The issue of anonymous parties engaging in defamation has been one of the primary issues in online defamation. In many cases, the matter is often resolved as soon as this identity is ascertained. As such, de-anonymizing has been one of the primary strategies employed by plaintiff’s counsel in such actions.
In Manson v John Doe, Justice Goldstein granted judgment against an anonymous blogger who had been noted in default, and stated,
 There are few things more cowardly and insidious than an anonymous blogger who posts spiteful and defamatory comments about reputable member of the public and then hides behind the electronic curtain provided by the Internet. The Defendant confuses freedom of speech with freedom of defamation. There are, undoubtedly, legitimate anonymous Internet posts: persons critical of autocratic or repressive regimes, for example, or legitimate whistleblowers. The Defendant is not one of those people. The law will afford his posts all the protection that they deserve, which is to say none.
The Law Commission of Ontario described this phenomenon in the 2018 Consultation Paper in their project, Defamation Law in the Internet Age, as follows,
1. Motions to Identify Anonymous Speakers (Norwich Orders)
The anonymity possible in internet communications may be illusory. Website hosts, including social media platforms, may have identifying information for users where this was provided as a requirement of registering for the service. Even where this information is not directly available, website hosts are usually able to link particular content to the internet protocol (IP) address of the computer or other device used to post the content. Once the IP address is known, the internet service provider (ISP) will be able to link that address to the particular subscriber who registered that computer or device.
Therefore, website hosts and ISPs will often have access to identifying information about an anonymous poster. Experience demonstrates, however, that they are unlikely to release this information to a complainant without a court order for two reasons. First, they will generally owe a duty of confidentiality to their customers under their terms of service. Also, they are subject to the Personal Information Protection and Electronic Documents Act (PIPEDA). Under PIPEDA, an ISP is not permitted to disclose a subscriber’s personal information without the subscriber’s knowledge or consent, except in certain specified circumstances, including where required by court order.
Therefore, a complainant faced with an allegedly defamatory anonymous comment must generally bring a motion for an order requiring the website host and ISP to release information identifying a subscriber. Ontario’s rule 31.10 provides for a court order for discovery of a non-party “who there is reason to believe has information relevant to a material issue in the action” so long as the information cannot otherwise be obtained and, on balance, fairness requires it.
It is also possible even before commencing an action to bring a motion for an equitable bill of discovery known as a Norwich order. The rationale behind the Norwich order was explained in the decision which gave it its name:
[I]f through no fault of his own a person gets mixed up in the tortious acts of others so as to facilitate their wrong-doing he may incur no personal liability, but he comes under a duty to assist the person who has been wronged by giving him full information and disclosing the identity of wrongdoers…[J]ustice requires that he should co-operate in righting the wrong if he unwittingly facilitated its perpetration.
Ontario courts have used both the Rules approach and the Norwich approach to support an order requiring intermediaries to disclose the identity of anonymous posters in certain circumstances. Indeed, it is not clear that these are separate procedures. Subsequent case law has treated these as a single approach and, for the purpose of this paper, we do likewise. In referring to the Norwich procedure, we refer to this amalgamated approach.
The leading case in Ontario is Warman v. Fournier. The court held that the Rules of Civil Procedure must be interpreted in a manner consistent with Charter rights and values, including the right of freedom of expression and privacy interests. It laid out a four part test for determining whether it could order a third party to disclose the identity of an anonymous online user. The Court must consider whether:
(1) the unknown alleged wrongdoer had a reasonable expectation of anonymity;
(2) the applicant had a prima facie case of defamation and was acting in good faith;
(3) the applicant had taken reasonable steps to identify the anonymous party and had been unable to do so; and
(4) whether the public interests favouring disclosure outweighed the freedom of expression and privacy interests of the unknown alleged wrongdoers.
These identification motions seek to balance the long recognized value of a broad discovery process that is evident throughout our common law system of legal procedure with the important Charter values underlying anonymous speech. As Robert J. Currie notes, our civil procedure generally reflects the “right to face one’s accuser” and this is contrary to the idea that anonymous defamation may lie beyond the reach of the law. But Currie goes on to say that “the deeply complex and potentially explosive relationship between technology and privacy is touching and sometimes reshaping many of the moving parts in the administration of justice, including procedure”.
The paper also highlights some of the contemporary challenges of obtaining a Norwich Order in these contexts, including the need of the courts to ensure appropriate freedom of expression and privacy interests for anonymous speakers online. Even once such orders are obtained, there are logistical challenges in identifying these defendants when a public or shared computer is used. The use of proxy servers complicate this identification even further.
Consequently, the paper says “no matter how meritorious a claim may be, pursuing a Norwich motion may be akin to rolling very expensive dice.” Although alternatives have been suggested, much of the debate has centered around who is responsible for initiating these reforms.
For example, the Supreme Court of Canada’s decision in Haaretz.com v. Goldhar refers to the Consultation Paper, and was very much characterized by a philosophical divide over legislative reform. Justice Wagner’s concurring reasons resisted the need for legislative initiatives, and instead encouraged to Court to take action at the immediate juncture,
 …As private international law in common law Canada is almost entirely judge-made law, I see no need to wait for legislative initiative in this area or for the completion by the Law Commission of Ontario of its reform project. In Éditions Écosociété Inc. v. Banro Corp., 2012 SCC 18,  1 S.C.R. 636, at paras. 58-62, LeBel J. noted that the importance of place of reputation has long been recognized in Canadian defamation law. Despite citing several commentators in favour of the idea, LeBel J. nonetheless left the question of whether to modify the choice of law rule for multijurisdictional defamation cases for “another day”. That day has now arrived.
