affiliated to the International Association of Constitutional Law
In the landmark legal decision in Miller v Secretary of State for Exiting the EU, the High Court ruled that the government cannot use the royal prerogative to trigger Article 50 EU Treaty, and so leave the EU, without Parliament’s authority. Since that ruling was given, a flurry of commentary on it has appeared. Some of the commentary in the press has been highly critical and unpleasantly vitriolic in nature, leading to the (eventual) defence by the Lord Chancellor of the independence of the judiciary. Commentary in academia and the legal blogosphere has been much more measured but ever growing in quantity, as the many postings on the UK Constitutional Law blog reveal. This is unsurprising. For many, this is the constitutional case of its generation, and which constitutional lawyer worth their salt would not wish to make their contribution? Clearly, I am not hesitating to make a contribution. However, I do so now in response to what I have found a somewhat surprising development – the large amount of bloggers and commentators who have been critical of the High Court judgement (see, for example, the postings on the Judicial Power Project) and seek to provide what are perceived to be stronger arguments, often in highly technical, elaborate detail, that the government might use.
I disagree with the arguments they make. As the rest of my post makes clear, I stand by the central holding of the High Court ruling, which I (borrowing the expression from Keith Ewing on this blog although not necessarily agreeing with everything he writes about Miller) call ‘Constitutional 101’. I believe the High Court judgement is correct, for what are very important reasons of UK constitutional law, deeply rooted in constitutional history. I do not believe there are strong arguments to be made against the judgment’s central holding. I recognise that others disagree. I also recognise that the uncodified, flexible British Constitution very often provides no clear answer to major questions, and this enables the elaboration of tenable arguments, as in this context. Nothing that I write approves any Parliamentary action that would ‘block Brexit’ and I do not question the good faith of those who write criticizing the High Court judgement. However, I am somewhat concerned by the volume of arguments geared to enabling government action, when the central argument of the litigation revolves around the protection of rights. Should lawyers be so zealous in elucidating arguments that make it easier for government to override rights?
Article 50 TEU states that
‘1. Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.’
The basis of the claimants’ case in the High Court was that the ‘constitutional requirements’ require Parliament’s authority for triggering Article 50 TEU. A starting point for their argument is that the UK enabled its accession to the (then) EEC to be ratified by means of the European Communities Act [ECA] 1972. Under British law, a statute may only be repealed by another statute, and not by the exercise of prerogative power. This is clear from Fire Brigades Union and other cases. However, as the actual effect of Article 50 notification would be to trigger a 2 year timeline at the end of which the UK would cease to be an EU member state (unless extended by unanimity of all EU Council members) triggering Article 50 in fact nullifies the effect of the ECA in UK law. The prerogative cannot be exercised in a manner which would ‘turn a statute into what is in substance a dead letter’ or ‘cut across the object and purpose of an existing statute’ (see again Fire Brigades Union).
Furthermore, because constitutional principle holds that only Parliament may limit or abrogate rights, it follows that Article 50 may, because of its ultimate effect on rights, only be triggered by Parliament. The Laker Airways case provides further support for this proposition. If the UK had not held a referendum, the Government could not have abrogated the rights and responsibilities in the ECA by withdrawing from the EU by unilateral, executive act, and the EU Referendum Act 2015 provides no explicit authority for amendment or repeal of the ECA. It was an advisory not a mandatory referendum. These rights-related arguments are confirmed by Lord Oliver, who stated in Rayner (Mincing Lane) v DTI: ‘as a matter of the constitutional law of the United Kingdom…the Royal Prerogative, whilst it embraces the making of treaties, does not extend to altering the law or conferring rights upon individuals or depriving individuals of rights which they enjoy in domestic law without the intervention of Parliament.’
It therefore follows that it would be necessary for an Act of Parliament to provide the authority for Article 50 notification (and prospective repeal of the ECA two years after making the notification).
Now there is resistance to these arguments. It is contended, by eg Elliott, that the argument that, by triggering Article 50, the Prime Minister would be taking power to ‘prospectively repeal’ the ECA, is incorrect. But as David Pannick argues, ‘the crucial point is that, as a matter of law, Article 50 notification commits the UK to withdrawal from the EU, and so is inconsistent with the 1972 act.’
