Integrity and Privative Clauses - The Third Lecture in the 2004 National Lecture Series
INTEGRITY AND PRIVATIVE CLAUSES
THE THIRD LECTURE IN THE 2004 NATIONAL LECTURE SERIES
FOR THE AUSTRALIAN INSTITUTE OF ADMINISTRATIVE LAW
BY THE HONOURABLE JAMES J SPIGELMAN AC
CHIEF JUSTICE OF NEW SOUTH WALES
BRISBANE, 2 SEPTEMBER 2004
The sumptuary rules of the Chinese Imperial Civil Service established a rigidly defined set of dress requirements for all public officials: from the black lacquer-treated hats with protruding wings and the black boots trimmed with white lacquer to the ceremonial belts backed with jade, rhinoceros horn, gold or silver. Each distinctive sub-unit or rank of the civil service also had a badge of rank in the form of a cloth chest piece embroidered, in the case of the civil hierarchy, with birds in pairs. The top rank had two stately cranes soaring above clouds. The lowest rank had a pair of earth-bound quails, pecking the grass. The military ranks wore breast patches carrying images of fierce animals such as lions, tigers, bears and panthers.
There was one distinct civil service unit with a unique system of badge identification. Western scholars, by an inaccurate analogy with the Roman administrative system, called this unit the "censorial" or "supervising" branch of government. Its role was to maintain the integrity of the mechanisms of governance. Civil officials in this branch had an embroidered breast patch, which was identical for all members of the branch, regardless of rank. It displayed a legendary animal called a Xiezhi which could detect good from evil and, allegedly, could smell an immoral character from a distance, whereupon the Xiezhi would leap upon the person and tear him or her to pieces[1].
The Chinese censorial system was organised as a separate branch of government to maintain surveillance over all other governmental activities and thereby enforce proper behaviour through processes of impeachment, censure and punishment. It also had the function of initiating recommendations for change of governmental policies, practices or personnel. The success of the Chinese Imperial tradition as a system of administration, manifest in its longevity, has been attributed to the power and vigilance of the censorate[2].
In the 1920s, Sun Yat-sen proposed that the Republic of China adopt a five yuan or branch system of government comprised of three branches from the Western governmental tradition - executive, legislative and judicial - and two from China's past: an examination branch and a control or integrity branch. In the 1930s such a branch was established and called the Control Yuan.
When an American constitutional lawyer recently proposed that modern constitutions should now incorporate a separate institutionalised integrity branch of government[3], another American scholar drew attention to the similarity between that proposal and the Chinese Imperial tradition adopted by Sun Yat-sen[4].
Integrity Institutions
In the first lecture in this series[5] I proposed recognition of an integrity branch of government as a fourth branch, equivalent to the legislative, executive or judicial branches.
I noted that in any stable polity there is a widely accepted concept of how governance should operate in practice. The role of the integrity branch is to ensure that that concept is realised, so that the performance of governmental functions is not corrupt, not merely in the narrow sense that officials do not take bribes, but in the broader sense of observing proper practice.
I put forward the idea of integrity as a useful way to conceptualise a universal governmental function, within which the body of law known as administrative law may find a place. The focus is on institutional integrity, rather than personal integrity, although the latter, as a requirement of conduct of occupants of public office, has implications for the former.
Institutional integrity goes beyond a narrow concept of illegality to encompass at least two additional considerations. First, the maintenance of fidelity to the public purposes for the pursuit of which an institution is created. Secondly, the application of the public values, including procedural values, which the institution was expected and/or required to obey.
This focus on fidelity to purpose and on applicable public values does, in my opinion, distinguish the integrity function from other governmental functions, including most executive, legislative and judicial decision-making, which are concerned with the quality of actual outcomes.
I summarise the integrity functions performed by various institutions of governance, set out more fully in the first lecture.
The traditional role of ministerial responsibility in a Westminster system can be understood, in part, as the performance by Parliament of an integrity function. The particular institutional manifestations of this function include the role of the formal Opposition, the significance that has always been attributed to daily question time, debating conventions such as those attendant on no confidence resolutions, and inquiries by Parliamentary committees. Whilst these procedures are often directed to the quality of outcomes, and therefore perform an executive function, or are concerned with legislative reform, and therefore can be seen as part of the legislative process, frequently they are directed to maintaining the integrity of government.
The Parliament also has the ultimate authority to remove judicial officers for proved misbehaviour. In this regard the Parliament performs an integrity function with respect to the judicial branch of government.
In a Westminster system of government, the head of state, the Queen in the United Kingdom and the Governor-General in Australia, is also part of the integrity branch. Walter Bagehot's threefold classification of the powers of the constitutional monarch in The English Constitution was: to be consulted, to encourage, and to warn. This, in large measure, is the performance of an integrity function.
Many of the institutions of the integrity branch of government are emanations of the executive. Nevertheless over the years many of these institutions have, by legislation and practice, developed an independence which has become institutionalised.
Perhaps the oldest such institution is the centralised audit office, which we in Australia generally call the Auditor-General. The focus on probity of governmental expenditure in the course of the financial audit is the performance of an integrity function.
Over recent decades new specialist institutions have emerged. In New South Wales the Hong Kong model of an Independent Commission Against Corruption was adopted and a separate Police Integrity Commission was created. There is a statutory Corruption Commission in Western Australia and, in 2002, two Queensland institutions were merged in the Crime and Misconduct Commission.
Public inquiries frequently focus on integrity matters. The operations of parliamentary committees and executive inquiries, notably Royal Commissions, have been of considerable significance in this respect. In Australia over recent decades important integrity issues have been considered by the Western Australian Royal Commission into the Commercial Activities of Government, referred to as the inquiry into "WA Inc", the Fitzgerald Inquiry into Corruption in the Queensland Police Force and the Wood Royal Commission into the New South Wales Police Force. In the United Kingdom the Scott Report on Matrix Churchill, the Phillips Report on mad cow disease and the Hutton inquiry about press reports concerning British participation in the Iraq War, all raised integrity issues.
Various complaint handling mechanisms have been established which often focus on integrity issues. I include in that respect the Ombudsmen, particularly in those states in which an independent corruption commission does not exist. Of course complaint mechanisms are designed, in part, to improve the quality of decisions and in that regard perform an executive function. Nevertheless, such organisations do perform integrity functions.
