SC 49/2006

IpsofactoJ.com: International Cases Part 3 Case 12 [NZSC]


SUPREME COURT OF NEW ZEALAND

Coram

Susan Couch

- vs -

The Attorney General

ELIAS CJ

BLANCHARD J

TIPPING J

McGRATH J

WILSON J

 

24 MARCH 2010


Judgment

Elias CJ

  1. In 2002 the Privy Council in Bottrill v A [2002] UKPC 44, [2003] 2 NZLR 721 (at [63] per Lord Nicholls, Lord Hope and Lord Rodger; Lord Hutton and Lord Millett dissenting):

    [U]nder the common law of New Zealand the Court‘s jurisdiction to award exemplary damages in cases of negligence is not rigidly confined to cases where the defendant intended to cause the harm or was consciously reckless as to the risks involved.

    In the present appeal the Supreme Court overrules that decision and reinstates the view of the Court of Appeal in Bottrill v A [2001] 3 NZLR 622 (CA) (per Richardson P, Gault, Blanchard and Tipping JJ; Thomas J dissenting). I dissent from the conclusion. It re-introduces a "cause of action" condition for exemplary damages despite earlier rejection in New Zealand, Australia, and Canada of similar attempts in the United Kingdom at such restrictions as unprincipled and arbitrary. It requires construction of a "species of negligence[1] in which intention or conscious recklessness is an element, in order to exclude a remedy of otherwise general application once liability in tort is established. The restriction is justified on the basis that exemplary damages are "anomalous". Such assessment rests in part on the erroneous but persistent view that making an example of the defendant is not a proper function of the law of torts. That view has been accurately characterised as question-begging.[2] It is unhistorical[3] and was rejected in New Zealand by all members of the Court of Appeal in Taylor v Beere [1982] 1 NZLR 81 (CA) (at 85 per Cooke J; at 90 per Richardson J; and at 95 per Somers J).[4] There, by rejecting confinement of exemplary damages to categories, the Court of Appeal affirmed that they are awarded under a principle of general application.

  2. All members of this Court are in agreement that exemplary damages are available in negligence and are not confined to the intentional torts, rejecting the high ground on which the Attorney-General sought to justify strike out of the proceedings. The majority imposes however the precondition of subjective recklessness proposed by the majority in the Court of Appeal in Bottrill but rejected by the Privy Council. This "species of negligence" arises where the defendant consciously appreciates the risk of causing harm and deliberately runs that risk: See Tipping J at [100]–[101]. Such subjective recklessness was described by Tipping J in his concurring opinion (at [173]) in the Court of Appeal Bottrill as achieving "a policy synthesis with the intentional tort approach, at least for personal injury cases.[5]

  3. The arguments for and against the position now adopted in this Court were rehearsed at length in the judgments in the Court of Appeal and the Privy Council in Bottrill. I am in agreement with the reasons given by Thomas J, dissenting in the Court of Appeal, and by Lord Nicholls, delivering the majority judgment of the Privy Council. It would be superfluous to repeat them. I concentrate, rather, on the reasons why I am unable to agree with the revision undertaken by the other members of this Court, both in terms of the substance of the argument and in terms of the occasion for departing from a recent decision of high authority.

  4. In summary, I would decline to impose as a matter of law a precondition for the award of exemplary damages that the defendant must consciously run the risk of harm to the plaintiff. Such precondition restricts the general exemplary jurisdiction to mark society‘s condemnation of outrageous behaviour by the defendant which is insufficiently addressed by other remedy, and is contrary to the general application of the exemplary principle recognised in Taylor v Beere. It treats the occasion for exemplary damages in negligence as depending on conscious appreciation of the harm likely to be suffered by the plaintiff rather than as arising more broadly out of the conduct of the tortfeasor and despite foreseeability of harm not being an element of the cause of action in negligence. It saps the vitality of the exemplary principle in meeting the needs of modern New Zealand society, and turns on the creation of a subcategory of the tort of negligence on no sound basis. These points are addressed in what follows under headings which reflect this summary. I deal first however with suggestions that Bottrill represents a deviation and is out of step with other Commonwealth jurisdictions.

  5. In addition, I consider there is no sufficient basis on which to depart from New Zealand decisions of recent and high authority in Taylor v Beere and Bottrill. No occasion to reassess the decision of the Privy Council in Bottrill arises on the basis of any misunderstanding about the operation of New Zealand‘s accident compensation system[6] because the statutory scheme operates outside the exemplary principle, as was made clear in  Donselaar v Donselaar [1982] 1 NZLR 97 (CA).[7] Since the requirement of subjective recklessness imposed by other members of this Court is not confined to cases of personal injury, reliance upon the New Zealand accident compensation system as justifying the restriction (see Blanchard J at [67]; Tipping J at [96]; McGrath J at [240]–[241]; and Wilson J at [252]) is perhaps surprising. "Floodgates" concerns (described by Blanchard J at [66] and Tipping J at [135]) are not substantiated and seem inconsistent with legislative endorsement of the exemplary principle.[8]

  6. Moreover, the preliminary hearing in the present case was not appropriate for reconsideration of the approach in Bottrill.[9] This case concerns a more complex setting of statutory responsibilities and vicarious liability, against a background of incomplete pleadings, and without factual context. The dangers are illustrated by statements which bear on the question of vicarious liability,[10] on which we heard no argument and which (because of the admission of vicarious liability) is not in issue. I develop the view that the appeal is not suitable for reconsideration of Bottrill notwithstanding that the point is effectively overtaken by the judgment of the majority on the substantive point, because of the importance of consideration of the circumstances in which it is proper for this Court to decline to follow a decision of its own or of the Privy Council on appeal from New Zealand. The claim for exemplary damages should in my view be allowed to proceed. Any questions of law that remain in contention after trial will then be considered after the facts have been found. Since the plaintiff has indicated that she will plead conscious recklessness, the course here adopted by the Court will not save the parties the expense of trial.

  7. I write separately only on the Bottrill point. I agree with the reasons given by Tipping J for holding that s 317 of the Accident Compensation Act 2001[11] is no bar to a claim for exemplary damages for personal injury. And I agree with the reasons he gives for holding that no immunity for the Crown can be derived from s 86 of the State Sector Act 1988, a most unlikely source for such sweeping immunity which would clash not only with the purpose of s 6(1) of the Crown Proceedings Act 1950 but also with s 27(3) of the New Zealand Bill of Rights Act 1990. I agree with Tipping J that the immunity s 86 confers is in respect of internal responsibilities within the Executive, reorganised under the State Sector Act, and does not provide an immunity for government employees from primary tortious liability.

    Background

  8. Susan Couch was the victim of a serious attack by a parolee under the supervision of the Probation Service. She brings a claim in the High Court against the Attorney-General seeking exemplary damages for failure by the Service to exercise reasonable care in the parolee‘s supervision. The proceedings were struck out in the Court of Appeal on the grounds that the claim in negligence could not succeed because no duty of care was owed by the Probation Service to Ms Couch: Hobson v Attorney-General [2007] 1 NZLR 374 (CA). That result was reversed by this Court in its judgment of 13 June 2008: Couch v Attorney-General (on appeal from Hobson v Attorney-General) [2008] NZSC 45, [2008] 3 NZLR 725. We held that a duty of care cannot confidently be excluded as a matter of law on a preliminary basis and will turn on the facts found at trial. An alternative ground for strike-out, that exemplary damages are not available for a claim in negligence for personal injury in New Zealand (a ground that the Court of Appeal did not have to consider because of the view it took that there was no duty of care), was not reached at the earlier hearing in this Court. The appeal was accordingly adjourned for further argument, and the claim was not formally reinstated pending determination of the alternative ground.

  9. The parties were invited by the judgment of 13 June 2008 to reconsider whether they wished to pursue the additional basis for strike-out ahead of trial, in the light of the discussion in the judgment of the principles upon which strike-out is available. At a directions hearing to consider whether the availability of exemplary damage was suitable for determination as a point of law before trial, counsel for the Attorney-General indicated that the arguments at the resumed hearing would be that s 317 of the Accident Compensation Act 2001 is a bar to the claim and that, in addition, exemplary damages are not available for the tort of negligence. The first argument entailed reconsideration of the decision of the Court of Appeal in Donselaar (which counsel for the Attorney-General suggested was to be distinguished because the cause of action there was in trespass). The second was contrary to the assumption upon which Bottrill was argued in the Court of Appeal and Privy Council. (The disagreement in that case turned on the further question whether exemplary damages for the tort of negligence could be awarded only in cases of intention or advertent recklessness.) As had been foreshadowed at the first hearing before us,[12] counsel advised that, if unsuccessful in these strike-out points, the Attorney-General would take the position at trial that exemplary damages would not be appropriate if the Probation Service was unaware of the risk of harm to Ms Couch, thereby inviting the court at trial to exclude on the facts the possibility left open as a matter of law by the Privy Council in Bottrill. That would not have precluded a challenge to Bottrill post trial to this Court should exemplary damages have been awarded on findings of fact that the Probation Service had not been conscious of the risk to Ms Couch.

  10. It was accepted at the directions hearing in this Court that the points concerning the effect of s 317 of the Accident Compensation Act 2001 and the availability of exemplary damages for the tort of negligence were points of law suitable for determination ahead of the facts. The first is a question of statutory interpretation. The second takes the high ground that exemplary damages are not available for this tort, a conclusion that depends on acceptance of "hardening of the categories.[13] If the Attorney-General were successful in either of these arguments, the claim would properly be struck out as disclosing no cause of action. The two points were accordingly set down for argument at a resumed hearing.

  11. At the hearing, the argument that exemplary damages are not available for negligence at all was not greatly developed by counsel for the Attorney-General. Instead, counsel argued that the Privy Council decision in Bottrill was wrong. With hindsight, I consider it would have been preferable to have adhered to the approach envisaged at the directions hearing and to have declined to hear counsel on the Bottrill argument. For reasons developed further below, it is my view the reconsideration of Bottrill was not suitable for determination before trial. I remain of that view notwithstanding the arguments addressed to us. Bottrill is consistent with previous New Zealand law and not shown to be inconsistent with Australian and Canadian law

  12. It is necessary to say immediately that I am unable to agree with suggestions that the present decision restores New Zealand law to a path from which the Privy Council decision in Bottrill caused it to deviate and brings New Zealand law "back into line with that of Australia and Canada (as Tipping J suggests at [171]).

  13. First, the application of an established principle of general application to new circumstances is not deviation. That would be to treat application of exemplary principle as restricted to the categories of case in which it has formerly been applied, the position rejected by Taylor v Beere in 1982 in New Zealand. Indeed, on any view, Bottrill in both the Court of Appeal and the Privy Council was novel application of the principle because the availability of exemplary damages for negligence had not previously been upheld by an appellate court in New Zealand. It was however unexceptional because the application of general principle to the circumstances of the particular case is the method of the common law. What was exceptional was the limitation of the general principle by the Court of Appeal‘s requirement that it applies in negligence only where harm results from conscious assumption of risk. The decision on the present appeal is similarly bold because it overthrows Taylor v Beere‘s rejection of the reasoning on exemplary damages in Rookes v Barnard [1964] AC 1129 (HL) and Broome v Cassell & Co Ltd [1972] AC 1027 (HL). In Rookes v Barnard the House of Lords came close to abolishing exemplary damages altogether, as Lord Reid made clear in the subsequent case of Broome v Cassell (at 1087) (in which the House of Lords crushed rebellion by a strong Court of Appeal (Broome v Cassell [1971] 2 QB 354 (CA) per Lord Denning MR, Salmon and Phillimore LJJ) to maintain the restrictions on exemplary damages introduced by the categories described in Rookes v Barnard). Since then, however, belief in the utility of the exemplary principle has been remarked upon judicially in New Zealand (Taylor v Beere at 91 per Richardson J), the United Kingdom (Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29, [2002] 2 AC 122 at [63] per Lord Nicholls), and Canada (Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 at [37] per McLachlin CJ and L‘Heureux-Dubé, Gonthier, Major, Binnie, and Arbour JJ; LeBel J dissenting). In Canada, the United Kingdom and Ireland law reform projects have confirmed the usefulness of exemplary damages.[14] In Canada, exemplary damages are available to respond to outrageous conduct by the defendant in cases of contract and for breach of fiduciary obligations as well as across all torts: Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 at [67] per McLachlin CJ and L‘Heureux-Dubé, Gonthier, Major, Binnie, and Arbour JJ; LeBel J dissenting. In New Zealand, legislation has recognised such damages (Accident Compensation Act 2001, s 319). It is therefore surprising to see in the present case arguments for restriction of the exemplary principle developed from judicial reasoning in Rookes v Barnard and Broome v Cassell which start from the view that such damages are anomalous and to be strictly confined by judicially created categories: See Tipping J at [136]–[143].

  14. Secondly, I do not think it can be said that the position preferred in this case by the majority brings New Zealand law into line with the common law of Canada or Australia. The final appellate courts in those jurisdictions have not yet been called upon to decide whether conscious appreciation of the risk to the plaintiff is a precondition of an award of exemplary damages in negligence. But their refusal to accept the restricted categories suggested in Rookes v Barnard suggests that further categorisation is likely to be subject to searching scrutiny: See Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118 and Vorvis v Insurance Corporation of British Columbia [1989] 1 SCR 1085.

  15. In Australia there are judgments which refer with apparent approval to the Privy Council approach. The reasons of Kirby J in the High Court of Australia in Gray v Motor Accident Commission [1998] HCA 70, (1998) 196 CLR 1 (with which the majority did not express disagreement) and the judgment of Spigelman CJ in the Court of Appeal of New South Wales in State of New South Wales v Ibbett [2005] NSWCA 445, (2005) 65 NSWLR 168 at [40]–[44] (which was not on this point directly addressed by the High Court) indicate the view that awards of exemplary damages should not be confined to cases where the defendant intended to cause harm or was consciously reckless as to the risks involved. Kirby J in Gray (at [86]) considered that exemplary damages were available "whatever the subjective intention of the tortfeasor if, objectively, the conduct involved was high-handed, calling for curial disapprobation addressed not only to the tortfeasor but to the world". In Gray (where the question whether conscious recklessness is a condition for exemplary damages in negligence did not have to be resolved) the majority of the High Court considered (at [14]) that "'conscious wrongdoing in contumelious disregard of another‘s rights‘ describes at least the greater part of the relevant field, a position consistent with the approach of the Privy Council in Bottrill. In the High Court in New South Wales v Ibbett [2006] HCA 57, (2006) 229 CLR 638, it is the case that in a footnote (Footnote 47) the Court indicated that it considered the views of the Privy Council in Bottrill and the Supreme Court of Canada in Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 were to be contrasted with the statement in Gray that "there can be cases, framed in negligence, in which the defendant can be shown to have acted consciously in contumelious disregard of the rights of the plaintiff or persons in the position of the plaintiff (at [22]). And it may be that the position in Australia in relation to negligence will yet move to requiring subjective recklessness in the sense of conscious assumption of risk of harm to the plaintiff. The Court of Appeal in New South Wales in State of New South Wales v Delly [2007] NSWCA 303, (2007) 70 NSWLR 125 at [91] per Tobias JA and at [115] per Basten JA,[15] a decision decided after Ibbett, was however prepared to allow that subjective advertence to harm is not always necessary for an award of exemplary damages. But neither Ibbett nor Delly were cases where the cause of action was in negligence.[16] Further cases in Australia will no doubt require consideration of the general approach indicated by the High Court in Lamb v Cotogno (1987) 164 CLR 1 (at [21]) that exemplary damages express the Court‘s condemnation of objectively reckless behaviour[17]. Lamb v Cotogno suggests that callousness provides sufficient justification for exemplary damages. As things stand, it is drawing a rather long bow to maintain that the present decision brings New Zealand law into line with Australian law.

  16. In Canada, the principal recent Supreme Court cases concerning exemplary damages have not arisen in negligence but in employment contracts. In Whiten, however, Binnie J (for himself and McLachlin CJ, L‘Heureux-Dubé, Gonthier, Major and Arbour JJ) reviewed the purpose of exemplary damages more generally. Punitive damages were available to exact "retribution, deterrence and denunciation" (at [43]) for "misconduct that represents a marked departure from ordinary standards of decent behaviour". (at [36]) They provided a "socially useful service" because only the plaintiff can be expected to invest the legal costs to establish that the defendant "behaved abominably" (at [37]). After reviewing the comparative position in England, Australia, New Zealand, Ireland and the United States, Binnie J concluded "the attempt to limit punitive damages by 'categories‘ does not work and was rightly rejected in Canada in [Vorvis v Insurance Corporation of British Columbia [1989] 1 SCR 1085 at 1104-1106] ..." (at [67] (citations omitted)):

    The control mechanism lies not in restricting the category of case but in rationally determining circumstances that warrant the addition of punishment to compensation in a civil action. It is in the nature of the remedy that punitive damages will largely be restricted to intentional torts, as in Hill, or breach of fiduciary duty as in M (K) v M(H), but Vorvis itself affirmed the availability of punitive damages in the exceptional case in contract. In Denison v Fawcett, the Ontario Court of Appeal asserted in obiter that on proper facts punitive damages would be available in negligence and nuisance as well. In Robitaille v Vancouver Hockey Club Ltd, the British Columbia Court of Appeal awarded punitive damages in a negligence case on the principle that they ought to be available whenever "the conduct of the defendant is such as to merit condemnation by the Court". This broader approach seems to be in line with most common law jurisdictions apart from England.

  17. The apparent approval with which Robitaille v Vancouver Hockey Club Ltd (1981) 124 DLR (3d) 228 (BCCA) is cited by the Supreme Court of Canada does not suggest insistence on conscious assumption of risk as a precondition for exemplary damages for breach of a duty of care under an employment contract. In that case the defendant hockey club believed that the player had psychological, not physical problems and so was not subjectively aware of the risk it exposed him to in requiring him to play. It was liable for exemplary damages when he was permanently disabled. In Whiten the Supreme Court (at [67]) appears to adopt the "broader approach" that exemplary damages are available whenever the conduct of the defendant merits condemnation, while acknowledging that "it is in the nature of the remedy" that intentional wrongdoing will usually be required. Later references in Whiten to the intent and motive of the defendant (at [113]), referred to by Tipping J as "intentional elements" at [167], do not detract from the general approach and are part of a list of factors relevant to assessing the level of damages. They do not suggest that objective recklessness, if sufficiently outrageous, is not sufficient basis for an award of exemplary damages. Still less do they suggest a requirement that known risk be consciously run.

  18. Whiten was applied in Honda Canada Inc v Keays 2008 SCC 39, [2008] 2 SCR 362 where the distinction between compensatory and punitive damages was again stressed. The Supreme Court indicated (at [62]) in the context of a contractual claim for wrongful dismissal that exemplary damages "are restricted to advertent wrongful acts that are so malicious and outrageous that they are deserving of punishment on their own". The requirement of "advertence" in relation to wrongful acts is not a requirement of conscious appreciation of the risk of harm, as is illustrated by the use of the same word by the Ontario Law Reform Commission (at 38 and 69) in connection with objective recklessness. In McIntyre v Grigg the Ontario Court of Appeal (citing with approval and emphasis a passage from Vlchek v Koshel (1988) 52 DLR (4th) 371 (BCSC) (at 375 per Gallaghan J)) rejected the suggestion that intention to cause injury was required before exemplary damages would be appropriate in negligence: McIntyre v Grigg (2006) 274 DLR (4th) 28 (ONCA) at [69]–[70]. In describing the Robitaille case the Ontario Court of Appeal noted it had held that the advertence necessary for an award of exemplary damages included malice and "recklessness that indicates an indifference to the safety of others" (at [65]). But, "alternatively", it allowed that the defendant may have "engaged in conduct that is so socially reprehensible that it justifies the award of punitive damages", Ibid. Such cases illustrate Thomas J‘s point about the "folly of clutching at the use of the word 'advertent‘ and seeking to convert it into a requirement that the wrongdoer be subjectively aware of the risk he or she is creating": Bottrill (CA) at [105]. Again, I do not think that the approach adopted by the Canadian Supreme Court provides support for a requirement of conscious assumption of risk before exemplary damages can be awarded for liability in negligence.

    The “exemplary principle” is one of general application

  19. Exemplary damages are a general remedy available irrespective of the grounds of liability in tort (which, depending on the elements of the particular tort, may or may not entail intention) wherever compensation to the plaintiff is inadequate to respond to the outrageousness of the defendant‘s conduct. The general application of the "exemplary principle" (as Richardson J described the underlying rationale - Taylor v Beere at 8863) follows from and was affirmed in the rejection of closed categories of case in which such damages are available.[18] Exemplary damages are distinct from compensatory and aggravated damages, although the distinction was not always observed before Rookes v Barnard imposed order in that respect. They are available only when the amount required to make good the plaintiff‘s loss and any additional affront to him (properly reflected in aggravated damages) is unequal to mark the affront to community values occasioned by the defendant‘s conduct. Compensatory damages and aggravated damages are assessed on what is due to the plaintiff, but exemplary damages mark the difference between that amount and the amount the defendant ought to pay,[19] in what Lord Devlin described as "vindicating the strength of the law": Rookes v Barnard at 1226. The policies which underlie compensatory and exemplary damages overlap however because both reflect the general ends of tort law. Those ends include punishment and deterrence, as has long been accepted in New Zealand law.

  20. In this connection, Cooke P and Richardson J in separate cases both had occasion (Re Chase [1989] 1 NZLR 325 (CA) at 333 per Cooke P; Taylor v Beere at 90 per Richardson J) to cite with approval the views of Lord Wilberforce in Broome v Cassell (at 1114):

    It cannot lightly be taken for granted, even as a matter of theory, that the purpose of the law of tort is compensation, still less that it ought to be, an issue of large social import, or that there is something inappropriate or illogical or anomalous (a question-begging word) in including a punitive element in civil damages, or, conversely, that the criminal law, rather than the civil law, is in these cases the better instrument for conveying social disapproval, or for redressing a wrong to the social fabric, or that damages in any case can be broken down into the two separate elements. As a matter of practice English law has not committed itself to any of these theories: it may have been wiser than it knew.

    So, Richardson J in Taylor v Beere took the view that tort law "serves various social purposes": "it is not simply a compensation device or a loss distribution mechanism" (at 90). It was not, he thought, inappropriate for a jury or judge "to register condemnation of outrageous conduct on the part of a defendant in tort proceedings beyond what is properly allowed for in the award of compensatory damages (including aggravated damages for the harm and insult to the plaintiff)", ibid.

  21. In Taylor v Beere in 1982 the Court of Appeal declined to follow the views expressed in Rookes v Barnard that exemplary damages are "anomalous" and their availability restricted to specified categories of case.[20] The hostility towards exemplary damages implicit in the reasons of Lord Devlin in Rookes v Barnard and demonstrated in the reasons of Lord Reid in Broome v Cassell has not been part of New Zealand law. In rejecting the narrow categories proposed in Rookes v Barnard to confine exemplary damages, Richardson J in Taylor v Beere (at 92) expressed the view that "any other categorisation which attempts to limit the generality of the application of the exemplary principle is likely to be susceptible to similar criticisms". The Court of Appeal in Taylor v Beere preferred the broader approach adopted by the High Court of Australia in Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118 and affirmed for that jurisdiction on appeal by the Privy Council: Australian Consolidated Press Ltd v Uren [1969] 1 AC 590 (PC). On this view, and as the historical review of the origin of exemplary damages in Uren explained, exemplary damages are not properly treated as "anomalous" (at 136–139 per Taylor J and at 152–153 per Windeyer J). In Canada (Vorvis v Insurance Corporation of British Columbia [1989] 1 SCR 1085), similarly, the Supreme Court declined to limit the availability of exemplary damages to the categories described by Lord Devlin in Rookes v Barnard. Nor did the Court of Appeal in New Zealand in Taylor v Beere accept suggestions made in Broome v Cassell (at 1076 per Lord Hailsham and at 1130–1131 per Lord Diplock) that the availability of exemplary damages should be confined to those torts in which they had been awarded prior to 1964, when Rookes v Barnard was decided. More recently in the United Kingdom, the House of Lords in Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29, [2002] 2 AC 122 has moved closer to the view taken in Canada, Australia and New Zealand (without yet striking off the chains of Rookes v Barnard completely) in holding that exemplary damages are not confined to those torts in which they had been awarded before 1964.[21] The judgments in Kuddus (at [7]–[8] per Lord Slynn; at [44] per Lord Mackay; and at [68] per Lord Nicholls) affirm that whether exemplary damages are available depends on the behaviour of the defendant, rather than on the cause of action.

