1. Summary and comment
Is a Canadian citizen, allegedly defamed in an on-line article published in his home province by a foreign newspaper, entitled to vindicate his reputation in the courts of the province where he lives and maintains his business, and where the sting of the article’s comments is felt? According to the majority of the Supreme Court of Canada in Haaretz.com v. Goldhar1, the answer to this question is “no”.
While a majority of the Court concluded that Ontario had jurisdiction simpliciter, it also found, contrary to the courts below, that Israel was the clearly more appropriate forum for trial proceedings. The plaintiff’s claim was thus stayed in favour of litigation in the courts of Israel.
In this comment, we review the various reasons of the Court and offer guidance for businesses and individuals faced with potential internet defamation and the choice of whether to commence – or resist – such proceeding in Canada.
The Supreme Court confirmed that it is relatively easy for Canadian courts to assert jurisdiction over disputes with international dimensions. Even where an allegedly defamatory post is authored overseas, jurisdiction will almost certainly be asserted over the foreign defendant if the article is read in Canada and relates to a person with reputational interests in Canada. In this regard, a majority of the Supreme Court refused to adopt the ‘most substantial harm’ test (i.e., whether the plaintiff suffered comparatively more or less harm in the foreign jurisdiction) to determine whether a court may have presumptive jurisdiction over a given dispute. The lex loci delecti (law of the place where the tort occurred) will continue to ground presumptive jurisdiction. The good news, then, is that the Court has maintained a jurisdiction simpliciter framework that remains relatively predictable to litigants.
On the other hand, the Supreme Court’s decision has likely created uncertainty in a number of other respects. Exercising his discretion, the motion judge had concluded that Israel was not a clearly more appropriate forum than Ontario. That decision was (in the absence of an error in principle, failure to apprehend or take into account material evidence, or failure to reach a reasonable decision) entitled to considerable deference by appellate courts. Despite this, the majority of the Supreme Court decided to reweigh the evidence and, based on its own view of the record, impose a new decision. It did so by reviewing the statement of claim in detail and by finding that, notwithstanding an undertaking to confine the claim to Ontario, the complaint was not in fact so confined. In addition, the Supreme Court’s decision arguably clouds the issue of which party bears the burden of proving that requiring the foreign defendant to answer a claim in Canada would result in unfairness.2 Regarding letters rogatory, in particular, parties may be required to lead expert evidence on whether the foreign jurisdiction will honour such requests. The consequences to any plaintiff in Canada contemplating a proceeding against a foreign defendant are more than trifling:
Canadian plaintiffs must now pay far greater attention when drafting their claim, as it may be subjected to line-by-line jurisdictional analysis;3
An undertaking by the plaintiff to limit the scope of a claim to only those damages sustained in Canada may no longer be accepted by Canadian judges – further reinforcing the need for tightly drafted pleadings that are mindful of jurisdiction;4
Expert evidence will now almost certainly be required on the issue of whether Canadian letters rogatory will be accepted in the foreign jurisdiction;5
Although decisions of motion judges regarding forum non conveniens have traditionally been entitled to deference, this decision appears to invite a de novo analysis where appellate courts disagree with the weighting applied by the motion judge to forum non conveniens factors;6 and
In the context of jurisdiction disputes, it may be advisable to prepare detailed proposed witness lists to summarize not only the place of residence of each proposed witness but also each witnesses’ expected evidence, with indication as to the relevance of the evidence to the overall dispute.7
In 2011 Haaretz, an Israeli newspaper, published an article criticising Mitchell Goldhar, a successful Canadian businessman and owner of the most decorated football club in Israel. A print version of the article was widely disseminated in Israel and an online version, published in both Hebrew and English, attracted approximately 200 unique views from users in Canada.
Goldhar brought an action in the Ontario Superior Court of Justice against Haaretz, alleging that the article was libellous. In response, Haaretz brought a motion to stay or dismiss the action on three grounds:
The Ontario court lacked jurisdiction (jurisdiction simpliciter) over the dispute;
Israel was clearly the more convenient forum to hear the action (forum non conveniens); and
The action was an abuse of process.
