Introduction

COVID-19 has proved to be a fertile ground for the spread of disinformation. Clear examples are the references to COVID-19 as a bioweapon created by China or the United States, the denial that the virus even existed (idea of it being a hoax), or that it was a plot from the pharma industry to profit from a health crisis. The rise of so-called fake news as ‘social media rumors, online political disinformation or state-sponsored internet propaganda’ (International Press Institute, 2020) coincided with the emergence of social and political movements COVID-19 containment measures. These included protests against quarantine periods or the use of masks or social distancing (Wolff & Ladi, 2020, see De Saint Phalle contribution). These movements further fuelled the spread of misinformation, by consuming such content online, but also by helping to promote it and by sharing it within their networks. Government officials from several countries, most notably the former president of the United States of America, Donald Trump, also contributed to downplay the role of the pandemic but also shared many of the claims based on misinformation, namely the statement that people could try injecting bleach or disinfectant to fight COVID-19.Footnote 1

Such statements, whether made by government officials or people from civil society, became controversial because of their role in challenging public action that could prevent the spread of the disease. But they also resonated strongly with claims about the extent to which an existential threat and its consequences, namely the distrust and fear that fake news generates, can legitimise political action against freedom of expression or press freedom (Daemen, 2022; Goh et al., 2022; Palmer, 2022; Committee to Protect Journalists, n.d.), a topic to which we will now turn.

Restricting Freedom in the Name of Public Health

COVID-19 placed strains and difficult challenges on national governments and international cooperation. In a globalised world, where economies are highly interdependent and there is a significant daily movement of goods and people, the pandemic highlighted the need for international cooperation to close borders and pool resources to find the pandemic, but also for political and scientific cooperation to produce treatments and a vaccine that could halt the global spread of the disease. But these challenges were accompanied by equally difficult ones, namely the legitimacy of governments to implement measures to contain the development of COVID-19, but also the willingness of citizens to accept and comply with such measures. On the one hand, governments were faced with what became obvious trade-offs between public health and the spread of the disease, and avoiding economic disaster, but also mitigating social challenges, such as ensuring children’s education, avoiding exacerbation of domestic abuse and violence, and monitoring mental health indicators, among others, until a vaccine was in circulation and a certain segment of the population (nationally and globally) was inoculated. On the other hand, governments were faced with the need to justify such measures—legally and politically to other representatives and heads of state, but also to their constituents. What emerged from these challenges has been discussed as a set of issues that have implications for discussions of the legitimacy of governments, the effectiveness of international cooperation, or the interplay between scientific knowledge and political agency.

But it has also led to discussions about freedom, especially freedom of expression. It is noteworthy that the movements that have emerged against national and global measures to combat the pandemic—whether the vaccination mandates or the vaccine imposed by the state (see Egger and Magni-Berton in this volume)—have often used concerns and freedom-based arguments to make their case. A case in point is the Portuguese ‘Médicos Pela Verdade’, a movement founded by health professionals against the measures taken by the Portuguese government, namely the quarantines imposed or the mandates for the use of masks. Since its creation, the right to express their demands, on social media, but also to their patients or in the media, has been an integral part of their action. Their manifesto, in which they claim to be moving away from public eyes (but hint at going ‘underground’), illustrates the extent to which freedom is part of the legitimisation of their actions:

Our will and determination to spread awareness of science and present alternatives is alive and will continue, despite the centralised, reactionary and repressive environment in which we currently live, and which has been growing since we were founded as an organisation. We will suspend our online page and website until we are all free again.

This is for everyone, even those who do not want to be free or do not know how to be free. Because freedom is like air, you breathe it; because thought is free, and the verb is the most important weapon you have.

Freedom of expression can be defined as a fundamental right "that contain both a personal and a social dimension. They are considered “indispensable conditions for the full development of the person”, “essential for any society” (...) Without free speech, the enjoyment of other rights is not possible.” (Howie, 2018, pp. 12–13).

