Aronson, Mark --- "Misfeasance in Public Office: a Very Peculiar Tort" [2011] UNSWLRS 21

Last Updated: 6 May 2011

MISFEASANCE IN PUBLIC OFFICE: A VERY PECULIAR TORT

Mark Aronson[∗]
Citation

This article is forthcoming in (2011) 35 Melbourne University Law Review.

Abstract

Misfeasance in public office is the common law's only public law tort, because only public officials can commit it, and they must have acted unlawfully in the sense that they exceeded or misused a public power or position. This article examines who might be treated as a public official for these purposes, and whether the tort might extend to government contractors performing public functions. The article also discusses the tort's expansion beyond the familiar administrative law context of abuse of public power, to abuse or misuse of public position.

Misfeasance tortfeasors must at the very least have been recklessly indifferent as to whether they were exceeding or abusing their public power or position and thereby risking harm. That parallels the mens rea ingredient of the common law's criminal offence of misconduct in public office, and reflects a further reason for restricting the tort's coverage to public officials, who must always put their self-interest aside and act in the public interest.

Upon proof of the tort's fault elements, there beckons a damages vista apparently unconstrained by negligence law's familiar limitations upon claims for purely economic loss. This article questions the capacity of the "recklessness" requirement to constrain claims for indeterminate sums from an indeterminate number of claimants, some of whom may have been only secondary (or even more remote) victims of the public official's misconduct. Finally, it questions (and finds wanting) the assumption common in Australia that government will not usually be vicariously liable for this tort. It argues that the personal wealth (or otherwise) of a public official should not set the boundary for a truly public tort.

The article undertakes a comparative analysis of the law of Australia, New Zealand, England and Canada.

I. Introduction

Misfeasance in public office is a very peculiar tort. It is generally regarded as the common law's only truly public tort,[1] because the only people who can commit it are those holding public office,[2] and the only occasions on which it can be committed are those in which public office-holders misuse their public power. Because government's tort liability is usually judged by private law principles, there is no generalised common law right of action for damages for loss caused by invalid administrative action.[3] That is an absence that some have lamented,[4] although most have recognised that government liability for invalidity per se would be financially crippling (particularly in light of the rapid expansion of judicial review's grounds of invalidity), as well as being counter-productive to good administration.

Law reformers have long sought to articulate factors additional to invalidity which might form a coherent and justifiable basis for a new right of action.[5] However, their calls for legislative reform along those lines have failed; indeed the political mood seems to be heading in the opposite direction.[6] Human rights legislation has created a new species of government liability for damages, but these are discretionary, and assessed according to principles[7] that are usually less generous than tort's aim of replacing the entirety of a loss with a monetary award. Common law developments have been mixed. At least in Australia and England, government liability for negligence seems to be in retreat, although the pace of retreat differs between the two countries, and their courts now have different approaches to the resolution of novel negligence claims.

The misfeasance tort, however, has risen to some prominence in the last 20 or so years. A sequence of four leading cases has sketched its most basic outline. Speaking very generally, misfeasance now offers damages on a tort scale for loss inflicted by public officials guilty of "conscious maladministration",[8] a concept which catches abuses of power by public officers who either knew they were breaking the law or recklessly decided not to care that this might be so. The judges in each of the four cases nodded to the tort's lengthy antecedents, but they did not proceed as legal historians disinterring an ancient and well-settled doctrine. Nor did they see themselves as law reformers, imposing new duties of care or new standards of administration on government officers. Their implicit premise was the need for a tort that recognises some of the things that private sector people simply cannot do – government officials regulate, license and coerce in ways that often have no private sector analogue nor any court-based remedy aside from judicial review. Explicitly, however, the cases insisted that they were not about to create a whole new compensation right for government incompetence or inertia – popularly regarded as the bureaucracy's chief pathologies. They have instead sharpened their focus on those hopefully exceptional cases where officials deliberately take the law into their own hands.

The four leading cases have worked mostly in unison across national boundaries, starting in Australia, and from there to New Zealand, England and Canada in that order. Their sketch of misfeasance was only ever intended as preliminary; its edges are blurred, there are several gaps, and a lot of the detail remains to be filled in. It is therefore an appropriate time to take stock, to speculate on the loose ends so far, and to point to some of the hard choices that must now be made.

There are, of course, more than four leading cases, but this article will start (in Part II) with the quartet of cases that laid the tort's modern foundations, before back-tracking to some early history (Part III). It will then investigate the tort's mental elements, starting first with the general place of malice in tort and public law (Part IV), followed by a discussion of the criminal offence of misfeasance (Part V), and the misfeasance tort's recent and potential borrowing from its criminal namesake (Part VI). In a sense, the discussion to that point will have reflected the principal preoccupations of the four modern leading cases, but the article will then turn to other issues. The cases have been at pains to tell us what the tort is not – it is neither an action for breach of duty (Part VII), nor a sub-set of negligence (Part VIII). Nor is it limited to the provision of compensation for government violation of common law or statutory rights, or at least, not "rights" in any narrow and legalistic sense (Part IX). It is a tort defined in large part by the state of mind of officials who either knew they were law-breakers or decided not to care about their legal constraints. But in the latter case, the decision must have been "reckless", which is a requirement with considerable potential that is yet to be explored (Part X). Although the cases all treat misfeasance as a purely public law tort, they have yet to define how closely, if at all, it must track the law of judicial review. The remedy is clearly a supplement of sorts to judicial review, but is it confined to powers or duties supervised by judicial review, or can it extend further to deliberate abuse of government's private law powers such as its commercial powers (Part XII)? Must its defendants hold public office, or can they be government contractors[9] exercising public functions (Part XIII)? And if misfeasance is indeed a public tort, why do the Australian cases doubt the ability to hold government vicariously liable for the misfeasance of its individual officers (Part XIV)?

II. The modern quartet of leading cases

The facts of the four leading cases need only a brief introduction here. Government stock inspectors in Northern Territory v Mengel[10] (Mengel) had told the heavily indebted Mengels not to take their cattle to market and the law-abiding Mengels had complied. The inspectors suspected (wrongly, as it transpired) that the herd had fallen prey to a particularly virulent stock disease. Both the inspectors and the Mengels had thought that the inspectors were acting validly under a statutory scheme empowering intervention in the event of a mere "suspicion", but that scheme had applied only to agreements between government and stock owners, and the Mengels' agreement had expired some time ago. There was a statutory power in default of agreement, but it was never used and in any event, it applied only where inspectors "believed" the stock to be infected, and their suspicions in this instance had not risen to the level of a positive belief. The inspectors had acted beyond their powers, but in good faith, and this was held to be sufficient to exclude liability in misfeasance.

Garrett v Attorney General,[11] (Garrett) was the next leading case of the quartet, this time from the New Zealand Court of Appeal. A police constable had raped the plaintiff, and his sergeant had covered it up until the evidence had gone cold. The sergeant's excuse was that the plaintiff had initially asked that the matter be dealt with unofficially, wanting only for her rapist to be relocated to any place that would be far away from her. The jury appears to have believed him, which might have raised an interesting question as to whether it is misfeasance knowingly to break the law at the plaintiff's request or in the mistaken belief that it was for her benefit. However, the court ruled out misfeasance for a different (and one would have thought less compelling) reason, namely, that there was no evidence that the sergeant either knew or actually suspected the reputational loss or psychiatric harm that the plaintiff would suffer from a cover-up. It was a small town, and the plaintiff's reputation was ruined for want of official acceptance of her story against the constable's denials.

The third case in the quartet was the decision of the House of Lords in Three Rivers District Council v Bank of England (No 3)[12] (Three Rivers). Building on Mengel and Garrett, their Lordships said that misfeasance required either deliberate harm or knowingly illegal conduct in circumstances where the official had suspected that the plaintiff would suffer harm and consciously chose to run that risk. The main concern in Three Rivers was to confine any talk of reasonable foreseeability or objective fault to negligence, and to insist that at the very least, misfeasance required that its defendants knew they were running risks that their action was illegal and harmful and recklessly went ahead anyway. Three Rivers was a class action against the Bank of England for not pulling the plug earlier on a dodgy bank that took a long time to fail, and like a lot of misfeasance claims, House of Lords guidance came not after a trial but on an application to strike out the pleadings.

The fourth member of the quartet also involved an allegation that the police had deliberately (and in breach of the rules) destroyed any chance of a proper investigation. The subject matter was a police killing of a bank robber whose family sought misfeasance damages for their distress at the lack of a proper investigation. The Supreme Court of Canada decided in Estate of Odhavji v Woodhouse[13] (Odhavji) to allow the misfeasance claim to go to trial, endorsing the requirement of Three Rivers that at the very least, the relevant police must have chosen not to care that they were risking the infliction of mental harm on the family members. Odhavji's principal concerns were to insist that misfeasance applied as much to omissions as to acts, and as much to acts or omissions beyond power as to acts or omissions that were illegal for having abused power.

III. Officials behaving outrageously – an introduction

Holt CJ's judgment in Ashby v White[14] is usually said to have created the earliest version of the tort of misfeasance in public office.[15] Four borough constables were ordered to pay £200 to a poor cobbler (Mr Ashby) because they had fraudulently and maliciously prevented him from casting his vote at a Parliamentary election. In this, they had acted in breach of electoral duties which the sheriff's writ had imposed upon them for election day.[16] The elements of moral outrage and punishment were very clear in Holt CJ's judgment:[17]

"If public officers will infringe men's rights, they ought to pay greater damages than other men, to deter and hinder other officers from the like offences. .... To allow this action will make public officers more careful to observe the constitution of cities and boroughs, and not to be so partial as they commonly are in all elections, which is indeed a great and growing mischief, and tends to the prejudice of the peace of the nation."

The principal difficulty for the Chief Justice was the lack of precedent for Mr Ashby's "right to vote", so he created it; Holt CJ had to fall back on "the reason of the law" for want of "particular instances and precedents".[18] Ashby is striking to modern readers for having retained so little of its doctrine but so much of its sense of a need for some sort of tort remedy for outrageous behaviour on the part of public officials.

Ashby's connection with antecedent rights has gone. If the quoted passage from Holt CJ's judgment meant that lesser damages would have been ordered against a private individual who had hindered Mr Ashby's right to vote, then the action would have lain against public and private defendants alike. Misfeasance nowadays no longer lies against private individuals, and the common law is now more tolerant of malice and selfishness in a private individual. Misfeasance no longer follows from a simple breach of duty, nor from a simple breach of public or statutory duties. The House of Lords in Watkins v Secretary of State for the Home Department[19] (Watkins) disowned Ashby's award of punitive damages in the absence of material loss, with the result that moral outrage per se warrants neither punishment nor condemnation via the misfeasance tort.[20] Even in Watkins, however, Lord Bingham admitted the validity of the same policy drivers that had under-pinned Lord Holt's judgment:[21]

"There is great force in the respondent's submission that if a public officer knowingly and deliberately acts in breach of his lawful duty he should be amenable to civil action at the suit of anyone who suffers at his hands. There is an obvious public interest in bringing public servants guilty of outrageous conduct to book. Those who act in such a way should not be free to do so with impunity."

What has emerged since Ashby is a set of countervailing policy concerns, calling for the courts to set a "balance". The balance in Sanders v Snell[22] was between providing compensation for abuse of public power, and avoiding a liability rule that "may deter officials from exercising powers conferred on them when their exercise would be for the public good." Watkins added a more general consideration said to apply to all torts, namely, that the "primary role" of tort is compensation, not punishment.[23] The tort's characterisation as a fault-free action for breach of duty was dropped for obvious reasons of economy – that would have been far too expensive in an age of statutes, let alone an administrative state. Its fault elements have been set at deliberate wrong-doing rather than objective (even spectacular) incompetence,[24] partly to avoid making public servants overly cautious about the legality of their actions,[25] but also (in Australia) out of a misplaced sense that individual officers would "ordinarily" obtain no indemnity from their employers.[26]

If these were the only issues in play, the misfeasance action would still apply to private and public defendants alike, as Lord Holt appeared to suppose in Ashby itself.[27] Its retraction to the public sphere therefore needs further explanation. For more than a century now, the common law grounds of judicial review of administrative action have been driven by a sense that public officials must act altruistically. Selfishness, greed, revenge, spite, malice, prejudice – it is said that private individuals can act from these and other base motives, and even for no reason at all, but that public officials are different.[28] One might cavil with that proposition at several points, the most obvious being that trustees must also put selfishness to one side. That this has some relevance to misfeasance appears clearly from Paul Finn's "epigrammatic statement of the essence of the fiduciary obligation",[29] namely:[30]

"A fiduciary must act honestly in what he alone considers to be the interests of his beneficiaries."

Lord Millett said much the same thing about misfeasance in Three Rivers, explicitly analogising to the law relating to a trustee's obligations.[31] The dual requirements of honesty and altruism were perhaps more prominent in the days before misfeasance expanded beyond deliberately inflicted harm for an improper purpose.[32]

The obvious questions are why misfeasance does not apply to trustees, and why some of equity's more demanding obligations placed upon trustees do not apply equally to public officials. In other words, why does the law provide two sets of rules for those who must act altruistically, depending on whether they (or, perhaps, their functions) belong to the public or private sector? No judgments have ever addressed that issue directly, although they do insist that misfeasance should not swamp the private sector's torts[33] and that the development of private law's economic torts should not swamp the misfeasance tort.[34] Brennan J said that if the existing dividing lines were not maintained, the law would "speak with a forked tongue", sending out conflicting messages depending only on the pleader's ability to select the most favourable tort.[35] These are largely concerns for doctrinal coherence,[36] and none of them has addressed the areas where legally mandated altruism crosses the divide between public and private. It is suggested that if the issue were to arise, the answers would necessarily be pragmatic. Generally speaking, there is no shortage of common law rules applying to those from either sector who must subordinate their own interests, and no obvious reason for supposing that doctrinal coherence across the sectors would produce a better result.[37] A further complication for any coherence project would arise from the fact that in Australia, at least, many public officials and their institutions have some measure of statutory protection from damages claims so long as they act in good faith.[38] Finally, trustees owe their duties to a defined class of beneficiaries, whilst public officials serve the public interest more generally.

IV. Malice in tort and public law

The role of malice in tort law varies between different torts, and also according to whether the defendant is public or private. That was not always the case. Holt CJ forgot to mention fraud or malice in his first report of Ashby v White, although they were clearly implied. He slipped them in to a revised report in time for consideration by the House of Lords on appeal.[39] The role of malice in tort law kept changing over the ensuing centuries. For example, we learnt around two centuries later, and in a nod to the free market's need for competitive selfishness,[40] that in actions against private sector defendants, and aside from conspiracy, malice does not make otherwise lawful competitive behaviour tortious, just as good faith does not exonerate conduct which would otherwise be tortious.[41]

The position has long been different for defendants who are public officials of some kind. Well before the common law of tort had rejected the contention that malice in a private defendant could tortify conduct that was otherwise lawful in a free market, the tort liability of any public official exercising a discretionary function[42] other than a superior court judge had become conditional on establishing fraud or malice.[43] Inferior court judges enjoyed that protection plus one more, namely, that they must also have acted wholly outside their "jurisdiction", taking that term in a very narrow sense.[44] In other words, by the early 1900s, a tort law gap had already opened up between public and private defendants. Malice was not a good plea against a superior court judge, and it would count against inferior court judges only in combination with a plea of want or excess of jurisdiction. Aside from the judiciary, however, malice would remove any lingering common law protection for all other public officials exercising discretionary powers, and it was irrelevant to the liability of a private individual's market-oriented conduct.

The gap between public and private defendants got wider over the course of the twentieth century. It would appear that the common law had in that period extended to inferior court judges the complete immunity previously enjoyed only by superior court judges, but it had also retracted immunity altogether from anyone else exercising a discretionary function.[45] One cannot be absolutely sure of this, however, because statutory protections have replaced the common law in most places.[46] The legislative trend has been to grant absolute immunity to all judicial officers, high or low,[47] and to grant many others in public life an immunity conditional on their acting in good faith. "Good faith protection clauses" abound in Australia's regulatory statutes, and were once common in Britain. Lord Steyn spoke of Ashby v White as the misfeasance tort's "first solid basis",[48] but also acknowledged the contributions to the current mix of the older cases concerning liability for discretionary and judicial acts.[49] As his Lordship noted, the net result is that misfeasance in public office now stands as an exception to the usual position that the presence of malice will not make a non-tortious act tortious, just as its absence will not excuse an otherwise tortious act.[50]

The reasons for distinguishing between private and public malice are not hard to find. Lord Hoffmann once wrote of private sector defendants that "there is little which can be said in defence of a right to cause loss out of sheer malice".[51] He wondered aloud whether England might adopt something approximating America's prima facie tort theory for harm caused deliberately or maliciously,[52] a theory similar to the civilian doctrine[53] of abuse of rights. Forty years later, his Lordship entertained no doubts in rejecting the American option so far as it might apply to private sector companies and individuals,[54] saying that in the cut-throat world of competitive markets, it would produce too many uncertainties for too little gain.[55]

Malice and an intention to harm in a public official are very different. The very essence of competitive behaviour in the market place is to gain an advantage at the expense of one's rivals, but the public official is not conceived as a free agent in a free market. Public officials are supposed always to be constrained to act in the public interest or, at least, not in their own personal interest. The archetypal government body or official has no commercial motivations, adheres to proper process and standards of propriety, and has long been required to act not selfishly but altruistically. As explained below, the misfeasance tort can be proved in alternative ways, one them (targeted malice) amounting in essence to a prima facie tort.

Lord Holt awarded £200 in damages, even though Mr Ashby's preferred candidate had won the election without Ashby's vote, and even though Ashby himself had suffered no material loss. The judgment spoke of deterring "like offences",[56] and the sheer size of the award[57] showed that it was clearly intended as a fine. Neither malice nor fraud would attract a figure of equivalent size through the tort system these days, because the dominant view is that only the criminal courts should deal with crime,[58] leaving tort focused on compensation. Indeed, some would say that tort law should focus solely on claims for compensation.[59] However, there have always been redoubts, areas which undeniably use tort for other means.

