Aronson, Mark --- "Some Australian Reflections on Roncarelli" [2010] UNSWLRS 5

Last Updated: 21 February 2010

Some Australian Reflections on Roncarelli

Mark Aronson[∗]

Citation

This article is forthcoming in volume 55 McGill Law Journal.

(Note This paper is an unedited manuscript that has been accepted for publication in volume 55 McGill Law Journal. It must not be cited without permission.)

Abstract

Roncarelli v Duplessis figures far more frequently in Australia's secondary literature than in its court decisions, and it is noted not for its invalidation of Premier Duplessis' actions, but for its award of damages where judicial declaration of invalidity would usually be the only remedy. Invalidating Duplessis' interference with Roncarelli's liquor licence would have been the easy part of the case had it been tried in Australia. Australian statutes afforded good protection to liquor licensees, and general administrative law principles confined seemingly unfettered discretionary powers in less solicitous statutory regimes. In addition, the constitutional abolition of internal trade barriers used to be taken as banning unfettered regulatory powers over inter-State traders. Duplessis' tort liability was the hard part. His assumption of legal power was not deliberate, but it was extraordinarily indifferent to questions of legality. Rand J characterised this as "malice", which in turn triggered liability to a uniquely public law tort known nowadays as misfeasance in public office. That tort is likely to cover more forms of non-deliberate official misconduct in Canada than in Australia, whose High Court usually avoids open-ended legal principles, particularly those according immediate operative force to substantive conceptions of the rule of law.

1. Introduction

For two days in September 2009, more than a dozen academics pored over different aspects of Roncarelli v Duplessis,[1] debating different visions of what it might first have meant, what it might now mean, its political significance, and its legal importance both normatively and doctrinally. My contribution will be to look for some of the doctrinal consequences of Rand J's deployment of a substantive understanding of the rule of law, a principle that his Honour said requires "recourse or remedy" for administrative action "dictated by the arbitrary likes, dislikes and irrelevant purposes of public officers acting beyond their duty".[2] It is a contribution, however, that I offer with some diffidence, because it is obviously fraught for an Australian lawyer to look at a famous old Canadian case, particularly where Australia's courts and tribunals have given it only the briefest attention.[3] A foreigner reading Roncarelli is ill-equipped fully to appreciate both its provenance and its trajectory, and is tempted to take its judgments at face value, as if each of its apparent issues had been of equal importance, novelty and difficulty in its day. It might therefore be more productive if I were to focus largely on Rand J's reasons for awarding damages to Mr Roncarelli, and compare those with the likely response (both then and nowadays) of the Australian High Court had it been faced with evidence of such an obvious abuse of power as had occurred in Roncarelli.

I look first at Rand J's need to deploy the rule of law to get around an Act that seemingly invested the Attorney General[4] with unfettered power. Australian courts would have got around that problem without invoking the rule of law. General administrative law principles had long established some inroads into statutory grants of discretionary powers; some statutory licensing regimes provided their own protective mechanisms; and business interests operating across State lines received constitutional protection which the High Court read as limiting administrative discretion. Secondly, I look at Rand J's appeal to the rule of law to deliver a damages remedy to Mr Roncarelli, a remedy whose implicit predicate was the need to develop specifically public law principles of tort liability to meet those exceptional cases of public officials whose gross abuse of their power harms individuals without violating any of their legally protected interests. The public tort has tracked in broadly similar fashion in Canada, Britain and Australia, but the latter's reluctance to use the language of the rule of law as an operative legal principle might soon see some significant divergence.

2. Legalism and the rule of law

Mr Roncarelli needed judicial protection because the relevant Act gave him none. His annually renewable liquor permit was at the mercy of a baldly stated bureaucratic discretion: "The Commission may cancel any permit at its discretion."[5] The Act required neither hearings nor reasons, and it stipulated no grounds. Despite this, Rand J deployed several standard interpretive techniques and one not-so-standard technique (the rule of law) to "supply the omission of the legislature".[6]

Mr Roncarelli would have had considerably more statutory protection if he had been trading in Australia's Sydney instead of Montréal.[7] He would have needed a liquor "permit", which was less regulated than a publican's "licence" and easier to obtain.[8] Annual renewals were virtually automatic.[9] Licences could be cancelled for serious criminal convictions or for at least four lesser convictions over the previous year, and permits were revocable on the grounds of the neighbourhood's interests or any other reasonable cause. Magistrates determined all grants and revocations, with very generous appeal rights. Licences and permits passed to their holders' spouse or adult children in the event of death. The same applied where licensees were imprisoned for felony; and appropriately adapted transmission provisions also applied for both licences and permits in the event of bankruptcy and insanity. Even though they needed annual renewal, therefore, the permits and licences were a good deal more secure in Sydney than in Montréal.

Although Sydney's restaurateurs had greater legislative protection than their Montréal counterparts, there were other regulatory domains which appeared on their face to be as discretionary as that in Roncarelli. Even these, however, would have been judicially construed so as to require that the discretions be exercised by reference only to considerations having some rational and functional connection to the legislation's regulatory objects. Furthermore, in its inimitable "dense grinding judicial style",[10] the High Court back in Roncarelli's time was interpreting a Constitutional guarantee of free trade across State borders as necessarily requiring statutory limits to administrative discretion. For its part, the High Court in these cases scarcely mentioned the highest ideals of the rule of law, but it made very clear its concerns about the potential for the executive's arbitrary interference with the rights of private property. Rand J's judgment style is less technical than the High Court's, but his invocation of the rule of law may have been triggered by similar concerns for middle class status and its members' property rights.

There are doubtless several reasons why those in the liquor trade had more legislated protections in Sydney than Montréal in the late 1940s, but it has been a long time since liquor and the rule of law formed part of the same debate in Australia. So long, in fact, that no-one used the language of the rule of law back then; rather, they spoke of the rights of Englishmen, and the foremost of these were the rights of person and property.

The present Chief Justice of New South Wales wrote that Australia has experienced only two periods of flagrant breach of the rule of law.[11] His history was too kind, but his first example did involve alcohol. The event was the military overthrow of Governor Bligh (his second mutiny) in 1808, followed by an interregnum of almost two years of serious instability. It eventually became known as the Rum Rebellion, although it in fact had almost nothing to do with alcohol and a lot to do with rights of property, speculative development, due process, and (being Sydney) Harbour views.[12]

Like any other country's ultimate court of appeal, Australia's High Court is not averse to talking about the rule of law, indeed to singing its praises. However, except where this is a rhetorical flourish, the songs are usually about identifying the institutional design features, processes and values which might be said to have been constitutionally embedded within the judicial branch of the Commonwealth. The immediate aim of the High Court's version of the rule of law is the protection of the courts themselves, although, of course, the trickle-down benefits for everyone else are always claimed.[13] Beyond the protection of the judicial branch, and in the absence of either an entrenched or a statutory Charter of Rights, the High Court's hymn sheet is necessarily more brief than that of any comparable appellate court. Judicial review, for example, may deliver on rule of law values, but they have no "immediate normative operation".[14] Similarly, the court has rejected an attempt to create a new constitutional tort of breaching the rule of law by causing intentional harm.[15] The court wants rules, and the smaller and more precise the rule, the more comfortable it feels.[16] It expresses hostility to "top down" reasoning,[17] and its commitment to doctrinal stability is so strong that on one view, counsel needs leave before questioning a High Court precedent, at least in constitutional cases.[18]

By the time that Roncarelli was decided, Australia already had a string of precedents for confining statutory discretions to considerations functionally relevant to the regulatory scheme in question. As importantly, many of its regulatory schemes contained the sort of protective detail so sadly lacking in Roncarelli. Indeed, so far as they affected multi-State businesses, many of Australia's regulatory laws tried to avoid wide discretionary power precisely because the High Court's highly formalistic constitutional learning took the breadth of a regulatory discretion as an important indicator of constitutional invalidity.

