Introduction
The Supreme Court of Canada’s 2011 decision in the case of Crookes v. Newton is a prime example of how individual rights as ephemeral as reputation must be balanced with other freedoms, such as expression, in this powerful, evolving medium of the Internet.
Facts
Website owner Jon Newton posted an article on his site that contained two hyperlinks to allegedly defamatory articles regarding Wayne Crookes, a British Columbia businessman. Newton did not convey any obvious or independent support for the content in the articles hyperlinked. Crookes and his lawyer contacted Newton to ask him to remove the hyperlinks, but Newton refused to take them down. Crookes sued, claiming that by posting the hyperlinks and not removing them after receiving notification that they were defamatory, Newton was responsible for publishing defamatory material and was therefore was liable to Crookes in damages.
Hyperlinks and Defamation
Hyperlinks are images or words that, when clicked, take one to another site. Deep hyperlinks direct a website user straight to the destination page, while surface or shallow hyperlinks only bring the user to a different website’s homepage.1 Newton’s article contained both a shallow hyperlink which brought a web user to a homepage of a website called OpenPolitics, and a deep hyperlink which sent users directly to an article about Crookes on the website www.USGovernetics.com.
A simple – yet indispensible – feature of the Internet, the hyperlink is something web users take for granted today, despite a litigious history. In 2000, British Telecom sued one of the first Internet Service Providers (ISPs) for patent infringement. BT claimed one of its decades-old patents embraced hyperlinking and sought compensation for the ISP’s use of these hyperlinks. The court eventually concluded the ISP’s use of hyperlinks did not infringe upon BT’s patent.2
Hyperlinks originated with the Internet but defamation as crime and a tort has a much longer record. In Canadian common law, defamation is a strict liability tort. This means one is liable even if one did not intend to defame or realize that the defamatory remarks were false. Liability flows from proof of purposely or negligently publishing defamatory material to a third party.3 Courts attempt to balance personal reputation and freedom of expression. Technology adds a twist to reconciling these conflicting interests.
The Judicial Decision
All judges of both lower courts concluded there was no publication of defamatory material on Newton’s part. Hyperlinks were mere references, devoid of defamatory content of their own. They simply facilitated users to find it themselves. While Newton’s article containing the hyperlinks had been viewed 1788 times, there were no data on the number of times the hyperlinks were activated or the defamatory articles viewed. This meant there was no proof any third party had actually accessed one of the articles. Like many websites, the ones that Newton linked to did not record the number of people who had opened and viewed the articles involving Crookes.
All nine judges of the Supreme Court of Canada determined Newton’s hyperlinks not to be publications, but they were divided on the reasons and on what would constitute a publication. The majority of the Court reasoned that hyperlinks are nothing more than references. The hyperlinker has no control over the content to which the hyperlink is connected. Links merely enable a reader to find something already made available to an Internet audience. They are essential to the Internet. The inevitable “chill” arising from liability could destroy the Internet’s functionality: the Court wrote:“strict application of the publication rule in these circumstances would be like trying to fit a square archaic peg into the hexagonal hole of modernity.”
The Court went on to say that the person writing and posting the content is the publisher of it. A hyperlink is “content neutral” in that “it expresses no opinion, nor does it have any control over, the content to which it refers.” Only if the hyperlink itself contains defamatory text or repeats any defamatory wording contained in the article to which it is linked could a hyperlinker be found liable for defamation. This perspective is consistent with two American cases where references to allegedly defamatory material – one mentioned in a printed newsletter and the other on the radio – were found not to be publications.
Analysis and Conclusion
The Supreme Court was called to apply the centuries old tort of defamation to the modern Internet context. The Internet compels such flexibility. The Court had respect for Crookes’ reputation, but it also had to be mindful of the impact this decision would have on Internet usage and the flood of potential lawsuits that might be generated by a more expansive liability.
The decision offered protection to hyperlinkers who only draw attention and allow access to content already published. A conclusion that hyperlinks are never a publication for defamation purposes unless they repeat or add to something that has already been published can accommodate the broad, multi-layered, public utility scope and function of the Internet. Rendering hyperlinkers liable would do nothing to discourage its creation or to remove the defamatory content from the Internet.
The Supreme Court did not clarify what level of endorsement would make hyperlinkers liable for defamation. One of the British Columbia Court of Appeal judges (Madam Justice Prowse) was of the view that Newton’s article did encourage readers to follow the hyperlinks. This judge, along with one of the dissenting Supreme Court judges (Deschamps J.), was willing to assume that at least one of the 1788 viewers of Newton’s article, other than Crookes, had hyperlinked to the defamatory articles. Clearly, endorsing Hyperlinkers can escape liability if plaintiffs cannot identify at least one reader of the defamatory material accessed from the hyperlink.
Even tacit endorsement and re-publication, as the law now stands, will not be obvious thresholds to liability for defamation. There is almost always some text to accompany a hyperlink, if only something as terse as “see.” To generally suggest that an article be read for whatever reason might yet constitute endorsement. Likewise, endorsement and re-publication might be reached where the hyperlinker describes a linked-to text as “interesting,” “relevant or “noteworthy.” Therefore, liability for endorsement and re-publication remains vague and may be elusive in practice. This awaits refinement in future cases.
The Crookes v. Newton case represents an excellent example of the modern challenge facing judges to support both the Internet and individual interests such as reputation. While all these Supreme Court judges arrived at the same conclusion, they differed on the essential role hyperlinks play in defamatory publication.
Notes
1 Hasan A. Deveci, “Hyperlinks Citations, Reproducing Original Works,” (2011) 27, Computer Law & Security Review 465 at 466-68.
2 Heather Rowe, Lovells Baseby, and Francesca Baseby, “Case Report – BT Hyperlink Dispute – Did BT Invent the Hyperlink? Obviously Not as the US Courts Have Said So” (2002) 18:6 Computer Law & Security Report 439.
3 Raymond E. Brown, Brown on Defamation: Canada, United Kingdom, Australia, New Zealand, United States, loose-leaf, 2d. ed., Vol. 1 (Toronto: Carswell, 1999). ch. 3 at 1.