The only issue in Ashley v Chief Constable of Sussex[60] was whether the police had killed a man lawfully, and in an attempt to avoid that issue going to trial, the police were prepared to admit the quantum of damages as well as liability for false imprisonment and negligence, but the majority let the family proceed to trial for assault and battery. They said that the family was entitled to use the court solely for vindication of their belief that a wrong greater than the admitted torts had been committed.[61]

Punitive damages can be claimed for misfeasance,[62] provided that the plaintiff has also sustained some material loss.[63] It is in fact difficult to find modern cases in which they have been awarded,[64] doubtless because it is so easy to disguise a punitive award as aggravated damages, but even their theoretical availability represents for England an acknowledgment of the special position reserved for claims against public officials for misconduct that is oppressive, arbitrary or unconstitutional. In most other respects, punitive damages are unavailable in that country,[65] and very much the exception in Australia,[66] New Zealand[67] and Canada.[68] Once again, special treatment is reserved for officials behaving very badly. Lord Devlin's context was not misfeasance, but his explanation for treating public officials differently has obvious resonance:[69]

"Where one man is more powerful than another, it is inevitable that he will try to use his power to gain his ends; and if his power is much greater than the other's, he might perhaps be said to be using it oppressively. If he uses his power illegally, he must of course pay for his illegality in the ordinary way; but he is not to be punished simply because he is the more powerful. In the case of the government it is different, for the servants of the government are also the servants of the people and the use of their power must always be subordinate to their duty of service. It is true that there is something repugnant about a big man bullying a small man and very likely the bullying will be a source of humiliation that makes the case one for aggravated damages, but it is not in my opinion punishable by [exemplary] damages."

One might question whether corporate bullying is any less of a social evil than government bullying,[70] and that, indeed, appears to have been one of the reasons why his Lordship's categorical restrictions on the availability of punitive damages failed to catch on in Australia.[71] Professor Birks, however, suggested that England's reluctant retention of punitive damages might reflect a deep-seated suspicion that government cannot always be trusted to exercise its prosecutorial discretion wisely:[72]

"At some point in the rise of the efficient and infinitely resourced ... state, the view took hold that the national bureaucracy could be relied upon to do all the punishing and avenging that was necessary. This manifested itself in the ostensibly logical doctrine concerning the natural role of the law of civil wrongs. Properly understood, civil wrongs should not meddle with punishment and deterrence. ... Declining resources now expose this as no more than a choice made at a time of faith in the power, efficiency and benignity of the state."

That observation has added strength when the wrongdoer is a public officer,[73] although Lord Devlin did not assign that as a reason for allowing punitive damages in such circumstances. When that issue emerged more directly in a subsequent case, two Law Lords warmly endorsed the value of punitive damages in the protection of civil liberties.[74] The context was important – it was a misfeasance case, providing for the first time an occasion for drawing together the common threads running through misfeasance and the wider tort law debates concerning punitive damages. It appears that only Lord Bingham has advocated trusting the state to launch criminal proceedings against its own officers.[75]

V. The criminal offence

If, as argued above, the tort of misfeasance is driven in part by a sense of moral outrage at the abuse of collective power, then its sense of punishment can never be far from the surface. It is therefore no digression to look briefly at the crime of misfeasance. Misconduct in public office is a common law indictable misdemeanour with a long history pre-dating Ashby v White.[76] Its intersection with what is now called the misfeasance tort was obvious in Ashby itself, where Holt CJ noted that the defendants could have been prosecuted on indictment. However, if the defendants had by a single act refused to take the votes of a large number of electors, then in order to avoid a multiplicity of actions, prosecution would have been the only remedy.[77]

This type of cross-over between crime and tort was once common, and it is worth adding that a prosecution was indeed a remedy of sorts before the state had assumed a virtual monopoly of policing and prosecution functions. The legal system used to place far greater reliance on economic punishments and rewards for the enforcement of the criminal law generally and public duties in particular. Those derelict in the performance of their public duties could be prosecuted, and compensation of sorts could come from granting to the prosecutor or common informer a share of the fine or reward, or an award of costs to be paid from County funds.[78] The detection, capture and prosecution of thieves was a profitable, and therefore corruptible, system, and its acceptance of market-based policing bears some parallels with the re-emergence in modern times of private policing with its concomitant rejection of the state's previous claim to a monopoly of lawful violence.[79]

The common law misfeasance offence retains a degree of flexibility that makes it less certain than statutory offences, and attempts to codify it have been resisted in England in the belief that the advantages of flexibility outweigh the gains of certainty.[80] The common law offence covers acts or omissions of public officers in the course of or in relation to their public office, which amount to misconduct with a degree of culpability that warrants public condemnation and criminal punishment.[81] Speaking for the Hong Kong Court of Final Appeal, Sir Anthony Mason NPJ said that whether the misconduct is sufficiently culpable depends on whether it is serious:[82]

"having regard to the responsibilities of the office and the office-holder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities."

Speaking for the English Court of Appeal, Pill LJ said that:[83]

"there must be a serious departure from proper standards before the criminal offence is committed; and a departure not merely negligent but amounting to an affront to the standing of the public office held. The threshold is a high one requiring conduct so far below acceptable standards as to amount to an abuse of the public's trust in the office holder. A mistake, even a serious one, will not suffice."

Misconduct is probably an offence defined by conduct rather than outcome, so that the actuality or risk of harmful consequences serve only as bases for inferring the relevant degree of seriousness,[84] which is a jury issue. In jurisdictions where the common law's general offence remains, it is charged as misfeasance (or misconduct) in a public office.[85]

Some Australian jurisdictions have replaced the common law offence with statutory offences which appear to be narrower, and are certainly more determinate.[86] Canada's codification is also narrow, in that its "breach of trust" offence has been held to have codified the crime of misfeasance, but not of nonfeasance.[87]

Clearly, there are differences between the tort and the crime – for example, the crime is a conduct offence, whilst the tort requires both conduct and material damage. There has nevertheless been some cross-referencing in some of the recent tort judgments to their relatives in criminal cases,[88] and some of the recent criminal judgments concerning the common law offence have returned the compliment.[89] The context was a debate as to the meaning of reckless indifference. The House of Lords in R v Caldwell[90] had elided criminal negligence with criminal recklessness by ruling that a person could be criminally reckless if they had given no thought whatsoever to a particular and objectively obvious circumstance or risk. Lord Steyn refused in Three Rivers[91] to import that precedent across to the misfeasance tort, and gave the broadest of hints that Caldwell needed overruling, an event that occurred just four years later.[92]

VI. The mental elements of the tort: a gradual process of dilution

Accounts of the misfeasance tort typically credit it with a long history, but add in the next breath that its reach, content and boundaries still need definition.[93] Both statements are correct; its history is long and tangled, with the result that what has emerged over the last 50 or so years is in reality nothing less than a new tort to meet the needs of people living in an administrative state. Most of the modern changes have occurred by a series of cases diluting the requirement of malice, whilst at the same time expressing confidence that this redefined "malice" will be sufficient to protect public officials from their ever-present fear of liability to an indeterminate class to an indeterminate extent.

The meaning of "malice" in the early cases was generally straight-forward, which were all about officials aiming straight for their plaintiffs, with every intention of harming them for reasons that they must have known were unlawful. The moral case for damages in such circumstances is very strong, but it would be rare indeed for a plaintiff to be able to plead and particularise such malice, and even more rare to prove it. It is probably small wonder, therefore, that misfeasance was virtually defunct in England (although not the commonwealth) by the beginning of the last century. Halsbury confused it with tort law's more general distinction between misfeasance and nonfeasance,[94] and the Court of Appeal forgot its existence entirely.[95]

Smith J took the first step beyond the old model of the intentional and malicious infliction of harm in Farrington v Thomson.[96] A publican had committed three licensing offences in quick succession, so that his second offence occurred before conviction for his first offence. The Act provided for forfeiture of the licence upon a third conviction. No-one was empowered to order the publican to cease trading, because the Act made forfeiture automatic. The police had known this when they ordered the publican to close down; they thought that the licence had been forfeited. If the forfeiture provision had in truth applied, then the publican could hardly have claimed that the police order had caused him compensable loss, but the police had misconstrued that provision in complete good faith. Because the publican's second offence had predated his first conviction, it did not count towards forfeiture, with the result that his licence remained lawfully intact. For this technicality,[97] Smith J held the police liable in misfeasance, which he repackaged into a form that is now familiar. He proposed an action for damages for misfeasance in public office where the public officer either "acted maliciously, in the sense of having an intention to injure", or (and this was the big step) caused damage to the plaintiff by "an act which, to his knowledge, amounts to an abuse of his office".[98]

Professor de Smith publicised Farrington for all commonwealth readers in 1968.[99] Almost a decade later, Professor Wade switched from sceptic to enthusiast.[100] The following year, an article appeared which explored the commonwealth literature in considerable detail, and urged the English courts to treat this rediscovered tort as more than just an "academic curiosity".[101] Shortly afterwards, and without citing a single authority, the Privy Council said that the tort was "well-established", and that its constituent mental elements were either malice or deliberate excess of power.[102] The English Court of Appeal confirmed that distinction in Bourgoin SA v Ministry of Agriculture, Fisheries and Food,[103] and applied for the first time the label of "targeted malice" to the old model.[104] As in Farrington, the defendant Minister in Bourgoin had borne the plaintiffs no grudge, but his deliberate illegality was hardly technical. He had banned the importation of French turkeys for Christmas on food safety grounds when he knew that these did not apply. His motive was the patriotic protection of English producers, and his plea that he was not aiming to hurt their competitors was to no avail. Bourgoin sought to minimise the step it was taking, by saying that the Minister's knowledge that his illegal action would harm the French was no different to him having sought that damage – in effect, an intention to harm was imputed from his knowledge.[105]

Bourgoin's expansion from deliberate harm to deliberate illegality was small compared to the next step. The quartet of leading cases[106] added a third alternative to the mental elements of misfeasance. They reasoned that there was no moral difference between knowing something on the one hand, and being aware of its possibility but not caring whether it might be true or might occur. This third variant is generally referred to as reckless indifference, but it is not to be imputed – the defendant must have consciously adverted to the relevant circumstance or risk and decided not to care about it. Analogising from criminal law principles, the indifference must be subjective.[107]

Two immediate problems that arose in consequence of watering down the minimum requisite mental elements for misfeasance to reckless indifference were whether to apply this variant to one or both of the principal elements of misfeasance (namely, illegality and resultant harm), and whether to insist on a correlation between the plaintiff's harm and the degree of its risk that the defendant consciously chose to disregard. It is clear that the reckless indifference component applies in New Zealand and England both to the harm that the plaintiff has suffered,[108] and to the illegality component of the tort.[109] It remains to be seen if misfeasance will require some degree of correlation between the extent of the risk of harm and the defendant's deliberate unconcern.

The House of Lords decided Three Rivers[110] in two parts; the first part considered the tort's elements, and the second part considered the revamped pleadings and a motion to strike them out on the ground that the claim had no reasonable prospects of success. Their Lordships were very clear in Part 1 that a claim based on reckless indifference had to plead that the harm which the defendant had chosen to disregard was a probable or likely harm.[111] With respect, that was an unnecessary complication, as Lord Hope seemed to realise in Part 2, when he rejected a challenge to a pleading that alleged that the defendant had consciously adverted merely to a risk of harm, rather than a probable or likely risk of harm. In dismissing the challenge, Lord Hope converted Part 1's "probable" and "likely" into the non-probabilistic "serious", with the result that an improbable risk of serious harm might sometimes suffice.[112] In effect, though not in words, Lord Hope ended up treating the probability of risk as one of the elements, along with its gravity should it eventuate, that goes not to indifference but to recklessness. Distinguishing between recklessness and indifference is not a mere semantic nicety – as discussed below, it has the potential to counter the possibility of misfeasance liability that is out of all proportion to the blameworthiness of the defendant's conduct.[113]

Mengel[114] had the least need to descend to this level of detail, because the plaintiffs had come nowhere close to establishing knowledge or conscious indifference on the part of the defendants' officers to the limits of their lawful powers.[115] The plurality judgment said that even if the plaintiffs' case had been stronger in this respect, they would also have needed to establish that their harm had been at least foreseeable, and perhaps even probable.[116] Although it is not entirely clear, it is likely that like Lord Hope in Part 2 of Three Rivers, their Honours wanted that at the very least, there be reckless indifference as to harm, but regarded the probability of risk as going to recklessness, rather than treating it as a super-added requirement.[117] However, that is not how Mengel has been interpreted in Australia, where the current rule applies the reckless indifference requirement only to the illegality issue, and not to the risk of harm. In Australia, the harm must have been foreseeable, but the defendant need not have adverted to its risk.[118]

Most modern cases and commentaries are content to give the standard two-limb definition of misfeasance, which splits it into "targeted malice" and "the rest". Using the same ingredients, Lord Hobhouse said that there are three limbs – purpose, knowledge and consciously reckless indifference.[119] Because it is clear, however, that there are not two (or three) torts but one,[120] it is necessary to consider the point of having alternative limbs. On one view of it,[121] the alternative mental elements (targeted malice, knowledge and conscious indifference) comprise a closed list of the types of fault sufficient to warrant an action for misfeasance. An alternative view is to see these mental elements not as a closed list, but rather as instances (or even evidence)[122] of the types of dishonesty or want of good faith that the tort requires.

In a much quoted passage, Brennan J said in Mengel[123] that the core of misfeasance lay in "the absence of an honest attempt to perform the functions of the office". His Honour said that there was such an absence if the defendant had acted invalidly and with malice, knowledge or reckless indifference, and he may well have intended that list to be exhaustive. There are passages in Three Rivers that could be interpreted as requiring proof of dishonesty or bad faith as an additional element in all cases.[124] In Australia, proof that defendants knew that they were acting beyond power is all that is needed to establish bad faith.[125]

Lord Millett's view, however, was that deliberate illegality is insufficient where defendants believe that they are acting in the best interests of the plaintiff. He distinguished those cases from the situation in Bourgoin, where the defendant knew that he had acted against the plaintiff's interests but had imagined that he was acting in the best interests of the general public.[126] Implicit in that distinction is a welcome recognition that deliberate breaches of the law sometimes merit different responses. In other contexts, judicial condemnation does not automatically follow from proof that public officers have deliberately broken the law,[127] and in their dealings with bureaucracy, the general public has often depended upon the willingness of public officials to bend the rules. Perhaps Lord Millett's concerns might be appropriately addressed by setting tighter limits to the scope of liability where the tortfeasor's wrong-doing was not self-serving. An official who accedes to the plaintiff's request to bend the rules should not necessarily incur liability unless, of course, the plaintiff was in a particularly fragile state.[128]

Even where the rules are bent against the plaintiff's interests, the loss suffered will not always be recognised as compensable in tort. The plaintiff in one case had been convicted on the basis of evidence that the police had deliberately broken the law to obtain from a third party. The plaintiff’s misfeasance claim failed at several points, one being that the police had only ever intended to obtain truthful evidence, and that it had in fact been truthful.[129] In effect, the plaintiff's loss was to have been convicted on reliable evidence after a fair trial – hardly a loss that tort should recognise.

Several of the leading judgments proceed as if liability will always result from proof of targeted malice,[130] but two obvious qualifications are needed. First, there will be no case of targeted malice if the plaintiff's harm is the very point of the relevant power, and if nothing more is shown than the defendant's intention to inflict that harm; the misfeasance claim will need in those circumstances to meet the requirements of knowledge of the illegality or reckless indifference to the risk of illegality.[131] Police, for example, often know that they will harm a person's interests, but that is no misfeasance provided they act in good faith.[132] The same point emerged from a case involving a public servant's dismissal, even though there had been bad blood between the defendant and the plaintiff.[133] Similarly, a decision taken in good faith to place the plaintiff in immigration detention could not qualify as targeted malice.[134]

Secondly, malice is not always a short-cut to liability in misfeasance, a point that Harper J made in Grimwade v Victoria.[135] His Honour hypothesised a parking officer giving a ticket with malicious glee to his worst enemy. If the car was in fact parked illegally, the officer’s malice would be irrelevant.

Most cases treat Roncarelli v Duplessis[136] as a modern paradigm of misfeasance, although the term itself played no part in the case. The assumption is that Mr Duplessis would still be found liable under today's more structured misfeasance tort, but I am not so sure. The famously autocratic Premier and Attorney General of Quebec had either cancelled or ordered the cancellation of Mr Roncarelli's liquor permit, attached to a profitable and well run restaurant that had been in the family for many years. That he wished harm to Roncarelli was indisputable, and if it was relevant, his reasons were also clearly ultra vires, although his intervention was even more clearly invalid because the power of permit cancellation belonged to the Liquor Commissioner, not the Premier. However, only Abbott J said that Duplessis knew he was breaking the law;[137] the other judgments treated Duplessis as having assumed that he was acting lawfully, which meant that he was not deliberately indifferent as to the law.[138] Admittedly, Rand J found Duplessis guilty of "malice", but his definition slid from deliberate illegality, to action taken in good faith but for an "improper purpose" in administrative law terms,[139] and that is a review ground that almost never involves personal dishonesty or impropriety. If targeted malice (meaning the deliberate infliction of harm) provides no more than a strong but rebuttable presumption[140] that the defendant knew that he or she was exceeding the law, and if Duplessis is to be taken as having broken the law entirely inadvertently, then he would nowadays escape liability under that head, and under the alternative limb because his illegality was not advertent. Nowadays Mr Duplessis should escape misfeasance liability for want of relevant fault; he was objectively reckless but not subjectively indifferent.[141]

VII. Actions for breach of duty: which duty?

The early precedents sometimes cited as part of the history of the misfeasance tort all discussed the common law's remedial reach for breaches of public duty. As explained below, that could be misleading these days, because it suggests the need either for an antecedent relationship between the parties or for a legal duty requiring the defendant to have the plaintiff in mind. If there is a duty nowadays, it is not deliberately to abuse a public power, and expressed at that level, it is not very useful. It is instructive to track the history of the search for a meaningful action for breach of duty.

Tort law has a long history of experimentation with actions for breach of duties which might in some sense be characterised as duties of a public nature. There have been actions upon the statute, actions for breach of public duty, actions for breach of statutory duty, a short-lived attempt to use Lord Cairns Act as a route to a damages award in lieu of a public law injunction, and even an attempt to extend damages for negligence wherever a public official's failure to confer a benefit was invalid for Wednesbury unreasonableness.

Some say that the story starts with the second Statute of Westminster.[142] Enacted in 1285, one of its chapters (in Consimili Casu) stated that in cases where no exact precedent for a writ could be found,[143]

"from henceforth, where in one case a writ is granted, in like case, requiring like remedy, the writ shall be made as hath been used before."

That certainly gave some impetus to the incremental development of the forms of action, but it is doubtful that it authorised actions for breach of any statutory duties,[144] let alone for breach of common law duties.