It is, therefore, worth looking at how restrictions on freedom have been the result of measures taken to mitigate the effects of the COVID-19 pandemic, before analysing empirical data to identify any trend in this respect, and concluding whether the pandemic legitimises restrictions on freedom of expression and whether this can hinder liberal democracies. We can distinguish between ‘direct and indirect measures’ that affect freedom of expression.

Direct action is linked to what became the COVID-19 messaging campaign. In many ways, governments were faced with the task of communicating the strict measures they wanted to enforce, which meant convincing citizens of the public health risks of not adopting such measures. It was an effort to balance scientific evidence and messaging with rhetoric and emotional communication. Particularly in the early stages of the pandemic, this meant communicating the threat of COVID-19 in a way that avoided panic, but also either instilling a degree of fear (or related emotions) or appealing to scientific rationality to contribute to public acceptance of the measures being taken. However, as noted above, the emergence of the pandemic was accompanied by a simultaneous rise in misinformation campaigns, fuelled by social groups, individuals on the Internet and even government officials. This led to calls for action to curb the spread of misinformation, with health authorities calling on governments and social media companies to limit the opportunities for fake news. These measures are what Engler and her co-authors call ‘public sphere restrictions’, namely by ‘weakening the media’ through the ‘introduction of so-called fake news laws, restricting media coverage of COVID-19 and government responses to it’ (Engler et al., 2021, pp. 1080–81). For example, a study by the International Press Institute (IPI) in 2020 found that at least 17 states have enacted regulations to control disinformation during the pandemic. While these may include small measures, such as imposing fines on those who spread unsourced and potentially dangerous information, in some countries, governments have passed significantly restrictive legislation on the issue.

It is worth noting that efforts to combat misinformation have also been taken up by other actors. The European Union, through the European External Action Service (EEAS), has begun to combat the various disinformation campaigns. Social media companies such as Facebook, Instagram and Twitter are also adjusting their apps and services, highlighting warnings and blocking the promotion of misinformation. In many ways, a triangulation between public, private and supranational spheres helped to mitigate the impact of misinformation during the COVID-19 waves (Amadio Viceré & Tercovich, 2020; Council of Europe, 2021).

In addition to what we have termed ‘direct measures’, that is, those relating to the management of pandemic-related information, there were also ‘indirect measures’ that affected freedom of expression. These included the quarantine measures, especially the social distancing measures, which made protests and rallies illegal for a period of time, or even the closure of parliaments, which reduced opportunities for scrutiny of government action (see Poyet in this volume). Such measures effectively shrank the public arena, limiting opportunities for debate and the expression of dissent. As such, they largely contributed to the fact that much of the debate took place either in private forums (i.e. among friends, families) or on social media.

In what follows, we will analyse the data from the EXCEPTIUS codebook, which examined the impact of COVID-19 measures on several variables, namely freedom of expression (through freedom of speech and the press), before discussing the significance of such results for the discussion of freedom and democracy.

Insights from EXCEPTIUS Data on Freedoms of Expression and Press

The present comparative analysis is based on EXCEPTIUS data focusing on the first, initial wave of the pandemic, from January to June 2020.Footnote 2 The analysis of this data is complemented with additional documentation on the 22 countries studied.

From the 22 European countries included, the results show that Romania is the only country that enforced restrictions on freedom of speech through ‘Decret no. 195’ in article 54, which states that public and private actors must contribute to the spread of information about protective measures against COVID-19, whilst containing the spread of fake news. To achieve that, public authorities should act and inform people with the correct information. Hosting service providers and content providers are obliged to immediately interrupt the transmission of information to their users if the respective content promotes false information regarding COVID-19 and jeopardizes the application of preventive measures. Moreover, the ‘ORDONANȚĂ DE URGENȚĂ no. 28’, foresees, in its article 326, sanctions for those who spread false information, resulting in ‘imprisonment from 6 months to 2 years or with a fine’.

It is worth pointing out that while EXEPTIUS only mentions one country that has taken direct measures to restrict freedom of expression, media coverage and international reports have exposed policies that appear to have had the same effect in countries other than Romania.