Abolition of trade barriers between the Australasian colonies was one of the principal drivers behind the decision to join a federal union, but the constitutional expression of that ideal was very poorly drafted. Section 92 of the Constitution provided that "trade, commerce and intercourse among the States ... shall be absolutely free," but for almost 90 years, the biggest question was: "free (indeed, 'absolutely free') of what?"[19] Tariffs, obviously, but what else? After roughly 140 attempts at answering that question, attempts that for the most part deliberately prioritised the purity of analytical doctrine over functionality of outcome, the High Court came up with a functional test justified by reference to the section's drafting history. Business was to be free of legislative and administrative measures that discriminated between in-State and inter-State trade or commerce for protectionist reasons or with protectionist effect.[20] Until that point, however, the court had propounded all sorts of confusing tests. It was obvious that s 92 was not meant to invalidate all laws and administrative practices that might apply to inter-State business. Even laissez faire economics of the kind leading up to the recent global financial crisis tolerated some law affecting markets and some law regulating the production and exchange of goods and services. Besides, the High Court always denied constitutionalising laissez faire economics,[21] or any other theory of economics.[22] The court often said that an "ordered society"[23] was obviously acceptable, indeed necessary, and with no difference in meaning, this was sometimes rendered in terms of "ordered liberty".[24] The real difficulties lay in defining what this might mean.

"Ordered liberty" was the term that Sir Owen Dixon used in his speech on the occasion of his swearing-in as Australia's Chief Justice in 1952. Sir Owen swore fealty to the rule of law, but in a passage that continues to puzzle and even outrage the academy, he claimed that the best way that the courts could enforce the rule of law was by shunning any attempt at producing "constructive" outcomes, and instead adopting a method of "strict and complete legalism".[25] In an obvious reference to s 92 of the Constitution, he said that the courts' administration of the rule of law "offers a reconciliation of ordered liberty with planned control".[26] In those days, that meant that the scales were weighted very heavily against the administrative state and in favour of "ordered liberty", which usually boiled down to meaning an absolute minimum of regulatory interference with inter-State trade or commerce. Despite the court's denial of a commitment to laissez faire, some of its decisions had that practical effect.[27] One of the extensions of "ordered liberty" that the cases had allowed in those days was for "regulation"; not all regulation, but only such as was shown to be minimally acceptable or necessary.[28] These distinctions were obviously fraught, but one approach was to invalidate any regulatory scheme that was too discretionary. To pass constitutional muster under that approach, regulatory schemes had to stipulate the considerations by which discretionary power was to be controlled, and those considerations were to exclude any reference to whether the relevant activity had an out-of-State origin.[29]

By the time that Roncarelli was decided, therefore, the Australian legal context had some well-established public law constraints on the exercise of regulatory discretion. That is not to deny the prevalent fear of an ever-expanding administrative state. Nor is it to deny concerns over absolute discretionary power – the common law had not yet acquired enough self-confidence (or hubris) to confront such a concept head-on.[30] However, the legal landscape was already populated by various regulatory regimes in which administrative discretion was not absolute but constrained. It was constrained by judicial supervision according to generalised principles that were paying less and less heed to distinctions between privileges and rights,[31] or between administrative and quasi-judicial powers or functions. As in the case of restaurateurs, some intra-State businesses dependent on licences or permits were well-protected by statute. Others whose regulatory regimes lacked explicit constraints were nevertheless protected by well-established techniques of statutory interpretation, whereby the judges implied functionally relevant limits to discretionary power from the subject matter, scope and purposes of the relevant legislation.[32] The same interpretive techniques were used to save legislation from constitutional invalidation for overreaching the ambit of federal legislative power. The Regulation in Shrimpton v Commonwealth,[33] for example, gave the Treasurer an "absolute discretion" to impose "such conditions as he thinks fit" when granting consent to certain property transfers. However, Latham CJ, McTiernan and Dixon JJ read that down so as to exclude arbitrary or personal considerations, or any other considerations functionally unrelated to the proper operation of the defence power, which was the only relevant head of legislative power available to federal legislators.

The High Court's restrictive approach to regulatory controls on inter-State business was therefore consistent with a more generalised leaning against broad discretionary powers, although in domains protected by s 92 of the Constitution, the court was able to go one step further by insisting that regulatory statutes had to minimise discretionary power. Where Acts challenged for violating s 92 lacked explicit restrictions on discretionary power, government counsel sometimes defended that the relevant discretion was not to be read literally, because it was subject to the common law's general administrative law principles.[34]

Considerations of class and property are not so prominent in current discussions of the rule of law,[35] but if the "rule of law" itself transitioned from the "rights of Englishmen", then it is right to accord some room to these considerations. One suspects that in Roncarelli's time, money, class and economic freedom figured at least as prominently as considerations of equality, free speech, freedom of association, and freedom from discrimination on religious grounds. Each of those is an important topic, of course, but not one of them scored an explicit mention in Roncarelli, although they did figure in earlier instalments of the mini-series dubbed the implied Bill of Rights.[36]

Rand J discussed only one of what we might now call Mr Roncarelli's fundamental rights or freedoms, and it was an economic freedom -- his freedom to continue in his chosen (indeed, inherited) vocation as a respectable upmarket restaurateur, so long as he complied with the law generally and with the rules pertaining specifically to licensed restaurants. If that looks like too narrow a description of Mr Roncarelli's economic rights, I would point to the judgment itself, which stressed his respectability and middle class qualities. He ran a restaurant of a "superior class",[37] and he was well-educated and of good repute.[38] He had leased a meeting hall to the Jehovah's Witnesses for their use in Sherbrooke, but that was categorically irrelevant (doubtless an unimpeachable right of property).[39] Importantly, it seems,[40] he had not himself distributed any of the tracts which had so upset the general public on religious grounds, nor the one which appeared to have upset the government on more self-serving grounds.[41] The only possible marks against his character, therefore, were the facts that he was himself a Witness and had stood bail for his co-religionists many times, both facts being perfectly legal and utterly irrelevant to the way he conducted his business.[42] The Attorney General had deliberately destroyed "the vital business interests of a citizen",[43] and sentenced him to "vocational outlawry".[44] As for the business perspective, the licences may legally have been a "privilege", but they were vital to a "superior class" restaurant, and licensees invested considerable money and effort into their businesses on the assumption that their licences would continue.[45] Importantly, Mr Roncarelli's citizenship status[46] would in previous times have allowed him to conduct his perfectly "ordinary"[47] and "legitimate"[48] business activities free of state interference, and that was a major driver of Rand J's judgment. Licensing legislation was steadily enveloping "occupations and businesses of this nature" previously free of it;[49] indeed, "economic activities" more generally were coming under regulatory control.[50] "Privileges" they may have been, but his Honour thought it essential that their grant, administration and revocation be not arbitrary but impartial, governed by considerations relevant only to the licensed or regulated activity in question.[51]

An ungenerous reader might read Rand J as having taken a stand for the middle class, the self-employed small business person so vital to economic life and so much in need of protection from the administrative state. And he had been able to do this without having to turn Mr Roncarelli's liquor permit into a species of "property".[52] In a telling qualification, he even added that his judgment was not intended to protect government workers.[53] Their rights were doubtless left to labour law, admittedly a more elaborately protective affair those days than now.

