When lawyers say that hard cases make bad law, they usually mean that extreme or unusual circumstances provide poor basis for making legal rule that would have to be applicable to a wider range of more common cases. Sometimes the phrase describes cases that involve a party whose hardship draws sympathy even if its legal case is weak. But sometime hard cases can make good law, when they present smart judges with difficult dilemmas and force them to think hard and deep on their ruling and its broader consequences. Yet courts don’t always choose the cases that come before them and the possibility of a hard case making bad law is an occupational hazard of the legal system.
One of the methods that courts use to mitigate this hazard is by refraining to deal with hard cases and to answer tough questions unless they must. Yesterday’s decision of the Supreme Court of Canada in Google v. Equustek might belong to the second type in cases, but in my view it belongs to a fourth one: cases that are made unnecessarily hard. These are relatively easy cases that for one reason or another appear to be very tough ones. In such cases courts confront avoidable dilemmas that should have awaited another time and they risk making bad law, avoidably.
Michael Geist provides the following concise summary of the underlying facts. He explains:
This case stems from claims by Equustek, a Canadian company [Datalink], that another company used its trade secrets to create a competing product and engaged in misleading tactics to trick users into purchasing it. After struggling to get the offending company’s website taken offline, Equustek obtained a British Columbia court order requiring Google to remove the site from its search index. Google voluntarily removed search results for the site from Google.ca search results, but was unwilling to block the sites from its worldwide index. The B.C. court affirmed that the order applied on an international basis, however, issuing what amounted to global takedown order. The case then proceeded to the Supreme Court of Canada.
The Supreme Court, by a majority of 7 to 2, upheld the worldwide injunction against Google.
As presented to the Court, the case involved two though questions: was it proper for the court to order Google to do certain things when Google, who is not a party to the underlying legal dispute between Equustek and Datalink, has done nothing wrong or unlawful (the non-party injunction question), and was it proper for the court to issue a worldwide injunction that orders Google to de-list Datalink’s websites and make them undiscoverable not only for users of Google in Canada but worldwide (the worldwide injunction order question). Justice Abealla, for the majority, answered “yes” to both questions, while Justices Côté and Rowe, in dissent, answered that while the court had the power to issue such an injunction against Google, the court should not have exercised that power in this case.
A lot of the commentary following the decision focused on the worldwide injunction question, how it might affect the governance of the internet, and the problems it may present to freedom of expression and other human rights if courts in countries with more oppressive regime might follow Canada’s lead and issue similar worldwide injunction in order to prevent access to content they deem unlawful. For excellent commentary on this issue see the posts of Michael Geist, Daphne Keller (former Associate General Counsel at Google), or Mike Masnick on Techdirt.
So far, the non-party injunction question hasn’t attracted much attention even though in my view it involves serious rule of law issues. The majority glossed over them and the dissent merely scratched their surface and I hope to write a separate piece that question in the future. The Court confronted very hard questions that don’t lend themselves to easy answers, no wonder it made problematic law.
But this case didn’t really have to be a hard one. It only became a hard one because almost no attention was given to the underlying dispute between Equustek and Datalink. I am confident that if the courts, at all three levels, paid a little more attention to Equustek’s claims against Datalink, they would have noticed a serious mismatch between Equustek’s legal claims and the orders that it sought, and they could have resolved this case by answering simple question. The hard ones could await another day.
It’s not that Equustek’s couldn’t possibly win on the basis of one or more of its claims against Datalink. Maybe it could. Maybe it couldn’t. Unfortunately, we might never know. Datalink abandoned its defence and fled the jurisdiction, and Google, who was not a party did not want to be a party, has chosen not to challenge the merit of the case between Equustek and Datalink and to focus on resisting the order that was sought against it. Google’s choice is understandable, but I can’t help thinking that the outcome of the case would have been very different if more attention was paid to the underlying intellectual property claims and their territorial scope instead of focusing on whether the court had personal jurisdiction over Google, whether it could ever grant a worldwide injunction, and what it could mean for freedom of expression and other human rights.
I have a strong feeling that if a little more time was spent on examining the nature and scope of the underlying IP claims, the courts would have easily realized that as a matter of IP law, the injunction that Equustek sought was overly broad and unjustified. If the court had been properly briefed on the underlying IP issues, they would not have to confront really tough questions and risk making bad new law. Instead, they could have resolved this case by applying tried and true good existing law.
One of the remarkable things about this case is how little we know about the underlying IP claims. Nowhere in the decision does the Court mention any Canadian IP statute, any specific provision of any such statute, or, in the case of common law claims, refer to the relevant doctrines using their standard legal terminology. The majority opinion says almost nothing. In para 1, Justice Abella explains that Datalink has been “using [its] websites to unlawfully sell the intellectual property of another company” (what does using a website to sell intellectual property even mean? The term “intellectual property” denotes certain legal rights. Equustek alleged that Datalink operated its websites in a way that infringed some of its intellectual property rights and sold products that infringed others, but saying that Datalink sold Equustek’s IP reflects how inattentive the court has been to the underlying dispute). Para 3 tells us a tiny bit more: “[Equustek] claimed that Datalink, while acting as a distributor of Equustek’s products, began to re-label one of the products and pass it off as its own. Datalink also acquired confidential information and trade secrets belonging to Equustek, using them to design and manufacture a competing product, the GW1000.”