I will presently examine the counter arguments in greater detail. But for the meantime, I want to stress the foundational and deeply historical nature of the key principle at issue in the claimants’ case. Phillipson, after a searching and highly detailed examination of caselaw, has contended that ‘arguably that normative concern typically generated by the Executive’s use of prerogative to frustrate statutory purpose is absent in this particular case.’ (G Phillipson, ‘A Dive into Deep Constitutional Waters: Article 50, the Prerogative and Parliament’, (2016) MLR 1087). Yet, what could be more fundamental and normatively compelling than the reasoning behind the key principle at issue here: namely, that the government cannot change legislation by executive fiat? As Keith Ewing writes, ‘As the High Court has made clear there is no question that the government cannot change domestic law by means of prerogative power. Constitutional Law 101.’
During the EU referendum, voters were constantly urged to ‘take back control’ and regain Parliamentary sovereignty from the EU. Yet in what sense would Parliament be ‘taking back control’ if the government is able, using its ancient prerogative powers, to manage the whole EU withdrawal process without any significant parliamentary involvement? That would be extremely undemocratic, and democracy is what we are told the EU referendum was about. Furthermore, the Court’s judgement makes clear that the exclusion of Parliament in the Article 50 process is not only undemocratic, it is illegal. There is a wealth of case law supporting the claimants’ case, some of it dating back to the 17th century and the English civil wars. These wars, and the ejection of two kings, established that Parliament is sovereign and the Executive cannot ignore it, where it has no legal authority to do so. This judgement makes clear that the government does not have any such legal authority in the context of triggering Article 50.
So I take the claimants’ central argument to rest on compelling, fundamental, historically supported principles of UK constitutional law. The High Court judgement upholds them. Nonetheless, the Government is appealing this decision. It will need stronger arguments if it is to succeed. There have certainly been plenty of lawyers willing to step up and provide arguments that might provide a compelling case for the government. If the government wins in the Supreme Court, then academic constitutional lawyers will have helped them win. The remainder of this blog considers a few such arguments, and attempts a critique of them.
One argument that has been aired and some consider persuasive is that of John Finnis. Finnis argues that the claimants’ argument in Miller contained a ‘glaringly fallacious syllogism’. This, according to Finnis, takes the following form: (1) statutory rights enacted by Parliament cannot be destroyed by executive action without Parliamentary statute/authorization; (2) myriad rights acquired by UK persons under EU Treaties given effect in UK by s. 2(1) ECA 1972 are statutory rights enacted by Parliament; (3) therefore no executive action can be taken to terminate the UK’s adherence to those Treaties without authorization by statute/other parliamentary authorization. Finnis rejects these conclusions as ‘fallacious’ because it is his argument that ‘rights acquired by virtue of s. 2(1) ECA are not “statutory rights enacted by Parliament”’, but instead ‘rights under EU law, as it stands “from time to time”’. To explain his argument, he draws an analogy between EU law and some provisions in double-taxation treaties, which are given statutory effect in UK law. Finnis argues that, although rights created by the ECA and Tax Acts (eg Taxation (International and Other Provisions) Act 2010, [TIOPA]) are statutory in the broad sense that domestic statutory provision is in both cases a necessary condition for those rights’ legal efficacy in the UK, they are non-statutory in the sense that Parliament was not the original source for the rights, which in fact may be changed, or eliminated, at international or EU level, without intervention by the UK Parliament. Thus he postulates a fundamental asymmetry between the introduction of treaty-based rights into UK law and their termination. Introduction can only be by statute. Termination, however, can be by executive action. ‘Constitutional law 101’ cannot apply, because these are not statutory rights. Finnis also contends this argument was not made sufficiently clearly during the High Court litigation: ‘The true relevance of the double tax treaties to the subject of this litigation thus disappeared from view.’
Mark Elliott has described Finnis’s argument as ‘outstandingly clear — and in my view compelling’, continuing that this ‘Shows exactly how Government *should* have argued Article 50 case’. Elliott and Hooper, in their ‘Critical reflections’ on the Miller judgement, also argue that ‘The issue is whether the rights whose removal it is said Article 50 would impact upon in effect form part of ‘the law of the land”, i.e. domestic law.’
