The Rule of Law Series

Melbourne University, 7 November, 2001

As a principle of government, the rule of law, like representative democracy, and the separation of powers, has both formal and aspirational aspects. It has a certain minimum content; but the principle is usually invoked in a manner that either assumes, or explicitly asserts, more. The definition of representative democracy given by John Stuart Mill in the 19th century [1] falls short of describing a system that would satisfy the expectations of citizens in a liberal democracy of the 21st century. We speak as though there are degrees of democracy, and we make contestable claims about those degrees. Is compulsory voting more or less democratic than optional voting? Is a first-past-the-post electoral system more or less democratic than a system of preferential voting? The minimum formal content of representative democracy may be generally accepted, but the extent of the principle may be in dispute in a given society, and its practical application varies with time and place. The same can be said of the separation of powers. The nature and degree of separation varies, even among societies that we regard as having comparable systems of government. The idea of the rule of law has a formal essence, but contestable claims are made about its substantive content in different places or circumstances [2] .

My purpose is to examine some features of the way in which the principle of the rule of law affects, and influences the role of, the judicial branch of government in Australia at the beginning of the 21st century. For that purpose, it is unnecessary to consider the difficult issues that may confront judges in a society in transition, as from an undemocratic to a democratic system, or vice-versa. But issues of that kind are not entirely foreign to us. Our own system of law and government has involved transitions. They have been gradual, and peaceful, but substantial. For example, the Supreme Court of New South Wales was established in 1824, when the colony, which then included the whole of the eastern part of mainland Australia, was in a process of change from military to civilian government. Chief Justice Forbes had to deal with Governors who were accustomed to command, and who regarded a court as something to be controlled. In 1827, in a letter to the Under-Secretary of State for the Colonies concerning the relationship between the Supreme Court and the Governor, Forbes wrote:

"The notion of control is inconsistent with the nature of a Supreme Court ; the judicial office stands uncontrolled and independent, and bowing to no power but the supremacy of law" [3] .

That assertion of the rule of law was made by a colonial Chief Justice, in a remote part of the British Empire, writing some 50 years before A. V. Dicey. Again, the gradual changes that occurred, over the 20th century, in Australia's relations with the United Kingdom, and the development of nationhood, presented issues about sovereignty and the source of our basic law. Some of those changes were considered recently by the High Court in Sue v Hill [4] .

As an idea about government, the essence of the rule of law is that all authority is subject to, and constrained by, law [5] . The opposing idea is of a state of affairs in which the will of an individual, or a group, (such as a Party), is the governing force in a society. The contrasting concepts are legitimacy and arbitrariness [6] . The word "legitimacy" implies an external legal rule or principle by reference to which authority is constituted, identified, and controlled.

In Australian legal and political discourse, a governing authority could not satisfy the requirements of the rule of law merely by being able to point to a fundamental law which empowered it to act in an arbitrary manner. The issue is unlikely to be of practical concern, because of our federal Constitution. Even so, it is possible to construct a theoretical example to raise the point. The Parliament of the Commonwealth has power to enact laws with respect to taxation. [7] Suppose legislation created an office of Tax Collector, and decreed that every person who derived income should pay to the Collector such percentage of that income as the Collector, in his or her absolute discretion, with uncontrolled power to discriminate, might think fit. That would be a tax. But would it be a law, within the meaning of a Constitution which assumes the rule of law?

The contrast between rules of general application, known in advance, and ad hoc decision-making, is a familiar aspect of the concept of law. F A Hayek, in Law, Legislation and Liberty, wrote:

"The thesis of this book is that a condition of liberty in which all are allowed to use their knowledge for their purposes, restrained only by rules of just conduct of universal application, is likely to produce for them the best conditions for achieving their aims; and that such a system is likely to be achieved and maintained only if all authority, including that of the majority of the people, is limited in the exercise of coercive power by general principles to which the community has committed itself." [8]

