1 Introduction

This study aimed to discuss the power of the counter-mobilization that emerged on Twitter during the coronavirus pandemic in Brazil, inflated by the denials supported by Bolsonaro. We did an analysis of the collective movement deflagrated with the use of #BolsonaroGenocida, as we understand that studying social and political demonstrations through the focus of Social Representation Theory—TRS enables a combined exploration of how dominant and hegemonic relations are maintained and disrupted through localised, everyday practices of political participation. Thus, it provides a path forward to investigating the possibilities for social change [1].

In Brazil, the first recorded case of Covid-19 appeared right after the 2020 carnival. The pandemic management in the country presented denialistFootnote 1 features from president Bolsonaro and his followers [2]; in addition to the informational polarization between the president and the former health minister (Luiz Henrique Mandetta), which resulted in a narrative dispute regarding the whole phenomenon [3].

The narrative dispute concerning Covid-19, the promotion of fake news, and the political polarization surrounding coronavirus, ended up contributing to the construction of polemical social representations (SR) among social groups in several countries in the world, including Brazil [5]. In this sense, studies observe that the political belonging of people influenced the SR about the pandemic, and consequently influenced their behavior toward the prevention and treatment methods [6, 7], besides impacting the intention to get vaccinated [8,9,10,11].

Questioned about the number of deaths, former president Bolsonaro’s statements about the pandemic and his responses to these questions were significant, and some had global repercussions. For example, when asked about the death toll, he answered “So what? I am sorry, I am Messiah, but I do not perform miracles”; “I am not a gravedigger”. Exalting masculinity to fight the virus he said, “We have to stop being a country of sissies”. As for the purchase of vaccines, he made comments regarding other public bodies, such as “look for another one to pay for your vaccine”, and addressed the population protesting about the theme - “go buy some vaccines at your mother’s house”. Also regarding the vaccines, he stated that they could have collateral effects - “If you take the vaccine and turn into an alligator, I have nothing to do with it”.

Finally, there were countless false statements that President Bolsonaro made about the pandemic in 2020 and 2021 [12], waging a real war against science and politicizing the virus, aiming to disqualify both the science and international bodies and their recommendations. The use of this rhetoric to create an identity in his core of supporters, resorting to fake newsFootnote 2 and post-truthFootnote 3 [2], has generated consequences such as the so-called “Bolsonaro effect”. Such an effect indicates bigger Covid-19 damages to the municipalities that supported him in the first turn of the 2018 elections, demonstrating a correlation between the preference for the president and the Covid-19 expansion. More specifically, for every 10 percentage points more votes for him, there was an 11% increase in the number of cases and a 12% increase in the number of deaths [13]. This is a result of the importance of political speech in the increase of political polarization and the dissemination of fake news. They can both create fake information and amplify their promotion. In this sense, elite polarization can raise mass political polarization [14].

The more he supported this anti-science narrative, the more President Bolsonaro’s popularity decreased throughout 2020 and 2021 [16]. In addition to denialism, the neglect regarding the deaths, and the promotion of fake news concerning coronavirus, in March 2020 the government issued a provisional presidential decree (PD) that enabled the suspension of employment contracts and wages for up to four months during the pandemic period—nicknamed as “MP da Morte” (Death PD) by opposers. Such a policy adopted by the Brazilian government during the pandemic wide opened and intensified the historical State violence in the Brazilian context [17, 18], demonstrating that it operates under the logic of “making live and letting die” [19] and necropolitics [20].

From then on, a counter-mobilization began to take place on social media with groups opposing this government position. The hashtag #BolsonaroGenocida (BolsonaroGenocidal) reached the trend topic mark on Twitter in Brazil and was used in more than 16 countries around the globe. Other significant hashtags used, generally jointly with this one, include #ForaBolsonaro (#OutWithBolsonaro), #BolsonaroVagabundo (#BolsonaroScoundrel), #BolsonaroMente (#BolsonaroLies). However, this work focuses on #BolsonaroGenocida for understanding its importance and complexity to discuss the power of this counter-mobilization in Brazil.

