Jump to ContentJump to Main Navigation
Law, Liberty, and JusticeThe Legal Foundations of British Constitutionalism$

T. R. S. Allan

Print publication date: 1994

Print ISBN-13: 9780198259916

Published to Oxford Scholarship Online: March 2012

DOI: 10.1093/acprof:oso/9780198259916.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 25 February 2017

British Constitutionalism

British Constitutionalism

(p.1) 1 British Constitutionalism
Law, Liberty, and Justice

T. R. S. Allan

Oxford University Press

Abstract and Keywords

Dicey's dogmatic distinction between law and convention, for all its continuing influence on public lawyers, is only another manifestation of the positivist outlook responsible for the rule of parliamentary omnipotence. This book attempts to defend a different Dicey — the constitutional theorist struggling to escape the shackles of the Hobbesian authoritarianism he learned from Lord Atkin. It holds that Dicey's doctrine of law seems a better starting point than parliamentary sovereignty for the analysis in this book. It seeks to understand constitutional doctrine as a reflection of the underlying political ideal of the rule of law. In trying to explain the meaning of the rule of law as a constitutional principle, and explore its implications for British public law, this book mixes public law and legal and political theory.

Keywords:   Dicey, William Blackstone, common law constitution, F. A. Hayek, Ronald Dworkin, legal positivism

Political…or civil, liberty, which is that of a member of society, is no other than natural liberty so far restrained by human laws (and no farther) as is necessary and expedient for the general advantage of the publick. Hence we may collect that…every wanton and causeless restraint of the will of the subject, whether practiced by a monarch, a nobility, or a popular assembly, is a degree of tyranny…But then…that constitution or frame of government, that system of laws, is alone calculated to maintain civil liberty, which leaves the subject entire master of his own conduct, except in those points wherein the public good requires some direction or restraint.

Sir William Blackstone, Commentaries on the Laws of England, i, ch.1.

In accordance with British jurisprudence no member of the executive can interfere with the liberty or property of a British subject except on the condition that he can support the legality of his action before a court of justice. And it is the tradition of British justice that judges should not shrink from deciding such issues in the face of the executive.

Eshugbayi Eleko v. Government of Nigeria [1931] AC 662, 670 (Lord Atkin).

Contemporary public law is rife with strange notions. And many of these result from taking Blackstone and Dicey rather too literally, overlooking the core of good sense in their writings beneath some more questionable dogmas. Potentially the most serious of these dogmas—and the most awesome—is that of parliamentary sovereignty. Notwithstanding his reflections on the nature of tyranny, Blackstone followed Coke in asserting the ‘transcendent and absolute’ power of Parliament, relegating any idea of a breach of the legislative ‘trust’, reposed in the legislature by the people, to the realm of political theory, devoid of practical legal significance.1 A. V. Dicey, of course, followed suit and established the modern orthodoxy.2

As if the idea of unbounded legislative supremacy—unlimited not merely as a matter of practical politics but even as a matter of legal principle—were not bad enough, British constitutional lawyers have also generally repudiated the doctrine of the separation of powers. It (p.2) has never been made clear quite how their hostility to this (arguably) fundamental feature of constitutional government is consistent with the claim that the executive, if not the legislature, is subject to legal control.3 Perhaps, as has been recently suggested, some form of the separation of powers is only a matter of convention, rather than law. But then the confusion is surely compounded. Dicey's dogmatic distinction between law and convention, for all its continuing influence on public lawyers, is only another manifestation of the positivist outlook responsible for the rule of parliamentary omnipotence.4

The stark separation of legal rule from political principle, which the law-convention dichotomy reflects, is ultimately incoherent. It is certainly not compatible with a developing system of administrative law, which involves the courts in making judgments about the fairness and reasonableness of the actions of public authorities. What is fair and reasonable—or what is unfair and unreasonable—can hardly be determined in deliberate disregard of political practice, and the settled expectations which may have arisen on the basis of it.5 Nor can such judgments be made in abstraction from the wider constitutional landscape within which public agency and individual citizen interact. Legal rule, that is to say, cannot be isolated from political principle, even if it would suit positivist legal theory.

This book will attempt to defend a different Dicey—the constitutional theorist struggling to escape the shackles of the Hobbesian authoritarianism he learned from Austin.6 Dicey's doctrine of the rule of law, with its emphasis on equality and civil liberties, seems a better starting point than parliamentary sovereignty.7 We shall seek to understand the constitutional doctrine as a reflection of, and intimately associated with, the underlying political ideal of the rule of law. No doubt, the great complexity of that ideal, as well as continuing philosophical controversy about both its nature and importance, explain the relative neglect of Dicey's doctrine, by comparison with his other principles. The practical lawyer's distrust of theory and desire for clear-cut rules is understandable. If, then, Dicey's work laid the foundations of a theory of British constitutionalism, little work has yet been done to build a viable structure by developing his conception of the rule of law. It may (p.3) now be necessary, however, to cloud the purity of ‘legal science’ by rather more abstract speculation.8

In striving to explain the meaning of the rule of law as a constitutional principle, and to explore its implications for British public law, this book is inevitably something of a mixture of public law and legal and political theory. Since public law involves fundamental questions about the distribution of power and the relations between citizen and state, it is hard to see how else it could be written. It is certainly my view that there is no useful division between constitutional law and constitutional theory. That view explains the ubiquity of such magisterial figures as F. A. Hayek and Ronald Dworkin, amongst other writers more briefly noticed.9 There are, of course, many differences between the conceptions of liberalism such writers espouse, and none can be accepted uncritically. If a constitutional lawyer is not bound slavishly to follow a single theorist, however, or embrace one set of ideas to the exclusion of every other, he must nevertheless acknowledge the profound influence which Dicey's writings still exert on contemporary perceptions of the subject. Hayek and Dworkin each assist us, in different ways, to extract the core of good sense in Dicey's discussion, and to adapt his insights to the demands of modern constitutionalism.

