Reinterpretation of the Requirements to Preserve the Autonomy of the EU Legal Order in Opinion 1/17
Reinterpretation of the Requirements to Preserve the Autonomy of the EU Legal Order in Opinion 1/17**
On 30 April 2019, the Court of Justice of the European Union (CJEU), sitting as a full court, delivered its long-awaited Opinion 1/17 on the compatibility of the new investor-State dispute settlement (ISDS) mechanism proposed in the Comprehensive Economic and Trade Agreement (CETA) between the EU and Canada with European Union (EU) law. The outcome of the Opinion is somewhat surprising, because the CJEU, in the past, seemed to be overly protective of its own jurisdiction when it came to the establishment of a new international court or the accession to an international agreement providing for its own court and usually rejected the participation of the EU and the Member States in the competing international dispute settlement mechanisms.
Indeed, the CJEU has laid down in its case-law extremely strict criteria that a dispute settlement mechanism has to fulfil in order to be found compatible with EU law. These criteria have been mostly spelled out in the CJEU’s opinions given under Article 218(11) TFEU, which allows the CJEU to rule on the compatibility of an international agreement with EU law prior to the conclusion of the agreement.
One of the main arguments used by the CJEU in the course of the assessment of the envisaged adjudicatory mechanisms, based on which the CJEU has usually established incompatibilities, was the protection of the autonomy of the EU legal order. Even though this notion is not elaborated on in the EU Treaties, the CJEU has developed an extensive case-law on the autonomy of EU law, which is usually accompanied by the notions of very foundations, very nature and essential characteristics. In other words, as long as a new dispute settlement mechanism has no adverse effect on the very core elements of EU law, the preservation of the autonomy of the EU legal order can be ensured and the envisaged mechanism is likely to be compatible with the EU Treaties.
However, particularly in the latest opinions of the CJEU, the autonomy of the EU legal order has gained a broader interpretation compared to the CJEU’s previous case-law on autonomy, which gave rise to uncertainties as regards the exact meaning and boundaries of the term. This shift has been heavily criticised in the legal literature, concluding that autonomy is still ‘partially nebulous’ and describing the CJEU as being ‘selfish’ and ‘jealous’.
In addition to the uncertainties as to what autonomy is supposed to protect exactly and how the autonomy of the EU legal order may be preserved within the framework of the new ISDS mechanism envisaged by the CETA, there was another event that added further aspects to the legal situation surrounding Opinion 1/17. This was the Achmea decision, in which the Grand Chamber of the CJEU held that the investment arbitration clause contained in the Dutch–Czech–Slovakian intra-EU Bilateral Investment Treaty had an adverse effect on the autonomy of the EU legal order and was, therefore, incompatible with EU law. Although the Achmea decision relates to the incompatibility of an intra-EU dispute settlement mechanism with EU law and, therefore, it has no direct implications on the assessment of an extra-EU adjudication, the CJEU, by referring to its previous opinions, noted in Achmea that
[…] an international agreement providing for the establishment of a court responsible for the interpretation of its provisions and whose decisions are binding on the institutions, including the Court of Justice, is not in principle incompatible with EU law. The competence of the EU in the field of international relations and its capacity to conclude international agreements necessarily entail the power to submit to the decisions of a court which is created or designated by such agreements as regards the interpretation and application of their provisions, provided that the autonomy of the EU and its legal order is respected […].
Consequently, the CJEU, having in mind the then ongoing procedure initiated for the opinion of the CJEU on the compatibility of the international investment court envisaged in the CETA, confirmed that as long as the autonomy of the EU legal order is respected by an international regime with a binding dispute settlement mechanism, this system, as a matter of principle, cannot be considered incompatible with EU law. Thus, Achmea may also be read as guidance as to how an ISDS mechanism may be lawfully integrated into the framework of the EU Treaties.
With all these considerations in mind, in Opinion 1/17 the CJEU gave a green light to the ISDS mechanism proposed in the new generation bilateral free trade agreements, stating that the envisaged international investment court ‘does not adversely affect the autonomy of the EU legal order’. With this decision, however, the CJEU has departed, to a certain extent, from its latest opinions and reinterpreted the requirements that must be met for the preservation of the autonomy of the EU legal order.
This paper seeks to examine how and to what extent the CJEU reconsidered its assessment on the autonomy of the EU legal order in Opinion 1/17 and how this recent opinion can be reconciled with the previous case-law of the CJEU in connection with other extra-EU dispute settlement mechanisms. First, we give a brief overview of the factual and legal background that led to the delivery of Opinion 1/17 (II). Subsequently, the development of the concept of autonomy will be discussed in more detail, by describing the CJEU’s most relevant opinions on the external dimension of autonomy (III). Following this, we will focus on Opinion 1/17 and the main elements of the CJEU’s argumentation on how the autonomy of the EU legal order may be preserved in relation to the new ISDS mechanism envisaged in the CETA (IV). Finally, we close this paper with our conclusions (V).
II The Way Leading to the CETA Opinion
Over the last decade, the prevailing system of international investment arbitration faced a ‘legitimacy crisis’, which necessitated radical changes to the traditional ISDS mechanism. The main concerns that have been raised against investor-State arbitration include the lack of consistency, coherence, and predictability of the awards and the lack of transparency, as well as the lack of an impartial and independent procedure flowing from the nature of the arbitrators’ appointment.
In response to the various criticisms, the EU, under its new competence conferred by the Lisbon Treaty, launched a two-step reform process with the aim of fundamentally reforming investor-State arbitration. As a first, preliminary step, investment arbitration is intended to be gradually replaced with a bilateral two-tier investment court system (ICS) envisaged in the investment agreements concluded with third States and, as a second, final move, this process is expected to culminate in the establishment of a multilateral investment court, which will be entitled, in the long run, to adjudicate investment disputes covered by the new generation agreements.
The details of the ICS mechanism have been spelled out in the investment chapter of the CETA and also appear in the free trade agreement negotiated with Mexico, as well as in the investment protection agreements concluded with Singapore and Vietnam. This new EU-led approach provides, for the first time in an investment protection context, an appellate mechanism in order to ensure the consistency and predictability of the ICS awards and, thereby, to remedy one of the main deficiencies of investor-State arbitration. This means that the ICS comprises a standing first instance Tribunal and an Appellate Tribunal, which may review the Tribunal’s awards for errors in the application or interpretation of the law as well as in the appreciation of facts.
In addition to the appellate review, the selection method of the members of the permanent tribunals is another remarkable feature of the new system because the disputing parties have no say in the appointment of their own adjudicators. Instead of this, a joint committee, consisting of representatives of the contracting parties, is responsible for the appointment of the members of the tribunals and the case-allocation to a division of the Tribunals occurs in a ‘random and unpredictable’ way.
Since it was not entirely clear under the Lisbon Treaty whether the EU had the necessary competence to sign and ratify the EU–Singapore free trade agreement, which contained an investment chapter similar to the one set forth in the CETA, on its own, the European Commission asked the CJEU to opine on this question pursuant to Article 218(11) TFEU. The CJEU made clear in its Opinion 2/15 that the agreement in question had the characteristics of a mixed agreement, which means that the ISDS provisions of the investment Chapter of the agreement ‘fall within the competence shared between the European Union and the Member States’. Consequently, the provisions on the ICS mechanism may not enter into force until each individual EU Member State has completed its own internal ratification procedure.
Since the Commission did not raise any question about the compatibility of the new ISDS mechanism with EU law in its request, the CJEU did not touch upon this delicate issue in Opinion 2/15 but rather left this compatibility question open and only stated that
this opinion of the Court relates only to the nature of the competence of the European Union to sign and conclude the envisaged agreement. It is entirely without prejudice to the question of whether the content of the agreement’s provisions is compatible with EU law.