On March 25, 2020, Google registered an almost 100% increase from one day to the next one in the search for the word “genocidal” in Brazil [21]. The term “genocidal”, created at the beginning of the 40 s by the polish Jewish attorney Raphael Lemkin is a junction of génos (from the Greek “family, tribe, or race) and caedere (from the Latin “to kill”). He used such a word in his work named Axis Rule in Occupied Europe to describe the systematic murder of the Jews in Europe by nazi Germany. In the following year, the International Military Court of Nuremberg used the term when ruling the crimes committed by the nazis during World War II, along with the accusations of “crimes against humanity”.

Some studies have been carried out mainly on the position of the president himself on Twitter, seeking to study the communication strategies he adopts [22], analyzing publications on vaccination [23], scientific denialism [24] and disinformation [25]. On the other hand, there are still few studies looking at the population’s reaction to these phenomena. In this sense, [26] highlight readers’ reactions on Twitter to news containing statements in newspapers containing disinformation. By understanding that SRs are constructed by common sense, we understand the importance of expanding knowledge about what the population thinks and what was conveyed in #bolsonarogenocida.

The use of this term associated with Bolsonaro during the pandemic is controversial among lawyers, but it is undeniable that #BolsonaroGenocida was frequently used during the pandemic. Therefore, there is a social association of this SR with president Bolsonaro, who has also been linked to the fascist aesthetic during his government [27]. In this sense, this paper aims to investigate the content of posts and images uploaded on Twitter marked with #BolsonaroGenocida in Brazil between March 13 and June 6, 2021; verify how the # becomes tangible (objectification), and how it is attributed to one or various meanings (anchoring) in such posts. Also we intend to analyse the communicative processes of emotional anchoring and objectification on these posts and images to understand the power of the counter-mobilization that emerged on Twitter during the coronavirus pandemic.

1.1 Pandemic, political and affective polarization in Brazil

According to Chauí [28], the rise of conservatism in Brazil stems from political protests in 2013 held on behalf of public safety and against corruption by Brazil’s so-called “new middle class” appalled by the national social problems. As an ideological strategy of discursive capture of the middle class, the dominant one designs an enemy causing all social ills and thus should be eliminated. If during the civil-military dictatorship, the communists were enemies, now the target turned to the Worker’s Party (Partido dos Trabalhadores), which was in power under corruption charges. According to Martín-Baró [29], the delimitation and objectification of any enemy are some of the central processes of political polarization, along with the restriction of perception supported by institutionalized lying and the practices of violence.

Therefore, we understand that political polarization is a psychosocial process with a objectification of “Them” in defence of “Us”; the reductionism and rigidity of perceptive schemes that result in social stereotypes and that disqualify “Them”, in a country already stained by countless forms of violence, especially against minority groups [18, 30,31,32,33].

The phenomenon of polarization can be understood in several ways, both on an individual level, as a polarized attitude, which occurs when the individual takes an extreme position [34], and on a societal level, which is interest of social psychology and can contribute to the understanding of polarization in the political environment for example, where polarization can be found in the cognitive (issue polarization) and affective (affective polarization) mode when members of opposing political parties see each other as one unsympathetic outgroup and attribute negative characteristics to opponents [35].

In this sense, 2013 journeys in Brazil were a result and catalysers of the Brazilian political polarization that lead to the legal-civil-parliamentary-mediatic coup of Dilma Rousseff’s Government [2, 28, 36, 37], making room to the rise of conservative narratives resulting in Bolsonaro’s election in 2018. In this scenario, we have a politically polarized Brazilian context [33, 38], stressed by specific details with the advent of the Covid-19 pandemic [2, 39].

According to Costa [37], between the Worker’s Party elections and the emergency of the 2020 pandemic, we lived through an intense process of political discursive disputes that antagonize and update identity and historical marks of Brazilian society, such as authoritarianism, conservatism, and anti-democracy. We understand that this polarized context therefore favored the emergence of “us and them”, or according to the theory of social identity [40], of ingroup and outgroup, separating Brazilians into those against and in favor of Bolsonaro. The views of the two groups became increasingly marked by emotions and this favored the emergence of counter-mobilization through social networks. In this sense, #BolsonaroGenocida emerged as a social and collective response in face of the actions of the Brazilian State Chief during such a tragic moment for the Brazilian population.