Then there was the relatively short-lived proposition that an action would lie for breach of a public duty. First aired in 1786,[145] it appears that this doctrine was intended to link up a number of previously distinct actions against those charged with duties towards the public for neglect, non-performance, or even abuse of office.[146] Although the House of Lords approved the rule twice in the early nineteenth century,[147] it was stated far too broadly to have survived. According to Paul Finn,[148] the courts voiced concerns that so unrestricted a basis of public liability would have ruinous consequences for public authorities, particularly in light of the massive expansion of statutory authorities with finite budgets charged with the performance of public duties.

The action for breach of public duty morphed into a narrower but always uncertain tort for breach of statutory duty.[149] This tort has survived, but with underlying policy concerns similar to those that had helped kill off the action for breach of public duty. In purely numerical terms, there are far more duties sourced to statute than to the common law, and the puzzle for the courts has always been how to set principled and predictively useful parameters to the statutory tort.

Holt CJ himself had said something in support of an action for breach of statutory duty,[150] but that was obiter, and the action cannot be said to have been firmly established until the last quarter of the nineteenth century.[151] Even now, however, its field of operation seems largely to be confined to actions for breach of statutory duties relating to the prevention of injury in places of work.[152] In earlier times, the tort was useful in helping employees to get around the common employment rule,[153] and (for a while) the defence of contributory negligence.[154] If the relevant statute prescribes a duty of strict liability, then the statutory tort will have a practical advantage over the common law negligence action. That advantage was reduced, however, by the advent of no-fault workers' compensation.[155] It may also be theoretically possible for the statutory tort occasionally to fill a gap left by the common law's denial of a general duty to conform to the morals of a Good Samaritan.[156]

The tort's basis is the implication of a statutory intention to grant a right of action to a plaintiff suffering loss from the defendant's breach of a statutory duty, if the duty was intended for the benefit of a limited class of persons of whom the plaintiff is one.[157] Although posed as an exercise in statutory construction, the legislative intention (if it be found) is fictional, and the list of factors going one way or the other has never been of much help, because none of them is dispositive. Some of the leading judgments read as if there were rebuttable presumptions of statutory construction, but others give more credit to the judge than to legislative drafters in determining whether the Act has created a cause of action.[158] It would appear that the real reason for the tort's ossification is a prevailing unwillingness on the part of the judiciary to indulge in such a haphazard divination of statutory torts, particularly where the plaintiff's purpose is to establish a strict liability tort. That certainly appears to have been one of the reasons why the Canadian Supreme Court abolished the tort, folding it into negligence.[159] An additional concern in Australia is that any treatment of the constructional exercise as an unconvincing cover for judicial creativity might threaten the constitutional separation of powers.[160] The tort survives in Australia, New Zealand and England, but cannot be said to have thrived.

Having regard to the evolution of the statutory tort from earlier doctrines that focused on breach of public duty, what is particularly noteworthy is how rarely the statutory tort is applied for what might loosely be called maladministration by public authorities.[161] Although it is theoretically available against private and public defendants alike, it appears to be more difficult to imply the statutory grant of a right of action against public authorities.

One of the big factors in favour of implying a right of action is that the statutory duty might otherwise lack an adequate sanction for its breach.[162] One of the difficulties here is that the alternative sanction is sometimes a criminal penalty,[163] sometimes an administrative enforcement mechanism or even a right of access to judicial review[164] or merits review.[165] Further, a test that turned upon the adequacy of the statutory sanctions for breach would involve a radical departure from text-based principles of statutory interpretation.[166]

The requirement that the statutory duty have been enacted for the benefit of the plaintiff or a class to which the plaintiff belongs also weighs heavily against its application to public authorities. Regulatory and social welfare schemes, for example, are said to have been passed for the benefit of society at large rather than for particular classes such as the homeless or the unemployed.[167] Lord Simonds took the provision of a criminal sanction as evidence that the duty was enacted for the public's benefit.[168]

It has been said that the action for breach of a public authority's statutory duty is confined to duties with little or no discretionary content,[169] because it would be inappropriate for a court to exercise a discretion that is legislatively vested in an administrative authority.[170] Two Australian States have legislated to remind their readers that breach of a public authority's statutory duty cannot lead to liability if that would contradict the relevant Act or its "policy".[171]

Mention might also be made of a short-lived attempt to use Lord Cairns Act[172] as a vehicle for delivering discretionary damages for violation of laws enacted for the public benefit. That Act (and its successors) empowered the court in its discretion to grant damages in lieu of an injunction if that had been sought for the prevention of a "wrongdoing". The development of the public law injunction was achieved by a liberalisation of the standing rules, such that plaintiffs could seek injunctions against public wrongs even if they had no legal or equitable rights under threat, provided that they suffered (or would suffer) damage peculiar to themselves.[173] In Australia, that standing rule was further relaxed by switching from "peculiar" to "special" damage,[174] and in England, the standing requirement became entirely discretionary.[175] The courts in New Zealand and Australia have decided, in effect, that the injunction's reach now extends considerably further than the power under Lord Cairns Act to award damages in lieu of an injunction, a power which should be confined to cases of threatened violations of a plaintiff's legal or equitable rights or interests.[176] It was reasoned that the contrary proposition would have been tantamount to making discretionary damages available for threatened or continuing breach of statute.

Finally, Lord Hoffmann wondered aloud whether plaintiffs might occasionally (albeit rarely) be entitled to damages in negligence, on the ground, in effect, of Wednesbury unreasonableness. His Lordship had thought this possible if a body exercising public power had failed to exercise a statutory power to confer a benefit upon the plaintiff in circumstances where the only lawful option on the facts had been to confer that benefit.[177] In effect, the idea was that if mandamus could have been granted for Wednesbury unreasonableness, then that might be the basis for a claim in negligence. However, the idea was abandoned shortly afterwards, in a case that reaffirmed the distinction between a statutory duty enforceable by mandamus and a common law duty of care which remained a precondition to an award of damages in negligence.[178]

VIII. The limits of negligence law

Although misfeasance is classified as an intentional tort,[179] it now stretches to include actual reckless indifference, in which the indifference is defined subjectively so as to exclude the defendant who failed to think of a risk that any reasonable person would have foreseen. Apart from that, however, one might question whether there is much practical difference between indifferent tortfeasors in misfeasance and negligence. The Law Commission for England and Wales had thought the two to be so close that if negligence were to be appropriately expanded, then misfeasance could be safely abolished.[180] The Commission eventually abandoned the whole idea,[181] because of wide-spread opposition from the profession and the academy on normative grounds, and from the government which declined to assist the Commission's speculation as to the cost of its proposals.

The Law Commission's methodology adapted the English common law's approach to novel "duty of care" issues in negligence law,[182] balancing normative issues with considerations designed to make its proposals cost-neutral so far as that could be predicted. The liability of public authorities in negligence is usually assessed on the same basis as the liability of anyone else, but where the complaint relates to an activity that has no close private sector analogue or raises issues best left to the political branches, the courts deploy a number of control devices that limit or even deny a duty of care, or that limit the amount of liability. In essence, the judges shape the contours of negligence in response to certain types of relational or cost-benefit considerations and in response to operational pressures on the defendant.

At common law, for example, authorities investigating possible child abuse owe no duty of care to suspect parents;[183] regulators of nursing homes might well owe a duty of care to the residents, but not to the operators;[184] in most places, the police owe no duty of care to potential victims in their task of crime prevention;[185] and litigants (even though they are public authorities) owe no duty of care to their opponents.[186] Negligence also has tight restrictions on claims for compensation for purely economic loss, and yet as far as one can tell, these are typically the only losses alleged in misfeasance cases.[187]

Lawyers can be liable for getting the law wrong because they are legal professionals; even so, their liability is usually limited to their clients or those the client seeks to benefit.[188] Negligence law rarely requires those not legally qualified to get the law right, and that holds for public authorities, who usually have no duty to take care that they act validly.[189] If public officials were under such a duty of care, then it would sometimes be the shortest way around the proposition that invalidity per se is no ground for a tort claim. Furthermore, one would expect the duty usually to be discharged by obtaining legal advice.[190] However, it would be difficult for a court to set standards as to when an administrative authority should seek legal advice.[191]

Misfeasance has adopted none of these strategies in response to fears that it might over-reach, burdening individual defendants or their governments with potentially unlimited liability for what might have been an entirely understandable (if wrong) excess of public zeal. It also lacks the protection of negligence law's troubled abstention from cases of so-called "pure omission".[192] Negligence law rarely requires defendants to take active steps in the protection of plaintiffs either from themselves[193] or from harm caused by third parties,[194] and that is usually explained in terms of respecting an individual defendant's autonomy, especially where the parties are strangers to each other.[195] If a public authority is already under a duty to assist, however, then misfeasance will cover its conscious decision to violate that duty by taking no action.[196] Negligence law would probably have imposed no liability on the constable in R v Dytham,[197] who had callously chosen to do nothing to help or summon help as he watched a man being beaten to death; his uniform would have made no difference there, but it would have made all the difference to a misfeasance action.[198]

These distinctions between negligence and misfeasance, therefore, go considerably beyond the difference between accidental and advertent risk-taking. Tort's pervasive fears of excessive liability are never far from the surface in misfeasance cases, but that particular tort lacks the usual tools for handling them. Without the interposition of a common law duty of care, and with actionability being possible for deliberate or consciously indifferent breach of statutory duties to assist or protect the public, the tort may yet show signs of strain. It currently relies on its very demanding mental elements to restrain liability, but if these prove insufficient, then the focus may well turn to a closer examination of the factors that might make indifference "reckless", to causation issues, and to the limits of vicarious liability.

IX. Rights, wrongs and duties

Reckless indifference in misfeasance law, therefore, is not about want of care because it has no predicate of a duty of care. Indeed, misfeasance can apply in the absence of any duty lying upon the defendant or correlative right inhering in the plaintiff, the most obvious example being the case where the defendant deliberately and unlawfully refuses to exercise a discretionary function. This was not always the case.

In Ashby v White[199] itself, Holt CJ's judgment was anchored to a theory of rights protection, Ashby's right to vote being cast in terms of a property right. Modern judges rightly find that entirely unconvincing,[200] and one might speculate as to whether even Lord Holt saw it as no more than a convenient device, designed to get him to his damages conclusion in an era when there could be no common law damages without rights. Many judgments still talk of abuse of "duty", but usually interchangeably with abuse of "power".[201]

In Australia, however, there is a troubling trickle of cases flowing from a Full Court decision in Tampion v Anderson,[202] which said quite plainly that the misfeasance tortfeasor must have owed the plaintiff "a duty not to commit the particular abuse complained of."[203] That was one of the reasons for excluding a misfeasance claim against counsel assisting a Board of Inquiry and its chairperson, where the gist of the complaint was that the defendants had deliberately exceeded their terms of reference in eliciting evidence from persons other than the plaintiff. In other words, if there was any misfeasance, its only compensable victims had been the third party witnesses, because only they were owed duties not to be improperly coerced.

The joint judgment in Mengel[204] quoted Tampion without apparent disapproval. Writing separately in Mengel, Brennan J was at pains to distance himself from Tampion's asserted need for an antecedent duty,[205] and Deane J was in general agreement with Brennan J.[206] The New South Wales Court of Appeal did not formally side with Brennan J in Leerdam v Noori,[207] but it came very close. Spigelman CJ said that misfeasance covers abuse of powers and duties,[208] and Macfarlan JA was disposed to say that if there must be a duty, then it need be owed only to the public generally, or perhaps to a class to which the plaintiff belongs.[209] Leerdam in its turn was responding to Cannon v Tahche,[210] which had lined up with Tampion. The issue in Leerdam and Cannon was whether misfeasance could be founded upon breaches by legal practitioners of their ethical duties towards a court or tribunal. If (as Cannon supposed) the "model litigant" obligations borne by government's legal representatives are purely ethical,[211] then it would follow that the defendant lawyers' alleged misbehaviour would not have constituted illegalities. Their misconduct would therefore be an insufficient basis for a misfeasance action unless (as discussed in Part XI below) the tort were to extend beyond illegality, to include abuse of position or influence.

That is not to deny the relevance of distinguishing duties from powers. A plaintiff's loss is more easily connected to a defendant's wilful breach of a tightly confined legal duty than to a defendant's wilful breach of a broad discretion. It will often be difficult to satisfy causation's prima facie "but for" test in cases where the defendant illegally chose not to perform a protective or interventionist duty for the plaintiff's benefit.[212] It will be even more difficult as the discretionary elements in the defendant's functions increase, and most difficult where the defendant has ruled the plaintiff out of consideration for the discretionary award of benefits in the future. In the case of lost future benefits that are contingent on the future exercise of a discretionary power, the court cannot determine how those discretions should be exercised.[213] To avoid usurping the administrative decision-maker's role, the court must confine itself to valuing the opportunities lost to the plaintiff.[214]

New Zealand agreed with Brennan J's decision (in Mengel) not to follow Tampion,[215] and Lord Hobhouse said in Three Rivers[216] that misfeasance in public office "does not, and does not need to, apply where the defendant has invaded a legally protected right of the plaintiff." As a purely descriptive matter, Lord Hobhouse was surely correct, and it appears that all but one of the other judgments came to the same view.[217] No misfeasance case, old or modern, has attempted a catalogue of the sorts of legal rights or interests that the tort might protect, the reality being that its "right" is no less (and possibly more) than the right to compensation for wilfully misusing state power.

There are some who regret the absence in misfeasance of a right/duty relationship between the parties. Those regrets are to an extent located in the broader debate between the corrective justice and distributive justice models of tort law,[218] but this is not the place (and I am not the author) to enter upon debates as to tort's taxonomy.[219]

X. Moral equivalence and the variability of "recklessness"

There is little if any need for doctrines of standing, proximity or remoteness in the case of targeted malice, where the moral case for full recovery is plain. The cases concerning the mental elements of the tort of misfeasance, however, now appear to have established a moral equivalence between on the one hand, consciously indifferent risk-taking (whether as to illegality or the potential for harm), and on the other, the knowing use of illegal means for the deliberate infliction of harm. That moral equivalence applies both to cases of power and duty.

It might now become necessary to set some additional criteria connecting the defendant and the plaintiff, and some limits to harms for which the defendant might be liable. The joint judgment in Mengel[220] said that the tort's reach must bear some correlation to the defendant's ability to pay, especially if (as it supposed was the case in Australia) the government might not indemnify its officers.[221] Lord Hobhouse spoke in Three Rivers of the need for a nexus between the defendant's state of mind and the type of harm for which he or she would be liable.[222]

The allegations against the Bank of England in Three Rivers were always highly implausible, but imagine if the plaintiffs had succeeded in establishing that at some point, the Bank had just dithered for some considerable time after it had reached the conclusion that the crooked bank it was meant to be regulating simply could not be saved. The only members of the class action were the failed bank's existing or potential depositors, but the Bank would have known that many others would also have lost money. Three Rivers rejected the Bank's call to require an antecedent legal right in a plaintiff, and it also rejected the Bank's argument for a "proximity" limit either to the class of plaintiffs or to their compensable losses. It held that these restraints were unnecessary in light of its requirement that a misfeasance tortfeasor must at the very least have consciously decided to risk inflicting probable harm on the plaintiff or on a class to which the plaintiff belonged.[223] At the same time, however, it hinted that some other limits might be needed in future, limits variously framed as requirements that there be standing to sue,[224] or special damage not suffered by the general public,[225] or membership of a class that the defendant was meant to have protected,[226] or a "direct" link between the defendant's misconduct and the plaintiff's loss.[227] It is difficult to place much faith in limits such as these. Judicial review's locus standi learning would be inadequate, especially in England which has in practical terms converted the lack of standing into no more than a discretionary bar.[228] Limiting the tort to damage caused "directly" would be equally vague.

The Court of Appeal declared that Three Rivers was mistaken in ever thinking that there was a problem. That was in Akenzua v Home Secretary,[229] whose facts[230] cried out for compensation. The estate of a murdered woman sued the defendant in misfeasance for her death. The authorities had twice secured her assailant's release from custody, because he was useful to them as a paid informer. He was at large at the time of the murder because the authorities had interfered with the proper handling of serious charges relating to drugs, offensive weapons, and rape, all of which they knew he had committed whilst improperly released. They had also known that he had entered the country illegally, but had procured a temporary visa to get over that problem. They had known that he was extremely dangerous, because he had told them that he had murdered people in his home country. In terms of Three Rivers, therefore, they had known of the probability of him committing a violent crime, and the issue was whether it mattered that the authorities could have had no idea who his next victim might be. The court held that any suggestion in Three Rivers that the defendant must have appreciated risk to a class of probable victims[231] was both obiter and misguided. It was morally misguided because it sought to draw lines that were "purely numerical",[232] lines drawn from negligence law, whereas the conduct of a misfeasance tortfeasor is "altogether more blameworthy".[233] The court said that misfeasance liability should apply equally to the bad faith release of three different types of murderers, namely, the man known to be intent on murdering his wife, Hannibal Lecter intent on murdering a few more than that, and a terrorist bent on a single act of mass destruction.[234]

The case for liability in all three instances supposed in Akenzua is indeed strong, but it fails to overcome the Three Rivers concern for articulating some limits to liability. Akenzua is almost unique in misfeasance law, which hardly ever sees claims for personal injury or death.[235] Misfeasance claimants typically assert purely economic loss, and fears of liability to an indeterminate class in an indeterminate amount for that type of loss are surely as valid there as in negligence claims.

To return to Lord Hope's implicit distinction between indifference and recklessness,[236] the concept of recklessness is not purely subjective. It includes an objective, normative judgment, which courts might need to use to set some limits to the extent of liability for purely economic loss. To return to Three Rivers, it would be easier to characterise as "reckless" the Bank's hypothetical indifference to depositors than its hypothetical indifference to sections of the wider community which would also suffer, even though the Bank's brief is wider than protection of depositors' interests.[237]

XI. Must the defendant be exercising a public function?

The moral case for a misfeasance tort is typically put as the need for protection from a deliberate abuse of state power over the subject, but in fact, the tort has always gone wider than the improper use of coercive power – it has always covered wilful refusals to perform duties to provide protection for the subject. In the old case of Henly v Mayor of Lyme,[238] it was a duty to keep a sea wall in good repair; in Three Rivers,[239] it was a duty to protect a trading bank's potential depositors (at least), and perhaps existing depositors; and in R v Dytham,[240] it was a duty to make some attempt to stop a murder taking place before the very eyes of an indifferent police officer.

The duty cases are harder to prove than the power cases, because a mere omission to perform a duty is not enough. The requirement of subjective fault means that defendants must either have decided not to perform a duty of which they were aware, or have consciously decided not to care whether they were under a duty.[241] Further, a discretionary power of protection will not become a duty unless the discretion has run out in the sense that no legally available reasons apply for declining to intervene.[242]

The duty cases have gone a step further in cases where the gist of the allegation is that the authorities have deliberately and illegally suppressed or distorted evidence, so that criminals go free to the distress of their victims and their families. The Canadian Supreme Court allowed such a claim to go forward in Odhavji,[243] leaving for later the difficult issue of whether the family had suffered tortiously recognised psychiatric harm. A cover-up of a police rape in Garrett[244] had effectively ruined the victim's reputation, because her assailant's lurid denials gained credibility for want of a proper investigation.

