Brazil has recently had its score downgraded on a report by Freedom House on freedom of expression online, the Freedom on the Net 2016 Report. The downgrade is a result of several actions which sought to restrict users’ rights online, especially their freedoms to post criticism, investigative news and to freely use such means of communication.
Amidst an ongoing political reform, the Brazilian Chamber of Deputies and Senate have approved a measure that might further threaten freedom of expression online. Based on Deputy Vicente Cândido’s report on the Political Reform, the inclusion of an amendment, Amendment 6, was accepted. It obligates platforms to remove, in 24 hours, anonymous posts considered “offensive” to candidates and political parties during the time of electoral campaigns. In practice, it would give a legal basis for the removal of comments, posts, and shares by users of social networks such as Twitter and Facebook.
What does paragraph 6 of Article 57-B say
“A complain of hate speech, spreading of false information or offense in favor of a party or candidate, done by a user of a social network application, through a channel made available by such means on the provider itself will imply on the suspension, in twenty four hours, of the denounced publication until said provider may certify the personal identity of the user who published it, without providing any of its personal data to the complainant, except under a judicial order.”
BIll 110/2017, forwarded for presidential approval after passing with outstanding celerity, goes directly against some Marco Civil da Internet (Brazil’s Internet Framework of Rights) provisions as well as against the Ten Internet Governance Principles established by the Brazilian Internet Steering Committee.
In such a context, said provision violates the Principle of the Unaccountability of the Network, since it is not aimed at the possible authors of illicit acts, but towards the Internet’s intermediaries. Besides, it clearly offends freedom of expression, as it proposes a mechanism for private censorship, perpetrated by private enterprises operating social network applications, summarily, in a short deadline, which, in practical terms, also ignores rights to a prior hearing and to defense.
The goal of such provision would be to make sure application providers could identify users responsible for hate speech, false information or even offenses to parties and candidates, so as to enable later judicial measures. Therefore, the provision obligates providers to suspend, in twenty-four hours, counting at the receipt of the complaint, of the complained content, until it can personally identify the user.
The provision, however, is absolutely unnecessary. A rule established by Article 15 of the Marco Civil already obligates providers to store connection logs for 6 months, which are enough to guarantee the identification of its users. Marco Civil has also established, in Article 22, legal requirements and a specific procedure for the judicial requisition of such logs.
The Superior Tribunal of Justice recognizes the obligation by application providers to collect and store only the application’s logs and not a single other data or information by its users. So much that article 11, § 1st, of Decree 8771/2016, which further regulates the Marco Civil, expressly releases application providers from collecting and storing registration data (Personal qualification, address, and filiation) of its users.
Another point that deserves criticism for its absence of any technical concern is the clarity of the terms “personal identity of the user”, which lacks any precise legal definition. The excessive amplitude and lack of clarity in the adopted language would bring too much legal uncertainty. On the terms of the specific legislation (The Marco Civil da Internet and its decree), Internet application providers are not obligated to collect and store “register data” but only “access and connection logs”, which won’t fit into any concept of “personal identity of the user”. Therefore, the amendment shows even further disregard for the current normative context on the matter.
The goals of such provision are already achieved by the obligations imposed on Internet application providers, whether by the Marco Civil da Internet or by the decree with further regulated it. Therefore, there is no valid justification for the summary suspension of any content until providers are able to identify its users, on the terms and limits of the Marco Civil and of the Superior Tribunal.
The provision also violates freedom of expression (Established by Article 5 of the Constitution) as it imposes a summary suspension of content on the Internet by private actors, independently from judicial evaluation, in a subjective and vague language and with an unreasonable deadline. Despite recent attacks by less democratic segments of the population, freedom of expression is still recognized as a fundamental human right by international treaties ratified by the country, such as the International Pact for Civil and Political Rights. Freedom of expression is also a fundamental principle for the use of the Internet in Brazil, as established by the Marco Civil da Internet legislation, as well as a fundamental right guaranteed by the Constitution.
According to the UN special rapporteur for freedom of expression, in a joint declaration with the Organization for Security and Cooperation in Europe, with the Organization of American States and the African Commission on Human and Peoples’ Rights, “general prohibitions to the dissemination of information, based on vague and ambiguous ideas, such as ‘fake news’ or ‘non-objective information’, are incompatible with international standards.”
The legal requirement for the summary suspension of content, after mere complaint, would violate not only the exercise of citizenship and freedom of expression but also the dynamic structure of Internet in Brazil. This practical provision establishes a “privatization” of the legal analysis of content, transferring to application providers the delicate responsibility of analysing whether the content of any post made through its networks is or isn’t false information or hate speech – which goes against the constitutional principle of non-obviation of analysis by the judiciary power (Article 5th, XXXV of the Constitution).
Wouldn’t it fall on the Judiciary Power to determine what is illicit or not, as its typical function? This is an important safeguard of abuse, which was reflected in the Article 19 of the Marco Civil da Internet, after an extensive debate, and which has been consolidated by the Superior Tribunal of Justice, the court responsible for the uniformization of Brazilian federal legislation. Should the provision not be vetoed, there will be a clear stimulus for the conduction of censorship during the electoral period, which would cause serious damages to the democratic debate in the country.
The Electoral Justice must act with the least interference possible in the democratic debate. Beyond that, the limitations imposed on Internet content are focused on the candidates, political parties and coalitions, not on the electorate. If even the Electoral Justice understands there must be minimal interference on Internet content, there is no basis for obligating application providers to suspend content based on any complaint it qualifies as “hate speech, dissemination of false information or offense against party or candidate”, under the justification that it must identify the user, and not because of any illegal content.
There is not, in the legislation, any criteria to determine what is hate speech or false information. Even if a user spreads false information in good faith, for example, he or she would have the content summarily suspended, without judicial appreciation. Even further, such provision has potential to suppress any criticism, something that should not be tolerated in any democratic regime, especially during the electoral period.
Once again, the Internet is being seen as a threat and a risk, and not as a tool that amplifies access to information and knowledge, connects people and interests, expand access to public services, creates opportunities for innovation and entrepreneurship and expands citizenship. The measure, if approved by Congress, would work as some form of virtual censorship, which is especially concerning during electoral periods, when the search for information and opinion formation occur, in great part, online. The exercise of political participation by citizens in the future of the country, nowadays, also occurs in social networks, in personal websites and in other platforms for publishing of personal content.
Sadly, the political reform bill was already approved by the Senate. Now, all it takes in a presidential sanction for the Law to be applied in 2018. That means it is time to mobilize. There are many reasons for such a provision so damaging to individual freedoms and to the online exercise of citizenship to be vetoed. For those reasons, the Institute for Research on Internet and Society recommends the integral veto of paragraph 6th of art. 57-B of Bill nº 110/2017.