3. Procedural history8
The motion judge held that the lawsuit should be allowed to proceed in Ontario, on two conditions: (1) that Goldhar’s damages would be limited to the reputational harm he suffered in Canada, and (2) that Goldhar would pay the travel and accommodation expenses of Haaretz’s witnesses, the majority of which resided in Israel. In allowing the matter to proceed in Ontario, the motion judge applied the framework previously established by the Supreme Court, in Club Reports Ltd. v. Van Breda,9 for determining whether a court should assume jurisdiction over a matter involving a foreign defendant:
Did the Ontario court have jurisdiction simpliciter over the defendant (i.e., was there a sufficient ‘real and substantial connection’ between the subject matter of the litigation and Canada sufficient to allow a Canadian court to assert jurisdiction over the defendant); and
If the Ontario court had jurisdiction simpliciter, should it nevertheless exercise its discretion to stay the action in favour of a proceeding in another clearly more appropriate forum?
The motion judge concluded that because the alleged tort occurred in Canada (through publication of the article on the internet and readership by individuals located in Ontario) jurisdiction simpliciter was established. The motion judge further concluded that the action should proceed in Ontario, as Israel was not a clearly more convenient forum.
The Court of Appeal upheld the motion judge’s order in a 2-1 decision. The fact that the article had been read by a small number of people in Ontario, relative to Israel, did not determine whether or not a Canadian court had jurisdiction. The dissenting judge found that Israel was a clearly more convenient forum and proposed that a ‘most substantial harm’ test should replace the lex loci delicti when determining choice of law in international defamation cases.
4. Supreme Court – majority decision
Six of the nine Supreme Court judges agreed that the Ontario proceeding should be stayed in favour of litigation in Israel. The judges that formed the majority arrived at this conclusion in different ways.
4a.The reasons of Justices Côté, Brown, Rowe, and Karakatsanis
The decision of Justice Côté (Brown J. and Rowe J. concurring) favoured the traditional two-step approach set out in Van Breda and the continued application of the lex loci delicti when determining whether there was a ‘real and substantial connection’ on which to base jurisdiction.
Côté J. held that because the alleged tort had occurred in Canada, a presumptive connecting factor to the jurisdiction had been established. This presumption can be rebutted, for instance if the plaintiff has no reputation in the jurisdiction.10 However, in this case, the presumption was not rebutted as Goldhar had a reputation in Ontario and the alleged defamatory article directly referenced his Canadian residency and businesses.11
Moving to consideration of the motion judge’s forum non conveniens analysis, Côté J. affirmed that the burden was on the defendant to satisfy the motion judge that an alternative forum was clearly more appropriate by establishing that it would be fairer and more efficient to proceed in that forum.12 Given the ease with which jurisdiction simpliciter can be established in cases of internet defamation, Côté J. cautioned that a robust forum non conveniens analysis must take place and that motion judges must be “particularly attuned to concerns about fairness and efficiency” at that stage.13
Côté J. confirmed that lower court decisions regarding forum non conveniens are entitled to deference, but an appellate court may intervene where there is an unreasonable conclusion, error in principle, or misapprehension or failure to take into account material evidence.14 With this in mind, Côté J. identified six errors in the motion judge’s forum non conveniens analysis15 and, in light of these errors, replaced the motion judge’s analysis with her own:
Comparative convenience for the parties: favoured Israel (not disputed by the parties).
