Agreements in the EC Legal Order after the Portuguese Textiles Cases’ (2000) 6 International Trade Law and Regulation, 42, 47; Martines (n 19), 141, 143; Kuijper and Bronckers (n 81), 1343–54. 154 Eileen Denza, ‘A Note on Intertanko’ (2008) 33 EL Rev, 870, 875–6. 155 Koutrakos (n 39), 282. 156 Sasha Prechal, Directives in European Community Law (Oxford: Oxford University Press, 1995), 276; Prechal, ‘Direct Effect, Indirect Effect’ (n 34), 37–8. 157 This observation is made in relation to WTO law; Van Vooren and Wessel (n 16), 221. 158Portugal v Council (n 15), para. 36–47. 159Van Parys (n 15), para. 53. 160 Klabbers (n 4), 276. 161Bresciani (n 88); Polydor (n 50); Kupferberg (n 59). 162Portugal v Council (n 15), paras 44–5. 163 Allan Rosas, ‘Case Comment: Case C-149/96, Portugal v Council’ (2000) 37 CML Rev, 797, 807. 164 Geert Zonnekeyn, ‘Mixed Feeling about the Hermes Judgment’ (1999) 5 International Trade Law and Regulation, 20, 24. 165 Opinion of Advocate General Alber, Case C-93/02 P Biret [2003] ER I-10497, para. 102; Lavranos (n 116), 40. 166Portugal v Council (n 15), para. 40. 167 Bebr (n 52), 58. For the analysis of the role dispute settlement mechanisms play in the finding of direct effect see further Beatrice I. Bonafé, ‘Direct Effect of International Agreements in the EU Legal Order: Does it Depend on the Existence of an International Dispute Settlement Mechanism?’ in Cannizzaro and others (n 40), 229–48. 168 Cannizzaro, ‘The Neo-Monism’ (n 40), 43. 169 Bronckers (n 132), 895. 170 See below Section V. 171 Cremona (n 4), 241. 172 Kuijper (n 68), 105, 114. 173 Fabri (n 95), 164 174 Gáspár-Szilágyi (n 31), 610. 175 Klabbers, ‘Völkerrechtsfreundlich‘ (n 2), 97. 176 Achilles Skordas, ‘Völkerrechtsfreundlichkeit as Comity and the Disquiet of Neoformalist: A Response to Jan Klabbers’ in Koutrakos (ed.) (n 2), 115–44, 129–30. 177 Skordas (n 176), 129–30. 178 Case C -69/89 Nakajima v Council [1991] ECR I-2069; Case 70/87 Fediol [1989] ECR 1781; see further Snyder (n 4), 342–7; Koutrakos (n 39), 301–7. It should be noted that certain authors view these exceptions as part of indirect effect; Peters (n 40), 71. 179Commission v Germany (n 131), para. 31; Koutrakos (n 39), 307–11; Eeckhout (n 9), 355–7. 180 Klabbers, ‘Völkerrechtsfreundlich‘ (n 2), 102. 181 Koutrakos (n 39), 264. 182Intertanko (n 3). 183Intertanko (n 3), paras 45–65. 184Intertanko (n 3), paras 54–65; Piet Eeckhout, ‘Case C-308/06, The Queen on the application of Intertanko and Others v Secretary of State for Transport, judgment of the Court of Justice (Grand Chamber) of 3 June 2008, nyr’ (2009) 46 Common Market Law Review 2041, 2054. 185 Eeckhout (n 184), 2054. 186International Fruit (n 15), paras 8, 19–20, 27. 187 Eeckhout (n 9), 382. 188 Van Vooren and Wessel (n 16), 232; Cannizzaro, The Neo-Monism of the European Legal Order (n 40), 49. 189 Denza (n 154), 875. 190 Cremona (n 4), 242. 191 Mendez points out the inconsistency with the Biotech and IATA judgments where the Court seemed to separate the issue of legal review from that of conferral of individual rights; C-377/98 Biotech [2001] ECR I-7079; C-344/04 IATA [2006] ECR I-403; Mendez (n 44), 246, 250, 273–81; Mendez (n 77), 1751. 192 Eeckhout, (n 184), 2055. 193 Eeckhout points out the importance of the multilateral nature of the agreement, many provisions of which codify customary international law; (n 184), 2041. 194 Eeckhout (n 80), 199. 195 Denza (n 154), 877–8. 196 Cremona (n 4), 243. 197 Gáspár-Szilágyi (n 31), 613. 198 Mendez (n 77), 1751. 199Air Transport Association of America and Others (n 3). 200 Jed Odermatt, ‘The Court of Justice of the European Union: International or Domestic Court’ (2014) 3 Cambridge Journal of International and Comparative Law, 696, 707. 201Air Transport Association of America and Others (n 3), paras 75–6. 202 Koutrakos (n 39), 260. 203Air Transport Association of America and Others (n 3), para. 77. 204 Gáspár-Szilágyi (n 31), 621–4. 205 Gáspár-Szilágyi (n 31),, 613. 206 Koutrakos (n 39), 266. 207 Mendez (n 44), 246, 250, 273–81; Eeckhout (n 9), 381–3. 208 For an overview of the application of this condition see Koutrakos (n 39) 267–70. 209 Gáspár-Szilágyi (n 31), 614–15. 210 Case C-18/90 Kziber [1991] ER I-199, para. 25; Case C-265/03 Simutenkov [2005] ECR I-2579, para 25. 211Case C-213/03 Pêcheurs de l'étang de Berre [2004] ECR I-07357, paras 41–42. 212Simutenkov (n 210), paras 26–28; see further Christophe Hillion, ‘Russian Federation’ in Stephen Blockmans and Adam Lazowski (eds), The European Union and its Neighbours: A Legal Appraisal of the EU’s Policies of Stabilisation, Partnership and Integration (The Hague: TMC Asser Press, 2006) 463–97. 213 Case C-162/00 Pokrzeptowicz-Meyer [2002] ECR I-1049, paras 20–27. 214 Kaddous (n 143), 304. 215 Mendez (n 44), 151; Francis G Jacobs, ‘Direct Effect and Interpretation of International Agreements in the Recent Case Law of the European Court of Justice’ in Allan Dashwood and Marc Maresceau (eds), Law and Practice of EU External Relations: Salient Features of a Changing Landscape (Cambridge: Cambridge University Press, 2008) 13–33, 32. 216 Koutrakos (n 39), 269. 217Simutenkov (n 210), para. 21. 218Simutenkov (n 210), para. 21. 219 Jacobs (n 215), 32. 220 Hillion particularly notes that no such obligation comes through in the Spanish version of the provision on which the applicant relied; Christophe Hillion, ‘Case C-265/03, Igor Simutenkov v. Ministerio de Educación y Cultura, Real Federación Española de Fútbol’ (2008) 45 CML Rev, 815, 824. 221 Hillion (n 220), 823, 826–7. 222Demirel (n 26), paras 17–25. 223 It should be noted that similar departures from Demriel were made in Kziber (n 210), para. 25. 224 In Pokrzeptowicz-Meyer itself the analysis followed the same pattern, that is finding that the provision was clear, unconditional, and not dependent on further action before considering the nature and purpose of the agreement; Porkzeptowicz-Meyer (n 213), paras 19–26; Kaddous (n 143), 304. 225Simutenkov (n 210), paras 26–28. 226Simutenkov (n 210), para 28. 227 Hillion (n 212), 474. 228Kziber (n 210); Tietje (n 40), 62–2; Bebr (n 52), 63. 229 The Court views it as ‘essentially an economic agreement’; Case C-162/96 Racke [1998] ECR I-3655, para. 20. 230Kziber (n 210), para. 21, Case C-113/97 Babahenini [1998] ECR I-183, para. 17; Case C-103/94 Krid v CNAVTS [1995] ECR I-719, paras 21–23; Case C-58/93 Yousfi v Belgian State [1994] ECR I-1353, paras 16–18; Case C-126/95 Hallouzi-Choho v Bestuur van de Sociale Verzekeringsbank [1996] ECR I-4807, para. 19; Racke (n 229), paras 34–36. 231 Gáspár-Szilágyi (n 31), 619. 232 Martines (n 19), 145. 233Lesoochranárske zoskupenie VLK (Brown Bear) (n 29), paras 44–45. 234 Jacobs (n 215), 32–33. 235 On the Court’s methods of interpretation see Eeckhout (n 9), 304–19. 236 Mendez (n 44), 110. 237 Van Wayenerge and Pecho (n 102), 757. 238Opinion 1/91 [1991] ECR I-6079, para. 14; Jacobs (n 215), 16. 239 Case 225/78 Bouhelier [1977] ECR 3151, para. 6; Polydor (n 50), paras 14–18; Kupferberg (n 59), paras 29–31; Case C-312/91 Metalsa [1993] ECR 3751, paras 11-19. 240Bresciani (n 88), para. 25; Case C-268/99 Jany [2001] ECR I-8615, paras 33–38; Cremona (n 146), 113. 241Metalsa (n 239), para. 11. 242 Peters (n 40), 23. 243 Delano Verwey, The European Community, the EU and the International Law of Treaties (The Hague: TMC Asser Press, 2004) 223–4. 244Bresciani (n 88), para. 25. 245 Kaddous (n 143), 312. 246 Van Wayenerge and Pecho (n 102), 758–62; Chrtistophe Hillion, ‘Cases C-63/99 Secretary of State for the Home Department ex parte Wiesław Głoszczuk and Elzbieta Głoszczuk; C-235/99 Secretary of State for the Home Department ex parte Eleanora Ivanova Kondova; C-257/99 Secretary of State for the Home Department ex parte Julius Barkoci and Marcel Malik; judgments of the Full Court of 27 September 2001; Case C-268/99 Aldona Małgorzata Jany e.a v. Staatssecretaris van Justitie, judgment of the Full Court of 20 November 2001; Case C-162/00 Land Nordrhein-Westfalen v. Beata Pokrzeptowicz-Meyer, judgment of the Full Court of 29 January 2002’ (2003) 40 CML Rev, 465, 489; see further Niamh Nic Shuibhne, ‘European Law—the European Court of Justice and the Europe Agreements: Shaping a Legal Framework’ (2001) 23 Dublin University Law Journal, 203. 247 Van Wayenerge and Pecho (n 102), 758–62. 248 Koutrakos (n 39), 274. 249 This inconsistency can be seen in contrasting Pabst and Richarz; Pokrzeptowicz-Meyer; Kolpak on the one hand, and, Gloszczuk; Kondova; Barkoci and Malik, on the other; Case C-17/81 Pabst and Richarz [1982] ECR 1331; Pokrzeptowicz-Meyer (n 213); Case C-438/00 Kolpak [2003] ECR I-04135; Case C-63/99 Gloszczuk [2001] ECR I-06369; Case C-235/99 Kondova [2001] ECR I-06427; Case C-257/99 Barkoci and Malik [2001] ECR I-06557; Hillion (n 246), 489–91; Antje Pedain, ‘A Hollow Victory: the ECJ Rules on Direct Effect of Freedom of Establishment Provisions in Europe Agreements’ (2002) 61 Cambridge Law Journal 284, 287. 250Gloszczuk (n 249), para. 86; Kondova (n 249), para. 91; Barkoci and Malik (n 249), para. 66. 251 Hillion (n 246), 490. 252 Barbara Bogusz, ‘Regulating the Right of Establishment for Accession State Nationals: Reinforcing the “Buffer Zone” or Improving Labour Market Flexibility’ (2002) 27 EL Rev, 472, 482. 253 Antje Pedain, ‘“With or Without Me”: The ECJ Adopts a Pose of Studied Neutrality towards EU Enlargement’ (2002) 51 ICLQ 981, 989; Bogusz (n 252), 482. 254 Pedain (n 249). 255 Case C-416/96 El Yassini [1999] ECR I-1209, para. 30. 256Kolpak (n 249), paras 25–26. 257Simutenkov (n 210), paras 21–24; Pokrzeptowicz-Meye (n 214), para. 22. 258 Hillion (n 220), 831. 259 Jacobs (n 215), 28. 260Simutenkov (n 210), paras 33, 36–38; Hillion (n 220), 829–830. 261 Nollkaemper identifies the dual function of direct effect which includes being used both as a ‘sword’ and a ‘shield’; Nollkaemper (n 94), 112–17. 262 De Witte (n 11), 336. 263 Pescatore (n 22), 149. 264 Mendez (n 77), 1720; Peters (n 40), 63–4, 77. 265 Klabbers, ‘Straddling the Fence’ (n 2), 64. 266 See note 99 above. 267 John H Jackson, ‘Direct Effect of Treaties in the US and the EU, the Case of the WTO: Some Perceptions and Proposals’ in Anthony Arnull, Piet Eeckhout, and Takis Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford: Oxford University Press, 2008) 361–82, 370. 268 Schütze (n 105), 52; Holdgaard (n 33), 251–71. 269 Cheyne, ‘Haegeman, Demirel’, (n 24), 41. 270 Kuijper as cited in Snyder (n 4), 335. In relation to the effect of WTO law, Advocate General Alber favoured seeking a multilateral solution instead of a unilateral finding of direct effect; Opinion of Advocate General Alber, C-27 &122/00 Omega Air [2002] ECR I-2569, para. 95. 271 Multilateral Agreement between the European Community and its Member States, the Republic of Albania, Bosnia and Herzegovina, the Republic of Bulgaria, the Republic of Croatia, the former Yugoslav Republic of Macedonia, the Republic of Iceland, the Republic of Montenegro, the Kingdom of Norway, Romania, the Republic of Serbia, and the United Nations Interim Administration Mission in Kosovo on the establishment of a European Common Aviation Area [2006] OJ L 285. 272 [1998] OJ L42/43. 273 [1998] OJ L42/43, Art. 17(3). 274 Andre Nollkaemper, ‘The Direct Effect of Public International Law’ in Jolande M Prinssen and Annette Schrauwen (eds), Direct Effect: Rethinking a Classic of EC Legal Doctrine (Amsterdam: Europa Law Publishing 2002) 157–80, 171. 275 According to Art. 336, ‘[n]othing in this agreement shall be construed as conferring rights or imposing obligations on persons, other than those created between the Parties under public international law’; Trade Agreement between the European Union and its Member States, on the one part, and Colombia and Peru on the other part [2012] OJ L354/3. 276 Art. 17.15 of EU–Singapore Free Trade Agreement is entitled ‘No direct effect’ and provides that ‘[f]or greater certainty, nothing in this Agreement shall be construed as conferring rights or imposing obligations on persons, other than those created between the Parties under public international law’. The Agreement was initialled, but following the CJEU’s Opinion 2/15 it cannot be concluded in its current form solely by the EU; Opinion 2/15, ECLI:EU:C:2017:376. 277 According to Art. 356, ‘[n]othing in this agreement shall be construed as conferring rights or imposing obligations on persons, other than those rights or obligations created by this agreement nor as obliging a party to permit that this agreement be directly invoked in its domestic legal system, unless otherwise provided in that party’s domestic legislation’; Agreement establishing an Association between the European Union and its Member States, on the one hand, and Central America on the other [2012] OJ L346/3. 278 Art. 30.6 provides that ‘[n]othing in this Agreement shall be construed as conferring rights or imposing obligations on persons other than those created between the Parties under public international law, nor as permitting this Agreement to be directly invoked in the domestic legal systems of the Parties’; Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union and its Member States, of the other part [2017] OJ L 11/23 279 Aliki Semertzi, ‘The Preclusion of Direct Effect in the Recently Concluded EU Free Trade Agreements’ (2014) 51 CML Rev, 1125, 1130. 280 For instance free trade agreements, cooperation agreements, partnership and cooperation agreements; Kupferberg (n 59); Simutenkov (n 210); Racke (n 229). 281 Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other part [2014] OJ L261; Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and the Republic of Moldova, of the other part [2014] OJ L260; Association Agreement between the European Union and its Member States, of the one part, and Ukraine, of the other part [2014] OJ L161. 282 Semertzi (n 279), 1134, 1141–2. 283 Guillaume van der Loo, Peter van Elsuwege, and Roman Petrov, ‘The EU–Ukraine Association Agreement: Assessment of an Innovative Legal Instrument’, EUI Working Papers No. 9 (2014), 26; Semertzi (n 279), 1155–6. 284 Kuijper and Bronckers (n 81), 1321. 285 Koutrakos (n 39), 258; Semertzi (n 279), 1129. 286 Art. 5, Council Decision 2016/838/EU of 23 May 2016 on the conclusion, on behalf of the European Union, of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other part [2016] OJ L 141/26; Art. 5, Council Decision 2014/295/EU on the signing, on behalf of the European Union, and provisional application of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Ukraine, of the other part, as regards the Preamble, Art. 1, and Titles I, II and VII thereof [2014] OJ L161/1; Art. 5, Council Decision 2016/839/EU of 23 May 2016 on the conclusion, on behalf of the European Union, of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and the Republic of Moldova, of the other part [2016] OJ L 141/28. 287 Schermers (n 111), 564. 288 Some agreements might require implementing measures, which would be adopted prior to or simultaneously with entering into the agreements; Mcleod, and others (n 53), 128–9. 289 Council Decision 94/800 [1994] OJ L 336/1, recital 14. 290 Peters (n 40), 60. 291 McGoldrick,(n 14), 133. 292Portugal v Council (n 15). 293 Opinion of AG Saggio in Portugal v Council (n 15), para. 20. 294 AG Saggio (n 293), para. 20. 295 AG Saggio (n 293), para. 20. 296 The CJEU considers the EU to be bound only by those provisions of the Vienna Convention which are part of customary international law (Case C-386/08 Brita [2010] ECR I-1289, para. 42). Art. 31 of the Vienna Convention is seen as part of customary international law; Crawford (n 35), 380. 297 AG Saggio (n 293), para. 20. 298Portugal v Council, (n 15), para. 35. 299Portugal v Council, (n 15), paras 36–48. 300 Rosas (n 163), 808. 301 Fabri (n 95), 162. 302 Rosas (n 163), 810; Mendez (n 44), 246. 303 The Commission’s accord with the Council was obvious in the case of the WTO Agreement, where it declared its unequivocal support for the exclusion of direct effect; Commission, Uruguay Round Implementing Legislation, COM(94) 414 final; Cremona (n 4), 241. 304 Kuijper and Bronckers (n 81), 1321. 305 André Nollkaemper (n 6), 363. 306 Lori F. Damrosch, ‘Role of the United States Senate Concerning Self-Executing and Non-Self-Executing Treaties’ (1991) 67 Chicago-Kent Law Review 515, 515, 518. 307 Vázquez, ‘The Four Doctrines of Self-Executing Treaties’ (n 20), 706–8; Vázquez, ‘Treaties as Law of the Land’ (n 20), 667–85; Carlos M. Vázquez, ‘Laughing at Treaties’ (1999) 99 Columbia Law Review 2154, 2174; Damrosch (n 306), 526–7; Thomas Buergenthal, ‘Modem Constitutions and Human Rights Treaties’ (1997) 36 Columbia Journal of Transnational Law 211, 221–2. 308 Vázquez, ‘Treaties as Law of the Land’ (n 307), 672; Stefan A Riesenfeld, ‘International Agreements’ (1989) 14 Yale Journal of International Law 455, 466. 309 The European Parliament rejected the Anti-Counterfeiting Trade Agreement in 2012; ‘European Parliament Rejects ACTA’ Press Release, 4 July 2012; Ricardo Passos, ‘The External Powers of the European Parliament’ in Eeckhout and Lopez-Escudero (eds) (n 80), 83–128. 310 Art. 218(6) TFEU. 311Biotech (n 191); Case C-479/04 Laserdisken [2006] ECR I-8089; IATA (n 144); Intertanko (n 3). 312 European Parliament, Resolution on the Relationship Between International Law, Community Law and the Constitutional Law of the Member States, 24 September 1997, Explanatory statement, para. 3.2. 313 European Parliament (n 312). 314 European Parliament (n 312), para. 14. 315 Snyder (n 40), 172. 316 Damrosch (n 306), 528–9. 317 AG Saggio (n 293), para. 20. 318 Art. 2(1)(d) of the Vienna Convention on the Law of Treaties. 319 Verwey (n 243), 225–6. 320 Oliver Dörr, ‘Art 31: General Rule of Interpretation’ in Oliver Dörr and Kirsten Schmalenbach (eds), Vienna Convention on the Law of Treaties: A Commentary (Springer 2012) 521–70, 552; Richard K Gardiner, Treaty Interpretation (Oxford: Oxford University Press, 2008) 211, 214–15. 321 Aust (n 83), 212. 322 Dörr (n 320), 550. 323 Mark E Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Leiden: Martinus Nijhoff 2009) 430. 324 Dörr (n 320), 552. 325 Verwey (n 243), 226. 326 Mark E Villiger, ‘The Rules on Interpretation: Misgivings, Misunderstandings, Miscarriage? The “Crucible” Intended by the International Law Commission’ in Enzo Cannizzaro (ed.), The Law of Treaties Beyond the Vienna Convention (Oxford: Oxford University Press, 2011) 105–22, 114. 327 Dörr (n 320), 549. 328 Jean-Marc Sorel and Valerie Bore Eveno, ‘Article 31: General Rule of Interpretation’ in Olivier Corten and Pierre Klein (eds), The Vienna Conventions on the Law of Treaties: A Commentary (Oxford: Oxford University Press, 2011) 804–37, 808. 329 Various commentators indeed dismissed the significance of the preambular statement as it was not part of the decision’s operative part; Kuijper and Bronckers (n 81), 1345; Jacques H J Bourgeois, ‘The Uruguay Round of GATT: Some General Comments from an EC Standpoint’ in Nicholas Emiliou and David O’Keeffe (eds), The European Union and World Trade Law After the GATT Uruguay Round (Oxford: Wiley, 1996) 81–90, 89; Dashwood and others (n 97), 957. 330 Van der Loo and others (n 283), 27. 331 Oddly, a similar provision was not included in the EU–Georgia and EU–Moldova Association Agreements. 332El-Yassini (n 255); Kziber (n 210). 333 Art. 23, Agreement on partnership and cooperation establishing a partnership between the European Communities and their Member States, of one part, and the Russian Federation, of the other part [1997] OJ L 327. 334 Van der Loo and others (n 283), 27. 335 European Council, Press Remarks by H Van Rompuy, President of the European Council, following the EU–Ukraine Summit, Brussels 25 February 2013, EUCO 48/13. On EU–Ukraine Association Agreement see further Guillaume van der Loo, The EU–Ukraine Association Agreement and Deep and Comprehensive Free Trade Area: A New Legal Instrument for EU Integration without Membership (Leiden: Brill 2016). 336 Art. 8, Council Decision 2011/265/EU of 16 September 2010 on the signing, on behalf of the European Union, and provisional application of the Free Trade Agreement between the European Union and its Member States, of the one part, and the Republic of Korea, of the other part [2011] OJ L 127/1. 337 Semertzi (n 279), 1135. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals-permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Yearbook of European Law Oxford University Press

Who Are the ‘Gatekeepers’?: In Continuation of the Debate on the Direct Applicability and the Direct Effect of EU International Agreements

Loading next page...