Therefore, during the ratification process of the CETA, Belgium submitted a request for an opinion of the CJEU pursuant to Article 218(11) TFEU on the compatibility of the ICS model with the EU Treaties, including with fundamental rights. The doubts Belgium raised as to the envisaged ISDS mechanism can be grouped into three categories, which are its compatibility with the autonomy of the EU legal order, its compatibility with the general principle of equal treatment and the requirement of effectiveness, as well as its compatibility with the right of access to an independent tribunal.
Although the CJEU in its Opinion 1/17 concluded that the ICS mechanism was compatible with all three requirements indicated by Belgium and, thus, gave the green light for the EU and the Member States to participate in the new extra-EU dispute settlement mechanism, only the conclusions of the CJEU on the compatibility of the ICS model with the autonomy of the EU legal order will be discussed in more detail in this paper. Before turning to the findings of the CETA Opinion, however, the subsequent section will focus on the case-law of the CJEU on the autonomy of the EU legal order, i.e. how this concept has been developed in a series of opinions of the CJEU in order to find incompatibilities with the rival extra-EU dispute settlement mechanisms.
III The Concept of the Autonomy of the EU Legal Order
1 The Development of Autonomy in the CJEU’s Case-Law
Even though the autonomy of the EU legal order is not mentioned in the text of the EU Treaties, the roots of this concept had already appeared in the case-law of the CJEU in the early 60s. Whereas in Van Gend en Loos the CJEU laid down the foundations for the principle of autonomy of the EU legal order by describing EU law as ‘a new legal order of international law’, in Costa, EU law was recognised as ‘an independent source of law’ which cannot ‘be overridden by domestic legal provisions’. At this time, the autonomy of the EU legal order had only been interpreted vis-à-vis the domestic legal order of the Member States and its main purpose was to guarantee the essential characteristics of EU law, such as primacy and direct effect, in relation to the legal order of the Member States across the whole EU.
Nevertheless, over time, the EU has gained more competences in the field of external relations and this brought up the question of how and to what extent the EU’s relationship with third States and other international organisations may affect EU law. The CJEU answered the question by referring to the external dimension of autonomy. Based on this, the autonomy of the EU legal order has been relied on to limit the effects of public international law on EU law and, thus, to safeguard the very core elements of EU law from any external influences. Hence, from an external relations law perspective, the principle of autonomy has been used to protect, inter alia, fundamental rights or the CJEU’s exclusive competence to interpret EU law.
This ultimate authority of the CJEU on the interpretation and application of EU law is of paramount importance in the context of EU constitutional law because this allows the CJEU to guarantee the uniform and consistent interpretation and application of EU law throughout the entire EU. This constitutional role of the CJEU can be derived from the joint application of Article 19(1) TEU, as well as Articles 267 and 344(1) TFEU. While Article 267 TFEU provides for the preliminary reference procedure that establishes direct cooperation between the CJEU and the domestic courts of the Member States in order to ensure the correct application and uniform interpretation of EU law, Article 19(1) TEU states that the CJEU ‘shall ensure that in the interpretation and application of the Treaties the law is observed’. In addition, Article 344 TFEU can be regarded as an ‘archetypal exclusive jurisdiction clause’, which provides that ‘Member States undertake not to submit a dispute concerning the interpretation or application of the Treaties to any method of settlement other than those provided for therein’.
Nevertheless, all these constitutional norms seem to interfere with the increasing political will set out in the Lisbon Treaty, based on which the EU intends to be a more active participant in the international scene and, thus, to contribute to the development of international law. This openness of the EU towards international law also implies that the EU should foster the various extra-EU dispute settlement mechanisms in order to strengthen the enforcement of rights set forth in the international treaties concluded with third States. The interference between ‘the exclusive jurisdiction [of the CJEU] over the definitive interpretation of EU law’ and the jurisdiction of these other mechanisms arises at this point, because the EU has been following the monist approach to international law since the CJEU’s judgment in the Haegeman case. In this case, the CJEU held that the provisions of an international agreement, ‘from the coming into force thereof, form an integral part of Community law’. In other words, the international agreements form part of the EU legal order and, therefore, the CJEU’s monopoly to interpret EU law extends to the provisions of the international agreements as well. Consequently, there is a clash between the exclusive jurisdiction of the CJEU to interpret and apply EU law and the jurisdiction of the other extra-EU dispute settlement mechanisms, which are, evidently, also entitled to interpret and apply the provisions of the international agreements.
In such a situation, when it comes to conferring jurisdiction on an external court or tribunal through an international agreement, the CJEU applies its own conflict rules in order to safeguard its prerogatives. This means that the CJEU sees no problem as long as the court or tribunal interprets and applies solely the provisions of the international agreement. Hence, an international agreement providing for the establishment of a court or tribunal responsible for the interpretation of its own provisions is not, in principle, incompatible with EU law. However, the cornerstone of the CJEU’s approach is that the decisions of the extra-EU courts or tribunals cannot result in spillover effects on the construction and the essential elements of the EU legal order, including the CJEU’s exclusive competence to interpret and apply EU law. If, notwithstanding the above, this is the case, the autonomy of the EU legal order is jeopardised and, therefore, the envisaged dispute settlement mechanism is not compatible with EU law.
In the CJEU’s case-law, the concept of the autonomy of the EU legal order has gained even greater importance over time, and it became a constitutional principle that has been interpreted by the CJEU, particularly in recent years, in a rather expansive way. Therefore, in the following sub-sections, the CJEU’s opinions on the different extra-EU mechanisms will be discussed in more detail, by describing how the principle of autonomy has been advanced prior to Opinion 1/17.
2 The Fund Tribunal Under Opinion 1/76
The first opinion in which the CJEU examined the compatibility of an international dispute settlement mechanism with Community law was Opinion 1/76. This Opinion relates to the draft international agreement establishing a European laying-up fund for inland waterway vessels amongst the Community, six Member States and Switzerland in order to compensate shippers using the Rhine and Moselle basins that withdraw their vessels at times of overcapacity. The agreement set up the Fund Tribunal which was entrusted with the task of ruling on the lawfulness of the decisions of the Fund organs and was also empowered to give preliminary rulings on the validity and interpretation of the Fund organs’ decisions, as well as on the interpretation of the agreement itself. The composition of the Fund Tribunal was also notable because six judges out of the seven were simultaneously members of the CJEU.
Although the CJEU marginally dealt with the potential conflict of jurisdiction between the Fund Tribunal and the CJEU, this issue has not yet been elaborated on in this Opinion. Since the Haegeman decision, it has been well known that an international agreement concluded by the Community could be considered an act of one of the Community’s institutions and, therefore, the CJEU had the competence to give preliminary rulings on the interpretation of the agreement. Nevertheless, based on the agreement in question, it was not clear whether the jurisdiction of the CJEU would be replaced by that of the Fund Tribunal to give a preliminary ruling or whether the two jurisdictions would work in parallel. The CJEU has not made a firm conclusion on the competing competences but only stated that ‘no one can rule out a priori the possibility that the legal organs in question might arrive at divergent interpretations with consequential effect on legal certainty’.
Although the possible conflict of jurisdiction of the Fund Tribunal with the jurisdiction of the CJEU was not entirely resolved in this Opinion, the finding of the CJEU implies that parallel jurisdictions should not be allowed, because it can give rise to diverging interpretations on the same matters. As such, this Opinion paved the way for more extensive reasoning on conflict of jurisdiction in the later opinions of the CJEU.
