Statutory interpretation and the administrative state: refocusing the purposivist/intentionalist debate

Abstract

Questions of statutory interpretation form a significant portion of administrative law cases. Accordingly, judicial methodology in this area requires careful consideration by public law commentators. The core aim of this paper is to question the general orthodoxy that statutory interpretation is invariably a question for judges. In recent times, one prominent argument has been that between jurists arguing that interpretation should be the realisation of Parliament's intention (‘intentionalist’), and those who prefer the closely related but subtly distinct focus on its objective purpose (‘purposivist’). I contend that this argument, framed as a question over the inter-institutional relationship between Parliament and the courts, focuses on how interpretation is carried out without considering who is best placed to do the interpreting. Given, as I demonstrate with a series of case studies, that in hard cases the distinction between interpretation and discretion/policymaking can become obscure to the point of nullity, the debate fails to consider institutional arguments for judicial deference on questions of law. Moreover, and counterintuitively, the arguments of the intentionalists/purposivists can be used to bolster an argument for deference. My conclusion adopts a wider lens, setting out a broader constitutional argument for, and addressing objections against, cautious and context-sensitive deference.

Introduction

Legal scholarship can suffer a disconnection between the issues receiving attention and those arising most frequently in practice.¹ While questions over the nature and application of judicial review doctrine or broader constitutional principle often preoccupy public law debate,² statute predominates in almost every area of public administration.³ Judicial interpretation of statute is thus an area in which the courts’ role in regulation of government has arguably its most significant impacts,⁴ but in which academic commentary has not paid the attention it might to the mechanics of judicial practice and its implications for administration. Certain issues have generated heat and light – the courts’ approach to section 3 of the Human Rights Act 1998, for example,⁵ and the application of the so-called ‘principle of legality’⁶ – but there has been limited consideration focusing on the application of general principles in a public law context.

The tide has begun to turn, with important recent studies highlighting that judicial review at the coalface is dominated by contextually focused questions of statutory interpretation.⁷ Bell and Fisher, in a recent study, note for example that much of judicial review practice involves negotiating ‘swathes’ of complex legislation.⁸ Yet the institutional implications of this are not afforded proper recognition in debates over interpretation's purpose and method. The core argument of this paper will be that the body best placed to construe a statute's meaning is not always and inevitably a court. Rather, there are circumstances in which a court should be willing to defer to an interpretation adopted by an administrator which the words of the statute can reasonably bear, provided the administrator possesses and has exercised relevant expertise in explicating statute's meaning. In order to make this argument, I engage with a prominent recent debate – notably involving members of the UK Supreme Court – turning on whether the courts’ role in interpreting statute is to determine Parliament's objective purpose or its actual intent.

My overall argument is at odds with constitutional orthodoxy, insofar as that orthodoxy holds the meaning of statute to be pre-eminently a matter of judicial determination. Yet, as Paul Daly argued persuasively over a decade ago, there are sound reasons to think that there is scope for judicial deference to administrative views on the meaning of statute, provided a cautious and institutionally sensitive approach is taken.⁹ In retreading Daly's footsteps here, in addition to critiquing the institutional limitations of current debates on interpretation, I consider a range of cases demonstrating the porous nature of the distinction between statutory interpretation and policymaking. From this perspective, I show that there are sound constitutional arguments – based on long-standing public law principles – to incorporate deference on questions of law.

The argument for judicial deference on questions of statutory interpretation comprises the following building blocks. Part 1 sets the scene by summarising the standard judicial approach to interpretation. Part 2 sets out the core arguments in debates between jurists arguing that interpretation's aim should be to realise Parliament's intention (‘intentionalist’), and those who prefer the closely related but very subtly distinct focus on its objective purpose (‘purposivist’). I show that while these competing attitudes will have a limited impact on outcomes, the arguments made by the conflicting sides in the debate expose the limitations of assuming that statutory interpretation is inevitably a task purely for judicial resolution. In particular, as I show in Part 3, in hard cases the distinction between interpretation and discretion (or policymaking) can become obscure to the point of nullity, raising questions over the institutional capacity of the courts to resolve such cases. Part 4 concludes by setting out a broader constitutional argument for, and addressing a range of objections against, judicial deference to administrative interpretation of law.

1. Interpretation in a nutshell¹⁰

This section sets out a generalised description (ie ignoring specialised regimes such as that under the Human Rights Act 1998) of judicial practice to establish a baseline for subsequent discussion. My intention is that this summary can be read neutrally, for which reason it intentionally suppresses questions around intention addressed in the next section. However, as we shall see when we turn to consider a range of examples from the case law, the core processes of ordinary, legitimate interpretative practice described here are susceptible to deployment – in hard cases – in a manner more akin to policymaking than interpretation. This insight – often asserted rather than demonstrated – will allow me to set out a constitutional argument for deference on questions of interpretation.

In the early twentieth century a literal approach to interpretation predominated,¹¹ in a judicial strategy of purported neutrality.¹² Text remains firmly at the heart of current practice, though in hard cases it tends to set the tramlines of permissible interpretation, rather than constituting a complete description of judicial method.¹³ The modern approach is the effectuation of Parliament's purpose or intention.¹⁴ This may be readily discoverable via textual analysis – and text remains preeminent – but may require broader investigation.

The hallmark of the current approach is to seek the true meaning of a statute in light of its context. The core idea is summarised by Lord Bingham in R (Quintavalle) v Secretary of State for Health:

The basic task of the court is to ascertain and give effect to the true meaning of what Parliament has said in the enactment to be construed. But that is not to say that attention should be confined and a literal interpretation given to the particular provisions which give rise to difficulty. … The court's task, within the permissible bounds of interpretation, is to give effect to Parliament's purpose.¹⁵

Identifying a provision's policy necessitates careful consideration of the context in which a statute was passed. The courts can look to a range of contextual materials to identify this, working outwards from the immediate textual context to a wider array of resources. The immediate textual context will ordinarily be most influential as constituting the words agreed by Parliament, and the Supreme Court has strongly emphasised this in recent cases.¹⁶ Where such text is unambiguous then this will ordinarily be determinative.¹⁷ However, a broader range of materials may be looked to in order to determine statute's meaning, and while such materials are secondary to text they are nonetheless frequently deployed to resolve hard cases. There is a live debate among judges about the legitimate width of the range of materials that may be looked to, but the following have all been deployed: the text and content of the statute in which a provision appears;¹⁸ the broader scheme of a statute;¹⁹ legislation on in pari materia topics,²⁰ background documentation such as Law Commission Reports,²¹ white or green papers;²² and (restrictively) statements in Hansard.²³

The immediate context of the statute and background policy discussions are not necessarily the limit of the courts’ approach. This method may also involve consideration of the statute's afterlife – the ways it has been judicially construed,²⁴ its practical application, semantic evolution, and broader societal change. In particular, a presumption sometimes applied is that legislation is always speaking; understood in its current context.²⁵

