Article, Case note

The Withdrawal of the Notification under Article 50, or How to Stop Brexit

The Editors

The United Kingdom is due to leave the European Union on 29 of March 2019. On what terms it will do so, it is yet uncertain when less than three weeks are left before Brexit day.

On Tuesday next week, 12 March 2019, the UK Parliament will cast a vote on the new deal reached by Theresa May and the EU – ‘new’ because the parties re-open the negotiation talks after the UK Parliament had rejected the previous deal on 15 January 2019, inflicting a historic defeat to Theresa May.

On 12 March, the deal with the EU on the conditions for leaving the European Union may either be approved or rejected. If it is rejected, the conservative party has promised to the Parliament the opportunity to vote on whether to go ahead in just over three weeks’ time without any kind of negotiated deal; or whether to ask the EU to push back Brexit day in order to extend negotiations. In the latter case, it is not excluded that the UK will hold new elections, a new referendum, or even eventually halt the Brexit process. We dedicate this blog post to the European Court of Justice’s decision on whether the UK can withdraw the notification of the intention to leave given in March 2017.

The Wightman case originates from a request for a preliminary ruling made by the Court of Session –  the Scottish Supreme Civil Court – to the European Court of Justice in October 2018. The case in front of the referring court concerned a petition for judicial review brought by some Members of the Scottish and the English Parliaments before the Scottish Court on the 19th December 2017. The referring Court asked the CJEU whether the notification to withdraw from the EU sent to the European Council by the UK government on the 29th March 2017 could be unilaterally revoked by the government itself before the expire of the 2 years period envisaged in article 50(3) TEU.

The Opinion of the Advocate General

Advocate General Campos Sánchez-Bordona delivered his opinion on 4 December 2018, only one week after the hearing of the case. Such extremely short time, unprecedented in the history of the Court, bears witness to the extent of the focus of the Court on this case – especially since, in that week, the opinion was not only written in Spanish, but also translated into other languages.

Contrary to the UK’s submission, the AG took the view that the question was admissible because, under EU law, the question referred for preliminary ruling enjoys a presumption of relevance (para 34). The AG was also satisfied that none of the conditions for the ECJ to refuse to answer where fulfilled. First, the question was referred by a court that complied with national law. Second, the dispute in the main proceedings was genuine. Third, the answer of the ECJ was necessary to enable the referring court to give a judgment. Fourth, the question was not merely academic. Fifthly, that the question was not premature. Sixthly, the question was not superfluous. Seventhly, the answer of the Court would not be merely an advisory opinion. These seven reasons are small variations over the conditions first established in Foglia Novello – sometimes with a remarkably original touch, as in the condition of the ‘ripeness’ of the question, phrased by the AG as ‘not premature’ (para 42). ‘Ripeness’ is a legal doctrine developed in the United States, which warrants that only cases that have matured into an existing judicial controversy, and not those that vert on future potential developments, may be adjudicated upon. Overall, the assessment of the AG followed the classic case law of the Court on admissibility, but one might still wonder whether it is convincing the reasoning that requesting clarifications over the revocability of Article 50 TEU does not amount to seeking an advisory opinion because it enables Member of the Parliament to cast their votes e.g. on draft withdrawal agreements. Many questions that may object of requests for advisory opinions pursuant to Article 218(11) TFEU can still ‘enable’ parliamentarians, either national or European, to cast their votes one way or the other. An example could be the calculations of a minority party in a domestic parliament, threatening to vote against the ratification of a mixed agreement. Clearly, this party’s strategy would only make sense if the Court pronounced that the agreement is indeed compatible with EU law.

On the substance, the AG considered that the rule contained in Article 68 of the Vienna Convention on the Law of Treaties, that a state may unilaterally withdraw from a Treaty, was not a rule of customary international law. It is not clear, however, why the AG would start its reasoning from the Vienna convention, especially since the conclusion was that Article 68 of the Vienna Convention was not binding on the EU but might only be used as an interpretative criterion. Particularly unconvincing is the statement (para 85) that ‘there is nothing to preclude recourse being had to Article 68 of the Vienna Convention even though it does not reflect, stricto sensu, a rule of customary international law’, since the AG has not explained on what grounds that Article should be taken into account in the first place. Was the AG suggesting that EU law should always be interpreted in the light of international law, even of those provisions not binding on the EU?

