Does Including a Forwarding Feature to Defamation Amount to Republication?

The Supreme Court of Canada in Crookes v Newton held that the mere linking to a web site that contained defamatory material did not make the linker liable for defamation. Adding content to the link might change that result.

The Supreme Court of British Columbia has recently held, however, that offering a link to an email program (e.g. ‘mailto:’) on a web page that contains defamatory material constitutes republication of that material, apparently whether or not anyone used it.

Weaver v Corcoran 2015 BCSC 165 (CanLII)

Here is the main passage on that point:

[261] The invitation to email the article to a friend meets the test set out in the jurisprudence that re-publication is the “natural and probable result of the original publication”. The invitation is to email an article, the content of which is known and indeed created by the defendants. This is unlike the situation in Crookes which involved a website concerning commentary on various issues with hyperlinks to other websites, the content of which is not controlled by the operator of the website. In this context, I am prepared to conclude that a reasonable inference could be drawn of publication to a third person of each of the articles at issue.

(You will see from that paragraph number that it’s a long decision.)

Most web browsers offer the capacity to send on either a link to the page being viewed or the page itself. Does that make the browser liable, under this ruling, or the publisher of the original page more deeply liable because it must be taken as general knowledge that anything on the web can be forwarded (not just linked to) through the browser used to view it? Or must there be an ‘invitation’? If so, does it make sense to treat what is probably a pretty routine feature of web pages of this sort (newspaper pages, but also magazines and others) as an automatic increase in aggravation of defamation?


  1. The ‘mailto:’ use in a hyperlink simply triggers the users email software. By default, there is no message. It could contain anything, depending on whether the coder of the page has affixed a message as an attribute within the tag.

    That a page is sharable, either by email software, or triggering a social media sharing function, is not only common place, but a standard feature of browser software. This was also the case in 2011 and likely considered in Crookes, yes?

    Including social media share buttons, or using javascript that triggers an article sharing function, are ‘choices’ of the website publisher. That might be an argument going the other way. But even then, if a feature is implemented as a standard design feature, across an entire website, I think it could be established that the sharing function preceded the content in question; or was implemented independently of any published content.

  2. The case also raises an interesting issue on forums/comments, finding that they can be but aren’t if you monitor and remove:

    [269] The parties are of the view this case raises, for the first time in Canada, the issue of whether one who operates an internet forum – in this case a reader comment area on the newspaper’s website – is liable for third-party postings.

    [275] I agree, as argued by the defendants, that the cases appear to establish the requirement to show an active or deliberate act in making defamatory information available to establish liability. (…)

    [287] Due to the prompt removal of the offending reader comments once known to the defendants, I have concluded the defendants are not publishers of the reader postings. Accordingly, I do not need to deal with the defence of innocent dissemination or fair comment.

  3. On James’s point, I don’t think the court said that the owner of the web page that is open for comments has to monitor the comments. It is sufficient if the owner removes defamatory comments once somebody draws his/her/its attention to them.

    I suppose that rule must make complaints operate a bit like a notice-and-takedown message under the DMCA in the US, though – the incentives are all in favour of removal, rather than standing fast on the ground that the comment is not defamatory or is justified in some way, just as content hosts in the US have no real incentive to decide that the works referred to in the DMCA notice are not subject to copyright or there is a legal justification for publishing them. Why expose yourself to the risk you’re wrong, or the costs of a lawsuit on the point? It’s cheaper and faster to remove them, and all that suffers is public discourse and freedom of expression…

    Maybe we need a notice-and-notice system for defamation as well as for copyright in this country. Or maybe a rule like the US one that immunizes third-party hosts from any liability for material they don’t create themselves (s. 230, Communications Decency Act.) Which brings us to my columns from last year on intermediary liability.