If Rand J had been concerned only with Mr Roncarelli's economic rights, then his judgment would be little more (and perhaps a little less) than the then-popular justification for a rapid expansion of judicial review doctrine to catch up with and counter-balance an administrative state which had grown fearfully large in most common law jurisdictions. But it would clearly be unfair to treat Rand as a Canadian version of Lord Denning, ever protective of middle class virtue whilst denying it to the working class (particularly if they were unionised).[54] Mr Roncarelli's middle class status and virtues, and even his ill-defined citizenship, can lead one to different places – most obviously, they might have placed him in the category of "deserving claimant". But Denning would never have thought him deserving; he rocked the boat, challenged police,[55] frustrated the forces of law and order, and did all this with profits derived from government beneficence in the form of a discretionary licence.[56]

Rand J's style was less concerned with analytical precision of doctrine than Dixon J's, although they each saw the rule of law as something that could help civilise the burgeoning regulatory state's interference with legitimate business interests previously free of bureaucratic impediment. And Rand J was more generous than Dixon J when it came to the question of remedies. Dixon J would have had no trouble in invalidating the Attorney General's actions. In Australia, Mr Roncarelli would have been spoilt for choice in terms of grounds of invalidity. The permit cancellation was based on impermissible considerations, taken for an impermissible purpose, and taken either by the wrong person or by the right person acting under the Attorney's dictation. The life-long ban on Mr Roncarelli would have been invalid on the same grounds, and also because it attempted to fetter the future exercise of a discretionary power. None of that would have been remarkable. But Rand J upheld Mr Roncarelli's right to damages, and for Australians, that was indeed exceptional. Indeed, I suspect that in both countries, damages for unlawful licence cancellation would still be exceptional.

3. Malice, damages and the rule of law

My purpose in this part of the paper is to ask why Rand J said that the Attorney General had behaved in bad faith or maliciously. Those are very serious conclusions and at first glance, one might wonder whether they were necessary, especially because his Honour defined them so broadly as to strip them of any necessary connection with dishonesty or lack of public purpose. Furthermore, the other majority judgments refrained from such conclusions.

Rand J's definition of malice started out unremarkably: "Malice in the proper sense is simply acting for a reason and purpose knowingly foreign to the administration ...."[57] But it was clear that the Attorney General thought that he was acting in the public interest and that his powers were unfettered. He did not know that his reasons and purpose were ultra vires, and so Rand J quickly expanded his definition to include conduct that the Attorney should have known was unauthorised:[58]

"'Good faith' in this context, applicable both to the respondent and the general manager, means carrying out the statute according to its intent and for its purpose; it means good faith in acting with a rational appreciation of that intent and purpose and not with an improper intent and for an alien purpose; it does not mean for the purposes of punishing a person for exercising an unchallengeable right; it does not mean arbitrarily and illegally attempting to divest a citizen of an incident of his civil status."

It is doubtful that Rand J invoked malice to outflank what would otherwise have been the immunity of quasi-judicial functions from damages actions,[59] because his preference was to treat the Attorney General's functions as administrative.[60] Further, only Cartwright J (in dissent) had thought that the characterisation of the Attorney's functions was important.[61] It is submitted that the importance of "malice" for Rand J was that he saw it as far more egregious in a public official than a private person, and it was that perception which under-pinned his assertion that Mr Roncarelli could avail himself of a wholly public tort.

Of the majority, only Rand J fully acknowledged the difficulty in theorising a basis for awarding damages to Mr Roncarelli, and even he left some fairly big question marks. I agree with David Mullan's observation in this collection that one cannot be entirely sure how his Honour connected the two components of an excess of discretionary power on the one hand (something more obviously relevant had the case been brought against the Liquor Commissioner), and on the other hand, the Attorney's usurpation of a power that belonged only to the Commissioner.

There was no obviously applicable private law tort, and the House of Lords had established in Allen v Flood[62] that in the absence of a conspiracy (and none was pleaded in Roncarelli), the intentional and malicious infliction of economic harm upon a plaintiff was perfectly acceptable market behaviour. But the Attorney had acted as a public official, not as a player in the market, and that was Rand J's critical distinction. When tort doctrine invoked a distinction between public and private functions (or defendants), it was usually in order to reduce government liability, but Rand J's distinction served to take him in the opposite direction.

Rand J distinguished Allen in one, and possibly two ways. His first distinction was that Roncarelli was a public law matter:[63]

"Here the act done was in relation to a public administration affecting the rights of a citizen to enjoy a public privilege, and a duty implied by the statute toward the victim was violated. The existing permit was an interest for which the appellant was entitled to protection against any unauthorized interference, and the illegal destruction of which gave rise to a remedy for the damages suffered. In Allen v Flood there were no such elements."

In other words, malice in the sense of the intentional infliction of harm was actionable if (but only if) there was either a conspiracy (Allen), or if the tortfeasor was a public official acting beyond his or her powers. The plaintiff had also argued that the defendant might have been liable for intentional interference with the plaintiff's permit even if there had been no excess of power. Apart from noting that such a tort would need to be confined to public functions if it were to avoid conflicting with Allen,[64] Rand J did not pursue this intriguing possibility, but his language in the passage quoted above betrays his Honour's uncertainty about the nature of his public law tort, and an ambivalence as to the subject matter of the tort's protection.

The "rights of a citizen to enjoy a public privilege" are odd rights indeed, and their correlative "duty implied by the statute" is even more curious. Rand J was not treating the liquor permit as a property right. It is suggested that his Honour might have had in mind the obvious assumptions and expectations underlying the commercial viability of any occupational licensing scheme. The licences might need annual renewal, but investors would naturally expect the scheme's administration to be run on commercially predictable (and therefore stable) lines. The stable (as opposed to arbitrary) administration of the licensing scheme might well be described as a right to enjoy such licences as have been granted, although it is suggested that Rand J's tort and its successor known as the tort of misfeasance in public office are best approached not as mechanisms to protect legal rights or interests, but as mechanisms to discipline public officers for abuses of public power which they knew were inexcusable.[65] Mr Roncarelli had no right to maintain or renew his permit, nor any right to a stable economic environment free of government interference; that is why his damages award focused on disciplining arbitrary public behaviour. It may also go some way to explaining why the quantum of that award was so obviously inadequate, and the reasoning offered in support of that quantum so unconvincing.[66] The court was obviously torn between a torts model of damages assessment that focuses on the plaintiff and seeks to put him or her in the same position as if they had not been wronged, and the legal form of the liquor scheme with its annual time lines and broad discretionary power. Mr Roncarelli's damages award may have sufficed to denounce the defendant's conduct, but it is hard to believe that it compensated him for his loss.