Hayek's work is of value to the legal theorist because it clarifies so much of the philosophical basis of liberal constitutionalism. In particular, it provides an account of the separation of powers which, although only an ideal to which we might aspire, is none the less unrivalled in elegance and clarity. I shall invoke Hayek's thought in defence of my view of the central importance of the separation of powers to the political ideal of the rule of law. Dicey neglected the separation of powers largely, it seems, because he associated it with French droit administratif, and the exemption of officials from the jurisdiction of the ordinary courts which that implied. The resuscitation of his rule of law doctrine, however, requires us to make good that deficiency.

A system of government which depends on the close co-operation of executive and legislature, and even permits the former to dominate the latter, as a matter of practical politics, seems superficially to flout the separation of powers. It certainly violates any ‘pure’ model of that doctrine, as it might be conceived by a political scientist for the purpose of describing or classifying constitutional forms. Dicey, however, was (p.4) intent on expounding the constitution as a matter of law, expressly distinguishing his task from both history and political science.10 And as a matter of legal theory, his principle of the rule of law is hardly comprehensible without some appropriate conception of separation of powers. If the executive were not subordinate to the existing law in some sense in which the legislature is not, Dicey's account of the equality of all before the law, including public officials, could make no sense.

The common law constitution

In the absence of a higher ‘constitutional’ law, proclaimed in a written Constitution and venerated as a source of unique legal authority, the rule of law serves in Britain as a form of constitution. It is in this fundamental sense that Britain has a common law constitution: the ideas and values of which the rule of law consists are reflected and embedded in the ordinary common law. If important liberties are given protection, and standards of justice and fairness accepted and upheld, it is ultimately because—and largely to the extent that—they find expression in the common law. The great influence of Dicey's Law of the Constitution rested mainly on his firm grasp of this fundamental idea. He was careful to juxtapose the principle of parliamentary sovereignty with that of the rule of law, even if his attempt to reconcile the two has hardly met with universal acclaim. The rule of law meant essentially the rule of common law, supplemented by statutes, which ‘being passed to meet special grievances, bear a close resemblance to judicial decisions, and are in effect judgments pronounced by the High Court of Parliament’.11

The central role of the common law in the constitutional scheme placed unique responsibility on the shoulders of the judges of the ordinary courts. If in those Continental countries which possessed written constitutions, individual rights were ‘deductions drawn from the principles of the constitution’, in England by contrast the principles of the constitution were ‘inductions or generalisations based upon particular decisions pronounced by the courts as to the rights of given individuals’.12 Dicey avoided the mistake of supposing that individual liberties could not be as well (or perhaps better) protected with a constitutional declaration of rights: he noted that the rule of law was as marked a feature of the United States as of England. But he was right to emphasize the inherent strength of common law adjudication as a (p.5) basis of protection: a common law right is intrinsically related to the means of its enforcement.

Dicey's account of the plainly subordinate role played by statute in the constitutional order now looks strange. We are accustomed to the use of legislation to achieve major programmes of social reform, and to the modern practice of granting extensive powers to government ministers to implement their policies by issuing rules, orders, and directions. Indeed, Dicey's other generalizations about the rule of law have had to be qualified in consequence. We can no longer accept, without reservation, his assertion that the rule of law excluded ‘the existence of arbitrariness, of prerogative, or even of wide discretionary authority on the part of the government’.13 If arbitrariness is excluded, and prerogative controlled, it is only because English law has belatedly developed a body of ‘administrative law’, which seems superficially to contradict Dicey's emphasis on ‘legal equality’ between official and ordinary citizen—the ‘universal subjection of all classes to one law administered by the ordinary courts’.14

However, this familiar criticism can be overdone. Dicey's rejection of ‘administrative law’ has been misunderstood. What he opposed was the exemption of officials from the constraints of the ordinary law, which he took to be the chief characteristic of droit administratif. His claim that every official was amenable to the jurisdiction of the ordinary courts accurately asserted the basic constitutional position. The modern law of judicial review testifies to the truth of his observation that disputes involving the government were not in England beyond the jurisdiction of the ordinary civil courts: that (French) idea was ‘fundamentally inconsistent with our traditions and customs’.15

Dicey's insistence on equality before the law, and the universal role of the ordinary courts, was consistent with his view that the constitution was the result of the ordinary law of the land—of private law, as it applied to the activities of public officials as well as private citizens. The underlying conception of constitutionalism here is precisely the one Hayek's work exemplifies. The rule of law entails the subjection of government to the law—rule by law as opposed to politicians or public officials—in the sense of its being bound to comply with ‘rules of just conduct’, which regulate the ordinary relations between citizens, and which are independent of any particular purpose or policy, public or private.16 Individual liberty is secured, and political oppression prevented, by denying government the power to violate these universal rules of just conduct in its relations with the governed.