  22. In New Zealand the "exemplary principle" is not confined to intentional torts. It prompts no "synthesis" (Bottrill (CA) at [173] per Tipping J) of the intentional torts and negligence because it is available on establishment of liability (whatever the elements of the cause of action) where the outrageous conduct of the defendant is insufficiently condemned by compensatory and aggravated damages. Because the ends of punishment and deterrence will usually be sufficiently achieved through compensatory and aggravated damages awards, exemplary damages have been rare in New Zealand. But their restriction in negligence to cases of subjective recklessness would "evade the underlying principle", as pointed out by Clement JA in the Appellate Division of the Alberta Supreme Court in respect of the restrictions attempted in Paragon Properties Ltd v Magna Envestments Ltd (1972) 24 DLR (3d) 156 (ABCA) at 167:

    The case recognizes the principle of exemplary damages, but in restricting its application it, in my opinion, does injustice to the principle. The basis of such an award is actionable injury to the plaintiff done in such a manner that it offends the ordinary standards of morality or decent conduct in the community in such marked degree that censure by way of damages is, in the opinion of the Court, warranted .... It is the reprehensible conduct of the wrongdoer which attracts the principle, not the legal category of the wrong out of which compensatory damages arise and in relation to which the conduct occurred. To place arbitrary limitations upon its application is to evade the underlying principle and replace it with an uncertain and debatable jurisdiction.

    These remarks were cited with approval by Richardson J in Taylor v Beere (at 91–92).

    The outrageous conduct of the defendant need not be in relation to conscious risk-taking

  23. It is not accurate to characterise the difference between the Privy Council and the Court of Appeal in Bottrill as turning on "whether the negligence was outrageous": See Tipping J at [91]. The jurisdiction to award exemplary damages arises where the conduct of the defendant is "so outrageous as to call for condemnation and punishment" (Bottrill (PC) at [4]). The conduct may however be in the "manner or circumstances" in which the tort was committed and which make it "particularly appalling" (at [23] per Lord Nicholls). The distinction is illustrated by Lamb v Cotogno where the High Court of Australia (at [21]) accepted that callous conduct following commission of the tort and not itself comprising an actionable wrong could occasion an award of exemplary damages if outrageous.[22] In Rookes v Barnard (at 1228) Lord Devlin had taken the view that everything which aggravates or mitigates the defendant‘s conduct is relevant to a claim for exemplary damages. So in defamation cases it is established that the relevant conduct continues until judgment: Praed v Graham (1889) 24 QBD 53 (CA); Broome v Cassell at 1071–1072 per Lord Hailsham. In Vorvis, there was disagreement between the members of the Supreme Court of Canada about whether exemplary damages for breach of conduct could be awarded if the conduct relied on did not constitute in itself an actionable wrong. The majority (at 1106–1107 per Beetz, McIntyre and Lamer JJ) suggested that exemplary damages might, unusually, be awarded in cases of breach of contract, although in such cases the misconduct would also amount to an actionable wrong. Wilson J (at 1129–1130), joined by L‘Heureux-Dubé J, took a broader approach, considering rather that the task of the Court was simply "to assess the conduct in the context of all the circumstances and determine whether it is deserving of punishment because of its shockingly harsh, vindictive, reprehensible or malicious nature":

    In his dissent in the Court of Appeal in the present case Anderson JA applied the principles set out by Linden J in Brown v Waterloo Regional Board of Commissioners of Police and concluded that the conduct of the defendant both before and after the wrongful dismissal should be considered. I agree. This broader approach is required if the court‘s purpose is to punish highhanded, vindictive or otherwise shocking and reprehensible conduct by the defendant.

    ....

    Undoubtedly some conduct found to be deserving of punishment will constitute an actionable wrong but other conduct might not.

    Wilson J then adopted the statement of Clement JA in Paragon Properties Ltd v Magna Envestments Ltd (1972) 24 DLR (3d) 156 (ABCA) set out above at [22]. In the subsequent case of Whiten the Supreme Court of Canada maintained the position that exemplary damages must depend on an actionable wrong in addition to breach of contract, but found it in that case in breach of the additional obligation of good faith to which the insurer was subject.[23]

  24. The approach to contract is not directly in point here. The broad approach adopted in Whiten seems to me however to be consistent with the views of Wilson J in Vorvis. In any event, I am in agreement with the reasons of Wilson J on the point and consider they are supported by the decision of the High Court of Australia in Lamb v Cotogno. They also accord with the approach preferred in the New Zealand cases of Taylor v Beere, Donselaar, and Re Chase [1989] 1 NZLR 325 (CA) and are consistent with the terms in which Lord Nicholls described the basis of the jurisdiction in Bottrill. I too would not exclude the application of the exemplary principle in cases where the reprehensible conduct of the defendant does not itself constitute the actionable wrong. In particular, the misconduct of the defendant need not in itself entail advertent assumption of risk in respect of the harm suffered by the plaintiff,[24] for the convincing reasons given by Thomas J in the Court of Appeal in Bottrill (at [145]). The Privy Council in Bottrill accepted that the standard of outrageousness was not likely to be reached by conduct which is "grossly negligent" (at [33] per Lord Nicholls) without more. Indeed, it is hard to conceive of a case in which advertent misconduct will not feature in a case appropriate for exemplary damages. Such advertence need not, however, in itself constitute the actionable wrong. In the case of systemic failure (such as is alleged here), for example, it may be sufficiently outrageous if there is consciousness of the inadequacy of the systems in place or deliberate indifference to responsibility, even if the harm that eventuates is not foreseen.[25]

  25. In Bottrill the wrongdoing relied on comprised the breach of the duty of care. It was claimed that the want of care was of such a degree as to be outrageous, warranting exemplary damages. Because the feature which was said to warrant exemplary damages was an element in the cause of action (breach of the duty of care), the Privy Council was not called upon to consider whether conduct additional to the elements of the tort but accompanying its commission or following it, could occasion an award of exemplary damages. Nor is there explicit discussion in the majority judgments in the Court of Appeal of the point. The focus is understandable, because the basis for the claim of exemplary damages was in that case put solely on gross negligence. Thomas J however pointed out the difference (Bottrill (CA) at [109]):

    A wrongdoer may deliberately act in such a way as to disregard the plaintiff‘s rights without intending to do so or being consciously aware of the risk he or she is creating. Professor Williams has pointed out that conduct which may consist of inadvertent negligence in one respect may be combined with deliberate conduct in another respect.

    See Glanville Williams and BA Hepple Foundations of the Law of Tort (London, 1976) at 120.

  26. It is often said that the conduct of the defendant justifying the award of exemplary damages must be "contumelious".[26] The adjective conveys the flavour that the conduct may be consciously high-handed or arrogant behaviour. As indicated, I consider it may be conduct accompanying or following the commission of the tort. It is not necessary that it be conduct in respect of an element of the actionable wrong, such as a requirement of conscious breach of a duty of care in the case of negligence. The test adopted by the majority in this case would however so limit it.

  27. In Rookes v Barnard (at 1226) the first category identified by Lord Devlin as supported by authority for the award of exemplary damages where conduct was sufficiently outrageous to justify exemplary damages was high-handed infringement of rights by servants of the government. The availability of exemplary damages to deter and punish "the arbitrary and outrageous use of executive power" (at 1223 per Lord Devlin) was a means of "vindicating the strength of the law" (at 1226 per Lord Devlin). Lord Wilberforce in Broome v Cassell expressed the view that in the circumstances of modern society citizens required as much protection as formerly against "[e]xcessive and insolent use of power" (at 1120). In Kuddus Lord Nicholls acknowledged the role played by exemplary damages in the protection of civil liberties (at [63]). In such cases the availability of the salutary remedy of exemplary damages may serve the public interest when the actions which constitute the legal wrong (and are the source of liability) are accompanied by behaviour which is scornful, insulting, callous, or insolent, if such behaviour is sufficiently outrageous. "Contumelious" action on behalf of the executive may often entail conscious interference with rights. But in modern society, in which the public may reasonably rely on the discharge of responsibilities by those entrusted with public power and resources, it may be that inaction which amounts to shrugging off responsibility in some circumstances is as accurately described as high-handed or arbitrary even if the harm that eventuates is not consciously foreseen. That possibility should not be foreclosed.

    Restriction of the exemplary principle is inconsistent with its continued vitality

  28. The vitality of the exemplary principle was remarked upon by Lord Wilberforce in Broome v Cassell (at 1114), by Cooke J in Donselaar (at 106–107) and by Lord Nicholls in Kuddus (at [63]) and Bottrill (at [41]). Lord Hutton in Kuddus (at [75]–[79]) illustrated the social utility of the exemplary principle by reference to experience in Northern Ireland.[27] The discussion in Whiten already referred to in [17] indicates acceptance of the "socially useful service" provided by exemplary damages in circumstances where only the plaintiff has the incentive to achieve formal correction of outrageous behaviour. Similarly, in New Zealand Richardson J in Taylor v Beere considered "the utility of the exemplary principle" and "the felt need for this kind of civil remedy" to mark "heinous conduct in the course of tortious activity", as he thought was demonstrated by decisions in a number of common law jurisdictions (at 91). The hostility of Lord Devlin in Rookes v Barnard and Lord Reid in Broome v Cassell (echoed more recently by Lord Scott in Kuddus (at [110]–[111])) seems to have been overtaken by the insight that exemplary damages serve the needs of modern society. That may be why law reform projects in Ontario, England and Ireland have recommended retention of the exemplary principle (in Ontario, on the basis of objective assessment of outrageousness).[28]

  29. In New Zealand the vitality of exemplary damages was affirmed in Donselaar,[29] notwithstanding statutory exclusion of compensatory damages for personal injury. Cooke J expressed the view (at 107) that the non-compensatory functions of tort law, formerly also achieved by compensatory damages, would for the future in cases of personal injury fall to be fulfilled by exemplary damages alone. That is a view sometimes misunderstood, as Thomas J pointed out in Bottrill (at [121]). Donselaar did not trigger legislative intervention, as might have been expected if it was thought inconsistent with the policy of accident compensation or with a wider public interest in the confinement of remedies in tort to compensation.

  30. It is significant that when the Legislature did move to enact what is now s 319 of the Accident Compensation Act 2001, to reverse the restriction imposed by the Court of Appeal in Daniels v Thompson,[30] it did not introduce any cause of action refinement, such as by restricting the s 319 relaxation to intentional torts or by excluding exemplary damages where the defendant‘s conduct, although objectively outrageous, is not subjectively advertent. With such legislative context, it is wrong to suggest that New Zealand‘s accident compensation system provides distinctive local context which the Privy Council in Bottrill was not well-placed to assess and which justifies this Court in restricting the availability of damages for liability in negligence to cases of subjective recklessness. As Cooke P explained in Re Chase (at 329), the purpose of exemplary damages "is to punish the perpetrator of a tort, and they are outside the scope of the statutory compensation scheme". Moreover, the condition of subjective recklessness adopted by the Court in this judgment on the reasoning of the majority is acknowledged to attach also to non-personal injury cases. It cannot therefore be justified by reference to New Zealand‘s accident compensation regime.

    A “species of negligence”

  31. A requirement that subjective recklessness is a necessary threshold for liability to exemplary damages in negligence would restrict exemplary damages to a subcategory or "species of negligence": See Tipping J at [75]. Tipping J in the Court of Appeal in Bottrill (at [173]) accepted the significance of the change in describing it as achieving a "synthesis" between the intentional and unintentional torts for the purpose of exemplary damages. As Thomas J pointed out in the same case (at [134]) and as is suggested by Lord Nicholls‘s rejection of the analogy with misfeasance in public office (at [31]), the "synthesis" is achieved through the creation of the new "species of negligence", based on conscious foresight. As Thomas J said (at [139]):

    [N]egligence is regularly concerned with what the defendant "ought" to have known or "ought" to have done. The majority‘s requirement of subjective awareness eliminates "ought" from the language of exemplary damages. No conduct, however heinous, that the defendant "ought" to have known would create a risk will be subject to condemnation and punishment by an award of exemplary damages.

    He points out (at [138]) that a consequence of the insistence on subjective awareness "will effectively put beyond the reach of exemplary damages all human conduct, however blameworthy and deserving of condemnation, involving negligent omission". I agree with Thomas J that the requirement of subjective consciousness should not be imposed to put beyond the reach of exemplary damages negligent conduct which, because of its quality or extent, or its duration or repetitiveness, or casualness or indifference, or any other reprehensible feature, is adjudged to be beyond the bounds of what the community is prepared to tolerate without legal redress.

    Bottrill should not be reconsidered on the present appeal

  32. It is open to this Court to depart from a decision of its own or of the Privy Council on appeal from New Zealand if it is right to do so because the rigid adherence to precedent would lead to injustice in the particular case or would unduly restrict the proper development of the law to meet the needs of New Zealand society (see Supreme Court Act 2003, ss 3(1) and 13). That could be the case where the Court comes to the view that an earlier decision is wrong or has become wrong. But it would seldom be appropriate to take that course because of a difference in intellectual preference. The Privy Council decision in Bottrill is binding on all New Zealand courts apart from this Court. Departure from such authority is warranted only for good reason. That is discipline accepted in all common law jurisdictions by judges of final appellate tribunals, even when they disagree with the earlier decision.[31] Such respect for precedent is an aspect of the rule of law because "no judicial system could do society‘s work if it eyed each issue afresh in every case that raised it".[32] Here, no new arguments are put forward for one position or the other. We are invited to come to a different conclusion from the Privy Council by rehearsing yet again the same arguments earlier weighed. In this exercise we act as first and last court, lacking the benefit of considered judgments from the High Court or Court of Appeal. That cannot be avoided in cases where the Court of Appeal is bound by higher authority and the matter must be determined.[33] But the circumstance that a first consideration will be final should prompt great care, especially when a final court is asked to restrict the jurisdiction of the courts for the future. Any such restriction of jurisdiction is to be viewed with suspicion, for the reasons given by Cooke P in Re Chase (at 333 (citation omitted)):

    As indicated in Donselaar v Donselaar and other cases, it does not seem to me a legitimate function of the High Court, or this Court on appeal, to renounce or narrow inherent jurisdiction. The same applies to wide discretionary jurisdiction conferred by statute. Attempts to foreclose the categories of cases in which such jurisdiction may appropriately be exercised can be equally short-sighted. It is given to no Judge to foresee all the possible kinds of case, or all the shifts in what the public interest will require from time to time. Restraint in exercising such jurisdiction is quite another matter ....

  33. I have expressed my agreement with the decision of the Privy Council in Bottrill, because other members of the Court have dealt with the argument that it should be overruled. My own view is that it is wrong to use this failed strike-out as an opportunity to overrule Bottrill. There are two reasons: first, experience since the Privy Council decision does not suggest the need for revision; secondly, it is undesirable that such significant revision be undertaken on preliminary argument and without the context of findings of fact.

    (i) New Zealand practical experience since Bottrill does not prompt reassessment

  34. It has not been shown that experience in the eight years since the Privy Council decision has thrown up any practical problem with its result, such as might prompt reassessment. The fear that unmeritorious or extravagant claims may be made (see Blanchard J at [66]; Tipping J at [135]; and McGrath J at [241]) is a floodgates argument put forward on the basis of impression. Like all such arguments, it is likely to be more alarming in future contemplation than in reality. Certainly the risk of extravagant awards was thought in Taylor v Beere to be one the courts were well-placed to contain (at 92–93 per Richardson J). The risk in a particular case of doubling up or topping up compensation, through failure to keep compensatory damages distinct from what is necessary to reflect the exemplary principle, is something the courts must be alert to in all cases, as Cooke J made clear in Donselaar (at 107). The risk arises equally when personal injuries are not the basis of claim. Despite the advantage Parliament might be thought to have in assessing the validity of floodgates concerns, no such considerations prompted a restrictive approach in the enactment of what is now s 319 of the Accident Compensation Act 2001.

    (ii) Reconsideration of Bottrill is inappropriate in the present appeal

  35. Cooke P‘s warning in Re Chase as to the dangers of limited foresight when abjuring jurisdiction applies in particular in a preliminary determination of law ahead of findings of fact. Unless there is a clear and fatal flaw (such as statutory impediment) warranting peremptory determination, the court in such a case is required to identify legal principles for the future exercise of jurisdiction without the factual context to provide points of distinction in the cases of tomorrow. There is particular risk in areas of developing or disputed common law. Ex cathedra statements about the organisation and limitation of jurisdiction too often end in confusion or worse.[34] The risk may be illustrated by Rookes v Barnard itself, despite the high standing of those who comprised the Judicial Committee of the House of Lords in that case. In Rookes v Barnard the House of Lords was not faced, as we are, with a clear precedent to the contrary. It was attempting, rather, to impose order on an area of common law then acknowledged to have been confused, particularly because of lack of clarity in the different functions served by aggravated and exemplary damages. Those are not the conditions today in New Zealand law.

  36. If undertaken, change should be in the context of actual facts wherever possible. That is the method of the common law. Such context is not always available. In Kuddus, for example, the lower courts had struck the claim out and so it was necessary for the House of Lords to consider the issues of law[35] ahead of findings of fact, a circumstance that Lord Slynn (at [1]) considered to have been unsatisfactory. In the present case, we are not in that position. The claim was struck out in the lower courts not on the question of the availability of exemplary damages but because it was held that the defendant owed no duty of care to the plaintiff. In accordance with the views expressed in the judgment of 13 June 2008, the claim must be reinstated.

  37. The blameworthy conduct of the defendant which is relied upon as justifying exemplary damages has not yet been properly pleaded. As is apparent from the reasons for judgment of 13 June 2008, even the facts giving rise to any duty of care and to the manner of its breach remain unclear and substantially unpleaded. The judgment of this Court concluded only that breach of a duty of care owed to the plaintiff cannot at this stage be excluded. As I indicated at [38] of my reasons in the earlier judgment, in identifying whether there was breach of a duty of care, knowledge of risk is likely to be key. The same consideration will be important and may well also be key on the claim for exemplary damages.

  38. It is acknowledged that the claim must proceed because conscious recklessness remains in issue in any event. In such circumstances, the view that we should entertain the legal point at this stage to save the parties the trouble of appeal after trial strikes me as justification that is stretched. It is the case that the trier of fact (whether judge or jury) may have to address any question of exemplary damages eventually reached in a sequenced way, according to whether it finds subjective recklessness or not, in order to preserve the ability to seek a reconsideration of the need for subjective recklessness. But the inconvenience of that course in the present case is outweighed by the undesirability of restricting jurisdiction and overruling high authority without at least the discipline of facts.[36] The facts as finally pleaded and as found at trial may well overtake the issue entirely. If the scope of exemplary damages is to be cut down and new limitations established, it is preferable that it be done on established facts and in a case where the issue is not conjectural.

  39. The background here of statutory duties and powers is an important point of distinction with Bottrill. The defendant is sued in respect of institutional failings in a government department. Bottrill was not a case concerning the conduct of public officials, with allegations of negligence amounting to both misfeasance and nonfeasance. I have suggested in my reasons in support of the judgment of 13 June 2008 that a defendant in such circumstances cannot be treated as though an indifferent bystander (Couch at [57]). I would not foreclose as a matter of law at this stage the argument that, similarly, such a defendant may be liable for exemplary damages on the basis of ignorance of circumstances of which the department or officials ought to have been aware. If Bottrill is to be reconsidered and we are to impose the view that such breach of duty, however outrageous, can never warrant the award of exemplary damages, such reconsideration should not simply apply the reasons that convinced in a much more simple case. Significant questions of legal policy here arise, not present or considered in Bottrill, concerning the purpose of exemplary damages, vicarious liability, and the liability in tort of public officials. Some are discussed in the reasons of the Court of Appeal in the present appeal but in my view they are not appropriate for consideration by this Court on preliminary hearing. These are areas of law which may be developing and in such a case consideration in a factual vacuum carries especial risk. In respect of such cases it would be wrong on preliminary argument to foreclose the jurisdiction to award exemplary damages for negligence accompanied by neglect of responsibility which is outrageous. Exemplary damages, as Lord Kilbrandon noted in Broome v Cassell (at 1134), are based on the footing that "there is an element of public interest to be protected". As the Supreme Court of Canada was prepared to allow in Whiten (at [37]), that public interest may only be addressed by a plaintiff with the incentive to establish that the defendant behaved abominably. The jurisdiction to award exemplary damages in a proper case, in vindication of a public interest otherwise not readily able to be addressed, should not be denied on the basis of the limited argument addressed to us and without a proper factual context.

  40. Bottrill was not a case complicated by vicarious liability. The pleadings in the present case are still not clear as to the extent of reliance on the admitted vicarious liability for any tort committed by the probation officer. As indicated in the judgment of 13 June 2008, some of the claims of systemic failings describe direct liability of the Probation Service. To the extent however that the Probation Service may be liable for the negligence of the probation officer on a vicarious basis, it may be an open question whether the outrageous conduct which could justify an award of exemplary damages is only that of the probation officer or may also extend to the conduct of the entity or persons vicariously responsible for her actions. The answer depends on legal policy, reflecting the public interests served by liability for exemplary damages and vicarious liability.[37] Some cases and commentators suggest that an employer vicariously liable for the tort of an employee, who has taken all reasonable steps to prevent harm (or perhaps more accurately to prevent the outrageous feature which warrants exemplary damages), should be liable only for compensatory damages.[38] Others recognise public interest in permitting such claims to promote higher standards: See McGregor on Damages (18th ed, London, 2009) at [11-048]–[11-049]. The matter has not been addressed in the present appeal, it seems because of the undertaking given to the probation officer (reflected in the draft statement of defence) that the Attorney-General accepts vicarious responsibility for any exemplary damages awarded against her and perhaps because the basis of the exemplary claim is not yet pleaded. In my view it would be wrong for this Court to express any view on this difficult point. Whether some fault on the part of the person vicariously responsible is required for an award of exemplary damages raises significant questions of policy which it would be inappropriate to enlarge upon here in the absence of argument. Nor could it be considered in the absence of identification of the wrongful conduct of the vicariously liable defendant said to warrant exemplary damages.

  41. The case then is quite unlike the claim in Bottrill and raises complexities not contemplated there, touching upon uncertain and developing law and the responsiveness of New Zealand law to modern conditions. What this means is that it is unsafe to view the disposition of the present appeal as turning on a simple choice between the approach of the majority of the Privy Council or the approach of the majority of the Court of Appeal in Bottrill. If reassessment of Bottrill is to be undertaken and the jurisdiction to award exemplary damages restricted in negligence to cases where the defendant is conscious of the risk of harm to the plaintiff, it should be in a wider context than was necessary in respect of the argument as to whether Ms A should be granted a rehearing of her claim to exemplary damages on the basis of new evidence as to the scale and seriousness of Dr Bottrill‘s errors.

    Conclusion

  42. Bottrill is not a precedent that is restrictive of New Zealand common law. The effect of the decision should not be exaggerated. Although the possibility of exemplary damages for conduct that does not amount to conscious recklessness as to risk is not excluded as a matter of law, absence of conscious wrongdoing remains a substantial element as a matter of fact in the assessment (one of degree) whether conduct is outrageous enough to warrant an award of exemplary damages. Bottrill cannot be said to limit the responsiveness of New Zealand law to the developing circumstances of New Zealand society, such as might call for more anxious assessment. No compelling reason has been advanced for departing from it.

  43. Tipping J (at [91]) suggests that the choice between the approaches preferred by the Privy Council and by the Court of Appeal in Bottrill is essentially a choice of "legal policy". Legal policy in this is not however at large. While acknowledging that the availability of exemplary damages for negligence had not previously arisen directly for determination in New Zealand, Lord Nicholls pointed out that judicial observations in a number of cases in New Zealand and other common law jurisdictions were consistent with exemplary damages being available across the range of torts with "the ultimate touchstone constantly mentioned .... that of outrageous conduct by the defendant which calls for punishment": Bottrill (PC) at [43]. He concluded (at [49]):

    Overall this summary suggests that Courts in other countries have not found it necessary in practice to restrict the scope of exemplary damages in cases of negligence to cases of intentional wrongdoing or conscious recklessness. Wisely, they have left the door ajar.

  44. I, too, would leave the door ajar. In agreement with Thomas J, I would not eliminate the "ought" from negligence for exemplary damages. I would let this claim proceed, leaving it to the trier of fact to determine whether the conduct of the defendant warrants an award of exemplary damages. It is not possible to be confident that the claim for exemplary damages is clearly untenable. Conduct of the defendant may be outrageous and deserving of denunciation through exemplary damages not because it entails advertent appreciation of risk, but because it should have. It may be outrageous or deserving of denunciation through exemplary damages not because it entailed subjective recklessness as to the harm suffered by the plaintiff, but because it was outrageously indifferent to responsibility. It may be deserving of denunciation through exemplary damages because, even though the risk to the plaintiff was not foreseen, the conduct of the defendant was outrageously high-handed or cruel or contemptuous. Such cases are likely to be rare. But it would be wrong to renounce the general jurisdiction to award exemplary damages wherever the conduct of the defendant, although outrageous, is not consciously reckless as to risk.

    Blanchard J

  45. I agree with the conclusions reached by Tipping and McGrath JJ. I write separately because the majority of this Court is declining to follow a decision of the Privy Council (see Bottrill v A [2002] UKPC 44, [2003] 2 NZLR 721) and because it is necessary to say why I do not accept that it is premature or inappropriate for this Court in this case to take a position on whether exemplary damages can ever be awarded for inadvertent negligence, where the tortfeasor has not been conscious of the risk posed to the plaintiff by the act or omission which has been causative of injury (the Bottrill issue).

