The Draft Law on Offenses against Public Order and Peace, prepared at the end of April this year by the Sarajevo Canton (SC) Government, which, among other things, defines the Internet and electronic media as public spaces, gave rise to a number of discussions in the general public, but also opened up space for many dilemmas and questions related primarily to its implementation. Most of the attention focused on the insufficiently defined part of the law regulating the offense of spreading fake news, whereas the media and professional community, as well as international organisations, expressed fear that the adoption of such a law could also affect freedom of speech.
Authors: Emir Zulejhić and Dalio Sijah
As SC Minister of the Interior Admir Katica explained, the Draft introduces online activity as a public place for the first time in Sarajevo Canton.
“This means that when someone expresses threats or behaves violently online, it will be possible to sanction that person for a misdemeanour, if the offence was committed in the SC territory or where its consequences occurred in the territory of our canton”, Katica explained.
Article 30 of the Draft defines the following as misdemeanour: “whoever presents or transmits fake news or claims that cause panic or seriously disturbs public order and peace or impedes or significantly hinders the implementation of decisions and measures of competent bodies and institutions that exercise public authority shall be punished with a fine in the amount of BAM 600 to 1,800, whereas whoever commits such an act during the state of natural and other disaster shall be fined in the amount of BAM 700 to 2,100”.
In 2020, an almost identical wording was introduced in the Decision Prohibiting Causing of Panic and Disorder During a State of Emergency in the Territory of the Republika Srpska, adopted by the entity government on 19 March 2020 after the state of emergency was introduced three days earlier due to Covid-19. In the following days, the first misdemeanour reports were filed, however, after objections from national and international officials and representatives of the media and civil society, the decision was revoked, and the MoI withdrew all issued misdemeanour orders.
Similarly, the provisions of the cantonal Draft Law on Misdemeanours, which foresee high penalties for sharing or transmitting fake news, met with the disapproval of the media community and national and international civil society organisations.
The first reactions and criticism refer to the content and (im)precision of the Draft. The Head of the OSCE Mission to Bosnia and Herzegovina, Brian Aggeler, sent a letter to the Prime Minister of Sarajevo Canton, Nihad Uk, the Speaker of the Cantonal Assembly, and the heads of political party caucuses, in which he emphasised that, taking into account the significant impact that disinformation and propaganda have on democracy, it is of crucial importance to be extremely careful when defining “fake news” and assessing its potential for violating public order and peace. Apart from this most problematic part of the law pointed out by Aggeler, which concerns the definition of the term “fake news” and the assessment of the potential for damage that has occurred or may occur, the second most problematic shortcoming of the law is the completely undefined methodology for establishing facts, i.e., fake news, as well as the methodology for selecting fake news that the institutions will handle.
Transparency International BiH pointed to another spectrum of problems brought about by the proposed Draft – namely, its inconsistency with the recommendations of European institutions. They emphasised that the Council of Europe recommendations encourage and call on states to improve freedom of expression online by enacting laws and practical measures that would prevent any form of state or private censorship.
“In April 2018, the European Commission proposed a pan-European approach in the fight against disinformation online, which includes an independent European fact-checking network, improving media literacy and other measures to combat abuses of online space in the direction of prevention through education and other forms of regulation, and not restricting freedom of expression”, Transparency International said.
It is noteworthy that the European Parliament in their Resolution on foreign influence, including disinformation, adopted in June 2023, also warn that politicians are very often the ones who spread disinformation or are used for increasing foreign influence. With this Resolution, the European Parliament called on Member States and the Commission to consider how to counter disinformation from individual actors inside the EU, such as influencers on social media or politicians promoting disinformation on behalf of high-risk states.
However, although the Cantonal Minister Admir Katica assured the public at the beginning of May that the SC MoI has the necessary capacity to implement the new law, in early July he confirmed to Radio Free Europe that the procedure for adopting the draft had been suspended, and that a working group had been formed to amend the draft in the part that specifically refers to offenses online and to redefine the relevant provisions.
According to lawyer Adnan Kadribašić, regulating the issue of online space is not controversial, it is even desirable. He states, however, that “the offenses that will certainly cause problems in practice” are disputable.
“The standards developed by the European Court of Human Rights emphasise that the state has an obligation to protect the rights of third parties from violations happening online and on social media. Therefore, the police officers can do that for a number of violations. However, this only applies to violations of rights that are established as such in the Convention. It is important to separate the fact that hate speech must be punishable (e.g., the decision in the case Beisaras and Levickas v. Lithuania, concerning hate speech on Facebook), while offensive speech against state authorities should not be punishable (e.g., decision in the case Savva Terentyev v. Russia). That is why, in my view, the issue of online space being regulated is not disputable (which is de facto covered by the criminal legislation), but the offences that will certainly cause problems in practice are”, Kadribašić said.