 
/lp/ou_press/who-are-the-gatekeepers-in-continuation-of-the-debate-on-the-direct-YuM22sU008
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals-permissions@oup.com
ISSN
0263-3264
eISSN
2045-0044
D.O.I.
10.1093/yel/yey003
Publisher site
See Article on Publisher Site

Abstract

Abstract The article revisits the debate on direct applicability and direct effect of EU international agreements by questioning the role of the so called gatekeepers. It considers the established role of the Court of Justice of the EU as the gatekeeper of the EU legal order through identifying the stages of gatekeeping and their implications. It further analyses the possibilities of sidelining the Court through various techniques, which include the agreement between the parties to the international agreement. A more controversial challenge to the Court’s position stems from a practice emerging from Council decisions concluding a number of international agreements. These decisions make a strong pronouncement on the exclusion of direct effect for the entire agreement. The status of such pronouncements is analysed with reference to CJEU’s jurisprudence as well as the relevant rules of international law. I. Introduction The reception and the status of international law in domestic legal orders is a problem which many modern states continue to grapple with. Similarly, in the European Union (EU), the last decade witnessed a renewed debate on the relationship between the EU legal order and international law, not least due to the Kadi saga and Opinion 2/13.1 The openness of the EU legal order to wider international law has been increasingly challenged.2 Certain recent cases, denying direct effect to such multilateral agreements as the United Nations Convention on the Law of the Sea (UNCLOS) and the Kyoto Protocol to the United Nations Framework Convention on Climate Change,3 also raised question marks overs the EU’s commitment to international law and emphasized the ‘gatekeeping’ exercised by the Court of Justice of the EU (CJEU).4 As far as international agreements are concerned, the gatekeeper role has been assumed by the Court in a legal vacuum. Despite the significant developments in international law propelling the role of the individual in terms of creation of rights and imposition of personal responsibility,5 there is no general rule in international law imposing an obligation on states or international organizations, such as the EU, to satisfy rights of individuals stemming from international treaties in domestic courts.6 Neither does international law establish rules on the incorporation of an international agreement into the domestic legal systems. These matters are decided internally on the basis of national constitutions or other foundational documents, or through institutional intervention. The original Rome Treaty and its successive revisions neither referred to the status of international agreements nor their effects merely stating that agreements were binding upon the EU institutions and Member States.7 The judicial intervention on the matter initially came through the seminal Haegeman judgment according to which the provisions of international agreements upon their ratification by the EU become part of its legal order.8 The Court broke further grounds by finding that international agreements can have direct effect by transposing its ‘internal’ concept of direct effect into its external relations.9 Established in Van Gend en Loos,10 direct effect recognized the justiciability of the provisions of the Rome Treaty and came to be viewed as a ‘defining characteristic of EU law’.11 Any provision of the Rome Treaty, which was clear, unconditional, and required no further implementing measures was capable of direct effect. The paradox of the extension of this doctrine to international agreements lies in the fact that the doctrine helped to ‘define’ the EU legal order ‘in opposition to international law’,12 largely due to reversing the presumption that international treaties were not capable of having direct effect.13 In simple terms, the extension of direct effect to international agreements renders their provisions justiciable without internal implementing measures.14 The externalization of direct effect led to a gradual moulding of its rather flexible criteria on the basis of which various bilateral and multilateral agreements were found to have justiciable provisions, with some notorious exceptions, namely the persistent finding of the lack of direct effect of agreements under the Gereral Agreement of Tariffs and Trade (GATT) or the World Trade Organization (WTO).15 Most significantly, the direct effect became a fixture of a vast amount of cases involving challenges to the actions of both the Member States and the EU institutions in light of the commitments undertaken through EU international agreements. The decades of jurisprudence appeared to settle the Court’s gatekeeper or ‘door opener’16 role. The latter, however, is not set in stone. While the Court’s findings on the incorporation of international agreements remain largely unchallenged by other institutions, the opposite seems to be true in relation to direct effect. A number of recently concluded agreements, in a rather unusual fashion, set out the effect of their provisions themselves, leaving no choice to the CJEU and signalling the intention of the treaty-making institutions to take back control over the matter. In addition, the Council of the EU in its decisions concluding certain agreements makes strong statements on the exclusion of direct effect for entire agreements.17 In the past, a similar statement made in the Council Decision adopting the WTO Agreement, did not receive any clarification as to its status by the Court of Justice.18 What weight should then be accorded to these decisions? This article aims to answer this and other questions by revisiting the issue of who the gatekeepers are. The discussion is restricted to international agreements (and decisions of bodies established under the latter), and does not include customary international law as the trends described above are particular to international agreements (and by implication to the bodies mentioned above). The role of the Court of Justice as the main gatekeeper to international law is considered first. For this purpose, the Court’s key jurisprudence on the issue of the incorporation of international agreements and direct effect is analysed. Next, the sidelining of the Court as a gatekeeper is questioned through two techniques: either through the agreement between the parties or through unilateral pronouncements by the Council of the EU. The article is concluded with a brief summary of findings. To start with, a clarification of the terms used in this article is in order. II. A brief terminological clarification The Court’s gatekeeper role evolved primarily through its findings on the issues of the incorporation of international agreements in the EU legal order and the justiciability of the provisions of international agreements. Various terms have been used in the past to deal with these two issues. While some scholars find the term ‘self-executing treaties’ helpful in analysing the effects of EU international agreements,19 the latter never became a fixture of EU law unlike in the US legal order.20 The concepts of ‘direct applicability’ and ‘direct effect’ are the ones that are used in EU law most commonly even though accompanied by lack of clarity inter alia due to their interchangeable use both in jurisprudence and scholarship. Early on, Winter cautioned against treating these concepts as equivalent.21 Many have debated the linkages and correlation between the two but their interchangeable use has continued.22 The reason for this might be in the very fact of interchangeable use of these terms by the Court to denote the same notion. This lead to many commentators distinguishing them within the parameters of one single concept. Winter himself viewed the concept of direct applicability as one relating to ‘the specific nature of the treaty contents’ which he advises should not be confused with the issue of incorporation, even though he does not object against using the term in the latter sense in the internal EU legal order.23 Cheyne also viewed both concepts of direct applicability and direct effect as relating to the effects of international agreements without necessarily separating the issue of their incorporation: direct applicability relates to such features of the specific provision, as clarity, precision, unconditionality and lack of further implementing measures, direct effect is a narrower concept denoting the possibility of individuals relying on a particular provision.24 These two issues, however, are both encompassed within the notion of direct effect: one is part of the concept itself, the other refers to its conditions. Most recently, Lenaerts, writing extrajudicially, also viewed direct applicability and direct effect as two distinct notions: direct applicability denotes whether an international agreement requires further implementing measures to be deduced from the parties’ intentions, most commonly by examining the nature and the logic of the agreement, while direct effect is simply a quest into the provision’s unconditional and precise characteristics.25 Such distinction is somewhat problematic if one turns to Demirel, considered to be the classic authority on what constitutes direct effect: A provision … must be directly applicable when, regard being had to its wording and the purpose and nature of the agreement itself, the provision contains a clear and precise obligation which is not subject, in its implementation or effects, to the adoption of any subsequent measures.26 The case sets the conditions for direct effect to include an inquiry into the nature and purpose of the agreement, and the provision itself to determine whether it is clear, precise and requires no further implementation in light of the agreement. Here, as an example of interchangeable use, direct applicability is applied to denote direct effect. While the Demirel definition is not without its critics,27 the latter and the subsequent jurisprudence of the Court does not necessarily make the distinction between direct applicability and direct effect in a manner suggested by Lenaerts. Instead, such delineation is somewhat problematic in view of the Court’s practice on applying the conditions noted above. Even though the application of these conditions is characterized by significant flexibility, discussed further below, such delineation fragments the concept of direct effect and is capable of adding further confusion. First, the analysis of the nature and the broad logic of the agreement focuses on a much wider range of factors than whether the agreement requires implementing measures or not.28 Besides, the need for further implementing measures is usually considered in relation to the second Demirel condition, that is whether the provision is clear, sufficiently precise and requires no further implementing measures, which indeed would be determined within the wider context of the agreement. A taxonomical clarification was also done in passing by the Advocate General Sharpston in Brown Bear according to which the term direct applicability—corresponding to the notion of ‘self-executing treaties’—denotes the instances where international agreements require no EU or national implementing legislation.29 It is not clear, however, whether the lack of the need of implementing measures is about the incorporation of the international agreement into the EU legal order or its effects in courts, which are two distinct matters.30 This terminological confusion even led to suggestions to dismiss the familiar terms in favour of new concepts such as ‘direct judicial enforceability’,31 ‘invocability’,32 or ‘direct invocability’ as ‘the capability of a legal subject to rely on (ie use or invoke) that norm in a particular context’.33 However, calls to replace the familiar terms remain unanswered, as direct applicability and direct effect are being continuously used. These terms, nonetheless, are well suited to denote two distinct but related issues: direct effect, pertaining to the justiciability of a particular provision (in its broader understanding),34 can be distinguished from direct applicability referring to a separate issue of incorporation or ‘automatic integration’ of the EU international agreements into the internal legal order of the EU and its Member States without the need for transposing measures.35 Distinguishing direct applicability from direct effect in this manner is preferable for a number of reasons. First, it is semantically more accurate. Second, it avoids fragmenting the concept of direct effect as applied by the CJEU. Finally, it is more compatible with the internal use of the term. The reference to regulations in Article 288 of the Treaty on the Functioning of the EU (TFEU) uses direct applicability not to denote its effects but rather the mechanism of its automatic transposition into the legal order of the Member States.36 Transposition measures are distinct from implementing measures, as individual provisions of regulations (as well as Treaties) might require implementing measures. The Court’s case law on the direct effect of directives is particularly instructive in the internal distinction made between direct applicability and direct effect.37 Making the distinction between these two terms in the fashion suggested above would be most faithful to the logic, the findings and the language of Van Gend en Loos. At the same time, we should keep a close sight of the relationship between the two concepts. Direct applicability is the precondition for direct effect as international agreements have to be incorporated into the internal legal order prior to the consideration of their effects. The Court’s has played a paramount role in developing both of these notions. III. The tales of Haegeman and direct applicability of EU international agreements as the primary stage of judicial gatekeeping The Court of Justice’s gatekeeper role was assumed first through its findings on direct applicability of international agreements. The latter is essentially about the binding nature of international agreements concluded by the EU and the EU and its Member States jointly. While in certain cases the EU can be bound by agreements it did not conclude or accede to,38 in the majority of the cases the question turns to the mechanism of incorporation of agreements concluded by the EU or the EU and its Member States jointly into the internal legal order of the EU. It is often with reference to the concepts of monism and dualism that the status and validity of international treaties in internal legal systems is determined. Many have been sceptical, however, about the ultimate utility of these concepts not only in relation to the EU, but more generally,39 since monism or dualism in their pure form are rather uncommon. Instead, alternative frameworks have been advanced to address the issue of coexistence of different legal systems.40 As the concepts of monism and dualism are not central to the discussion that follows, they are referred to when relevant for considering the pathway to the Court’s gatekeeper role. We start with the Haegeman story concerning the Court’s jurisprudence to interpret the provisions of an Association Agreement with Greece in a preliminary reference procedure. The Court in an apparent straightforward manner proclaimed that upon entering into force the provisions of EU international agreements formed part of EU legal order.41 An EU international agreement enters into force after its ratification according to the procedure set out in the Treaty,42 currently found in Article 218 TFEU. The Court’s reasoning was minimal and simplistic. It was based on the assertion that the Council decision on the conclusion of an international treaty is ‘an act of one of the institutions of the [Union]’,43 and therefore part of the EU legal order. This shortcut, however, left a glaring gap in the judgment, that is that the agreement itself is not an act of an EU institution, but rather an act of the EU as an organization.44 This approach might have been driven by the determination to comply with the wording of Article 267 whereby the Court’s interpretative function is limited to Treaties and ‘acts of the institutions, bodies, offices or agencies of the Union’.45 It, nonetheless, paved the way for conflicting interpretations. On the one hand, the reasoning was interpreted by many as an indication of the adoption of a monist system,46 or automatic incorporation.47 Others pointed to the failure of the judgment to provide clarity on whether the treaty automatically becomes part of internal EU law or by virtue of the Council’s act.48 The phrasing used by the Court—it is the provisions of the international agreement, rather than the agreement itself, that become the integral part of the Union law—is viewed as pointing towards the possibility that it is due to the Council’s act and not by virtue of the agreement on its own.49 This would be indicative of dualist characteristics. Another omission in Haegeman might indicate otherwise, though. The Court’s disregard for the nature and the form of the Council’s act (decision in the case) can be interpreted as a testimony to the latter’s insignificance. In its subsequent case law, the Court similarly did not pursue the apparent importance the national courts attached to the form and nature of the act approving the agreement.50 Besides, the emphasis put on the fact that the provisions of an international treaty become an integral part of EU legal order from the moment the latter enters into force also suggests the lesser importance of the decision.51 Various agreements were concluded both by Council decisions and regulations,52 fuelling assumptions that regulations are relied upon for agreements capable of direct effect, while decisions are used for agreements with no such effect.53 Such attempted distinctions, however, conflate the issues of direct applicability and direct effect.54 Only as late as in 2010 did the Court confirm that the form of the act is of no consequence for the issue of direct effect.55 In any event, currently Article 218 TFEU does not leave much choice: the Council ‘shall adopt a decision concluding the agreement’. This would have no bearing on the Haegeman findings. Returning to the limited reasoning in Haegeman, why was such an important question given such a facile and shallow answer? Perhaps, because the case evolved around a related but a different question. It would be recalled that Haegeman involved a challenge to the Court’s jurisdiction to interpret the provisions of the EC–Greece Association Agreement. The main findings, therefore, do not intend to address primarily the issue of the status of international law in the EU legal system or its doctrinal underpinnings. Rather, it is a shortcut to establish the Court’s jurisdiction to interpret the agreement, later also extended to decisions of bodies established under international agreements.56 The finding on the Court’s interpretative role was paramount as ‘whoever controls the process of interpretation, therewith controls the truth, or at least the meaning to be given to the text subject to interpretation’.57Haegeman, therefore, was the stepping stone for the Court’s gatekeeper role laying the foundations for further stages of gatekeeping related to the effects of agreements. Most importantly, the case signalled a clear openness to international law, despite leaving much scope for speculation on how exactly international agreements became part of EU law. Further clarification was due; although instead of answering the question how, the Court opted to answer the question why. The subsequent judgment in Kupferberg has been viewed as offering a ‘sounder’ analysis of the issue of direct applicability.58 Building upon Haegeman, the Court adds a new rationale to the automatic incorporation of EU international agreements: it derives from the assumption of obligations by the EU and its Member States towards third countries, as well as the obligations assumed by the Member States towards the EU.59 An outwards- and inwards-looking rationale should be distinguished here. The outwards-looking rationale is tied to the principle of pacta sunt servanda entrenched in the current Article 216(2) TFEU.60 The inwards-looking rationale, even though it upholds the same principle of pacta sunt servanda, acknowledges the internal dynamics of the EU. Rather than being preoccupied with the concepts of monism and dualism, the Court’s basic concern is the implementation of the Union’s commitments which might depend on the Member States.61 The latter should not be ‘undermined by recalcitrant Member States’ and stems from ‘the perceived necessity of protecting the autonomy of the [Union] legal order’.62 This obligation reinforces the internal commitments required from the Member States based on Article 4(3) of the Treaty of European Union (TEU) on duty of cooperation enabling the EU to perform its obligations internationally.63 Such clarification of the nature of the obligations of the Member States also solidifies the Court’s ‘full control’ over the application of EU international agreements in line with Haegeman.64 Both Haegeman and Kupferberg, hence, can be seen as being concerned with the internal dynamics of the EU legal system. Kupferberg, nonetheless, failed to address the shortcomings of Haegeman related to the mechanism of incorporation of agreements. Even if one interprets the cases above in favour of a monist view, it has been suggested that the Haegeman formula does not lead to automatic incorporation as there are certain qualifications to be made, including the respect for the constitutional values of the EU, and