Judicial review, ouster clauses, and the democratic credentials of the judiciary in the United Kingdom

Abstract

The aim of this paper is to challenge the argument that says, as judges are not elected, they have weaker or no democratic legitimacy when compared to legislators. This paper draws on dicta from Laws LJ, as he then was, in the Divisional Court case of Cart v Upper Tribunal, to offer two reasons why this is false. Call these the efficacy and equality principles of representation. The claim here is that without an independent judiciary, legislators cannot legislate or legislate in a way that applies equally. So, without an independent judiciary, the democratic legitimacy of a legislature is weakened or disappears. This argument makes a legal difference, but the kind of legal difference it makes varies between jurisdictions. This paper focuses on one difference the democratic legitimacy of judges makes in the UK: the extent to which Parliament can oust judicial review for error of jurisdiction.

Introduction

Do judges have democratic legitimacy? The answer usually given to this question is no, both in theory and in law. Judges’ legitimacy is thought to come instead from their independence from the usual processes of democracy and politics. ¹ It is elections that give institutions democratic legitimacy. ² This paper argues that elections are neither a necessary nor a sufficient condition for a legislature to have democratic legitimacy. It claims further that some of what is required for the democratic legitimacy of a legislature, in particular the efficacy of statutes and equality in the application of statutes, cannot exist absent an independent judiciary. This is not to say that the democratic legitimacy of the judiciary and the legislature are identical. It is right to say these institutions are composed of different kinds of representatives, and that the rationales of those representatives are not the same. ³ The argument is instead that these two processes of representation are inextricably linked and reliant on each other, and so (at least some of) their democratic legitimacy is shared too. Rather than seeing democratic legitimacy as singular, the argument in this paper is that there necessarily must be different types of legitimacy in a democracy. One type of democratic legitimacy is responsive, one is independent, and one is shared. The focus of this paper is the shared type of democratic legitimacy. These three types of democratic legitimacy may be visualised as follows in Figure 1:

Figure 1. Democratic legitimacy.

The shared democratic legitimacy of judges and legislators makes a difference to all kinds of legal questions. What is the appropriate scope of judicial review? ⁴ Should judges make ‘political’ ⁵ decisions? Can judicial review ever be ousted through legislation? ⁶ Further, the relationship between the judiciary and the legislature will vary across jurisdictions, and so the legal implications of the democratic legitimacy of judges will vary across jurisdictions too. This paper focuses on one legal difference the democratic legitimacy of judges makes specifically in the UK: the extent to which Parliament can oust judicial review for error of jurisdiction. ⁷

This paper has four sections. Section 1 outlines two slightly different versions of the standard view that judges lack democratic legitimacy because they are not elected, those offered by Waldron and Bellamy. These two accounts are the focus because they argue that political equality must logically precede legal equality, and the principles of efficacy and equality, offered here, respond directly to this claim. Section 2 of the paper explains why elections are neither a necessary nor a sufficient condition for democratic legitimacy. Section 3 identifies two conditions that are necessary, albeit not sufficient, for democratic legitimacy: the efficacy and equality principles of representation. Section 4 demonstrates the difference the shared democratic legitimacy of judges makes in law, particularly the limits on the ouster of judicial review in the UK. The aim of this paper is to say this: the claim that ‘judges are not elected therefore they have no democratic legitimacy’ is unhelpful, both as a matter of theory and also as a matter of law, in the UK.

Before turning to the substance of the paper, two quick caveats are necessary to clarify the contribution this paper is making. First, this is certainly not the first paper to argue that parliamentary sovereignty and the rule of law are interconnected constitutional principles in the UK. Hunt ⁸ and Young ⁹ and even Dicey ¹⁰ emphasise this point. This paper’s contribution is to develop the efficacy and equality principles of representation to provide further support for this way of thinking about all constitutions, particularly the constitution of the UK. This leads to the second qualification, this time on a point of law. The argument here is not that appellate courts should refuse to give effect to ouster clauses, irrespective of how they are worded. The aim of the paper is instead to push back against the proposition that judges cannot restrictively interpret ouster clauses because they lack the democratic legitimacy to do so. Further, the paper argues, what cannot be ousted either conceptually or practically is determining what the law requires, about what the law requires. This means, at a minimum, a court must be able to check that: (1) the ouster clause exists; (2) the body claiming its benefit is the body in the ouster; (3) more controversially, that the decision is within the public body’s jurisdiction; (4) there has not been extreme procedural unfairness which undermines equality before the law. ¹¹

1. The standard view of democratic legitimacy

The standard view of democratic legitimacy, both in law and theory, is that it is secured through elections. This standard view is not confined to the UK, of course, but the argument takes on a distinctive salience in a country where so much of the constitution is uncodified, without strong form judicial review. This section of the paper traces two persuasive versions of the argument that judges are not elected and therefore lack democratic credentials: those offered by Waldron and Bellamy. These views were chosen because they are influential, compelling, and connected. They were also chosen because they argue that political equality must logically precede legal equality. The equality and efficacy principles of representation, advocated here, respond directly to this claim. This paper argues that political and legal equality are inextricably linked. The political equality on which democracy relies reinforces legal equality. Legal equality, through the protection of rights of participation in election law and judicial review, sustains political equality as well.