Over recent decades a concern with personal integrity of public officials, particularly focussed on the existence of conflicts of interest, has also taken new institutional forms. Formal codes of ethics identifying standards of behaviour have been promulgated in many areas of the public sector. In some places separate institutions have emerged, for example, the Integrity Commissioner in Ontario and the Integrity Commissioner in Queensland.
Integrity and Administrative Law
The role of the courts in supervising administrative action has frequently given rise to controversy with respect to the proper role of a judiciary in a democratic polity. What constitutes transgressing that proper role is a matter on which reasonable minds may differ. In the first lecture I expressed the view that the idea that judicial review is part of an integrity branch of government, expresses the limits upon the permissible scope of judicial review in a manner which may be useful. When the courts review matters which do not give rise to integrity issues, it is likely, I said, that they have gone too far. The issue is one of "judicial legitimacy", to adopt Chief Justice Gleeson's phrase[6].
The distinction between judicial review and merits review is a fundamental principle of Australian administrative law. Judicial review is a manifestation of the integrity branch of government. Merits review is a manifestation of the executive branch.
Australian administrative law continues to adhere to the proposition that there is no error of law in making a wrong finding of fact[7], unless the fact is jurisdictional.
The task of policing the boundaries of the legality/merits dichotomy is a continuing one. Over recent decades, as I indicated in the first lecture, Australian administrative law has diverged from that of England in important respects with respect to the identification of the boundary and the rigour with which it is enforced. This is a function of our strong tradition of the separation of powers, manifest primarily in the force of the High Court's Chapter III jurisprudence[8]. In the second lecture, I discussed the gravitational pull of High Court jurisprudence about the "constitutional writs" upon the common law[9].
Subject to what may emerge from the High Court's further consideration of Kable, the difference between the constitutional writs and the common law applicable to State administrative law appears to be well established by Craig. The constitutional writs are not necessarily attended by the same incidents as the prerogative writs[10].
For the foreseeable future we will have three distinct, but interrelated, bodies of administrative law: the constitutional writs, the common law and the ADJR Act and its State progeny. Nevertheless, the constitutional jurisprudence will exercise a gravitational pull on both the common law and on the statutory jurisdiction.
In the administrative law literature integrative concepts have emerged which threaten the legality/merits distinction. One is the suggestion that the specific rules of administrative law are part of a broader principle of preventing an "abuse of power" or of curing "serious administrative injustice"[11]. Another posits that these rules are simply "principles of good administration". Such general concepts are beguiling.
It is a short step from stating that all of the particular rules which are in fact recognised in the case law can be so categorised, to saying that the results of a particular dispute should be determined by the judge's opinion as to whether the conduct constitutes an "abuse of power", "serious administrative injustice" or that in some manner, the "principles of good administration" require judicial intervention. Such concepts are more likely to lead to judicial decisions which transgress the proper limits of judicial review in a democratic polity, than the integrative concept which I propound, namely, the performance of an integrity function.
The court system cannot supervise the broad stream of discretionary administrative decision-making, even by the application of a standard of "legality", unless that standard is narrowly confined. Nor in a democratic society should judges attempt any such task where what is criticised, as a matter of substance, is the quality of an outcome of a decision-making process. It is, however, appropriate for the judiciary to ensure the fidelity of decision-makers to their jurisdiction, so that the integrity of the institutions within which those individual decision-makers operate is maintained.
In the first lecture, I expressed my opinion, that the English cases which appeared to recognise substantive legitimate expectations, such as Coughlan and Begbie[12], the effect of which is indistinguishable from merits review, would not be followed in Australia, where legitimate expectations give rise only to procedural rights[13].
Similarly, the emergence in England of a doctrine of proportionality, under the influence of European Community law, particularly by reason of the adoption of the European Convention on Human Rights in the Human Rights Act 1998 (UK), trespasses on the merits side of the legality/merits boundary. A proportionality test requires a court to assess the weight a decision-maker has given to particular considerations and to assess the balance struck between conflicting considerations[14]. Such a process involves an executive rather than an integrity function of government[15].
In the second lecture I noted that earlier this year, the English Court of Appeal took the decisive step and recognised material mistake of fact as a basis for overturning administrative decisions[16]. This new ground of judicial review, or of appeal for error of law, will probably lead to further divergence between Australian and English administrative law. The most Australian law presently permits is an appeal from a perverse finding of fact[17] and in New South Wales not even that[18]. These cases may need to be reviewed in the light of the High Court decision in Applicant S20/2002. Judicial review for factual error is constrained by Wednesbury unreasonableness, save in the case of jurisdictional facts. The High Court declined an invitation to decide whether "material error of fact" was a ground of review in Australian law[19].
England has moved a long way from maintaining institutional integrity. The focus of attention has moved from the function being performed to the quality of the decisions made. That has not happened in Australia.
The Concept of Jurisdiction
In the second lecture, delivered in Adelaide last month, I focussed on the concept of jurisdiction which bears a close resemblance to the idea of "integrity" which, I acknowledged, is neither capable of precise definition nor of uncomplicated application. At the heart of each concept is a notion of significance. Each represents a recognition that not every error in administrative procedure can be the subject of judicial review. At the top of the spectrum of significance is the concept of "essentiality".
I expressed the view that the distinction between permissible and impermissible conduct, as manifest in the difference between jurisdictional and non-jurisdictional error, is real, indeed fundamental. I considered and rejected the criticisms that have sometimes been advanced about the utility of the concept of jurisdiction.
In the second lecture I explored how the idea of an integrity function assists in the understanding of the distinction between jurisdictional and non-jurisdictional errors of law, in determining when a fact is jurisdictional and in the rejection of a doctrine of deference in Australian administrative law.
A jurisdictional error of law or of fact, I suggested, raises issues of integrity. A non-jurisdictional error of law or of fact raises issues of competence and correctness.
Jurisdictional error of law can take different forms. The power may be misinterpreted by the decision-maker. A jurisdictional fact, sometimes called a "collateral fact", may be absent. A procedural defect may be such as to invalidate the decision, which requirement was once described as "mandatory" rather than "directory"[20]. A consideration that a decision-maker was obliged to take into account may have been ignored[21]. All of these tests serve an integrity function.