the application of this formula only in the areas of exclusive competence in the cases of mixed agreements.65 However, rather than affecting the incorporation of the agreement into the EU legal order, these factors are relevant for the issue of the validity, rather than the incorporation of the agreement, and as such do not impact the Haegeman findings. While the jury is still out on whether the EU legal order displays monist or dualist features (some argue that both monist and dualist interpretations can be advanced),66 the cases have been commonly interpreted to signal the ‘openness’ of EU law to international law. Such openness had significant implications for other actors both at supranational and national levels. It has an exclusionary impact on other EU institutions as it allows only for ex-post assessment of the essence of the rules of international law.67 The ability of other institutions to adapt to international law is curtailed,68 albeit that at the time the impact would have been greater for the Council than the Parliament which had limited powers in EU foreign relations. The exclusionary effect, however, guarantees that the application of the agreement cannot be undermined by the refusal of political institutions to transpose it into internal law.69 On the other hand, even bigger losers in this process are the Member States, as such moulding of the EU legal order is at the expense of modifications to their constitutional rules on external relations.70 While direct applicability at its core is about ‘the transfer of a provision between [two legal] systems’,71 in this case it is instead about three legal orders. The findings in Haegemen and Kupferberg determined not only the relationship between international law and EU law, but also between international law and the national legal orders of the Member States.72 This has been branded as the ‘Europeanization’ of international law which introduces a European element to the ‘“classical” dual legal relationship international law/national law’ turning it into ‘a new triangular relationship, international law/EU law/national law’.73 Such ‘indirect’ reception of international law in the legal order of the Member State,74 in combination with the principle of supremacy in EU law, introduced a detectable ‘openness’ in the internal legal orders of the Member States irrespective of their monist or dualist traditions,75 ensuring a uniformity as far as the supremacy of EU international agreements was concerned.76 While for Member States with monist legal systems, such intervention at the supranational level might not signal any drastic changes at first sight, for those with clear dualist features, this demonstrates a radical departure in terms of guardedness towards international agreements, and introduces a bifurcation between agreements accepted via EU law and those concluded by the Member States through their usual procedures requiring transposing measures.77 Others, however, argue that the openness to international law is not settled with direct applicability alone. This is where the direct effect of international agreements makes an entrance and takes central stage in this debate by building upon the Haegeman findings which provided ‘a basis for recognizing’ the direct effect of international agreements within the EU legal order.78 Indeed, direct effect requires prior rules settling the matter of the incorporation of internal law into an internal legal order.79 Perhaps, because the case law was not sufficiently or substantively clear on the matter of the incorporation of international law into the EU legal order, it is the concept of direct effect that has come to be seen as adding another layer of understanding to the position international agreements occupy in the EU legal order. Thus, various commentators, even though distinguishing the effects of the agreements from the issue of their incorporation, nonetheless interpret direct effect as a qualifying factor for the issue of incorporation of international treaties,80 possibly introducing a dualist element to this exercise.81 These views can be justified if one considers monism and dualism to be notions determining the relationship between different legal systems.82 More commonly, however, monism and dualism denote the process of the incorporation of a treaty into a domestic legal order.83 From this perspective, caution is required against conflating the issues of direct effect or ‘invocability of a treaty provision’ with the classification of the system as monist or dualist.84 Indeed, in this latter sense, the rejection of direct effect does not reverse the finding that EU international agreements are an integral part of the EU legal order.85 Neither does it affect the manner in which the Court arrived at that conclusion. However, rejection of direct effect can certainly cast a shadow over the exposure of EU law to international law. In this respect, direct applicability can be seen as the initial stage of gatekeeping that is static in nature, as the outcomes of individual cases cannot vary: all agreements concluded by the EU are directly applicable. The Court’s stance here can be interpreted as signalling an unconditional openness to international law. It is the further stages of gatekeeping, revolving around the concept of direct effect, that are capable of closing the proverbial gates and rendering international agreements toothless as far as their enforcement in national courts is concerned. IV. Judicial gatekeeping through direct effect: implications and phases The extension of the internal doctrine of direct effect to EU international agreements (including those concluded jointly with Member States) as a category to a wide range of agreements was seen as its ‘second major expansion’ (alongside directives) after its extension to a vast number of treaty provisions.86 Early on, the case of International Fruit demonstrated that the Court was willing to open the EU legal order to international law by considering the possibility of recognizing direct effect, although denying the latter to the GATT.87 This possibility materialized in Bresciani, the first case positively acknowledging the direct effect of the Yaoundé Convention.88 This positive finding was also extended to a wide range of agreements through the course of the next decades.89 The recognition of direct effect in principle was a progressive development as the justiciability of norms of international treaties is the exception rather than the default,90 contrary to Pescatore’s well-known remark that direct effect was ‘the normal condition of any rule of law’.91 There has even been a suggestion that in view of the proliferation of international instruments setting rights or remedies for individuals, the development of international law might lead towards ‘a more general obligation to allow invocation of treaty norms by individuals in national courts in situations where the treaty contains provisions that are protective of individuals’, stimulated by the same considerations as those that motivated the Court of Justice in finding direct effect.92 The finding that international agreements are capable of direct effect was, thus, the next significant phase in signalling openness to international law. This, however, was a two-sided coin for this general finding was accompanied by an intricate and fluid jurisprudence allowing the Court to close the curtain at different sub-phases of this exercise. We shall start with the general finding of direct effect by the Court and its implications before turning to its sub-phases. A. The implications of granting direct effect to international agreements The extension of an internally developed legal technique to international agreements was bound to have political implications. While political considerations already transpire at the stage of direct applicability, the politics is even starker with direct effect as an ‘inherently political’ concept used for political purposes,93 and according the court with a political power.94 Even the arguments made to revisit the notion of direct effect are political in nature.95 Direct effect is said to be ‘about the separation of powers, and specifically about the extent of the judicial power to enforce the obligations of the state’.96 In the context of the EU, it is the judicial power to determine the EU’s obligations, namely those of its political institutions and Member States, that is at stake. While the positive finding of direct effect was not imperative,97 as noted above, the Court opted to grant direct effect to international agreements in principle. The earlier cases, referred to in the previous paragraph, did not shed much light on the reasons behind the Court’s assumption of responsibility on the matter of the effects of international agreements. In this respect, the already familiar to us later case of Kupferberg was more significant. Here, Article 21 of the EEC–Portugal Trade Agreement prohibiting discriminatory internal fiscal measures was found to be directly effective.98 The Court dismissed the argument against direct effect on the basis of the relationship between EU institutions: In conformity with the principles of public international law [Union] institutions which have power to negotiate and conclude an agreement with a non-member country are free to agree with that country what effect the provisions of the agreement are to have in the internal legal order of the contracting parties. Only if that question has not been settled by the agreement does it fall for decision by the courts having jurisdiction in the matter, and in particular by the Court of Justice within the framework of its jurisdiction under the Treaty, in the same manner as any question of interpretation relating to the application of the agreement in the [Union].99 At the outset, the Court takes a step back. It does not claim to have an inherent jurisdiction to decide on the matter: basing its analysis on the premise of the principles of international law, the Court’s role would be ‘residual’100 or secondary to the agreement between the parties. The acknowledgement that it is the prerogative of the parties to decide is seen as homage to Danzig guidance in international law, where the object of the agreement ‘according to the intention of the contracting Parties, may be the adoption by the Parties of some definite rules creating individual rights and obligations and enforceable by the national courts’.101 However, the Court’s argument has been criticized for being ‘naïve and inconsistent’ since issues related to the internal effects of international agreements do not have a significant part in the negotiations.102 This aspect of the Court’s finding is discussed further in the article. Next, the Court’s analysis moves to the domain of EU law to substantiate its competence: this is done with reference to its jurisdiction to interpret EU law.103 Having established that international agreements upon their conclusion form part of EU legal order, by implication the Court’s jurisdiction extends to their interpretation as well. Thus, the judgment in Haegeman laid ‘the legal foundation of the Court’s competence in all cases where international obligations have been accepted by an act of the Council under treaty-concluding powers’.104 The Court’s jurisdiction to decide on the effects of international agreements was tied to the exclusivity of this exercise. In the name of guaranteeing the uniformity of EU law, the Court effectively ‘monopolised’ this issue.105 One of the most forceful arguments in favour of this finding was put forward by Advocate General Mayras in International Fruit: The unity and, it can be said, the very existence of [Union] law require that the Court is alone empowered to say, with the force of law, whether an agreement binding the [Union] or all the Member States is or is not [directly effective] within the territory of the [Union] and, if it is, whether or not a measure emanating from a [Union] institution conforms to that external agreement.106 This can be linked to one of the two, at times rivalling considerations, which motivated the Court in its finding of direct effect, that is the inclination to advance international law and its desire to preserve the ‘unique character’ and the autonomous legal order of the EU.107 The first would suggest that the very finding of the possibility of direct effect would indicate advancement of international law and ultimately its efficiency.108 Thus, one rationale in the opening of the gates to EU international treaties by according them direct effect is linked to the arguments of the efficient enforcement of law, direct effect being one of the underlying notions of the principle of effectiveness in EU law,109inter alia to mitigate the limitations of public enforcement.110 Individual reliance on international treaties strengthens their enforcement by adding a further element of supervision.111 The enforcement of international agreements through granting them direct effect is seen as an example of the Court of Justice’s so called ‘maximalist’ approach to the implementation of international law.112 The Court, therefore, assumed the role of the guarantor of the application of international law in the EU legal order. From this perspective, the Court’s leading role was needed to guarantee not only the uniformity of interpretation, but also the efficiency of Union law of which the agreements formed an integral part.113 Returning to the second consideration of preserving the ‘unique character’ and the autonomous legal order of the EC, vertical and horizontal dimensions of the institutional or power balance can be identified, that is between the EU and its Member States vertically and between the EU institutions horizontally.114 The ‘door-opening’ through direct effect furthers the impact of direct applicability on the national legal orders of the Member States. The finding that provisions of agreements concluded by the EU and third countries can be invoked in national courts on the basis of centrally set criteria sidelines domestic constitutional arrangements pertaining to the effects of international treaties with varying impact depending on the constitutional traditions of the Member States.115 But even for monist states, it signals a significant limitation of the powers of national courts since the Union concept of direct effect limits their ability to determine how far the effects of international agreements can reach.116 In this respect, the ‘Europeanization’ of international law entails the central role of the CJEU not only in relation to the status, but also to the effects of ‘Europeanized’ agreements.117 In fact, it is the ‘centralization’ of the issue of direct effect that is seen at times as the ‘crucial contribution’ of the doctrine of direct effect.118 Besides, Kupferberg was decided at a time when the Court played a ‘dominant’ role in the EU, and in view of the relative weaknesses of the EU as an emerging external actor, it preferred to be ‘closely guarded’ by (international) law.119 Should the Court have ruled out a direct effect of EU international agreements, it would have clearly undermined them as a source of EU law. Thus, the extension of direct effect to international agreements should also be placed within the context of the Court’s wider findings on external relations in this period, including its paramount role in drawing up the external competences of the EU.120 At the same time, the Court was not the only institution whose profile was to be boosted by the finding of direct effect. By giving direct effect to international agreements, the Court in addition ‘elevated the [EU] institutions as a whole in their power struggle with the Member States’ through an emphasis on the role of the Union as an international actor ‘capable of concluding treaties with direct effect and supremacy in the domestic legal systems of the Member States’.121 The finding of direct effect, therefore, might have been motivated by ‘the desire to provide an effective way of enforcing agreements against Member States’.122 Particularly in the areas of emerging EU competences, direct effect is seen as a means of affirming the powers of the EU vis-à-vis the Member States as ‘competing internal actors’.123 Another development of the doctrine should be noted here. Significantly, the Court also stretched its finding that EU international agreements are capable of having direct effect to ‘mixed agreements’ which are concluded jointly by the EU and the Member States.124 This finding is, of course, relevant for the parts of the agreements falling solely under EU competences. On the downside, while mixed agreements permit avoiding strict delimitation of competences, the possible finding of direct effect invites the Court to do exactly that.125 This might result in an interpretation of competence unfavourable to the Member States. For instance, in Demirel, the Court interpreted the EC–Turkey Association Agreement to conclude that the provision on the free movement of workers fell under EU competences against the objections of the Member States.126 Notwithstanding the affirmation of the EU’s overall position vis-à-vis the Member States, through its finding of direct effect the Court also pitched itself horizontally against other EU institutions. While in some jurisdictions, the national constitutions might provide for direct effect (even though rarely),127 in all others the choice of the institution is about the separation of powers as noted earlier.128 The judicial finding of direct effect can have implications pre- and post-conclusion of the international agreement. It can present a potential ‘threat’ during the process of negotiations undermining the position and bargaining strength of other EU institutions.129 This observation is particularly made in relation to WTO agreements. For the majority of EU agreements, however, it is the post-conclusion finding of direct effect that has a restrictive impact on institutional choices. It ‘entails a decisive shift of responsibility from the executive to the judiciary’ whereby the Court determines issues which are usually part of the responsibility of the executive.130 The legislative choices are also impacted due to the principle of supremacy where international agreements rank below the Treaties, but above secondary legislation.131 Hence, the legislation adopted subsequent to the international agreement cannot override the latter,132 rendering the legislature unable to compensate for the shift of the responsibility from the executive to the judiciary.133 Despite the encroachment upon the executive and legislative powers, there are arguments in favour of the judicial lead on this matter. Since international agreements are part of EU law and it is the Court’s task to interpret EU law, ‘it would be contrary to the structure of the [EU] to leave the auto-interpretation of international obligations to an institution other than the Court’.134 The judiciary is seen as more objective in its interpretative task in comparison with the executive,135 and is more mindful of upholding the rule of law.136 Nonetheless, as noted above, by assuming this role the Court casts itself into the political realm. The politics of judicial gatekeeping comes across most prominently through the various sub-phases of what constitutes the direct effect exercise rooted in the setting and the application of the conditions on direct effect, as well as interpreting the directly effective provisions considered in turn below. B. The setting and application of the conditions for direct effect as the sub-phases of the direct effect exercise It was clear at the outset that any automatic transposition of an internally developed concept to EU international agreements would be problematic in terms of the establishing of the conditions for international agreements’ direct effect.137 Caution was required due to the difference in context and given the political nature of the negotiation and conclusion of international agreements.138 The conditions set in Van Gend en Loos, that of the clarity, precision, and unconditional nature of the provisions, would, therefore, have to be supplemented by additional conditions.139 These conditions did not crystallize immediately, however, and a string of early case law, including cases recognizing direct effect lacked clarity in this regard.140 Two conditions can be deduced from Demirel, cited above. The first relates to the entire agreement, its nature and purpose, whereas the second focuses on a specific provision to establish its ‘normative intensity’ through ‘a positive test’.141 The initial point of the Court’s flexibility is the uncertain relationship between the two conditions. Commentators even diverge on the sequence of the conditions,142 which is reflective of the schizophrenic case law of the court, potentially changing the nature of the exercise.143 The application of the first condition will be addressed prior to returning to the issue of the relationship between the two conditions. Even though at various times the Court deployed different terminology to denote the condition on the nature and logic of the agreement,144 it is essentially a ‘policy test’,145 aimed at determining the intentions of the parties in accordance with the principles of international law.146 While the level of scrutiny of the nature, logic, structure, scheme, the spirit, etc, varies from case to case, the Court is also at liberty to imply different factors within its analysis.147 The political underpinnings of this condition were particularly evident in the WTO law-related case law which has been extensively analysed and commented upon.148 Suffice it to refer here to the factors that led to the rejection of direct effect and the reasons behind it. (i) The GATT and WTO saga Early cases denying direct effect to the GATT agreement