9.1 Introduction

The European Court of Human Rights (ECtHR) has long established that freedom of expression is crucial to the existence of a democratic society. A free society is impossible without a free exchange of information and opinions. Freedom of expression also entails both the imparting as well as receiving of information. The internet has given unprecedented power for new platforms to realise this right.

Information society services and especially intermediary services, including social media platforms, have become an important part of our daily lives. At the same time, the digital transformation and increased use of those services has also resulted in new risks and challenges, both for individual users and for society as a whole.Footnote ¹

The ECtHR and other bodies in numerous cases have pointed to a positive obligation for the state to protect the right to freedom of expression. At the same time, such an obligation has so far not been expressly attributed to the online environment by the ECtHR, while some other human rights protection bodies and institutions have pointed towards it only indirectly.

Therefore, the aim of this chapter is to analyse the positive obligations of states to protect freedom of expression in the online environment, especially on social media platforms. To achieve this aim, the first part of the chapter will set the background by analysing the general positive obligations of the state in the context of freedom of expression as found by various international bodies, most notably the ECtHR. The second part of the chapter will look at how the current regulation of social media and internet intermediaries impacts freedom of expression. Consequently, the third part will evaluate the requirements of the state in terms of its positive obligations and current regulation in the context of freedom of expression.

It must be noted that this chapter is not intended as a general overview of the definition and characteristics of positive obligations under human rights law or as exhaustive research on the rules governing the liability of social media platforms and internet intermediaries. It is rather a merging of both of these topics, drawing attention to problematic situations and indicating their possible solutions.

9.2 Freedom of Expression and Positive Obligations of the State

Almost all human rights protection documents recognise the positive obligation of the state to ensure the protection of the rights enshrined in each particular document. For example, Article 1 of the European Convention on Human Rights (ECHR) provides that the Member States have a duty to secure for everyone within their jurisdiction the rights and freedoms defined therein. In addition, Article 13 of the ECHR guarantees the availability, at the national level, of a remedy to enforce the substance of ECHR rights and freedoms in whatever form they might happen to be secured in the domestic legal order. This requires the provision of a domestic remedy to deal with the substance of a complaint under the ECHR and to grant appropriate relief.Footnote ² Therefore, states have a positive obligation to investigate allegations of a human rights infringement. The procedures followed must enable the competent body to decide on the merits of the complaint of the violation of the Convention and to sanction any violation found but also to guarantee the execution of decisions taken.Footnote ³

Similarly, under Article 2 (3) of the International Covenant on Civil and Political Rights (ICCPR), state parties must ensure that persons whose rights under the Covenant have been violated have an effective remedy.Footnote ⁴ The Charter of Fundamental Rights of the European Union (CFREU) contains a rule that is comparable to the previously mentioned Article in the ICCPR and the ECHR. Namely Article 51 of the CFREU requires EU institutions and the Member States to ‘respect the rights, observe the principles and promote the application thereof’. At the same time, it must be noted that obligations contained in the CFREU have some limitations. The same article prescribes that this obligation is applicable only as far as EU law is being implemented and does not extend the field of the application of EU law.

It is the ECtHR that can be thought of as the most prominent advocate of imposing positive obligations on the state to protect particular human rights. In addition, it holds that although the essential object of many provisions of the ECHR is to protect the individual against arbitrary interference by public authorities, there may in addition be positive obligations inherent in an effective respect of the rights concerned. The Court has emphasised that the effective exercise of certain freedoms does not depend merely on the state’s duty not to interfere, but may require positive measures of protection.Footnote ⁵ Although the ECtHR has not provided a general definition of the concept of positive obligation, from its case law it can be deduced that the prime characteristic of positive obligations is that they require in practice national authorities to take the necessary measures to safeguard a right.Footnote ⁶

The positive obligations of the state can be divided into several groups. First, there are substantive positive obligations and procedural positive obligations. Second, there are positive obligations of a vertical kind or those that protect the individual from the state and positive obligations of a horizontal kind or those that protect individuals against other individuals. Third, there are positive obligations that relate to the legal and administrative frameworks and those that encompass more practical measures that states need to take.Footnote ⁷

These positive obligations can be found in relation to almost every human right laid down in the Convention, including freedom of expression. As the ECtHR has put it: ‘Genuine, effective exercise of this freedom does not depend merely on the State’s duty not to interfere, but may require positive measures of protection, even in the sphere of relations between individuals.’Footnote ⁸ This is because of the key importance of freedom of expression as one of the preconditions for a functioning democracy; therefore, states must ensure that private individuals can effectively exercise the right of communication between themselves.Footnote ⁹

In deciding whether a positive obligation relating to the freedom of expression exists, the ECtHR has emphasised that there must be regard for the kind of rights of expression at stake, their capability to contribute to public debate, the nature and scope of restrictions on expression rights, the ability of alternative venues for expression, and the weight of the countervailing rights of others or the public.Footnote ¹⁰ It should be noted, however, that the Court itself looks at these criteria cumulatively in each particular case, attributing more or less weight to any one or set of these criteria, depending on the circumstances.

The ECtHR has emphasised several positive obligations of the state in the context of freedom of expression. For example, the state has to protect the right to freedom of expression by ensuring a reasonable opportunity to exercise the right of reply and an opportunity to contest a newspaper’s refusal in suing for a right to reply in the courts.Footnote ¹¹ It has also recognised a rather broad obligation to create a favourable environment for participation in public debate for all persons concerned, enabling them to express their opinions and ideas without fear.Footnote ¹² Moreover, the state has a positive obligation to protect speakers, especially journalists,Footnote ¹³ from physical attacks from other individuals in connection with the exercise of its freedom of expression.Footnote ¹⁴ Recognising that there can be no democracy without pluralism,Footnote ¹⁵ in the context of access to the broadcast market, the ECtHR has emphasised that states must put in place an appropriate legislative and administrative framework to guarantee effective pluralism.Footnote ¹⁶

Other human rights protection bodies have also recognised that states not only have a duty to refrain from limiting the right to freedom of expression, but also have a positive obligation to protect and guarantee it. Although not explicitly recognising the concept of positive obligations under Article 11 of the CFREU, the Court of Justice of the EU has argued that it must be possible for national courts to check that interference of the information rights of internet users is justified.Footnote ¹⁷

Furthermore, the United Nations (UN) Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression in his report has emphasised that governments must adopt and implement laws and policies that protect private development and the provision of technical measures, products, and services that advance freedom of expression.Footnote ¹⁸

Therefore, it can be argued that freedom of expression not only prohibits unjustified interferences from states, but may also require an active participation on behalf of them. At the same time, the positive obligations of the state should not be understood too broadly. They have been recognised as applicable to a limited number of situations touching upon the very essence of the freedom of expression. In addition, they are more often than not formulated in a manner that provides states with a rather wide margin of appreciation.

In the second part of this chapter, I will analyse whether the existing rules on regulating social media platforms and internet intermediaries in general ensure the positive obligations of the state in protecting and ensuring freedom of expression in the online environment.

9.3 Regulation of Social Media Platforms and Internet Intermediaries: A Threat to Freedom of Expression?

There are several types of actors participating in the flow of information in the online environment. First, there are internet users – natural persons – who access information stored online. Sometimes these also share information, thereby in essence becoming content producers, which are the second category of online actors. Possibly the most complex of the three is the third category – the internet intermediaries, including social media platforms. They differ from content producers as those individuals or organisations who are responsible for producing information in the first place and posting it online.Footnote ¹⁹ Internet intermediaries give access to, host, transmit, and index content originated by third parties or provide internet-based services to third parties.Footnote ²⁰ The means by which they do this is almost without exception technical, meaning that the intermediaries are generally not aware of the content of the information they process and provide accessibility to. Such a role is unprecedented in the offline environment. Exactly because of this complexity, the regulation of internet intermediaries, including social media platforms, and their impact on freedom of expression is one of the focal points of this chapter.