The first two of the three aspects of the rule of law identified by Dicey [9] , regularity as opposed to arbitrariness or unconfined discretion, and equal subjection of all, the governors as well as the governed, to law, also reflect a view of the nature of law. Judgments in the High Court of Australia contain numerous assertions of practical conclusions said to be required by the principle of the rule of law. They include the following:

. that there must be some minimum capacity for judicial review of administrative action [10] ;

. that courts may not grant the executive dispensation from the criminal law [11] ;

. that there must be separation between executive and judicial functions [12] ;

. that judicial decisions are to be made according to legal standards rather than undirected considerations of fairness [13] ;

. that citizens have a right to a fair trial [14] ;

. that citizens have a right to privileged communications with legal advisers [15] ;

. that the content of the law should be accessible to the public [16] ;

. that access to the courts should be available to citizens who seek to prevent the law from being ignored or violated, subject to reasonable requirements as to standing [17] ;

. that courts have a duty to exercise a jurisdiction which is regularly invoked [18] ;

. that citizens are equal before the law [19] ; and

. that the criminal law should operate uniformly in circumstances which are not materially different [20] .

The rule of law is such a powerful rhetorical weapon, both in legal and political argument, that care is needed in its deployment. Nevertheless, the examples just given show the extent to which the principle has been extended judicially beyond its minimum content. One reason for this may be that the existence of a written Constitution, which established a federal system of government, has accustomed Australian courts, and, in particular, the High Court, to the application and interpretation of a basic law that defines and limits all governmental power. Dicey said that Federal government is weak government [21] . The essence of federalism is an agreed division, and therefore limitation, of powers; legislative, executive, and judicial. That agreement is embodied in an instrument which is legally anterior to, and which confines, all governmental authority.

Dicey associated Federalism not only with weakness but also with judicial dominance. He wrote:

"Federalism means legalism - the predominance of the judiciary in the constitution the prevalence of a spirit of legality among the people." [22]

Those were not intended as words of commendation. Justice Gummow pointed out in his Clarendon Law Lectures that Dicey was a strong supporter of the Unionists during the movement for Irish Home Rule, and "helped give federalism a bad press in the United Kingdom for over a century" [23] . He also pointed out that Dicey was writing well before the New Deal, which might have cast a different light on the supposed weakness of federalism. Dicey had one other criticism to offer. "Federalism", he wrote, "tends to produce conservatism" [24] .

Judicial review of legislative action

As Alfred Deakin told Parliament in 1902, in the course of a debate about establishing the High Court, the Australian colonies, like those of North America, were accustomed to legislatures with limited powers, and to judicial decision-making about the limits of those powers [25] . When our Constitution was being framed, the principle of judicial review, established in Marbury v Madison [26] and later cases in the United States, was taken for granted [27] . Disputes between citizens and governments, or between governments, about the meaning of the Constitution frequently involve a claim that a legislative body has acted in excess of power, or that an executive officer has acted unlawfully, or that a court or tribunal has acted beyond jurisdiction. It is the province of the judicial branch of government to determine the law, including the meaning of the Constitution, and it has the ultimate power to make a binding decision in such disputes [28] .

In one respect, the existence of a written Constitution as the basic law of a federal democratic society has a specific implication for the substantive content of the rule of law. The source of law-making authority is the Constitution, and the law, including the common law as developed by the courts, must conform to the Constitution. The Federal Parliament and the High Court both owe their existence to the Constitution. However law is made in Australia, it must be consistent with the Constitution. And so must any substantive principle said to flow from the rule of law itself.