The distinction between jurisdictional and non-jurisdictional error of law remains a critical distinction in Australia. If there had ever been a prospect that the distinction would become attenuated, that prospect disappeared with the renewed emphasis given to the constitutional dimension of the relevant principles, as the High Court developed its 75(v) jurisprudence over recent years. During the course of this development, the High Court emphatically confirmed the traditional common law distinction in Craig.
In the second lecture I expressed the opinion that it is open for the High Court to determine that, subject to the particular statute, an error of law by an administrative decision-maker is usually jurisdictional. There is nothing inconsistent with performance of an integrity function if this proposition is understood as a rebuttable presumption of statutory interpretation rather than, as the House of Lords concluded, as effectively abolishing the distinction between jurisdictional and non-jurisdictional errors of law at common law. The idea of institutional integrity can guide the development of the law in respects such as this.
Apprehension is sometimes expressed that judges may too readily find facts to be "jurisdictional" and thereby intrude too far into executive decision-making. The concept of integrity can serve as a guide to ensure that any such tendency is kept in check. The focus of attention must always be fidelity to the purposes of the power and the maintenance of the public values to be served. The focus is not the quality of the outcome.
In the second lecture I emphasised that the determination of when a fact or event is jurisdictional is a principled process. It does not trespass on the merits side of the legality/merits distinction. The appellation "jurisdictional fact" is, in my opinion, a convenient way of expressing a conclusion, which is the result of a process of statutory interpretation. That process leads to the determination that, on the proper construction of the relevant power, a fact referred to must exist in fact, a test of objectivity, and that the Parliament intended that the absence or presence of the fact would invalidate action under the statute, a test of essentiality[22]. The position is similar to that expressed by the High Court with respect to procedural requirements in Project Blue Sky.
The language of essentiality, including words like "mandatory" and "jurisdictional", directs attention to matters that are appropriately described as issues of institutional integrity. It directs attention away from the quality of the actual outcome which, save in exceptional circumstances, is not relevant to the inquiry.
A good indication of the rigour with which Australian administrative law is restricted to an integrity function is the decisive rejection in the City of Enfield case of any idea of judicial deference[23]. Where intervention by a court is designed to ensure the institutional integrity of the decision-making process, it should be clear that "deference" is entirely inappropriate.
In this third lecture I propose to continue some of the themes developed in the first two lectures. The legality/merits dichotomy remains at the heart of Australian administrative law. That dichotomy has not been attenuated, as appears to have occurred in other common law jurisdictions, by the recognition of a doctrine of substantive legitimate expectations, by the application of a proportionality doctrine or by the recognition of material mistake of fact as a basis for review. The critical concept identifying the scope of review is the concept of "jurisdiction", both in terms of jurisdictional errors of law and jurisdictional facts. In this third lecture I propose to apply this analysis to the particular case of statutory provisions restricting access to the courts which are sometimes called ouster clauses but, in Australia, generally called privative clauses.
Privative Clauses
Privative clauses take different forms, though there are certain commonly occurring formulations. The number of cases in which a privative clause has been found not to protect a particular decision-making process is very large. The cases are not necessarily reconcilable. Nevertheless, certain broad themes emerge.
Perhaps the most important factor is the application of the well-known principle of the law of statutory interpretation that Parliament is presumed not to intend to reduce the jurisdiction of a court save to the extent that the legislation expressly so states or necessarily implies[24]. This is a specific application of a more general principle that Parliament did not intend to abrogate fundamental rights or freedoms, of which a citizen's rights of access to the courts is one. This principle and its application has recently been reaffirmed in the authoritative joint judgment of the High Court in the Plaintiff S157 case[25]. Privative clauses are, accordingly, construed strictly.
A number of alternative, but equivalent, formulations have been propounded to identify the level of strictness appropriate to the construction of provisions which may have the effect of invading common law rights: "express words of plaint intendment"[26], or "clear and unambiguous words"[27], or "unmistakeable or unambiguous"[28] or "irresistible clearness"[29] or "with a clearness which admits of no doubt"[30] or "clearly manifested by unambiguous language"[31]
General words contained in a statute are frequently read down in accordance with longstanding principles of statutory interpretation of which one can find an expression, in surprisingly contemporary reasoning, as long ago at 1560, when the Barons of the Court of the Exchequer said:
"... The sages of the law heretofore have construed statutes quite contrary to the letter in some appearance, and those statutes which comprehend all things in the letter, they have expounded to extend but to some things, and those which generally prohibit all people from doing such an act, they have interpreted to permit some people to do it, and those which include every person in the letter they have adjudged to reach some persons only, which expositions have always been founded upon the intent of the legislature which they have collected some times by considering the cause and necessity of making the Act, sometimes by comparing one part of the Act with another and sometimes by foreign circumstances. So that they have ever been guided by the intent of the Legislature, which they have always taken according to the necessity of the matter, and according to that which is consonant to reason and good discretion."[32]
No parliamentary drafter can be in any doubt that this approach will be applied to a privative clause. Subject to constitutional limitations, the drafter can maximise the efficacy of a provision, but only by words of 'irresistible clarity'. In view of the longevity of the principle of strict construction of privative clauses, the Courts are entitled to proceed on the basis that legislation has been drafted on the assumption that that will occur.
A privative clause generally consists of two component parts. The first identifies the decision or event to which the relevant restriction applies, e.g. in a frequently appearing formulation "a decision under this Act". The second is the specification of the restriction upon a court which may otherwise have supervisory jurisdiction, e.g. language of 'finality', "no appeal", no "certiorari or quashing" and no "calling into question", etc.
A strict construction applies to each of these two distinct aspects of a privative clause. Of particular significance for present purposes is the first, i.e. the identification of the decision or event which triggers the application of the restriction found in the second part of the clause. The High Court decision in Plaintiff S157 turned on this factor.
The Court held that s474 of the Migration Act 1958, which purported to protect from review "decisions ... made under this Act", referred only to decisions which were not infected by jurisdictional error. An administrative decision which involved jurisdictional error is "regarded, in law, as no decision at all"[33]. In part this conclusion was determined by the Constitutional context under consideration in Plaintiff S157, specifically the inability of the Parliament of the Commonwealth to pass legislation inconsistent with the conferral by s75(v) upon the High Court of jurisdiction to supervise officers of the Commonwealth by the constitutional writs and also by other express and implied provisions of Chapter III of the Constitution.
