1. Changing UK law or rights

The proposition that the executive cannot use the prerogative to change the law has a respectable pedigree (see e.g. Lord Coke in Case of Proclamations (1610) 12 Co. Rep. 74, 76; Bill of Rights 1689 (England and Wales); Claim of Rights 1689 (Scotland)). This applies a fortiori to the use of the foreign affairs prerogative, because, in the constitutional law of the UK, international treaties do not have effect in municipal law unless or until legislation authorises such effect. On the face of it, however, giving notice under Article 50 does not change the law in any way. It merely sets in motion negotiations which might (depending on their outcome) lead to the UK leaving the EU two years (or perhaps more) afterwards, and as a consequence there might (depending on decisions taken later by the executive and the Houses of Parliament) be a change in municipal law in the UK.

Counsel for the claimants, however, characterised the process differently, likening an Article 50 notification to firing a bullet which would inevitably lead to the UK ceasing to be a Member State of the EU. This, they said, would automatically change the law of the UK, because EU law, which (said counsel) was part of the law of the UK, would cease to have effect. UK legislation might be made to preserve some of its effects, and some requirements of EU law are already contained in UK legislation; but some provisions, such as those relating to elections to the European Parliament, would inevitably cease to have any effect once the UK ceased to be entitled to be a Member State. Simultaneously, certain acts of EU institutions would cease to be sources of UK law.

That argument assumed that the Government would not be allowed to withdraw its Article 50 notice before the end of the period which Article 50 allows for negotiations (two years, unless extended by the European Council acting unanimously, in agreement with the UK). All parties were prepared to accept this, but it is a controversial proposition. As leading counsel for Miller has accepted (D. Pannick, “Brexit: Is It a One-Way Street or Could the UK Change Its Mind?”, The Times, 30 March 2017), the normal rule in relation to withdrawing from a treaty is that “a notification or instrument” given by a state party to a treaty, communicating to other parties to the treaty that it intends to withdraw from the treaty, “may be revoked at any time before it takes effect”: Vienna Convention on the Law of Treaties (1969), Article 68. Article 50 TEU contains nothing to displace that rule. It might be thought absurd if the UK, having instigated the arduous process of negotiating withdrawal from the EU Treaties and finding the result of the negotiations unsatisfactory as they approached the time-limit, could withdraw the notification, then serve a new Article 50 notification and so re-start the clock until negotiations yielded an outcome more to her liking. On the other hand, it would be absurd if a sovereign state could be prevented from acting in the best interests of the state as it sees them from time to time, or if a new government, elected during the negotiations with a manifesto pledge to halt the process of withdrawal, were nevertheless to be forced to follow the policy of its defeated predecessor. With absurdities either way, there is no compelling reason to imply into Article 50 TEU a provision displacing the normal rule contained in Article 68 of the Vienna Convention on the Law of Treaties. But asserting the possibility of revoking the notification might have made the Government seem less than fully committed to leaving the EU, and the Government did not want the question to be referred to the Court of Justice of the EU (CJEU) for a preliminary ruling, causing delay and giving the CJEU an unpalatable role in setting the Government's legal powers. Nevertheless, the better view, it is submitted, is that giving notice under Article 50 is the start of a tortuous journey with no certain end-point, not the shooting of a bullet directly to its target.

The claimants’ argument further assumed that EU law and EU rights are part of the municipal law of the UK. The majority of the Court accepted this, reasoning (1) that “the EU Treaties, EU legislation and the interpretations placed on these instruments by the Court of Justice are direct sources of UK law” and “an independent and overriding source of domestic law” (at [61], [65]), and (2) that the ECA is the means whereby they are able to have effect, and overriding effect, in UK law notwithstanding the normal constitutional principle of parliamentary sovereignty; EU law is not delegated legislation (at [68]). The overriding quality of EU law is, however, limited: it does not override “legislation which alters the domestic constitutional status of EU institutions or of EU law”, because the status of EU law in domestic law depends on the operation of the fundamental constitutional principle of parliamentary sovereignty (at [67]). The Court did not elaborate on the implications of this, but seems implicitly to have been arguing that an Article 50 notice would lead to a change in “domestic” law and rights (i.e. EU law and rights which take effect in the UK by virtue of the ECA) by rendering it no longer applicable in the UK. On this view, the prerogative could not be used to produce such a result, and an Act of Parliament would be needed to authorise the notice.

This reasoning is flawed, as Lord Reed and Lord Carnwath, dissenting, argued at [216]–[218] and [262], respectively. So far as substantive requirements of EU law have been given effect by way of UK legislation, they can properly be regarded as domestic law, which would continue to have effect unless and until UK legislation amended or repealed them. But those parts of EU law which have direct effect or are directly applicable in the UK are not properly regarded as domestic law merely on account of their being enforceable in UK courts and tribunals by virtue of the ECA. EU law and rights are autonomous. As R. v Secretary of State for Transport, ex parte Factortame Ltd. (No. 2) [1991] A.C. 603 made clear, they cannot be amended, suspended, revoked or repealed by an Act of Parliament, and in the case of certain rights (such as those of UK citizens in other Member States) could never be created by an Act of the UK's Parliament. Only by withdrawing from the EU, or, perhaps, repealing the operative parts of the ECA, can EU citizens in the UK be deprived of their rights under EU law before UK courts and tribunals.