Besides the above, what the CJEU considered essential from the perspective of compatibility of the envisaged dispute settlement mechanism was the composition of the Fund Tribunal. With regard to the fact that the two adjudicatory bodies consisted of the same members, the CJEU held that the judges were not in a position to ‘give a completely impartial ruling on contentious questions’ because the same legal questions might come before the CJEU after being brought before the Tribunal or vice versa. Broadly speaking, this means that, from an EU law perspective, no personal link is welcome between the CJEU and another international court or tribunal.
Although in Opinion 1/76 the CJEU has not yet expressly referred to the protection of the autonomy of the Community legal order to qualify an international mechanism of dispute settlement as incompatible with Community law, this Opinion is of great importance because it can be regarded as a precursor to the subsequent opinions in which the CJEU has elaborated on the external dimension of autonomy in respect of other extra-EU dispute settlement mechanisms.
3 The EEA Court Under Opinion 1/91
Opinion 1/91 concerned the compatibility of the envisaged European Economic Area (EEA) Court, which was intended to be set up under the agreement on the creation of the EEA amongst the Community, its Member States and the countries of the European Free Trade Association (EFTA). This Opinion is of high importance because it spells out, for the first time, in detail what the autonomy of the Community legal order shall mean from an external relations law perspective and, thus, the CJEU usually relies on the conclusions laid down in this Opinion in its subsequent case-law.
The objective of the agreement was to establish a homogeneous EEA by extending the existing and future Community internal market rules covering the free movement of goods, persons, services and capital, and competition to the entire territory of the EEA. This means in practice that the majority of internal market rules were taken over in the agreement with identically worded provisions. The agreement provided for the establishment of the EEA Court composed of eight judges, including five members of the CJEU.
As a preliminary remark, the CJEU made clear at the beginning of the Opinion that even if the provisions of the agreement were textually identical to the corresponding provisions of Community law, this did not mean that they should necessarily be interpreted identically. On the contrary, the CJEU made a distinction between the Treaties which had ‘established a new legal order for the benefit of which the States have limited their sovereign rights’ and the EEA agreement which had created ‘rights and obligations as between the Contracting Parties’ and provided for ‘no transfer of sovereign rights to the intergovernmental institutions’. Consequently, the CJEU concluded that the contradictions between the objectives and context of the agreement and those of Community law did not secure the aim of homogeneity of the law throughout the EEA.
After these considerations, the CJEU held that the proposed judicial system may undermine the autonomy of the Community legal order for several reasons.
First of all, the EEA agreement qualified as a mixed agreement, meaning that some of the topics covered by the agreement belonged to the shared competence of the Community and the Member States. This means that when the EEA Court settled disputes between the ‘Contracting Parties’, the Court first had to interpret the expression ‘Contracting Party’ in order to determine whether the Community and the Member States, or the Community, or the Member States were covered by the case before the Court. This implied that the Court had to necessarily rule on the respective competences of the Community and the Member States. This task of the Court, however, was contrary to the allocation of responsibilities set forth in the Treaties and, thus, the autonomy of the Community legal order, because this task was exclusively assigned to the CJEU under the Treaties. In other words, the CJEU considered the allocation of responsibilities laid down in the Treaties an essential element of Community law, the alteration of which adversely affected the autonomy of the Community legal order. In addition, the CJEU noted that the exclusive jurisdiction of the CJEU was also confirmed in Article 219 of the EEC Treaty (currently Article 344 TFEU) as regards disputes concerning the interpretation and application of the EEC Treaty between Member States.
The second argument of the CJEU concerned the interpretative power of the EEA Court over the provisions of the EEA agreement, which were strongly identical to the internal market rules of the EEC Treaty. Based on the Haegeman doctrine, the decisions of the EEA Court on the interpretation of the provisions of the EEA agreement were regarded as part of Community law and, thus, binding on the Community institutions, including the CJEU. In connection with this, the CJEU laid down its general statement on the compatibility of an extra-EU dispute settlement mechanism with Community law, which has been repeated in several subsequent opinions:
An international agreement providing for such a system of courts is in principle compatible with Community law. The Community’s competence in the field of international relations and its capacity to conclude international agreements necessarily entails the power to submit to the decisions of a court which is created or designated by such an agreement as regards the interpretation and application of its provisions.
Nevertheless, in Opinion 1/91, the CJEU has placed great emphasis on the fact that the EEA agreement took over the Community internal market rules that were the very core provisions of Community law. Although the EEA agreement’s main objective was to ensure uniform application and homogeneity of the law throughout the entire EEA, the EEA Court had a duty to interpret the provisions of the agreement in the light of the CJEU’s case-law given only prior to the date of signature of the agreement but not after that date. This means that it was not guaranteed under the EEA agreement that the interpretation of the EEA rules would be identical to that of the Community internal market rules. In other words, the EEA Court would have been in a position to interpret not only the provisions of the agreement itself but also the corresponding rules of Community law, which constituted fundamental provisions of the Community legal order and, thus, their interpretation belonged to the exclusive competence of the CJEU.
Consequently, an international court can be considered compatible with EU law as long as its decisions concern only the interpretation and application of the respective international agreement and do not produce any ‘spillover effects’ affecting the fundamental provisions of the EU legal order.
Third, the CJEU went on to say that ‘the organic links between the EEA Court and the Court of Justice by providing that judges from the Court of Justice are to sit on the EEA Court’ would not eliminate the problem. On the contrary, there was a fear that the fact that the same judges should ‘apply and interpret the same provisions but using different approaches, methods, and concepts in order to take account of the nature of each treaty and of its particular objectives’ would even accentuate the problem because the judges would not be able to decide impartially on questions that have already come up before the EEA Court.
Fourth, the CJEU examined the possibility of the EFTA States to authorise their courts or tribunals to refer questions to the CJEU for a preliminary ruling. Although the CJEU concluded that, in principle, there was nothing in the EEC Treaty which would prevent an international agreement from conferring jurisdiction on the CJEU to interpret the provisions of the agreement, it was not acceptable that the answers given by the CJEU were purely advisory and without any binding effects on the courts and tribunals in the EFTA States. The CJEU found that this system, which was capable of having an adverse impact on legal certainty, would have undermined the proper operation of the preliminary ruling procedure within the Community and was, therefore, contrary to the autonomy of the Community legal order.
As a result of Opinion 1/91, the EEA agreement was amended in order to address all the concerns raised by the CJEU. The new version of the agreement has abandoned the idea of the EEA Court and set up a separate court created only for the EFTA States. Since the EFTA Court did not hear cases between the Contracting Parties but its jurisdiction was restricted only to the EFTA States and had ‘no personal or functional links with the Court of Justice’, the CJEU was not bound by the interpretations given by the EFTA Court. Consequently, this new system was able to preserve the autonomy of the EU legal order and, therefore, was approved in Opinion 1/92.
4 The European and Community Patents Court Under Opinion 1/09
The concept of the autonomy of the EU legal order has been further interpreted by the CJEU in Opinion 1/09 and a new aspect has been added to the existing case-law of the CJEU on autonomy, namely the importance of the role that the national courts and tribunals of the Member States play in the correct application and uniform interpretation of EU law.
In this Opinion, the CJEU assessed the compatibility of a European-wide patent court system, the European and Community Patents Court with EU law. The origins of the Opinion can be traced back to the European Patent Convention, which provides for a unitary procedure for European patents to be granted by the European Patent Office, located in Munich. The Convention was signed in 1973 and covered a number of European countries, including all the Member States of the EU. Although the patents granted under the Convention have a Europe-wide validity, the Convention did not provide for an international dispute settlement mechanism. Additionally, the EU has attempted to build up a truly harmonised EU patent system, relying on the already existing and well-functioning regime under the Convention.
