THE AUSTRALIAN BAR ASSOCIATION CONFERENCE,
LONDON, ENGLAND

MONDAY, 5 JULY 1998

DEMOCRACY AND THE LAW

The Hon Justice M H McHugh AC

THE JUDICIAL METHOD 1

INTRODUCTION
Even a cursory reading of the reasons for judicial decisions given at any time during the last 400 years reveals that judges construct judgments in widely differing ways. The methodology employed by one judge is not necessarily the same as that employed by other judges deciding similar cases. But from at the least the early 17 th century, judges trained in the common law tradition have followed broadly similar patterns of reasoning in reaching their decisions. Notwithstanding the variations in judicial style - and they are very great - the similarity of these patterns has made it appropriate and natural for common lawyers to speak of judicial method as a field of special knowledge available for study and exposition. At all events, Sir Owen Dixon thought so. On the occasion of receiving the Henry E Howland Memorial Prize at Yale in September 1955, he entitled his Address - "Concerning Judicial Method" 2 . He used the occasion to argue that the common law judicial method was one of "high technique and strict logic" 3 .

The necessity for an identifiable judicial method arises from the nature of judicial power. When parties submit their disputes to a court for determination, they assume that the dispute will be decided in accordance with principles or rules that exist independently of the personal beliefs of the judge or judges who sit in that court 4 . They assume that, so far as the court can, it will ascertain the facts of the dispute and will determine the legal significance of those facts by reference to a rule or principle that is drawn from existing legal materials. To give effect to these assumptions, a judicial decision must be a reasoned decision arrived at by finding the relevant facts and by applying the relevant legal rule or principle.

Reasons are therefore essential to the common law judicial method. Moreover, those reasons must be so structured that they serve at least three purposes. First, they must permit the parties to see the extent to which their arguments have been understood and accepted. Lord MacMillan thought that the main object of a reasoned judgment "is not only to do but to seem to do justice." 5 Second, they must be reasons that clearly explain the reasoning process and thereby further judicial accountability 6 . Third, those reasons must declare and apply a principle or rule at a level of generality that transcends the facts of the case and enables other courts to decide other cases, identical in principle but not in detail, in the same way. It is necessary for a court to declare a rule or principle at this level of generality because, under the common law system of adjudication, courts not only resolve disputes but formulate rules and principles that can be used to decide comparable cases.

Formulating the governing principle or rule in a case sometimes involves a law-making function on the part of the judge, and it is often social, economic and even political 7 factors that stimulate or dictate the terms of this law-making function 8 . That judges at least change the law and not merely apply it was acknowledged by Sir Owen Dixon, Australia's greatest lawyer and a judge who believed in strict legalism, in his Address "Concerning Judicial Method" to which I have referred. He accepted that the common law changes to "meet the demands which changing conceptions of justice and convenience make." 9 However, he seems to have believed that the law was an autonomous discipline and that judicial method cannot legitimately be influenced by political, social or economic factors. How conceptions of justice or convenience could change without any input from social, economic or political factors was not explained. Sir Owen Dixon argued that to abandon the common law method of "high technique and strict logic" would place the common law system of justice at risk of destruction. He concluded his Address with the following passionate peroration:

"The purpose of this extended and technical discussion is to show by example that it is an error, if it is believed that the technique of the common law cannot meet the demands which changing conceptions of justice and convenience make. The demands made in the name of justice must not be arbitrary or fanciful. They must proceed, not from political or sociological propensities, but from deeper, more ordered, more philosophical and perhaps more enduring conceptions of justice. Impatience at the pace with which legal developments proceed must be restrained because of graver issues. For if the alternative to the judicial administration of the law according to a received technique and by the use of the logical faculties is the abrupt change of conceptions according to personal standards or theories of justice and convenience which the judge sets up, then the Anglo-American system would seem to be placed at risk. The better judges would be set adrift with neither moorings nor chart. The courts would come to exercise an unregulated authority over the fate of men and their affairs which would leave our system undistinguishable from the system which we least admire."

No one can read that Address without being struck by Sir Owen Dixon's powerful intelligence and absolute mastery of the common law. Yet the whole argument of that Address "sounds nowadays like a voice from another world" to use a statement of Lord Cross of Chelsea in Director of Public Prosecutions v Boardman 10 . Perhaps the subject matter of the Address - demonstrating how the common law method of "high technique and strict logic" could outflank the rule that payment of a lesser sum does not discharge a debt - accentuates the otherworldliness of the Address. Even so, Sir Owen Dixon plainly believed that the common law judicial method was not compatible with a judge altering a legal rule or principle because it no longer meets or conforms with existing social needs. Change in legal principles had to be generated from existing doctrine. A principle could be extended, modified or subsumed in a wider principle only if a logical analysis of existing doctrine showed that a particular branch of the law required rationalization 11 . Sir Owen did not object, of course, to the application of an existing principle to new factual situations if analogy to existing examples or the rationale or logical development of the principle required it. But to him the modification, amplification or rejection of a principle on the ground that today's social conditions required it was unthinkable.

It is a near certainty that Sir Owen Dixon's perception of the common law judicial method reflected the view of most common law judges who held office in the period 1820-1970. But my clear impression from reading current judgments and talking to judges is that the majority of common law judges today accept that social conditions can generate changes in legal principles and rules and that the judicial method is not so restrictive as Sir Owen's analysis indicates. Most common law judges today accept that it is an inevitable incident of the judicial process that, from time to time, a judge has to make, and not merely declare or restate, legal rules and principles. In the frequently cited words of Lord Reid speaking extra-judicially in 1972 12 :

"There was a time when it was thought almost indecent to suggest that judges make law - they only declare it. Those with a taste for fairy tales seem to have thought that in some Aladdin's cave there is hidden the Common Law in all its splendour and that on a judge's appointment there descends on him [or her] knowledge of the magic words Open Sesame. Bad decisions are given when the judge has muddled the pass word and the wrong door opens. But we do not believe in fairy tales any more."

THE ATTITUDE OF HIGH COURT JUSTICES TO THEIR LAW-MAKING FUNCTION

At the time I delivered a paper on the judiciary's law-making function in 1987 13 , few High Court justices had acknowledged the law-making function of courts. The judgment of Sir Garfield Barwick in MLC Assurance Co Ltd v Evatt 14 was a notable exception. However, during and since the time of the Mason Court (as some have described it), justices of the High Court have acknowledged the Court's law-making role and made numerous observations concerning the parameters within which the law-making function of the judiciary operates. As Chief Justice Doyle has noted, "[i]t is not that previously the High Court did not make law but now it does. It is simply that during [Sir Anthony Mason's] time as Chief Justice that role was more openly avowed and ... more frequently exercised" 15 .

