Barker, Kit --- "Unfamiliar Waters: Negligent Advocates, Egregious Errors and Lost Chances of Acquittal" [2005] UQLRS 1; (2005) 24(2) University of Queensland Law Journal 469-479

Last Updated: 10 December 2008

UNFAMILIAR WATERS:
NEGLIGENT ADVOCATES, EGREGIOUS ERRORS AND LOST
CHANCES OF ACQUITTAL

KIT BARKER*

Abstract: The negligent misrepresentation of a client’s case at criminal trial can lead to wrongful incarceration. Although Australia persists in extending an immunity to criminal advocates in such instances, more liberal jurisdictions offer cases which are instructive in demonstrating the way in which difficult issues relating to breach of duty and causation can be handled. This article critically appraises the approach taken to these issues in the Canadian case of Folland v Reardon [2005] 194 OAC 201 (Ont CA), suggesting both that a standard of reasonable competence (not egregious error) is the appropriate one to apply to criminal advocates; and that damages for the lost chance of an acquittal may be appropriate in at least some cases in which a client’s conviction has been set aside.

I INTRODUCTION

In the recent case of Folland v Reardon,[1] the Ontario Court of Appeal considered two key issues relating to the negligence liability of criminal advocates for the misconduct of cases in court, neither of which had been clearly determined in Canadian law. Firstly, what standard of care do they owe their clients? Is this the normal standard of reasonable competence we have come to expect of professionals generally, or is it set at some lower level, in view of the particular, critical role they play in the administration of justice? Secondly, is it open to a wrongfully convicted client to frame her claim against the advocate not in terms of the harm she has suffered through the conviction itself, but rather in terms of the loss of a substantial chance of acquittal? The impact of the answer to the latter question will be felt most acutely in cases in which she is unable to prove that the mishandling of her defence caused the jury to convict, but where she can nonetheless show that it made that conviction significantly more probable. It will also affect the way in which her damages are quantified where the notional chance of acquittal she lost is greater than fifty per cent.[2]

Australian observers may consider both questions somewhat academic in the light of the High Court’s recent decision to affirm advocates’ immunity against suit in criminal trials.[3] But the pressures against this immunity are increasing all the time and getting ever closer to home.[4] The way in they are answered in jurisdictions in which immunity is now a thing of the past will prove central to the arguments of those who seek to persuade the High Court to change its mind in the future, as they almost certainly will. Australian judges will then need to be assured that the lower-level concepts of breach, causation and damage are capable of accommodating the various policy concerns which currently underpin their protective attitude towards trial lawyers, if that attitude is to be relaxed in favour of a more case-sensitive approach. From this point of view, the instant case not only addresses questions of contemporary significance for Canadian and English courts, but represents an important opportunity for Australia to test unfamiliar waters. It illustrates both the strengths of the case-sensitive approach and some of the complexities it entails.

II FACTS AND DECISION

Reardon was engaged to act for Folland at his trial for sexual assault, where the issue of the identity of the victim’s attacker proved central. Folland claimed that his defence had been bungled in several key respects, one of which related to his advocate’s failure to collect DNA evidence from the victim’s boyfriend, Harris, who had also been in the building at the time of the attack and who was therefore another potential suspect. Folland was convicted, but his conviction was subsequently set aside on the basis of fresh DNA evidence of the very sort which it was alleged ought to have formed part of his defence. The Crown elected not to pursue a retrial. The disgruntled Folland sued Reardon for three years’ wrongful incarceration.

This is precisely the type of controversial fact-configuration which would attract immunity in Australia. In Canada, there is no immunity as such,[5] but Folland’s claim was initially struck out on the grounds that it failed to disclose genuine issues for trial, being bound to fail on the issues of breach and causation.[6] The Court of Appeal disagreed and allowed the case to proceed. In doing so, it concluded that the trial judge had been guilty of two significant errors. Firstly, he had assumed it was necessary for Folland to prove some ‘egregious error’ on Reardon’s part in order to establish his negligence, as opposed to simply a departure from the standards of the ordinary, competent defence counsel. Secondly, he had underestimated the potential significance of the evidence on the causation question, though (importantly) he had been right to assume that Folland needed to prove that he would not have been convicted, if his defence had been properly conducted. The Court of Appeal thus expressly rejected any possibility of clients suing their advocates for the lost chance of an acquittal. In this detail, it declined to follow a recent English lead[7] and may, with respect, have fallen into error.