A further set of presumptions covers intrusion upon fundamental rights or important constitutional norms.²⁶ Parliament may do this expressly but very clear terms are required.²⁷ This ‘principle of legality’ is considered a tenet of the common law constitution, wherein the judicial role is not solely to effect democratic will, but to ensure the delivery of public policy within a framework of liberal constitutional values.²⁸ Examples of protected interests are: individual liberty;²⁹ property rights;³⁰ the presumption of mens rea in criminal offences;³¹ fairness;³² a right to notice of certain decisions;³³ rights to legal professional privilege;³⁴ and access to a court.³⁵

A core point requires emphasis given the arguments that follow. Orthodoxy remains that statutory meaning is pre-eminently for judicial determination.³⁶ This orthodox position does not mean that the courts ignore the functions of administrative bodies impacted by statutory interpretation.³⁷ Indeed, the process of determining Parliament's aims frequently involves consideration of the aims and functions of the bodies to which a statutory term applies.³⁸ The courts have, too, been willing to afford administrators leeway in the scope of an unclear provision's application, provided this meets a reasonableness standard.³⁹ The Supreme Court has also shown willingness in certain circumstances to defer to administrative tribunals.⁴⁰ In Cart the Court held, for example, that while there needs to be possibility of judicial review of legal issues decided by the Upper Tribunal, claimants would need to satisfy a restrictive test in order to do this.⁴¹ The Court took this a step further in Jones, where Lord Carnwath showed willingness to give weight to an expert tribunal's view on the meaning of the law. These examples represent an important evolution in the jurisprudence, and may yet develop further. But we need to keep in mind that this strand in the case law is an exception that proves the rule, given the increasing extent to which tribunals are both conceptually and functionally judicial.⁴²

In what follows, I will suggest that, where context permits, there is greater room for respect (indeed, deference) on statutory interpretation carried out by non-judicial administrative agencies. To do that, I now turn to a prominent recent debate regarding the proper approach to judicial determination of Parliament's intention, demonstrating that the competing perspectives in the debate can be deployed to critique the largely unquestioned assumption in the UK that courts are always best placed to determine questions of legal interpretation.

2. Parliament's purpose: subjective or objective?

(a) Introduction

Statutory interpretation interrelates with constitutional theory; its objectives and practice both sustain and develop understandings of the state.⁴³ In the UK the constitutional primacy of the legislature means that interpretative practice turns on achieving Parliament's aims, but this leaves open the question of how and by whom that meaning is to be established (and, moreover, the presumptions that are to made in the realisation of those aims).⁴⁴ While, as I have noted, the question of who interprets the law is largely settled, there is a live debate – notably now involving members of the UK Supreme Court – as to how Parliament's intention should be identified.

This turns on whether interpretation involves determining the objective purpose of statute (‘purposivist’), or whether the goal is to determine Parliament's intention (‘intentionalist’). The latter is more conservative, insofar as an attitudinal model focused on what Parliament specifically desired may suppress judicial creativity. Accordingly, the debate is not entirely academic – as demonstrated by the contrasting approaches of Lord Leggatt, on one hand, and Lady Arden and Lord Burrows on the other, in the recent Supreme Court case of Kostal.⁴⁵ Similarly, in Maughan Lady Arden made greater use of a consultation response published by the Government as relevant contextual material in determining the purpose of a statute than her colleagues were willing to allow.⁴⁶ Recent cases showing some judicial antipathy to the principle of legality make the same point.⁴⁷ Yet there remains broad agreement on how the interpretative task should be undertaken – and in a majority of cases the distance between intentionalist and purposivist approaches is minimal. In R v Luckhurst, for example, the Supreme Court confirmed that the aim is to identify the meaning of the words used, with context and purpose key factors in the process.⁴⁸ The arguments made in the debate between the two schools do, however, help expose institutional and functional limitations with the interpretative method. The competing perspectives allow us to deconstruct the underlying premise of the debate, namely that courts are always best place to interpret statute.

The upshot of this assumption is that the most prominent debate around statutory interpretation perpetuates a constraining dynamic wherein the practice is framed purely as a relationship between Parliament and the courts.⁴⁹ This focus inevitably occludes potentially valuable administrative expertise in the explication of statutory meaning. This section sets out the contours of the debate, critiquing both intentionalist and purposivist perspectives for omissions, respectively, to acknowledge the extent or the implications of the parallels (explored later in the paper) between interpretation and policymaking. Given that questions of policy or discretion are generally treated as matters for which administrators are best placed in terms of institutional competence to resolve, the debate thus distracts from the need to consider on a case-by-case basis whether a court is functionally best placed to determine Parliament's intention/purpose.⁵⁰ Paradoxically, the critiques made by each side in these debates demonstrate the flaws and assumptions underpinning the argument itself.

(b) Purposivism

The purposivist argument, predicated on seeking the objective purpose of legislation rather than trying to pinpoint what Parliament intended, is best explained via the series of objections its advocates make to an intentionalist approach.⁵¹ The first purposivist objection is the conceptual difficulty of attributing a single intention to a group.⁵² Individual members of Parliament will have different reasons for voting in a Bill's favour. Some will genuinely support a measure, others will vote for it to avoid harming the reputation of the government or to avoid the displeasure of party whips.⁵³ Others, of course, will have voted against the legislation. Hence, the argument goes, it is impossible to conceptualise legislation passed by a multi-member assembly as being underpinned by a single intention. To frame interpretation in terms of a search for that intention is misconceived.

The second objection is that legislation inevitably falls to be applied in situations and contexts that the legislature could not possibly have envisaged.⁵⁴ Legislation may be intrinsically vague, or contain aporias not identified at the time the Bill was passed. It is, accordingly, illogical to attribute intention to an entity that could not have foreseen every context in which the law it had passed would be applied.

The third objection, related but not identical to the second, is that presenting interpretation as primarily a question of determining legislative intention obscures the reality that interpretation – to greater or lesser extent depending on context – supplements and develops a statute's original text. The creativity of the interpretative process means that, as Lord Burrows puts it, intention becomes a conclusion for reasons based on other grounds.⁵⁵ Cass Sunstein has demonstrated here that the methodologies of contextual interpretation (practised by both purposivists and intentionalists) are flawed in terms of their ostensible objectivity. Structuralist approaches (ie determining intention from the way in which statute is laid out) assume a coherence that does not exist; and extrapolating purpose invariably involves judicial invention.⁵⁶ One might add, since purpose is derived from a range of contextual factors, that the potential for privileging particular sources over others, or combining those sources in novel ways, undermines the notion of any ‘true’ intention. Eskridge and Frickey thus conclude that intentionism fails to deal completely with the practical implications of ambiguity.⁵⁷ Vagueness and indeterminacy mean that intention in hard cases is potentially impossible to find.⁵⁸

