At issue in Torgerson et al. v. Nijem was the limitation of a claim arising from internet defamation where the impugned words remain available indefinitely. The plaintiff argued that each day the words remain available gives rise to a new cause of action, a new claim, and a new limitation period. Justice Schabas rejected this argument: “When a plaintiff becomes aware of a specific posting on a website, time begins to run for the purpose of applying limitation periods. To hold otherwise, and allow plaintiffs to wait indefinitely to sue simply because the libel remains available, would render limitations defences meaningless and have serious implications for freedom of expression.”. The decision is well-reasoned and persuasive; I have no quibbles. This is the analysis:
[12] Section 4 of the Limitations Act, 2002 creates a general limitation period of two years from “the day on which the claim was discovered.” Section 5(2) of the Act creates a presumption that the claim is discovered when the act or omission took place, “unless the contrary is proved.” Here, the defendant’s action in publishing the defamatory statements took place on January 18 and 19, 2016, and the plaintiff became aware of them within days, and no later than the end of January, 2016. The defendant therefore submits that it is plain and obvious that the action cannot succeed as the plaintiff did not commence this action until April 23, 2018, more than two years after the actions of the defendant were known, or “discovered”, by the plaintiff.
[13] The plaintiff, however, relies on the “multiple publication” rule, that each publication of a defamatory communication is a separate and distinct libel, subject to its own limitation period. In the context of an electronic communication or publication on the internet, the argument goes, every time a person accesses or downloads the information there is a new claim and a new limitation period applies. In my view, the plaintiff’s position misunderstands the “multiple publication” rule and how it has been applied to the internet.
[14] The “multiple publication” rule dates back to the somewhat infamous case of Duke of Brunswick v. Harmer (1849), 14 Q.B. 185, in which a cause of action arose after the Duke dispatched his manservant to purchase a back issue of a newspaper from the publisher in order to sue for a libel first published 17 years earlier. This, it was held, constituted a republication by the newspaper, allowing the plaintiff to avoid the limitation period. The “multiple publication” rule established in that case has been subject to criticism, especially in light of the development of the internet. American courts have rejected it, adopting a “single publication rule” in which limitation periods begin to run from the date of first publication: see, e.g., Firth v. State of New York, 775 N.E.2d 463 (NY Ct App 2002); Canatella v. Van de Kamp, 486 F. 3d 1128 (9th Cir. 2007). And the British have now adopted a somewhat similar approach in s. 8 of the Defamation Act 2013, (U.K.) 2013, c. 26.
[15] The multiple publication rule continues to find acceptance in Canada: see, e.g.,Carter v. B.C. Federation of Foster Parents Assn., 2005 BCCA 398 (CanLII); Shtaif v. Toronto Life Publishing Co. Ltd., 2014 ONCA 405 (CanLII). In this case it is not necessary to address the validity of the rule, other than to note that it may be seen as another way of expressing the “repetition rule”, where someone has chosen to repeat or re-publish a defamatory statement, as the law provides that every person who repeats a libel is liable as if he or she is the original speaker: see Downard, Libel (Lexis Nexis, 3rd Ed., 2014) at paras. 5.32–5.44.
[16] Another form of republication is found in Breeden v. Black, 2012 SCC 19 (CanLII), on which the plaintiff relies. Statements posted on the internet in the United States were read, downloaded and republished by newspapers in Ontario, creating a cause of action against the American defendants in Ontario, as the tort of defamation occurs when and where the words are read by a third party, and the republication in Ontario was foreseeable as a natural and probable consequence of the posting of the statements on the internet. This does not mean, however, that every time a person reads the defamatory words in those newspapers a new cause of action arises and limitation periods can be avoided. Applying the discoverability principle, a plaintiff must sue within two years of when he or she becomes aware of the defendant’s action in publishing the defamatory words, whether in print or on the internet.
[17] The Ontario Court of Appeal considered republication and the multiple publication rule in Shtaif. There, the plaintiffs had not provided notice of their intention to sue over the print article within six weeks of becoming aware of it, as required by s. 5 of the Libel and Slander Act. However, they became aware of the internet publication later and did provide notice of it within the required time. The plaintiffs then sought to include their complaint about the print version in their action against the magazine, relying on s. 6 of the Act which permits adding other claims for libel against the defendant that occurred within the previous year.
[18] The Court considered and rejected applying the single publication rule, as it would be inconsistent with s. 6 of the Act. The analysis in Shtaif must be seen in the context of that case, which dealt with separate publication in two different mediums – print and online. As the Court of Appeal observed more recently in John v Ballingall, at para. 35, Shtaif “does not mean that each day of online publication grounds a new cause of action.”
[19] In John v. Ballingall the plaintiff had missed the notice and limitation periods for publications by the media under the Libel and Slander Act, and argued that “for every day the defamatory words are published online, a new and distinct cause of action accrues and a new limitation period begins to run.” The Court of Appeal rejected that position, stating, at para. 35:
The appellant seeks to rely on an incorrect interpretation of the “multiple publication rule”. That concept provides that when an alleged libel is republished across different mediums, including the Internet, those republications are treated as distinct libels. In Shtaif, the court rejected the notion that the limitation period for a suit about an online magazine article starts to run when the plaintiff becomes aware of the printed version. This was the basis for the conflicting evidence on discoverability in Shtaif. This decision does not mean that each day of online publication grounds a new cause of action. The court in Vachon v. Canada Revenue Agency, 2015 ONSC 6096 (CanLII), expressly rejected this interpretation of Shtaif. I concur with Hackland J., who said, at para. 22:
The plaintiff argues that the alleged defamation should be taken as having been republished every day [while it] remained accessible on the internet … Shtaif does not support that proposition … any limitation period based on discoverability will run from the point where the internet defamation is discovered. [emphasis added]
[20] The plaintiff seeks to distinguish John v. Ballingall on the basis that it only deals with notice periods under the Libel and Slander Act. While that was the context, the case also engaged the 3 month limitation period under the Libel and Slander Act, and the same reasoning must apply. The Libel and Slander Act simply creates a shorter limitation period, together with a notice requirement, in recognition of the special position of the media, allowing it the opportunity to publish timely corrections to minimize damage and to prepare defences when facts remain fresh.
[21] In my view, the Court of Appeal in John v. Ballingall clearly rejected the proposition that just because defamatory words remain online and are available to be downloaded and read indefinitely, there is continuous publication allowing limitation periods to be ignored. As the Court noted, the multiple publication rule applies when something is “republished across different mediums”. Such republications require specific acts by the publisher, or republisher, to further disseminate, or repeat, the libel, as was the case in Shtaif.
[22] Furthermore, to give effect to the amendment in the Statement of Claim would create the potential for endless retriggering of limitation periods simply because the words remain on the internet. This would allow plaintiffs to sit on their rights until it suited them to take action, rather than sue when they become aware of the wrong. This would be unfair to defendants who would be subject to lawsuits indefinitely, and raises concerns about freedom of expression.