III BREACH- THE FALLACY OF EGREGIOUS ERROR

If the occasional references in the Canadian case law to ‘egregious error’[8] posited a lower standard of care for trial advocates that that of reasonable care, this would tend to replicate within the breach inquiry a strategy of protectionism almost as complete as the Australian immunity, albeit in a less visible guise. In fact, however, close reading reveals that their intention has been to distinguish mere errors of judgement (reasonable mistakes) on the one hand from unreasonable incompetence on the other, not, actually, to lower the required standard below that of the reasonably competent defence counsel.[9] Since there will always be a range of reasonable responses to questions of professional judgement, the concept of egregiousness merely sought to identify a set of decisions lying outside the range reasonably open to the advocate. It did not mean that she is negligent only if she selects from the range of unreasonable responses one which is of a different order of magnitude to the rest.

At the policy level, there is of course scope for debate as to whether particular concerns pertain to the work of advocates, which require either a general distinction between them and other professionals, or between in-court and out-of-court work. These arguments mirror those which have arisen in the wrangle over immunity and it is scarcely surprising that, having failed to convince Canadian judges to immunise the advocate against liability completely, the same arguments have also failed to persuade them of the merits of a lower standard of care either for the profession generally, or for the conduct of cases in court. The case for a distinct approach towards the profession as a whole is virtually impossible to make, given the parallels drawn between the respective positions of legal and medical practitioners in Arthur Hall v Simons[10] and Demarco v Ungaro.[11] Furthermore, even if Krever J in Demarco did contemplate the possibility of a lower standard of care in relation to in-court work[12] (perhaps by way of a trade off for the lack of any immunity), any distinction between ‘solicitor’ and ‘barrister’ work would be difficult to draw in practice and is highly questionable in principle. The vulnerability of this distinction is patent in Ryan[13] itself, where McHugh J cannily spotted the need to align the scope of any protective strategy (in that case in its immunity guise) with the reasons underpinning it. If, as the High Court now maintains, the reason for containing liability is to ensure finality in litigation, even out-of court work should logically attract protection, where a claim would tend to re-open issues previously litigated.[14] In the event, then, even if Ryan’s ‘finality’ logic is accepted as a valid rationale for restricting advocate liability, any hard and fast distinction between in-court and out-of-court work seems impossible to justify.

The Court of Appeal is therefore right both to affirm that the ordinary standard of reasonable competence is applicable to criminal advocacy work of all types,[15] and to suggest that the potentially misleading language of ‘egregious error’ be dropped.[16] This is consistent with dicta in Arthur Hall v Simons[17] and avoids the unhealthy implication that lawyers are somehow treated preferentially in matters of professional standards. Residual concerns will no doubt persist that liability for in-court work may precipitate defensive tactics, which will in turn work against the interests of the administration of justice by prolonging trials. However, the availability of liability insurance, combined with advocates’ duties to the court and their educated appreciation of the realistic nature of the legal standards expected of them should safeguard sufficiently against this possibility. If there is still a ‘defensive practice’ problem after all of this, because the levels of insurance available economically are insufficient to meet likely liabilities, ethics rules could be changed to permit advocates to refuse to take a case in such circumstances,[18] or (better perhaps, though controversial in a different way) mechanisms could be found to fund the additional insurance cover either from clients or, if they are pubicly funded, the public purse. The distinct problem of vexatious litigation, if it exists at all, is better addressed via procedures for the summary disposal of claims, than by lowering professional standards or applying blanket immunities. Both of these devices tend to undercut the very confidence in the administration of justice which it is sometimes thought they might exist to promote. From this point of view, the Court of Appeal’s readiness in the instant case to allow the question of negligence at least to be seriously examined stands in marked contrast to the closed-door policy of Australian courts and, it is submitted, casts Canadian law in a rather more respectable light.

III LOST CHANCES OF ACQUITTAL

More controversial is the Court of Appeal’s assumption that a client must prove on a balance of probabilities that he would have been acquitted, but for his advocate’s negligence.[19] On the facts, it was assumed that Folland might do this. It is also true to say that he did not plead the loss of any chance of acquittal. Nonetheless, the Court of Appeal felt it important to address the argument because of its potential consequences for his case at trial. If accepted, it would entitle him to some damages, even if he was unable to prove Reardon’s negligence caused the jury to convict him. Conversely (and here is the double-edge of this particular sword) any damages award would be calculated in proportion to the percentage chance of acquittal lost, even if this proved to be higher than fifty per cent.

The availability of an action for the lost chance of an acquittal has never formally been determined in Canada,[20] though it is fair to say that the reformulation of ‘gist’ damage in terms of lost chances has received a much cooler reception in that jurisdiction[21] than in either England[22] or Australia.[23] It has met significant resistance in the context of claims for personal injury and medical negligence, but commands broader, cross-jurisdictional support in cases involving claims against negligent solicitors and others for lost civil litigation, or lost commercial opportunities.[24] Its clear acceptance in these instances and in one closely analogous criminal case in England[25] begs the question why it was rejected so firmly here.

