The judgment of the High Court in R (Miller) v Secretary of State for Exiting the European Union  EWHC 2768 (Admin), decided on 3 November 2016, has been subjected to astonishing and wholly misguided criticism by politicians and the press. Let us be clear, then, at the outset of this post that we entertain no doubt whatever about the constitutional appropriateness of the Court ruling on the question that was put to it in Miller. The question for the court was solely a legal question about the extent of executive authority. It was not, as some tabloid newspaper outlets have suggested, an opportunity seized by an elite judiciary to usurp the will of the people. Questions on the legal limits of governmental authority are par excellence questions for courts in any democracy based on the rule of law.
To criticise the Miller judgment — and the judges who delivered it — as undemocratic is not only wrong, it is dangerous. The independence of the judiciary, and its resulting capacity to determine legal questions, is imperative. Politicians and the media act grossly irresponsibly when they criticise judges for fearlessly and impartially adjudicating upon legal disputes. An independent and impartial judiciary is of course a cornerstone of any functioning democracy. In short, Miller was solely concerned with whether the Government has the power to trigger Article 50. The Court’s declaratory ruling simply states that ‘the Secretary of State does not have the power under the Crown’s prerogative to give notice pursuant to Article 50 of the TEU for the for the United Kingdom to withdraw from the European Union.’ The ruling has nothing to do with the wisdom of triggering Article 50, nor does it affect, let alone displace, the outcome of the referendum on EU membership. The highly charged political context should not detract from the plain fact that the decision in Miller is an example the Court discharging its ordinary constitutional function.
However, these considerations do not (and should not) immunise judicial decisions against critical legal analysis. In this post we offer such an analysis. To this end we focus upon on three issues raised by the case. First, we argue that the Court was wrong to conclude that there is a category of domestic statutory rights flowing from EU membership which would be displaced by the operation of the prerogative in the Article 50 context, and that this would in turn be contrary to the operation of the constitutional principles governing the extent of the prerogative and its relationship with statute. Further to this, we contend that the Court is insufficiently sensitive to questions about the source of the relevant rights, and the implications for legal doctrine regarding whether a given right is sourced in a domestic statute, in the common law, or emanates from an international treaty. Secondly, we question the Court’s conclusion that, in light of ‘background constitutional principles’, the European Communities Act 1972 (‘ECA’) is to be read as having relevantly displaced the executive’s prerogative authority to trigger Article 50. Third, and relatedly, we examine the notion of ‘sovereignty’, contrasting parliamentary and popular notions thereof and suggesting that the Court may have been insufficiently sensitive to the relationship between those two phenomena.
This piece has two overarching aims. One aim is to argue that in certain respects the Court simply goes astray and gets the law wrong. The other aim is to demonstrate that the Court enters into highly contestable constitutional territory in respect of certain aspects of its reasoning. While the very contestability of that territory makes it difficult to argue conclusively that the Court did indeed go wrong, the Court’s failure to openly acknowledge and engage with the contestable nature of the relevant issues is itself a concern about the judgment.
The nature of the relevant rights
The decision in Miller turns upon the application of the constitutional principle that (as the Court put it) ‘the Crown has no power to alter the law of the land, whether it be common law or contained in legislation’. Triggering Article 50 plainly will not alter or repeal the ECA itself. That will, quite properly, be a matter for Parliament, via the so-called ‘Great Repeal Bill’. But will Article 50 nevertheless alter ‘the law of the land’ by ultimately causing a removal of at least some of the legal rights presently enjoyed by individuals as a matter of EU law?
The answer to that question turns in part on whether — assuming relevant rights are at stake at all — triggering Article 50 will itself relevantly impact upon them. Arguably it will not. Triggering Article 50 will do no more than initiate the negotiation process on the diplomatic (and international) plane, the outcome of which cannot be known. A further reason for doubting that triggering Article 50 relevantly impacts upon any pertinent rights concerns the revocability question. This amounted to the elephant in the room in the Miller case, the Court simply assuming — because the parties conceded — that a notification under Article 50 cannot be unilaterally withdrawn by the notifying Member State. We do not propose to address that question in any detail here, except to note that the matter is contested, that leading authorities — including Paul Craig — have argued that a notification can be unilaterally withdrawn, and that the Court proceeded simply on the assumption that unilateral withdrawal is not possible. The nature of the Article 50 process may yet form the substance of a reference for a preliminary ruling to the European Union judicature when the Miller appeal is heard by the Supreme Court early next month. While it is not hard to fathom the political and tactical reasons for the parties’ reluctance to press this point it remains unfortunate in legal terms that the case was decided on the basis of an assumption that might well be wrong.