In this way, this context of political polarization contributed to an intense affective polarization in social networks and, as Twitter is known for its emotional and forceful language [41], we can consider affective polarization as an important concept to describe the polarization of social media. In this article we consider that emotions are discursive phenomena that never exist alone, but are natural and necessary parts of all meaning-making, important in all kinds of social representation. Also we can consider that emotions are cultural-cognitive products related to value and social norms and that representations held by individuals as well as textual and visual objects carry emotions as an inseparable part of them [42]. So there is an interest in bridging SRs and emotions because both are socially constituted, constitutive and sensitive to context immediacy [43].

1.2 Mobilizing social representations in social media

SR is a typical phenomenon of modern society, in which mass communication enables the promotion of different forms of knowledge, therefore contributing to forming a thinking society capable of articulating various pieces of knowledge [44, 45]. Thus, communication plays a crucial role in representative processes, since it is a vector for language transmission, concerning structural and formal aspects of social thinking [46], and social reality [47, 48].

SR consists of social knowledge shared daily, which partakes in a series of beliefs, images, and metaphors forming such a representation. Individuals generally adopt images and representations as a result of a collective process of interpreting and understanding new objects [49]. This stems from the fact that every representation has an objectified image, which is a figurative dimension of the social object that simplifies it and is naturalized as the object itself. Objectification establishes a core of images reflecting a complex set of ideas regarding the social phenomenon [50]. This process is guided by interpersonal and media communication. Dynamically, some phenomena become a social problem, acquiring notoriety in the mass media [49], such as in the case of the hashtag BolsonaroGenocida. In addition, it is important to notice that representations do not only influence people’s daily practices but also constitute them [51]. In this way, we can say that the experience lived by Brazilians during the pandemic provided strong emotions of fear, grief, and threat. Feelings and experiences lived bodily. What ended up mobilizing a counter-mobilization through the internet. For (Martikainen e Sakki, 2023) while bodily experiences constitute social (psychological) processes, the social and cultural meanings available in the social environment guide the perception and construction of meaning of bodily experiences. So it’s important to consider the role of sensory experiences and embodiment in construction of SR.

The SR structure has a double nature - one conceptual and the other one figurative. The concept contributes to attributing meaning to an object, symbolizing it; the realization recovers this object, concretising it and making it tangible. This process operates through anchoring and objectification. Anchoring seeks to anchor foreign ideas, allocating them to a familiar context. Such a mechanism aims to approximate unknown situations, categorising phenomena into already known classes and, based on this process, attribute meaning to them. Objectification brings a figurative dimension to the object; therefore the former unknown becomes objective, assuming the contour of real, with concreteness, palpable [50]. These two processes occur as two sides of the same story [52] intertwined and partially simultaneous [53]. To these two processes (anchoring and objectification), Höijer [42] added emotional anchoring and objectification, that refers to communicative processes by which a new phenomenon is attached to well-known positive or negative emotions, in the case of this research we are talking about a social mobilization that anchored and objectified strong and negative emotions in the image of Bolsonaro.

It is vital to study the SR constructed through social media, for insofar as the users upload images, memes, and texts, we have anchoring and objectification. Besides that, some studies show that the use of memes and humor during the pandemic served as a coping mechanism, assisting individuals in navigating the challenges and uncertainties of the period, thereby promoting mental well-being. Humor emerged as a powerful tool to critique government actions, uncover inefficiencies, and shape public discourse. Linguistic devices, including wordplay, irony, satire, and parody, play a crucial role in conveying messages and emotions related to the pandemic [54, 55].