Although the Constitutional basis would have been a sufficient basis to read down the word "decision" so as not to apply to a decision affected by jurisdictional error, the High Court regarded the conclusion as flowing equally from the application of the law of statutory interpretation. In this regard the same result would apply to State legislation which, subject to future development of the Kable principle in the High Court, does not have the same constitutional overlay.
The conclusion that words identifying the decision or event which triggers the application of a privative clause does not encompass a decision or event infected by jurisdictional error, is a longstanding principle. Notwithstanding the convolutions that the Commonwealth went through in formulating and explaining s474 of the Migration Act, the result should not have been unexpected. As far back as volume 1 of the Commonwealth Law Reports an attempt to insulate decisions of the then New South Wales Arbitration Court by means of a privative clause protecting any "award, order or proceeding" in that court from any 'challenge, appeal, review, quashing or calling in question ... on any account whatsoever' was rejected on the basis that such provisions "have always been construed as not extending to cases in which a court with limited jurisdiction has exceeded its jurisdiction"[34].
The same approach has been applied consistently in a long line of authority, notably in the arbitration area, so that reference to "decision" was not generally protected from review for jurisdictional error[35].
A comprehensive privative clause will be effective to oust review for non-jurisdictional error of law[36]. Indeed, that is permitted under the Commonwealth Constitution[37]. However, for those matters which are essential, in the sense that they are jurisdictional, clear and express language (in the various formulations to which I have referred) is required before Parliament can be said, even in the case of State legislation not subject to a constitutional restriction, to have intended to restrict judicial review.
In one High Court judgment dealing with New South Wales legislation a contrast was drawn between the formulation "a decision under this Act" and an alternative formulation, not in fact used in the legislation then under consideration: "under or purporting to be under this Act"[38].
No doubt inspired by this implicit suggestion, the New South Wales Parliament did pass legislation protecting the Industrial Commission from any form of review with respect to "a decision or purported decision of the Commission". Subject to a constitutional challenge to the validity of this provision, which it was not necessary to decide[39], this extension has been held to be effective to protect from jurisdictional error[40]. However, there does have to be a "decision" of some character by the Commission, so that a challenge mounted as soon as proceedings are instituted can, the New South Wales Court of Appeal has held, be entertained[41].
Constitutional restrictions would prevent a privative clause in Commonwealth legislation extending to a 'purported decision'.
The Hickman Principle
Lord Diplock once said:
"Any judicial statements on matters of public law if made before 1950 are likely to be a misleading guide to what the law is today."[42]
Whatever may be the position in England that statement cannot apply to the observations of Dixon J in R v Hickman; Ex parte Fox & Clinton[43]. In Hickman Dixon J identified a core content of supervisory jurisdiction in the form of three provisos to the operation of a privative clause. Those three provisos were: the decision must be a bona fide attempt to exercise the power, that it must relate to the subject matter of the legislation and that it is reasonably capable of reference to the power given to the decision-maker.
In the Darling Casino case, Gaudron and Gummow JJ contrasted the position under Commonwealth law and State law in the following terms:
"... A privative clause in a valid State enactment may preclude review for errors of any kind. And if it does, the decision in question is entirely beyond review so long as it satisfies the Hickman principle."[44]
This passage appears to give the Hickman principle a binding force that does not derive from the Constitution. What is it that gives the matters identified force of this character?
The principle applied to the interpretation of privative clauses to the effect that a Parliament is presumed not to deny citizens access to the courts is a rebuttable presumption. In the context of State legislation it cannot be an irrebuttable presumption.
If I may be permitted the sin of self-quotation, I sought to explain the passage from the judgment of Gaudron and Gummow JJ in the following way:
"The final qualification with reference to the 'Hickman principle' is not, as I understand it, a suggestion that in some manner that 'principle' is immune from legislative overruling rather, once the intention appears from the legislative scheme that the privative clause has in fact extended to jurisdictional error, then a final principle of statutory construction reflected in the Hickman principle must be applied. This is because the strict construction, appropriate for all such clauses, is applied with particular stringency to those core matters. Indeed it may be so difficult to conceive of a form of words capable of satisfying a 'necessary intention' test, that express words are, as a practical matter, required."[45]
To say that a statutory provision must be strictly construed does not invoke a specific body of rules. There are degrees of strictness[46]. To similar effect are the observations of Gleeson CJ in Plaintiff S157 where his Honour referred to various adjectives which are often used to emphasise the significance of an alleged defect in jurisdiction. Gleeson CJ said:
"Unless adjectives such as 'palpable', 'incontrovertible', 'plain' or 'manifest' are used only for rhetorical effect, then in the context of review of decision-making, whether judicial or administrative, they convey an idea that there are degrees of strictness of scrutiny to which decisions may be subject."[47]
Notwithstanding the iconic status that has been given to the threefold Hickman principle it is not appropriate to treat it as if it were enshrined in legislation. However, that was what the Commonwealth effectively sought to do in Plaintiff S157. For some reason, which has never been fully explained so far as I am aware, the Commonwealth did not write the Hickman principle into the legislative scheme, but sought to have s474 of the Migration Act construed as if it had on the basis of expressions of intention in the Explanatory Memorandum and the Second Reading Speech.
The unsuccessful Commonwealth submission was to the effect that the privative clause should be understood to expand the jurisdiction of the relevant decision-maker subject only to the three Hickman provisos. This submission was rejected on the basis that the Hickman principle was simply a rule of construction and that, accordingly, "there can be no general rule as to the meaning or effect of privative clauses"[48].