A recent statement of the judiciary's role in the development of the common law was made by Justice Gummow in Esanda Finance Corporation Ltd v Peat Marwick Hungerfords 16 . His Honour noted that the very essence of accepted judicial method is the combined purposes of developing the law, maintaining its continuity and preserving its coherence. His Honour continued 17 :

"Accepted means of effecting those purposes include (i) extending the application of accepted principles to new cases, (ii) reasoning from the more fundamental of settled legal principles to new conclusions, and (iii) subsuming unforseen instances under a category which, in reason, is not closed against them."

Justice Gummow made similar comments in Wik Peoples v Queensland 18 . After referring to a statement by Lord Radcliffe in 1956 to the effect that the common law is a body of law which develops over time, his Honour said 19 :

"Here is a broad vision of gradual change by judicial decision, expressive of improvement by consensus, and of continuity rather than rupture .... Movement also may plainly be perceptible, and there may be an explicit change of direction, where, in the perception of appellate courts, a previously understood principle of the common law has become ill adapted to modern circumstances .... Again, it may emerge that the rationale of a particular cause of action is the product of a procedural fiction (eg, an implied promise to pay) which should no longer be supported after the demise of old forms of action 20 . In those cases, the perceived reason for change stems from alterations in the legal system itself .... More simply, upon analysis it may appear that a particular principle (eg, as to the irrecoverability of payments made under a mistake of law) rests upon a dubious foundation in the case law which has not been accepted in this court. 21 "

His Honour then referred to Mabo [No.2] as being a case where non-recognition of native title "rested upon past assumptions of historical fact, now shown then to have been false" 22 .

Today, the law-making function of the court is accepted by the overwhelming majority of lawyers. I have examined the historical path toward this recognition elsewhere 23 , and it is unnecessary to re-trace it here.

PREVALENCE OF LAW-MAKING

To recognise that judges make law, however, is not to imply that most judges spend their time making law. Lord Devlin once expressed 24 the view that 90 per cent of the time of English judges was spent in the "disinterested application of known law" 25 . That is almost certainly true of Australian judges. In the majority of cases that come before the courts, judges have no discretion to make law. This is especially so in trial courts where the judges are bound by precedent and have room to make law only in purely novel cases 26 . However, the scope for judicial law-making increases as cases go up the judicial hierarchy. In the intermediate appellate courts, although much of the work concerns the application of established rules, the determination of questions of evidence and procedure and the evaluation of evidence, there is more scope for judicial law-making. But it is confined by the binding precedents of the ultimate appellate court in the system. The creative role of intermediate appellate courts in Australia is therefore limited to novel cases and the re-examination of their own precedents 27 .

It is the apex of the judicial hierarchy - in Australia, the High Court - which has the most apparent and significant law-making role. Because of the appellate structure and the special leave procedures required to bring a matter before the High Court, most cases which come to the Court on appeal are generally hard cases which can be plausibly decided either way. Because of its position at the top of the judicial hierarchy, the High Court is forced to make law in many cases.

THE NEED FOR JUDICIAL LAW-MAKING

Why do courts need to make law? As I said in my 1987 paper 28 , drawing on the analysis of Arthur T Von Mehren 29 , there are four types of (sometimes overlapping) situations which constantly recur and place pressure on the courts to make law.

1. The relevant interests are accommodated by existing precepts, but the court disagrees on policy grounds

The first situation is where the court recognises that the interests pressing for recognition in a particular case are authoritatively accommodated by the existing body of precepts, but disagrees, on policy grounds, with the consequences of this accommodation. An example is the High Court's decision in Trident General Insurance Co Ltd v McNiece Bros Pty Ltd 30 where the majority of the Court modified the doctrine of privity of contract to enable a third party to enforce an insurance contract. The majority decision was inconsistent with a strict or logical application of the doctrine of privity. Established authority, including decisions of the High Court, had formulated the doctrine in a way that covered all contracts. However, the majority modified the doctrine on the ground that to deny the third party a right of recovery would be unjust in circumstances where the insurer had accepted premiums on the basis that third parties were covered by the policy. A further example is the Court's recent decision in Northern Sandblasting Pty Ltd v Harris 31 where the Court (in separate judgments) expressly or implicitly disapproved of the rule in Cavalier v Pope 32 that a landlord owes no common law duty to a stranger to the contract of tenancy. That rule was a relic of the doctrine of privity of contract and is inconsistent with the modern doctrine of liability for negligence 33 .

2. The relevant interests are historically accommodated by precepts, but political and ethical ideas have changed

The second situation that places pressure on courts to make law is where the interests now pressing for recognition are covered by the existing body of precepts, but the court believes that social and ethical ideas about their proper accommodation have changed since these precepts were formulated. There is no better example than the decision of the High Court in Mabo v Queensland [No 2] 34 where a majority of the Court held that indigenous title to land survived the colonisation of Australia by the Crown in 1788, notwithstanding that the Privy Council had not accepted that view a century earlier in Cooper v Stuart 35 . The majority in Mabo held that the extinguishment of indigenous rights and interests, based upon the doctrine of terra nullius, "was justified by a policy which has no place in the contemporary law of this country." 36

A further example of this situation is the High Court's decision in R v L 37 . The Court held that, despite earlier English authority to the contrary, the presumption that, upon marriage, a wife gave to her husband irrevocable consent to sexual intercourse, could not be reconciled with modern attitudes and, thus, could not form part of the common law in Australia.

3. Interests similar to those involved in the case are accommodated by existing principles

The third situation that places pressure on courts to make law is where the interests recognised by the existing body of principles are similar to, but not identical with, the interests now before the Court. Examples in this category include David Securities Pty Ltd v Commonwealth Bank of Australia 38 , in which case the Court extended the principles of restitution beyond cases of mistake of fact to circumstances where there has been a mistake of law, and Bryan v Maloney 39 where the Court extended a house builder's liability for economic loss arising from inadequate footings to subsequent purchasers of the house.