The general and so far traditional rules of liability for internet intermediaries in the EU are set out in the E-Commerce Directive and recently in the Digital Services Act (DSA). The Directive exempts intermediaries from liability for the content they manage if they fulfil certain conditions. First, the service providers hosting illegal content need to remove it or disable access to it as fast as possible once they are aware of the illegal nature of the content. Second, only services that play a neutral, merely technical and passive role towards the hosted content are covered by the liability exemption.Footnote ²¹ For the Directive to become effective, it was transposed by the EU Member States into their national laws. Therefore, the E-Commerce Directive laid the groundwork for a notice-and-take-down procedure but did not provide any additional guidelines with regard to its implementation. Instead, the Directive left the subject matter to the discretion of the Member States.Footnote ²²

Certain Member States have developed more detailed, formal notice-and-take-down procedures. Possibly the most known of these is the German Network Enforcement Act, which was adopted in 2017. Among other rules, it orders big social networks to remove or block access to content that is manifestly unlawful within twenty-four hours of receiving a complaint about it.Footnote ²³ Administrative penalties of up to €5 million can be applied in cases where the social network fails to comply with the rules set out in the Network Enforcement Act.

The majority of the Member States opted for a verbatim transposition of the Directive, resulting in a lack of any firm safeguards for the content removal procedures in most EU countries.Footnote ²⁴ For example, the Latvian Law on Information Society Services in Article 10 provides rules for the liability of an intermediary service provider. Nevertheless, like the E-Commerce Directive, it does not specify what amounts to actual knowledge or immediate action of the service provider.

Recognising these issues and the need to harmonise intermediary liability rules throughout the EU, the Commission of the EU has proposed new rules for digital platforms.Footnote ²⁵ The Regulation of the European Parliament and of the Council on a Single Market For Digital Services (DSA) amending the E-Commerce Directive, as stated by the Commission, strives to maintain the core principles of the E-Commerce Directive and at the same time provide more protection for fundamental rights in the online environment, as well as online anonymity wherever technically possible.Footnote ²⁶ Although the proposed regulation clarifies some aspects that were previously unclear under the guidance of the E-Commerce Directive, the general rules on intermediary liability have been maintained.Footnote ²⁷ For instance, the DSA maintains the exemption of online platforms from liability provided the social media platform was not actively involved in the transmission or took action to delete the illegal information upon obtaining knowledge. However, one noticeable distinction in regard to liability rules is the fact that the DSA further develops the due diligence obligations applicable to social media platforms and includes new rules related to illegal content, content moderation, and algorithm oversight. While such due diligence obligations are dependent on the role, size, and impact of social media platforms, the fines for non-compliance are high and can reach a maximum of 6 per cent of a company’s annual worldwide turnover.Footnote ²⁸

Other pieces of EU legislation also impose an obligation on intermediaries deciding on which content should and which should not be available online. For example, the Audiovisual Media Services Directive promotes co-regulation and self-regulation.Footnote ²⁹ It also imposes duties on video sharing platforms to eliminate such harmful content from their platforms as incitement to violence and hatred.Footnote ³⁰ Under the General Data Protection Regulation, internet search engines are required to balance freedom of expression and privacy rights in applying the right to be forgotten.Footnote ³¹ The Code of Conduct on Countering Illegal Hate Speech Online announced by the Commission in May 2016 incentivises information technology companies to tackle hate speech online on their own initiative.Footnote ³² Therefore, for now, and it seems also for the foreseeable future, the balancing of various interests at stake is delegated to internet intermediaries.

Therefore, essentially it is the internet intermediaries, not the judiciary of states, that must take decisions on balancing freedom of expression, privacy, and other rights online. Although such a mechanism comes with some bonuses, such as speedy procedures and reduction of workload for the courts, there are several issues that arise from it, most visibly the threats to the freedom of expression of users of social media platforms and access to information removed by intermediaries by the public at large. Furthermore, after social media platforms imposed indefinite suspensions of former US President Trump, one can add prior censorship concerns. The one issue that is closely related to the positive obligations of the state to ensure protection mechanisms for freedom of expression is that in general this type of intermediary liability regime puts private intermediaries in the position of having to make decisions about the lawfulness or unlawfulness of the content and create incentives for private censorship.Footnote ³³ Several international organisations,Footnote ³⁴ and other stakeholders,Footnote ³⁵ have pointed to the fact that intermediaries should not be expected to conduct a quasi-adjudicatory exercise that weighs the rights of their users. They have argued that the fact that intermediaries have the technical means to prevent access to content does not mean that they are the best placed to evaluate the legality of the content in question, and whether measures affecting fundamental rights should be applied by an independent court rather than by private bodies.Footnote ³⁶ Intermediaries are commercial entities whose fear of potential liability, or lack of resources to fully address requests for removal of information, may motivate an overzealous response to individual requests that information be delisted.Footnote ³⁷ As private actors, intermediaries are not necessarily going to consider the value of freedom of expression when making decisions about content created by third parties for which they might be held liable.Footnote ³⁸

Another criticism of entrusting the balancing of fundamental rights to internet intermediaries is that by enlisting internet intermediaries as watchdogs, governments delegate online enforcement to algorithmic tools with limited or no accountability. Due process and fundamental guarantees are mauled by technological enforcement, curbing fair uses of content online and silencing speech according to the mainstream ethical discourse.Footnote ³⁹ In addition, until the obligations contained in the DSA are implemented, the algorithms in most cases are known to the intermediary alone or are even considered to be a commercial secret. Such a lack of transparency in the intermediaries’ decision-making processes can obscure discriminatory practices or political pressure affecting the companies’ decisions. Consequently, transferring regulation and adjudication of rights to freedom on the internet to private actors does jeopardise fundamental rights in general – such as freedom of information, freedom of expression, and freedom of business – by limiting access to information, causing chilling effects, or curbing due process.Footnote ⁴⁰ More particularly and closer to the topic of this chapter, it also influences the capacity of the states to carry out their positive obligations.

In the first part of this chapter, it was concluded that states must create a favourable environment for participation in public debate, protect speakers, and put in place an appropriate legislative and administrative framework to guarantee effective pluralism. We must remember that the ECtHR has recognised the state as an object of these duties, and human rights law does not, as a general matter, directly govern the activities or responsibilities of private business.Footnote ⁴¹ Yet, as emphasised earlier, the EU rules on intermediary liability follow different logics.

Since internet intermediaries through advertising or the payment of subscription fees benefit from disseminating third-party content online, it seems only fair that they should bear responsibility for preventing access to illegal or harmful material.Footnote ⁴² At the same time, while seeking the most effective enforcement mechanism, the regulation introduced by EU law and transposed into national law, shifts the balancing duty from the state to private companies – social media platforms. That means that not the states, but online platforms are the ones creating the environment for participation in public debate. Until recently, the limited potential to review decisions, often made using algorithms, is inevitably linked with limited possibilities to ensure the protection of speakers. It also involves very few options for influencing pluralism in the online environment, which is becoming the principal source of information and communication in many countries.

Therefore, although the intermediary liability regime itself does not interfere with the freedom of expression, it creates a situation of horizontal interference resulting from a failure of the legislature to effectively protect the right to freedom of expression – a form of ‘State interference by proxy’.Footnote ⁴³ The measures introduced by the DSA, which focus on ensuring more transparency and better protection of citizens’ fundamental rights online, including the obligation to provide information to users and establish a complaint and redress mechanism, is a huge step in the right direction. However, these steps should not be considered a substitute for the positive obligations of the state arising from the human rights standards set by the ECtHR and other international bodies.

9.4 Safeguards for Freedom of Expression

As just concluded, states have an obligation to effectively protect human rights from interference by other private individuals. Although not specifically mentioning the positive obligations of the state, the ECtHR has already argued that the state should protect freedom of expression online, not only by avoiding any limitations to it, but also by creating an appropriate legal framework. In Ahmet Yıldırım v. Turkey, one of the most important cases dealing with the accessibility of information online,Footnote ⁴⁴ it indicated that Article 10 of the ECHR requires a law to provide safeguards that are intended to protect against the over-removal of information from the internet.Footnote ⁴⁵

In addition, according to the UN Guiding Principles on Business and Human Rights, which requires businesses to avoid causing or contributing to adverse impacts on human rights, the duty to protect and to provide access to an effective remedy for violations of human rights is essentially incumbent on states.Footnote ⁴⁶ Therefore, even though the document emphasises the duties of private stakeholders, it still underlines the role of the state in the protection of human rights.