The rule of law, in Hayek's account, requires a thoroughgoing (p.6) separation of powers, which denies even the legislature the right to alter the law as it sees fit. A legislative assembly should exist to improve and supplement the existing corpus of rules of just conduct: it should not act, in the manner of a governmental body, to direct the performance of particular tasks or authorize, by changes in the general law, the pursuit of specific policies. Hayek distinguished between nomos—those rules of just conduct which have come to be accepted (by long usage) as the framework of social co-operation—and theseis, or those instructions which command the execution of particular tasks.17 He considered that the principle of separation of powers had been subverted by confusing these different senses of ‘law’. Instead of institutions charged with the function of enacting general rules of just conduct—changing the law, in the primary sense of that expression— modern legislatures had become merely representative assemblies chiefly occupied with directing or controlling government.18

The distinction between nomos and theseis, or between rules of just conduct and administrative ‘measures’, illuminates the distinction between law and policy which (I argue below19) is fundamental to an understanding of modern public law. In the sense of ‘administrative powers over persons and property’, administrative law entails the kinds of discriminations and discretion which the rule of law is supposed to inhibit.20 It is only to the extent that public authorities are subject to the ordinary private law, in their dealings with the citizen, that the rule of law, in Hayek's illuminating conception of that ideal, truly exists as a safeguard against the suppression of liberty. It does little harm to the public lawyer's pride—and probably much good for his humility—to be reminded that a constitution ‘is essentially a superstructure erected over a pre-existing system of law to organise the enforcement of that law’.21 It is possible to exaggerate the ‘particular dignity and fundamental character’.22 of constitutional law. As Hayek reminds us, and the doctrine of the primacy of European Community law perhaps affirms, even when the structure of government changes—or the location of ‘sovereignty’ alters—‘most of the rules of just conduct, the civil and criminal law, will remain in force’.23

Nevertheless, we must acknowledge the peculiar inadequacies of Dicey's theory of the rule of law to the circumstances of the modern administrative state; and here Hayek—whose principal concern, of course, was to limit that state in the interests of liberty—will offer us little assistance. It is therefore part of the present project to offer an account of the separation of powers which accommodates the exercise of government power beyond the ordinary realm of private law. For (p.7) this purpose, we will take up the distinction between law and policy, suggested by Hayek's distinction between nomos and theseis, and argue that it explains the legitimacy of judicial review. Modern administrative law has recognized the need for general principles of law which, while they apply only to the ‘public sphere’, serve to constrain the exercise of administrative powers in the interests of justice or fairness to those most directly affected.24

I shall argue that Ronald Dworkin's account of the distinction between principle and policy makes a helpful contribution to the task of defining the nature and limits of public law.25 Questions of principle are those which concern the scope and content of individual rights, as opposed to the general welfare or the public interest. Matters of public interest or public policy should be determined by the ‘political’ branches of government—executive or legislature. Questions of right, by contrast, are peculiarly the province of the courts. As counter-majoritarian entitlements, or ‘trumps’ over general utility or the public interest, the relative insulation of the judges from the ordinary political process ought to be especially conducive to their protection and enforcement.26 I shall suggest that administrative law may be helpfully interpreted as a system of public law rights; and that the legitimate boundaries of judicial review may be found in the process of defining and enforcing those rights.27 Issues of standing and justiciability may usefully be considered in the light of a rights-based approach to public law.

It will not be suggested, however, that Dworkin's distinction may be applied in any mechanical way, or that the division between principle and policy is always self-evident or uncontroversial.28 Although the doctrine of legitimate expectation plainly amounts to the recognition of rights against the state, in the manner Dworkin's analysis would appear to commend, it may involve the court in detailed review of the general merits of an application for relief—a review which can scarcely escape an appraisal of the reasons given in justification of the administrative action in question. The doctrine expresses a principle of justice or fairness, which is ultimately dependent (I argue below29) on all the circumstances of the particular case. That seems an embarrassing (p.8) conclusion, if it is correct, for the traditional theory of judicial review, which distinguishes sharply between the merits of a public agency's action or decision, on the one hand, and its legality, on the other.

Admittedly, the distinction between the merits of an action and its legality roughly reflects Dworkin's distinction between policy and principle—a correspondence which confirms the appropriateness of a Dworkinian analysis. While fundamental to the constitutional legitimacy of judicial review, however, the merits-legality dichotomy must be applied with sensitivity to particular cases. In particular, the traditional distinction between appeal and review—corresponding to that between the merits of an action and its legal validity—will in practice sometimes be only one of degree.30

If it were possible to identify and isolate a finite set of discrete public law rights, whose satisfaction ensured the legality of administrative action, the principle-policy dichotomy could perhaps be applied with greater precision. (Of course, controversy over the exact nature and scope of such rights would inevitably remain.) Reflection on the role of the courts as defenders of the ordinary citizen against abuse of executive power, however, indicates that no such set of rights could ever be exhaustive, even in principle. The circumstances of practical politics, and the infinite variety of (generally legitimate) public projects and purposes, suggest the need to acknowledge a more general, open-ended right to fair treatment at the hands of the state.

Many readers will instinctively recoil from a doctrine which places so much trust in the judges, apparently so dependent on the wise exercise of judicial power. It seems necessary, however, to match executive discretion with judicial discretion. The discretions are complementary, rather than opposed; and exercise of the latter, if not the former, must be fully reasoned and open to public scrutiny.31 Modern government cannot be entirely conducted within the constraints of purpose-independent general rules. And the scope and style of judicial review—if it is to constitute a genuine protection from abuse of executive power—must reflect the discretionary nature of contemporary government.

It is also true that many legal commentators have doubted whether it is accurate, even if desirable, to ascribe to the courts a primary concern with rights in the public law field. In an interesting essay, for example, it was recently suggested that the dominant political theory in the House of Lords is a form of ‘democratic élitism’, coupled with an (p.9) exaggerated deference to professional expertise.32 This may be accompanied by a similar élitism in respect of judgments about the requirements of the law itself, as I argue in Chapter 5. In seeking to indicate a coherent theoretical grounding for administrative law, however, it is necessary to look beyond the idiosyncrasies and predilections of particular judges, even those in the highest court. As an interpretative exercise in political theory, our task must be one of evaluation, as much as description.33 Moreover, we should recognize that theoretical analysis may contribute to constitutional change by affecting the style and categories of legal thought. It has been aptly observed that ‘there can be no value-free facts about the British constitution’.34 Legal analysis, then, must necessarily be creative: to analyse is to reinterpret and therefore—within reasonable limits—to re-create.