  46. At the outset, I should make it clear that I am proceeding on the basis that intentional wrongdoing, including subjective (advertent) recklessness, may be characterised as a form of negligent behaviour: Gray v Motor Accident Commission [1998] HCA 70, (1998) 196 CLR 1 and B(M) v British Columbia 2001 BCCA 227, (2001) 197 DLR (4th) 385.

    Should a ruling be deferred?

  47. Should the point of law upon which the majority of the Privy Council reversed the majority decision of the Court of Appeal in that case (the Bottrill issue) be revisited at this stage of the present proceeding? Is it suitable for determination before completion of the pleadings and of discovery and, indeed, before the facts have been found at trial? I would answer these questions affirmatively. I am not persuaded that any refinement of the relevant pleading or any wider factual context is required for the resolution of the legal issue which is now before us. We are not being asked to say what particular conduct of the probation officer or of her superiors was or was not in breach of a duty of care owed to the plaintiff/appellant. It can for present purposes be assumed that there has been gross negligence. The issue is simply whether, no matter how gross it was – if there can indeed be degrees of grossness – it is nevertheless insufficient for an award of exemplary damages if neither the probation officer nor her superiors had an appreciation that a failure to take general or particular steps or precautions in the supervision of Mr Bell posed a risk to Ms Couch or to someone in a like position. That does not require any advance determination of whether and in what respect there has been any gross negligence.[39]

  48. In my view it is desirable to determine the Bottrill issue at this stage because the parties need to know where they stand on that legal issue before going to trial. It would be unfortunate, if the matter were to come to this Court again after trial, and the Court then declined to follow Bottrill (PC) and were consequently obliged to order a new trial because the jury had not been asked to say whether, if it found for the plaintiff, it did so on the basis of advertent or of inadvertent conduct by the defendant. Even if it were asked to make such a finding of fact, a further appeal would be inevitable if it found against the defendant on the basis of inadvertent negligent conduct. It is better that this complication, in what may be a lengthy and complex trial, is removed by a decision of this Court at this stage.

  49. It is true that we do not have the benefit of the views of the trial Judge and the Court of Appeal on the Bottrill issue but those Courts are bound by precedent to follow the Privy Council decision, and very properly can be expected to refrain from expressing their own opinions.

    Privy Council precedent

  50. The Chief Justice takes the view that it would be wrong for the Court to depart from the recent Privy Council precedent of Bottrill; that it has caused no difficulties in practice; and that the position taken by the majority of the Privy Council is "a tenable view" on which no new arguments have emerged.

  51. It is right of course that this Court should be slow to overturn a Privy Council precedent. Occasions justifying doing so will be comparatively rare. It would be unsettling for New Zealand law if it were thought by litigants and lawyers that the Supreme Court would easily be persuaded to depart from law apparently settled by a decision of the Privy Council. It is certainly not enough that some criticisms may have been made of the decision in question. This Court is not bound by a Privy Council decision, even one on appeal from this country, but, to adopt what has been said in the High Court of Australia, the Court should depart from a Privy Council decision only "if in the proper performance of its duty it feels that it should do so": Barns v Barns (2003) 214 CLR 169 at [101] per Gummow and Hayne JJ (quoting Aickin J in Viro v The Queen (1978) 141 CLR 88 at 174). In my view, it should not do so unless it is satisfied that the Privy Council decision was not only in error but also inappropriate for the proper development of New Zealand law in New Zealand conditions. In the earlier decision of John v Commissioner of Taxation of the Commonwealth of Australia, five members of the High Court of Australia gave some guidance about when that Court regarded itself as having power to depart from its own earlier decision, adding that such a course is not lightly undertaken: John v Commissioner of Taxation of the Commonwealth of Australia (1989) 166 CLR 417 at 438. That guidance seems equally applicable to earlier decisions of this Court and of the Privy Council in New Zealand appeals. Four considerations derived from The Commonwealth of Australia v The Hospital Contribution Fund of Australia (1982) 150 CLR 49 (at 56–58) are endorsed in the joint judgment in John:

    1. Whether there is already a principle carefully worked out in a significant succession of earlier cases;

    2. Whether in an earlier case there are differences between the reasons given by the Judges who were in the majority;

    3. Whether the earlier case has achieved no useful result but has led to considerable inconvenience; and

    4. Whether or not the earlier case has been acted on in a manner which militates against reconsideration.

    These considerations are of some present relevance and I will come back to them after discussing the Bottrill case.

    The Bottrill decision

  52. My reading of the case law leads me to the conclusion that the Privy Council‘s decision in Bottrill is out of step with the way in which the law has developed both in New Zealand and in the jurisdictions with which we usually compare ourselves. It appears to be the only decision of a senior court, directly addressed to the issue, which expressly countenances exemplary damages for an inadvertent act of negligence. That of course does not make it wrong or unsuitable for New Zealand but it is an unpromising beginning.

  53. Little need be said of the prior law in this country and in Britain, which is fully described in the reasons of Tipping and McGrath JJ. It is, however, quite plain that prior to Bottrill in the Privy Council the New Zealand Court of Appeal had set its face against an award of exemplary damages where there was no consciousness of wrongdoing on the part of the defendant, for the reasons given by the Court of Appeal majority in that case. The Privy Council decision must also have come as a surprise in England, if intended to guide the courts there, not only because of its inconsistency with the first of the categories to be found in Rookes v Barnard[40] but even more so because it markedly diverged from what had recently then been said in Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29, [2002] 2 AC 122, even by Lord Nicholls himself (at [63]). It came as a particular surprise in this country, too, when the Privy Council departed from its practice of not interfering with decisions of local courts based on policy grounds articulated in relation to local conditions.[41] The decision was less than rapturously received by most commentators in this country.[42]

  54. The position in Australia is not as Lord Nicholls thought it to be. The central passage in the joint judgment in the decision of the High Court of Australia in Gray v Motor Accident Commission (1998) 196 CLR 1 at 9[43] makes that quite clear:

    No question arises here of an intentional wrong being committed by inadvertence. For present purposes it is enough to note two things. First, exemplary damages could not properly be awarded in a case of alleged negligence in which there was no conscious wrongdoing by the defendant. Ordinarily, then, questions of exemplary damages will not arise in most negligence cases be they motor accident or other kinds of case. But there can be cases, framed in negligence, in which the defendant can be shown to have acted consciously in contumelious disregard of the rights of the plaintiff or persons in the position of the plaintiff.

    [emphasis added]

    An earlier passage cited by Lord Nicholls, in which it is said that conscious wrongdoing in contumelious disregard of another‘s rights describes "at least the greater part" of the relevant field, was actually directed at the whole field of exemplary damages. It was not directed towards the narrower question of claims made in negligence and must be read in light of the central passage quoted above, which was. If that requires any confirmation, it is to be found in a subsequent judgment of the High Court, joined by one of the plurality Judges in Gray, Gummow J, in New South Wales v Ibbett (2006) 229 CLR 638. There the Court recollected that it had previously said that there may be cases, framed in negligence, in which the defendant could be shown to have acted consciously in contumelious disregard of the rights of the plaintiff or persons in the position of the plaintiff. The significance comes from a footnote (footnote 56 at [47]) which cites the passage from Gray quoted above and contrasts it with Bottrill (PC) and with the Canadian case of Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 at 634–635. By making the latter contrast in the judgment in Ibbett the High Court can only be taken to be indicating that the Supreme Court of Canada had not fully appreciated the intended effect of the passage from Gray, that is that "exemplary damages [cannot] properly be awarded in a case of alleged negligence in which there was no conscious wrongdoing by the defendant".

  55. In his reasons, Tipping J describes the position in Canada, noting that Whiten involved a claim in contract where there was deliberate disregard for the rights of the plaintiff when she made a claim on her insurance policy. It was not a negligence claim. A passing reference in which the Supreme Court appeared to approve of the possibility of an award of punitive damages in a case of inadvertent negligence was accompanied by the comment that such an approach seemed to be in line with most common law jurisdictions apart from England. With respect, that was an erroneous observation in so far as it was based on a survey of cases from Australasia and may have been influenced by decisions in the United States which I regard as problematical as a precedent for courts in New Zealand.

  56. To the Canadian references given by Tipping J may be added the summary of the position in that country given before Whiten by another text writer, Fridman (The Law of Torts in Canada (2nd ed, Carswell, Ottawa, 2002) at 506):

    Crucial to the granting of punitive damages is the fact that the defendant‘s behaviour was outrageous or subjected the plaintiff to a humiliating or highly unpleasant experience that transcends the usual hurt, injury or loss to be expected from the kind of tortious behaviour involved. This imports some degree of intent, malice, or deliberate infliction of harm on the part of the defendant. Where negligence is the cause of action, such features are normally absent. The defendant may have been careless, to the extent that he is liable in tort, but he has not behaved with such disdain for the rights of the plaintiff, or such evil intent, that he merits some further penalty. Hence, as often stated by the courts, awards of punitive damages for negligence will be rare. Before such damages will be awarded in a negligence action there must be something more than mere negligence. There must be proof of such entire want of care as to raise a presumption that the defendant was conscious of the probable consequences of his carelessness and was indifferent, or worse, to the danger of the injury to other persons. The test is whether the defendant‘s conduct was so "reprehensible" as to warrant punitive damages over and above compensation. This requires proof of some high-handed conduct, with the intention of disregarding the plaintiff‘s rights. Unless such circumstances are present, there is no rational purpose in awarding punitive damages.

    [footnotes omitted]

  57. But although Bottrill (PC) may be out of step, is it nonetheless the better approach and suited to New Zealand conditions, which of course differ from those in other jurisdictions because of the accident compensation scheme‘s prohibition on almost all claims for compensatory damages in negligence for bodily injury, and many claims for mental injury?

  58. The purpose of exemplary damages is to punish the defendant for outrageous conduct which has harmed the plaintiff and, by denouncing that conduct and marking it out for punishment, to deter the defendant from any repetition and to deter others from behaving in a similar manner. The focus should therefore be on the character of the defendant‘s conduct, not on the loss or suffering of the plaintiff. The subject of the punishment is the private wrong committed against the plaintiff, not a public wrong against the State. The exemplary damages go to the plaintiff, not to the State. That may be why, in jurisdictions other than New Zealand, they can be claimed only when private injury or loss has been proved, and therefore only in conjunction with compensatory damages, upon which they are parasitic. In this jurisdiction, in order to justify exemplary awards since the accident compensation scheme commenced, the courts have had to decouple them from compensatory awards.

  59. To my mind, there is a real question whether a civil court should involve itself in punishing people,[44] especially those who have genuinely not been conscious of doing any wrong.[45] However, although I do not feel comfortable with the idea that a court in its civil jurisdiction should, in the absence of any statutory power, engage in meting out punishment, I can nevertheless accept that there is a proper moral role for exemplary damages as a deterrent to outrageous harmful behaviour.[46] In performing this role such damages may in this country provide a useful counter-balance to the absence of the sanction of compensatory awards for tortious infliction of personal injury. They may have especial value in cases of systemic fault which causes such injury.

  60. But if the defendant was in truth not conscious of creating a hazard for others, the deterrent role of exemplary damages is much diminished. The deterrent purpose will be largely superfluous in relation to such a defendant. That must count against them when their justification is examined in an overall context. It is to be remembered that the behaviour in question, in order to qualify for the sanction of exemplary damages under Bottrill (PC), must be a very extreme departure from the standard expected of a reasonable person. The rare defendant who was quite unconscious of risk caused by such grossly negligent behaviour will not have wanted to cause harm and, having seen the consequences of that behaviour, is also unlikely to repeat it, even in the absence of a sanction, simply because of seeing the harm which has resulted. And where significant personal injury has been caused by gross negligence the wrongdoer will very likely have suffered a stringent criminal and/or disciplinary penalty, as well as a public shaming. An additional penalty by way of a relatively modest award of damages is unnecessary to achieve the aim of future deterrence of the wrongdoer, especially as it will have to be reduced to take account of any criminal penalty in order to avoid a double penalty. That it may arguably be a means of appeasement for the injured person is not, in my view, a sufficient reason for permitting such awards where they are not in reality needed as a deterrent to a defendant who has not acted deliberately or with a conscious appreciation of putting others at risk.

  61. The case for awarding exemplary damages as a deterrent to others is, I accept, rather more arguable. An award of exemplary damages for inadvertent negligence may in theory provide some incentive for others who learn of it to refrain from similar behaviour. They may be brought to an awareness of risk by becoming aware of the award against the defendant. It seems hardly principled, however, to punish an inadvertent wrongdoer by means of a damages award in order to provide a deterrent to someone else when a sufficient deterrent for others exists in the form of the almost inevitable criminal prosecution of the wrongdoer for negligence causing injury or death and, in the case of a professional, there will have been a sanction imposed by a disciplinary body. For its likely minimal consequential effect on others as a deterrent, it may be thought not to merit the allocation of the scarce resources of the civil courts.

  62. Furthermore, the existence of the possibility of an award in admittedly rare circumstances – on the "never say never" approach of Lord Nicholls – gives rise to several problems. First, a jury or trial judge will be called upon to make a difficult and unpredictable value judgment on the defendant‘s inadvertent conduct in order to be able to say whether in their opinion it crossed the line into outrageousness. The standard of outrageousness itself can fairly be said to be uncertain in application even where the defendant was a conscious wrongdoer.[47] That uncertainty and unpredictability is at least contained to some extent if it must also be shown that the defendant acted with consciousness of risk. Contrary to the view of Lord Nicholls, the need to determine whether the defendant was conscious of creating a risk is not a distraction or requiring of an especially hard factual assessment. It can be deduced from the nature of the defendant‘s conduct and from any statements which shed light on the defendant‘s mental processes at the time. That is familiar territory for both juries and judges.

  63. It has also been suggested that drawing a line between advertent and inadvertent negligence is difficult in another way, because conduct which may consist of inadvertent negligence in one respect may be combined with deliberate conduct in another respect. But the question in any such case is merely whether a defendant was conscious of creating a risk of injury or loss to the plaintiff. Take the examples given by Thomas J in  Bottrill v A [2001] 3 NZLR 622 (CA) at [145]. The first was posited by Glanville Williams and BA Hepple (Foundations of the Law of Tort (Butterworths, London, 1976) at 120), of a defendant pointing and firing a gun believing it to be unloaded. Surely the defendant who acted without checking on the accuracy of that belief cannot be heard to say that he was unaware that there was some degree of risk that there was a bullet in the chamber. If, when the trigger was pressed, the gun was knowingly pointed at someone, or in a direction where someone might be present, the act was advertently reckless. Likewise, a medical practitioner who does not bother to read medical journals must surely be taken to be aware of running the risk of not keeping up to date and thus doing something which may needlessly harm his or her patients. This is why the minority in the Privy Council in Bottrill was, I now believe correctly, able on the facts, and despite confirming the law to be as stated by the Court of Appeal, to concur in allowing the appeal.

  64. It is not easy to conceive in realistic terms of some act of negligence accompanied by conduct which could ever merit being characterised as outrageous where that conduct does not in itself demonstrate, like the examples just given, that the defendant must have been aware of the risk to the plaintiff or someone in the position of the plaintiff. One could perhaps conjecture a grossly negligent act causing injury where afterwards the defendant took advantage of what had occurred, or decamped leaving the plaintiff to suffer. In relation to grossly negligent behaviour, either of those subsequent reactions would go a long way towards demonstrating that, at the very least, the defendant must throughout have been conscious of the grossly negligent quality of the act or omission.

  65. It is possible that there may be shown to be a consciousness by a defendant of inadequacy of its systems or deliberate indifference to its responsibilities, but not consciousness as to the harm that may eventuate. That does however seem rather unlikely when the present issue simply does not arise unless the negligence was very extreme. And, if it were the case, and the possibility of some injury was really not seen by the defendant as a risk, its conduct could not in my view fairly be characterised as outrageous, arbitrary or high-handed in the context of a claim for damages.

  66. The second problem is that, if Bottrill (PC) stands, then notwithstanding the Judicial Committee‘s admonition that successful claims for exemplary damages for inadvertent negligence will be rare and damages awards moderate, the remote possibility of success is likely to lead to a not insignificant number of hopeful claims from unrealistic litigants (or their unrealistic advisers) which will almost certainly fail, but will be costly to defend and unnecessarily add to the congestion of the court system. That is of no benefit to anyone, certainly not to a plaintiff who is effectively chasing a will-o'-the-wisp. Observation of some claims made in the lower courts leads me to the view that this problem has already manifested itself.

  67. The third problem relates to the accident compensation scheme. Allowing claims for inadvertent negligence may encourage the belief in some plaintiffs and their lawyers that exemplary damages are a means of topping up what they perceive to be inadequate payments under the scheme. Furthermore, some potential defendants may because of the nature of their businesses feel the need not only to pay for accident compensation cover but also to have and pay for third party insurance against the consequences of negligence in the form of personal injury. Many such potential defendants would not see that need if exemplary damages were confined to intentional acts, including subjective recklessness, for which in any event insurance cover may not be available, at least for the actual wrongdoer.[48] It is most unlikely that an insurance company would offer cover just for inadvertent gross negligence.[49] So the extra cover will have to be for all negligence causing personal injury.[50] No doubt the premiums will be relatively modest because few claims will actually require indemnification but there will certainly be some extra cost if this Court confirms Bottrill (PC), where the Law Lords appear to have paid no attention to this consequence of their decision. I am not of course suggesting that the present defendant would be likely in the future to take out insurance cover against the type of claim now being faced, but many other potential defendants, such as drivers of motor vehicles and engineers, may think it necessary to do so in differing factual situations where there is an exposure to a negligence claim for causing personal injury.

  68. For these reasons, and applying an overall cost/benefit analysis, I conclude that it is distinctly preferable to remove from our law the ability to claim exemplary damages against an unconscious wrongdoer for causing personal injury.[51] On balance, therefore, I continue to prefer the view taken in Bottrill by the Court of Appeal majority, of which I was a member, and by the Privy Council minority.

    Declining to follow Bottrill

  69. I am also satisfied that this is a proper occasion for this Court to decline to follow the decision of the Privy Council. For the reasons I have given, I believe that it does not represent an appropriate statement of the law for this country. Referring again to the first consideration mentioned by the High Court of Australia in John, far from having been worked out in a significant succession of earlier cases, Bottrill (PC) in fact was a departure from precedent and, as I have shown, out of step with the law as previously declared in England and Australia. A second consideration is whether there were differences between the reasons given by the majority Judges. That is not likely to be a feature of Privy Council decisions, where concurrent judgments have been rare. But it must be of some moment that in Bottrill there was a dissenting opinion of two Law Lords who strongly favoured the position on the legal issue taken by the majority of the Court of Appeal. I cannot say, as I have indicated, whether, as yet, there has been "considerable inconvenience in practice" although I understand that a number of claims for exemplary damages are pending in the High Court. But, as I believe emerges from the above discussion, it seems to me that the existence of a possibility of an award for inadvertent negligence creates inappropriate incentives which outweigh the advantage of preserving Bottrill (PC) for the very rare case which might be successful on the basis of that decision. And, finally, Bottrill (PC) does not appear to have been acted on in a manner which militates against reconsideration. It cannot be said that anyone will have arranged their affairs in advance on the strength of it, save by taking out insurance cover which, consistently with the accident compensation scheme, should not be necessary.

  70. I therefore consider that this Court should decline to follow the Privy Council‘s decision in Bottrill.

    No statutory obstacles

  71. I agree with Tipping J, for the reasons he gives, that exemplary damages are not barred by s 317 of the Accident Compensation Act 2001 and that Ms Couch‘s claim is not defeated by s 86 of the State Sector Act 1988. At [173] of his reasons Tipping J touches upon the question of whether the Crown can be liable for negligence directly through attribution as well as vicariously. The law on that topic, which we are not called upon to consider on this appeal, is uncertain. Statutory reform is overdue, as Professor Anderson says in his paper to which Tipping J refers. I would not want it to be thought, however, that I am presently persuaded that a direct claim is necessarily precluded by the Crown Proceedings Act 1950, when read against a background of prior statutory history unique to this jurisdiction.

    Result

  72. I agree with the disposition of this appeal proposed by Tipping J.

    Tipping J

    Introduction

  73. The two main issues which arise on the resumed hearing of this appeal both concern exemplary damages.

    • The first is whether, in cases of personal injury covered by the Accident Compensation Act 2001, recovery of exemplary damages is barred by s 317 of the Act.

    • The second arises if there is no such bar and concerns the ingredients of the test for recovery of exemplary damages in respect of the negligent infliction of personal injury. The question is whether it is necessary for the defendant to be subjectively reckless in the sense of having a conscious appreciation of the risk of causing harm as a component of acting in an outrageous manner, before exemplary damages can be awarded. This second issue reduces to whether this Court should adopt the approach of the majority of the Privy Council (Bottrill v A [2002] UKPC 44, [2003] 2 NZLR 721) or that of the majority of the Court of Appeal (Bottrill v A [2001] 3 NZLR 622 (CA)) in the Bottrill litigation.

    The circumstances in which these issues arise are set out in the reasons of the members of the Court when dealing with the first stage of the proceedings in this Court: See Couch v Attorney-General (on appeal from Hobson v Attorney-General) [2008] NZSC 45, [2008] 3 NZLR 725. They need not be repeated.

  74. I should add that my description of the second issue is based on the rejection of what was a logically prior point advanced by the respondent, namely that there should be no capacity to award exemplary damages in any case of negligence. This submission must be addressed in the light of the Crown‘s acceptance, as I shall address below, that subjective recklessness should be regarded as a species of negligence. On that footing I would reject the proposition that exemplary damages should never be available for the tort of negligence. As I will elaborate later, subjective recklessness is much closer, at least for present purposes, to intention than it is to high level, that is, gross negligence.

  75. The Crown‘s proposition that there should be no room for exemplary damages in negligence cases has force in cases which fall short of deliberately running the risk of causing harm. But once that kind of case is regarded as a species of negligence, it is going too far to say that exemplary damages should not be available in any case of negligence. The policy considerations to be discussed later in these reasons support the availability of exemplary damages in cases of subjective recklessness. The second general issue, as I have described it, is therefore based on the premise that the Crown‘s logically prior contention should be rejected.

  76. The first issue arises because, if exemplary damages in personal injury cases are barred by s 317 of the Accident Compensation Act, the claim brought by Ms Couch cannot succeed and must be struck out. The second issue arises for two reasons. The first is that the viability of aspects of Ms Couch‘s present pleading depends on the answer to it. The second is that we were informed Ms Couch will seek to have her case heard by a jury. If that is what transpires the trial Judge would be obliged, as the law stands at present, to direct the jury in terms of the decision of the majority of the Privy Council in Bottrill. If this Court were later to decide that this was not appropriate, there would have to be a retrial on the correct basis.[52] Far better that the point be resolved before the trial so that all concerned have the security of knowing the correct legal basis on which the claim should be addressed.

    Need for ruling

  77. That was all I was proposing to say about the reasons why this Court is seized of the second issue at this stage of the proceeding. In light of the views expressed by the Chief Justice, I consider it would be appropriate to elaborate a little. Ms Couch‘s present pleading involves allegations of both subjective and objective recklessness. Mr Henry advised the Court that, after discovery and any other necessary interlocutory processes are completed, he will be amending his pleading to allege discretely, with appropriate particulars, subjective recklessness and, as an alternative, objective recklessness.

  78. The Crown‘s application to strike out the claim was based on two limbs;

    • first, want of a viable cause of action and

    • second, lack of support in the pleading for a claim for exemplary damages.

    We ruled in favour of Ms Couch on the first point in our first judgment. We are now concerned with the second point. As I have already indicated, the Crown‘s argument that the claim should be struck out entirely on the ground that neither subjective recklessness nor objective recklessness, as two different species of negligence can, in law, give rise to exemplary damages, fails because subjective recklessness is capable of giving rise to a claim for exemplary damages. But the Crown is entitled to a decision on whether the alternative basis on which the claim is framed is also capable of giving rise to a claim for exemplary damages. That is why I said above that the viability of aspects of Ms Couch‘s present pleading depends on the answer to that question. It cannot therefore be said that giving that answer is neither necessary nor desirable. This Court has a responsibility to provide an answer.

  79. The next point concerns the suggestion that the answer to the question is fact-specific and therefore giving the answer should wait until the facts have been identified with precision by means of a trial. On a strike-out application the facts, as alleged, are taken to be established. For the purpose of assessing the viability of Ms Couch‘s alternative allegation, objective recklessness must be taken as being established in the circumstances set out in the pleading. In any event, I do not consider that what is essentially a point of legal principle and policy is, or should be, influenced by the particular circumstances in which the issue arises. Subjective recklessness is either necessary or it is not.

  80. Furthermore, as this case may be heard by a jury, it is appropriate to point out that civil juries do not find facts in the same way as Judges do. Juries answer specific issues. The settling of issues is often difficult. The inherent difficulties should not be exacerbated by the law being uncertain. This would require the issues to be framed on the basis of one or more alternative legal premises – a most unsatisfactory state of affairs for both Judge and jury.