4. No similar interests are accommodated by existing principles

The fourth situation that places pressure on courts to make law is where the interests before the court cannot fairly be considered similar to the interests already recognised by the existing body of principles. However, the formulation of those principles may be sufficiently flexible or abstract to allow an interest pressing for recognition to be accommodated. A recent example of this kind of case is the High Court's decision in Dietrich v The Queen 40 where the Court applied the fair trial principle and declared that a trial court has power to stay criminal proceedings where lack of legal representation would lead to an unfair trial. Many would regard the recognition of native title in Mabo [No 2] as another example of this situation. My own view is that Mabo [No.2] was merely a belated recognition in this country of an interest long recognised by the common law of England and other countries.

5. Rationalisation of existing principles

To these four categories I would add a fifth (although it is probably already encompassed to a large degree by the other categories), and that is the "rationalisation of general principle with a view to bringing more unity and symmetry to the general law" 41 . The most recent example of this kind of law-making is the decision in Northern Sandblasting Pty Ltd v Harris 42 where the Court integrated the liability of landlords into the general law of negligence. The Court did the same in Burnie Port Authority v General Jones Pty Ltd 43 by incorporating the well established rule in Rylands v Fletcher into the general law of negligence and in Australian Safeway Stores Pty Ltd v Zaluzna 44 by integrating occupier's liability for negligent acts into the general law of negligence. The movement towards overarching principles has also been favoured in relation to estoppel 45 , unconscionability 46 and unjust enrichment 47 .

In these cases, there were established principles regulating liability. However, the Court subsumed these older principles under more general principles of liability.

Another example of the rationalisation of existing principles is Hatzimanolis v ANI Corporation Ltd 48 where the High Court reformulated the principles for determining when an injury, occurring in the course of an interval between work, occurs in the course of employment. After concluding that the current principles had led to findings which, on the proper application of those principles were fictitious, the Court said 49 :

"Consequently, the rational development of this area of law requires a reformulation of the principles which determine whether an injury occurring between periods of actual work is within the course of the employment so that their application will accord with the current conception of the course of employment as demonstrated by the recent cases".

The common denominator in all five categories is social change 50 . When legal rules and principles are no longer efficient or do not meet social needs, they must be reviewed and sometimes revised or extended. The law is a social instrument - a means, not an end. It changes as society changes. As Justice Cardozo recognised, law may well be influenced by logic, historical development, or tradition, but "[t]he end which the law serves will dominate them all." 51 In Justice Cardozo's view, "[n]ot the origin, but the goal is the main thing." 52

However, as I have maintained elsewhere, much evidence, analysis and argument is required before a court can determine whether changed social conditions justify the development of the law by the court 53 . An assessment of the requisite level of information is in the hands of each judge, and views will inevitably differ as to whether that threshold has been crossed in individual cases 54 . The existence of this level of discretion is one of the reasons why it is essential for the judiciary to identify the values which are being recognised in individual cases and to explain candidly why those values justify a development in the law.

Notwithstanding the sometimes formidable difficulties in determining whether changed social conditions justify the development of the law, the judiciary has continually extended and modified the common law for 800 years. In 1981, Sir Ninian Stephen declared 55 :

"Development has always been the life-blood of the common law and the more swiftly our society changes, the greater the need for developments in the law to keep pace with those changing mores."

Lord Radcliffe pointed out in 1964 that there "was never a more sterile controversy than that upon the question whether a judge makes law. Of course he does. How can he help it?" 56 . Whether a judge is interpreting the Constitution or an ordinary statute, or resolving a common law or equity dispute, he or she is forced to make choices between two or more competing arguments 57 . Where no precedent binds, the judge's decision - whatever it is - becomes a precedent which creates a new proposition of law and which may provide a foundation for further development of the law 58 . Whether a development is the result of an incremental change in the law or the sudden departure from a previously accepted direction in the law, the courts are exercising their law-making function in response to a change in the social context.

Moreover, the courts have little alternative to maintaining their law-making function. As Justice Brennan pointed out in Gala v Preston 59 :

"The purpose of judicial development of legal principle is to keep the law in good repair as an instrument of resolving disputes according to justice as it is understood in contemporary society, subject to statute .... In a society where values change and where the relationships affected by law become increasingly complex, judicial development of the law is a duty of the courts - more especially when legislative law reform languishes."

Modern legislatures simply do not have the time to continually monitor and amend legal rules, particularly those rules dealing with the relationships between private citizens. Furthermore, legislators cannot foresee all the circumstances which may call for the application of a rule. Because that is so, they cannot formulate rules that are so exact and yet so comprehensive that they plainly cover every dispute that comes before the courts. If judges were to become reluctant to adapt the law to a changing society, public confidence in the rule of law would be seriously impaired because large areas of law would be out of touch with the needs of society. If the law did not remain the appropriate mechanism by which citizens regulate their affairs and resolve their disputes, it would become largely irrelevant.

THE JUDICIAL DILEMMA

The spectre of the law's potential irrelevancy creates a dilemma for the Australian judiciary at the present time. If law is to remain relevant, social change requires the courts, and the High Court in particular, to develop the law to meet these changes. Yet if the courts develop the law, they are criticised for exceeding their role and usurping the role of the legislature.

My impression is that most Australian politicians and members of the public think that judges should only interpret the law. Paradoxically, they want results which accord with what Lord Devlin called "the aequum et bonum" 60 . If the community or an important section of it like a result, it will applaud a decision, no matter how far the law has been stretched to achieve that result. If it does not like the result, it will criticise the decision, no matter how impeccable the legal reasoning behind it. As more and more novel situations come before the courts, judges are forced to develop the law to an extent that was unthinkable even 30 years ago. Because many of these cases give rise to controversial issues, the development of the law has brought the courts under increasing political, media and academic criticism 61 .

Admittedly, the public misapprehension of the law-making role of the courts has been generated largely by the judiciary maintaining the declaratory theory of judgment. That is to say, the theory that judges merely find and do not make any part of the law. The lowering of the shield by the judiciary - the judiciary's open acknowledgment of its creative role - has exposed it to criticism from an unsuspecting and unprepared public, a criticism fuelled, as Justice Sackville has pointed out, by greater public scepticism and cynicism towards established institutions 62 . The present climate of public opinion makes it incumbent on the judiciary, when giving judgment in cases which break new legal ground, to explain clearly how and why the change has occurred 63 . As Sir Daryl Dawson speaking extra-judicially, has observed 64 :

"A judicial decision is not merely the resolution of a case in favour of one party or another, but a decision based upon articulated reasons which are consistent with established principle. If an apparently applicable principle is not applied, the judge ought to explain how and why something else is to be preferred."

