Wightman and the Perils of Britain’s Withdrawal

Abstract

On 10 December 2018, the Court of Justice (CJEU) delivered the Wightman judgment and recognized the unilateral revocability of the notification ex Art. 50 Treaty on European Union (TEU). This article offers a critical analysis of the decision by insisting above all on the national background of the ruling and the political risks stemming from the decision. The article is structured as follows. Firstly, it analyses the legal questions of the Scottish case, which constituted the ground for the admissibility of the preliminary ruling and showed the perils for the exercise of national sovereign rights embedded in the lack of clarity on revocation options. It thus reconstructs the critical aspects of the preliminary ruling of the CJEU. Subsequently, the article examines the implications of the ruling for the EU legal order. On the one hand, the analysis considers the conception of the EU membership by comparing the approach of the CJEU and that of Advocate General Campos Sánchez Bordona in Wightman.

A. Introduction

On December 10, 2018, the Court of Justice (CJEU) delivered the Wightman judgment¹ and recognized the unilateral revocability of a notification under Article 50 of the Treaty on European Union (TEU).² The CJEU tackled the wider question of its own foundations and decided on a number of crucial aspects governing the most incremental of the challenges—the departure of a Member State from the European Union (EU).

The process of the United Kingdom (UK) withdrawing its membership from the EU has revealed itself to be a constitutional crisis affecting both the identity of the European integration, and that of the British Constitution.³ Alongside the Brexit negotiation process, the difficult management and controversial interpretation of its legal implications called for the intervention of the CJEU itself.

In this Article, we will offer a reflection on some of the questions raised—and not resolved—by the Wightman decision. The respect for national sovereignty is the key concern emerging from both the Scottish reference and the CJEU’s ruling. Undoubtedly, while most of the judgment is reasonable, the risk of the strategic use of Article 50 TEU has not been, in our opinion, completely averted. To stem this danger, the CJEU defined this potential revocation as “unequivocal and unconditional.”⁴ Against this background, the unconditional nature of the revocation is to be understood as meaning that its purpose is “to confirm the EU membership of the Member State concerned under terms that are unchanged as regards its status as a Member State, and that revocation brings the withdrawal procedure to an end.”⁵ This decision’s insistence on the concept of national sovereignty—a mantra of today’s populisms that also characterized the referendum campaign—is less convincing, however.⁶ This is the background to a judgment of the CJEU, which itself is characterized by an ambiguous terminology and strange omissions.

When offering a legal device to rescind Brexit, the relevance of Wightman goes beyond the British case and points to the very essence of the EU legal order, and the nature of the EU membership. The case pinpoints unilateral revocability as a sovereign right that is not subject to any conditions, but is instead a genuine acceptance and commitment to the values and the rules of the EU legal order.

Despite this, the Wightman case was problematic both in its genesis and outcomes. Firstly, the path that led to the preliminary ruling was controversial. The legal relevance of the questions under the national Scottish case struggled to be recognized, and only the Court of Session, Inner House, First Division (Scotland) on appeal raised the preliminary reference. Secondly, the CJEU’s ruling—despite its continuity with its previous case law—did not unravel the perils of unilaterality. By highlighting the relevant factors that contributed to shaping the preliminary question, this Article aims to discuss the relevant difficult aspects of the Wightman case.

This Article offers a critical analysis of the decision by insisting, above all, on the national background of the ruling and the political risks stemming from the decision. The Article is structured as follows: Firstly, it analyzes the legal questions of the Scottish case, which constituted the ground for the admissibility of the preliminary ruling and showed the perils for the exercise of national sovereign rights embedded in the lack of clarity on revocation options. Thus, it reconstructs the critical aspects of the preliminary ruling of the CJEU. Subsequently, this Article examines the implications of the ruling for the EU legal order. On the one hand, the analysis considers the conception of the EU membership by comparing the approach of the CJEU and that of Advocate General Campos Sánchez-Bordona in Wightman. On the other hand, unilateral revocability is analyzed and its perils are highlighted. We conclude the Article by highlighting some of the questions that remain unanswered after Wightman.

B. The Domestic Background: The Scottish Case for Revocation

In the conundrum of Brexit negotiations, some members of the Scottish Parliament, the UK Parliament and the European Parliament felt a responsibility to explore all the potential consequences of Article 50 TEU. They thus petitioned the Scottish Court of Session for a declarator clarifying “whether, when and how the notification … can unilaterally be revoked.”⁷ They essentially sought a legally valid alternative to both deal and non-deal Brexit. The declarator is a public law remedy aimed at advising a public body on the meaning of the law and its application to the specific case. Is there a legal remedy to stop the cut of the “conduit pipe” established by the European Communities Act (ECA) 1972 that changed the British Constitution?⁸ By addressing this key constitutional question, the action aimed at unravelling the Brexit conundrum. Seeing that the existence of such a remedy was calling for a uniform interpretation of Article 50 TEU, the petitioners sought a reference to the CJEU under Article 267 TFEU.