The Court of Appeal’s judgement discloses four reasons, none of which, with respect, withstands detailed scrutiny. More generally, if one draws down distinctions from the great volume of judgements and academic writings on loss of chance, none of them supports its rejection in criminal cases, at least in cases such as the present one, where the claimant’s conviction has been set aside.

A Error 1 - Chances of acquittal of less than fifty per cent are worthless

A first objection was that any chance of acquittal was valueless to Folland unless he could also prove on a balance of probabilities that he would have been acquitted, if Reardon had acted properly:

A mere chance of an acquittal had no real value to Folland in that it would not have avoided conviction, imprisonment and the damages flowing from those events. If Reardon’s negligence only decreased Folland’s chance of acquittal but was not a “but for” cause of his conviction, ...[it] resulted in nominal damages only to Folland.[26]

This situation is contrasted with that in which a client loses a prospect of success in a civil trial,[27] where the chance of success is said to have a ‘settlement value’ of its own, independent of the merits of the case at trial.

Precisely this type of reasoning was rejected by the High Court of Australia in the Poseidon[28] case. It errs in its premise that chances have no independent value of their own, but become valuable only where the end consequence of which they represent a possibility is itself probable at a level above fifty per cent. That premise is not only logically suspect, but is contradicted by the case law.[29] Moreover, the distinction between lost civil and criminal litigation is spurious: if the valuation argument were logically correct, civil cases could not help but be caught by it. The truth is that clients value chances of acquittal, whatever their prospects. That is why they are happy to pay for them.

B Error 2 - Chances of acquittal of less than fifty per cent are ‘unmeritorious’?

Reading between the lines, it seems probable that the court’s intention in denying value to chances of acquittal of less than fifty per cent was to avoid the potential embarrassment of a client claiming for the loss of an opportunity to advance an unmeritorious defence:[30]

If Folland were to set out to demonstrate that he would likely have been acquitted, but were to only establish a less than 50 per cent chance of an acquittal, by implication the trier of fact would have found it more likely than not that Folland had been properly convicted of sexual assault. Public policy would not countenance a damage award...when, on the findings of the trial court, Folland probably committed the crime with which he was charged.[31]

The proposed solution to this potential problem is that the court trying the negligence action should engage in a full examination of the merits of the criminal case as best it can and determine for itself whether the party seeking damages was guilty or innocent. If it concludes he was most likely guilty, he should obtain nominal damages only; if innocent, he can claim his full loss.

The policy behind this all-or-nothing approach is understandable, but there are several difficulties with it. Firstly, it has not generally prevailed in the context of civil litigation, where the danger of unmeritorious claims is equally applicable in logic, but where the loss of chances of less than fifty per cent is regularly compensated.[32] If a claimant can recover for the loss of a forty per cent chance of winning a potentially ‘unmeritorious’ civil case, why should he not be allowed to recover for the loss of a forty per cent chance of asserting an ‘unmeritorious’ defence to a criminal prosecution ?

Secondly, even if criminal cases are distinct in terms of public policy,[33] the danger of allowing a criminal to claim compensation for the consequences of a crime is not strictly confined to instances in which the client can prove a chance of acquittal of less than fifty per cent. It remains a risk even where his chance of acquittal was sixty, seventy or even eighty per cent. This is because it is in the nature of any adversarial system of justice that parties stand a (sometimes very good) chance of winning cases which, in truth, they probably ought to lose. Whilst the Court of Appeal’s approach laudably attempts to limit this risk as regards claimants with less than a fifty per cent chance of acquittal, it awards full damages to those whose (potentially also unmeritorious) defences had a more than fifty per cent chance. Some might consider that the award of full damages in the latter instance is at least as undesirable in public policy terms as apportioning all awards by reference to the chance of acquittal which can be shown to have been lost.

Thirdly (and this seems to be the crux) it is questionable in any case whether it is appropriate for a civil court trying an action against a negligent advocate to engage in a full determination of the merits of his client’s criminal defence. Strictly speaking, the court’s concern is whether the client has lost anything of substantial value as a consequence of negligent representation at trial, not whether or not he ought to have been convicted. The decision as to guilt or innocence on the ‘merits’ is one for the criminal process alone and cannot legitimately be replicated by civil courts, even if it were practical to do so. Whilst there is therefore a very good public policy reason for denying damages to a client whose defence has been conclusively rejected after full consideration by a criminal court, it is not immediately clear why a party whose conviction has been set aside as wrongful and who was denied the opportunity to put his defence to a jury should not be compensated for his loss of a substantial[34] chance of acquittal.