We can glean a bit more from the dissenting opinion, where it describes that Equustek sought a temporary and permanent injunction restraining the Defendants from:
a. using the Plaintiffs’ trademarks and free-riding on the goodwill of any Equustek products on any website;
b. making statements disparaging or in any way referring to the Equustek products;
c. distributing the offending manuals and displaying images of the Plaintiff’s products on any website; and
d. selling the GW1000 line of products which were created by the theft of the Plaintiff’s trade secrets; and obliging them to:
e. immediately disclose all hidden websites;
f. display a page on all websites correcting [their] misrepresentations about the source and continuing availability of the Equustek products and directing customers to Equustek.
Clauses a, b, and f allude to possible claims under the Trade-Marks Act or the common law torts of passing off and trade libel. They activities described in clause c sound like alluding to claims of copyright infringement, and clause d refers to trade secrets. None of those claims, even if substantiated, would normally support the wide injunction that was granted.
For example, if Equustek alleged infringement a registered trademark and the courts hearing this case spent some time trying to understand its legal claims, they would have surely noticed that under Section 19 of the Trade-marks Act, Equustek would have an exclusive right to use the mark in association with the goods “throughout Canada”, bot not “worldwide”. This would immediately highlight the fact that no Canadian law is violated when someone uses a Canadian trademark when selling goods outside Canada. In turn, this would have forced the courts to acknowledge the territorial nature of trademark law and consider the appropriateness of an order enjoining other persons from doing certain things could not violate Canadian law.
If Equustek’s claim concerned an unregistered trademark, then Equustek could still have a claim for passing off, but then the right and the remedy would have to be limited to protecting Equustek’s local goodwill in BC. If the court’s attention was directed towards that issue, it might have recognized that even before it dealt with the difficult question of an injunction that seeks to regulate conduct outside Canada, it would have to grapple with the problem of an injunction that attempts to regulate conduct in another province.
Clause b seeks to prohibit making statements disparaging or in any way referring to the Equustek products. Making disparaging statements on a competitor’s product could be actionable, but not any type of disparagement. Section 7(a) of the Trade-marks Act prohibits only making “a false or misleading statement tending to discredit the business, goods or services of a competitor”, and the common law of trade libel covers the same acts if done “willfully”. Section 22 of the Trade-marks Act “prohibits the use of another person’s trademark “in a manner that is likely to have the effect of depreciating the value of the goodwill attaching thereto” but it applies only to registered trademarks. Canadian trademark law does not prohibit other forms of disparagement. And in general, referring to the products of another person is not only totally legal, but laws purporting to limit them might as well be challenged as unconstitutional limitation on freedom of expression. Instead of raising hypothetical scenarios of extraterritorial injunctions that might violate constitutionally protected speech in other countries, focusing on the claims in this specific case would have allowed Google to draw the courts’ attention to how the injunction at issue might conflict with the Canadian Constitution. I don’t know how the courts would decide those issues, but I am confident that they would have been more reluctant to grant a broad worldwide injunction if their invited to think about those issues.
Clause c refers to actions that might constitute copyright infringement, but Canadian courts can only enforce Canadian copyrights and prohibit acts that take place in Canada. It doesn’t mean that an operator of a website outside Canada could never infringe a Canadian copyright, but again, I am confident that if the courts in this case were properly briefed on those issues, they wouldn’t have issued a worldwide sweeping injunction.
Clause d seeks an interim and permanent injunction prohibiting the sale of products that embody misappropriated trade secrets. But alas, in Cadbury Schweppes Inc. v. FBI Foods Ltd., the Supreme Court of Canada vacated a permanent injunction granted by the BC Court of Appeal, which prohibited the continued use of the confidential information, or products derived therefrom. While the Court did not rule out the possibility of injunctive relief in trade secret cases, it signaled that this should only be done in rare circumstances. Among other reasons, the Court noted that
“Imposing a disability on a party in possession of confidential information from participating in a market in which there is room for more than one participant may be unreasonable, such as where the information relates to a manufacturing process or a design detail. In such cases, it may be that the obligation on the confidee is not to use the confidential information in its possession without paying compensation for it or sharing the benefit derived from it.”
In other parts of that case, the Court highlighted the difference between the remedies available for an infringement of a patent and those for trade secret misappropriation. The Court warned against an approach that would give the holders of trade secrets the same type of remedies available to patentees, when protection of trade secrets does not require anything that resembles non-obviousness, and does not involve the disclosure of the information, which is fundamental to the patent bargain. Now, maybe the facts of this case would justify an injunctive relief, even under the standards of Cadbury Schweppes. But then maybe they won’t. I can’t tell, because nobody asked the question. I am confident, however, that if the Court would have been more reluctant to uphold a worldwide injunction against Google, when its own precedent suggests that even the injunction against Datalink should not have been issued.
Apparently, Google never disputed the merit of Equustek’s claims. Maybe it didn’t want to be seen as siding with the defendants, though I think that it could have explained the underlying IP laws without taking sides. Maybe it thought that it would secure a bigger victory if it framed the case as involving hard questions. Evidently, though, this choice didn’t work well. But the problem isn’t Google. The problem is that the Court never paused to consider whether Equustek was, as a matter of Canadian IP law, entitled to the extraordinary injunction that it sought. As Howard Knopf reminded me, the International Trademark Association sought leave to intervene in the case in order to draw the Court’s attention to some of the underlying IP issues, but the Court, strangely and unfortunately, was not interested. The Court, perhaps fascinated by the hard questions, just took it for granted that Equustek was entitled to the extraordinary injunction against Datalink, and then held that it was appropriate to require Google to help enforcing it.
Hard cases make bad law. Unnecessarily hard cases make unnecessarily bad law.