Comparative convenience and expense for the witnesses: heavily favoured Israel, particularly as almost all of Haaretz’s proposed witnesses resided in Israel and the evidence “did not allow the courts below to ensure that Haaretz would be able to compel its witnesses to testify if the trial proceeded in Ontario”.16 Further concern was raised with respect to Goldhar’s undertaking to pay the travel expenses of Haaretz’s witnesses, as this sort of undertaking might permit a wealthy plaintiff to sway the analysis in a manner “inimical to the foundational principles of fairness and efficiency undermining this doctrine.”17
Loss of legitimate juridical advantage: favoured Ontario, but should not weigh heavily in the analysis.18
Fairness: favoured Israel. Goldhar had a multijurisdictional reputation and would suffer “no significant unfairness by having to bring a libel claim in Israel for comments that were written and researched in Israel and that pertain primarily to his reputation and business in that jurisdiction.”19 In this regard, the motion judge had failed to ensure that both parties were treated fairly by not weighing Goldhar’s interest in vindicating his reputation in Ontario against the significant unfairness that trial in Ontario would impose on Haaretz.20
Enforcement: slightly favoured Israel, as Haaretz had no assets in Ontario.
Applicable law: favoured Ontario, but should be given little weight. Significantly, Côté J. held that the analysis of applicable law, given the comparative nature of the forum non conveniens analysis, should consider the law of the alternative forum in addition to the law of the chosen forum.21 Côté J. refused to apply the proposed ‘most substantial harm’ test due to her reluctance to change the existing private international law framework.22 In any event, according to Côté J., the most substantial harm test would not have clearly favoured either forum as Goldhar had a substantial reputation in both Israel and Ontario.23
In the result, Côté J. held that Goldhar’s action should be stayed in favour of the clearly more convenient forum: Israel.
Karakatsanis J., in separate reasons, largely adopted the reasoning of Côté J., with one exception: as part of the forum non conveniens analysis, Karakatsanis J. held that consideration of applicable law should not be expanded to include consideration of the applicable law of the alternative forum – such consideration would unduly lengthen the analysis and not aid in determining whether a domestic court would have to apply foreign law at trial.24
4b. The reasons of Justices Abella and Wagner
Justice Abella held that the ‘most substantial’ harm should replace the lex loci delicti as a presumptive connecting factor in the jurisdiction simpliciter analysis for internet defamation cases. This test, according to Abella J., was more realistic in the internet age and helped avoid the risk of lawsuits being brought in whatever jurisdiction an allegedly defamatory online article was opened. Wagner J. did not agree with the application of this proposed test to the jurisdiction simpliciter stage.25
In considering the forum non conveniens analysis factors, Abella J. was concerned that a single download might be sufficient to establish applicable law in internet defamation cases.26 As with jurisdiction simpliciter, her preference was to consider the place where the plaintiff suffered the greatest harm to their reputation (as opposed to law of the place of publication).27 With reference to Australia and European jurisdictions, Abella J. called for a multi-factor contextual approach to determining where the most substantial harm was suffered.28
Wagner J. agreed with Abella J.’s application of a ‘most substantial harm’ test to the choice of law analysis: while there will be cases in which it will be challenging to identify the place of most substantial harm, “the range of possibly applicable law for a given dispute will be much narrower than with lex loci delicti and will be determined on a more principled basis.”29
Ultimately, on their respective analyses, Justices Abella and Wagner each favoured a trial in Israel, particularly given the financial strain a trial in Ontario would place on Haaretz.30
5. Supreme Court – dissenting reasons of Justices McLachlin, Moldaver, and Gascon
The dissent, written by McLachliln C.J., rejected much of the majority’s analysis and held that the only forum non conveniens factors that favoured a trial in Israel were comparative convenience, expense for the parties, and convenience for the witnesses.31
The dissent disagreed with Côté J.’s characterization of errors made by the motion judge: some of these errors, according to McLachlin C.J., were simple differences in the weighing of factors and were at odds with the deferential standard of review to be afforded to the motion judge.32 In general, the dissent disagreed with the approach of the majority, which, they stated, undermined stability and increased costs and uncertainty by setting a tone for further appellate intervention in jurisdiction decisions.33
On the issue of jurisdiction simpliciter, the dissent was of the view that the situs of the tort was sufficient to establish a presumptive jurisdictional connection: even in the case of libel on the internet, “the sting of the defamation is felt in the place where it is read.”34
Haaretz, according to McLachlin C.J., could not rebut the presumption of jurisdiction in this case: it was reasonably foreseeable that Goldhar, a Canadian, might seek to vindicate his reputation in Ontario. The existing standard for jurisdiction simpliciter was far from the automatic assumption of jurisdiction based on a single download, as was framed by Abella J.35
On the dissent’s forum non conveniens analysis, considerable emphasis was given to the history and selection of a “clearly more appropriate forum” as being the standard for exercising judicial discretion to stay a proceeding.
