Defamation / Reputation
Johnson v. Steele
Nominations Are Now Open for the 2024 Columbia Global Freedom of Expression Prizes. Learn more and nominate here.
Closed Mixed Outcome
Global Freedom of Expression is an academic initiative and therefore, we encourage you to share and republish excerpts of our content so long as they are not used for commercial purposes and you respect the following policy:
Attribution, copyright, and license information for media used by Global Freedom of Expression is available on our Credits page.
The Supreme Court of Canada ruled that Israel was “clearly the most appropriate forum” for a multi-jurisdictional defamation suit brought by Canadian businessman Mitchell Goldhar against the Israeli media group Haaretz. The case concerned an article published by Haaretz which was critical of Goldhar’s management of a professional Israeli football team he owned. The article was published in Israel but was also available online in English and hence accessible to Canadians. Although Goldhar had reputational interests in both jurisdictions, he attempted to sue for defamation in the Canadian courts. On appeal, the Supreme Court of Canada applied the simpliciter test and the forum non conveniens analysis to determine the most fair and efficient way to proceed based on convenience, expense, and ease of enforcement would be to move the hearing to Israel.
In 2011, Israeli media group Haaretz published an article criticizing Canadian businessman Mitchell Goldhar for his management of the professional Israeli football club he owns. When the article was published, Haaretz had a daily print circulation of about 70,000 in Israel. Haaretz also made the content available online in both Hebrew and English which meant that it was accessible in Canada. It appeared that between 200 and 300 Canadians accessed the article – including employees at Goldhar’s Canadian business (para. 8).
Goldhar, believing that the content was libellous, brought suit before the Ontario, Superior Court of Justice in Canada. Haaretz filed a motion to dismiss, arguing that Canada did not have jurisdiction over the matter or that, alternatively, Israel was clearly a more appropriate forum. That Court dismissed the motion finding that Canada did have jurisdiction over the matter and that Israel was not clearly a more appropriate forum. Haaretz appealed this decision to the Ontario Court of Appeal, which dismissed the appeal. Haaretz then appealed to the Supreme Court and the matter was heard before the Supreme Court on November 29, 2017.
In a five-to-three decision, the Supreme Court ruled in favor of Haaretz stating that the appeal should be allowed and granted Haaretz’s motion to stay the action. Côté, Brown and Rowe J.J. wrote the majority judgment. Karakatsanis J, Abella J, and Wagner J wrote separate concurring judgments, and McLachlin C.J. and Moldaver and Gascon J.J. wrote a joint dissenting judgment. The central issue of the case was whether the Canadian courts did have jurisdiction to hear the matter and, even if Canadian courts did have jurisdiction, whether Israel is a more appropriate forum for the claim to be held.
In the majority judgment, Côté explains that she held that Goldhar had not limited his claim to “libellous statements pertaining to his Canadian business or damage to his Canadian reputation” and that the majority judgment differed from the dissenting judgment in this respect (para. 20).
Côté set out the “fundamental principles underlying the conflict of laws” and defined the two different standards to be met in order for a court to have jurisdiction over a case: the jurisdiction simpliciter test and forum non conveniens doctrine. The jurisdiction simpliciter test establishes if the court has jurisdiction and requires that a “real and substantial connection” exist between the forum and the subject matter of the litigation. The forum non conveniens doctrine gives the Court the ability to dismiss a case, despite the existence of jurisdiction, because a more appropriate forum exists.
Côté stated that while the lower court judge had properly utilized the jurisdiction simpliciter test, he had made multiple errors in the forum non conveniens analysis. Côté explained that in determining whether there was a “real and substantial connection” the Court must determine whether there has been a presumptive connecting factor and then when that presumption has been rebutted. In determining this connection it must be “objective factors” that are taken into account to establish the “relationship between the subject matter of the litigation and the chosen forum” (para. 36). With reference to the Canadian case of Crookes v. Holloway 2007 BCSC 1325, 75 B.C.L.R. (4th) 316, Côté held that “the situs of Internet-based defamation is the place where the defamatory statements are read, accessed or downloaded by the third party” (para. 36).
Accordingly, Côté held that there was a presumptive connecting factor between the subject matter of this case – the publication of alleged defamatory material – and the forum chosen by Goldhar (para. 38). However, Côté explained that this was not the end of the analysis, and that it was necessary to determine whether this presumptive connecting factor had been rebutted.
Côté addressed the concerns that Haaretz had raised “about the ease with which a presumptive connecting factor may be established in Internet defamation cases” (para. 39). She mentioned that this concern had also been noted in the Canadian case of Éditions Écosociété Inc. v. Banro Corp.,  SCC 18. Côté explained that the implication of effect of this is that a “careful examination” of the rebuttal is “therefore of particular importance in Internet defamation cases” (para. 40). She added that “[t]he ability to rebut the presumption of jurisdiction where there is only a weak relationship between the subject matter of the litigation and the forum serves as an important check on jurisdiction” (para. 40).
The manner in which the presumptive connecting factor can be rebutted is by showing that “a given connection is inappropriate in the circumstances of the case” and that “the relationship between the forum and the subject matter of the litigation is such that it would ‘not be reasonable to expect that the defendant would be called to answer proceedings in that jurisdiction’.” (para. 43). Côté gave the example of a defendant showing that a “plaintiff has no reputation in the chosen forum” as a way of rebutting the presumption of jurisdiction in a defamation action (para. 44).
Côté held that as Haaretz knew that Goldhar lives and operates his business in Ontario – and that the alleged libellous article referenced this fact – Haaretz had not succeeded in proving that the Canadian jurisdiction had been rebutted (para. 45).