To clarify what Waldron and Bellamy think: their view is, ultimately, that judges do not have ‘capital-D’ democratic credentials. They argue this because they are, entirely rightly, concerned about the view that courts are exclusively, or even ‘primarily’, ¹² the institutions that can save or protect democracy. Both Waldron and Bellamy have good reasons to say that voters, not judges, are the ultimate backstop in a democracy. Waldron and Bellamy are clearly correct to clarify that it is not the courts’ job alone to protect a constitution. This cannot be solely the courts’ job, either in the UK or elsewhere. This section will argue, however, that it is misleading to argue that ‘political equality precedes legal equality’ ¹³ and majority decision-making offers ‘value and respect’ to individuals ¹⁴ which is not possible through judicial review. Political and legal equality are inextricably interconnected. Political equality is at least partly established and sustained through legal equality, and political equality helps maintain legal equality as well. Rights to vote, of suffrage and participation, and to challenge the lawfulness of Government decisions through judicial review, are not self-sustaining in democracies by majoritarianism alone. These rights, which ensure political equality, cannot be tenable without the equal protection of law. Understanding judges as giving effect to the equality and efficacy principles of representation, rather than having no democratic legitimacy at all, better captures both the normative rationale for democracy, and some of the most compelling cases ruling on these questions as well.

(a) Waldron on the democratic credentials of the judiciary

At best, argues Waldron, judges have weaker democratic legitimacy than legislators. ¹⁵ Waldron thinks that representation is a culture as well as an event. He argues that courts are not representative because their processes are not oriented around removability. To make this point, Waldron says that ‘the judiciary is not an elective or representative institution … it is not permeated with an ethos of elections, representation, and electoral accountability the way that the legislature is’. ¹⁶ A consequence of this approach is that there is some space on Waldron’s view for judges to have democratic legitimacy if they are elected, but even then these will be minimal because of the kind of representatives they are. He says that while ‘Defenders of judicial review insist that judges do have democratic credentials … these credentials are not remotely competitive with the democratic credentials of elected legislators.’ ¹⁷ Here, Waldron is unimpressed by those who think judges have democratic credentials. He thinks they are trying to have it both ways. It cannot both be the case, he says, that judges have democratic credentials because they are independent, and that they have the same kinds of democratic legitimacy as directly responsive actors like legislators do. ¹⁸

Waldron’s scepticism about the democratic legitimacy of judges reflects his belief in the inherent equality, fairness, and opportunity for self-determination offered by majoritarianism. ¹⁹ Waldron rightly argues that the case for a strong judiciary in a democracy is often rooted in a worry about the ‘tyranny of the majority’. ²⁰ But on Waldron’s view there is nothing necessarily tyrannical about the majority, provided that it did not ‘exclude certain people from their participation as equals’. ²¹ For Waldron, while majoritarianism is imperfect, ²² it is the best way to give effect to the connected principles of equality and self-determination. ²³ Majoritarianism, through broad participation, gives legislation a status that judgments of courts cannot. Legislation demonstrates respect for persons. He says that: ‘According equal weight or equal potential decisiveness to individual votes is a way of respecting persons’ ²⁴ and that this respect for persons translates into respect for legislation itself. He claims ‘a piece of legislation deserves respect because of the achievement it represents in the circumstances of politics: action-in-concert in the face of disagreement’. ²⁵ In essence, Waldron argues that judges do not have democratic legitimacy because voters do not select judges in a way that affords everyone an equal opportunity to participate. It is this input legitimacy offered by majoritarianism that gives legislatures their democratic legitimacy. Whatever the merits of an independent judiciary, for Waldron its legitimacy is not grounded in democracy. ²⁶

(b) Bellamy on the democratic credentials of the judiciary

The core of Bellamy’s view is that judges do have democratic legitimacy. The weaker legitimacy judges have is derivative of the larger democratic culture of which judges are a part. He says that:

Courts supplement democracy in valuable ways – not least in ensuring the impartial and equitable application of democratically determined legislation. However, the necessary qualities they possess in doing so derive from the legal and judicial system being part of a democratic system characterized by political equality. ²⁷

On Bellamy’s view there is a lexical priority of the democratic legitimacy of legislatures and those of judges. He says again that: ‘[Judges’] representativeness stems from a variety of indirect democratic pressures’ ²⁸ and, with added emphasis this time, that ‘the democratic legitimacy of court arises from its being within an electoral political system rather than independent of it… embedded in and dependent on’ ²⁹ a democratic system.