Let us turn, then, to the question whether any relevant rights are actually in play. 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.
It is plainly the case that EU law rights become domestic law rights when, for instance, separate primary legislation provides for such rights. However, rights enshrined in domestic law in this form are not threatened by the triggering of Article 50: they will disappear only if Parliament repeals the relevant legislation, or if Ministers to do exercising authority given to them (e.g. under the Great Repeal Bill). The key issue therefore concerns the legal status of those EU law rights which are enforceable in the UK only by operation of the ECA 1972. There are two ways of looking at this. The first sees the ECA as a conduit for the effect in domestic law of rights that are EU law rights. This, we contend, is the correct view. The second view — preferred by the Court — sees the ECA as the creator of domestic rights. By taking this view, the Court triggers the well-established constitutional principles concerning the extent of prerogative power. The upshot, on this view, is that the Government cannot use the prerogative to remove the rights in question — for that would entail using prerogative authority to ‘alter the law of the land’.
John Finnis has argued strongly in favour of the first view. He says that ‘rights acquired by virtue of s. 2(1) ECA are not [as counsel for the lead claimant put it] statutory rights enacted by Parliament’. Rather, says Finnis, they are ‘rights under the treaty law we call EU law, as it stands “from time to time”’. There are therefore, he continues, ‘two necessary conditions’ for the existence and operation of such rights: ‘One is that they be rights from time to time created or arising by or under the Treaties. The other is the silent operation of s. 2(1) ECA.’ We find this view, which is consistent with the analysis that one of us put forward shortly after the referendum, persuasive.
Finnis goes on to draw an analogy between the present context and the operation of ‘double tax treaties’, whereby domestic entitlements to tax deductions take effect through a combination of treaty arrangements and UK legislation. The UK legislation creates domestically enforceable rights to tax deductions which are provided for in treaties designated by Order in Council. In this context, Finnis regards the distinction between executive action on the international plane and the scope of domestically enforceable legal rights to be well understood. The consequence of this arrangement is that executive-instigated changes on the international level (e.g. by means of terminating relevant treaty arrangements) can be effected by using the prerogative. This means that the relevant statutory provisions give legal effect only ‘contingently’ to such rights as relevant treaty arrangements provide for at any given point in time. Finnis reasons that since — on his argument — it is uncontroversial that such treaty arrangements remain vulnerable to the exercise of prerogative power, the same must be true of the treaty arrangements to which the ECA 1972 gives effect in domestic law.
The position adopted by the Court in Miller suggests one of two things in respect of the argument concerning double tax treaties. One possibility is that Finnis’s view that such arrangements are vulnerable to the exercise of prerogative authority on the international plane no longer holds. Another possibility, however, is that the Court might consider the EU context to be unique, such that the kind of analysis applied by Finnis in the double tax context is considered inappropriate in the EU context. It might be thought that such a form of EU-exceptionalism is warranted by its sui generis nature, including the direct effect doctrine’s capacity to blur — or to appear to blur — the lines of demarcation between international and domestic law. On this view, the UK’s membership of the EU might be considered to have resulted in some form of legal alchemy, whereby the distinction between treaty-based rights and the UK legislation that operates as the vehicle for giving effect to those rights has somehow been dissolved. If this perspective was taken, and the Court did in fact see the impact of EU law as sui generis, then it would provide grounds for challenging the relevance to the EU context of Finnis’s argument founded upon the operation of double tax treaties. However, it is perfectly clear that no such legal alchemy is occurred.
