Professor Mark Elliott’s ‘The Government’s case in the Article 50 litigation: A critique’ has also been cross posted to Public Law for Everybody


Thanks to a court order, the Government’s case — its “detailed grounds of resistance” — in the Article 50 litigation currently pending before the High Court has been published. I have written before on this matter, arguing that the better view is that legislation is not needed for the purpose of triggering Article 50, but that there are strong policy arguments in favour of involving Parliament — including in relation to the initial decision concerning the initiation of the withdrawal process. The newly published Government case is — putting the point as diplomatically as possible — something of a mixed bag. There are some sound points in it, but also some astonishing and fundamental misapprehensions.

The core issue

The Government, in my view, is on the strongest ground when it focusses on the core issue concerning the interaction of statute law and the prerogative with regard to the operation of Article 50. The essence of that argument can be relatively simply stated, and must necessarily exist, at least in part, as a response to the position advocated by Nick Barber, Tom Hickman and Jeff King in their post on the UK Constitutional Law Association Blog. They argue that any prerogative power that would otherwise be available to the Government in order to make an Article 50 withdrawal decision is displaced by legislation enacted by Parliament, including the European Communities Act 1972 (ECA). Since that Act operates as a source of rights for individuals, they contend, the prerogative cannot be used to remove them.

A response to this argument must focus on the fact that there is, in fact, no relevant conflict between the ECA and the prerogative. Triggering Article 50 does not in itself “change any part of the common law, or statute law, or the customs of the realm”, and therefore does not breach the principle — famously laid down in the Case of Proclamations — that the prerogative cannot be used to do such things. Nor is the principle in Re De Keyser’s Royal Hotel Ltd [1920] AC 508 flouted. As Paul Craig has explained the argument advanced by Barber et al goes far beyond the De Keyser principle. Rather, says Craig, it implies a “new principle” to the effect that “the executive could not exercise the prerogative power to begin the process of amending or withdrawing from a treaty, because this very initiation would impact on, or cut across, the legislation through which that treaty had earlier been incorporated into UK law”. Craig concludes: “There is to my knowledge no case that comes close to establishing this proposition.”

This is a powerful rejoinder to the contention that legislation is necessary, and it finds a degree of expression in the Government’s case. However, even in the Government’s presentation of the core argument, there are some notable missteps. For instance, the Government states that the “decision” for the purpose of Article 50(1) is “the UK’s decision to leave the EU, as articulated in the referendum result”. This proposition is then used to support the characterisation of the Article 50 process as a purely “administrative” or procedural matter — one that takes effect on the “international law plane”, and about which “complaint cannot be made by any individual in the domestic courts”. This, however, seriously mischaracterises what is actually at stake here. The reality is that the outcome of the advisory referendum does not constitute a “decision” for the purpose of Article 50(1), meaning that there is presently no such decision that must or can be notified under Article 50(2). It follows that the matter of notification is far from all that is in issue. The prior, and fundamental, question is whether the Government possesses the legal authority necessary to decide to withdraw from the EU — a decision that, once taken, will trigger the obligation to notify.

However, the Government’s attempt to sidestep that question by suggesting that the “decision” has in fact already been taken is inconsistent with the decision of the Court of Appeal in Shindler v Chancellor of the Duchy of Lancaster [2016] EWCA Civ 469. There, Lord Dyson MR said that “by passing the [European Union Referendum Act 2015], Parliament decided that one of the constitutional requirements that had to be satisfied as a condition of a withdrawal from the EU was a referendum” and that the Act, by providing for a referendum, “contains part of the constitutional requirements of the UK as to how it may decide to withdraw from the EU” (my emphasis). The clear implication is that the holding of the referendum is not exhaustive of the constitutional requirements for the making of an Article 50(1) decision, and that the mere fact that the referendum has been held and has yielded a result does not establish that a decision has been made.

A further oddity in the Government’s treatment of the core issues relates to its argument that “the possibility of the exercise of prerogative powers” can be removed only by “an express restriction” in primary legislation. This proposition is advanced on the authority of a case decided in 1615. More recent authorities, however, adopt a very different view. Take, for instance, the following classical statement of principle found in Lord Dunedin’s speech in De Keyser, in which there is no hint that the prerogative is subject only to “express restriction”:

Inasmuch as the Crown is a party to every Act of Parliament it is logical enough to consider that when the Act deals with something which before the Act could be effected by the prerogative, and specially empowers the Crown to do the same thing, but subject to conditions, the Crown assents to that, and by that Act, to the prerogative being curtailed.

The risk arises that by presenting the core arguments in an exaggerated fashion, the essence of those arguments — which remain sound — become obscured and devalued. Much of the Government’s published case, however, dwells not on what I consider to be the core issues, but on peripheral ones. It is here that the case enters upon the weakest ground. In that regard, I draw attention to three such aspects of the Government’s case in the remainder of this post.

The European Union Referendum Act 2015

The Government says that it would be

constitutionally proper to give effect to the referendum result by the use of prerogative powers. It was clearly understood that the Government would give effect to the result of the referendum for which the [European Union Referendum Act 2015] provided, and that was the basis on which the electorate voted in the referendum.

