At the time of writing – less than 3 weeks until the current ‘Brexit day’ of 31 October 2019 – all options relating to the UK’s departure from the European Union appear to be on the table. Leaving with a deal, ‘crashing out’, not leaving at all, or anything in between seem equally possible. Much attention has been paid to the UK’s constitutional requirements governing the executive’s actions in relation to Brexit, as well as the domestic legal consequences of flouting them. The possibility of Prime Minister Johnson going to jail for violating these requirements has even been considered. However, not much has been said about the potential international law consequences. Here I explore whether international law could prevent a No-Deal Brexit – or, more precisely, whether a failure to comply with domestic constitutional requirements may prevent the UK’s withdrawal from the EU from taking effect in international law. This discussion draws on my recent work exploring the role of domestic law in the international legal validity of treaty withdrawal more generally.
The starting point for this discussion is Article 50 of the Treaty on European Union (TEU) – by now, likely the most famous treaty exit clause in legal history. Art 50 states, in part:
Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.
A Member State which decides to withdraw shall notify the European Council of its intention. In the light of the guidelines provided by the European Council, the Union shall negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union. …
The Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period.
For our purposes, Art 50(1) is the key provision. This, unusually, makes the triggering of withdrawal from the EU explicitly contingent on compliance with the State’s domestic constitutional requirements. Thus, in principle, if there is a failure to comply with those constitutional requirements, the decision to withdraw is invalid according to the TEU. In the 2018 Wightman decision, the ECJ affirmed that “the decision to withdraw is for that Member State alone to take, in accordance with its constitutional requirements, and therefore depends solely on its sovereign choice” (at para. 50).
While a full dissection of the UK’s constitutional requirements for leaving the EU is not possible here, there are two clear domestic law limitations constraining the UK executive’s prerogative in relation to Art 50. First, per the UK Supreme Court in Miller (No. 1), triggering Art 50 required Parliamentary approval through legislation. This was provided in the European Union (Notification of Withdrawal) Act 2017. Second, subsequent legislation has required the executive to request an extension of the withdrawal period as referred to in Art 50(3). Of these, the European Union (Withdrawal) (No. 2) Act 2019 (known as the ‘Benn Act’), currently requires the Prime Minister to write a letter in prescribed form requesting a further extension if there is no deal agreed by 19 October 2019, and there is no Parliamentary approval of a no-deal Brexit.
Thus, had Parliament not approved the Art 50(2) notification sent by the executive through the 2017 Act, it arguably would have been invalid – not only under domestic law, but also international law according to Art 50(1). Similarly, had Parliament included any conditions on the triggering of Art 50 in that legislation (for instance, that it could only be triggered after a successful withdrawal deal had been reached), failure to meet those conditions may have also resulted in the invalidity of the notification of the withdrawal. However, that particular ship has sailed. The 2017 legislation ensured that the UK had decided to withdraw from the EU ‘in accordance with its own constitutional requirements’.
So, given that the Art 50(2) notification has not yet taken effect, do options remain for Parliament to render the withdrawal decision unconstitutional and thus invalid per Art 50(1)? It may be thought that the Benn Act could constitute an amendment of Parliament’s approval of withdrawal in the 2017 Act, with Parliamentary approval now conditional on obtaining a deal. Indeed, it has been described as a ‘law to stop no-deal Brexit’. However, that does not appear to be so: all that it requires is that the executive request an extension to the withdrawal date, which may or may not be agreed by the EU; there is no requirement that, in the absence of agreement by the EU, the Art 50 notification is then withdrawn. There is no change to the legal default that, if no deal nor extension are agreed, the withdrawal will take effect on 31 October 2019 by operation of Art 50(3). Thus, as it stands, the Parliamentary approval of withdrawal – deal or no deal – remains intact.
In contrast, should Parliament pass legislation requiring the Art 50 notification to be revoked, I suggest that would fundamentally undermine its international legal validity under Art 50(1). This would be so whether the revocation was unconditional, or contingent on certain specified conditions (such as a failure to achieve a deal approved by Parliament). As recognised by the ECJ, a Member State may unilaterally revoke its Art 50(2) notification at any time before it takes effect. Crucially, the ECJ held in Wightman that this revocation also needs to be done according to the Member State’s own constitutional requirements:
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 Article 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. … [I]t 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 European Union in accordance with its constitutional requirements and following a democratic process, decides to revoke the notification of that intention through a democratic process
As emphasised by the ECJ, an Art 50(2) notification is merely a declaration of intention, subject to reversal according to the State’s constitutionally-expressed will. This suggests that if the UK Parliament passed legislation annulling its approval of the Art 50(2) notification before it takes legal effect on 31 October (or subsequent Brexit day), this would render the notification invalid under Art 50(1), and thus ineffective under international law. For, as Miller (No.1) clearly decided that Parliamentary approval of EU withdrawal was required, surely the revocation of that approval would mean that the decision to withdraw was no longer ‘in accordance with its own constitutional requirements’ – and therefore would not satisfy the terms of the TEU.
