"That Is Not How the Common Law Works": Paths to Tort Liability for Harassment.

• Author: Priel, Dan
• Position: Canada, Ontario

CONTENTS I. Introduction 89 II. Merrifield v Canada (AG) 91 III. Making a Case for Tort Liability for Harassing Behaviour 92 A. Why Tort Liability for Harassing Behaviour Makes Sense 93 B. Countering Potential Objections 97 IV. How the Common Law Changes 100 V. Doctrinal Routes to Tort Liability for Harassment 105 A. The Meaning of Legislative Silence 105 B. Limited Solutions 109 1. Nuisance 109 2. Intimidation 110 3. Assault 111 4. Defamation 112 C. More Promising Possibilities 113 1. Updating IIMS 115 2. Incorporation within the Tort of Negligence 120 D. Balancing IIMS with Negligent Infliction of Mental Injury 127 VI. Conclusion: The Limits of Judicial Restraint 130 I. INTRODUCTION

Harassment happens everywhere. It happens in the workplace, it happens on the streets, it happens in schools, and in recent years, it happens at alarming rates on the Internet as well. It affects millions. There have been various statutory responses to harassment in Canada, but it remained unclear whether there is a common law tort giving rise to civil liability for harassment. At least as far as Ontario is concerned, that uncertainty has now been resolved after the Court of Appeal for Ontario decided against recognizing such a tort in Merrifield v Canada (AG). (1) In its decision, the Court of Appeal concluded that a tort of harassment has not been previously recognized in Ontario and that there are no compelling reasons for changing the law. One of the Court's main reasons for its conclusion is the incremental nature of common law change, which is inconsistent with the establishment of a new tort "anytime [a court] considers it appropriate to do so," because, the Court reasoned, "that is not how the common law works." (2)

In this article I wish to challenge this decision and its reasoning. Specifically, I argue that by focusing its attention on the question of the Court's power to create a new tort, it ignored the possibility of extending existing torts, in line with traditional common law methods. I hope to show that there are ample resources within existing law to support tort liability for harassing behaviour, without having to create a new tort. Indeed, paying attention to existing law shows that, contrary to its message of modesty, Merrifield constitutes a rather bold rejection of a recent Supreme Court precedent, where a unanimous Court held that any categorical distinction between physical and mental injury cannot be sustained. (3) This article can be read, then, as a critique of Merrifield; it can also be read as suggesting ways of blunting its holding. Even accepting as settled law (for now) that there is no independent tort of harassment in Ontario, I argue that there are ways for plaintiffs seeking tort redress for harassing behaviour to seek redress on the basis of existing torts.

The argument proceeds as follows: Part II summarizes Merrifield and its main reasons. Part III starts outlining the case for tort liability for harassment by arguing that there are good policy reasons for it. The argument there is grounded in empirical data, much of it drawn from official Canadian sources, on the prevalence of harassment and the harm it causes. Acknowledging that that is not enough for justifying a change in the law through the courts, the remainder of the article is dedicated to showing respectable doctrinal paths to such liability. Part IV opens this discussion by showing that courts have long made significant changes to tort law, and that by the standards that Merrifield accepted for judge-made innovation in tort law, a judicial development of tort liability for harassing behaviour is justified. Part V then turns to showing several possible paths to a judicially created tort liability for harassment without creating a "new" tort. It asks specifically whether the fact that the legislature addressed harassment in various contexts but did not establish civil liability by statute, should be interpreted as an implicit rejection of such liability. After concluding that it should not be, it considers six possible routes to such an end, four that I consider relatively unpromising and two that are more successful. The conclusion that follows returns to the general question of common law change and frames it in terms of the limits of judicial restraint.

This outline helps identify the scope of this article and its aims. My main aim is practical: I aim to show how courts could, and in my view should, establish tort liability for harassment. But along the way I wish to contribute to discussions of the role of courts as norm creators. As my title suggests, one particular focus of this article is on the paths to liability for harassment by a common law court developing and adapting long-accepted bases for liability to new circumstances.

