Judges are ordering unwarranted or overly broad injunctions to 'tame' the internet's wild west
Canadian courts have set their sights on online defamers. In LeBlanc v City of Fredericton, the Court of Queen’s Bench of New Brunswick referred to the “wild west” nature of the internet and the need for courts to “tame” online defamation. In Caplan v Atas, the Ontario Superior Court of Justice created a new online harassment tort when a defamation defendant proved judgment-proof. Frustrated by the law’s limitations, some courts are ordering unwarranted or overly broad injunctions – essentially unjustified state censorship.
Defamation injunctions used to be rare for two reasons. First, they were less useful in the print and broadcast era, where yesterday’s newspaper was today’s fish and chips wrapper; the harm was done, and there was often nothing to enjoin. Second, prior restraint of speech (preventing it from being said at all) severely interferes with freedom of expression, and judges were reluctant to make such orders. It is one thing to make someone pay compensation for their unlawful communications. It’s another to order them not to speak in the first place.
At the interlocutory stage, the law set the bar high for defamation injunctions because of the risk of censoring lawful speech. But even after a finding of liability, permanent injunctions used to be exceptional remedies that needed to be clearly justified and narrowly tailored.
That is no longer the case: interlocutory and permanent defamation injunctions are now easier to get.
In the internet age, defamation injunctions can be more helpful. They can prevent harmful lies from spreading, persisting, and being constantly tied to a plaintiff through search engines. This damage may sometimes justify injunctions to take down content or not to repeat it. However, the importance of freedom of expression is undiminished, and courts have sometimes lost sight of this public and private interest. They are embracing defamation injunctions to an unjustifiable degree.
For interlocutory injunctions, courts reject the longstanding and exceptional rule (not the RJR-MacDonald approach) that defamatory speech shouldn’t be unless it’s unlawful. Courts are ignoring recent Supreme Court jurisprudence such as Canada (Human Rights Commission) v Canadian Liberty Net which restated the importance of a high threshold for interlocutory injunctions censoring speech. For example, in Pilot Insurance Co v Jessome, the judge refused to apply the prevailing rule that no injunction should be ordered unless defences would “inevitably fail.” Instead, he found that “the rights of the parties” would more “fairly be balanced” by applying a “substantial issue to be tried” accessibility threshold and requiring courts to consider the public interest in free speech. While I understand judges’ frustration with nasty online speech, this is contrary to the rule of law and ignores aspects of a complex policy issue.
Reasonable people can disagree about the threshold for an interlocutory injunction in defamation. Still, the Supreme Court of Canada has warned in Canadian Liberty Net against applying tests like those proposed in Jessome. When irreparable harm and the balance of convenience are applied to defamation cases where there is no commercial interest in the speech, the test will almost always result in an injunction being granted, which inadequately protects freedom of expression.
The test for obtaining permanent injunctions in defamation is understandably easier to meet since the speech at issue will have been found to be defamatory. While one might think censoring such speech is unproblematic, there are several reasons for caution. For example, defamation’s presumption of falsity means that the speech grounding liability may not be false, especially if defendants aren’t represented or don’t appear.
Historically the approach was to permanently enjoin defamatory speech only where truly necessary to prevent further harm that couldn’t adequately be compensated in damages. Yet, in 2015 the Ontario Court of Appeal misstated the test in St. Lewis v Rancourt such that it is now easy to get a permanent injunction where there is liability. They are available solely because a defendant won’t pay damages, regardless of the risk of repetition. This approach not only under-protects speech but means greater censorship of the poor than the wealthy (including corporations). And it is divorced from basic tenets of equitable relief that limit injunctions to circumstances where they are necessary to prevent harm. This area of the law needs a return to first principles.
And when crafting both interlocutory and permanent injunctions, courts are often absurdly broad. In studying the scope of defamation injunctions, I found that 16 percent of defamation injunctions ordered the defendant never to speak about the plaintiff again. It is always possible to craft a narrower injunction than one not to talk to anyone about the plaintiff again. This approach cannot be justified. More than half of the remaining defamation injunctions were also unjustifiably broad in the circumstances. If such orders were subject to the Charter, they would be unconstitutional. Yet perhaps the many self-represented defendants and default judgments mean that these orders are insufficiently scrutinized.
To be sure, online libel, like other online harms, is pernicious, worthy of sanction and hard to prevent. As the Ontario Court of Appeal put it in the 2014 Barrick Gold Corp v Lopehandia decision, judges don’t want to “throw up their collective hands in despair.” But that doesn’t justify unnecessary or overbroad injunctions, nor is it evident that interlocutory injunctions should be easier to get. The latter is a complex issue that requires careful balancing of public and private interests. Recent case law suggests a concerning disregard for longstanding rules limiting the availability of injunctions that censor allegedly defamatory speech.
I have written about interlocutory and permanent defamation injunctions and their scope in other published work. Get in touch if you’d like a copy.