It was on the substance of this third argument that Wade MacLauchlan critiqued intentionalist approaches for their excessive formalism (ie searching ‘through a reading of the language of the statute for the intent of the legislature’).⁵⁹ For MacLauchlan, taking this approach in cases involving administrative decision-making constituted an effective denial of the administrative state's existence.⁶⁰ It was in his view vital to acknowledge ‘the dynamic nature of the interpretative enterprise, the vital role of the interpreter, and the contingent status of the text’.⁶¹ MacLauchlan recommended as an alternative ‘a purposive interpretive process which takes account of field-related factors’.⁶²

(c) Intentionism

The intentionalist response to these criticisms is grounded in a constitutional argument made with typical power and clarity by Joseph Raz: without some notion of an identifiable intention the conferral of constitutional power on a deliberative legislature is futile.⁶³ For Lord Hodge, it is of paramount constitutional importance that legislators can be taken to understand the text of the measure that is placed before them for approval.⁶⁴ Lord Sales too prioritises an inference of intention because this best respects the purpose of legislative debate about questions of policy and law.⁶⁵

This is an ostensibly powerful argument for the intentionalists, but limited unless it can address purposivist scepticism of group intention. Richard Ekins has developed a sophisticated theoretical model of group intention for this purpose. Building on the work of Michael Bratman, he argues that it is natural to treat a group as possessing intention where it comprises rational agents acting on a shared plan. Parliament is such a group, acting under agreed procedures to achieve a rational plan to change the law.⁶⁶ These arguments pose a strong response to the problem of group intention; as John Gardner noted, while individual legislators may not intend to change law in a particular way, or even know what they are voting for, they do understand the process they are involved in.⁶⁷

Another argument for intention is the inculcation of judicial self-awareness, in the sense of maintaining a cognisance that constitutional propriety prioritises Parliamentary, not judicial, aims. Focusing on intention fosters due constitutional respect for Parliament, thereby helping restrict judicial invention. For Lord Sales, a refusal to recognise the importance of intention risks too much judicial law-making.⁶⁸ Lord Hodge notes that judges come at the bottom of a list of a statute's potential audiences, and should thus avoid overestimating the scope for adopting interpretations which stray too far from legislative text.⁶⁹ Ekins’ own commentary on decided cases likewise demonstrates concern about judicial creativity.⁷⁰ There are strong constitutional and normative considerations here, rightly militating against excessive judicial creativity. It would be demonstrably offensive to the rule of law here for judges to start adopting unwarranted or outrageous readings of statute.

How should statutory interpretation operate on an intentionalist view? Lord Sales holds that where statutory text is clear this settles the matter.⁷¹ In other cases, the right approach is to ‘proceed by reference to what a reasonable legislator would have wished to do, if he or she had notice of the problem; and to call on a wider range of aids to interpretation which offer clues to answering that question’.⁷² This may, but does not necessarily, involve consideration of the Parliamentary process of which the legislation in question was the end result.⁷³ For Lord Hodge, the court must look to the words of the relevant statute, ‘established assumptions and presumptions’, internal aids to construction and, with reticence and circumspection, external aids.⁷⁴ He allows the possibility of referring to proceedings in Parliament, but with significant hesitation.⁷⁵

(d) The debate's unarticulated premise

As noted above, given broad agreement over the fundamentals of interpretative practice, the purposivist/interpretivist distinction may have limited impact in most cases.⁷⁶ Attitude might influence the scope of judicial willingness to take into account contextual factors at a distant remove from a statute's text.⁷⁷ Lady Arden's willingness in Maugham, for example, to take into account material in a Government consultation response to interpret a statute which was not the subject of the consultation in question was criticised by some of her colleagues.⁷⁸ Yet the practical impacts of the argument are less important, in the context of my argument, than the potential for the competing perspectives to facilitate broader critique of the orthodoxy that statutory interpretation is always and entirely a matter for judicial determination. Both perspectives have developed important arguments and objections not fully answered by the other side. This is because the dynamics of the debate – focused on legislative intention at one end and an approach affording greater scope for judicial creativity at the other – preclude wider consideration of whether some other institution is better placed functionally to determine statute's meaning. As MacLauchlan's classic article reminds us, it is important not to lose sight of the question of who interprets when considering how interpretation is to be carried out. The current debate is predicated on arguments over the scope of judicial discretion,⁷⁹ but this negates the capacity for textual exegesis by agencies.⁸⁰ The following consideration of the competing positions will thus help build the case for cautious judicial deference.

The purposivist arguments regarding the difficulty – in hard cases – of identifying intention have not been satisfactorily answered. Textual vagueness, Parliamentary blindspots and inattention, and the limits of foresight mean there will be cases where interpretative canons cannot sensibly settle a case. The distinction between interpretation and policymaking becomes vanishingly thin, to the point of obsolescence, when uncertainty remains after textual and contextual analysis.⁸¹ In such cases a court is effectively left to determine the matter taking into account the general context, the overarching aims of the relevant legislation, and other relevant norms and principles. However sincerely a judge attempts to see the matter through the eyes of the legislature itself, it is unrealistic (as we shall see in the next section) to suggest that they can free themselves from (bounded) policymaking.⁸² Ultimately, to treat interpretation as entirely a search for intention is to conceal the reality that to interpret the law is to change the law.⁸³ The process of statutory interpretation involves ‘elaborating, supplementing, modifying and developing statutory meaning’.⁸⁴

It is critical here to recognise that interpretation and discretion exist on a continuum. Legal interpretation is the conclusive establishment of the meaning of propositions of law. Discretion, in the exercise of powers conferred by statute, means making decisions in a given context subject to relevant criteria.⁸⁵ Jerry Mashaw (writing in a US context) thus rightly argues that interpretation of a statute's purpose and scope, and the practical realisation of its ends – in unclear cases – approximates policymaking.⁸⁶ The point applies in a UK context. While the general aims of a statute may be evident to (and set limits upon) a government department making secondary legislation, or an administrator in an agency seeking to understand and apply a statutory provision, the determination of legislation's meaning involves the elucidation of policy insofar as the interpreting body will seek to realise the statute's meaning in the way it best sees fit. In cases of genuine statutory ambiguity a court can find itself in effectively the same position.⁸⁷ If and to the extent that this is the case, the question arises as to whether a court remains the best interpreter of the statute, or whether judges should be more ready to defer to administrators. This insight remains largely overlooked in judicial practice in the UK.

Intentionalist arguments are particularly hard to sustain in specific cases. Raz and Gardner are right – in general – that intention is the very point of legislative deliberation, but such arguments falter in the face of genuine ambiguity. Similarly, while Ekins’ use of group intention demonstrates the possibility of groups forming and acting on shared intention succeeds at a general level, it yields limited practical guidance for a judge faced with statutory ambiguity.⁸⁸ Ekins points out that the semantic range of legislation is often thinner than assumed,⁸⁹ but as we shall see when we turn to the case law, this argument is not decisive. In short, by fetishising intention out of respect to Parliament's constitutional status, intentionalist arguments can elide the real difficulties posed by statutory ambiguity.

