In the companion cases of Breeden v Black, 2012 SCC 19 (“Breeden”) and Éditions Écosociété Inc. et al. v Banro Corp., 2012 SCC 18 (“Banro”), the Supreme Court of Canada clarified the manner in which courts should determine whether to exercise jurisdiction over multijurisdictional defamation claims involving foreign defendants.

Although the decisions support the ability of plaintiffs to advance defamation claims in any Canadian jurisdiction in which allegedly defamatory material is published, the decisions also leave open the possibility that the law will evolve to reduce the potential for forum shopping.


In Breeden, the plaintiff commenced defamation actions in Ontario against the defendants, who were certain directors, advisors, and a vice president of a corporation headquartered in the United States. The plaintiff alleged that statements issued by the defendants and posted on the internet were defamatory and were published in Ontario when they were downloaded, read, and republished in Ontario by Canadian newspapers. The defendants brought a motion to have the defamation actions stayed on the grounds that the Ontario court should not exercise jurisdiction because there was no real and substantial connection between the actions and Ontario or, alternatively, because an American court was the more appropriate forum.

The facts in Banro are similar. There, the defendants, who were based in Québec, published a book which commented on the international mining activities of certain corporations, including the plaintiff. Copies of the allegedly defamatory book were available to be purchased or read in Ontario. The plaintiff brought an action in Ontario against the defendants alleging that the book was defamatory. The defendants moved to stay the Ontario action on the basis that there was no real and substantial connection between the action and Ontario, and that a Québec court was the more appropriate forum.


In both cases, Justice LeBel, writing for the Court, applied a new analytical framework for determining whether a court should exercise its jurisdiction. That analytical framework was established in the companion case of Club Resorts Ltd. v Van Breda, 2012 SCC 17 (“Club Resorts”), reasons of which were issued at the same time as those in Breeden and Banro.

The analytical framework involves a two-stage analysis. In the first stage, which depends on the application of the real and substantial connection test, the plaintiff must demonstrate a “presumptive connecting factor” that links the subject matter of the litigation with the jurisdiction. If the plaintiff demonstrates a presumptive connecting factor, then there will be a presumption of jurisdiction unless the defendant rebuts the presumption. In Club Resorts, the Court identified a non-exhaustive list of presumptive connecting factors. The most important of those presumptive connecting factors, for the purposes of Breeden and Banro, is the commission of a tort in the jurisdiction.

The defendant may rebut a presumption of jurisdiction by establishing “facts which demonstrate that the presumptive connecting factor does not point to any real relationship between the subject matter of the litigation and the forum or points only to a weak relationship between them.” For example, where the presumptive connecting factor is the commission of a tort in the jurisdiction, rebutting the presumption may be possible “where only a relatively minor element of the tort has occurred in the province.” If no presumptive connecting factor applies in the circumstances of the case, or if the presumption of jurisdiction resulting from such a factor is rebutted, the court cannot assume jurisdiction.

If the plaintiff establishes jurisdiction, the court will proceed to the second stage of the analysis, which involves application of the doctrine of forum non conveniens. At this stage, the burden shifts to the defendant to demonstrate that the court should not exercise jurisdiction because the court of another jurisdiction is the more appropriate forum for the hearing of the action. To succeed, a defendant must show that the other forum is “clearly more appropriate” because it is better suited to “fairly and efficiently” resolve the dispute. The defendant may point to a variety of factors, including the locations of the parties and witnesses, the possibility of conflicting judgments, and the substantive law that should apply to determine the claims.

In Breeden and Banro, the Court concluded that jurisdiction had been properly assumed. There was a real and substantial connection between Ontario and the defamation actions based on the fact that the alleged torts had been committed in Ontario. The Court was not convinced that the defendants in either case had rebutted the presumption of a real and substantial connection.

Significantly, the Court recognized that the analytical framework raises concerns about libel tourism, which is a variety of forum shopping in which a plaintiff brings a defamation action in the jurisdiction most likely to provide a favorable result. The prospect of libel tourism arises because the tort of defamation “crystalizes” upon publication of defamatory material.

Defamatory material is “published” whenever it is viewed or read by a third party, and is presumed to be “published” when it is printed in a book. As a result, where allegedly defamatory material is published in multiple jurisdictions – a feat easily achievable, even inadvertently, due to the ubiquity, universality, and accessibly of the internet – the courts of multiple jurisdictions will generally be able to exercise jurisdiction over the same claim.