Misfeasance clearly draws no distinction between powers or duties according to whether their provenance is statutory or prerogative,[245] but uncertainty remains as to whether they must bear a "public" aspect and if so, how one might define that limitation. Current doctrine extends only to "public" officers, a limitation discussed separately below. In this Part, however, it is necessary to discuss whether the conduct of public officers to which the tort extends has an additional "public" aspect, defined perhaps in the same or similar terms that are commonly used to describe the reach of judicial review of administrative action.

Macfarlan JA explained his reasons for wanting a "public" aspect in terms that assumed that the tort might extend beyond public servants:[246]

"If the tort were not limited to the abuse of public powers and authorities, its scope would be wide indeed. There would be the potential for a multitude of actions to be brought by members of the public in relation to the conduct by public servants and public contractors of their day to day duties. This would in my view involve an unwarranted extension of the tort well beyond what has been treated thus far as its scope."

However, Slade LJ said in Jones v Swansea City Council[247] that what matters is not whether the power is public or private, but whether it is exercised by a public body. Jones was a tenant of commercial property that the Council owned. She alleged that the Council had imposed a restriction upon her use of that property out of spite and revenge. Slade LJ’s statement lost some of its force, however, because the court also said that the Council's property powers could be exercised only in the public interest.[248] The case went to the House of Lords, which did not rule on that issue.[249] It is now clear that the Human Rights Act 1998 (UK) can occasionally trump the property rights of local authorities.[250]

The parallel criminal offence of misfeasance or misconduct in public office is unconcerned with whether the power is public or private, focusing instead on the status of the public officer and the gravity of his or her misconduct. The gist of the statutory offence is the improper use of either the power or the influence that come with the position,[251] and the common law offence can be committed by serious misconduct that disgraces the office.[252] Brennan J said in Mengel that the misfeasance tort might flow from a deliberate abuse of "position or power".[253]

It is difficult to know what to make of Jones. In terms of principle, it is hard to justify a liability rule whose imposition turns solely on whether the defendant is public or private. In terms of precedent, it is also difficult to square with Calveley v Chief Constable of Merseyside Police,[254] in which the House of Lords held that a deliberately misleading internal affairs report about the plaintiff police officer could not be the basis of a misfeasance claim, because the investigating officer had neither needed nor used any special powers to make his report. The English Court of Appeal, however, followed Jones in Cornelius v Hackney London Borough Council,[255] even though Calveley has not been overruled. Cornelius held that misfeasance extends beyond abuse of power to abuse of public position, with the result that a council officer could be liable for blackening a whistle-blower's name with press releases he knew to be false.

Neasey J said in Pemberton v Attorney General[256] that a school teacher in the public sector could not claim in misfeasance for wrongful dismissal because the defendant was not exercising a "public" power, and the teacher was not a member of the "public", even though the dismissal power was wholly governed by statute law. The case is relatively old, and the other two judgments dismissed the misfeasance claim on the uncontentious ground that the defendant had acted in good faith.[257]

A Victorian court drew a line similar to that in Calveley, saying that so far as a Board of Inquiry exercised no special or coercive common law or statutory powers, but simply exceeded its terms of reference, it was not guilty of an abuse of "power" for the purposes of the tort, even if it had acted maliciously; it may have abused its position or influence, but not its power.[258] The Full Court of the Federal Court did not deal satisfactorily with the issue in Emanuele v Hedley,[259] a case involving the sale of government real estate. The court was prepared to assume that misfeasance could catch a government officer's conduct in touting the sale, but ruled out the same officer's internal report to his superior of his subordinate's misconduct.

Decisions such as Odhavji and Garrett prompt speculation as to the assumption made in most other cases that the exercise of the power or refusal to perform the duty must occur in circumstances warranting a judicial review declaration of invalidity. Invalidity was a critical requirement in Three Rivers and Mengel, but it would be a stretch to describe the deliberate suppression of evidence as "invalid" conduct. Invalidity is needed to repel a defence of lawful authority,[260] and that will usually be the case where (as in Three Rivers) the breadth of the defendant's discretionary power is the plaintiff's chief obstacle. Invalidity, however, is only one way of establishing unlawfulness. According to Lord Hobhouse in Three Rivers, it is also established "from a straightforward breach of the relevant statutory provisions".[261] Invalidity is clearly irrelevant to a claim that police have breached their statutory obligation to launch (Garrett) or co-operate with (Odhavji) an internal inquiry, or have wilfully refused to preserve the peace, in the knowledge that the plaintiff will suffer in consequence.[262] It is submitted that for misfeasance to apply in those circumstances, the question is whether the defendant's illegality consisted in breach of a requirement that applied only to public officials. Hitching misfeasance to invalidity in these cases is no more than an unintended hangover from the failed campaign to extend the negligence action to loss caused by careless invalidity.

Regardless of whether misfeasance applies only to the performance of public functions, a question remains as to whether it can apply to the judiciary. New Zealand's Court of Appeal held that misfeasance extended beyond the abuse of administrative functions to include misconduct of a District Court judge,[263] and Victoria's Court of Appeal was inclined to agree.[264] The issue is unlike to arise very often, because most judges now enjoy complete immunity from tort actions regarding their official conduct.[265]

XII. The defendant must be a public officer

The cases require not just that defendants be guilty of an abuse of a public power, duty or possibly office, but also that they have been public officers. The precedents undoubtedly support this additional requirement, but it is otherwise difficult to justify. Its abolition would accord with the tort's moral basis that the function is public in the sense that it must be exercised with the public interest in mind, and probably that it be a function which members of the public generally have no power or responsibility to perform. Abolition, however, is an option available only to the highest courts,[266] who have declared that the defendant must be a public officer.[267] All other courts must require that both the power and the person who exercises it be relevantly "public".

There are many cases discussing the extent to which private bodies or individuals may be subject to the processes of judicial review if they can be said in some sense to be performing public functions. Speaking generally, Australia has yet to resolve the issue of a government contractor's susceptibility to the processes of judicial review,[268] whilst English law occasionally extends those processes beyond institutionally public bodies to persons whose powers are functionally public.[269] The issue is often one of mere process in administrative law, as where, for example, private parties might be contractually obliged to accord procedural fairness even though they might not be subject to judicial review. In misfeasance, however, the issue is likely to be more important, because its principles currently apply only to defendants who not only exercise public power but are also public officers.

Lord Steyn said that "office" should be understood in a relatively wide sense",[270] and Lord Hobhouse said that it "is a broad concept",[271] applying to "those vested with governmental authority and the exercise of executive powers".[272] There is very little authority on whether a person or body is a public officer for the purposes of misfeasance. Buxton LJ said in Society of Lloyd's v Henderson[273] that this was because the answer is usually obvious.[274] His Lordship said that the critical points were whether the defendant was exercising powers for a public, governmental purpose as opposed (in that case) to Lloyds, which existed solely for profit.

Brennan J referred to an old definition of public officers, which in essence contained two elements. First, they must have been appointed to perform a public duty. Secondly, they must be remunerated, although that may come in the form of money or land from the Crown, or fees from the public.[275] The second element would not apply nowadays;[276] it reflects a bygone era when public offices were treated as property[277] that could be bought and sold, an era which had a very different conception of corruption.

Not all public servants fall within the scope of the misfeasance tort, even though their salaries come from public funds, and even though they may be "officers" for administrative purposes and have to take an oath on appointment. A person might be a public employee but not a public officer.[278] There is in fact no single definition of "public officer" across all contexts.[279]

Lawyers from the private profession increasingly conduct court and tribunal work for government clients, and some concern has been raised as to whether their contractual status should make any difference. It is unlikely that a person whose only powers and responsibilities flow from contract would be treated as a public officer for misfeasance purposes.[280] In Australia, this would appear to be one of two reasons for holding that lawyers drawn from the private sector are not amenable to the misfeasance tort, the other (and more straight forward) reason being that they are unlikely to be exercising specifically public powers or duties.[281] Hodgson JA concluded that the contractor must not only be carrying out "some task", but also "himself or herself hold a public office".[282] In contrast to these Australian decisions, it was suggested that officers working for England's Crown Prosecution Service might be liable in misfeasance,[283] a suggestion distinguished in Australia on the ground that the CPS is a creature of statute with statutory functions.[284]

The decision of the Full Court of the Federal Court in Emanuele v Hedley,[285] presents a problem. In brief, the alleged misconduct was criminal entrapment to pay a bribe, an entrapment authorised at very senior levels within the public sector. The court said that it was "a legal nonsense" to suggest that the Commonwealth itself could commit misfeasance, because it could act only through agents, and because it does not itself hold public office. The first reason would suggest that institutions could only ever be liable vicariously and never as a principal through their agents, a suggestion at odds with decisions of the Privy Council and the House of Lords.[286] It is true that only natural persons have mental states, but it is no nonsense to suggest that a principal has in law the mental states of its agents, or an organisation the mental state of the individuals on its governing body. The second reason puts form over over substance. Of course the Commonwealth "holds" no "office", but that is to return to a conception that made sense when offices were seen as ways of making money.

It would be preferable to rebadge the tort as abuse of public power (or position, if it were to extend beyond instances of public law invalidity); the deliberate abuse of public power is no less serious if committed by either a public employee who holds no office or a government contractor performing a governmental function. The tort's concern is with conscious maladministration, which is a wrong that can be committed by people high and low in the public sector and by some contractors to the public sector. Any concern about imposing misfeasance liability on low-level is misplaced. Low-ranking staff are unlikely to have enough power to abuse; they are even less likely to know whether they are abusing it; and if they are consciously indifferent about that, they are highly unlikely to be adjudged reckless for having decided to give no further thought to the legal limits to their power.

XIII. Vicarious liability issues: a public purse for a public tort?

Unlike their individual members, companies and public bodies have no minds. That does not prevent public bodies from being directly liable for misfeasance. In the right circumstances, the mental state of their staff can be imputed to the organisations themselves. For example, if all the members of a local government council voted unanimously to cancel a land-use development consent, and if they did that entirely out of a sense of revenge and self-interest, and in the knowledge that they were acting illegally, then they will have been guilty of bad faith sufficient for the purposes of misfeasance. However, they will not as individuals have caused the property developer's loss, because only the council itself could cancel the development consent. In such a case, the council will be fixed with the mental state of the individuals comprising its governing body, and will be directly rather than vicariously liable in misfeasance for the harm it caused.[287] In many misfeasance cases, however, only individual staff members will be directly liable because causal responsibility and the requisite mental states resided only in them. The issue then becomes whether the public bodies for which they worked can be fixed with vicarious liability.[288]

Two concerns have been raised concerning the possibility of vicarious liability. Because the minimum requisite mental state comes close to dishonesty, and because illegality is required, doubts have arisen as to whether the misconduct of misfeasance tortfeasors can properly be regarded as an incident of their employment. Further, many (if not most) of the individual tortfeasors will have been abusing a statutory or common law function invested directly in them by virtue of their office or position, with the result that they will have been performing discretionary functions free of their employers' control or direction. The common law does not impose vicarious liability on employers where the relevant tort was committed in the exercise of an independent discretion. The second concern is more formidable than the first.

There have long been difficulties in formulating the basis of vicarious liability for deliberately illegal conduct committed without the employer's de facto authority or ratification.[289] The difficulties increase where the primary tortfeasors act in their own interests and against those of their employers. The courts have recently propounded different tests for resolving claims against schools for their teachers' sexual abuse of students, tests turning on whether the illegal act bore a sufficiently close connection to the employee's job, or whether the tortious conduct was no more than an unauthorised mode of performing authorised acts.[290] Vicarious liability for misfeasance, however, should present fewer problems than in the sexual assault cases. The individual tortfeasors' conduct in cases of targeted malice would usually have been lawful if only their motives had been proper; the acts themselves will have been of the type that those individuals could usually have performed with propriety. The same applies to most other instances of misfeasance. The absence of targeted malice diminishes the tortfeasors' impropriety, but their misbehaviour will usually be of the kind that they could have performed with propriety. And if the test be whether the misconduct was "closely connected" to the misfeasance tortfeasors' jobs, then once again, it will usually be easy enough to establish vicarious liability. Indeed, the sort of misconduct alleged in most misfeasance cases can only be committed "on the job". As a private individual, I can neither cancel a trading bank's licence (Three Rivers) nor impose regional restrictions on cattle droving (Mengel).

A more difficult problem arises from the common law protection of employers from vicarious liability for tortious behaviour that occurs in the exercise of an independent discretion. That protection is difficult to defend in policy terms, and has the potential to be raised in many (if not most) cases against public officers, whose discretionary powers are typically independent.[291] Statute abolished the protection in England,[292] New South Wales,[293] New Zealand, [294] and Canada.[295] It is unclear whether that protection remains elsewhere in Australia. South Australia used to have an Act that abolished most of the protection,[296] and it seems to have been assumed that when that Act was replaced by a more modern and slimmer version that made no mention of the independent discretion rule,[297] the common law rule was not revived.[298]

The English cases appear to have accepted that individual officers will transmit vicarious liability for their misfeasance torts.[299] The Law Commission for England and Wales wanted to abolish misfeasance entirely.[300] It believed that individual liability for misfeasance was inappropriate and (in the Three Rivers case), intrusive and distressing. Its preference was that government alone should be liable, and that the criminal offence of misconduct could be used to achieve any necessary disciplinary or denunciation effect on individual miscreants.

The Australian position presents a marked contrast between doctrine and practice. The High Court has not explored the issue,[301] but its analysis in Mengel[302] of the tort's structure and principles proceeded on the premise that "ordinarily", individual misfeasance tortfeasors would receive no indemnity or contribution from their employing authorities.[303] One must doubt, however, whether that accurately describes the position. There are no official figures, but as far as I am aware, governments have always borne the defence costs in misfeasance claims, and have also paid the bills in those few cases which they have settled or lost. Indeed, it would be surprising if the position were otherwise. Few (if any) public servants would have professional indemnity insurance, and one might doubt whether PI insurance would respond to misfeasance. If the financial burden of misfeasance liability were indeed "ordinarily" to remain with the individual, the larger claims of recent times would never have been brought. Billions of pounds were claimed in Three Rivers. More recently, the Australian government settled two individual claims and a related class action regarding the misfeasance of its pharmaceuticals regulator for a total of roughly $127.5m.[304] Those Australian settlements appear to have conformed to the government's published guidelines, which basically offer the prospect of complete indemnities for public servants who act responsibly, and without "serious or wilful misconduct or culpable negligence".[305]

There appears to be a similar gap between doctrine and practice in the case of police torts. All Australian governments bear sole responsibility to plaintiffs who sue for police torts, but with the exceptions of the Commonwealth and Queensland, the police must have been acting honestly or in good faith.[306] In practice, however, governments appear to have conceded vicarious liability for police torts in circumstances where they might have been able to establish bad faith.[307] Perhaps the police associations have wrung this concession from their governments but if so, it would appear to be no more than the concessions wrung by the broader union movement from employers more than a generation earlier. In the mid-1950s, Lister v Romford Ice & Cold Storage Ltd[308] had allowed an employer whose tort liability to the plaintiff was wholly vicarious to seek a full indemnity from its employed tortfeasor. However, legislation in some places has overturned Lister on the proviso that the employee was not guilty of serious or wilful misconduct.[309] On the same proviso, Australian insurers have no right of subrogation to an employer's rights against its employee.[310] Whilst gaps clearly remain, it appears that governments are cautious about how they are used.[311]

These matters are mentioned here not to question the obvious (albeit complex) inter-dependence of tort and insurance, but to challenge Mengel's assertion that things are (or should be) different for misfeasance – that individual defendants are (or should) "ordinarily" be unable to obtain indemnities from their governments. It is submitted that the assertion is not just descriptively dubious, but also normatively. Back in 1963, the Privy Council said, in effect, that the misfeasance tort was one part of the judiciary's answer (along with an expansion of judicial review) to the growth of the administrative state.[312] If that is right, then the tort is more about holding the state to account than its individual officers, and if that last proposition is right, then the independent discretion rule should have no place in the tort's operation. If misfeasance is a public tort, its damages should come from the public purse.

XIV. Suggestions

Articles commonly start with Introductions and end with Conclusions, but in the case of misfeasance, "Conclusions" might be misleading; the cases have too many loose ends. What follows is a list of suggestions, some more tentative than others, as to how some of the larger issues should be resolved.

This article has not touched on what is perhaps the largest issue, which is whether it might be best to stop calling misfeasance a tort. It would certainly be an attractive law reform option to introduce greater flexibility into its assessment of damages. Small sums could be available for "vindicatory" purposes where the plaintiffs are secondary or more remote victims and their losses are purely economic, and even then, perhaps only where the state has failed to undertake genuine disciplinary and remedial action of its own.[313] Reform along those lines would need to be achieved by legislation, but it would be a risky venture in the current political climate. It could end up granting public authorities the same protections from their primary victims as from their secondary victims, and even apply equally to claims for personal injury and pure economic loss.

On the assumption that misfeasance remains as a tort, one must then ask whether it should continue to be a purely public law tort, available only against public actors. Misfeasance responds to malice, but only when that comes from public actors, the underlying rationale being that public actors are different, as indeed they usually are. Destructive motivations towards competitors are regarded as an inherent part of a free market, and the common law has left their regulation to the legislature. However, it is going too far to say that only public actors need to put their personal feelings to one side because that fails to consider the position of trustees who must also act altruistically. It would nevertheless be a non sequitur to say that one should therefore attempt a merger of the rules for the pecuniary liability of trustees and public officials. In the absence of evidence that trustee liability is radically dysfunctional, the real issue is not whether trustees and public actors must both act altruistically, but the sufficiency of the sanctions against public actors who behave maliciously. Whilst public actors are subject to judicial review and the regular torts, and to a range of disciplinary measures that are not court-based, misfeasance performs valuable compensatory and punitive functions beyond the capacity of other remedies. In virtually all of the cases considered in this article, judicial review would have been a meaningless remedy from the plaintiffs' perspective because they had already suffered irreparable damage. Furthermore, it seems highly likely that the individual tortfeasors in most cases were not subjected to internal disciplinary measures. Ultimately, public power is exercised in our name, and without misfeasance liability, public actors (and their governments) would get away with its deliberate abuse to inflict loss that they know is wrong. Misfeasance therefore has a good claim to existence as a public law tort.