Roncarelli appears in most histories of the evolution of a specifically public law tort of misfeasance in public office, although no such label appeared in the case itself. It will be recalled that Rand J's version of the rule of law required the court to afford Mr Roncarelli "recourse or remedy"[67] for the defendant's extraordinary behaviour. There is an obvious similarity to Holt CJ's famous reasoning in Ashby v White[68] that the plaintiff's rights having been breached, the law simply had to create a damages remedy as a means of vindication.[69] The restrictions inherent in the more traditional ("private law") torts combined in Rand J's judgment with a strong sense of the need to vindicate Mr Roncarelli's rights to drive the development of what we would now call the tort of misfeasance in public office. This public law tort is a fall-back for those hopefully rare cases where private law torts are manifestly inadequate. Australia also sees misfeasance as a residual, back-up tort.[70]

Australian tort law operates for the most part on a private sector model, whose prime function is to focus on and adjust the competing interests of plaintiffs and defendants to the extent that their activities collide. Government parties to tort actions can usually make the same claims and are usually subject to the same liability rules as would be made or applied in an action between subjects. This equality principle is highly prized. Dicey treated it as a critical aspect of his rule of law,[71] and more often than not, its common law and statutory exceptions are bitterly contested,[72] usually on the ground that they are unjustified reductions of the government's exposure to liability principles or damages awards. The equality principle's flip-side, however, is that with only one exception, there is no special tort that only governments commit.[73] The exception is the tort of misfeasance in public office, an exception that seeks its justification on the basis that there is something especially wrong about malice or dishonesty when it comes from a public official.

Individuals who suffer loss as a result of government action found to have been unconstitutional cannot get damages in Australia on that score alone – plaintiffs must bring their claims within the standard private law causes of action, because the High Court refused to create constitutional torts. That was in James v Commonwealth,[74] in which the plaintiff had sought compensation following his successful challenge[75] to the validity of Commonwealth legislation that severely limited the inter-State sale of his dried fruits. The court allowed him compensation for the seizure of goods en route for inter-State sale, but that was because the facts fell within the established tort of conversion. There had been a direct interference with his rights of property, and the good faith of the public officials was no defence because conversion is a tort of strict liability.

However, the court denied the greater part of Mr James' claim for compensation. He had lost inter-State business opportunities because he had found it difficult to secure the services of common carriers, who feared the prospect of prosecution. But the implicit threats to the carriers constituted no tort. Even if one assumed conspiracy's "agreement",[76] the threats would not have amounted to a tortious conspiracy, because "unlawful" means had not been used; threatening a prosecution was not relevantly unlawful. Nor did the threats amount to the tort of intentionally inducing common carriers to breach their common law duties to accept Mr James' business. First, the passage of an Act was not an "inducement" by the Executive, but a prior step taken by the legislative branch for which the Executive should not be held tortiously responsible.[77] Secondly, inducements would not by themselves have been sufficient; there should have been more evidence that the inducements succeeded in the sense that they resulted in actual breaches of the carriers' duties.[78] Thirdly, no express or implied threats to prosecute carriers could count as inducements unless they had been made in the knowledge that the legislative measures were unconstitutional.[79] As for the threats not to his carriers, but to Mr James personally, the Commonwealth might in other circumstances have been liable for intentionally harming his business through threats of further illegal seizures.[80] But the trouble for Mr James was that he had been too plucky – the implicit threats had not in fact diminished his determination.[81] Although not addressed directly, it is clear that the Commonwealth would not have been liable for threatening to take Mr James to court, provided it had acted in good faith.[82]

If one were to fast-forward to the present, one would have to conclude that certainly in England and probably in Australia, the relevant private law torts are at least as demanding as they were in the times of James and Roncarelli.

In its recent magisterial review of two of the more bewildering economic torts in a trilogy of appeals generally cited as OBG Ltd v Allan,[83] the House of Lords unanimously and decisively rejected more than a century of speculative judgments and scholarship which had urged various extensions of the torts in question. OBG rejected the contention that there was (or even should be) a single, grand principle to unite the tort of knowingly inducing a breach of contract with the tort of intentionally harming the plaintiff's trade or business interests by unlawful means. The two torts had too many principled differences to allow them to be repackaged beneath a unified (and grander) principle. A majority also rejected a proposal to extend the strict liability tort of conversion beyond its protection of chattel interests, to include the misappropriation of choses in action. A larger majority also tightened the definition of what might amount to unlawful means for the purposes of the tort of deliberately harming the plaintiff's interests by unlawful means. Such means must in future not only be actionable at the suit of the third person against whom they were directed;[84] they must also interfere with that person's freedom to deal with the plaintiff.[85] Lord Hoffmann stressed that the elements of intention and unlawfulness were important control devices built into the economic torts in order to minimise the common law role in "devising rules of fair competition", and its role in stipulating "basic standards of civilised behaviour in economic competition, between traders or between employers and labour".[86] That is why his definition of unlawful means required that they be both actionable at the instance of the third person (thereby excluding means that were unlawful only because they were in breach of a criminal or regulatory statute),[87] and used for the very purpose of harming the plaintiff's economic interests.[88]

It might be interpolated at this point that to some extent, the Australian High Court anticipated OBG's retraction of "unlawful means" by almost a decade.[89] It held that if there is an economic tort of interfering with a person's trade or business interests by unlawful means, then the means will not be relevantly unlawful simply because they were unauthorised or ultra vires. One reason was that were it otherwise, the economic tort would render redundant the tort of misfeasance in public office. Another reason was that it would extend a public officer's liability for the putative economic tort beyond that of private actors, who must also know that they are acting unlawfully or be recklessly indifferent about that. More importantly, the High Court pointed out that the economic and misfeasance torts have different defendants, and their tortfeasors have different intentions. Individual economic tortfeasors will usually transmit vicarious liability to their employers without much difficulty, because they will have acted for their employers' benefit. Individual misfeasance tortfeasors, on the other hand, will by definition have acted beyond their authority and will be more likely to have acted in pursuit of their own ends (as opposed to those of their employers). That will make it more difficult (but not impossible)[90] to pin their employers with vicarious liability.[91] As for different intentions, economic tortfeasors usually intend to inflict economic loss on their competitors, and that is perfectly acceptable, because it "is central to competition".[92] By contrast, the intentional (and unlawful) infliction of harm by government actors is often said to be sufficient in itself to establish misfeasance.[93]

If OBG's economic torts are all about devising "basic standards of civilised behaviour" in the market place, the misfeasance tort is (or at least, should be) all about defining the exceptional circumstances in which government illegalities will be actionable in tort, not just because they were unlawful (because judicial review is normally sufficient), but because they violated basic standards of civilised behaviour in the exercise of public power. That, surely, is why Rand J fell back on something as broad as the rule of law, and also why he fudged his definition of malice so that it went beyond spite, dishonesty, or a deliberate excess of power to something equally objectionable in a public officer if less morally reprehensible at a personal level – intentional harm at the hands of someone behaving like an autocrat, someone whose ignorance of his legal limits was so spectacular as to be entirely unforgiveable.