  81. At the conclusion of the hearing held to determine whether the present point would be examined before trial, the issue was framed as being whether exemplary damages were available for negligence. This formulation in no way envisaged that the Court might simply answer that question in the affirmative, but would not, at this stage, state what the criteria for such awards should be. On that basis the parties, and in particular the Crown, would have to wait until after trial and the matter returned to this Court before knowing for sure whether Ms Couch‘s alternative claim was viable in law.

  82. Until that time both the High Court and the Court of Appeal would be bound to apply the law as stated by the Privy Council. It is suggested that this Court suffers at the moment from the impediment of not having the views of the Courts below on the question at issue. But that is likely to be the position even after trial. As the High Court and Court of Appeal are bound by the decision of the Privy Council in Bottrill, neither Court is likely to offer its views gratuitously. Indeed, it would be unconventional in this kind of situation for them to do so.

  83. For these reasons I consider this Court should, indeed is obliged to, rule on what in [73] I have described as the second issue.

    Are exemplary damages barred by section 317?

  84. Section 317(1) of the Act provides:

    317.

    Proceedings for personal injury

    (1)

    No person may bring proceedings independently of this Act, whether under any rule of law or any enactment, in any court in New Zealand, for damages arising directly or indirectly out of—

    (a)

    personal injury covered by this Act; or

    (b)

    personal injury covered by the former Acts.

  85. In the present case Ms Couch claims exemplary damages arising out of personal injury covered by the Act. Her claim is framed in tort on the basis of breach of a duty of care. In other words, her claim is based on tortious negligence. The first question is whether the "damages" referred to in s 317(1) include exemplary damages. The second is whether, since no claim can be brought for compensatory damages arising out of personal injury covered by the Act, exemplary damages may nevertheless be awarded on a stand-alone basis. The Attorney-General‘s contention is that without actionable compensatory damages the tort of negligence is incomplete as damage is an essential ingredient. Hence there is no tort for which exemplary damages may be awarded.

  86. In Donselaar v Donselaar [1982] 1 NZLR 97 (CA), which involved the similarly worded s 5(1) of the Accident Compensation Act 1972, the Court of Appeal held that the damages which were barred by the section were compensatory damages. Exemplary damages were not barred. This involved a narrow reading of the word "damages" and may have been a debatable conclusion at the time, but far too much water has gone under the bridge since then to contemplate taking a different view. Since Donselaar was decided Parliament has had several opportunities to include exemplary damages expressly within the bar and has not done so.

  87. That must represent an acceptance of the Court of Appeal‘s conclusion that the policy of the accident compensation legislation is not undermined by permitting exemplary damages to be claimed in circumstances defined by the courts. This proposition is underlined by s 319 of the Act which was introduced into the legislation in 1998 to reverse the effect of the judgment of the Court of Appeal in Daniels v Thompson [1998] 3 NZLR 22 (CA). Section 319(1) provides:

    319.

    Exemplary damages

    (1)

    Nothing in this Act, and no rule of law, prevents any person from bringing proceedings in any court in New Zealand for exemplary damages for conduct by the defendant that has resulted in—

    (a)

    personal injury covered by this Act; or

    (b)

    personal injury covered by the former Acts.

  88. The following subsections provide that exemplary damages may be awarded despite the fact that the defendant‘s conduct has resulted in a criminal charge leading to either a conviction or an acquittal. These subsections do not cut down the general reach of subs (1). They simply make it clear that the circumstances mentioned in them do not preclude awards of exemplary damages. The effect is that Parliament has expressly recognised and reinforced the Donselaar principle whereby exemplary damages may be awarded in cases of personal injury covered by the Act. Hence the word "damages" in s 317 must be construed as referring only to compensatory damages (which include so-called aggravated damages). I cannot accept the Crown‘s argument that s 319 implicitly confines exemplary damages to circumstances which amount to crimes. The purpose of the section was to eliminate the double jeopardy bar created by Daniels v Thompson. There is nothing to suggest the section was designed to circumscribe the role of exemplary damages in the way the Crown suggested.

  89. Section 319 also has relevance to the second aspect of this first issue. Parliament has expressly provided that exemplary damages may be awarded despite the Court‘s inability to award compensatory damages. It is therefore no bar to a claim for exemplary damages for personal injury that no compensatory damages may be awarded.179 The ordinary need for there to be damage to complete the tort of negligence must for this purpose be satisfied by there being a case for the award of exemplary damages. The consequence is that there is no actionable tort of negligence for causing personal injury in New Zealand unless the case justifies exemplary damages.

  90. Section 319 says nothing, however, about the circumstances in which exemplary damages are available. What it does recognise is that they are available in circumstances to be determined by the court and if the case is one for exemplary damages an actionable tort has been committed despite the harm suffered by the plaintiff amounting to personal injury covered by the Act. That brings me to the second and more substantial general issue. In what circumstances may exemplary damages for negligently causing personal injury be awarded?

    179 The statutory context makes the present case clearly distinguishable from Watkins v Secretary of State for the Home Department [2006] UKHL 17, [2006] 2 AC 395.

    The second issue – the criteria for exemplary damages

  91. As earlier foreshadowed, this question is whether the law of New Zealand on this subject should be as formulated by the majority of the Court of Appeal or by the majority of the Privy Council in Bottrill. The crucial difference between the two formulations is that the Privy Council‘s sole criterion was whether the negligence was outrageous, whereas the Court of Appeal required not only that the negligence be outrageous, but that the defendant must have been subjectively reckless as to the harm caused. The choice between these two approaches is essentially one of principle and legal policy.

    The history and purpose of exemplary damages

  92. Awards of exemplary damages at common law can be traced back to the mid-18th century: See Huckle v Money (1763) 2 Wils KB 206, 95 ER 768 and Wilkes v Wood (1763) Lofft 1, 98 ER 489. In the Wilkes v Wood case the reasons for such awards were said to be a combination of satisfaction to the injured person, punishment of the guilty and proof of the detestation of the jury of the defendant‘s conduct. The history of exemplary damages was extensively and helpfully reviewed in 2002 by Thomas Colby ("Beyond the Multiple Punishment Problem: Punitive Damages as Punishment for Individual, Private Wrongs" (2003) 87 Minn L Rev 583). In its early stages the focus of the common law was on redress for causing direct and tangible harm. As the common law developed with the development of society, it came to recognise the need for redress in cases of intangible harm and harm which was indirectly caused. That recognition overtook one of the earlier purposes of exemplary damages, namely compensation for the insult. With these developments came an increasing focus on the punitive purpose of exemplary damages. The idea that the insult (a form of indirect harm) done to the plaintiff was deserving of punishment as well as compensation came to the fore and punishment, albeit anomalous in the civil field, came to be recognised as the essential underpinning of the power to award exemplary damages.

  93. The common law nevertheless insisted that this was not punishment in the abstract for morally wicked conduct; it was punishment for the commission of a private civil wrong in a particularly reprehensible way – in a way which was deserving of punishment. Exemplary damages could be awarded only if the defendant had committed a civil wrong for which compensatory damages, which might include compensation for elements of aggravation, were available. Only if the award of those damages was not enough to exact the necessary punishment should further damages of an exemplary kind be added. This is why exemplary damages were often described as being parasitic on compensatory damages. The punitive focus of exemplary damages was reinforced by the fact that exemplary damages were and still are sometimes referred to as punitive damages.

  94. In W v W [1999] 2 NZLR 1 (PC) Lord Hoffmann, writing for the Privy Council, stated simply that the main purpose of exemplary damages was to punish the defendant. Similar comments have been made in our Court of Appeal.[54] Very recently the Privy Council has confirmed this view, saying in Takitota v Attorney General [2009] UKPC 11, (2009) 26 BHRC 578 at [12] that the purpose of exemplary damages is to punish the defendant for outrageous behaviour, thereby deterring him and others from repeating it.

  95. There have been dicta in some recent cases, principally those of Thomas J in Daniels and Bottrill suggesting that the purpose of exemplary damages is substantially wider than to punish. Discrete purposes such as vindication, education, appeasement of the victim, therapeutic effect, and expressing general societal disapproval have been mentioned. I consider these features are best regarded as the consequences of a punitive award rather than as purposes of the award in their own right. Some of them represent purposes of punishment, and are therefore inherent in the concept of punishment. We punish to achieve societal goals and to reinforce societal norms. It conduces to clarity and simplicity in the field of exemplary damages to concentrate on their punitive purpose and to let such other purposes as may be implicit do their work silently.

  96. This concentration reinforces the point that exemplary damages are not a surrogate way of awarding greater compensation. By allowing other purposes than punishment to feature directly in the assessment, we increase the risk of at least an unconscious blurring of the line between compensation and punishment. This line is particularly important because of the presence in New Zealand law of the accident compensation scheme and because of the anomaly that what is really a fine goes to the plaintiff and not to the State. Such concentration will also steer the court‘s consideration away from the difficult approach, evident in the statement made by Thomas J in Bottrill (at [98]) that included in the concepts of vindication and appeasement is the "therapeutic effect of a civil trial in which the victim is an equal participant with the perpetrator of the wrongful conduct".

  97. No doubt plaintiffs and some legal advisors may view exemplary damages as having an additional compensatory effect, but that is not their purpose. They are not designed to top up any perceived shortcomings in what is available to those suffering personal injury under the accident compensation scheme. In this respect it is also worth pointing out that so-called aggravated damages are compensatory and the idea floated by Cooke J in Donselaar (at 107) that exemplary damages would have to take on part of the role of aggravated damages is problematical. To the extent that this remark has encouraged an expansive role for awards of exemplary damages, I would not adopt it.

  98. It would, in my view, also assist clarity of analysis to abandon the idea that aggravated damages are a category of damages in their own right. Properly viewed, the concept of aggravation is simply an element in assessing an appropriate amount of compensatory damages. If the wrong has been committed in a way or in circumstances which aggravate the effect of the wrong on the plaintiff, it is appropriate to award more by way of compensation to recognise that fact.[55]

  99. The following discussion therefore proceeds on the basis that the purpose of exemplary damages is to punish.

    Negligence and recklessness

  100. The English language encompasses two distinct states of mind within the single concept of recklessness. The law also recognises the distinction. A person may be described as reckless who does not appreciate an obvious risk of causing harm and proceeds to cause the harm without appreciation of the risk. This is what in law is known as objective recklessness. It is the practical equivalent of a high level of negligence. On the other hand, a person may appreciate the risk of causing harm and proceed nevertheless deliberately to run that risk and end up causing the harm. That is subjective recklessness. Subjective recklessness is generally seen as more culpable and deserving of punishment than objective recklessness. In the case of subjective recklessness there is a conscious appreciation of the risk that one‘s conduct may cause harm and a deliberate decision to run that risk. The greater the risk and the greater the harm which is likely to ensue, the more culpable the person‘s conduct will be and the more appropriate it may be to describe it as outrageous.

  101. The purpose of this discussion is to point out that, while objective recklessness sits comfortably with the concept of negligence, subjective recklessness does not. The inadvertence in objective recklessness is a paradigm of negligence; the advertence in subjective recklessness is closer to intention than to negligence. The point does not have to be explored any further because of the Solicitor-General‘s acceptance, for the purposes of the present issue, that no distinction should be drawn between the two types of recklessness. They are both a species of negligence.

  102. I should immediately add, however, that the Solicitor-General did suggest, and this was his central proposition, that there was a crucial distinction between the two types of recklessness for the different purpose of deciding when exemplary damages should be available. The submission in this respect was, as I will develop below, that exemplary damages should not be awarded unless the defendant had a conscious appreciation of the risk of causing harm and ran that known risk; in other words exemplary damages should be available for subjective and outrageous recklessness but not for objective recklessness, even if capable of being described as outrageous.

    Current state of the law

  103. Using the distinction just made, it is fair to say that the Court of Appeal in Bottrill adopted as the test for exemplary damages the standard of subjective recklessness of an outrageous kind, whereas the Privy Council adopted the lesser standard of objective recklessness of an outrageous kind. The majority of their Lordships rejected the proposition that before exemplary damages could be awarded the defendant must have consciously appreciated the risk of harm to the plaintiff his conduct was causing. Outrageous negligence was sufficient. That is how New Zealand law stands at present.

    Departing from the Privy Council

  104. As the Attorney-General contends that this Court should not follow the decision of the Privy Council in Bottrill, it is appropriate at this point to address, to the extent necessary, how this Court should, in general terms, approach such a submission. The Supreme Court now performs the same role as that previously performed by the Privy Council. It is logical therefore that we should regard a decision of the Privy Council as being for precedent purposes the same as a previous decision of our own. It would not be appropriate for this Court to regard itself as absolutely bound by its own previous decisions. No other comparable court now takes that approach to its own decisions. It is, however, highly desirable for the stability of the law, and the ability of citizens to order their affairs with confidence, that previous decisions of a final appellate court be departed from only in compelling circumstances.

  105. How compelling those circumstances need to be will depend, among other things, on the type of case under consideration and the length of time for which the previous decision has stood. Some cases are intrinsically of a kind upon which reliance is placed by people making choices about how to order their affairs; others are not relied on in the same way. It is a moot point whether cases in the first category should be departed from other than by Parliament which is best placed to make the necessary transitional arrangements. The strength of a previous case may vary depending on whether the decision was unanimous or by a majority; but again the need for stability and certainty in the law must be firmly borne in mind. Some decisions may become out of date in terms of policy or social circumstances and thus more amenable to change provided no serious injustice is involved in the change. All in all the touchstone should be caution, often considerable caution when it comes to suggestions that this Court should depart from one of its own decisions or a decision of the Privy Council. It must usually be evident that the previous decision was or has become clearly wrong, rather than simply representing a preferred choice with which the current bench does not agree.

  106. When a decision of the Privy Council is in issue there may be circumstances in which this Court feels compelled to say that the Privy Council has made a policy choice which was not the right policy choice for New Zealand conditions. A principal reason for the establishment of this Court was to enable New Zealand law to be settled by New Zealand Judges, they being more familiar with New Zealand society, its legal system and its social conditions and aspirations than English and Scottish Judges could ever be.

  107. In the present case this Court is asked to choose between the approach of the New Zealand Court of Appeal and that of the Privy Council. Both decisions were reached by a majority. In the New Zealand Court of Appeal the decision was that of four Judges out of five. In the Privy Council the contrary decision was that of three Judges out of five. The view taken by the majority in the Privy Council became the law by dint of the hierarchical system of our courts and the effect that hierarchy has on the doctrine of precedent.

  108. The subject matter of the present issue is not one in respect of which citizens will have ordered their affairs on the basis of the decision of the Privy Council. The persuasive force of the decision of the Privy Council is in the circumstances reduced because of the strong dissent by two of the five Judges sitting. It is the sort of issue of which it can reasonably be said that the outcome in the Privy Council could well have gone the other way if the Board had been differently constituted. There is also force in the view that the majority of their Lordships may not have appreciated that the relationship between exemplary damages and New Zealand‘s accident compensation scheme means it is particularly desirable that our legal system have a clear and principled line between cases of negligence causing personal injury which justify exemplary damages and those which do not. This feature of our legal system also means that we must keep conceptually as clear as possible the line between punishment and compensation. When all these factors and the more general points made earlier are borne in mind I do not consider this Court would be acting inappropriately in departing from the decision of the majority of the Privy Council in Bottrill if we are otherwise persuaded that it is desirable to do so.

    Submissions on test for exemplary damages

  109. As mentioned above, the Solicitor-General contended that this Court should restore New Zealand law to the position it was in before the Privy Council altered the stance of the Court of Appeal in Bottrill. He made a number of submissions in support of that proposition which I will examine in the course of the following analysis. Mr Henry pointed out that the Privy Council had said that most cases deserving exemplary damages would be cases involving subjective recklessness but room should be left for awards of exemplary damages in rare cases of outrageous negligence where the evidence fell short of demonstrating subjective recklessness. Mr Henry submitted that this was necessary, both as a matter of principle and, in particular, to deal appropriately with cases where the defendant was a government department or other large organisation. In those cases it might be difficult to establish subjective recklessness on the part of any one individual but overall the failings of the body as a whole might be outrageous and thus merit an award of exemplary damages.

  110. I am of the view that the submissions of the Solicitor-General are to be preferred to those of Mr Henry. In short, for the reasons I will develop, I consider exemplary damages should not be available in cases of negligence causing personal injury unless the defendant consciously appreciated the risk the conduct in question posed to the safety of the plaintiff and proceeded deliberately and outrageously to run that risk and thereby caused the harm suffered by the plaintiff.

    Analysis

    Introduction

  111. The first point that supports this conclusion is, as the Solicitor-General submitted, that punishment is generally meted out only to those who have deliberately caused harm or who have deliberately run the risk of doing so, rather than to those who have been inadvertent, even grossly inadvertent. In his book The Common Law Oliver (Little, Brown, and Co, Boston, 1881) Wendell Holmes observed (at 3) that punishment "can hardly go very far beyond the case of a harm intentionally inflicted: even a dog distinguishes between being stumbled over and being kicked". I recognise that Parliament has, over the years, created a number of offences which do not require a guilty mind in the conventional sense. But the anomalous nature of exemplary damages at common law is such that they should be restricted to circumstances which are analogous to the conventional offences which require conscious appreciation of wrongdoing. At least in the present context a conscious appreciation of the risk of causing harm should be a necessary precondition to the infliction of punishment. Otherwise the distinction between crime and tort becomes even further blurred.

  112. An allied point is that punishment should not be inflicted unless the person concerned is able to determine in advance with some certainty when their conduct is liable to punishment. The need for conscious appreciation of risk brings into the inquiry the state of the defendant‘s mind. The presence of the necessary state of mind, as a pre-existing matter of fact, must be shown before punishment by way of exemplary damages should be inflicted. A test for exemplary damages which did not have this ingredient would depend entirely on an after the event assessment by a Judge or jury of whether the relevant conduct should be viewed as outrageous. That is a subjective and inherently uncertain criterion on which minds may well differ. Assessing whether the defendant consciously appreciated the risk inherent in the relevant conduct involves an objective and conceptually certain inquiry into the defendant‘s state of mind which is preferable to a test based on the adjudicator‘s subjective reaction to what has occurred.

  113. It could be said that requiring subjective recklessness gives an advantage to those who fail to appreciate the risks inherent in their conduct. But the answer is that those in that category are less deserving of punishment for the harm they cause than those who cause that harm deliberately or with subjective recklessness. A person who consciously chooses to run the risk of causing harm is more blameworthy than a person who causes harm without choosing to do so. The law should recognise that distinction and use it in the present context as a conceptually sound and clear dividing line.

    The Bottrill judgments

    The Court of Appeal

  114. With those points in mind, I turn to the judgments in Bottrill. In the Court of Appeal four of the five Judges favoured a requirement of subjective recklessness. The fifth Judge, Thomas J, did not. I make the preliminary observation that it was Thomas J who in Bottrill, as he had in Daniels, took an expansive view of the purpose of exemplary damages. I have already discussed this subject in general terms. I consider Thomas J transferred, without sufficient analysis, concepts which are undoubtedly conventional underpinnings of tort law generally into his exemplary damages discussion.[56] Taking appeasement and vindication as examples, it is all too easy, when viewing those factors as purposes rather than consequences of exemplary damages, to slip, unconsciously perhaps, into a mindset which includes as an ingredient in the decision whether to award exemplary damages and, if so, at what level, a need for the defendant indirectly to compensate the plaintiff by means of appeasement and vindication.

  115. This plaintiff-focussed approach is inconsistent with the whole concept of exemplary damages. The focus should be on the defendant, the question being what the defendant should pay, not what the plaintiff should receive. The plaintiff receives vindication and appeasement only indirectly through the punishment of the defendant. The anomaly that exemplary damages go to the plaintiff will be compounded if there is mental slippage from the purpose of the remedy. I therefore respectfully consider Thomas J‘s reasoning proceeded from an insecure starting point.

  116. I should also mention that Thomas J espoused, as a key point in his reasoning, a proposition which was later taken up in the reasoning of the majority of the Privy Council. His Honour said (at [76]) that he could see no reason to restrict what he regarded as the general principle that exemplary damages may be awarded in those exceptional cases where the defendant‘s conduct is so outrageous and contumelious as to deserve condemnation. He did not, however, identify the source of the general principle upon which he was relying to support the view that objective recklessness of an outrageous kind sufficed. His Honour made a similar comment (at [80]) when he said that one of the main respects in which he differed from the majority was that he adhered to the general principle earlier described. I do not consider the correct outcome should be viewed as mandated by any such general principle. In any event Thomas J‘s approach and the similar approach of the majority in the Privy Council tend to beg the very question which arose for determination in the case, namely what the general principle should be. It is only after securely establishing the general principle that the question of whether to make exceptions or adjustments to it can be considered.

  117. The majority (Richardson P, Gault, Blanchard and Tipping JJ) in the Court of Appeal made a number of points in support of the view that subjective recklessness was a necessary ingredient. On reassessment, I still agree with them. In summary they were:

    1. That the object of exemplary damages is to punish;

    2. Where the harm is not intentional, the state of the defendant‘s mind must closely approach intentional infliction of harm in order to justify punishment. Deliberate risk taking does closely approach intentional causing of harm. Causing harm as a result of a high level of negligence does not;

    3. The test should avoid any risk of hidden or indirect compensation;

    4. There should be no indirect subversion of the accident compensation scheme;

    5. The various adjectival descriptions of the qualifying conduct point clearly towards conscious wrongdoing. They do not signal inadvertence, even at a very high level. There is therefore no general principle that high level inadvertent negligence justifies exemplary damages;

    6. From the point of view of moral blameworthiness there is a conceptual analogy in the fact that subjective recklessness is equated with intention in the definition of murder; and

    7. A test for exemplary damages which incorporated high level negligence would result in uncertainty and unpredictability, and would clash with the proposition that punishment should not be inflicted if there is no conscious wrongdoing.

    The Privy Council

  118. I turn now to the judgment of Lord Nicholls on behalf of the majority of the Privy Council (Himself, Lord Hope and Lord Rodger). It will be recalled that their Lordships regarded exemplary damages as available if the negligent conduct of the tortfeasor satisfied the sole criterion of outrageousness. They considered this to be a principle of general application, departures from which had to be justified. But there was no reasoned analysis demonstrating the existence of the so-called general principle. As I have already said, the assumption of such a principle begs the question of what the general principle should be. The assumption of this general principle represented both the essential underpinning of the reasoning of the majority and, if I may respectfully say so, its fundamental weakness. The general rule was said to spring from the rationale of exemplary damages; but the rationale for including high level objective recklessness was not discussed. Their Lordships said (at [22]):

    In principle the limits of the Court‘s jurisdiction to award exemplary damages can be expected to be coextensive with this broad-based rationale. The Court‘s discretionary jurisdiction may be expected to extend to all cases of tortious wrongdoing where the defendant‘s conduct satisfies this criterion of outrageousness. Any departure from this principle needs to be justified. Otherwise the law lacks coherence. It could not be right that certain types of outrageous conduct as described above should attract the Court‘s jurisdiction to award exemplary damages and other types of conduct, satisfying the same test of outrageousness, should not, unless there exists between these types a rational distinction sufficient to justify such a significant difference in treatment.
  119. The law lacks coherence only if the underlying rationale supports the inclusion of objective recklessness, and it is not included. But it is far from clear that the underlying punitive rationale does support the inclusion of objective recklessness. Their Lordships in the minority did not think so.

  120. Earlier (at [20]), under the heading of principle, the Privy Council majority had said that exemplary damages were appropriate where something more was required than compensatory damages, even if they included an element of compensation for aggravation. That "something more" was to demonstrate that the conduct in question was "altogether unacceptable to society". In that situation the wrongdoer could be ordered to make a further payment by way of condemnation and punishment. The fundamental problem with this reasoning is the correlation between punishment and conduct which is altogether unacceptable to society. For reasons earlier discussed that allows punishment on an after the event appraisal of the quality of conduct, which is an inherently imprecise and uncertain exercise.[57]

  121. Their Lordships next (at [26]) deployed the "never say never" argument. They observed that it was wrong to conclude that negligent conduct, in the absence of intentional wrongdoing or conscious recklessness, would never give rise to "a justifiable feeling of outrage". But that again begs the question whether a justified feeling of outrage should be the sole criterion. There are times when a principled approach to a legal problem does require the law to say never. To take the opposite view risks elevating uncertain judicial assessment over clarity and principle.

  122. When addressing the argument that to include objective recklessness would create imprecision and uncertainty, their Lordships said, and this is clearly correct, that differences of degree in negligence cases are real. But they then added that courts are "well able" to identify the presence or absence of "that something extra" which turns a case of grossly negligent conduct into conduct which is quite outrageous. I do not share that confidence and respectfully consider that the "something extra", unless it is the conscious risk taking element, is likely to be inherently elusive, for want of both conceptual and practical clarity.

