A checagem de fatos no Brasil: uma abordagem semiótica

Abstract

This thesis aims to determine whether the confrontation between misinformation and fact-checking during the Covid-19 pandemic has led to a consistent improvement in the critical skills of fact-checking journalists. The theoretical framework is based on Charles Sanders Peirce’s pragmatism, which posits that through semiosis, the evolution of logical interpretants—and consequently, of thought—can occur via the scientific method. The central hypothesis is that, in the interpretative processes surrounding the confrontations between misinformation and fact-checking, certain limitations arise that hinder the consistent improvement of the critical capacity of professional fact-checkers. This hypothesis is further divided into three subhypotheses, aligned with Peirce’s categories of phaneroscopy. First, we propose that the lack of transparency regarding the origin of misinformation—whether robotic or human—impacts the verification process, hindering journalists' perceptual understanding of the qualitative constitution of the news. Second, we suggest that the tension between the insistence on accurately depicting facts and the "packaging" of news by journalism, coupled with the commercial and political interests of factchecking networks, creates conflicts in journalists' interpretative processes regarding the relationship between facts and news. Finally, we hypothesize that robotic influence and working conditions limit the potential expansion of critical skills. To empirically test these hypotheses, we analyzed 20 pieces of content produced by the fact-checking agencies Aos Fatos and Lupa, using Peirce’s analytical framework, and conducted interviews with eight journalists involved in these productions. Our findings indicate that, although the fact-checks do not explicitly address the use of bots, journalists acknowledge the influence of qualisigns associated with robotic operations on their perceptual judgment. These qualisigns could serve as a starting point for deeper investigations, potentially expanding the understanding of the misinformation phenomenon. Additionally, we observed that journalists prioritize dicent indexical sinsigns to convey certainty rather than doubt, and through collateral experience, they strive to approach the dynamic object. This reflects an ethical concern for raising readers' awareness of existing misinformation processes. The content adheres to a journalistic writing style that applies standard journalistic criteria and is sponsored by Silicon Valley companies. The interviewed professionals confirmed that they use these journalistic criteria to select which content to fact-check and how to phrase their findings. However, they asserted that their association with Meta does not influence the content they choose to verify or the manner in which they conduct their factchecking; rather, Meta provides a platform where misinformation processes can be identified. They also noted that other tools are used for this purpose. Finally, the interviews revealed that there is a continuous practice of self-criticism and peer critique among the journalists, and although there are no formal evaluation seminars, daily reflections enable them to refine their verification methods and content presentation practices.