Does the tort of “harassment” exist at Ontario law? And, if not, is it time for Ontario’s court to recognize the existence of such tort? The answer to both of those questions, as provided by the Court of Appeal for Ontario in Merrifield v. Canada (Attorney General), 2019 ONCA 205, is “no.”
I reviewed the underlying trial decision in my post Ontario Superior Court Awards $100,000 in General Damages for Tort of Harassment.
In that post, I wrote the following commentary:
My sense of this decision is really threefold: (1) bad facts make bad law, and it was likely hard for the judge to overlook how the RCMP’s behaviour given the attention that has recently been brought to it; (2) I believe the finding that the plaintiff could not sue for breach of contract, and therefore access the Honda model very likely drove this decision; and (3) this is a trial decision, which will likely be appealed.
Of those three points, it is the second that I think most important. To my mind, the court in this case was looking for a way to compensate the plaintiff for what it saw as horrible behaviour. Given the very peculiar circumstances of Mr. Merrifield’s employment, i.e. that his employment was governed by statute and not contract, the court had to find a way to compensate him via a route other than that available to most workers in Ontario and the tort of harassment was it.
Given such circumstances, and this was not made specific by the trial judge, I am of the sense that this tort may be found to have limited application, i.e., it may only be available to employees in those case where they cannot claim a breach of contract. Maybe I am wrong, but if the Court of Appeal chooses to uphold this decision, that may be a distinction that they amplify.
What is curious, however, is that the judge found it necessary to entertain the tort of harassment analysis at all. Given that she was prepared to award Mr. Merrifield damages for intentional infliction of mental suffering, notwithstanding the fact that Mr. Merrifield declined to call any of his treating physicians to corroborate his claims that he either suffered from the alleged psychiatric and/or physical harms or to testify that the harms are medically and causally linked to the alleged instances of harassment that comprise his claim, and that the amount of general damages awarded account for both torts, one queries why she did not just take the path of least resistance and simply award damages on the basis of intentional infliction of mental suffering.
Do I think that this decision will withstand appellate scrutiny? If it does, then I believe the court will amplify the fact that, on the facts of this case, the plaintiff could not have sued for breach of contract. As to the fact that the plaintiff declined to provide any medical evidence, that will be a different issue.
I am hesitant to suggest that the takeaway for employees is that they can now sue for the tort of harassment. I am not yet convinced, notwithstanding this trial decision, that such a thing exists. If such a thing does exist, I am skeptical of the limits of its application.
Decision of the Court of Appeal for Ontario
On March 15, 2019, the Court of Appeal for Ontario (Juriansz, Brown and Huscroft JJ.A.) released its reasons for decision to allow the government’s appeal.
In allowing the appeal, the court started by noting that “authority does not support the existence of a tort of harassment” and then went on to ask the fundamental question “should this court nevertheless recognize such a new tort?”
In answering “no” to that question, the court reasoned as follows:
 To pose the question in this way is to suggest that the recognition of new torts is, in essence, a matter of judicial discretion – that the court can create a new tort anytime it considers it appropriate to do so. But that is not how the common law works, nor is it the way the common law should work.
 At the outset, it is important to recognize that this is not a case like Tsige, which, as we have said, is best understood as a culmination of a number of related legal developments. As we have explained, current Canadian legal authority does not support the recognition of a tort of harassment.
 We were not provided with any foreign judicial authority that would support the recognition of a new tort. Nor were we provided with any academic authority or compelling policy rationale for recognizing a new tort and its requisite elements.
 This is not a case whose facts cry out for the creation of a novel legal remedy …
 … In this case, there are legal remedies available to redress conduct that is alleged to constitute harassment. The tort of IIMS [Intentional Infliction Of Mental Suffering] is one of these remedies, and it is discussed below.
 In summary, the case for recognizing the proposed tort of harassment has not been made. On the contrary, as we will explain, there are good reasons opposing the recognition of the proposed tort at this time.
 The tort of IIMS is well established in Ontario and may be asserted as a basis for claiming damages for mental suffering in the employment context.
 A comparison of the elements of the proposed tort of harassment accepted by the trial judge and the elements of the existing tort of IIMS is instructive.
 Whereas IIMS requires flagrant and outrageous conduct, the proposed harassment tort would require only outrageous conduct. More significant, IIMS is an intentional tort, requiring an intention to cause the kind of harm that occurred or knowledge that it was almost certain to occur. This is a purely subjective test … whereas the proposed tort of harassment would require either intention or objectively-defined reckless disregard. Finally, IIMS requires conduct that is the proximate cause of a visible and provable illness, whereas causing severe or extreme emotional distress is sufficient for the proposed tort of harassment.