In order to ensure the appropriate application and enforcement of this bifurcated regime, the Council drew up a draft international agreement on the establishment of a two-tier European and Community Patents Court (Patents Court) amongst the Member States, the EU and third countries which were parties to the Convention. The Patents Court was given exclusive jurisdiction to hear actions related to European and Community Patents, which necessarily involved the interpretation and application of EU law as well. With regard to this, the Court of First Instance was entitled, while the Court of Appeal was obliged to refer a question to the CJEU for a preliminary ruling if a question of interpretation of EU law arises. Learning from Opinion 1/91, the decision of the CJEU on the interpretation of EU law was binding on the Patents Court.
With regard to the fact that the national courts of the contracting States, including those of the Member States, retained jurisdiction only to the extent that was not subject to the exclusive jurisdiction of the Patents Court, the main legal question that arose in Opinion 1/09 was whether the Member States were allowed to outsource the jurisdiction of their national courts to an international judicial regime that was ‘outside the institutional and judicial framework of the EU’.
The CJEU based its main line of argument on Article 19(1) TEU and Article 267 TFEU to find the Patents Court incompatible with EU law. According to Article 19(1) TEU, the CJEU and the courts and tribunals of the Member States are the guardians of the European legal order and make sure that ‘in the interpretation and application of the Treaties the law is observed’. In other words, the judicial system of the EU, consisting of the CJEU and the domestic courts of the Member States, is a ‘complete system of legal remedies and procedures’ where the national courts, with the assistance of the CJEU, secure the uniform interpretation of EU law in each Member State.
However, the envisaged Patents Court would have altered this essential character of the powers laid down in the Treaties that is indispensable to the preservation of the very nature of EU law. More precisely, under this new dispute settlement system, the Patents Court would have taken the place of national courts and tribunals in the field of its exclusive jurisdiction and would have been called upon to interpret and apply not only the provisions of the agreement in question but also EU patent law and other related instruments of EU law. This implies that the Patents Court would have deprived the national courts of the power to submit questions for a preliminary ruling, which can be considered essential for the preservation of the autonomy of the EU legal order.
In addition, if the Patents Court had misinterpreted or misapplied EU law, no redress would have been available, given that the Patents Court lays outside the institutional and judicial framework of the EU. In other words, in the event of a breach of EU law, neither the infringement proceedings set forth in Articles 258 to 260 TFEU nor the Köbler liability action could have been used to correct the mistake.
In Opinion 1/09, the CJEU concentrated on the essential role that the national courts play in the correct application and uniform interpretation of EU law, as well as the preliminary ruling procedure. The CJEU made it clear that these elements of the EU constitutional order are indispensable; they belong to the very core elements of EU law and, therefore, an extra-EU dispute settlement mechanism that intends to alter these keystones of the EU legal order cannot be compatible with EU law.
5 Compatibility of the EU Accession to the European Convention on Human Rights with EU Law Under Opinion 2/13
Opinion 2/13 is one of the most controversial of the CJEU’s opinions where the protection of the autonomy of the EU legal order has been used to conclude that recourse to an international court was not compatible with EU law. This Opinion was already the second one in which the CJEU dealt with the compatibility of the EU accession to the European Convention on Human Rights (ECHR). In Opinion 2/94, the CJEU already examined this question and concluded that the Community had ‘no competence to accede to the Convention’ and the accession ‘could be brought about only by way of Treaty amendment’. With regard to this, the Lisbon Treaty introduced Article 6(2) TEU, which expressly empowered the EU to accede to the Convention.
Upon the entry into force of the Lisbon Treaty, the negotiations started between the EU and the Council of Europe in relation to an accession agreement in order to agree on all the provisions which were considered necessary for the EU’s accession to the Convention. During the drafting process, special attention was paid to the instruments ensuring the preservation of the autonomy of the EU legal order. Thus, two institutional innovations, i.e. the co-respondent mechanism and the prior involvement procedure, were also introduced in order to make sure that the accession complied with the requirements laid down in the previous opinions of the CJEU.
Nevertheless, despite all efforts, the CJEU was not convinced that the draft accession agreement would preserve the autonomy of the EU legal order and, therefore, concluded that the accession agreement was not compatible with EU law. Although the CJEU based its conclusion on seven grounds, for the purposes of the present paper three of them will be discussed in more detail.
First, the draft accession agreement introduced the co-respondent mechanism in order to address the CJEU’s concern raised in Opinion 1/91 in relation to the division of powers between the EU and its Member States. This procedure allowed both the EU and the Member States to become parties to a procedure initiated before the European Court of Human Rights (ECtHR) if applications were not correctly addressed to the Member States and/or the EU. Even if the purpose of the co-respondent procedure was to prevent the ECtHR from assessing the rules of EU law governing the competences between the EU and its Member States, the CJEU concluded that the design of the co-respondent mechanism was still able to adversely affect the autonomy of the EU legal order. According to the CJEU, the ECtHR still would have had the possibility to indirectly assess EU law on the division of powers because the ECtHR itself would have decided on a request to intervene as co-respondent in a case.
Second, the procedure for the prior involvement of the CJEU was the other institutional innovation of the draft accession agreement. This procedure sought to permit the CJEU to first rule on a question of EU law, thereby producing a binding decision on the ECtHR in a procedure pending before the Strasbourg Court. Thus, the main purpose of this newly introduced procedure was to ensure that ‘the competences of the EU and the powers of its institutions, notably the Court of Justice’, are preserved.
However, the CJEU took the position that the prior involvement procedure foreseen in the accession agreement was not able to achieve this purpose. First, it was not guaranteed that the competent EU institutions could assess whether the CJEU had already given a ruling on the question at issue before the ECtHR but, instead of this, the ECtHR itself was allowed to rule on this question. The CJEU interpreted this feature as conferring jurisdiction on the ECtHR to interpret the case-law of the CJEU. Second, based on the wording of the draft accession agreement, the CJEU drew the conclusion that the scope of the prior involvement procedure was limited, ‘in the case of secondary law, solely to questions of validity’ but it did not include the interpretation of secondary law.
Third, the CJEU examined the relationship between the preliminary ruling procedure set forth in Article 267 TFEU and the advisory opinion mechanism established by Protocol 16. Confirming that the dialogue set up between the CJEU and the national courts and tribunals of the Member States pursuant to Article 267 TFEU is the keystone of the EU judicial system, the CJEU held that by failing to make any provision on the relationship between these two procedures, there was a risk that, by way of the prior involvement procedure, the preliminary ruling procedure might be circumvented. Consequently, the effectiveness of the preliminary reference procedure and, thus, the autonomy of the EU legal order were adversely affected.
Although in this regard the CJEU intended to reinforce Opinion 1/09, it is important to keep in mind that the main justification of the prior involvement procedure was to enable the CJEU to rule on EU law when the procedure under Article 267 had not been triggered. This means that the preliminary ruling procedure would not have been circumvented by way of the prior involvement procedure but the CJEU would instead have been provided with the possibility to determine the correct interpretation and application of EU law.