The Constitution has been said to assume the rule of law. That was said in a case which provides a powerful example of the rule of law, and judicial review of legislative action, at work. The case is Australian Communist Party v Commonwealth [29] . During the Cold War, the Federal Parliament enacted legislation dissolving the Australian Communist Party and empowering the Executive Government to dissolve other associations. The Federal Parliament has no power to make laws on the subject of unincorporated associations; that power rests with the States. However, the Parliament has power to make laws with respect to the naval and military defence of the Commonwealth and the States; and the legislation contained a preamble reciting the reasons why Parliament thought this law was necessary for defence, even though Australia was not at war. The High Court was adamant that it was for the Court, and not the Parliament, to decide whether the law bore the character of a law with respect to the naval and military defence of the Commonwealth. Dixon J, after observing that the Constitution was framed in accordance with "many traditional conceptions", including the separation of powers, and the rule of law, firmly rejected any argument that "would have the effect of making the conclusion of the legislature final and so the measure of the operation of its own power" [30] . McTiernan J said that the preamble was "in no way decisive of the question whether the Act is valid or invalid, for that is a judicial question which only the judicature has the power to decide finally and conclusively" [31] . Kitto J said that under a unitary system of government the judgment of the legislature as to whether the law was a law with respect to defence could not be challenged but under a Federal system the central legislature is equipped with limited powers only, and the duty is cast upon the courts to determine whether laws which that legislature thinks necessary for the security of the country are within the scope of its powers [32] .

In the reasoning of the majority, it is not easy to discern what is referred to, by some jurists, as deference. The Court treated as irrelevant to its decision what was, in effect, an explanation by the Parliament of the connection between the law and a danger to national security. Parliament could not be permitted to be the judge of the extent of its own power. That would be inconsistent with the division of powers in the Constitution, by which judicial power (including the ultimate power to decide the meaning of the Constitution) was assigned to the Court. And it would be inconsistent with the rule of law. Whether a law was within power was for the Court to decide. If a law were within power, then the question whether there was a need for the law was entirely a question for Parliament, and was no concern of the Court. The Constitution marked out the territory for each branch of government. The only deference required of each was to keep out of the other's territory. That meant that judicial decisions were for the Court; and political decisions were for Parliament. The Court did not engage in any dialogue with the Parliament. It held that the legislation was invalid and therefore not part of the law. The final say was within the people of Australia. They rejected a proposal for a constitutional amendment. That was not deferential dialogue [33] . That was due process of (constitutional) law.

There have been many similar examples, over the century since Federation, of the High Court's insistence upon its right, and constitutional responsibility, to decide whether laws enacted by the Federal Parliament, or State Parliaments, conform to the limitations upon legislative power found in the Constitution. One of the most striking of these concerned an attempt by the Parliament to establish what was undoubtedly envisaged by the framers of the Constitution as an important and influential agency of government, to be concerned with regulating aspects of finance and trade of federal significance. Section 101 of the Constitution provides:

"There shall be an Inter-State Commission, with such powers of adjudication and administration as the Parliament deems necessary for the execution and maintenance, within the Commonwealth, of the provisions of this Constitution relating to trade and commerce, and of all laws made thereunder."

This was intended to follow the precedents of the Inter-State Commerce Commission created in the United States in 1887, and the Commission constituted in England by the Railway and Canal Traffic Act 1888. The kind of function contemplated for the Commission is indicated by s 102, which empowers the Federal Parliament to make laws forbidding unreasonable preference or discrimination by States in relation to railways. Quick and Garran, writing in 1901 [34] , foresaw that, as an administrative body, the Commission would supervise the execution and prevent the violation of laws relating to inter-state and foreign commerce, and in addition, that it would have wide powers of adjudication.

An Inter-State Commission was set up, pursuant to the constitutional mandate, by an Act of 1912. Part V of the Act purported to invest the Commission with judicial power. The High Court, in the Wheat Case [35] , held that, notwithstanding the explicit reference to "powers of adjudication" in s 101, the Constitution was framed on the fundamental principle of the separation of powers, and that the judicial power of the Commonwealth could be validly invested only under Ch III of the Constitution, in a court there referred to. The Constitution, in s 101, was said to have provided only for an administrative and consultative organ, with incidental quasi-judicial functions, of the kind exercised by a Commissioner of Patents or a Collector of Customs.

The decision had major implications for the principles of the separation of powers and the rule of law.