Nor should we expect the most persuasive account of the judicial function always to be readily apparent on the face of court judgments. If Lord Bridge's speech in the Gillick case35 illustrates his adherence to ‘democratic élitism’, it may also be consistent with a broadly coherent conception of rights. It is hard to know whether Lord Scarman's contrasting speech comes closer to the ‘ideal of accountability under law’, enabling the application of ‘legal principles of good administration’,36 unless we can make more explicit our theory of ‘law’. Without an explanation of the legitimacy of judicial review, within a democratic polity, we cannot assess the respective merits of judicial ‘activism’ or passivity. A rights-based view may indeed reveal that Lord Scarman's approach serves only ‘to substitute the view of a legal élite for that of a politically accountable political élite’.37

In the same way, one cannot defend an elastic conception of locus standi, on the basis of the public interest in keeping public agencies within the law, without further analysis of the rule of law. The problem of legitimacy in judicial review concerns precisely the nature of the courts' superior claim to determine the law. A conception of public law as protecting rights—individual entitlements which fetter the exercise of government power—may explain the court's ability to substitute its view of the applicable law for that of the public agency. We cannot escape the need for an explicit conception of the rule of law. In the absence of a normative theory of law—evaluative as well as descriptive—one can give no account of the real nature of judicial (p.10) review. Appeal to the ‘rule of law’ is empty rhetoric in the absence of further elaboration.

Chapters 6 to 9 below are devoted to a discussion of public law rights, in an attempt to give more substance to Dicey's doctrine of the rule of law. No doubt, the attempt marks only the start of an enterprise whose validity many will question. It none the less forms part of a broader endeavour to elicit the merits and advantages of traditional common law reasoning, and to demonstrate its potential for the renewal of British constitutionalism. Although the courts have been rightly condemned by many commentators and critics for failing to develop a public law which is adequate to contemporary needs, we should guard against the danger of identifying such failure with inherent weakness in the common law. As a constitutional framework for securing justice and fairness, the common law is self-evidently adaptable to new insights and fresh demands. In the absence of a radical ‘new constitutional settlement’,38 but while absorbing the full implications of European Community membership, the common law must be developed with imagination to meet the needs of modern constitutionalism.

Common law and statute

The primacy Dicey accorded the common law reflects our basic constitutional arrangements much more closely than is usually understood. First, legislation obtains its force from the doctrine of parliamentary sovereignty, which is itself a creature of the common law and whose detailed content and limits are therefore matters of judicial law-making. (It could hardly, without circularity, be a doctrine based on statutory authority.) Parliament is sovereign because the judges acknowledge its legal and political supremacy; and the rule of implied repeal, whereby a later statute always abrogates an earlier one to the extent of any conflict, is a doctrine developed by the courts to preserve the supremacy of the contemporary legislature.

Secondly, for all its intrusive impact on the detailed content of the modern law—in almost all fields of government and public affairs—legislation is in one important sense inferior to judicial precedent as a source of constitutional law. (It is usually asserted that statute is the superior source because common law rules may be abrogated by statute.) The point here is not simply that a statute is always necessarily subject to judicial interpretation—a process inevitably dependent on common law standards and traditions. Dicey's attempted reconciliation of fundamental principles rested largely on this claim: ‘Powers, however (p.11) extraordinary, which are conferred or sanctioned by statute, are never really unlimited, for they are confined by the words of the Act itself, and what is more, by the interpretation put upon the statute by the judges.’39

It is also the case that while the common law embodies many of the values traditionally associated with the rule of law, statute is necessarily more piecemeal and technical. It may supplement common law principles in specified classes of cases, or operate to overturn and defeat them (usually within fairly narrow boundaries). What it cannot do is displace the common law by providing a rival vision of the constitutional order. For that, if it is desired, we are largely dependent on developing the common law. Sir Owen Dixon's tribute to the common law as ‘an ultimate constitutional foundation’ rested primarily on that perception.40 The form and method of modern legislation imposed real limitations on its content and scope: ‘A rhetorical question may be enough to make this clear. Would it be within the capacity of a parliamentary draftsman to frame, for example, a provision replacing a deep-rooted legal doctrine with a new one?’41

Dicey's account is not as dated as it superficially appears. Statutes may, of course, be interpreted with the aim of achieving consistency of principle across a wide range of public affairs. But the criteria of interpretation can hardly be supplied wholly by statute. (Which principles are to be taken as fundamental when statutes seem to conflict?) Sir Owen Dixon's rhetorical question points us towards important limitations on legislation as a source of legal principle. The relationship between statute and common law has not generally been well understood, and we shall need to explore it in some detail.42 It is of central importance to the account of British constitutionalism which I wish to advance.

The nature of the rule of law as a political ideal is illuminated by reflection on the theory of law defended by Ronald Dworkin in Law's Empire.43 Dworkin's proposal that a judge should follow the guide of ‘integrity’ in his decisions echoes Dicey's emphasis on the role of general principles of law applied by the ordinary courts. Dicey, like Dworkin, was concerned with the overall coherence and unity of the legal order.44 I shall argue, however, that Dworkin fails to distinguish sufficiently between the nature of the principal sources of law. Although his insistence on the close interdependence of law and morality is persuasive, his attempt to reconcile the conflicting requirements of (p.12) political morality gives exaggerated weight to ‘fairness’ at the expense of ‘justice’. I shall suggest that fairness—meaning here the appropriate distribution of political power—finds its constitutional expression in the enactment and application of statute. It has no relevance to adjudication at common law.45

None the less, Dworkin's vision of the legal order as an integrated whole, in which both common law and statute may be viewed as harmonious aspirations of a ‘community of principle’, makes an important contribution to an understanding of modern constitutionalism. In particular, we should accept his invitation to treat legislation as ‘flowing from the community's present commitment to a background scheme of political morality’.46 It would be mistaken, however, to understand this idea as a proposal to give equal weight to the requirements of legislation and common law, taken at face value and without regard to the nature of those requirements in the circumstances of particular cases. I shall argue instead that the common law embodies principles of equality and reasonableness, and traditional conceptions of individual rights, which—within appropriate limits—must govern the interpretation of statute.