Hence, the interrelation between icons and symbols present in the images is crucial in the SR investigations [56, 57]. In this sense, images can be: (1) an activation or construction source for SR; (2) a product, the “icon-symbolical” synthesis of representations, stemming from the objectification processes; or (3) a transmission means for thinking content, conveyed in different communication channels [58]. Additionally, the use of memes in social media has been a means of sharing beliefs, values, and the sense of belonging to groups through the use of images using humour or irony. Saint Laurent et al. [59] state that memes are created and shared with a strong emotional component (anger, fear, or happiness) which strengthens or rejects a social identity. Due to that association, it is used for political purposes in an empowering or derogatory way in polarising contexts. Studies that include images and SR have focused on analysing social media posts, such as Instagram or Twitter, for they are wide-use platforms, and fertile soil to analyse the SR [18, 60,61,62]. More specifically, Twitter has become a space for discussion of political topics without the effect of social desirability, given that all posts are spontaneous. Social desirability emerges in discourse with the purpose of avoiding negative impressions, presenting what is imagined the other person wishes to hear or what is socially accepted. Therefore, social media platforms in general, and in this study, Twitter in particular, allow for anonymity, should the user choose, enabling the sharing of opinions with little or no influence from social desirability [61].

Furthermore, it’s important to consider the impact of Twitter on public opinion, highlighting how social media platforms shape and reflect societal views. Twitter can amplify polarization by reinforcing pre-existing beliefs and providing a platform for extreme viewpoints. In this way, social media plays a significant role in shaping political discourse and public perception, often through the spread of emotionally charged and polarizing content [63].

1.3 This study

From this study we seek to understand how political and affective polarization contributed to the counter-mobilization of part of the Brazilian public during the coronavirus pandemic, evidenced by #BolsonaroGenocida. We collected tweets using that hashtag, as well as images associated with it to answer the following questions: What is the main content of Twitter’s posts tagged with the hashtag BolsonaroGenocida in Brazil? What images regarding the president and his policies are emotional anchored and objectified through such #?

2 Method

It is a qualitative study in the modality of documentary research. The documentary study is characterized by various styles of written materials, such as legal, journalistic, scientific, and also social media materials [64]. The study occurred in two stages - in the first one we collected posts in text form associated with #BolsonaroGenocida, and in the second, we collected images related to the same hashtag.

To define the collection period, we researched Google Trends to verify the search rates for the expression “Bolsonaro genocida” (Bolsonaro genocidal) in Brazil from the advent of the pandemic to the collection moment (January 2022). According to Fig. 1, we observed two important search peaks, especially in the northeast region of the country. Based on this, we chose to collect the hashtags in the period that encompassed these two peaks: between March 13, 2021, and June 6, 2021, a period of intense political demonstrations in Brazil against the Bolsonaro government.

Fig. 1
figure 1

Search for Bolsonaro Genocida on google trends

Full size image

The tool used to consult and collect the material was the Twitter API v2, made available by Twitter itself through the Twitter Developer Platform, so that algorithms can integrate with the platform programmatically. It is worth noting that API is the abbreviation used for Application Programming Interface, which is a set of rules and protocols that allows different software to communicate with each other. To access this API, it was necessary to request academic researcher credentials, after submitting the research data. After analyzing the submitted data, the platform granted access and enabled researchers to develop a programming algorithm in the TypeScriptFootnote 4 language, which made requests via the HTTPFootnote 5 protocol to the API, requesting tweets using filters such as date, keyword, hashtag and language. After receiving the filtered tweets in JSONFootnote 6 format in return, the algorithm exported a list in spreadsheet format (CSV) for further analysis.

Data collection from the search period found 106.454 tweets from #BolsonaroGenocida, with 15.825 images. From then on, we established as inclusion criteria the tweets that had more than 100 likes, resulting in a sample of 1243 tweets with respectively 503 images. This numerical criterion of 100 likes was established considering the large number of tweets; this way, it was possible to select texts and images with greater ’approval’ on the social media.

As for text analysis, the texts of the tweets were inserted into a file with command lines, containing the variables of “profile type” (female, male, group/pages, company, social/political movements, political party, and non-identified) and “tweet date” (day, month, and year), which were subsequently analysed through software IRaMUTeQ - Interface de R pour les Analyses Multidimensionnelles de Textes et de Questionnaires [65] version 0.7 alpha 2, through the Descending Hierarchical Classification (DHC). It is a lexical analysis allowing one to comprehend the vocabulary contexts and text segments sharing words and according to the strength of association between them (chi-square, x²). In this analysis modality, the software automatically generates classes composed of segments of texts with similar vocabularies. The classes and their most prominent words are visually presented in a dendrogram [66]. Each cluster presents significant lexical components in the corpus and allows the understanding of socially shared informational elements. At this stage, researchers actively participate in identifying the elements and meanings of the partitions and clusters generated by Iramuteq.