Because the law of defamation varies between jurisdictions such that it is easier or more difficult for plaintiffs to establish their claims depending on their choice of jurisdiction, plaintiffs can strategically advance their actions in the jurisdictions in which they have the greatest juridical advantage. For example, American defamation law may require some plaintiffs to demonstrate malice on the part of the defendant as a pre-requisite to establishing liability. Because no such requirement exists in Canada, plaintiffs may enjoy a juridical advantage by pursuing their defamation claims in Canada rather than in the United States.

The Court’s reasons in Banro may provide courts in future cases with a way to restrain libel tourism. After concluding that jurisdiction had been properly assumed, the Court turned to determine whether the court of another jurisdiction was a more appropriate forum for the hearing of the action. In the course of applying the doctrine of forum non conveniens, the Court considered the question of which substantive law should be applied to determine the claim. Courts have traditionally applied the lex loci delicti rule (“the place where the tort occurred”) to decide which law applies to determine tort claims. The rationale for the application of the lex loci delicti rule is that, in the case of most torts, the occurrence of the wrong constituting the tort occurs in the same jurisdiction in which the consequences of the wrong are suffered.

The Court recognized that the lex loci delicti rule may not be appropriate in all defamation cases. In certain cases, the reputational harm caused by the publication of defamatory material may be suffered in a jurisdiction other than the one in which the defamatory material was published. The Court suggested that in those circumstances it may be more appropriate to apply a rule based on the “place of most substantial harm to reputation.” According to that rule, the applicable law would be that of the jurisdiction most closely connected to the harm occasioned by the publication. Such an approach could eliminate the strategic advantage to libel tourism by providing that the same law would apply regardless of where the matter was heard.

The Court concluded that it did not need to decide whether the lex loci delicti rule ought to be abandoned as the choice of law rule in multijurisdictional defamation cases in favour of an approach based on the location of the most substantial harm to reputation. The Court observed that, on the facts of both cases, application of either rule had the same effect. Under a rule based on the location of the most substantial harm to reputation, Ontario law would apply. Alternatively, under the lex loci delicti rule, Ontario law would also apply because the alleged torts were committed in Ontario.


Breeden and Banro challenged the Court to consider the appropriate balance between the protection of reputation, freedom of expression, and jurisdictional restraint. The decisions clarify that Canadian courts will have presumptive jurisdiction over defamation cases involving foreign defendants if the defamatory statements are published to at least one person in the jurisdiction. For example, if a person in Hong Kong were to create an allegedly defamatory website, an Ontario court would have presumptive jurisdiction over an action brought by the plaintiff against the person in Hong Kong if the plaintiff demonstrated that at least one other person in Ontario viewed the website.

This precedent will likely have significant consequences, particularly given the ubiquity, universality, and accessibility of the internet. As Lebel J. recognized, with a “globalized world comes the sometimes poisonous gift of ubiquity.” Statements published in one location may, with the aid of the internet, be widely disseminated and viewed in a multitude of jurisdictions all over the world. Given the ease by which allegedly defamatory material may be published in Canadian jurisdictions through the use of the internet, plaintiffs in cases involving internet defamation will likely face little difficulty establishing a presumption of jurisdiction.

As a result, litigation involving jurisdictional disputes in defamation cases will likely turn on whether the foreign defendant is able to rebut a presumption of jurisdiction or demonstrate that another jurisdiction is a more appropriate forum. Foreign defendants may succeed in rebutting a presumption of jurisdiction by demonstrating that only a relatively minor element of the tort of defamation, such as publication, occurred in the jurisdiction.

Although the analytical framework applied in Breeden and Banro may create a heightened risk of libel tourism in future cases, the Court’s remarks with respect to the appropriateness of a choice of law rule based on the location of most substantial harm to reputation may provide lower courts with a legal foundation to curb libel tourism.

Finally, it must be recognized that the analytical framework applied in Breeden and Banro is subject to legislation in certain provinces that governs the assumption of jurisdiction and the doctrine of forum non conveniens: see e.g. Court Jurisdiction and Proceedings Transfer Act, S.B.C. 2003, c. 28. However, because those statutes contemplate an approach similar to the analytical framework applied in Breeden and Banro, the reasoning in those cases is likely to influence the manner in which courts in those provinces determine whether to exercise jurisdiction over defamation cases involving foreign defendants.