 Plainly, the elements of the tort of harassment recognized by the trial judge are similar to, but less onerous than, the elements of IIMS. Put another way, it is more difficult to establish the tort of IIMS than the proposed tort of harassment, not least because IIMS is an intentional tort, whereas harassment would operate as a negligence-based tort.
 Given the similarities between IIMS and the proposed tort of harassment, and the availability of IIMS in employment law contexts, what is the rationale for creating the new tort?
 Merrifield submits that the new tort must be created because there is an increased societal recognition that harassment is wrongful conduct. He notes that moral damages for mental distress can be awarded only at termination of employment, leaving a gap that the tort of harassment should fill. He asserts that the decision of the Supreme Court in Saadati v. Moorhead, 2017 SCC 28, supports the creation of the tort of harassment, and that the test the trial judge recognized for the tort is sufficiently stringent to limit the reach of the tort.
 We disagree.
 Saadati is concerned with proof of mental injury in the context of a known cause of action. Although it may make damages for mental injury more readily available in negligence actions, it does not require the recognition of a new tort. Moreover, this court has not allowed negligence to ground a claim for mental suffering in the employment context: Piresferreira v. Ayotte, 2010 ONCA 384.
 In summary, while we do not foreclose the development of a properly conceived tort of harassment that might apply in appropriate contexts, we conclude that Merrifield has presented no compelling reason to recognize a new tort of harassment in this case.
In summary, the Court of Appeal found that the trial judge had erred in concluding that the tort of harassment exists in Ontario and was not persuaded that the tort should be recognized. The Court of Appeal trial also found that the trial judge erred in concluding that the tort of IIMS was made out. The appeal was dismissed with costs of $750,000 awarded to the appellants.
Although the decision is said to be authored by “the Court” it is impossible for me to overlook the similarities that exist as between this decision and that of Piresferreira v. Ayotte, 2010 ONCA 384, which was authored by Justice Juriansz.
In Piresferreira Justice Juriansz wrote the following:
 Could the trial judge have rested her conclusion on a tort duty of care? Since no Canadian appellate court has recognized a free standing cause of action in tort against an employer for negligent infliction of mental suffering by an employee, it is necessary to apply the Anns test. The two-part test … involves asking first whether the relationship between the plaintiff and the defendant is sufficiently close or “proximate” to render damages reasonably foreseeable and justify the imposition of a duty of care, and second, whether there are countervailing policy considerations why a duty of care should be limited or not recognized ….
 The relationship of employer and employee puts the parties in a relationship of proximity. As Le Dain J. explained in Rafuse, a relationship of proximity that would not have arisen but for a contract can support the creation of a common law duty of care. The question is whether there is a relationship of sufficient proximity, not how that relationship arose.
 The trial judge found the damage was foreseeable. She noted as follows:In my view, it is reasonably foreseeable that a person of ordinary fortitude would suffer serious psychological injury if that person was regularly yelled and sworn at by her manager/supervisor/boss, was told by the manager/supervisor/boss that she did not know what she was doing, was not given the opportunity to explain her actions or defend herself, was pushed by the manager/supervisor/boss who at the time was clearly angry and out of control, and was immediately told that she would be put on probation or issued a PIP.
 The appellants submit that this foreseeability finding is incompatible with the Supreme Court’s judgment in Honda. In Honda, the Supreme Court reiterated the general rule that damages in a wrongful dismissal action are confined to the loss suffered from the employer’s failure to give proper notice and that no damages are available for the mental suffering the employee may have suffered unless the parties contemplated at the time of the employment contract that a breach of the contract might cause the plaintiff mental distress. The proposition that mental suffering is generally not within the contemplation of the parties to an employment contract does not mean, as the Supreme Court makes clear, that it is not contemplated by the parties to a particular employment contract. As well, the principle enunciated in Honda is directed to mental suffering as a consequence of the manner of termination, as opposed to mental suffering that results from mistreatment during the employment relationship.
 I see no reason to resist the finding of the trial judge that it was reasonably foreseeable that Piresferreira would experience mental suffering from the abusive manner in which Ayotte supervised her during her employment.
 Given that the parties had a relationship of proximity and that the damages suffered were reasonably foreseeable, it must be determined whether policy considerations foreclose the recognition of a duty of care. In my view, they do.