The analysis of the images followed Moliner’s [67] three-step model. Initially, we evaluated the corpus and elaborated a systematic inventory with 142 forms representing the elements present in the corpus image. Based on the inventory, two arbiters identified separately the forms in each of the images, followed by a comparison of the analysis resulting in an agreement rate of 98,52%. With this information, we calculated the frequency of each form and the number of different forms to identify the corpus’ diversity. The second stage consisted of a cut of the most frequent forms in the images to recognize combinations and identify patterns. In the third and last step, we created hypotheses about how the image patterns are associated with the study objectives.

3 Results

The results are presented in two steps, first those resulting from the textual analysis of the tweets and, then, the analysis of the images.

3.1 Textual analysis

The textual corpus consisted of 1243 tweets, from which 567 were posted on men’s profiles, 427 were posted on women’s profiles, 168 by groups or pages’ profiles, especially with humorous intent, 36 by profiles linked to political or social movements, 25 by political parties’ profiles, 11 by non-identified profiles, and 8 by companies’ profiles. From the analysis with IRaMuTeQ software, this corpus unfolded into 1291 text segments (TS) employing the Descending Hierarchical Classification (DHC) lexicographic analysis, which retained 88,82% of the TS. In the first partition, DHC divided the corpus splitting class 6 from the rest and, subsequently in the second partition, the analysis split class 5 from the rest. Lastly, classes 4 and 3 and classes 1 and 2 were also divided. Figure 2 presents the dendrogram with classes and words with the highest association rate with each class and their frequency. Classes are presented in a descriptive manner to provide the reader with an understanding of the Brazilian context. Subsequently, these classes will be associated with the Theory of Social Representations.

Fig. 2
figure 2

Dendrogram

Full size image

Class 1 contains 10.33% of the text segments (TS) and presented a relevant statistical relation with the date of May 29, 2021; other dates from May and April presented a relation with the class, however on a lesser scale (p< 0.001), as well as with social and political movements (p = 0.0013). The segments involve a calling for the people to participate in the street protests against Bolsonaro on May 29, 2021, known as 29 m, and took place in several Brazilian cities. 29 m was also a tool to promote the images of the demonstrations. The name of the politician Ciro Gomes appears on tweets anchoring and objectifying the internet users’ position against Bolsonaro, either as a person being sued by the president or exalting his speeches as a political competitor of Bolsonaro. These callings for demonstrations anchored on the discontent regarding the president’s opinions concerning the pandemic, which culminated in what is deemed by internet users as a genocide of the population, that should be shouted down (promoted). The term people was used to indicate that the population will not tolerate either the president’s opinions or the direction his opinions are driving Brazil any longer, as perceived from the following excerpt:

For this to happen in the middle of a pandemic, it is due to people being very angry with the scoundrel #BolsonaroGenocida. I work in public health and I cannot look at a crowd like this, in the middle of the pandemic, and be calm, but I have to admit that the only reason for people to do this is that they cannot take it anymore! (tweet 0080, male profile, 05/30/2021).

In this class, counter-mobilization emerged from a position in favor of science, vaccines and the appropriate use of masks, as a way of positioning itself contrary to presidential opinions and decisions. This is, therefore, the meta-representational aspect, a way of marking recognition between different groups.