Bellamy concludes his argument for the indirectness of judges’ democratic legitimacy with the claim that ‘no court can save democracy’. ³⁰ This point is at the heart of his agreement with Waldron, and it turns on the meaning of equality. Bellamy thinks that ‘political equality promotes equality before the law, not vice versa’. ³¹ Bellamy goes onto say:

Instead of viewing courts as defining and maintaining democratic qualities of the political system, it proves more appropriate to reverse this perspective and note how courts possess important and necessary democratic qualities precisely because they are embedded in and framed by a democratic political system … characterized by political equality. ³²

In this passage from Bellamy, there is the same resistance as Waldron to the supposed tyranny of the majority and, in the final words, that Bellamy’s resistance too comes from his view of political equality. Bellamy thinks it is politics, not law, that ultimately establishes equality. Because they think that equality is primarily a political proposition, Bellamy and Waldron are arguing against the proposition that unelected judges are the guardians of democracy. While Waldron and Bellamy are right that it is unhelpful to see judges alone as the guardians of democracy against a tyrannical legislature, it does not follow from their arguments that judges have no ‘capital-D’ democratic legitimacy at all. Rather than argue that judges have a weaker kind of legitimacy, which is disconnected from democratic legitimacy, this paper will argue instead that there are different kinds of democratic legitimacy, and one type of democratic legitimacy is inextricably shared between the legislature and the judiciary.

A reader may still be wondering, after reading this analysis, why it matters that judges have ‘capital-D’ democratic, as opposed to a weaker kind of legitimacy derived from the broader culture of democracy. If judges do not have capital-D democratic credentials, and political equality is lexically prior to legal equality, then the case for restrictive interpretation of ouster clauses is weaker. On what basis can courts then constructively limit the legislative intentions of Parliament? If judges have capital-D democratic legitimacy, however, the case for the restrictive interpretation of ouster clauses is stronger. With capital-D democratic legitimacy, judges cease to be secondary, passive interpreters of legislation from Parliament. With capital-D democratic credentials, an independent judiciary becomes part of what it means to legislate at all.

2. The necessary and sufficient conditions for democratic legitimacy

Elections alone are neither a necessary nor a sufficient condition for democratic legitimacy. Elections are the selection of representatives. The selection of representatives, as Gardner argues, is not a necessary condition for democracy because elections are not the only means of public political participation. ³³ Referendums and lottocracies can plainly be democratic too. Elections are not a sufficient condition for democracy because elections require an institutional system in order to be free and fair, as dictatorships constantly demonstrate. More fundamentally, though, there is a deeper difficulty with the picture of democratic legitimacy offered by Waldron and Bellamy. It concerns their approach to equality. The reason that elections alone cannot give rise to democratic legitimacy is that democracy cannot mean just majoritarianism. Without an independent judiciary, the conception of equality on which Waldron and Bellamy’s views are built collapses internally in on itself and so then too do their conceptions of majoritarianism and democracy in turn. Political equality cannot be guaranteed without independent election law, and the opportunity to challenge the decisions of a government through judicial review.

Majoritarianism is a powerful idea. The case for it is intuitive. In broad terms, Weale puts its rationale like this: ‘Democracy implies political equality and political equality implies majority rule, since majority rule is impartial with respect to individuals and alternatives’. ³⁴ It is not obviously true that majority rule is impartial with respect to individuals and to alternatives; the agenda-setting and shaping roles of individuals are not the same. ³⁵ Accept for sake of argument, though, that majoritarianism is impartial with respect to individuals and alternatives. Majoritarianism alone cannot sustain the conception of political equality on which is rests. Put more simply: the conditions of equality of participation on which majoritarianism relies cannot be sustained or defined by majoritarianism alone. Beitz rightly argues that the conception of equality on which majoritarianism is based is ‘implausibly narrow’. ³⁶ He says that the move from equal respect to majority decision-making ‘reflect[s] an implausibly narrow understanding of the more basic principle [ie equal respect], from which substantive concerns regarding the content of political outcomes have been excluded’. ³⁷ Responding to Beitz, Waldron bites this bullet. He accepts the idea that he is making a minimalist, formal case for equality, even one that is ‘implausibly narrow’. Waldron argues that under the current circumstances of politics, an implausibly narrow equality is the most that can be hoped for. He responds that:

Thus, in the circumstances of politics, all one can work with is the ‘implausibly narrow’ understanding of equal respect … majority decision is the only decision-procedure consistent with equal respect in this necessarily impoverished sense. ³⁸

In this passage, Waldron is saying that democracy must be a purely procedural ideal because it is impossible under the current circumstances of politics to agree to anything else. This cannot be right. Even the minimalist, formalist conception of equality Waldron employs rests on a substantive account of what equality entails; it must mean something like equality of participation. As Dworkin argues, there is a ‘gap’ ³⁹ between the concept of equality in the abstract and articulating it in a political community. ⁴⁰ Waldron is right to say that how this plays out is contested, and right as well that process of contestation is a necessary and ongoing part of democratic life. But the prior right to participate in that process of contestation, and what that right to participation entails, is a substantively demanding conception of equality. While it may not be likely, in theory it is conceptually possible that the majority will act to disenfranchise the minority. That possibility requires substantive agreement about what equality requires in terms of participation. It can both be true that Waldron is right that there are deep disagreements about equality in a democracy, and that there is some broad agreement on what it requires. But even if the paper is wrong about all of this, and Waldron is right that majoritarianism can get off the ground without any kind of substantive conception of equality, the next section will argue that the nature of rules is such that without an independent judiciary, the decisions of those elected by the majority cannot be applied in a way that maintains the basic equal respect for persons on which Waldron’s view relies.