did so on the basis of certain of its features, such as the principle of negotiations, characterized by flexibility, including the possibility of derogations and the special dispute settlement mechanism.149 This was criticized for being an ‘unsatisfactory legal test’ due to the uncertainty embedded in the idea of ‘flexibility’ as many international agreements would include derogations, as well as procedures for reserving conflicts.150 It is significant that these early cases concerned challenges to EU law, whereby the denial of direct effect effectively meant ensuring the validity of the then Community acts.151 It should be noted here, that while the finding that international agreements are capable of direct effect signified an openness to international law, the Court simultaneously took a significant step in the opposite direction by linking the issue of validity of secondary EU law in light of obligations assumed under international agreements to the latter’s direct effect. The Court’s jurisprudence is seen here to be based on an ‘(unspoken) assumption’ that EU law should be presumed to be compatible with international law.152 Despite continuous criticism and judicial challenges against such an extension of the doctrine,153 this position is entrenched with no signs of a reversal in sight. It can, however, be defended on the ground that invalidating EU legislation in actions by individuals who themselves are incapable of relying on the agreement would be ‘a draconian step’ hardly envisaged by the Treaties.154 The linking of legality actions to the direct effect of international agreements is not restricted to challenges by individuals and includes those by Member States which is seen as justified due to the ‘broad construction’ of the principle of direct effect.155 Due to this wider function of direct effect, some have suggested a wider definition for the latter to highlight its function as a measure for legality review.156 In this context, direct effect assumed an additional function to those noted earlier becoming also a means of solving the problem of a ‘collision of norms’.157 Such a ‘collision of norms’ also took place beyond the GATT, where the direct effect of the WTO agreement was similarly ruled out despite the differences in GATT and WTO arrangements. The decisive factors for the rejection of any direct effect of the WTO agreement were the centrality of negotiations, the dispute settlement mechanism which might have an impact on the negotiation position of the EU legislative and executive institutions, as well as the so called reciprocity.158 Reciprocity raises the issue of whether the other parties to the agreement have granted direct effect.159 This particular feature is seen as introducing ‘a clear political element’ to the matter,160 as it allows a determination of the effect of international law vis-à-vis the position of other parties. Even though a stumbling block for direct effect, the Court ignored this argument as far as bilateral agreements are concerned.161 The Court itself acknowledges this inconsistency in Portugal v Council but justifies it by noting that a lack of reciprocity would lead to ‘disuniform application of the WTO rules’.162 Another justification for this dichotomy was proposed by Rosas whereby the lack of recognition of direct effect by other parties was not problematic per se, but the explicit exclusion of direct effect by them can be.163 Indeed, the direct effect of WTO agreement is excluded by most other members.164 Others, however, criticize this reasoning for belonging to the realm of economics than more law.165 As to the WTO dispute settlement mechanism, despite its mandatory nature, the Court linked it to the negotiating freedom of the legislative or executive organs.166 The presence of the dispute settlement mechanism does not necessarily bar direct effect in non-WTO law related case law,167 and at times does not even merit the Court’s attention.168 However, maintaining such distinct approaches has become unsustainable.169 Recently, a number of trade and other agreements include dispute settlement mechanisms which are modelled after the WTO and are capable of ruling on WTO related obligations. The Court, however, will not be required to reconsider its position as the matter has been decided by explicitly precluding direct effect as discussed further below.170 Most significantly, the features of the WTO rules chosen to negate direct effect demonstrate a political concern for the position of the EU and its institutions. The combined reasoning of the Court demonstrates a conscious limitation of its own role in recognition that in certain circumstances the obligation to comply with international agreements is a matter for the legislative and the executive institutions.171 In particular, it is the Council’s role as a legislator (later a co-legislator with the European Parliament) and the Commission’s role as a negotiator that is at stake.172 The interests of the political institutions of the EU are protected not only externally, but also internally. By linking the validity of internal EU legislation to the conditions of direct effect, the Court guarantees the latter’s intactness retaining the prerogatives of the legislative. Besides, secondary EU legislation often represents an, at times painstaking, institutional consensus which the Court would be unwilling to strike down. This is particularly the case when challenges to EU legislation are brought by the Member States: by denying direct effect the Member States are being directed ‘to the place where they are supposed to exert their influence, through the political institutions’.173 The Court’s approach attracted much criticism. Its ‘purposive interpretation’ has been seen as falling short of the principles of good faith interpretation in international law.174 Despite acknowledging the political sensitivity entailed by granting direct effect to WTO law, the Court’s position is also interpreted as lacking openness to international law.175 On the other hand, the denial of direct effect to WTO law does necessarily indicate that the Court’s case law is at odds with ‘the structural principles of the world trade system’.176 Moreover, the alternative could have been counterproductive as not only it would have failed in reforming the WTO law, but it could also have undermined the interests of EU producers facing increased challenges by individuals from other WTO members.177 The harshness of the exclusion of direct effect for WTO law was somewhat mitigated through he Nakajima and Fediol exceptions, which allow for challenges against the legality of EU secondary legislation against WTO law in case of either a clear reference or transposition.178 In addition, the duty of consistent interpretation, the so called indirect effect of international agreements, was developed to oblige the European courts to interpret EU secondary legislation in light of the relevant international agreements.179 The indirect effect is enabled precisely because of the direct applicability of international agreements and their ranking above the secondary legislation as mentioned earlier. Neither the technique of indirect effect, nor the limited application of the exceptions, however, can fully compensate for the exclusion of direct effect. Until recently, the lack of openness characterizing the GATT/WTO line of case law was seen as exceptional, even though it qualified for ‘a substantial part of the empirical material’ on the subject.180 Currently, it is safe to say that WTO law is not exceptional in its rejection of direct effect on the basis of the first conditions as it had been denied to two other—notably multilateral—agreements. Even though in all of these cases direct effect was denied through the application of the first condition, the latter is applied differently in all three cases.181 (ii) Intertanko and lack of direct effect of the UNCLOS Intertanko raised inter alia the issue of the validity of secondary EU legislation in light of the MARPOL Convention (the International Convention for the Prevention of Pollution from Ships) and the UNCLOS.182 While the challenge against the MARPOL Convention failed on a different ground, the UNCLOS was found to lack direct effect precluding the legality review of EU legislation. Even though direct effect is not mentioned in the case, the Court focused on the nature and broad logic of the agreement,183 and it is here that the judgment ‘innovated’ by focusing on the issue of conferral of rights.184 Accordingly, in setting a wide regulatory regime the UNCLOS aims to achieve a balance between the interests of various states and does not grant any individual rights or freedoms. While the conferral of rights has been largely dormant within the case law on WTO and bilateral agreements,185 it is nonetheless not a complete novelty, as the issue of conferral of rights originally featured within International Fruit.186 While some have suggested incorporating the conferral of rights into the analysis of direct effect,187 others view this with caution since this would most probably preclude the direct effect of the majority of international agreements.188 It would also entail the narrowing of the concept of direct effect from its wider understanding as the justiciability of the norm. While some praised the ‘correctness’ of Intertanko,189 others noted the lack of effort on the part of the Court to address any of the aspects of its reasoning which were the stumbling blocks for the direct effect of WTO law.190 Even if one views the WTO law as ‘a case apart’ deserving exceptional treatment, Intertanko unnecessarily relies on the issue of conferral rights injecting further inconsistency in the case law.191 The particularly narrow view of conferral of rights adopted in the case has also be criticized by the proponents for incorporating this factor into the relevant analysis.192 Some, however, see parallels with the case law on WTO law. Apart from the multilateral nature of the agreement,193 it has been suggested that, similar to WTO law, concerns about binding the hands of the EU executive and legislative would have played a part,194 even though the judgment itself avoids such reasoning. Intertanko is also seen as shifting the emphasis to the Member States to ensure compliance with international law when drafting legislation.195 In both Portugal v Council and Intertanko, the Court’s position is seen as cautionary taking stock of ‘the structures and processes established by the agreement, the role played in them by the EU and its Member States, and the need to avoid fragmentation in the presentation of the Union interest in such international regulatory regimes’.196 Furthermore, some have interpreted Intertanko as suggesting that the Court was protecting its own interests to avoid following the rulings of such a powerful international court as the International Tribunal on the Law of the Sea.197 If such considerations indeed played a part in the Court’s reasoning, then they are masked by its reliance on the conferral of rights. By denying direct effect through the latter as part of the first condition results in a ‘general immunisation of EU norms from review vis-à-vis UNCLOS’,198 which would also include any subsequent norms. (iii) ATTA and the Kyoto Protocol The third instance of denying direct effect on the basis of the first conditions in Air Transport Association of America and Others (ATAA) concerned the issue of the validity of EU Directive 2008/101/EC including aviation in the EU emission trading scheme in light of the Chicago Convention, the Kyoto Protocol, and the Open Skies Agreement between the European Union and the United States of America.199 For the purpose of this discussion it is the findings on the Kyoto Protocol to the United Nations Framework Convention on Climate Change that are most relevant. Even though the case is not viewed as ‘groundbreaking’ in setting any new rules,200 it nevertheless introduced certain new features as far as the first condition of direct effect is concerned. Instead of relying on its previous case law on multilateral treaties to draw on the factors decisive in the ruling out of direct effect, here the Court focuses on a different issue—that of the flexibility of the implementation in the obligations of the parties.201 The flexibility relates to the manner and the speed of fulfilling the relevant obligations depending on the parties’ agreement. In addition to such novel element, another peculiarity in ATAA relates to the way the Court appears to incorporate the analysis of the second condition into the first one.202 In support of its reasoning on the flexibility available to the parties in terms of the manner and speed of meeting their obligations, the Court cites Article 2(2) of the Kyoto Protocol which ‘cannot in any event be considered to be unconditional and sufficiently precise so as to confer on individuals the right to rely on it in legal proceedings’ contesting internal EU law.203 While the emphasis on conferral of rights appears to be in accord with Intertanko, it nonetheless appears to blur the lines between the two conditions. The case, therefore, does not in any way help to resolve the confusion as to whether direct effect requires conferral of individual rights.204 It can be argued that what the case has in common with WTO law is deference to the political institutions of the EU which will be in charge of implementing the provisions of the Kyoto Protocol, including via the International Civil Aviation Organization (ICAO). Another possible explanation for the judgment is that the Court attempted, in this way, to protect a more developed internal EU regulation in comparison with international norms.205 While, on the one hand, a clear inconsistency emerges in the application of the first condition of direct effect, the position described above demonstrates the flexibility injected into its application. Although this might be justified in view of the necessity to make ‘various adjustments in the light of factors subject to constant evolution’,206 the lack of convincing reasoning as to the introduction of new factors undermines the coherence and the continuity of the case law.207 The flexibility in the application of conditions of direct effect is not solely restricted to the first condition. (iv) The flexibility embedded in the application of the second condition and the reversal of conditions The application of the second condition on clarity and unconditionality of specific provisions is also characterized by a high level of flexibility and often leniency,208 as well as a certain subjective elements present in its assessment.209 The leniency can manifest itself in various forms. The Court can declare a provision in an agreement to be unconditional and requiring no further implementation even though the latter is subject to recommendations by an institution established under the agreement.210 The Court’s lenient approach at times also surfaces in a scarce and fleeting analysis.211 Leniency is also apparent where the Court supports its positive findings on the second condition with past precedents on other types of agreements without having analysed the nature and logic of the agreement first.212 Furthermore, as noted above, there is also a certain flexibility in terms of the choice of which condition to apply first, which will potentially impact the nature of the exercise. In many cases, the features of the specific provision are analysed first whereas the first condition on the nature and the purpose of the agreement is given scarce attention post factum to determine that it does not preclude direct effect.213 This approach has been criticized as being potentially counter-productive, bacause ruling on the specific provision prior to establishing the purpose of the agreement would be useless if the purpose of the agreement were not such as to allow for direct effect.214 On the other hand, the reversal of the conditions’ order might suggest that the Court is predisposed to granting direct effect. Indeed, cases of this kind characterize the Court’s approach at its most permeable with a high rate of success.215 In some cases, where a specific type of agreement is concerned, this approach might not be as controversial. For instance, due to the extensive case law on association agreements, it is accepted that their nature is such as to afford direct effect. Excluding direct effect for such agreements would be equal to ‘ignoring [their] raison d’être as the foundation for gradual integration between parties’.216 However, the practice of the reversal of the conditions is not restricted to the types of agreements which were perviously found to be directly effective. A useful example in terms of the flexibility embedded in the application of the second condition as well as the order of the conditions is Simutenkov involving a non-discrimination clause in the EU–Russia Partnership and Cooperation Agreement (PCA).217 According to Simutenkov, as long as a ‘clear and precise obligation’ is found which is not subject to any implementing measures, then the provision can be directly effective if the nature of the agreement does not preclude it.218 Jacobs notes that the essential issue here was whether the agreement includes provisions which can ‘directly govern the position of individuals’ in which case the analysis should turn to the specific features of the relevant provision.219 In this instance, it appears to be manifested through the imposition of an obligation despite the lack of clarity in the phrasing of the provision.220 The reversal of the order of the conditions, and the limited emphasis on the first condition meant the latter played only ‘a marginal role’ in the Court’s findings, which is strikingly at odds with the case law on WTO law.221 In addition to the reversal of the order of the conditions, Simutenkov demonstrates the flexibility the Court injects into the application of the second condition. The principle of non-discrimination was to be implemented on the basis of the recommendations of the PCA Cooperation Council. In earlier Demirel, the Court ruled out direct effect for a provision of the EEC–Turkey Association Agreement due to the measures to be adopted by the Association Council.222 In a departure from Demirel, the Court found that the Cooperation Council recommendations were intended solely to ‘facilitate’ the compliance with the principle of non-discrimination, rather than limit the immediate application of the prohibition on discrimination.223 What is remarkable is the Court’s reliance on case law related to the EC–Poland Association Agreement where a positive finding on direct effect was made in relation to the principle of non-discrimination.224 This is significant because the Court at that stage did not consider the nature and the purpose of the agreement, and therefore made no distinction between the two agreements. Only after finding that the provision is clear, unconditional, and does not require further measures, did the Court move to the first condition. Despite the PCA falling short of promising association, offering less advantageous cooperation and less ambitious objectives, the Court confirmed that the lack of close links similar to association does not preclude direct effect.225 Furthermore, in a circular motion, the Court relied on the agreement’s general ability to govern the position of individuals.226 Thus, the apparent differences in the status of the partners and their different prospects in forging close relations with the EU did not have much impact on the CJEU’s findings.227 Indeed, the early case law on other types of agreements had already demonstrated that looser integration prospects or weaker links with the EU did not necessarily have a bearing on the finding of direct effect.228 Even in bilateral agreements, merely providing for cooperation,229 the Court undertook a similarly structured analysis focusing on the second condition.230 On the one hand, this would suggest the lesser significance accorded to the first condition, whereby the gatekeeping takes place predominantly through the application of the second condition and is of a less intense character. The reversal of the conditions is not restricted to bilateral agreements only. Indeed, there is a suggestion that if the Court commenced its analysis with the second condition in the WTO line of cases, the outcome might have been different.231 Nonetheless, as discussed above, other fundamental considerations played their part in the rejection of the direct effect of the WTO law. In addition, the preference for which condition to start with is itself indicative of the nature of the gatekeeping exercise the Court is about to undertake. The Court exercises an assessment as to whether the agreement in question might require a legislative and executive intervention: ‘the [Court] operates a sort of ex post control … closely related to the mechanism of automatic incorporation of international treaty into the EU legal system’.232 Following this approach in the case of a multilateral agreement will not necessarily lead to a positive outcome. In Brown Bear, the Court commenced its analysis of the direct effect of the Aarhus Convention with the second condition but eventually rejected direct effect.233 By deciding initially that the relevant provision was subject to further implementing measures, the Court dismissed the need for evaluating the nature and the purpose of the agreement. This can be seen as an open-minded form of gatekeeping, leaving open the possibility of finding other provisions of the Aarhus Convention directly effective. While this approach might raise the issue of the continuous relevance of the first condition, the judicial choice as to which condition to consider first will depend on the nature of the agreement.234 As seen in Intertanko and ATAA, the