11 April 2020

Fighting Fake News or Fighting Inconvenient Truths?

On the Amended Hungarian Crime of Scaremongering

Last week, the Hungarian Parliament adopted the Defence Against the Coronavirus Act (Act Nr. 12 of 2020, henceforth Coronavirus Act) which authorizes the government to rule by decree without jurisdictional or temporal limits (for details, see here and here). The Act also amended the Criminal Code: it created the new crime of “obstructing epidemic prevention” and amended the already existing crime of scaremongering (rémhírterjesztés).

Critics, both inside and outside Hungary expressed fear that the crime of scaremongering in its amended form can be abused to stifle legitimate criticism of the government and help to cover up inconvenient truths. The government claims that the amendment brings necessary clarifications to a crime that had been left vague all too long. This position also found its support, some more qualified than others, among academics.

It is certainly right to point out that, as opposed to what is reported widely, the crime itself, including some of its now controversial elements, is not created by the amendment but has been in force for decades. There are also convincing arguments for the view that in its old form, it did not encompass the spreading of fake news during an epidemic and thus an amendment was therefore warranted. Yet the now adopted modification addresses none of the previously existing problems and makes the crime more susceptible to abuse by the authorities.

Scaremongering: For Long, an Ambiguous Crime of Complete Practical Irrelevance

The definition of the crime of scaremongering which the Coronavirus Act now amends was adopted in 2001, after the Constitutional Court had declared the previous version unconstitutional (Section 270 of the – old – Criminal Code, Act Nr. 4 of 1978). The new Criminal Code left this unchanged (Section 337, Act Nr. 100 of 2012, henceforth the Criminal Code). According to this pre-amendment formulation, scaremongering constituted the act of

“claiming or spreading a falsehood or claiming or spreading a distorted fact before a large public, which is suitable for alarming or agitating a large group of people at the site of a public emergency.”

There are concerns with this formulation. First, the definition is, arguably, rather ambiguous at places, which means that the crime might breach the nullum crimen sine lege requirement enshrined the Constitution (Art XXVIII. (4)). The second, more specific concern is whether this crime, due to its choice of specific terms, can be applied to stop the spread of fake news during an epidemic. There are convincing arguments that it is not.

The first of the problems is the term “at the site of a public emergency”. Public emergency (közveszély) in Hungarian criminal law generally means a factual (existing) situation where the physical integrity of a large number of persons, and/or things of considerable value might be endangered (BH 1988. 304). The formulation “at the site” leads us to conclude that the public emergency must be some sort of geographically and temporally localized phenomenon, and that the spreading or claiming of falsehoods or distorted facts must happen at the physical location of the emergency.

But is it possible, in an epidemic, to demarcate the area that is the actual site of the public emergency? Localized epidemic breakouts arguably do qualify as sites of a public emergency, but the existence and geographic scope of these could be hard to determine, especially with the lack of mass testing and data on infection rates being generally unreliable.

Some, like the Hungarian Helsinki Committee argue that in the case of an epidemic, the whole country can be deemed a site of public emergency (concluding from this that the law does not need updating and is already applicable to spreading fake news about the epidemic). This is not necessarily a convincing argument. The plain meaning of the word “site” is a physical location. Furthermore, such an extensive interpretation of the word “site”, while it might appear to be an acceptable interpretative leap to make the crime applicable to an epidemic, would also extend the applicability of other crimes, too. For example, committing theft (section 370. Subsections 3,4,5,6 of the Criminal Code), embezzlement (section 372. (1) of the Criminal Code), and fraud (section 373. subsection (1)) at the site of a public emergency is an aggravating circumstance in these crimes entailing longer imprisonment. With such an extended interpretation, committing a theft during an epidemic anywhere in the country would theoretically qualify to be a more serious crime.

Another problem with the definition is the term “distorted fact”. Not necessarily in itself. The distinction between falsehood and distorted facts is an important one in criminal law, and used in many jurisdictions. In German criminal law, for example, the crime of fraud (section 263 of the German Criminal Code) also requires that the perpetrator, in order to obtain unlawful pecuniary benefits, damages the assets of another by, among others, “distorting or supressing true facts” (“Entstellung oder Unterdrückung wahrer Tatsachen”). The definition of fraud in many other jurisdictions, including the UK, and US federal law also acknowledge that fraud by misrepresentation can be committed by the distortion of actual facts.

This is based on the age-old realization that crimes involving misrepresentation such as fraud or libel are more often committed by making assertions that do have a factual core than by a bare-faced lie. Distorted facts thus usually mean factual information that is somehow taken out of context, or substantially altered by omitting certain related information or by adding false information. While there remains a core of truth to the assertion, it is distorted to such a degree that the statement qualifies as false as a whole. This interpretation is echoed in Hungarian commentaries, too.1)

But the analogy of fraud is misleading. In a fraud case, at least most of the time, it is easy to distinguish facts from falsehood. The concerns of run-of-the-mill fraud cases, such as what an advertised household appliance can do, can be relatively easily established in a criminal procedure.

The epidemic is a different story, as are many other possible situations where a “state of public emergency” develops. In such a state of events, everything is in a flux. The situation changes dynamically, and so do the facts. If scientists and experts themselves cannot agree on basic truths (about the effective ways of prevention, for example), how can we expect of layman citizens, journalists or politicians to make determinations as to what the facts are and how to present them? While there will always be clear and simple cases, the grey zone between outright falsehood and unquestionable truth will be indefensibly large.