P. S. Atiyah's study of the relation between common law and statute helps to identify the necessary qualifications to Dworkin's account of law as integrity.47 Doubting the notion that legislation might be invoked as a source of analogy, where it was not directly applicable, he pointed out that, ‘if statute law and common law really constituted one integral body of law, there seems no reason why our sovereign Parliament should not be able to declare the common law to be something different from what the judges say it is’.48 He was inclined to think that ‘most lawyers and even most parliamentarians would today think this improper, if not indeed in some sense conceptually impossible’. I shall seek to support those conclusions, contending that legislation is inherently incapable of displacing the common law as a body of legal principle, and defending the interpretative role of the courts as essential to the separation of powers and, therefore, to the rule of law.

At the heart of the rule of law ideal is a conception of adjudication which treats legislation as the outcome of a democratic process whose legitimacy is ultimately dependent on its respecting minimum standards of justice. A judge defers to the popular will, as represented by statutes duly enacted, because his conception of political morality includes (we may reasonably suppose) a commitment to democracy. But it is necessarily a qualified commitment. He would be unlikely, for example, to accept that the (existing) legal rights of individual litigants could be (p.13) determined, in cases of dispute, by popular opinion.49 And the virtues of democracy will not persuade a judge—if he is rational—of the necessity to give automatic and unqualified allegiance to every parliamentary enactment, whatever its content. A wise judge will be reluctant to accept at face value legislation which violates important civil rights, and will strive to interpret it consistently with traditional (common law) values of individual liberty and autonomy.

Dworkin's ideal of integrity should therefore be placed in the context of well-established presumptions of legislative intent, which impose necessary, if limited, constraints on majoritarian government in the interests of liberty and equality, and in defence of individual rights and expectations. The requirements of those rights and interests are not immutable, and their content always vulnerable to fresh appraisal, but they are none the less fundamental to liberal democracy. The ‘community of principle’ which the legal order should seek to reflect is essentially dependent—in the British context—on the principles of justice embedded in the common law.

David Dyzenhaus has shown how the resources of the common law have provided certain, bolder judges in South Africa with the means of administering justice in the teeth of laws enforcing apartheid.50 He ably defends the ‘common law approach’ of judges who, embracing an ideal of law akin to Dworkin's, have refused to co-operate with legislative policies which violate fundamental standards of justice and equality— except in so far as the clear terms of unambiguous legislation have left them no choice. Their stance is contrasted with that of ‘plain fact judges’—an outlook bolstered by legal positivism51—who have given largely uncritical allegiance to the goals of repression and inequality which have undeniably informed the legislative process.

Dyzenhaus shows how some judges have sought to preserve the separation of powers by attributing to the legislature an underlying commitment to fundamental principles of justice consonant with common law values—a fiction fully justified by their conception of the legal order. Legislation is then viewed, not as a simple expression of the wishes of a majority of legislators—unrepresentative of the population as a whole—but as something to be reconciled, so far as possible, with (p.14) the values which underpin the rule of law. In refusing to relinquish the court's jurisdiction to protect detainees from illegal detention, or to acknowledge any greater deprivation of legal rights than is explicitly enacted, such judges work to ‘make statutes cohere with the fundamental purposes or ideals of their legal system, which are expressed in the common law’.52

The study of administrative law in modern South Africa is complemented by an instructive analysis of Liversidge v. Anderson.53 Dyzenhaus explains how, as in the decisions of the South African Appellate Division, divergent styles of interpretation produced very different reactions to radical departures from common law principles. For the majority in Liversidge the extraordinary nature of the powers of detention conferred on the Home Secretary, and the perceived requirements of national security in time of war, weakened the relevance of traditional modes of interpretation. Echoing Lord Finlay LC's refusal, in 1917, to apply a presumption in favour of liberty to an executive measure intended to avert a ‘public danger’,54 the majority thought that the immediate context of the Emergency Powers (Defence) Act 1939 justified a purposive construction of Defence Regulation 18B in the interests of national security.55 It followed that the minister's decision to detain a man without trial, on the basis of his ‘hostile origin or association’, was not subject to judicial control. There could ‘plainly be no presumption applicable to a regulation made under this extraordinary power that the liberty of the person in question will not be interfered with, and equally no presumption that the detention must not be made to depend…on the unchallengeable opinion of the Secretary of State’.56

When, in his forceful dissent, Lord Atkin insisted that the regulation's requirement of ‘reasonable cause’ for the minister's order invited judicial scrutiny, he took the unusual nature of the legislation as justification for especial vigilance in defence of individual liberty. His famous restatement of common law orthodoxy—that ‘the judges are no respecters of persons and stand between the subject and any attempted encroachment on his liberty by the executive, alert to see that any coercive action is justified in law’57—was grounded in an understanding of the moral complexity of the interpretative process. Drastic invasions of liberty call for more urgent appraisal as a necessary check on the operation of legislation which is clearly capable of producing great injustice and suffering. And if that is a political judgment, it is also (p.15) a legal one—entailed by judicial responsibility under an adequate conception of the rule of law.