In Hungary, since March 2020, the government of Prime Minister Viktor Orbán has adopted measures that criminalise the dissemination of fake news. In addition, journalists faced fines and even prison sentences of up to five years and were targeted for coercion through unlawful denial of access to medical personnel (testimonial sources) and places (in this case, hospitals). Poland supported pro-government media outlets and restricted organisations as foreign media, while at the same time the government and state-owned companies brought cases against journalists. In addition, the Bulgarian government implemented a measure stating that any ‘fake news’ related to the pandemic would be punishable by imprisonment (Council on Foreign Relations, 2021).

In addition, efforts to combat misinformation have also been taken up by other actors. Although this was not the focus of the EXCEPTIUS study, including them allows us to better understand the scope of restrictions that have been implemented so far. The European Union, through the European External Action Service (EEAS), took steps to fight the various disinformation campaigns. Social media companies such as Facebook, Instagram and Twitter have also adjusted their apps and services to highlight warnings and block the promotion of misinformation. In many ways, a triangulation between public, private and supranational spheres helped to mitigate the impact of misinformation during the COVID-19 waves (Amadio Viceré & Tercovich, 2020; Council of Europe, 2021).

In summary, there is evidence that at least Romania introduced restrictions on freedom of speech and the press in the early stages of the pandemic, while other countries, such as Poland and Hungary, also appear to have introduced such restrictions at the beginning of the pandemic, although these were not systematically recorded by the EXCEPTIUS study. This small set of countries justify the claim that governments were cautious in implementing restrictions on freedom of expression during the early stages of the pandemic. Before moving on to debating the legitimacy of such measures, it is worth considering possible explanations for these findings.

One possible reason for these results stem from the difficulties most European countries face when it comes to restricting freedom of the press and freedom of speech. Since the entry into force of the Lisbon Treaty, which incorporated the Charter of Fundamental Rights of the European Union into European law, every EU state is legally obliged to respect freedom of expression in its laws and in the exercise of its powers and competences. Alongside national constitutions, which in themselves often oblige states to protect freedom of expression, EU law has made such an obligation more binding, with sanctions for states that choose not to comply. Thus, constitutional and EU protection of freedom of expression may explain why, at the onset of the pandemic, 21 of the 22 countries analysed in the study appear not to have adopted any restrictions on freedom of expression, despite the surge in misinformation, notably through fake news.

However, there may be a second reason, which also helps to explain why such results appear to have been observed at the start of the pandemic. Degerman and his co-authors (2020) discuss the ways in which the COVID-19 pandemic illustrates the role of fear in the juggling of scientific and political authority and decision-making. Countering Nussbaum’s assertion that fear is an irrational emotion, they instead discuss how, in the early stages of the pandemic, a lack of fear on the part of government officials was detrimental in enabling governments to make decisions that would benefit the public, but also promote greater citizen engagement with the measures that the government might need to implement. They highlight the cases of Sweden and the UK to argue that the lack of fear characterising such governments led to a delayed response to the virus in March 2020, but also ‘contributed to public complacency about the outbreak’. Thus, they argue, “the apparent issue was not that government and citizens acted in fear instead of waiting for scientific evidence; it was that they did not experience sufficient fear in the face of the available evidence” (Degerman et al., 2020, p. 6).

We can certainly remember the early stages of COVID-19, when most governments urged against panic and often downplayed the potential consequences of the pandemic, not only as a response to public fear (expressed in forms of panic buying), but also as a way of trying to outweigh the potential economic consequences of closing borders, schools or businesses. To the extent that this is true, it may also help explain the results we discussed above: the health threat of COVID-19 took some time to sink in, and consequently the phenomenon of misinformation, or at least the potential threat of misinformation, and the political debate about whether or not fake news should be restricted in some way.

