Post-Svensson Stress Disorder #1: Does it matter whether linked content is lawful?

Last week's decision of the Court of Justice of the European Union (CJEU) in Case C-466/12 Svensson was not just keenly awaited but has already been keenly discussed on this blog [here and here, and to some extent here, and above all see several readers' comments] and elsewhere [herehereherehereherehere, here, here...].

The CJEU held that the owner of a website may, without the authorisation of the copyright holders, redirect internet users, via hyperlinks, to protected works available on a freely accessible basis on another site. This is so even if the internet users who click on the link have the impression that the work is appearing on the site that contains the link.

Despite its apparent clarity, the neat 42-paragraph judgment has already raised several doubts as to the practical implications that the principles laid down by the CJEU might and will have.

This is why the IPKat has decided to run a short Post-Svensson Stress Disorder series to address in more detail some of the issues that every European's (no matter whether from the EU or outside) favourite Court made not so clear in this decision.

Let's start today with a sudden question that has arisen in the minds and hearts of these two Kats:

Does it make a difference whether the content one hyperlinks to has been placed on a certain website without the authorisation of the relevant copyright holder?

Put otherwise: Can a hyperlink to content that has been placed on a certain website without the authorisation of the relevant rightholder be considered tantamount to an (unauthorised) act of communication to the public?

The first time these Kats read the decision’s conclusions, they were under the impression that that the answer of the CJEU was in the negative.

Is it OK to hyperlink
to unauthorised content?
By stating that “the provision on a website of clickable links to works freely available on another website does not constitute an ‘act of communication to the public’”, it could appear that “the owner of a website may, without the authorisation of the copyright holders, redirect internet users, via hyperlinks, to protected works available on … another site”, and do so irrespective of whether the copyright holder’s had authorised that “site” to host his or her work(s).

However, it would seem that this conclusion may not be correct. 

As all copyright enthusiasts already know by heart, the CJEU solved the issue of whether a hyperlink may amount to communication to the public by focusing on the notion of “new public”.

Having observed that “the provision of clickable links to protected works must be considered to be ‘making available’ and, therefore, an ‘act of communication’” and that the “indeterminate and fairly large number of recipients” who could potentially use the link can well be considered a “public” within the meaning of Article 3(1) of Directive 2001/29, the CJEU excluded that links to freely available works (news articles in this case) could be considered as falling within the notion of communication to the public. This is because:

[I]n order to be covered by the concept of ‘communication to the public’, within the meaning of Article 3(1) of Directive 2001/29, a communication … must also be directed at a new public, that is to say, at a public that was not taken into account by the copyright holders when they authorised the initial communication to the public

The notion of “new public” appears as a subjective criterion, rather than an objective requirement.

A “new public” is the “public that was not taken into account by the copyright holders when they authorised the initial communication to the public”.

Thus, what becomes really crucial is the following:

What public does the copyright holder take into account when he or she authorises the initial communication of his or her work to the public on the internet?

Earlier CJEU case law may provide some guidance in this respect.

In Case C-306/05 SGAE, the CJEU made reference to the WIPO’s Guide to the Berne Convention stating that, when the holder of copyright in a musical work “authorises the broadcast of his work, he considers only direct users, that is, the owners of reception equipment who, either personally or within their own private or family circles, receive the programme”. Accordingly, the CJEU concluded that use of television sets for playing music in a hotel goes beyond such direct users’ circle.

In later decisions the CJEU applied the same principle to TV broadcasts. In Case C-136/09 Organismos Sillogikis Diacheirisis Dimiourgon Theatrikon kai Optikoakoustikon Ergon, and Joined Cases C403/08 and C429/08 FAPL the CJEU found that, when authorising the initial communication to the public of their works, copyright holders would not consider those watching TV in hotels and pubs as the direct (initial) public. Rather the “direct public” in such rightholders’ minds would include “only the owners of television sets who, either personally or within their own private or family circles, receive the signal and follow the broadcasts”.
What kind of public does he have in mind?

In Joined Cases C431/09 and C-432/09 Airfield NV Canal Digitaal BV, the CJEU considered the case of satellite broadcastings, and stated that, when the initial communication to the public is not carried out by the copyright holder, the “new public” is the public which was not taken into account by the authors of the protected works within the framework of an authorisation given to another person”.

Svensson is the first decision in which the CJEU took into consideration the case of a copyright holder that authorised the initial communication of his or her work(s) on the internet.

Comparatively, it would seem that the Svensson copyright holder is much more “open-minded” than the copyright holders in earlier case law. Whilst the latter appeared to only consider personal or family users, the copyright holder that authorised the initial communication on the internet seems to know that his or her direct public is not so narrow, but is rather composed by “all Internet users [who] could have free access to them” (provided, of course, that the content is freely available).