After Opinion 2/13, the CJEU was heavily criticised as being overly formalistic and protective of its own jurisdiction vis-à-vis other international courts and tribunals, stating that the CJEU has been ‘building up Luxembourg into an excessively armored constitutional fortress’. With regard to the fact that the requirements to preserve the autonomy of the EU legal order has gained, over time, an ever broader interpretation in the opinions of the CJEU, the compatibility of the new ISDS mechanism envisaged in the CETA with EU law was highly questionable. However, the CJEU seems to have revisited its previous hostile attitude towards the external dispute settlement mechanisms and found, in Opinion 1/17, the ICS to be compatible with the autonomy of the EU legal order.
IV The Compatibility of the Investment Court System with the Autonomy of the EU Legal Order
Although Opinion 1/17 concerns the compatibility of the new ISDS mechanism envisaged in the CETA with EU law, the conclusions of the CJEU will also most probably be applicable to future international agreements providing for a dispute settlement mechanism. Thus, since the investor-State dispute resolution provisions set forth in the other new generation bilateral agreements, such as the EU – Singapore and EU – Vietnam Investment Protection Agreements, are highly similar, the conclusions made in Opinion 1/17 are equally applicable to the dispute resolution regime envisaged in these agreements.
The CJEU assessed the compatibility of the new system with the autonomy of the EU legal order from two perspectives. While in the first part of its examination, the CJEU concentrated on the question as to whether the CETA confers on the envisaged tribunals any power to interpret and apply EU law as well as to rule on the division of powers, in the second part, the potential effects of an award rendered by the ICS on the EU constitutional framework were considered. Accordingly, we follow the CJEU’s reasoning and start with the first aspect of the CJEU’s examination.
1 No Jurisdiction to Interpret and Apply EU Law
What was essential to the CJEU in determining as to whether the CETA Tribunal would be empowered to interpret and apply EU law other than the provisions of the CETA was the governing law provision set forth in Article 8.31 of the CETA. Article 8.31.1 of the CETA states that
[w]hen rendering its decision, the Tribunal established under this Section shall apply this Agreement as interpreted in accordance with the Vienna Convention on the Law of Treaties, and other rules and principles of international law applicable between the Parties.
Additionally, Article 8.31.2 of the CETA goes on to say that the ‘Tribunal shall not have jurisdiction to determine the legality of a measure, alleged to constitute a breach of this Agreement, under the domestic law of the disputing Party’.
This means that the CETA Tribunal has no jurisdiction to interpret and apply the domestic law of the disputing Party, including EU law, and the legality of a measure cannot be reviewed under domestic and/or EU law. Instead of this, the Tribunal’s power of interpretation and application is restricted to the provisions of the CETA and the other rules and principles of international law. This is how the Commission sought to preserve the CJEU’s interpretative monopoly over EU law: the ICS, unlike the European and Community Patents Court, does not apply domestic law.
However, when the CETA Tribunal has to render a decision on an infringement of the provisions of the CETA that was committed by a Member State or the EU, it is difficult to imagine how it is possible without applying and interpreting domestic and/or EU law. In response to this concern, Article 8.31.2 of the CETA gives the answer, according to which
[f]or greater certainty, in determining the consistency of a measure with this Agreement, the Tribunal may consider, as appropriate, the domestic law of the disputing Party as a matter of fact. In doing so, the Tribunal shall follow the prevailing interpretation given to the domestic law by the courts or authorities of that Party and any meaning given to domestic law by the Tribunal shall not be binding upon the courts or the authorities of that Party.
Consequently, the CETA Tribunal, when it is called upon to examine the compliance with the CETA of a measure adopted by a Member State or the EU, the Tribunal should consider domestic and/or EU law as a matter of fact, which ‘cannot be classified as equivalent to an interpretation’ of domestic law by the Tribunal.
What is somewhat surprising in relation to Opinion 1/17, compared to the previous opinions of the CJEU, is that the CJEU has accepted this innovative move brought by the Commission and concluded that domestic law being taken into account as a matter of fact suffices to preserve the autonomy of the EU legal order.
This approach, however, raises the question as to what the CETA Tribunal is supposed to do during the consideration of domestic law as a factual matter. Pursuant to the second sentence of Article 8.31.2 of the CETA, the Tribunal shall follow the prevailing interpretation of domestic law given by the courts or authorities of the disputing Party. However, at this point, two further questions may be asked: what happens if there is no prevailing interpretation, or if there is one but the CETA Tribunal does not follow it?
As regards the first question, the CETA has a big deficiency that was not addressed in Opinion 1/17. The drafters of the CETA did not see any need for a similar mechanism to the preliminary ruling or the prior involvement procedure because they intended to solve the question of autonomy of the EU legal order by excluding domestic and/or EU law from the applicable law. Nevertheless, an investor-State dispute under the CETA is likely to involve the interpretation of domestic and/or EU law. This means that if EU law is not clear enough and a determinative interpretation given by the CJEU is not yet available, the CJEU cannot be requested to give a preliminary ruling on the proper interpretation of EU law.
This was accepted by the CJEU in Opinion 1/17, even if it was not in line with its previous jurisprudence. Although, in Opinions 1/09 and 2/13, the CJEU considered the preliminary ruling procedure the keystone of the EU judicial system, which
has the object of securing uniform interpretation of EU law […], thereby serving to ensure its consistency, its full effect and its autonomy as well as, ultimately, the particular nature of the law established by the Treaties
in Opinion 1/17 the CJEU has departed from its long-standing characterisation of the preliminary ruling procedure and only stated that it was
consistent that the CETA makes no provision for the prior involvement of the Court that would permit or oblige that Tribunal or Appellate Tribunal to make a reference for a preliminary ruling to the Court.
Consequently, in Opinion 1/17, the CJEU has not put too much emphasis on the proper operation of the preliminary ruling procedure, even though it is considered, based on its previous opinions, an essential element of the EU constitutional framework that is necessary for the preservation of the autonomy of the EU legal order.
As regards the second question, even if there exists a prevailing interpretation of domestic law, there is no guarantee that the Tribunal will follow that interpretation correctly. In this case, it is the task of the Appellate Tribunal under Article 8.28.2(b) of the CETA to remedy the mistake that occurred in the first instance procedure. Pursuant to this provision, the
Appellate Tribunal may uphold, modify or reverse a Tribunal’s award based on […] (b) manifest errors in the appreciation of the facts, including the appreciation of relevant domestic law.
Although this two-tier mechanism intends to ensure that the well-established case-law of the CJEU will be respected by the CETA Tribunals, it cannot be excluded that EU law and the corresponding jurisprudence, as a matter of fact, will be taken into account wrongly, even by the Appellate Tribunal. Whereas, in Opinion 1/09, this problem appeared in the CJEU’s argumentation and served as a reason to find the European and Community Patents Court incompatible with EU law, in Opinion 1/17 this concern was not even addressed. The problem, however, exists, given the fact that ‘the envisaged ISDS mechanism stands outside the EU judicial system’. This means that even if an award were to be in breach of EU law, the misinterpretation of EU law by the CETA Tribunals could not be sanctioned by way of a financial liability claim or infringement proceedings. Nevertheless, the CJEU, contrary to Opinion 1/09, has not extended the scope of its examination to this question, given its understanding that the CETA Tribunals have no jurisdiction to apply EU law.
The only response that can be given to this concern on the basis of the third sentence of Article 8.31.2 of the CETA is that the interpretative spillover effects on domestic and/or EU law are precluded. In other words, this provision is aimed at making sure that any meaning given to the domestic and/or EU law by the CETA Tribunals will not be binding on the EU and its Member States. Although that is true that, based on this provision, the decisions of the Tribunals will not have any binding effects on the EU and its institutions, the basic problem has not been resolved. In the specific case, the interpretation of EU law as a fact will be binding on the disputing Parties, even if it is contrary to the prevailing jurisprudence of the CJEU.