In fact, English courts, including the Supreme Court, have denied that EU law has acquired a sui generis status as a matter of constitutional theory. For example, in Thoburn v Sunderland City Council  EWHC 195 (Admin),  QB 151, such an argument about the nature of EU law was firmly rejected. Counsel argued that (as Laws LJ paraphrased) ‘the EC Treaty’s effect in domestic law does not depend, merely at least, upon the terms of its incorporation by the ECA, but, in part at least (and to a decisive extent), upon principles of EU law itself’. Laws LJ was unconvinced. In rejecting this submission, Laws LJ’s underlying reasoning treats the ECA as the conduit through which EU law has effect in the UK, thus signalling that EU law has no legal status or effect in the UK separate from that provided for by domestic legislation. In the Supreme Court, Lord Reed took a similar view in R (HS2 Action Alliance Ltd) v Secretary of State for Transport  UKSC 3,  1 WLR 324, when he said that questions about the relative priority of EU and domestic law ‘cannot be resolved simply by applying the doctrine developed by the Court of Justice of the supremacy of EU law, since the application of that doctrine in our law itself depends upon the 1972 Act’. Such dicta are consistent with Parliament’s own understanding of the juristic basis upon which EU law has effect in the UK. Section 18 of the European Union Act 2011 provides that:
Directly applicable or directly effective EU law (that is, the rights, powers, liabilities, obligations, restrictions, remedies and procedures referred to in section 2(1) of the European Communities Act 1972) falls to be recognised and available in law in the United Kingdom only by virtue of that Act or where it is required to be recognised and available in law by virtue of any other Act.
This makes it entirely clear that, as a matter of constitutional theory, EU law has not acquired a sui generis status. It follows that no argument based upon its having such a status can be advanced as a means of denying the pertinence of Finnis’s argument in the EU context. And from this it follows that the Court in Miller goes wrong by evidencing insufficient sensitivity to the nature of the rights in question. Those rights, as treaty-based, EU law rights, are therefore not the form of rights to which the limitations upon the prerogative apply.
‘Constitutional statutes’ and ‘background constitutional principles’: General matters
If domestic rights are in play which could be destabilised by the triggering of Article 50 (which is contrary to what we argue) then the focus of the analysis shifts to the question whether the constitutional principles concerning the extent of the prerogative and its relationship to statute preclude the triggering of Article 50 via prerogative power. The answer to that question must be ‘no’ if the ECA is interpreted in such a way that it leaves the prerogative power to conduct international relations or foreign affairs relevantly intact. The crucial question in Miller then becomes one of the statutory construction of the ECA. The Court’s analysis of this matter is fundamentally shaped by its perception of two key aspects of the constitutional backdrop against which the ECA exists. The first aspect concerns the ECA’s status as a ‘constitutional statute’ — a term coined by Laws LJ in Thoburn.
The second concerns the operation of what the Court calls ‘background constitutional principles’. The Court rightly notes that such principles influence statutory interpretation: ‘Where background constitutional principles are strong, there is a presumption that Parliament intended to legislate in conformity with them and not to undermine them. One reads the text of the statute in the light of constitutional principle.’ This dictum raises questions as to what the relevant constitutional principles are, and whether, and if so how, they operate to determine whether the ECA is to be read as leaving the royal Pprerogative intact in the present context. There are, of course, many ‘background’ — or ‘fundamental’ — constitutional principles. Among them must be the sovereignty of Parliament, the rule of law, and the separation of powers. Indeed, many — perhaps most — background or fundamental constitutional principles consist of some version of those three foundational principles, taken either in isolation or in combination. That is certainly true of the principles that govern the way in which statute and prerogative interact with one another.
One of the ‘background’, or ‘fundamental’, constitutional principles that shapes the interpretation of legislation is the ‘principle of legality’ — as articulated in decisions such as Simms, Pierson and Witham. These cases create a common law presumption of statutory interpretation which amounts to the fact that ‘fundamental rights cannot be overridden by general or ambiguous words’ (per Lord Hoffmann in R v Secretary of State for the Home Department, ex parte Simms  2 AC 115). In other words, if Parliament wishes to legislate to remove fundamental (common law) rights, it must do so explicitly. The Court’s appeal to this principle in the context of the interpretation of the ECA in this context, however, is misguided. As it focuses upon common law constitutional rights, it cannot assist the Court in interpreting the ECA, because, in the Court’s eyes, the EU law rights at stake in the Article 50 process are statutory rights. These are not the common law rights that the principle of legality as articulated in Simms and Pierson are generally understood to protect. That is not to deny that the content of the body of common law rights — and so the field of application of the principle of legality — can and may be influenced by international law, including EU law and the ECHR. For example, as Lord Reed put it in Osborn v Parole Board  UKSC 61,  1 AC 1115, the common law of human rights falls to be developed ‘in accordance with’ Convention rights when appropriate. Be that as it may, it is upon common law constitutional rights that the principle of legality centrally bites.