Similarly, Bellamy argues that the democratic legitimacy of judges must follow derivatively from elected institutions because he thinks that ‘political equality promotes equality before the law, not vice versa’. ⁴¹ There is not only one direction of travel here. Political and legal equality are symbiotic. Progress in one advances progress in the other, and setbacks in one lead to setbacks in the other. ⁴² This is what the gap that Dworkin identifies is about: the navigation and constant evolution of the many relationships between political and legal equality in a democracy. Waldron and Bellamy are right to say that majoritarian concerns about legislatures are overstated. ⁴³ But even if legislatures never acted against minorities in practice, the argument made in the next section will still hold. It is impossible for legislatures to legislate, and legislate in a way that applies equally, without an independent judiciary. This is not because legislatures seek to oppress minorities, but because Parliament is not the kind of body that can apply its rules equally to all parties. This is all to say: majoritarianism may be necessary for democracy, although this paper identified doubts about that, ⁴⁴ but it cannot on its own be sufficient. ⁴⁵ Elections, including those held on a majoritarian basis, are neither necessary or sufficient for democracy, and so democratic legitimacy – and the conception of political equality on which majoritarianism rests – cannot be established by majoritarianism alone.

3. The efficacy and equality principles of representation

Some of what is required for a legislature to have democratic legitimacy it cannot establish on its own or through elections. It needs help from an independent judiciary. Call this support that is needed from the judiciary for democratic legitimacy the efficacy and equality principles of representation. In making the case for these principles, particularly the efficacy principle, the paper will rely heavily on the judgment of Laws LJ, as he then was, in the case of Cart v Upper Tribunal ⁴⁶ in the Divisional Court. While the dicta Laws LJ offers in Cart are a significant contribution to public law in the UK, it is worth stressing that the case drawn on here is the Divisional Court decision rather than the Supreme Court one. The Supreme Court eventually rejected parts of Laws’ reasoning, and some of his conclusions, and areas of disagreement between the Supreme Court and Divisional Court decisions are explained below. The first sub-section below explains what Laws LJ said in Cart; it then turns to why his statements demonstrate the equality and efficacy principles of representation.

(a) Cart v Upper Tribunal

Cart arose out of reforms to the system of administrative law in the UK through the Tribunals, Courts, and Enforcement Act 2007. The primary purpose of these reforms was to reduce pressure on the process of judicial review by reorganising the tribunal system in the UK, which is meant to be a more efficient alternative to judicial review. ⁴⁷ The question in Cart was whether decisions of the newly created apex of the tribunal system, the Upper Tribunal (UT), should be subject to judicial review. Essentially, the question at issue in Cart was: ‘does the rule of law require that decisions of the UT are subject to judicial review, or does the UT satisfy the requirements of the rule of law?’ ⁴⁸ In answering this question, Laws LJ made four connected points. He argued that:

1. (1) legislation is necessarily textual: laws require interpretation;

2. (2) without independent interpretation, there is the potential for arbitrariness;

3. (3) interpretation is necessary for the efficacy of laws;

4. (4) therefore, an independent judiciary is an ‘affirmation’ of parliamentary sovereignty – a necessary corollary of it.

The conclusions Laws LJ drew from these claims, the section argues, demonstrate two necessary principles of democratic representation. ⁴⁹ These two principles are:

1. P1 If my actions are not efficacious then I am not really standing for and acting for others. My democratic legitimacy is weakened. Call this the efficacy principle.

2. P2 If my decisions do not apply equally to all (similarly situated) legal subjects, then they cease to be rules. If they cease to be rules, my democratic legitimacy is weakened. Call this the equality principle.

Consider now why the efficacy and equality principles of representation follow from Laws LJ’s account.

(b) The efficacy and equality principles of representation

Lord Justice Laws, rightly, thought that if decisions of legislatures cannot be applied by an independent judiciary, then they are not efficacious. Interpretation is a means to deliver statute and policy to people. ⁵⁰ He began drawing the link between democratic representation and efficacy through the idea of the rule of law. While accepting that the rule of law is contested, Laws LJ did think some of its principles are fixed:

The sense of the rule of law with which we are concerned rests in this principle, that statute law has to be mediated by an authoritative judicial source, independent both of the legislature which made the statute, the executive government which procured its making, and the public body by which the statute is administered. ⁵¹

This principle of the rule of law, Laws LJ argued, follows logically from features of law and language:

The principle I have suggested has its genesis in the self-evident fact that legislation consists in texts. Often – and in every case of dispute or difficulty – the texts cannot speak for themselves. Unless their meaning is mediated to the public they are letters on a page. They have to be interpreted. The interpreter’s role cannot be filled by the legislature or the executive: for in that case either of them would be a judge in their own cause, with the ills of arbitrary government which that would entail. ⁵²

Laws’ first claim that ‘words on a page…can have no effect without an interpreter’ ⁵³ is right. Through interpretation statutes are ‘give[n] life’. ⁵⁴ Indeed, he could put the point even more strongly. Parliament is not the type of institution that can adjudicate the meaning of its decisions in the contexts of different sets of facts. It can, of course, revise and revisit legislation, and it does and should do this all the time, but that is not the same kind of act as applying the law to different sets of facts. The paper will return to the nature of application and the evils of arbitrary government below, but first observe how Laws LJ connected his first point about efficacy to his conclusion about the sovereignty of Parliament. He said:

The need for such an authoritative judicial source cannot be dispensed with by Parliament. This is not a denial of legislative sovereignty but an affirmation of it: as is the old rule that Parliament cannot bind itself. The old rule means that successive Parliaments are always free to make what laws they choose; that is the one condition of Parliament’s sovereignty. The requirement of an authoritative judicial source for the interpretation of law means that Parliament’s statutes are always effective; that is another. ⁵⁵

The genius of this argumentative move, drawing a connection between efficacy and parliamentary sovereignty, is the efficacy principle. The point of having Parliament is to make decisions. If those decisions cannot be applied, there is no point in having them. A legislature is incomplete without an independent judiciary. An independent judiciary is an affirmation, and a corollary of parliamentary sovereignty. As Laws says ‘The sovereignty of Parliament itself only has life by means of the common law’s methods’. ⁵⁶ If I am selected to represent people but I either cannot make decisions, or those decisions have no meaning or effect, then I cease to be democratically representative not in all ways, but in at least some significant ways. The discussion here is of democratic representation, but this principle holds for all kinds of representation. No matter how I am purporting to stand for and act for others, if my efforts and actions in doing so are not efficacious then my representative legitimacy is weakened. The democratic legitimacy of a legislature rests on its efficacy, and it cannot be efficacious without a system of independent interpretation.

It was crucial for Laws LJ too that this system of interpretation be independent. If it is not independent, it is just a matter of opinion:

If the meaning of a statutory text is not controlled by such a judicial authority, it would at length be degraded to nothing more than opinion. Its scope and content would become muddied and unclear. Public bodies would not … be kept within the confines of their powers prescribed by statute. The very effectiveness of statute law, Parliament’s law, requires that none of these things happen. ⁵⁷

To see why Laws is right about this, consider an example to illustrate how statutes can become muddled. ⁵⁸ Say that Parliament passes a law giving X category of persons a taxable benefit, but there is no independent system of interpretation to establish whether this applies to Xa and Xb. So, Parliament passes another law saying Xa is eligible for this taxable benefit but not Xb, and then another Act saying Xb is also eligible, but only on two limited sets of facts. Laws’ point is that then these all of these endless decisions by Parliament are not laws, but opinions. They are statements about the content of the law, but without an independent judiciary they are not articulated and applied in the form of rules. This is the equality principle of representation. Parliament is not the kind of institution that can articulate rules in a way that apply to all legal subjects equally, because they are constantly revising the content of the rule. Consider another example to illustrate the nature of the equality and the efficacy principles. Imagine I say to students on the first day of term: I will not accept late essays. Suppose I then make exceptions for students based on illness and technical difficulties, or if I make an exception for a student because she submitted her other essays on time, or has blue eyes. Here, I am moving the goalposts for the students at every stage. I do not have a rule about essays: I have opinions on a case-by-case basis about the acceptability of submitting essays late, depending on the circumstances of the student. Those statements are authoritative because of my position, but I do not have a rule. My statement about essays is not really a rule because, among other things: (a) it is not efficacious; and (b) it does not apply to all students equally. If, however, I leave it to my colleague or supervisor to determine what I meant by the statement ‘I will not accept late essays’ and he applies the same method of interpretation equally with respect to all students, my start of term statement then becomes a rule. Rules of all kinds, as Schauer says, apply to ‘types and not to particulars’. ⁵⁹ If rules apply to types, and not to particulars, then it must be the case that they apply equally to those who are of the same type.

Consider further how independence is connected to efficacy and equality. Laws says that if the content of the law becomes muddied or unclear it is then not efficacious either for Parliament or those to whom Parliament’s laws applies. Laws is worried that in the absence of a court, Parliament would be a ‘judge in its own cause’. I am not sure this is the best phrasing. Gardner makes the same point in a better way when he places the emphasis on democracy as rule by the people. ⁶⁰ Gardner says:

Democracy is rule by the people, not the implementation of the people’s will (whatever that means). If popular influence is exerted other than through a system in which authoritative general rules are complemented by authoritative independent adjudication of what counts as a breach of those rules, that isn’t rule by the people because it isn’t rule at all. ⁶¹

