Guest post by prof. Dirk Voorhoof, Ghent University.*
In its decision of 9 March 2017 in Rolf Anders Daniel Pihl v. Sweden, the ECtHR has clarified the limited liability of operators of websites or online platforms containing defamatory user-generated content. The Court’s decision is also to be situated in the current discussion on how to prevent or react on “fake news”, and the policy to involve online platforms in terms of liability for posting such messages. Although the Court’s ruling expresses concerns about imposing liability on internet intermediaries that would amount to requiring excessive and impractical forethought capable of undermining the right to impart information via internet, the decision in Pihl v. Sweden itself guarantees only minimal protection for the rights of internet intermediaries and users’ rights.
The decision in Pihl v. Sweden builds on the judgments in Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary (see our blog) and the Grand Chamber’s judgment in Delfi AS v. Estonia (see our blog). This time however the applicant was not the owner or operator of an online platform with users’ comments complaining about an interference with its right to freedom of expression under Article 10 of the Convention. In Pihl v. Sweden the applicant, a private person, complained of a breach of his right to privacy and reputation under Article 8 ECHR, because the Swedish authorities had refused to hold the operator of a website liable for a defamatory blog post and an anonymous online comment.
A blog post on a Swedish website was published accusing Mr. Pihl of being involved in a Nazi party. The blog on which the post appeared was a small one run by a non-profit association. Although the blog allowed comments to be posted, it was clearly stated by the operating association that such comments were not checked before publication and that commentators were responsible for their own statements. Commentators were therefore requested to “display good manners and obey the law”. The day after publication of the post, an anonymous person posted a comment stating that “that guy Pihl is also a real hash-junkie according to several people I have spoken to”.
Nine days later Pihl posted a comment on the blog in reply to the above comment and blog post about him, stating that the information in the blog post and comment was wrong and should immediately be removed. The following day the blog post and the comment were removed and a new post was added on the blog by the association – stating that the earlier post had been wrong and based on inaccurate information – and it apologised for the mistake. However, Pihl sued the association and claimed symbolic damages of 1 Swedish krona (SEK), approximately 0.10 euro (EUR). He submitted that the post and the comment constituted defamation, and that the association was responsible for the fact that both the blog and the comment remained on the website for nine days.
The District Court rejected Pihl’s claim. It found that the comment constituted defamation, but it found no legal grounds on which to hold the association responsible for failing to remove the blog post and comment sooner than it had done. The Court of appeal confirmed this judgment and the Supreme Court refused Pihl’s leave to appeal. A short time later, Pihl lodged an application with the Chancellor of Justice for payment of damages on the basis that the Swedish State had failed in its positive obligations under Article 8 of the Convention through the national courts’ decision not to hold the association responsible for the defamatory comment against him. In 2015 the Chancellor of Justice rejected the application as he found that it could not be deduced from the European Court’s case-law – Delfi AS v. Estonia was pending before the Grand Chamber at that time – that there was an absolute obligation on States to have legislation in place, in each individual case, enabling the person responsible for a blog and the comments on it, to be held accountable.
It is worth mentioning that defamation in Sweden is still a criminal offence under Chapter 5, Section 1 of the Penal Code. However there is also specific legislation on liability and obligations for removal upon notice of certain content on online platforms. Section 5 of the Act on Responsibility for Electronic Bulletin Boards, concerning the obligation to erase certain messages, states:
“If a user submits a message to an electronic bulletin board, the supplier of the service must remove the message from the service, or in some other way prevent its further dissemination, if
- the message content is obviously such as is referred to in the Penal Code, Chapter 16, Section 5, about inciting rebellion, Chapter 16, Section 8, about agitation against a national ethnic group, Chapter 16, Section 10a, about child pornography crime, or Chapter 16, Section 10b, about unlawful depiction of violence, or
- it is obvious that the user has, by submitting the message, infringed the copyright or other right protected by Section 5 of the Copyright (Artistic and Literary Works) Act (1960:729).
As the duty to erase some types of obvious illegal content, such as hate speech, child pornography, incitement to violence and copyright infringements, did not include defamation or breach of privacy, Section 5 of the Act on Responsibility for Electronic Bulletin Boards was not applicable in the case at issue, while the association, or its legal representative could not be convicted of defamation, either as the principal or as an accomplice, according to the Penal Code or Section 5 of the Act.
Pihl complained under Article 8 of the Convention that the fact that Swedish legislation prevented him from holding the association responsible for the defamatory comment had violated his right to respect for his private life.