There were a number of statements in previous High Court decisions suggesting that, subject to the Hickman provisos, the effect of a privative clause was to extend jurisdiction. However, those statements were often, but not always, expressed in the form that that was the "effect" of the application of the privative clause[49]. The Commonwealth's submission in Plaintiff S157 confused practical effect with legal effect. The earlier statements must now be understood as subject to the views expressed in the joint judgment in Plaintiff S157:
"... It is inaccurate to describe the outcome in a situation where the provisos are satisfied as an 'expansion' or 'extension' of the powers of the decision-makers in question."[50]
The joint judgment went on to state:
"To understand the three Hickman provisos as qualifying the powers of those who make privative clause decisions, rather than qualifying the protection which the privative clause affords, either assumes the Act on its true construction provides no other jurisdictional limitation on the relevant decision making or other power or it assumes that the repository of the power can decide the limits of its own jurisdiction ... [T]he first assumption is wrong. The alternative assumption would contravene Chapter III."[51]
Save to the extent that Chapter III impinges on State courts, the extent of which we will probably know when decisions are handed down in cases presently before the High Court, the second "assumption" does not apply to State legislation. The position with regard to State decisions, particularly statutory courts, may be different. Nevertheless, a process of statutory interpretation of a significant level of strictness is still required with respect at least, to the three Hickman provisos, to inviolable limitations and manifest defects.
Beyond the Hickman provisos
Notwithstanding the terms in which Dixon J stated and subsequently restated the Hickman principle[52], it appears now to be clear that it was not a comprehensive statement. It has been proposed that a fourth proviso is discernible in the authorities[53]. The fourth principle is that a decision must not "display a jurisdictional error on its face". However, that is the way in which a number of High Court authorities, including Plaintiff S157 itself, explains the third Hickman proviso, i.e. whether or not a decision is reasonably capable of reference to the power[54].
Of greater significance is the strong line of authority, affirmed in S157, that the process of statutory interpretation, required when a privative provision is included in legislation, may lead to the conclusion that certain kinds of limitations or requirements in the legislative scheme are, as variously expressed in the authorities, "essential", "indispensable", "imperative" or "inviolable". This was once advanced as a fourth proviso[55]. However, it has a different quality. It is not in fact a "proviso" at all.
The approach to interpretation first identified by Dixon J in Hickman is now authoritatively established in Australia. The task is one of reconciliation of conflicting provisions, namely the provision imposing some form of limitation on jurisdiction and the provision preventing judicial review[56]. This is the starting point of analysis. It is a very revealing starting point. There is an unexpressed major premise that there is no such thing as unlimited executive authority.
Contrary to our monarchical tradition, the executive arm of government cannot act merely on the basis of its power. This is a rule of law assumption and in Plaintiff S157 Gleeson CJ expressly invoked the rule of law as one of the factors required to be taken into account in the process of statutory construction[57]. I can see no reason why rule of law assumptions would not lead to the same result for a State Parliament. A parliament cannot confer unlimited power on the executive branch.
The joint judgment in Plaintiff S157 rejected the Commonwealth submission that the Parliament could confer an entirely open ended discretion[58]. However difficult it may be to identify the limits, there is a point where a parliament is delegating legislative power itself and that is not a proper exercise of legislative power[59]. This important constitutional principle is the starting point for a Hickman analysis.
The interpretation of a privative clause cannot proceed on the basis that it stands alone. It is not permissible to focus only on the fact that it states there will be no judicial review and to conclude, therefore that there can be no judicial review. A privative provision must be construed in its total context. As the joint judgment in Plaintiff S157 said:
"The process of construction for which (the Commonwealth) contends is not a process of construing the legislation as a whole. It is a process which places a construction on one provision, the privative clause, and asserts that all other provisions may be disregarded."[60]
Their Honours went on to refer to the formulation of Dixon J in Murray; ex parte Proctor that certain limitations may be "essential" and that a privative clause could not, therefore, apply to them. The same process is described in many other ways, perhaps most frequently in terms of an "inviolable limitation"[61]. It is convenient to use that term.
The characterisation of a limitation as "inviolable" represents the culmination of a process of statutory interpretation by means of the reconciliation between the privative provision and the relevant restriction or restraint. This process is an alternative to the three-fold proviso of the Hickman principle. It is not a proviso at all. Dixon J referred to the three-fold proviso as a "first step" and the "inviolable restriction" as a "second step"[62].
Terminology such as "imperative duties" and "inviolable limitations or restraints" clearly points to jurisdictional error[63]. It may be, however, that the conception of an "inviolable" limitation does not exhaust the field of jurisdictional error, particularly as that somewhat elusive concept has expanded over recent decades. Plainly not all limitations or restrictions in a statutory scheme can be described as "inviolable", nor all duties as "imperative".
Everything will turn on the specific legislative scheme under consideration in any case. Nevertheless, what appears to be involved is a heightened level of strictness of scrutiny with respect to matters which a consideration of the legislative scheme as a whole - namely the total context in which the privative provision which has to be construed appears - requires or imposes. There is an idea of essentiality underlying this process.
The position is the same as that which the High Court determined to be the case when discussing the terminology often applied to breach of procedural conditions in the Project Blue Sky case[64]:
"The classification of a statutory provision as mandatory or directory records a result which has been reached on other grounds. The classification is the end of the inquiry, not the beginning ... A better test for determining the issue of validity is to ask whether it was a purpose of the legislation that an act done in breach of the provisions should be invalid."
However, as the joint judgment went on to immediately note "a finding of purpose or no purpose in this context often reflects a contestable judgment"[65]. The same is true when determining when a "limitation" is "inviolable" or a duty "imperative".
I think we are all now used to the idea that the appellation "mandatory" and "directory" is a statement of a conclusion. So long as we recognise that fact, I find the terminology useful and would tentatively encourage its reintroduction into the administrative law lexicon, as does Professor Aronson and his co-authors[66]. It seems to me to be a useful shorthand.
It is not finally settled whether strict scrutiny of a privative clause leads to the conclusion that the clause does not protect every form of error that may be classified as jurisdictional for other purposes of administrative law. The process of reconciliation may involve a different form of analysis from that which is appropriate for determining categories of jurisdictional error.
Plaintiff S157 was concerned with an alleged breach of the requirement of procedural fairness. Such a breach had generally been assimilated with jurisdictional error in its original narrow sense[67].
Over recent years, at least in the case of administrative decision-makers, the concept of jurisdictional error has expanded. In Craig the idea of jurisdictional error was found to encompass virtually all of the matters listed in the Administrative Decisions Judicial Review Act, as picked up in the Queensland Judicial Review Act, except the catch all of "abuse of power"[68]. Furthermore in Yusuf the High Court indicated that even this list was not exhaustive[69].