This articulation of reasons also serves a separate, but related function. Although unelected, the judiciary remains accountable to the community for its decisions. The provision of written reasons facilitates that accountability. It provides practising lawyers, academic lawyers, the media and other interested members of the public with an opportunity to engage in informed debate. Thus, as Justice Davies has pointed out 65 :

"Informed criticism should assist in determining whether or not a decision and the reasons provided for it reflect community values. If it becomes clear that those values have not been reflected, criticism should lead to the overruling of the decision. In this sense, judges can become accountable to the public through their reasons for decision."
THE LIMITS OF THE LAW-MAKING FUNCTION

The law-making function of the judiciary is not unfettered. There is a real difference between judge-made law and the creation of law by a popularly elected legislature. Any encroachment into the legislative sphere is constitutionally impermissible and democratically unpalatable. Although this ideal is easy to state, the dividing line is not so easy to draw. As Justice Stephen Breyer of the Supreme Court of the United States has been reported as saying, a "brightline" between permissible and impermissible judicial creativity does not exist 66 . That said, there is no doubt that "judicial law-making is of a different nature and order from legislative enactment. It occurs in different circumstances, in response to different stimuli, and is subject to restrictions that do not constrain the legislature's freedom of action" 67 . Rather than a usurpation of the legislative role, the judiciary's law-making function may be seen as a complementary dimension to governmental law-making as a whole.

First, courts only make law in the context of determining a legal dispute which is initiated by the parties to the dispute. The courts resolve issues which litigants define. In the words of Chief Justice Doyle, "the law making [of courts] is opportunistic" 68 .

Second, the natural inclination of most judges is to place a premium on certainty and predictability which are important characteristics of a stable legal system. Stability instils confidence in the institution of the judiciary and in the law. Because judicial law-making operates retrospectively, the rule of law would be seriously threatened if law-making was a routine function of courts.

Third, in most cases, judges make law only when changes in society require the law to be developed to meet the consequences of those changes 69 . Changing social conditions are central to the development of the law by the judiciary. Unlike political parties, judges have no agenda to be implemented. Moreover, judges know that it is no easy task to identify and measure social change or to assess the effect of an alteration of the law which responds to that change. The courts are largely dependent on litigants, interveners and amici curiae to provide relevant information for the determination of the issues in dispute. Additionally, it is the appellate courts that are the principal judicial law-makers and their procedures are not geared towards the elaboration of relevant non-legal material. Awareness of these difficulties naturally makes the judiciary develop the law cautiously and only when it is clear that the needs of society demand it or the state of the law requires that the existing rules or principles be rationalised.

It was awareness of these difficulties that led me in Burnie Port Authority v General Jones Pty Ltd 70 to caution against a too ready willingness to depart from the settled rules of common law. After referring to Justice Mason's comments in State Government Insurance Commission (SA) v Trigwell 71 to the effect that the Court is neither a legislature nor a law reform agency, I said:

"No doubt courts in general, and this Court in particular, are more ready to alter the rules of the common law and equity than they were in 1979 when Trigwell was decided. But the law-making function of a court is different from that of a legislature. It is merely an incident of the duty to adjudicate disputes between litigants. It arises from the necessity to do justice between the parties and those who stand in similar situations. A judge-made rule is legitimate only when it can be effectively integrated into the mass of principles, rules and standards which constitute the common law and equity. A rule which will not 'fit' into the general body of the established law cannot be the subject of judge-made law." 72

JUDICIAL METHODOLOGY

However, it is methodology that really distinguishes judicial law-making from that of law-making by the legislature or the executive government when it implements subordinate legislation. The difference in approach is so great that judicial law-making belongs to a separate genus of law-making, notwithstanding that most Australian courts no longer act as if the law is an autonomous discipline.

Dr Horrigan has observed that "[u]ntil the early 1980's, many Australian lawyers approached the study of law and legal problems in practice with a preconceived bias towards the idea that law was a completely logical, self-contained set of clear rules of determinate meaning which supplied one right answer to every legal problem by a logical process of induction and deduction from case law and legislation (including constitutions)." 73 This mechanical approach assumed that the rule that decides a case is the last link in a logical legal chain made up of precedents, legal principles, concepts, rules or other authoritative legal texts or writings. It required law-making to have a strictly legal pedigree, a line of reasoning colorfully illustrated by the statement of Bagnall J in Cowcher v Cowcher 74 "that equity is [not] past childbearing; simply that its progeny must be legitimate - by precedent out of principle."

Although induction and deduction from established precedents and principles are involved in much judicial law-making - in probably all cases where the courts create a new or extend an old rule - induction and deduction only explain part of the law-making function of the judicial process. That function is not mechanical. In novel cases, the precedents will usually yield competing premises. The competing arguments, drawn from those premises, may be supported by the processes of deduction and induction. The choice between those arguments can only be made, however, by applying a theory, principle or value, representing some policy, which persuades the reader that one choice is better than other. Logic may take a judge a long way in determining a novel case. But usually it cannot take him or her all the way. Lord Buckmaster's dissenting speech in Donoghue v Stevenson 75 was as much a logical deduction from the precedents as was Lord Atkin's majority speech. Lord Atkin said that there must be some general principle of negligence that explained the cases. But why? Lord Buckmaster's dissent showed that the cases were logically explicable without a general principle of negligence. Lord Atkin's decision was based on what he thought justice required, not logical compulsion. It was his sense of justice and not logic that gave rise to the general principle that has dominated the law of negligence since 1932.

Neither can reasoning by analogy, a form of logical reasoning much used in legal argument, take a judge all the way in a novel case. A judge who reasons by analogy has to have an organising theory that explains why the facts of the instant case are similar or dissimilar to the facts of the precedent cases. The choice of that theory will seldom depend on induction or deduction; more often as not it will depend on some policy that promotes or protects some goal, interest or value external to the precedents.