The problem for the purposivist argument, on the other hand, is its failure to fully answer the intentionalist criticism that disavowing intention risks placing too much (or inappropriate) power into judicial hands. As Lord Sales points out, sidestepping intention leaves the difficulty that documents have neither intention nor agency.⁹⁰ The intentionalist concern here is both constitutional and institutional, in that it relates to fears over judicial legislation; but the same reasoning can be applied to judicial engagement in the process of bounded policymaking ordinarily entrusted to the executive branch. As MacLauchlan notes, ‘the “definitive” interpretation ought to be that of the interpreter who is best situated to assess the text, its tradition, and its contemporary context in a purposive fashion’.⁹¹ Thus, while MacLauchlan's key argument was a move away from textual formalism, a corollary of his argument is that the best person best situated is not necessarily a judge.

Purposivists that do not consider the scope for interpretation to be carried out other than by a court thus fail to fully address the implications of their own analysis. A purposivist may perhaps point to the generally accepted and understood processes of legal interpretation to legitimate judicial pre-eminence. The demands of the rule of law – including stability and predictability – dictate that even if statute's meaning is itself unclear we still understand the process by which this lack of clarity is to be resolved. There are, moreover, constitutional arguments for delegating interpretation to a neutral arbiter. These points will be addressed further below. Nonetheless, in cases where real uncertainty remains after the ordinary processes of textual and contextual analysis have been carried out, the question of whether judges are best placed institutionally to have the final word on the meaning of a statute is overlooked both within the framework of the intentionalist/positivist debate, and in judicial practice more generally.

This is an unfortunate oversight. Adrian Vermeule argues that identifying the ends of interpretation leaves open the question of the ways in which that end is realised.⁹² Thus, identifying that achievement of Parliament's policy objectives is the end goal of interpretation does not settle how this is best achieved. This can be addressed only by consideration of the institutional facilities and capabilities of the relevant actors.⁹³ The arguments of the purposivists and the intentionalist assume that courts are always best placed to do this. However, as we have seen, the content of the arguments on either side combine to undermine this largely unquestioned assumption. In particular, the debate helps expose and critique the ways in which administrative expertise can be excluded from orthodox approaches to interpretation. Intentionalist approaches are rightly wary of judicial policymaking; judges are often neither functionally nor constitutionally fit for this. But such approaches tend to underplay the extent – in certain contexts and circumstances – to which interpretation involves policymaking. Purposivists, on the other hand, are rightly critical of the more formalist attitude of the intentionalist school and readier to accept a more creative role for the judiciary. However, they fail to acknowledge that once the creativity of interpretation has been accepted, arguments around institutional competence need to be addressed. In particular, as I shall argue, this includes the argument I make here for (cautious) judicial deference to administrative interpretation of law. The next step in that argument requires consideration of the nature and extent of discretion involved in interpretative practice, because it is only if interpretation can approximate policymaking that an argument for judicial deference can be sustained. In the next section, via consideration of the case law, I will demonstrate that in hard cases the lawmaking/policymaking distinction can break down.

3. Interpretation, policymaking and the harrowing of the administrative state

(a) Interpretation and policymaking: an unstable binary

Paul Daly has argued that for reasons of relative expertise, complexity, accountability, democratic legitimacy, and possible level of participation in the decision-making process, administrators can be relatively better placed than courts to interpret statute.⁹⁴ If and to the extent that the processes of interpretation and discretion elide – provided that an administrative interpretation of a statute falls with the range of reasonable interpretations – there are thus sound institutional arguments for deference to administrative views on Parliament's purpose. In this light, current arguments around appropriate interpretative method can lead us to overlook the question of who is best placed to reify Parliament's policy aims. This oversight reflects a core assumption of interpretative practice in the UK which I set out to challenge in this paper. However, the basic claim that statutory interpretation resembles policymaking is at times asserted rather than demonstrated. This is a necessary bedrock for my argument.

Four examples demonstrate judicial deployment of accepted norms of interpretation encroaching on ground potentially better understood and navigated by administrators: (i) contextual source manipulation; (ii) competing purposes; (iii) practical consequences; and (iv) differential diagnosis.⁹⁵ The examples that follow – illustrative UK Supreme Court cases decided in the last decade – are important because of the nature of judicial reasoning and (in some cases) disagreement. In each case, the line between discretion and interpretation becomes – to greater or lesser extent – indistinct. I will also set out – where appropriate – examples where judicial deference to administrative perspectives could have been helpful.

(i) Contextual source manipulation

Selecting and deploying contextual material is, as set out in Part 1, a normal part of legitimate interpretative practice. In hard cases, the range of discretion afforded by this process can be substantial – supporting Stanley Fish's argument that guidelines for determining the meaning of texts themselves require interpretation.⁹⁶ In R (N) v Lewisham London Borough Council, for example, the Supreme Court considered whether the Protection from Eviction Act 1977 (the 1977 Act) provided protection for homeless persons placed in interim accommodation. If the 1977 Act applied, the authority would need a court order to obtain possession. Lord Hodge found for Lewisham, on the basis of Parliament's intention, having considered a range of contextual factors: the 1977 Act's predecessor legislation;⁹⁷ immediate statutory context;⁹⁸ the potential need for claimants to be moved;⁹⁹ relevant case law;¹⁰⁰ and the need to ensure that authorities could fulfil their legal duties.¹⁰¹ He also referred to the established practice of local authorities in this context, which had been impliedly ‘endorsed’ by Parliament, though he held that this might be relevant only if the statute itself is unclear (which he did not consider to be the case).¹⁰²

In dissent, Lord Neuberger critiqued the use by the majority of the 1977 Act's predecessor statutes, and cases in which the courts had considered those statutes.¹⁰³ His alternative focus was the wording of the statutory provision in question taking account of the wider statutory context.¹⁰⁴ He also took into account case law on statutes in pari materia. ¹⁰⁵ This alternative contextual focus led Lord Neuberger to adopt a different understanding of the 1977 Act's purposes and, in turn, a wider meaning of ‘dwelling’.¹⁰⁶ He was highly critical of the idea that ‘implied’ legislation (ie where Parliament is taken to impliedly endorse established local authority practice).¹⁰⁷ Notably, given my overall argument here, Lord Neuberger accused the majority in terms of being swayed by policy concerns.¹⁰⁸

The reality, however, is that both majority and minority approaches, in a case of legislative ambiguity with significant implications for local authorities, constituted quasi-policymaking. Lord Neuberger's criticism of the majority is well made, since it is clear that their concerns around impacts on local authority resourcing influenced their views on intention. But in suppressing ‘tacit’ legislation and customary meaning, and focusing on the statute's supposed purpose, the dissenters’ approach also demonstrates that selective use of contextual material to identify statutory purpose can be used to prioritise a particular set of values. In their case, the value of enhancing protection for vulnerable individuals led them to characterise the 1977 Act's disputable purpose (in this instance) differently to the majority. A key aim of the 1977 Act, of course, is to provide protection for tenants from a range of abusive behaviours by landlords, including from eviction in certain circumstances. However, this does not mean that it is blind to the need to strike a balance between the legitimate interests of landlords (including public sector landlords) and those of tenants. To the extent that the Act strikes this balance, in the context of public sector housing management, on the question before the court its meaning was obscure. Crudely put, for the majority the needs of local authorities to readily obtain access to short-term accommodation weighed more heavily; for the dissenters the balance tipped in favour of the need to protect homeless people housed in short-term accommodation.