Prior to S157, the majority in a five judge bench of the Full Federal Court had concluded that s474 of the Migration Act was effective with respect to the Craig type list of jurisdictional errors[70]. In the immediate wake of S157 the view was expressed that S157 was concerned only with the principle of natural justice and that the privative clause may still be effective for some other matters constituting jurisdictional error[71]. This was doubted and eventually overruled by subsequent Full Federal Court decisions[72]. The High Court has yet to determine this issue.
It may be of significance that in S157 the joint judgment did not simply use the terminology of "jurisdictional error", which would have been a clear reference back to the broad concept of jurisdictional error as it has developed with respect to administrative decision-makers over recent years. The formulation adopted at [76] was that the privative clause "must be read so as to refer to decisions which involve neither a failure to exercise jurisdiction nor an excess of the jurisdiction conferred by the Act". The only examples of relevant jurisdictional error given in that paragraph of the joint judgment were "imperative duties" and "inviolable limitations".
It is not clear that a particular defect capable of constituting jurisdictional error, e.g. a failure to take into account a particular relevant consideration, necessarily constitutes a "failure to exercise jurisdiction". The test suggests a broader inquiry than simply asking and answering the question has there been a failure to take into account a relevant consideration? There is a suggestion that an additional level of significance has entered the equation.
The particular formulation found in the critical passage in Plaintiff S157 appears in a number of other places including on at least one occasion in a passage where a distinction is drawn between that formulation and the full Craig type list of factors capable of constituting jurisdictional error. In Darling Casino, the joint judgment of Gaudron and Gummow JJ said at 633-634:
"... A clause which provides only that a decision may not be called into question in a court of law is construed so as not excluding review on the ground that the decision involved jurisdictional error, at least in the sense that it involved the refusal to exercise jurisdiction or that it exceeded the jurisdiction of the decision-maker." [Emphasis added]
The footnote attached to this sentence was:
"Note the wider use of the expression 'jurisdiction' referred to in Anisminic Limited v Foreign Compensation Commission (1969) 2 AC 147 at 171 per Lord Reid; see also at 195, per Lord Pearce."
The two references to Anisminic, which their Honours described as a "wider use of the expression 'jurisdiction'", suggested a Craig type list of factors.
To similar effect is the reasoning of Gaudron J in her dissenting judgment in Abebe v The Commonwealth[73] where her Honour stated at [107] that "jurisdictional error is not confined to situations which a tribunal either lacks jurisdiction or exceed its jurisdiction" and added at [108]: "Not every failure to have regard to relevant matters or to disregard relevant matters constitute jurisdictional error". Similarly in Plaintiff S157, Callinan J left open the question whether the remedies in s75(v) required a "grave or serious breach" in the case of procedural fairness[74].
In Vanmeld I raised the possibility that the extension of the concept of jurisdictional error may require a review of the proposition that privative clauses do not protect against jurisdictional error[75]. The significance of this review is even clearer in view of the High Court judgment in Applicant S20/2002 which appears to recognise irrational or illogical decisions as constituting jurisdictional error[76].
The reference in Plaintiff S157 to the process of construction leading to a conclusion that a particular limitation is "inviolable" or a particular defect is "manifest" or that a particular duty is "imperative"[77], does suggest that it is not sufficient to simply categorise an error under one of the Craig type factors. Rather, the overall process of interpretation must be conducted to determine the element of essentiality in the circumstances of the case. The process of statutory construction must lead to the conclusion that the relevant error has resulted in 'a failure to exercise jurisdiction or an excess of jurisdiction'.
The approach of the joint judgment in Plaintiff S157 was affirmed in a subsequent High Court decision[78]. There will of course be clear cases in which a detailed investigation of the legislative scheme is not required. However, it appears that in some cases categorisation of the nature of the error will be insufficient.
A process of reconciliation may still be required even if the privative provision extends to a "decision or purported decision"[79]. What is involved in such a case is two similarly forceful expressions of parliamentary intention which make the process of reconciliation more acute, but do not resolve it. In such a context it is likely that some, but not necessarily all, examples of jurisdictional error are protected. Some restrictions may still be "inviolable".
The Integrity Function
Strict scrutiny of legislation is appropriate in a range of situations, frequently with the effect that general words are read down[80]. The strict construction applied to privative clauses has the effect of ensuring that decisions are made for the purposes for which they were intended to be made and in accordance with the public values which the decision-making process was expected to obey. The idea of essentiality at the heart of the jurisdictional/non-jurisdictional distinction coincides to a substantial degree with what I have described as institutional integrity. As applied to privative clauses, this principle of statutory interpretation is a clear manifestation of the integrity function of administrative law.
This perspective is reinforced by the heightened strictness of scrutiny in the case of "inviolable limitations" or "imperative duties". The concept of institutional integrity may be of assistance in the task of statutory interpretation directed to identifying which restrictions and/or duties are of sufficient essentiality to answer the description of an "inviolable limitation" so that the privative provision is read down. There is an idea of 'essentiality' underlying this process. When matters involving institutional integrity arise with respect to a particular decision-making process, a conclusion that the fact or event is "essential" will more readily be drawn.
I have referred to the unstated major premise of the Hickman analysis that there is no such thing as unlimited executive authority. Powers are conferred for a purpose. The purpose, if nothing else does so, identifies limits. Once one attaches a purposive dimension to a power, then the assumption that the limits of that power are to be determined by a process of judicial review manifests a concern with institutional integrity.
The constitutionalisation of administrative law at a Commonwealth level proceeds apace. The joint judgment in Plaintiff S157 emphasises the integrity function of the High Court's role. Their Honours said at [104]:
"The reservation to this Court by the Constitution of the jurisdiction in all matters in which the named constitutional writs or an injunction are sought against an officer of the Commonwealth is a means of assuring to all people affected that officers of the Commonwealth obey the law and neither exceed nor neglect any jurisdiction which the law confers on them."
The jurisdiction of other courts at common law serves precisely the same purpose of maintaining institutional integrity.
As Kirby and Callinan JJ put in a joint judgment, referring to all Australian Parliaments:
"[69] ... where a discretion is conferred by statute, it must be exercised in accordance with the language by which it is conferred and to achieve the purposes for which the power has been granted. To talk of 'absolute' judicial discretions; at least where such discretions are conferred by an Australian statute, involves a contradiction in terms. Absolute discretions are a form of tyranny.