Legal formalism and the jurisprudence of Professor Ronald Dworkin reject the use of theories, principles or values that come from outside the legal system. Under those doctrines, cases can only be decided by the analysis and logical application of existing legal materials - precedents, concepts, doctrines, principles, rules, statutes and authoritative writings. But few lawyers today doubt the truth of the statement of Oliver Wendell Holmes jnr that the "life of the law has not been logic: it has been experience." 76 Lord Reid said that, when a judge has "some freedom to go in one or another direction", he or she should have regard to "common sense, legal principle and public policy in that order." 77 This was a candid admission by one of the greatest judges of this century that legal materials will carry the judge only part of the way in the novel case. Commonsense is a value which reflects the community's current thinking on a subject. It has nothing to do with legal doctrine. Moreover, commonsense views change as the community's knowledge and understanding of a subject change.

When judges extend the scope of a legal rule, change its content, or reject it altogether, more often than not, they rely, sometimes unconsciously, on values or practical considerations outside the legal system. In Victoria v The Commonwealth 78 , Windeyer J said that the Engineers' Case 79 "looked at as an event in legal and constitutional history, was a consequence of developments that had occurred outside the law courts as well as a cause of further development there." That was "because in 1920 the Constitution was read in a new light, a light reflected from events that had, over twenty years, led to a growing realization that Australians were now one people and Australia one country and that national laws might meet national needs." 80

Numerous illustrations can be given of the effect of external influences on the content of legal rules and principles. Mabo [No 2] 81 is the classic Australian example. But the influence of values outside the legal system occurs in all areas of the law. In Bankstown Foundry Pty Ltd v Braistina 82 , Mason, Wilson and Dawson JJ accepted that the content of reasonable care is influenced by "changing ideas of justice and increasing concern with safety in the community." In the same case, Brennan and Deane JJ listed among the reasons why higher standards of care were now required of employers "the contemporary tendency to reject the discounting of any real risk of injury to an employee in the assessment of what is reasonable in the pursuit by an employer of pecuniary profit." 83 In Trident General Insurance Co Ltd v McNeice Bros Pty Ltd 84 Mason CJ and Wilson J held that a third party beneficiary should be able to sue on a policy of insurance. One of the reasons they gave was the likelihood of the third party relying on the expected benefit. That reliance was "so tangible that the common law rule should be shaped with that likelihood in mind." In R v L 85 where the accused had been charged with raping his wife, Mason CJ, Deane and Toohey JJ were even prepared to use external values to extend the scope of the criminal law. Their Honours said:

"[E]ven if the respondent could, by reference to compelling early authority, support the proposition that ... by reason of marriage there is an irrevocable consent to sexual intercourse, this Court would be justified in refusing to accept a notion that is so out of keeping with the view society now takes of the relationship between the parties to a marriage."

These illustrations show that neither logic nor reasoning by analogy from decided cases is the only factor in judicial law-making. The process is much more pragmatic. Values and the practical working of legal rules have as much a part to play in creating, extending or modifying a legal rule as logic does. No doubt many of the values invoked to develop or modify the law derive from the legal system itself. Values such as freedom of the individual, equality before the law, certainty and predictability, unconscionability, good faith, reasonableness and, in recent years, fairness 86 permeate the legal system. They shape judicial law-making. But these legal values do not exhaust the materials which judges use when making law. Extrinsic values and practical experience derived from democracy, economics, science, social and political forces, public morality and contemporary conceptions of justice are often relevant factors in shaping the development of the law. More than 60 years ago, Justice Stone of the United States Supreme Court pointed out 87 :

"We are coming to realize more completely that law is not an end, but a means to an end ... [and] that within the limits lying between the command of statutes on the one hand and the restraint of precedents and doctrines, by common consent regarded as binding, on the other, the judge has liberty of choice of the rule which he applies, and that his choice will rightly depend upon the relative weights of the social and economic advantages which will finally turn the scales of judgment in favour of one rule rather than another.

It is true that statements can be found in Australian judgments and extra judicial writings that might suggest that only values, and enduring values at that, can be taken into account in developing or modifying the law. Thus, Justice Brennan observed in Mabo [No.2] that it is to "contemporary values" 88 which judges refer when developing the law. In Dietrich , his Honour spoke of "the relatively permanent values of the Australian community" 89 . Sir Anthony Mason has spoken of "enduring values" as opposed to "transient community attitudes" 90 . But these remarks have to be read in their context. I do not think that their Honours intended these statements to be exhaustive of the extra-legal considerations that can generate judicial law-making. Tort, contract and administrative law, for example, are frequently developed and modified by practical considerations rather than values 91 . In Esanda Finance Corporation Ltd v Peat Marwick Hungerfords 92 , for example, in rejecting a claim that auditors should be liable to third parties whom they could reasonably foresee might rely on the auditor's accounts, I considered such factors as the problematic nature of such claims, the withdrawal of insurance protection for auditors, the potential decline in the availability of audit services, the capacity of plaintiffs in the likely class to protect themselves and the effect of long trials on the justice system.

Moreover, if judicial law-making was dependent on the ascertainment of permanent or enduring community values outside the legal system, present day Australian judges would make little law. As Justice Kirby 93 has pointed out, Australia is a multi-cultural society, which is constantly undergoing rapid social and economic change. It is extremely difficult for present day judges to know what are the permanent or enduring values in contemporary Australian society. Bedrock values of the nation have been overturned in recent years. Writing in 1992, Paul Kelly asserted 94 :

"Australia was founded on: faith in government authority; belief in egalitarianism; a method of judicial determination in centralised wage fixation; protection of its industry and its jobs; dependence upon a great power (first Britain, then America), for its security and its finance; and, above all, hostility to its geographical location, exhibited in fear of external domination and internal contamination from the peoples of Asia/Pacific. Its bedrock ideology was protection; its solution, a Fortress Australia, guaranteed as part of an impregnable Empire spanning the globe. This framework - introspective, defensive, dependent - is undergoing an irresistible demolition."

It is still continuing.

For this reason, I suspect that in the future extra-legal values will have only a small role in judicial law-making. It is likely that what will drive judicial law-making in the future are the legal values to which I have referred and practical considerations based on a cost/benefit analysis. Indeed, I suspect that what has been called community values has usually been a reference to values such as freedom, equality before the law, good faith and reasonableness which already inhere in the legal system. Whether that be true or not, however, if the extra-legal values of the community are clear and the issues are "relatively discrete and manageable" 95 , there is no reason to disregard them as legitimate sources of the judicial law-making function 96 .