In the light of this, the Government goes on to contend that characterising the referendum as merely “advisory” is “incomplete and inappropriate” because it implies a “lack of Parliamentary permission to give effect to the result [of the referendum]” or the existence of “some Parliamentary requirement to return by primary legislation before beginning that process in the event of a vote to leave” (my emphasis). These arguments are problematic.

The argument as to a lack of “Parliamentary permission” appears implicitly to concede that such permission is required, while contending that such permission is supplied by the 2015 Act. (This argument, it is worth noting in passing, sits in tension with the Government’s core argument that its authority to trigger Article 50 lies in the prerogative.) However, the Government attempts to conjure the permission that the 2015 Act supposedly provides out of largely thin air. It argues, for instance, that the court should avoid any decision “which would prevent the Government from implementing the result of the referendum, which was provided for by Act of Parliament”. However, the implication — that the 2015 legislation itself constitutes a legal basis for the triggering of Article 50 — is unwarranted. The fact that people expected the Government to trigger Article 50 in the event of a Leave vote, and the fact that the Government said it would do so, cannot go to the distinct question of whether, in the first place, the Act authorises the Government to trigger Article 50.

Then there is the related but distinct argument that characterising the referendum as “advisory” implies some “Parliamentary requirement” of further legislation. The suggestion appears to be that if — as the Government urges — we characterise the referendum as other than advisory, then the need for further legislation evaporates. But this misses the point. No-one is suggesting that any requirement for further legislation is a requirement imposed as such by Parliament. The argument, rather, is that legislation triggering (or authorising the triggering) of Article 50 is a legal requirement that exists independently of the 2015 Act as an underlying and fundamental matter of constitutional law. It is true that that argument would cease to be relevant if the 2015 Act were itself considered to supply executive authority to trigger Article 50, but, for reasons already given, that view is unsustainable.


The Government baldly asserts that:

The decision to withdraw from the EU is not justiciable. Like the decision to join the EEC (as it then was), it is a matter of the highest policy reserved to the Crown. Equally, the appropriate point at which the UK should begin the procedure required by Article 50(2) to give effect to that decision (that is, the notification) is a matter of high, if not the highest, policy; a polycentric decision based upon a multitude of domestic and foreign policy and political concerns for which the expertise of Ministers and their officials are particularly well suited and the Courts ill-suited.

This is a quite bewildering argument that is fatally undermined by its conflation of what are in fact two entirely distinct questions. First, there are the questions whether the UK should withdraw from the EU and when that decision should be notified. (These questions are barely distinct, given that a legal obligation to notify accrues once a decision to withdraw has been taken — hence the need to stick to the line that no Article 50 “decision” to withdraw has yet been taken.) The second question concerns whether the executive has prerogative authority to do either of these things, or whether they could be done only under authority conferred by Parliament.

If these questions are kept distinct — as they must be — the Government’s argument as set out above collapses. Assume, for instance, that the court were to conclude that the executive did have the necessary prerogative authority to trigger Article 50. Then further assume that there was then a challenge alleging that an executive decision to withdraw from the EU was unlawful on some ground or other. In those circumstances, concerning a challenge to the lawfulness of the exercise of an acknowledged prerogative power, arguments about justiciability, high policy, polycentricity and so on would be entirely in point. (Although even then it cannot be ruled out that a court would intervene if a Prime Minister were to purport to use the prerogative so as to trigger Article 50 on nothing more than a whim.)

But the matter that is being litigated does not relate to the questions whether and, if so, when the UK should withdraw. Rather, the central question is whether, in the first place, the executive possesses the relevant prerogative authority to initiate the withdrawal process. Judicial determination of the boundaries of the executive’s power cannot logically be stopped in its tracks by assertions about the policy-laden nature of the decisions that the executive might take if it were found to possess the relevant power. The question in this litigation is a hard question of law, and is indisputably one for the court. The Government’s case is therefore correct but irrelevant: decisions about whether, and if so when, to withdraw are questions of high policy. That observation, however, has no bearing on whether, to begin with, the Government is legally equipped to make and notify such a decision.

Constitutional impermissibility of relief sought

The final point in the Government’s case to which I draw attention is its assertion that the relief sought in these proceedings is “constitutionally impermissible”. The desired relief is a declaration that it would be unlawful to issue an Article 50 notification “without an Act of Parliament authorising such a notification”. (It seems implicit in this argument that it would be equally unlawful for the Government, absent parliamentary authorisation, to make a decision to withdraw from the EU that would trigger the requirement to notify such a decision.) The Government’s argument, such as it is, is that:

If relief were granted it would impermissibly impinge upon the territory of Parliament by compelling the Secretary of State to introduce legislation into Parliament, failing which he would be prevented from giving effect to the outcome of the statutory referendum. [However,] [i]t is well-established that the Court may not grant relief which trespasses on proceedings in Parliament.