In contrast, the joint dissenting reasons by McLachlin C.J. and Moldaver and Gascon JJ. resisted changing the law and applying a policy preference involving the most substantial harm rule around jurisdictional issues in online defamation, citing numerous complexities around its interpretation and concerns of the proliferation of “mini-trials” at a preliminary stage of proceedings.
A new decision by Justice Myers of the Ontario Superior Court of Justice in Theralase Technologies Inc. v. Lanter has provided additional judicial led reform in this area. In this default judgment motion, Justice Myers concluded the court has jurisdiction to grant a judgement against anonymous online defamers,
 …I find that where a form of service can reasonably be expected to bring court proceedings to the attention of an unidentified defendant at whom the litigation finger has been appropriately pointed, judgment can issue against the defendant despite the fact that the plaintiffs and the court do not know the defendant’s name.
The consequence of this is that plaintiffs in online defamation cases have another approach to deal with anonymity, one that will likely cost less and may have more certainty than a Norwich Order.
The case involves defamatory statements made in an online forum, which focuses on investors in North America. The website contained information related to companies that are traded on North American exchanges. The forum portion of the website is populated by pseudonyms, and these accounts are created by email address. These accounts also include a messaging function.
The plaintiffs were successful in obtaining a Master’s order to serve the defendants by email and private message on their accounts for the site. All but one of the defendants failed to respond to the statement of claim, resulting in a deemed conclusion under Rule 19.02(1)(a) that the defendants admit the allegations against them. The plaintiffs then moved for default judgment.
Justice Myers noted that civil proceedings are assumed to involve claims and judgments in personam (against a person), which is why “John Doe” placeholders are used until a defendant is identified. But nothing in the Rules of Civil Procedure contemplates a final judgment against an unidentified defendant.
Relying on the 2019 Supreme Court of the United Kingdom decision in Cameron v. Liverpool Victoria Insurance Co Ltd., Justice Myers adopted their reasoning and framework to extend this procedure to Ontario to allow the court the ability to grant judgment against an unidentified defendant, specifically to allow for the plaintiffs to reasonably bring the proceedings to the attention of specific identifiable defendants,
 …Where, for example, people are hiding behind internet anonymity to make allegedly defamatory comments on a website, service through the website using the coordinates and the identifiers that the users themselves provided to the website operator strikes me as both reasonable and just. If notice does not reach the users, it is because they choose not to access the accounts from which they made their comments or the email addresses that they provided to the website operator. Where there is evidence that a person is actively evading service, such as by shutting down a previously active email address or website account after learning that an action exists, correspondingly less certainty of service may be required as long as it remains conceptually possible…
There are future logistical challenges about amending the title of proceedings for enforcement purposes, but Justice Myers did not make any determination on those issues at this time.
This approach is an exciting one, not only for the access to justice implications due to streamlining of process alluded to above, but also because of the tantalizing possibility of better results. While it might be relatively easy or convenient to ignore the service of pleadings that are served by substitute service under Rule 16.04, it’s an entirely different matter to receive a judgment and then face collections on that undefended claim.
Of course there is always the concern that there are legitimate defendants that have not checked a specific email address, perhaps one that was created for the exclusive purpose of participation in a single online forum, or that the messages sent to an account were never received because a user never returned to a site after posting defamatory content. Fortunately, the threshold for setting aside a default judgment is a low one in Ontario.
The Ontario Court of Appeal set out the test for setting aside a default judgement in 2014 in Mountain View Farms Ltd. v. McQueen as follows:
- whether the motion was brought promptly after the defendant learned of the default judgment;
- whether there is a plausible excuse or explanation for the defendant’s default in complying with the Rules of Civil Procedure;
- whether the facts establish that the defendant has an arguable defence on the merits;
- the potential prejudice to the defendant should the motion be dismissed, and the potential prejudice to the plaintiff should the motion be allowed; and
- the effect of any order the court might make on the overall integrity of the administration of justice.
This test has recently been applied in Kaur v. Janeallam to illustrate that the motion to set aside can be brought on an urgent basis, and in 2355305 Ontario Inc. v. Savannah Wells Holdings Inc. to illustrate how low threshold an arguable defence is needed. The prejudice suffered by an anonymous defendant who attempts to set aside a default judgment for online defamation is relatively small given the public interests at stake, and the prejudice experienced can typically be compensated through costs.
What this procedure does is ensure that anonymous defendants in online defamation cases actually step forward. Their participation, or lack thereof in proceedings, have been the most significant obstacle to preventing defamatory content online.
Once they are involved formally, the parties have an opportunity to discuss further resolution, which typically includes prioritizing a removal of the defamatory content even beyond claims of monetary damages. Default judgment against anonymous defendants therefore has the potential to address the concerns of plaintiffs in more cost effective manner, and strengthens the ability of the courts to protect reputations in the digital era.