The last aspect of Opinion 1/17 that demonstrates a further contradiction with the previous opinions of the CJEU is set forth in paragraph 133 of the Opinion which states that
[w]hile Article 8.28.2(b) of the CETA adds that the Appellate Tribunal may also identify ‘manifest errors in the appreciation of the facts, including the appreciation of relevant domestic law’, it is nonetheless clear from the preceding provisions that it was in no way the intention of the Parties to confer on the Appellate Tribunal jurisdiction to interpret domestic law.
This means that while in Opinion 2/13, despite the firm intention of the parties, the CJEU rejected the accession of the EU to the ECHR, in Opinion 1/17, the intention of the parties played the main role in the argumentation of the CJEU to justify that the Appellate Tribunal would have no jurisdiction to interpret EU law. Consequently, this is, again, a new element of the CJEU’s argumentation that has never appeared before in its jurisprudence on autonomy.
To summarise, when the CJEU found the new ISDS mechanism compatible with the autonomy of the EU legal order, its line of argument was based on the premise that the CETA Tribunals would have no jurisdiction to interpret and apply EU law. However, as demonstrated above, the potential interpretation of EU law cannot be ruled out under the CETA, even if EU law should be taken into account only as a matter of fact. Consequently, if the CJEU had followed its rigid case-law on autonomy, the envisaged ICS could not have been qualified as being compatible with EU law. Nevertheless, in Opinion 1/17, the CJEU has reinterpreted the requirements necessary for the preservation of the autonomy of the EU legal order and, thus, saved the Commission’s project to fundamentally reform the field of ISDS mechanisms.
2 No Jurisdiction to Rule on the Division of Powers
The question of who is entitled to determine the proper respondent in a specific dispute in the case of mixed agreements has been a constantly recurring topic in the case-law of the CJEU on autonomy from the early 1990s. Considering the fact that the CETA is a mixed agreement, the CJEU also touched upon this question in paragraph 132 of Opinion 1/17.
The legal provision that the CJEU relied on during its examination was Article 8.21 of the CETA which expressly confers on the EU the power to determine ‘whether the dispute is, in the light of the rules on the division of powers between the Union and its Member States, to be brought against [the] Member State or against the Union’. Consequently, it is not the CETA Tribunal but solely the EU who can decide on the proper respondent in each case and this decision of the EU is binding on the Tribunal. Since this is exactly what the previous opinions, such as Opinion 1/91 and Opinion 2/13 required, it is not surprising that the CJEU reaffirmed the compatibility of this system with EU law. Thus, in this respect, Opinion 1/17 is consistent with the previous jurisprudence on autonomy.
3 The Level of Protection of the Public Interest
Finally, in the second part of its examination, the CJEU addressed the question as to whether the awards of the Tribunal had ‘the effect of preventing the EU institutions from operating in accordance with the EU constitutional framework’. In other words, the question concerned the right to regulate, namely whether the parties had the right under the CETA to regulate within their territories to achieve legitimate policy objectives, inter alia, the protection of public order, public safety or the protection of public morals.
This question added a further, substantive aspect to the concept of autonomy because, in the previous opinions, the CJEU usually examined the preservation of the autonomy of the EU legal order from a jurisdictional/procedural perspective and did not extend the scope of its review to substantive issues.
Based on several substantive provisions of the CETA, the CJEU concluded that the CETA standards of protection, similarly to the EU investment treaty practice, respected state sovereignty because the power of the CETA Tribunals did not ‘call into question the level of protection of public interest determined by the Union following a democratic process’ but rather allowed ‘the Union to operate autonomously within its unique constitutional framework’. Consequently, the CJEU reached the conclusion that the investment Chapter of the CETA did not adversely affect the autonomy of the EU legal order.
Although the autonomy of the EU legal order is not mentioned in the EU Treaties, in the jurisprudence of the CJEU this concept has become a constitutional principle that has been used to safeguard the essential elements of EU law from interference flowing from both national and international legal orders. While it is self-evident that the CJEU sought to protect the very foundations of the EU constitutional framework from any external influences and, to this end, relied on the principle of autonomy in its argumentation, over the past years, the CJEU has taken a quite restrictive approach when it came to the participation of the EU and its Member States in an international dispute settlement mechanism. Hence, it was questionable whether the new ISDS mechanism proposed in the CETA would be able to reach the high threshold which had been set up by the CJEU in a series of opinions.
However, the CJEU, in Opinion 1/17, reconsidered its previous jurisprudence on autonomy and, with a new line of argument, placed the EU back into the path of international dispute settlement. With regard to the obvious turnaround, right after the CJEU had delivered Opinion 1/17, several authors wondered why the CJEU had departed from its previous decisions and held that the ICS did not adversely affect the autonomy of the EU legal system.
In our view, it must be accepted that the CJEU is not only a legal but also a political institution which is pursuing its political programme. In other words, the CJEU made a clear political decision in Opinion 1/17, namely that the protection of the investors’ individual rights should take precedence over autonomy. This political consideration clearly appears in the Opinion of Advocate General Bot, who highlighted that
[i]n order to rule on the compatibility of the dispute settlement mechanism provided for in Section F of Chapter 8 of the CETA with EU primary law, it is […] necessary to broaden the perspective and to take account of the need to protect EU investors when they invest in third States.
Consequently, with Opinion 1/17, the CJEU wanted to ensure that there would be a neutral and independent forum from the domestic courts and tribunals of the host State where the EU investors may enforce their rights arising from the CETA.
Although the purpose that the CJEU wished to achieve is welcome, one must conclude that, after Opinion 1/17, the concept of autonomy no longer has a uniform interpretation but it must be interpreted differently depending on the actual context surrounding it. This means that autonomy has remained ‘partially nebulous’ even after Opinion 1/17 and its exact meaning can be determined only on a case-by-case basis.
* Aliz Káposznyák is doctoral candidate and assistant lecturer at the Department of Civil Procedure, Eötvös Loránd University (ELTE), Faculty of Law, Budapest, Hungary (email address: email@example.com).
** The present paper is based on the presentation given by the author at the 12th Annual Legal Research Network Conference in Budapest on 19 September 2019.
 Opinion 1/17 of the Court (Full Court) of 30 April 2019  OJ C220/02 (Opinion 1/17).
 Comprehensive Economic and Trade Agreement between Canada, of the one part, and the European Union and its Member States, of the other part, Chapter 8, Section F <http://ec.europa.eu/trade/policy/in-focus/ceta/ceta-chapter-by-chapter/> accessed 3 May 2020 (CETA).
 Although at first glance the CJEU’s opinion may seem only advisory, in case of an adverse opinion, ‘the agreement envisaged may not enter into force unless it is amended or the Treaties are revised’.
 Consolidated version of the Treaty on the Functioning of the European Union  OJ C326/1.
 Opinion 1/91 of the Court of 14 December 1991  ECR I-6079, paras 46, 71 (Opinion 1/91).
 Opinion 1/09 of the Court (Full Court) of 8 March 2011  OJ C211/03, paras 85, 89 (Opinion 1/09); Opinion 2/13 of the Court (Full Court) of 18 December 2014,  OJ C065/02, para 212 (Opinion 2/13).
 Opinion 1/91 (n 5) para 21; Opinion 1/00 of the Court of 18 April 2002  ECR I-3493 paras 14, 18, 21, 23, 26 (Opinion 1/00).
 Cristina Contartese, ‘The autonomy of the EU legal order in the ECJ’s external relations case law: From the “essential” to the “specific characteristics” of the Union and back again’ (2017) 54 (6) Common Market Law Review 1627–1672, 1627.