Thus, I agree with Professor Mark Elliot that ‘Legislation along these lines [revoking the Article 50 notification], if enacted, would provide a cast-iron guarantee against a no-deal Brexit because it would, presumably, impose an unqualified and immediate duty upon the Prime Minister to revoke the UK’s notification under Article 50.’ However, I would go further and argue that this would be so even if the Prime Minister did not comply with such a duty. The removal of Parliamentary approval of itself is sufficient to negate the constitutionality of the Art 50(2) notification, and thus its international legal effectiveness under Art 50(1).
In fact, this might be the case even if Art 50 did not expressly refer to the need to comply with the UK’s constitutional requirements. The VCLT provides that a ‘manifest violation’ of a ‘rule of its internal law of fundamental importance’ can exceptionally vitiate a State’s consent under international law when joining a treaty (Art 46). I have argued in more detail elsewhere that, on the basis of a textual and principled interpretation, this exception to the executive’s international treaty-making authority should apply by analogy to both making and unmaking treaties. As Brexit has starkly illustrated, leaving a treaty can be equally central to the expression of national sovereignty as joining. It is inconsistent with modern expectations of constitutional government that violations of domestic law may invalidate the executive’s consent when becoming a treaty party, but are simply irrelevant in relation to withdrawal.
There are two necessary elements to satisfy the ‘manifest violation’ exception: there must be a breach of a domestic law of fundamental importance concerning treaty-making capacity; and that breach must be ‘objectively evident to any state conducting itself in the matter in accordance with normal practice and in good faith.’ The ICJ has held that there is no general duty on States to know the domestic legal requirements of their treaty partners; such requirements must therefore be ‘at least properly publicised’ in order for their violation to invalidate a state’s treaty consent. Both elements would likely be satisfied here – particularly given the unprecedented publicity surrounding the UK’s domestic law requirement of parliamentary approval of withdrawal from the TEU.
Thus, whether because of the express requirement of constitutional compliance in Art 50(1), or the proposed application of Art 46 VCLT to treaty withdrawals by analogy, the passage of legislation revoking Parliament’s consent to leaving the EU may prevent the termination of the UK’s membership of the EU under international law. Finally, though, if the UK executive did not comply with such legislation, would other EU Member States, or perhaps the ECJ, accept that the UK was still an EU Member State in international law?
That would remain to be seen. While some argue that Art 50(1) makes the ECJ the final arbiter of the question of constitutionality of EU withdrawal, Wyrozumska has argued that “the fulfilment of constitutional requirements can only be verified by the Member State itself, not by the CJEU or other Member States”. The ECJ’s 1965 decision in the Acciaierie San Michele case may support the latter view. The ECJ held that there must be ‘complete and uniform application’ of the ECSC Treaty for all Italian citizens, notwithstanding an ongoing domestic court challenge to the constitutionality of Italy’s ratification of the treaty. This was despite Art 99 ECSC Treaty which provides: “the present Treaty shall be ratified by all the member States in accordance with their with their respective constitutional rules.” This may indicate that the ECJ was (implicitly) of the view that it did not have the authority to examine the constitutionality of ratification, though it did not examine Art 99 directly.
According to this approach, it would be for the UK itself to determine whether domestic requirements have been fulfilled as prescribed by Art 50(1) – regardless of whether a violation is objectively obvious on the basis of widely-reported judicial decisions or parliamentary legislation. Since it is the executive’s role to represent the UK’s position to the EU, this would likely result in the ECJ accepting the executive’s view of the domestic constitutionality of withdrawal.
However, it is normal practice that international legal requirements for valid treaty exit – especially those expressly included in the treaty itself – are the subject of international evaluation and assessment, by other Member States and any international courts with jurisdiction. The same should apply to the requirements in Art 50(1), even though they concern matters of domestic law. It has long been accepted, including in the ICJ’s Nicaragua decision (see para. 258-259), that the inclusion of such matters in an international treaty mean they are no longer part of the State’s exclusive domestic jurisdiction. Further, given the extraordinary publicity surrounding the UK’s constitutional requirement of parliamentary approval of withdrawal from the EU, the ECJ would be justified in taking a more robust approach to that in the Acciaeirie case, decided more than 50 years ago.
Thus, it is at least arguable that international law gives a powerful tool to the UK Parliament to prevent a No-Deal Brexit. However, whether this can be used will depend on whether Parliament is willing or able to introduce legislation rescinding its approval of the UK’s departure from the EU, before the Art 50(2) notification takes effect in international law. Ultimately, then, the question is one of political, rather than legal, possibility.