Yet, their case struggled to be heard in Scottish courts and was referred to the CJEU only after appeal in the national court.⁹ When the case was decided in the first instance before the Outer House of the Scottish Court of Session on June 8, 2018, the petitioners were unsuccessful because Lord Boyd of Duncansby declined to refer the case on three different grounds: The hypothetical nature of the declaration sought; the encroachment of the question on parliamentary sovereignty; and the impossibility to ascertain the facts on which the case was filed.¹⁰ Lord Boyd of Duncansby held that there was no proposal nor intention to revoke the notification from the Government.¹¹ Petitioners’ action was based on parliamentary discussion, so that in the Lord Ordinary’s view,¹² it was up to Parliament itself and not to the courts to decide the options on Brexit.¹³ In addition, according to his reading of the admissibility test before the CJEU in Gauweiler ¹⁴ and American Express,¹⁵ the case did not meet the necessary factual requirements, nor the interpretation of EU law being critical to the decision of the case.¹⁶

The appeal judges sitting in the Inner House of the Scottish Court of Session, however, noted that the circumstances of the case had significantly changed after the decision in the Outer House. With the royal assent on the EU Withdrawal Act of June 26, 2018, the legal framework under which Brexit was to be domestically managed and negotiated changed. Under Section 13 of the EU Withdrawal Act 2018, the UK Parliament was required to express a “meaningful vote” on the international agreement between the UK Government and the European Council.¹⁷ If no approval were obtained, the Government would have needed to explore new solutions, such as engaging in new negotiations, proposing non-deal Brexit and—why not?—revoking the notification if it were a legally viable option. The petitioners claimed that “first, the Lord Ordinary erred in holding that the issues were academic or hypothetical.” They argued “the issue was of great constitutional importance” in so far as “the issue was directly relevant to the parliamentarians’ decisions on how to vote.” For example, “if a decision to remain was available as a matter of EU law, the UK Parliament could pursue that option irrespective of Government policy. The Lord Ordinary’s decision would mean that parliamentarians would have to vote in ignorance of this.”¹⁸

By understanding how many strings MPs have to their bows, the petitioners first pointed to the nature of national sovereignty under EU law. The issue of legality of unilateral revocation can be raised with no consequence on Parliamentary sovereignty, because it is a matter of EU law.¹⁹ The declarator would comply with the principle of primacy of EU law, which does not compromise Parliamentary sovereignty.²⁰ Conversely, the respondent Secretary of State for Exiting the European Union claimed that in the absence of any “real prospect of the Government seeking to revoke the notification,” there was “no genuine dispute about the proper construction of Article 50(2).” Thus, Parliament would be required to vote on the negotiated withdrawal agreement under Section 13 of the EU Withdrawal Act, without any “need to resolve any legal uncertainty.”²¹ The interpretative question is able to be raised only at the international level, by the Member States, and not by individuals. In this case, Parliament did not intend to ask Government to raise such an issue. Therefore, this situation did not meet the requirements for advisory declarators “to ensure that a decision maker did not exceed or abuse his powers.”²²

Yet, the petitioners claimed that “secondly, the Lord Ordinary had erred in holding that the court could not provide an advisory declarator on the legality of future or contingent action,” because the court’s interpretation would not interfere with parliamentary sovereign choice and proceedings. Thus, “the court would be fulfilling its constitutional obligation to maintain the rule of law.” This included the concern surrounding the meaning of the applicable EU law and avoiding a “pointless” Parliamentary decision in contrast with EU law.²³

The petitioners also submitted that “thirdly, the Lord Ordinary had erred in holding that referring to statements in Parliament was unconstitutional, unlawful or incompetent,” because the petition did not question those statements in courts against Parliamentary privilege.²⁴ “Fourthly, the Lord Ordinary had erred in determining that the CJEU would not entertain a reference,” because it aims to “ensure access to justice” and only exceptionally denies such a reference.²⁵.