Class 2 contains 29.35% of the TS and relates to the dates March 16, 17, and 23, 2021. The term “good day” appears ironically to assert the position against Bolsonaro, demarcating antagonistic groups, such as in the following excerpt: “Good day for those who understand that today being alive is a privilege #forabolsonarogenocida #bolsonarogenocida (tweet 488, female profile, 13/04/2021). However, the word day is also used to indicate the daily death toll, anchoring the arguments regarding the use of the term genocidal to define the president, since Bolsonaro does not like the use of this word associated with him. Here, the ideological position is strengthened based on controversial social representations, marking groups with distinct identities. The term Genocidal is also a form of anchoring, associating government decisions with an already known concept, objectifying the concept of genocidal to the management of the pandemic that resulted in a high number of deaths. Irony emerges as an emotional aspect in the communication of representational arguments.

Class 3 represents 32.07% of the corpus’ TS and is significantly associated with the dates April 6 and April 29, 2021 (p< 0.001). Similarly, to class 2, the element deaths appears to indicate the number of deaths in a lament for the situation in Brazil attributed to Bolsonaro’s decisions. The name of the country also appears because it has become world news due to the pandemic and other problems faced by the population, such as hunger, poverty, and vaccination, indicating the vaccine and Bolsonaro’s removal from power as a means to save lives. In this class, the anchoring and objectification of the figure of Bolsonaro as the character of death occurs. Furthermore, the class presents the experiences with deaths resulting from the pandemic for the group that joined the hashtag, demarcating the social context of Brazil with Covid-19 in light of Bolsonaro’s decisions that reinforced the political polarization of the pandemic and the deaths. These elements give RS a functional value. An example of this class is presented in the following excerpt:

While Bolsonaro tries to stop the bloodletting in his government, Brazil sinks into death and hunger. Today we set a new record - 3668 lost lives. The person responsible has a first and last name. And he will be judged for this tragedy. #BolsonaroGenocida #ForaBolsonaro #UrgentImpeachment (tweet 0612, male profile, 30/03/2021).

Class 4 comprises 7.88% of the TS and relates to the date March 25, 2021 \(\textsc {(p < 0.001)}\). In this class, the tweets refer to the firing and constant changes of health ministers who disagreed with the president’s positions, with the names of ministers Nelson Teich and Marcelo Queiroga standing out. The element education arises to indicate the different government sectors that are under a scrapping process. This class highlights the experiences shared by Brazilians in relation to the frequent changes of health ministers who disagreed with the president, reinforcing informational and political polarization. This antagonism between groups provides the social and identity basis on which these SR are anchored.The following excerpt demonstrates the contents of this class: “In 1 year: this will be the fourth Minister of Health: Mandetta, Nelson Teich, Pazuello, and now, with Ludmilla Hajjar’s refusal, it seems it will be Marcelo Queiroga. But the only one who should be replaced is #Bolsonarogenocida” (tweet 1154, female profile, 15/03/2021).

Class 5 corresponds to 7.88% of TS and presented a significant association with May 13, 2021 \(\textsc {(p < 0.001)}\). This class is composed of tweets highlighting the doses of the Pfizer vaccine offered to the Brazilian government but were not bought, which favoured the peak of half a million dead Brazilians by covid. Other excerpts comprehend other vaccines, such as Coronavac, and the lack of supplies to produce the vaccines at the Butantan institute due to Bolsonaro’s insults directed at China, the exporter of the necessary materials. The deaths and Bolsonaro’s refusal to buy vaccines anchor and objectify mismanagement and give meaning to the term ’Genocidal’ to define the country’s main manager. “Pfizer did not send eleven, but rather 53 emails offering the vaccine and did not get any response. Do you have any idea how terrible that is? This government has sentenced more than half a million to death!” (tweet 009, female profile, 04/06/2021).

Class 6 corresponds to 12.5% of corpus TS and relates to May 15, 18, and 31, 2021 \(\textsc {(p < 0.001)}\). Such a class includes tweets that pay solidarity to the artist Felipe Neto after he called Bolsonaro a genocide and was indicted for a crime against national security. In addition, the outrage and request for liberty for Arquidones Bites, arrested by the police for refusing to remove the sticker “Fora Bolsonaro Genocida”. The events of persecution of people contrary to government decisions were used as arguments to reinforce controversial SR, identity and positioning between groups. An example of this class is: “I cannot believe that this GENOCIDE GOVERNMENT use the National Security Law against Felipe Neto because he spoke the truth. Our solidarity, Felipe!” (tweet 0986, political party’s profile, 16/03/2021).

