Top-Down and Bottom-Up Disinformation in the 2022 Brazilian Coup Attempt

The next session at the AoIR 2024 conference conference is a session that I co-organised which focusses on controversies, and starts with a presentation by Felipe Soares. His focus is on the 2022 Brazilian presidential election, which finally brought the reign of far-right President Jair Bolsonaro to an end. The election was beset by the dissemination of disinformation on social media, especially about the integrity of the electoral process, and this also led to calls for military intervention in the political system, and coup attempt by Bolsonaro supporters in Brasilia on 8 January 2022.

What is difficult here is that such disinformation isn’t always entirely true or false; it often takes details out of context and shares unverifiable narratives that capture the imagination of far-right supporters – the nebulous threat of a ‘communist takeover’, which was the justification for the 1970s military takeover – still resonates with some in this group. There is often a top-to-bottom cascade of disinformation from leading politicians to the grassroots, but the opposite may also occur, where grassroots disinformation percolates to the top in a ‘reverse influence’ process.

This project sought to explore these dynamics, and focussed on three case studies: visual disinformation on social media (especially posts with images on Facebook); calls for a military coup by elected representatives (focussing on tweets from their accounts here); and in-depth interviews with Brazilian fact-checkers working in these contexts.

Visual disinformation on Facebook was often top-down influence material, but some bottom-up disinformation was also evident. A post calling Lula a thief and featuring this in braille as well moved from fringe actors to versions shared by major political actors, for instance. Such generic statements are not technically disinformation as it cannot be easily verified, of course. Tweets by politicians calling for a military coup are shaped by a collective identity and imagination: they connect disinformation narratives with anti-government messaging. Fact-checkers struggle with all this.

Reverse influence operations are therefore also in evidence, and this represents a form of participatory disinformation. It involves shared imaginaries and beliefs, and creates a blur between straightforward disinformation and collective narratives directed against opponent groups. This is a challenge for the study of disinformation as well as for fact-checkers, and feeds on institutional distrust in political parties and institutions as well as in electoral institutions.