How then should the Court have approached this issue? Constitutional orthodoxy means that the Court would not defer to authority interpretation of statute, but Lord Carnwath's approach in this case is nonetheless instructive. Lord Carnwath, like Lord Hodge, emphasised the importance of text.¹⁰⁹ However, he took a slightly different approach to the relevance of established authority practice. While for Lord Hodge this could potentially be relevant but was not on the facts because the statute was clear, Lord Carnwath wanted the Court to confirm that settled practice is a relevant contextual factor as part of the normal process of interpretation.¹¹⁰ For reasons of stability, authorities should be able to rely on the legality of well-established practice carried out ‘for a significant period without serious problems or injustice’.¹¹¹ Such practice would, Lord Carnwath warned, be lawful only insofar as it did not go against the grain of the legislation,¹¹² but is it noteworthy that he was willing by implication to give weight to authority readings of statute.

(ii) Competing purposes

Cases involving identification of statutory purpose have become a key argument for those arguing against judicial overreach.¹¹³ But the difficulties of identifying purpose, readily and frequently admitted by the courts,¹¹⁴ demonstrate in practice Mashaw's point that interpretation can collapse into policymaking. In re Recovery of Medical Costs for Asbestos Diseases (Wales) Bill, for example, the Supreme Court considered whether the Government of Wales Act 2006 (the 2006 Act) provided authority for the National Assembly for Wales (as it was) to pass liability to tortfeasors who had caused asbestos-related diseases for funding treatment. The central question was whether, in light of the 2006 Act's purpose, this change related to the ‘organisation or funding of the National Health Service’. Lord Mance, giving judgment for the majority, who determined that this was outwith National Assembly competence, held that the question was the ‘natural meaning’ of the statute.¹¹⁵ This required careful examination of context, which here included the framework established by the 2006 Act and the statutory framework governing healthcare in the UK prior to the 2006 Act's enactment.¹¹⁶ Lord Thomas, in dissent with Lady Hale, agreed that the proper role of the courts was to determine the statute's ordinary meaning, with reference to the relevant context.¹¹⁷ In Lord Thomas's view that did not involve any consideration of healthcare legislation pre-dating the 2006 Act's enactment.¹¹⁸ These contexts examined by the majority/minority were both entirely reasonable. The key point in the present context is that, where determining purpose can be as contingent as this case demonstrates, it becomes clear that the question will turn a great deal on values and the balancing of competing priorities.

Now, the scope of the powers of a devolved legislature is a point of high constitutional importance and thus prima facie a matter for judicial determination. But highly contestable questions of purpose also emerge in hard cases with significant policy content less obviously suited to judicial determination. In such cases the interpreter best placed to find a solution that best serves Parliament's aims may not be a court. In R (Cornwall Council) v Secretary of State for Health, for example, the Supreme Court had to determine in which of three contender local authority areas a person (PH) was ‘ordinarily resident’ for purposes of the National Assistance Act 1948 (the 1948 Act), a decision of some importance for the allocation of financial resources.¹¹⁹ The three authorities, in accordance with the procedure in section 32(3) of the 1948 Act, referred the matter to the Secretary of State, who duly determined that Cornwall Council was responsible for PH. Writing for the majority, Lord Carnwath held that while the Secretary of State's decision was perfectly justifiable as a policy choice, it did not comport with the wording of the statute.¹²⁰ In order to unveil the ‘policy’ or purpose of the 1948 Act and thereby define ‘ordinary residence’, Lord Carnwath considered a range of factors: the content of the 1948 Act itself, the legislative background, Law Commission research, and relevant case law.¹²¹ The led him to conclude that Wiltshire Council, which had made most of the decisions about PH, was the authority for the area in which he was ordinarily resident.

Lord Wilson's dissent in this case is important here. Just as the majority had characterised the Secretary of State's decision as a policy choice, Lord Wilson in turn critiqued the approach of the majority as policymaking masquerading as interpretation. In his view, the majority's characterisation of the 1948 Act's purpose was swayed by legally irrelevant points suggesting that Wiltshire Council bore responsibility for PH. He is critical of this on the basis that courts are not legislators and therefore should not be indulging in decision-making predicated on what should happen rather than what legislation dictates.¹²² Taking a self-consciously more legalistic approach, he held that Parliament had adopted a phrase with a ‘well known’ meaning in the light of decided cases, and had determined not to derogate from that well known meaning during the legislative history of amendments to the 1948 Act.¹²³ The problem with Lord Wilson's more formal approach is that it relies on a series of unverifiable presumptions about Parliament's knowledge when agreeing to the relevant provisions in the 1948 Act. Ultimately the point is unclear – judges in the higher courts, all expert statutory interpreters, had determined that Cornwall Council (Mr Justice Beatson, as he was), Wiltshire Council (Lord Carnwath, Lady Hale, Lord Hughes and Lord Toulson), and South Gloucestershire Council (Lord Wilson, Lord Justice Elias, Lord Justice Lewison and Lord Justice Floyd) were responsible for PH.

How might the Court have better addressed this issue? The statutory background was intimidatingly complex, and the case turned on complex facts. Normal processes of statutory interpretation led to a range of defensible conclusions. The context is social policy, an area where the courts are ordinarily deferential to executive decision-making in light of the sensitive questions of resource management that arise.¹²⁴ The approach of Mr Justice Beatson (as he was) in the High Court is thus notable. He considered much of the same material as the justices in the Supreme Court.¹²⁵ A key point of difference was that he paid greater heed to the careful nature of the Secretary of State's decision-making process. While, he noted, the Secretary of State's conclusion was somewhat ‘artificial’ in terms of the wording of the 1948 Act, it was not an unreasonable interpretation.¹²⁶ In short, Justice Beatson's approach comes close – in circumstances of significant statutory ambiguity in an area where judges are generally respectful of executive discretion – to deference to a reasonable interpretation carefully adopted by an expert administrator.