The result is that, if courts wish to draw clear a clear line in the sand to safeguard against the possibility of compensating criminals from the pockets of lawyers, they may be better to do so by denying claims (whether for loss of chance or for wrongful conviction) to all those whose convictions remain in place, rather than by rejecting claims for the loss of chances of acquittal of less than fifty per cent. In the vast majority of cases, this result is achieved in any case via the rules on abuse of process, which prevent any claim being brought which seeks to re-litigate a prior criminal conviction.[35] No distinct rejection of the concept of loss of chance seems to be needed on public policy grounds.[36]

C Error 3- No loss of chance claim where claimant can prove end-damage on a balance of probabilities

A third reason given for rejecting loss of chance on the facts was that Folland was likely in any event to be able to meet the normal burden of proving his conviction was caused by Reardon’s’ negligence on a balance of probabilities.[37] This raises serious questions about the scope of the argument, but is again questionable. It is true that in Acton,[38] the claimant’s chance of acquittal was ultimately assessed at only fifty per cent. There are, however, a host of English and Australian cases involving lost civil litigation occasioned by solicitor negligence, where damages have been awarded in respect of the loss of chances of greater than fifty per cent.[39] Whilst courts are understandably reluctant to accept the argument in the ‘standard’ negligence case on account of the additional complexities this would entail, it is therefore inapposite to assume that the fifty per cent marker is an arbitrary boundary beyond which the strategy may not pass. There will certainly be limits to the type of case in which is acceptable to reformulate ‘gist’ damage in this way, but these limits are not currently set at any particular percentage level of chance lost, nor is there any apparent reason why they should be. Though the jury is still very much out on this question, such limits are more likely to be determined by the nature and source of a court’s uncertainty about causal issues and by the underlying purposes of negligence rules.

D Error 4- Case not sufficiently indeterminate - retrial of issues possible

The final reason for rejecting loss of chance on the facts was that the issue of causation was insufficiently indeterminate to justify its use. The analysis should be confined to cases entailing ‘diffuse elements of pure chance, analogous to the non-specific factors of fate or fortune’[40] or to cases in which ‘perhaps because of the complexity of the variables involved or the unavailability of crucial evidence, it will be impossible to realistically assess what would have happened but for the defendant’s misconduct.’[41] This was not the case here because, even though the judgement as to whether or not Reardon’s negligence had caused Folland’s conviction entailed speculation about a hypothetical (how the jury would have decided his fate, had the defence been conducted properly) ‘the outcome of a criminal trial is knowable in the sense that an informed, objective, reasonable assessment can be made of what the outcome would be.’[42]

The distinction between deterministic and indeterministic cases was first advocated as a guideline for applying the loss of chance strategy in an acclaimed article by Helen Reece in 1996.[43] It has proven helpful, though not necessarily watertight in determining when the approach is used.[44] Importantly, however, the instant case seems to be precisely the type of ‘indeterministic’ case in which the author argues the approach should be deployed, because - involving as it does speculation about how humans would have made decisions in circumstances which did not in fact happen - its uncertainties are ‘objective’ (insurmountable even under full informational assumptions) not, as Doherty JA seems to assume, purely epistemological. The point is that no matter what information were to be made available to the Court of Appeal to assist it in assessing the merits or otherwise of Folland’s defence, it would still be impossible for him to show that a jury which never considered that defence in full would have decided the issue of his guilt or innocence any differently. Epistemological completeness - full knowledge of the facts and evidence - does not help to bridge this type of gap and is not, in fact, the central issue. This much seems to have been accepted by a number of authorities in both England and Canada, which suggest that the loss of chance analysis can be used even where it is supposedly possible to access all relevant facts and to hold a full, notional retrial of the original action.[45] Indeed, one of the strengths of the approach highlighted in the Acton case, is that it avoids the attempt to hold a full (and procedurally inapposite) retrial and considers the facts and evidence only indirectly, with a view to estimating the extent of the chance which has been lost.

