DONATE
>>Hyperlinking and Infringement: The CJEU Decides (sort of)

Hyperlinking and Infringement: The CJEU Decides (sort of)

Prof. Jane C. Ginsburg*, Columbia University School of Law
March 17, 2014

On February 13 of this year, the Court of Justice of the European Union rendered its decision in Nils Svensson, Sten Sjögren, Madelaine Sahlman, Pia Gadd v. Retreiver Sverige AB (Case C-466/12).  The case concerned a news-clipping service, Retriever Sverige AB, which linked to and framed articles appearing on the website of the Göteborgs-Posten newspaper.  For purposes of the CJEU decision, the articles were freely available on the newspaper’s site, although on remand to the Swedish courts the journalists may claim the articles did not appear without restriction.  Similarly, the parties (the journalists on the one hand and the online clipping service on the other) differed as to whether, if a client clicks on one of those links, it is or is not apparent to him that he has been redirected to another site in order to access the work.

The case was closely followed because the stakes on all sides seemed perilously high.  For Internet entrepreneurs, any claim that hyperlinking, even by “deep” or “framing” links, was a compensable communication to the public sparked concern that the practice of linking, fundamental to the Internet, could be ruled infringing, with deleterious consequences for the Web.  This concern led some to urge that the right of “making available to the public,” comprised within the right of communication to the public, was not triggered unless there was an actual transmission to the member of the public who requested the communication of a work.  For the authors and publishers, the narrow reading of the communication right urged by the defendant and its supporters, limiting the right to communications actually effected, rather than encompassing offers to communicate works of authorship, rendered authors’ rights essentially ineffective in the digital environment.  Were the making available right restricted to actual transmissions of the offered content, the enforcement of the right, for example against pirate websites, would be severely curtailed because proof of actual unauthorized transmissions may be difficult to obtain, even where the unauthorized offer is blatantly proclaimed.  By the same token, limiting the right to actual transmissions misconceives the right in its affirmative guise: The copyright owner or its licensee is offering access to the works to the public on an on-demand basis.  The license entitles the offeror to propose the transaction to the public; the license’s legal effect is not held in suspension until (or unless) a member of the public in fact takes up the offer.

The CJEU affirmed that the making available right extends to the offer to communicate, and not only to communications actually effected.  Per para. 19: “a work is made available to a public in such a way that the persons forming that public may access it, irrespective of whether they avail themselves of that opportunity” and para 20: “It follows that, in circumstances such as those in the case in the main proceedings, the provision of clickable links to protected works must be considered to be ‘making available’ and, therefore, an ‘act of communication.’”  It is therefore now clear that in the EU, “making available” covers potential communications as well as those that have been completed or that are in the course of transmission.

But in the recent caselaw of the CJEU, for there to be a “communication to the public,” an additional condition pertains.  It is not enough that a communication of the work be offered to the public; when the communication is an offer to retransmit content already made available for transmission, the offer must be made to a “new public.”  In this regard, it did not matter whether the content was offered by a deep link directly to the targeted content, or by framing that content.  (Para 29: “Such a finding cannot be called in question were the referring court to find, although this is not clear from the documents before the Court, that when Internet users click on the link at issue, the work appears in such a way as to give the impression that it is appearing on the site on which that link is found, whereas in fact that work comes from another site.”)

It is not, however, entirely clear who the “new public” is, or when the condition in fact applies.

Paragraph 24 of Svensson informs us that “the same works as those covered by the initial communication and made, as in the case of the initial communication, on the Internet, and therefore by the same technical means, must also be directed at a new public.”  Thus, there appear to be two elements, which, if both are present, mean that the “public” is not “new”:

(1)  The initial communication was made without restrictions as to possible recipients (or, if there was a restriction, there is a complete identity between the original intended recipients, and members of the public who may access the work by means of the link; the link must respect the same conditions of access).  In other words, there must be an identity of the intended potential recipients of the communication.