But is Finnis’s argument really compelling? I think at least 4 arguments can be made against it.
First, Finnis’s analogy does not hold. Bilateral double taxation treaties are not comparable to multilateral EU treaties and do not give rise to rights comparable to rights acquired through EU membership. Section 2 TIOPA ‘gives effect’ to double tax arrangements made with other territories. However, there the analogy ends. In what sense are these ‘rights’? Double tax treaties are concluded to ensure people do not pay tax twice on the same income. But these tax treaties do not themselves impose taxes, nor create rights. EU treaties do create individual rights. Tax treaties accord at most a privilege to UK taxpayers to get relief from UK tax if there are also obligations to pay tax on that income elsewhere. The domestic obligation continues, but is suspended, and, if the treaty is terminated, an individual becomes liable to pay tax to UK authorities. This is not the same as having an entitlement. Individuals concerned do not lose a ‘right’. In fact, if there are any rights at issue, then these are the ‘rights’ of states to tax individuals.
Second, Finnis’s analogy does not take account of the special and singular nature of EU membership. Even if every single EU right is not individually itemized and detailed in the ECA, this does not mean that EU rights are not part of national law. EU law is, according to the Court of Justice in van Gend en Loos, a ‘new legal order’ which gives individuals directly effective, enforceable rights which become part of their national legal heritage. That is the point of direct effect, a crucial doctrine in EU law whereby individual rights can be enforced in national courts. EU law and EU rights are not a foreign body, a distinct and separate element, in the way most international treaties are regarded by UK law. Direct effect has allowed EU law to pierce the veil of domestic law and for EU law to become directly enforceable in national law, as part of national law. The ECA provides a statutory basis for this (and a mechanism to deal with the demands of parliamentary sovereignty) but direct effect also ensures a direct and unbroken link between enforceable national rights and their ultimate EU source. The according of vertical direct effect to unimplemented directives recognises this. This means that State action or inaction (ie failing to implement an EU Directive) alone cannot override individual EU rights.
Third, Finnis also seeks to base his case on the argument that EU rights already can be removed by Executive action alone, ie if EU law is changed by EU Institutions. This he believes, illustrates that UK Parliament involvement is not needed for their elimination. But in what sense is this so? Repeal or elimination of rights in the EU treaties would have to be by treaty amendment, a serious enough repeal to require approval by the UK Parliament. Absent such approval, the Treaty could not come into force. In fact, I would ask how many such rights have been repealed by the EU in this way? Not any of the fundamental principles to be found in the EU treaties, such as free movement of persons, or equal pay for men and women. However, in cases outside of treaty amendment (where an EU Regulation is repealed for example) where rights are eliminated, the House of Commons European Scrutiny committee has the possibility of assessing the relevant EU measure, which should be deposited in Parliament by the Government. In this way, EU law is again different from double taxation treaties, because of the special role accorded to scrutiny of EU law by the UK Parliament. That is why there exists a dedicated EU Scrutiny Committee and also the House of Lords EU and sub-EU committees. In particular, the European Scrutiny Committee has the remit to look at the significance of EU proposals (which should include repeals of significance) and decide whether to clear the document from scrutiny or withhold clearance and ask questions of the Government. All documents deemed politically or legally important are reported on in the Committee’s weekly Reports. Therefore, if the EU Scrutiny Committee is doing its job, Parliament should have a role in the repeal of EU rights.
Finally, even if, contrary to what I have just argued, the ECA does not create statutory rights in the sense required by Finnis, it is undeniable that the European Parliamentary Elections Act 2002 does provide statutory rights, and so is in no sense comparable to the double taxation relief treaties to which Finnis makes reference. Section 8 of this Act states that ‘(1) A person is entitled to vote as an elector at an election to the European Parliament in an electoral region if he is within any of subsections (2) to (5).’ This provision clearly creates, not just gives legal recognition to a right, and this is clearly a right that will never be restored by Parliament if the UK leaves the EU.
Therefore, I think there are (at least) 4 reasons why the Finnis argument should not defeat ‘Constitutional law 101’.