 First, the Supreme Court has already strongly intimated that the recognition of such a tort in the employment context is better left to the legislature. In Wallace v. United Grain Growers Ltd.,  3 S.C.R. 701, Iacobucci J. writing for the majority rejected the notion that a tort existed for breach of a good faith and fair dealing obligation by employers in dismissing employees. He wrote at para. 77, “To create such a tort in this case would therefore constitute a radical shift in the law, again a step better left to be taken by the legislatures.” The further evolution of the law in Honda is completely consistent with this view.
 The duty rejected in Wallace is not exactly the same duty postulated in this case. This, however, provides more, not less, reason to reject a duty in this case. The duty of care put forward in this case is broader than the duty that was rejected in Wallace. A general duty to take care to shield an employee during the entire course of his or her employment from acts in the workplace that might cause mental suffering strikes me as far more expansive than a duty to act fairly and in good faith during just the termination process. The duty rejected in Wallace would have applied only at the time of termination and to the manner of termination. The duty put forward in this case would apply in the course of employment as well as to its termination. The general duty postulated would require employers to take care to shield employees from the acts of other employees that might cause mental suffering.
 In this sense, the asserted duty of care would have a far greater impact on settled jurisprudence than would the duty of good faith and fair dealing that Iacobucci J. described in Wallace at para. 76 as “overly intrusive and inconsistent with established principles of employment law”.
 The findings of the trial judge in this case provide an apt example. The trial judge assessed Piresferreira’s damages for wrongful dismissal as comprised of her salary and benefits for 12 months from May 24, 2005 and an additional $45,000 for mental distress under the framework established by Honda. By contrast, the tort damages awarded by the trial judge included lost income between May 24, 2005 and July 2009, less a 10% discount for contingencies. In BG Checo, the Supreme Court noted at p. 38 that “it would seem anomalous to award a different level of damages for what is essentially the same wrong on the sole basis of the form of action chosen, though, of course, particular circumstances or policy may dictate such a course.”
 No particular circumstances or policy that would support different scales of damages for mental suffering in tort and contract in the employment context were brought to the court’s attention. Rather, there are good reasons for avoiding such a scenario.
 In a case in which the employer’s allegedly tortious behaviour includes the termination of the employee, compensation for mental distress is available under the framework the Supreme Court has set out in Honda. In a case in which the employer does not terminate the employee, the employee who is caused mental distress by the employer’s abusive conduct can claim constructive dismissal and still have recourse to damages under the Honda framework. Recognizing the tort in the employment relationship would overtake and supplant that framework and all of the employment law jurisprudence from which it evolved. In other words, in the dismissal context, the law already provides a remedy in respect of the loss complained of here. The recognition of the tort is not necessary.
 That leaves the category of cases in which the employee suffers mental distress from employer conduct that would not provide the grounds for a claim of constructive dismissal. Perhaps it can be said, as the respondents submit, that it is not foreseeable that an employee would suffer mental distress from criticism of poor work performance that is constructive. However, much disagreement can be anticipated as to whether criticism is “constructive”, whether work performance is “poor”, and whether the tone of the former was appropriate to the latter. The existence of the tort would require the resolution of such disputes. The court is often called upon to review the work performance of employees and the content and manner of their supervision in dismissal cases. It is unnecessary and undesirable to expand the court’s involvement in such questions. It is unnecessary because if the employees are sufficiently aggrieved, they can claim constructive dismissal. It is undesirable because it would be a considerable intrusion by the courts into the workplace, it has a real potential to constrain efforts to achieve increased efficiencies, and the postulated duty of care is so general and broad it could apply indeterminately.
 I would set aside the trial judge’s finding that the appellants were liable to the respondents for committing the tort of negligent infliction of mental suffering as the tort is not available in the employment context. Employees, of course, can sue their employers or supervisors for the intentional infliction of mental suffering.
The takeaway I take from both the Merrifield and Piresferreira decisions is that the Court of Appeal for Ontario is simply not prepared to allow claims for mental suffering caused in the workplace, unless such claim is either: (a) attached to a claim of wrongful or constructive dismissal, and limited to the Honda analysis; or (b) advanced as a claim for intentional infliction of mental suffering. It would appear to me that the court views the tort of IIMS as a panacea for all workplace wrongs. Whether that is an appropriate approach to the issue of workplace harassment, especially given the development of the law in other areas, is a question best left to others.
What I do find interesting about the Merrifield decision is the fact that the court paid no attention to the fact that Merrifield did not have access to the constructive dismissal / Honda model, given the statutory nature of his employment. Perhaps the court felt that the same was unnecessary, given its other findings.
As always, everyone’s situation is different. The above is not intended to be legal advice for any particular situation. It is always prudent to seek professional legal advice before making any decisions with respect to your own case.
Image (c) istock/designer491