Lord Atkin's outspoken rebuke to the majority for being ‘more executive minded than the executive’—a rebuke vindicated by subsequent cases58—obtained its authority from his grasp of the fundamental nature of ordinary common law reasoning from established principle. Where legislative intention is too readily identified with the immediate purposes of government or parliamentary majorities—even broadly representative majorities—judicial co-operation with the statutory purpose can undermine rather than uphold the rule of law. As Dyzenhaus explains the point, on Lord Atkin's approach, ‘Parliament's intentions are best seen as a construct of reasoning in accordance with common law principles’.59 For Lord Atkin, then, ‘the common law interpretative context is the master over legal meaning’.60

In the result, the dissenting speech in Liversidge v. Anderson is a fine illustration of adjudication in the common law tradition, which enables judges to control the executive by interpreting statutes in accordance with the ‘rule and reason of the common law’.61 They employ the ‘principles of critical morality protective of…liberty already embedded in the law’.62 Sir Edward Coke's maxim was rightly applauded by C. K. Allen as an ‘essential guiding rule’, which ensured the continuity of legal development, adjusting the impact of new measures to fit the broader constitutional scheme. In Allen's view, it was a ‘dominant principle’ that ‘the common law is wider and more fundamental than statute’ and that ‘wherever possible, legislative enactments should be construed in harmony with established common law principles rather than in antagonism to them’.63

It is, of course, often suggested that the traditional precedence given to common law values is inconsistent with a modern, progressive legal order, in which legislation deserves enhanced respect. I shall argue, none the less, that the common law is a constantly evolving apparatus for protecting basic values. It is inherently open to changing perceptions of the requirements of justice and the demands of basic principle. It follows that the content of the common law is ultimately a matter of reason. If the common law is primarily constituted by general principles, moreover, it follows that lawyers' perceptions of justice have no unique authority.64 Principles cannot be imposed by authority but must be (p.16) argued for and understood. The common law constitution therefore enshrines a ‘protestant’ conception of the rule of law, which might usefully be understood as a continuing process of argument about the requirements of justice and reason—a process in which every citizen should be encouraged to participate as an integral part of conscientious citizenship.65

The legacy of legal positivism

Dicey's failure to provide a consistent and coherent theory of the constitution was mainly attributable to his adherence to Austinian legal positivism. It was ultimately impossible to reconcile his emphasis on the rule of law with the unlimited sovereignty of Parliament. If there are truly no limits to legislative supremacy, common law rights and liberties can always be overridden. An insistence on there being a source of ultimate political authority, which is free from all legal restraint and from which every legal rule derives its validity, is incompatible with constitutionalism. It envisages the legitimate exercise of absolute power, albeit according to constitutional forms, regardless of its consequences for established rights or settled expectations. According to F. A. Hayek, ‘the whole history of constitutionalism, at least since John Locke…is that of a struggle against the positivist conception of sovereignty and the allied conception of the omnipotent state’.66

Dicey's conception of legislative supremacy has become so ingrained amongst English lawyers—Scots lawyers have wisely been more cautious—that it is hard to question his doctrine without appearing to lose touch with practical reality. Until very recently, it was almost unthinkable that the courts would ever refuse to apply an Act of Parliament; and attempts to indicate necessary exceptions to the doctrine were understandably thought to be somewhat unreal, addressing ‘improbable extremes’.67 But Dicey's positivist assumptions—and those of his adherents—have skewed our wider constitutional vision, and the confused condition of contemporary theory seems to be the consequence.

It is a question of the nature of public law and the legitimacy of the judicial function. If all authority ultimately derives from Parliament, on the ground of its connections to the electorate, even the interpretative function must be seriously constrained. If judicial review must be (p.17) understood—as standard explanations would have it—as essentially a means of ensuring compliance with the legislative will, the traditional role of the courts in defence of individual rights is undermined. Inevitably, it comes to appear an illegitimate usurpation of political power, because individual rights—if they are genuine—necessarily inhibit the freedom of majorities or governments to enforce their will.

The truth seems to be that there is no straightforward or objective distinction between the application of statutes and their interpretation. A restrictive interpretation of statute, in defence of individual rights, necessarily limits the field of its operation; and the most restrictive construction may reduce it to practical impotence. The legitimacy of such interpretative approaches to duly enacted legislation is obviously a question of great importance and lively controversy. The point remains, however, that the traditional role of the common law in defence of justice and liberty—as those ideals have been understood—is radically inconsistent with a notion of unlimited legislative supremacy. The interpretative function itself denies that the only source of legal authority is Parliament, even in respect of statute; or at least entails that, in the course of applying statutory injunctions to particular cases, the legislative will must be tempered with (judicial) reason.68

Other writers have drawn attention to the ‘pragmatic contradiction’ in the work of Dicey and his contemporary disciples.69 It is particularly plain in Sir William Wade's critique of the judicial attitude to privative clauses, which seek to oust the court's jurisdiction in obvious violation of the rule of law doctrine. On the one hand, the judges are described as ‘disobeying’ Parliament in service of a higher constitutional mandate; on the other, unlimited legislative supremacy is not denied. Wade therefore seems to claim that ‘judges act in violation of legal norms both when they disobey Parliament's will, as he conceives it, and when they do not’.70 What is missing here is an account of the constitutional basis of judicial ‘disobedience’.