However, the Court was then required to determine the forum non conveniens question. Here, Côté explained that Haaretz bore the burden of demonstrating that “the alternative forum is ‘clearly more appropriate’.” (para. 46). Côté indicated that she agreed with Pepall J.A. in the Banro case that because jurisdiction is easy to establish in Internet defamation cases, “a motion judge must conduct a robust and carefully scrutinized review of the issue of forum non conveniens” (para. 48). Côté noted that “the establishment of a presumptive connecting factor is ‘virtually automatic’ in Internet defamation cases (para. 48). Here, Côté gave the example of a defendant having a reputation in numerous jurisdictions, and that it is only through the forum non conveniens analysis that it can be determined where it would be more fair and efficient to proceed (para. 48).
Côté found that the lower court had erred in answering the forum non conveniens question, and that, inter alia, the Court had erred by “failing to weigh Goldhar’s interest in vindicating his reputation in Ontario against the significant unfairness that a trial in Ontario would impose on Haaretz” (para. 50). Côté explained that there were six factors to take into account: the comparative convenience and expense for the parties; the comparative convenience and expense for the witnesses; the loss of legitimate juridicial advantage (such as whether there is a jury trial in one jurisdiction and not the other); fairness; the likelihood of enforcement; and the applicable law.
Côté held that “Haaretz would “face substantial unfairness and inefficiency if a trial were held in Ontario” (para. 95), and that a comparison of the convenience, expense, and ease of enforcement would also favor a hearing in Israel. Côté noted Goldhar’s significant business and reputation in Israel, and although she found that there were factors that would have favored a hearing in Ontario she held that these factors should not be given as much weight as those that favored the hearing in Israel (para. 96).
In respect of the applicable law factor, Côté acknowledged that, under the lex loci delicti rule (that is, the general principle for determining the choice of law is the place where the tort occurs), Ontario law applies to the present case but that “the lex loci delicti rule may allow courts in multiple forums to assume jurisdiction and apply their own law” (para. 92). She noted that “[i]n an interconnected world where international players with global reputations are defamed through global publications, this is unsurprising” (para. 92). However, she cautioned against utilizing a “substantial harm” test rather than an “applicable law” test to determine which forum should be favored because, as in this case with Goldhar, it would be difficult to determine where the most damage to a reputation would occur (para. 94).
Accordingly, Côté held that “Israel is clearly a more appropriate forum” (para. 97).
The concurring judgments disagreed with Côté on the weight given to the factors in the forum non conveniens analysis. Karakatsanis J. argued that the applicable law factor is to determine if the plaintiff’s chosen court would apply a different foreign law, and not an assessment of “what law would apply in the alternative jurisdiction” (para. 100). He also disagreed that Goldhar’s reputation in Israel is relevant to the fairness analysis and states that a plaintiff has a right to vindicate his reputation in the jurisdiction in which he enjoys that reputation – in this case, Canada (para. 101).
In the second concurring opinion, Abella J. explained why she believes that a substantial harm analysis should replace the lex loci delicti analysis. She stated that “adopting ‘most substantial harm’ for choice of law would ensure that the choice of law rule reflects what is at the core of the tort of defamation — protection of reputation” (para. 113). Abella expressed concern that the Canadian approach allows for an assumption of jurisdiction to exist based on a “single download” (para. 117). She concluded that “[a]dopting the place of most substantial harm to the plaintiff’s reputation when deciding the applicable choice of law would also arguably strike a better balance between freedom of expression and harm to reputation concerns” (para. 119). Accordingly, Abella stated that the alleged libellous article would have caused greater harm to Goldhar’s reputation in Israel (para. 135). She agreed that an analysis of the other factors in the forum non conveniens analysis would also support a finding that Israel was the more appropriate jurisdiction.
Wagner J. agreed with Abella that the substantial harm test is more appropriate in internet defamation cases, and stated that “this Court should make such a modification to the choice of law rule in the specific context of Internet defamation” (para. 144).
In their dissenting judgment, McLachlin C.J. and Moldaver and Gascon JJ. explained why they would have dismissed the appeal and held that Israel was not the more appropriate forum. They would have held that the “high threshold set by the ‘clearly more appropriate’ test is not met in this case” (para. 161), and disagreed with Abella and Wagner JJ that a “substantial harm” test should be adopted in Canada.
The dissenting judges noted that Goldhar “has a real and long-standing reputational interest in Ontario” and that “[t]he sting of the article relates to his reputation in Ontario” (para. 213). They would have found that “his reputation in Israel is not material to the analysis” (para. 218) and that it would have been fair to allow Goldhar to vindicate his reputation in Ontario (para. 214). The dissenting judges concluded the forum non conveniens analysis by stating that the “key factors of applicable law and fairness to the parties weigh heavily in favour of Ontario, while the factor of loss of legitimate juridical advantage also weighs in favour of Ontario” (para. 238). They acknowledged that the factors of comparative convenience and expense for the parties and witnesses did favor Israel, but “only slightly” (para. 238). While acknowledging that it would not be unreasonable to hold the trial in Israel, they denied that it had been established that Israel was “a clearly more appropriate forum” (para. 239) and would have held that Haaretz had not met the high threshold required to find that the trial should have been held in Israel.
Decision Direction indicates whether the decision expands or contracts expression based on an analysis of the case.
This case does not create a hard and fast rule in dealing with internet defamation in Canada but does address some of the legal issues associated with a globally accessible internet, such as choice of law and jurisdiction issues surrounding content posted on the internet. This case is important for what it did not do – it did not force a newspaper to be subject to a foreign court because their content was available online and, therefore, accessible to individuals in that foreign country.
Global Perspective demonstrates how the court’s decision was influenced by standards from one or many regions.
Case significance refers to how influential the case is and how its significance changes over time.
Let us know if you notice errors or if the case analysis needs revision.