Conceived of in broader terms, however, the principle of legality can be understood as something that does not simply protect common law rights, but which operates as a general background norm of statutory construction so as to preserve a range of fundamental constitutional values, whether or not those values specifically take the form of common law rights. If such an interpretive approach is relevant in the present context, its relevance must lie in that broader sense. The Court reasons that when the ECA is interpreted in the light of relevant background constitutional principles, the argument that it leaves undisturbed the relevant prerogative power becomes untenable. Indeed, the Court says that the Government’s case is ‘flawed’ at a ‘basic level’ by the fact that it ‘gave no value to the usual constitutional principle that, unless Parliament legislates to the contrary, the Crown should not have the power to vary the law of the land by the exercise of its prerogative powers’. The Court notes that this principle evolved through the ‘long struggle … to assert parliamentary sovereignty’. And the Court goes on to say that the status of the ECA as a constitutional statute indicates Parliament’s desire to exempt it from implied repeal. This, the Court reasons, makes it unlikely that Parliament would be content to allow its legal effects to be removed by the Crown via the prerogative. These aspects of the judgment seem to us to be not without difficulty. Three matters are particularly worth noting.
‘Constitutional statutes’ and ‘background constitutional principles’: Three particular matters
The first matter concerns the relationship between constitutional statutes and what the court calls background constitutional principles. The notion of constitutional statutes was developed by Laws LJ in Thoburn. He articulated a test for identifying those statutes which the common law properly holds to be ‘constitutional statutes’. A statute will be a constitutional statute — rendering it immune from implied repeal — if it (a) ‘conditions the legal relationship between citizen and State in some general, overarching manner’ or (b) ‘enlarges or diminishes the scope of what we would now regard as fundamental constitutional rights.’ For Laws LJ the ECA was ‘by force of the common law, a constitutional statute’. The consequence of categorising a statute as constitutional means that it is immune from the doctrine of implied repeal. In short, if Parliament wishes to repeal a constitutional statute it must do so explicitly. In Miller the Court suggests that the significance of statute being regarded as constitutional is that background constitutional principles are peculiarly relevant to their construction: ‘statutory interpretation, particularly of a constitutional statute…must proceed having regard to background constitutional principles’.
There are admittedly certainly examples — Robinson v Secretary of State for Northern Ireland  UKHL 32,  NI 390, for instance — of the constitutional dimension of legislation significantly affecting its interpretation. In Robinson the classification of the Northern Ireland Act 1998 (‘NIA’) as a constitutional statute led the court to interpret it in the light of the general (including factual), not just the legal, constitutional background in Northern Ireland. The House of Lords referred to the NIA as a ‘constitution for Northern Ireland, framed to create a continuing form of government against the background of the history of the territory and the principles agreed in [the] Belfast [Agreement].’ Thus the court took account of the need to prevent the collapse of the Northern Ireland Government, which would have directly threatened the peace process. Against that background, the unwillingness of the Court in Miller to take greater account of the broader constitutional context — of which the referendum formed an axiomatic component — is striking to say the least. We return to this issue below when we consider the relative relevance of popular sovereignty to parliamentary sovereignty. In Robinson, context included the factual background, but in Miller this was conspicuous by its absence. For now, we simply note the dissonance between the interpretive approaches respectively generated by the constitutional status ascribed to the NIA in Robinson and the ECA in Miller.
The second noteworthy matter in this context relates to the consequences that, according to the Court in Miller, flow from the ECA’s designation as a ‘constitutional statute’ in the Thoburn sense. As already noted, according to Thoburn the consequence that flows from a statute being ‘constitutional’ is immunity from the doctrine of implied repeal. The Court in Miller, however, ascribes far broader consequences to the ‘constitutional’ designation of the statute in question, including consequences that operate so as to preclude the use of the prerogative to trigger Article 50. This represents a highly novel reading of the notion of constitutional statutes (a notion that is itself relatively novel, and not without difficulty, in the first place).