He also points out that a specific concern is the possible overlap with criminal offences, because by issuing a misdemeanour order for some of the offences, such as hate speech, further criminal prosecution would be prevented due to the ne bis in idem principle, which could create a number of new problems. According to Kadribašić, this part of the Draft is incomplete, because for some of the offences, the police officers would first have to consult with the competent prosecutor before issuing a misdemeanour order.
Fact-Checking in Practice
Although countering fake news would be very welcome, as legal sciences PhD candidate Lejla Gačanica believes, not all regulation is inherently good. This, as she wrote in the analysis for Mediacentar relating to Draft CS Law on Misdemeanours, specifically refers to the lack of definition of fake news. The question of who determines what fake news is one of the essential questions that Gačanica asked while analysing the draft. The draft law does not offer any definitions, including the definition of “fake news”. Considering that there are different interpretations and definitions of this term, that we do not have legal definitions, but also that the Draft fails to offer any or assume their elaboration, there is a reason for the question what fake news actually means or what news or claims Sarajevo Canton seeks to sanction.
According to the methodology of the only BiH media fact-checking portal Raskrinkavanje.ba, the rating “fake news” is given to “an original media report (fully produced by the media that published it) that contains factually incorrect statements or information”. Raskrinkavanje, however, does not rate all factually incorrect information as “fake news”.
Disinformation, as defined by the work methodology of this portal, constitutes “media reports that contain a ‘mix’ of facts and inaccurate or semi-true content”. In addition to the above ratings, Raskrinkavanje.ba also uses the “error” rating to indicate factually incorrect information created and published based on obvious mistakes of the sources, without the intention of disinforming the public.
“Pseudoscience” and “conspiracy theory” are two other ratings Raskrinkavanje.ba uses to label factually incorrect content. The former is an attempt to inaccurately present opinions, attitudes, values, or findings obtained by a non-scientific method as facts and scientific discoveries, while the latter is an untrue or unverifiable description of a phenomenon, event, or person that presents them as a part or result of some hidden plan or conspiracy. These, in essence, constitute factually incorrect information that has the same effect, arises from different motives, is created and disseminated using different techniques, and to a greater or lesser extent incorrect.
In addition, experiences of countering harmful, inaccurate and misleading content show that factually incorrect content is not the biggest problem of the contaminated information space, but that there are more numerous, subtle, as well as more difficult to recognise, verify or prove, reports rated as biased reporting and manipulation of facts. In neither case does the released information have to be completely factually incorrect. Biased reporting means reporting that favours facts, attitudes, and conclusions that fit a particular narrative, often disregards the other side rule, and presents facts selectively. Fact manipulation, on the other hand, according to the Raskrinkavanje.ba methodology, means drawing explicit incorrect conclusions based on accurate information, as well as subtly leading conclusions in the wrong direction. Helena Mandić, Assistant Director for broadcasting at the Communications Regulatory Agency, in an interview for the research on disinformation in BiH, pointed out that the majority of cases in which CRA sanctioned broadcasters for disseminating disinformation were related to biased reporting. It is unclear from the Draft whether Sarajevo Canton only wants to address factually incorrect information or whether the intention is to sanction semi-true, manipulative and misleading information, which can also cause damage. If it is the former, the additional question arises as to what is fake news: just fake news or is it also a conspiracy theory, pseudoscience, edited photo and/or something else?
In addition to the question of what exactly will SC address, the answer to the question of how the institutions will implement the law if it is adopted in the proposed form is even more unclear. It is also unclear from the law whether the institutions in charge of implementation will carry out continuous monitoring or act on reports. CRA, the agency responsible for supervising the work of and sanctioning electronic media (TV and radio stations) for non-compliance with the Code, issued by the same institution, has the possibility of monitoring and acting ex officio. To what extent this possibility is used and in what way is monitoring carried out are topics for another research text. On the other hand, the Press and Online Media Council, a non-governmental organisation that brings together the media community and represents a self-regulatory body, has no possibility of supervision and acts exclusively on reports. Which approach will SC decide on is not discernible from the Draft, whereas defining any of the two options in the way the other aspects of it are (un)defined only gives rise to additional questions, ambiguities and imprecisions.
In any transparent and professional process of establishing facts, according to the methodology, after defining fake news and how to identify fake news – which was not done by the Draft – there is another step that the Draft completely fails to address. No matter how the implementing institution finds potential fake news, it is utterly unclear and undefined which fake news, of all the potential ones, they will actually address. If, for example, police officers identify 50 potential fake news from monitoring or reports, what will be the criteria to select the more urgent, relevant and potentially harmful items from the less relevant or less harmful ones?