  123. After reference (at [36]) to my refinement of the test in McLaren Transport Ltd v Somerville [1996] 3 NZLR 424 (HC), their Lordships said (at [37]):

    As a matter of principle, the Court‘s ability to award exemplary damages does not turn on niceties such as this. As Thomas J noted at pages 657-658, the necessity to observe the distinction between advertent and inadvertent conduct will distract Courts from making a decision in accordance with the fundamental rationale of exemplary damages. If, having heard all the evidence, a Judge firmly believes the case is so truly exceptional and outrageous that an award of exemplary damages is called for, his power to make an award is not dependent upon his being able conscientiously to find that the defendant was subjectively reckless. The absence of intentional wrongdoing and conscious recklessness will always point strongly away from the case being apt for an award of exemplary damages. That is a very important factor to be taken into account by the Judge. But if the Judge decides that, although the case is not one of intentional wrongdoing or conscious recklessness, the defendant‘s conduct satisfies the outrageous test and condemnation is called for, in principle the Judge has the same power to award exemplary damages as in any other case satisfying this test.

  124. But the distraction arises only if the premise upon which it is based is sound; and for reasons already given the premise is not sound. The need for advertence to risk is consistent rather than inconsistent with the rationale for exemplary damages. In any event the so-called nicety, namely the difference between direct evidence of advertence and inferring advertence from all the circumstances, has nothing to do with the principle involved; it is simply an evidentiary matter. The difference can hardly be described as a nicety, but, in any event, it is not a nicety affecting what principles should guide the court‘s ability to award exemplary damages. The fundamental point of principle is the need for proof of advertent risk taking, not how that proof is derived.

  125. Their Lordships‘ conclusion on matters of principle was expressed in this way (at [40]):

    Their Lordships are of the view that, considered as a matter of legal principle, the arguments against restricting the jurisdiction to cases of intentional or consciously reckless conduct are to be preferred. The fundamental flaw in such a rigid limitation is that it fails to treat like cases alike. For the purposes of exemplary damages the basic question is always whether the defendant‘s conduct satisfies the outrageous conduct criterion. The suggested rigid limitation treats some cases satisfying this criterion in a different way: if the harm was done intentionally or resulted from conscious recklessness, exemplary damages may be awarded. It treats other cases satisfying the same criterion in another way: in the absence of intentional wrongdoing or conscious recklessness, exemplary damages can never be awarded, whatever the circumstances. As a matter of principle, this is not a sound distinction.
  126. For reasons already given, there are difficulties with this reasoning and they spill over into their Lordships‘ discussion of why they should not defer to a policy choice made for New Zealand by New Zealand Judges. After saying that they would ordinarily do so (at [55]), their Lordships observed (at [56]):

    There is a complication in the present case. The Court of Appeal evaluated the policy considerations on the (mistaken) basis that considerations of principle require the advertent conduct only limitation. The correct approach is to evaluate the policy considerations against the background that, far from according with principle, this limitation would represent a departure from basic principle. The Court of Appeal did not carry out this exercise.

  127. It was in the context of the existence in New Zealand of the accident compensation scheme that the majority said they would ordinarily defer to the views of New Zealand Judges but because of the "complication" which they perceived in the present case, they felt unable to do so. The complication was, with respect, of their Lordships‘ own making. Once it is removed from the equation the difficulty perceived by their Lordships in the Court of Appeal‘s policy approach is itself removed.

  128. I end this analysis of the majority‘s reasoning by observing that their reasons written by Lord Nicholls were built substantially on his Lordship‘s own speech in Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29, [2002] 2 AC 122. There are distinct conceptual and linguistic parallels between the two. Kuddus was decided about 15 months before Bottrill. What is striking is that an important point made by Lord Nicholls in Kuddus was omitted from the reasons of the majority in Bottrill. What his Lordship said in Kuddus (at [63]) was this:

    From time to time cases do arise where awards of compensatory damages are perceived as inadequate to achieve a just result between the parties. The nature of the defendant's conduct calls for a further response from the courts. On occasion conscious wrongdoing by a defendant is so outrageous, his disregard of the plaintiff's rights so contumelious, that something more is needed to show that the law will not tolerate such behaviour. Without an award of exemplary damages, justice will not have been done. Exemplary damages, as a remedy of last resort, fill what otherwise would be a regrettable lacuna.

    [emphasis added]

  129. I have emphasised his Lordship‘s reference to "conscious wrongdoing". It was the feature of conscious wrongdoing and disregard of the plaintiff‘s rights[58] that Lord Nicholls considered could be so outrageous or contumelious as to demand more than compensatory damages. But two paragraphs later in Kuddus his Lordship said (at [65]):

    Stated in its broadest form, the relevant principle is tolerably clear: the availability of exemplary damages should be co-extensive with its rationale. As already indicated, the underlying rationale lies in the sense of outrage which a defendant's conduct sometimes evokes, a sense which is not always assuaged fully by a compensatory award of damages, even when the damages are increased to reflect emotional distress.

  130. The problem with this statement, which the majority adopted in Bottrill, is that it is inconsistent with the earlier paragraph. The phrase "as already indicated" does not reflect the earlier paragraph in which the feature of conscious wrongdoing and disregard of the plaintiff‘s rights was a central element. That necessarily involves a focus on the state of mind of the defendant. This is by no means the same as the defendant‘s conduct evoking a sense of outrage. That looks to the state of mind of the adjudicator. This again indicates the insecurity of the premise on which the reasons of the majority in Bottrill were constructed.

  131. That brings me to the dissenting judgment of Lord Hutton and Lord Millett in which they supported the conclusion of the Court of Appeal. They considered (at [76]), in agreement with the Court of Appeal, that if the primary purpose of exemplary damages is to punish, punishment should not be imposed unless the defendant intended to cause harm to the plaintiff or was subjectively reckless as to whether his conduct would cause harm. It would be contrary to well-established principles to punish someone who did not have a guilty mind, "no matter how gross his negligence". They made the useful practical point that in any event a court may well infer subjective recklessness where it finds gross negligence.

  132. A little later their Lordships said (at [77]) that the rationale of exemplary damages was not to mark the court‘s disapproval of outrageous conduct, but rather to punish the defendant for outrageous conduct. This rationale did not apply if the defendant had no intention to harm or was not subjectively reckless.

  133. Their Lordships (at [78]) drew on the report of the English Law Commission which had supported the need for subjective recklessness. They (at [80]) also drew on the American Law Institute‘s suggestion that exemplary damages are inappropriate in the absence of evil motive or reckless indifference to the rights of others. Their Lordships saw considerable force in the concerns of the Court of Appeal in relation to uncertainty and unpredictability. They emphasised that outrage is a subjective concept upon which minds can easily differ. While subjective recklessness will unavoidably give rise to some uncertainty, objective recklessness will give rise to much greater uncertainty. Their Lordships concluded that the Court of Appeal had been right to hold, for reasons of policy and principle, that awards of exemplary damages should be confined to cases of deliberate infliction of harm or harm inflicted with subjective recklessness as to consequences.

  134. It is immediately apparent from their Lordships‘ approach that they did not subscribe to the view that there was some general principle from which the Court of Appeal had carved out an exception. They did not fall into the error of assuming what they were setting out to establish. With respect, their reasoning is sounder than that of the majority.

  135. Any fundamental review of the law‘s approach to exemplary damages should be examined as one of principle and policy without any a priori assumption of a general principle one way or the other. The more is this so in New Zealand where the demarcation between compensation and punishment is so fundamental because of the accident compensation regime. There is an ever-present danger, if exemplary damages were available for outrageous but inadvertent negligence, that they would be perceived as a means of obtaining extra compensation outside the accident compensation system when the negligence was bad enough. The fact that the damages go to the plaintiff would encourage that perspective and would also encourage claims for exemplary damages where the punitive rationale was less than persuasive.

    A little more history

  136. I propose now to examine the key decisions of the House of Lords in respect of exemplary damages over the last 50 years, namely Rookes v Barnard [1964] AC 1129 (HL), Broome v Cassell & Co Ltd [1972] AC 1027 (HL) and Kuddus. My purpose is to see how they fit with the approach taken by the majority of the Privy Council in Bottrill. The majority was of the view that its approach represented the mainstream of principle as generally understood in English law, and the Court of Appeal had not appreciated that mainstream or had departed from it for insufficient reasons.

  137. Rookes v Barnard marked a major rationalisation of the law of exemplary damages in England. Lord Devlin delivered the only speech on this aspect of the case. The other Judges sitting (Lord Reid, Lord Evershed, Lord Hodson and Lord Pearce) all expressed their concurrence with Lord Devlin‘s speech. In short, the House decided, after a detailed review of English case law going back 200 years, that, for the future, exemplary damages were to be available in three classes of case only:

    1. where government servants had acted in an oppressive, arbitrary or unconstitutional manner;

    2. where a defendant‘s wrongful conduct had been deliberately calculated to make a profit exceeding any compensation payable to the plaintiff; and

    3. where authorised by statute.

    This restrictive approach was based in no small part on the view that exemplary damages were an anomaly which could not properly be abolished by the House but which could and should be kept within narrow bounds.

  138. Lord Devlin (at 1221) expressed the view that exemplary damages could well be thought to confuse the civil and criminal functions of the law. His Lordship also described the exemplary principle (at 1226) as one which ought logically to belong to the criminal law. He reiterated the anomalous nature of exemplary damages (at 1227) and recorded that Judges had used numerous epithets down the years to describe the sort of conduct which would attract exemplary damages (at 1229). Included were the adjectives wilful, wanton, high-handed, oppressive, malicious and outrageous. His Lordship observed that these descriptions had been used in judgments largely by way of comment on the facts of particular cases. He then added an important point about these descriptive words, ibid:

    It would, on any view, be a mistake to suppose that any of them can be selected as definitive, and a jury directed, for example, that it can award exemplary damages whenever it finds conduct that is wilful or wanton.

  139. I turn now to Broome v Cassell. Lord Reid‘s speech in that case is of particular significance because he alone, of the seven Judges who sat in Broome, had also sat in Rookes v Barnard. His Lordship said (at 1086):

    On the other hand when we came [in Rookes v Barnard] to examine the old cases we found a number which could not be explained in that way. The sums awarded as damages were more — sometimes much more — than could on any view be justified as compensatory, and courts, perhaps without fully realising what they were doing, appeared to have permitted damages to be measured not by what the plaintiff was fairly entitled to receive but by what the defendant ought to be made to pay as punishment for his outrageous conduct.

    That meant that the plaintiff, by being given more than on any view could be justified as compensation, was being given a pure and undeserved windfall at the expense of the defendant, and that in so far as the defendant was being required to pay more than could possibly be regarded as compensation he was being subjected to pure punishment.

    I thought and still think that that is highly anomalous. It is confusing the function of the civil law which is to compensate with the function of the criminal law which is to inflict deterrent and punitive penalties. Some objection has been taken to the use of the word "fine" to denote the amount by which punitive or exemplary damages exceed anything justly due to the plaintiff. In my view the word "fine" is an entirely accurate description of that part of any award which goes beyond anything justly due to the plaintiff and is purely punitive.

    Those of us who sat in Rookes v Barnard thought that the loose and confused use of words like punitive and exemplary and the failure to recognise the difference between damages which are compensatory and damages which go beyond that and are purely punitive had led to serious abuses, so we took what we thought was the best course open to us to limit those abuses.

    Theoretically we might have held that as purely punitive damages had never been sanctioned by any decision of this House (as to which I shall say more later) there was no right under English law to award them. But that would have been going beyond the proper function of this House. There are many well-established doctrines of the law which have not been the subject of any decision by this House. We thought we had to recognise that it had become an established custom in certain classes of case to permit awards of damages which could not be justified as compensatory, and that that must remain the law. But we thought and I still think it well within the province of this House to say that that undesirable anomaly should not be permitted in any class of case where its use was not covered by authority.

    In order to determine the classes of case in which this anomaly had become established it was of little use to look merely at the words which had been used by judges because, as I have said, words like punitive and- exemplary were often used with regard to damages which were truly compensatory. We had to take a broad view of the whole circumstances.

  140. A little later Lord Reid continued his speech in the following way (at 1087):

    We were confronted with an undesirable anomaly. We could not abolish it. We had to choose between confining it strictly to classes of cases where it was firmly established, although that produced an illogical result, or permitting it to be extended so as to produce a logical result. In my view it is better in such cases to be content with an illogical result than to allow any extension.

    It will be seen that I do not agree with Lord Devlin‘s view that in certain classes of case exemplary damages serve a useful purpose in vindicating the strength of the law. That view did not form an essential step in his argument. Concurrence with the speech of a colleague does not mean acceptance of every word which he has said. If it did there would be far fewer concurrences than there are. So I did not regard disagreement on this side issue as preventing me from giving my concurrence.

    I think that the objections to allowing juries to go beyond compensatory damages are overwhelming. To allow pure punishment in this way contravenes almost every principle which has been evolved for the protection of offenders. There is no definition of the offence except that the conduct punished must be oppressive, high-handed, malicious, wanton or its like — terms far too vague to be admitted to any criminal code worthy of the name. There is no limit to the punishment except that it must not be unreasonable. The punishment is not inflicted by a judge who has experience and at least tries not to be influenced by emotion: it is inflicted by a jury without experience of law or punishment and often swayed by considerations which every judge would put out of his mind.

  141. Lord Diplock (at 1127–1128) expressed views similar to those of Lord Reid. He described (at 1129) as obsolete and properly discarded a category of cases in which exemplary damages were based on conduct which could be described as wilful, wanton, high-handed, oppressive, malicious or outrageous: a whole gamut of dyslogistic judicial epithets, as he famously called them. The role of this sort of conduct should be confined to when it served to aggravate compensatory damages. His Lordship was firmly opposed to restoring to English law "the anomaly of awarding exemplary damages" simply on the premise that the defendant‘s conduct might attract one or more of the epithets previously listed, including of course the epithet 'outrageous‘ (at 1130).

  142. Other members of the House sitting in Broome v Cassell were not quite so trenchant in their views but all seemed to be of similar mind about exemplary damages save Lord Wilberforce who expressed the opinion (at 1114):

    It cannot lightly be taken for granted, even as a matter of theory, that the purpose of the law of tort is compensation, still less that it ought to be, an issue of large social import, or that there is something inappropriate or illogical or anomalous (a question-begging word) in including a punitive element in civil damages, or, conversely, that the criminal law, rather than the civil law, is in these cases the better instrument for conveying social disapproval, or for redressing a wrong to the social fabric, or that damages in any case can be broken down into the two separate elements. As a matter of practice English law has not committed itself to any of these theories: it may have been wiser than it knew.

  143. It is perhaps of some significance that, while Lord Wilberforce‘s views are always entitled to the highest respect, he alone of the eleven Judges who sat in Rookes v Barnard and Broome v Cassell considered that a more expansive role for exemplary damages might be appropriate.

  144. I come now to Kuddus. I have already mentioned the inconsistent approach taken by Lord Nicholls to consciousness of wrongdoing or risk. That said, the House confirmed Rookes v Barnard, albeit with the rider that the categories should be viewed as factual rather than based on causes of action already recognised as at 1964. Lord Slynn made the point (at [22]) that it was not until Rookes v Barnard that the distinction between so-called aggravated damages and exemplary damages was clearly articulated. Lord Mackay reiterated the anomalous nature of exemplary damages (at [42]) and so did Lord Hutton (at [73]). Lord Scott made the same point (at [95]) in the course of a speech which expounded the view that there was no longer any place for exemplary damages in the civil law – "victims of tortious conduct should receive due compensation for their injuries, not windfalls at public expense" (at [121]). That approach has support from the view now widely held internationally that an effective remedy in cases where human rights have been breached should seldom, if ever, include a punitive element.[59]

  145. What emerges from this review of these three leading decisions in England is that there is no support for and indeed substantial opposition to the view that exemplary damages may be awarded simply on the basis that the defendant‘s conduct can be described as outrageous. It is, with respect, difficult to discern the source of the conclusion reached by the majority of the Privy Council in Bottrill that the New Zealand Court of Appeal, by insisting on conscious appreciation of risk, had not appreciated a well-established principle of the common law and had thereby departed from it for no sufficient reason. The fact that New Zealand law has not followed Rookes v Barnard‘s categorisation approach to exemplary damages does not diminish the force of the rejection of outrageous conduct as a sufficient test in Rookes and the cases which followed it.

    An academic viewpoint

  146. I draw support for the conclusion I favour from New Zealand‘s leading tort law scholar, Professor Stephen Todd. Writing in 1998 (Stephen Todd "Exemplary Damages" (1998) 18 NZULR 145 at 160–161), after the decision in McLaren but before the Bottrill litigation, Professor Todd said:

    Serious negligence which nonetheless is unthinking or inadvertent, especially where it causes grave injury, might be thought to be included [within the compass of exemplary damages], yet it is strongly arguable that exemplary damages are not appropriate. Seemingly in this kind of case any element of deterrence is minimal or non-existent. It is difficult to see how defendants can be deterred from making a mistake. And once mere inadvertence, however gross, is admitted as the basis for an award it is hard to identify any particular point at which to draw the line. Certainly there is an obvious risk of a multiplicity of actions. Thus emphasis needs to be put on the element of deliberate or conscious taking of risks.

  147. After the Bottrill litigation, Professor Todd wrote another article in which he observed that the majority in the Court of Appeal and the minority in the Privy Council had appropriately put the focus squarely on the defendant‘s conduct: Stephen Todd "A New Zealand Perspective on Exemplary Damages" (2004) 33 CLWR 255 at 264. The plaintiff should be required to show intentional or consciously reckless misconduct. The Professor referred in support of this proposition to the decision of the High Court of Australia in Gray v Motor Accident Commission [1998] HCA 70, (1998) 196 CLR 1 at [22], which I will discuss later, and said that there were a number of reasons for preferring the view of the majority in the Court of Appeal to that of the majority in the Privy Council which ultimately prevailed. He too made the point that the reasoning of the majority in the Privy Council seemed to suffer from assuming what they were seeking to prove.

  148. In this article Professor Todd also expressed concern about the elusive nature of the factors needed to identify what the majority of the Privy Council called the "rare and exceptional" case where exemplary damages might be awarded for high level negligence. He added that one consideration which permeated much of the debate, without its true significance being clearly articulated, was the nature of the consequences to the plaintiff of the conduct in question. He expressed the valid view that in principle the fact that the consequences are very serious cannot in itself be relevant. Both gross and trifling negligence may each have either grave or minor effects, or may cause no damage at all. He added that the potential for conduct to have exceptionally grave consequences could have a bearing on the defendant‘s culpability but actual consequences (perhaps with the benefit of hindsight) have, at most, marginal significance.

  149. Professor Todd also made a point which I have already mentioned, namely that outrageousness as the governing test lacks objective content. By contrast, advertence or conscious recklessness as to consequences raise familiar and well understood questions. Such tests are conceptually coherent. They focus on the conduct and state of mind of the defendant. Obviously, as with any test, there can be difficulties about application and the problem of a degree of uncertainty is unavoidable but, in Professor Todd‘s view, which I share, the problem is exacerbated if outrageousness is the sole criterion.

    Conclusion

  150. It follows that I agree with the Crown‘s submission that the concept of outrageousness is unsatisfactory as the sole criterion. It is far better to have a test which focuses objectively on the state of mind of the defendant as a precursor to any subjective assessment by the Judge or jury of whether the defendant‘s conduct was outrageous. This focus on the state of the defendant‘s mind also helps to underline that the consequences of the defendant‘s conduct are not the primary question. The more important question is the level of risk to the plaintiff‘s safety which the defendant consciously ran and the level of harm which the defendant appreciated would eventuate if the risk he was consciously running became a reality. This is a more principled basis for deciding whether exemplary damages should be awarded than the uncertain and amorphous concept of the defendant‘s conduct being outrageous.

  151. I am not saying that outrageousness should be excluded from the inquiry. What I am saying is that the inherently elusive concept of outrageousness should not be the sole criterion. It has a part to play in conjunction with the need for subjective recklessness. Not all states of subjective recklessness can fairly be characterised as outrageous. It may not be outrageous deliberately to run a small risk of causing harm when the probable consequences of the risk maturing are not great. On the other hand a person who deliberately runs a high risk of causing harm, which is likely to be substantial, may well be said to have acted outrageously. This conjunction of level of risk and level of likely harm gives a desirably objective aspect to the concept of outrageousness. It also helps to focus the inquiry on the defendant‘s conduct and state of mind, rather than the actual consequences of that conduct for the plaintiff. While actual consequences are certainly relevant in conventional culpability terms, they are not, at least in present circumstances, the principal ingredient in assessing blameworthiness.

  152. The Chief Justice has expressed concern that the conclusion I favour introduces a "cause of action" condition for exemplary damages and involves subcategories of the tort of negligence in a way which creates different "species" of negligence. I do not regard the distinction between the kind of negligence which qualifies for exemplary damages (subjective recklessness) and that which does not (objective recklessness) as creating a condition as to the necessary cause of action (negligence). Both states of mind involve negligence, but of different qualities.

  153. I agree it is possible to describe the qualifying kind of negligence as a species or subcategory of the tort; but that description is no more than a facet of distinguishing between the kind of inadvertence that should qualify for exemplary damages and that which should not. The objection is the same as saying that outrageous negligence is a species of negligence or a subcategory of the tort. That objection could just as well be levelled at the conclusion favoured by the majority of the Privy Council.

  154. The Chief Justice also relies substantially on the decision of the Court of Appeal in the defamation case of Taylor v Beere [1982] 1 NZLR 81 (CA). She suggests in particular that all three judges in that case rejected the proposition that exemplary damages are anomalous. But Cooke J (at 84) simply acknowledged that in defamation cases the defendant‘s conduct before action, after action and at trial can be relevant to damages generally. He added that punishment for malice or high handed conduct was historically regarded as a legitimate purpose of libel damages. He did not engage with the wider question whether exemplary damages in general are anomalous.

  155. Richardson J (at 90) recorded that the soundness of the premise that punishment is the exclusive concern of the criminal law was a matter of dispute. He gave the impression that he himself did not regard the premise as sound. But it was Richardson J who, nearly twenty years later, wrote the leading judgment for the majority of the Court of Appeal in Bottrill favouring the requirement of subjective recklessness.

  156. The third judge in Taylor v Beere, Somers J (at 95), rehearsed the competing views about whether exemplary damages are anomalous without expressing any conclusion on the point. But, on a related point, in his judgment (at 117) in the companion case of Donselaar, Somers J significantly observed that without some additional feature as, for example, abuse of power or the invasion of other rights of the plaintiff - meaning in context intentional invasion - it was not easy to envisage a case of personal injury which would not be adequately met by compensatory or aggravated compensatory damages, which are of course barred by the Accident Compensation Act. Somers J thereby implied that there was a case for confining exemplary damages to intentional wrongdoing, or, I would add, the moral equivalent of deliberate risk taking.

    The corporate issue

  157. Mr Henry argued that requiring subjective recklessness would make it too difficult for plaintiffs to obtain exemplary damages from large organisations such as a department of state or a major corporation. In such instances a number of different people may have combined to produce an outrageous outcome but no one individual could be shown to have been subjectively reckless as to that outcome. This was the most cogent of Mr Henry‘s submissions but I do not consider it is of sufficient force to outweigh the combined strength of the arguments which support the necessity for subjective recklessness. Indeed I consider the suggested difficulty is more theoretical than real.

  158. The law recognises two ways in which one person may be liable for the conduct of another. The first is by means of vicarious liability; the second is by attribution. Vicarious liability is imposed largely as a matter of policy. When one person is acting on behalf of another, the wrong is seen for the purposes of compensation as having been committed both by that person and by the other. But the policy behind awarding compensatory damages against a person on the basis of vicarious liability does not mean that exemplary damages should be similarly awarded. There is no policy basis for punishing someone for the conduct of another unless the first person‘s conduct itself also qualifies for punishment. People should not be vicariously liable for punishment on account of the conduct of someone else just because that conduct renders them liable to compensate the plaintiff.

  159. The concept of attribution operates in the case of a corporate body having a separate legal existence from its members and employees. A body such as a department of state or a corporation can act only through human beings. When a human being acts on behalf of the corporation or department their conduct may be attributed to that body. Their conduct and their state of mind becomes that of the corporation or department itself. They act as rather than for the corporation or department: See Giltrap City Ltd v Commerce Commission [2004] 1 NZLR 608 (CA) at [52]; and Tesco Supermarkets Ltd v Nattrass [1972] AC 153 at 170 per Lord Reid (HL).

  160. When one applies these concepts to the present issue there is no doctrinal difficulty in attributing the subjective recklessness of one person to another person or body. What may pose some difficulty is when no one person can be said to have been subjectively reckless but, after adding the knowledge and attitudes of two or more persons together, the organisation as a whole is said to have been subjectively reckless. That represents something of a construct, and there may be jurisprudential difficulty in combining more than one mind and ascribing the combination to the body as a whole. There may also be some evidential challenges in the process. These issues, upon which it is unnecessary to dwell, are not however sufficient to overtake the force of all the other factors which support the need for subjective recklessness.