Open-minded or not, in assessing the existence of a “new public” (and, as result, the occurrence of an act of communication to the public), both Svensson and earlier CJEU case appear to agree that it is the copyright holder’s idea of the public to which the work is made available at the time of the first communication that matters.

If the copyright holder does not perform or authorise the initial communication, he or she would logically not take into account any public.

Accordingly, if works have been initially made available on the internet without the consent of the copyright holder, any subsequent act of communication of the infringing work – including hyperlinking to it – makes the work available to a new public.

Thus - and this a conclusion that these Kats find less sexy than the apparent exciting simplicity of the response of the CJEU in Svensson – consent of the copyright holder, or lack thereof, is material in order to assess whether a certain hyperlink amounts to an act of communication to the public.

Consequently, there appear to be two cumulative conditions for linking not to be considered an act of communication to the public:

(i)         The work has been initially communicated by the copyright holder or with his or her consent;

(ii)        The work is freely available on a certain website, this meaning that it is accessible potentially to all internet users without restrictions.

Tolomeo begins to think that 
Svensson might be 
much ado about nothing
As regards the first condition, does this entail a burden on the subject who links to a certain work to make sure that the content has been made initially available by the copyright holder or with his/her consent? If this interpretation of Svensson was correct, then we could say that the devil is not just in the detail but also in one’s link. How could one determine that a certain work has been made initially available with the relevant rightholder’s consent?

In addition, there might be instances when the “new public” requirement may be difficult to apply in practice, especially in the “subjective” terms embraced by the CJEU. Let’s take, for instance, the case of a work (eg a tale published on its author’s blog) that has been initially made freely available by the relevant rightholder on a certain website and has been illicitly reproduced by a third party on his/her website. If you hyperlink to the illicitly reproduced content are you infringing the relevant rightholder’s copyright? You might argue that there is no “new public” in this case because the relevant rightholder made the work initially available on a certain website for anybody to enjoy. However, there would be “new public” if you considered this requirement in objective terms as referring to any public that might access authorised copies of one’s works, not also unauthorised reproductions.

Having said all this, there seems that two further questions may arise:

(1)  Did we need a CJEU’s decision to know that one can lawfully link to freely available content made available by the copyright holder or with his/her consent?

(2)  What does “freely available” mean? Is it just a matter of technical availability or do also possible contractual restrictions (or lack thereof) play a role?

While the latter is going to be central to the next Post-Svensson Stress Disorder episode, one may wonder whether, in light of the first question, Svensson may not be so seminal after all … Or are we missing something? What do the readers think?
Post-Svensson Stress Disorder #1: Does it matter whether linked content is lawful? Post-Svensson Stress Disorder #1: Does it matter whether linked content is lawful? Reviewed by Unknown on Friday, February 21, 2014 Rating: 5


  1. What! Three days and no comments?
    I had better start the ball rolling. First an observation. I don't think it is helpful to compare Svensson to the earlier decisions about TV programmes (Airfiled) etc, since we are not dealing with the same thing. The TV cases were about broadcasting and the viewing of a broadcast (and not recording the broadcast by the consumer) and therefore the InfoSoc Directive is barely engaged by such cases.
    In contrast, Svensson was entirely about Article 3 of the Infosoc Directive. And of course it is worth remembering that Svensson was a preliminary ruling on the narrow aspect of how Article 3(1) should be interpreted.
    So the the first question posed by Alberto and Eleonora. Yes of course it matters if the links lead to content which has not been authorised by the rights owner. Indeed part of the Svensson ruling touches on this at [31] where the CJEU deals with a case where a hyperlink circumvents a technological measure (like a paywall) on the target site. The court says that this would be communication to a different public from the public envisaged by the host site (ie the subscribers) and thus the link would be infringing. From there it e easy to show that if the rights owner had not envisaged the public which the site hosting the unauthorised content was making their content available to, then again infringement would occur if another site linked to the unauthorised content. The fact that the (unauthorised) content was freely available on this site would have no bearing on the fact that because the rights owner had not licensed the infringing site (ie there was primary infringement by this hosting) by definition the owner had not envisaged that particular public, that is, the wider audience than a site which the owner had authorised.
    This of course has some resonance with the Meltwater case (although not the aspect which is currently being referred to the CJEU by the UK Supreme Court). In Meltwater I, and to a lesser extent Meltwater II, the Claimants (the NLA and various newspaper titles) established that the terms and conditions found on the newspapers' websites effectively created a specific public who can freely access their content (namely private individuals) and another public (commercial bodies, such as Meltwater and its clients) who were subject to NLA licensing before they could view the contents.
    It is interesting that the Svensson case is also about a similar activity involving the Götebborgs-Posten and Retriever Sverige, yet as far as we are told the Götebborgs-Posten website did not seem to employ similar terms and conditions, otherwise I suspect the CJEU ruling might have been worded differently, so as to answer the questions posed by Sweden, but making it clear that the facts in Svensson meant that it fell outside their ruling.