In Thoburn, Laws LJ was at pains to emphasise that whether something is a constitutional statute is not a matter of parliamentary intention. Rather, it is a conclusion reached, and a status ascribed to legislation by, the common law. ‘The ECA is, by force of the common law, a constitutional statute.’ Thus it is not for Parliament to intend that a statute be regarded as constitutional; it is for judges applying the principles of the common law to decide for themselves. In Miller, however, the Court overlooks these common law roots of the notion of constitutional legislation. Indeed, it positively misreads Thoburn, asserting that ‘the status of the ECA 1972 as a constitutional statute is such that Parliament is taken to have made it exempt from the operation of the usual doctrine of implied repeal by enactment of later inconsistent legislation’.
But Laws LJ said nothing of the sort. On his analysis, at least, the ECA’s status as a constitutional statute signifies nothing of Parliament’s intention, and reflects only a common law determination as to the constitutional status of the legislation. Moreover, according to Thoburn, the doctrine of implied repeal itself ‘was always the common law’s own creature.’ Properly understood, therefore, the ECA’s constitutional status does not cast any light on the intention of Parliament as to the conditions that must be fulfilled before it can be repealed or amended by a future Parliament. But, more generally, the common law’s designation of a statute as ‘constitutional’ does not tell us anything whatever about legislative intention, because that designation is in the first place a matter of common law. For this reason, the ECA’s constitutional status cannot — contrary to what the Court in Miller appears to think — be leveraged so as to infer that Parliament intended to limit or displace the prerogative. Parliament might have had that intention; but that is a question to which the ECA’s constitutional status — at least as that concept has been understood hitherto — is irrelevant.
A third question raised by the Court’s ‘background constitutional principles’ relates to the content of those principles. As noted above, many of those principles consist of versions or combinations of the foundational principles of parliamentary sovereignty, the rule of law, and the separation of powers. The Court chooses to focus particularly, and perhaps understandably, on the sovereignty of Parliament, which forms the root of the further constitutional principle governing the relationship between statute and prerogative. However, it need hardly be pointed out that the content, meaning and interaction of the UK’s fundamental constitutional principles is contestable. For instance, when the principle governing the relationship between statute and prerogative is viewed through the prism of parliamentary sovereignty, it is perhaps lent an expansive character, the focus being upon the preservation of parliamentary power and the subjugation of executive authority wielded through the exercise of the prerogative. However, this represents an unsubtle and questionable reading of the significance of parliamentary sovereignty in this context. That is so in part because of the questions that arise in this context about the interaction of popular and parliamentary sovereignty — questions to which we turn below.
A further issue concerns the role to be played by other constitutional principles, including the separation of powers. The Court’s approach in Miller seems to turn the ordinary principles of statutory interpretation in relation to the prerogative on their head, and indeed the separation of powers on its head. It is well known that an Act of Parliament which is clearly in the same subject area as the prerogative displaces the executive’s power. In Fire Brigades Union the House of Lords held that it was unlawful for the Minister to use the royal prerogative to enact his preferred compensation scheme where an Act of Parliament already existed in this area and created a clear duty on the part of the Minister to exercise the power as Parliament intended. The House of Lords said ‘it would be most surprising if, at the present day, prerogative powers could be validly exercised by the executive so as to frustrate the will of Parliament expressed in a statute’. Quite right. This would indeed violate the constitutional principle of the separation of powers if the executive could dis-apply an Act of Parliament on a whim. But Fire Brigades says nothing of a situation where the will of Parliament in a particular statute is either (a) unclear, or (b) simply has not been expressed on a particular point. It remains unclear as to whether Parliament really did contemplate the process of leaving the EU when section 2(1) of the ECA 1972 was drafted. It is not immediately (let alone literally) clear whether the wording of section 2(1) which reads: ‘All such rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaties… shall be read as referring to one to which this subsection applies’ indicates any particular parliamentary intention relating to the process of EU withdrawal. Nonetheless, it seems that in Miller the Court has subtly stretched the Fire Brigades principle by selectively using ‘background constitutional principles’ to confect a clear parliamentary intention in respect of an unclear statutory provision, so as to rule that use the prerogative power was in fact prohibited by statute.