This article was originally published at The Court (Osgoode Hall Law School).

Subscribe to this blog by email, in a reader, or follow it on Twitter

The use of the internet as a tool in the commission of crime has given rise to new search and seizure issues. When individuals use the internet, their personal information may be transmitted to various online service providers, such as social networking websites, email service providers, and internet service providers. In many cases, online service providers impose terms of service agreements on their users which require them to agree to the disclosure of their personal information to the authorities for the purpose of criminal investigations. Recent decisions indicate that such terms of service agreements are a key factor in assessing the legality of warrantless disclosure in the internet context under s. 8 of the Canadian Charter of Rights and Freedoms.

These decisions may contribute to an erosion of privacy rights as the internet becomes increasingly central to daily life. Individuals use the internet to perform a variety of personal activities, including writing and receiving correspondence, storing personal files, and developing social networks. However, in order to use these increasingly vital services, individuals must trust their information to online service providers. In doing so, users often unknowingly subject themselves to non-negotiated terms of service agreements that may limit their privacy expectations. As computing trends fuel a migration of information from personal computers to remote servers controlled by online service providers, more of the public’s information may become exposed to warrantless seizure by the state. This article surveys the law, discusses the effect of terms of service agreements, and considers the privacy implications.

Read the full article here. It was published in:

Matthew Nied, “The Internet, Cloud Computing, and the Charter Right to Privacy: The Effect of Terms of Service Agreements on Reasonable Expectations of Privacy” (2011) 69:5 The Advocate (Magazine of the Vancouver Bar Association) 701. Also published in (2011) 12:5 Internet and E-Commerce Law in Canada 40.

Subscribe to this blog by email, in a reader, or follow it on Twitter

Earlier this week, the Ontario Superior Court of Justice released its decision in Baglow v. Smith, 2011 ONSC 5131. The decision suggests that an allegedly defamatory statement made in a debate on blogs or internet forums may not be found to be defamatory if the plaintiff previously engaged in the debate but did not respond to the statement despite having the opportunity to do so.

The plaintiff claimed that the defendants defamed him by making statements that exceeded the boundaries of their normally acrimonious political debate on the internet. In particular, the plaintiff complained that the defendants defamed him by branding him “one of the Taliban’s more vocal supporters” on an internet message board. The words complained of referred back to an ongoing discussion, largely on the plaintiff’s blog, where the parties had debated the validity of the trial of Omar Khadr. The parties had aggressively berated each other, and often employed colourful derogatory characterizations. Although the plaintiff had the opportunity to respond to the impugned statements on the internet message board, he did not do so. The defendants brought a summary judgment motion to dismiss the action on the basis that the statements were not defamatory or, alternatively, that the defence of fair comment applied.

Mr. Justice Annis concluded that the impugned statements were not defamatory and granted summary judgment dismissing the action. Significantly, the Court proceeded to remark in obiter that the conclusion that the statements were not defamatory was supported by the fact that the statements were made “in the context of an ongoing blogging thread over the Internet” that provided each party with the opportunity to “respond to disparaging comments before the same audience in an immediate or relatively contemporaneous time frame.” According to the Court, “a statement is not derogatory when made in a context that provides an opportunity to challenge the comment and the rules of the debate anticipate a rejoinder, unless the statement is wholly outside the scope of the debate or otherwise so outrageous as to prevent meaningful argument from continuing.”

In the Court’s view, the fact that the statements were made in the context of an internet debate forum was a contextual factor to consider in determining whether the statements were defamatory:

[58] Although I am satisfied that the words complained [of] are not capable of damaging the reputation of the plaintiff, I am of the view that there is another contextual factor that would further bolster this conclusion, namely that the alleged defamatory words were made in the context of an ongoing blogging thread over the Internet.

[59] Internet blogging is a form of public conversation. By the back and forth character it provides an opportunity for each party to respond to disparaging comments before the same audience in an immediate or a relatively contemporaneous time frame.

[60] This distinguishes the context of blogging from other forms of publication of defamatory statements. One exception could be the live debate, of which blogging constitutes the modern written form.

