Richard Lang: Consequences of the Extension Deal: A Reply to Professor Allott

I would like to thank Professor Allott for his fascinating post, ‘Unexpected Denouement’. The UK Remains in the EU by Mistake. The Brexit Saga Could Run and Run’ from Tuesday.  I just wanted to be advocatus diaboli on a couple of points, although some extremely able advocates, commenting below the original post, should also be heeded.

While I have no quibble with Professor Allott’s view of Wightman, my concern is centred on the paragraphs discussing how the UK would come to stay indefinitely within the EU.  Professor Allot states:

The agreement of 21 March between the Prime Minister and the European Council extended the period of negotiation and thereby prevented automatic withdrawal on 29 March.  As a result of the agreement of 21 March, the two-year time-limit in Article 50 measured from March 2017 ceased to have legal significance.

Thus, if I may paraphrase, Mrs May’s deal of last week nullified – as far as the UK was concerned – that part of Article 50(3) TEU dealing with the time limit by which withdrawing Member States must leave failing the adoption of a withdrawal agreement.  By way of reminder, Article 50(3) reads:

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.

Since the proviso beginning ‘unless the European Council’ has been overcome, the preceding text of Article 50(3), or at least all the text after ‘failing that’ is rendered if not otiose then certainly in casu redundant, to be replaced by whatever the new ‘unanimously decide[d]’ leaving date (or as it turned out dates) may be.

Professor Allott goes on to criticize the Council’s choice of new dates, suggesting among other things that this choice is an abuse of power, that it amounts to a ‘virtual amendment of the time-limit in Article 50,’ and that it cannot conceivably be lawful.  He goes on to imagine a future legal challenge in the national court or at the Court of Justice, or indeed a challenge which involved both fora, and the length of time which such a challenge would take.  Indeed, he supposes, the resolution of the case might actually post-date the date or dates impugned.  He concludes this section as follows:

If the time-limits in the agreement of 21 March were held to be unlawful and invalid under EU law and UK law, the UK would then continue to be, or would retroactively become again, a member state of the EU.

It is this sentence which I find – with the greatest possible respect – puzzling.  For if the agreement of last week (the resultant Council Decision is actually dated the 22nd) was operational for the purposes of erasing the ‘two years after the notification’ stipulation in Article 50(3), why would it not also be operational in all other respects?  A contrario, we could ask, if the agreement was not operational in all other respects (that is, the actual choice of the new dates), why would it be operational to bring about the erasure?  I feel that this is a sleight of hand causing the UK first to lose its old leaving date, then to lose its new ones, and finally to find itself not leaving at all.

Just to take one of Professor Allott’s arguments, a perusal of the travaux préparatoires for the Constitutional Treaty (CT) – which eventually gave way to the Lisbon Treaty, albeit with new Article 50 TEU more or less intact from its CT days – reveals little to support the proposition that the extension mechanism was intended ‘to allow [the withdrawing State] to avoid the disaster of an automatic withdrawal’ or indeed to facilitate further negotiation, genuine or otherwise.  The one fact that comes across considerably more than any other is how unpopular the two-year deadline limb of (what is now) Article 50(3) really was, with an enormous number of delegates calling for the third paragraph to end with the words ‘entry into force of the withdrawal agreement.’  Many others did not want a voluntary withdrawal provision in the Treaty at all.

Might other arguments be available?  Certainly the fact of making these alternatives dependant on a matter falling entirely within national political competence is worthy of note.  Aside from the areas of EU Enlargement and occasionally state aid, the national parliaments rarely feature in EU legal discourse.  They have never been elevated to the role of a ‘third chamber,’ and so, outside of the ratification of Treaties, their say-so is not usually required under EU law, even if their objections have taken on more significance in recent years.  It is thus hard to think of a previous EU legal act giving a Member State parliament such a central, indeed crucial, role.  The EU’s famous foray into Austrian politics during the Haider Affair comes to mind, as do EU incursions into Hungary’s internal policies and Greece’s finances, but while during those episodes the EU was directing events and the Member State following, here the roles are reversed (at least on the surface).  And although this is not a reason to impugn the Council Decision, one does worry that these two choices of deadline might sway national parliamentarians in their decision-making, undermining the EU’s own rule-of-law credentials.

Of course in carrying out EU legal responsibilities, the European Council must act in accordance with EU general principles, and perhaps a more promising line of enquiry, although not necessarily more fruitful, would be to consider whether the use of two alternative deadlines might not have strained, if not breached, the principle of legal certainty.

Let us suppose that there was an arguable case that the EU had acted beyond its powers.  Then the correct thing to do then would be to bring an action under Article 263 TFEU to challenge the Council Decision.  But were it later struck down, all that would do is return us to the position quo ante – we leave on Friday 29 March 2019.  I respectfully do not follow Professor Allott’s thesis that we stay indefinitely.  I suppose that it is possible that, if passed, the Statutory Instrument drafted to introduce the new time-limits into UK law (not technically needed thanks to direct applicability) could be judicially reviewed, but if the argument was that its EU origins were somehow unlawful, that would require a reference to the Court of Justice under Article 267 TFEU, with the result of a strike-down in ultimate effect no different to that described above in relation to Article 263.

Theresa May’s fourteen (or fifty four) day extension deal may have taken eight hours to negotiate, but will probably ultimately feature only as a small footnote in the Brexit saga, of which I suspect we now only at the end of Volume One.  The episode nonetheless illustrates once again how uncomfortably the law’s certainty sits alongside the dynamic, shifting realities of the Brexit experiment.

Richard Lang, Senior Lecturer in EU Law, University of Brighton and member of the EU Committee of the Law Society of England & Wales

(Suggested citation: R. Lang, ‘Consequences of the Extension Deal: A Reply to Professor Allott’, U.K. Const. L. Blog (29th Mar. 2019) (available at