However, judges have no authority to change the law merely because they find the precedents of earlier generations unpalatable. Judges are not in the business of repudiating the past, although sometimes, as Mabo [No 2] 97 and R v L 98 show, they must repudiate rules developed in earlier times when those rules have become out of touch with contemporary notions of justice. In Dietrich v The Queen 99 , Justice Brennan said that the Court is under a "responsibility for keeping the common law consonant with contemporary values" 100 . However, he pointed out that this does not mean that "the courts have a general power to mould society and its institutions according to judicial perceptions of what is conducive to the attainment of those values" 101 . His Honour continued 102 :

"Changes in the common law are not made whenever a judge thinks a change desirable. There must be constraints on the exercise of the power, or else the courts would cross 'the Rubicon that divides the judicial and the legislative powers' .... In ultimate courts of appeal, the chief constraints are found in the traditional methods of judicial reasoning which ensure that judicial developments 103 remain consonant not only with contemporary values but also with what I described in Mabo v Queensland [No.2] 104 as 'the skeleton of principle which gives the body of our law its shape and internal consistency' .... In practical terms, the courts are aware that rejection or discounting of the authority of precedent not only disturbs the law established by a particular precedent but infuses some uncertainty into the general body of the common law. The tension between legal development and legal certainty is continuous and it has to be resolved from case to case by a prudence derived from experience and governed by judicial methods of reasoning."

As this passage indicates, judicial law-making is not legitimate unless it can be connected to existing doctrines. For that reason, as Lord Diplock once pointed out 105 , common law judges could not "have created the welfare State". In Breen v Williams 106 , the High Court refused to recognise a patient's entitlement to inspect or obtain his or her medical records. Justice Gaudron and I emphasised that judicial law-making must be tied to existing doctrines. We said 107 :

"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. Any changes in legal doctrine, brought about by judicial creativity, must 'fit' within the body of accepted rules and principles. The judges of Australia cannot, so to speak, 'make it up' as they go along. It is a serious constitutional mistake to think that the common law courts have authority to 'provide a solvent' 108 for every social, political or economic problem. The role of the common law courts is a far more modest one."

An example of a "new" principle which did not fit with existing doctrine was the decision in Beaudesert Shire Council v Smith 109 . The principle laid down in that case was rejected in the United Kingdom and New Zealand and widely criticised as unsupported by authority 110 . In overruling the Beaudesert principle in Northern Territory of Australia v Mengel 111 , Mason CJ, Dawson, Toohey, Gaudron JJ and I (with Brennan and Deane JJ agreeing to its overruling in separate judgments) considered that the principle was "not in harmony with the law as it was when the decision was given and as it has since developed." 112 Although judicially formulated principles necessarily require a first formulation, development of the common law as attempted in Beaudesert 113 needs to proceed "on the basis of the identification and enunciation of principles that unify and explain earlier decisions" 114 .

The judicial law-making function arises from the necessity to do justice between the parties who have brought a dispute before the court and between those who stand in similar situations. In discharging that duty, it is the web of established legal principle which provides the first point of reference for a judge. By reference to established principles, concepts and precedents and authoritative texts, the judge may be able to reason by induction and deduction to determine the issue in dispute. However, the use of strictly legal materials will often be insufficient to achieve the result which the needs of society require. In that event, the judge is entitled to refer to non-legal materials to determine whether that result can be achieved in a manner that is compatible with the exercise of judicial power.

If contemporary social justice requires that a doctrine be overturned, it may be incumbent upon the court do so. If society does not tolerate the view that a woman irrevocably consents to legal intercourse with her husband, for example, it would be abhorrent to uphold legal precedent to the contrary. In such a case, it is incumbent on a court, not bound by the precedent, to overrule that authority. But few cases are so clear. If the social or economic consequences of extending or modifying liability are problematic, the extension or modification should not be made. Courts are not law reform commissions. They cannot "provide a solvent" 115 for every social or economic problem. They do not have the resources to undertake a comprehensive survey of permanent social values or the social or economic ramifications of judicial changes to the law and, even if they did, that exercise is incompatible with their constitutional function.

JUDICIAL METHOD IN CASES CALLING FOR THE LAW-MAKING FUNCTION

Law-making is primarily the function of the legislature. It is legitimate for the judiciary to make law only in the limited circumstances to which I have referred. What then does judicial method require the judge to do when faced with the novel case or an argument that a particular rule or principle is out of touch with society's needs?

The starting point is to determine as precisely as possible what is the present state of the law. Next, the judge will have to evaluate the alleged need for change. He or she will have to make at least a preliminary assessment as to whether there is a social need for change and whether the proposed change in the law is likely to meet that need. At this stage, the judge may be convinced that an extension or modification of the law is out of the question for any one of a number of reasons. Thus, the proposed change may not fit with existing doctrine or may be contrary to a statutory command or policy or it may have become apparent that it is problematic whether the alleged social need exists or whether the proposed change in the law would alleviate it. In the case of the lower courts, it may become apparent that the proposed change is precluded by binding precedent.

If the judge concludes that the argument in favour of change raises "a serious issue to be tried", to use the language of the Chancery lawyer, the judge will probably need to examine arguably relevant legal material outside the area of immediate legal concern. Statutes and cases in allied areas of law and general legal principles and jurisprudential concepts may give valuable guidance as to general legal policies or experience in other fields that ought to be taken into account. Finally, the social and economic consequences of the proposed change and its alternatives will need to be examined. It is at this juncture that the judicial law-maker is most likely to fall into error for the reasons to which I have already referred. The more uncertain the consequences, the greater is the need for caution.

Whether or not the judge decides to extend, modify or reformulate the law, his or her decision will be the result of examining and weighing the effect of the present law and its proposed change on many interests, values and factors, some of which will point in different directions. They will include the need to maintain the historical continuity and coherence of legal doctrine, the need for a stable and predictable legal system, the extent to which change will affect existing rights and liabilities, the need to maintain the rule of law, the social and economic costs and benefits of the current rule and its change, the impact of change on the efficient administration of justice and the fairness of results under the present and proposed rule according to contemporary notions of justice. Other interests and factors will also be relevant, depending on the circumstances of the particular case.