2. MERRIFIELD V CANADA (AG)

   Peter Merrifield was a junior constable in the Royal Canadian Mounted Police (RCMP). In 2007, while still employed, he filed a lawsuit against the RCMP, in which he alleged he had been subjected to ongoing harassment by his workplace supervisors. In his statement of claim, Merrifield made a host of allegations, from breach of contract and breach of a fiduciary duty to a violation of his Charter rights. After a trial, Justice Vallee of the Ontario Superior Court of Justice dismissed most of these claims, but accepted that the actions of Merrifield's supervisors constituted either harassment or intentional infliction of mental suffering (IIMS), both of which she took to be established torts in Ontario. (4)

   The defendant successfully appealed this decision. Unusually for an appellate decision, the Court of Appeal dedicated about half of its reasons to a re-evaluation of the facts, (5) and its disagreements with the trial court on this score were profound. Where the trial court's decision tells a story of a dedicated public servant, who gave it all but was constantly rebuffed by insensitive superiors, the Court of Appeal's decision depicts Merrifield as dishonest and insubordinate. Rather than repressed or harassed, the Court of Appeal describes an employee who repeatedly flouted workplace regulations, lied to his supervisors, misused the credit card he was given for work purposes, and more.

   Without access to the full case record, it is difficult to evaluate these competing narratives, and I will not attempt to adjudicate between them. It is possible, of course, that Merrifield was harassed by his bosses and that he was a less than stellar employee. Had the Court of Appeal recognized a tort of harassment, it might have been worthwhile to consider whether Merrifield's behaviour should in some way affect his remedies; but the Court of Appeal also disagreed with the trial court on the law, rendering such a discussion moot. The Court of Appeal held that there is no existing tort of harassment in Ontario and that there were no good reasons for a common law court to create one.

   It is possible that the Court of Appeal's lack of sympathy for Merrifield coloured its legal conclusions. It is also possible that the Court did not want to make a major statement on a new tort of harassment only to dismiss the case afterwards on the facts. (6) That the case could have been dismissed on its facts may lead some to consider its discussion of the tort of harassment as obiter. But Merrifield is significant. The Court could have dismissed the case without making any general statement on the law ("we need not decide on whether there is a tort of harassment in Ontario; we can assume, without deciding, that such a tort exists, but still conclude that his claim fails"). That it chose not to do so is a clear indication that it intended its decision to have an impact, one that it is likely to have. By the imperfect measure of citation counts, the Court of Appeal for Ontario is the second-most important court in Canada, (7) leading some commentators to call it Canada's "junior Supreme Court." (8) So there can be little doubt that Merrifield will be taken seriously, obviously by the courts in Canada's most populous province, and probably by other courts across the country as well.

3. MAKING A CASE FOR TORT LIABILITY FOR HARASSING BEHAVIOUR

   The first question to address in considering tort liability for harassment is whether there is a need for it. The Court of Appeal in Merrifield referred to the institutional limits of courts, the nature of common law change, and the absence of supportive judicial authority in support of such liability--all issues I will take up in detail in below. But it is quite clear that it was not just the absence of doctrinal grounding for a tort of harassment that troubled the Court, as the Court explicitly stated it found no "compelling policy rationale" (9) for such a tort. My analysis, in line with the Supreme Court's general approach when tasked with the question of whether to extend or restrict common law tort liability, is to assess the question both in terms of "policy," i.e. whether the benefits of the legal change are greater than its costs, and in terms of doctrinal fit. (10) In this Part, I argue that, in fact, the prima facie case for such a tort is very strong.

   1. Why Tort Liability for Harassing Behaviour Makes Sense

      To set the stage, it will help to have a clear sense of what we mean by harassment. Ontario's Human Right Code defines harassment as "engaging in a course of vexatious comment or conduct that is known or ought reasonably to be known to be unwelcome." (11) A recently passed federal statute added a definition of harassment to the Canada Labour Code: "any action, conduct or comment, including of a sexual nature, that can reasonably be expected to cause offence, humiliation or other physical or psychological injury or illness to an employee, including any prescribed action, conduct or comment." (12) As the latter definition suggests with its reference to sexual harassment, a lot of harassment is directed at vulnerable groups, and the definition...