Arguments supporting the positive obligations of the state in the context of freedom of expression and intermediary liability can also be found in legal doctrine. These arguments are based upon the idea that as many powerful social media platforms have become central to communication and information exchange, the legal framework in which they exist must be compatible with human rights standards.Footnote ⁴⁷ Therefore, the content removal mechanisms should have a sufficient basis in law. To meet this requirement, the legislature should introduce specific legal provisions to clarify removal procedures. Legislation providing for content removal procedures should meet the requirement of ‘quality’. This means that rules should be clear and sufficiently precise for those subject to them to foresee the consequences and adjust their behaviour accordingly.Footnote ⁴⁸

The positive obligations of states to protect freedom of expression are perhaps of even more relevance if such interferences are accepted, or even encouraged by the states,Footnote ⁴⁹ as is the case in the rules dealing with intermediary liability. The current intermediary liability regime, the most notable documents of which are the E-Commerce Directive and the DSA, includes (especially in the latter) some safeguards that could ensure the protection of the right to freedom of expression.Footnote ⁵⁰ But its application across Europe is yet to be seen and it does not relieve the state as an ultimate actor to ensure fundamental rights also online. As stated earlier, the doctrine of the positive obligations of states in the context of freedom of expression may provide for the further legal protection of the users of social media platforms, their right to express themselves, and their right to have access to information.

9.5 Effective Remedies and Procedural Safeguards

When freedom of expression is violated, appropriate remedies to this situation may include access to information about the violation and grievance mechanisms.Footnote ⁵¹ In order for these safeguards for freedom of expression to be effective, the states should ensure that the principles of due process and access to independent and accountable redress mechanisms are respected in the application of them.Footnote ⁵²

First of all, users should be provided with the right to learn about the removal of the information they have published online.Footnote ⁵³ Otherwise, they practically cannot protect their right to freedom of expression.Footnote ⁵⁴

Although there is a chance that such a practice could backfire and once removed information could resurface again through different channels, a notification to webmasters and content providers should be issued by internet intermediaries whenever they restrict access to information created by these actors. Such an approach is supported, for example, by the UN Special Rapporteur on the promotion and protection of freedom of opinion and expression.Footnote ⁵⁵ It is also prevalent in the doctrine applied by the courts in cases concerning intellectual property rights in the US and Brazil.Footnote ⁵⁶ The notification should, as far as possible, include the reasoning of the decision that, if necessary, can be later challenged. Even more, in accordance with the Manila Principles of intermediary liability, before such a removal becomes permanent, the intermediary should weigh the arguments of the author of the information in question.Footnote ⁵⁷ The DSA apparently ensures this remedy by introducing the concept of ‘notice and action’ and obliging social media platforms to inform users of the removal of their information. Consequently, this rectifies the existing lacuna in the EU legal framework. However, the DSA only applies as far as to intermediaries who offer their services in the EU single market.

Second, users who are notified by the service provider that their content has been flagged as unlawful should have the option of challenging the block or filtering of their content and to seek clarifications and remedies.Footnote ⁵⁸ The possibility to appeal could come in various forms – using procedures provided by the intermediary or by a competent judicial authority.Footnote ⁵⁹

The DSA notes that first and foremost the internet intermediaries themselves should have a review procedure in place. Such an idea is nothing new and some intermediaries have worked on this already. In 2018, Facebook already announced that it will create an independent oversight body to adjudicate appeals on content moderation issues.Footnote ⁶⁰ The Board reviews a select number of highly emblematic cases and determines if decisions were made in accordance with Facebook’s stated values and policies.Footnote ⁶¹ Individual users can bring appeals to the Board, and Facebook as a company is able to refer cases for expedited review if they could have urgent real-world consequences.Footnote ⁶² Most notably the Oversight Board accepted a case referral from Facebook to examine their decision to indefinitely suspend former US President Donald Trump’s access to post content on Facebook and Instagram. Facebook has also requested policy recommendations from the Board on suspensions when the user is a political leader.Footnote ⁶³

At the same time, the opportunities to contest the decisions of online platforms should complement, yet leave unaffected in all respects, the possibility to seek judicial redress.Footnote ⁶⁴ There should always exist the option of judicial redress to ensure effective legal protection of the right to freedom of expression.Footnote ⁶⁵ Indeed, the safeguards would become redundant if there were no option to receive a judicial overview of the decisions made by internet intermediaries. The liability placed upon private companies to remove third-party content without judicial oversight would not be compatible with international human rights law and freedom of expression specifically.Footnote ⁶⁶

Although internet intermediaries are technically the best placed to evaluate applications for content removal and act upon them, such a mechanism might not be completely compatible with the requirements of legality and quality of law. While the practice of implementing the transparency rules provided by the DSA on online platforms is unclear, so far the internal decision-making procedures of the intermediaries are not always very transparent. Therefore, removal orders issued by independent and impartial bodies provide a much greater degree of legal certainty.Footnote ⁶⁷

Third, the independence of internal complaint and redress mechanisms and the readiness of social media platforms to follow the decisions of oversight boards depend also on the goodwill of the platforms. The change in ownership in 2022 of one of the biggest social media platforms, Twitter, illustrates how rapidly the existing policies of the social media platform can change and how fragile reliance only on internal redress mechanisms created by the platform can be. A court or similar authority would operate with greater safeguards for independence, autonomy, and impartiality,Footnote ⁶⁸ and would have greater capacity to evaluate the rights at stake and offer the necessary assurances to the user.Footnote ⁶⁹ Such authorities as public bodies (not private) would also be better placed to make the determination of whether a particular content is illegal, which requires careful balancing of competing interests and consideration of defences.Footnote ⁷⁰

The positive obligation of states to protect freedom of expression entails the creation of a system that would allow individuals and other content creators to protect their freedom of expression also in case it is exercised through social media platforms and internet intermediaries. These intermediaries should abide by the rules of freedom of expression themselves, but at the end of the day, the state cannot rely on these private players to protect speech online. The DSA envisages a new enforcement mechanism, which will complement the internal complaint and redress tools established by social media platforms. This mechanism will consist of the Commission and independent national authorities to supervise how online intermediaries, including social media platforms, adapt their systems to new requirements. To safeguard the freedom of expression, the independence of these bodies and the possibility of challenging their decisions in the courts are of crucial importance. The criticism so far has been correctly expressed that the oversight of very large platforms is subject to the European Commission, which is not an independent regulator but the executive arm of the EU.Footnote ⁷¹

9.6 Rules on Balancing

At this point, it is clear that monitoring and restricting access to information online would be impossible without the involvement of internet intermediaries. They act as gatekeepers with direct access to the keys needed to close the gates to malicious or otherwise illegal information. The keys available to states are much more robust and take longer to activate. This means that to ensure effective protection of human rights online, the states have to cooperate with internet intermediaries. Therefore, besides the creation of a mechanism for protecting freedom of expression online, states have to set clear rules for internet intermediaries to follow in deciding which information remains accessible and which should be removed. The importance of clear rules at the national level is especially relevant, since only some illegal content is defined at the EU level, while some content might be found illegal and subject to removal only in certain states.

Internet users should always be able to understand why certain information has been removed.Footnote ⁷² The changes introduced by the DSA apparently ensure that social media platforms notify users of the reasons for removing their content and the option to contest such decisions.

The states should ensure that removal of information or blocking access should be undertaken in the observance of the principle of freedom of expression.Footnote ⁷³ In mediating the public interest and individual rights online, a delicate regulatory balance is required.Footnote ⁷⁴ Therefore, the state should provide guidance for intermediaries on how to achieve this.Footnote ⁷⁵ The DSA could play a role in establishing standards and best practices in this area as it tries to balance measures imposed on platforms to remove harmful information against the restrictions to freedom of expression. Yet the application of the DSA both at the EU and national level is yet to be seen.

