Lorna Woods, Professor of Internet Law, University of Essex

A recent case before the European Court of Human Rights (ECHR) raises the question of whether internet site providers should be held accountable for comments made by users. This issue gained significance following the court’s ruling in MTE v. Hungary.

Considering the importance of intermediaries in the internet’s growth, there are arguments for not treating them like traditional publishers. Instead, they could be granted some immunity regarding content posted by others. However, in the Delfi case, the ECHR’s Grand Chamber decided that holding a news portal responsible for user comments didn’t necessarily violate freedom of expression under Article 10 of the ECHR. This decision faced criticism.

The ECHR recently revisited the liability of online newspapers for reader comments. While referencing the Delfi case, the court reached a different conclusion. This begs the questions: Did the court deviate from its previous stance, or did the case’s specifics warrant a different outcome? The MTE case, while not the first on intermediary liability, reveals how the court’s thinking on internet-related issues is developing.

Facts

Two applicants, a news portal company and a Hungarian internet content self-regulatory body, published articles criticizing an online estate agency’s consumer protection practices. Both articles attracted user comments, some of which were negative. The agency sued both applicants for the articles and certain comments, prompting the removal of the content. The applicants argued against being held responsible for third-party comments, but the highest national court disagreed despite varying approaches and outcomes in lower courts.

Judgment

The ECHR acknowledged an interference with the applicants’ freedom of expression, focusing on whether it was justifiable under Article 10(2) ECHR. Despite the national courts’ varied justifications for imposing liability, the ECHR found the law sufficiently foreseeable due to the applicants’ professional context. The court also accepted the legitimacy of protecting others’ interests. The key issue was whether holding the applicants liable was ’necessary in a democratic society’ and proportionate.

The court reiterated that ’necessary’ implied ‘a pressing social need’, emphasizing the press’s crucial watchdog role in democracies, referencing the Delfi case. It stressed the internet’s unique nature as an information source before summarizing existing case law on reputation protection under Article 8. Essentially, both rights are equal; neither holds precedence, and the outcome should be consistent whether considered under Article 8 or Article 10.

The court then likened the applicants to the press due to their role in facilitating public debate, but noted their duties differed from traditional publishers given the internet’s nature. It cited the Delfi case, highlighting the ‘clearly unlawful’ nature of the speech in that instance. The court identified key differences: in this case, the speech was offensive but not clearly unlawful, and the regulatory body wasn’t profit-driven.

The court criticized the national courts for readily accepting harm without investigation, noting a distinction between reputational harm to individuals and companies. Nevertheless, it applied principles from cases like von Hannover II, Axel Springer, and Hachette Filipacchi to balance Article 8 (right to privacy) and Article 10. Referencing Delfi, it added that additional factors were relevant due to the company’s intermediary role:

The comments’ context, the company’s efforts to prevent or remove defamatory content, the liability of the authors as an alternative to the intermediaries, and the impact of the domestic proceedings on the company.

While Delfi involved hate speech, the court stated that these criteria also applied when assessing proportionality in non-hate speech cases.

The court concluded that comments concerning the estate agents’ malpractice were in the public interest and not gratuitous, distinguishing between the news portal and the self-regulatory body due to their different audiences. It criticized the national courts for disregarding the applicants’ role in generating comments and their impact on them. It suggested that while the applicants could have foreseen issues by allowing unfiltered comments, this essentially demanded “excessive and impractical forethought,” potentially hindering the freedom to share information online.

Comment

The court didn’t explicitly address the nature of the free speech rights at stake, as both sides seemingly agreed. This raises a question: Do those providing a platform for speech have their own expressive rights, even if they aren’t the ones speaking? This question has surfaced in cases like Pirate Bay without a clear answer.

The MTE case offers some clarification, albeit implicit. The court considered the applicants’ expression rights based on their communication role, noting they provided a platform for exercising expression rights and public information sharing. It then equated their position to the press.

While offering a starting point for evaluating expression rights, it remains unclear what constitutes a platform covered by Article 10. Telecommunications and letters fall under Article 8, suggesting a shift from the platform provider to the speaker’s protection at some point. The link between the platform and the press hints at content curation as a key factor in driving public discourse. The court seems to view user comments akin to interviews or letters to the editor:

Holding the applicant liable contradicts existing case law, stating that ‘punishing a journalist for helping disseminate another person’s statements in an interview would significantly hinder the press’s contribution to public interest discussions and shouldn’t be considered without strong justification.’

However, the court also seems concerned about users’ ability to speak freely, reflecting concerns about the audience’s interests in journalistic case law. It’s debatable whether one news forum’s case impacts the entire internet or whether the same protective reasoning applies to non-public interest matters.

Refocusing this question, one might ask if this reasoning applies to all intermediaries or just those closely linked to content. This is crucial, as the court applied press principles to MTE. Generally, the media enjoys substantial protection, though in Delfi the court focused on publishers’ duties rather than rights, acknowledging their potential difference from traditional publishers. It also aimed for a narrow judgment, avoiding setting precedents for all platforms.

Assuming a news portal falls under Article 10, how did the court differentiate between MTE and Delfi? The answer seems to lie in the weighting during the assessment of whether the restriction was “necessary in a democratic society.” As in Delfi, the court emphasized not overturning national courts’ decisions but rather assessing their balancing of Articles 8 and 10, expressing reservations about the Hungarian courts’ approach. The judgment frequently references Delfi and broader Article 8 and 10 case law. However, it’s debatable if the court truly refrained from second-guessing the national courts to justify the factual distinction from Delfi.

According to the court, Delfi involved clearly illegal hate speech, while the language in this case was merely “offensive and vulgar.” While the Grand Chamber categorized the speech in Delfi as hate speech, the boundary between that and “offensive and vulgar” in MTE seems blurry. Delfi contains a single instance of “sick Jew” used offensively amidst multiple instances of comparing a person to a pig. In MTE, a translated comment reads, “people like this should go and s**t a hedgehog and spend all their money on their mothers’ tombs until they drop dead.”

Furthermore, there are instances where the court upholds the Delfi approach but then justifies a different, narrower application. For example, the court reiterated Delfi’s approach to defamation’s effects but then argues that online communication styles lessen offense.

Finding the Hungarian courts’ inadequate balancing of freedom of expression and Article 8, the court considered the notice-and-take-down regime. It restated the Delfi approach, suggesting that such a system, coupled with rapid response procedures, “could function in many cases as an appropriate tool for balancing the rights and interests of all those involved.” However, endorsing the Grand Chamber’s stance also means endorsing the notion that clearly unlawful speech (specifically hate speech) necessitates immediate action. It’s difficult to envision a system enabling such swift action without requiring content monitoring. Yet, the court criticizes mandatory filtering as requiring “excessive and impracticable forethought” that could hinder online information sharing.

Lastly, the parties’ commercial motivations are noteworthy. While the court distinguished between the applicants, it didn’t explore the consequences. Commercial motivation in Delfi contributed to the no-infringement finding. This is concerning, as many intermediary services are profit-driven, directly or indirectly.

Judge Kuris’s concurring opinion states that the judgment shouldn’t be used by internet providers, especially those profiting from comments regardless of their content, to evade liability for failing to address harmful statements. Judge Kuris warns that such an outcome could legitimize an internet business model focused on profit at any cost.

In conclusion, while MTE moderates the Delfi approach, it doesn’t offer strong support for intermediary safe harbor. The court hasn’t developed a consistent theory for dealing with intermediaries, and questions about online speech styles and the balance of protection remain. As Judge Kuris noted, MTE is just the beginning of post-Delfi judgments.

Photo credit: universitypost.dk