Viral Misinformation and the Freedom of Expression: Part II

Editors’ note: this post is part of a series – see here for Part I and Part III.

We have seen in Part I how any restrictions on the freedom of expression to curb the spread of viral misinformation must be used only in compelling circumstances, when the misinformation directly caused significant social harms, and how must be necessary and be as narrowly tailored as possible. In this part I will try to map out the sources of viral misinformation, and then discuss the responsibility of states which promote such misinformation.

Sources of viral misinformation

There are many different types of Covid-19 misinformation on the basis of its content – one monitoring effort lists six categories (stories relating inaccurate information about cures or preventative measures or treatments that make it worse; relating inaccurate information about the nature of the virus; relating conspiracies; relating false diagnosis procedures; relating inaccurate information about emergency responses; relating inaccurate but funny or feel good stories) with more than a hundred individual examples.

But legally more relevant is the nature of the source (and also of any subsequent spreader) of the misinformation. These can logically be grouped into three categories:

• State actors – this category includes state officials at all levels, from central government ministries and intelligence agencies to local authorities, as well as other actors over which the state exercises control (e.g. broadcast media).
• Organized non-state actors – this again can be anything from corporate actors, especially media companies of all kinds, political parties, to armed opposition groups, hacker groups, and various activist organisations or movements. The key distinguishing criteria here are that these actors are organized, which means that they can be both committed to the spread of particular misinformation as a policy and that they can have a large capacity to engage in such activity, and that these actors are not controlled by the state (although they can act in collaboration with the state).
• Individuals acting spontaneously or organically, at least initially not in concert with others.

Various other distinctions can usefully be made. Sources of misinformation can be overt or covert. The source can spread the misinformation sincerely believing that it is actually true, knowing that it is not true, or simply by being indifferent as to its truth value. Similarly, the source can spread the misinformation with the sincere (even if misguided) purpose of contributing to public debate or helping our their friends and family, or can do so with the insidious purpose of manipulating others. The same goes for all subsequent spreaders of the misinformation further down the chain. Sources and spreaders can clearly be of varying degrees of infectivity (think, for example, of the power of a state-owned TV station with a broadcasting monopoly or of a celebrity ‘influencer’ with millions of followers on social media). Sources can act with the primary purpose of influencing their own in-groups (e.g. members of their social circle or religious community or country as a whole) or out-groups (e.g. an intelligence agency of one state spreading misinformation about Covid-19 within the population of some other state).

Thus, depending on the nature and purpose of the source, the dissemination of misinformation exists on a broad spectrum, ranging from purely innocent, spontaneous spread to organized, malicious propaganda or disinformation.

It is important in that regard to reflect on the motives of the sources and spreaders of misinformation. Most of the spread itself is innocent and organic, and occurs within the community – it would not work otherwise. This is simply because most of us would not knowingly infect another person with harmful misinformation any more than we would infect them with the coronavirus. Many of the original sources may also be acting innocently, sincerely believing in the truthfulness of the information whose spread they initiated, and spreading it without any desire to cause harm. But even when the original source is not innocent, most of the spreaders will be: ‘[f]or a lot of people, participation in the process is altruistic.’ The key legal point here is that innocent spread cannot be adequately addressed by criminal sanctions, even if these were a good idea in principle, because of deficits in the culpability of the individuals concerned.

Collaboration and overlap between the categories of sources

It is clear that much of Covid-19-related misinformation arose spontaneously and spread organically in the affected community. This appears to be true for one of the examples with which I started this post – there is no evidence that the false information that led to hundreds of deaths in Iran from alcohol poisoning had any kind of organized source or that the misinformation is being spread maliciously.

But the other example, that of the spread of conspiracy theories on 5G networks and the coronavirus, is particularly instructive in how it demonstrates overlaps between different types of sources and spreaders, acting with different motives. At least initially the 5G misinformation arose and spread organically, in many different variants, by individuals already predisposed to believe in alleged ill-effects of telecommunications equipment on human health, acting over social media or local radio stations.

Then that spread got amplified. Much of the amplification came from presumably honest and sincere (but super-spreading) celebrities with mass followings on social media, such as the actors John Cusack and Woody Harrelson. But at least some of the amplification came from a disinformation operation by organized state or non-state actors using networks of inauthentic social media accounts, as well as from one state-controlled media organization.

Similarly, it is perfectly possible for misinformation to start spreading spontaneously and for the spreaders to gradually organize into a more coherent and potent movement. This is arguably what has happened over the years with anti-vaccination campaigners, who are very systematically expanding and utilising their online presence. Obviously, the dissemination of misinformation by certain older actors, e.g. media organizations, can be extremely impactful even without active collaboration. Consider, for example, how parts of the right-wing media in the United States, like Fox News, have systematically downplayed and minimized the threat of the pandemic, especially in its early stages. This is sometime done in amplificatory symbiosis with political and state actors, sometimes not. The most recent such bout of misinformation, for example, tries to cast doubts on official death tolls, arguing that these are inflated.