(2)  Identity of the technical means of the communication.  This requirement follows from para. 39 of the CJEU’s TV Catchup decision (Case C‑607/11, ITV Broadcasting and Others [2013] ECR, cited for this proposition in Svensson at para. 24), where the Court specified that the “new public” requirement did not apply if the means of communication were different from the original communication (in that case, the EU version of Aereo, the recipients were the same – persons entitled to view free-to-air television broadcasts – but the means changed from over-the-air to Internet transmissions).

The “new public” condition allowed the CJEU to reject liability in the Svensson case (thus saving the architecture of the Internet) while affirming the general scope of the making available right, most importantly, that it covers offering works and not just delivering them.  In effect, as applied in Svensson, the condition operates rather like an implied license: If the author made the work freely available on the website, then it follows that the author consents to third-party links to that website.

If the “new public” condition should be conceived as a kind of implied license, the condition should apply differently if the content on the source website, albeit made available to an unrestricted public, were infringing, i.e., should not have been made available to any public in the first place.  In those instances, there will have been no license, implied or otherwise.  In Svensson, the national court on remand will have to determine whether the journalists’ articles were on the source websites with the journalists’ authorization (this appears to be disputed; in one case, the source website may not have been authorized, but in the others, the content was available behind a paywall).  If the journalists had already invited the general public to view their articles without restriction, the “new public” criterion would not be met.  But if the source website was not authorized to make the content available, or if restrictions did apply, it cannot be said that the authors invited anyone to access their works from that site.  Thus, before we get to whether the linker is making the work available to a “new” public, we should first ask whether the author had authorized any public to access the work from that site.  If, as to the source website, no public was authorized, then it should not matter whether the linker was making available to a “new” public.

One might derive support for this approach from para. 31 of the CJEU judgment, which distinguishes source sites that make the work available on a restricted basis, where the linker makes the work available to a wider public.  If making the work available to a public that was not supposed to be able to access the work would render the linker liable for making available to a new public, then perhaps collaborating (by providing links) with a source that was not supposed to be making the work available to any public could be analyzed as a kind of joint and several liability for making available.

Similarly, para. 24’s reference to “a public that was not taken into account by the copyright holders when they authorised the initial communication to the public” implies that the copyright owner has already been remunerated (or had the opportunity to be remunerated, whether or not it availed itself) for an initial communication.  In addition, one should further infer, the subsequent communication does not create a new market for the work, hence the court’s non-application of the “new public” requirement when the “technical means” of the communication are different.  Subjecting the second communication to the copyright holder’s authorization would create an unseemly situation of “double dipping.”  But if the copyright owner hasn’t been remunerated because the initial communication was infringing, then the second communication cannot result in “double dipping.”

While it may be possible to restate the “new public” criterion as a variant, perhaps a more elegant one, of the implied license mechanism to restrain copyright overreaching, it remains controversial on several levels.  For example, in the case of aggregators of links to lawful websites, the link aggregator may not be providing access to a public that would not otherwise have had access, but as a practical matter is increasing access for those members of the public who may otherwise have had difficulty finding the source websites.  When a link aggregator renders a site available to thousands of viewers where the source site attracted only dozens, is it really convincing to contend that the aggregators’ viewers are not a “new public”?

Moreover, one should inquire whether a “new public” requirement (whatever it means) is compatible with WIPO Copyright Treaty Article 8 and Berne Convention Articles 11 and 11bis, which bring at least some retransmissions by third parties within the scope of the right of communication to the public.  Arguably, moreover, the requirement creates a kind of “exhaustion” doctrine for the right of communication to the public.  In addition, if the author adds a notice that she does not wish commercial sites to link to hers, does such a notice constitute the necessary restriction to rule that a commercial public would be “new”?  But if the author’s notice is effective, is it also a formality inconsistent with international norms?  A future column will explore those and other questions raising the meaning and scope of “new public.”


By |2018-07-03T17:50:26+00:00March 17th, 2014|Intellectual Property Issues|Comments Off on Hyperlinking and Infringement: The CJEU Decides (sort of)