C. From Edinburgh to Luxembourg: The Content of the Preliminary Reference

When responding to these pleas, the Inner Court of Session recognized their validity and admitted the preliminary reference on the following grounds: The relevance of the remedy sought; the judicial competence to ascertain its existence; and the need to settle the case by asking the CJEU for the correct interpretation of the provision under Article 50 TEU. Once the Inner Court of Session ascertained that parliamentary sovereignty was not compromised by the legal action, the exercise of sovereign rights in Parliament depended on the existence of limits under EU law for the revocation of the withdrawal. The action of the declarator was found to be “neither academic nor premature” under the EU Withdrawal Act 2018, as “the answer will have the effect of clarifying the options open to MPs in the lead up to what is now an inevitable vote.”²⁶

Lord Menzies particularly stressed the constitutional relevance of the case and the need that the judiciary make certain the legal meaning of the provision in order to enable the decision-makers to make their choice.²⁷ The fact that the Government did not intend to revoke the notification of withdrawal, or that Parliament had not instructed it to do so, did not change the practical need for the declarator. In fact, “constitutionally, and in terms of the 2018 Act, it is a matter for Parliament to decide, not for the UK Government. The intention of the Government may be relevant in terms of factual background, but it cannot be determinative of this issue.”²⁸ Lord Drummond Young also emphasized that the position of the Government might change because of unforeseen developments and pragmatically stressed that “so far as the withdrawal from the European Union is concerned, at this stage it appears impossible to be certain as to what will happen in the coming months.”²⁹ In his view, no deal as a default option would have severe legal consequences, whose impact was hard to predict “from a professional legal perspective.”³⁰ Lord Drummond Young also recognized that Brexit is “plainly a matter of enormous importance for the United Kingdom, constitutionally, economically and in numerous other ways. … In these circumstances, ascertaining the legal principles that apply to the use of Article 50 and its consequences are a matter of great practical importance.”³¹

The petitioners had an interest in identifying the correct meaning of Article 50(2) TEU. Following Lord Reed’s wider approach to accessing justice in the 2011 Supreme Court case AXA v Lord Advocate,³² Lordship Carloway clarified that the role of the courts is “to provide rulings on what the law is and how it should be applied” and the existence of “sufficient interest” in the solution of a case, depending upon whether the context justifies standing.³³ The Court of Session considered that the existence of sufficient interest in the case—which might not be identifiable with an individual right—was enough to act in courts, even if it did not cover the existence of such an individual right.

Having ascertained the existence of such an interest and the relevance of the action, the Court of Session considered that the issue of a declarator in such a matter would not be in breach of parliamentary privilege:

The Court is not advising Parliament on what it must, or ought to, do. It is not otherwise seeking to influence Parliament’s direction of travel. It is merely declaring the law as part of its central function. How Parliament chooses to react to that declarator is entirely a matter for that institution.³⁴

According to Lord Drummond Young, under the British Constitution, the judiciary shall “decide the law as it now exists” and “their decision is binding on both Parliament and the government; that is an elementary application of the rule of law.”³⁵ Parliamentary privilege aims to protect free speech in Parliament and the ability of MPs to vote, while ensuring the separation of powers.³⁶ Every constitutional body holds specific constitutional functions, which shall be exercised with mutual respect.³⁷ Insofar as no change of law is involved in the interpretation of Article 50 TEU, the decision is thus not for Parliament. Nor is the decision for Government because its absence of any intention to revoke the notification is only a political matter.³⁸ Instead, the courts must decide on the interpretation of the law.

Thus, the Court of Session emphasized that the case “can only be answered definitively by the CJEU” and only after the preliminary ruling could the Scottish court “grant the appropriate declarator.”³⁹ The Court of Session held the reference as necessary and determined that in the spirit of cooperation between courts, the CJEU should give the preliminary ruling. By relying on American Express and Gauweiler—the cases already mentioned in the first instance case as precedents on the admissibility of references to the CJEU—the Court of Session held that it is for the national court to assess both the need for a preliminary ruling and the relevance of the questions to solve the national case, while the CJEU is in principle bound to give a ruling.⁴⁰ However, on the top of the legal admissibility of the question, the Court of Session also recalled a pragmatic approach to judicial cooperation. Insofar as Scottish courts have always been careful when referring cases to the CJEU—only doing so ten times in 45 years—and have worked to solve disputes involving EU law on their own “without troubling the CJEU … it would be disappointing if a rare request for assistance were to be met with a negative response.”⁴¹

The CJEU responded positively to such a cooperation request, by pointing out the responsibility of the national court referring the case “to determine in the light of the particular circumstances of the case, both the need for a preliminary ruling in order to enable it to deliver judgment and the relevance of the questions which it submits to the Court.” Thus, the CJEU, recalling the same Gauweiler and American Express cases mentioned by the Court of Session, held that it “is in principle bound to give a ruling.”⁴² In the CJEU’s words:

It follows that questions relating to EU law enjoy a presumption of relevance. The Court may refuse to rule on a question referred for a preliminary ruling by a national court only where it is quite obvious that the interpretation of EU law that is sought bears no relation to the actual facts of the main action or its purpose, where the problem is hypothetical, or where the Court does not have before it the factual or legal material necessary to give a useful answer to the questions submitted to it (judgments of 16 June 2015, Gauweiler and Others, C-62/14, EU:C:2015:400, paragraph 25, and of 7 February 2018, American Express, C-304/16, EU:C:2018:66, paragraph 32).⁴³

As a result, the CJEU reconnected the general or hypothetical nature of the preliminary reference to its irrelevance for “the effective resolution of a dispute.”⁴⁴ Insofar as the Edinburgh court recognized in the question of law “a genuine and live issue, of considerable practical importance,”⁴⁵ linked to the exercise of MPs’ rights in Parliament—it is not for the CJEU to question such assessments of the national court. Nor is the CJEU to consider the legitimacy of national legal proceedings. As long as the national court identified the relevance of the issue for the exercise of parliamentary sovereignty, the CJEU deemed the action admissible.⁴⁶

D. The Substance of the Decision of the CJEU

Wightman was the first occasion for the Luxembourg Court to offer an interpretation of Article 50 TEU and, as in other “historical” decisions, it tried to give continuity to its case law by reading Article 50 TEU in combination with other provisions. There, the Luxembourg Court openly used the constitutional jargon inaugurated with the Les Verts case.⁴⁷ Not by chance, Les Verts, together with the Kadi judgment⁴⁸ and Opinion 2/13,⁴⁹ is mentioned in paragraph 44 in order to recall the constitutional and sui generis nature of the European Treaties.⁵⁰

The Court offered a reading of Article 50 TEU in context, by attempting, in other words, to interpret this provision in light of the values that characterize the integration process. For example, the reference to the impact of the possible exit of the United Kingdom on the EU citizenship⁵¹ is confirmation of the systematic reading given by the Court.⁵²

It is interesting to note that, in Wightman, on the specific issue of the revocability of the notification, the CJEU opted for a construction other than that endorsed by the High Court of England and Wales⁵³ and by the Supreme Court of the United Kingdom in the Miller I saga.

On those occasions, both British courts denied the Government’s ability to exercise the power of notification of the intention to leave the Union ex Article 50 TEU to the European Council without parliamentary involvement. The non-revocability of the notification had been one of the fundamental pillars of the judges’ reasoning and served as an external—referring to a non-national norm—premise of that judgment. Indeed, Article 50 TEU offered margins for either options as scholars had already pointed out:

The point is arguable either way. It could be argued that since a notification to withdraw is subject to a Member State’s constitutional requirements, the Treaty therefore leaves to each Member State the possibility of rescinding that notification in accordance with those requirements. On the other hand, it could also be argued that Article 50 only provides for two possibilities to delay the withdrawal of a Member State from the EU once notification has been given (an extension of the time limit, or a different date in the withdrawal agreement). There’s no suggestion that this is a non-exhaustive list. Therefore, the notification of withdrawal can’t be rescinded.⁵⁴

We think that, on those occasions, the reference to Article 50 TEU was appropriate. Over the years, the UK membership in the Union has created complex rules in the etymological sense of the term:—a set of norms that are the result of a tangled knot between domestic law and European law.⁵⁵ This made the reference to the law of the EU Treaties inevitable for the correct reading of the question posed to the British courts in the case of the Miller I saga. The wording itself of Article 50 TEU, moreover, refers to domestic law when it reads that “[a]ny Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.” This confirms the legal interlacement created by the European integration process. A similar choice can be found in other provisions of the European Treaties. For example, it is possible to recall the reference included in Article 48(4) TEU and in many of the provisions of the EU Charter of Fundamental Rights. However, precisely because of the impact that the question of revocability of the notification had on the margin of maneuver of the UK Parliament, commentators stressed the excessive brusqueness of some of the relevant passages in the Miller I saga. They also criticized the choice not to raise a preliminary question to the CJEU.⁵⁶ Returning to Wightman of the CJEU, we can see how the Luxembourg Court started from a premise similar to that employed by the British courts—the inextricability of national and supranational legal systems. This was then reaffirmed by the CJEU itself in paragraph 45 of the Wightman judgment, which also included a reference to the Achmea ⁵⁷ case.⁵⁸

Indeed, Wightman is interesting as it offers a confirmation of the difficult distinguishability between national and supranational legal systems and the complex nature of European law. Another example of the complex context triggered by the Europeanization of the national system is the Miller I decisions of the High Court and the UK Supreme Court, especially in the passages devoted to the importance of the EU integration for the genesis of a new category of rights.⁵⁹