There is no scale upon which the various consequences can be weighed. In the end, the judge can do no more than make a judgment as to which of the policies underlying the present rule and its proposed change will best serve the public interest in having a stable and predictable legal system which is nevertheless responsive to social needs. That is the way that the common law has developed over the last 800 years. As Oliver Wendell Holmes jnr pointed out 110 years ago, the life of the law has not been logic but experience. Because the law was developed pragmatically - even in the days when judges denied that they made law - the common law remains relevant even in this age of statutes. If it is to continue to remain relevant, it must continue to be developed pragmatically in response to social needs. To those who crave for certainty in the law, that will not seem very satisfactory. But the alternative is even less satisfactory. It is to delude ourselves into believing in what Lord Reid called the fairy tale that in some Aladdin's Cave there is hidden the Common Law in all its splendour and that on a judge's appointment there descends on him or her knowledge of the magic words "Open Sesame".

1 I am indebted to James Stellios, Senior Research Officer of the High Court Library for research assistance for this paper.

2 "Concerning Judicial Method", Jesting Pilate , (1965) 152 at 153-165.

3 "Concerning Judicial Method", Jesting Pilate , (1965) 152 at 157.

4 Gewirtz, "On 'I Know It When I See It'", (1996) 105 Yale Law Journal 1023 at 1025.

5 "The Writing of Judgments", (1948) 26 Canadian Bar Review 491 at 491.

6 "In Defense of Judicial Candor", (1987) 100 Harvard Law Review 731 at 737.

7 Using that term in a non-party sense.

8 I have made only a passing reference to the need to dispose of disputed questions of fact. This is not because of any prejudice against determining those questions. There never has been a time where it could be said of High Court justices, as Lord Holt CJ said of the House of Lords jurisdiction in cases of writs of error, "all causes generally consist more of matters of fact, than of law, and it is beneath the dignity of their Lordships, to be troubled with matters of fact." ( R v Earl of Banbury (1695) Skinner 517 at 523; [90 ER 231 at 235]).

9 Dixon, "Concerning Judicial Method", Jesting Pilate , (1965) 152 at 165.

10 [1975] AC 421 at 458.

11 Commissioner for Railways v Cardy (1960) 104 CLR 274 at 285 per Dixon CJ.

12 "The Judge as Lawmaker", (1972) 12 Journal of the Society of Public Teachers of Law 22 at 22.

13 See "The Law-making Function of the Judicial Process", (1988) 62 Australian Law Journal 15 and 116.

14 (1968) 122 CLR 556 at 563: "the common law is what the Court, so informed, decides that it should be, subject of course to correction by the Judicial Committee in a case in which Her Majesty's Privy Council retains jurisdiction. For, where no authority binds or current of acceptable decision compels, it is not enough, nor indeed apposite, to say that the function of the Court in general is to declare what the law is and not to decide what it ought to be."

15 Doyle, "Implications of Judicial Law-Making", in Saunders (ed), Courts of Final Jurisdiction - The Mason Court in Australia , (1996) 84 at 84.

16 (1997) 188 CLR 241.

17 (1997) 188 CLR 241 at 298.

18 (1996) 187 CLR 1.

19 (1996) 187 CLR 1 at 179.

20 Pavey & Matthews Pty Ltd v Paul (1987) 162 CLR 221 at 253-255.

21 David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353 at 370-376.

22 Wik Peoples v Queensland (1996) 187 CLR 1 at 180.

23 McHugh, "The Law-making Function of the Judicial Process", (1988) 62 Australian Law Journal 15 at 18-24.

24 "The Judge as Lawmaker" in The Judge (1979), Ch 1 at 3.

25 "The Judge as Lawmaker" in The Judge (1979), Ch 1 at 3 citing Jaffe, English And American Judges As Law-makers , (1969) at 13.

26 This is not to ignore the opportunities available to judges of lower courts to navigate around precedent by distinguishing a decision on its particular facts. However, the judicial appeal process ordinarily restrains any inappropriate use of the law-making power in this regard.

27 As to the ability of the appeal courts of the Supreme Courts and the Federal Court to depart from their own decisions, see Nguyen v Nguyen (1990) 169 CLR 245 at 268-270 per Dawson, Toohey and McHugh JJ.

28 McHugh, "The Law-making Function of the Judicial Process", (1988) 62 Australian Law Journal 116 at 116.

29 The Civil Law System , (1957) at 836.

30 (1988) 165 CLR 107.

31 (1997) 188 CLR 313.

32 [1906] AC 428.

33 Northern Sandblasting Pty Ltd v Harris (1997) 188 CLR 313 at 339-340 per Brennan CJ, at 342 per Dawson J, at 347 per Toohey J, at 357 per Gaudron J, at 365-366 per McHugh J, at 391-392 per Kirby J.

34 (1992) 175 CLR 1.

35 (1889) 14 App Cas 286.

36 Mabo v Queensland [No.2] (1992) 175 CLR 1 at 42 per Brennan J (with Mason CJ and McHugh J agreeing), at 109 per Deane and Gaudron JJ, at 182-184 per Toohey J.

37 (1991) 174 CLR 379.

38 (1992) 175 CLR 353.

39 (1995) 182 CLR 609.

40 (1992) 177 CLR 292.

41 Mason, "The Judge as Law-maker", (1996) 3 James Cook University Law Review 1 at 3.

42 (1997) 188 CLR 313.

43 (1994) 179 CLR 520.

44 (1987) 162 CLR 479.

45 See, for example, Waltons Stores (Interstate) Ltd v Maher (1988) 164 CLR 387; Commonwealth of Australia v Verwayen (1990) 170 CLR 394.

46 See, for example, Muschinski v Dodds (1985) 160 CLR 583; Baumgartner v Baumgartner (1987) 164 CLR 137; Stern v McArthur (1988) 165 CLR 489; Louth v Diprose (1992) 175 CLR 621.

47 See, for example, Pavey & Matthews Pty Ltd v Paul (1987) 162 CLR 221; ANZ Banking Group Ltd v Westpac Banking Corporation (1988) 164 CLR 662; David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353.

48 (1992) 173 CLR 473.

49 (1992) 173 CLR 473 at 482.

50 See the discussion by Professor Julius Stone in Precedent and Law , (1985) at 109-112.

51 Cardozo, The Nature of the Judicial Process , (1921) at 66.

52 Cardozo, The Nature of the Judicial Process , (1921) at 102.

53 For example, see Burnie Port Authority v General Jones Pty Ltd (1994) 179 CLR 520 at 594.

54 See, for example, the difference between the majority and minority judgments in Trident General Insurance Co Ltd v McNiece Bros Pty Ltd (1988) 165 CLR 107; Burnie Port Authority v General Jones Pty Ltd (1994) 179 CLR 520.