  161. In some cases where the conduct of a multi-person organisation is in issue the general systems in operation in the organisation, its resourcing and its perception of priorities are likely to have a bearing on the question whether the organisation itself has been subjectively reckless. The present case is said to involve so-called systemic negligence of this kind. In the case of governmental organisations there is a danger that such cases may become an examination of general governmental policies, priorities and funding decisions into which the legal system should be very reluctant to go.[60] The remedy for issues of this kind is generally political rather than legal. Nevertheless an issue may remain whether, despite the constraints imposed upon the organisation, its management was subjectively reckless as regards a particular person‘s safety or welfare. In that situation the focus will be on those who control the use and allocation of the organisation‘s resources. But, again, any difficulties in this respect are not such as to justify outrageousness as the sole criterion against the weight of the factors which suggest otherwise. If an inference of subjective recklessness cannot properly be drawn from all the objective circumstances, the case should not qualify for exemplary damages. Society‘s punitive instincts should be restrained by principle. It is part of the function of the law to achieve that outcome.

    Comparable jurisdictions

  162. The preponderance of the authorities and other legal writings in England and Wales, and in Australia and Canada support the view that conscious appreciation of wrongdoing is a necessary ingredient in the criteria for awarding exemplary damages. Subjective recklessness is regarded for this purpose as involving conscious appreciation of wrongdoing. It is equated with intentionally doing something known to be wrong.

    England

  163. The position in England and Wales is the same for present purposes as it was when Bottrill was decided. The only development has been the decision of the House of Lords in Kuddus which rejected the so-called existing causes of action criterion said to have been established by Rookes v Barnard. That, however, has no bearing on the present issue. A key feature of the legal landscape in England and Wales therefore remains the Report of the Law Commission for England and Wales published in 1997 (Report on Aggravated, Exemplary and Restitutionary Damages (LC247, 1997)). The Commission recommended that, in respect of claims in negligence, exemplary damages be limited to cases of subjective recklessness. No action has been taken in England on the Commission‘s report. The report was, however, mentioned in Kuddus without any apparent disagreement on this point (at [35] per Lord Mackay and at [61] per Lord Nicholls).

    Australia

  164. The general position in Australia supports the need for subjective recklessness. It does so, at least indirectly, by reason of the general test of contumelious disregard of the plaintiff‘s rights (see Lamb v Cotogno (1987) 164 CLR 1). The most significant authorities in Australia are still the decisions of the High Court in Whitfeld v De Lauret & Co Ltd (1920) 29 CLR 71[61] and Gray v Motor Accident Commission. In their joint judgment in Gray four members of the Court (i.e. Gleeson CJ, McHugh, Gummow and Hayne JJ) said that exemplary damages could not properly be awarded in a negligence case if there was "no conscious wrongdoing by the defendant" (at [22]). Their Honours went on to say that ordinarily questions of exemplary damages would not arise in negligence cases, but there could be cases "framed in negligence" in which the defendant might have acted in contumelious disregard for the plaintiff‘s rights. They thereby equated contumelious disregard with conscious wrongdoing in the form of deliberately running a known risk of causing harm to the plaintiff. Subsequent cases in Australia have applied this principle as laid down by the High Court.[62]

  165. Recently, however, a line of cases has developed in New South Wales based on an apparently contradictory sentence in the joint judgment in Gray whereby their Honours said (at [14]) that conscious wrongdoing in contumelious disregard of the plaintiff‘s rights described "at least the greater part of the relevant field". In another passage in the joint judgment (at [20]) their Honours said that the remedy of exemplary damages was exceptional as it arose "chiefly if not exclusively" in cases of contumelious disregard. I consider, with respect, that their Honours were correct in their first statement when they said that conscious wrongdoing was a necessary ingredient. The view represented by this more recent line of authority builds also on Kirby J‘s dissenting judgment in Gray. The most significant case is State of New South Wales v Ibbett [2005] NSWCA 445, (2005) 65 NSWLR 168 which involved a claim against police officers for assault and trespass.[63] Hence the issue of the test for negligence did not strictly arise.

  166. Ibbett was appealed to the High Court of Australia (see New South Wales v Ibbett [2006] HCA 57, (2006) 229 CLR 638)). In a unanimous judgment the High Court (at [47]) in reference to Gray‘s case said "[m]oreover, in this Court, it has been said that there may be cases, framed in negligence, in which the defendant can be shown to have acted consciously in contumelious disregard of the rights of the plaintiff or persons in the position of the plaintiff". Significantly the Court also implied disagreement with the decision of the Privy Council in Bottrill by saying, after its reference to Gray in a footnote (at footnote 56), "cf A v Bottrill [2003] 1 AC 449 at 463-464". This seems to confirm that in Australia conscious appreciation of wrongdoing, ie subjective recklessness, is a necessary ingredient in cases framed in negligence.

    Canada

  167. In Canada an award of exemplary damages in negligence cases would seem most unlikely unless there has been conscious wrongdoing by the defendant. In recent decisions the Supreme Court of Canada has said that punitive damages will only be awarded in exceptional cases, for "malicious, oppressive and high-handed" misconduct that offends the court‘s sense of decency: Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 and Fidler v Sun Life Assurance Co of Canada 2006 SCC 30, [2006] 2 SCR 3. In Whiten v Pilot Insurance Co[64] the Court set out a number of factors which go toward assessing the blameworthiness of the defendant, including whether the defendant‘s misconduct was planned or deliberate, and what the intent and motive of the defendant was. That case did not concern negligence, but the inclusion of these intentional elements in a breach of contract case suggests they would be applied to negligence cases as well.

  168. A recent general summary of the law relating to exemplary damages for negligence in Canada is to be found in the decision of the Ontario Court of Appeal in McIntyre v Grigg (2006) 274 DLR (4th) 28 (ONCA). In that case the majority of the Court said that negligent conduct could only attract punitive damages if it was "intentional and deliberate". The Court suggested (at [62]) that the requirement for deliberate action or omission "is clear from the relatively few cases in which courts have awarded punitive damages for negligent conduct". The conjunction of intent and deliberation with negligence is not at first sight an easy one; but the context suggests that the Judges must have been focusing on the concept of intentional and deliberate disregard for the rights of the plaintiff. That is the equivalent of subjective recklessness in the sense of intentionally and deliberately running a known risk that your conduct will harm the plaintiff.

  169. Waddams (The Law of Damages (4th ed, Canada Law Book, Toronto, 2004) at [11.219]) states that generally exemplary damages are not awarded for negligence. The author adds "but where the defendant deliberately exposes the plaintiff to a risk without justification – which can be said to amount to recklessness – some Courts have awarded exemplary damages". Waddams‘ use of the concept of recklessness clearly amounts, in context, to what I have been calling subjective recklessness.

  170. There are earlier views the other way, which suggest that the absence of intent to injure goes to mitigation of the award of damages, and not to prevent exemplary damages altogether: Klein v Jenoves [1932] 3 DLR 571 (ONCA) per Masten JA. However, this is not the modern approach. In an overview of the Canadian approach to exemplary damages, Feldthusen says that punitive damages are available in tort "provided the defendant‘s actions were advertently wrongful and met a threshold of exceptionally objectionable conduct" ("Punitive Damages: Hard Choices and High Stakes" [1998] NZ Law Review 741 at 748). The concept of conduct being advertently wrongful in Feldthusen‘s terminology denotes conscious awareness on the part of the defendant that the conduct was wrong. Feldthusen disapproves of cases which have held that reckless conduct or outrageous negligence, short of advertent wrongdoing, are sufficient to award exemplary damages. His objection, which I share, is that this standard is the practical equivalent of gross negligence, which is notoriously difficult to define or apply consistently. Punishment should be predicated on a clear legal standard.

  171. It is therefore clear that, in adopting a requirement for subjective recklessness, this Court is not taking New Zealand law out of line with the law in the three countries I have surveyed. We are, if anything, bringing New Zealand law back into line with that of Australia and Canada and into line with the conceptual underpinning of the three House of Lords cases and the recommendations of the English Law Commission, which reached its conclusions after a very thorough and persuasive examination of the whole issue.

    Section 86 of the State Sector Act 1988

  172. The Solicitor-General argued, as a separate and subsidiary point, that Ms Couch‘s claim should be struck out by reason of the immunity granted by s 86 of the State Sector Act 1988, in combination with s 6(1) of the Crown Proceedings Act 1950. Sections 86 and 6(1) provide as follows:

    86.

    Protection from liability

    No chief executive, or .... employee, shall be personally liable for any liability of the Department, or for any act done or omitted by the Department or by the chief executive or any .... employee of the Department or of the chief executive in good faith in pursuance or intended pursuance of the functions or powers of the Department or of the chief executive.

    6.

    Liability of the Crown in tort

    (1)

    Subject to the provisions of this Act and any other Act, the Crown shall be subject to all those liabilities in tort to which, if it were a private person of full age and capacity, it would be subject—

    (a)

    In respect of torts committed by its servants or agents;

    ....

    Provided that no proceedings shall lie against the Crown by virtue of paragraph (a) of this subsection in respect of any act or omission of a servant or agent of the Crown unless the act or omission would apart from the provisions of this Act have given rise to a cause of action in tort against that servant or agent or his estate.

  173. The scope and purpose of s 6(1) is to make the Crown liable for torts committed by its servants or agents in the same way as any other person of full age and capacity. The same policy is to be found in s 27(3) of the New Zealand Bill of Rights Act 1990. The Crown liability recognised by s 6 is subject to the reservations expressed in the section. One of those reservations is that the Crown is not liable unless the servant or agent is liable independently of the Act. This, it has been said, suggests that the Crown can be liable vicariously but not by attribution.[65] The Crown‘s proposition is that s 86 removes the liability of the servant or agent, and hence the Crown is not liable by virtue of the proviso to s 6(1). That is not, however, the effect of s 86. It removes the capacity for a chief executive or employee to be liable on a secondary basis. It cannot be construed as removing the capacity of the chief executive or employee to be liable in a primary way. The chief executive or employee may well be entitled to be indemnified by the department, but that does not remove their primary liability.

  174. Section 86 is not well worded because its purpose must have been to render chief executives and employees immune from any liability of the department or any other employee. In other words, it prevents the department from seeking indemnity from its chief executive or employee for a liability of the department on account of the good faith acts or omissions of the chief executive or employee. Semantically the section is capable of being read as providing that no chief executive or employee shall be personally liable to the plaintiff for a good faith act or omission. That however would be a most unlikely construction to place on a provision, the context of which does not suggest this external kind of immunity. The reading which accords better with the purpose of the provision in its context is an internal one, namely that the department cannot recover from its chief executive or employee provided they have acted in good faith, despite the fact that it is their conduct which has rendered the department liable. Read in that way s 86 does not prevent the act or omission spoken of in the proviso to s 6(1) from making the servant or agent of the Crown personally liable to the plaintiff. For these reasons Ms Couch‘s claim cannot be struck out on the basis argued by the Crown.

  175. Since writing the foregoing I have had the opportunity to consider McGrath J's approach to this issue. The difficulty I have with it is that McGrath J makes a distinction which I do not consider can be found in the terms of the section. He is satisfied a major change to the liability of the Crown in tort was not intended by s 86. To this extent we are in agreement. But once that point is reached I do not see how it can be said that s 86 and s 6, when read together, relieve public servants from personal liability while preserving the Crown's vicarious liability.

  176. Once the personal liability in tort of public servants is removed the proviso to s 6(1) comes into operation. This is because the act or omission of the servant or agent does not give rise to a cause of action in tort against that servant or agent independently of the Crown Proceedings Act. Hence the Crown has no vicarious liability ("no proceedings shall lie against the Crown") by virtue of the proviso to s 6(1). The consequence is that the Crown would be vicariously liable in tort only for the acts or omissions of its servants or agents if those acts or omissions were not in good faith. This would amount to a major change to the liability of the Crown for the torts of its servants or agents, the very thing we are in agreement was not envisaged by the section.

  177. The difficulty I have identified is reinforced by s 6(4) of the Crown Proceedings Act which provides that the Crown is entitled to the same statutory immunities and limitations in respect of a tort committed by a department or one of its officers as the department or officer is entitled to. I do not find it easy to reconcile that provision with the proposition that the proviso to s 6(1) does not exclude Crown liability where the servant or agent of the Crown is immunised on account of being a Crown servant or agent.

    Summary of exemplary damages conclusion

  178. Exemplary damages are anomalous. Civil remedies are not generally designed to punish. The reach of exemplary damages should therefore be confined rather than expanded. Outrageousness is not a satisfactory sole criterion. The concept lacks objective content and does not contain sufficient certainty or predictability. Exemplary damages should be confined to torts which are committed intentionally or with subjective recklessness, which is the close moral equivalent of intention.

  179. Applying that principle to the case of negligently caused personal injury (that is, injury caused through breach of a duty of care), exemplary damages may be awarded if, but only if, the defendant deliberately and outrageously ran a consciously appreciated risk of causing personal injury to the plaintiff. Whether running such a risk should be regarded as outrageous will depend on the degree of risk that was appreciated and the seriousness of the personal injury that was foreseen as likely to ensue if the risk materialised.

    Disposition of appeal

  180. The conclusions reached in these reasons, coupled with those expressed at the first stage of this appeal, mean that the Courts below should not have ordered the striking out of Ms Couch‘s claim. The appeal should therefore be allowed and the orders made below set aside. The application for strike-out should be dismissed with costs in Ms Couch‘s favour both in this Court and in both Courts below. I would make orders to this effect. The case will now proceed in the High Court in the ordinary way. Ms Couch should, as soon as possible, file an amended statement of claim consistent with these reasons.

    McGrath J

    Introduction

  181. The appellant, Ms Couch, has brought a proceeding in the High Court against the Attorney-General claiming exemplary damages for negligence in relation to the Department of Corrections, and a probation officer. The claim related to their administration of the parole of William Bell. In 2001, in the course of a robbery of the licensed premises at which they had both been employed, Bell had attacked the appellant, causing her grievous bodily harm. The appellant confines the relief she seeks from the Attorney-General to exemplary damages because she accepts that she is barred by the Accident Compensation Act 2001 (section 317) from claiming any damages that would compensate her for the extensive personal injuries she suffered.

  182. The Attorney-General applied to strike out the appellant‘s claim in negligence on two grounds. The first was that no duty of care was owed by the Department or its employees to the appellant which covered the circumstances alleged to have caused her losses and suffering. In an earlier judgment, we refused to strike out the proceeding on that ground and adjourned the appeal for hearing of argument on an alternative ground for striking out: Couch v Attorney-General [2008] NZSC 45, [2008] 3 NZLR 725 (on appeal from Hobson v Attorney-General). This ground puts in issue whether exemplary damages are available at all in the pleaded circumstances. If they are, the Court is asked to state what the legal test is for an award. We heard argument from counsel on these points at a resumed hearing.

  183. The questions before the Court are:

    1. Is the combined effect of s 86 of the State Sector Act 1988 and s 6(1) of the Crown Proceedings Act 1950 to immunise the Crown from liability in tort for the acts of its servants?

    2. Does s 317 of the 2001 Act bar recovery of exemplary damages for negligent conduct of a defendant that is causative of personal injury by accident under that Act?

    3. If not, what test must a plaintiff meet to be entitled to an award of exemplary damages?

    Section 86 of the State Sector Act 1988

  184. The Solicitor-General submitted that s 86 of the State Sector Act 1988, in combination with s 6(1) of the Crown Proceedings Act 1950, confers an immunity on the Crown for acts of servants performed in good faith.

  185. Sections 86 and 6(1) provide as follows:

    86.

    Protection from liability

    No chief executive, or .... employee, shall be personally liable for any liability of the Department, or for any act done or omitted by the Department or by the chief executive or any .... employee of the Department or of the chief executive in good faith in pursuance or intended pursuance of the functions or powers of the Department or of the chief executive.

    6.

    Liability of the Crown in tort

    (1)

    Subject to the provisions of this Act and any other Act, the Crown shall be subject to all those liabilities in tort to which, if it were a private person of full age and capacity, it would be subject—

    (a)

    In respect of torts committed by its servants or agents;

    ....

    Provided that no proceedings shall lie against the Crown by virtue of paragraph (a) of this subsection in respect of any act or omission of a servant or agent of the Crown unless the act or omission would apart from the provisions of this Act have given rise to a cause of action in tort against that servant or agent or his estate.

  186. The purpose of s 6(1) of the Crown Proceedings Act is to make the Crown liable for the torts committed by its servants or agents in the same way as any other person of full age and capacity subject to the expressed exceptions. One exception is that the Crown is not liable unless the claimant can point to a Crown servant who would be liable independently of the Act. The Solicitor-General‘s argument was that when s 6 is read with s 86 of the State Sector Act, the Crown cannot be held liable in negligence under s 6. Because s 86 removes the tortious character of wrongful acts by public servants made in good faith, the Crown‘s liability under s 6(1) of the Crown Proceedings Act is excluded by the proviso to that subsection. The reason is that there is no servant who is personally liable.

  187. The Solicitor-General‘s argument is based on a literal reading of the two provisions but I am satisfied that, reading the text of s 86 in light of the purposes of the State Sector Act, it is plain that such a major change to the liability of the Crown in tort was not the purpose of that statute. That would also be inconsistent with s 27(3) of the New Zealand Bill of Rights Act 1990, which provides that:

    Every person has the right to bring civil proceedings against, and to defend civil proceedings brought by, the Crown, and to have those proceedings heard, according to law, in the same way as civil proceedings between individuals.

    Section 27(3) reflects the purpose of s 6(1) of the Crown Proceedings Act. Section 27(3) likewise is intended to place the Crown in the same position in relation to litigation as private individuals: Knight v Commissioner of Inland Revenue [1991] 2 NZLR 30 (CA) at 37 per Cooke P. The White Paper on the Bill of Rights stated the purpose of s 27(3) was:

    [T]o give constitutional status to the core principle recognised in the Crown Proceedings Act 1950: that the individual should be able to bring legal proceedings against the Government, and more generally to engage in civil litigation with it, without the Government enjoying any procedural or jurisdictional privileges. This is central to the rule of law.

    See Geoffrey Palmer "A Bill of Rights for New Zealand: A White Paper" [1984–1985] 1 AJHR A6 at [10.176].

  188. In Canada legislative provisions which are similar to s 86 have been read in this way. In Hill v British Columbia (1997) 148 DLR (4th) 337 (BCCA) at [24], the Court of Appeal for British Columbia held that a provision preventing personal liability of a Crown employee, for an act done in good faith, could not have been intended to be a "back door" for the Crown to escape liability under the equivalent of s 6 of the Crown Proceedings Act. The Court held that the equivalent provision to the proviso to s 6(1) entitles the Crown to rely only on defences that would be available if the proceedings were "between persons"[66] and not to defences available only to employees of the Crown generally. There is no material difference between the language of the Canadian statutory provisions and s 6. As well, s 27(3) of the Bill of Rights Act provides that litigation involving the Crown is to be treated as though the proceedings were "between individuals". Sections 27(3) and 6(1) require that the Crown is placed in the same position as a private employer. A private employer does not have Crown status nor the power or ability to rely on defences that are conferred only on employees who are Crown servants.

  189. This supports giving the proviso to s 6(1) a contextual interpretation by reference to the whole of that subsection. Section 6(1) imposes on the Crown liability in tort, for the actions of its servants and agents, to which it would be subject if it were a person of full age and capacity. That, of course, means a private person. In that context the proviso should be read as excluding Crown liability only where the defence can be relied on by the Crown servant or agent as a private person. It does not exclude Crown liability where the servant is immunised on account of being a Crown servant, which is the effect of s 86.

  190. One of the purposes of the 1988 Act was to provide for employment relationships in the state sector. A further purpose was to provide greater responsibilities in that respect on chief executives of government departments who became employers. With such greater responsibility came greater exposure to legal proceedings and another purpose of the Act was to protect against that exposure.

  191. Consistently with its plain terms and these purposes, s 86 of the State Sector Act must be read as relieving public servants from personal liability. It does so by preventing chief executives and employees from incurring any personal liability for negligent acts committed in good faith. But giving s 6(1) of the Crown Proceedings Act a contextual interpretation, which is consistent with s 27(3) of the Bill of Rights Act, that does not affect Crown liability.

  192. As indicated, I consider my view better fits the ordinary meaning of s 86. That ordinary meaning is consistent with the policy of immunity from personal liability summarised by Hogg & Monahan (Liability of the Crown (3rd ed, Carswell, Ontario, 2000) at 191):

    Many commentators take the view that some degree of immunity from tortious liability should be conferred by statute upon individual Crown servants. It is argued that a damages award against a Crown servant is an unpredictable and usually disproportionately severe penalty to impose on a person who has acted in good faith in the intended execution of his or her duties. It is also argued that the risk of personal liability could lead to overly cautious (risk-averse) behaviour on the part of Crown servants whose jobs call for vigorous action but who are fearful of being sued.

  193. For these reasons, and in agreement with other members of the Court, I do not accept the Crown submission that s 86 removes Crown liability in tort. I disagree, however, with the majority‘s view of s 86.[67] I read that provision as excluding personal liability of chief executives and Crown servants for all negligent acts committed in good faith. That view is shared by Wilson J. The majority‘s interpretation is that s 86 removes tortious liability from individual chief executives and employees for the wrongful acts of others, but not for those of the chief executives or Crown servants themselves. The point of importance to the present case, however, is that all judges are agreed that s 86 does not affect Crown liability.

    Exemplary damages in New Zealand

  194. In general, a court will award damages to compensate a plaintiff for loss suffered as a result of the wrongful conduct of a defendant in breach of the plaintiff‘s rights. The common law has, however, long recognised that in appropriate circumstances a court may award exemplary damages in addition, not to compensate the plaintiff for the harm suffered, but to punish the defendant for the wrongful conduct: Stephen Todd (ed) The Law of Torts in New Zealand (5th ed, Brookers Ltd, Wellington, 2009) at [25.3.03 (1)] and [25.3.03 (2)].

  195. Over the years the legitimacy of awards of exemplary damages has been questioned. The main argument that critics advance is that punishment is not properly imposed in civil proceedings and, because they are awarded in that context, exemplary damages are anomalous. In 1964 the House of Lords confronted this issue of principle in Rookes v Barnard [1964] AC 1129 (HL). The leading judgment on this point, with which all other Judges agreed, was delivered by Lord Devlin. The House of Lords accepted that the punitive purpose of exemplary damages confused the civil and criminal functions of the law (at 1221), but decided that, nevertheless, there were situations in which there was "practical justification for admitting into the civil law a principle which ought logically to belong to the criminal" (at 1226). Lord Devlin confined the situations in which common law exemplary damages could be awarded to three categories of cases which he identified. Subsequently, in Broome v Cassell & Co Ltd [1972] AC 1027 (HL), the House of Lords rejected criticisms of the judgment in Rookes v Barnard, affirming that it was the considered precedent on the basis of which English law on exemplary damages should go forward (at 1083 per Lord Hailsham).

  196. In 1969 in Australian Consolidated Press Ltd v Uren [1969] 1 AC 590 (PC), the Privy Council accepted the view of the High Court of Australia that Australian courts were not required to apply the restrictive approach to exemplary damages in Rookes v Barnard in place of the settled broader approaches being applied by Australian courts.

  197. In 1982, in Taylor v Beere [1982] 1 NZLR 81 (CA), the Court of Appeal of New Zealand followed the Australian approach. The Court reaffirmed an earlier line of authority which said that damages having a purpose beyond compensation would, in appropriate circumstances, be available to the courts as an additional remedy for wrongful conduct in breach of a plaintiff‘s rights. Taylor v Beere was a defamation case, where exemplary damages had been awarded by a jury in addition to damages compensating the plaintiff. The Court of Appeal decided not to adopt the categorisation approach applied in Rookes v Barnard. Rather, in appropriate cases, New Zealand courts should continue to make awards of exemplary damages to punish the wrongdoers, in addition to damages that compensated the plaintiff for losses suffered. The award of exemplary damages in the particular case was upheld.

  198. On the same day as it decided Taylor v Beere, the Court of Appeal also delivered judgment in Donselaar v Donselaar [1982] 1 NZLR 97 (CA), a case in which the plaintiff claimed exemplary damages for assault and battery. The statement of claim alleged that the defendant had caused harm to the plaintiff of a kind which amounted to personal injury by accident in terms of the accident compensation legislation. The Court held that, nevertheless, the bar under the legislation (Accident Compensation Act 1972, s 5(1)) on proceedings for damages arising out of personal injury by accident did not apply to the claim for exemplary damages. Exemplary damages were not compensation for the injury. As well, they did not "arise" out of the injury suffered but from the outrageous manner in which the defendant had acted in committing the tort (at 109 per Richardson J). On the correct interpretation of the 1972 Act the bar did not apply to them. Because compensatory damages were unavailable, and the statutory compensation did not have any punitive element, the need for the additional remedy to punish high-handed and illegal conduct, public or private, remained (at 104 per Cooke J). The claim for exemplary damages was accordingly allowed to stand.

    Section 317 – the statutory bar

  199. At the resumed hearing the Attorney-General invited this Court to hold that proceedings in negligence seeking exemplary damages for an injured plaintiff were barred by s 317 of the 2001 Act. That provision bars "proceedings .... in any court .... for damages arising directly or indirectly out of .... personal injury covered by (the) Act". Section 317 is the successor of s 5 of the Accident Compensation Act 1972[68] which was considered by the Court of Appeal in Donselaar. There is no material difference in the language used to express the bar in the successive provisions.