Sir Owen Dixon's statement, in the Communist Party Case, that the Constitution was framed in accordance with traditional conceptions such as the separation of powers and the rule of law, raises an issue as to how those conceptions may be used to resolve questions of constitutional interpretation. As was noted earlier, the content of the conceptions is a matter of contention. The same may be said of representative democracy. Is it only the minimum content, or essence, that can be used to inform a proper understanding of the Constitution? To what extent can contestable opinions about what the rule of law entails in a liberal democracy of the 21st century provide a legitimate basis for the interpretation of an instrument of government? Sir Owen Dixon was referring to conceptions that he described as "traditional", by reference to which the Constitution was framed. Traditional as those conceptions may be, they are far from precise. To what extent is it consistent with the proper function of a court interpreting the Constitution to go beyond their essential, and generally agreed content, as a guide to the meaning of that text? This is a perennial problem, which may have significance in relation to the powers of the Parliament concerning judicial review of administrative action.

Judicial review of administrative action

Judicial review of administrative action is a familiar example of the application of the rule of law. Whether, in a given case, its basis is constitutional, as in an application for an order under s 75(v) of the Constitution against an officer of the Commonwealth, or statutory, as in proceedings under the Administrative Decisions (Judicial Review) Act 1977 (Cth) or the Migration Act 1958 (Cth), or the common law, as in an application to a Supreme Court for a prerogative writ against a lower court or administrator amenable to such a writ, the essence of what is involved is to compel those invested with governmental power to exercise such power according to law. Where what is in question is a decision of an administrative tribunal, or a court of limited jurisdiction, in Australia the distinction between jurisdictional and non-jurisdictional error remains significant, although its practical content may depend upon the nature of the decision-making body [36] .

Section 75(v), included in the Constitution to ensure that Federal officials did not exceed their authority, now operates as an important source of power in the Federal judiciary, especially the High Court, to require officers (including judicial officers) of the Commonwealth to act within the law. Of course, it is for the Parliament, in the exercise of its legislative power, to enact the law to which such officers must conform, but the Parliament's legislative power is limited by the Constitution itself. The debate as to whether ultra vires is a complete explanation of the basis of judicial review, the courts measuring the conduct of the decision-maker against the statute pursuant to which impugned action is taken, construed in the light of established principles, or whether the common law is a separate source of judicial power, must in any event take account of the power of Parliament (acting within the Constitution) to alter the common law.

In the development of administrative law, the focus was upon the powers and duties of administrators. More recently, there has been a tendency to focus upon the rights of citizens. This tendency is strongest in countries which, unlike Australia, have formal Charters or Declarations of Rights. Where declared rights are part of the law, then they are part of the scheme to which administrators must conform. Even in Australia, a rights-conscious community is not slow to resort to litigation aimed at keeping the executive within the law.

This brings me to the subject of privative causes. A statutory provision that effectively limits or excludes judicial review of administrative action is as much a part of the law as a provision that empowers administrative action. But Parliament's capacity to empower administrative action is fettered by the limits imposed by the Constitution upon its powers. And, as the Constitution assumes the rule of law, a question may arise as to the consequences of that assumption in this context. In Australia, privative clauses were originally used by parliaments most commonly in the field of industrial law, for the purpose of confining the capacity of the ordinary courts to interfere in the decision-making of specialist industrial tribunals. Much of the learning on the subject developed in that area. More recently, immigration law has been the growth area for litigation seeking to challenge administrative decisions, and for parliamentary response in the form of limitations on the scope for curial intervention.

As the basis for judicial supervision of administrative conduct is the need to ensure that an official who is given, by statute, a certain power, acts within that power, and conforms to the express and implied conditions which are imposed upon its exercise, a privative clause presents a conceptual problem. There is an apparent inconsistency between a provision defining and limiting power, and a provision which appears to say that such a limitation may not be invoked as a ground of challenge to a decision made in the exercise of such a power. The approach that has prevailed to date in Australia has been to treat the problem as one of statutory construction, and to seek to resolve the inconsistency in that manner. It was formulated by Dixon J in The King v Hickman; Ex parte Fox and Clinton [37] , and has since been followed in many cases, although some aspects remain to be explored fully. In brief, the statute in that case was construed to mean that the decision in question would not be invalidated on the ground of failure to conform to the limitations on power or authority, or the manner of its exercise, contained in the statute, provided that the decision was a bona fide attempt to exercise the power, that it related to the subject matter of the legislation, and that it was reasonably capable of reference to the power. That may be regarded as a qualified amplification of the power.