(iii) Practical consequences

Statutory interpretation's interrelationship with administrative policymaking is further discernible in the use of practical consequences. Real world outcomes may well help identify Parliament's intention – an absurd practical result is unlikely to have been intended by reasonable legislators. However, as a matter of institutional competence and resourcing, predicting the outcomes of conflicting interpretations is not necessarily best undertaken by the courts.¹²⁷ As Lady Hale noted in Doogan v Greater Glasgow and Clyde Health Board, for example, it was inappropriate for the Court to predict the outcomes of wide or narrow readings of provisions relating to abortion and conscientious objection.¹²⁸

In practice, however, avoiding absurd outcomes shades into the pre-eminently administrative role of predicting the extent to which different approaches contribute to the achievement of a given purpose. In HM Inspector of Health and Safety v Chevron North Sea Ltd, for example, in which the Supreme Court considered the permissibility of employment tribunals taking account of material unavailable to inspectors from the Health and Safety Executive in health and safety appeals, Lady Black rightly took account of the serious impracticalities of preventing this.¹²⁹ In other cases, however, questions of practical consequences appear more tightly bound with policy issues. In R (Eastenders Cash & Carry Plc) v Revenue and Customs Commissioners, for example, the question was whether the Customs and Excise Management Act 1979 allowed for temporary detention of goods pending an investigation by customs officers. Lord Sumption and Lord Reed looked to consequentialist arguments in holding that officers did possess such powers, in circumstances where there was room for reasonable debate on whether this was a part of Parliament's purpose.¹³⁰ In R (N) v Lewisham London Borough Council, a social housing case concerning the extent of the statutory protection given to homeless persons in temporary accommodation in which the court strongly split on Parliament's aims, Lord Hodge took account of the practical consequences which differing procedural requirements would have for local authorities.¹³¹

The point here is not to argue about the merits of these decisions (in each case the practical matters considered by the Court in fact militated in favour of the relevant public body). Rather, these cases demonstrate the courts’ use of practical outcomes to evaluate competing readings. The predictability and, indeed, desirability of particular outcomes is often – context depending – inextricably bound up with questions of social or economic aim, classically the province of administrators. If the nature and extent of the impacts of different reasonable readings of a statute are going to be decisive of its meaning, then an administrative perspective arrived at via the deployment of expert faculties to the question may well be better than that of a court.

(iv) Differential diagnosis

Differential diagnosis is a related but distinct way of using practical outcomes. It is an approach which clearly recognises that statute can bear multiple reasonable readings, and proceeds by systematically testing each of those against relevant contextual sources and practical outcomes. An example of this approach is found in MS (Uganda) v Secretary of State for the Home Department.¹³² The case concerned the scope of appeal rights conferred on asylum applicants by section 83 of the Nationality, Immigration and Asylum Act 2002. The section is notably unclear, and in order to resolve the issue Lord Hughes posited four possible interpretations. Some adhered closer to the statute's text than others, but all were reasonable and plausible.¹³³ The best reading was determined on the basis of its likely practical consequences (on which see above) and relative rationality.¹³⁴ Again, once the question turns on practical consequences (an empirical question) and substantive rationality this process begins to feel more akin to the exercise of a discretion.

It is certainly not the case that a differential approach inevitably takes a court into the realms of policymaking, and the approach can be used to tease out differences between modes of legal analysis. In Romein v Advocate General for Scotland, for example, Lord Sumption used this approach to evaluate the competing claims of textualist and purposive approaches.¹³⁵ However, in cases where significant ambiguity remains after the application of accepted interpretative tools and approaches, and where meaning turns to a significant extent on questions involving prediction and evaluation of substantive outcomes, it is at least questionable whether a judge is functionally best placed to have the final say.¹³⁶

(b) Undermining administration: judicial and administrative policymaking

The previous section elucidated the potentially porous boundary between discretion and interpretation; between policymaking and legal analysis. While these processes are of course distinct in aim and method, in hard cases they can become functionally very similar. This problematises the general acceptance that interpretative issues are best resolved by judges, substantively inexpert in the relevant policy area but highly adept at unravelling obscure legal texts. It also adds substance to my critique of the intentionalist/purposivist debates discussed in Part 2. In short, orthodox perspectives on statutory interpretation collapse what should be a tertiary relationship (Parliament, courts and administrators) into a binary one (Parliament and the courts).

One immediate challenge is to ask why this framing is problematic. After all, statutory ambiguity poses difficulties for regulated persons, and delegating its resolution to an independent third party with interpretative expertise is an efficient manner of removing uncertainty. Moreover, courts by no means ignore the implications of particular readings of statute for administrators. And there are sound constitutional reasons relating to protection from abusive executive power (see further below) in favour of courts having the final say on interpretive questions. It is certainly unsound to argue that merely because a particular interpretation of statute poses difficulties for administrators, it should be avoided. However, an unquestioned preference for judicial interpretation of law can lead us to overlook institutional arguments otherwise deemed to have salience and weight in administrative law doctrine. The academic literature on deference is vast, but the core point is that on policy questions of substance judges may need to defer to the reasoned decisions of legislators or administrators for epistemic reasons (ie the judge knows less than the decision-maker), for reasons based on relative expertise, or for constitutional reasons (eg Parliament may have instructed a decision-maker to carry out some particular role, and courts need to avoid usurping that).¹³⁷ If and to the extent that interpretation shades into policymaking, the reasoning underpinning deference starts to become more relevant. Thus, while the argument for cautious deference on questions of law may appear heretical as first sight, counterintuitively it would be consistent with wider constitutional norms.¹³⁸

A brief illustrative case study, examining the UK Supreme Court's approach in a series of social housing cases, demonstrates the incongruence here. This is an area of administration suffused with delicate policy issues involving human needs and broader resource implications, a point to which the appellate courts have given recognition in the development and deployment of doctrine. For example, the courts have shown consistent deference in the application of judicial review grounds in cases involving resource allocation.¹³⁹ In the application of the Human Rights Act 1998 the Supreme Court's steadfast limitation of the application of Article 6 ECHR to welfare questions recognises the risks of judicialising administrative issues.¹⁴⁰ Even in questions of interpretation, when interpreting non-statutory materials, such as housing officer decision letters, the courts take a generous approach in recognition of the practical difficulties and sensitivities at large.¹⁴¹ On questions of statutory interpretation, however, the Supreme Court has resolutely followed an orthodox line on the law/policy distinction.