Sovereignty: Parliamentary or Popular?
Expressions of popular sovereignty in the British constitution are few and far between, but the referendum to leave the EU was arguably a clear manifestation of sovereignty in that form. How far popular sovereignty is a ‘background constitutional principle’ and what the relationship might be between that principle and parliamentary sovereignty is as yet unclear. However, the in the last few paragraphs of its judgment in Miller, the Court goes some way towards engaging with this question. The Court commented that the EU Referendum Act 2015 (which legally authorised the Brexit vote):
falls to be interpreted in light of the basic constitutional principles of parliamentary sovereignty and representative parliamentary democracy which apply in the United Kingdom, which lead to the conclusion that a referendum … can only be advisory for the lawmakers in Parliament.
The Court then continued by stating that the 2015 Act was passed against a clear background ‘including a clear briefing paper to parliamentarians explaining that the referendum would have advisory effect only.’ In fact, in the space of two paragraphs, the Court emphasises the referendum’s advisory status three times, although it conceded that this did not detract from ‘the status of the referendum as a political event’. The implication, it seems, is that the principle of parliamentary sovereignty can act as a trump on popular sovereignty. In real terms, this means that it would be constitutionally and legally permissible for Parliament to refuse to trigger Article 50, the result of the referendum notwithstanding.
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. On 3 November 1774 —242 years to the day before the High Court delivered its judgment in Miller — Burke cautioned his audience that parliamentarians did not abrogate their independent judgement in order to fulfil the wishes of their electors:
But authoritative instructions; mandates issued, which the member is bound blindly and implicitly to obey, to vote, and to argue for, though contrary to the clearest conviction of his judgment and conscience –these are things utterly unknown to the laws of this land, and which arise from a fundamental mistake of the whole order and tenor of our constitution.
Parliament is not a congress of ambassadors from different and hostile interests; which interests each must maintain, as an agent and advocate, against other agents and advocates; but parliament is a deliberative assembly of one nation, with one interest, that of the whole; where, not local purposes, not local prejudices, ought to guide, but the general good, resulting from the general reason of the whole. (Emphasis added).
For Burke, once parliamentarians were elected, it was their constitutional duty to reach an independent judgement on behalf of the common good. They were not bound by any given expression of popular will. This seems to be what the Court’s dictum also implies about the referendum result, and the relative weight the constitution accords to popular sovereignty on the one hand, and parliamentary sovereignty on the other. But right now we are sailing in uncharted seas. For that reason, it is not straightforwardly clear that the principle of parliamentary sovereignty — once it is weighed in the light of the relevant circumstances — exerts the degree of pull that the court assumes. The relative weight to be accorded to fundamental constitutional principles — including when they are used, as in Miller, to shape statutory interpretation — cannot be divorced from the context in which those principles fall to be applied, and it is at least arguable that the Court paid insufficient attention to the referendum itself as a key element of that context.
The Court’s determination to examine the constitutional appropriateness of executive action triggering Article 50 without any reference to the broader context lends the decision a highly artificial air. It also, as we noted earlier, sits in contrast to the approach adopted in Robinson, in which the underlying factual — as distinct from legal — constitutional context was the principal driver of the court’s interpretation of the Northern Ireland Act. It is one thing to argue that executive action which jeopardises domestic rights granted by Parliament (if that is indeed what the relevant rights are) is an affront to the doctrine of parliamentary sovereignty. But that proposition surely does not have equal purchase in respect of (on the one hand) executive action undertaken on a whim and (on the other hand) executive action that implements the will of the electorate expressed in a referendum that was itself legislated for by Parliament. The principle of parliamentary sovereignty might, on one view, be the primary principle of the UK constitution, but that does not mean that it should be understood as an abstraction whose implications fall to be understood without any reference to the surrounding constitutional context.
Against this background, the Court’s judgment is striking in its muscularity, there being surprisingly little evidence of sensitivity to the contestable nature and effects of the constitutional principles that — far from being in the ‘background’ — occupy centre-stage in the Court’s reasoning. The result is that the Court constructs the legislative intention underpinning the ECA in a very particular way that generates a very particular result. In the end, we are told — with absolute certainty — that Parliament in 1972 intended to displace the prerogative. But it is far from clear that that certainty is warranted.