State misinformation

By far the most pernicious, however, is misinformation by state actors. Such misinformation inherently attracts more attention, and its impact is inevitably amplified by the media. State misinformation can be especially damaging to the information ecosystem as a whole, destroys the public trust necessary for combatting the pandemic, and is often coupled with direct and indirect forms of censorship. In other words, state actors have the ability to construct and promote entire false narratives by simultaneously spreading misinformation and suppressing accurate information.

State actors can engage in such practices for various different motives. They can do so externally, by conducting disinformation operations in adversary states to thereby sow discontent in these societies. Or, more commonly, they can do so internally, against their own populations. Misinformation can thus be deployed to project a sense of power, authority and competence of the regime, to blame some other actor for the state’s failures in stopping the epidemic or simply as a convenient distraction. This is fundamentally a very old playbook. Obviously, as with any other actor, states can act out of mixed motives, and some of their spread of misinformation might not be knowing or malicious. 

Because managing the coronavirus epidemic requires the population at large to willingly adopt measures such as handwashing and social distancing, any state misinformation that actively minimizes the threat posed by the virus is particularly harmful. This can range from downplaying the virulence or danger of the Covid-19 disease (for example in Brazil), to complete (if implausible) denials that the virus is even present in the country or that is being transmitted in the community (e.g. in Nicaragua and Turkmenistan).

Also particularly harmful is the spread by state agents of misinformation about specific medicines and treatments, e.g. by promoting ineffective or unproven treatments, especially when it is coupled with the suppression of accurate information. Perhaps the most notorious such example is the promotion by the then-President of South Africa, Thabo Mbeki, and by  his government, of the idea that beetroot juice, lemon and garlic were an effective treatment for AIDS, while rejecting the use of antiretroviral medication. That policy according to some studies led to more than 300,000 deaths.

An equivalent during the current pandemic, if on a much smaller scale, is the promotion by President Trump and some other world leaders of the anti-malarial drug hydroxychloroquine as an effective treatment for Covid-19. The issue here is not that the drug has no therapeutic value – it might have some – but that its effectiveness has not been proven in sufficiently rigorous trials. Widespread claims about its purported effectiveness started with a very small (and flawed) French study. These were amplified by various influential private actors on social media  and over mass media, and were then taken up by the US President as a miracle cure (“What do you have to lose? Take it,” Trump said).

The social harm here is not, as in the South African AIDS example above, in the displacement through misinformation of some existing, proven effective treatment for Covid-19 (as there is none). Rather, on the one hand, hydroxychloroquine is an effective treatment for other conditions (not just malaria, but also rheumatoid arthritis and lupus) and bulk buying of the drug for ineffective Covid-19 use may lead to shortages for those patients who actually need it for those other conditions. On the other hand, hydroxychloroquine also has serious potential side effects, including sudden cardiac death and damage to eyesight. People do, in fact, have something to lose by taking it, as confirmed by official reports from French hospitals using the drug empirically. (As a side note, at least one person died after self-medicating with chloroquine phosphate, a different substance used for the cleaning of fish tanks, after watching President Trump promote hydroxychloroquine on TV).

State misinformation as a violation of human rights

How, then, is state misinformation to be analysed from the standpoint of human rights law? Its characterization will depend primarily on the nature and magnitude of the social harms that it causes, on the directness of the causal nexus between the speech and the harm, and on the purpose of the relevant state agents who engaged in the spread of misinformation. That analysis will necessarily be highly contextual, but two broad conclusions can nonetheless be made.

First, state agents who systematically pollute the information space through the dissemination of falsehoods may actually be denying individuals’ right to seek and receive information, which is part and parcel of the freedom of expression as protected by international human rights law. This is because state action will make it difficult, if not impossible, for individuals to have access to accurate information which they are seeking. This will especially be the case (as it often is) when states spread misinformation and simultaneously actively suppress accurate information, through direct or indirect forms of censorship.

Second, state agents who spread misinformation that directly causes harms to health or expose individuals to significantly elevated risks to their health, do so in violation of state obligations to respect and protect the right to life and the right to health, as guaranteed by international human rights instruments. There is no doubt that the right to life extends to ‘general conditions in society that may give rise to direct threats to life … [including] the prevalence of life threatening diseases.’ (Human Rights Committee, General Comment No 36, para 26). There is also no doubt that in order to respect the right to health states should refrain ‘from censoring, withholding or intentionally misrepresenting health-related information’ and that as part of their core obligations states must ‘take measures to prevent, treat and control epidemic and endemic diseases’ and ‘provide education and access to information concerning the main health problems in the community.’ (Committee on Economic, Social and Cultural Rights, General Comment No 14, paras 34 and 44). The Committee is explicit, and rightly so, that the ‘deliberate withholding or misrepresentation of information vital to health protection or treatment’ is a violation of the duty to respect the right to health (ibid, para 50, emphasis added).

In sum, state agents have a negative duty under human rights law to refrain from spreading misinformation that causes harms to human health. Such a duty will clearly apply if the misinformation is being spread knowingly or deliberately. It will, moreover, in my view also apply if the spread is external, i.e. if a state is spreading Covid-19-related misinformation in other states and causes extraterritorial harms to their populations. Finally, that negative duty is coupled with the positive obligation to promote accurate information, because such information is indispensable for effectively combatting the pandemic.