Gardner is not only right about this, he is right about the consequences he draws from this point too. He says that on his view there is only room for marginal conflict between parliamentary sovereignty and the rule of law. ⁶² This is because the point of Parliament is not to make statements of the content of law, it is to make rules that apply to (all similarly situated legal subjects) equally. This is not a democratic task that Parliament can perform on its own. Parliamentary sovereignty is defensible on the basis of its democratic legitimacy; if that democratic legitimacy is weakened, the case for parliamentary sovereignty is weakened too. Gardner and Laws are, of course, not the only theorists to make this point, Dicey makes it in a related way. Dicey says too that the interpretation of laws by an independent judiciary does not make Parliament weaker, but rather stronger and more authoritative. He says, ‘The sovereignty of Parliament … favours the supremacy the law, whilst the predominance of rigid legality throughout our institutions evokes the exercise, and thus increases the authority, of parliamentary sovereignty’. ⁶³ Consequently, Dicey says, parliamentary sovereignty and the supremacy of law are not in tension with each other – they are part of the same exercise. ⁶⁴

The link Laws draws between efficacy and parliamentary sovereignty is a powerful statement, not only about the nature of democratic representation in the UK, but of the conditions of democratic rule more broadly. Feldman rightly frames Laws’ idea as being that ‘aspects of the rule of law [are] part of parliamentary sovereignty in action’. ⁶⁵ Feldman’s framing articulates how closely on Laws’ view the principles of parliamentary sovereignty and the rule of law are connected in their exercise in the UK. While Laws is arguing about the UK in particular, two of his claims in Cart hold conceptually. First, legislation is necessarily textual: laws require interpretation. Secondly, interpretation is necessary for the efficacy of laws. Legislatures do not exist in a vacuum. Their decisions exist to be applied. Legislatures are not the kinds of bodies that can apply laws to sets of facts and so, on their own, they cannot articulate laws in the form of rules. Consequently, democratic legislatures cannot be representative – fulfil the purpose they were created to fulfil – without an independent judiciary to interpret them.

4. The legal significance of judges’ democratic legitimacy

The democratic legitimacy of judges makes a difference to all kinds of legal questions. How should parliamentary intention be interpreted? What degree of deference should be given to the executive? Should judges not answer policy questions? The focus here is on one legal difference that the judges’ democratic legitimacy makes in the UK. In particular, the democratic legitimacy of judges makes a difference to the question of whether, or to what extent, judicial review can be ousted.

Judicial review is the legal way that legal actions of public bodies are reviewed by the senior courts. An ouster clause is a provision of an Act of Parliament which purports to exclude a public body from being judicially reviewed. Ouster clauses are of constitutional significance because they often appear to present a clash between two constitutional principles: parliamentary sovereignty; and the rule of law. On the one hand, the law is what Parliament enacts; on the other, the rule of law requires both the policing of public bodies for compliance with the law, and that citizens should have option of challenging the legality of the decisions of public bodies. This section argues that the democratic legitimacy of Parliament constitutively requires that judicial review can never be ousted for error of jurisdiction. That, as Laws says, is an affirmation of Parliamentary sovereignty, not a denial of it. To see why all this matters, and why the democratic legitimacy of judges makes a difference for the distinction between error of jurisdiction and error of law, ⁶⁶ it is necessary to consider a case that draws in great detail on the judgment of Laws LJ in Cart, the case of Privacy International.

(a) Privacy International v the Investigatory Powers Tribunal

Privacy International ⁶⁷ is a case, like Cart, about the judicial review in the High Court of decisions by tribunals. ⁶⁸ Parliament passed the Regulation of Investigatory Powers Act in 2000 (RIPA 2000), which created the Investigatory Powers Tribunal (IPT) to oversee the actions of the intelligence services. RIPA 2000 includes an ouster clause, in section 67(8), which purports to exclude judicial review of the IPT by the High Court. Section 67(8) of RIPA 2000 states: ‘Determinations, awards, orders and other decisions of the Tribunal (including decisions as to whether they have jurisdiction) shall not be subject to appeal or be liable to be questioned in any court.’ The approach judges have taken to the interpretation of ouster clauses is to make an interpretative presumption that if Parliament is going to oust judicial review then it has to be absolutely beyond doubt in the legislation what it is doing. This clarity forces governments to bear the political consequences through the passage of the Act. ⁶⁹ Privacy International is a case about determining the scope of this ouster clause. Did Parliament make it absolutely beyond any doubt that it intended to oust judicial review in passing RIPA 2000? The judgment in Privacy International is significant too, in that in the leading judgment, Lord Carnwath goes further and asks if it is ever possible to oust judicial review of a tribunal of limited jurisdiction. He says the question is:

Whether, and, if so, in accordance with what principles, Parliament may by statute ‘oust’ the supervisory jurisdiction of the High Court to quash the decision of an inferior court or tribunal of limited statutory jurisdiction? ⁷⁰

Although in the end Lord Carnwath concludes it is not necessary to answer this question, and there is not a majority in the case for his tentative answer, he nevertheless concludes:

Whether for excess or abuse of jurisdiction, or error of law. In all cases, regardless of the words used, it should remain ultimately a matter for the court to determine the extent to which such a clause should be upheld, having regard to its purpose and statutory context, and the nature and importance of the legal issue in question; and to determine the level of scrutiny required by the rule of law. ⁷¹

Feldman argues that this conclusion by Lord Carnwath is vindication of Laws LJ’s judgment in Cart, that Laws thought that ‘it was for the courts, not Parliament, to decide when the rule of law required that a decision-maker was subject to judicial review’. ⁷² As Feldman rightly says, all four of the judgments in Privacy International draw on Laws LJ’s judgment in their arguments ⁷³ but Wilberg is right too that it is Lord Sumption (with whom Lord Reed agrees) who explicitly adopts Laws LJ’s reasoning. ⁷⁴ The approach of Laws LJ has not been adopted, however, in the interpretation of further ouster clauses, most notably with respect to section 11 of the Courts, Tribunals and Enforcement Act 2007. That section and those cases are considered in detail next.

(b) The Judicial Review and Courts Act 2022 and the IMA ouster clauses

This paper has argued that what cannot be ousted either conceptually or practically is determining what the law requires, about what the law requires. At a minimum, a court must be able to check that: (1) the ouster clause exists; (2) the body claiming its benefit is the body in the ouster; (3) the decision is within the public body’s jurisdiction; and (4) there is no procedural unfairness which undermines equality before the law. These kinds of limitations on ouster of judicial review are partially captured in the Judicial Review and Courts Act 2022 (JRCA 2022), a piece of legislation which responds directly to Cart. The JRCA 2022 amended the Tribunals, Courts and Enforcement Act 2007, making it much more difficult to challenge refusals of appeal by the UT.

There are, however, still some exceptions to ouster of judicial review. In the most exceptional cases, refusal of leave to appeal by the UT can still be challenged through High Court judicial review. The Tribunals, Courts and Enforcement Act 2007, section 11A(4)(a)–(c), as inserted by the JRCA 2022, lays out the exceptions to ouster, which require that ‘(a) the Upper Tribunal has or had a valid application before it under section 11(4)(b)’ and the UT ‘(b) is or was properly constituted for the purpose of dealing with this application’. Additionally, section 11A(4)(c) of the Tribunals, Courts and Enforcement Act 2007 does allow for High Court review in the most exceptional reasons such as bad faith and extreme procedural unfairness. This procedural fairness exception is clearly essential for democracy. Extreme procedural unfairness speaks to the ways, argued above, in which legal equality and political equality are interconnected. Without an equal opportunity to challenge decisions of the government, protected through procedural fairness, Parliament cannot be legislating in a way that applies equally and efficaciously. The section 11A(4)(a)–(b) exceptions, however, must be interpreted robustly to satisfy the efficacy and equality principles of representation. Particularly section 11A(4)(b), which requires that the UT is or was ‘properly constituted’, may usefully be interpreted in a demanding way to mean something closer to error of jurisdiction. The idea here is that if a tribunal is considering a question well outside its jurisdiction – the UT considering pure, substantive questions of tort or criminal law, for instance – that is unlawful. While it is correct to say that a Tribunal is not properly constituted in those circumstances, that is only part of the story. Bound up in the question of the constitution of the tribunal is the question of whether it has properly exercised its jurisdiction. What it means to be a properly constituted tribunal is to exercise its jurisdiction properly.

While the ouster clause introduced by the JRCA 2022 can be restrictively interpreted to be in line with the efficacy and equality principles of representation, this is not true of the ouster proposed in section 51 of the Illegal Migration Act (IMA) 2023. The IMA 2023 was passed in 2023, but the Regulations to bring the ouster into effect are not yet in force. Some of the exceptions to ouster in the IMA 2023 resemble those found in the amendments to the Tribunals, Courts and Enforcement Act 2007. There is a natural justice exception in the IMA 2023, and section 51(4)(c) of the 2023 Act has the same wording as section 11A(4)(c) of the Tribunals, Courts and Enforcement Act 2007. Much worse in the IMA 2023, however, is the ouster of review of decisions of the Secretary of State or immigration officer in section 13(3A)(4)(a)–(b) which say ‘(a) the powers of the immigration officer or the Secretary of State (as the case may be) are not to be regarded as having been exceeded by reason of any error made in reaching the decision’ and ‘(b) the supervisory jurisdiction does not extend to, and no application or petition for judicial review may be made or brought in relation to, the decision’. These and other proposed ouster clauses in the IMA 2023 cannot be reconciled with the efficacy and equality principles of representation put forward above. Section 51(3)(b) of the IMA 2023 holds that: ‘the supervisory jurisdiction does not extend to, and no application or petition for judicial review may be made or brought in relation to, the decision’. This provision, for the reasons put forward above, is both conceptually and legally incoherent. It is not possible for Parliament to give a tribunal of limited powers unlimited jurisdiction. This logic applies even more strongly when the decision-maker is not a tribunal, but an individual decision-maker such as the Secretary of State or an immigration officer.