Perhaps the most fundamental choice confronting the House of Lords in OBG was whether to describe the limits of the private law tort of interference by unlawful means in terms that were tightly defined or open-ended. The open-ended model would have amounted to an endorsement of what is known in America as a prima facie tort, defined as an intention to inflict harm without just cause or excuse.[94] Despite the eminence of some of its proponents,[95] their Lordships baulked at an approach whose definition of unlawful means would have required the judges to determine in every case what is unjust or inexcusable – that was thought to be too uncertain for a tort designed to regulate market-place behaviour.[96]

The case for certainty is less compelling when it comes to misfeasance, and that might explain why Rand J fudged his definition of malice. Malice currently figures prominently in most accounts of misfeasance, but the tort's expansion from deliberate wrong-doing to reckless indifference has made "malice" a slippery word. It has been said that "malice" has been subject to a "regrettable exuberance of definition",[97] and that the word has caused "more confusion in English law than any judge can hope to dispel".[98] That is certainly true of its use in the misfeasance cases, where its definitions range from narrow to broad. Starting with improper motives (such as ill will, spite and revenge), one can also find definitions that extend by degrees to improper intentions as to outcomes (such as the intentional infliction of harm),[99] action taken in the knowledge that it was beyond lawful authority, and circumstances in which public officials were recklessly indifferent as to whether they had the necessary lawful authority.

One currently popular definition of misfeasance in public office splits into two alternative limbs. Various two-limbed versions have been propounded, but a broad overview is sufficient for present purposes.[100] In each limb, the relevant act or omission of the public officer must have been a purported exercise of a public power or function that was invalid either for being in excess of power for any of the standard administrative law reasons, or because it was wholly lacking in lawful authority from the outset. The difficulties of detail in that proposition will not be pursued. Nor is it necessary here to examine the implications of Canada's extension of the tort beyond coercive powers, to encompass deliberate breach of statutory duties.[101] The concern of this part of the paper is to focus on the tort's mental elements.

The only mental element in the first limb is that the officer must have specifically intended to harm the plaintiff, and that is sometimes called "targeted malice". The second limb is more complicated, because it addresses two issues – the extent of the officer's awareness that his or her act lacked lawful authority, and the extent of the officer's awareness of the relevant act's probable consequences for the plaintiff.[102] On each of those second-limb issues, the plaintiff must establish either actual knowledge or reckless indifference.

Although the cases repeat it endlessly, not everyone subscribes to a two-limb account. In England, for example, Lord Hobhouse treated the first limb not as an ingredient of the misfeasance tort, but as a proposition of evidence. In his Lordship's view, an officer who sought to inflict harm on a particular person was "extremely" unlikely to have believed that he or she was acting with lawful authority.[103] Lord Millett went further; he treated both limbs as no more than propositions of evidence. For him, the ultimate fact in issue was "abuse of power", which "involves other concepts, such as dishonesty, bad faith and improper purpose".[104] Analogising to trust law, Lord Millett said that the plaintiff hadt in one way or another to show that the defendant officer did not believe that he or she was acting for the benefit of at least some of those for whom such officer was were required to act. His Lordship acknowledged that one consequence of his approach would be to exonerate an officer who deliberately acted in excess of power if he or she believed that it was for the plaintiff's benefit.[105]

In Australia, an intermediate appellate court unearthed another problem with the first limb.[106] The defendant Minister's acts fell squarely within a literal reading of the first limb. His good faith breach of natural justice had meant that he had acted unlawfully, and he had definitely intended to harm the plaintiff – he had sought his dismissal from a statutory agency. The trouble was that harming the plaintiff was the very purpose of the power; perhaps unusually, that was lawful.[107] The court exonerated the Minister because he had not acted dishonestly. Rather, he had sought the plaintiff's dismissal for the permissible purpose of cleaning up what he had believed to have been an incompetent agency within the Minister's portfolio of responsibilities. It appears likely that the Canadian Supreme Court would have reached the same result on those facts.[108]

Perhaps the first-limb problems could be somewhat reduced by insisting on the distinction between improper motives and improper purposes. That would be consistent with administrative law doctrine, which focuses on purposes rather than motives. But these are more than just tricky doctrinal difficulties with the two-limb approach. They flow from the difference between defining misfeasance in terms of tight rules or more flexible principles. There is an obvious parallel with OBG's policy choice between closed rules or a prima facie tort, but the call of principle is stronger in the misfeasance arena, in which government is meant at all times to behave with more altruism and personal detachment. Rand J's principle was the rule of law, which warranted a damages award for the intentional infliction of harm by a public officer who he said was malicious, but the Attorney's malice consisted only of honest yet egregious ingredients. First, there was hubris or stupidity (nowadays called reckless indifference) – even a cub lawyer should have realised that the power lay with the Commissioner, not the Attorney. And secondly, powers that were all about liquor were used for purposes that were all about crushing lawful political and religious dissent. To make matters worse, they were used to close down a perfectly respectable small business whose regulation might well be regarded as an exercise in red tape.

The leading misfeasance cases all make frequent appeals to principle. Perhaps the principles most frequently voiced are knowing want of power, dishonesty, and the failure to make an "honest attempt"[109] to use power lawfully. These are all principles focused on the putative tortfeasor's mental state, a state ranging fairly narrowly from knowledge and the deliberate infliction of harm, to "reckless indifference", which probably translates as knowledge of the risk (of illegality or harm to the defendant or both) and indifference as to whether the risk will materialise.[110] One could offer several reasons why misfeasance claims are exceptional, and why their success is even more exceptional. However, the centrality of the defendant's mental state must surely count as one of the principal explanations. This is partly because its proof will usually be extraordinarily difficult. Defendants are unlikely to acknowledge moral impropriety,[111] and reliable written proof will be rare. Furthermore, proof of the requisite mental state might well carry with it an implicit suggestion that the defendant is unfit for office,[112] which would mean that in practical terms, there will be a tougher standard of proof.[113]

The rule of law also figures in the English[114] and Canadian[115] judgments as one of the principles informing the misfeasance tort, but the Australian High Court was distinctly hostile[116] to a lower court's attempt[117] to side-step the niceties of the misfeasance tort by creating a new tort -- a constitutional tort of breaching the rule of law. As formulated, the new tort would have rendered actionable every harm intentionally inflicted by officers unwittingly acting beyond their powers. Further, it would have been more demanding of government than a negligence-standard duty of care to stay within its lawful authority, a duty, incidentally, that the Australian courts reject as a step too far for the tort of negligence.[118] On the surface, the High Court's objections were not to the rule of law language, but to a judgment that had slipped too easily from a requirement of government legality to a requirement of a remedy in damages, a judgment, incidentally, that had understandably[119] drawn support in this respect from the majority judgments in Roncarelli other than that of Rand J. At a deeper level, however, I suspect that Rand J and the Australian High Court had different understandings of the rule of law. In Mark Walters' typology,[120] Rand J went beyond the rule of law as order, as an abstraction from the mass of common law cases validated largely by its descriptive accuracy, its explanatory power and its coherence. Rather, his Honour's understanding conformed to Walters' model of the rule of law as reason, a distinctly substantive and normative conception that drew on legal and political ideals, which is why it had (and still has) such transformative potential.

4. Conclusions

It would be easy to exaggerate the differences between Rand J's approach to the issues in Roncarelli and the way in which the High Court would have dealt with the same issues. Small differences in judicial styles, however, can produce larger consequences over time. A court that is reluctant to give normative bite to rule of law principles will find it correspondingly difficult to define a role for a tort designed only for the public sphere.

So far as the case involved the issue of the validity of the decision to cancel Mr Roncarelli's liquor permit, Rand J used the language of the rule of law where an Australian court would have seen no need. Had he been in Sydney, Mr Roncarelli's decision-makers would have been courts, not a Minister; the criteria for making decisions would have been narrower; and he would have had back-up protection on any one of a number of grounds via judicial review. In the judicial review arena, Australia's High Court had grown accustomed to reading down statutory grants of broad discretionary powers, sometimes because that was the only way to save a federal law from invalidation, but usually because the court's general administrative law principles were increasingly invoked to read down apparently unlimited grants of discretionary power.