Legal theory does not require the conclusion that all forms of restriction upon the capacity of the judiciary to override executive action on legal grounds necessary involve a derogation from the rule of law. Subject to any limits on legislative power imposed by the Constitution, it is for Parliament to define the power and jurisdiction of administrators and tribunals. The essential supervisory role of the courts is to ensure that the recipients of the power or jurisdiction conform to the terms and legal conditions upon which it is conferred, and by which it is confined. But not all courts have that role, and most courts have a jurisdiction which is created, and may be limited, by Parliament. To the extent to which a privative clause, properly construed, lawfully amplifies power or limits jurisdiction, then respect for the rule of law requires courts to give effect to that expression of legislative will. Subject to the Constitution, the Parliament, in the exercise of its legislative power, is not obliged to maximise the area of potential justiciability of disputes between citizen and government. In this context, an appeal to the rule of law may be to its aspirational rather than its formal content. It may be an appropriate use of political rhetoric to contend that a privative clause is a derogation from the rule of law, but that is not a substitute for legal analysis. And the primary focus of legal analysis will be the legislative competence of the Parliament. If such competence exists, the rule of law requires that its exercise be respected by the judiciary. The two most obvious potential constraints upon the capacity of the Parliament to enact privative clauses are s 75(v) of the Constitution, which confers upon the High Court a jurisdiction that cannot be diminished by Parliament, and the limitations upon the subject matters with respect to which the Parliament has power to enact laws. The ultimate bounds are set by the limits upon the power of Parliament itself. This is a point at which the Constitution's assumption of the rule of law may be significant. The extent to which it is within the competence of Parliament to exclude all forms of judicial review of administrative action remains to be defined.

The complexities of the interplay between legislative will, executive action, and judicial power will continue to evolve. The ultimate principle underlying the role of judicial supervision, however, is simple. It was expressed by Lord Denning, in words that related to the jurisdiction of tribunals:

"If tribunals were at liberty to exceed their jurisdiction without any check by the courts, the rule of law would be at an end" [38] .

Judicial action

The rule of law is not just a principle that, in a variety of ways, is enforced by courts. It controls the operation of courts themselves.

Contrary to an assumption popular among some lawyers, the rule of law does not necessitate an ever-expanding role for the courts and the legal profession in the affairs of governments and citizens. And the rule of law does not mean rule by lawyers. Subject to any constitutional limitations on their powers, it is for parliaments to decide what controversies are justiciable, and to create, and, where appropriate, limit, the facilities for the resolution of justiciable controversies. Parliaments regularly expand and contract the subjects of justiciable controversy. That is what much law-making entails. In Australia, the most common form of litigation, providing a substantial part of the caseload of courts of general jurisdiction, is the common law action for damages, framed in the tort of negligence, arising out of a motor vehicle accident. In New Zealand, there is no such litigation. A scheme of no-fault compensation, dealt with administratively, has replaced the action at common law. That was not a derogation from the rule of law. It was a change in the law. Whether it was good policy is irrelevant. The point is that the rule of law does not require all possible disputes to be justiciable, or all grievances to be resolved by litigation. The capacity of the legislature, by altering the law, to determine what claims are to be resolved by the litigious process, and to regulate the manner in which those claims are to be resolved, has a major practical effect upon the business of the courts. Issues of funding, and allocation of resources, may have a powerful effect on policy. Most Australian courts were themselves created by a parliament, and their jurisdiction is defined, and may be altered, by legislation.