E Other Possible Distinctions

It would be inappropriate in this, brief space to consider the merits of all current ideas as to when loss of chance claims ought to be available in private law,[46] but it is striking that it not just Reece’s analysis which supports the strategy’s use in the current type of case. Firstly, though the criterion is almost certainly misguided, there was clearly a contract between the parties.[47] Secondly, entailing, as it did, speculation about the hypothetical behaviour of third parties (the jury), the case sits comfortably with those theories which insist on maintaining the normal ‘all or nothing’ approach where the relevant uncertainties relate to hypothetical behaviour of the claimant himself,[48] or to questions of ‘past fact.’[49] Thirdly, to the extent that the content of the defendant’s primary duty is relevant in determining whether and when a claimant’s chances can constitute protected interests in tort, it is clearly arguable that the primary function of an advocate is to afford his client the possibility of a successful outcome in a criminal trial, not the outcome itself.[50] Fourthly, (though this distinction is again dubious), the defendant’s negligence in this case consisted at least in part in omission, not commission, which is reconcilable with some admittedly rather confusing dicta in the Maples[51] case. Fifthly and finally (an argument sometimes put in the context of medical negligence cases) application of the approach would have clear hortatory effects for barristers and help to reinforce professional legal standards.[52] The only distinction which might set this type of case apart from others in which the argument has succeeded, is that it did not involve a chance of a purely economic outcome,[53] but this distinction has itself been criticised and clearly rejected in Australia.[54] Far, then, from being a case in which it was obviously inapposite to apply the analysis, both the authorities and the arguments suggest that Folland was about as clear a case as one can get on current thinking.

IV CONCLUSIONS

The immunity which still attaches to advocates in Australia constitutes a blunt device for accommodating a variety of policy concerns which are dealt with in a more case-sensitive way in Canada. These concerns are at their height in cases involving the negligent conduct of criminal litigation in court, but the general strategy evident in Folland, of allowing individual issues relating to breach, causation and damage to proceed to trial seems preferable not simply because it targets these concerns more accurately, but also because it inspires greater public confidence in our systems and agents of justice. At the same time, it is clearly (and rightly) no guarantee of success for a disgruntled client. It will still be difficult for him to prove that the way in which a case is conducted in court was negligent, even applying the normal standard of reasonable care, and he will not be permitted to bring any claim to trial which seeks to re-litigate his conviction in another forum.

Once a conviction has been set aside, however, the above analysis suggests that it may be overly restrictive, given the insuperable objective difficulties that otherwise face the client in proving causation, to insist that he prove on a balance of probabilities that he would have been acquitted, if his defence had been properly conducted. Though the limits of arguments about loss of chance in private law are yet to be determined, cases of negligently conducted criminal litigation seem insufficiently distinguishable from their civil counterparts to justify a different approach, particularly given the fact that the nature and source of causal uncertainties are exactly the same. Though there are clearly added concerns about the fact or perception of criminals being entitled to claim damages, such concerns should not obstruct the civil law rights of those whose convictions have been set aside by due process of law and who have received sub-standard representation. In the end, then, it seems that Canadian law may have as much to gain from Australian and English learning on the question of loss of chance, as it has valuable insights to offer in relation the need for advocate immunity. Whether the possibilities it offers prove tempting enough to the High Court in the longer run to persuade it to change its mind, don its bathing suit and join the other swimmers remains to be seen. In the meantime, the waters do not seem as cold or as dangerous as it has perhaps been led to believe.