Another point, possibly helpful to the Government appeal in the Supreme Court, is that employed by Robert Craig. Craig argues that there is no need for any legislation before Article 50 may be triggered because there already exists UK legislation which provides just the statutory authority necessary for executive action in this area. This legislation (originally EU (Amendment) Act 2008, now EU Act 2011) suspends any prerogative source of authority to act. Craig argues that the EU Act 2008 incorporated the Lisbon Treaty into UK law and inserted it, in terms, directly into s. 1(2) ECA. This means, argues Craig, that Article 50 is now incorporated in UK law by primary legislation, and so executive discretion to use it is already authorised and approved by Parliament. This is quite contrary to the submission made by the Secretary of State in Miller, who argued that neither the 2008 Act nor the 2011 Act restricted the Crown’s prerogative power to give notice under Article 50, so the Government would have to engage in a clear change of course if it were to embrace Craig’s argument.
Craig’s argument is highly detailed and expanded elsewhere. Keith Ewing also appears to support it in this blog (‘I had thought that a general power to this effect already existed in the 1972 Act (as amended by the European Union (Amendment) Act 2008’… by incorporating Article 50, Parliament has provided that authority and government power to withdraw as a matter of domestic law.’)
However, rather like Finnis’s argument, I believe Craig’s contention fails to do justice to the nature of EU law, and its relationship with UK law. First, I argue that incorporating the Lisbon Treaty into UK law, and thus recognizing the new Article 50 procedure in UK law, is not the same thing as, nor is it sufficient, to authorise the executive to trigger Article 50 according to the UK’s ‘constitutional requirements’. For a start, there is certainly nothing in the EU Act 2008 which explicitly does anything like provide such authorization. Given that to do so would have a considerable impact on UK constitutional law, giving it a power to override statute, one might expect it to have been done explicitly and in clear language, in the detailed way one finds, for example, in the Fixed Term Parliaments Act 2011, which switched the power to dissolve parliament from a prerogative to a statutory basis. The EU Act 2008 did nothing like this.
Secondly, Craig’s argument entails that Article 50 confers a right (according to s.2(1) ECA) for the UK to exit the EU. He writes: ‘Furthermore, s 2(1) ECA gives direct effect to that right without the need for further enactment’ (Craig, ‘Casting Aside Clanking Medieval Chains: Prerogative, Statute and Article 50 after the EU Referendum’ (2016) MLR 1057). I disagree. Article 50 is not directly effective — it is not clear, precise and without need for further implementation, according to relevant European Court of Justice [ECJ] caselaw, going back to van Gend en Loos. Craig contends that the meaning of direct effect has changed in EU law. Yet, even if it has, at a bare minimum to be directly effective, an EU provision should impose clearly defined obligations on Member States, and provide enforceable, individual rights in domestic law, as already discussed earlier in this post. In fact, many EU treaty articles are not directly effective, because they do not do this. They concern EU Institutional procedures, and do not relate to individual claims under national law. Much of Article 50 is like this, in that most of it relates to procedures in the EU Council or Parliament, and is certainly not enforceable at the behest of individuals. Only Article 50(1) explicitly relates to national law, in its reference to national ‘constitutional requirements’, but in this case the reference to these requirements is far too vague to satisfy any test of direct effect, for which the provision must be clear and precise. For example, in Francovich, the ECJ found that Directive 80/987 lacked direct effect because it did not provide sufficient certainty on the nature of the organization, operation and financing of the guarantee institutions required to provide minimum protection to employees in the event of the insolvency of their employer. Indeed, to describe Article 50 as directly effective seems to misunderstand the purpose of direct effect, which is to provide individuals with enforceable rights. The doctrine was elaborated in the ECJ to ensure that states could not undermine individual rights by failing to comply with EU obligations. Can states then benefit from direct effect? This seems to set the doctrine on its head. Therefore, I would argue that Craig, along with Finnis, misunderstands the relationship of EU law and national law, and the very nature of EU law itself. It is a misnomer to use the term ‘direct effect’ in the context of Article 50 in order to circumvent any explicit national statutory authorization for the government to trigger withdrawal proceedings.