The dissent would not have adopted a most substantial harm test for applicable law in defamation cases, out of concern that such a rule would be highly subjective and would result in the proliferation of costly mini-trials. 36 The dissenting reasons also expressed concern regarding the majority’s analysis: that one party might be able to manipulate the witness factor by providing a list of foreign-domiciled witnesses, in order to sway the forum non conveniens analysis, without providing any further indication of their relevance or expected evidence.37
Haaretz.com v. Goldhar helpfully affirms the test for jurisdiction simpliciter in internet defamation cases. At the same time, however, it muddies the waters regarding those circumstances in which a court might exercise its discretion to stay a proceeding in Canada in favour of trial in another jurisdiction, including when an appellate court might deem it appropriate to overturn a motion judge’s decision on that issue.
The Supreme Court’s decision shows that the Canadian judiciary remains divided on how best to adapt existing private international law principles to the modern reality of borderless communications. While Haaretz.com v. Goldhar may not be the final word in this debate, it stands for now. In the result, litigants in matters involving foreign defendants are left with increased uncertainty regarding the appropriate forum for their claim and would be wise to examine jurisdictional issues with increased scrutiny and vigilance.
1 Haaretz.com v. Goldhar, 2018 SCC 28, para 151.
2 Ibid, para 63.
3Ibid, paras 20-22, 24, 78, and 132.
4Ibid, para 23.
5Ibid, paras 50 and 64.
6Ibid, para 179.
7Ibid, paras 223-224 and 227.
8For full commentary on the procedural history of this case, please see: (1) “Ontario Court Assumes Jurisdiction Over Israeli Publisher in Online Defamation Claim”, International Law Office (also published in Internet and E-Commerce Law in Canada, v.16, no. 8, December 2015) and (2) “Court of Appeal Affirms Jurisdiction Over Israeli Publisher”, International Law Office.
92012 SCC 17.
10Haaretz, supra, para 44.
11Ibid, para 45.
12Ibid, para 46.
13Ibid, para 48.
14Ibid, para 49.
15The errors identified by Côté J. were: (1) finding that letters rogatory could be used to compel Israeli witnesses to testify in Ontario; (2) giving too much weight to Goldhar’s undertaking to fund travel and accommodation expenses of foreign witnesses in accordance with rates provided in the Rules; (3) unreasonably discounting Haaretz’s proposed witnesses and the relevance of their evidence; (4) failing to consider Goldhar’s significant reputation in Israel; (5) failing to weigh Goldhar’s interest in vindicating his reputation in Ontario against the unfairness a trial in Ontario would impose on Haaretz; and (6) failing to consider the question of enforcement.
16Haaretz, supra, para 63.
17Ibid, para 66.
18Ibid, para 76.
19Ibid, para 78.
20Ibid, para 79.
21Ibid, para 89.
22Ibid, para 91.
23Ibid, para 94.
24Ibid, para 100.
25Ibid, para 147.
26Ibid, para 117.
27Ibid, para 109.
28Ibid, para 114-116.
29Ibid, para 145.
30Ibid, paras 133-139.
31Ibid, para 238.
32Ibid, para 179.
33Ibid, para 180.
34Ibid, para 166.
35Ibid, para 171.
36Ibid, paras 196 and 199-201.
37Ibid, para 227.