However, while the data do not suggest massive legal restrictions directly targeting freedom of speech or the press, it is possible to consider other measures that have had an indirect impact on these freedoms, namely legal restrictions implemented since the onset of the pandemic that have reduced the size of the public arena and thus reduced the opportunities for public speech and debate, which we have termed ‘indirect measures’. The shrinking of the public arena helped to further increase the relevance of social media. Combined with the absence of restrictions on freedom of speech and the press, namely the lack of a regulatory framework to monitor and/or restrict information online, fake news has flourished. Moreover, early evidence suggests that while such direct measures were not implemented in the early stages of the pandemic, they became more prevalent as the second and third waves emerged, and misinformation was also more damaging to governments’ efforts at social distancing and masking mandates, as well as the implementation of vaccination plans.

In conclusion, this movement cannot be understood as a constant or fixed event, because since July 2020, most European countries have applied restrictions on the press and speech or referred to fake news as ‘infodemics’ (Borrell, 2020). In what follows, we will discuss whether we can look at such direct and indirect measures and assess their legitimacy.

COVID-19 Misinformation, Freedom of Speech and Democracy

In what follows, we will discuss some normative arguments connecting the observed restrictions with conceptions about freedom of speech and press, and in general, liberal democracies. We start with the main insights provided by EXCEPTIUS but also consider, March 2020, countries implemented indirect measures, which impacted freedom of expression.

Let’s start by considering the position that freedom of expression should be limited or restricted to limit misinformation about COVID-19.

Such a position can be justified on several grounds. One can argue that not restricting instances of misinformation can contribute to instances that can promote harm. If one considers the former president of the United States, Donald Trump, and his claims that people should try drinking bleach to cure COVID-19, it is not difficult to conclude that such a statement, on a major media outlet, by someone with political authority may lead many people to drink bleach and risk their lives. If this were the case, then even a negative conception of freedom, such as John Stuart Mill’s one, based on non-interference and respect for basic liberties (2001), could accept such a restriction on such speech and conduct, and even could condemn the author of such discourse. Mill’s harm principle would be applied, and restrictions on freedom of speech would be deemed legitimate in these cases.

However, it is unclear whether such an exception based on the harm principle would be sufficient to maintain restrictions on misinformation in the context of COVID-19. In fact, the harm principle might be too limited and leave a lot of misinformation unchecked. Take, for example, all the conspiracy theories surrounding COVID-19. There has been a lot of misinformation about the source and origin of the virus. Particularly in the early stages of the pandemic, many argued that COVID-19 was the ‘Chinese’ or ‘Wuhan’ virus, which helped to fuel prejudice and hatred against Asian minorities, especially in places where anti-Asian racism was already entrenched, such as the United States (Degerman et al., 2020, p. 11). Such narratives do not necessarily lead to physical harm in Mill’s sense, but they can contribute to increased hatred and fear of certain minorities, even to violence, or certainly to social condemnation and exclusion of these groups.

Justifying restrictions on freedom of expression on the basis of a thin rule, such as the harm principle, can also leave unaddressed cases of misinformation that could have dire consequences for public health but were unlikely to be easy to appeal to the harm principle. Let’s take the mask mandates that have been implemented around the world since March 2020. Misinformation has often targeted such mandates, arguing that masks are not effective, or that they may even endanger lung health and be more harmful than beneficial. Not using masks is mostly a collective issue: you may be sick, and by not covering your face you may be contributing to the spread of the virus. While it might protect us individually, it is above all a protection for other people. But not all COVID-19 was fatal or caused significant physical harm, and it would be difficult to say with certainty that those who chose not to wear masks on one occasion caused direct physical harm to others. Therefore, if one chooses to use the harm principle only to limit misinformation, such narratives are likely to be accepted.