---

*Senior Lecturer in Law, University of Southampton.
1 [2005] 194 OAC 201 (Ont CA).
[2] Awards for chances of over fifty per cent will be proportionate the percentage chance lost. See generally McGregor on Damages (17th ed, 2003) [8-024]-[8-059].
[3] Ryan D’Orta-Ekenaike v Victoria Legal Aid [2005] HCA 12 (‘Ryan’).
[4] Immunity has never existed in Canada (see Demarco v Ungaro (1979) 95 DLR (3d) 385) (‘Demarco’) and was recently abolished in England in Arthur JS Hall & Co v Simons [2001] 1 AC 615 (HL). There are now signs of it crumbling in New Zealand: Sun Po Lai v Chamberlains [2005] NZCA 37 (immunity abolished in civil trials, left open in criminal proceedings).
[5] Ibid.
[6] [2004] OJ No 434 (Ont SCJ).
[7] Acton v Graham Pearce & Co [1997] 3 All ER 909 (Eng HC) (‘Acton’). Although decided before immunity was abolished in England, this case fell outside its boundaries.
[8] See, notably, Bertucci v Marchioni [2001] OJ No 2198 (Ont SCJ), aff’d [2002] OJ No 914 (Ont CA); Blackburn v Lapkin [1996] 28 OR (3d) 292; Garrant v Moskal [1985] 2 WWR 80, aff’d [1985] 6 WWR 21 (Sask CA). The source of the phrase is Krever J in Demarco (1979) 95 DLR (3d) 385.
[9] [2005] 194 OAC 201 (Ont CA) [38], [39]; Hagblom v Henderson (2003) 232 Sask R 81 (Sask CA) (Jackson JA).
[10] [2001] 1 AC 615 (HL), 680 (Lord Steyn), 609 (Lord Hoffman), 740 (Lord Hobhouse). The possible parallels relate to the need in either instance to ensure the maintenance of professional standards, dangers of defensive practice, possible conflicts of duty (legal or ethical) which may be encountered in the performance of professional functions, and the particular difficulties associated with on-the-spot decision making in pressured environments.
[11] (1979) 95 DLR (3d) 385, 405 (Krever J).
[12] Ibid. This seems unlikely: Hagblom v Henderson (2003) 232 Sask R 81 (Sask CA).
[13] [2005] HCA 12.
[14] Ibid [168].
[15] [2005] 194 OAC 201 (Ont CA), [41].
[16] Ibid [43].
[17] [2001] 1 AC 615, 726 (Lord Hope), 737 (Lord Hobhouse). See also Rondel v Worsley [1969] 1 AC 191 (HL), 248a-b, (Lord Morris - ‘reasonable degree of care and skill’), 248d; Saif Ali v Sydney Mitchell [1980] AC 198 (HL), 221a (Lord Diplock -‘outside the range of possible courses of action that in the circumsta nces reasonably competent members of the profession might have chosen to take’); Hett v Pun Pong (1890) 18 SCR 290, 292; Central Trust Co v Rafuse [1986] 2 SCR 147 (SCC), 208; Boudreau v Bennaiah (1998) 154 DLR (4th) 650 (Ont Gen Div), 693 (‘ordinary competent criminal counsel’) (appeal allowed in part on certain damages issues [2000] 46 OR (3d) 737 (Ont CA).
[18] In England and Wales, the Bar Council Professional Standards Committee is seeking permission to amend Bar Council rules to allow a barrister to refuse to take a case where possible negligence liabilities exceed levels of available insurance, in response to the abolition of barristers’ immunity for in-court work. See Causton, ‘Legal Profession Worth the Risk ?’ Legal Week (2004) 6 (43) 14.
[19] [2005] 194 OAC 201 (Ont CA) [62], [69]-[93]. See also Saif Ali v Sydney Mitchell [1980] AC 198 (HL), 222 (Lord Diplock).
[20] The issue appears to have been left open in the Saskatchewan Court of Appeal in Fischer v Halyk (2003) 229 DLR (4th) 67 (claim struck out as abuse of process). In England, claims are allowed: Acton [1997] 3 All ER 909 (Eng HC).
[21] See Lawson v Lafferiere [1991] 1 SCR 541 (SCC); Arndt v Smith [1997] 2 SCR 539 (SCC) (both cases of medical negligence/personal injury). As was observed in Lawson, this resistance has long existed in Quebec, even in cases against solicitors for lost civil litigation: Beaupre v Joly [1971] CS 199; Lacouciere v Laplante [1976] CA 433. The reception in similar cases in Ontario has been mixed: Fyk v Millar [1973] 2 OR (2d) 39 (Ont HC) and Banks v Reid [1974] 6 OR (2d) 404 (Ont HC) seem hostile, Gouzenko v Harris (1976) 1 CCLT 37 (100% damages) is equivocal, and Prior v McNab (1976) 1 CCLT 137 (Ont HC) is encouraging (albeit that 100% damages were awarded on the facts). In British Columbia, the analysis has been accepted in a similar case involving lost civil litigation (Cridge v de Voot [2004] BCSC 101) (0% damages) and left open in medical cases, where the duty of care is contractual (De la Giroday v Brough [1997] 6 WWR (BCCA) (retrial ordered). It is accepted in cases involving lost civil litigation in Saskatchewan: Hagblom v Henderson (2003) 232 Sask R 81 (Sask CA) (75% damages).