D. J. Galligan has sought to distinguish between two judicial functions.71 The courts must both settle jurisdictional, or ‘demarcation’, disputes, keeping public authorities within their allotted spheres of action, and ensure the observance of principles of good administration by such authorities in exercising their powers. While the former function may be dependent on dutiful adherence to the legislative mandate, the latter must be independently justified: ‘the primary source of principles (p.18) of good administration lies beyond Parliament, and their justification depends on values in the constitutional order that precede the doctrine of sovereignty’.72 In practice, these different functions are likely to be closely related and intertwined. If principles of good administration are ultimately dependent, as I shall suggest, on underlying convictions about justice and fairness, their application will be sensitive to the nature of official action and its impact on individual interests. Those considerations will inevitably affect the court's view of the legitimate scope of an agency's jurisdiction, and the appropriate construction of its statutory grant of power.73

It follows that the court's adherence to legislative supremacy must, if rational, be a qualified one. On some questions Parliament must have full authority. It is not the courts' function to formulate conceptions of the public interest in opposition to legislative requirements. In supervising the execution of public policy in particular cases, however, the judicial function must rely on constitutional principles which are accepted for their intrinsic value, and not because they have legislative approval—unless such approval is purely notional, and therefore mainly fictitious. It is therefore worth contesting the scope of parliamentary sovereignty, not merely on account of the ultimate contradiction between unlimited legislative supremacy and constitutional government, but also because by taking Dicey's doctrine literally public lawyers have confused the nature of public law.74

Lon Fuller repudiated the notion of parliamentary omnicompetence in the course of his critique of legal positivism.75 Like the conception of the legal order as a hierarchical system of rules or commands, which positivism assumed, Dicey's theory mistook the structure of legal authority for the nature of law itself. Dicey's observation that retrospective laws—Acts of Indemnity—‘being as it were the legalisation of illegality’, were the ‘highest exertion and crowning proof of sovereign power’,76 demonstrated a strange detachment from the realities of constitutional government. His doctrine of sovereignty derived from theories of law which displayed a ‘fatal abstraction from the enterprise of creating and administering a system of rules for the control of human conduct’.77

(p.19) F. A. Hayek made a similar point in contesting the positivist thesis that both validity and content of law derive ultimately from the will of a legislator.78 Hayek considered that legal positivism, which denied any necessary connection between law and justice, and emphasized the dependence of the content of law on the choice of the law-giver, was founded on a ‘constructivist fallacy’. Ignoring the possibilities of spontaneous social order, positivism wrongly assumed that all law was the deliberate product of somebody's will.79 A legislator who wishes to maintain a functioning legal order cannot pick and choose any rules he likes, and thereby confer validity on them: ‘His power is not unlimited because it rests on the fact that some of the rules which he makes enforceable are regarded as right by the citizens, and the acceptance by him of these rules necessarily limits his powers of making other rules enforceable.’80

Fuller adds to his analysis an appropriate contribution to the tradition of testing assertions of legislative supremacy which are ‘extreme to the point of absurdity’ by ‘illustrations that are equally absurd’.81 The best justification for that tradition, however, is that such illustrations may help to bring more familiar and mundane uncertainties and tensions into high relief. It is the philosophic unity of the practical problems of the interpretation and application of statutes, and of legal and political theory, which precludes analysis which eschews the worst eventualities. The thesis is that the deference due to the parliamentary will is necessarily a qualified one, and its interpretation in particular cases rightly sensitive to the requirements of justice and fairness. The ‘practical lawyer's’ objection to testing the limits of sovereignty—that the content of law is always subject in the last resort to the exercise of political power—assumes the unqualified, uncritical deference to statutes which better political theory condemns.


(1) Commentaries, 15th edn. (London, 1809), i. 160–2.

(2) The Law of the Constitution, 10th edn. (London, 1959; first published in 1885), Part 1.

(3) See further Ch. 3, below.

(4) This distinction is considered in Ch. 10.

(5) For the doctrine of legitimate expectation and principle of fairness, see Ch. 8.

(6) Dicey discusses John Austin's work in Law of the Constitution, 71–6, distinguishing legal from political sovereignty. For a similar analysis, showing conflicting influences on Dicey's thought, see Martin Loughlin, Public Law and Political Theory (Oxford, 1992), 140–59.

(7) Law of the Constitution, Part 2.

(8) Dicey followed Austin's ‘scientific’ approach to legal analysis: see Loughlin, Public Law and Political Theory, 13–23.

(9) See esp. F. A. Hayek, The Constitution of Liberty (Chicago, 1960), and Law, Legislation and Liberty (London, 1982); Ronald Dworkin, Taking Rights Seriously (London, 1977), A Matter of Principle (Oxford, 1985), and Law's Empire (London, 1986). For discussion of the role of theory in public law, see also P. P. Craig, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, 1990), ch. 1.

(10) Law of the Constitution, 1–35, ‘The True Nature of Constitutional Law’.

(11) Ibid. 197.

(12) Ibid. 197–8.

(13) Ibid. 202.

(14) Ibid. 193.

(15) Ibid. 203.

(16) Hayek, Law, Legislation and Liberty, i and ii.

(17) Ibid. i. 126–8.

(18) Ibid. 129–31.

(19) Chs. 89.

(20) Law, Legislation and Liberty, i. 137–40.

(21) Ibid. 134.

(22) Ibid.

(23) Ibid. 135.

(24) For the different senses of ‘administrative law’ see Hayek, Law, Legislation and Liberty, i. 137–8. Cf. D. J. Galligan, ‘Judicial Review and the Textbook Writers’ (1982) 2 OJLS 257 at 258. Definition of the ‘public sphere’ is subject to change in response to novel modes of exercising governmental (or quasi-governmental) power: see Datafin case [1987] QB 815.

(25) Taking Rights Seriously, ch. 4. See Ch. 3, below.

(26) Cf. Dworkin, A Matter of Principle, ch. 1.

(27) For analysis of the nature of legal interpretation, see Dworkin, Law's Empire, esp. ch. 2.

(28) For discussion, see Taking Rights Seriously, Appendix (Reply to Critics), and Marshall Cohen (ed.) Ronald Dworkin and Contemporary Jurisprudence (London, 1984), Part 2.