Will Brexit imply a perfect return to the starting condition of the British legal system? We think not. The simple fact that the European Union (Withdrawal) Act 2018 also aims to save EU-derived domestic legislation, while incorporating EU legislation,⁶⁰ shows that the future of the UK will build upon the legacy of its European membership—instead of radically denying the added value produced by these years of integration. In other words, “Brexit does not mean that all these rights will be lost.”⁶¹

Nevertheless, if in the Miller I saga the British courts opted for the non-revocability of the notification to justify the involvement of the Parliament,⁶² the CJEU instead came to a different conclusion in Wightman. Indeed, in Miller I litigation, the non-revocability of the Article 50 notification was the agreed assumption on which the courts had proceeded. The CJEU followed a different path and did so by relying on the systematic reading of Article 50 TEU,⁶³ and recalling the aims of the integration process.

In so doing, it also paid particular attention to the values of liberty and democracy, especially in the part of the judgment recalling the Kadi case.⁶⁴ Moreover, after discarding the parallel with the decision to request an extension of the negotiations pursuant to Article 50(3) TEU paragraph 60, the CJEU focused on Article 49 TEU in order to reiterate that the decision to join the EU is based on a free and voluntary commitment.⁶⁵ Finally, by illustrating the parallels between Article 49 TEU and Article 50 TEU, the CJEU concluded, regarding the revocability of the notification:

In those circumstances, given that a State cannot be forced to accede to the EU against its will, neither can it be forced to withdraw from the EU against its will. However, if the notification of the intention to withdraw were to lead inevitably to the withdrawal of the Member State concerned from the EU at the end of the period laid down in Article 50(3) TEU, that Member State could be forced to leave the EU despite its wish—as expressed through its democratic process in accordance with its constitutional requirements—to reverse its decision to withdraw and, accordingly, to remain a Member of the EU. Such a result would be inconsistent with the aims and values referred to in paragraphs 61 and 62 of the present judgment. In particular, it would be inconsistent with the Treaties’ purpose of creating an ever closer union among the peoples of Europe to force the withdrawal of a Member State which, having notified its intention to withdraw from the EU in accordance with its constitutional requirements and following a democratic process, decides to revoke the notification of that intention through a democratic process.⁶⁶

Another interesting passage in the judgment is that—and this is a rare thing in the case law of the Court—the CJEU looked at the preparatory works of the second Convention (concerning the Treaty establishing a Constitution for Europe)⁶⁷ in order to support its decision. In particular, it recalled that all amendments proposed about the possibility of expelling a Member State in order to avoid the risk of abuse during the withdrawal procedure had been rejected.⁶⁸ This is a curious “exhumation” of the preparatory works of a text that never came into force, which is the Treaty establishing a Constitution for Europe. At the same time, the CJEU, aware of the risk of an excessively unilateral reading of this provision, tried to limit the time frame in which such a decision could be taken, when it says (repeatedly, as if to reassure itself) that:

As the Advocate General stated in points 94 and 95 of his Opinion, the sovereign nature of the right of withdrawal enshrined in Article 50(1) TEU supports the conclusion that the Member State concerned has a right to revoke the notification of its intention to withdraw from the EU, for as long as a withdrawal agreement concluded between the EU and that Member State has not entered into force or, if no such agreement has been concluded, for as long as the two-year period laid down in Article 50(3) TEU, possibly extended in accordance with that provision, has not expired.⁶⁹

In so doing, the Court seemed to limit the risks associated with the unilateral nature of the revocability of the intention to leave in the two-year period—except for what is established by Article 50(3) TEU. However, even during this period, a situation of evident uncertainty could be generated, increasing the confusion surrounding the activity of the EU institutions and the instability of the markets. These considerations of the Court were followed by a procedural clarification when it added that “in the absence of an express provision governing revocation of the notification of the intention to withdraw, that revocation is subject to the rules laid down in Art[icle] 50(1) TEU for the withdrawal itself, with the result that it may be decided upon unilaterally, in accordance with the constitutional requirements of the Member State concerned.”⁷⁰

The lack of reference to the concept of national identity has been one of the fundamental differences between the Opinion of the Advocate General and the ruling of the CJEU. There is another difference, however, that should be emphasized. While the Advocate General tried to find, in the principle of good faith and sincere cooperation—Article 4 TEU—a limit to this sovereign choice of the Member State concerned, this reference disappeared in the judgment of the CJEU. These are probably the most critical issues of the Wightman case that we will analyze in the following sections.