9.7 Conclusions

Social media platforms and internet intermediaries in general are becoming, and in many societies already are, the dominant actors in providing an environment for freedom of expression and communication. Yet the legal regulation of platforms has not always followed technological developments and changes in society so as to ensure respect for freedom of expression and other fundamental rights. The idea during the emergence of the online environment that it should be left free and unregulated to ensure the marketplace of ideas was gradually changed with increasing regulation from the states in order to protect other fundamental rights, for instance human dignity, and ensure non-discrimination. The signs of this shifting attitude of states and international bodies was evidenced by the judgement of the ECtHR in the case of Delfi v Estonia,Footnote ⁷⁶ which for the first time imposed the liability on a news platform for not removing derogatory comments created by its users in a timely manner. The Court came to such a decision despite the fact that the platform acted quickly after the notice was received. This judgment was followed by laws restricting freedom of expression and imposing more obligations on social medial platforms in a number of European countries,Footnote ⁷⁷ and even debates in the US about changes to section 230 of the Communications Act of 1934, which grants websites legal immunity for much of the content posted by their users.Footnote ⁷⁸

While such moves by states are understandable in order to protect other fundamental rights and currently provide measures against the spread of misinformation and harmful content in the online environment, the safeguards for freedom of expression seems to have been forgotten in this process. Online intermediaries, which have implemented different technical tools to block and to remove information in response to demands by states, are actually now in charge of setting the boundaries of freedom of expression – a role that should be performed by states. The role of freedom of expression in democracies is essential, and states should not shirk their obligations in ensuring freedom of expression and deciding about its limits by outsourcing this function to private corporations. Social media platforms and internet intermediaries in general have neither democratic legitimacy nor do they in many cases share the goal of protecting this fundamental right, and the process of limiting expressions is in most cases accompanied by a lack of clear rules, transparency, and access to remedies. We must remember that the right to freedom of expression entails not only the negative obligations of the state, but also positive ones. The doctrine of these positive duties is still developing and awaits clarification, especially in the context of information published online. However, it can be argued that it can already be applied in order to improve the safeguards for freedom of expression for users of social media platforms, and discourage the platforms from the over-removal of content to avoid liability. The DSA is certainly a step forward; however, its scope is limited and its application is yet to be seen.

Last but not least, the majority of existing legal standards focus on the development of content moderation rules, which would balance freedom of expression and the removal of illegal content by social media platforms. Still, another and similarly important threat to freedom of expression in the online environment is the fact that the vast majority of public discourse takes place on a very small number of platforms, which hold excessive power over information flow.Footnote ⁷⁹ As stated earlier, the ECtHR has emphasised that states must put in place an appropriate legislative and administrative framework to guarantee effective pluralism in the media ecosystem.Footnote ⁸⁰ Therefore, states are obliged to decentralise the channels of public discourse and develop rules that favour the development of new social media platforms.

10.1 Introduction

Democracy and freedom of expression are intertwined in the European Court of Human Rights (hereinafter, ‘the ECtHR’ or ‘the Court’) case law. Article 10 (Freedom of Expression) of the European Convention on Human Rights (hereinafter, ‘the ECHR’ or ‘the Convention’) introduces obligations on Member States to protect the right to freedom of expression. The scope of the right is broad, and as such, it clarifies that any limitation on the freedom of expression must be prescribed by law, pursue a legitimate aim, and be necessary in a democratic society (e.g., protecting the reputation or rights of others). In addition, Articles 17 (Prohibition of abuse of rights) and 18 (Limitation on use of restrictions on rights) of the ECHR are also relevant in assessing the scope and legitimacy of restrictions against the freedom of expression. In protecting the right to the freedom of expression and robust public debate, which are the foundations and pre-conditions for a democratic society, strong safeguards have been developed by the ECtHR in its case law interpreting the Convention.

Within this case law, the defence of democracy as a legitimate aim of limiting the freedom of expression is a relatively unnoticed part of European case law. More often, the discussion has been on the abuse of rights provision, which is closely related to extreme political movements that try to promote undemocratic ideologies. In this connection, the Court has reiterated that the general purpose of Article 17 is to prevent individuals or groups with totalitarian aims from exploiting the principles enunciated in the Convention in their own interests.Footnote ¹

There has been consensus on certain fundamental doctrines related to freedom of expression and defending democracy. However, these are more designed for a traditional media environment that does not reflect the prevailing internet-based system of expression in our current society. In the contemporary system, the defence of democracy is a more complex and multifaceted phenomenon than just guaranteeing the protection of democracy from extreme movements and the most serious hate speech. That is, pre-emptive measures have become a key part of the tools for defending democracy. This means that there need to be strong foundations for securing public debate, including a diverse and pluralist media environment that would enable safeguards against disinformation and other attacks on the democratic system. In contemporary society, the concept of the media as a public watchdog has also evolved into more of a diverse network of watchdogs, consisting of different actors, from the media to non-governmental organisations (NGOs) and even academics.

In addition, there is a fear of abuses of the defence of democracy argument. Sanctions and other restrictive measures could be used to silence critical voices and democratic opposition. This development in applying ulterior purposes is inevitable when looking at how Article 18 of the Convention has emerged in the Court’s case law. There is more evidence of using different sanctions for ulterior motives, especially against those in opposition.

Against this background, this chapter tries to systematise current interpretative doctrines that the ECtHR applies in cases related to defending democracy and the rule of law under Article 10. In addition, it looks at the approach to contextualism taken by the Facebook’s Oversight Board. The idea is to analyse these doctrines in light of present-day conditions and to evaluate how the doctrines are adaptable to the new media environment. In so doing, this chapter first focuses on the question of how European history and the historico-political context have been taken into account and what the key principles have been in assessing the width of the margin of appreciation in relation to the right to freedom of expression in the Court’s interpretation in particular regard to defending democracy. Section 2 will then further outline established doctrines of defending democracy in the Court’s case law. Section 3 of the chapter delves more into how the emergence of internet-based media and enhanced access to a space of social debate has further complicated the case law, where the focus has shifted more to encompassing freedom of expression online for upholding democracy as a system rather than merely restricting extremists or hate speech. The section also considers the threats of disproportionately restrictive measures on public debate, which might undermine democracy and fall under Article 18. The section also looks into the positive obligations and how a regulatory framework requiring pluralism would be essential as a pre-emptive measure against threats to democracy. The final section provides a concluding observation on how the ECtHR has dealt with the defence of democracy argumentation in relation to freedom of expression and the evolving nature of such established doctrines in the face of the current media environment and democratic society.

10.2 Historical Context and Stability of Democracy Argumentation

According to the Court’s case law, the defence of democracy argumentation must be interpreted in light of European history. Consequently, the argumentation of protecting democracy is founded on the history of the twentieth century and especially on how fragile democracy can be in times of transition and where certain resurfacing ideologies pose a threat to democracy, referring to the ideologies and even symbols of authoritarian regimes, whether under Nazism or Communism.

Context is part of the defending democracy doctrines, and the Court emphasises a careful examination of context,Footnote ² particularly when symbols with multiple meanings are used for the expression of political views.Footnote ³ Therefore, the Court does not exclude that certain symbols are not always ‘equally permissible in all places and [at] all times’, stating that ‘the display of a contextually ambiguous symbol at the specific site [of mass murders] may in certain circumstances express identification with the perpetrators of those crimes’ and that a ban on demonstrations when held on a specific day of remembrance may ‘represent a pressing social need’ and thus ‘necessitate an interference with the right to freedom of expression’.Footnote ⁴

In the case of Vogt v. Germany, history was essential for assessing whether the restriction on the freedom of expression in the name of defending democracy was in accordance with the Convention. The Court analysed whether a legitimate aim existed to restrict the rights of civil servants when the political loyalty requirement was applied to a teacher in a public school who was a member of a communist party and was dismissed on account of her political activities. The political loyalty requirement requires that state employees are loyal in actively upholding a free democratic constitutional system. The Court noted that the notion of the civil service being ‘the guarantor of the Constitution and democracy has special importance in Germany because of that country’s experience under the Weimar Republic, which, when the Federal Republic was founded after the nightmare of Nazism, led to its constitution being based on the principle of a “democracy capable of defending itself” (wehrhafte Demokratie)’.Footnote ⁵