[70] All repositories of public power in Australia, certainly those exercising such power under laws made by an Australian legislature, are confined in the performance of their functions to achieving the objects for which they have been afforded such power. No Parliament of Australia could confer absolute powers on anyone ... [t]here are legal controls which it is the duty of the courts to uphold when their jurisdiction is invoked for that purpose."[81]
I conclude with an observation of Brennan J which reflects the concern with institutional integrity of administrative law:
"Judicial review is neither more nor less than the enforcement of the rule of law over executive action; it is the means by which executive action is prevented from exceeding the powers and functions assigned to the executive by law and the interests of the individual are protected accordingly."[82]
1 See Ray Huang, 1587, A Year of No Significance: The Ming Dynasty in Decline, Yale Uni P, Newhaven (1981) pp53-54.
2 See Charles O Hucker, The Censorial System of Ming China, Stanford Uni P, Stanford, California (1966) p3.
3 See Bruce Ackerman, "The New Separation of Powers" (2000) 113 Harvard Law Review 633 at 694-696.
4 See William P Alford, "Book Review" (2000) 113 Harvard Law Review 1677 at n 96.
5 The Integrity Branch of Government, accessible at www.courts.nsw.gov.au/speeches.
6 Murray Gleeson "Judicial Legitimacy" (2000) 20 Aust Bar Rev 4.
7 Waterford v The Commonwealth (1987) 163 CLR 54 at 77; Corporation of the City of Enfield v Development Assessment Commission (1999) 199 CLR 135 at [44].
8 See, e.g. Re Minister for Immigration and Multicultural Affairs; Ex parte Lam (2003) 77 ALJR 699 at [76]-[77]; Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002 (2003) 77 ALJR 1165 at [59].
9 Jurisdiction and Integrity accessible at www.courts.nsw.gov.au/speeches.
10 Re Grimshaw; Ex parte Australian Telephone and Phonogram Officer's Association (1986) 60 ALJR 588 at 594; Abebe supra at [21]; Re Refugee Tribunal; Ex parte Aala (2000) 204 CLR 82 at [21]-[23], [162], [164]-[166].
11 Ex parte Applicant S20/2002 supra at [170] and [161] per Kirby J.
12 R v North & East Devon Health Authority; Ex parte Coughlan [2001] QB 213; R v Secretary of State for Education & Employment; Ex parte Begbie [2001] 1 WLR 1115.
13 See Ex parte Lam supra 699 at [81]-[83], [111], [143], [148]. See also Applicant NAFF of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs [2004] HCA Trans 64, granting special leave from the decision in NAFF of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 127 FCR 259.
14 See e.g. R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532 at 547E-G.
15 On the response to proportionality in Australia see Aronson, Groves and Dyer Judicial Review of Administrative Action (3rd ed) Sydney 2004 at 343-348; Allars "Proportionality, Tradition and Constitutional Framework: Borrowing Foreign Legal Notions in Australian Public Law" in Doeker-Mach and Ziegert (eds) Law, Legal Culture and Politics in the Twenty First Century Franz Steiner Verlaz, Stuttgart 2004.
16 E v Secretary of State for the Home Department [2004] 1 WLR 1179 at [66], see also at [63]. This case involved an appeal limited to a question of law, but the reasoning extends to judicial review.
17 Puhlhofer Hillingdon London Borough Council [1986] AC 486 at 518; Minister for Immigration and Multicultural Affairs v Eshetu (1999) 197 CLR 611 at [41].
18 Azzopardi v Tasman UEB Industries Ltd (1985) 4 NSWLR 139 at 156. See also Bruce v Cole (1998) 45 NSWLR 163 at 189.
19 In Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S134/2002 (2003) 211 CLR 461 at [35]-[42].
20 See Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355.
21 See Minister for Aboriginal Affairs v Peko Wallsend Ltd (1986) 162 CLR 24; see also Craig v South Australia (1995) 184 CLR 163 at 177-179.
22 See Timbarra supra at [37]-[39]. See also SDAV v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 199 ALR 43 at [27].
23 Enfield supra at [39]-[48].
24 See, e.g. Magrath v Goldsbrough Mort & Co Ltd (1932) 47 CLR 121 at 134; Johnson v Director General of Social Welfare (Vic) (1936) 135 CLR 92 at 97.
25 Plaintiff S157/2002 v The Commonwealth (2003) 211 CLR 476 at [72].
26 Commissioner of Police v Tanos (1958) 98 CLR 383 at 396; R v Lieschke (1987) 162 CLR 4447 at 463; Ainsworth v Criminal Justice Commission (1992) 175 CLR 564 at 576.
27 Bropho v Western Australia (1991) 171 CLR 1 at 17.
28 Coco v R (1994) 179 CLR 427 at 436-438; Kartinyeri v The Commonwealth (1998) 195 CLR 337 at 381; Plaintiff S157 supra at [30].
29 Potter v Minahan (1908) 7 CLR 277 at 304.
30 McGrath v Goldsborough Mort & Co Ltd (1932) 47 CLR 121 at 128.
31 Al Kateb v Godwin [2004] HCA 37 at [19].
32 Straddling v Morgan (1560) 75 ER 305 at 315. See also Bowtell v Goldsborough Mort & Co Limited (1906) 3 CLR 444 at 456-457.
33 Plaintiff S157 at [76] quoting Minister for Immigration and Multicultural Affairs v Bhardwaj (2002) 209 CLR 597 at [5], [63] and [152].
34 Clancy v Butchers' Shop Employees Union (1904) 1 CLR 181 at 197.
35 See, e.g. Baxter v New South Wales Clickers' Association (1909) 10 CLR 114 esp at 131; Brown v Rezitis (1971) 27 CLR 157 esp at 172.
36 See e.g. Hockey v Yelland (1984) 157 CLR 124; Houssein v Under Secretary Department of Industrial Relations & Technology (1982) 148 CLR 88. See references in Aronson et al (3rd ed) at 837-838.
37 See S157 at [81].
38 See Darling Casino Limited v New South Wales Casino Control Authority (1997) 191 CLR 602 at 635. See also Anisminic Ltd v Foreign Compensation Commission [1969] 2 AC 147 at 171, 195.