Court can opts to commence its analysis with the first condition depending on the agreement in question. Thus, the first condition is still part of the legal test, albeit a very flexible one. To sum up the setting and the application of the conditions of direct effect, this stage of gatekeeping demonstrates that the Court has not bound itself to being either open or closed, rather it has created a wide margin of discretion for ad hoc solutions. The Court’s flexible application of the conditions of direct effect demonstrates that the EU legal order can be successfully closed to international agreements which are capable of impacting the interests of the EU as a political organization. Such interests would pertain to the external and internal roles of EU institutions, as well as the intactness of secondary EU law. The more permissive approach towards bilateral agreements, however, does not necessarily mean ultimate openness to their ability to have an impact on individuals. This openness can be rebutted by the application of the second condition, where a negative answer leaves the international agreement toothless as far as the individual is concerned. In addition, even a positive finding in favour of the second condition does not guarantee a change in the legal position of the individual. This is due to the Court’s interpretative freedom in relation to the scope of the provision considered next. C. The interpretation of directly effective provisions as the final stage of gatekeeping Having found a provision to be directly effective, the Court’s gatekeeping culminates with the interpretation of the scope of a particular provision,235 save for cases where the Court directly turns to interpretation without establishing direct effect first.236 At this final stage, the Court ultimately secures or denies the impact of an international agreement on a given individual’s legal position as the finding of direct effect is fruitless unless the relevant provision receives a ‘favourable interpretation’ by the Court.237 The gatekeeping at this stage also ranges from rather sparing to strict attitudes. The Court relies on international law, in particular Article 31 of the Vienna Convention, to interpret the provisions of the agreement ‘in their context and in the light of its object and purpose’.238 One of the main issues pertinent to the gatekeeping through interpretation is whether the provisions of international agreements resembling those found in the TFEU (previously EC Treaty) should be granted a similar interpretation. Despite establishing that provisions resembling those found in the EC Treaty should not necessarily be accorded with the same meaning,239 the Court, nonetheless, accorded similar interpretations to international agreements in a range of cases.240 The comparative analysis of the context of the TFEU and other agreements had ‘considerable importance’ in the Court’s findings.241 On the one hand, comparing international agreements to the EC Treaty/TFEU might appear to be devoid of controversy as it merely represents the outcome of considering the objectives of each Treaty.242 On the other hand, Article 31 of the Vienna Convention ‘itself does not permit … comparative analysis’ between different agreements, and it is the context and the object and purpose of each treaty that should be separately analysed.243 It can be argued, that the comparison with the TFEU can indeed cast a shadow over the Court’s perception of the international agreement. The most straightforward transposition of internal interpretations to provisions of international agreements were found in instances of direct references to the EC Treaty. For instance, the Yaoundé Convention, in its Article 2, directly referenced then Article 13 EC on the abolition of charges having equivalent effect leading to a uniform interpretation.244 Opting for a similar interpretation entails an ‘extension’ of the EU legal order which can be justified for agreements with an element of EU acquis transposition.245 However, having an element of EU acquis transposition itself does not guarantee similar interpretation. Neither does a promise of association always secure a homogenous interpretation. It is, therefore, difficult to deduce a clear and consistent pattern in the Court’s case law.246 Some suggest that a successful outcome would often depend on the ‘amenability’ of the ‘scope, historical and legal context’ of the relevant agreement, while the unsuccessful outcome is tied to the provisions themselves, commonly on the rights of third country nationals to reside on the territory of the EU.247 For the seeming inconsistency between the cases with transposition of internal interpretation and cases denying the latter, a justification is found in the context of each agreement.248 The focus of the interpretation, however, can hover from the context of the agreement to the provision itself, or the Court could put more weight on one than the other. Lack of consistency is seen most acutely in relation to association agreements with candidate countries: in some cases, the accession prospects were factors favouring uniform interpretation, while in others, the accession factor was not given sufficient weight to influence the outcome of the interpretation.249 In the latter category of cases, the directly effective provisions on freedom of establishment and free movement of workers in certain Europe Agreements were not accorded a uniform interpretation with the the equivalent TFEU provisions, as a result of which the Member States could impose their own rules on entry, stay, and establishment, including their immigration rules.250 Here, the Court chose to limit ‘its own creative reading of the association’ by granting less significance to the political and historical context of the relevant agreements and the parties’ accession intentions.251 Some viewed the Court’s approach as merely respecting the differences between the Europe Agreements and the EC Treaty and avoiding a potential backlash against the accession strategy.252 This in itself manifests the making of a political choice. It is expressed in particular in the refusal to advance the enlargement agenda and in the tribute to the Member States’ concerns about migration.253 Ultimately, the finding of direct effect here turned into ‘a hollow victory’.254 The comparative approach is not restricted to the TFEU. In El-Yassini, the EC–Morocco Agreement was compared to the EC–Turkey Association Agreement: since the former did not provide for a prospect of association, the scope of its non-discrimination provision was narrowly interpreted.255 Such comparison might be appropriate for agreements which were concluded as part of a regional approach, as seen in Kolpak, where the Europe Agreement with Poland was compared to the Europe Agreement with Slovakia in order to draw out their similarities ‘to their objectives or the context in which they were adopted’.256 There was, however, no such regional context in Simutenkov, where the Court transferred its interpretation of the non-discrimination provision from the Europe Agreement with Poland to the EU–Russia PCA without unpicking the distinctions between the context of the two agreements,257 thus diluting the ‘differentiation’ the conclusion of distinct agreements was meant to signify.258 Moreover, despite the lack of comparison between the PCA and the EC Treaty,259 the Court also relied on Bosman to transpose its internal interpretation of the principle of non-discrimination to the PCA, without clearly demarcating the limits of such transposition.260 Ultimately, the Court is not an unequivocal gate-opener: despite the far-reaching implications of direct applicability and direct effect, the application of the conditions of direct effect provides a wide margin for limiting the permeability of EU law towards international agreements. Each case of denying direct effect or interpreting the scope of a directly effective provision to exclude a particular right is an example of shutting the proverbial gates. Direct effect is used, in particular, as a ‘shield’ to protect EU institutions and legislation.261 Indeed, the application of direct effect to international agreements has at times been assessed as ‘less generous’262 or ‘more reserved’263 in comparison with internal EU legal acts. The Court’s generosity or the lack thereof has been linked to the nature of the action whereby challenges against Member States’s actions have a higher likelihood of success than those against the EU.264 As far as individual reliance on provisions of international agreements is concerned, it has been noted that the successful cases even though constituting ‘a broad category’, inclusive of association agreements and trade agreements, are nonetheless the exception.265 In any case, it appears that other EU institutions do perceive the Court’s approach to be generous as they are keen to challenge the status quo. V. Sidelining the court as the gatekeeper?: Alternative gatekeeping techniques As noted earlier, the Court carved out its role on determining the effects of international agreements in the legal vacuum created by the lack of an agreement between the parties: only if the effect of the provisions is not settled by the parties would it ‘fall’ to the Court’s jurisdiction, as set out in Kupferberg.266 If the parties have come to a consensus on the effects of the agreement, the Court will be required to give it full effect by enforcing the intentions of the parties. At this stage the inquiry is within the domain of international law.267 If such a consensus is absent, the inquiry shifts to the level of European law as discussed above. The main question to ask here is whether the treaty-making institutions can pre-empt a judicial inquiry into direct effect. Both scenarios are considered in turn below. A. The agreement between the parties The agreement between the parties is viewed at times as one of the conditions for direct effect.268 Some qualifications are in order here. If the agreement is non-existent, then there can be no talk of it being a condition. If there is an agreement, then it can take two forms—positive or negative, both creating implications for the Court’s role. If the agreement positively sets out direct effect, the Court is precluded from finding otherwise. Alternatively, if the agreement excludes direct effect, the Court cannot find to the contrary. Cheyne refers to this as ‘the pre-emptive rights of the executive institutions to determine whether [the agreement’s] provisions should be given [direct effect]’.269 As noted earlier, most commonly the issue of effects does not occupy an important role in international negotiations. Furthermore, in some cases open opposition to any such prescriptions has been recorded. For instance any such possibility in relation to WTO law was firmly rejected by its members.270 Positive prescriptions of direct effect are extremely rare in practice. An example of such practice can be found in the agreement establishing the European Common Aviation Area which, in its Article 15, obliges its parties to ‘ensure that the rights which devolve from this agreement may be invoked before national courts’.271 This positive setting of direct effect can be linked, perhaps, to the purpose of the agreement aiming to create a single aviation market between the EU and certain European states, non-members of the EU. Providing for direct effect in the agreement itself enables challenges by individual travellers and members of the aviation industry which could speed up the removal of barriers to movement. As for the negative exclusion of direct effect, in departure from the established view that the effects of the agreements are not usually part of the negotiations, examples of negative preclusion can be found. Until recently, they were rare. An isolated example can be found in the 1990s in the Agreement on international humane trappings standards between the EU, Canada, and Russia.272 In particular, it established that ‘[the] Agreement is not self-executing’ and requires implementation by each party.273 It is not clear whether the term ‘self-executing’ is used here to refer to direct applicability or direct effect, however, the reference to the need for implementing measures ensures that the conditions for direct effect will not be satisfied.274 More recently, a trend of a more systematic exclusion of direct effect appears to be emerging. This negative preclusion can take two forms either by denying direct effect to the entire agreement or to its specific provisions. The first form is more extreme, examples of which, with a variety of formulations, can be found in recent trade agreements with Colombia and Peru,275 the somewhat ill-fated free trade agreement with Singapore,276 an association agreement with Central America,277 and the Comprehensive Economic and Trade Agreement with Canada (CETA).278 None of these agreements has an ‘integrationist potential’.279 Nonetheless, as such agreements have been found to be directly effective in the past,280 the effect of such provisions is to remove any possibility of making such a finding by the Court of Justice. Excluding direct effect for an entire agreement would by implication extend also to the decision of the bodies established thereunder. It can be argued that in bilateral agreements denying direct effect in the agreement itself is more warranted by the EU, in particular its treaty making institutions and its Member States, since in the case of the other party its ordinary constitutional arrangements would apply in any case. The same conclusion, however, does not stand for multilateral agreements: the other parties might also be interested in restricting the obligations to intergovernmental level without creating judicially enforceable rights for private parties. In terms of the variations in formulations, it should be noted that the agreement with Colombia and Peru, as well as the EU–Singapore trade agreement and CETA clarify that no rights or obligations are created for individuals beyond those created between parties under international law. This secures an outcome whereby the Court of Justice can no longer rule on the possible direct effect of such agreements. However, this does not mean that the Court’s interpretative function is surrendered. Presumably, the Court can still rely on indirect effect to interpret any relevant legislation in light of the international commitments. Unlike the agreements noted above, the Association Agreement with the Central American countries goes further to bar the justiciability of its provisions by private parties ‘unless otherwise provided in that party’s domestic legislation’. Such a formulation, besides sidelining the Court, ensures that control over the matter is firmly with the legislative institutions of the EU should they opt for conferring certain rights to individuals. Examples of the second type of exclusionary practice limited to specific provisions are found in the Association Agreements with Ukraine, Georgia, and Moldova respectively.281 The relevant provisions relate to the dispute settlement mechanisms, as well as the schedules of commitments in service laid down in annexes to the agreement. The exclusion of direct effect here is linked to WTO law-related content: this includes the dispute settlement mechanism based on the WTO model (with some adjustments) with a jurisdiction to rule on WTO-related obligations, and the schedules for commitments in service areas based on The General Agreement on Trade in Services (GATS).282 Precluding direct effect for these provisions is due to substantive incorporation of WTO law guaranteeing that the lack of direct effect of WTO law cannot be bypassed.283 As such, this practice does not aim to sideline the Court. On the contrary, it can be suggested that it confirms the Court’s jurisprudence. At the same time, while the EU and the relevant third country have agreed to exclude the direct effect of specific provisions, by implication this would entail that other provisions of the agreement—those meeting the relevant criteria—are capable of having direct effect. As noted earlier, at this stage a transition is made from the domain of international law to EU law where the Court steps in with its ‘monopolized’ role to determine the effects of the other provisions. But what if the Council decision adopting such agreements explicitly excludes the direct effect for the entire agreement? B. Internal institutional challenges to judicial gatekeeping Over a decade ago, doubts have been expressed over the tenability of the status quo predicting a certain tension over the Court’s power, particularly on the part of the legislature.284 The current challenges against the Court’s lead role is manifested in the Council’s practice precluding direct effect in its decisions on the conclusion of certain international treaties. Such decisions have been viewed as one ‘form’ or ‘way’ of excluding direct effect.285 Examples include the decisions concluding the Association Agreements between the EU and Ukraine, Georgia, and Moldova respectively which rule out direct effect for the entire agreement.286 This section aims to enquire into whether such decisions can indeed be seen as a means of excluding direct effect by aborting the Court’s jurisdiction. As noted earlier, international law is characterized by it permissiveness to the domestic legal effects of international agreements; it is the prerogative of each state to fulfil its obligations according to its interests.287 This implies that a state, or in this case the EU, would establish its own rules on the implementation and the effects of the agreement. The Council decisions concluding the agreement cannot be equated with constitutional rules, however. Neither are they measures implementing the agreement.288 What, then, is their effect? Can an argument be made under EU law or international law for giving a decisive weight to the Council’s decision? (i) The Council decisions in view of the Kupferberg formula? While the Kupferberg formula cited above made no mention of the Council’s decision adopting an international treaty, on one occasion the matter did come to the Court’s attention. In 1994, the Council’s Decision adopting the WTO Agreement included the following statement in its preamble: the agreement is ‘not susceptible to being directly invoked in [EU] or Member States’.289 Initial assessments viewed it as indecisive in setting the effects of the agreement on individuals or EU institutions and Member States, which was a matter for the Court to decide.290 There were also predictions as to the decision’s potential to lead to a more cautious approach by the Court.291 The issue soon came to the Court’s attention in Portugal v Council.292 Advocate General Saggio in his opinion was unequivocal against the Council’s decision having an effect on the Court’s competence to rule on direct effect: it was merely a ‘policy statement’.293 His argument was twofold. The first stems from international law. Invoking Articles 31–33 on the rules of interpretation of the 1969 Vienna Convention on the Law of Treaties, the Advocate General ruled against a unilateral institutional declaration being seen as decisive at the level of international law in terms of limiting the direct effect of the agreements.294 Rather than according primacy to such declaration, the issue of direct effect should be resolved through the interpretation of ‘the objective content of the textual provisions of the agreement’ in accordance with the rules of customary international law,295 including Article 31 of the Vienna Convention binding the EU.296 The latter sets the general rule of interpretation according to which treaties should be interpreted ‘in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. The second argument derives from the EU legal order. The Council cannot by secondary legislation limit the Court’s (or national courts’) competence to establish the effects of international agreements: ‘a unilateral interpretation of the agreement made in the context of an internal adoption procedure cannot—outside the system of reservations—limit the effects of the agreement itself’.297 The Court in turn acknowledged that it was the parties’ prerogative to establish the means of implementation of the agreement, however in the absence of such an accord the parties are at liberty to choose the means of implementation.298 Instead of clarifying the status of the Council’s decision, the Court then proceeded with setting its arguments for denying direct effect to the WTO agreement and, only having established the latter, did it refer to the Council’s decision noting that its own finding ‘corresponded’ to the decision.299 First, it is clear that the Court finds the decision of relevance.300 Secondly, the Court only refers to it post factum, and not as a starting premise for its findings. Thirdly, there are no suggestions that the Court views the Decision as a measure implementing the agreement. In fact, the CJEU does not specify the capacity in which the Council’s decision was taken into account. The Court’s approach has been interpreted differently. According to Fabri, the Council’s decision did not play a significant part in the outcome of the case: the negative preferences of other institutions cannot bind the Court in a manner that primary law would.301 Others saw the decision as having impacted the Court’s approach as evidence of the intention of the EU as a party.302 Can a negative pronouncement in a Council decision be equated with the expression of the intention of the EU as a party, however? It is safe to say that the preamble of the decision demonstrates the views of both the Council and the Commission,303 ‘the treaty-making institutions’.304 It might be suggested that in dualist countries the requirement to implement the agreement