Now it was rightfully pointed out that it is a well-established principle that dissemination of true, factual information, even if it is suitable to agitate a large number of people, cannot be viewed as a criminal act. Likewise, the expressions of mere opinions are not criminalized either. All these principles, however, are of little help during the fluid situation of an epidemic caused by a virus about which we have only scant scientific knowledge.

In conclusion, the crime of scaremongering has been vague, possibly even unconstitutionally so. Moreover, it is doubtful whether it could be applied during an epidemic at all. In this sense, an amendment was unquestionably warranted. Regrettably, the current amendment solves none of the old problems, and creates new ones.

The New Amendment Avoids Solving the Old Problems and Creates New Ones

The new amendment brings two changes, one minor and one major. The minor change is an arguably superfluous clarification to the original definition cited above, specifying that the falsehood or distorted fact must be in relation of the public emergency (Section 337 (1)) of the Criminal Code).

The second, major change brought about by the amendment is the introduction of an aggravated form of the offense (Section 337 (2)). This can be committed by

“claiming or spreading a falsehood or a distorted truth before a large public during an emergency legal regime in a way that is suitable for obstructing or preventing the successful defence.”

The aggravated form thus switches two elements of the basic offense: First, instead of the site of a public emergency it applies to acts committed while an emergency legal regime is in effect. Second, instead of the suitability to alarming or agitating a large group of people it requires the act must be suitable for obstructing or preventing the successful defence.

At the first sight, the amendment version does make things clearer by simply referring to the emergency legal regime. The latter is a question of law not a question of fact (as the “site of a public emergency” is) and much easier established in a criminal procedure. Moreover, as an emergency legal regime is a very rare phenomenon, the aggravated form is actually much more restrictive than it appears.

But alas, there is an emergency legal regime now, one in which the Constitution explicitly allows such restrictions on the freedom of expression that in normal times would be unconstitutional (Art 54 (1)).

In this light the other inserted element of the aggravated form appears especially worrisome. This requires that the act must be “suitable for obstructing or endangering a successful defence”. This, however, can only be determined retrospectively. With our understanding of the epidemic changing almost daily, so does the catalogue of possible acts that might obstruct the successful defence.

Moreover, such a determination is, even retrospectively, inevitably subjective. At the end of the day, what needs to be established is not an actual causation chain (an obstruction that occurred), but simply a potential one (what could have occurred).

This, combined with the ambiguities coded into the definition even before the amendment, creates a crime that is indefensibly vague, and as such, very likely unconstitutional. How can it be determined with a reasonable objectivity, for example, what statements of fact are distorted enough to qualify as false, all the while suitable to obstruct the successful defence in a situation in which both the facts and the acts that might hamper the defence themselves are in flux? This is especially problematic because even the assertion of actual facts can be suitable to undermine the successful defence. Spreading the verified fact of, for example, the lack of equipment or security lapses in isolation wards and hospitals might lead people to refuse to comply with government orders to relocate to these, thus frustrating the successful control of the epidemic. The determination of criminal liability in such a case could hinge on what “lack of equipment” actually means. And this is inevitably contextual and subjective.

These are, of course, theoretical concerns. In response to them, one might argue that the crime, despite its possible unconstitutionality and ambiguity, can prove to be less troublesome in practice, with proper judicial restraints and prosecutorial decisions that strictly adhere to rule of law and human rights standards. But, alas, this is not everywhere the case.

The Danger of a Chilling Effect

It is a natural reflex of any government, democratic and authoritarian alike, to supress the evidence of its own failures. This urge can be even stronger when those failures actually cost lives and cause significant economic harm, as it is during a pandemic. While citizens tend to rally around the government during times of the crisis, mid or long-term, such support can weaken, or the tide of public opinion can even turn against the government. By all accounts, most countries are facing uncertain times: thousands of deaths, prolonged restrictions, and a deep recession could prove to become a toxic mix capable of weakening government support and creating social unrest.

In such circumstances, crimes that are as ambiguous as that of scaremongering in its current form are especially open to abuse. This is particularly worrisome in Hungary, a country where the freedom of the press is already compromised and the political independence of the prosecutorial service is regularly questioned. Authoritarian regimes like China and Russia are both using similar statutes to silence whistle-blowers about their faulty response to COVID-19 and already provide chilling examples how such anti-fake news laws can be abused, especially during an emergency legal regime where the usual fundamental rights guarantees do not apply.

The aggressive posturing of the government, which already lashes out at every critical, however innocent, journalistic question is hardly reassuring. Whether consciously or not, Zoltan Kovacs, the government spokesman, regularly uses language of the new formulation of scaremongering and depicts anyone who raises alarm or doubt about certain measures taken as obstructor of the epidemic defence – even if it is the Hungarian Medical Chamber,.

The problem, however, goes even deeper: for such a crime to have negative effects, it does not even have to be prosecuted. Just the simple fact that it exists can easily create a chilling effect on information flows, leading to self-censorship and heightened cautiousness. A mayor, for example, learning of an outbreak in his or her town will now think twice whether to make such an information public. Editors might err on the side of caution when clearing articles for publication, possibly preventing the public of learning of important information.

This is where the most immediate danger of the new crime lies.


1 Belovics Ervin: Rémhírterjesztés. In Belovics Ervin, Sinku Pál: Büntetőjog különös rész. HVG ORAC, p. 543.