We could take a different reading of freedom. John Stuart Mill’s view of freedom of speech is based on a negative conception of freedom. Isaiah Berlin’s essay on freedom famously distinguished between positive and negative freedom: the latter is “simply the area within which a man can act unobstructed by others” (Berlin, 1969, p. 16). Negative freedom is, therefore, about avoiding interference. Positive freedom, on the other hand, is about autonomy: “I wish to be a subject, not an object; to be moved by reasons, by conscious purposes, which are my own, not by causes which affect me, as it were, from outside” (Berlin, 2014 [original publication 1969], p. 231). Positive freedom is, therefore, associated with the notion of freedom to do what we want, what is best for us, that is, self-fulfilment and development, while negative freedom is about freedom from coercion and interference. Mill’s conception is based on a notion of basic liberties, namely a defence of individual freedom of conscience, expression, association and occupation as something that should not be interfered with by anyone or by the state, except where the harm principle could be applied. But one could consider other readings of freedom that might help to justify restrictions on basic liberties, namely freedom of expression. Daemen discusses how a republican reading of freedom as non-domination might justify restrictions on freedom of expression that, while restricting people’s choices, do not necessarily constitute unlawful interference. Indeed, she argues that if such measures are implemented against a background where “checks on state power are in place and policies are the product of democratic decision-making, they do not entail domination” (Daemen, 2022, p. 11), then one could legitimise such restrictions even through a negative conception of freedom. However, it would be necessary to ensure that restrictions on fake news or public speech that spreads misinformation are debated and subjected to a rigorous democratic decision-making process to be deemed acceptable. In many cases where such measures have been adopted, these processes have not taken place, either because parliaments have been constrained in their activities, or because governments have seen the pandemic as an opportunity to implement these restrictions—on speech or on the press—an issue we will discuss below. Regardless, it is worth pointing out that while it is not easy to justify restricting freedom of expression, there are ways of doing so, even when appealing to a negative conception of freedom.

This leads us to our second position, which argues that freedom of expression should not be limited or restricted, despite misinformation about COVID-19.

Let’s start by considering one of the more familiar arguments used by those who want to restrict fake news. They have often said that the problem with such misinformation is that it is ‘fake’. The reason why we should restrict it is because it spreads information that is factually untrue, thus justifying its restriction. However, such a justification seems less robust to us, and can become an argument in favour of no restrictions. Similar to John Stuart Mill, Thomas Scanlon in “Theory of Freedom of Expression” (2000) argues for a “Millian Principle” based firmly on respect for the autonomy of the individual. Thus, the fact that a particular message conveys falsehoods that lead others to believe them to be true could not be a justification for imposing legal restrictions on it. To do so would be to restrict our freedom to make judgements about what we believe and want to do.

For Mill, the same is true, because being able to form an independent judgement is essential for self-development and flourishing. Individuality can only be realised through the nurturing of the capacities for reasoning, and the pursuit of truth. If doing so demands open debate in order for truth to be able to make “such head as to withstand all subsequent attempts to suppress it” (Mill, 2001, p. 29), one needs open debate and deliberation, where all conceptions, including those considered false or absurd, in order for truth to shine light to their falsehood or illogical nature.

Another argument in favour of unrestricted freedom of expression is based on appeals to the health of democracy. It is not only the pursuit of truth that can benefit from unfettered debate, but also the health of democracy’s deliberative process. Henry Voigt, for example, has argued that political demonstrations and impassioned debates on social media about government decisions may “actually enlarge the scope of politics in these societies. Such incidences make the breadth of democratic alternatives evident, however weird or non-scientific the protest opinions may occasionally be” (Voigt, 2021, p. 552; Daemen, 2022, p. 10).Footnote 3

Finally, a further justification for non-intervention is the possibility of democratic backsliding or the promotion of an ‘authoritarian slip’. Degerman et al. discuss Hannah Arendt’s conception of fear as an “‘anti-political’ principle that destroys the possibility of political action by citizens, which for Arendt was synonymous with freedom” (2020, p. 13). Implementing restrictions on fundamental freedoms based on fear of an existential threat, such as COVID-19, could demobilise citizens and promote the opposite of political action, thus paving the way for totalitarian state solutions (idem). Furthermore, the securitisation of COVID-19 and the tight grip of governments like China, whose authoritarian nature has allowed them to impose tight restrictions on social distancing or mask mandates, has also been portrayed as successful—considering the metric of security—and as such may further lead to the legitimisation of more stringent measures, many of them illiberal in nature, by presenting and arguing for a trade-off between freedom and security. As Voigt points out: “This could also lead us to believe that the narrative of the efficiency with which authoritarian societies have been able to control the pandemic—and not just the pandemic, but many other problems as well—is the right and desirable one” (2021, p. 552), thus contributing to the legitimisation of authoritarian government solutions.