(29) Ch. 8.

(30) Ibid.

(31) Cf. Galligan, ‘Judicial Review’, 269–76; id., Discretionary Powers (Oxford, 1986), ch. 6. There is no general rule at common law requiring reasons to be given by public authorities, but see Ch. 8, below.

(32) David Feldman, ‘Public Law Values in the House of Lords’ (1990) 106 LQR 246.

(33) Cf. Loughlin, Public Law and Political Theory, ch. 3.

(34) Ibid. 50.

(35) Gillick v. W. Norfolk AHA [1986] AC 112.

(36) Feldman, ‘Public Law Values’, 249–50.

(37) Ibid. 251.

(38) Cf. Sir Leslie Scarman, English LawThe New Dimension, Hamlyn Lectures, 26th series (London, 1974), 75.

(39) Law of the Constitution, 413.

(40) ‘The Common Law as an Ultimate Constitutional Foundation’ (1957) 31 ALJ 240.

(41) Ibid. 241.

(42) Ch. 4, below.

(43) London, 1986.

(44) Cf. T. R. S. Allan, ‘Dworkin and Dicey: The Rule of Law as Integrity’ (1988) 8 OJLS 266.

(45) Ch. 4, below.

(46) Law's Empire, 346.

(47) P. S. Atiyah, ‘Common Law and Statute Law’ (1985) 48 MLR 1.

(48) Ibid. 19.

(49) For criticism of Law's Empire in this respect, see Ch. 4 below.

(50) Hard Cases in Wicked Legal Systems: South African Law in the Perspective of Legal Philosophy (Oxford, 1991).

(51) ‘Plain fact judges hold that the judicial role is not to make law in accordance with their convictions about what morality requires, but to apply the law as it, on a particular conception of fact, exists…If it is obvious from the public record…how legislators responsible for enacting the statute would have wanted it interpreted, the judges' doctrine of judicial responsibility compels them—a matter of legal duty—to decide the case in accordance with that want’ (Ibid. 57–8). Dyzenhaus argues that, notwithstanding its notion of judicial ‘discretion’ in the absence of unambiguous law, positivism encourages the plain fact approach.

(52) Ibid. 119.

(53) [1942] AC 206.

(54) R. v. Halliday, ex p. Zadig [1917] AC 260, 270.

(55) Reg. 18B provided: ‘If the Secretary of State has reasonable cause to believe any person to be of hostile origin or associations…and that by reason thereof it is necessary to exercise control over him, he may make an order against that person directing that he be detained.’

(56) [1942] AC 206, 219 (Lord Maugham).

(57) Ibid. 244.

(58) See Ridge v. Baldwin [1963] 2 All ER 66, 76; R. v. IRC, ex p. Rossminster [1980] AC 952, 1011, 1024–5.

(59) Hard Cases, 95.

(60) Ibid.

(61) Sir Edward Coke: ‘The surest construction of a statute is by the rule and reason of the common law’ (cited by C. K. Allen, Law in the Making (Oxford, 1978), 456).

(62) Dyzenhaus, Hard Cases, 94.

(63) Law in the Making (Oxford, 1978), 456.

(64) Here I part company with Coke's conception of the common law as ‘artificial reason’: see Prohibitions del Roy (1607) 12 Co. Rep. 63.

(65) This theme is taken up in Ch. 5. Dworkin defends a ‘protestant’ conception: Law's Empire, 413. Cf. Dyzenhaus, Hard Cases, 267: ‘The conception of the common law advanced…is a conception of a process. The process is one such that the law remains an arena for participation in a public debate about what law is after the primary legislative decisions in the arena of parliamentary politics have been made.’

(66) Law, Legislation and Liberty, ii. 61.

(67) Simon Lee, Comment on Allan, ‘Limits of Parliamentary Sovereignty’ [1985] PL 632.

(68) For interesting analysis of this dichotomy, see M. J. Detmold, ‘Law as Practical Reason’ [1989] CLJ 436; id., The Australian Commonwealth (Sydney, 1985), ch. 13.

(69) Cf. Dyzenhaus, Hard Cases, 236–8. See also Geoffrey de Q. Walker, The Rule of Law, Foundation of Constitutional Democracy (Melbourne, 1988), ch. 4.

(70) Dyzenhaus, Hard Cases, 238.

(71) ‘Judicial Review and the Textbook Writers’ (1982) 2 OJLS 257 at 261–3.

(72) Ibid. 262.

(73) Galligan accepts that the two functions are related, noting the ‘blurring’ of the distinction in cases like Anisminic [1969] 2 AC 147: Ibid. 262–3.

(74) Cf. Geoffrey de Q. Walker, ‘Dicey's Dubious Dogma of Parliamentary Sovereignty: A Recent Fray with Freedom of Religion’ (1985) 59 ALJ 276 at 283–4: ‘It seems that Dicey's theory is like some huge, ugly Victorian monument that dominates the legal and constitutional landscape and exerts a hypnotic effect on legal perception.’ See further id., The Rule of Law, ch. 5.

(75) The Morality of Law, revised edn. (New Haven, Conn, and London, 1969), 113–17.

(76) Law of the Constitution, 50.

(77) The Morality of Law, 115.

(78) Law, Legislation and Liberty, ii. 61.

(79) Ibid. i. 28–9.

(80) Ibid. ii. 61.

(81) The Morality of Law, 117. Fuller denies that Parliament could enact secret laws, or assign its powers to a dictator, and provides a colourful example of an ‘invalid’ statute: ‘At some point we take leave of the gravitational field within which the distinction between law and not-law makes sense.’ That point would be reached, he thought, ‘far short’ of his illustrations of invalidity.