E. The EU Legal System Under the Light of the Right Under Article 50 TEU: The Role of Autonomy?

The Wightman case of the CJEU contributed not only to clarifying the viable options concerning Brexit, but also offering several insights on the nature of the EU legal order itself. Both Advocate General Campos Sánchez-Bordona and the Grand Chamber of the Court recognized that Article 50 TEU does not expressly prohibit, nor authorize, any form of revocation of the withdrawal notification.⁷¹ Despite this, they approached the legal minimalism of the provision on this very issue by relying on different visions of the relationship between the departing Member State and the EU legal order. Their interpretative approach to filling the lacuna in the black letter of the law differed sensibly with regard to the conceptualization of the sovereign right to reverse the withdrawal process and the conditions for its exercise. Advocate General Campos Sánchez-Bordona filled the lacuna through international law, looking at Article 50 TEU as “lex specialis, in respect of the general rules of international law on withdrawal from treaties, but not a self-contained provision which exhaustively governs each and every detail of the withdrawal process.”⁷² Conversely, the CJEU relied on the autonomy of the EU legal order and justified its interpretation on the grounds of the “structured network of principles, rules and mutually interdependent legal relations binding the EU and its Member States reciprocally as well as binding its Member States to each other.”⁷³

Although both Advocate General Campos Sánchez-Bordona and the Court reached the same positive conclusion about the unilateral revocability of the notification under Article 50 TEU, they reached it from different routes. The Advocate General analyzed withdrawal as “a typical international law issue.”⁷⁴ Thus the Advocate General used international law and the Vienna Convention on the Law of the Treaties “to provide interpretative guidelines to assist dispelling doubts about issues that are not expressly dealt with in Article 50 TEU.”⁷⁵

Predictably, the CJEU refused to treat withdrawal and its reversal as an international law issue related to the participation of States in a Treaty. It also reaffirmed its willingness to note the difference between the EU legal order and “ordinary international treaties.”⁷⁶ The Court built its reasoning off the idea that the EU is a new legal order, autonomous from the Member States and international law, with its own institutions and independent sources of law—which have primacy over the laws of the Member States and may confer rights with direct effects.⁷⁷ By emphasizing the autonomy of EU law, in Wightman the CJEU reiterated the consistency of its own case law since the Van Gend and Loos case aiming to “expressly cut the umbilical cord with classic international law.”⁷⁸ This interpretation underscores the distinctiveness of EU law and dualism in the application of international law. As the Court did in both Kadi, when assessing the validity of restrictions on suspected terrorists,⁷⁹ and Opinion 2/13, when evaluating the EU accession to the ECHR,⁸⁰ the autonomy of EU law brings about the assessment of the relevant question in light of the Treaties.⁸¹ In Wightman, the CJEU analyzed the question of revocation in light of EU law, holding that its conclusion “is only corroborated by the provisions of the Vienna Convention on the Law of the Treaties.”⁸² The same approach of autonomy has also been upheld in the recent Opinion 1/17 on the Comprehensive Economic and Trade Agreement between Canada and the European Union and its Member States (CETA).⁸³ This characterization of the autonomy of EU law creates a self-protecting barrier against external interference from international law that may affect the functioning of the EU law’s fortress. This approach extends to the foundational moment of the establishment of that fortress, including the withdrawal stage.

The contrasting approaches of the Advocate General and the Court represent different backdrops against which the unilateral revocation of the notification is contextualized and evaluated. According to the Advocate General, the right of withdrawal under Article 50 TEU is:

Based on the rules of international law … because withdrawal from an international treaty is by definition a unilateral act of a State party. Like a treaty-making power, the right no longer to be bound (withdrawal or denunciation) by a treaty to which a State is a party is a manifestation of that State’s sovereignty. If a State’s decision to conclude a treaty is unilateral, so is its decision to withdraw from it.⁸⁴

As a consequence of the unilateral nature of the withdrawal, “unilateral revocation would also be a manifestation of the sovereignty of the departing Member State, which chooses to reverse its initial decision.”⁸⁵ The CJEU shared the idea that the right of withdrawal expresses a sovereign decision of the State, but it reconnected this to the right of the State “to retain its status as a Member State of the European Union, a status which is not suspended or altered by that notification”.⁸⁶