The same idea of historical experiences and using democratic processes to promote totalitarian movements was also mentioned in the Refah Partisi case (2003).Footnote ⁶ The communist authoritarian past was relevant in the Ždanoka v. Latvia case, where the Court established and consolidated the current criteria for the ‘self-defence’ of democracy. The Ždanoka judgment includes an extended doctrinal summary under the title Democracy and its protection in the Convention system, where it acknowledges that ‘democracy constitutes a fundamental element of “European public order”’ and the Preamble of the Convention makes a connection between the realisation and the maintenance of human rights and fundamental freedoms with ‘effective political democracy’.Footnote ⁷ It also refers to the common heritage of political traditions, ideals, freedom, and the rule of law of European countries enshrined within the Convention, where, in fact, on many occasions, the Court has reiterated that the Convention exists ‘to maintain and promote the ideals and values of a democratic society’.Footnote ⁸ According to the Ždanoka judgment, democracy is the only political model contemplated by the Convention and, accordingly, the only one compatible with it.Footnote ⁹

In the Ždanoka case, the Court referred to the principle that:

The Court applies very careful reference to the historico-political background, which makes the same measure acceptable in one country and yet incompatible in another with different historico-political contexts.Footnote ¹¹ Regarding a state with a less established democratic tradition or institutions, or those with a totalitarian past, therefore, the Court considers that the margin of appreciation is different, noting in such a state, national authorities would be better placed to ‘assess the difficulties faced in establishing and safeguarding the democratic order’.Footnote ¹² Under the Court’s scrutiny, the Latvian context was different because the Court acknowledged that there might be a threat against the new democratic order and a ‘resurgence of ideas which, if allowed to gain ground, might appear capable of restoring the former regime’.Footnote ¹³

One of the key arguments in this contextual approach was that in the field of election-related rights, there was a wide margin of appreciation, but also in the Latvian context, the Court affirmed that the Parliament should keep statutory restrictions under constant review.Footnote ¹⁴ The Court also referred to European integration as one of the key elements to be taken into account as a contributing factor for Latvia’s stability. Therefore, the election context impacts the proportionality test and gives more discretion to national authorities.

In this section, we have looked at the Court’s deliberation so far based on European history and historico-political contextuality in its judgments in relation to defending democracy and protecting the freedom of expression of individuals, especially when political views are expressed. We can say that the foundational correlation between democracy and the Convention has been clearly demarcated in its interpretations, and democratic society is a central concept in the ECHR. Moreover, the ECtHR has applied the democracy argumentation since the start of its work, particularly when talking about the general principles of interpretation, as the Court founded its argumentation in the Soering case that as ‘[t]he general spirit of the Convention, [is] an instrument designed to maintain and promote the ideals and values of a democratic society’.Footnote ¹⁵

Furthermore, ‘an effective political democracy’ is mentioned in the Preamble and ‘democratic society’ appears in six of the Articles (6, 8, 9, 10, 11, and 2 of Protocol 4) of the ECHR. The emphasis on the value of protecting the democratic system is deeply rooted in European history and human rights atrocities, and the reason to draft the Convention and establish its supervisory system in the spirit of the Universal Declaration of Human Rights was to ensure that such events would never be repeated in Europe.

10.3 Established Doctrines on Defending Democracy in the ECtHR Case Law and Heckler’s Veto

Now that we have looked at the approach that the Court has taken in regard to the defence of democracy and the freedom of expression in its case law, in this section, we will highlight some of the established doctrines and principles related to defending democracy.

The key doctrinal element in defending democracy is the Prohibition of the Abuse of Rights, enshrined in Article 17 of the Convention. Against the background of the provision is the idea that a state, group, or person does not have a right to activities aimed at the destruction of the rights and freedoms of others. For example, the Court has, in the case of Refah Partisi et al. v. Turkey, reiterated that ‘no one should be authorised to rely on the Convention’s provisions in order to weaken or destroy the ideals and values of a democratic society’.Footnote ¹⁶ In the case of Ždanoka v. Latvia, furthermore, by referring to the consideration during the drafting of the Convention, the Court acknowledged that certain rights could not be used in order to ‘destroy’ certain other rights set forth by the Convention, stating:

The Court has also set forth the principles for assessing the breadth of the margin of appreciation afforded to national authorities in its freedom of expression case law. Freedom of expression is closely related to democracy. According to the Court, freedom of expression constitutes one of the essential foundations of a democratic society, and the basic conditions for its progress and each individual’s self-fulfilment.Footnote ¹⁸ As such, the Court states, ‘in a democratic system, the actions or omissions of the government must be subject to the close scrutiny not only of the legislative and judicial authorities but also of public opinion’.Footnote ¹⁹ This means that authorities should tolerate even harder criticisms from the public, yet the key principle is set that ‘where such remarks incite violence against an individual or a public official or a sector of the population, the State authorities enjoy a wider margin of appreciation when examining the need for interference with freedom of expression’.Footnote ²⁰ As stated in the joint dissenting opinion in Sürek and Özdemir v. Turkey, judges have expressed that even though freedom of expression constitutes one of the foundations of democratic society, speech inciting violence undermines democracy.Footnote ²¹

As such, the Court has accepted that states can limit rights guaranteed under the Convention in order to defend democracy.Footnote ²² Furthermore, these measures can also be implemented in a pre-emptive manner. That is, states cannot be required to wait for authoritarian forces ‘to take steps that might prejudice civil peace and the country’s democratic regime’ before intervening.Footnote ²³ The Court referred to the possibility of preventing attempts against democratic regimes even before they occur. It approached the situation with the purpose of deference to national courts and other authorities. Therefore, the national courts, after detailed scrutiny of European supervision, can take restrictive measures to protect democracy. However, there are different contexts to be taken into account in the discourse of defending democracy; that is, how much the line should be stretched to tolerate minority opinions.

In contemplating this question, let us examine the Vajnai v. Hungary case, where the Court deliberated over disproportionate restrictions in the name of defending democracy and was concerned that free speech and opinion would be subjected to a heckler’s veto, which would negate freedom of expression.Footnote ²⁴ The heckler’s veto is invoked in situations where hecklers or demonstrators silence a speaker without the intervention of the law. The heckler’s veto is thought to happen when the government considers restrictions on a speech with the anticipated or actual reactions of the opponents of the speech.Footnote ²⁵

In this case, Attila Vajnai was convicted of using a symbol of an international workers’ movement, which was considered a totalitarian symbol and prohibited in Hungary. The Court recognised the systematic terror of communist rule in several countries and regarded that ‘it remains a serious scar in the mind and heart of Europe’.Footnote ²⁶ It is understandable that these symbols can cause uneasiness or seem disrespectful. Such sentiments cannot limit the freedom of expression. However, the Court went on to consider that this was about feelings rather than rational fears. The Court reviewed the proportionality of restrictions against Vajnai and considered that such a restriction based on satisfying ‘the dictates of public feeling – real or imaginary – cannot be meeting the pressing social need requirement in a democratic society’.Footnote ²⁷

The disproportionate conviction obviously reflects the legislative framework that did not previously provide any further qualification for wearing certain symbols. The authorities did not have to weigh wearing the symbol and other interests in order to find a fair balance. In their report to the Committee of Ministers, they referred to the new formulation of legislation, which does not base criminal conviction on the mere display of a symbol, but also includes an additional formulation, stating ‘in a manner that is capable of disturbing public peace, in particular that violates the human dignity of victims of totalitarian systems or the due reverence for the dead’.Footnote ²⁸ The national legislation now determines more precisely the condition under which wearing such a symbol is prohibited and leads to a conviction.

So far, the established doctrine reflects an era where democracy was organised by political parties. However, a contemporary democratic system has a more complex structure with different layers. In addition to the model of a representative democratic system, there is also a direct democracy element, with NGOs having a major impact on the public debate on specific questions of interest.