The AG proceeded to a literal, contextual, teleological and historical interpretation of Article 50 TEU, to argue that unilateral revocability was the correct interpretation. This was so because the ‘reasons which I set out below, in favour of withdrawal notifications being revocable, can, I believe, be put forward with greater cogency than those against such revocation’ (para 96). Those reasons are, in particular, that there are few procedural obligations on the exiting Member State; that an intention is ‘not definitive and subject to change’ (para 99), and not allowing the opportunity to unilaterally withdraw the notification means compelling to exit; and that the symmetry between before and after notification points to the direction that the Member State can act unilaterally.

The Judgment of the Court


The UK Government’s submitted that the case was merely hypothetical or of advisory nature since no draft act of revocation had been put in place or even conceived. It was the UK Government’s view that advisory opinions of the Court may be obtained only under Article 218(11) TFEU, concerning the compatibility of a proposed international agreement with the Treaties. Therefore, were the UK to revoke Article 50 notification, the only possible remedy would be a direct action against the Member States. Interestingly, the Commission joined the UK Government in suggesting that the question was hypothetical, but it did so on different grounds. Notably, the Commission’s argument was that the judgment issued by the Court would have not been binding upon the UK Government and was thus hypothetical.

The Court did not uphold the inadmissibility plea. Instead, it first recalled its settled case law, whereby the preliminary ruling questions are presumed to be relevant for the resolution of disputes pending before national courts (para 27 and 28). The Court noted that, as a matter of fact, a dispute was pending and involved MPs which had to vote for the withdrawal of the UK from the EU and on the ratification of the agreement negotiated between the UK and the EU under Article 50 TEU. It followed that the question was admissible and for the Court to determine, since the Members of the Scottish and English Parliaments have an interest in the answer to that question of EU law. Indeed, the response would clarify which options are open to them in exercising their parliamentary mandates. The Court was responsive to the pressing need to interpret the effects of Article 50 TEU and provide clarity in the chaos brought by the Brexit shadow.


On the substance, the ECJ addressed two issues: first, are Member States entitled to revoke the notification under Article 50 TEU and, second, if so, under what conditions?

In their submissions, the Council and the Commission were mainly concerned to ensure that, if a right to revoke Article 50 notification existed, it could not lead to abuse of rights by the Member States. Both institutions argued that such right to revoke had to be subject to a unanimous vote in the European Council. The views from the institutions are understandable: years of negotiations should not be forgotten to follow the whims of politics in a Member State. Also, in case of unilateral right to revoke the Article 50 notification, the exiting Member State might start exerting their political pressure by triggering and subsequently revoking their decision to leave the EU: the remaining Member States would become hostage of the ‘indecisive’ Member State.

In this respect, the Court delivered a powerful judgment offering a reconstruction of the nature of the EU and its legal system. In the legislative silence on the right to revoke Article 50 TEU, the main EU features in the light of which a such right could be conceptualised are the autonomy of EU law, its primacy over the laws of the Member States and the direct effect of a whole series of provisions which are applicable to their nationals and to the Member States themselves. The uniqueness of the EU lies in the rights and obligations of the Member States and their citizens, as result of the Treaties. In the light of the above, Article 50 is to be interpreted as protecting the sovereignty of the Member States to leave the EU as well as the objective of ensuring orderly exit from the EU. In agreement with AG Campos Sánchez-Bordona, the Full Court ruled that a right to revoke Article 50 TEU exists and may be exerted before the withdrawal agreement enters into force and, if no such agreement has been concluded, for as long as the two-year period laid down in Article 50(3) TEU has not expired. The Court went on by holding that the revocation could be made unilaterally by the Member States as an expression of their sovereign powers to remain in the EU (para 59). If Member States were to be forced to leave the EU, the democratic processes according to which they determined also their willingness to remain within the EU would be made void. As for its consequences, the revocation of the notification would bring the withdrawal procedure to an end, thus leaving the exiting state’s status within the European Union unchanged. Hence, the Court also identified conditions for the revocation to be valid, i.e. it must, firstly, be submitted in writing to the European Council and, secondly, be unequivocal and unconditional, that is to say that the purpose of that revocation 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.