The teleological interpretation of Article 50 TEU is, therefore, based on different premises. The Advocate General identified in the unilateral revocability of the notification of the withdrawal a means to reconcile the respect for the national identities of the Member States under Article 4(2) TEU, and the integration objective under Article 1 TEU, including the protection of citizens’ rights.⁸⁷ Conversely, the CJEU did not refer to the respect for the constitutional identities of the Member States, but instead emphasized the Member States’ commitment to the common values of integration, specifically excluding that a Member State can be forced to leave against its own will.⁸⁸ This implicitly highlights the pragmatic challenges that withdrawal entails. Insofar as EU and domestic legally binding relations are deeply interrelated, it is pragmatically difficult to disentangle such constitutional interdependence and separate a Member State’s legal order from the EU. As Eeckout and Frantziou underlined, such interdependence means that a complex system of rights is at stake in the withdrawal process, and this requires a specific constitutional reading of Article 50 TEU.⁸⁹ In short, while the Advocate General aimed to strike a balance between national sovereignty and the European project, the CJEU focused on the goals of the EU legal order and the persistent willingness of the State to be part of that project.

The different characterization of the sovereign right of the Member State to commit to and withdraw from the EU legal system is accompanied by different conditions for its exercise. Both the Advocate General and the CJEU considered the revocability of the withdrawal to be the expression of a sovereign right. Yet, Advocate General Campos Sánchez-Bordona attached a number of reasonable conditions, while the CJEU rejected the imposition of constraints on the exercise of the sovereign right.

Insofar as revocation is unilateral, in principle, the constitutional requirements of the States are the only legal conditions attached to the exercise of such a right. The Advocate General argued that the unilateral exercise of the right should be reasonably tempered to protect from procedural abuse. Thus, the Advocate General emphasized that the duty to give reasons for the revocation would be reasonable, as the State in question “runs counter to its previous actions;”⁹⁰ the temporal limit covering the period before the subscription of the agreement of the parties “logically” applies to the revocation;⁹¹ and the exercise of the right shall be subject to the principles of good faith and sincere cooperation.⁹² All these common sense requirements should protect against “tactical revocations,”⁹³ while ensuring the legitimate right of the State to change its position. According to the Advocate General, making the revocation unanimous “would increase the risk of the Member State leaving the European Union against its will”—changing a sovereign choice into potential expulsion.⁹⁴ In the interpretation of the Advocate General, revocation should be exercised by ensuring mutual trust between the departing State and the EU, and not mutual consent as required by the Council and the Commission. Mutual trust requires reasonable justification to ensure that collateral legal tactics do not drain negotiations and do not turn a right into a privilege.

As previously mentioned, the Court did not impose any condition on the exercise of this right. The Court requires only that the revocation be “unequivocal and unconditional, that is to say that purpose of that revocation is to confirm the EU membership of the Member State concerned under the terms that are unchanged as regards its status as a Member State, and that revocation brings the withdrawal procedure to an end.”⁹⁵

To avoid the risk of undue interference and preserve the unconditional characteristics of the right, the CJEU rejected the condition of the unanimity of the European Council required by the Council and the Commission, and did not request any additional evidence in the exercise of the revocation to demonstrate that there is no tactical use of the right. Accordingly, the burden is upon the other Member States to trust the declaration of the State to reverse its withdrawal intentions unequivocally and unconditionally. The conclusion of the CJEU focused on the good faith of the decision of the Member State to genuinely reverse its previous withdrawal intention, while binding the other Member States to that decision. This means that the principle of mutual trust still applies in the relations between the un-departing Member State and the other Member States, because they all commit to the same values.

F. Sovereign Perils. What Might this Decision Represent for the Rest of the EU?

The wording of Article 50 TEU does not say much about the revocability of the notification and scholars⁹⁶ argue that a possible bona fide change of mind should be admissible under EU law. However, we are of the opinion that the content of Article 50 TEU must be read narrowly in order to avoid the risk of abuse. The abuse in question could result in a strategic use of the threat of exit in order to renegotiate more favorable conditions. This almost became a reality—and if not for the vote of June 23, 2016—“A New Settlement for the United Kingdom within the European Union” would have entered into force and this would have created several issues, as scholars pointed out when commenting upon the norms of that problematic deal.⁹⁷ In this sense it can be argued that the EU is experiencing a sort of inevitable crisis, because not even the victory of Bremain would have avoided a necessary rethinking of the integration process and a season of conflicts. This leads to the second point of our argument: In our opinion, we should favor a “sustainable” reading of Article 50 TEU, precisely in order to avoid the integration process being held hostage to possible instrumental uses of this provision.