It is suggested that two consequences should flow from an unequivocal acceptance of misfeasance as a public tort. First, there can be no excuse for denying government's vicarious liability for the misfeasance torts of its officials. Indeed, one could make good policy arguments for legislative intervention to impose liability solely upon the government except in the case of harm caused deliberately with knowing illegality and for reasons personal to the individual tortfeasor. The state's vicarious liability is needed to make the tort meaningful. The common law's "independent discretion" rule currently stands in the way of vicarious liability in most places in Australia, but that is a rule which seems not to have been applied in practice in misfeasance cases, and is in any event ripe for repeal in those jurisdictions where that step remains to be taken.

The second consequence flowing from the tort's public law status is perhaps more controversial, but it is nevertheless suggested that the tort should apply as much to government contractors who abuse public power as it does to those who are on the full-time public payroll. The essence of misfeasance is surely that it is a deliberate abuse of public power, and it should be no excuse that a particular defendant is not subject to the internal disciplinary processes of the public service. If anything, that should be seen as an argument for liability, because there are fewer alternative remedies against the contractor.

Further, there can be no plausible defence of some of the fine distinctions between public servants who are or are not "holders" of "public office". Those distinctions open up a gap between public power and tort responsibility for its deliberate abuse that may have been justifiable in the days when holding office was a way of making money. Those days have gone, as should the distinctions which went with them in this context. There is a legitimate concern to avoid imposing liability on low-level public servants who simply cannot be expected to know the law, but the tort's mental elements provide adequate protection. Low-level public servants who do not know the legal limits to their power will be liable only if they were reckless in consciously adverting to the risk of illegality and then deciding not to care about that issue. There will be very few low-level public actors who consciously advert to such a risk, and even fewer whose decision not to care could be described as reckless.

Misfeasance is designed to redress deliberate abuse of "public power", but the cases are in some disarray as to how that might be defined. The tort clearly extends to wilful refusals to perform public duties, but there is an issue as to whether it extends (and if so, how far) beyond the sorts of powers and duties that are judicially reviewable. It is suggested that the tort must travel beyond the reach of judicial review – it can be a useful supplement to judicial review, but that should not be its defining feature. Like its criminal law forebear, the misfeasance tort should apply to abuses of public power[314] or position. There can be no difference between an elected local government authority that adversely rezones land to get back at its political rivals, and the same authority that determines its leases to those rivals for the same reason.

The great majority of the modern cases have focused their attention on the tort's requisite mental elements, but even in that area, some issues need clarification. The cases have now relaxed the original requirements of deliberate illegality and deliberate harm, introducing the lesser alternative of reckless indifference provided that it, too, was deliberate. The minimum requirement of subjective indifference applies in most places as much to the prospect of harm as to the risk that the action taken will be illegal. However, Australia currently favours plaintiffs in requiring conscious indifference to the illegality issue but not to the prospect of harm, which need only be foreseeable. It is suggested that there is no sound argument for maintaining different mental elements for the illegality and harm components. As an intentional tort, misfeasance should set equally demanding fault standards at both points. Fault might occasionally be negated, however, where defendants bend the rules at the plaintiffs' request.

The cases have yet to decide whether defendants who are guilty of targeted malice should escape liability if they neither knew nor subjectively suspected that they were breaking the law. It is suggested that there should be no misfeasance liability in such cases. A subjective intent to inflict loss is part of the job description for some public officers (police and the managers of immigration detention centres are the most obvious examples), and there is no good reason to make them assume the risk of accidental illegality. I am aware that this entails the prospect of Mr Duplessis beating a misfeasance claim if Mr Roncarelli were to sue him today, but only if the court were to believe that Duplessis neither knew nor suspected his want of lawful power. Defendants who intentionally harm their plaintiffs should bear at least an evidential burden of proof.

Finally, it is reasonably clear that if misfeasance vindicates "rights", they are rights in only the loosest sense of the term – the political right to be free of deliberate abuse of public power. That admittedly raises difficulties for a tort that has foresworn some of the constraints so well-known to negligence law – limits designed to avoid the risk of indeterminate liability to an indeterminate number of plaintiffs for purely economic loss. The only protection from that risk that this article has identified lies in the requirement that where liability is premised on conscious indifference, that indifference must also be reckless. A requirement of recklessness has elements that are both objective and subjective.

---

[∗] Law Faculty, University of New South Wales. I am grateful for the advice and assistance I received from Roderick Bagshaw, John Basten, Erika Chamberlain, Jim Davis, Matthew Groves, Carol Harlow, Susanna Khouri, Keith Mason, Martin Matthews, David Mullan, Jason Neyers, Robert Stevens, Prue Vines, Greg Weeks, Lucia Zedner, and this Journal's anonymous referees. My thanks go also to Corpus Christi College, Oxford, for generously providing me with a Visiting Fellowship for a term, during which I was able to write the bulk of this article.

[1] Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330 at 376 [124], Gummow J.

[2] It is therefore an exception to a fundamental Diceyan principle of holding government to the same liability principles as apply to private individuals. See A V Dicey, The Law of the Constitution (1st ed, 1885, Macmillan & Co, London) pp 178-179 and 200-201. The principle is necessarily aspirational, because "perfect equality is not attainable"; Graham Barclay Oysters Pty Ltd v Ryan [2002] HCA 54; (2001) 211 CLR 540 at 556, Gleeson CJ.

[3] R v Secretary of State for Transport; Ex parte Factortame Ltd (No 2) (1990) UKHL 13; [1991] 1 AC 603 at 672; X (Minors) v Bedfordshire County Council [1995] UKHL 9; [1995] 2 AC 633 at 730; and Three Rivers District Council v Bank of England (No 3) [2003] 2 AC 1 at 229.

[4] See, eg, P Cane, "Remedies Available in Judicial Review Proceedings" in D Feldman (ed) English Public Law (OUP, Oxford, 2004) 915 at 949 : "The absence of a right of damages for losses sustained as a consequence of public law wrongs is widely recognised as one of the most serious of the remaining gaps in our remedial system. It is a gap that does not exist in more developed systems." M Sunkin assumed responsibility for that chapter in the second edition (OUP, Oxford, 2009), and added (793 at 820) the following to Cane's observation's: "This gap has been widely criticised over the years by judges, by legal commentators, and by the Law Commission. This is an issue that now cries out for reform." (References omitted from each quotation.)

[5] The prominent report of the Justice-All Souls Committee, Administrative Justice – Some Necessary Reforms (1988) proposed legislative reform with no general limits. See also Law Commission, Administrative Redress: Public Bodies and the Citizen – a Consultation Paper (Consultation Paper No 187, 2008), (Administrative Redress) at [4.27]-[4.30] for references to other criticisms of the limited availability of damages when administrative authorities act invalidly.

[6] All recent legislative interventions in Australia have restricted the common law liability of public authorities; see M Aronson, "Government Liability in Negligence" [2008] MelbULawRw 2; (2008) 32 Melbourne University Law Review 44

[7] eg, Human Rights Act 1998 (UK), s 8(3)(b) – "just satisfaction".

[8] Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330 at 376 [124], Gummow J.

[9] The Australian government even outsourced the assessment of some claims for asylum: Plaintiff M61/2010E v Commonwealth [2010] HCA 41; (2010) 272 ALR 14 at [51].

[10] (1995) 185 CLR 307.

[11] [1997] 2 NZLR 332. The facts given here are taken from the pleadings, the misfeasance jury's answers to a number of specific questions, and the Court of Appeal's inferences from the way that plaintiff's counsel had conducted the claim. The constable was taken to have raped the plaintiff, although he had left the country and was never brought to trial.

[12] [2003] 2 AC 1.

[13] [2003] 3 SCR 263.

[14] (1703) 2 Ld Raymond 938; 14 How St Tr 695; 1 Smith's LC (13th ed, 1929, by T W Chitty, A T Denning and C P Harvey) 253 (Smith's LC). It is generally acknowledged that various reports of the case (including three by Holt CJ himself), and of its aftermath in the House of Commons and the House of Lords, are patchy. I will use the report in Smith's LC, partly because of its accompanying essay. This appears to be the favoured report in the UK (Three Rivers n 12 at 190; and Watkins v Secretary of State for the Home Department [2006] 2 AC 395 at 416), whereas the Australian High Court usually uses Lord Raymond's report, which is now freely available online through Commonlii.org.

[15] Most accounts of the misfeasance tort start with Lord Holt CJ's judgment, although it has also been traced to common law cases on the liability of judicial officers in inferior courts; see R P Balkin and J L R Davis, The Law of Torts (4th ed, 2009, LexisNexis, Sydney) 705, and that element certainly figured prominently in some of the judgments in Ashby itself. Lord Steyn in Three Rivers n 12 at 189-190 went back to Turner v Sterling [1726] EngR 585; (1671) 2 Vent 25, which had allowed a claim that the defendant had maliciously refused to accept the result of a vote for installing the plaintiff to a vacancy in the remunerated office of a London bridgemaster; his Lordship went on to say that Ashby provided "the first solid basis for this new head of liability": at 190.

[16] The case is part of English constitutional history's canon. Ashby and a number of other voting cases sparked what 1 Smith's LC called (at 281) "one of the most furious controversies between the Houses of Lords and Commons of which there is any example in English history". The plaintiffs and most of their lawyers were arrested on warrants from the Speaker, the Commons believing that the matter was within its sole jurisdiction. One lawyer escaped, using sheets and a rope to climb down from a rear window in his second floor chambers in the Temple. The dispute between the two Houses came to an end only with the prorogation of the Parliament. Ashby was a poor cobbler backed by the Whigs; the Tories backed the constables: Watkins v Secretary of State for the Home Department [2006] 2 AC 395 at 416, per Lord Rodger, citing E Cruikshanks, "The case of the men of Aylesbury 1701-4" in C Jones (ed), Party and Management in Parliament, 1660-1784 (Leicester University Press, New York, 1984) 87.

[17] (1703) 1 Smith's LC 153 at 276.

[18] 1 Smith's LC 253 at 277.

[19] [2006] 2 AC 395.

[20] Cf de Montigny v Brossard (Succession) 2010 SCC 51, upholding the availability of punitive damages for Charter breaches in the absence of material loss.

[21] n 19 at 403 [8].

[22] (1998) 196 CLR 329 at 344 [37], per Gleeson CJ, Gaudron, Kirby and Hayne JJ.

[23] n 19 at 403 [9], Lord Bingham. Cf C Harlow, "A Punitive Role for Tort Law?", in L Pearson, C Harlow and M Taggart (eds) Administrative Law in a Changing State (Hart Publishing, Oxford, 2008) (Harlow) 247.

[24] Lord Millett said in Three Rivers n 12 at 292 [188] that misfeasance was not made out by proof of "honest incompetence" or "gross negligence". Gummow J said in Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330 at 376 [124] that misfeasance "concerns conscious maladministration rather than careless administration". The Court of Appeal upheld the defence of honest but gross incompetence in Muuse v Secretary of State for the Home Office [2010] EWCA Civ 453.

[25] Mengel n 10 at 358, Brennan J; Garrett n 11 at 350; and Sanders v Snell (1998) 196 CLR 329 at 344 [37]. Lord Steyn added in Three Rivers n 12 at 196 that one should avoid a liability rule that would see public officers being "assailed by unmeritorious actions", but that begged the question as to when an action would be unmeritorious.

[26] Mengel n 10 at 347, Mason CJ, Dawson, Toohey, Gaudron and McHugh JJ. See below, Part XIII.

[27] Cf Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330 at 376 [124], where Gummow J said that the tort's roots lay within public law.

[28] The proposition first appeared in the fourth edition of H W R Wade, Administrative Law (Clarendon Press, Oxford, 4th ed, 1977) at 340-341, and had become classic by the time that appeared unaltered in H W R Wade and C F Forsyth, Administrative Law (OUP, Oxford, 10th ed, 2009) at 296-297.

[29] L Smith, "The Motive, not the Deed", in J Getzler (ed), Rationalizing Property, Equity and Trusts: Essays in Honour of Edward Burn (LexisNexis, London, 2003) 53 at 67.

[30] P D Finn, Fiduciary Obligations (Law Book Co, Sydney, 1977) 15.

[31] n 11 at 235-236.

[32] See Jones v Swansea City Council [1990] 1 WLR 54 at 71 (Slade LJ); and Sanders v Snell [2003] FCAFC 150; (2003) 130 FCR 149 at 173 [89].

[33] Mengel n 10 at 348;

[34] Sanders v Snell (1998) 196 CLR 329 at 344 [36]-[37] and 345-346 [39].

[35] Mengel n 10 at 359.

[36] Lord Steyn said in Three Rivers n 12 at 190 that misfeasance needed to fit within the "general scheme of the law of tort", in order to ensure the "coherent development of the law".

[37] See also E Chamberlain, "Misfeasance in Public Office: a Justifiable Anomaly to the Rights-Based Approach?", in A Robertson and D Nolan (eds), Rights and Private Law (Oxford, Hart Publishing, 2011, forthcoming).

[38] Statutes protecting officials who acted unlawfully but in good faith used to be extremely common in England, but electronic searches suggest that they might now be less common. They remain standard drafting policy in Australia. Section 136N of the Policing and Crime Act 2009 (UK) is probably unique. It applies to all police acting in good faith in the performance or purported performance of their functions relating to the enforcement of orders to close certain premises. The section provides that no police or police force shall be liable in damages "in proceedings for judicial review or for the tort of negligence or misfeasance in public office".

[39] See Mengel n 10 at 356 per Brennan J, adopting the view expressed in 1 Smith's LC 153 at 283. The New Zealand Court of Appeal agreed with Brennan J: Rawlinson v Rice [1998] 1 NZLR 454 at 458. In Tozer v Child [1856] EngR 450; (1856) 6 El & Bl 289 at 295, Lord Campbell CJ, Wightman, Erle and Crompton JJ interpreted Ashby as treating malice as an essential element of actionability.

[40] Cf Bradford Corporation v Pickles [1895] UKHL 1; [1895] AC 587 at 601, where Lord McNaghten could see no malice in Mr Pickles' conduct, even if it was "churlish, selfish and grasping", or morally "shocking". Mr Pickles had diverted underground water flowing in no defined course from entering the city's water catchment, presumably to drive a hard bargain. For a book-length case study, see M Taggart, Private Property and Abuse of Rights in Victorian England (OUP, Oxford, 2002). Taggart was unsure about heaping all of the criticisms upon Mr Pickles, noting (at 72) that the Town Clerk had also been quite intransigent.

[41] See: Bradford Corporation v Pickles [1895] UKHL 1; [1895] AC 587 at 594, 598-599 and 601; Allen v Flood [1898] AC 1 at 92 and 152-153; McKernan v Fraser [1931] HCA 54; (1931) 46 CLR 343 at 380; Crofter Hand Woven Harris Tweed Co Ltd v Veitch [1941] UKHL 2; [1942] AC 435 at 442; O'Brien v Dawson [1942] HCA 8; (1942) 66 CLR 18 at 28; Sanders v Snell (1998) 196 CLR 329 at 342; Perre v Apand Pty Ltd [1999] HCA 36; (1999) 198 CLR 180 at 306; Three Rivers n 12 at 190; and OBG Ltd v Allan [2007] UKHL 21; [2008] 1 AC 1 at 22.

[42] Discretionary functions were usually labelled judicial or quasi-judicial. Non-discretionary functions (ie, duties to act, as opposed to duties to consider how to act) were usually labelled ministerial.

[43] See S A de Smith, Judicial Review of Administrative Action (Stevens, London, 1959) 201, citing Everett v Griffiths [1921] 1 AC 631 at 695.

[44] See: J W Salmond, The Law of Torts (2nd ed, 1910, Stevens and Haynes, London), 478-480; J F Clerk and W H B Lindsell, The Law of Torts (6th ed by W Paine, Sweet and Maxwell, London, 1912) Chapter 22; and M Aronson and H Whitmore, Public Torts and Contracts (Law Book Co, Sydney, 1982) 138-147. The propositions in the text are stated with some diffidence. This corner of the law was never neat, and the leading cases struggled to accommodate several principles that pulled in different directions, some dealing with the provision of damages for breach of duty, or for breach of statutory duty, and others dealing with exceptions in relation to discretionary or judicial duties. Retreat was inevitable in the face of the welter of statutory duties being imposed upon both private and public actors; see P D Finn, "Public Officers: Some Personal Liabilities" (1977) 51 Australian Law Journal 313; and P D Finn, "A Road not Taken: the Boyce Plaintiff and Lord Cairns' Act" (1983) 57 Australian Law Journal 493 (Part I) and 572 (Part II) (Finn, The Boyce Plaintiff).

[45] See: Everett v Griffiths [1921] 1 AC 631; Sirros v Moore [1975] QB 118; Maharaj v Attorney General (Trinidad and Tobago) [No 2] [1978] UKPC 3; [1979] AC 385 at 404; Re McC [1985] AC 528; Attorney General (NSW) v Agarsky (1986) 6 NSWLR 38 at 40; Rajski v Powell (1987) 11 NSWLR 522 at 527-530; and Fingleton v R [2005] HCA 34; (2005) 227 CLR 166 at 184-187 and 214.

[46] Brennan J put these issues aside in Mengel n 10 at 355-356. Member states of the EU (but not their individual officers) are liable when their supreme courts commit manifest and inexcusable breaches of EU law: Köbler v Republik Österreich [2004] QB 848.

[47] See, for example: Judicial Officers Act 1986 (NSW), s 44A (granting all judicial officers the same "protection and immunity" afforded to Supreme Court judges); and Courts Act 2003 (UK), s 32(1) (actions lie against justices of the peace acting beyond or without jurisdiction only on proof of bad faith). The legislation in New South Wales has taken care to include protection for conduct in the performance of "ministerial duties": ss 44A-44C. Statutory protection for inferior court judicial officers used formerly to vary according to whether they had acted in good faith. See Constables Protection Act 1750 (24 Geo 2, c 44); and Justices Protection Act 1848 (11 & 12 Vict, c 44).

[48] Three Rivers n 12 at 190.

[49] Three Rivers n 12 at 191.

[50] Three Rivers n 12 at 190.

[51] L H Hoffmann, "Rookes v Barnard" (1965) 81 Law Quarterly Review 116 at 138.

[52] L H Hoffmann, "Rookes v Barnard" (1965) 81 Law Quarterly Review 116 at 140-141. Lord Devlin entertained a similar possibility: Rookes v Barnard [1964] UKHL 1; [1964] AC 1129 at 1215-1216.

