From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Sunday, 7 December 2014

What happened to Svensson and his friends after the CJEU decision?

Jan Rosén
Over the past few months a good part of this Kat's thoughts has been devoted to hyperlinking and copyright, courtesy of the reference that the Svea Court of Appeal (Sweden) made to the Court of Justice of the European Union (CJEU) in Svensson [here and here; the Court confirmed its value as a precedent a few weeks ago in its quick order in BestWater, a case concerning embedded videos].

As IPKat readers will remember, also the European Copyright Society [hereand the Association Littéraire et Artistique Internationale (ALAI) [here] had their say on this reference, and produced two fairly diverging contributions while the case was still pending. Readers will also remember that the latter recently issued another Opinion, which builds upon its earlier statement and comments critically on the CJEU decision.

Among all the questions raised by Svensson, this Kat could not help but wonder what happened to the national proceedings following the decision of the CJEU. So she thought of asking Swedish Katfriend and Chair of the Study Group that proposed the latest ALAI Opinion, Jan Rosén (Stockholm University), to help.

Here's what Jan writes:

The Svensson case: CJEU collides with international copyright law - still, Svensson et al. better off than suspected!

The CJEU’s milestone decision in Svensson, about the question of whether hyperlinking to subject-matter which is protected by copyright requires the permission of the rightholder, offers a ruling elucidating some important issues on contemporary communication and making available to the public on the internet. In doing so, the CJEU certainly gave a correct and positive response to the basic question of whether hyperlinking is (interactive) making available/communication.

The Court stressed that the concept of communication to the public includes two cumulative criteria - an ‘act of communication’ of a work and the communication of that work to a ‘public’ - and that the existence of an ‘act of communication’ must be construed broadly (referring to FAPL, [193]), in order to ensure, in accordance with recitals 4 and 9 in the preamble to Directive 2001/29, a high level of protection for copyright holders (at [16], quoting TVCatchup, [21] and [31]).

A cuter type of linking
 than the one addressed by the CJEU
At [18] in Svensson the court laconically noted that, under the circumstances of the case, the provision on a website of clickable links to protected works, published without any access restrictions on another site, affords users of the first site direct access to those works.

As later confirmed by the CJEU  in BestWater, the technical means by which a linking measure is accomplished are as a matter of principle irrelevant, thereby supporting a somewhat technology neutral approach to communication to the public, be it standard in-line links, framing or embedded measures.

Further (at [19]) the CJEU stated that it “is apparent from Article 3(1) of Directive 2001/29, for there to be an ‘act of communication’, [that] it is sufficient, in particular, that 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 (see, by analogy, Case C‑306/05 SGAE [2006] ECR I‑11519, paragraph 43).

Hereby, the CJEU confirmed that the enabling effect of a measure to potentially address a public is sufficient for it to trigger use of the basic right of communication to the public; posting a work on a website, just as rigging a link to a work thereby accessible by the public, falls within the scope of the exclusive right of communication to the public. It is irrelevant if anyone actually avails himself of this opportunity. No actual “transfer” of the work to the public is needed in order to make a communication to the public.

In this respect the Svensson decision is completely in line with the Opinion on hypertext linking, agreed by the Executive Committee of the ALAI - representing copyright experts of some 30 national groups - on 16 September 2013.

Problems arise when the CJEU in Svensson introduced, as to the question of whether an act concerns a communication to "the public", a rigid “new public” criterion, qualified by, inter alia, an equally problematic new “specific technical means” criterion. 

The “new public” criterion was somewhat developed in CJEU earlier case law (see SGAE; Del Corso; TVCatchup), but became somehow of a more strict nature in Svensson and BestWater.

Is this the public the relevant
rightholder had in mind?
As applied in Svensson, the “new public” criterion has the effect of inappropriate exhaustion of the exclusive right of communication to the public of works which their authors or other rightowners have made available over generally accessible websites. This criterion is obviously inconsistent with the right of communication to the public within the Berne Convention and the WIPO Copyright Treaty, as well as with the provisions of Directive 2001/29.

Moreover, to the extent Svensson indicates that a the “new public” criterion will not apply if restrictions accompany the making available of a work, the decision risks establishing an obligation to reserve rights or protect works etc. by technical protection measures, in violation of the Berne Convention’s prohibition of formalities that condition the exercise of exclusive rights.

In plain English: if authors do not prevent free access to works posted on a website, they no longer have the right to prevent others to communicate it to the public via links pointing at those freely accessible works.