eliminates the need for any such proclamations. In monist countries, the stance of the executive can play a part in the judicial finding of direct effect. Paying homage to the views of the executive, in deciding on the matter of direct effect, forms part of the judicial inquiry into the intentions of parties in the Netherlands, for instance.305 The Council decisions can also be compared to the declarations by the US President and two-thirds of the Senate on the non-self-executing nature of an international agreement. This practice has been criticized for its ‘neo-isolationist preferences’ depriving the judiciary of the opportunity of ‘contributing to the evolution of international practice’.306 Not only the status of such declarations in the US legal order is uncertain, both in courts and in scholarship,307 the comparison itself is somewhat problematic. First, the US declarations as to non-self-execution are about the need for further implementing measures in the US domestic legal order: individuals can derive rights stemming from the international treaty only if the Congress passes a legislative measure.308 In contradistinction, the Council’s decision merely excludes direct effect which does not necessarily imply that an implementing measure is required. Another consideration should also be born in mind when embarking upon such comparison. The US President and the Senate are the actors behind the treaty ratification: the US President signs it after the consent of two-thirds of the Senate. The other legislative chamber of the Congress, the House of Representatives, does not take part in this process. In the context of the EU, a principled objection can emerge towards treating the Council’s decision as reflective of the EU’s intention as a party for the reason of the involvement of the European Parliament. The latter, as a co-legislator with the Council, has seen an expansion of its involvement in the process of conclusion of international agreements to the extent of influencing their content or even rejecting them.309 As a party, the EU concludes agreements through the process established in Article 218 TFEU requiring the consent or consultation of the European Parliament.310 When an agreement within its provisions excludes direct effect for its entirety or for few provisions, the Parliament’s participation would thus make the intention of the EU as a party complete. Would a similar conclusion be applicable in relation to those cases where direct effect is excluded in a Council decision only? What is the evidence to suggest that the Parliament would support such restrictive practices? Support of this nature could have been found in case law where the Parliament became involved to argue against direct effect. However, only a handful of such examples can be found,311 and the case law offers no systematic conclusion on the Parliament’s position on the issue of direct effect more generally. On the other hand, a compelling argument can be made to suggest that the Parliament, in effect, acquiesces in this type of exclusionary practice. The preambles of parliamentary resolutions expressing consent to the conclusion of international agreements, refer first and foremost to the respective draft Council decisions. The Parliament, therefore, tacitly endorses the preferences of the Commission and the Council. This conformism, however, might be rooted in the Parliament’s own disapproval of the openness to international law expressed on occasion. In its 1997 resolution on the relationship between international law, Union law and the constitutional law of the Member States, the Parliament called for a provision in the EC Treaty setting out the process of the transposition of international law into the EC legal order.312 The resolution, thus, focused on the issue of direct applicability. Even though the Parliament appears not to have challenged the Court’s leading role, it nevertheless is discontent with the Court’s ‘solutions’: EU law is more ‘permeable’ to international law than domestic legal orders of the Member States.313 The Parliament evidently favoured a non-automatic transposition of agreements which should be directly applicable only if it ‘has been declared applicable by an internal legal act of the [EU] or after its substance has been transposed into [EU] legislation’.314 Even though these objections were aimed at direct applicability, clearly the Parliament would be interested in moulding the provisions which could potentially be invoked by individuals. This can be linked to the ‘democratic’ argument where such issues should be decided by representative institutions instead of leaving them to the fate of sporadic developments through individual claims.315 The Council decision, however, does not suggest any meaningful return of control over this issue of the conferral of rights to the Parliament as it does not necessarily imply any further legislative measures. This can be juxtaposed with the US practice, where the relevant declaration passes control to Congress.316 In the EU, if the Council decision is to be upheld on its effect of barring direct effect, it should rather be viewed as returning control to Member States. If one is to view the Council’s decision as representative of the EU’s intentions as a party, it does not necessarily lead to an argument that it should prevent the Court from ruling on the matter. As prescribed by the Kupferberg formula, the Court’s jurisdiction to determine the effects of the agreement is secondary only to the agreement between the parties. The Court made no such reservations for the acts of the Council concluding the agreement, neither did it consider it necessary to give a clear weight to the Council decision in Portugal v Council, as noted above. Any suggestions of treating the Council’s decision as definitive on the matter of direct effect will require the Court to revisit its settled jurisprudence and to make a new constitutional argument to determine its own role and potential limitations to it. Portugal v Council, however, would suggest that by mentioning the Council decision within its analysis of the first condition of direct effect the Court indeed viewed it as relevant for establishing the intentions of the parties. Crucially, it was merely used as one of the factors relevant for establishing the intentions of the parties. It can also be questioned whether any further weight could be granted to the relevant decisions under the provisions of the Vienna Convention on the Law of Treaties. (ii) Council decisions as an interpretative tool under the Vienna Convention on the Law of Treaties? At the outset, it should be noted that Council decisions excluding direct effect would not qualify for a so called ‘reservation’ under the Vienna Convention on the Law of Treaties.317 The latter defines reservations as ‘a unilateral statement, however phrased or named, made by a State, when signing, ratifying, accepting, approving, or acceding to a treaty, whereby it purports to exclude or to modify the legal effect of certain provisions of the treaty in their application to that State’.318 The stipulations on preclusion of direct effect in the Council decisions do not qualify for this definition as they are not aimed at excluding or otherwise modifying the legal effect of a particular agreement. Rather, they make a comprehensive pronouncement on the entire agreement in relation to domestic legal effects. So, what capacity can be accorded to such decisions? According to Verwey, even though non-decisive per se, the Council’s decision can be taken as an ‘additional source’ for the purposes of interpreting a provision in the agreement.319 But what type of ‘additional source’ would it be? Returning to Article 31 of the 1969 Vienna Convention, this allows the context of the agreement to be taken into account when interpreting the treaty. According to Article 31(2)(b), the context, in addition to the text, the preamble, and annexes, can include inter alia ‘any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty’. While the Vienna Convention itself does not clarify what the ‘conclusion’ of the agreement precisely refers to, it is accepted that it would include instruments ratifying or expressing the parties’ ‘consent to be bound’.320 The Council’s decision is indeed an instrument expressing the EU’s will to be bound by a particular agreement. That leaves the issue of ‘acceptance’. It is suggested that ‘the need for acceptance’ would distinguish the instrument envisaged in the provision above from ‘unilateral interpretative declarations made by a state when signing or ratifying’ the agreement.321 The acceptance by the other party is a condition,322 requiring the opposite parties at least to have ‘acquiesced in the instrument’.323 Even though the acceptance can be ‘informal’ or ‘tacit’, any party suggesting a particular interpretation on the basis of such extrinsic instrument will be required to demonstrate that other parties have accepted the position declared in the instrument.324 Verwey suggests that an explicit acceptance or rejection of the Council’s statements by the other party is not necessarily referring to Bresciani, where reciprocity, that is reciprocal recognition of direct effect, had no bearing on the Court’s finding of direct effect.325 However, these are two distinct matters. In assessing reciprocity as a factor indicative of the nature of the agreement, the Court is not affected by denial or recognition of direct effect by other parties (with the exception of the WTO case law). But when evaluating whether the Council’s decision precluding direct effect for an entire agreement is part of the context of the agreement under Article 31 of the Vienna Convention, the Court should require some form of acceptance of this exclusionary effect of the Council’s decision. If the Court is to view the Council’s decision as ‘any instrument’ comprising the context of the agreement under Article 31(2), there should be some evidence of some form of acceptance by other parties. Furthermore, there is no hierarchy among the sources used for interpretation: the order ‘appears to be that of logic, proceeding from the intrinsic to the extrinsic, from the immediate to the remote’.326 ‘Any instrument’ referred to above is ‘extrinsic to the treaty’,327 suggesting that it cannot per se set aside the text of the Treaty which may well indicate in favour of direct effect. Ultimately, under Article 31(1) of the Vienna Convention the interpretation of the agreement is undertaken on the basis of textual and teleological approaches in addition to the subjective quest for the intentions of the parties.328 (iii) Taking stock The discussion above is not a matter of historical debate, but it is pertinent due to a number of Council decisions excluding direct effect in relation to a range of agreements, including free trade agreements and association agreements with neighbouring countries, as noted above. For those agreements which incorporate WTO law to a certain extent, precluding direct effect in the Council decision can similarly be interpreted as an insurance policy for not bypassing the established practice of excluding direct effect for WTO law. This would not appear to create any issues for those agreements which themselves preclude direct effect in their entirety. However, where direct effect is precluded for specific provisions only, or which contain no provisions about direct effect, the Council’s decision can be viewed as problematic. As discussed earlier, the decisions concluding the Association Agreements between the EU and Ukraine, Georgia, and Moldova, respectively, rule out direct effect for the entire agreement, while the agreements themselves rule out direct effect for specific provisions only. Unlike the Council Decision on the WTO Agreement where the relevant limitation was included in the preamble, in the above cases it is found among the main provisions of the decision. This might perhaps lead to a suggestion that a more significant weight should be attached to the latter as opposed to a preambular statement.329 On the other hand, in his forceful position in Portugal v Council, Advocate General Saggio referred to the inability of the secondary legislation to exclude the Court’s jurisdiction generally, without making a distinction between the relevant pronouncement being made in a preamble or in the text of the act. Furthermore, irrespective of the location of such limitations, the decisions are problematic for the following reasons. First, as noted above, these agreements themselves preclude direct effect for specific provisions implying that other provisions are capable of direct effect. In this context, it can even be suggested that the Council decisions are against the intentions of ‘the parties’. Secondly, such unilateral declarations by the Council, outside of WTO law, are at odds with past jurisprudence, especially if the agreement contains provisions identical to those which were found to be directly effective in the past in cases of comparable bilateral agreements.330 Van der Loo and others note in this connection Article 17 in the EU–Ukraine Association Agreement on the principle of non-discrimination in relation to workers.331 Similar provisions in Association Agreements with EU neighbouring countries in the south were found to be capable of direct effect in the past.332 In addition, this provision is identical to the non-discrimination provision in the EU–Russia PCA in Simutenkov.333 As discussed earlier, the Court in its analysis of the condition on the nature and the logic of the agreement embraces a comparative approach. The PCA with Russia offers much narrower integration prospects vis-à-vis the Association Agreement with Ukraine.334 The Ukrainian agreement is ‘the most advanced agreement of its kind ever negotiated by the [EU]’,335 offering the closest possible links to the EU falling short of membership, which has also served as a template for the respective agreements with Georgia and Moldova. In this light, the Council’s decision indeed appears to signal a clear preference for a departure from previous case law. It might be suggested that it is perhaps due to the integrationist agenda of such agreements that the Council felt compelled to restrict their direct effect. Ukraine, Georgia, and Moldova entertain European aspirations and view these agreements as a stepping stone on their future membership path. By denying direct effect to these Agreements, the Council sends a clear signal that they remain at an intergovernmental level falling short of creating a union of people. But this cannot be the sole logic driving the Council to exclude direct effect for EU bilateral agreements. A similar decision precluding direct effect was adopted for the EU–Korea free trade agreement.336 The latter cannot be compared to the Association Agreements noted above and contains no similar integrationist agenda. It might, therefore, indicate a general trend of restricting direct effect for agreements which were found to be directly effective in the past. The EU–Korea agreement, for instance, contains standstill provisions which could be directly effective as seen in earlier jurisprudence.337 The Council decisions, therefore, signal a manifest preference for a departure from the Court’s practice. It might only be a matter of time before the CJEU is called to clarify the status of such pronouncements by the Council. In particular, a clarification will need to be made whether excluding direct effect in a provision of the decision gives the latter more legal weight than a preambular statement, as in the WTO Decision, so as to exclude the Court’s jurisdiction. Alternatively, if the Court maintains the Kupferberg formula, it should interpret international treaties by giving ordinary meaning to their terms which might indicate in favour of direct effect according to the conditions set by the Court. If the Council is to insist that its decision should be viewed as part of the context of the agreement, then some form of ‘acceptance’ on behalf of other parties should be demonstrated. VI. Conclusion The Court of Justice has meticulously carved outs its role of gatekeeper, which is not one of an unequivocal ‘door opener’. The Haegeman ruling establishing the direct applicability of international treaties signalled an almost automatic openness to international law. It laid the foundations for the next opening act which is the general finding that intentional agreements are capable of having direct effect. Paradoxically, these developments were affected more by internal than external considerations. Apart from affirming the crucial role of the Court of Justice in EU international relations vis-à-vis other institutions, they also propelled the EU’s external actorness against the Member States. The Court, having placed itself at the core of the direct effect exercise, became a flexible gatekeeper. The manner in which the conditions of direct effect have been set and applied allowed for ample scope for both generous and ungenerous findings. Ridden with inconsistency, the setting and the application of the conditions of direct effect permitted the Court to limit the extent to which international agreements could empower individuals or affect the legality of secondary EU legislation, in particular when the external or internal interests of the EU institutions were at stake. The CJEU’s flexibility has been carried forward to the last phase of gatekeeping through the interpretation of the provisions of the agreement ultimately determining whether the individual would find the relief sought. Here, additional concerns—those of Member States—can also be taken into account. The reign of the Court is being challenged though. While challenges to direct applicability expressed by the European Parliament in its 1997 resolution did not lead to any consequences, the Council and the Commission are willing to mitigate the openness created by direct applicability through limitations to direct effect. The dual practice of excluding direct effect either for the entire agreement or for specific provisions reveal the ease with which the Court can be sidelined. The situation is less clear with the Council’s attempt to exclude direct effect for entire agreements in its decisions on their adoption. Despite mentioning the exclusionary preambular pronouncement in the Council’s 1994 decision in Portugal v Council in support of its rejection of the direct effect of the WTO agreement, the Court failed to clarify the weight it attached to the latter. In the recent decisions, the Council has made a stronger statement by relocating the exclusion of direct effect to the text of the decision from the preamble. If the issue of direct effect of any of these agreements is ever raised, the Court will be called upon to revisit its gatekeeper role and the possible concessions to it. The Court will have to clarify whether the decisions can be viewed as expressing the intention of the EU as a party, and, if so, whether it would be capable of setting the Court’s jurisdiction aside. Unless the Kupferberg formula is revised, Portugal v Council suggests that the Court would interpret the international agreement, ie would preserve its jurisdiction over the matter of direct effect, but would, nonetheless, take into account the Council’s decision as part of its analysis of the first condition of direct effect. The status of the pronouncements by the Council can also be clarified under international law. It is only a matter of time before this issue is raised in front of the Court. I would like to thank Jeffrey Kenner, Dominic McGoldrick, Daria Davitti, Marko Milanovic, Mary Footer and Or Bassok for their helpful comments on earlier versions of this article. Footnotes 1 Case C-402/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-06351; Opinion 2/13, ECLI:EU:C:2014:2454. 2 See for instance Jan Klabbers, ‘Völkerrechtsfreundlich? International Law and the Union Legal Order’, in Panos Koutrakos (ed.), European Foreign Policy: Legal and Political Perspectives (Cheltenham: Edward Elgar 2011) 95–114; Jan Klabbers, ‘Straddling the Fence: The EU and International Law’ in Anthony Arnull and Damian Chalmers (eds), The Oxford Handbook of European Union Law (Oxford: Oxford University Press, 2017) 52–71, 62–3; Gráinne de Búrca, ‘The European Court of Justice and the International Legal Order after Kadi’ (2010) 51 Harvard International Law Journal, 1. 3 Case C-308/06 Intertanko [2008] ECR I-4057; Case C-366/10 Air Transport Association of America and Others (ATAA) [ECR] 2011 I-13755. 4 The term ‘gatekeeper’ is borrowed form literature; Jan Klabbers, ‘International Law in Community Law: the Law and Politics of Direct Effect’ (2001) 21 Yearbook of European Law, 263, 296; Francis Snyder, ‘The Gatekeepers: The European Courts and WTO Law’ (2003) 40 CML Rev, 313; Marise Cremona, ‘External Relations and External Competence of the European Union: The Emergence of an Integrated Policy’ in Paul Craig and Gráinne de Búrca (eds), The Evolution of EU Law (2nd edn, Oxford: Oxford University Press, 2011) 215–68, 234. 5 Robert McCorquodale, ‘The Individual and The International Legal System’ in Malcolm D Evans (ed.), International Law (4th edn, Oxford: Oxford University Press, 2014) 280–305. 6 Sean D Murphy, ‘Does International Law Obligate States to Open Their National Courts to Persons for the Invocation of Treaty Norms That Protect or Benefit Persons?’ in David Sloss (ed.), The Role of Domestic Courts in Treaty Enforcement: A Comparative Study (Cambridge: Cambridge University Press, 2009) 61–119, 118; André Nollkaemper, ‘The Effects of Treaties in Domestic Law’ in Chrtistian J Tams and others (eds), Research Handbook on the Law of Treaties (Cheltenham: Edward Elgar, 2014) 123–50, 145. 7 Current Art. 216(2) TFEU [2012] OJ C326/47. 8 Case 181/73 Haegeman (Haegeman II) [1974] ECR 449. 