3.2 Analysis of the images

The image corpus is composed of 503 images, in which we identified 142 different forms. The forms represent elements building the images. We analysed the forms of each of the images through a systematic inventory applied by two researchers and submitted to the analysis of arbiters, as noted above. After this stage, we could observe that the forms with a frequency higher than 5% (16) represented a higher saturation in the corpus, present in 90.05% (n=453) of the images, in which they are selected for the next steps of the analysis and described in the following table  1.

Table 1 Forms with higher saturation in the images’ corpus
Full size table

The following step consisted of identifying existing combinations among forms with the highest saturation to outline patterns structuring the corpus, known as “factors”. We identified 4 factors from the grouping of forms, as presented in Table  2 and illustrated in Fig. 3. Factor 1 relates to forms with the main theme “Movements against Bolsonaro and/or his government”; factor 2 refers to forms related to the topic named “Humour”; factor 3 relates to “Pandemic negligence”; factor 4 relates to shapes representing the “Ultranationalism”Footnote 7 feature.

Table 2 Scope of the factors and frequency of forms
Full size table

Factor 1 presents the largest coverage of the corpus while factor 4 is the smallest. Regarding factor 1, named “Movements against Bolsonaro and/or his government”, we observed a high frequency of the shape “Phrases against Bolsonaro/government” jointly with “Reference to Genocidal”, “T-shirts/posters supporting the movement”, and “Protests/demonstrations/actions against the government”, signalling the strong association of the term genocidal to President Bolsonaro in several public demonstrations against his government. We also noticed, through the presence of the form “Proper use of Mask” in this factor, the representation of people who wore masks correctly during the demonstrations performed as a protective resource, and also as a symbol of protest, contrary to the speeches and incentives of the improper use of masks by Bolsonaro and his supporters. Another form present was “Food”, whose word appeared as a claim on “T-shirts/posters supporting the movement” or with elements in the images that demonstrated the distribution of food in the protests, considering the return of hunger and food insecurity to the Brazilian scenario during the Bolsonaro government.

Factor 2, named “Humour”, is built from the form “Meme/cartoon/comics” associated with “Quote/photo/drawing of Bolsonaro” and “Symbols/words related to death”. This association demonstrates the relevant use of irony in the tweets as a way to reinforce the population’s disagreement with the government’s actions regarding the pandemic. Some examples bring caricatures of Bolsonaro’s supporters, nicknamed “cattle” by the opposition, and others images of similar content. In addition, the association of Bolsonaro’s figure with symbols that represent death is noteworthy, such as tombs, skulls, crosses, the mythological character of death itself, and his image associated with the devil, corroborating factor 1 by reinforcing the Brazilian government’s accountability towards the deaths of the Covid-19 pandemic in the country.

Factor 3 presents the topic “Pandemic negligence” with associations with “Phrases against Bolsonaro/government”, “Symbols/words related to death”, “Number of deaths by Covid-19 in Brazil”, “Protests/demonstrations/actions against the government”, “Quote/photo/drawing of Bolsonaro”, and “Request for vaccine”. These associations refer to the images representing both the management negligence during the pandemic and the delay in purchasing the vaccines, the government incentive to the use of medicines without any proof of efficiency, and the minimisation of the severity of the pandemic and its complications. This category also presents images of various requests for “Vaccine now” by the Brazilian population before the beginning of vaccinations and images of people getting vaccinated while holding posters with the number of deaths by COVID-19 in the country, or other phrases protesting against the government.

Lastly, factor 4, named “Ultranationalism”, presents the association of “Quotes/photos/drawings of Bolsonaro” with the shapes “Colours of Brazil” and “Brazil’s flag”. This association highlights the appropriation of patriotic symbols by Bolsonaro, his government, and his supporters. The images in this factor portray the use of the Brazilian flag, the colours of Brazil, and the jerseys of the soccer team in pro-government protests, and are used as elements of irony regarding this appropriation in the form of “Meme/cartoon/comics”. Factor 4 also shows the presence of requests for military intervention and dictatorship by Bolsonaro supporters, holding posters with such requests in their supporting demonstrations.