The Court’s reasoning and decision
First the Court reiterates that a person’s right to protection of his or her reputation is encompassed by Article 8 as part of the right to respect for private life, while in order for Article 8 to come into play, the attack on personal honour and reputation must attain a certain level of seriousness and must have been carried out in a manner causing prejudice to personal enjoyment of the right to respect for private life. The Court considers that the comment, although offensive, certainly did not amount to hate speech or incitement to violence, but it accepts the national courts’ finding that the comments at issue constituted defamation and, consequently, fell within the scope of Article 8.
Next, the Court refers to its Grand Chamber judgment in Delfi AS v. Estonia in which it explained how to balance the conflicting rights protected under Article 8 and 10, including its approach that “where the balancing exercise between those two rights has been undertaken by the national authorities in conformity with the criteria laid down in the Court’s case-law, the Court would require strong reasons to substitute its view for that of the domestic courts” (see also Axel Springer AG v. Germany and Von Hannover (no. 2) v. Germany). Referring to Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary and protagonists playing an intermediary role on internet, the Court sums up a set of specific aspects that are relevant for the concrete assessment of the interference in question: “the context of the comments, the measures applied by the company in order to prevent or remove defamatory comments, the liability of the actual authors of the comments as an alternative to the intermediary’s liability, and the consequences of the domestic proceedings for the company” (§ 28).
Analysing these factors step by step, the ECtHR scrutinises whether the Swedish judicial authorities achieved a fair balance between Pihl’s right to respect for his private life under Article 8 and the association’s right to freedom of expression guaranteed by Article 10 of the Convention.
As regards the context of the comment, the Court notes that the underlying blog post accused Pihl, incorrectly, of being involved in a Nazi party. However, the post was removed and an apology published when the applicant notified the association of the inaccuracy of the post. The comment about Phil being a “real hash-junkie” did not concern his political views and had nothing to do with the content of the blog post. It could therefore hardly have been anticipated by the association. The Court attaches particular importance to the fact that the association is a small non-profit association, unknown to the wider public, and it was thus unlikely that it would attract a large number of comments or that the comment about Pihl would be widely read. The ECtHR also considers that “expecting the association to assume that some unfiltered comments might be in breach of the law would amount to requiring excessive and impractical forethought capable of undermining the right to impart information via internet” (§ 31). As regards the measures taken by the association to prevent or remove defamatory comments, the Court notes that it was clearly stated on the blog that the association did not check such comments before they were published and that commentators were responsible for their own statements. Commentators were also requested to display good manners and obey the law. Moreover, the Court observes that the association removed the blog post and the comment one day after being notified by Pihl that the post was incorrect and that he wanted the post and the comment removed. The association furthermore posted a new blog post with an explanation for the error and an apology. The Court also refers to its earlier case law in which it held that “liability for third-party comments may have negative consequences on the comment-related environment of an internet portal and thus a chilling effect on freedom of expression via internet. This effect could be particularly detrimental for a non-commercial website”. Turning to the liability of the originator of the comment, the Court observes that Pihl obtained the IP-address of the computer used to submit the comment. However, he has not stated that he took any further measures to try to obtain the identity of the author of the comment. Lastly the Court notes that Pihl’s case was considered on its merits by two judicial instances at the domestic level before the Supreme Court refused leave to appeal. Moreover, the Chancellor of Justice examined Pihl’s complaint under Article 8 of the Convention, referring to the Court’s case-law and the need to balance the interests under Article 8 and Article 10, before finding that the case did not disclose a violation of his rights under Article 8. The Court further observes that the scope of responsibility of those running blogs is regulated by domestic law and that, had the comment been of a different and more severe nature, the association could have been found responsible for not removing it sooner.
In its overall conclusion the ECtHR emphasises the fact that the comment, although offensive, did not amount to hate speech or incitement to violence and was posted on a small blog run by a non-profit association which took it down the day after the applicant’s request and nine days after it had been posted. In view of this, the Court finds that the domestic courts acted within their margin of appreciation and struck a fair balance between the applicant’s rights under Article 8 and the association’s opposing right to freedom of expression under Article 10. Therefore the Court finds the application manifestly ill-founded.