The rule of law does not require that the entire apparatus of the judicial system be brought to bear upon all disputes, or even upon all disputes about legal rights and obligations. It does not require that everyone disgruntled about a library fine should have a right of appeal to a court, and, if necessary, ultimately to the High Court. In fact, as a result of legislation enacted by Parliament, no one has a right of appeal to the High Court. All appeals require special leave which may be granted or withheld upon grounds defined by statute. The legislation which thus limits appeals to the High Court provides an example of Parliament, consistently with the Constitution, limiting access to the courts. It is a method of rationing scare judicial resources. When a parliament creates a new area of potential controversy, as, for example, by enacting laws concerning privacy, or various forms of discrimination, it may decide that the appropriate machinery for resolution of such controversies is administrative, rather than curial. Town planning issues, which often involve balancing rights of private property and the public interest, can be dealt with politically, or administratively, or judicially. Parliaments have a wide discretion as to the appropriate method of resolving such issues, and different Australian jurisdictions provide a variety of solutions. This is not inconsistent with the rule of law provided, of course, it conforms to the Constitution.

Theories about the role of judges, and the relationship between that role and the rule of law, need to take account of the differences between judges. There are 976 judicial officers in Australia, and only 7 of them are judges whose decisions are never the subject of a potential appeal to a higher court or some other form of judicial review. The appellate system is a powerful instrument for ensuring adherence to the principle of legality by the judiciary. The possibility of appellate review means that, even in that small minority of cases where judges might be called upon to break new legal ground, or in areas where they are invested with substantial discretion, judges must conform to a legal discipline by which their powers are circumscribed. Only a relatively small number of cases go on appeal, and all but a few appeals are finally disposed of by an intermediate appeal court. But the very existence of the appeal system, and of an ultimate court of appeal, is a powerful influence for judicial conformity to law.

Within an ultimate court of appeal, collegiate pressures, decision-making by majority, the techniques by which legal development must be explained and justified, and the capacity of Parliament to reverse the effect of a decision, are all factors which inhibit deviancy. Even when such courts bring down what is regarded as a radical decision, the process of reasoning by which the decision is justified is likely to be one that would be regarded in most other disciplines as conservative. The decision may be explained as based upon well entrenched values in the common law, and precedents that are swept aside may be characterised as departures from the purity of previously established principle. Substantial changes in the common law are often accompanied by elaborate deference to precedent and settled principle. The best evidence of what judges themselves regard as legitimacy is to be found in the techniques by which they justify their decisions.

It is unfortunate that the process by which judges, usually judges of courts of appeal, develop and refine the common law, is often described as "making law" in a manner that implies that the process is legislative. The judicial method is, or ought to be, different from the legislative method [39] . In Breen v Williams [40] , Gaudron and McHugh JJ said:

"Advances in the common law must begin from a baseline of accepted principle and proceed by conventional methods of legal reasoning. Judges have no authority to invent legal doctrine that distorts or does not extend or modify accepted legal rules and principles."

Of course, individual judges, perhaps because of differences in legal philosophy, or personal temperament, may be more or less "activist", or more or less "conservative", in their approach to such matters as the weight to be given to precedent, the importance of legal certainty and predictability, or the proper relationship between the courts and parliaments in an area of legal change. But in truth, for all the excitement that erupts occasionally about activism, the capacity for judicial creativity is, by comparison with other forms of human inventiveness, limited.

Sir Owen Dixon, in a letter to Justice Frankfurter, expressed his disapproval of judicial adventurers by saying that a judge:

"ought to appear to believe that he has some external guidance even if in his ignorance he regards it as untrue. In the Darwinian process of adaptation to environment such a bird as the honey-sucker ought not consciously to enlarge its bill by stretching it even if reaching for the honey causes it to do so. In any case law-making ought not to be regarded as honey [41] ."