The majority recognised the separate sources of legal authority for EU law and municipal law when rejecting the argument that EU law was comparable to delegated legislation, made by EU institutions under powers delegated by the ECA (at [68]); but they failed to recognise the corollary, that EU law and rights exist in a different legal order from the UK legal orders. The role of the ECA is to require courts and tribunals in the UK to recognise and give effect to them despite their foreign legal character. The Act is best seen as a channel running between two legal systems, allowing EU law and rights to be drawn on in Member States’ systems, subject to various constitutional filters which may differ between Member States. (See further J. Jowell, D. Oliver and C. O'Cinneide (eds), The Changing Constitution, 8th ed. (Oxford, 2015), ch. 5.)

The majority's view that EU rights are domestic rights might be expected to have led them to conclude that the prerogative could, unless limited by statute, be used to restrict or remove them, in line with the general approach they adopted at [52]–[53]. They seem to have reasoned, however, that the prerogative could not be used in such a way as to bring about a constitutional change as significant as that caused by leaving the EU. This change would be different in kind, not merely in degree, from the abrogation of particular rights (at [81]), and was so significant that it required legislation. This is discussed in depth by Professor Mark Elliott in his illuminating article in this volume, to which I am greatly indebted (see [2017] C.L.J. 257–288). Here it suffices to say that any legal rule founded on the category of “significant constitutional change” is so uncertain in its scope and effect that it offers inadequate guidance to Government and courts, and is unjustified.

2. Statutory authority for, or exclusion of, use of the prerogative

In the light of their conclusion concerning the scope of the prerogative, the majority need not have considered the effect of the ECA on the prerogative. Nevertheless, they discussed the issue at length. The Act says nothing about withdrawing from the EU (unsurprisingly, since the EU Treaties contained no mechanism for withdrawal until 2009), yet the majority read it as prohibiting, by necessary implication, the use of the prerogative to give notice under Article 50. The Act, they wrote, contemplated the UK's continuing membership of the EU, not its departure from it (at [88]). It thereby excluded the prerogative as a source of authority to start the process of withdrawal (at [89]). “It would scarcely be compatible with” the provisions of the ECA “if, in reliance on prerogative powers, ministers could unilaterally withdraw from the EU Treaties, thereby reducing the volume and extent of EU law which takes effect domestically to nil without the need for parliamentary approval” (at [88]).

The majority relied (at [88]) particularly on three matters: the long title to the ECA, which described it as being to make provision for the enlargement of what became the EU; the side-note to s. 2 of the Act, referring to general implementation of the treaties; and the provisions requiring parliamentary approval for new arrangements before the Government committed itself to them. The majority thought that these pointed away from there being a prerogative power to take steps leading to the UK leaving the EU.

These are tenuous grounds for treating the Act as impliedly supplanting the prerogative. The long title of a Bill is a mechanism for ensuring that extraneous matter is not inserted during the Bill's passage through Parliament. The side-note is a drafter's guide, not part of the text, and is not open to amendment during a Bill's passage. The need for parliamentary approval under the Act arises before the Government commits itself to a new arrangement, not before discussions begin as to what the new arrangement is to be. None of these can make up for the absence of any express provision excluding the royal prerogative; the majority effectively stood on its head the important constitutional rule that only clear statutory provisions can displace a prerogative power.

In earlier cases on which the claimants relied, such as Attorney General v De Keyser's Royal Hotel [1920] A.C. 508, Laker Airways Ltd. v Department of Trade [1977] Q.B. 643 and R. v Secretary of State for the Home Department, ex part Fire Brigades Union [1995] 2 A.C. 513, it was held to be an abuse of power to use the prerogative to evade rules or procedures which had been clearly laid down in statute to cover the situation in question. But the ECA provided no such rules or procedures.

The reasoning of the dissenting Justices is more persuasive. Lord Reed J.S.C., with whom Lords Carnwath and Hughes agreed, concluded that:

(See [177] et seq., [243], [281], and also discussion of the Divisional Court's decision at < https://ukconstitutionallaw.org/2016/11/08/david-feldman-brexit-the-royal-prerogative-and-parliamentary-sovereignty/>.)

We should beware of courts magically producing statutory prohibitions where no such prohibition appears on the face of the Act, particularly in areas of constitutional sensitivity.

The majority thought that it was upholding parliamentary supremacy, but that would have been maintained by recognising that parliamentary legislation (which the Government calls the “Great Repeal Bill”) would be needed to amend or repeal the ECA before withdrawal from the EU could take effect, and that Parliament could keep itself involved in the meantime by exerting the usual mechanisms of ministerial responsibility. The majority, however, entirely discounted the political accountability of the executive to Parliament under our Constitution, as Lords Reed and Carnwath, dissenting, pointed out at [240], [248] et seq.