More notable for Australians, therefore, was Roncarelli's decision to uphold the trial judge's decision to award damages. Based on the rule of law and appealing to a normative distinction between liability for private and public action, Canada's misfeasance tort is now heading towards an open-textured concept of morally blameworthy public behaviour. Australia's misfeasance tort is probably heading in the opposite direction. Although this is not yet clear, the odds are that it will end up focusing only on the abuse of public power, and being defined in minimalist and rule-focused terms, as would befit its perceived function as a very occasional fall-back for the private torts.

---

[∗] Law Faculty, University of New South Wales. I would like to thank Jim Davis, Carol Harlow and John Randall for their assistance at different stages of this paper. I would also like to thank all of the contributors to the September 2009 workshop on Roncarelli, but especially David Mullan.

[1] [1959] SCR 121 [Roncarelli].

[2] Roncarelli, n 1, at 142.

[3] An electronic search of Australia's largest case base (www.Austlii.edu.au ) produced only 7 court and tribunal decisions mentioning Roncarelli.

[4] The defendant was both Provincial Premier and Attorney General, but his most obvious legal error was as Attorney, in assuming that he could exercise a power vested by statute in another.

[5] Alcoholic Liquor Act RSQ 1941, c 255, s 35(1).

[6] As in Cooper v Wandsworth Board of Works [1863] EngR 424; (1863) 14 CBNS 180 at 194, per Byles J.

[7] What follows are broad summaries of two Acts as they were at the time of the events which gave rise to the Roncarelli litigation, namely, the Liquor Act 1912 (NSW) and the appeal provisions of the Justices Act 1902 (NSW).

[8] The criteria related to local needs and amenity, the condition of the premises, applicants' compliance with the relevant laws, and their character and reputation.

[9] Although the police, local residents and commercial competitors could lodge objections on the same grounds that applied to objections to first-time applications.

[10] A Mason, "Justice of the High Court", in T L H McCormack and C Saunders (eds) Sir Ninian Stephen: A Tribute (Miegunyah Press, Carlton, Australia, 2007) 3 at 5.

[11] J Spigelman, "Bicentenary of the Coup of 1808" (2008) 30 Legal History 1.

[12] See also H V Evatt, Rum Rebellion: A Study of the Overthrow of Governor Bligh by John Macarthur and the New South WalesCorps (Angus & Robertson, Sydney, 1938). The second event involved a racist colonial Premier appealing to popular prejudice by refusing, for a short time, to obey a habeas corpus designed to let Chinese passengers disembark.

[13] The High Court's most famous invocation of the rule of law was in Australian Communist Party v Commonwealth [1951] HCA 5; (1951) 83 CLR 1 at 193, a case which held that an Act with appalling consequences for civil liberties was invalid for a number of reasons, but largely for its attempt to prevent the court from having the last word on a particular issue. The court's more recent response to an ouster clause was similar in Plaintiff S157/2002 v Commonwealth (2003) 211 CLR 476 at 482 [5] and 513-514 [103]-[104]; [2003] HCA 2 (Plaintiff S157).

[14] Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Lam (2003) 214 CLR 1 at 23 [72]; [2003] HCA 6, per McHugh and Gummow JJ.

[15] Northern Territory v Mengel (1995) 185 CLR 307 [Mengel], discussed in Part 3 below.

[16] See J D Heydon, "Judicial Activism and the Death of the Rule of Law" (2003) 47 Quadrant 9, also published in [2004] OtaLawRw 2; (2004) 10 Otago Law Review 493.

[17] See: Bofinger v Kingsway Group Ltd [2009] HCA 44; (2009) 239 CLR 269 at 300-301 [90]- [94]; [2009] HCA 44; and M Aronson, "Process, Quality and Variable Standards: Responding to an Agent Provocateur" in D Dyzenhaus, M Hunt and G Huscroft (eds) A Simple Common Lawyer: Essays in Honour of Michael Taggart (Hart Publishing, Oxford, 2009) 5 at 22-28.

[18] See Permanent Trustee Australia Ltd v Commissioner of State Revenue (Vic) (2004) 220 CLR 388 at 451-453 [176]-[181]; [2004] HCA 53.

[19] For the most entertaining (and thoroughly political) history of the High Court's treatment of s 92, one cannot go past D Marr, Barwick (Allen and Unwin, Sydney, 1980), esp chs 6-7, 11 and 17.

[20] Cole v Whitfield [1988] HCA 18; (1988) 165 CLR 360 [Cole].

[21] See, eg: Home Benefits Pty Ltd v Crafter [1939] HCA 12; (1939) 61 CLR 701 at 731-732; and Milk Board (NSW) v Metropolitan Cream Pty Ltd [1939] HCA 28; (1939) 62 CLR 116 at 151.

[22] North Eastern Dairy Co Ltd v Dairy Industry Authority (NSW) [1975] HCA 45; (1975) 134 CLR 559 at 615.

[23] See eg Hughes and Vale Pty Ltd v New South Wales [No 2] [1955] HCA 28; (1955) 93 CLR 127 at 159, 171 and 219, [Hughes 2]; Uebergang v Australian Wheat Board [1980] HCA 40; (1980) 145 CLR 266 at 281 and 305, [Uebergang]; Miller v TCN Channel Nine Pty Ltd [1986] HCA 60; (1986) 161 CLR 556 at 629, [Miller]; and Cole, n 20 at 403.

[24] Duncan v Queensland (1916) 22 CLR 556 at 592; and Mikasa (NSW) Pty Ltd v Festival Stores [1972] HCA 69; (1972) 127 CLR 617 at 653.

[25] "Swearing in of Sir Owen Dixon as Chief Justice" (1952) 85 CLR xi at xiv [Swearing in].

[26] Swearing in, n 25 at xiv.

[27] See: Uebergang, n 23 at 309-310; Miller, n 23 at 571 and 618; and Cole, n 20 at 403.

[28] Miller, n 23 at 600.

[29] See: Hughes and Vale Pty Ltd v New South Wales [1954] UKPCHCA 5; (1954) 93 CLR 1 at 26-27; Hughes 2, n 23 at 162-163; and Ackroyd v McKechnie [1986] HCA 43; (1986) 161 CLR 60 at 68. Dawson J summarised thus in Miller, n 23 at 628: "[I]t is clearly established that a prohibition, subject only to an unfettered executive discretion to issue or refuse a licence, goes beyond regulation which may be permissible having regard to the guarantee afforded by s 92 ...".

[30] These days, the High Court doubts that there can ever be a truly unfettered discretion in the federal arena, because that would exceed constitutional limitations on legislative competency: Plaintiff S157, n 13, at 512-513 [102]-[103]. State legislatures have fewer limitations on legislative competency, but "very plain" words would be needed before State grants of unfettered power were to be read literally (Hot Holdings Pty Ltd v Creasy [1996] HCA 44; (1996) 185 CLR 149 at 171), particularly where fundamental rights or freedoms are involved: K-Generation Pty Ltd v Liquor Licensing Court [2009] HCA 4; (2009) 237 CLR 501 at 520 [47]; [2009] HCA 4; and R & R Fazzarolari Pty Ltd v Parramatta City Council [2009] HCA 12; (2009) 237 CLR 603 at 619 [43]; [2009] HCA 12. Kirby J thought that no Australian Parliament could legislate for "absolute discretions", because they are "a form of tyranny": Gerlach v Clifton Bricks Pty Ltd (2002) 209 CLR 478 at 503 [69]; [2002] HCA 22.