 Bruno de Witte, ‘A Selfish Court? The Court of Justice and the Design of International Dispute Settlement Beyond the European Union’ in Marise Cremona, Anne Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Hart 2014, Oxford-Portland, 33–46) 33ff.
 Paul Gragl, ‘The Reasonableness of Jealousy: Opinion 2/13 and EU Accession to the ECHR’ in Wolfgang Benedek and others (eds), European Yearbook of Human Rights (Neuer Wissenschaftlicher Verlag 2015, Vienna, 27–50) 27ff.
 Case C-284/16 Slowakische Republik v Achmea BV  ECLI:EU:C:2018:158.
 Ibid, para 57, emphasis added.
 Burkhard Hess, ‘The Fate of Investment Dispute Resolution after the Achmea Decision of the European Court of Justice’ MPILux Research Paper 3/2018, 18 <https://www.mpi.lu/fileadmin/mpi/medien/events/2018/4/3/WPS3_2018_The_Fate_of_Investment_Dispute_Resolution_after_the_Achmea_Decision_of_the_European_Court_of_Justice> accessed 3 May 2020.
 Opinion 1/17 (n 1) para 161.
 UNCTAD, ‘World Investment Report Reforming International Investment Governance’ (2015) 128 <http://unctad.org/en/PublicationsLibrary/wir2015_en.pdf> accessed 3 May 2020 (World Investment Report 2015).
 Gabrielle Kaufmann-Kohler, Michele Potestà, ‘Can the Mauritius Convention serve as a model for the reform of investor-State arbitration in connection with the introduction of a permanent investment tribunal or an appeal mechanism? – Analysis and roadmap’ (2016) CIDS – Geneva Center for International Dispute Settlement Research Paper (2016), 12–14 <http://www.uncitral.org/pdf/english/CIDS_Research_Paper_Mauritius.pdf> accessed 3 May 2020; Michael Waibel and others, ‘The Backlash against Investment Arbitration: Perceptions and Reality’ in Michael Waibel and others (eds), The Backlash against Investment Arbitration (Kluwer Law International 2010, xxxvii-li) xxxviii; Richard Happ, Sebastian Wuschka, ‘From the Jay Treaty Commissions Towards a Multilateral Investment Court: Addressing the Enforcement Dilemma’ (2017) VI Indian Journal of Arbitration Law 113–132, 113; World Investment Report 2015 (n 15) 128; Giovanni Zarra, ‘Orderliness and Coherence in International Investment Law and Arbitration: An Analysis Through the Lens of State of Necessity’ (2017) 34 (4) Journal of International Arbitration 653–678, 656.
 Art 3(1) e) and 207(1) TFEU.
 Art 8.29 CETA.
 Draft Text of the EU-Mexico Trade Agreement, Section [X] Resolution of Investment Disputes, <https://trade.ec.europa.eu/doclib/docs/2018/april/tradoc_156814.pdf> accessed 3 May 2020.
 Investment Protection Agreement between the European Union and its Member States, of the one part, and the Republic of Singapore, of the other part, Chapter 3 <https://trade.ec.europa.eu/doclib/press/index.cfm?id=961> accessed 3 May 2020.
 Investment Protection Agreement between the European Union and its Member States, of the one part, and the Socialist Republic of Vietnam, of the other part, Chapter 3 <https://trade.ec.europa.eu/doclib/press/index.cfm?id=1437> accessed 3 May 2020.
 Elsa Sardinha, ‘The New EU-Led Approach to Investor-State Arbitration: The Investment Tribunal System in the Comprehensive Economic Trade Agreement (CETA) and the EU-Vietnam Free Trade Agreement’ (2017) 32 (3) ICSID Review – Foreign Investment Law Journal 625–672, 626; N. Jansen Calamita, ‘The Challenge of Establishing a Multilateral Investment Tribunal at ICSID’ (2017) 32 (3) ICSID Review – Foreign Investment Law Journal 611–624, 612; Art 8.28 2. a)–b) CETA.
 Art 26.1(1) CETA.
 Art 8.27(7) CETA; Michele Potestà, ‘Chapter IV: Investment Arbitration, Challenges And Prospects For The Establishment Of A Multilateral Investment Court: Quo Vadis Enforcement?’ in Christian Klausegger and others (eds), Austrian Yearbook on International Arbitration (Manz’sche Verlags- und Universitätsbuchhandlung 2018, Vienna, 157–178) 162.
 Commission, ‘Singapore: The Commission to Request a Court of Justice Opinion on the trade deal’ (Press Release) IP/14/1235 <https://ec.europa.eu/commission/presscorner/detail/en/IP_14_1235> accessed 3 May 2020.
 Opinion 2/15 of the Court (Full Court) of 16 May 2017  OJ C239/03, para 305.
 Ibid para 30, repeated in paras 290, 300.
 Request for an opinion submitted by the Kingdom of Belgium pursuant to Article 218(11) TFEU (Opinion 1/17)  OJ C369/2.
 Case C-26-62 NV Algemene Transport- en Expeditie Onderneming van Gend & Loos v Netherlands Inland Revenue Administration  ECR 1.
 Case C-6/64 Flaminio Costa kontra E.N.E.L.  ECR 585.
 Steffen Hindelang, ‘The Autonomy of the European Legal Order – EU Constitutional Limits to Investor-State Arbitration on the Basis of Future EU Investment-Related Agreements’ in Marc Bungenberg, Christoph Herrmann (eds), Common Commercial Policy after Lisbon, Special Issue to the European Yearbook of International Economic Law (Springer 2013, Berlin–Heidelberg, 187–198) 189; Szilárd Gáspár Szilágyi, ‘A Standing Investment Court under TTIP from the Perspective of the Court of Justice of the European Union’ (2016) 17 (5) Journal of World Investment & Trade 701–742, 704.
 Jed Odermatt, ‘When a Fence Becomes a Cage: The Principle of Autonomy in EU External Relations Law’ (2016) European University Institute Working Papers, Max Weber Programme 2016/17, 1 <https://cadmus.eui.eu/bitstream/handle/1814/41046/MWP_2016_07.pdf?sequence=1> accessed 3 May 2020.
 Joined cases C-402/05 P and C-415/05 P Yassin Abdullah Kadi and Al Barakaat International Foundation v Council of the European Union and Commission of the European Communities  ECR I-06351.
 Opinion 1/91 (n 5); Opinion 1/00 (n 7); Opinion 1/09 (n 6); Opinion 2/13 (n 6); Christina Eckes, ‘International Rulings and the EU Legal Order: Autonomy as Legitimacy?’ (2016) CLEER PAPERS 2016/2, 12 <https://www.asser.nl/media/3002/cleer16-2_complete_web.pdf> accessed 3 May 2020.
 Consolidated version of the Treaty on European Union  OJ C326/1.
 Yuval Shany, The Competing Jurisdictions of International Courts and Tribunals (Oxford University Press 2003, Oxford) 180.
 Art. 3(5) and 21(1) TEU. Besides, this political will is very well illustrated by Art. 6(2) TEU, which, taking into account the findings of the CJEU in Opinion 2/94, contains the necessary competence for the EU to accede to the European Convention on Human Rights.
 de Witte (n 9) 34; Szilágyi (n 31) 705.
 Opinion 2/13 (n 6) para 246; Opinion 1/17 (n 1) para 111.
 C-181/73 R. & V. Haegeman v Belgian State  ECR 449.
 de Witte (n 9) 34–35; Jan Wouters, Jed Odermatt, Thomas Ramopoulos, ‘Worlds Apart? Comparing the Approaches of the European Court of Justice and the EU Legislature to International Law’ in Marise Cremona, Anne Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Hart 2014, Oxford–Portland, 249–280) 253.