  200. The Crown‘s submission is based on the proposition that exemplary damages may be recovered only for conduct amounting to a legal wrong. The appellant‘s claim relies on negligence but that tort, the Crown says, is incomplete without proof that actual damage has been sustained. As s 317 bars the appellant from claiming damages arising out of personal injury, the appellant is unable to claim exemplary damages in cases where negligence has caused personal injury. The argument does not directly confront the Court of Appeal‘s reasoning in Donselaar as the claim for exemplary damages was based in battery. If the Crown‘s argument is correct, it would probably confine awards of exemplary damages to the intentional torts.

  201. A similar point was discussed by Richardson J in Donselaar. After pointing out that a cause of action in battery was complete without proof of actual damage he said (at 111):

    And s 5(1) does not preclude recognition in proceedings for damages not barred under its provisions of the existence of nominal damage arising out of a personal injury. What it bars is proceedings for recovery. Moreover, if I am wrong in the conclusion I have reached and exemplary damages must be founded on the existence of actual rather than nominal damage, it does not follow that s 5(1) bars proceedings for exemplary damages. It does not exclude proof that the plaintiff suffered a loss which is compensable under the legislation: it merely prohibits the bringing of proceedings for damages in respect of that loss.

  202. Likewise in the present case, if satisfying the elements of the tort of negligence requires proof of actual loss, s 317 does not preclude calling evidence that loss has been suffered in order to establish liability in proceedings that are not barred. On that basis the issue is, as it was in Donselaar, whether the appellant‘s proceedings are for damages arising indirectly out of personal injury covered by the Act. The need to prove harm for which damages cannot be recovered does not provide a basis for distinction.

  203. I am satisfied that Donselaar was correctly decided on this point. The judgment in Donselaar was the outcome of a purposive construction of the language of the statutory bar in the 1972 Act which, like its successors, makes provision for statutory compensation for personal injury. Cooke J was satisfied that the reformers responsible for accident compensation legislation had not deliberately set out to do away with exemplary damages. Preserving them was consistent with the social philosophy of the legislation. Two views on the meaning of the statutory bar provision were open to the Court. The preference for the view that the general language expressing the bar did not extend its application beyond proceedings for compensation reflected the statutory purpose. As well, there was a continuing social need for what Cooke J described (at 106) as "constitutional remedies for high-handed and illegal conduct, public or private". In Donselaar the Court stated the law of exemplary damages in a way that enabled the statutory bar to operate consistently with its purpose while at the same time meeting the continuing need for the special remedy. I see no reason to depart from the Court of Appeal‘s approach in Donselaar .

    Reconsidering decisions of the Privy Council

  204. The Crown‘s remaining ground for its strike-out application is that the law does not allow claims for exemplary damages for negligence in cases of personal injury. A New Zealand Court first held that exemplary damages could be awarded in those circumstances in 1996 - McLaren Transport Ltd v Somerville [1996] 3 NZLR 424 (HC) where Tipping J disagreed with dicta to the contrary of Somers J in Donselaar at 113. In 1998 the Court of Appeal did not decide, but accepted for the sake of argument, that in some cases of negligence, exemplary damages might be awarded: Ellison v L [1998] 1 NZLR 416 (CA). The Court emphasised (at 419) that the cases were likely to be rare because negligence was an unintentional tort. In 2001 in Bottrill v A [2001] 3 NZLR 622 (CA), a majority of the Court of Appeal decided that exemplary damages were not confined to intentional torts, and were available within strict limits for claims in negligence involving conscious risk taking. On appeal, the Privy Council, also by a majority, reduced the restrictions on that availability: Bottrill v A [2002] UKPC 44, [2003] 2 NZLR 721.

  205. It follows that, in order for the Crown‘s application to succeed, this Court must reconsider and overrule the ratio of the Privy Council‘s judgment. This raises for the first time the issue of when the Court should reconsider judgments on New Zealand law of the Privy Council.

  206. In 2003, Parliament enacted legislation which established this Court as the final court of appeal for New Zealand and at the same time ended appeals from New Zealand courts to the Privy Council (Supreme Court Act 2003, ss 3(c), 6 and 42). Its purpose was:

    3.

    Purpose

    (1)

    The purpose of this Act is—

    (a)

    to establish within New Zealand a new court of final appeal comprising New Zealand judges—

    (i)

    to recognise that New Zealand is an independent nation with its own history and traditions; and

    (ii)

    to enable important legal matters .... to be resolved with an understanding of New Zealand conditions, history, and traditions; and

    (iii)

    to improve access to justice; and

    (b)

    to provide for the court‘s jurisdiction and related matters; and

    (c)

    to end appeals to the Judicial Committee of the Privy Council from decisions of New Zealand courts ....

  207. The common law rule of precedent requires that every court is bound to follow any case decided by a court above it in the hierarchy: Rupert Cross and JW Harris Precedent in English Law (4th ed, Oxford University Press, Oxford, 1991) at 6. Under this rule, until 2004, all New Zealand-based courts applied judgments of the Privy Council in appeals from New Zealand as authoritative statements of New Zealand law. Such judgments remain authoritative precedent which is binding on those courts, subject, however, to the position of the Supreme Court of New Zealand, which was established on 1 January 2004. The Supreme Court is the final court of appeal and has succeeded to the position of the Privy Council at the apex of the New Zealand court system. It is not bound by the Privy Council‘s judgments and, if a lower court perceives any conflict between the Supreme Court‘s decisions and those of the Privy Council, it must follow those of the former.

  208. In exercising its responsibility for determining the content of New Zealand law in any case before it, the Supreme Court decides as a matter of course in every case whether decisions of other courts, overseas and in New Zealand, should be followed. It is guided by the statutory policy that important legal matters are to be resolved with an understanding of New Zealand‘s conditions, history and traditions. The knowledge and appreciation that members of the Court have of these matters must be brought to bear in articulating and applying New Zealand law, whether its source is the common law or statute.[69]

  209. This does not, however, mean that this Court should take an unrestrained approach when invited to depart from a decision of the Privy Council. The use of precedent remains a foundation on which to decide the content of the law and determine how it applies in any case. The fact that this Court has power to depart from the Privy Council‘s decisions does not mean that it should do so merely because, if the matter were being decided afresh, the Court might take a different view. On the contrary, a policy in this Court of general adherence to precedent in respect of Privy Council judgments, even though it is not bound to do so, will have beneficial force in achieving stability, consistency, orderly development and a degree of certainty in the law.

  210. Accordingly, statements of the law of New Zealand by the Privy Council should stand as authoritative unless this Court considers there are cogent reasons for reconsideration of and departure from them. But, in deciding whether there are such reasons, the Court must recognise that, at times, from a New Zealand perspective, rigid adherence to precedent would perpetuate inappropriate legal rules, continuation of which is not justified by the advantages of certainty: See Dahya v Dahya [1991] 2 NZLR 150 (CA) at 160 per Richardson J.

    Should Bottrill be revisited?

  211. Is it appropriate for this Court now to depart from the principles stated in the Privy Council majority‘s judgment in Bottrill? No legislative amendments have addressed the issue since that judgment was delivered in 2002. The area of exemplary damages is, however, one which the legislature has largely left to the courts, both before and since Parliament first abolished compensatory damages for personal injury in 1972. The legislature‘s silence since 2002 does not indicate approbation of the Privy Council‘s view.

  212. The scope of the remedy is an issue addressed by the courts of many jurisdictions in which it has been seen as one for determination in the context of local conditions. In New Zealand, the question has a particular local dimension because of the prohibition placed on compensatory damages by accident compensation legislation. The Privy Council recognised in Bottrill that the local court was better placed to make the evaluation of the relevant social considerations that a decision concerning the test for exemplary damages requires. The Judges in the majority were clear that they were making their own evaluation of the policy considerations only because they disagreed with the Court of Appeal‘s view that, as a matter of principle, exemplary damages in negligence were confined to advertent conduct (at [57]). Of relevance also is that the Judges of the Privy Council were divided, with two joining what became the majority judgment delivered by Lord Nicholls, and two approving the approach taken by the Court of Appeal. Together, these considerations provide a strong argument for the Court to consider whether to depart from Bottrill.

  213. The issue is before us directly as the remaining ground of the Crown‘s application, which is effectively a strike-out application. That context does not warrant our refusing to determine it. We must do so assuming that the facts pleaded by the appellant can be proved. In this respect the position this Court faces is the same as that faced by the Court of Appeal when it decided Donselaar. Nor does the lack of a determination of the Court of Appeal or High Court on the issue in this case provide a reason for declining to reconsider Bottrill. That will invariably be the position because those courts will be bound to apply the Privy Council decision concerned. Indeed, the procedure has advantages in the present case. If the claim survives the strike-out application, this Court‘s judgment will clarify the basis on which an action is to be tried, avoiding the risk of appeals possibly making a retrial necessary.

  214. These are cogent reasons that persuade me the Court should address the arguments on the scope of availability of exemplary damages in negligence cases, involving personal injury.

    The test for exemplary damages

  215. Although New Zealand departed from Rookes v Barnard in relation to its restriction of the categories of tortious conduct for which an award of exemplary damages might be made, the distinction Lord Devlin drew between aggravated damages and exemplary damages has been applied in New Zealand. Aggravated damages may be awarded to compensate for additional suffering for injury to feelings and dignity that results from the manner of a defendant‘s conduct. Exemplary damages are separate and are awarded to punish and deter a wrongdoer; as Lord Devlin put it (at 1227), "to teach a wrongdoer that tort does not pay". The clarification sets limits on the scope of exemplary damages which emphasise that they have no element of compensation. Lord Devlin also emphasised, ibid, that awards of exemplary damages may only be made to a plaintiff who is the victim of the behaviour that is being punished. A person who is unaffected has no claim.

  216. Determining the circumstances in which exemplary damages should be awarded in negligence has, however, been a controversial issue for the New Zealand courts. Negligent conduct covers a range of culpability and it has been difficult to specify when it reaches the threshold that makes a punitive award of damages appropriate. This difficulty is well demonstrated by the different views of the Privy Council and Court of Appeal in Bottrill, the judgments of each Court being reached by a majority.

  217. In Bottrill, a plurality of three of the five Judges of the Court of Appeal held that exemplary damages could be awarded for negligence only where the defendant is subjectively aware of the risk to which the negligent conduct exposed the plaintiff and acts deliberately or recklessly in taking that risk (at [41], [62] and [63] per Richardson P, Gault and Blanchard JJ). They could not be awarded if the conduct was merely inadvertent. This test required an objective assessment of whether there was deliberate or reckless risk taking, the latter turning on whether the defendant was subjectively reckless. The test was described as one of "conscious risk taking", which was satisfied (at [62]):

    where on an objective assessment the defendant had an actual appreciation of the risk or was recklessly indifferent to the consequences and must be taken to have been content for the consequences to happen as they did.

  218. The three Judges decided that the element of the defendant‘s state of mind under this test could be satisfied by proving circumstances which justified an inference of the defendant‘s awareness of the risk and acceptance it could well happen. They emphasised (at [43]) that for there to be conscious risk taking, where there is no intention to harm, the "quality" of the defendant‘s conduct has to so closely approach that involved in an intentional harming that the appropriate response is civil punishment. An actual appreciation of the risk and taking it while hoping no harm would ensue would meet this test. So would a case where, viewed objectively, the defendant must have appreciated the risk and gone ahead indifferent to consequences.

  219. Tipping J, in a concurring judgment, stated the test in similar terms. He said that the level and kind of negligence had to amount "to a conscious, outrageous and flagrant disregard for the plaintiff‘s safety, meriting condemnation and punishment" (at [174]). He agreed (at [175]) that the subjective state of mind involved in conscious appreciation of the risk to the plaintiff‘s safety would usually be inferred from objective circumstances.

  220. On appeal the Privy Council reversed the Court of Appeal‘s judgment. The Judges in the minority (Lord Hutton and Lord Millett) reasoned that, as the rationale for exemplary damages was punishment, rather than marking disapproval of outrageous conduct, exemplary damages should not be imposed, in the absence of intent to cause harm, unless the defendant had been subjectively reckless. They added as a qualification that the Court might infer subjective negligence where it found gross negligence.

  221. The Privy Council majority (Lord Nicholls, Lord Hope and Lord Rodger) said that, ordinarily, a defendant who committed a tort would be required to make good the plaintiff‘s loss including, where appropriate, by payment of aggravated damages. Exceptionally, exemplary damages could be awarded, in addition, to punish a defendant for conduct that was so outrageous that compensation was an inadequate response (at [20]). The cases involving negligence in which exemplary damages were appropriate would be exceptional and overwhelmingly cases in which the conduct was intentional or involved conscious recklessness closely approaching intentional wrongdoing (at [24]). The exceptional cases meeting the outrageousness test would also involve either intentional wrongdoing, with additional features making the conduct particularly appalling, or reckless indifference evoking a sense of outrage.[70] The Privy Council said (at [23]) that such conscious recklessness very closely approaches intentional wrongdoing.

  222. The majority also said that absence of intentional wrongdoing or conscious recklessness in any case pointed strongly away from the circumstances being appropriate for an award of exemplary damages. But even so, if the defendant‘s conduct satisfied the outrageous test and condemnation was called for, the judge had power to award exemplary damages (at [37]).

  223. The basis for this additional element in the test is the attitude "never say never". It was expressed this way (at [26]):

    However, if experience in the law teaches anything, it is that sooner or later the unexpected and exceptional event is bound to occur. It would be imprudent to assume that, in the absence of intentional wrongdoing or conscious recklessness, a defendant‘s negligent conduct will never give rise to a justifiable feeling of outrage calling for an award of exemplary damages. "Never say never" is a sound judicial admonition. There may be the rare case where the defendant departed so far and so flagrantly from the dictates of ordinary or professional precepts of prudence, or standards of care, that his conduct satisfied this test even though he was not consciously reckless.

    [original emphasis]

  224. Thomas J, who dissented in the Court of Appeal, had proposed a more straightforward test. Exemplary damages would be awarded if the negligent conduct was so outrageous and contumelious as to warrant an award (at [77]). In rare cases features of negligent conduct would be so flagrant that exemplary damages should be warranted notwithstanding the defendant was not subjectively aware of the risk created and did not deliberately take the risk (at [80]). It would be sufficient if the wrongdoer ought to have known that the conduct posed an unreasonable risk. This reflected the Judge‘s view that exemplary damages served other functions of the law of torts as well as punishment (at [95] and [101]).

  225. As well, Thomas J considered that the distinction, drawn by the Court of Appeal majority, between inadvertent and advertent negligence would not be easy to draw in practice. It would necessarily distract the courts from focusing on reprehensible features said to warrant the remedy (at [144]).

  226. The Privy Council majority‘s threshold for an award of exemplary damages accordingly adopts an ultimate criterion of outrageousness. Lord Nicholls (at [33]) referred to a difference in degree, involving the presence or absence of something extra which turns a case of grossly negligent conduct into conduct that is "quite outrageous". That conduct would qualify for an award of exemplary damages.

  227. At the same time, however, the majority also recognised the importance of an intentional or subjectively reckless element in deciding if wrongful conduct of a defendant should be subject of an exemplary damages award. Lord Nicholls said that it would be rare that cases not exhibiting either of these features will justify awards of exemplary damages and that their absence pointed strongly against any award.

  228. Indeed, this factor is given special emphasis in the judgment (at [64]):

    Their Lordships cannot overemphasise what has already been indicated more than once. The cases where it is appropriate to make an award of exemplary damages are exceptional. The cases where it is appropriate to make an award of exemplary damages in the absence of intentional wrongdoing or conscious recklessness will be exceptional and rare indeed. It must always be kept in mind that compensation is not the purpose of exemplary damages. A perceived need for compensation, or further compensation, is not a proper basis for making an award of exemplary damages.

  229. These factors indicate that the Privy Council‘s approach was largely an endorsement of the Court of Appeal‘s view of the need for elements of intention or subjective recklessness before misconduct would qualify for exemplary damages. But as well, the Judges wanted to leave open as an additional possibility an award for outrageous conduct outside of that area. This was on the basis it seems that courts should "never say never".

  230. On this analysis, which reflects by far the greater part of the Privy Council majority‘s reasoning, there is probably little practical difference between the test those Judges favour and that favoured by the majority of the Court of Appeal.

  231. There is, however, a passage in the Privy Council‘s judgment which, to my mind, is not consistent with this analysis. After referring to the "broad-based rationale" for exemplary damages, being that conduct committing a civil wrong was so outrageous that something more than an order to pay compensation was required, Lord Nicholls said (at [22]):

    The Court‘s discretionary jurisdiction may be expected to extend to all cases of tortious wrongdoing where the defendant‘s conduct satisfies this criterion of outrageousness. Any departure from this principle needs to be justified. Otherwise the law lacks coherence. It could not be right that certain types of outrageous conduct as described above should attract the Court‘s jurisdiction to award exemplary damages and other types of conduct, satisfying the same test of outrageousness, should not, unless there exists between these types a rational distinction sufficient to justify such a significant difference in treatment.

  232. This passage appears to set out a much broader approach to exemplary damages than is generally contemplated by the remainder of the judgment. Under it, exemplary damages may be awarded in tort wherever the wrongful conduct is outrageous and contumelious. This is not consistent with the general emphasis on an exceptional remedy that is rarely awarded in the absence of subjective recklessness. Nor is this test consistent with the theme in the majority judgment of allowing a narrow exception for an unexpected case.

  233. The majority (at [46]) referred to the judgment of the Supreme Court of Canada in Whiten v Pilot Insurance Co 2002 SCC 18, [2002] 1 SCR 595 at [67], observing that the Supreme Court had approved a "broader approach", earlier adopted by the British Columbia Court of Appeal (in Robitaille v Vancouver Hockey Club Ltd (1981) 124 DLR (3d) 228 (BCCA) at 250), that punitive damages ought to be available whenever the conduct of the defendant was such as to merit condemnation by the court. The Supreme Court‘s observation was, however, made in a passage in its judgment in which it had declined to take the categorisation approach to exemplary damages adopted in Rookes v Barnard: Whiten at [67]. The Supreme Court‘s approval of a "broader approach" was confined to that context and had no wider significance.

  234. The Supreme Court of Canada saw the mechanism for control of exemplary damages as lying not in restrictions of the categories of cases but in rationally determining circumstances that warrant punishment in addition to compensation in a civil action. It concluded that, ibid:

    It is in the nature of the remedy that punitive damages will largely be restricted to intentional torts.

    As well, it emphasised the importance of both the state of mind and conduct of the defendant. A principled approach was desirable (at [70]). The position taken in subsequent decisions of the Supreme Court of Canada also gives no support to the inclusion of a separate criterion of outrageousness as a sufficient basis for an award of exemplary damages. In Honda Canada Inc v Keays 2008 SCC 39, [2008] 2 SCR 362 at [62], the majority judgment said:[71]

    [P]unitive damages are restricted to advertent wrongful acts that are so malicious and outrageous that they are deserving of punishment on their own.

    And later, in emphasising the exceptional nature of an award of exemplary damages (at [68]):[72]

    Another important thing to be considered is that conduct meriting punitive damages awards must be "harsh, vindictive, reprehensible and malicious", as well as "extreme in its nature and such that by any reasonable standard it is deserving of full condemnation and punishment".

  235. As I have earlier indicated, in 1982 New Zealand‘s position on exemplary damages was aligned with the law in Australia rather than the more restrictive approach traditionally taken in England. The leading decision on the availability of exemplary damages in negligence in that jurisdiction is the High Court of Australia‘s judgment in Gray v Motor Accident Commission [1998] HCA 70, (1998) 196 CLR 1.

  236. In Bottrill, the Privy Council majority said that Gray provided an instance of its view that, while other jurisdictions place emphasis on the presence of intentional misconduct or conscious recklessness, they "invariably qualify their remarks by leaving open the possibility of an award of exemplary damages in other cases" (at [45]). It is true that in Gray the majority‘s judgment (per Gleeson CJ, McHugh, Gummow and Hayne JJ) observes that the phrase "conscious wrongdoing in contumelious disregard of another‘s rights" describes "at least the greater part of" the field and that the remedy is said to arise "chiefly, if not exclusively" in such cases (at [14] and [20]). But these observations are respectively made in the course of a general discussion of the kinds of cases in which exemplary damages are awarded and the exceptional nature of the remedy. The judgment then identifies as a different and narrower issue whether exemplary damages are available for negligence, rather than some intentional wrong, on which it concludes (at [22], Kirby J dissented at [84]–[86]):

    [E]xemplary damages could not properly be awarded in a case of alleged negligence in which there was no conscious wrongdoing by the defendant. Ordinarily, then, questions of exemplary damages will not arise in most negligence cases be they motor accident or other kinds of case. But there can be cases, framed in negligence, in which the defendant can be shown to have acted consciously in contumelious disregard of the rights of the plaintiff or persons in the position of the plaintiff. Cases of an employer‘s failure to provide a safe system of work for employees in which it is demonstrated that the employer, well knowing of an extreme danger thus created, persisted in employing the unsafe system might, perhaps, be of that latter kind. No doubt other examples can be found.
  237. I regard it as clear that the High Court of Australia in Gray did not wish to leave open the possibility of an award of exemplary damages in negligence actions, not involving conscious wrongdoing.[73] A decision by this Court not to follow the Privy Council, and to approve the Court of Appeal‘s formulation of the test for liability for exemplary damages in negligence would bring New Zealand law into line with the view of the High Court of Australia.

  238. Two linked considerations which underpinned the Court of Appeal‘s reasoning in Bottrill are, in my opinion, of fundamental importance in deciding when exemplary damages are to be awarded. The first is that the primary purpose of exemplary damages is to punish a defendant for wrongful conduct. Deterrence of the offender is likely to be the effect of an award, as is vindication of the plaintiff who suffers harm and receives the damages. But these are both incidental consequences and should not divert the courts from the punitive purpose of the remedy.

  239. Secondly, because the focus of the courts is on punishment, it is the culpability of the defendant‘s conduct that justifies an award of exemplary damages. Assessment of the degree of culpability is straightforward where a defendant intentionally causes harm. Where that is not the case, it is more problematic. If the claim is based on negligent acts, the defendant‘s conduct must be sufficiently egregious to be deserving of the punitive remedy. The Court of Appeal‘s approach in Bottrill was to set the threshold at acts taken with conscious recklessness as to the consequences. This test provides for the mental element that is appropriate for actions to be punished. That is not the case, however, with a test based on perceptions of outrageousness.[74]

  240. The other consideration of particular importance, which arises in the New Zealand context, is the accident compensation scheme and in particular its legislative bar on recovery of damages as compensation for personal injury. The Privy Council majority doubted that an increase in the number of awards of exemplary damages would follow from the addition of the test stated in its decision, because the settled practice in New Zealand was for judges to exercise the discretion to make awards of exemplary damages with great restraint.

  241. The conclusion I reach on the likely impact of continuing the Privy Council‘s approach is to the contrary. I consider it will be destructive of the exceptional nature of the remedy. This is for two reasons. First, since 1992, the accident compensation legislation has not provided lump sum benefits for non-pecuniary loss. Although this development cannot, as a matter of principle, justify lowering the threshold for awards, it has unsurprisingly resulted in an increased number of claims for exemplary damages. The reality is that the only way those suffering personal injury can obtain capital payments in civil proceedings is through bringing claims for exemplary damages when they are available and, as earlier discussed, plaintiffs may bring proceedings claiming such damages as the sole remedy.

  242. The second reason is a point made by the minority Judges of the Privy Council. If plaintiffs are not required to establish intent, or subjective recklessness, of a wrongdoer but can recover exemplary damages if the conduct is outrageous, the standard becomes a very uncertain one. As the minority said (at [81]):

    Outrage is a subjective concept and if a sense of outrage is the principal factor to guide a Judge, individual Judges might well take differing views as to whether grossly negligent conduct in a particular case should be viewed as outrageous.

  243. The problem with a stand-alone outrageousness test is that it requires the court to make an assessment without reference to precise criteria. Judges must apply the remedy if, as a matter of impression, they consider it appropriate in the case. The absence of principle tends to put the remedy in the gift of the trial court. If this test is retained, there will be considerable uncertainty within the legal system as to when such damages are available. Like cases will not be decided alike.[75] This, coupled with the strictures of the regime of limited statutory compensation, will lead to many claims and multiple appeals against decisions where the minds of judges will reasonably differ.

  244. One such case is McDermott v Wallace. In the District Court, the Judge was "certain" that the defendant‘s conduct met the test of outrageousness: McDermott v Wallace DC Wellington NP296/97, 30 May 2003 at [74]. On appeal, the High Court Judge decided that the defendant‘s conduct "fell far short" of outrageous conduct: Wallace v McDermott [2004] NZAR 747 (HC) at [27]. The High Court Judge did, however, grant leave to appeal to the Court of Appeal, recognising that the Privy Council‘s test was not easy to apply in practice: Wallace v McDermott HC Wellington CIV-2003-485-1427, 20 September 2004 at [5]. The Court of Appeal allowed the appeal on liability and upheld the District Court‘s decision, observing that the District Court Judge was undertaking an evaluative role which required compelling reason for her views to be departed from: McDermott v Wallace [2005] 3 NZLR 661 (CA). It had not been shown that the Judge‘s decision was plainly wrong.