This debate reminds us of the discussion about the codification of some secession clause in federal and regional systems. Building upon this similarity, one could argue that risks like these exist even in the absence of a secession clause as the Spanish case shows. In that context, the lack of an explicit provision on secession has led to the explosion of a political crisis in the absence of constitutional paths other than the constitutional amendment.⁹⁸ More in general, silence does not seem to work when dealing with secession because it creates ambiguity, as Ginsburg and Versteeg have recently pointed out.⁹⁹

To remedy this ambiguity, in 1998 the Canadian Supreme Court devised a normative path despite the silence of the relevant constitutional sources and in light of some unwritten “underlying principles.”¹⁰⁰ Stepping back to the EU, the conclusion reached in Wightman could be seen as a reasonable choice, but it risks exposing Article 50 TEU to dangerous unilateral readings. Indeed, there is something that we do not find convincing in this judgment. First of all, the frequent mention made by the CJEU of the concept of sovereignty represents a linguistic choice at odds with that of national identity used by the Advocate General in his Opinion.¹⁰¹ The adjective “sovereign” was repeated six times in the English version of the judgment and the concept of sovereignty was undoubtedly one of the keywords of the decision. Echoing the concerns of the Scottish reference on the protection of national sovereign rights in the Brexit’s vote, the CJEU’s ruling focused on enabling the State to make a fully informed choice.

Among the flaws of the judgment, there is certainly the lack of reference to the principle of sincere cooperation, which had instead been mentioned by the Advocate General as a “further limit on the exercise of the right of unilateral revocation” in paragraph 148 of his Opinion. Along with the principle of sincere cooperation, the reference to national identity also vanished in the Wightman ruling. In this way, another piece of the interpretative mosaic proposed by the Advocate General disappeared.¹⁰² In other words, it seems to us that the CJEU has excessively focused on an approach that looks at national sovereignty, while the Advocate General was more careful in coming up with possible remedies to prevent the Member States from exercising their sovereignty in an abusive manner.

In our view, the CJEU tried to guarantee the sovereign choice at the costs of the multilateral rationale of Article 50 TEU. While the Advocate General tried to mitigate such a choice with the recognition of loyal cooperation, this kind of balance does not seem to belong to the reconstruction advanced by the CJEU. The CJEU instead kicked Article 4 TEU out of the picture, by framing the issue in light of national sovereignty. The reference to identity in this sense was much more promising. It is no coincidence that both national identity and loyal cooperation are recalled in this provision. Recently, national identity has been under siege as if it were per se a dirty word¹⁰³—a bad concept that would inevitably lead to the disintegration of the EU. There is no doubt that Article 4(2) TEU has been abused, and scholars have already identified several examples of this.¹⁰⁴ At the supranational level, Advocates General have sometimes opposed national identity to fundamental rights,¹⁰⁵ subsequently offering a misleading dichotomy. In our view, however, these abuses or misunderstandings do not exclude the possibility of a proper use of this clause. Our first recommendation regards the need for a systematic interpretation of the clause. In other words, this clause should not be read in splendid isolation, but in context, because it is merely a star in a broader constellation of values guaranteed by the EU Treaties.

Our second point directly stems from the first—Article 4(2) should not be detached from the reference to the moment of commonality recalled in Articles 2 and 6 TEU. This was clear in the wording of Article 6 pre-Lisbon formula, where both national identity and common constitutional traditions were disciplined in the same provision. It is likely that the decoupling that occurred in Lisbon was a mistake, galvanizing the temptations of a unilateral and decontextualized reading of the identity clause.

Finally, as suggested earlier, Article 50 TEU can be compared to secession clauses, although scholars have distinguished between withdrawal and secession.¹⁰⁶ Secession clauses are traditionally ambiguous and rarely aim at facilitating the exit.¹⁰⁷ On the contrary, they are frequently drafted to make secession difficult and conditional upon the respect of some procedures and values—thus preserving axiological continuity and respect of the rule of law.¹⁰⁸ Article 50 TEU represents such ambiguity in both the lack of an explicit mention of the possibility of revoking the notification and in clarifying how many times the extension can be requested. In giving such interpretation in Wightman the CJEU exposed the rest of the EU to uncertainty and unilateral choice that have little to do with the reference to “its own constitutional requirements” included in Article 50(2) TEU.

Clearly, the CJEU knew that it had to rule on a very sensitive question and knew sovereigntists were waiting for its final decision. In this sense, the insistence on the concept of sovereignty could be seen as a strategic, rather than an obliged, choice. The wording itself of Article 50 TEU, after all, refers to the national legal systems in its first paragraph, and this could justify the deference of the Court. Will this interpretation be enough to avoid the risk of instrumental abuse? That is hard to say. For the time being, Brexit has been a fundamental test bench mainly for the constitutional system of the United Kingdom. However, the risk of a domino effect, according to which other EU countries might be tempted to follow a similar path, cannot be excluded. Wightman is going to be a reference point for future exit related cases, but its ambiguities might still trigger unexpected developments.