The suggestion that declaratory relief in these proceedings would involve the court “trespassing on proceedings in Parliament” is nothing short of absurd. The argument that the Government seeks to advance is that if it is denied prerogative authority to trigger Article 50, then Parliament will be compelled to legislate. By the logic of this argument, it must also be the case that a child who wants to buy sweets but who is denied pocket money by her parents is thereby compelled to resort to theft. Such logic, of course, collapses under the slightest scrutiny.

The reality is that if the Government were to lose this case on the ground that it lacked relevant prerogative power, Parliament would not be required to do anything whatever. Nor would the Government. All that would follow is that the Government, lacking the legal power necessary to trigger Article 50, would have to ask Parliament to give it such authority if the Government wanted to go ahead and trigger Article 50. It would be for the Government to decide whether to seek to procure such legislation. And it would be for Parliament to decide whether to enact it. But no-one would be compelled to do anything. Indeed, if the Government were correct in its argument that relief in this case would be constitutionally impermissible, it would follow that a court could never declare that the Government lacked the power to do something it wanted to do — because the Government would then be forced to procure primary legislation if it wanted to go ahead and do the thing in question.

In support of its “constitutional impermissibility” argument, the Government cites three cases. Analysis, however, reveals that none of them actually provides the support that the Government claims.

First, R (Wheeler) v Office of the Prime Minister [2008] EWHC 1409 (Admin) is cited, in which the court refused to order the Government to introduce a Bill into Parliament in order to give effect to the legitimate expectation asserted by the claimant. However, a court ordering the Government to introduce a Bill is a far cry from a court declaring that the Government cannot do something unless a Bill authorising the action in question is introduced and enacted.

Second, the Government cites R (UNISON) v Secretary of State for Health [2010] EWHC 2655 (Admin), in which the claimant argued that it had a legitimate expectation of consultation before a Bill was introduced into Parliament. The claim was dismissed. Mitting J stated that courts “cannot forbid a Member of Parliament from introducing a Bill”, noting that “[t]o do so would be just as much an interference with Parliamentary proceedings as to require the introduction of a Bill”. Those principles, he held, extended to judicial relief that would delay the introduction of a Bill. However, the circumstances of Unison are very different from those of the present case, in which — for the reasons given above — there is no question of the Government being ordered, directly or indirectly, by the court to introduce a Bill.

The third case, Wheeler v Office of the Prime Minister [2014] EWHC 3815 (Admin), is superficially the most helpful to the Government. The claimant argued that it would unlawfully breach a legitimate expectation for the Government to opt into the European Arrest Warrant (EAW) scheme without a vote in Parliament first being held. Holding that no legitimate expectation had arisen, Sir Brian Leveson P said that in substance the claim was that “unless the House of Commons organises its business in a particular way, and arranges for a vote in a particular form, the courts must intervene and either grant a declaration or issue an order prohibiting the government from taking certain steps unless and until there is such a vote”. That, he said, “would involve the courts impermissibly straying from the legal into the political realm”. He went on to say that granting relief would involve a breach of Article 9 of the Bill of Rights 1689 because “the courts would be called upon to determine whether particular votes held in the House of Commons were, or were not, sufficient to satisfy the alleged expectation”.

Although this reasoning may appear to support the Government’s position in the present proceedings, closer scrutiny reveals that Wheeler and the Article 50 litigation are not on all fours. Wheeler is concerned with the interaction of (on the one hand) parliamentary proceedings and (on the other hand) the placing of public law restraints (by means of the legitimate expectation doctrine) upon the exercise of extant executive authority. In contrast, the central question in the Article 50 litigation is whether it is necessary for Parliament to confer authority upon the Government because, to begin with, the Government lacks the relevant authority. Even if it can be argued at some high level of abstraction that this has implications for parliamentary proceedings — because the Government, if it wants to trigger Article 50, might be put to the trouble of having to ask Parliament to authorise such a step — that point surely pales into insignificance beside the axiomatic constitutional principle that Government action must rest on a satisfactory legal foundation. The upshot is that even if one were willing to buy the highly dubious argument that granting (merely declaratory) relief would involve some form of curial incursion into parliamentary proceedings, it would be necessary to balance that constitutional evil against abdication of the core judicial role of ensuring that the Government actually possesses the legal authority it says it has. Faced with such a choice, one would hardly find oneself on the horns of an insoluble constitutional dilemma.

A final thought

The constitutional issue raised by this litigation is about as fundamental as it gets. It goes to the division of authority between the constitution’s legislative and executive limbs against the backdrop of a referendum and in respect of a matter whose importance cannot be overstated. The essence of the Government’s case concerning the interaction of legislation and the prerogative is, in my view, sound. Against that background, it is surprising, to say the least, to find that core case so heavily obscured by arguments that are not merely peripheral, but weak verging on absurd. It is to be hoped, therefore, that as the case moves forward the central arguments will be sharpened and given the prominence they deserve.

I am grateful to Tom Fairclough, Hayley Hooper and Alison Young for their very helpful comments on an earlier draft. The usual disclaimer applies.

Professor Mark Elliott
University of Cambridge