[22] The analysis has been rejected by a bare majority of the House of Lords in medical/personal injury cases in Gregg v Scott [2005] UK HL 2 (having technically been left open in Hotson v East Berkshire Health Authority [1988] UKHL 1; [1987] AC 750 (HL)), but it is widely used in cases against solicitors involving lost civil and (now) criminal litigation: Kitchen v Royal Air Forces Association [1958] 1 WLR 563 (Eng CA) (66% damages); Rey v Graham & Oldham (a firm) 2000 BPIR 354 (80% damages); Hanif v Middleweeks [2000] Lloyd's Rep PN 920 (Eng CA)(20% damages); Dixon v Clement Jones [2004] EWCA Civ 1005 (Eng CA)(30% damages); Acton [1997] 3 All ER 909 (Eng HC) (50% damages). See also cases involving negligent misstatements or advice by solicitors and others leading to pure financial loss: Spring v Guardian Assurance Plc [1995] 2 AC 296 (HL) (damages to be assessed); Maples [1995] EWCA Civ 17; [1995] 1 WLR 1602 (Eng CA) (‘Maples’) (preliminary issue); Stovold v Barlows [1996] 1 PNLR 91 (Eng CA) (50% damages); First Interstate Bank of California v Cohen Arnold [1996] 1 PNLR 17 (Eng CA) (66% damages); Motor Crown Petroleum Ltd v SJ Berwin & Co [2000] Lloyds Rep PN 438 (Eng CA) (40% damages); Normans Bay v Coudert [2003] EWCA Civ 215 (Eng CA) (40% damages). The convincing reconciliation of these lines of authority is still awaited. For an early, seminal analysis, see Stapleton, ‘The Gist of Negligence’ (1988) 104 Law Quarterly Review 215, 389. .
[23] In medical/personal injury cases, the availability of the strategy has never been conclusively determined by the High Court, but now appears to be gaining acceptance: see Malec v Hutton (1990) 169 CLR 638 (HCA) (approach accepted, but probably not a case of ‘pure’ loss of chance)(damages to be assessed); Chappel v Hart (1998) 195 CLR 32 (HCA) (preferring a robust approach to establishing causal connection between breach and injury) (full damages); Naxakis v Western General Hospital (1999) 197 CLR 269 (HCA) (split reasoning, all comments obiter: Gaudron J against, Callinan J in favour of loss of chance); Gavalas v Singh [2001] VSCA 23; [2001] 3 VR 404 (CA) (availability of approach assumed in at least some medical cases, damages remitted for reassessment); Rufo v Hosking [2004] NSWCA 391 (availability of approach accepted by all parties as a matter of law) (damages of less than 50% to be assessed). It has also be used in cases against solicitors involving civil litigation: Nikolaou v Papasavas Phillips & Co [1988] VicRp 68; [1988] VR 682 (FC), (1988) 166 CLR 394 (CA) (75%damages); Golec v Scott [1995] 38 NSWLR 168 (NSW CA); Williams v Bodewes (1997) Aust Torts Reps 81-449 (NSW CA); Green v Berry [2000] QCA 133 (unreported, Qld CA, 5th May, 2000). See also cases involving the loss of commercial opportunities flowing from breach of contract (Commonwealth v Amann Aviation Pty Ltd (1991) 174 CLR 332) or deceptive trade practices (Sellars Ltd & Poseidon Ltd v Adelaide Petroleum NL [1994] 68 AJLR 313 (HCA) (40% damages). For further analysis, see Luntz, ‘Loss of Chance’ in Freckelton, Mendelson (eds), Causation in Law and Medicine (2002), 152.
[24] See above nn 21-23.
[25] Acton [1997] 3 All ER 909 (Eng HC).
[26] [2005] 194 OAC 201 (Ont CA), [87].
[27] Ibid.
[28] Sellars Ltd & Poseidon Ltd v Adelaide Petroleum NL [1994] 68 AJLR 313 (HCA).
[29] See cases cited above at nn 22-23.
[30] See Smith, ‘Liability for the Negligent Conduct of Litigation: The Legacy of Rondel v Worsley’ (1982-3) 47 Saskatchewan Law Review 211, for whom this policy concern logically entails rejection of loss of chance claims in all cases of lost litigation (both criminal and civil), at least where is possible for a court to hold a retrial of the merits.
[31] [2005] 194 OAC 201 (Ont CA), [92].
[32] See above nn 22-24. Admittedly, in Canada, there appear to be no cases in which less than 50% damages have been awarded. This possibility was, however, left open in Hagblom v Henderson (2003) 232 Sask R 81 (Sask CA) and English cases are quite clear.
[33] The risk of compensating criminals is likely to be thought to undermine the integrity of our system of justice in a more fundamental way that the risk of allowing unmeritorious civil claims.
[34] The chance must be real and substantial, not merely speculative: Maples [1995] EWCA Civ 17; [1995] 1 WLR 1602 (Eng CA), 1611.
[35] Fischer v Halyk (2003) 229 DLR (4th) 67 (Sask CA); Smith v Linskells [1996] 1 WLR 763 (Eng CA). See Evans, ‘Hall v Simons and Abuse of Process’ (2001) 17 Professional Negligence 218; Lawyers Liabilities (2nd ed, 2002), ch 5. On the facts, Folland’s claim was not deemed an abuse of process, since his conviction had been set aside: [2005] 194 OAC 201 (Ont CA), [98].