In Nzolameso v Westminster City Council, for example, the Court determined that authorities’ duties under section 208(1) of the Housing Act 1996 (the 1996 Act) to provide certain homeless persons with ‘suitable’ accommodation within its district required by implication consideration of a range of factors neither expressly nor by necessary implication referred to by the statute.¹⁴² In Hotak v Southwark London Borough Council the Court considered the appropriate comparator for determining whether a homeless person is ‘vulnerable’ for purposes of section 189(1) of the 1996 Act.¹⁴³ Lord Neuberger departed from the leading (and long-standing) authority of R v Camden LBC, ex p Pereira,¹⁴⁴ holding that the relevant comparison is with an ordinary person if made homeless (ie rather than an ordinary homeless person).¹⁴⁵ In so doing he expressly held irrelevant the views and practice of decision-makers, since this would undermine statutory intention.¹⁴⁶ Finally, in Haile v Waltham Forest London Borough Council, the Court reconsidered when someone is to be considered ‘intentionally’ homeless under section 191(1) of the 1996 Act.¹⁴⁷ The council here had, correctly, followed the long-standing authority of Din (Taj) v Wandsworth LBC.¹⁴⁸ Yet Lord Reed reinterpreted Din's meaning to find for the claimant.¹⁴⁹ Lord Carnwath dissented in this case on the basis that authorities had planned their work in the basis of Din for two decades.¹⁵⁰ While his perspective is grounded in concerns about law's stability and predictability rather than institutional competence, it nonetheless highlights the problems of downplaying administrative perspectives in interpretation.

Ian Loveland has framed these cases as problematic instances of judicial legislation.¹⁵¹ But we may extend his useful analysis of a series of cases arising in a particular policy area beyond the uncertainty arising from judicial usurpation of the legislative function. The cases also help demonstrate the inconsistency of treating interpretative questions of law entirely for judicial resolution. The permeability of the law/policy boundary means that the reasons underpinning judicial deference should be considered when interpreting statute, yet the orthodoxy around judicial supremacy on questions of law means that they are ignored. In Hotak and Din, as noted above, the Court was willing to depart from long-standing authority with little concern for administrative practice. In Nzolameso the Court demonstrated willingness to use a question of interpretation to lay down criteria for policy delivery (it should, in fairness, be noted that the housing authority's decision-making in that case had been poor, but the case nonetheless demonstrates the current approach to questions of function in this context). A heightened sensitivity to the policy content of interpretative processes can engage the kinds of concern around institutional competence arising in the context of, for example, substantive review at common law.

4. From dialectics to trialectics: the consitutional argument for limited deference to administrative interpretation

(a) Introduction

Thus far, I have argued that prominent debates on the nature of interpretation overlook the extent of the permeability between interpretation and policymaking. I have suggested that there may be circumstances in which administrators are better placed than judges to interpret statute. I have implied that there is room, against orthodoxy, for (cautious) judicial deference on questions of law. In this concluding section I take a wider perspective, embedding my central contention here among a broader set of constitutional arguments in favour of deference to administrative interpretations of law arrived at via a robust reasoning process. I also address a series of potential objections which, while properly reminding us that constitutional evolution here must cautiously avoid undermining an important check on the UK's already strong executive, do not undermine the central argument.

The discussion here is usefully prefaced in terms of the tendency in British legal constitutional thought to conceptualise the state via what Matthew Lewans terms the ‘Diceyan dialectic’.¹⁵² On this reading, a Diceyan framing of constitutional authority in terms of legislative supremacy and the rule of law can lead to a blind-spot with regard to the administrative state. Both Dicey himself,¹⁵³ and the jurisprudence, moved beyond this dialectic, with the growth of the dynamic administration and the concomitant evolution of a discrete administrative law throughout the twentieth century.¹⁵⁴ But it succinctly encapsulates the constitutional limitations of an interpretative model which, while by no means institutionally blind, is nonetheless constrained by its inability to conceive of interpretation as anything other than a uniquely judicial role. This model, grounded in the twin constitutional principles of Parliamentary sovereignty and the rule of law policed by the ordinary courts, conceptualises the interpretative endeavour as pre-eminently judicial in nature. While, as noted above, the Supreme Court has shown a willingness to defer to expert tribunals (which would have been anathema for Dicey), in truth this may well be as a result of the heavy judicialisation of such bodies. Such a limited vision of the constitution was overly simplistic even when articulated by Dicey. While in other areas constitutional thought has incorporated due respect for expert decision-making carried out by administrators institutionally equipped to manage decision-making in a large, modern state, the constitutional theory implied by dominant approaches to interpretative questions (and assumed by purposivist/intentionalist debates) remains incomplete.

(b) The functional case for deference

An argument for cautious deference on questions of law – beyond the limits of current practice – has limited but weighty academic support. Jack Beatson, for example, has critiqued judicial manipulation of the law/fact distinction in order to modulate intensity of review.¹⁵⁵ Rebecca Williams has argued for a principled qualification of the width of judicial discretion heralded by the demise of the collateral fact doctrine.¹⁵⁶ Paul Daly, as mentioned already, argues that Parliament may intend to delegate questions of law to agencies.¹⁵⁷ Mark Aronson has argued for a variable error of law standard.¹⁵⁸ And Paul Craig has also suggested that there is no reason, in principle or practice, for assuming that statutory interpretation should not be delegated to agencies.¹⁵⁹ There are a range of additional arguments in favour of a limited deference doctrine.

(i) Parliament's intention

There is no sound reason to infer from vague or open-texted statutory terminology that Parliament intended ambiguity to be settled by courts. Depending on context, it might be presumed Parliament intended for fine detail to be left to administrative discretion. This is supported by the demonstration in Part 3 that interpretation can collapse into discretion; the very fact of irresolvable questions of interpretation suggests – in hard cases – that we may be looking at the conferral of a discretion rather than the laying down of a definitive rule. Light touch review on questions of substance is a long-standing principle of UK administrative law, on the basis that a discretion conferred by Parliament should be exercised by the persons tasked with its exercise.¹⁶⁰ Indeed, one fundamental principle of UK administrative law is that such persons must not subdelegate that discretion to some third party.¹⁶¹ For that reason, it is arguable that the bedrock of the UK constitution militates in favour of deference; in certain circumstances it may best reflect Parliament's purpose.¹⁶²

(ii) Expertise

Deference to agency interpretations of law is often grounded in the relative expertise of courts and administrators.¹⁶³ Once the policymaking aspects of interpretative process have been elucidated, reasons of function, epistemic advantage and expertise weigh in favour of deference to administrative interpretations of law.¹⁶⁴ Judicial reticence on questions of substance in the discharge of a statutory power has been primarily shaped in the UK by ultra vires doctrine rather than deference on the basis of expertise. Courts intervene to ensure that administrators act within the scope of the powers conferred by Parliament, but otherwise take a relatively light touch approach on the substantive exercise of a discretion.¹⁶⁵ Nonetheless, as the scope and intensity of review has developed, functional considerations have come to play an increasing role. Since incorporation of the Human Rights Act 1998, for example, a specific legal doctrine of deference predicated in part on institutional competence has arguably emerged.¹⁶⁶ From this perspective, it might be argued that cautious deference on questions of law would better reflect a more refined, context-sensitive model of judicial review. Aileen Kavanagh has recently argued for a ‘relational’ concept of the separation of powers, which moves away from a ‘pure’ theory of separation of powers (ie in which ‘the rule of law is parcelled out to the independent judiciary, and the democratic principle is allocated to the elected legislature’).¹⁶⁷ This idea assists here – there may be questions of law in which relevant administrative expertise should give a judge pause for thought before substituting judgment.