  4. “all that suffers is public discourse and freedom of expression…” Indeed. American journalist Barrett Brown had been facing a potential 105 years for linking to material hacked by Jeremy Hammond. ( Jeremy’s info was already in the public domain when Barrett linked to it.) In Jan. 2015, BB was given 63 months prison, (of which he has spent 3 yrs., I think, already), and $890,000 restitution to be paid. (And in J’s case, the judge did not recuse herself despite it being her husband’s private security firm that J. had hacked.) Many have been targetted for revealing info.

  5. I think the judge got it wrong in this case. She did not appear to have had a full appreciation and understanding of how “story tools” work on mainstream media sites work. The “story tools” the judge refers to earlier in her judgment are what every media outlet includes with their articles – icons that allow a reader to share an article digitally through email and various social media platforms like Facebook, Twitter, LinkedIn, etc. At paragraph 251 of the judgment, the judge notes the following:

    “[251] The evidence of re-publication includes numerous downloads of Weaver’s Web; Weaver’s Web II; Climate Agency Going up in Flames; and some of So Much for Pure Science from various websites. The defendants maintain, however, republication cannot be a mere hyperlink as stated in Crookes. Thus, the ‘story tools’ on the National Post website do not qualify as authorization. The evidence established these are merely permitted hyperlinks back to the original article.”

    She makes a finding that using these story tools are “merely permitted hyperlinks back to the original article,” and therefore that type of publication does not constitute a separate republication (and additional liability for the original publisher), in accordance with Crookes. But for some reason, the judge draws a distinction between those types of story tools that permit sharing on social media sites (by providing a link to the original story) and the link/icon that allows a reader to email the article to a friend. The judge fails to examine how the “email to a friend” link/icon actually works. Based on her finding, she seems to think that the invitation to email link allows a reader to make a whole new copy of the text of the article and send it in an email to a friend (not just provide a link to the original story). Therefore, she finds that, since emailing a whole new copy of the story is a natural and probable result of the original publication, the National Post is also liable for those subsequent “republications.”

    However, there is no distinction between sharing an article by email or by social media. When you click on the email link/icon on the Post’s site, it sends an email to your friend that includes any message you write, plus the first couple of lines of the story and a link to the original story on the Post’s site. I believe the other mainstream media outlets work the same way. It’s the same practical effect as if you shared the story using the Facebook icon – it provides a link back to the original story. According to the judgment, those type of social media story tools are “permitted hyperlinks back to the original article,” which the Post would not be liable for, based on Crookes. Yet she finds the Post liable for the republications because of the invitation to email the story.

    The republications in question were not situations where the articles in question were simply shared via social media story tools on the Post’s site. They appear to be situations where individuals may have simply cut and pasted the text of the article to an online forum, blog, or other site, or even into an email. She mistakenly uses the invitation to email link as the basis for finding additional liability for republication in those situations, despite the fact that using that link/icon is no different than the links/icons for sharing the story on social media. The trial evidence showed that the Post would sometimes send cease and desist letters to people who reproduce their content without permission because it violates their copyright, but they could only do that once they became aware of the violation:

    “[258] The contents of the impugned articles were however reproduced numerous times over the Internet. The National Post says this is in violation of its copyright. John Racovali, the Assistant Managing Editor of the National Post who is in charge of legal matters, said the National Post may occasionally send a ‘cease and desist’ letter but not much more as the National Post may not be aware of the violation.”

    But those appear to be the republications for which she held the Post liable – where the Post may not have even been aware of the republications and violations of their copyright.

    At the end of the day, liability of an original publisher for republication that is the natural and probable result of the original publication should only affect the quantum of damages because it potentially increases the number of readers of the defamatory comments. There needs to be a finding of liability of the original publication for liability of republication to come into play. In this case, the judge appears to have taken the republications into account in her assessment of damages, but does not indicate how much of an impact it had on the total damages. General damages in defamation cases are such a crapshoot and far from an exact science that it is difficult to know whether that finding of liability for republication had any real consequence on the damages awarded. But it appears to be one factor in her decision that a “significant award” was appropriate. But the Post should not have been liable for any additional damages she tacked on for those republications on the basis of the Post’s email link on its site.