Even if there were clear criteria and procedures for selecting information whose accuracy is to be checked by police officers, the question arises as to what standards, procedures, tools and techniques will be used to determine the facts. Raskrinkavanje.ba, but also BiH’s political fact-checking portal Istinomjer.ba, like any other fact-checking portal that wants to be recognised locally and globally as professional and credible, undergo an independent external evaluation of their work and standards every year. The International Fact-Checking Network (IFCN) is a global organisation of fact-checkers that sets professional standards and annually checks how well its members adhere to those standards. Only when the evaluation shows that there is a clear, transparent and public methodology in accordance with the highest standards of the profession, as well as that the portal adheres to these standards in its work, IFCN issues a badge, i.e., a verification that confirms all of the above.
To arrive to facts, the journalists of the Raskrinkavanje and Istinomjer portals daily use advanced Google search, Yandex and other available search engines, social media searches, reverse image search, meta-analysis of images and videos, geolocation, as well as dozens of other tools among hundreds and thousands of them. They contact national, regional and foreign interlocutors, experts, institutions and other actors every day, regularly read scientific and social research, monitor the socio-political situation and analyse statements, laws and articles. To answer the question of which of these tools are used in which situations and in which way, we would need another text longer than this one. Which of the mentioned tools and techniques are known to the police officers who will establish facts according to the law if the Draft is adopted, as well as who, when, how and from what resources will enable them to develop and improve if they need to, is completely unclear.
Even if the numerous questions posed in the text above are answered and there are definitions and an elaborate system available (known as methodology in the research and scientific world), everything that police officers establish as facts, as well as the way in which they arrive at those facts, will remain completely secret and unknown to the public, considering that the Draft does not envisage the publication of any of the aforementioned information. Whether something posted on a private Facebook profile, unseen by anyone except friends of the person who posted it, is a public space or not, is a question that also needs to be asked of the Sarajevo Canton Government.
Disinformation and Practices in the EU
Given that harmful content is most often spread online and through privately owned social media, certain activities have been undertaken at the EU level to regulate harmful content online.
In June 2022, the Digital Services Act (DSA) was passed, which will gradually enter into force from the end of August this year until February 2024. According to the European Commission, the Digital Services Act and the Digital Markets Act form a single set of rules that apply across the whole EU, and their goals are to create a safer digital space in which the fundamental rights of all users of digital services are protected and to establish a level playing field to foster innovation, growth, and competitiveness, both in the European Single Market and globally. However, this package does not prescribe any sanctions for citizens, but risk management. On the other hand, under this Act, companies like Meta and Facebook will have to comply with a series of transparency and content rules, including a ban on targeted advertising to children. Violations of the DSA will be punishable by up to 6% of the company’s global annual revenue.
In 2018, the EU also established a voluntary Code of Practice on Disinformation, which obliges tech platforms to demonetise advertising that includes disinformation, flag political ads, give independent researchers greater access to platform data, but also, inter alia, to cooperate with fact-checking initiatives. The Code was strengthened and improved in 2022, and it also defines four key terms relating to harmful disinformation content, which were taken from the European Democracy Action Plan.
The term “fake news” is not used in the Code, but the terms “misinformation”, “disinformation”, “information influence operations” and “foreign interference in the information space”. So far, tech companies applied the Code to regulate harmful content in two cases of crisis – during Covid-19 and Russia’s invasion of Ukraine. This code could soon become an implementing regulation under the DSA package.
After the publication of the first set of reports on the implementation of the Code in early 2023, the European Fact-Checking Network (EFCSN) announced da that “the platforms have fallen short of their commitments”.
Some European Union states have their own regulations pertaining to fake news. Thus, at the beginning of 2023, amendments were adopted to the Law on Offenses against Public Order and Peace in Croatia, where Article 16 reads: “whoever invents or spreads fake news disturbing peace and tranquillity of citizens, shall be punished with a fine in the amount of from EUR 20.00 to 100.00 or imprisonment for up to 30 days”.
According to Poynter, several other European states have some form of regulation to counter fake news, such as Italy, where citizens can report disinformation to the authorities through a special portal, or France, which has a law, but also Spain and Sweden, which established working groups thereon.
Although, according to information from the SC Government, it is evident that the Draft Law on Offenses against Public Order and Peace has been sent back for revision, in the past two months, since it was presented to the public, it has rightly opened a debate. According to experts, there are numerous flaws in its text, which raise doubts about its implementation – from the problem of overlap with criminal legislation to the definition of fake news – raising new questions regarding the restriction of freedom of speech. European practices, as presented above, focus more on risk management rather than repressive policies directed at citizens.
The views expressed in this text do not necessarily reflect the views of the Initiative for Monitoring European Integration’s members or the Initiative itself.