As Jakob’s post from yesterday notes, last week’s Svensson ruling came as a relief to copyright watchers everywhere when it concluded that a hyperlink to a publicly available work did not constitute a ‘communication to the public’ in a manner that exercised the copyright holder’s exclusive right in that work. (Let us not ask, at the moment, how copyright law could come so close to doing something so outlandish as requiring every hyperlink on the World Wide Web to be licensed.)
The Court of Justice for the EU (CJEU) arrived at that point through a curious analysis: a hyperlink to another server is communicating the work (notwithstanding the fact that a hyperlink is merely an instruction, and the actual content upload is done by the hosting party), but a link is not communicating to the public because the rights-holder has already done that, and the hyperlink merely makes the work available to the same “public taken into account by the copyright holders when they authorised the initial communication.” (para. 27).
Stated simply, the conclusion was: “since there is no new public [resulting from a hyperlink], the authorisation of the copyright holders is in any event not required for such a communication to the public.” (para. 30).
Generally speaking, in most cases this line of reasoning will produce unsurprising results. It is not entirely clear, however, what the decision means for improperly licensed content. If the “initial communication” to the public was not “authorised”, how should we treat third parties who unknowingly link to that unauthorized communication?
Consider, for example, how this new rule might have played out in the recent legal dispute between photographer Daniel Morel and Agence France-Presse (AFP). Morel photographed the immediate aftermath of the 2010 Haitian earthquake. His photos today remain some of the most distinctive images of the earthquake. How these photos came to be so widely recognized, however, was unfortunate: a foreign Twitter account identified as belonging to one Lisandro Suero swiped and reposted Morel’s photographs, claiming credit. The AFP and Getty Images, which had an arrangement with Twitter, redistributed Morel’s photographs to major newspapers around the world, crediting Suero. Getty and AFP went to trial arguing they had been deceived, but the evidence apparently led a jury to believe the companies had not acted in good faith; in November 2013, the jury awarded Morel $1.22 million USD. Ultimately, several other global news organizations also settled with Morel.
At the trial it became evident that months after the earthquake, improperly licensed versions of Morel’s photos had remained on more than a dozen sites. This one New York Times image, for example, still today credits Morel via AFP and Getty, when in fact neither had the right to distribute his images. At one time (and indeed – perhaps still today) that image likely infringed Morel’s rights under U.S. copyright law. Today, over 2700 indexed links on the World Wide Web point at it. Presumably, thousands of other sites linked to other images credited to Suero on CNN, TIME, and the Washington Post in the days following the Haitian earthquake, heedless of the fact that the image was unlicensed.
So, let us return to Svensson: One interpretation of the Svensson opinion could be this: since Morel did not authorize the publication of his picture on the N.Y. Times in 2010, then the 2700 sites that link to it today (including, potentially, this blog post) are also directly infringing Morel’s right to communicate to the public. It would be a bizarre result, however, that labels as infringers anyone who tweeted or linked to purportedly lawful content on NYTimes.com, CNN.com, TIME.com, or Washington Post pictures. A legal theory under which an individual linking to the most prominent news institutions in the world is directly liable for copyright infringement cannot possibly be right.
For this reason, the more logical interpretation of Svensson would be this: insofar as the N.Y. Times infringed Morel’s right to communicate to the public, Hypothetical Blogger John Smith who unknowingly links to the infringing image did not similarly infringe, since Blogger John’s hyperlink does not increase the audience beyond the audience to which the N.Y. Times already made the work available. This logic would apply equally whether Blogger John is linking to a large entity like AFP, Getty, or the N.Y. Times, or an individual, like Suero. Thus, as long as Blogger John’s set of potential recipients is the same as that of the N.Y. Times, and on the same terms, no new exercise of the right occurs, whether the underlying content was licensed or not.
Certainly, this is not the only interpretation of Svensson, but it would seem that any other interpretation would impose potential liability on thousands of sites who are entirely unaware of the fact that they infringe. Whether courts adopt this more logical interpretation, however, remains to be seen.