UK Constitutional Law Association

Alison Young: Brexit, Article 50 and the ‘Joys’ of a Flexible, Evolving, Un-codified Constitution

There has been a lot of speculation as to the meaning of the ‘constitutional requirements’ to trigger Article 50. Nature abhors a vacuum and sometimes, it seems, so do constitutions. This vacuum arises because, unlike other legislation providing for a referendum, the European Union Referendum Act 2015 makes no mention of any legal obligation triggered by the referendum result. Most, if not all, British constitutional lawyers would accept the proposition that referenda do not generate legally binding obligations upon government to implement their results, and the lack of provisions in the 2015 Act contrasts starkly with the Northern Ireland Act 1998 and the Parliamentary Voting and Constituencies Act 2011. Given the relative silence of the European Union Referendum Act 2015, one can only infer that the result of the referendum is advisory only, and does not trigger Article 50 in and of itself. If any legal obligations follow from the Referendum they are to be found elsewhere, or are to be considered to be purely political.

Given this, what are the ‘constitutional requirements’ governing a decision of the UK to withdraw from the European Union? There would appear to be a number of possibilities:

The argument in favour of a legal requirement for legislation before an exercise of prerogative powers has been made by Nick Barber, Tom Hickman and Jeff King. Their argument turns on an interpretation of legal requirements governing the existence and the exercise of prerogative powers. It is clear that, to the extent to which the two directly conflict, statutory provisions override prerogative powers. It is also clear that, where there are clear statutory provisions governing the way in which prerogative powers is exercised, it is unlawful to exercise these powers in a manner which contradicts these statutory provisions. Barber, Hickman and King argue for a different interpretation of the nature of the relationship between statutory provisions and prerogative powers, essentially arguing that as statute always overrides the prerogative, prerogative powers cannot be exercised in a manner which would ‘turn a statute into what is in substance a dead letter’ or to ‘cut across the object and purpose of an existing statute’. They argue in addition that the European Communities Act 1972 is the means by which all EU rights and obligations are incorporated into European Union law. As such, to use the prerogative power to initiate Article 50, which will then inevitably remove these rights, would be to cut across the purpose of the European Communities Act 1972.

This is a difficult argument to accept. First, it is clear that their interpretation, though clever, is not widely accepted. Second, it is by no means clear that the triggering of Article 50 automatically means that all of the EU rights currently enjoyed by EU citizens will be removed. Unless and until a withdrawal agreement is reached between the UK and the European Union, it is impossible to predict which EU rights will and will not continue to be enjoyed by UK citizens. Even though we can point to some that definitely will be removed – the right of UK citizens to vote in EU elections, for example – it is hard to argue that the removal of one right found in existing UK legislation means that prerogative powers cannot be exercised so as to restrict that right without legislation. Third, it is hard to read Article 50 as requiring that, once commenced, the only possible outcome is a withdrawal agreement or the removal of the United Kingdom from the EU. Article 50 makes it clear that ‘the Treaties cease to apply’ to the UK ‘from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification’ unless there is an extension of the negotiation period. It is silent on whether a Member State may withdraw its decision to leave the EU before the end of that period. Admittedly, to read a power to withdraw the notification may be abused – a Member State being held in a constant state of limbo by notifying the EU of its intention to withdraw, failing to reach an agreement within 2 years and failing to obtain permission to extend the negotiation period, then withdrawing its notification only to trigger this again to obtain at least a further two years for negotiations. However, this potential for abuse can be countered by potential legal action before the European Court of Justice, not to mention the political difficulties of adopting this course of action. Fourth, a prerogative power to enter into Treaties does not in and of itself create rights and obligations in UK law, these rights and obligations need to be ratified – normally through legislation – into UK law.

It is easier to argue for a requirement of parliamentary deliberation prior to the exercise of a prerogative power. The easier argument would be to argue in favour of a convention requiring parliamentary deliberation prior to the exercise of Article 50. It is difficult to point to a specific convention to this effect. However, that is not in and of itself conclusive – new conventions have to start somewhere. If we adopt Jennings’s definition of a constitutional convention, they arise from a series of past actions, coupled with the belief of the actors that they were bound to act in this manner, and a reason for acting in this way. There is no past practice of the use of the prerogative power to initiate Article 50. Nor is there any past practice of a requirement of parliamentary deliberation prior to the use of the prerogative to leave an international Treaty. However, it may be possible to argue that a new convention has arisen by analogy with the convention, recognised in the Cabinet Manual, although subject to different interpretations, of providing the House of Commons with the ‘opportunity to debate’ before the commitment of troops to military action. Arguably, this convention has since evolved into a requirement for a parliamentary vote in favour of military action, following the failure to obtain a vote in favour of military action in Syria in 2013, the prerogative power then being exercised following a later vote in favour of military action in 2015.