[31] Although Barwick CJ protected a taxi owner's licence by treating it as a species of property: Banks v Transport Regulation Board (Vic) [1968] HCA 23; (1968) 119 CLR 222 at 231.

[32] See, eg: R v Trebilco; Ex parte F S Falkiner & Sons Ltd [1936] HCA 63; (1936) 56 CLR 20 at 32; and Swan Hill Corporation v Bradbury [1937] HCA 15; (1937) 56 CLR 746 at 757-758.

[33] [1945] HCA 4; (1945) 69 CLR 613.

[34] See, for example, Miller, n 23 at 562-563.

[35] Murphy J was the only High Court judge to attack legal doctrine on overtly class lines. See: Attorney General (Cth); Ex rel McKinlay v Commonwealth [1975] HCA 53; (1975) 135 CLR 1 at 76; Heatley v Tasmanian Racing and Gaming Commission [1977] HCA 39; (1977) 137 CLR 487 at 496; Forbes v Trotting Club (NSW) [1979] HCA 27; (1979) 143 CLR 242; and Neal v R [1982] HCA 55; (1982) 149 CLR 305 at 316-317.

[36] See Eric Adams' contribution to this collection.

[37] Roncarelli, n 1 at 130.

[38] Roncarelli, n 1 at 130.

[39] Roncarelli, n 1 at 132.

[40] Roncarelli, n 1 at 132.

[41] This was the tract which accused the government of "savage persecution": Roncarelli, n 1 at 133.

[42] Roncarelli, n 1 at 132.

[43] Roncarelli, n 1 at 137.

[44] Roncarelli, n 1 at 141.

[45] Roncarelli, n 1 at 139-140.

[46] Roncarelli, n 1 at 141.

[47] Roncarelli, n 1 at 144.

[48] Roncarelli, n 1 at 140.

[49] Roncarelli, n 1 at 140.

[50] Roncarelli, n 1 at 142.

[51] Roncarelli, n 1 at 140.

[52] However, his Honour did state that the permit's transferrability (usually a hallmark of "property") was "most pertinent": Roncarelli, n 1 at 140. Cf Martland J at 156 and Cartwright J at 168.

[53] Roncarelli, n 1 at 142. They still rank very low in administrative law's pecking order: Dunsmuir v New Brunswick [2008] 1 SCR 190; 2008 SCC 9.

[54] See P Davies and M Freedland, "Labour Law", in J L Jowell and P P W B McAuslan (eds), Lord Denning: the Judge and the Law (London, Sweet & Maxwell, 1984) ch 8, 367-438.

[55] At least in the context of the Birmingham Six and the Guildford Four, Denning thought it better to maintain public confidence in the justice system than to acknowledge that innocent people were in prison because of police lies. He even suggested that it would have been better to hang them all. All the major media ran the stories; it suffices here to refer to articles in The Guardian (17 and 18 August 1990), The Independent (17 August 1990), and The Times (23 August 1990).

[56] Cartwright J in dissent thought that this last consideration was permissible; Roncarelli, n 1 at 164.

[57] Roncarelli, n 1 at 141, emphasis added.

[58] Roncarelli, n 1 at 143.

[59] See Roncarelli, n 1 at 141, where his Honour referred to the dissentients in McGillivray v Kimber (1915) 52 SCR 146.

[60] Roncarelli, n 1 at 141.

[61] Indeed, for Cartwright J, the characterisation game presented Mr Roncarelli with a Catch-22 dilemma. The defendant's discretion was "administrative" and therefore unfettered, or else it was quasi-judicial and therefore immune from a damages action in the absence of malice.

[62] [1898] AC 1 [Allen].

[63] Roncarelli, n 1 at 143.

[64] Roncarelli, n 1 at 144.

[65] See R Stevens, Torts and Rights (OUP, Oxford, 2007) [Stevens] pp 218 and 242-3, where it is argued that the misfeasance tort is not a regular tort at all, because it it protects no pre-existing rights or interests. Stevens regards it as a "regulatory" mechanism. The tort remains exceptional even if one were to see torts as creating (as well as vindicating) rights and correlative duties; see Peter Cane's review of Stevens' book in (2008) 71 Modern Law Review 641.

[66] David Mullan's contribution to this collection notes the cursory reference to compensation for lost "goodwill". The overall sum awarded came nowhere near matching what would have been needed if Mr Roncarelli were truly to be compensated for what Rand J had called "vocational outlawry", and there was no pretence at estimating a discount for future vicissitudes.

[67] Roncarelli, n 1 at 142.

[68] (1703) 1 Sm LC 253 at 273.

[69] See also C Harlow, "A Punitive Role for Tort Law?", in L Pearson, C Harlow and M Taggart (eds) Administrative Law in a Changing State: Essays in Honour of Mark Aronson (Hart Publishing, Oxford, 2008) 246.

[70] Mengel, n 15 at 345 and 348.

[71] His version of the rule of law required such disputes to be adjudicated in the "ordinary courts": A V Dicey, Introduction to the Study of the Law of the Constitution (MacMillan, London, 10th ed, 1959) p 193.

[72] For negligence, see M Aronson, "Government Liability in Negligence" [2008] MelbULawRw 2; (2008) 32 Melbourne University Law Review 44.

[73] Public officials are the usual defendants in actions for malicious prosecution, but it remains a private law tort.

[74] [1939] HCA 9; (1939) 62 CLR 339, [James] in which Dixon J said (at 362) that it would be "ridiculous" to analogise from the action for breach of statutory duty. The High Court continues to deny the availability of actions for breach of constitutional protections: Lange v Australian Broadcasting Corporation [1997] HCA 25; (1997) 189 CLR 520 at 563; Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1 at 124-126 and 146-148; British American Tobacco Australia Ltd v Western Australia [2003] HCA 47; (2003) 217 CLR 30 at 52 [40]; [2003] HCA 47; and Mulholland v Australian Electoral Commission [2004] HCA 41; (2004) 220 CLR 181 at 245 [180]; [2004] HCA 41. Kirby J accepted in British American (at 75-81 [118]-[137]) that current authority was against constitutional causes of action, but protested his disagreement as a matter of principle. The High Court acknowledges that its position is at odds with: The State (Quinn) v Ryan [1965] IR 70; Bivens v Six Unknown Named Agents of the Federal Bureau of Narcotics 403 US 388 (1971); and Simpson v Attorney General (NZ) (Baigent's Case) [1994] 3 NZLR 667.

[75] James, n 74.

[76] The rationales for the two conspiracy torts (namely, lawful but malicious acts, and intent to harm by unlawful means) are much-debated. Accepting for the moment that an agreement between two or more people rightly "tortifies" something that would not be actionable if done solo, one might question why proof of agreement is needed in the case of large organisations, such as corporations or bureaucracies.

[77] James, n 74 at 371-372.

[78] James, n 74 at 372.

[79] James, n 74 at 372-373.