9 Eeckhout distinguishes between ‘internal’ and ‘external’ direct effect; Piet Eeckhout, EU External Relations Law (2nd edn, Oxford: Oxford University Press, 2011) 229–330. 10 Case C-26/62 Van Gend en Loos [1963] ECR 6. 11 Bruno De Witte, ‘Direct Effect, Primacy, and the Nature of the Legal Order’ in Paul Craig and Gráinne de Búrca (eds), The Evolution of EU Law (2nd edn, Oxford: Oxford University Press, 2011) 323–62, 324. 12 Eeckhout (n 9), 324. 13 Bruno de Witte, ‘The Continuous Significance of Van Gend en Loos’ in Miguel P. Maduro and Loïc Azoulai (eds), The Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford: Hart, 2010) 9–15, 10. 14 If the international agreement has been implemented through regulations or directives, then the internal rules on direct effect will be relevant; Dominic McGoldrick, International Relations Law of the European Union (Harlow: Longman, 1997) 125. 15 Cases 21-24/72 International Fruit [1972] ECR 1219; Case C-280/93 Germany v Council [1994] ECR I-4973; Case C-149/96 Portugal v Council [1999] ECR I-8395; Case C-377/02 Van Parys v BIRB [2005] ECR I-1465; Joined Cases C-300/98 and C-392/98 Dior [2000] ECR I-11307. 16 Mayer as cited in Bart Van Vooren and Ramses A. Wessel, EU External Relations Law (Cambridge: Cambridge University Press, 2014) 218. 17 See nn 287, 337 below. 18Portugal v Council (n 15). 19 See for instance, Francesca Martines, ‘Direct Effect of International Agreements of the European Union’ (2014) 25 The European Journal of International Law, 129; Stefan A Riesenfeld, ‘The Doctrine of Self-executing Treaties and Community Law’ (1973) 67 American Journal of International Law, 504. 20 Carlos M. Vázquez, ‘The Four Doctrines of Self-Executing Treaties’ (1995) 89 American Journal of International Law 695; Carlos M. Vázquez, ‘Treaties as Law of the Land: the Supremacy Clause and the Judicial Enforcement of Treaties’ (2008) 122 Harvard Law Review 600; David Sloss, ‘United States’ in Sloss, (n 6), 504–54. 21 J A Winter, ‘Direct Applicability and Direct Effect: Two Distinct and Different Concepts in Community Law’ (1972) 9 CML Rev 425, 425. 22 See for instance Winter (n 21); Pierre Pescatore, ‘The Doctrine of “Direct Effect”: An Infant Disease of Community Law’ (1983) 8 EL Rev, 135; David Edward, ‘Direct Effect, The Separation of Powers and the Judicial Enforcement of Obligations’ in Scritti in onore di Guiseppe Federico Mancini (Essays in honour of Guiseppe Federico Mancini) (Milan: Dott. A. Giuffrè Editore, 1998), Volume II, 423–443; Ilona Cheyne, ‘International Agreements and the European Community legal system’ (1994) EL Rev, 581. 23 Winter (n 21), 427, 438. 24 Cheyne (n 22), 585–8. Elsewhere Cheyne analyses the Haegeman formula in relation to the effects of international agreements; Ilona Cheyne, ‘Haegeman, Demirel and their Progeny’ in Alan Dashwood and Cristophe Hillion (eds), The General Law of EC External Relations (London: Sweet and Maxwell, 2000) 20–41, 26. 25 Koen Lenaerts, ‘Direct Applicability and Direct Effect of International Law in the EU Legal Order’ in Inge Govaere and others (eds), The European Union in the World: Essays in Honour of Marc Marescegau (Leiden: Brill; Nijhoff 2013) 45–64, 46. 26 Case 12/86 Demirel [1987] ECR 3719, para. 14. 27 According to Eeckhout, the definition ‘does not put in much relief the distinction between the structure and nature of the agreement as such, and the conditions for direct effect of specific provisions’, Eeckhout (n 9), 337. 28 See for instance International Fruit (n 15); Portugal v Council (n 15); Intertanko (n 3); ATAA (n 3). 29 Opinion of AG Sharpston, Case C-240/09 Lesoochranárske zoskupenie VLK v Ministerstvo životného prostredia Slovenskej republiky (Brown Bear) [2011] ECR I–1255, para. 38, footnote 19. 30 Edward (n 22), 426. 31 Szilárd Gáspár-Szilágyi, ‘EU International Agreements through a US Lens: Different Methods of Interpretations, Tests and the Issue of Rights’ (2014) 39 EL Rev, 601, 602, 607. 32 Jean Groux and Philippe Manin, The European Communities in the International Order (European Perspectives) (European Commission, 1985) 118. 33 Holdgaard uses ‘direct invocability’ in a wider sense to incorporate not only direct effect as the ability of the individual to rely on a particular provision, but also other legal effects of international agreements; Rass Holdgaard, External Relations of the European Community: Legal Reasoning and Legal Discourses (Alphen aan den Rijn: Kluwer Law International, 2008) 244. 34 The justiciability of the norm—‘the capacity of a norm of Union law to be applied in domestic court proceedings’, should be distinguished from a narrower understanding of conferral of individual rights; De Witte (n 11), 323; Sasha Prechal, ‘Does Direct Effect Still Matter?’ (2000) 37 CML Rev, 1047, 1050; Sasha Prechal, ‘Direct Effect, Indirect Effect, Supremacy and the Evolving Constitution of the European Union’ in Catherine Barnard (ed), The Fundamentals of EU Law Revisited: Assessing the Impact of the Constitutional Debate (Oxford: Oxford University Press, 2007) 35–69, 37. 35 Edward (n 22), 426; Van Vooren and Wessel (n 16), 229; by analogy ‘direct application’ in Kees J. Kuilwijk, The European Court of Justice and the GATT Dilemma: Public Interest versus Individual Rights? (Nexed Editions, 1996) 82–103. The term ‘direct applicability’ is also used in scholarship on international law to denote the same concept: James Crawford, Brownlie’s Principles of Public International Law (8th edn, Oxford: Oxford University Press, 2012), 58; Nollkaemper (n 6), 138–42. 36 Art. 288(1) TFEU; Paul Craig and Gráinne de Búrca, EU Law: Text, Cases and Materials (6th edn, Oxford: Oxford University Press, 2015) 107; Robert Schütze, European Union Law (Cambridge: Cambridge University Press, 2015) 91; Damian Chalmers, Gareth Davies and Giorgio Monti, European Union Law (3rd edn, Cambridge: Cambridge University Press, 2014) 112. 37 Case C-41/74 Van Duyn [1974] ECR 01337, para. 12; Case C-51/76 Nederlandse Ondernemingen [1977] ECR 00113, para. 21; Case C-148/78 Ratti [1979] ECR 01629, para. 19. 38 For instance in International Fruit, the Court found that the EC was bound by the GATT even without being a party to the latter; International Fruit (n 15), paras 10–18. 39 Klabbers (n 4), 294–5; Armin von Bogdandy, ‘Pluralism, Direct Effect and the Ultimate Say: On the Relationship between International and Domestic Constitutional Law’ (2008) International Journal of Constitutional Law 397, 400; Van Vooren and Wessel (n 16), 222; Panos Koutrakos, EU International Relations Law (2nd edn, Oxford: Hart, 2015) 257. 40 See for instance constitutionalism and pluralism in Ramses A. Wessel, ‘Reconsidering the Relationship Between International and EU Law: Towards a Content-Based Approach?’ in Enzo Cannizzaro, Paolo Palchetti, and Ramses A. Wessel (eds), International Law as Law of the European Union (Leiden: Martinus Nijhoff Publishers 2012) 7–33; Von Bogdandy (n 39); pluralism in Francis Snyder, The EU, the WTO and China: Legal Pluralism and International Trade Regulation (Oxford: Hart, 2010); neo-monism in Enzo Cannizzaro, ‘The Neo-Monism of the European Legal Order’ in Cannizzaro and others, ibid, 35–58; communitarisation in Anne Peters, ‘The Position of International Law Within the European Community Legal Order’ (1997) 40 German Yearbook of International Law, 9, 34–5; and Christian Tietje, ‘The Status of International Law in the European Legal Order: The Case of International Treaties and Non-Binding International Instruments’ in Jan Wouters, André Nollkaemper, and Erika de Wet (eds), The Europeanisation of International Law: The Status of International Law in the EU and its Member States (The Hague: TMC Asser Press, 2008) 55–69, 58–9. 41Haegeman (n 8), para. 5. 42Haegeman (n 8), para. 5; Case C-301/08 Bogiatzi [2009] ECR I-10185. 43Haegeman (n 8), para. 3–4. 44 Trevor C Hartley, ‘International Agreements and the Community Legal System: Some Recent Developments’ (1983) 8 EL Rev, 383, 391; Trevor C Hartley, The Foundations of European Community Law (4th edn, Oxford: Oxford University Press, 1998) 263; Mario Mendez, The Legal Effects of EU International Agreements (Oxford: Oxford University Press, 2013) 64. 45 Giorgio Gaja, ‘Trends in Judicial Activism and Judicial Self-Restraint Relating to Community Agreements’ in Enzo Cannizzaro (ed.), The European Union as an Actor in International Relations (Alphen aan den Rijn: Kluwer Law International, 2002) 117–34, 119. 46 Cheyne (n 22), 586–7; De Witte (n 11), 336; Allan Rosas, ‘The European Court of Justice and Public International Law’ in Wouters and others, (n 40), 71–85, 75; Joost Pauwelyn, ‘Europe, America and the “Unity” of International Law’ in Wouters and others (n 40), 205–25, 222, Kuilwijk (n 35), 101. 47 Mendez, (n 44), 63. 48 Klabbers (n 4), 264. 49 Klabbers ( n 4), 276. 50 For instance, in Polydor the Court of Appeal of England and Wales appeared to link the issue of direct enforceability by individuals to the fact that the agreement was adopted by regulation; Case 270/80 Polydor [1982] ECR 329, para. 10; Bundesfinanzhov, judgment of 5 August 1980 (1980) RIW 786 as cited in Geert A Zonnekeyn, ‘The Direct Effect of GATT in Community Law: from International Fruit Company to the Banana Cases’ (1996) 2 International Trade Law and Regulation 63, 64; Peters (n 40), 22. 51 Klabbers (n 4), 275. 52 Gerhard Bebr, ‘Agreements Concluded by the Community and Their Possible Direct Effect: From International Fruit Company to Kupferberg’ (1983) 20 CML Rev, 35, 38–9; Riesenfeld (n 19), 506. 53 Iain Mcleod, Ian D. Hendry, and Stephen Hyett, The External Relations of the European Communities: A Manual of Law and Practice (Oxford: Clarendon Press, 1996) 81; Jacques H J Bourgeois, ‘Effects of International Agreements in European Community Law: Are the Dice Cast?’ (1984) 82 Michigan Law Review, 1250, 1257. 54 Groux and Manin (n 32), 115–16. 55 Case C-160/09 Katsivardas [2010] ECR I-4591, para. 34. 56 Case C-192/89 Sevince [1990] ECR I-3461, paras 8–11; Case 30/88 Greece v Commission [1989] ECR 3711, para. 13. 57 Jan Klabbers, ‘Virtuous Interpretation’ in Malgosia Fitzmaurice, Olufemi Elias, and Panos Merkouris (eds), Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years On (Leiden: Martinus Nijhoff Publishers 2010) 17–37, 20. 58 Edwards (n 22), 435. 59 Case 104/81 Hauptzollamt Mainz v Kupferberg [1982 ] ECR 364, paras 11–13. 60 According to this maxim, treaties are binding upon parties and must be performed in good faith; Art. 26, Vienna Convention on the Law of Treaties, 23 May 1969, United Nations, Treaty Series, vol. 1155, p. 331. 61 Klabbers (n 4), 281; Hartley, The Foundations of Community Law (n 44), 217. 62 Klabbers, Völkerrechtsfreundlich (n 2), 100. 63 Mcleod and others, (n 53), 126–7. 64 Klabbers, Völkerrechtsfreundlich (n 2), 100. 65 Gáspár-Szilágyi (n 31), 605–6; Lenaerts (n 25), 55. 66 Pieter Verloren van Themaat, ‘The Impact of the Case Law of the Court of Justice of the European Communities on the Economic World Order’ (1984) 82 Michigan Law Review, 1422, 1435–6. 67 Martines (n 19), 134. 68 Pieter J. Kuijper, ‘The Case Law of the Court of the Court of Justice of the EU and the Allocation of External Relations Powers: Wither the Traditional Role of the Executive in EU Foreign Relations?’ in Marise Cremona and Anne Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Oxford: Hart, 2013) 95–114, 103. 69 John H. Jackson, ‘Status of Treaties in Domestic Legal Systems: A Policy Analysis’ (1992) 86 American Journal of International Law, 310, 322. 70 Mendez, ‘The Legal Effect of Community Agreements: Maximalist Treaty Enforcement and Judicial Avoidance Techniques’ (2010) European Journal of International Law, 83, 88; Mendez, The Legal Effects of EU International Agreements (n 44), 73–6. 71 Hartley refers to ‘direct effect’ as corresponding to the concept of direct applicability as used in this article; Trevor C Hartley, ‘The Constitutional Foundations of the European Union’ (2001) 117 Law Quarterly Review, 225, 237. 72 Cremona (n 4), 234. 73 Jan Wouters, André Nollkaemper, and Erika de Wet, ‘Introduction: The “Europeanisation” of International Law’ in Wouters and others (n 40), 1–3, 8–9. 74 André Nollkaemper, ‘The Netherlands’ in Sloss (n 6), 366. 75 Wessel (n 40), 12. 76 Peters (n 40), 29. 77 Mario Mendez, ‘The Enforcement of EU Agreements: Bolstering the Effectiveness of Treaty Law’ (2010) 47 CML Rev, 1719, 1724. 78 Emphasis added; Pescatore (n 22), 173. 79 Gerrit Betlem and André Nollkaemper, ‘Giving Effect to Public International Law and European Community Law before Domestic Courts: A Comparative Analysis of the Practice of Consistent Interpretation’ (2003) 14 European Journal of International Law, 569, 573. 80 Per Bourgeois, direct effect is part of ‘a broader problem of the status of international agreements’. Klabbers notes that in relation to at least the early case law ‘the idea of direct effect is inescapable when thinking about the reception of international law in the Community legal order’. According to Eeckhout, direct effect can serve as a ‘limitation to integration of international law’. Van Vooren and Wessel, while noting that the issue of the status and validity (matters for monism/dualism) of international law should at least ‘formally’ be distinguished from the issue of direct effect, also perceive that the concept of direct effect as applied in the case law can cast a shadow over the monist nature of the EU system; Bourgeois (n 53), 1255; Klabbers (n 4), 282–83; Piet Eeckhout, ‘The Integration of Public International Law in EU Law: Analytical and Normative Questions’ in Piet Eeckhout and Manuel Lopez-Escudero (eds), The European Union’s External Action in Times of Crisis (Oxford: Hart, 2016) 190–204, 204; Van Vooren and Wessel (n 16), 218–20, 231; Ernst-Ulrich Petersmann, ‘Application of GATT by the Court of Justice of the European Communities’ (1983) 20 CML Rev, 397, 402. 81 See for instance Klabbers (n 4), 292–4, 296–7; Pieter J. Kuijper and Marco Bronckers, ‘The WTO in the European Court of Justice’ (2005) 42 CML Rev, 1316, 1354; Christina Eckes, ‘International Law as Law of the EU: The Role of the ECJ’ in Cannizzaro, Palchetti, and Wessel (n 40), 353–77; Thomas Cottier, ‘International Trade Law: The Impact of Justiciability and Separation of Powers in EC Law’ NCCR Trade Working Paper No 2009/18, 8; Antonis Antoniadis, ‘The European Union and WTO Law: A Nexus of Reactive, Coactive and Proactive Approaches’ (2007) 6 World Trade Review, 45, 83. 82 Crawford (n 35), 48–50; David Sloss, ‘Treaty Enforcement in Domestic Courts: A Comparative Analysis’ in Sloss (n 6), 1–60, 6. 83 Anthony Aust, Modern Treaty Law and Practice (3rd edn, Cambridge: Cambridge University Press, 2013) 163, 167. 84 Groux and Manin (n 32), 119; Wessel (n 40), 13. 85 AG Maduro remarked in FIAMM that ‘the fact that WTO law cannot be relied upon before a court does not mean that it does not form pat of the Community legal system’; Opinion of Advocate General Maduro, Case C-120/06 FIAMM [2008] ECR I-06513, para. 37; Bourgeois (n 53), 1260. 86 de Witte (n 13), 11. 87International Fruit (n 15). 88 Case 87/75 Bresciani [1976] ECR 129. 89 For an overview see Holdgaard (n 33), 288–98. 90 Martines (n 19), 132. 91 Pescatore (n 22), 155. 92 Murphy (n 6), 109. 93 Klabbers (n 4), 264; Joel Trachtman, ‘Bananas, Direct Effect and Compliance’ (1999) 10 European Journal of International Law, 655, 664. 94 André Nollkaemper, ‘The Duality of Direct Effect of International Law’ (2014) 25 The European Journal of International Law, 105, 124. 95 Hélène R. Fabri, ‘Is There a Case—Legally and Politically—For Direct Effect of WTO Obligations?’ (2014) 25 European Journal of International Law, 151, 151. 96 Edwards (n 22), 425. 97 Hartley, ‘International Agreements and the Community Legal System’ (n 44), 386–7; Alan Dashwood and others, Wyatt and Dashwood’s European Union Law (6th edn, Oxford: Oxford University Press, 2011) 954–5. 98Kupferberg (n 59), para 26. 99Kupferberg (n 59), paras 16–17. 100 Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 90. 101 Jurisdiction of the Courts of Danzig, 1928 PCIJ Series B, No 15, at 17–18. 102 Kuijper and Bronckers (n 81), 1320; Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 90; Arnaud Van Wayenerge and Peter Pecho, ‘Free Trade Agreements after the Treaty of Lisbon in the Light of the Case Law of the Court of Justice of the European Union’ (2014) 20 European Law Journal, 749, 753. 103 Art. 19(1) TEU [2012] OJ C 326. 104 van Themaat (n 65), 1428–9. 105 Robert Schütze, Foreign Affairs and the EU Constitution (Cambridge: Cambridge University Press, 2014) 51. 106 Opinion of Advocate General Mayras, International Fruit (n 15), 1234. 107 Klabbers (n 4), 271. 108 Nollkaemper (n 94), 118. 109 Armin Von Bogdandy, ‘Founding Principles’ in Armin von Bogdandy and Jürgen Bast (eds), Principles of European Constitutional Law (Oxford: Hart, 2006) 11–54, 29–30; Prechal, ‘Direct Effect, Indirect Effect’ (n 34), 38–9. 110 Paul Craig, ‘Once upon a Time in the West: Direct Effect and the Federalisation of EEC Law’ (1992) 12 OJLS 454, 458–63. 111 Henry G Schermers, ‘The Direct Application of Treaties with Third States: Note Concerning the Polydor and Pabst Cases’ (1982) 19 CML Rev, 563, 564–5. 112 Mendez, ‘The Legal Effect of Community Agreements’ (n 70), 89–93; Mendez (n 44), 107–3. 113 Bebr (n 52), 41. 114 Fabri (n 95), 161. 115 Nollkaemper (n 74), 367. 116 Mendez (n 77), 1725; Nikos Lavranos, Decisions of International Organisations in the European and Domestic Legal Orders of Selected EU Member States (Amsterdam: Europa Law Publishing, 2004) 45. 117 Wouters, and others (n 73), 9. 118 De Witte (n 11), 327. 119 Kuijper and Bronckers (n 81), 1322. 120 Case 22/70 ERTA [1971] ECR 00263; C-3/76 Kramer [1976] ECR 1279; Opinion 1/76 [1977] ECR 741. 121 Pauwelyn (n 46), 219. 122 Emphasis added; Hartley, ‘International Agreements’ (n 44), 386–7; see also Kuijper and Bronckers (n 81), 1317–8. 123 Pauwelyn (n 46), 222. 124Demirel (n 26); See further Panos Koutrakos, ‘Interpretation of Mixed Agreements’ in Christophe Hillion and Panos Koutrakos (eds), Mixed Agreements Revisited: The EU and its Member States in the World (Oxford: Hart, 2010) 116–37, 118; Andrea F Gagliardi, ‘The Right of Individuals to Invoke the Provisions of Mixed Agreements before the National Courts: A New Message from Luxembourg?’ (1999) 24 EL Rev, 276. 125 Nanette Neuwahl, ‘Freedom of Movement for Workers under the EEC Treaty Association Agreement’ (1988) 13 EL Rev, 360, 365. 126Demirel (n 26), para. 9. 127 The Dutch constitution is the example often cited in scholarship; Constitution of the Kingdom of the Netherlands, June 2002. 128 Pauwelyn (n 46), 222; Snyder (n 4), 326–35, 362; Prechal, ‘Does Direct Effect Still Matter’ (n 34), 1065; Geert de Baere, Constitutional Principles of EU External Relations (Oxford: Oxford University Press, 2012) 35. 129 Kuijper and Bronckers (n 81), 1317–23. 130 Claus-Diete Ehlermann, ‘Application of GATT Rules in the European Community’ in Meinhard Hilf, Francis G Jacobs, and Ernst-Ulrich Petersmann (eds), The European Community and Gatt (Alphen aan den Rijn: Kluwer 1986) 127–40, 137. 131 Case C-61/94 Commission v Germany [1996] ECR I-3989, para. 52; Joined Cases C-402/05P and C-415/05 P Yassin Abdullah Kadi v Council [2008] ECR I-06351, para. 307; Simon Marsden, ‘Invoking Direct Application and Effect of International Treaties by the European Court of Justice: Implications for International Environmental Law in the European Union’ (2011) 60 ICLQ 737, 742. 132 Marco Bronckers, ‘From “Direct Effect” to “Muted Dialogue”: Recent Developments in the European Courts’ Case Law on the WTO and Beyond’ (2008) 11 Journal of International Economic Law 885, 896; Kuijper (n 68), 103. 133 Ehlermann (n 130), 137. 134 Schermers (n 111), 566–7. 135 Schermers (n 111), 564. 136 Martines (n 19), 147. 137 Bebr (n 52), 36–7. 138 Gáspár-Szilágyi (n 31), 606–7. 139 AG Mayras suggested a more elaborate test in comparison with the direct effect of internal EU law to include such conditions as the analysis of the provisions, evaluation of the context and general scheme of the treaty, the aim of the provision in question, having regard to overall objective of the measure, and lastly the circumstances in which the contracting parties have decided to apply the treaty; International Fruit (n 15), 1235. 140International Fruit (n 15); Bresciani (n 88); Kupferberg (n 59); Holdgaard (n 33), 246; Cheyne, ‘Haegeman, Demirel’, (n 24), 24. 141 Van Wayenerge and Pecho (n 102), 754–5. 142 For instance, Martines and Zonnekeyn consider that the condition related to the nature and the broad logic of the agreement comes first followed by the attributes of the provision, whereas Holdgaard views the attributes of the provision to be the first condition followed by the purpose and the nature of the agreement; Martines (n 19), 138; Zonnekeyn (n 50), 66; Holdgaard (n 33), 251–71. 143 For criticism of this approach see Christine Kaddous, ‘Effects of International Agreements in the EU Legal Order’ in Marise Cremona and Bruno de Witte (eds), EU Foreign Relations Law: Constitutional Fundamentals (Oxford: Hart, 2008) 291–312. 144 For instance, ‘the purpose, the spirit, and the general scheme’ in International Fruit, ‘the spirit, the general scheme and the wording’ in Bresciani; the ‘nature and structure’ of the agreement in Portugal v Council, ‘the nature and broad logic’ in IATA; International Fruit (n 15), para. 8, 19–20, 27; Bresciani (n 88), para. 16; Portugal v Council (n 15), para. 47; Case C-344/04 IATA [2006] ECR I-403, para. 35. 145 Schütze (n 105), 52. 146 Cheyne (n 22), 594; Marise Cremona, ‘Who Can Make Treaties? The European Union’ in Duncan B. Hollis (ed.), The Oxford Guide to Treaties (Oxford: Oxford University Press, 2012) 93–124, 112; Klabbers (n 4), 279–80. 