The ultimate limitation on the power of judges is the same as that on the power of Parliament: the Constitution. Covering clause 5 provides that the Constitution, and all laws made by the Parliament of the Commonwealth under the Constitution, shall be binding "on the courts, judges, and people of every State and of every part of the Commonwealth." This expresses what is otherwise necessarily implied. The principle that courts are bound by the Constitution, and all other laws, defines the relationship between judges and the other arms of government, and between judges and the community.

Dicey contrasted the rule of law with discretionary power [42] . Much of the power exercised by courts, whether given by statute or common law, involves discretionary decision-making. Discretion implies choice between legally available alternatives. The law limits the judge's area of choice. From the point of view of a litigant, the rule of law suggests that the outcome of the litigation should depend as little as reasonably possible upon the identity of the judge who hears the case. It also suggests that Parliament, in enacting law, and appellate courts, in developing the common law, should pay attention to the importance of establishing principles of general application rather than widening the scope for ad hoc discretionary judgment. The concept of laws as rules of general application, capable of being known in advance by citizens who may exercise choice, and order their affairs, accordingly, is part of the idea of the rule of law.

The common law judicial method, whether applied by trial judges, judges of intermediate appeal courts, or judges of courts of final resort, is a method of legalism. Justice Ginsburg, in her paper in this series [43] , referred to the "decision-making mores to which legions of federal judges adhere: restraint, economy, prudence, respect for other agencies of decision reasoned judgment, and, above all, fidelity to the law." I see those observations as related to a point made by Professor Troper in his paper [44] . He pointed out that the power and influence of a supreme court is greater, the more its behaviour is constrained by past decisions. The point is valid, and extends to the other constraints mentioned by Justice Ginsburg. Furthermore, it applies to the entire judiciary. For the judicial arm of government, restraint and discipline are sources of strength, not weakness.

The importance of the rule of law lies partly in the power it denies to people and to governments, and in the discipline to which it subjects all authority. That denial, and that discipline, are conditions of the exercise of power, which in a democracy, comes from the community which all government serves. Judicial prestige and authority are at their greatest when the judiciary is seen by the community, and the other branches of government, to conform to the discipline of the law which it administers. The rule of law is not enforced by an army. It depends upon public confidence in lawfully constituted authority. The judiciary claims the ultimate capacity to decide what the law is. Public confidence demands that the rule of law be respected, above all, by the judiciary.

* Chief Justice of Australia.

[1] Mill, Considerations on Representative Government (1861) at 557.

[2] For a discussion of some of the competing theories, see Craig, Formal and Substantive Conceptions of the Rule of Law: An Analytical Framework; [1977] Public Law 467; Hutchison, "The Rule of Law Revisited: Democracy and Courts" in Recrafting the Rule of Law: the Limits of Legal Order, (1999) ed. D Dyzenhaus at 196.

[3] J M Bennett (ed), Some Papers of Sir Francis Forbes, at 143.

[4] (1999) 199 CLR 462.

[5] For an account of various expressions of the principle of the rule of law, see K Mason, "The Rule of Law", in Essay on Law and Government, (1995) Vol 1 (ed) PD Finn, at p. 114.

[6] J A G Griffith, The Common Law and the Political Constitution, [2001] 17 LQR 42 at 46.

[7] Constitution, s 51(ii).

[8] Hayek, Law, Legislation and Liberty, Vol 1, at 55.

[9] A V Dicey, Introduction to the Study of the Law of the Constitution, 10th ed. (1959) at 185 193.

[10] Church of Scientology v Woodward (1982) 154 CLR 25 at 70-71 per Brennan J.

[11] A v Hayden (1984) 156 CLR 532 at 595 per Deane J; Ridgeway v The Queen (1995) 184 CLR 19 at 39, 44 per Mason CJ, Deane and Dawson JJ.

[12] Abebe v The Commonwealth (1999) 197 CLR 510 at 560 per Gummow and Hayne JJ; Enfield City Council v Development Assessment Commission (2000) 199 CLR 135 at 157 per Gaudron J.

[13] Federal Commissioner of Taxation v Westraders Pty Ltd (1980) 144 CLR 55 at 60 per Barwick CJ.