[80] See Mengel, n 15 at 351. The intimidation tort is usually tripartite, in that the unlawful action (or threat of it) is against a third person, with the intent of causing harm to the plaintiff. Two-party intimidation presents fewer challenges of principle, and the House of Lords put it to one side in OBG Ltd v Allan [2007] UKHL 21; [2008] 1 AC 1 at 35 [61]; [2007] UKHL 21 [OBG].

[81] James, n 74 at 375.

[82] Mengel, n 15 at 351 and 371-373.

[83] OBG, n 80.

[84] One qualification was added, namely, that the means (eg, intimidation) might be unlawful even if the third person suffered no loss only because he or she yielded to the defendant's pressure: OBG, n 80 at 32 [49].

[85] OBG, n 80 at 32-33 [51].

[86] OBG, n 80 at 34 [56].

[87] OBG, n 80 at 34 [57].

[88] OBG, n 80 at 34-35 [56]-[60]. See also S Deakin and J Randall, "Rethinking the Economic Torts" (2009) 72 Modern Law Review 519 at 544-550 [Deakin and Randall] for criticism of OBG's requirement that the unlawful means be independently actionable. Aside from the criticism that it will be difficult to square OBG with earlier precedent that it failed to overrule, the authors said that the requirement narrowed the tort unduly, and lost sight of its function which was to set limits to direct interference with recognised trade or business interests.

[89] Sanders v Snell (1998) 196 CLR 329 at 344 [36]-[37]; [1998] HCA 64, [Sanders].

[90] It was obvious that if the Bank of England's individual officers had been misfeasance tortfeasors in Three Rivers District Council v Bank of England (No 3) [2003] 2 AC 1 [Three Rivers], then the Bank itself would also have been liable because none of the relevant individuals had been pursuing their own personal interests. The House of Lords may be more disposed than the Australian High Court to allowing vicarious liability for deliberate misconduct; see: Racz v Home Office [1994] 2 AC 45; and Three Rivers at 191.

[91] Mengel, n 15 at 347; and Sanders, n 89 at 345 [38]. J L R Davis discussed the issue in "Misfeasance in Public Office, Exemplary Damages and Vicarious Liability" to be published in 2010 in the Australian Institute of Administrative Law Forum. The High Court was unable to propound an authoritative test for determining the extent of vicarious liability for misbehaving employed school teachers in New South Wales v Lepore (2003) 212 CLR 511; [2003] HCA 4.

[92] Sanders, n 89 at 344 [37], per Gleeson CJ, Gaudron, Kirby and Hayne JJ.

[93] This is "targeted malice", the so-called first limb of the misfeasance tort, discussed below.

[94] See H Carty, An Analysis of the Economic Torts (Oxford University Press, Oxford, 2001) p 109, where it is also noted that the American doctrine has a limited application.

[95] See, eg, J D Heydon Economic Torts (Sweet & Maxwell, London, 2nd ed, 1978), [Heydon] p 28, who argued that the toleration in Allen n 62 of intentional harm caused by "intolerable conduct" had been a short-term expedient to protect trade unions. He argued that had Allen gone the other way, "[m]uch race relations legislation would have been less necessary". See also P Sales and D Stilitz, "Intentional Infliction of Harm by Unlawful Means" (1999) 115 Law Quarterly Review 411.

[96] OBG, n 80 at 22 [14].

[97] British Railway Traffic and Electric Co Ltd v CRC Co Ltd [1922] 2 KB 260 at 268 per McCardie J.

[98] Shapiro v La Morta (1923) 40 TLR 201 at 203, per Scrutton LJ. These quotations from McCardie J and Scrutton LJ appear in Heydon, n 95 p 83. See also P Cane, "Mens Rea in Tort Law" (2000) 20 Oxford Journal of Legal Studies 533 [Cane].

[99] See P Cane, The Anatomy of Tort Law (Hart Publishing, Oxford, 1997) p 35, where people's motives are their reasons for acting, and their intentions are the consequences that they hope to cause. The two need not coincide.

[100] This account draws upon only a handful of leading appellate decisions: Mengel, n 15; Garrett v Attorney General [1997] 2 NZLR 332, [Garrett]; Sanders, n 89; Three Rivers, n 90; Odhavji Estate v Woodhouse [2003] 3 SCR 263; 2003 SCC 69 [Odhavji]; Hobson v Attorney General [2006] NZCA 409; [2007] 1 NZLR 374; and New Zealand Defence Force v Berryman [2008] NZCA 392 [Berryman]. Hobson went to the Supreme Court, but not on misfeasance: Couch v Attorney General [2008] NZSC 45.

[101] Odhavji, n 100, which involved the duty of police officers to cooperate with an internal investigation. This issue was discussed inconclusively in Berryman, n 100, and went unnoticed in Garrett, n 100.

[102] Being an intentional tort, defendants are liable only for the harm which they either intend or deliberately choose to ignore. Odhavji, n 100 at 285 [29] put this differently, in terms of requiring a nexus between the parties such that the defendant owes the plaintiff a duty. The plurality in Mengel, n 15 at 346-347 did not resolve whether the misfeasance defendant needs to have breached a duty owed to the plaintiff. Brennan J opposed the idea: at 357.

[103] Three Rivers, n 90 at 230.

[104] Three Rivers, n 90 at 235 emphasis added. See also Watkins v Home Office [2006] 2 AC 395 at 404 [12] and 423 [73]; [2006] UKHL 17.

[105] Three Rivers, n 90 at 235. The same view appears in Garrett, n 100 at 350, where a policeman had deliberately broken the rules, but quite possibly for what he had mistakenly seen as the plaintiff's benefit.

[106] Sanders v Snell (No 2) (2003) 130 FCR 149; [2003] FCAFC 150. This was the end of the line in an extremely protracted litigation saga.

[107] Brennan J foresaw this possibility in Mengel, n 15 at 356.

[108] Odhavji, n 100 at 284-285 [28].

[109] Mengel, n 15 at 357, Brennan J.

[110] Cane, n 98 at 535.

[111] Few officials would publicly boast as Mr Duplessis had done. Indeed, it was evidently difficult to determine whether his boasts had claimed too much credit for himself, and too little for the Liquor Commissioner.

[112] There is a parallel with judicial review challenges for bias. Australian doctrine allows but actively discourages challenges for actual bias, diverting most challengers to the less pejorative ground of "reasonable apprehension" of bias: M Aronson, B Dyer and M Groves, Judicial Review of Administrative Action (4th ed, 2009, Thomson Reuters, Sydney) pp 640-646.

[113] See Briginshaw v Briginshaw [1938] HCA 34; (1938) 60 CLR 336. Deakin and Randall, n 88 at 538-540, were critical of the failure of the House of Lords in OBG n 80 to use a "presumption" that people intend the probable consequences of their acts. The misfeasance torts, however, have never wavered in their insistence on strict proof.

[114] Three Rivers, n 90 at 190.

[115] Odhavji, n 100 at 283 [26].

[116] Mengel, n 15 at 352-353.

[117] In Northern Territory v Mengel (1994) 95 NTR 8 at 12-15 [12]-[24]; [1994] NTSC 37, Angel J.

[118] The court said in Precision Products (NSW) Pty Ltd v Hawkesbury City Council [2008] NSWCA 278 that a duty to take care to act intra vires would have required the courts to set public sector management standards, which would be both difficult and a violation of the separation of powers.

[119] See David Mullan's contribution to this collection.

[120] M Walters, "Dicey and Rand and the Rule of Law", this collection.