–Pedro Marques Neto, Professor of Law at Sorocaba Law School and Associate Researcher with the Constitution, Politics, and Institutions Research Group at the University of São Paulo
Contemporary democracies have increasingly restricted or debated the restriction of false or misleading claims, particularly in the context of online speech during election periods. In Europe, Germany’s Network Enforcement Law (NetzDG) and the European Union’s Digital Service Act (DSA) are among the most prominent recent examples. Similar regulatory approaches have been proposed, adopted or discussed across the world. In South America, Brazil’s regulation of electoral disinformation has likewise emerged as a prominent and contested regulatory model. Brazil’s recent experience illustrates the importance of carefully distinguishing between different forms of falsity in political speech. Failure to do so risks not only undermining electoral integrity, but overregulating political expression through excessively elitist or technocratic conceptions of truth.
In recent years, Brazilian electoral legislation and case law have increasingly imposed new restrictions on both the form and content of political expression. Article 323 of the Brazilian Electoral Code, for example, makes it a criminal offense to disseminate ‘knowingly false facts’ in campaign advertising or during an electoral campaign.
Similarly, Resolution nº 23,732/2024, issued by the Superior Electoral Court (TSE), (i) prohibits ‘the use, in electoral advertising, regardless of its form or modality, of fabricated or manipulated content to disseminate notoriously false or decontextualized facts’; (ii) establishes that internet application providers have a ‘duty to adopt and publicize measures to prevent or reduce the circulation of notoriously false or seriously decontextualized facts capable of affecting the integrity of the electoral process’; and (iii) provides for the liability of application providers that fail to immediately remove content and accounts during the electoral period in cases involving risks such as the dissemination of anti-democratic information or the sharing of ‘notoriously false or seriously decontextualized facts’.
Such rules raise serious constitutional concerns, both in their design and in their application. Traditionally, restrictions on suffrage and freedom of expression have been regarded as particularly serious interferences with the exercise of political rights. The right to vote and to stand for office, as well as the freedom to express one’s views on political matters, uniquely ensure that electoral competition remains open to new actors and ideas. For this reason, the state bears a strong burden of justification when it infringes on these constitutional rights. In practice, this means that the state must demonstrate not only that the restrictive measures pursue a legitimate and sufficiently important interest, but also that the means employed are proportionate and appropriately tailored to the objective pursued.
In Brazil, restrictions on false electoral speech are generally justified with reference to the protection of democracy and the integrity of the electoral process. Although these objectives certainly constitute legitimate and weighty interest, both the legislation itself and its judicial application have not always ensured the proportionality of the means employed. More specifically, Brazilian electoral law has often failed to distinguish between two fundamentally different forms of falsity in political speech and, consequently, between two distinct forms of speech regulation: (i) restrictions targeting the viewpoint expressed by the speaker, and (ii) restrictions directed at the mode, manner or mechanisms through which political messages are communicated. This distinction is crucial, however, because each category raises different constitutional concerns and, therefore, imposes on the state a different burden of justification for the infringement of the right.
Consider, first, a statement whose alleged falsity derives from the evaluative judgment expressed by the speaker. In this case, the claim is said to be false not because the underlying facts are fraudulent or clearly incorrect, but because the speaker’s understanding of those facts is regarded as erroneous. This is the type of falsity commonly attributed to statements through which individuals express their understanding and evaluation of political reality. In electoral contexts, such statements usually reflect different personal views of the technical competence, moral character, or political intentions of candidates, political parties, or government programs.
Common examples include statements such as: ‘Candidate A is corrupt!’, ‘Candidate B is a coward!’, or ‘Candidate C wants to abolish workers’ rights!’. Speech of this nature is inherently subject to deep, often irreconcilable disagreement among voters. In pluralist democracies, a significant portion of the electorate may regard Candidate A as an honest person or may not perceive Candidate C’s proposals as posing any genuine threat to workers’ rights. For this reason, content-based restrictions of political speech are generally considered constitutionally suspect and are typically permissible only in truly exceptional circumstances, if at all.