[53] Discussed in M Taggart, Private Property and Abuse of Rights in Victorian England (OUP, Oxford, 2002), pp 144-149.

[54] The unsatisfactory tort of conspiracy by lawful means was left untouched, as to which, see R Bagshaw, "Can the Economic Torts be Unified?" (1998) 18 Oxford Journal of Legal Studies 729 at 731.

[55] OBG Ltd v Allan [2007] UKHL 21; [2008] 1 AC 1 at 22 and 34. Lord Nicholls thought (at 54-55) that these were complex issues best left to statute, the competition regulators and the industrial relations regulators. Australia had rejected the prima facie tort option a few years earlier: Sanders v Snell (1998) 196 CLR 329 at 342; Perre v Apand Pty Ltd [1999] HCA 36; (1999) 198 CLR 180 at 306; and Zhu v Treasurer of NSW [2004] HCA 56; (2004) 218 CLR 530 at 566 and 570-571. A prima facie tort theory presumes the tortiousness of harms caused deliberately or maliciously unless the defendants can persuade the court that their conduct was justified.

[56] (1703) 1 Smith's LC 253 at 276.

[57] Measured in terms of average wages, the Bank of England's rough estimate was that £200 in 1694 (when the Bank was founded) equated to £80,000 in 1994: H MacFarlane and P Mortimer-Lee, "Inflation over 300 years", in Bank of England Quarterly Bulletin, May 1994, pp 156-162, accessed on 30 January 2011 at http://www.bankofengland.co.uk/publications/quarterlybulletin/qb940201.pdf. If from 1994 one uses the the index for average weekly earnings in the Bank’s quarterly Inflation Reports (published online since 1993), the overall result is that £200 in 1694 equated to roughly £148,500 by the end of January 2011.

[58] As if regulators, tribunals and administrative agencies need not be considered. Gleeson CJ, McHugh, Gummow and Hayne JJ said in Gray v Motor Accident Commission (1998) 196 CLR 1 at 7-8, that the old intermingling of criminal and civil law has proceeded apace in recent times, with statutory schemes for criminal injury compensation and civil penalties. Tribunals and regulatory agencies typically administer such schemes, although appeals and judicial review lie to the superior courts.

[59] The author is indebted to Harlow n 23. See also P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press, Oxford, 1996): chapters 1 (P S Atiyah, "Personal Injuries in the Twenty-First Century: Thinking the Unthinkable"), 7 (A Burrows, "Reforming Exemplary Damages") and 13 (P Cane, "Exceptional Measures of Damages: a Search for Principles").

[60] [2008] UKHL 25; [2008] 1 AC 962.

[61] Lord Scott said that the assault and battery claim was legitmately rights-centred, not loss-centred: [2008] UKHL 25; [2008] 1 AC 962 at 976 [22]. Lumba v Secretary of State for the Home Department [2011] UKSC 12 held by majority that there should be no separate measure of "vindicatory" or "conventional" damages that were not nominal, compensatory or exemplary.

[62] Kuddus v Chief Constable of Leicestershire [2001] UKHL 29; [2002] 2 AC 122 (Kuddus).

[63] Watkins n 19. New Zealand does not require material damage: Donselaar v Donselaar [1982] NZCA 13; [1982] 1 NZLR 97.

[64] Fernando v Commonwealth (No 4) [2010] FCA 1475 is an exception.

[65] Giving the leading judgment in Rookes v Barnard [1964] UKHL 1; [1964] AC 1129, Lord Devlin conceded the possibility of exemplary damages in only three situations, one of which was where legislation makes specific provision in that regard. His first general law category comprised cases of "oppressive, arbitrary or unconstitutional action by servants of the government" (at 1226). No-one has taken that category as restricted to people directly engaged in the Civil Service, or even to people working only for central government. His second general law category was for cases where defendants (whether public or private) have cynically calculated on making a net profit from their wrongdoing after allowing for anything they may be legally obliged to pay the plaintiff by way of compensation. The subsequent development in England of restitutionary principles has diminished the force of the second category. See also Broome v Cassell & Co Ltd [1972] UKHL 3; [1972] AC 1027. The Law Commission could not enlist sufficient support for its proposal to abolish exemplary damages, and ended up recommending not abolition, but legislative reform: Law Commission, Aggravated, Exemplary and Restitutionary Damages (HMSO, London, 1997). No legislation followed.

[66] See: Uren v John Fairfax & Sons Pty Ltd [1966] HCA 40; (1966) 117 CLR 118; XL Petroleum (NSW) Pty Ltd v Caltex Oil (Aust) Pty Ltd [1985] HCA 12; (1985) 155 CLR 448; Lamb v Cotogno [1987] HCA 47; (1987) 164 CLR 1; and Gray v Motor Accident Commission (1998) 196 CLR 1 at 12-13. A relatively recent Australian attempt to add punitive damages to Equity's remedies (or at least, to its remedies for breach of fiduciary duty) was rebuffed with extraordinary hostility and an emphatic insistence that punishment be awarded only by the criminal courts: Harris v Digital Pulse Pty Ltd [2003] NSWCA 10; (2003) 56 NSWLR 298, Heydon JA. This article will not discuss the considerable number of legislative provisions in Australia for, against, or regulating the award of exemplary damages. It must suffice to mention that all jurisdictions have banned them in defamation actions (eg Defamation Act 2005 (NSW), s 37); some jurisdictions have banned them in negligence claims for personal injury or death (eg Civil Liability Act 2001 (NSW), s 21); and some have come to the relief of motor vehicle insurers but not their insured drivers (eg Motor Vehicles Act 1959 (SA), s 113A).

[67] See Taylor v Beere [1982] NZCA 15; [1982] 1 NZLR 81. New Zealand's position has wavered since then. First, it allowed stand-alone exemplary damages to get around legislative abolition of compensatory damages: Donselaar v Donselaar [1982] NZCA 13; [1982] 1 NZLR 97. Secondly, it restricted exemplary damages to cases where the defendant had acted either deliberately or with subjective recklessness: Bottrill v A [2001] 3 NZLR 622. Thirdly, the Privy Council reversed the latter decision: Bottrill v A [2002] UKPC 44; [2003] 2 NZLR 721, [2003] 1 AC 449. Fourthly, the Supreme Court recently overturned the Privy Council decision: Couch v Attorney General [2010] NZSC 27; [2010] 3 NZLR 149.

[68] See: Vorvis v Insurance Corporation of British Columbia [1989] 1 SCR 1085; and Whiten v Pilot Insurance Co [2002] 1 SCR 595.

[69] Rookes v Barnard [1964] UKHL 1; [1964] AC 1129 at 1226.

[70] Kuddus n 62 at 145, Lord Nicholls.

[71] Uren v John Fairfax & Sons Pty Ltd [1966] HCA 40; (1966) 117 CLR 118 at 132-133 and 160.

[72] P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press, Oxford, 1996) at viii. Another factor might well be the re-emergence of classical economic assumptions as to the causes of crime and the ways it may be managed; see L Zedner, "Policing Before and After the Police: the Historical Antecedents of Contemporary Crime Control" (2006) 46 British Journal of Criminology 78.

[73] See A Burrows, "Reforming Exemplary Damages", in P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press, Oxford, 1996) 153 at 168-169.

[74] Kuddus n 62 at 144-145 and 147-149, Lords Nicholls and Hutton respectively.

[75] Watkins n 19 insisted that misfeasance in public office was actionable only by plaintiffs who had sustained material loss, even if their defendants had breached a "constitutional" protection. One of Lord Bingham's reasons (at 409) was that the offending officers could be prosecuted for the indictable misdemeanour of misconduct in public office. Lord Walker (at 422) was sceptical. The defendants were prison officers who had flagrantly flouted a rule protecting a prisoner's right of privacy for communications covered by legal professional privilege. Lord Scott's position in Kuddus n 62 was not as explicit as Lord Bingham's in Watkins, but it came close.

[76] (1703) 1 Smith's LC 153. The leading English work on the offence is C Nicholls, T Daniel, M Polaine and J Hatchard, Corruption and Misuse of Public Office (OUP, Oxford, 2006) Chapter 3, 65-93 (Nicholls). That work (at 66) traces the offence back at least as far as Crouther's Case (1600) Cro Eliz 654. Modern cases tend not to go further back than Lord Mansfield CJ's judgment in R v Bembridge (1783) 3 Doug 327.

[77] (1703) 1 Smith's LC 153 at 275. The stated reason was that the law needed to avoid a multiplicity of civil actions for the same event. The applicability of the representative (or class) action to the misfeasance claim in Three Rivers seems to have passed without comment.

[78] See: L Radzinowicz, A History of English Criminal Law and its Administration from 1750, vol 2, The Clash Between Private Initiative and Public Interest in the Enforcement of the Law (London, Stevens, 1956). See also Downey v Acting District Court Judge Boulton (No 5) [2010] NSWCA 240; (2010) 272 ALR 705 at [67].

[79] L Zedner, "Policing Before and After the Police: the Historical Antecedents of Contemporary Crime Control" (2006) 46 British Journal of Criminology 78.

[80] The DPP for England and Wales had urged codification in 2003, but both the Joint Committee and the government disagreed. See: Report of the Joint Committee on the Draft Corruption Bill (2002-03, HL Paper 157, HC 705) [40]-[45], and [73]-[81]; and Nicholls n 76, 65 at 93 and 764-803. The Bill failed to pass, principally because it had sought to adapt "misconduct" to a Corruption Bill built on a model of agent and principal, in which the essence of corruption was the agent's disloyalty to the principal's interests. The Bribery Act 2010 (UK) overcame that problem by abandoning the principal/agent structure.

[81] This is an amalgam drawn from R v Dytham [1979] QB 722; R v Bowden [1996] 1 WLR 98; Attorney General's Reference (No 3 of 2003) [2005] QB 73; Wah v Hong Kong Special Administrative Region [2005] HKCFA 29; [2005] 2 HKLRD 375; R v Quach [2010] VSCA 106; and Nicholls 65 at 66-71.

[82] Wah v Hong Kong Special Administrative Region [2005] HKCFA 29; [2005] 2 HKLRD 375 at 391.

[83] Attorney General's Reference (No 3 of 2003) [2005] QB 73 at 90.

[84] See: Nicholls at 83-87; Attorney General's Reference (No 3 of 2003) [2005] QB 73; Wah v Hong Kong Special Administrative Region [2005] HKCFA 29; [2005] 2 HKLRD 375; R v Boulanger [2006] 2 SCR 49; and R v Quach [2010] VSCA 106

[85] For England, see P J Richardson (ed) Archbold: Criminal Pleading, Evidence and Practice 2010 (London, Sweet & Maxwell, 2010), [25-381] citing Attorney General's Reference (No 3 of 2003) [2005] QB 73. For Australia and Hong Kong, see: Wah v Hong Kong Special Administrative Region [2005] HKCFA 29; [2005] 2 HKLRD 375; and R v Quach [2010] VSCA 106. See also: P D Finn, "Public Officers Some Personal Liabilities" (1977) 51 Australian Law Journal 313; and P D Finn, "Official Misconduct" [1978] 2 Criminal Law Journal 307.

[86] The Criminal Code Act 1995 (Cth), s 142.2 makes it a criminal offence for Commonwealth public officials to misuse their powers either to benefit or to harm someone dishonestly. Section 359 of the Criminal Code 2002 (ACT) is the same. Western Australia's offence is similar, except that it requires corruption, rather than dishonesty: Criminal Code Act 1913 (WA), s 83. South Australia's offence is narrower in that it contains the "benefit" ingredient but no "harm" alternative. It is wider in so far as the prohibited "benefit" motivation need only be "improper", which is defined very broadly. See: Criminal Law Consolidation Act 1935 (SA), ss 238 and 251; and Question of Law Reserved (No 2 of 1996) [1996] SASC 5674; (1996) 67 SASR 63. Queensland has equivalents, but it also has an offence called "abuse of office". Section 92 of the Criminal Code Act 1899 (Qld) makes it an offence for anyone employed in the public service to do anything in abuse of their authority that is an arbitrary act prejudicial to the "rights" of another. The Northern Territory has the same offence of "abuse of office": Criminal Code Act (NT), s 82.

[87] Speaking for the Canadian Supreme Court in R v Boulanger [2006] 2 SCR 49, McLachlin CJ said that there had been two common law offences. The first offence was misfeasance, and it had required the accused to have acted (at 73-74 [58]) "for a purpose other than the public good, for example, for a dishonest, partial, corrupt, or oppressive purpose." The second offence (nonfeasance) related to advertent and very serious neglect of duty, and had no requirement as to motive. Boulanger held that s 122 of the Criminal Code 1985 (Can) imported only the first offence, so that nonfeasance was no longer a general offence.

[88] Eg, Three Rivers n 12 at 191-193, Lord Steyn referring to R v Bowden [1996] 1 WLR 98; R v Cunningham [1957] 2 QB 396; R v Caldwell [1982] UKHL 1; [1982] AC 341; and R v Lawrence [1982] AC 510.

[89] Eg, Attorney General's Reference (No 3 of 2003) [2005] QB 73 at 88-90; and R v Quach [2010] VSCA 106 at [28].

[90] [1982] UKHL 1; [1982] AC 341.

[91] Three Rivers n 12 at 192-193.

[92] In R v G [2003] UKHL 50; [2004] 1 AC 1034.

[93] It suffices to refer to Mengel n 10 at 345; and Three Rivers n 12 at 190.

[94] Halsbury's Laws of England (Butterworth, London, 1912) vol 23, "Public authorities and Public Officers", p 316 par [659].

[95] See below, n 213.

[96] [1959] VicRp 49; [1959] VR 286.

[97] What should have mattered was the fact that the police had intended no illegal harm.

[98] [1959] VicRp 49; [1959] VR 286 at 293. Little v Law Institute of Victoria [1990] VicRp 25; [1990] VR 257 at 270 confirmed this extension of misfeasance.

[99] S A de Smith, Judicial Review of Administrative Action (2nd ed, 1968) (Stevens, London) 19 and 319.

[100] Noting the first instance decision of Roncarelli v Duplessis [1952] 1 DLR 680, the first edition of H W R Wade, Administrative Law (Clarendon Press, Oxford, 1961) 83 said that an English plaintiff would not have succeeded. The switch came in H W R Wade, Administrative Law (Clarendon Press, Oxford, 4th ed, 1977) 636-640, which gave enthusiastic coverage to Farrington and to Roncarelli v Duplessis [1959] SCR 121 (SCC).

[101] J McBride, "Damages as a Remedy for Unlawful Administrative Action" [1979] Cambridge Law Journal 323 at 343.

[102] Dunlop v Woollahra Municipal Council [1982] AC 158 at 172.

[103] [1986] 1 QB 716.

[104] [1986] 1 QB 716 at 776.

[105] [1986] 1 QB 716 at 777, per Oliver LJ: "If an act is done deliberately and with knowledge of its consequences, I do not think that the actor can sensibly say that he did not 'intend' the consequences or that the act was not 'aimed' at the person who, it is known, will suffer them."

[106] Mengel n 10; Garrett n 11; Three Rivers n 12; and Odhavji n 13.

[107] Three Rivers n 12 at 192-193. The joint judgment of Gleeson CJ, Gummow, Hayne and Heydon JJ said in Ruddock v Taylor [2005] HCA 48; (2005) 222 CLR 612 at 622-623 [28] that misfeasance might lie against a Minister who "knew or ought to have known" that he was acting beyond power (emphasis added). It was a throw-away line that should not be taken seriously.

[108] Garrett n 11 at 350; and Three Rivers n 12 at 196, 227-228 and 231 and 235.

[109] Rawlinson v Rice [1997] 2 NZLR 651 at 658, 663 and 665; Three Rivers n 12 at 196, 230-231 and 236; and Watkins n 19 at 413 [39].

[110] n 12.

[111] n 12 at 196, 223, and 231 and 236.

[112] n 12 at 247 [46] and 251 [60].

[113] See Part X below.

[114] n 10.

[115] Strictly speaking, Mengel did not accept that it would be sufficient to establish reckless indifference as to the lawful limits of a defendant's power, because there was no such indifference. That might explain the High Court's subsequent reference to reckless indifference being an acceptable alternative in England; see Federal Commissioner of Taxation v Futuris Corporation Ltd [2008] HCA 32; (2008) 237 CLR 146 at 153 [11], Gummow, Hayne, Heydon and Crennan JJ.

[116] See n 10 at 347.

[117] The ambiguities flow from the references (joint judgment at 347, and Brennan J at 357) to harm "calculated" to flow from the defendant's intentional acts. That language was borrowed from Wilkinson v Downton [1897] 2 QB 57 at 59, in which Wright J had treated an obvious risk of harm as having been intended by a defendant who had given it no thought. Lord Hoffmann rejected Wilkinson's fictionalised intent in Wainwright v Home Office [2003] UKHL 53; [2004] 2 AC 406 at 425. Cf D Réaume, "The Role of Intention in the Tort in Wilkinson v Downton", in J Neyers, E Chamberlain and S Pitel (eds) Emerging Issues in Tort Law (Hart Publishing, Portland, Oregon, 2007) 533-556.

[118] South Australia v Lampard-Trevorrow [2010] SASC 56; (2010) 106 SASR 331 at 388 [263]- [264].

[119] Three Rivers n 12 at 230-231 and 286 [167].

[120] Three Rivers n 12 at 192 and 223; Odhavji n 13 at 281 [22]; and Watkins n 19 at 423 [73] (2).

[121] A view that is almost always implicit. Deane J's pithy summary in Mengel n 10 at 370-371 probably comes the closest to an explicit affirmation of this view.

[122] eg Three Rivers n 12 at 230 (Lord Hobhouse) and 235 (Lord Millett). See also Watkins n 19 at 407 [22].

[123] n 9 at 357.

[124] See n 12 at 246 [41]-[42] (Lord Hope), 267 [121] (Lord Hutton), 289 [175] (Lord Hobhouse), and 290-291 [179]-[182] (Lord Millett).

[125] Federal Commissioner of Taxation v Futuris Corporation Ltd [2008] HCA 32; (2008) 237 CLR 146 at 153 [11], Gummow, Hayne, Heydon and Crennan JJ.

[126] Three Rivers n 12 at 235-236.

[127] Australian criminal trial judges, for example, have a discretion whether to exclude illegally obtained evidence proferred against the accused, even where the illegality was deliberate. See: Bunning v Cross (1978) 141 CLR 54; Evidence Act 1995 (Cth), s 138.

[128] The sergeant should surely have overridden the plaintiff's objections and made a timely internal report of the police rape in Garrett n 11. The criminal misfeasance offence applied to a policeman who had consensual sex with a woman he was meant to be protecting and who he knew to be in a particularly fragile state: R v Quach [2010] VSCA 106.

