Ahead of a hearing before the Italian Court of Cassation in the case of Fabio Butera, a journalist who was found liable for comments made by others on a post on his Facebook page, ARTICLE 19 reiterates its belief that the court must protect freedom of expression online. The lower courts ordered Butera to pay damages for failing to remove third-party comments, reasoning that he must have been aware of them but chose not to act. Upholding the lower court’s decision would be deeply concerning, as it would effectively impose content moderation obligations on ordinary users, posing serious risks to public debate and incentivising self-censorship. ARTICLE 19 calls on the Court of Cassation to overturn the lower court’s judgment and confirm that users cannot be held liable for third-party comments.
This case, which has taken several turns, centres on a fundamental challenge to freedom of expression online.
In August 2018, Fabio Butera, a journalist, published a post on his Facebook page criticising a news report by the journalist Valentino Gonzato for claiming that asylum seekers in Vicenza were protesting for a pay-TV subscription to watch football matches. Gonzato subsequently sued Butera for damages in civil defamation proceedings.
In April 2023, the Tribunal of Verona ruled that Butera’s post was not defamatory, as it was based on documented research in the public interest and factually accurate. However, the court ordered Butera to pay EUR 33,000 in damages (including legal fees) to Gonzato for failing to remove defamatory comments posted by third parties under his Facebook post. The court ordered this despite the fact that Butera had not been asked by any user or other party to remove any comments from his page. The Venice Court of Appeal upheld this ruling, reasoning that Butera’s publication of additional content on the same Facebook page a few days later proved that he had read the comments and failed to delete them.
ARTICLE 19 finds that the rulings of the Tribunal of Verona and the Venice Court of Appeal, which held Butera liable for third-party comments, effectively impose content moderation obligations on social media users for activity by third parties on their own accounts, with serious repercussions for freedom of expression online.
Importantly, we believe that the lower courts failed to consider international standards on freedom of expression, applicable in the case. We would like to highlight especially the following issues.
1. The lower courts erroneously extended intermediary liability standards to an ordinary social media user
In the European Union, internet intermediaries (such as web hosting companies, internet service providers, search engines, and social media platforms) enjoy robust exemptions from liability for content posted by their users. This well-established legal framework aims to shield these entities from being held responsible for unlawful content they host unless they have actual knowledge of it and fail to act, thus protecting freedom of expression while still establishing some content moderation responsibilities.
Although the case concerns an ordinary social media user rather than a commercial platform, understanding these intermediary liability protections is essential.
ARTICLE 19 has long argued that internet intermediaries should enjoy broad immunity from such liability, as they would otherwise effectively be required to monitor all user content and make determinations about its legality, when such decisions should rest with independent judicial authorities. If intermediaries faced liability for user-generated content, they would have strong incentives to remove material that is perfectly lawful and protected under international human rights law, simply to avoid potential legal risk. Therefore, intermediaries should only be required to remove content following an order from an independent and impartial court or other adjudicatory body that has determined that the material at issue is unlawful. This position is reflected in the Manila Principles on Intermediary Liability, developed by a consortium of civil society organisations with the aim of protecting freedom of expression online.
It is also to some extent enshrined in EU law. While the main EU legislation governing intermediary services, the Digital Services Act (DSA), imposes obligations on certain intermediaries, including very large online platforms, regarding how they should manage content moderation, it preserves the principle of conditional liability exemption established by the E-Commerce Directive. This means that service providers are not liable for user-generated content unless they have actual knowledge of its illegality. Such actual knowledge may be established if another user notifies the intermediary of the unlawful content.
Therefore, by finding that Butera, an ordinary social media user, can be held liable for comments posted under his content creates the absurd outcome of holding him to a higher standard than intermediaries. We note that many intermediaries are companies that operate the platforms commercially and maintain centralised content moderation systems with far greater resources.
While there are some questions regarding how intermediary liability rules may apply in the context of community-driven, decentralised forms of content moderation – where some users might act themselves as content moderators – it is clear that ordinary users are out of scope of EU-wide liability regimes. Beyond issues of fairness and resourcing, just as holding internet intermediaries liable for third-party content has serious implications for freedom of expression, holding individual social media users liable would be equally problematic.
2. By imposing liability for third-party comments, lower courts restricted freedom of expression online
ARTICLE 19 believes that upholding the decision of the lower courts in the Butera case and, thus, imposing liability on social media users for comments made by others, will carry numerous negative consequences for freedom of expression. In particular:
- Users lack the competence to judge content: Like companies, individual users are neither equipped nor legitimate arbiters of whether content posted by others on their accounts is ‘legal’. For example, in cases of defamation, as in Butera’s case, it is particularly difficult to see how a journalist, or any user, could be expected to assess the accuracy of third-party statements without access to all potentially relevant facts. This is precisely why liability should rest only on those who made the contested statements.