  245. It seems very unlikely on the Privy Council‘s approach that the established approach of restraint in awarding exemplary damages could be maintained. That approach is built on the principle that there is a general need for intentional or conscious wrongdoing for exemplary damages to be awarded. The tendency will be for courts to apply a test which is solely that of outrageousness. Conscious recklessness will be an alternative test that ceases to be applied.

  246. For these reasons, I am satisfied that New Zealand should adopt an approach that confines all awards of exemplary damages to cases where the defendant acted intentionally or was subjectively reckless. I would not confine this principle to cases involving claims for such damages arising from personal injury.

    Conclusion

  247. It follows that I am in agreement with Tipping J as to the test to be applied in respect of the appellant‘s claim for exemplary damages, as to the disposition of the appeal, and as to the future course of the proceeding.

    Wilson J

  248. This appeal raises three issues. Is the appellant‘s claim for exemplary damages barred by s 317 of the Accident Compensation Act 2001[76] or by s 86 of the State Sector Act 1988, in combination with s 6(1) of the Crown Proceedings Act 1950? If not, should this Court revisit the decision of the Privy Council in Bottrill v A [2002] UKPC 44, [2003] 2 NZLR 721? If it should, what test should be applied in determining whether the appellant can recover exemplary damages?

  249. These issues are all comprehensively discussed in the judgments of the other members of the Court. I will not repeat what they have said in any detail but, because of the importance of the issues, I set out as follows my views and why, in summary, I have reached them.

  250. I agree with Tipping and McGrath JJ[77] that neither s 317 nor s 86 and s 6(1) bar the present claim. It appears to me however that, in order to reconcile s 86 and s 6, it is necessary either to read s 86 as leaving employees who act in good faith liable to be sued, but not liable to the Crown, or to read s 6(4) as not applying to s 86. Neither interpretation appears to me to accord with the plain meaning of the relevant words. My view, like that of McGrath J, is that the conflict between the sections should be resolved by giving full effect to the words of s 86, with the consequence that public servants acting in good faith do not incur any personal liability but the Crown is liable for their acts or omissions. To leave those employees exposed to being sued seems to me to be not only contrary to the words of s 86 but also to be an outcome which Parliament would not have intended in enacting that section against the background of s 6.

  251. Judgments of the Privy Council on New Zealand appeals bind the other courts in this country. Other Privy Council decisions prior to the establishment of the Supreme Court are probably also binding although, as the Court of Appeal noted in R v Chilton [2006] 2 NZLR 341 (CA) at [112]–[113], the point is not free from doubt. What is not in doubt however is that Privy Council decisions, whether on New Zealand appeals or otherwise, are not binding on this Court. Indeed, as McGrath J points out (at [208]), one of the purposes of establishing the Supreme Court was to enable important legal matters "to be resolved with an understanding of New Zealand conditions, history, and traditions" (Supreme Court Act 2003, s 3(1)(a)(ii)). To fulfil that purpose, the Judges of this Court must be prepared to depart, if they think it right to do so, from a position taken by Judges from the other side of the world whose understanding of New Zealand conditions was necessarily limited. This Court will always have regard to the value of certainty and consistency in the law and will give great weight to a relevant Privy Council decision, but is free to depart from it. Because the courts which have considered a matter on appeal to the Supreme Court will have been bound to apply a relevant Privy Council judgment, they are unlikely to have expressed their views on whether it should be followed and this Court will therefore not have the benefit of those views. That is regrettable but unavoidable.

  252. There are a number of reasons which in combination make the present appeal a paradigm for consideration of whether or not to follow the Privy Council.

    • First, their Lordships reversed in Bottrill a judgment of the Court of Appeal (Bottrill v A [2001] 3 NZLR 622 (CA)) reached by a majority of four to one (Richardson P and Gault, Blanchard and Tipping JJ; Thomas J dissented).

    • Secondly, the advice of the Board was given by a narrow majority, three to two (Lord Nicholls, Lord Hope and Lord Rodger were the majority; Lord Hutton and Lord Millett the minority).

    • Thirdly, the divergence of judicial thought in this area of the law has been such that the views of the majority and minority were each supported by persuasive authority.

    • Fourthly, Bottrill involved a consideration of the accident compensation scheme, which is unique to New Zealand.

    • Fifthly, Bottrill has not so far as I am aware been relied on as a precedent to any significant degree, in part because the hearing of proceedings in which it would have been relevant has been deferred pending the outcome of the present appeal.

  253. What then should be the test for the award of exemplary damages? This question comes down to a choice between two possible approaches. Exemplary damages could be awarded whenever the conduct of the defendant can be characterised as "outrageous", as the Chief Justice has concluded, or they could be confined to where the conduct of the defendant is intentional or subjectively (consciously) reckless, as Blanchard, Tipping and McGrath JJ have held. For the following reasons, I prefer the latter approach.

  254. Exemplary damages are awarded to punish the defendant. The focus should therefore be on the mind of the defendant, in order to decide whether punishment is deserved. This focus is achieved by adopting a test of intention or subjective recklessness, but not by a test of outrageousness which requires an assessment of how the defendant‘s conduct appears to others. That assessment may well be influenced by how those making it view the consequences, possibly unintended and unforeseeable, of the conduct.

  255. As a related point, whether conduct appears outrageous is a question of impression. In contrast, whether the defendant acted intentionally or with subjective recklessness, by deliberately running a risk, is a more precise and certain standard. While different triers of facts, be they judges or juries, may reach different conclusions on the same facts as to whether that standard is satisfied, the possibility of like cases not being decided alike will be significantly increased if the test is whether those making the assessment are outraged by what the defendant has done or failed to do: McDermott v Wallace [2005] 3 NZLR 661 (CA), referred to by McGrath J at [244], provides a striking example. Undesirably, the outcome of the trial will therefore be much less predictable and settlement will be more difficult.

  256. As I have already noted, there has been a wide divergence in the views expressed by judges as to what the test should be. It seems to me however that the weight of authority supports a test of intention or subjective recklessness.[78]

  257. The value of deterrence is a possible justification for exemplary damages. That consideration cannot however support the adoption of a test of "outrageousness". An award of exemplary damages might deter the defendant from intentionally repeating the punished conduct or knowingly running the same risk again, or deter others who were aware of the award from knowingly acting in a similar way. Those who were not conscious of the risk could not however be deterred because the question of whether they should proceed, notwithstanding the risk, would not even cross their mind.

  258. Exemplary damages are anomalous in that, if awarded, they confer a windfall on a plaintiff who has already been compensated to the extent prescribed by law (the accident compensation scheme where there is personal injury, otherwise the common law). To the extent that the conduct of a defendant has aggravated the damage to a plaintiff entitled to compensatory damages, the plaintiff can and should be compensated through an award of aggravated damages. If exemplary damages were now being introduced into our law, there would be a good argument that, if awarded, they should be paid to the State. Being punitive in nature, such damages are closely analogous to fines, which are paid to the State (in contrast to sentences of reparation which are paid to victims and are compensatory in nature) (Sentencing Act 2002, s 32). A restrictive rather than an expansive approach should therefore be adopted in considering the scope of exemplary damages.

  259. For these reasons, I think that the test should be whether a defendant has acted intentionally or with subjective recklessness. I would not confine this principle to claims resulting from personal injury. I cannot see any reason for doing so, and it would I think be undesirable to leave undecided the question of whether the principle is of more general application.

  260. Finally, I record my view that it is right that the test to be applied has been settled by this Court prior to the trial of the appellant‘s claim. The question of what should be the test does not turn in any way on the resolution at trial of questions of fact. If the Bottrill test remained the law, it would have to be applied at trial. If the appellant succeeded and this Court subsequently changed the test, a new trial would be required. The parties might agree that the claim should be considered at trial on the alternative bases of "subjective recklessness" and "outrageousness" but that would further complicate what will in any event be a complex trial. If there is to be a jury trial, settling the issues on the alternative bases and explaining them to the jury would be extraordinarily difficult exercises.

  261. I therefore agree that the appeal should be allowed, with the consequences proposed by Tipping J (at [180]).


[1] As it was described by the Solicitor-General in argument and as adopted by Tipping J at [75].

[2] By Lord Wilberforce in Broome v Cassell & Co Ltd [1972] AC 1027 (HL) at 1114; and Cooke P in Re Chase [1989] 1 NZLR 325 (CA) at 332–333. See Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118 at 130–131 per Taylor J and at 149–150 per Windeyer J. And see Nicholas McBride "Punitive Damages" in Peter Birks (ed) Wrongs and Remedies in the Twenty-First Century (New York, 1996) 175, who says at 195 that it is "a conclusion masquerading as an argument".

[3] As described by Taylor J and Windeyer J in Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118 at 136–139 and 152–153.

[4] See also Cooke J in Donselaar v Donselaar [1982] 1 NZLR 97 (CA) at 106.

[5] Although this case concerns personal injury, as other members of the Court acknowledge, their reasoning is equally applicable to all cases of negligence. See Blanchard J at footnote 172 at [68]; Tipping J at [178]; McGrath J at [246]; and Wilson J at [259].

[6] Compare Tipping J at [108]; and see McGrath J at [212].

[7] This is further explained below at [29].

[8] As further explained below at [30].

[9] For reasons explained below at [32]–[41].

[10] See Tipping J at [158]–[161].

[11] The Act formerly entitled the Injury Prevention, Rehabilitation, and Compensation Act 2001.

[12] And recorded in footnote 11 of my reasons for that judgment.

[13] Williams and Hepple consider this condition to be "one of the worst that can afflict a legal system": Glanville Williams and BA Hepple Foundations of the Law of Tort (London, 1976) at 28.

[14] Ontario Law Reform Commission Report on Exemplary Damages (1991); Law Commission for England and Wales Aggravated, Exemplary and Restitutionary Damages (LC247, 1997); Irish Law Reform Commission Report on Aggravated, Exemplary and Restitutionary Damages (2000).

[15] In another decision of that Court, Port Stephens Shire Council v Tellamist [2004] NSWCA 353, (2004) 135 LGERA 98, Ipp JA at [402] considered that exemplary damages could be appropriate "where the conduct of the defendant is neither malicious nor conscious wrongdoing".

[16] Ibbett arose out of trespass to land and assault; Delly was a case of unlawful arrest.

[17] A point made by Spigelman J in the New South Wales Court of Appeal in Ibbett at [41].

[18] In Australia in Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118; in Canada in Vorvis; in New Zealand in Taylor v Beere, Donselaar and Re Chase [1989] 1 NZLR 325 (CA).

[19] This distinction between "what the plaintiff ought to receive" and "what the defendant ought to pay" was made by Lord Hailsham in Broome v Cassell at 1077–1078.

[20] In addition to statutory jurisdiction, the categories of exemplary damages were restricted to cases of oppressive, arbitrary or unconstitutional action by servants of the government and cases in which the defendant‘s conduct was calculated to make a profit.

[21] The House of Lords reinstated a claim for exemplary damages for misfeasance in public office which had been struck out on the basis that no such award had before been made in a case of misfeasance in public office.

[22] See also, in relation to defamation The Herald & Weekly Times Ltd v Popovic [2003] VSCA 161, (2003) 9 VR 1 at [425], and Taylor v Beere at 84 per Cooke J.

[23] In Vorvis there are suggestions the actionable wrong would need to constitute a tort, but in Whiten Binnie J considered that the requirement for an actionable wrong was not restricted to tortious conduct (at [79]).

[24] Compare Tipping J at [110].

[25] Compare Blanchard J at [65].

[26] See for example Taylor v Beere at 90 per Richardson J; Donselaar at 115 per Somers J; and Kuddus at [63] per Lord Nicholls.

[27] Lord Hutton observed that in cases where security forces, exercising powers which necessarily reduce the freedom of individuals in order to combat terrorism, act in gross breach of discipline and commit an unlawful act which is oppressive or arbitrary "the power to award exemplary damages .... serves to uphold and vindicate the rule of law because it makes clear that the courts will not tolerate such conduct". Such awards, he thought, "bring home to officers in command of individual units that discipline must be maintained at all times".

[28] Ontario Law Reform Commission at 38 and 69; Law Commission for England and Wales at 100; Irish Law Reform Commission at [1.73]–[1.75]. See also the discussion of Thomas J in Bottrill (CA) at [102]–[110].

[29] See also Re Chase and Auckland City Council v Blundell [1986] 1 NZLR 732 (CA).

[30] That exemplary damages could not be awarded in respect of conduct constituting a crime because civil punishment would cut across criminal punishment: Daniels v Thompson [1998] 3 NZLR 22 (CA).

[31] See, for example, Knuller (Publishing, Printing and Promotions) Ltd v Director of Public Prosecutions [1973] AC 435 (HL) at 455 per Lord Reid and R v Kansal (No 2) [2001] UKHL 62, [2002] 2 AC 69 at [27] per Lord Steyn.

[32] Planned Parenthood of Southeastern Pennsylvania v Casey 505 US 833 (1992) at 854 per O‘Connor, Kennedy and Souter JJ for the Court, drawing on Benjamin Cardozo The Nature of the Judicial Process (New Haven, 1991) at 149.

[33] As indicated at [36] I do not consider that is the position here, where the claim cannot be struck out and the point of law can be raised after trial if it in fact arises.

[34] In Broome v Cassell at 1085 Lord Reid made the point that it is not the function of judges "to frame definitions or to lay down hard and fast rules". Their function is rather to enunciate principles "and much that they say is intended to be illustrative or explanatory and not to be definitive".

[35] Whether exemplary damages were available only in respect of causes of action in which they had been awarded before Rookes v Barnard in 1964.

[36] Compare Blanchard J at [47]; Tipping J at [79]; and McGrath J at [213].

[37] Thus, Lord Kilbrandon in Broome v Cassell at 1134 explained the rationale of exemplary damages as protection of the public interest. And justification for vicarious liability is expressed by Glanville Williams and BA Hepple Foundations of the Law of Tort (London, 1976) at 114 as being based on the ability of the employer "to change his system of work, or to change his staff, in such a way to minimise acts of negligence". On such views, the public interest may be in deterrence and incentivisation of those vicariously liable.

[38] Thus the American Law Institute‘s Restatement of the Law of Torts took the view that it would be improper "to award punitive damages against one who himself is personally innocent and therefore liable only vicariously" (2nd ed, Minnesota, 1979) § 909. And in Kuddus two of the members of the House of Lords considered that vicarious liability should be modified in respect of exemplary damages (at [47] per Lord Mackay; and at [108] per Lord Scott).

[39] It is admitted that there was negligence on the part of the probation officer.

[40] Rookes v Barnard [1964] AC 1129 (HL) at 1226 (oppressive, arbitrary or unconstitutional action by the servants of the government). The New Zealand courts have not confined exemplary damages awards to the Rookes v Barnard categories.

[41] See for example Australian Consolidated Press Ltd v Uren [1969] 1 AC 590 (PC); Invercargill City Council v Hamlin [1996] 1 NZLR 513 (PC); W v W [1999] 2 NZLR 1 (PC) and Lange v Atkinson [2000] 1 NZLR 257 (PC).

[42] Stephen Todd "A New Zealand Perspective on Exemplary Damages" (2004) 33 CLWR 255; Allan Beever "Punishing the innocent" [2002] NZLJ 359; Andrew Beck "Exemplary damages: the snake uncoils" [2002] NZLJ 361; and see in England Andrew Phang and Pey-Woan Lee "Exemplary Damages – Two Commonwealth Cases" (2003) 62 CLJ 32. Contrast J Manning "'Never say never‘: exemplary damages in negligence" (2003) 119 LQR 24.

[43] This passage is entirely consistent with the approach earlier taken in Lamb v Cotogno (1987) 164 CLR 1 at 9 where the judgment of the Court cited with approval Brennan J‘s statement in XL Petroleum (NSW) Pty Ltd v Caltex Oil (Australia) Pty Ltd (1985) 15 CLR 448 at 471 that "an award of exemplary damages is intended to punish the defendant for conduct showing a conscious and contumelious disregard for the plaintiff‘s rights and to deter him from committing like conduct again". Further back, the same requirement, drawn from the first edition of JW Salmond The Law of Torts (Stevens & Haynes, London, 1907) at 102 had been endorsed by Windeyer J in Uren v John Fairfax & Sons Pty Ltd (1966) 117 CLR 118 at 154.

[44] See the contrasting views in Broome v Cassell & Co Ltd [1972] AC 1027 (HL) at 1087 per Lord Reid and at 1114 per Lord Wilberforce.

[45] See Glanville Williams and BA Hepple Foundations of the Law of Tort (Butterworths, London, 1976) at 72 ("Exemplary damages have no place in the unintentional torts, in particular negligence."). The position of the criminal courts is of course very different for it is obviously necessary in the public interest that there should be criminal sanctions inflicted on behalf of the State for negligent conduct which causes injury or the risk of injury to others.

[46] But noting Atiyah‘s concern about the effects of over-deterrence: Patrick Atiyah The Damages Lottery (Hart Publishing, Oxford, 1997) at 164.

[47] Intentional conduct which causes harm may not be of an outrageous character.

[48] Glanville Williams "The Aims of the Law of Tort" (1951) 4 CLP 137 at 166. Employers might still want cover for their vicarious liability for such advertent acts of their employees, pending an authoritative decision on whether and when exemplary damages are claimable in this country against an employer or principal.

[49] It seems that it is not contrary to public policy for insurance to cover claims for indemnity against exemplary damages: Lancashire County Council v Municipal Mutual Insurance Ltd [1997] QB 897 (CA) at 908.

[50] That would be contrary to the philosophy that tort law should generally operate in a way which promotes efficiency. See, for example, David Allen, John Hartshorne and Robyn Martin Damages in Tort (Sweet & Maxwell, London, 2000) at [2-036].

[51] It is only because this case is concerned solely with personal injury that I have expressed my conclusion in that limited way. It follows logically that exemplary damages should not be available as well against an unconscious wrongdoer for negligently causing other forms of injury.

[52] It would not be appropriate in what would already be a difficult case for the Judge to ask the jury to answer issues on an alternative basis.

[53] As Cooke P put it in Auckland City Council v Blundell [1986] 1 NZLR 732 (CA) exemplary and punitive damages are different words for the same thing. The damages are meant to make an example of the defendant by punishing him.

[53] For example in s 28 of the Defamation Act 1992.

[54] See, for example, Richardson J in Donselaar at 109. See also Somers J in the same case at 113 where his Honour said that exemplary damages are intended to punish the defendant and their award seeks to achieve recognised objects of the criminal law – deterrence and retribution.

[55] See Stephen Todd (ed) The Law of Torts in New Zealand (5th ed, Brookers Ltd, Wellington, 2009) at [25.2.10] citing Midland Metals Overseas Pte Ltd v The Christchurch Press Co Ltd [2002] 2 NZLR 289 (CA). See also Attorney General v Niania [1994] 3 NZLR 106 (HC) at 111.

[56] See Allan Beever "The Structure of Aggravated and Exemplary Damages" [2003] 23 OJLS 87 at 97ff. The author makes the point that while some of the additional perceived purposes of exemplary damages are purposes of tort law generally, they are not shown to have a sufficiently clear connection with the award of exemplary damages.

[57] Is there, for example, to be a difference between conduct which is unacceptable to society and conduct which is altogether unacceptable to society?

[58] The concept of contumelious disregard must imply at least advertence to risk. It is not consistent with inadvertence, however outrageous.

[59] See Taunoa v Attorney-General [2007] NZSC 70, [2008] 1 NZLR 429 at [321]. See also Orhan v Turkey (25656/94) Section I, ECHR June 18, 2002 at [448] where it was said "The Court notes that it has rejected on a number of occasions, recently and in Grand Chamber, requests by applicants for exemplary and punitive damages"; and R Carnwath "ECHR Remedies from a Common Law Perspective" (2000) 49 ICLQ 517 at 521: "The Strasbourg jurisprudence does not support the award of .... 'exemplary‘ damages".

[60] I am not persuaded by Geoff McLay‘s article ("The New Zealand Supreme Court, the Couch case and the future of governmental liability" (2009) 17 TLJ 77) that this traditional reluctance should be cast aside, as he suggests (at 98), in favour of using exemplary damages as a way of "getting the government to do its job better". That strikes me as a most difficult suggestion and no basis upon which to decide between the competing contentions in this case.

[61] In that case Knox CJ (at 77) emphasised the need for "conscious wrongdoing". It was that kind of wrongdoing that had to be in contumelious disregard of the plaintiff‘s rights.

[62] Port Stephens Shire Council v Tellamist Pty Ltd [2004] NSWCA 353, (2004) 135 LGERA 98; Ali v Hartley Poynton Ltd [2002] VSC 113, (2002) 20 ACLC 1006; TCN Channel Nine Pty Ltd v Ilvariy Pty Ltd [2008] NSWCA 9, (2008) 71 NSWLR 323; Harris v Digital Pulse Pty Ltd [2003] NSWCA 10, (2003) 56 NSWLR 298; Crump v Equine Nutrition Systems Pty Limited Trading As Horsepower [2006] NSWSC 512; SB v State of New South Wales [2004] VSC 514, (2004) 13 VR 527.

[63] Other cases include State of New South Wales v Riley [2003] NSWCA 208, (2003) 57 NSWLR 496 and State of New South Wales v Delly [2007] NSWCA 303, (2007) 70 NSWLR 125.

[64] That case concerned punitive damages for breach of contract; however it has been broadly applied, including in negligence cases.

[65] On this point, which does not have to be resolved here, see the discussion by Professor Stuart Anderson in the F W Guest Memorial Lecture on 6 August 2008: "'Grave injustice‘, 'despotic privilege‘: the insecure foundations of crown liability for torts in New Zealand" (2009) 12 Otago L Rev 1.

[66] The words used in the statutory provision equivalent to s 6(1).

[67] Tipping J refers at [177] to s 6(4) of the Crown Proceedings Act. I consider that provision to be of limited scope. The purpose and effect of s 6(4) is no more than to treat proceedings against the Crown as though the proceedings were against the responsible government department or officer of the Crown. Just as a government department cannot rely on an employee‘s immunity under s 86 to escape tortious liability, neither can the Crown escape tortious liability by relying on s 86.

[68] Section 5 was amended prior to commencement of the 1972 Act and, in that form, was considered in Donselaar .

[69] A similar view was taken in Australia when appeals to the Privy Council against decisions of courts exercising federal jurisdiction were terminated: Viro v The Queen (1978) 141 CLR 88 at 120 per Gibbs J and at 135 per Mason J.

[70] As this case does not concern the question of whether exemplary damages could be awarded where deliberate but non-tortious outrageous conduct follows the commission of an inadvertent tort, that issue is better left to be determined another day.

[71] See also Prebushewski v Dodge City Auto (1984) Ltd 2005 SCC 28, [2005] 1 SCR 649 at [25] and [37], where the Supreme Court held that a statutory provision which allowed for awards of exemplary damages for "wilful" violations of the Act was less onerous than the "more exacting" common law test for exemplary damages which required outrageousness in addition to advertent conduct.

[72] See also Performance Industries Ltd v Sylvan Lake Golf & Tennis Club Ltd 2002 SCC 19, [2002] 1 SCR 678, where the Supreme Court reiterated that exemplary damages would only be awarded in exceptional cases and where the award would serve a rational purpose.

[73] This is confirmed by the High Court‘s subsequent judgment in New South Wales v Ibbett [2006] HCA 57, (2006) 229 CLR 638 at [47] and in particular footnote 56.

[74] Carol Harlow "A Punitive Role for Tort Law?" in Linda Pearson, Carol Harlow and Michael Taggart (eds) Administrative Law in a Changing State (Hart Publishing Ltd, Oxford, 2008) 247 points out that the test for exemplary damages should be co-extensive with its rationale of punishment and that it would not be right to punish a defendant in the absence of a guilty mind: see 252–253, 269 and 270.

[75] These points are made in the valuable article: Peter Birks "Three Kinds of Objection to Discretionary Remedialism" (2000) 29 UWAL Rev 1 at 7 and 16.

[76] For convenience, I use the current title of the accident compensation legislation, the name of which has come full circle from the Accident Compensation Acts 1972 and 1982 to the Accident Rehabilitation and Compensation Insurance Act 1992, the Accident Insurance Act 1998, the Injury Prevention, Rehabilitation, and Compensation Act 2001, and now back to the Accident Compensation Act 2001.

[77] Tipping J at [84]–[90] (s 317) and [173]–[174] (s 86/s 6(1)); and McGrath J at [184]–[193] (s 86/s 6(1)) and [199]–[203] (s 317).

[78] See the discussion by Blanchard J at [52]–[56]; Tipping J at [92]–[94], [136]–[149] and [162]–[171]; and McGrath J at [233]–[237].


Representations

B P Henry (M/s Dennis Gates, Whangaparaoa) for Appellant.

D B Collins QC Solicitor-General, J C Pike and R Kirkness (Crown Law Office, Wellington) for Respondent.


all rights reserved