55 Stephen, "Southey Memorial Lecture 1981: Judicial Independence - A Fragile Bastion", (1981) 13 Melbourne University Law Review 334 at 344.

56 Lord Radcliffe, "Law and Order", (1964) 61 Law Society's Gazette 821.

57 See the instructive discussion by Professor Julius Stone in Precedent and Law , (1985) (especially at 81-83, 111).

58 Lupton v FA & AB Ltd [1972] AC 634 at 658-659 per Lord Simon of Glaisdale.

59 (1991) 172 CLR 243 at 262.

60 "The Judge and the Aequum et Bonum" in The Judge , (1979) Ch 4 at 85.

61 The controversy surrounding the proper parameters of judicial law-making is not exclusive to Australian courts. The debate has raged in the United States since the creation of the Supreme Court and has recently ignited at the birthplace of our common law, in the United Kingdom. See a survey of the debate in the United States, United Kingdom and India in Kirby, "Judicial Activism", (1997) 16 Australian Bar Review 10 at 13 and (1997) 27 University of Western Australia Law Review 1 at 14-18.

62 Sackville, "Continuity and Judicial Creativity - Some Observations", (1997) 20 University of New South Wales Law Journal 145 at 151.

63 Doyle, "Judicial Law-making - Is Honesty the Best Policy?", (1995) 17 Adelaide Law Review 161 at 163.

64 Dawson, "Do Judges Make Law? Too Much?", (1996) 3 Judicial Review 1 at 4.

65 Davies, "The Judiciary-Maintaining the Balance", in Finn (ed), Essays on Law and Government Vol 1 , (1995) 267 at 281.

66 See the comments of the Honourable Justice Kirby in "Judicial Activism", (1997) 16 Australian Bar Review 10 at 11 and (1997) 27 University of Western Australia Law Review 1 at 15.

67 Davies, "The Judiciary-Maintaining the Balance", in Finn (ed), Essays on Law and Government Vol 1 , (1995) 267 at 278.

68 Doyle, "Judicial Law-making - Is Honesty the Best Policy?", (1995) 17 Adelaide Law Review 161 at 179 and "Implications of Judicial Law-Making", in Saunders (ed), Courts of Final Jurisdiction - The Mason Court in Australia , (1996) 84 at 89.

69 The exception arises where the incoherent or unsatisfactory state of a legal doctrine requires a rationalisation which often results in an extension or modification of the reach of the doctrine. Cases may also come before the court arising out of conditions of long standing but which have not been the subject of prior legislative or judicial determination.

70 (1994) 179 CLR 520.

71 (1979) 142 CLR 617 at 633.

72 Burnie Port Authority v General Jones Pty Ltd (1994) 179 CLR 520 at 593. Again, in Hill v Van Erp (1997) 188 CLR 159, I was concerned that the development of the law, which will probably result from the decision of the majority, will involve too great a departure from accepted doctrine and must inevitably extend the frontiers of legal liability in a way that departs from its basic doctrines.

73 Horrigan, "Is the High Court Crossing the Rubicon? - A Framework for Balanced Debate", (1995) 6 Public Law Review 284 at 293-294.

74 [1972] 1 WLR 425 at 430.

75 [1932] AC 562.

76 The Common Law , (1881) at 1.

77 "The Judge as Lawmaker", (1972) 12 Journal of the Society of Public Teachers of Law 22 at 25-29.

78 (1971) 122 CLR 353 at 396.

79 (1920) 28 CLR 129.

80 (1971) 122 CLR 353 at 396.

81 (1992) 175 CLR 1.

82 (1986) 160 CLR 301 at 309.

83 (1986) 160 CLR 301 at 314.

84 (1988) 165 CLR 107 at 123-124.

85 (1991) 174 CLR 379 at 390.

86 It may be that the common law has long recognised "fairness" as one of its inherent values. In Australasian United Steam Navigation Co Ltd v Hiskens (1914) 18 CLR 646 at 671, Isaacs J expressed the view that the common law rules concerning shippers and carriers were "based on reasonableness and fairness".

87 (1936) 50 Harvard Law Review 4 at 20.

88 Mabo v Queensland [No.2] (1992) 175 CLR 1 at 42.

89 Dietrich v The Queen (1992) 177 CLR 292 at 319.

90 Mason, "The Judge as Law-maker", (1996) 3 James Cook University Law Review 1 at 12.

91 cf the discussion in Finn, "Of Power and the People: Ends and Methods in Australian Judge-Made Law", (1994) 1 Judicial Review 255 at 275 concerning "values", "standards", "circumstances" and "needs" of "the community".

92 (1997) 188 CLR 241.

93 "Judicial Activism", (1997) 16 Australian Bar Review 10 at 13 and (1997) 27 University of Western Australia Law Review 1 at 16.

94 The End of Certainty , 1st ed (1992), at 2.

95 Kirby, "Judicial Activism", (1997) 16 Australian Bar Review 10 at 13 and (1997) 27 University of Western Australia Law Review 1 at 16.

96 As in R v L (1991) 174 CLR 379.

97 (1992) 175 CLR 1.

98 (1991) 174 CLR 379.

99 (1992) 177 CLR 292.

100 Dietrich v The Queen (1992) 177 CLR 292 at 319.

101 Dietrich v The Queen (1992) 177 CLR 292 at 319-320.

102 Dietrich v The Queen (1992) 177 CLR 292 at 320-321.

103 "A constant process of innovation and amelioration" to use Lord Wright's phrase: "The Study of Law", Law Quarterly Review , (1938) vol 54, 185 at 188.

104 (1992) 175 CLR 1 at 29.

105 The Lawyer and Justice , (1978) at 279.

106 (1996) 186 CLR 71.

107 (1996) 186 CLR 71 at 115.

108 Tucker v US Department of Commerce (1992) 958 F 2d 1411 at 1413.

109 Beaudesert Shire Council v Smith (1966) 120 CLR 145.

110 Northern Territory of Australia v Mengel (1995) 185 CLR 307 at 338.

111 (1995) 185 CLR 307.

112 (1995) 185 CLR 307 at 343.

113 (1966) 120 CLR 145.

114 (1995) 185 CLR 307 at 339.

115 Breen v Williams (1996) 186 CLR 71 at 115 per Gaudron and McHugh JJ quoting from Tucker v US Department of Commerce (1992) 958 F 2d 1411 at 1413.