[129] Poynder v Kent [2008] VSCA 245 at [119]- [126]. The officers in Mengel n 10 knew that they would be inflicting loss on the plaintiffs, but they had believed that they were acting lawfully; see the comment by Blanchard J in Garrett n 11 at 346.

[130] Cf Three Rivers n 12 at 230, where Lord Hobhouse treated targeted malice as no more than an extremely good evidentiary basis for inferring that the defendant lacked an honest belief that he or she was acting lawfully.

[131] Debelle J's decision in Rowan v Cornwall (No 5) [2002] SASC 160; (2002) 82 SASR 152 at 357-362 appears to be inconsistent with the analysis in this paragraph. The defendant Minister was held liable for action intended to harm the plaintiff because it was unlawful for having been taken in breach of natural justice and for improper purposes. The difficulty with Rowan is that the Minister had not known that he was acting illegally; nor had he been recklessly indifferent as to the limits of his lawful powers.

[132] Odhavji n 13 at 284-285 [28].

[133] Sanders v Snell [2003] FCAFC 150; (2003) 130 FCR 149 at 177-178 [106]- [108].

[134] Goldie v Commonwealth (2002) 117 FCR 566 at 586 [74].

[135] (1997) 90 A Crim R 526 at 566.

[136] [1959] SCR 121.

[137] [1959] SCR 121 at 185.

[138] It appears likely that Duplessis’s only concern was not the law, but whether he was aiming at the right man. He had ordered a detective to check whether Roncarelli was indeed the same man who both held the permit and was acting as bail bondsman. The court in Odhavji n 13 at 285 [30] glossed over the problem, saying that Duplessis knew that he had acted illegally.

[139] [1959] SCR 121 at 141 and 143 respectively.

[140] If the presumption were irrebuttable, there would be a significant increase in the risk of liability for police and others with lawful (albeit limited) power to inflict harm.

[141] One could argue that Rand J found Duplessis liable for a gross violation of the rule of law; see [1959] SCR 121 at 142, where Rand J said that it was a fundamental postulate of the rule of law that the victim of arbitrary and ultra vires action on the part of administrative officials should have "recourse or remedy". The rule of law's abhorrence of arbitrary conduct certainly informs misfeasance: Three Rivers n 12 at 190. Its violation is not, however, a tort in itself: Mengel n 10 at 352-353, by Mason CJ, Dawson, Toohey, Gaudron and McHugh JJ, with Brennan and Deane JJ agreeing at 354 and 373 respectively.

[142] See Finn, The Boyce Plaintiff n 44 at 495.

[143] (1285) 13 Edw 1, Westm Sec, c xxiv, "Of Writs in Consimili Casu".

[144] K Stanton, P Skidmore, M Harris and J Wright, Statutory Torts (Thomson, Sweet & Maxwell, London, 2003) 17.

[145] Sutton v Johnstone [1786] EngR 18; (1786) 1 TR 493 at 509 per Baron Eyre: "... every breach of a public duty, working wrong and loss to another, is an injury, and actionable ...". This was a malicious prosecution case against George III for action he had taken whilst still a Prince.

[146] Finn, The Boyce Plaintiff n 44 at 493-494.

[147] Mayor of Lyme Regis v Henley (1834) 8 Bli NS 690 at 714; and Ferguson v Kinnoull (1842) 9 Cl & Finn 251 at 279.

[148] Finn, The Boyce Plaintiff n 44 at 494-495, citing Atkinson v Newcastle and Gateshead Waterworks Co (1877) 2 Ex D 441 at 445-446, and Glossop v Heston and Isleworth Local Board (1879) 12 Ch D 102 at 109-110.

[149] Cf Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330 at 376 [124], where Gummow J said that the action for breach of statutory duty sprang originally "from the relationship between the legislature and the promoters of private Acts".

[150] In Ashby v White (1703) 1 Smith's LC 253 at 273: "Where a new act of parliament is made for the benefit of the subject, if a man be hindered from the enjoyment of it, he shall have an action against such person who so obstructed him."

[151] K Stanton, P Skidmore, M Harris and J Wright, Statutory Torts (Thomson, Sweet & Maxwell, London, 2003) 17-19.

[152] F Trindade, P Cane and M Lunney, The Law of Torts in Australia (4th ed, 2007, OUP, Melbourne) pp 165-166; and N Foster, "Breach of Statutory Duty and Risk Management in Occupational Health and Safety Law: New Law in Old Wineskins?" (2006) 14 Torts Law Review 79.

[153] Groves v Wimbourne [1898] 2 QB 402.

[154] Bourke v Butterfield & Lewis Ltd [1926] HCA 38; (1926) 38 CLR 354 held that contributory negligence was no defence, but the House of Lords disagreed in Caswell v Powell Duffryn Associated Collieries Ltd [1940] AC 152, and the High Court fell into line in Piro v W Foster & Co Ltd [1943] HCA 32; (1943) 68 CLR 313.

[155] A Wednesbury unreasonableness standard of care applies to actions brought in several Australian jurisdictions against public authorities for breach of statutory duty other than employees' claims for work-place injuries. See: Civil Law (Wrongs) Act 2002 (ACT), s 111; Civil Liability Act 2001 (NSW), s 43; Civil Liability Act 2003 (Qld), s 36; and Civil Liability Act 2002 (Tas), s 40. Victoria also imposes a Wednesbury standard, but only where the statutory duty is imposed specifically on the defendant in its public capacity, and not if it is a duty of strict liability: Wrongs Act 1958 (Vic), s 84(2).

[156] K M Stanton, Breach of Statutory Duty in Tort (Sweet & Maxwell, London, 1986) p 26, citing Monk v Warbey [1935] 1 KB 75. The replacement work does not pursue the issue; see K Stanton, P Skidmore, M Harris and J Wright, Statutory Torts (Thomson, Sweet & Maxwell, London, 2003).

[157] Stuart v Kirkland-Veenstra [2009] HCA 15; (2009) 237 CLR 215 at 263-264; and Morrison Sports Ltd v Scottish Power [2010] UKSC 37; [2010] 1 WLR 1934 at 1943-1944 [28]- [29] (SC).

[158] Lord Denning MR thought that his only honest alternative to tossing a coin was to follow his sense of justice: Ex parte Island Records Ltd [1978] Ch 122 at 135. Dixon J thought that some judges were deciding according to their own preferences rather than genuinely treating the matter as a constructional issue: O'Connor v S P Bray Ltd [1937] HCA 18; (1937) 56 CLR 464 at 478. Kitto J acknowledged the constructional difficulties, but insisted that the exercise remained a genuine issue of statutory interpretation: Sovar v Henry Lane Pty Ltd [1967] HCA 31; (1967) 116 CLR 397 at 405. McHugh and Gummow JJ said in Byrne v Australian Airlines Ltd (1995) 185 CLR 410 at 459 that legislative "intentions" were illusory, particularly as regards Acts that represented compromises or deals between different forces. Rather, one should approach the matter on a principled basis as a matter of statutory construction. See also: Gardiner v Victoria [1999] VSCA 100; [1999] 2 VR 461 at 468-469 [23]; and Miller v Miller [2011] HCA 9 at [29].

[159] R (Canada) v Saskatchewan Wheat Pool [1983] 1 SCR 205.

[160] Byrne v Australian Airlines Ltd (1995) 185 CLR 410 at 458; and Crimmins v Stevedoring Industry Finance Committee [1999] HCA 59; (1999) 200 CLR 1 at 58.

[161] See K Stanton, P Skidmore, M Harris and J Wright, Statutory Torts (Thomson, Sweet & Maxwell, London, 2003) 449-450; and D Fairgrieve, State Liability in Tort: A Comparative Law Study (OUP, Oxford, 2003) pp 36-41. Fairgrieve pointed to leading judgments emphasising the unlikelihood of the legislature impliedly creating a new cause of action with huge budgetary costs.

[162] See: Doe dem Murray v Bridges [1831] EngR 57; (1831) 1 B & Ad 847 at 859; and Pasmore v Oswaldtwistle Urban District Council [1898] AC 387 at 394.

[163] Cutler v Wandsworth Stadium Ltd [1949] AC 398 at 408; Sovar v Henry Lane Pty Ltd [1967] HCA 31; (1967) 116 CLR 397 at 405-406; and Gardiner v Victoria [1999] VSCA 100; [1999] 2 VR 461 at 469 [25]. The criminal penalty in the legislation under consideration in Groves v Wimbourne [1898] 2 QB 402 would have varied according to the defendant's fault rather than the plaintiff's loss, and would not necessarily have gone to the plaintiff.

[164] See: O'Rourke v Camden London Borough Council [1997] UKHL 24; [1998] AC 188 at 194 Lord Browne-Wilkinson; and Cullen v Chief Constable of the Royal Ulster Constabulary [2003] UKHL 39; [2003] 1 WLR 1763 (HL).

[165] See Repacholi Aviation Pty Ltd v Civil Aviation Safety Authority [2010] FCA 994 at [30].

[166] See Miller v Miller [2011] HCA 9 at [65]- [68].

[167] X (Minors) v Bedfordshire County Council [1995] UKHL 9; [1995] 2 AC 633 at 731-732; and O'Rourke v Camden London Borough Council [1997] UKHL 24; [1998] AC 188 at 193-194. See also: Phelps v Hillingdon London Borough Council [2000] UKHL 47; [2001] 2 AC 619 at 652; and Attorney General v Daniels [2003] 2 NZLR 742 at 767-768 [87]-[88].

[168] Cutler v Wandsworth Stadium Ltd [1949] AC 398 at 408.

[169] O'Rourke v Camden London Borough Council [1997] UKHL 24; [1998] AC 188 at 194.

[170] Semantic distinctions are sometimes drawn between duties on the one hand, and prohibitions or powers on the other hand; see, eg, Lonrho Ltd v Shell Petroleum Co Ltd [1982] AC 173 at 186. Even in the case of a statutory discretion, however, there might occasionally be a relevant "duty" in circumstances where there is only one way in which the discretion can lawfully be exercised: Stuart v Kirkland-Veenstra [2009] HCA 15; (2009) 237 CLR 215 at 265 [146]. The idea that flagrant abuse of discretionary power should be actionable has a long history. See Sutton v Johnstone [1786] EngR 18; (1786) 1 TR 493 at 504 per Baron Eyre: "And one may observe in general, in respect of what is done under powers incident to situations, that there is a wide difference between indulging to situation a latitude touching the extent of power, and touching the abuse of it. Cases may be put of situations so critical that the power ought to be unbounded: but it is impossible to state a case where it is necessary that it should be abused; and it is the felicity of those who live under a free constitution of Government, that it is equally impossible to state a case where it can be abused with impunity."

[171] Wrongs Act 1958 (Vic), s 84(3); and Civil Liability Act 2002 (WA), s 5Y.

[172] Chancery Amendment Act 1858, 21 & 22 Vict c 27, s 2.

[173] Boyce v Paddington Borough Council [1903] 1 Ch 109 at 114.

[174] Australian Conservation Foundation Inc v Commonwealth (1980) 146 CLR 493; and Onus v Alcoa of Australia Ltd [1981] HCA 50; (1981) 149 CLR 27.

[175] See R v Inland Revenue Commissioners; Ex parte National Federation of Self-Employed and Small Businesses Ltd [1981] UKHL 2; [1982] AC 617.

[176] Stininato v Auckland Boxing Association (Inc) [1978] 1 NZLR 1 at p 23; and Wentworth v Woollahra Municipal Council [1982] HCA 41; (1982) 149 CLR 672. See Finn, The Boyce Plaintiff n 44 at 494-495.

[177] Stovin v Wise [1996] UKHL 15; [1996] AC 923 at 952-953.

[178] See Gorringe v Calderdale Metropolitan Borough Council [2004] UKHL 15; [2004] 2 All ER 326.

[179] Mengel n 10 at 347.

[180] Law Commission, Administrative Redress: Public Bodies and the Citizen – a Consultation Paper (Consultation Paper No 187, 2008). It is admittedly a slight exaggeration to characterise the Commission's reform proposals as having focused on negligence law, although that forms the bulk of so much of its discussion as is court-centred. Liability would have attached to the acts and even "pure omissions" of persons exercising truly public functions. There must have been "serious fault" on the defendant's part in a manner which frustrated the legislative intent to confer a benefit on the plaintiff or a class to which the plaintiff belonged, raising the same constructional problems that currently plague actions for breach of statutory duty, which the Commission also wanted to abolish. The new compensation right would have extended into territories previously reserved for government "policy" or "discretion".

[181] See Administrative Redress n 5. The Commission’s proposals had met with wide-spread opposition from most quarters. Most importantly, all government agencies and relevant Ministers were opposed: at [1.3].

[182] The differences in the approaches of Australian and Canadian courts are not relevant to this discussion.

[183] See: Sullivan v Moody (2001) 207 CLR 562; and JD v East Berkshire Community Health NHS Trust [2005] UKHL 23; [2005] 2 AC 373.

[184] Jain v Trent Strategic Health Authority [2009] EWCA Civ 1357; [2009] 1 AC 853.

[185] See: Tame v New South Wales [2002] HCA 35; (2002) 211 CLR 317; Brooks v Metropolitan Police Commissioner [2005] UKHL 24; [2005] 1 WLR 1495; and Smith v Chief Constable of Sussex Police [2008] UKHL 50; [2009] 1 AC 225. Canada, however, does have a tort of negligent investigation: Hill v Hamilton-Wentworth Regional Police Services Board [2007] 3 SCR 129.

[186] Elguzouli-Daf v Metropolitan Police Commissioner [1994] EWCA Civ 4; [1995] QB 335 at 348; and Brooks v Metropolitan Police Commissioner [2005] UKHL 24; [2005] 1 WLR 1495 at [38].

[187] See Perre v Apand Pty Ltd [1999] HCA 36; (1999) 198 CLR 180 at 245 [178]. The position might be changing. See P Vines, "Misfeasance in Public Office: Old Torts, New Tricks?" (forthcoming), which lists several cases where the harms were reputational, personal injury, or false imprisonment. Vines suggests that part of the upsurge in Australian misfeasance claims might be the result of the recent imposition of legislative restrictions of the negligence liability of public authorities. Where other torts are available, it is difficult to see the advantage in pursuing a misfeasance claim; cf Habib v Commonwealth (No 2) [2009] FCA 228; (2009) 175 FCR 350; and Habib v Commonwealth [2010] FCAFC 12; (2010) 183 FCR 62. In England, a murdered woman's estate sued for misfeasance in Akenzua v Home Secretary [2003] 1 WLR 741, in circumstances where there was no proximity and therefore no common law duty of care in negligence. It has been suggested that Canadian pleading rules permit allegations of bad faith without genuine particulars, and that a misfeasance claim occasionally has the procedural advantage over other torts of affording broader (and potentially more embarrassing) discovery: E Chamberlain, "What is the Role of Misfeasance in Public Office in Modern Canadian Tort Law?" (2009) 88 Canadian Bar Review 579 at 597-599. Chamberlain also argues (at 593-596) that the lack of a broad judicial discretion to limit individual misfeasance claims by reference to public policy considerations sometimes makes it a more attractive tort than negligence.

[188] White v Jones [1995] UKHL 5; [1995] 2 AC 207; and Hill v Van Erp (1997) 188 CLR 159.

[189] Three Rivers n 12 at 229 per Lord Hobhouse: "There is no principle in English law that an official is the guarantor of the legality of everything he does ...". Cf Mengel n 10 at 353, where the joint judgment speculated that there might be "very many circumstances" in which government officials have a duty of care to ensure that their subordinates know and stay within the limits of their power. In such cases, their Honours said that there would usually be an additional duty on those officials, to ascertain their power. Deane J appeared to agree at 373-374.

[190] See: Dunlop v Woollahra Municipal Council [1982] AC 158 at 171-172; and Rowling v Takaro Properties Ltd [1988] 1 AC 473 at 500 and 502-503.

[191] Allsop P said in Precision Products (NSW) Pty Ltd v Hawkesbury City Council [2008] NSWCA 278; (2008) 74 NSWLR 102 at 128 [119] that a duty to take care to act intra vires would require the courts to set public sector management standards, which would be both difficult and a violation of the separation of powers.

[192] See G McLay, "Book Comment: What are we to do with the Public Law of Torts?" (2009) 7 New Zealand Journal of Public International Law 373 at 375-376.

[193] See, eg, Stuart v Kirkland-Veenstra [2009] HCA 15; (2009) 237 CLR 215, which held that police owe no common law duty of care towards would-be suicides. The situation is obviously different as regards special relationships, such as parent and child, jailer and prisoner, or teacher and pupil.

[194] See, eg: Modbury Triangle Shopping Centre Pty Ltd v Anzil [2000] HCA 61; (2000) 205 CLR 254; and Smith v Chief Constable of Sussex Police [2008] UKHL 50; [2009] 1 AC 225.

[195] Cf P Benson, "Misfeasance as an Organizing Normative Idea in Private Law" (2010) 60 University of Toronto Law Journal 731, arguing that the distinction between misfeasance and nonfeasance is a normative principle permeating much of private law.

[196] Three Rivers n 12 at 230 (Lord Hobhouse) and 236-237 (Lord Millett).

[197] [1979] QB 722. The Court of Appeal upheld Mr Dytham's conviction for the indictable offence of misconduct in public office.

[198] Three Rivers n 12 at 230 (Lord Hobhouse) and 236-237(Lord Millett). Lords Bingham and Brown said in Smith v Chief Constable of Sussex Police [2008] UKHL 50; [2009] 1 AC 225 at 264 [53] and 282 [120] respectively that the constable in Dytham would now be said to have owed the victim a common law duty of care. It is curious that they made no mention of the possibility of an action for misfeasance.

[199] (1703) 1 Smith's LC 153 at 276.

[200] See Watkins n 19 at 404-405 and 415-417. See also Attorney General (Cth) (Ex rel McKinlay) v Commonwealth [1975] HCA 53; (1975) 135 CLR 1 at 76 per Murphy J: "Enforcement of constitutional political rights does not have to be justified by characterizing them as rights of property. This degrades the political right. The exaltation of property rights over civic and political rights is a reflection of the values of a bygone era."

[201] Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 at 554-555 [4].

[202] [1973] VicRp 70; [1973] VR 715 at 720.

[203] McInerney J at first instance dealt at greater length with the requirement that there be a duty, but his point was that unless it was owed to the public, the defendant would not be a "public officer" for the purposes of the tort: Tampion v Anderson [1973] VicRp 32; [1973] VR 321 at 336-337.