10.4 Distinguishing between Traditional and Internet-Based Media and Freedom of Expression

In terms of assessing the balance of journalistic freedom and the legitimate restriction of the freedom of expression by a state in the context of traditional media, the Court identified four key elements to be considered in the hate speech case of Jersild v. Denmark in 1994. Those were (a) the manner in which the feature was prepared, (b) its contents, (c) the context in which it was broadcast, and (d) the purpose of the programme.Footnote ²⁹ In this case, the applicant, a journalist, made a documentary containing extracts from a television interview he had conducted with three members of a group of young people calling themselves the Greenjackets, who had made abusive and derogatory remarks about immigrants and ethnic groups in Denmark. The applicant was convicted of aiding and abetting the dissemination of racist remarks, and he alleged a breach of his right to freedom of expression.

While the national courts laid emphasis on issues different from those of the Court, considering that the applicant had intended to include the racist remarks and edited the feature in that way, the Court took a different view and considered that the objective purpose of the feature was not the propagation of racist views and ideas. The Court drew a distinction between the members of the Greenjackets, who had made openly racist remarks, and the applicant, who had sought to expose, analyse, and explain this particular group of youths and to deal with ‘specific aspects of a matter that already then was of great public concern’.Footnote ³⁰ It was opined that the documentary as a whole was not aimed at propagating racist views and ideas but at informing the public about a social issue. Accordingly, the Court held that there had been a violation of Article 10 of the Convention.

What was interesting in the different approaches taken by the national courts and the European Court was the consideration of the scope of the assessment. That is, the Court considered that there were editorial decisions, especially related to the news value of the items that were broadcast by the media. According to the Court, there was no reason to question the editorial staff’s own appreciation of the news or information value of the impugned item, and emphasis was also placed on the fact that it was part of a serious news programme.Footnote ³¹

Now, how has the established case law since the Jersild case reflected the situation in social media and the responsibility of social media giants such as Twitter/X and Facebook? How can we apply the four factors mentioned in the Jersild case to internet-based media? First of all, the ‘context in which something is broadcast’ is one of the trickiest parts of transforming the broadcast context to social media because one of the key features of social media has been its ability to provide everyone with access to the social discussion in a more or less public sphere. It should not matter whether a person is mostly providing his or her own personal comments or whether a person is representing his or her institutional position. But are there, or should there be, different requirements for those with a more formal media status?

For some time, the internet’s role in the media environment and public debate has been acknowledged in the Court’s case law. Internet-based media has become one of the key focal points in the Court’s interpretation of freedom of expression in recent years, as seen in the case of Delfi v. Estonia. In this case, the Court reminded us of the important role played by the internet in enhancing the public’s access to news and facilitating the dissemination of information,Footnote ³² and later the Court elaborated further, noting that the function of bloggers and popular users of social media may also be assimilated to that of public watchdogs,Footnote ³³ insofar as the protection afforded by Article 10 is concerned. An essential factor in the Court’s argumentation was the interrelationship between the functioning of the political system and the role of the press and NGOs.Footnote ³⁴

The Court has also articulated a need for a legislative framework applicable to the imparting of information received from the internet. For example, in the Ukrainian case of the Editorial Board of Pravoye Delo and Shtekel, the Court considered that the absence of a sufficient legal framework could have negative consequences for the freedom of expression. It found that the lack of legislation at the domestic level allowing journalists to use information obtained from the internet without fear of sanctions for such use seriously hinders the exercise of the vital function of the press as a public watchdog.Footnote ³⁵

Online media outlets are also often now applicants before the Court. One recent interesting judgment was in the case of OOO Memo v. Russia, where an executive body of the state instigated the defamation case. The Court made a new interpretative approach and considered that executive bodies do not have the same rights as legal entities. In previous cases, the question was more about proportionality, and in the OOO Memo case, the Court chose a different line of interpretation. That is, the argumentation comes from the need for the close scrutiny of public power in order to prevent the abuse of power and corruption of public office in a democratic system. The Court saw that if executive bodies could file civil defamation cases, it would place an excessive and disproportionate burden on the media, creating an inevitable chilling effect. The conclusion was that restricting the freedom of expression on the basis of the defamation inflicted on the executive body did not have a legitimate aim.Footnote ³⁶

In another case, Cicad v. Switzerland, the Court considered it proportionate that, through a court order, the applicant was to remove an article from a website and publish the main findings on its website.Footnote ³⁷ In the Hungarian case of Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt, other special features of the internet were considered necessary to be taken into account when deciding whether certain measures are proportionate or not, including expressions in the comment area. According to the Court, ‘comments, albeit belonging to a low register of style, are common in communication on many internet portals – a consideration that reduces the impact that can be attributed to those expressions’.Footnote ³⁸

Another aspect that is typical of the contemporary media environment is that there are actors who deliberately aim to use different forms of disinformation as a way of destabilising democratic systems and societies through abusing the freedom of expression. The flood of disinformation and other hybrid tactics used in the new media environment can be described as, according to some scholars, information disorder and information pollution.Footnote ³⁹ According to the Council of Europe report, the forms of disinformation are (a) false context, (b) imposter content, (c) manipulated content, and (d) fabricated content. In these distributions of disinformation, the role of different types of new media has become an essential factor to be taken into account.Footnote ⁴⁰ Often, this means that social media platforms, for example, are taking a key role in the dissemination process and spreading distrust in the democratic system. The Joint Declaration from the United Nations (UN) and the Organization for Security and Co-operation in Europe (OSCE), African Union, and Organization of American States rapporteurs have emphasised the positive duties of states especially in promoting a free, independent, and diverse media environment.Footnote ⁴¹

As we have previously seen, the defence of democracy doctrine has been developed in the context of protecting the democratic system from those with totalitarian beliefs and a totalitarian history, especially in relation to electoral candidates or the loyalty requirements set for civil servants, but also in relation to pre-emptively dissolving political parties with objectives that are contrary to, and thus weakening or destructive to, the democratic system.

In the course of this development, the question of anti-democratic forces and their imminent threat to democracy has led to accepting pre-emptive measures against such forces, even accepting the dissolution of political parties. Some scholars have warned about the risks of abuse and unclear limits of defending democracy. This was raised, especially in the context of peace-time limitations.Footnote ⁴² Of course, given the current situation, this criticism has to be reviewed in light of Russia’s war against Ukraine. The threat of using disinformation to overthrow democratic governments, and even more generally to destabilise democratic processes that might have been previously seen as highly unlikely, is now very concrete in countries neighbouring Russia, and even elsewhere where Russian influence and hybrid tactics can cause imminent danger to the democratic system.

In a new media environment, which needs to be distinguished from earlier interpretative lines of print and audiovisual media, one of the significant issues involves anonymity. It has become an important part of social media that people are allowed to express views while maintaining their anonymity. Its value in social media posts and different digital platforms has been widely accepted. The idea behind anonymity is that it provides the opportunity to avoid reprisals or unwanted attention and thus promotes the free flow of opinions, ideas, and information, as was mentioned in one of the key cases in this context, that of Standard Verlagsgesellschaft mbH v. Austria (no. 3).

This is, however, one extension that also caused criticism within the Court. Judge Eicke saw that the Court’s line of argument could cause problems for victims of abusive speech, stating that:

Another interesting case is that of Ecodefence and 60 others v. Russia, where one of the most elaborate attempts by a state to create a legal regime placed a significant chilling effect on different NGOs and their choice to seek or accept any amount of foreign funding, especially in respect of politically sensitive or domestically unpopular topics. The case concerned the Foreign Agent Act and its concept of political activities carried out by so-called foreign agents, which could also include any views expressed in interviews or even social media posts by those ‘foreign agents’.Footnote ⁴⁵

There is a more rigorous approach to online media hatred that also emphasises the role of governments in criminalising hate speech. This has especially been seen in recent cases against minority groups. Hate speech related to a person’s sexual orientation is a key factor in the case of Oganezova v. Armenia, where an inadequate response was considered a violation of Article 3 of the Convention together with Article 14.Footnote ⁴⁶

10.5 Concluding Observations

So far, the development of the defending democracy doctrine has been considered to involve only very narrow extreme political movements pursuing totalitarian ideologies. The relevance of the historico-political context, and especially the history of events in Europe, have made it necessary for the Court to establish a doctrine based on the ‘self-defence’ of democracy. The Court has repeatedly reminded us of the scars related to Nazism and Communism. It has also reminded us about specific threats related to religious extremism. The historico-political context in different countries, especially the stability of the democratic system, constitutes one of the relevant issues to be assessed in the process of achieving a fair balance. On the one hand, this has allowed different standards in countries that have close historical connections to human rights atrocities in the past; for example, those who are denying historical events such as the Holocaust are considered under Article 17, thus examining their conduct as an abuse of rights. On the other hand, historical context is relevant in states with a short democratic history and, in some cases, forces with support for old authoritarian regimes.