[61] I am not suggesting that defamation can never occur in a live debate. I do say however, that the live debate forum should be considered as a contextual factor to determine whether the statement is defamatory in so far as whether it is complete.

The Court suggested that the defamatory sting arising from statements made on the internet may be substantially reduced or eliminated by responding to the statements:

[62] An example that does not in any manner reflect the Court’s views on these issues, but which might serve to explain how derogatory, even defamatory remarks are expected to be parried in a live debate so as to remove the “sting of the libel” and attenuate any threats of diminution of reputation might be as follows:

[The defendant] knows full well that I abhor what the Taliban stand for. His calling me one of their supporters because I think they should be entitled to due process in accordance with International law would be like me calling him (some derogatory descriptor, e.g. “a Nazi fascist”) because he wants to trample the rights that Canadians cherish, etc. [Example provided by the Court]

[63] Given that the plaintiff pleads his belief that “there is a reasonable likelihood of damage to my reputation if it became generally believed that I supported the enemies of the Canadian Forces”, it seems that the tendency of the comment to lower his reputation, particularly when arising in the form of a comment in a debate, could have been quickly nipped in the bud by a simple rejoinder in the fashion described above. This would have had the additional benefit of allowing him to score some points of his own.

The Court’s comments were based on the principle that a statement is defamatory if it tends to injure the reputation of the person to whom it refers by lowering him or her in the estimation of right-thinking members of society: Baglow, at para. 11.  Accordingly, the issue of whether the statements on the blog were defamatory was to be judged through the eyes of its readers:

[64] More importantly to the issue of context, the blogging audience is expecting and would indeed want to hear a rejoinder of this nature where the parry and thrust of the debaters is appreciated as much as the substance of what they say.

[65] In essence, I am suggesting that the Court, in construing alleged defamatory words in an ongoing debate, should determine whether the context of the comment from the perspective of the reasonable reader or listener is one that anticipates a rejoinder, which would eliminate the possible consequence of a statement lowering the reputation of the plaintiff in their eyes.

[66] To some extent the Court is attempting to decide whether the debate should have gone forward, such that walking off the blogging stage, so to speak, is a form of “gotcha” contrary to the rules governing the debate.

[67] I realize that this sounds like a form of defence of mitigation of a defamatory comment. But I see it more as an uncompleted comment, something akin to a plaintiff arguing that he or she has been defamed by a question, when the response was what the audience was expecting.

It appears that the Court’s view was a response, in part, to the concern that the threat of legal action may chill debate on the internet:

[70] Bringing an action on the comment in mid-debate runs contrary to the rules and has the effect of chilling discussion. If allowed, it places the opposing party in a defensive mode, rather than an offensive one, strategically putting that party at a disadvantage.

[74] The comment of the defendant Smith was on topic and generally consistent with the language and positions taken in the on-going debate. Accordingly, in no sense was it one that would have had any different effect on the plaintiff’s reputation from other derogatory remarks made throughout the blogs. Like those comments, it should have been answered to remove the sting, if any, and to comply with expectations of readers of these blogs.

It is interesting to note that the impugned statements were made on an internet message board that was distinct from the blogs on which much of the previous debate had occurred. The Court appears to have considered the comments made in all of these forums as a whole rather than concentrating on isolated comments in determining whether the impugned statements were defamatory: Baglow, at para. 27. However, one might question whether the forums had the same audience, and whether a reasonable reader will  anticipate a rejoinder in a place on the internet that differs from the one where the previous debate occurred.

In any event, the Court’s comments should not be read to suggest that persons defamed on the internet should necessarily enter the fray and respond to defamatory comments if given the opportunity to do so. The allegedly defamatory statements in this case were made in the context of an acrimonious debate in which the plaintiff was found to have participated. This is distinguishable from circumstances in which a plaintiff finds themselves defamed by statements made on a blog or message board in which they have not participated. In these cases, the context of the comment from the perspective of the reasonable reader will not be one that anticipates a rejoinder. It may be advisable for victims of internet defamation in these circumstances to avoid responding to defamatory comments in order to avoid inviting further attention to the matter and increasing the harm to their reputation.

Subscribe to this blog by email, in a reader, or follow it on Twitter

Quoted on Slaw, the Heenan Blaikie LLP Entertainment and Media Law Signal, and also posted on the International Forum for Responsible Media.


Get every new post delivered to your Inbox.