147 For observations on earlier case law see Bebr (n 52), 58; for a more general overview of the application of the first condition see Koutrakos (n 39), 259–67. 148 See further Zonnekeyn (n 50); Hilf and others (n 130); Piet Eeckhout, ‘Judicial Enforcement of WTO Law in the European Union: Some Further Reflections’ (2002) 5 Journal of International Economic Law, 91; Snyder (n 4); Nicholas Emiliou and David O’Keeffe (eds), The European Union and World Trade Law After the GATT Uruguay Round (Oxford: Wiley 1996); Kuijper and Bronckers (n 81); Fabri (n 95); Bronckers (n 132); Koutrakos (n 39), 280–301. 149 Even though the EC did not accede to the GATT, the Court established its binding effect by relying on succession theory; International Fruit (n 15), paras 10–18, 21; Case 9/73 Schlüter [1973] ECR 1135. 150 Cheyne, ‘Haegeman, Demirel’, (n 24), 22–3. 151 Bebr (n 52), 46. 152 Jan Klabbers, ‘The Validity of EU Norms Conflicting with International Obligations’ in Cannizzaro and others (n 40), 111–31, 125. 153 In Germany v Council, Germany argued that the issue of direct effect is irrelevant for judging the validity of EU law in light of GATT 1947; Case C-280/93 Germany v Council (Bananas) [1994] ECR I-4973, para. 103; AG Saggio in Portugal v Council argued against using direct effect for legality challenges (n 15); Geert Zonnekeyn, ‘The Status of the WTO Agreements in the EC Legal Order after the Portuguese Textiles Cases’ (2000) 6 International Trade Law and Regulation, 42, 47; Martines (n 19), 141, 143; Kuijper and Bronckers (n 81), 1343–54. 154 Eileen Denza, ‘A Note on Intertanko’ (2008) 33 EL Rev, 870, 875–6. 155 Koutrakos (n 39), 282. 156 Sasha Prechal, Directives in European Community Law (Oxford: Oxford University Press, 1995), 276; Prechal, ‘Direct Effect, Indirect Effect’ (n 34), 37–8. 157 This observation is made in relation to WTO law; Van Vooren and Wessel (n 16), 221. 158Portugal v Council (n 15), para. 36–47. 159Van Parys (n 15), para. 53. 160 Klabbers (n 4), 276. 161Bresciani (n 88); Polydor (n 50); Kupferberg (n 59). 162Portugal v Council (n 15), paras 44–5. 163 Allan Rosas, ‘Case Comment: Case C-149/96, Portugal v Council’ (2000) 37 CML Rev, 797, 807. 164 Geert Zonnekeyn, ‘Mixed Feeling about the Hermes Judgment’ (1999) 5 International Trade Law and Regulation, 20, 24. 165 Opinion of Advocate General Alber, Case C-93/02 P Biret [2003] ER I-10497, para. 102; Lavranos (n 116), 40. 166Portugal v Council (n 15), para. 40. 167 Bebr (n 52), 58. For the analysis of the role dispute settlement mechanisms play in the finding of direct effect see further Beatrice I. Bonafé, ‘Direct Effect of International Agreements in the EU Legal Order: Does it Depend on the Existence of an International Dispute Settlement Mechanism?’ in Cannizzaro and others (n 40), 229–48. 168 Cannizzaro, ‘The Neo-Monism’ (n 40), 43. 169 Bronckers (n 132), 895. 170 See below Section V. 171 Cremona (n 4), 241. 172 Kuijper (n 68), 105, 114. 173 Fabri (n 95), 164 174 Gáspár-Szilágyi (n 31), 610. 175 Klabbers, ‘Völkerrechtsfreundlich‘ (n 2), 97. 176 Achilles Skordas, ‘Völkerrechtsfreundlichkeit as Comity and the Disquiet of Neoformalist: A Response to Jan Klabbers’ in Koutrakos (ed.) (n 2), 115–44, 129–30. 177 Skordas (n 176), 129–30. 178 Case C -69/89 Nakajima v Council [1991] ECR I-2069; Case 70/87 Fediol [1989] ECR 1781; see further Snyder (n 4), 342–7; Koutrakos (n 39), 301–7. It should be noted that certain authors view these exceptions as part of indirect effect; Peters (n 40), 71. 179Commission v Germany (n 131), para. 31; Koutrakos (n 39), 307–11; Eeckhout (n 9), 355–7. 180 Klabbers, ‘Völkerrechtsfreundlich‘ (n 2), 102. 181 Koutrakos (n 39), 264. 182Intertanko (n 3). 183Intertanko (n 3), paras 45–65. 184Intertanko (n 3), paras 54–65; Piet Eeckhout, ‘Case C-308/06, The Queen on the application of Intertanko and Others v Secretary of State for Transport, judgment of the Court of Justice (Grand Chamber) of 3 June 2008, nyr’ (2009) 46 Common Market Law Review 2041, 2054. 185 Eeckhout (n 184), 2054. 186International Fruit (n 15), paras 8, 19–20, 27. 187 Eeckhout (n 9), 382. 188 Van Vooren and Wessel (n 16), 232; Cannizzaro, The Neo-Monism of the European Legal Order (n 40), 49. 189 Denza (n 154), 875. 190 Cremona (n 4), 242. 191 Mendez points out the inconsistency with the Biotech and IATA judgments where the Court seemed to separate the issue of legal review from that of conferral of individual rights; C-377/98 Biotech [2001] ECR I-7079; C-344/04 IATA [2006] ECR I-403; Mendez (n 44), 246, 250, 273–81; Mendez (n 77), 1751. 192 Eeckhout, (n 184), 2055. 193 Eeckhout points out the importance of the multilateral nature of the agreement, many provisions of which codify customary international law; (n 184), 2041. 194 Eeckhout (n 80), 199. 195 Denza (n 154), 877–8. 196 Cremona (n 4), 243. 197 Gáspár-Szilágyi (n 31), 613. 198 Mendez (n 77), 1751. 199Air Transport Association of America and Others (n 3). 200 Jed Odermatt, ‘The Court of Justice of the European Union: International or Domestic Court’ (2014) 3 Cambridge Journal of International and Comparative Law, 696, 707. 201Air Transport Association of America and Others (n 3), paras 75–6. 202 Koutrakos (n 39), 260. 203Air Transport Association of America and Others (n 3), para. 77. 204 Gáspár-Szilágyi (n 31), 621–4. 205 Gáspár-Szilágyi (n 31),, 613. 206 Koutrakos (n 39), 266. 207 Mendez (n 44), 246, 250, 273–81; Eeckhout (n 9), 381–3. 208 For an overview of the application of this condition see Koutrakos (n 39) 267–70. 209 Gáspár-Szilágyi (n 31), 614–15. 210 Case C-18/90 Kziber [1991] ER I-199, para. 25; Case C-265/03 Simutenkov [2005] ECR I-2579, para 25. 211Case C-213/03 Pêcheurs de l'étang de Berre [2004] ECR I-07357, paras 41–42. 212Simutenkov (n 210), paras 26–28; see further Christophe Hillion, ‘Russian Federation’ in Stephen Blockmans and Adam Lazowski (eds), The European Union and its Neighbours: A Legal Appraisal of the EU’s Policies of Stabilisation, Partnership and Integration (The Hague: TMC Asser Press, 2006) 463–97. 213 Case C-162/00 Pokrzeptowicz-Meyer [2002] ECR I-1049, paras 20–27. 214 Kaddous (n 143), 304. 215 Mendez (n 44), 151; Francis G Jacobs, ‘Direct Effect and Interpretation of International Agreements in the Recent Case Law of the European Court of Justice’ in Allan Dashwood and Marc Maresceau (eds), Law and Practice of EU External Relations: Salient Features of a Changing Landscape (Cambridge: Cambridge University Press, 2008) 13–33, 32. 216 Koutrakos (n 39), 269. 217Simutenkov (n 210), para. 21. 218Simutenkov (n 210), para. 21. 219 Jacobs (n 215), 32. 220 Hillion particularly notes that no such obligation comes through in the Spanish version of the provision on which the applicant relied; Christophe Hillion, ‘Case C-265/03, Igor Simutenkov v. Ministerio de Educación y Cultura, Real Federación Española de Fútbol’ (2008) 45 CML Rev, 815, 824. 221 Hillion (n 220), 823, 826–7. 222Demirel (n 26), paras 17–25. 223 It should be noted that similar departures from Demriel were made in Kziber (n 210), para. 25. 224 In Pokrzeptowicz-Meyer itself the analysis followed the same pattern, that is finding that the provision was clear, unconditional, and not dependent on further action before considering the nature and purpose of the agreement; Porkzeptowicz-Meyer (n 213), paras 19–26; Kaddous (n 143), 304. 225Simutenkov (n 210), paras 26–28. 226Simutenkov (n 210), para 28. 227 Hillion (n 212), 474. 228Kziber (n 210); Tietje (n 40), 62–2; Bebr (n 52), 63. 229 The Court views it as ‘essentially an economic agreement’; Case C-162/96 Racke [1998] ECR I-3655, para. 20. 230Kziber (n 210), para. 21, Case C-113/97 Babahenini [1998] ECR I-183, para. 17; Case C-103/94 Krid v CNAVTS [1995] ECR I-719, paras 21–23; Case C-58/93 Yousfi v Belgian State [1994] ECR I-1353, paras 16–18; Case C-126/95 Hallouzi-Choho v Bestuur van de Sociale Verzekeringsbank [1996] ECR I-4807, para. 19; Racke (n 229), paras 34–36. 231 Gáspár-Szilágyi (n 31), 619. 232 Martines (n 19), 145. 233Lesoochranárske zoskupenie VLK (Brown Bear) (n 29), paras 44–45. 234 Jacobs (n 215), 32–33. 235 On the Court’s methods of interpretation see Eeckhout (n 9), 304–19. 236 Mendez (n 44), 110. 237 Van Wayenerge and Pecho (n 102), 757. 238Opinion 1/91 [1991] ECR I-6079, para. 14; Jacobs (n 215), 16. 239 Case 225/78 Bouhelier [1977] ECR 3151, para. 6; Polydor (n 50), paras 14–18; Kupferberg (n 59), paras 29–31; Case C-312/91 Metalsa [1993] ECR 3751, paras 11-19. 240Bresciani (n 88), para. 25; Case C-268/99 Jany [2001] ECR I-8615, paras 33–38; Cremona (n 146), 113. 241Metalsa (n 239), para. 11. 242 Peters (n 40), 23. 243 Delano Verwey, The European Community, the EU and the International Law of Treaties (The Hague: TMC Asser Press, 2004) 223–4. 244Bresciani (n 88), para. 25. 245 Kaddous (n 143), 312. 246 Van Wayenerge and Pecho (n 102), 758–62; Chrtistophe Hillion, ‘Cases C-63/99 Secretary of State for the Home Department ex parte Wiesław Głoszczuk and Elzbieta Głoszczuk; C-235/99 Secretary of State for the Home Department ex parte Eleanora Ivanova Kondova; C-257/99 Secretary of State for the Home Department ex parte Julius Barkoci and Marcel Malik; judgments of the Full Court of 27 September 2001; Case C-268/99 Aldona Małgorzata Jany e.a v. Staatssecretaris van Justitie, judgment of the Full Court of 20 November 2001; Case C-162/00 Land Nordrhein-Westfalen v. Beata Pokrzeptowicz-Meyer, judgment of the Full Court of 29 January 2002’ (2003) 40 CML Rev, 465, 489; see further Niamh Nic Shuibhne, ‘European Law—the European Court of Justice and the Europe Agreements: Shaping a Legal Framework’ (2001) 23 Dublin University Law Journal, 203. 247 Van Wayenerge and Pecho (n 102), 758–62. 248 Koutrakos (n 39), 274. 249 This inconsistency can be seen in contrasting Pabst and Richarz; Pokrzeptowicz-Meyer; Kolpak on the one hand, and, Gloszczuk; Kondova; Barkoci and Malik, on the other; Case C-17/81 Pabst and Richarz [1982] ECR 1331; Pokrzeptowicz-Meyer (n 213); Case C-438/00 Kolpak [2003] ECR I-04135; Case C-63/99 Gloszczuk [2001] ECR I-06369; Case C-235/99 Kondova [2001] ECR I-06427; Case C-257/99 Barkoci and Malik [2001] ECR I-06557; Hillion (n 246), 489–91; Antje Pedain, ‘A Hollow Victory: the ECJ Rules on Direct Effect of Freedom of Establishment Provisions in Europe Agreements’ (2002) 61 Cambridge Law Journal 284, 287. 250Gloszczuk (n 249), para. 86; Kondova (n 249), para. 91; Barkoci and Malik (n 249), para. 66. 251 Hillion (n 246), 490. 252 Barbara Bogusz, ‘Regulating the Right of Establishment for Accession State Nationals: Reinforcing the “Buffer Zone” or Improving Labour Market Flexibility’ (2002) 27 EL Rev, 472, 482. 253 Antje Pedain, ‘“With or Without Me”: The ECJ Adopts a Pose of Studied Neutrality towards EU Enlargement’ (2002) 51 ICLQ 981, 989; Bogusz (n 252), 482. 254 Pedain (n 249). 255 Case C-416/96 El Yassini [1999] ECR I-1209, para. 30. 256Kolpak (n 249), paras 25–26. 257Simutenkov (n 210), paras 21–24; Pokrzeptowicz-Meye (n 214), para. 22. 258 Hillion (n 220), 831. 259 Jacobs (n 215), 28. 260Simutenkov (n 210), paras 33, 36–38; Hillion (n 220), 829–830. 261 Nollkaemper identifies the dual function of direct effect which includes being used both as a ‘sword’ and a ‘shield’; Nollkaemper (n 94), 112–17. 262 De Witte (n 11), 336. 263 Pescatore (n 22), 149. 264 Mendez (n 77), 1720; Peters (n 40), 63–4, 77. 265 Klabbers, ‘Straddling the Fence’ (n 2), 64. 266 See note 99 above. 267 John H Jackson, ‘Direct Effect of Treaties in the US and the EU, the Case of the WTO: Some Perceptions and Proposals’ in Anthony Arnull, Piet Eeckhout, and Takis Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford: Oxford University Press, 2008) 361–82, 370. 268 Schütze (n 105), 52; Holdgaard (n 33), 251–71. 269 Cheyne, ‘Haegeman, Demirel’, (n 24), 41. 270 Kuijper as cited in Snyder (n 4), 335. In relation to the effect of WTO law, Advocate General Alber favoured seeking a multilateral solution instead of a unilateral finding of direct effect; Opinion of Advocate General Alber, C-27 &122/00 Omega Air [2002] ECR I-2569, para. 95. 271 Multilateral Agreement between the European Community and its Member States, the Republic of Albania, Bosnia and Herzegovina, the Republic of Bulgaria, the Republic of Croatia, the former Yugoslav Republic of Macedonia, the Republic of Iceland, the Republic of Montenegro, the Kingdom of Norway, Romania, the Republic of Serbia, and the United Nations Interim Administration Mission in Kosovo on the establishment of a European Common Aviation Area [2006] OJ L 285. 272 [1998] OJ L42/43. 273 [1998] OJ L42/43, Art. 17(3). 274 Andre Nollkaemper, ‘The Direct Effect of Public International Law’ in Jolande M Prinssen and Annette Schrauwen (eds), Direct Effect: Rethinking a Classic of EC Legal Doctrine (Amsterdam: Europa Law Publishing 2002) 157–80, 171. 275 According to Art. 336, ‘[n]othing in this agreement shall be construed as conferring rights or imposing obligations on persons, other than those created between the Parties under public international law’; Trade Agreement between the European Union and its Member States, on the one part, and Colombia and Peru on the other part [2012] OJ L354/3. 276 Art. 17.15 of EU–Singapore Free Trade Agreement is entitled ‘No direct effect’ and provides that ‘[f]or greater certainty, nothing in this Agreement shall be construed as conferring rights or imposing obligations on persons, other than those created between the Parties under public international law’. The Agreement was initialled, but following the CJEU’s Opinion 2/15 it cannot be concluded in its current form solely by the EU; Opinion 2/15, ECLI:EU:C:2017:376. 277 According to Art. 356, ‘[n]othing in this agreement shall be construed as conferring rights or imposing obligations on persons, other than those rights or obligations created by this agreement nor as obliging a party to permit that this agreement be directly invoked in its domestic legal system, unless otherwise provided in that party’s domestic legislation’; Agreement establishing an Association between the European Union and its Member States, on the one hand, and Central America on the other [2012] OJ L346/3. 278 Art. 30.6 provides that ‘[n]othing in this Agreement shall be construed as conferring rights or imposing obligations on persons other than those created between the Parties under public international law, nor as permitting this Agreement to be directly invoked in the domestic legal systems of the Parties’; Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union and its Member States, of the other part [2017] OJ L 11/23 279 Aliki Semertzi, ‘The Preclusion of Direct Effect in the Recently Concluded EU Free Trade Agreements’ (2014) 51 CML Rev, 1125, 1130. 280 For instance free trade agreements, cooperation agreements, partnership and cooperation agreements; Kupferberg (n 59); Simutenkov (n 210); Racke (n 229). 281 Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other part [2014] OJ L261; Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and the Republic of Moldova, of the other part [2014] OJ L260; Association Agreement between the European Union and its Member States, of the one part, and Ukraine, of the other part [2014] OJ L161. 282 Semertzi (n 279), 1134, 1141–2. 283 Guillaume van der Loo, Peter van Elsuwege, and Roman Petrov, ‘The EU–Ukraine Association Agreement: Assessment of an Innovative Legal Instrument’, EUI Working Papers No. 9 (2014), 26; Semertzi (n 279), 1155–6. 284 Kuijper and Bronckers (n 81), 1321. 285 Koutrakos (n 39), 258; Semertzi (n 279), 1129. 286 Art. 5, Council Decision 2016/838/EU of 23 May 2016 on the conclusion, on behalf of the European Union, of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Georgia, of the other part [2016] OJ L 141/26; Art. 5, Council Decision 2014/295/EU on the signing, on behalf of the European Union, and provisional application of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Ukraine, of the other part, as regards the Preamble, Art. 1, and Titles I, II and VII thereof [2014] OJ L161/1; Art. 5, Council Decision 2016/839/EU of 23 May 2016 on the conclusion, on behalf of the European Union, of the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and the Republic of Moldova, of the other part [2016] OJ L 141/28. 287 Schermers (n 111), 564. 288 Some agreements might require implementing measures, which would be adopted prior to or simultaneously with entering into the agreements; Mcleod, and others (n 53), 128–9. 289 Council Decision 94/800 [1994] OJ L 336/1, recital 14. 290 Peters (n 40), 60. 291 McGoldrick,(n 14), 133. 292Portugal v Council (n 15). 293 Opinion of AG Saggio in Portugal v Council (n 15), para. 20. 294 AG Saggio (n 293), para. 20. 295 AG Saggio (n 293), para. 20. 296 The CJEU considers the EU to be bound only by those provisions of the Vienna Convention which are part of customary international law (Case C-386/08 Brita [2010] ECR I-1289, para. 42). Art. 31 of the Vienna Convention is seen as part of customary international law; Crawford (n 35), 380. 297 AG Saggio (n 293), para. 20. 298Portugal v Council, (n 15), para. 35. 299Portugal v Council, (n 15), paras 36–48. 300 Rosas (n 163), 808. 301 Fabri (n 95), 162. 302 Rosas (n 163), 810; Mendez (n 44), 246. 303 The Commission’s accord with the Council was obvious in the case of the WTO Agreement, where it declared its unequivocal support for the exclusion of direct effect; Commission, Uruguay Round Implementing Legislation, COM(94) 414 final; Cremona (n 4), 241. 304 Kuijper and Bronckers (n 81), 1321. 305 André Nollkaemper (n 6), 363. 306 Lori F. Damrosch, ‘Role of the United States Senate Concerning Self-Executing and Non-Self-Executing Treaties’ (1991) 67 Chicago-Kent Law Review 515, 515, 518. 307 Vázquez, ‘The Four Doctrines of Self-Executing Treaties’ (n 20), 706–8; Vázquez, ‘Treaties as Law of the Land’ (n 20), 667–85; Carlos M. Vázquez, ‘Laughing at Treaties’ (1999) 99 Columbia Law Review 2154, 2174; Damrosch (n 306), 526–7; Thomas Buergenthal, ‘Modem Constitutions and Human Rights Treaties’ (1997) 36 Columbia Journal of Transnational Law 211, 221–2. 308 Vázquez, ‘Treaties as Law of the Land’ (n 307), 672; Stefan A Riesenfeld, ‘International Agreements’ (1989) 14 Yale Journal of International Law 455, 466. 309 The European Parliament rejected the Anti-Counterfeiting Trade Agreement in 2012; ‘European Parliament Rejects ACTA’ Press Release, 4 July 2012; Ricardo Passos, ‘The External Powers of the European Parliament’ in Eeckhout and Lopez-Escudero (eds) (n 80), 83–128. 310 Art. 218(6) TFEU. 311Biotech (n 191); Case C-479/04 Laserdisken [2006] ECR I-8089; IATA (n 144); Intertanko (n 3). 312 European Parliament, Resolution on the Relationship Between International Law, Community Law and the Constitutional Law of the Member States, 24 September 1997, Explanatory statement, para. 3.2. 313 European Parliament (n 312). 314 European Parliament (n 312), para. 14. 315 Snyder (n 40), 172. 316 Damrosch (n 306), 528–9. 317 AG Saggio (n 293), para. 20. 318 Art. 2(1)(d) of the Vienna Convention on the Law of Treaties. 319 Verwey (n 243), 225–6. 320 Oliver Dörr, ‘Art 31: General Rule of Interpretation’ in Oliver Dörr and Kirsten Schmalenbach (eds), Vienna Convention on the Law of Treaties: A Commentary (Springer 2012) 521–70, 552; Richard K Gardiner, Treaty Interpretation (Oxford: Oxford University Press, 2008) 211, 214–15. 321 Aust (n 83), 212. 322 Dörr (n 320), 550. 323 Mark E Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Leiden: Martinus Nijhoff 2009) 430. 324 Dörr (n 320), 552. 325 Verwey (n 243), 226. 326 Mark E Villiger, ‘The Rules on Interpretation: Misgivings, Misunderstandings, Miscarriage? The “Crucible” Intended by the International Law Commission’ in Enzo Cannizzaro (ed.), The Law of Treaties Beyond the Vienna Convention (Oxford: Oxford University Press, 2011) 105–22, 114. 327 Dörr (n 320), 549. 328 Jean-Marc Sorel and Valerie Bore Eveno, ‘Article 31: General Rule of Interpretation’ in Olivier Corten and Pierre Klein (eds), The Vienna Conventions on the Law of Treaties: A Commentary (Oxford: Oxford University Press, 2011) 804–37, 808. 329 Various commentators indeed dismissed the significance of the preambular statement as it was not part of the decision’s operative part; Kuijper and Bronckers (n 81), 1345; Jacques H J Bourgeois, ‘The Uruguay Round of GATT: Some General Comments from an EC Standpoint’ in Nicholas Emiliou and David O’Keeffe (eds), The European Union and World Trade Law After the GATT Uruguay Round (Oxford: Wiley, 1996) 81–90, 89; Dashwood and others (n 97), 957. 330 Van der Loo and others (n 283), 27. 331 Oddly, a similar provision was not included in the EU–Georgia and EU–Moldova Association Agreements. 332El-Yassini (n 255); Kziber (n 210). 333 Art. 23, Agreement on partnership and cooperation establishing a partnership between the European Communities and their Member States, of one part, and the Russian Federation, of the other part [1997] OJ L 327. 334 Van der Loo and others (n 283), 27. 335 European Council, Press Remarks by H Van Rompuy, President of the European Council, following the EU–Ukraine Summit, Brussels 25 February 2013, EUCO 48/13. On EU–Ukraine Association Agreement see further Guillaume van der Loo, The EU–Ukraine Association Agreement and Deep and Comprehensive Free Trade Area: A New Legal Instrument for EU Integration without Membership (Leiden: Brill 2016). 336 Art. 8, Council Decision 2011/265/EU of 16 September 2010 on the signing, on behalf of the European Union, and provisional application of the Free Trade Agreement between the European Union and its Member States, of the one part, and the Republic of Korea, of the other part [2011] OJ L 127/1. 337 Semertzi (n 279), 1135. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals-permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Yearbook of European LawOxford University Press

Published: May 30, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off