Hooper and Elliott also argue in their ‘Critical Reflections’ on the Miller judgement that ‘The Court’s approach reflects a particular (and contested) notion of “representative parliamentary democracy” that arguably traces its origins to the writing of Edmund Burke’. I disagree with Elliott and Hooper’s critique of the Court on the parliamentary/popular sovereignty issue. However, I recognise that issues of representative democracy, popular and parliamentary sovereignty are immense in nature and cannot be properly dealt with here. Therefore I make just two comments – one very general, and the second of particular relevance to litigation of Miller in the Supreme Court.
First, I believe the High Court was correct in its treatment of the issues of parliamentary and popular sovereignty. The EU Referendum Act 2015 was an advisory referendum. The position of referendums is somewhat unclear in UK constitutional law, but there is no authority that they can override parliamentary sovereignty. The referendum vote provided no mandate for the government to override rights provided by UK statute without first obtaining parliamentary authority.
UK constitutional law (Scotland may be different) has not acknowledged the principle of popular sovereignty as a legally determinative principle. (I write ‘legally determinative’ because I do not argue that popular sovereignty, or referendums, are of no significance at all.) Instead, sovereignty is seen as resting with the Crown in Parliament, and UK politics as based on representative democracy. Although, in 1653, Oliver Cromwell’s Instrument of Government stated that legislative power resided in the person of Lord Protector ‘and the people’ this doctrine did not continue after Cromwell’s fall. England did not develop a tradition of popular sovereignty in constitutional terms, even if ‘sovereignty of the electorate’ has been used in political terms. However, even here, sovereignty of the electorate enables the people to choose the Members of Parliament who will form the government, rather than grounding the British Constitution in the authority of the people. In this way, the people is constitutionally embodied in Parliament. Parliamentary sovereignty possesses a political dimension, given that Parliament provides a forum in which complex issues of public governance are debated in a (hopefully) ordered and rational manner.
UK constitutional law simply lacks the resources to argue that popular sovereignty, namely the referendum vote, can somehow trump parliamentary sovereignty and provide a mandate for the government to trigger Article 50 without parliamentary authority. Some have argued that ‘popular sovereignty’ is highly relevant in the context of the EU referendum, and High Court litigation. Vernon Bogdanor, for example, writing in the Sunday Times, argued that the constitutional issue in the High Court ruling is that of Parliament versus the people, not government versus parliament. Yet how can Bogdanor reach this conclusion? The Miller case was about the powers of the executive versus Parliament, the question being whether the executive can act without parliamentary approval or at least discussion. There was no legal framing of this as an issue of parliamentary versus popular sovereignty.
However, should popular sovereignty be argued as germane to the Miller litigation, the government should be careful how it makes the case. Litigation of Miller in the Supreme Court will include interventions from all 4 nations of the UK, each making its own submissions. The doctrine of popular sovereignty, even if not able to override parliamentary sovereignty in the English constitutional setting, is seen as being of particular relevance in Scotland where it has a different resonance. In MacCormick v Lord Advocate, Lord Cooper famously stated that: ‘[the] principle of the unlimited sovereignty of parliament is a distinctively English principle which has no counterpart in Scottish constitutional law.’ Instead, reference is often made to a Scottish tradition of popular sovereignty claimed to date back to the 1320 Declaration of Arbroath. The claim is that, prior to the 1707 Act of Union, sovereignty rested in the Scottish people, and that it still does so, in spite of Diceyan notions of parliamentary sovereignty. Given that the 1707 Union brought into being the British state, it is argued that we should recognise that the British constitution is not simply the English constitution, but that Scottish constitutional principles (and Northern Irish, and Welsh) play their part in a multiple constitutional order. Reflection on the entirety of Union, and its constitutional basis, poses the question of whether, at least in Scotland, popular sovereignty might form the basis of Scotland’s own right to determine whether or not it exits the EU. Scotland voted (62% of those voting) for the UK to remain in EU. Nicola Sturgeon argued that to withdraw the UK from the EU, against Scotland’s will, would be ‘democratically indefensible’.