(iii) Political accountability

A corollary of my central point that interpretative questions may have significant policy content,¹⁶⁸ is that accountability is not necessarily best achieved via judicial scrutiny. One key insight of the political constitutionalist school is that outwith a hardcore bundle of rights and legal principles, accountability on questions of policy may best be achieved via the political process.¹⁶⁹ There are, naturally, limits on the capacity (in terms of both capability and time) of such processes, but this will depend on context.

(iv) Participation

The courts have shown increased willingness over time to allow interventions from interested groups to facilitate effective decision-making in public cases.¹⁷⁰ However, the tripartite structure of litigation is inevitably limited in its scope for determination of polycentric issues.¹⁷¹ Agency deliberation, applying relevant expertise and incorporating relevant consultation where necessary is, conversely, ideally suited to such questions. Interpretative questions in highly complex areas of administration may well give rise to questions best suited to the latter type of decision-making. Agency capture is a risk here,¹⁷² but since judicial deference does not preclude judicial oversight, this can be managed on a case-by-case basis.

(v) Transparency

While deference on questions of law is very limited in the UK, the courts can make use of a law/fact distinction to modulate the intensity of review.¹⁷³ Rather than relying on the notoriously complex law/fact distinction for this purpose, it would be better to deploy a specific doctrine of deference, thereby facilitating better scrutiny and understanding of judicial decision-making.

(c) Some objections and responses

Judicial deference on questions of law naturally gives rise to a range of objections. A general answer is that in public law context is all.¹⁷⁴ My recommendation is emphatically not for deference as a matter of course in all cases – indeed, the relevance and scope of any deference here will depend on context, including whether the agency in question has actively taken steps which enable a court to place faith in its perspective. Daly has set out a set of circumstances in which the courts should be more ready to defer, and I largely align with his analysis. It is more likely to be apt to give weight to administrative interpretations if: (1) statute relates to a subject matter on which deference would be the norm in the exercise of a discretion (eg social policy); (2) where the statute is open-textured or unclear; (3) where the authority interpreting the stature has relevant institutional competence; and (4) where the authority shows it has deployed its expertise. Even with these caveats, a series of objections arises.

(i) Parliament's intention

One constitutional objection here is that Parliament is supreme, and the role of the independent courts is to give effect to its intention without fear or favour. This point is sharpened in public law cases by the added imperative of restraining arbitrary or excessive executive power. Yet the interpretative process, as we saw in Part 2, is not a transmission belt between Parliamentary aims and implementation. Parliament's intention may be unclear, and once the ordinary tools of interpretative method fail then arguments about intention diminish, potentially to the point of nullity. Provided review is available to check that administrators are not acting at odds with Parliament's intention or otherwise abusing their power, the objection is of limited force.

(ii) Rule of law/separation of powers

A related but distinct bundle of concerns is that allowing the executive to determine the meaning of law could offend the demands of certainty, equality and justice demanded by the rule of law.¹⁷⁵ As Lord Simon put it in Black Clawson: ‘in statutory construction, the court is not solely concerned with what the citizens, through their parliamentary representatives, meant to say; it is also concerned with the reasonable expectation of those citizens who are affected by the statute…’.¹⁷⁶ To allow the executive to apply and interpret law gives rise to tyrannical potential that separation of powers theorists aim to design out of legal systems.¹⁷⁷ The UK's particular fusion of executive and legislative powers strengthens the need for independent control of government.¹⁷⁸ Recent critiques of skeleton bills and overuse of secondary legislation are relevant here,¹⁷⁹ since it might be feared that the government will draft vague legislation to enhance the scope of its powers.

The gravamen of this objection is that in a Parliamentary democracy lacking constitutional review, with significant executive control of the legislature, liberal constitutionalism demands judicial dominance of interpretation as a necessary check on unfettered or arbitrary power. This is a weighty criticism, but again not one that inevitably outweighs arguments for cautious deference. The weight of executive views on statutory meaning, where relevant, can be calibrated to take account of the demands of certainty and justice – in much the same way that rights-based review responds to the respective claims of protected rights, public interests, and institutional responsibility.¹⁸⁰ In cases involving the principle of legality, for example, deference is highly unlikely to be apt. However, while some of the normative content of separation of powers theory derives from its ability to protect rights and liberties, more sophisticated thinking in this area also focuses on ensuring the efficient allocation of responsibility to institutions best placed to deliver effective government.¹⁸¹ Where an interpretative question turns to a significant extent on substantive policy content, the executive might be best placed to answer it. As to vague legislation, concerns about the Parliament's servility in the legislative process have been strongly challenged by recent research.¹⁸²

(iii) Consistency of application

Proper concerns here about inconsistency, both geographically or across time can, again, be taken into account by the courts on a case-by-case basis. An interpretation adopted by an agency with sole or primary responsibility for a particular statutory scheme which is consistent across time (eg the Environment Agency with respect to the Environment Act 1995) may demand more respect than a case where multiple agencies take a range of approaches (eg local planning authorities).

(iv) Predictability of interpretative method

Interested actors (including Parliament itself) will have or may obtain an understanding of the courts’ approach to interpretation (and obtain legal advice based on established judicial methodology). A further objection here is therefore that even where legislation is unclear, we may at least can understand the process by which this will be resolved (and seek legal advice if necessary). Again, the objection is of limited weight. I do not advocate departure from accepted interpretative doctrine – rather, the argument here is that in genuinely hard cases there may be ambiguity not sensibly or best resolved by judicial method. Indeed, in such cases a limited doctrine of deference may in fact aid certainty, since if legal advice unveils genuine ambiguity in a context when an administrative agency has expressed a view on the meaning of a statute, it might be presumed that the agency's view is likely to prevail (or at least be given significant weight).

Conclusion

Judicial supremacy on the interpretation of law remains standard. The courts reflect institutional issues as part of that practice. They will respect executive expertise in the application of law, provided this is reasonable. They have even been willing to defer to tribunal interpretations. Nonetheless, the core principle is that questions of law are the preserve of the ordinary courts. This process can, however, incorporate significant policy content – and as Fisher and Bell note, questions of statutory interpretation predominate in judicial review of executive action.¹⁸³ Accordingly, there are sound constitutional reasons for greater incorporation – where context and principle permit – of administrator/agency views on the interpretation of statute. While there are some strong objections – which have proved consistently compelling to date – none of these cannot be adequately addressed as part of the approach I have propounded. Thus, current debates around intentionalist and purposivist approaches to interpretation occlude questions of institutional capability and function, warranting a willingness to consider a wider range of legal interpreters. As Kavanagh puts it, sensitive constitutionalism requires input from all three branches of the state in an ‘institutionally-specific way’ – this may, I suggest, require a (vigilant) loosening of doctrinal fetters.¹⁸⁴