There may be a way circumventing the incoherent approach in the IMA 2023. It could be that errors of jurisdiction must be addressed through the exceptional procedural fairness requirement. It could potentially be extremely procedurally unfair if a decision is made by a tribunal that does not have the proper jurisdiction to make that decision. So, it may be possible to bring in the kinds of questions introduced by the JRCA 2022, regarding whether the tribunal is properly constituted, through procedural fairness by the back door. This is a restrictive and demanding approach to the interpretation of ouster clauses, but necessary to satisfy the equality and efficacy principles of representation on which parliamentary sovereignty relies.

It is worth repeating, lest this argument be interpreted as anti-democratic, that the efficacy and equality principles of representation are affirmations of parliamentary sovereignty, not a denial of it. To see why this is the case, consider a thought experiment. Could Parliament pass an Act that states ‘this Act is not to be interpreted’? ⁷⁵ No, Gardner argues, because judges must be ‘committed to upholding an Act of Parliament as law even when it demands to not be upheld’. ⁷⁶ This is ‘the least that Parliamentary sovereignty requires’. ⁷⁷ Laws are necessarily textual, and their meaning is incomplete without interpretation and application by an independent judiciary. This meta-question cannot be ousted practically because, without this kind of interpretation, statements by Parliament are not rules or laws but just matters of opinion. They are not efficacious. Their democratic legitimacy is weakened or disappears. As Hart says, it is part of what is required for a legal system: that there are legal officials that consider themselves bound to interpret and apply laws from the internal point of view. ⁷⁸ Without interpretation by judges, there is no point of having parliamentary sovereignty at all, because those decisions are not efficacious, and so there is no parliamentary sovereignty in the first place.

(c) Oceana and LA (Albania)

The view challenged in this paper – that judges have capital-D democratic credentials – has been present in two recent further cases on the interpretation of ouster clauses: Oceana ⁷⁹ and LA(Albania). ⁸⁰ Both cases have taken the line challenged here: judges cannot restrictively interpret ouster clauses because they lack the democratic legitimacy to do so.

In Oceana, the Administrative Court dismissed a claim for judicial review of a refusal of permission to appeal to the UT. The Court held that the claim was impeded by section 11A of the Tribunals, Courts and Enforcement Act 2007 considered above. Saini J held – and it is necessary to quote at length to see his rationale – that:

The starting point is that the courts must always be the authoritative interpreters of all legislation including ouster clauses. That is a fundamental requirement of the rule of law and the courts jealously guard this role. However, the rule of law applies as much to courts as it does to anyone else. That means under our constitutional system, effect must be given to Parliament’s will expressed in legislation. ⁸¹

Parts of this passage are laudable, and consistent with the arguments offered here. It begins by affirming the indispensable role that courts play in interpreting legislation, including ouster clauses. Saini J then argues, however, that the rule of law ‘applies as much to courts as it does to everyone else’. This is both correct and misleading. The rule of law does not govern courts and Parliament separately; it governs their relationships and interactions together. Part of giving effect to Parliament’s intention is ensuring equality before the law, efficacy of the law, through the opportunity for judicial review. Restrictively interpreting ouster clauses, demanding clarity and equality before the law, is an affirmation of parliamentary sovereignty not a denial of it.

Similarly, LA (Albania), concerns section 11A of the Tribunals, Courts and Enforcement Act 2007. In this case, the Court of Appeal is similarly hands-off in its approach to interpreting ouster clauses, holding that ‘it is the duty of Courts to give effect to the clear words used by Parliament because no one, including a Court, is above the law’. ⁸² It would be difficult find a statement more antithetical to the argument offered in this paper. An argument for the restrictive interpretation of ouster clauses is not courts acting ‘above the law’; it is courts giving effect to the principles on which parliamentary sovereignty relies. Parliament cannot legislate, or legislate in a way that applies equally, without the independent interpretation of Acts. This interpretation is what turns Acts of Parliament into rule by the many. This transformation of Acts into rules cannot be accomplished by Parliament alone. The political equality on which majoritarianism rests is supported, and therefore sustained, by the equal protection of the law.

Conclusion

This paper challenged a familiar idea. It argued that it does not follow from the fact that judges are not elected that they have no democratic legitimacy. It argued instead that democracy, insofar as it is a form of rule, requires at least ⁸³ one other kind of representation in the form of independent interpretation. This independent interpretation is required in order for Acts of legislatures such as the Westminster Parliament to be efficacious and to apply equally. Without these efficacy and equality principles of representation, the democratic legitimacy of Parliament is weakened and so too is parliamentary sovereignty. This makes a difference in law because Parliament then cannot – either conceptually or practically – oust the prior question of whether laws or actions of public bodies are reviewable in the first place. It cannot oust judicial review for errors of jurisdiction of a tribunal of limited jurisdiction. This meta-interpretative requirement is a necessary precondition for a demos to rule, and so for Parliament to be sovereign at all.