  2. Interesting comment Andy J.

    For some vague reason (which will come to me after my second cup of coffee this morning) I am reminded of the recent US Kirtsang case.

    It appears that the 'specific audience' argument has a direct parallel with attempted (and failed) market segmentation argument there.

    Of course, the analogy is not perfect, but the case may offer some illumination.

  3. As we talking about linking, in case other readers have missed it here is a link to an interesting post on this subject from Graham Smith. Not only does he address the Alberto/Eleonora First Question, but I suspect he may also have addressed some if not all of their subsequent questions: Cyberlegal blog.

  4. Dear Andy,

    thank you for your comments!

    In my view, the earlier CJEU's decisions upon the issue of new public are worth being considered. All of them address the very question of whether a certain act has to be considered communication to the public (and, namely, to a new public) in light of Article 3(1) Infosoc, and so does Svensson.The way in which CJEU considers the right holder's thoughts at the moment of initial communication is intriguing and intriguing is also that, in the CJEU's view, the right holder thinks different depending upon the media involved in the initial communication.

    We the IP people are rather used to put ourselves in strange characters' shoes such as average consumers, european circles' members, informed users, persons skilled in different kinds of arts. Apparently we have a new one here.

    Thank you again!


  5. It is precisely that 'new one' that the Kirtsang decision touches upon.

    There, the existence of the secondary market - clearly a different and thus "new" 'public' from which the publishers envisioned when they released their materials into the 'public' of one specific country - failed to be allowed to distinguish the act of the rights holder, and the material was deemed to have exhausted its rights.

    Clearly, the notion of 'new public' alone cannot be determinative. At least not in how the US has ruled on this related topic.

  6. I respectfully disagree with the Kats' logic on the application of the ECJ's decision to the legality of linking to illegally published works. To be clear, the decision does not relate to such a situation and any close-reading exercises and extrapolations are therefore inherently counterfactual. We don't know how the Court would rule in such a case.

    Taking into account the "public that was taken into account by the copyright holders when they authorised the initial communication to the public" makes sense as a crition for establishing a (separate) communication to the public, in that it protects against additional economic or moral impact. However, this only works where such a public was actually taken into account. Where the right holders did not authorise the initial communication, they had no intent whatsoever and therefore it is not logical or useful to ask whether the hyperlink is directed at a "new" public.

    "New public" is indeed a subjective criterion. In the context of linking to illegally communicated works, moreover, it can only be made to make sense, if it is applied to the infringing communicator. In other words, a hyperlink that reaches a new public that the original infringing communication did not (intend to) reach is a new (and equally infringing) communication. So, for example, if the original publication is not freely available to all internet users but password-protected, and the hyperlink contains the password, then the hyperlink may be an infringing communication to a new public.

    This approach is consistent with the purpose of the "new public" criterion, i.e. to protect right holders against additional economic or moral impact. For the public (intended to be) reached by the initial publisher, the right holders can file suit against that publisher; for the additional public (intended to be) reached by the hyperlink, the right holders van address the hyperlinker.

    For a hyperlinker to establish the intentions of the rightholder in regard to a publication on a third site will often be impossible. How is the hyperlinker expected deal with situations where the publishing site has a license to do so, but is somehow exceeding the terms of that license (e.g. publishing too much, or for too long)? Or with the situation where the right holders have not authorised publication, but the publishing site can rely on a copyright exception (e.g. for quotation or news purposes) - would the hyperlinker need to meet all the criteria for the exception too? This is not practically applicable and would lead to unintended and/or undesirable consequences.

    Again, the Svensson decision doesn't address the question of whether the copyright status of a hyperlink (communication to the public or not) depends on the copyright status of the souce (infringing or not). It doesn't definitively support either view. On balance, though, the Kats initial (negative) answer makes more sense.

  7. I am disappointed that no one seems to grasp that the US Kirtsang decision has a direct impact on the notion of selected public (one public being somehow different than another).

    Clearly the copyright holders in the Kirtsang case aimed their foreign editions at a specific and distinct public than the second public to whom Kirtsang sold his legitimately bought copy.

    The copyright holders argument of worldwide segmentation - which is nothing more than a specific example of the multiple-public doctrine - was explicitly rejected by the US Court.


All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

Learn more here:

Powered by Blogger.