 Haegeman case, para 5.
 de Witte (n 9) 35.
 Opinion 1/91 (n 5) para 40; Opinion 1/09 (n 6) para 74; Opinion 2/13 (n 6) para 182.
 Steffen Hindelang, ‘Repellent Forces: The CJEU and Investor-State Dispute Settlement’ (2015) 53 (1) Archiv des Völkerrechts 68–89, 73.
 Christian Riffel, ‘The CETA Opinion of the European Court of Justice and its Implications – Not that Selfish After All’ (2019) 22 (3) Journal of International Economic Law 503–521, 505.
 Opinion 1/76 of the Court of 26 April 1977  OJ C107/4 (Opinion 1/76).
 de Witte (n 9) 35.
 Opinion 1/76 (n 47) para 17.
 Barbara Brandtner, ‘The “Drama” of the EEA – Comments on Opinions 1/91 and 1/92’ (1992) 3 (2) European Journal of International Law 300–328, 312.
 Ibid, para 20.
 Ibid, para 22.
 Opinion 1/91 (n 5) para 4.
 Ibid, para 21.
 Ibid, para 20.
 Ibid, paras 28–29.
 Marco Bronckers, ‘The Relationship of the EC Courts with Other International Tribunals: Non-committal, Respectful or Submissive?’ (2007) 44 Common Market Law Review 601–627, 606.
 Opinion 1/91 (n 5) paras 33–34.
 Ibid, para 35; Henry G. Schermers, ‘Opinion 1/91 of the Court of Justice, 14 December 1991; Opinion 1/92 of the Court of Justice, 10 April 1992’ (1992) 29 (5) Common Market Law Review 991–1009, 996; Bronckers (n 57) 606.
 Opinion 1/91 (n 5) para 39.
 Ibid, para 40.
 Ibid, paras 43–44.
 Hindelang (n 31) 190; Szilágyi (n 31) 711.
 Ibid, para 47.
 Ibid, para 51.
 Ibid, para 52; Schermers (n 59) 998.
 Opinion 1/91, (n 5) paras 59, 61.
 Opinion 1/92 of the Court of 10 April 1992  ECR I-2821, paras 13, 19.
 de Witte (n 9) 42.
 Ibid; Opinion 1/09 (n 6) paras 7, 73.
 Opinion 1/09 (n 6) para 71; Allan Rosas, ‘The National Judge as EU Judge: Opinion 1/09’ in Pascal Cardonnel, Allan Rosas, Nils Wahl (eds), Constitutionalising the EU Judicial System: Essays in Honour of Pernilla Lindh (Hart 2012, Oxford, 105–121) 112.
 Opinion 1/09 (n 6) paras 66, 69.
 Ibid, para 70.
 Ibid, paras 85, 89.
 Ibid, para 79.
 Ibid, paras 86–88; Nikolaos Lavranos, ‘Designing an International Investor-to-State Arbitration System After Opinion 1/09’ in Bungenberg, Herrmann, Common Commercial Policy after Lisbon (n 31) 214.
 Case C-224/01 Gerhard Köbler v Republik Österreich  ECR I-10239, paras 31, 33, 36.
 Opinion 2/94 of the Court of 28 March 1996  ECR I-1759, paras 34–36.
 Although Opinion 2/13 had postponed the EU accession to the ECHR, the negotiations on this topic between the EU and the Council of Europe have recently restarted and an ad hoc group, composed of representatives of the 47 Member States of the Council of Europe and a representative of the EU, has been set up in order to find solutions to all the legal issues raised by the CJEU in Opinion 2/13. For more details, see <https://www.coe.int/en/web/human-rights-intergovernmental-cooperation/accession-of-the-european-union-to-the-european-convention-on-human-rights> accessed 3 May 2020.
 Opinion 2/13 (n 6) paras 215–216.
 Ibid, paras 222–225.
 Ibid, paras 236–238.
 Ibid, paras 239–241; Stephan W. Schill, ‘Editorial: Opinion 2/13 – The End for Dispute Settlement in EU Trade and Investment Agreements?’(2015) 16 (3) The Journal of World Investment & Trade 379–388, 382.
 Opinion 2/13 (n 6) paras 242–247.
 Ibid, paras 176, 198.
 Ibid, para 199.
 Contartese (n 8) 1656.
 Schill (n 83) 379.
 Marc Bungenberg, Catharine Titi, ‘CETA Opinion – Setting Conditions for the Future of ISDS’ (5 June 2019) EJIL: Talk! <https://www.ejiltalk.org/ceta-opinion-setting-conditions-for-the-future-of-isds/> accessed 3 May 2020; Riffel (n 46) 507.
 Opinion 1/17 (n 1) para 119.
 Emphasis added.
 Daniele Gallo, Fernanda G. Nicola, ‘The External Dimension of EU Investment Law: Jurisdictional Clashes and Transformative Adjudication’ (2016) 39 (5) Fordham International Law Journal 1081–1152, 1125–1126 <https://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=2454&context=ilj> accessed 3 May 2020.
 Emphasis added.
 Opinion 1/17 (n 1) para 131.
 Ibid, para 131; Francisco de Abreu Duarte, ‘Autonomy and Opinion 1/17 – a matter of coherence?’ (31 May 2019) European Law Blog <https://europeanlawblog.eu/2019/05/31/autonomy-and-opinion-1-17-a-matter-of-coherence/> accessed 3 May 2020.
 Gallo, Nicola (n 92) 1132.
 Riffel (n 46) 516.
 Opinion 2/13 (n 6) para 176.
 Opinion 1/17 (n 1) para 134.
 Emphasis added.
 Opinion 1/17 (n 1) para 113; de Abreu Duarte (n 95).
 Riffel (n 46) 517; de Abreu Duarte (n 95).
 Emphasis added.
 de Abreu Duarte (n 95).
 Carlo Favaretto, ‘Beyond Selfishness: The Court of Justice in Opinion 1/17 on CETA’ (2019) Diritti Comparati <http://www.diritticomparati.it/beyond-selfishness-court-justice-opinion-1-17-ceta/> accessed 3 May 2020.
 Opinion 1/17 (n 1) para 132.
 Ibid, para 119.
 Panos Koutrakos, ‘More on Autonomy – Opinion 1/17 (CETA)’ (2019) (3) European law review <https://dialnet.unirioja.es/servlet/articulo?codigo=6976614> accessed 3 May 2020; Riffel (n 46) 519.
 See, for example, Articles 8.9 – 8.10 of the CETA, Annex 8-A to the CETA.
 Opinion 1/17 (n 1) paras 150, 156; Bungenberg, Titi (n 89).
 See, for example, Gisèle Uwera, ‘Investor-State Dispute Settlement (ISDS) in Future EU Investment-Related Agreements: Is the Autonomy of the EU Legal Order an Obstacle?’ (2016) 15 (1) The Law and Practice of International Courts and Tribunals 102–151, 150–151; Szilágyi (n 31) 723–739; Contartese (n 8) 1671.
 Bungenberg, Titi (n 89).
 See, for example, de Abreu Duarte (n 95); Favaretto (n 105).
 Riffel (n 46) 520.
 Opinion 1/17 – Request for an opinion by the Kingdom of Belgium, Opinion of Advocate General Bot delivered on 29 January 2019, ECLI:EU:C:2019:72, para 89, emphasis added.
 de Abreu Duarte (n 95); Koutrakos (n 108).