[36] In this respect, there seems to be an obvious tension between the Court of Appeal’s underlying concern that Folland’s defence might be unmeritorious and its observation that his conviction had been set aside.
[37] Ibid [92]. This is doubtful, for the reasons indicated in section D, below - proof of this causal link would be an ‘objective’ impossibility. What the court must have meant, was that it was prepared to make a causal inference on his behalf.
[38] [1997] 3 All ER 909 (Eng HC)
[39] See nn 21-23 above.
[40] [2005] 194 OAC 201 (Ont CA), [89], citing Lawson v Lafferiere [1991] 1 SCR 541, 605 (SCC).
[41] Ibid [88].
[42] Ibid [90].
[43] Reece, ‘Losses of Chances in the Law’ (1996) 59 Modern Law Review 188.
[44] The commonest cases used by authors as counter-examples are those involving speculation about hypothetical first party actions (classified as indeterministic because they entail speculation about the imponderables of human decision making), in which the onus apparently remains on the claimant to prove on the balance of probabilities what he/she would have done had it not been for the negligence. See Jones, Textbook on Torts (8th ed, 2002), 248. Other writers, with some justification, point to the unlikelihood of legal judgements adhering exactly to the distinction between deterministic and indeterministic situations, as opposed to deploying a more flexible strategy of ‘enlightened pragmatism,’ See Miller, Coal Dust, Causation and Common Sense’ (2000) 65 Modern Law Review 763.
[45] For a detailed review of these authorities, (adopting a ‘broad’ interpretation of Kitchen v Royal Air Forces Association [1958] 1 WLR 563) see Hagblom v Henderson (2003) 232 Sask R 81 (Sask CA).
[46] The literature is vast, but includes: Stapleton, above, n 22; Lutz, above n 23, Evans, above n 34; Cooper, ‘Loss of a Chance or Chance of a Loss?’ (1972-3) 37 Saskatchewan Law Review 193; Fleming, ‘Probabilistic Causation in Tort Law’ (1989) 68 Canadian Bar Review 661; Hill, 'A Lost Chance for Compensation in the Tort of Negligence by the House of Lords' (1991) 54 Modern Law Review 511; Lunney, ‘What Price a Chance?’ (1995) 15 Legal Studies 1; Stauch, ‘Causation, Risk & Loss of Chance in Medical Negligence’ (1997) 17 Oxford Juornal of Legal Studies 205; Waddams, ‘Damages: Assessment of Uncertainties’ (1998) 13 Journal of Contract Law 55; Jansen, ‘The Idea of a Lost Chance’ (1999) 19 Oxford Journal of Legal Studies 271; Reid, ‘The Hypothetical Outcome in Professional Negligence Claims’ (2001) 17 Professional Negligence 129.
[47] As in Maples [1995] EWCA Civ 17; [1995] 1 WLR 1602 (Eng CA). The rigid distinction between contract and tort is generally disavowed for excellent reasons: see eg Reece, above n 42, 189-90.
[48] See, eg, Bolitho v City & Hackney Health Authority [1997] UKHL 46; [1998] AC 232 (HL). The distinction between hypothetical questions relating to first party and third party actions is embedded in Maples, ibid. It has been criticised by Stapleton as anomalous in 'Cause-in-Fact and the Scope of Liability for Consequences' (2003) 119 Law Quarterly Review 388, but in fact carries considerable support both with courts and writers: McGregor, above, n 2, 8-035; Burrows, Remedies for Torts and Breach of Contract (3rd ed, 2003) 56-7.
[49] For the view that matters of ‘past fact’ should be dealt with on the ‘all-or-nothing’ approach, see eg, Mallett v McMonagle [1970] AC 166 (HL), 174 (Lord Diplock); Davies v Taylor [1974] AC 207 (HL), 212-213 (Lord Reid); Hotson v East Berkshire Health Authority [1988] UKHL 1; [1987] AC 750 (HL), 785 (Lord Mackay); Maples [1995] EWCA Civ 17; [1995] 1 WLR 1602 (Eng CA), 1610 (Stuart Smith LJ). The distinction between matters of fact and hypotheticals is malleable and therefore vulnerable for reasons explained by Jones, above n 43, 245 and Reece, above n 42, 191-2). It is also not clear that it is desirable in terms of policy: Burrows, above n 47, 60-61.
[50] For allusions to this type of distinction and the importance of identifying the content of the primary tort duty, see Peel, ‘Loss of Chance Revisited’ (2003) 66 Modern Law Review 623; Gregg v Scott [2005] UK HL 2, [42] (Lord Nicholls, dissenting); Rufo v Hosking [2004] NSWCA 391; Gavalas v Singh [2001] VSCA 23; [2001] 3 VR 404 (CA).
[51] [1995] EWCA Civ 17; [1995] 1 WLR 1602, 1610. These dicta inappropriately elide the distinction between acts and omissions with the distinction between past facts and hypothetical questions.
[52] Luntz, above n 23, 195.
[53] Weir, Tort Law, (2002), 76.
[54] Rufo v Hosking [2004] NSWCA 391; Burrows, above n 48, 60.