In Pihl v. Sweden, in contrast with Delfi AS v. Estonia and Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary, the ECtHR does not seem to require, at least not from small non-profit associations operating a website or online platform open for users’ comments, the pre-monitoring of all content, nor to have an effective notice-and-take-down system installed. Indeed in the case at issue it was clearly stated by the operating association that the content of users’ comments was not checked before publication and that commentators were responsible for their own statements, while only being requested to “display good manners and obey the law”. There are no indications that a procedure for notice-and-take-down was installed, although from the facts it can be derived that upon notice, in casu the platform reacted promptly to remove the incorrect, offensive and defamatory messages, even accompanied with an apology by the association. This reaction in itself was enough to exonerate the association from liability, in the context of the case at issue. The Court’s decision in the case of Pihl v. Sweden leaves open the question whether such a removal, eventually accompanied with an apology, is necessary to exonerate the operators from liability, or whether in other situations a rectification, right of reply or other way of correcting the “false” allegations might be a more appropriate, sufficient and proportionate way in respect of the internet intermediaries’ and their users’ right to freedom of expression and information.
A crucial, if not decisive element in the Court’s ruling, is derived from the content itself of the contested online messages, which according to the ECtHR “did not amount to hate speech or incitement to violence”. This approach by the Court, in line with Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary, confirms the need to make a distinction in the levels of liability for internet intermediaries. Indeed online internet platforms should only be held liable when they have failed to act expeditiously when it concerns “clearly unlawful content”, and more precisely when illegal hate speech and incitement to violence has been posted on their website or platform. This means that in cases of breach of privacy, libel or defamation internet intermediaries are not to be held liable for users’ comments, when upon notice the messages at issue have been promptly removed. The decision in Pihl v. Sweden makes this exoneration categorical, even in cases of clear defamatory comments without any public interest, while in Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary the Court still referred to the specific characteristics of the comments at issue, holding that they were related to a matter of public interest, using a style that was common in communication on many Internet portals and only concerned the commercial reputational interests of a company.
The consequence of the decision in Pihl v. Sweden remains however the same as in the Court’s previous judgments in Delfi AS v. Estonia and Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary: an expeditious removal-upon-notice of comments with “clearly unlawful content”, such as illegal hate speech and incitement to violence, will not exonerate internet intermediaries from liability, as the exoneration is only valid in cases where it does not concern such content. This means that operators of such websites can only fully protect themselves against criminal liability for incitement to hatred, discrimination or violence, by installing a system of pre-monitoring of all users’ comments and by filtering or removing clearly unlawful content on their own initiative.
The most important characteristic of the decision in Pihl v. Sweden however is that it seems to reserve the limited liability for defamatory users’ comments only for small non-profit websites. While in its judgment of 2 February 2016 in Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary the ECtHR did not connect decisive consequences to the different characteristics of the online platforms at issue (Index.hu Zrt being run by a commercial company and being one of the major Internet news portals in Hungary, while MTE is a non-commercial website of a self-regulatory body of Internet content providers), in Pihl v. Sweden the ECtHR emphasises the small and non-profit character of the association at issue as a crucial aspect in limiting or even conditionally excluding its liability. By narrowing the exoneration to only small not-profit operators of online platforms, the Court’s case law leaves a broad opening for the member states to impose liability on all other online platforms with user generated content, including defamatory content, even in cases of expeditious removal-upon-notice by the operators.
In an earlier blog we have expressed concern about pushing internet intermediaries further in the direction of private censorship, and that the burden on private actors to pre-monitor user generated content and eventually remove some of it, with a lack of clear criteria, a lack of transparency and no effective procedural guarantees creates a clear and present danger for the right to freedom of expression on the Internet. This concern is amplified in the context of current policies imposing more liability on internet intermediaries for content that can be considered as “fake news”, propaganda and hate speech. Similar concerns have recently been uttered in the Joint Declaration on Freedom of Expression and “Fake News”, Disinformation and Propaganda by the United Nations (UN) Special Rapporteur on Freedom of Opinion and Expression, the Organization for Security and Co-operation in Europe (OSCE) Representative on Freedom of the Media, the Organization of American States (OAS) Special Rapporteur on Freedom of Expression and the African Commission on Human and Peoples’ Rights (ACHPR) Special Rapporteur on Freedom of Expression and Access to Information. This joint declaration of 3 March 2017 put forward as a general principle that “intermediaries should never be liable for any third party content relating to those services unless they specifically intervene in that content or refuse to obey an order adopted in accordance with due process guarantees by an independent, impartial, authoritative oversight body (such as a court) to remove it and they have the technical capacity to do that”.
Dirk Voorhoof, Human Rights Centre Ghent University (Belgium), Copenhagen University (Denmark), Legal Human Academy and member of the Executive Board of the European Centre for Press and Media Freedom (ECPMF, Germany)