And while such a threat exists in liberal democracies, it is even more prevalent in contexts where democracy is already ‘backsliding’. While countries like Bangladesh, China and Egypt have laws that initially only targeted fake news, their broad application can be manipulated to develop a unique censorship of critical reporting. For illiberal leaders, such crises as health threats or ‘infodemics’ present a golden opportunity to apply laws without scrutiny and expand their arsenal of political tools. A Human Rights Watch study (2021) of 83 countries that used COVID-19 to justify controlling the exercise of freedom of expression showed that authorities used such manipulation to prosecute and criminalise journalists, activists and protesters. The authorities also closed media outlets and, as noted above, introduced vague laws to criminalise speech that could threaten public health. The survey highlighted, for example, the role of Bangladesh, China and Egypt, where people were imprisoned simply for criticising the government’s actions against COVID-19. Furthermore, 24 countries “have enacted vague laws and measures that criminalize spreading alleged misinformation or other coverage of COVID-19”, while 51 countries ‘have used laws to prevent COVID-19, as well as counterterrorism “to arrest, detain, and prosecute critics of government response to the coronavirus” (Human Rights Watch, 2021). Finally, the study also highlighted how measures to restrict freedom of expression were common, but only 44 of the 83 countries surveyed declared a state of emergency, which usually includes restrictions on the spread of the virus (i.e. social distancing). Thus, creating a sense of constant emergency and threat from the pandemic or misinformation can become a “permanent mechanism of freedom-curtailing domination” and thus a threat to democracies.

The above arguments highlight how limiting restrictions on freedom of expression can be a necessary condition for uncovering the trust and ensuring democracy’s deliberative process (let’s consider them argument 1.1 and 1.2), but also to avoid democratic backsliding (argument 2). In what follows, we will consider a range of objections to all three.

Let’s consider arguments 1.1 and 1.2: While Mill’s point about the importance of uncovering truth through open debate is worth considering, the unrestrained nature of misinformation does not seem to have contributed to a more salutary democratic debate or the uncovering of truth. Instead, fake news has often been found to have nothing to do with an intention to pursue or uncover a particular truth. As Brown points out: “In some cases, originators of fake news diffuse articles which contain claims they know to be false or misleading. In other cases, they simply diffuse claims they have no good reason to believe are accurate because such claims are not the product of a process that reliably leads to truth” (Brown, 2023, p. 4).

Thus, it is not a commitment to truth that tends to be behind the spread of misinformation. Moreover, beyond the motivation of those who spread misinformation, fake news seems to undermine trust in the media. Brown summarises a recent Pew Research Centre survey, which shows that “68% of Americans get their news from social media (primarily on Facebook), but over half of them (57%) expect this news to be inaccurate given the spread of misinformation. Moreover, the percentage of Americans who reported reducing the amount of news they get overall in response to made-up news varies between 31% (for highly politically aware respondents) and 50% (for less politically aware respondents)” (Brown, 2023). Thus, fake news undermines individuals’ trust in the media, contributes to a reduction in the amount of news consumed, and thus reduces the amount of information individuals use to make independent judgements. Therefore, not only does it undermine trust in the press, but more fundamentally, an open debate in which fake news is common does not lead to more information being consumed and used for decision-making, but quite the opposite.

Second, fake news may also jeopardise our ability to make accurate, independent judgements. Fake news not only leads us to mistake falsehoods for truths, but also truths for falsehoods. Our picture or reality can thus be manipulated by misinformation, making it difficult to make decisions and judgements (Brown, 2023), thus violating autonomy and undermining the argument for unrestricted information based on Scanlon’s ‘Millian principle’. For example, we might receive information based on a candidate’s values or past record, which is not true, and be led to vote for them in a local election. After winning, we might realise that we have supported a candidate who does not reflect our political preferences or political values.