But the contextual discussion is also currently bringing another side of contextualism. The proportionality test should also take into account the prevailing level of protection of the freedom of expression in the country. This is apparent in the recent cases before Facebook’s Oversight Board, considering that weak protection should also have an impact on applying community standards and considering whether removing posts is proportionate. Obviously, the Board has acknowledged that in countries such as Russia and Turkey, there should be a higher threshold for removing Facebook posts.

The European Court’s doctrines on the breadth of the margin of appreciation can provide a clear sign that regulating a complex and multifaceted issue requires a wider margin. Still, when authorities enter into the questions at the core of journalistic activities, such as the choice of journalistic methods, the margin is obviously narrow.

Democracy should be founded on pluralistic media, which, as one of the public watchdogs, ensures critical public debate on issues of public interest. When the authorities start making decisions on what is an acceptable issue to be discussed and what is not, this will ultimately lead to restrictions that jeopardise the core of freedom of expression. The Court states that it is not for the authorities or the Court ‘to review the press’s own appreciation of the news or information value of an item or to substitute their views for those of the press on what methods of objective and balanced reporting should be adopted by journalists’.Footnote ⁸⁹

When regulating audiovisual media, it has been necessary to take into account practical issues related to limited frequencies. Case law has continued to adopt, little by little, new elements related to a contemporary democratic system, in which democracy is something beyond elections and more about a comprehensive democratic system. On a global level, we are already seeing increasing repressions of freedom of expression and media integrity under the guise of fighting disinformation and protecting infrastructure against cyber-attacks.Footnote ⁹⁰ This kind of trend can also be detrimental to European countries.

Threats in the new media environment are, to a certain extent, similar to traditional print or audiovisual media. The Court founded its case law on the idea of the chilling effect that different sanctions can have on the media. It was important that the media and journalists were able to impart information to the public without censorship. The media’s access to information was key to performing as a public watchdog over those in power. With the democratic system becoming more complex, including multiple other actors in addition to the media, who are directly imparting information, it is necessary to talk of public watchdogs in the plural, including those who challenge the government and corporations, whether they be individual activists or networks of activists.

However, an even more significant development in the case law is related to understanding that the defence of democracy is related to understanding how modern democracy works and especially how one of the foundations of democratic society, freedom of expression, has transformed in the new media environment. There is ultimately a fear that restrictions become rules and protection comes as an exception. The Oversight Board has revealed many fundamental failings in respect to social media giants such as Facebook and Instagram removing posts, leading to a quest for transparency and specified rules on removing posts because the policy currently applied is in danger of silencing social debate.

Freedom of expression has changed substantially in the sense that governments are acting in bad faith against opposition parties. From Article 17 and extreme groups, it is now Article 18 that is not just an infrequent curiosity of the Convention but trending in cases such as Navalnyy. At the same time, more restrictions towards NGOs are being discussed, as in the case of Ecodefence and 60 others v. Russia. The linking issue in these cases is that the aim is to have a chilling effect and control the social debate.

There is a lack of clear interpretative guidance, especially concerning social media posts and applying traditional freedom of expression doctrines to them. It is not simply a case of proceeding with the Jersild doctrine, when at that time the situation was much clearer as to ‘informing the public about a social issue’. In the new media environment, there is a strong need to develop criteria for freedom of expression and how best to apply these in different contexts. The Court has already decided several cases since Delfi v. Estonia. Despite the idea of source neutrality, a different context must be taken into account when modifying doctrines to be relevant to cases where the role of digital media has become central.

11.1 Introduction

The dissemination of false, incomplete, erroneous, distorted, or intrusive information is a phenomenon as old as humanity, but since the development of information and communication technologies (ICTs), it has re-emerged with a vigour never seen before. The causes and effects have varied considerably to the point of becoming a very dangerous threat not only to human rights but also to the stability of the democratic system.

As the United Nations (UN) Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression, Irene Khan, explains:

Among the many examples of the use of false information with the purpose of manipulating the population (or certain sectors of it), the alleged impact fake news (selectively or widely disseminated through social media) could have on recent elections is usually cited. More concretely, the 2016 referendum in the UK on whether or not it should continue to be part of the European Union (EU), so-called Brexit, the presidential election in the same year held in the US where Donald Trump was elected, the resulting Facebook/Cambridge Analytica scandal, and the Italian elections in 2018. Even though there is no consensus that the impact of fake news on these processes has been significant, the tight results obtained suggested that it was decisive.Footnote ²

The use of fake news for political and electoral purposes was not the only preoccupying reality. Soon other uses appeared and turned the phenomenon of disinformation into a much more worrying matter, as can be recognised in the context of the COVID-19 pandemic, during which a large amount of false information circulated – even without the objective of manipulating the population – with mournful results.

In this regard, Khan held:

These misconducts by high-ranking public officials, deputies, and even presidents were also verified in many cases in Latin America.Footnote ⁴ This worrying reality has been favoured particularly by the appearance, within the framework of Web 2.0, of social media (websites and computer programs that allow people to communicate and share information on the internet using a computer or a mobile phone), where the use of social networks and of deepfakes emerged as the preferred and most effective way of communicating fake news to people likely to believe them.Footnote ⁵ But the phenomenon of fake news is not only due to the technological factor but to other epistemological, economic, affective, and political factors.

As Melo points out when considering those factors:

In this context, owing to the importance of public and private interests at stake, neither states and the international community nor the mass media and other social media actors (websites and applications that enable users to create and share content or to participate in social networking) have remained static, and have gradually come to face the need to set limits on freedom of expression in order to prevent or even to reduce the pernicious effects or spread of false information. As Warren and Brandeis pointed out, when new inventions converged at the end of the nineteenth century (the telegraph, the telephone, photography, and the rotary press) and competition between newspapers became more acute (causing the appearance of ‘yellow journalism’ or ‘infotainment’, as it is called today):

In the most relevant declarations and constitutions of the late nineteenth century, freedom of expression has been recognised as a preferred freedom, and as a consequence of this, even false or erroneous speech has been protected. However, following the ideas of Warren and Brandeis,Footnote ⁸ this freedom to inform – originally conceived almost absolutely – began to be subject to restrictions. Mechanisms such as the right of reply or the right to be forgotten in its pre-internet version (conceived not as a right to de-index content but as a right to be compensated for the consequences of publishing old information without current relevance that unjustifiably harms a person) are two clear examples. More recently, the new factual configurations claim new responses from governments, the international community, and the private sector, especially from information society services, who have a special responsibility when providing them. This need was seen much more clearly in the last decade, when false information was disseminated as knowingly false to alter electoral processes or influence in the context of the war between Russia and Ukraine and even in the COVID-19 pandemic scenario. As was clearly stated:

The main problem in adopting those concrete and multiple measures in the more recent and complex context to combat disinformation in general and fake news in particular, is how to evaluate the veracity of information and which kinds of measures, adopted both from public and private sectors, could be compatible with democracy and human rights.

The purpose of this chapter is precisely to analyse in general terms the phenomenon of disinformation in the new technological context (which added to its already wide field of action of traditional media the vaster field of the internet and social networks), and more particularly of fake news, which has become even more popular in recent years as a weapon to achieve political or economic gains, leading to the ‘disinfodemic’ and to interesting and relevant conflicts between freedom of expression and various individual and collective rights (e.g., access to information, privacy, data protection).