At [23] the CJEU stated that the manager of linking measures is factually making a communication to a public under the circumstances of the Svensson case. But the links did not address [24] “a public that was not taken into account by the copyright holders when they authorised the initial communication to the public” (at 24) - “all Internet users could therefore have free access to them” [25] - and were accomplished “by the same technical means” as in the case of the initial communication. Accordingly, those measures did not lead to the works in question being communicated to a new public.

The CJEU laconically concluded [28] that “since there is no new public, the authorisation of the copyright holders is not required for a communication to the public such as that in the main proceedings.”

However, the reference to the circumstances at hand in the main proceedings was not adequate.

The CJEU must have understood those to mean that works of Svensson et al. actually were freely accessible for that period under which the linking measures in question were active.

Kat-exclusive image
of the settlement negotiations
In fact, this was not the case!

All works posted on the source website,, run by the morning paper GP (Göteborgs Posten), were openly accessible merely for a period of three weeks from the day of first upload on the source website. After such a period a personal code was needed to access the works on the source website, which is a possibility normally offered to subscribers of the paper edition of GP. Hence, Retriever’s links to works at the website normally passed through a wall to a restricted area.

Retriever and Svensson et al. settled the Swedish case out of court. As the linking measures in question offered access to works on the website that was subject to restrictive measures, thus afforded a communication to a new public, a consent from the rightsowners had been necessary. This must have formed the basis for the settlement (which was and is secret), certainly leaving Svensson et al. with a very attractive end of the story.

Svensson et al. may have been, on their part, content with the outcome of the case. Still, the introduction by the CJEU of the “new public” criterion is in conflict with international copyright law, even if the “new public” criterion has gone through two corrections by the CJEU itself; there is a new public if communication is made by different specific technical means and if access to the communication through the internet is restricted.

As the ALAI stated in its latest Opinion:

The ---  application of the "new public" criterion in the Svensson decision is contrary to
·   Articles 11(1)(ii), 11bis(1), 11ter(1)(ii), 14(1) and 14bis(1) of the Berne Convention
·   Article 8 of the WCT
·   Articles 2, 10, 14 and 15 of the WPPT
·   Article 3 of the EU Information Society Directive
·   previous CJEU decisions and
·   interpretation rules of  Articles 31 and 32 of the Vienna Convention on the Law of Treaties.

The Svensson decision is also based on an misinterpretation of the old (1978) Guide to the Berne Convention.

It may lead to a WTO dispute settlement procedure and liability under the TRIPS Agreement for its inconsistency with the Berne Convention.


Anonymous said...

In my view the CJEU's approach leaves a lot to be desired, unnecessarily complicates copyright law and may well run into a dead end in future cases, but even the most fervent pro-copyright scholar should realise that one cannot expect a court to rule that posting a link as a rule infringes the copyright on the work linked to. No court is going to outlaw the Internet. Judges have common sense and ALAI better gets some, too.

Mark said...

I think the Court of Justice's approach isn't all that bad, and they have every right to interpret the Directive and Convention in this way. The idea that someone who has made their work available to the general public by placing it on the internet should then be able to claim infringement when someone simply post a link to that work is absurd. The copyright maximalists of the ALAI may dream of a world where the internet is illegal, but most of us are quite happy with it, thanks.

Francis Davey said...

Worse, the idea that a link to a work which is protected with an access code is a "communication to the public" is nonsense, surely? If the public can't access it, it isn't communicated to them.

The logic of the ALAI opinion applies equally well to, say, references in an academic paper or in fact any explanation of how to obtain a copyright work lawfully from its owner.

There can be essentially no principled reason why the copyright owner needs the incentive of being able to control such a mention of its work. I find it extremely hard to enter the mindset of someone who thinks otherwise.

Anonymous said...

I agree with Francis and would like to see a distinction between URL links which are a bit like book references stating "here you can find this information" from embedding/framing content which shows the material in a new context on another web page. Embedding deprives the holder of the control of the context and since the URL is hidden the origin of the material is also hidden.
In the urge to be technically neutral this important non-technical different is lost.

Anonymous said...


I have to ask: would it be parsing too finely to draw a distinction between general communication - which implicitly involves a back and forth capability - and communicating to someone - which does not implicitly involve 'the back' response?

If the distinction is allowable, then the ability to respond back is immaterial, and the fact that what is communicated is so communicated within a 'locked box' does not matter , as the item is still communicated to the public.

Francis Davey said...

I don't think that can work. What we are talking about here is the "making available" part of "communication to the public" which has the well known definition:

"the making available to the public of the work by electronic transmission in such a way that members of the public may access it from a place and at a time individually chosen by them"

If the member of the public can't access the work (because it is in a locked box) then it is not communicated to *them*.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':