Brazilian electoral legislation, however, does not clearly recognize this category of speech as deserving heightened legal protection. Yet such recognition should be explicit, because a large portion of the most significant controversies involving ‘notoriously false’ or ‘seriously decontextualized’ facts in electoral advertisings or campaigns are, in reality, disputes over competing interpretations of political reality, rather than disputes over objectively fabricated facts.
A well-known decision concerning the 2022 Brazilian presidential election illustrates this problem. In that case, the TSE ordered the removal of a YouTube video entitled ‘Remember the Scandals of Candidate Y’s Government’. The decision became notable for its reliance on the concept of ‘informational disorder’ and for its criticism of the ‘syllogism’ employed by the video’s authors. According to the majority, the video engaged in ‘the manipulation of certain true premises’, whereby ‘several true pieces of information are put together, and a false conclusion is drawn’.
The purpose here is not to revisit all the details of the case. Rather, the point is that, with respect to the syllogism argument, what the TSE ultimately disputed was not the occurrence of the underlying events, but the interpretive conclusion that could legitimately be drawn from them. The Court did not dispute that certain corruption scandals occurred during Candidate Y’s period in office. The disagreement concerned whether those scandals could properly be characterized as scandals of Candidate Y’s government. The facts themselves were not denied (‘true premises’); what was contested was the appropriate interpretation of those facts (‘false conclusions’).
This differs significantly from statements whose alleged falsity stems not from a disputed interpretation of reality but from the use of mechanisms that distort or compromise their authenticity or reliability. Here, regulation does not seek to suppress a particular viewpoint. Rather, it seeks to govern the manner in which information is communicated.
One example is TSE’s regulation of AI in electoral campaigns. Resolution 23,732/2024 requires electoral advertisements to ‘disclose, in an explicit, prominent, or accessible manner, whether the content has been fabricated or manipulated and which technology was used’. Similarly, one might consider fabricated videos falsely depicting electronic voting machines malfunctioning during elections. Restrictions of this kind remain subject to constitutional requirements of legitimacy and proportionality. Nevertheless, they generally raise less severe concerns related to freedom of expression, because they do not seek to determine which interpretations of reality are permissible. Instead, they seek to preserve the informational conditions necessary for democratic deliberation during election periods.
The foregoing examples illustrate the difficulty of regulating freedom of expression in electoral contexts. Although they are insufficient to provide a comprehensive framework for addressing the problem, they highlight the need for regulators and courts to develop conceptual tools capable of distinguishing among different forms of falsity and the distinct constitutional concerns associated with each of them. Broad legal categories such as ‘knowingly or notoriously false’ or ‘seriously decontextualized’ are often incapable, on their own, of providing the analytical framework necessary to resolve these issues in a manner consistent with constitutional principles. Questions concerning the speaker’s intent, the extent and probability of the alleged harm, and the relationship between the alleged harm and the viewpoint expressed must all be carefully weighed when regulating political speech.
In her reconstruction of the historical relationship between democracy and truth, Sophia Rosenfeld highlights the dangers associated both with populist revolts against the very idea of truth and truthfulness and with excessively elitist-technocratic approaches to its determination and dissemination. That same paradox is present in current attempts to regulate truth within the Brazilian electoral process. While an excessively permissive regulatory approach may undermine democracy and electoral integrity, an excessively restrictive approach may gradually transform democratic government into a form of electoral ‘pedantocracy’ — a regime in which a bureaucratic elite, supposedly endowed with superior knowledge, acquires excessive authority to determine the acceptable boundaries of public debate during elections. Distinguishing between different forms of falsity and the distinct constitutional concerns they raise is a necessary first step in preventing such a degeneration.
Suggested citation: Pedro Marques Neto, Lest Democracy Degenerate into Pedantocracy? Challenges in Regulating Truth in Electoral Contexts, A View from Brazil, Int’l J. Const. L. Blog, Jul. 21, 2026, at: http://www.iconnectblog.com/lest-democracy-degenerate-into-pedantocracy-challenges-in-regulating-truth-in-electoral-contexts/