There are, of course, other serious issues that have been generated and will continue to be generated by the new information technologies that are closely linked to the problem of false news, which we will not discuss here for reasons of space, but which must at least be mentioned: dangerous speech (including hate speech, which can be combated by the same tools that we will describe here); the effects on the rights of a specific person caused by the permanence on the internet of negative information that is true and that, although it was lawfully published at the time, has lost relevance owing to the passage of time (which finds a partial remedy in the right to be forgotten), and the appearance of cancel or call out culture, which condemns those who are deemed to have acted or spoken in an unacceptable manner aimed to ostracise, boycott, or shun (situations for which the right to be forgotten is clearly insufficient and requires other forms of mitigation).Footnote ¹⁰

11.2 Real and False News: Information, Disinformation, Misinformation, Malinformation, Propaganda, Fake, and Fabricated News

According to the Cambridge dictionary, information means ‘facts about a situation, person, event, etc.’ (UK), ‘news, facts or acknowledge’ (USA) and in business lexicon ‘facts or details about a person, company, product, etc.’.Footnote ¹¹

In the same dictionary, disinformation is ‘false information spread in order to deceive people’;Footnote ¹² misinformation is ‘wrong information, or the fact that people are misinformed’;Footnote ¹³ fake news is ‘false stories that appear to be news, spread on the Internet or using other media, usually created to influence political views or as a joke’;Footnote ¹⁴ and propaganda is defined as ‘information, ideas, opinions, or images, often only giving one part of an argument, that are broadcast, published, or in some other way spread with the intention of influencing people’s opinions’.Footnote ¹⁵

Malinformation and real, false, and fabricated news are not defined in the dictionary. Beyond its association with the concept of news, the meaning of ‘information’ is not exempt from debate because some authors think that ‘information encapsulates truth, and hence that false information fails to qualify as information at all’, and explain that the distinction between ‘information as true, and misinformation and disinformation as false, collapses due to the possibility of true misinformation and true disinformation’.Footnote ¹⁶

In short, the defined terms are few and, furthermore, these definitions do not have unanimous agreement in the academic field, so we will address them here in order to draw on such other sources.

11.3 Brief Conclusions

As initially stated, fake news is not a new phenomenon, but in the heat of the unthinkable technological changes that have taken place in recent decades (especially in the last two), they have become a huge problem for societies in general and particularly in the functioning of democracies, constituting an enormous challenge both for the international community and for governments, companies, and civil society.

The regulation of fake news became essential for various political reasons and owing to public health, both events that emerged with great force in the last half of the past decade, but regulation is still incomplete and in some cases confusing.

What is not in doubt, beyond the conflicts when dealing with freedom of expression, is the need for the regulation that has existed up to now, whether international, national, or from the private sector, to deepen, with clear and homogeneous guidelines across the globe, good practices, and intelligent initiatives, and in this, international cooperation becomes essential.

The task is not easy, but it is not impossible to carry out either. Let us keep working on it.

12.1 Introduction

The term ‘European imperialism’ in reference to the European Union (EU) legislator’s actions emerged in the context of internet law concerning data protection.Footnote ¹ Europeans have recognised the futility of creating one of the world’s most robust data subject protection systems without ensuring its effectiveness beyond their borders. Consequently, the EU legislator introduced various mechanisms explicitly aimed at ensuring the global applicability of the EU General Data Protection Regulation (GDPR). These mechanisms include controlling the transnational flow of data and extending the GDPR’s reach to any online platform actively seeking to collect personal data from European citizens. As a side effect, European data protection regulations not only impacted the digital lives of millions of Europeans but also indirectly influenced the legal frameworks of numerous countries seeking to maintain e-commerce ties with the EU, primarily the US,Footnote ² notwithstanding the role of China as a third pillar in this game of internet governance,Footnote ³ an analysis of its role and influence would exceed the limits of this research.

Such an evolution was inevitable because, by its very nature, the internet has global reach. Regulating the internet on a regional basis is feasible but will invariably have worldwide implications. Similarly, freedom of expression and its boundaries constitutes another global struggle involving various actors. The American perspective on freedom of expression holds sway on the internet by default, considering that all five ‘GAFAM’ (Google, Amazon, Facebook – now Meta, Apple, Microsoft) companies are headquartered in California. Indeed, the fight for digital sovereignty is, first of all, a fight of states against companies.Footnote ⁴ The EU legislator is then faced with an impossible dilemma – since a regional response in a globalised environment is doomed to fail, should it grant its legislation global reach, falling into a sort of digital imperialism and provoking economic conflict with its partners (mainly, the US), or should it restrict itself to regional scope?

This chapter will first examine the various interventions made by the EU legislator in the field of internet law that have directly affected freedom of expression worldwide (Section 12.2). This examination is crucial not only because it establishes substantial elements of the law regarding legitimate restrictions on freedom of expression, but also because it expressly focuses on the responsibilities of online service providers. Subsequently, it will demonstrate how this perspective clashes with the American philosophy on freedom of expression (Section 12.3). In light of the imminent prospect of a fragmented and divided internet, there is a pressing need to rethink online freedom of expression on an international scale and to seek consensus (Section 12.4).

12.2 The Birth of an EU Internet Law and Its Impact on Online Freedom of Expression

In this section, some examples of recent interventions from the EU legislator or developments from the European courts that have an impact on freedom of expression will be described, with emphasis on their explicit or implicit, and regional or global, reach. More specifically, the following topics will be analysed as legitimate interferences with freedom of expression from the European perspective: the rise of filters in online copyright law (Section 2.1), the question of the reach of injunctions in the context of the protection of reputation (Section 2.2), the innovations inserted by the Digital Service Act (Section 2.3), the measures taken at European level to fight the propagation of hate speech in digital environments (Section 2.4), the right to be forgotten (Section 2.5), and pictures of sexual content (Section 2.6).

12.3 Freedom of Expression versus Freedom of Speech: The False Twins

This section conducts a comparative analysis of the concept of freedom of expression in the EU and the US. As previously discussed, the US serves as the de facto model for information technology (IT) companies regarding freedom of expression, operating under the belief that this notion possesses a consensual and universally accepted definition in all modern democracies. First, the conceptual differences between European freedom of expression and the American freedom of speech will be explored (Section 3.1), and then the practical example of the recent policies adopted by X (formerly Twitter) will be used as a case study to illustrate these divergences (Section 3.2).

12.4 The Need for a Solution through International Consensus

This section examines the consequences of EU interventions into online freedom of expression in the context of divergences with the US on the scope of this human right. The primary concern is the potential risk of a fragmented digital landscape (Section 4.1), which would undermine the internet’s purpose and philosophical basis. Given the significant economic implications of this situation, the World Trade Organization (WTO) is proposed as an adequate forum of discussion for potential attempts at solutions (Section 4.2). Nonetheless, any attempt at solving the divergences on the boundaries and limitations of the human right must involve a reformulation of the concept itself. Online freedom of expression is effective only when its enforcement fully embraces the concept of the digital public space (Section 4.3).

12.5 Conclusion

The term ‘imperialism’ typically carries a historical and negative connotation. However, in the context of this chapter, it does not imply a deliberate intent of hegemony on the part of European institutions. Rather, it is an automatic consequence of safeguarding digital sovereignty in a globalised environment. This sheds light on fundamental conceptual differences regarding the interpretation of freedom of expression. Nevertheless, these differences give rise to political struggles with economic implications, highlighting the need for a solution to bridge these divides.

It must be accepted, then, that the EU institutions indeed endorse a certain amount of ‘digital imperialism’. While the thorny issue of the governance of a globalised network is as old as the internet itself, the quest for the digital sovereignty of the EU is a novel concept that encompasses a will to promote the defence of European values online more actively.Footnote ⁸¹

The argument here is that while this policy is justified from a democratic perspective, it may result in increasing confrontations that could hinder internet growth and potentially lead to network fragmentation. Instead, this chapter suggests that international negotiations with trusted partners, based on shared core elements of the concept of freedom of expression, should be prioritised. Such a solution could find support within the framework of the WTO, necessitating a re-evaluation of online freedom of expression. One of the main potential dangers of this clash between the US and the EU on the legitimate boundaries of freedom of expression may not be the potential fragmentation of the internet, but the occultation of a debate on the need to re-evaluate this human right in a way that enhances its scope in digital public spaces.