[14] Kingswell v The Queen (1985) 159 CLR 264 at 300 per Deane J; Krakover v The Queen (1998) 194 CLR 202 at 224 per McHugh J.

[15] Baker v Campbell (1983) 153 CLR 52 at 71 per Gibbs CJ; Corporate Affairs Commission v Yuill (1991) 172 CLR 319 at 346 per McHugh J; Carter v Northmore Hale Davy & Leake (1995) 183 CLR 121 at 161 per McHugh J.

[16] Incorporated Council of Law Reporting (Q) v Federal Commissioner of Taxation (1971) 125 CLR 659 at 672 per Windeyer J.

[17] Onus v Alcoa of Australia Ltd (1981) 149 CLR 27 at 35 per Gibbs CJ.

[18] Oceanic Sun Line Special Shipping Company Inc v Fay (1988) 165 CLR 197 at 239 per Brennan J; Jago v District Court (NSW) (1989) 168 CLR 23 at 76 per Gaudron J.

[19] R v Shrestha (1991) 173 CLR 48 at 60 per Brennan and McHugh JJ; Leeth v The Commonwealth (1992) 174 CLR 455 at 485 per Deane and Toohey JJ.

[20] Taikato v The Queen (1996) 186 CLR 454 at 465-466 per Brennan CJ, Toohey, McHugh and Gummow JJ.

[21] A V Dicey, Introduction to the Study of the Law of the Constitution 10th ed. (1959) at 171.

[22] Ibid, at 175.

[23] W M C Gumow, Change and Continuity, Statute, Equity and Federalism, at 72.

[24] A V Dicey, Introduction to the Study of the Law of the Constitution, 10th ed (1959) at 173.

[25] Australia, House of Representatives, Parliamentary Debates, (Hansard), 18 March 1902 at 1969.

[26] (1803) 1 Cranch 137; 5 US 87 (1803).

[27] For example, in the Convention Debates at Melbourne in 1898, Isaac Isaacs said:

"We are taking infinite trouble to express what we mean in this Constitution; but as in America so it will be here, that the makers of the Constitution were not merely the Conventions who sat, and the states who ratified their conclusions, but the Judges of the Supreme Court."

Official Record of the Debates of the Australasian Federal Convention (Melbourne), 28 January 1898, p 283.

[28] Marshall CJ said, in Marbury v Madison, "It is, emphatically, the province and duty of the judicial department to say what the law is". (1803) 1 Cranch 137 at 177; 5 US 87 (1803) at 111.

[29] (1951) 83 CLR 1.

[30] (1951) 83 CLR 1 at 193.

[31] (1951) 83 CLR 1 at 205.

[32] (1951) 83 CLR 1 at 271.

[33] cf D Dyzenhaus "The Justice of the Common Law: Judges, Democracy and the Limits of the Rule of Law" (Paper in this series pp 18, 24).

[34] Quick and Garran, The Annotated Constitution of the Australian Commonwealth (1901) at 898 899.

[35] New South Wales v Commonwealth (1915) 20 CLR 54.

[36] Craig v South Australia (1995) 184 CLR 163; Minister for Immigration and Multicultural Affairs v Yusuf [2001] HCA 30.

[37] (1943) 70 CLR 598 at 615.

[38] R v Medical Appeal Tribunal; Ex parte Gilmore [1957] 1 QB 574 at 586.

[39] See The Hon. M. H. McHugh, The Judicial Method (1999) 73 ALJ 37.

[40] (1996) 186 CLR 71 at 115.

[41] Letter of 14 January 1959 reproduced in The Oxford Book of Australian Letters, (1998) ed Niall and Thompson, at 255 257.

[42] A V Dicey, Introduction to the Study of the Law of the Constitution, 10th ed (1959) at 188.

[43] The Hon Ruth Bader Ginsburg, Remarks on Judicial Independence: The Situation of the US Judiciary, at 7-8.

[44] M Troper, The Rechtsstaat and the problem of obedience to the Law, at 13-15.