- Placing unacceptable burden on users: Imposing such a duty would place an unacceptable burden on users. To avoid potential liability, users might disable comments altogether or delete any content that could be seen as risky, including content that should be protected under international freedom of expression standards. Yet comment sections on social media are an essential part of online engagement. They allow users to share information, exchange ideas and opinions, and engage in discussion with others. Comment sections also expose users to diverse viewpoints by enabling direct responses and challenges to the content posted, serving both as spaces for debate and sources of information. Therefore, if users were to remove these sections to reduce exposure, this could qualify as a form of self-censorship insofar as they would restrict their right to receive information and opinions at the same time as restricting other users’ right to freedom of expression.
- Requiring resource-intensive monitoring: When it comes to journalists posting content, those who share material on matters of public interest are more likely to generate significant public debate, sometimes attracting thousands of comments. The effective monitoring of all comments, especially in the case of a post that generates significant debate, can require the availability of significant resources. While it is true that in such cases it would appear particularly unreasonable for a court to conclude that the author of the post had actual knowledge of all third-party comments, the threshold for establishing such knowledge, and the associated burden on the social media user, remains unclear. Given the ever-accelerating shift towards news consumption via social media and journalists’ reliance on those platforms for visibility of their content, it would be particularly problematic should journalists be forced to choose between visibility of and engagement with their work and facing legal exposure for posting public interest content, as the court recognised in this case.
Finally, there is a real risk that such liability could be weaponised to target journalists, civil society organisations or ordinary social media users. Individuals could deliberately post comments under someone’s content to expose that person or organisation to legal liability. Similar tactics have been used in the past, with actors exploiting copyright and privacy laws to claim that journalistic articles infringed intellectual property or data protection rules, leading to the removal of critical reporting from the internet and news archives.
3. The lower courts’ decision would likely be found in violation of Article 10 of the European Convention on Human Rights
The European Court of Human Rights (the Court) has pronounced itself a number of times on whether imposing liability for third-party comments in online fora is compatible with the right to freedom of expression under Article 10 of the European Convention on Human Rights (the Convention). Its jurisprudence on the issue is still evolving.
While some judgments, such as Delfi AS v. Estonia, criticised by ARTICLE 19, found no violation where a large, commercial news portal was held liable for failing to remove unlawful comments quickly, the Court set out critical proportionality criteria. These include examining the context of the comments, the measures taken to remove them, the possibility of holding the actual authors liable, and the consequences for the defendant. Notably, the Court emphasised the professional management and commercial nature of Delfi, distinguishing it from other, less structured forums, which indicates that ordinary social media users should not bear the same liability burdens.
Accordingly, the case does not concern other forums on the internet where third-party comments can be disseminated, for example an internet discussion forum or a bulletin board where users can freely set out their ideas on any topic without the discussion being channelled by any input from the forum’s manager; or a social media platform where the platform provider does not offer any content and where the content provider may be a private person running the website or blog as a hobby.
In subsequent cases, such as MTE and Index.hu v. Hungary and Jezior v. Poland, the Court found violations of Article 10 of the ECHR when liability was imposed in less serious contexts or on non-commercial sites, emphasising the risk of overburdening intermediaries and chilling freedom of expression.
A more troubling ruling is the decision of the European Court in Sanchez v. France. In this case, the Court upheld criminal liability on a local politician for comments on his Facebook feed but did so based on his role and responsibilities as a public figure engaged in an electoral campaign, circumstances distinct from journalists or ordinary users.
Importantly, in a recent decision on Pătraşcu v. Romania, which bears some similarity to Butera’s case, the Court found that the freedom of expression rights of an individual social media user had been violated in finding them liable for defamation for third-party comments. The case involved an applicant who was ordered by the High Court of Romania to delete comments on his blog and Facebook page and pay damages to two opera employees after reporting on a scandal at the Bucharest National Opera, which had generated significant public debate online. The Court found that the applicant’s right to freedom of expression had been violated. Although this was based on the lack of a sufficiently clear and detailed legal basis under Romanian domestic law rather than on an engagement with the Court’s prior case law, this case signals that the Court remains wary of extending strict liability regimes to users without clear legal frameworks.
Taken together, ARTICLE 19 finds that the Court’s jurisprudence rejects a one-size-fits-all approach that would impose on ordinary social media users like Butera the same liabilities applied to intermediaries and commercial platforms.
We believe that given Butera’s profile as a journalist using Facebook for commentary, his use of Facebook to share commentary cannot reasonably be equated with operating a commercial news outlet equipped with far more resources to engage in content moderation.
Moreover, although the topic of migration is often sensitive and politically charged, Butera was not running an election campaign that might reasonably be linked to a heightened risk of inflammatory or otherwise harmful comments. Finally, the payment imposed, EUR 33,000, is significant and far exceeds the amounts deemed proportionate in Delfi (EUR 320) and Sanchez (EUR 4,000, including payment to the opposing party).
ARTICLE 19’s recommendation
Irrespective of how the European Court’s complex case law might apply to Butera’s case, ARTICLE 19 believes that the Italian Court of Cassation should overturn the Court of Appeal’s judgment and affirm that social media users cannot be held liable for comments made by others. While social media today undoubtedly poses challenges for public discourse, it remains an essential space for democratic debate, public engagement, and for journalists to interact with their audiences.
The Court of Cassation should safeguard these spaces, reject the notion that individuals must police the speech of their peers, and uphold the right to freedom of expression.