Therefore, what reason is there to argue for a legally enforceable popular sovereignty in the EU referendum context, when the UK has no recognizable tradition of popular sovereignty, nor any great tradition of holding referendums, if popular sovereignty is ignored in context of the Scottish vote, where there exists a much clearer tradition of sovereignty of the people? If the referendum vote is to be analysed as a feature that reflects a growing recognition of a principle of popular sovereignty, even if there is no clear legal authority for this, then surely the argument must apply in the case of the a Scottish vote, where there was a clear vote of a different nature? Is this an argument that the UK government would wish to open up?
In Miller, the claimants asserted, and the government conceded, that a notification under Article 50 cannot be unilaterally withdrawn by the notifying Member State. However, such a position is not accepted by everyone. Authorities such as Jean-Claude Piris, Lord Kerr and Paul Craig have argued that Article 50 may well be reversible. Indeed it is possible that the government might try to change its position, and argue that a notification under Article 50 could be revoked (although this is certainly not indicated in the government’s Grounds of Appeal). This would be the opposite of its High Court concession that triggering Article 50 would inevitably result in the withdrawal of the UK from the EU (and so lead to loss of individual rights). In this way, the causal chain between triggering Article 50 and automatic loss of rights in 2 years would be broken. This is an important point. The irreversibility or not of Article 50 was certainly taken very seriously in the High Court. The Transcript of that hearing reveals the Lord Chief Justice as stating (at 193): ‘From our perspective, it seems to me absolutely essential for us to decide that it is irrevocable.’
An argument that Article 50 is reversible might be a strong argument against the claimants’ case. Yet such a change of argument brings powerful disadvantages for the government. First, it would involve a question of EU law, because the issue of the revocability or not of Article 50 could only be conclusively determined by the European Court of Justice. For this to happen, a reference would have to be made to the European Court in Luxembourg, and this would take time, further delaying the date on which Article 50 could actually be triggered. Such a reference would need to be made under Article 267 of the Treaty on the Functioning of the European Union [TFEU], which provides for preliminary reference from national courts to the European Court in Luxembourg. Under Article 267 TFEU, a domestic court of last instance, such as the UK Supreme Court, is under a duty to make a reference if a decision on an issue of the interpretation of the EU Treaties, in this case the revocability of Article 50, is necessary for the domestic court to give judgment. If the government maintains its previous argument that Article 50 is irrevocable, then arguably, a decision on this point is not necessary for the Supreme Court to give a judgment, although this issue is for the Court to decide. It is the Court, and not the parties, that decides whether to make the reference. If the government changes its arguments, a reference becomes more likely.
Moreover, such a reversal of the government’s arguments would be politically risky, as it would amount to acknowledging that the UK might decide not to leave the EU, and that Brexit does not mean Brexit after all, that Brexit is not unstoppable. This might not seem a very likely position for the Government to take. Yet given that such reversal must be the government’s best chance of winning in the Supreme Court, might it nonetheless take this course? For example, in a statement made in the House of Commons on 7 November, the Secretary of State for Exiting the EU suggests a route to this. In response to a question why government lawyers had conceded that an Article 50 notification would be irrevocable, David Davis said: ‘The reason was not really a point of law so much as a point of constitutional and political reality. I did not see it as possible that we could reverse the decision of the British people.’
Clearly law and politics can be separated, but how ethical would it be for the government and their lawyers to do so in this case? For the government to shift their position would be tantamount to arguing that although they now believe they argued a position wrong in law in the High Court, and now wish to make the opposite argument, even if that new argument is upheld, they have no intention of acting on it, no intention of ever revoking any withdrawal notice.
In conclusion, I argue that the strength of the High Court holding, ‘constitutional law 101,’ along with the reasons I give for defeating some new arguments the government may seek to advance in the Supreme Court, mean that it should lose its appeal. Of course, I have not considered the myriad of other arguments which have been advanced as to why the High Court judgement is wrong, and no doubt more are still to come. Yet, at base, I still believe this case to be about the point that the government should not be able to repeal legislation (and especially not legislation that provides individuals with rights) by executive fiat. We should not lose sight of this point.
Sionaidh Douglas-Scott, Anniversary Chair in Law, Queen Mary University of London
(Suggested citation: S. Douglas-Scott, ‘Miller: Why the Government Should Still Lose in the Supreme Court (even with new arguments)’, U.K. Const. L. Blog (15 Oct 2016) (available at https://ukconstitutionallaw.org/))