[204] n 9 at 347, Mason CJ, Dawson, Toohey, Gaudron and McHugh JJ.

[205] n 9 at 357.

[206] n 9 at 371.

[207] [2009] NSWCA 90; (2009) 255 ALR 553.

[208] [2009] NSWCA 90; (2009) 255 ALR 553 at 555 [4]- [6].

[209] [2009] NSWCA 90; (2009) 255 ALR 553 at 577 [116]. Western Australia's Court of Appeal took the same view in Neilson v City of Swan (2006) 147 LGERA 136 at 158-159 [66].

[210] Cannon v Tahche [2002] VSCA 84; (2002) 5 VR 317 at 347 [77].

[211] This would fit with the fact that they owe no duty of care to their adversaries: Elguzouli-Daf v Metropolitan Police Commissioner [1994] EWCA Civ 4; [1995] QB 335 at 348; and Brooks v Metropolitan Police Commissioner [2005] UKHL 24; [2005] 1 WLR 1495 at [38].

[212] The claimants discontinued in Three Rivers, for example, before getting to the stage of proving that their losses were as much the Bank of England's fault as that of the fraudsters who had run the failed bank.

[213] Vaughan Williams LJ thought it "perfectly clear" in Davis v Bromley Corporation [1908] 1 KB 170 at 173 that a builder could not get damages for a council's malicious refusal of a building permit; even a mandamus would have gone no further than compelling reconsideration according to law. Davis is treated nowadays as a case of judicial amnesia with regard to the misfeasance tort more generally; see: Three Rivers n 12 at 190, Lord Steyn; and Watkins n 19 at 404, Lord Bingham.

[214] See: Sellars v Adelaide Petroleum NL [1994] HCA 4; (1994) 179 CLR 332; and Commonwealth v Cornwell (2007) 229 CLR 519. The Privy Council said in David v Abdul Cader [1963] 1 WLR 834 at 839-840 that Davis v Bromley Corporation [1908] 1 KB 170 no longer applies. The Canadian Supreme Court in Roncarelli v Duplessis [1959] SCR 121 fudged the problem that the plaintiff's permit was entirely discretionary and renewable annually; the court awarded a considerable sum for loss of the goodwill attaching to a wrongfully cancelled permit that had very little time left to run. Cf G McLay, "Book Comment: What are we to do with the Public Law of Torts?" (2009) 7 New Zealand Journal of Public International Law 373 at 379.

[215] Garrett n 11 at 346, Blanchard J.

[216] n 11 at 229.

[217] See Three Rivers n 12 at 193 (Lord Steyn, who added that there was also no requirement of proximity), 197 (Lord Hope agreeing with Lords Steyn and Hutton), and 228 (Lord Hutton, dismissing the need for a separate and additional showing of "some other link or relationship between [the plaintiff] and the officer"). Lord Millett said (at 237) that it was unnecessary to determine that issue.

[218] See: E Chamberlain, "The Need for a 'Standing' Rule in Misfeasance in Public Office" (2007) 7 Oxford University Commonwealth Law Journal 215; J Neyers, "The Economic Torts as Corrective Justice" (2009) 17 Torts Law Journal 162; but cf P Cane, "The Anatomy of Private Law Theory: a 25th Anniversary Essay" (2005) 25 Oxford Journal of Legal Studies 203; and P Cane, Book review of R Stevens Torts and Rights, (2008) 71 Modern Law Review 641. See also E Chamberlain, "Misfeasance in Public Office: a Justifiable Anomaly to the Rights-Based Approach?", and J Neyers, "Explaining the Inexplicable? Four Manifestations of Abuse of Rights in English Law", both in A Robertson and D Nolan (eds), Rights and Private Law (Oxford, Hart Publishing, 2011, forthcoming).

[219] Corrective justice theorists might respond either that misfeasance is a private law tort responding (unusually) to an abuse of rights in a non-Hohfeldian sense, or even that it is best left to public law. See R Stevens, Torts and Rights (Oxford University Press, Oxford, 2007) at 218 and 242-43; and G McLay, "Book Comment: What are we to do with the Public Law of Torts?" (2009) 7 New Zealand Journal of Public International Law 373 at 375. Stevens regards it as a “regulatory” mechanism (ibid. at 218). It appears that OUP's prestigious volumes on public and private law could not decide where to deal with misfeasance, with the result that it is scarcely mentioned. For private law, see: J Edelman and J Davies, "Torts and Equitable Wrongs", and A Burrows, "Judicial Remedies", in A Burrows (ed), English Private Law (2nd ed, 2007) (OUP, Oxford) 1185 at 1315-1316, and 1617 at 1677-1679 respectively. For public law, see A W Bradley, "The Constitutional Position of the Judiciary", in D Feldman (ed), English Public Law (2nd ed, 2009) (OUP, Oxford) 281 at 310.

[220] n 9 at 347.

[221] Shades of punitive damages, which Lord Hailsham quantified in terms of what the defendant should pay, rather than what the plaintiff should receive: Broome v Cassell & Co Ltd [1972] UKHL 3; [1972] AC 1027 at 1077.

[222] n 11 at 229-230.

[223] n 11 at 194, 196, 228, 229, 235 and 237.

[224] n 11 at 193 (Lord Steyn) and 246-247 [43] (Lord Hope).

[225] n 11 at 231, Lord Hobhouse.

[226] n 11 at 237, Lord Millett.

[227] n 11 at 230-231.

[228] See M Aronson, B Dyer and M Groves, Judicial Review of Administrative Action (4th ed, 2009) (Thomson Reuters, Sydney) 772-774 [11.115].

[229] [2003] 1 WLR 741.

[230] This was a strike-out application, so that strictly speaking, the facts had not yet been established. The court nevertheless painted a compelling picture, drawn largely from remarks at sentencing.

[231] Sedley LJ said that any suggestion that the victim's class was 3.5m London women was "a spurious endeavour to give specificity to the unspecific.": [2003] 1 WLR 741 at 746 [15].

[232] [2003] 1 WLR 741 at 747 [20], Sedley LJ.

[233] [2003] 1 WLR 741 at 751 [35], Brown LJ.

[234] The wife-murderer and terrorist were Sedley LJ's examples, and Hannibal Lecter was Brown LJ's: [2003] 1 WLR 741 at 747 [16] and 750 [33] respectively.

[235] See above, n 187.

[236] See above, n 112.

[237] See, eg: Banking Act 1959 (Cth), s 9(2); Australian Prudential Regulation Authority Act 1998 (Cth), s 8(2); and Banking Act 2009 (UK), s 4.

[238] [1828] EngR 701; (1828) 5 Bing 91. Brennan J in Mengel n 10 at 355 treated Henly as an early misfeasance case, but Lord Millett in Three Rivers n 12 at 237 treated it as an action for breach of statutory duty. It is submitted that it was an action for breach of non-statutory duty.

[239] n 12.

[240] [1979] QB 722.

[241] Three Rivers n 12 at 228, 230 and 237; and Neilson v City of Swan (2006) 147 LGERA 136 at 152-153 [47] and 162 [84].

[242] Three Rivers n 12 at 237.

[243] n 13.

[244] n 11.

[245] Mengel n 10 at 355, Brennan J.

[246] Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 at 576 [109].

[247] [1990] 1 WLR 54 at 71.

[248] [1990] 1 WLR 54 at 70-71 and 85.

[249] Jones v Swansea City Council [1990] 1 WLR 1453 at 1458-1459.

[250] Manchester City Council v Pinnock [2010] UKSC 45; [2010] 3 WLR 1441 at [54].

[251] See, eg: Criminal Code Act 1995 (Cth), s 142.2(1)(a)(i); Criminal Code Act 1899 (Qld), s 92A(5); and Criminal Code Act 2002 (ACT), s 359(1)(a)(i).

[252] See, eg, R v Quach [2010] VSCA 106, where the charge related to an off-duty policeman having sex with a woman he knew to be in a particularly fragile state, having conducted an on-duty "welfare check" on her earlier in the day because she had attempted suicide.

[253] Mengel n 10 at 355, emphasis added.

[254] [1989] 1 AC 1228 at 1240-1241.

[255] [2002] EWCA Civ 1073 at [17].

[256] [1978] TASStRp 1; [1978] Tas SR 1 at 14.

[257] An employee's dismissal in McGuirk v University of New South Wales [2010] NSWSC 1471 was held not to be an exercise of public power, but it was not governed by statute.

[258] See Tampion v Anderson [1973] VicRp 70; [1973] VR 715 at 720.

[259] [1998] FCA 709.

[260] See: M Aronson, "Government Liability in Negligence" [2008] MelbULawRw 2; (2008) 32 Melbourne University Law Review 44 at 53.

[261] n 11 at 230. His Lordship added at 286 [167] that it must be unlawful not because it was itself tortious, but because it was unauthorised or forbidden by law. The House of Lords in Kuddus n 62 accepted without demur the defence concession that it was misfeasance for a police officer to forge a document dropping a theft complaint.

[262] The facts in R v Dytham [1979] QB 722. The deceased in Akenzua v Home Secretary [2003] 1 WLR 741 was murdered by an extremely dangerous man who was at large because the authorities had perverted the criminal justice system. His release on bail should have been revoked, but it was not a nullity.

[263] Rawlinson v Rice [1998] 1 NZLR 454. Brennan J noted the issue in Mengel n 10 at 355.

[264] Cannon v Tahche [2002] VSCA 84; (2002) 5 VR 317 at 336 [48].

[265] See above, Part IV.

[266] Spigelman CJ insisted in Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 at 555 [9] that the requirement that the defendant be a public officer is "quite distinct" from the requirement that the power or duty that was abused be public.

[267] Mengel n 10 at 355 and 370; and Three Rivers n 12 at 191 and 229-230.

[268] The issue and cases are discussed at length in Chase Oyster Bar Pty Ltd v Hamo Industries Pty Ltd [2010] NSWCA 190; (2010) 272 ALR 750, and noted in Plaintiff M61/2010E v Commonwealth [2010] HCA 41; (2010) 272 ALR 14 at [51].

[269] The leading case is R v Panel on Take-overs and Mergers, Ex parte Datafin Plc [1986] EWCA Civ 8; [1987] QB 815.

[270] Three Rivers n 12 at 191.

[271] Three Rivers n 12 at 230.

[272] Three Rivers n 12 at 229. See also Percy v Church of Scotland [2005] UKHL 73; [2006] 2 AC 28 at 38 [17], and 46 [54].

[273] [2007] EWCA Civ 930; [2008] 1 WLR 2255 at 2263 [23].

[274] Spigelman CJ agreed: Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 at 554 [3].

[275] Mengel n 10 at 355, referring to Best CJ in Henly v Mayor of Lyme [1828] EngR 701; (1828) 5 Bing 91 at 107-108.

[276] See R v McCann [1998] 2 Qd R 56 at 72, where Byrne J gave as examples justices of the peace and university chancellors.

[277] Marks v Commonwealth [1964] HCA 45; (1964) 111 CLR 549 at 567-568, Windeyer J.

[278] In R v McCann [1998] 2 Qd R 56 at 68, Byrne J listed cases taking publicly paid engine drivers, porters, fettlers, cleaners, railway clerks and temporary lecturers outside the definition of public officer.

[279] Byrne J helpfully analysed a large number of cases from a range of contexts and jurisdictions in R v McCann [1998] 2 Qd R 56 at 67-74.

[280] Cannon v Tahche [2002] VSCA 84; (2002) 5 VR 317; Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 at 556 [17]- [18], 562 [51] and 576 [109]; and Stewart v Ronalds [2009] NSWCA 277; (2009) 76 NSWLR 99 at 121 [106].

[281] Cannon v Tahche [2002] VSCA 84; (2002) 5 VR 317 gave both reasons, but emphasised the second, whilst Leerdam v Noori [2009] NSWCA 90; (2009) 255 ALR 553 emphasised the first.

[282] Stewart v Ronalds [2009] NSWCA 277; (2009) 76 NSWLR 99 at 121 [106].

[283] Elguzouli-Daf v Metropolitan Police Commissioner [1994] EWCA Civ 4; [1995] QB 335 at 347 and 352.

[284] Cannon v Tahche [2002] VSCA 84; (2002) 5 VR 317 at 344 [66].

[285] [1998] FCA 709.

[286] Dunlop v Woollahra Municipal Council [1982] AC 158 at 172; Jones v Swansea City Council [1990] 1 WLR 1453 at 1458-1459; and Three Rivers n 12 at 191.

[287] The example is drawn from Jones v Swansea City Council [1990] 1 WLR 54 (CA) and [1990] 1 WLR 1453 (HL). The claim failed in Jones for want of proof that a majority of councillors had been improperly motivated. It appears that in the Court of Appeal, the plaintiffs in Three Rivers n 12 at 54 had sought to make the Bank of England directly liable for the conduct of its senior officials. The same position probably applied in the appeal (see at 240 [15], and 270 [126]), but other passages suggest that the Bank's liability may have been alleged to have been vicarious; see at 191 and 230.

[288] This Part draws on J Davis, "Misfeasance in Public Office, Exemplary Damages and Vicarious Liability" [2010] AIAdminLawF 29; (2010) 64 Australian Institute of Administrative Law Forum 59.

[289] De facto authority suffices where an employer could not legally have authorised a breach of the law: James v Commonwealth [1939] HCA 9; (1939) 62 CLR 339 at 359-360; and Mengel n 10 at 347.

[290] See: Bazley v Curry (1999) 174 DLR (4th) 45; Jacobi v Griffiths (1999) 174 DLR (4th) 71; Lister v Lesley Hall Ltd [2002] 1 AC 215; and New South Wales v Lepore [2003] HCA 4; (2003) 212 CLR 511.

[291] See: Little v Commonwealth [1947] HCA 24; (1947) 75 CLR 94 at 114; Grimwade v Victoria (1997) 90 A Crim R 526 at 567-570; and R Balkin and J Davis, Law of Torts (LexisNexis, Sydney, 4th ed, 2009) [26.56]-[26.59].

[292] Crown Proceedings Act 1947 (UK), s 2(3).

[293] Law Reform (Vicarious Liability) Act 1983 (NSW).

[294] Crown Proceedings Act 1950 (NZ), s 6(3).

[295] Crown Liability and Proceedings Act 1985 (Can) (RSC 1985, c C-50), s 3(b)(i).

[296] Crown Proceedings Act 1972 (SA), s 10(2) precluded Crown resort to the "independent discretion" rule for acts or omissions that only Crown servants (and possibly other public bodies) had authority to perform: South Australia v Kubicki (1987) 46 SASR 282; and De Bruyn v South Australia (1990) 54 SASR 231 at 244-245.

[297] Crown Proceedings Act 1992 (SA).

[298] R Balkin and J Davis, Law of Torts (LexisNexis, Sydney, 4th ed, 2009) states at [26.56] that the rule was not revived. There have been several cases since 1992 in which the independent discretion rule was not raised, even though the torts pleaded clearly involved the exercise of independent discretions. See, eg, X v South Australia (No 3) [2007] SASC 125; (2007) 97 SASR 180 (Parole Board); and South Australia v Lampard-Trevorrow [2010] SASC 56; (2010) 106 SASR 331 (Aborigines Protection Board). See also TransAdelaide v Evans [2005] SASC 175; (2005) 98 SASR 394 at [16].

[299] Racz v Home Office [1994] 2 AC 45 held that there were no special rules governing vicarious liability in misfeasance, making such liability possible. See also Three Rivers n 12 at 191 and 230.

[300] See Administrative Redress n 5, [3.66]-[3.72].

[301] Indeed, it revoked special leave to appeal on the question of vicarious liability in Tepko Pty Ltd v Water Board (NSW) [2001] HCA 19; (2001) 206 CLR 1; see the transcript at [2000] HCATrans 697.

[302] n 9 at 347.

[303] See also Rogers v Legal Services Commission (1995) 64 SASR 572 at 587.

[304] The allegations of misfeasance were outlined in Pharm-a-Care Laboratories Pty Ltd v Commonwealth (No 3) [2010] FCA 361; (2010) 267 ALR 494 and Pharm-a-Care Laboratories Pty Ltd v Commonwealth (No 5) [2010] FCA 1204. Both the individual action and the class action received widespread media coverage; see, for example Business Review Weekly 21 August 2008 (reporting settlement of the individual action for around $55m), and The Australian Financial Review 26 March 2011 (reporting settlement of the class action for $67.5m). For the Federal Court's approval of the class action settlement, see Pharm-a-Care Laboratories Pty Ltd v Commonweralth (No 6) [2011] FCA 277. The Commonwealth paid $5m to Vita Life Sciences Ltd in settlement of a separate claim brought in the NSW Supreme Court; see that company's report to the stock exchange, reprinted at ASX ComNews for 25 February 2011, accessed online on 1 April 2011 at http://global.factiva.com/aa/?ref=ASXTEX0020110225e72p000sb&pp=1&fcpil=en&napc=S&sa_from= .

[305] See Legal Services Directions 2005 (Cth), Appendix E, paras [5]-[6].

[306] See: Law Reform (Vicarious Liability) Act 1983 (NSW) Part 4, and Police Act 1990 (NSW), s 213; Police Service Act 2003 (Tas), s 84; Police Act 1998 (SA), s 65; Police Act 1892 (WA), s 137; Police Administration Act (NT) Part VIIA; Police Service Administration Act 1990 (Qld), s 10.5; Police Regulation Act 1958 (Vic), s 123; and Australian Federal Police Act 1979 (Cth), s 64B.

[307] See: De Reus v Gray [2003] VSCA 84; (2003) 9 VR 432 at 438; and New South Wales v Ibbett [2006] HCA 57; (2006) 229 CLR 638.

[308] [1956] UKHL 6; [1957] AC 555.

[309] Employees Liability Act 1991 (NSW), ss 3 and 5; Law Reform (Contributory Negligence and Apportionment of Liability) Act 2001 (SA), s 6(9)(c); and Law Reform (Miscellaneous Provisions) Act (NT), s 22A.

[310] Insurance Contracts Act 1984 (Cth), s 66.

[311] R Balkin and J Davis, Law of Torts (LexisNexis, Sydney, 4th ed, 2009) [29.39].

[312] David v Abdul Cader [1963] 1 WLR 834 at 839-840.

[313] Harlow n 23 argued for a new tort that I suspect would cross the public-private divide. It would apply to conduct that is contumelious, deliberate and outrageous. It would also allow punitive damages in the absence of material loss, because it would reverse the common law's parsimonious attitude to dignitarian loss. A majority ruled against the creation of a separate measure of damages for "vindicatory" purposes in Lumba v Secretary of State for the Home Department [2011] UKSC 12.

[314] Actual or (as in Odhavji n 13) assumed.