39 Mitchforce Pty Ltd v Industrial Relations Commission of NSW (2003) 57 NSWLR 212 at [120]-[133].
40 See Mitchforce supra.
41 See Solution 6 Holdings Limited v Industrial Relations Commission of NSW [2004] NSWCA 200 applying Ex parte the Caterers & Restaurant Keepers Association (1903) 3 SR NSW 19; R v Australian Stevedoring Industry Board; Ex parte Melbourne Stevedoring Co Pty Ltd (1953) 88 CLR 100 at 118-119; Belmore Property Pty Ltd v Allen (1958) CLR 191 at 196.
42 See R v Inland Revenue Commissioners; Ex parte National Federation of Self Employed and Small Businesses Limited [1982] AC 617 at 640.
43 (1945) 70 CLR 598.
44 Darling Casino supra at 634.
45 See Vanmeld Pty Ltd v Fairfield City Council (1999) 46 NSWLR 78 at [137]; see also Mitchforce supra at [72].
46 See Vanmeld supra at [151].
47 Plaintiff S157 supra at [13].
48 Plaintiff S157 at [60]. See also at [35] per Gleeson CJ.
49 See, e.g. O'Toole v Charles David Pty Ltd (1991) 171 CLR 232 at 275; Darling Casino Limited supra at 631. See S157 at [133] per Callinan J.
50 Plaintiff S157 supra at [64].
51 S157 at [99]. Note that the reference to "the three Hickman provisos" reflects a submission put in those terms.
52 See also R v Murray; Ex Parte Proctor (1949) 77 CLR 387 at 398; Coal Miners Industrial Union of Workers of Western Australia v Amalgamated Collieries of Western Australia Limited (1960) 104 CLR 437 at 442-443.
53 See Aronson Dyer & Groves Judicial Review of Administrative Action (3rd ed) Sydney 2004 at p853 and cf Mitchforce supra at [77].
54 See R v Commonwealth Conciliation and Arbitration Commission; Ex parte Amalgamated Engineering (Australian Section) (1967) 118 CLR 219 at 252-253; R v Coldham; Ex parte Australian Workers Union (1983) 153 CLR 405 at 414; and O'Toole v Charles David supra at 285, 287; Mitchforce supra at [77]-[79], [210]-[213]; S157 supra at [57].
55 See, e.g. O'Toole v Charles David supra at 273.
56 See, e.g. S157 at [10], [17], [19], [58], [59] and [60].
57 See S157 at [31].
58 See at [101]-[102].
59 See Meyerson "Rethinking the constitutionality of delegated legislation" (2003) 11 Aust J of Admin Law 45.
60 See S157 supra at [65]. See also at [33] per Gleeson CJ.
61 See, for example, the various authorities and statements to this effect in S157 itself at [20], [21], [26], [65], [66], [70], [76], [157], [159], [160]. See also Charles David supra at 274; R v Metal Trade Employees Association; Ex parte Amalgamated Engineering Union Australian Section (1951) 82 CLR 208 at 248 referred to with approval in Coldham supra at 419, Darling Casino at 632. See also "final or definitive limitation": R v Central Reference Board; Ex parte Theiss (Repairs) Pty Ltd (1948) 77 CLR 123 and also the formulation "definitely ... not exercisable in other cases" R v Commonwealth Rent Controller; Ex parte National Mutual Life Association of Australia (1947) 75 CLR 361 at 369.
62 See Mitchforce supra at [86]-[87] referring to R v Murray; Ex parte Proctor at 399-400 and Plaintiff S157 at [20].
63 See Plaintiff S157 supra at [76].
64 Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 at [93].
65 Project Blue Sky supra at [91].
66 Aronson Dyer & Groves supra at 325, 854.
67 See the authorities collected in Vanmeld supra at par [160].
68 See Craig v South Australia (1995) 184 CLR 163 at 179.
69 See Minister for Immigration and Multicultural Affairs v Yusuf (2004) 206 CLR 323 at 351.
70 See NAAV v Minister for Immigration and Multicultural and Indigenous Affairs (2002) 123 FCR 298 esp [636].
71 See Lobo v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCA 144 per Giles J.
72 See WADK v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCAFC 48 at [26]; Applicant NALU of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCAFC 31 at [13]; Minister for Immigration and Multicultural and Indigenous Affairs v WAAG [2003] FCAFC 60 at [5]; Scargill v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 129 FCR 259 at [4]; SBBG v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 199 ALR 281 at [20]; SDAV v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 199 ALR 423 at [27], [30], [33], [34].
73 (1999) 197 CLR 520.
74 S157 at [159].
75 See Vanmeld supra at [134]; See also Aronson et al (3rd ed) at 867.
76 Applicant S20/2002 supra at [34], [37], [52] and [173] cf [9]. Note the qualification by Callinan J at [173] with respect to the 'sufficiency' of the error to justify relief. This conclusion may also require reconsideration in Azzopardi supra.
77 See S157 at [13], [18], [20], [21], [26], [56], [57], [65], [66], [70], [76], [157], [159], [160].
78 See Minister for Immigration and Multicultural and Indigenous Affairs v SGLB (2004) 78 ALJR 992 esp at [49]-[51].
79 See Mitchforce supra at [90]-[92].
80 See, e.g. Board of Fire Commissioners v Ardouin (1961) 109 CLR 105 at 116, Australian National Airlines v Newman (1987) 162 CLR 466 at 471, 476 and Puntoriero v Water Administration Ministerial Corp (1999) 199 CLR 575 at [33]-[37], [59]-[68], [113] (exception of liability provisions); Potter v Minahan supra at 304; Bropho supra at 17; Coco supra at 437; Al Kateb supra at [19]-[20]; S157 supra at 494; Daniels Corporation International Pty Ltd v ACCC (202) 213 CLR 543 at [11], [106]-[107] and [134]; Durham Holdings Pty Ltd v NSW (2001) 205 CLR 399 at 415; Hot Holdings Pty Ltd v Creasy (1995) 185 CLR 149 at 171 (invasion of common law rights). On reading down general words see cases collected in R v Young (1999) 46 NSWLR 681 at [22]-[32].
81 Gerlach v Clifton Bricks Pty Ltd (2002) 209 CLR 478 at [69]-[70].
82 Church of Scientology Inc v Woodward (1982) 154 CLR 25 at 70, quoted with approval in S157 at [31] by Gleeson CJ.
|
|