The analogy is appropriate for a number of reasons. Both would provide a democratic mandate before a prerogative power is exercised which may have large constitutional repercussions; be they the deployment of troops or a potential fundamental change in the rights and freedoms of UK nations and the potential threat to the unity of the United Kingdom. Both concern the external sovereignty of the UK. Both concern issues in which the emotions of the public run high, demonstrated through public protests against military action and the signing of the petition questioning the result of the referendum and public protests. Both concern issues on which emotions run high and there may be the risk of a snap judgment that may, or may not, be regretted upon later reflection. Both may also have originated from a perceived divergence between popular and parliamentary sovereignty, with Parliament having earlier appeared to have engaged in military action when this was not approved by the public, or where the public has voted to leave the EU when major political parties were campaigning for the UK to remain. It is hard not to see the justification of a convention that provides greater democratic scrutiny over the exercise of prerogative powers. Moreover, even if the case for a convention of parliamentary deliberation cannot be sustained, there may nevertheless be ample support for the Government to initiate this convention, a requirement of deliberation as opposed to parliamentary approval being easier to justify in the face of the referendum decision.

If we can make a case for a constitutional convention for parliamentary approval, can we argue that this convention could be legally enforced? Whether we would regard it is the crystalisation of convention into law, or the adoption of a new principle of the common law mirroring the convention, it is hard to argue that a convention requiring parliamentary consultation before the exercise of a prerogative power should be enforced by the courts. First, this may cause issues for parliamentary privilege, found in Article 9 of the Bill of Rights 1689, particularly if this required the court to evaluate the sufficiency of parliamentary deliberation, although admittedly there is some suggestion by Lord Reed in HS2 that this may be possible. It may also be non-justiciable as concerning a matter of high foreign policy Second, it is no exaggeration to classify the result of the referendum as creating a constitutional crisis. In these circumstances, the UK may be wise to learn from the experience of Canada, where the Canadian Supreme Court concluded that, whilst not illegal, it was unconstitutional to seek a constitutional amendment without consent from the Provinces of Canada. This position allowed courts to provide an informed view as to the principled reasons for the existence of the convention, whilst stopping short of requiring the Government to abide by a legal obligation. This compromise has at least the advantage of aiming to provide for further negotiation and deliberation, which may be more needed in such situations. The same may be achieved by the court recognising, though not legally enforcing, a constitutional convention. The stronger argument is that, if a convention does exist, it should not be enforced by the courts.

If there is no ability to find a legal or a conventional constitutional obligation for legislation or parliamentary consultation, there is at least a good argument, in terms of political expediency, for wider consultation by the Prime Minister before an exercise of the prerogative power to trigger Article 50. The people may have voted to leave the European Union, but there is no popular consensus as to what this entails, or the relative desirability of any particular withdrawal agreement. Parliamentary debate may help to further inform us about the benefits and burdens of EU membership – something which many commentators feel the referendum debate lacked. This may lead to further reflection on the possible aims of a withdrawal agreement, providing more democratic legitimacy for the Government charged with taking the negotiations forward. Parliamentary discussion may be particularly important given the lack of an electoral mandate to withdraw from the EU, the manifesto promise being only to hold a referendum. Moreover, the political expediency argument should extend to the need for parliamentary debate in the devolved legislatures, particularly given the even split with two of the component nations of the United Kingdom voting to leave and two voting to remain in the European Union.

Whilst the political expediency argument is the easiest to satisfy, it poses a potential danger to the constitutional stability of the United Kingdom and the European Union. ‘Political expediency’ is not a constitutional requirement. As such, if there is a long delay between the outcome of the referendum and the triggering of Article 50, the delay may prompt the Commission or another Member State to initiate an enforcement action against the UK under Article 258 TFEU, or Article 259 TFEU. By failing to notify the Council of an intention to withdraw, despite fulfilling the constitutional requirements to withdraw from the EU, the UK has arguably failed to fulfil its obligation under Article 250 TFEU – especially as ‘shall’ may well be interpreted to include a requirement of ‘within a reasonable timeframe’. The constitutional implications for the EU and the UK of such actions would be gargantuan, not to mention the impact of a further erosion of goodwill between the UK and the other 27 Member States when entering into negotiations. It would be wise, if delay is sought, to find a constitutional requirement – the easiest being a constitutional convention of parliamentary deliberation.


Alison Young is an Associate Professor of Law and a Fellow of  Hertford College, University of Oxford.

(Suggested Citation: A.L. Young, ‘Brexit, Article 50 and the ‘Joys’ of a Flexible, Evolving, Un-codified Constitution’ U.K. Const. L. Blog (1 Jul 2016) (available at