Finally, let’s consider argument 2 about the possibility of restrictions leading to democratic backsliding. As noted above, it seems to be true that some governments have used COVID-19 as an opportunity to introduce restrictions, particularly on freedom of speech and the press, which have helped to strengthen their discretionary power. But while this has been the case, the political responses of ‘democratic systems’ have been extremely varied. As Engler et al. point out, “political institutions constrain policy choices in times of crisis: previous decisions on the design of political institutions create legacy effects by decreasing the decision space for political actors. Countries that are more dedicated to individual liberties continue to do so even in moments of emergency” (Engler et al., 2021, p. 1095). Their findings, therefore, provide a compelling case for considering the extent to which political responses were not so much a direct result of the fear and demands of the pandemic, but rather a response constrained by political institutions. As such, the case for the role of restrictions during COVID-19 in leading to democratic backsliding may be too hasty. While restrictions may have opened this possibility in countries with authoritarian tendencies, less robust democratic political institutions and that had already experienced democratic backsliding, the opposite is true in other cases. In fact, what the EXCEPTIUS results seem to indicate is that constitutional principles seem to have worked in most European countries, at least in the early stages of the pandemic. Not only “executive overreach has generally been avoided” but “mechanisms of legal oversight have, by and large, proved sufficiently strong” (Voigt, 2021, p. 552).

Conclusion

This chapter aims to contribute to existing debates on COVID-19 restrictions on freedom of speech and expression and their impact on democracy. Using EXCEPTIUS data for the first pandemic wave as well as other relevant sources, we found that, with three exceptions, European countries have not introduced laws directly restricting freedom of speech and the press. We argued that this may be a result of constitutional principles protecting basic liberties, namely freedom of expression, or simply a reflection of the time frame of the study. If the latter is the case, new waves of EXCEPTIUS data may lead us to different conclusions.

Given these results, but also the likelihood that later waves of data would show more common cases of restrictions on freedom of expression in several countries, we wondered in what way these could be considered legitimate. Furthermore, we considered not only laws that directly restrict freedom of expression, namely the press (i.e. laws restricting fake news), but also those that indirectly contribute to shrinking the space for debate (i.e. laws on social distancing).

Overall, the case for not restricting freedom of expression, especially through direct measures, seems strong. Open debate not only promotes individual deliberative capacity, but also protects autonomy by ensuring that individuals are able to form independent judgements about the pandemic and the government’s measures to contain the spread of the virus. This is Scanlon’s ‘Millian principle’, inspired by the work of John Stuart Mill. It also contributes to democratic decision-making and deliberation, and, as Mill pointed out, it is essential to ensure that trust is not only exposed but also seen as robust, withstanding open confrontation with falsehoods.

However, we would like to conclude by asserting that, while one should be sensitive to the arguments about the relevance of an open debate, for the most part the COVID-19 pandemic and the tandem restrictions on freedom of expression, namely those restricting fake news, seem to be justified to some extent, particularly in contexts where strong constitutional principles still protect press freedom and limit what laws can be applied and to whom. For one thing, in such contexts it does not appear that governments have used the pandemic to increase their executive power. Concerns about democratic backsliding do not seem to apply in such cases. Moreover, arguments about the importance of not limiting misinformation for the integrity and preservation of deliberative decision-making seem to be unfounded. As fake news undermines trust in the media, thereby reducing news consumption, but also distorts our perception of reality by turning truth into falsehood and falsehood into truth, it becomes difficult to make independent judgements. Finally, we should consider whether private companies, namely those in charge of social media, should be in charge of deciding who sees what and when. While we consider whether governments should restrict fake news and misinformation, private companies already significantly restrict the information we receive, often contributing to the spread of tailored misinformation and the omission of counterfactuals. Therefore, while not all restrictions need to be justified, and in certain contexts any restriction may prove to be another step towards authoritarian government solutions, it is worth considering some mechanisms that could be the outcome of a democratic decision-making process and contribute to limiting the spread of fake news and misinformation.