Fig. 3
figure 3

Examples of images(Factor 1—Photo of poster with swastika design and Bolsonaro with horns and tail referring to the devil, and the phrase written “Bolsonaro Genocidal—Let’s invade another one”/Factor 2—Cartoon of Bolsonaro lying on skulls saying the phrase “Why this hurry with the vaccine? Brazil can wait, ok?”, while two other people, with names of countries written on their backs, running in the opposite direction with vaccines under their arms/Factor 3—Cartoon of the Ministry of Health building with dead bodies inside/Factor 4—Photo of pro-government protests with people dressed in green and yellow, holding up the country’s flags and posters with writings “Military intervention now, out with communism”.) presented in each factor

Full size image

4 Discussion

In this study, we intend to analyse the power of social counter-mobilization during the coronavirus pandemic in Brazil, focusing on emotional anchoring and objectification of Bolsonaro’s image through Twitter. As for the tweets, the content contemplates the social and political demonstration of a portion of the Brazilian public against the measures taken by the president of Brazil during the pandemic. The president’s neglect for the disease and the number of deaths in the country jointly with the constant changes of health ministers due to disagreements with their ideas on how to manage the health crises triggered the collective movement through social interaction via social networks and the call for demonstrations in the Brazilian streets through the use of this hashtag. According to De Rosa et al. [68], it is vital to rescue one of the fundamental epistemological assumptions of TRS - to understand the interaction between social agents (and their positioning), the social objects, and the multiple forms, channels, tools, contexts, and scenarios of communication, as a set of interrelated systems of social representations dynamically co-constructed and circulating in society, for they are essential to explore the social process of knowledge construction.

The images and memes associated with the hashtag demonstrated harsh criticism against the government’s actions regarding the pandemic, but also the emotional anchoring and objectification of Bolsonaro’s image as the death character. This is revealed because any SR inserted into a certain historical moment meets the experience lived by the development of a group history. The representation can be activated by these experiences, which gives the SR a functional value. Therefore, experience is a functioning representation engine [69]. We can understand that the experience of Brazilians during the Covid-19 pandemic was so intense that it affected mental health [70] and mobilised this social movement through Twitter. Burgos and Oliveira [27] focused on understanding this association of the Bolsonaro government with death, through a suicidal drive encouraged by the exacerbation of the masculinity of the Brazilian people (“stop being a country of sissies”). In this sense, the authors highlight that the suicidal State calls for the people to face death, denying them the possibility to grieve, as grief depends on an expression of individual vulnerability. As pointed out by Gomes [17], the pandemic in Brazil was stained by the industrialisation of deaths, by the reification of life, and by the state’s arbitrariness, turning many biographies to just numbers, and precluding the population from the grief process. Emotions in tweets are strategies to reinforce social identity through WE-THEY polarization, attributing representations and positions to specific groups [71].

We also noticed a strong demarcation of a group positioning favouring science, vaccines, and the proper use of masks through the use of the analysed images and texts, demonstrating that such a group wore masks correctly and was vaccinated even during the street demonstrations against the government, in a logic of meta-representation [72]. This occurs because the SRs enable discussion and argumentation; support the community identities and feed intergroup conflicts; and inform the daily politics and normative politics of legal, institutional, and political debates, which can change the social reality [1].

Communication is an integrating part of SRs, through which one can share beliefs. Social networks are the current mass media platform, allowing people to access the form others interpret the events, and to verify if it makes sense in their belief system or other groups’ systems, enabling a communicational and construction repertoire of arguments about reality. #BolsonaroGenocida and its related elements determined the criticism and distancing from an opposite group, in addition to the association with people with a similar opinion and sense of group affiliation, strengthening the ideological positions regarding Brazilian political reality concerning a sanitary crisis such as the Covid-19 pandemic. Additionally, textual and figurative content provide a satire of the elements featuring Bolsonaro and his supporters, reinforcing the issue of meta-representations, which are crucial when studying SRs and political psychology [48].

We also detected irony in the texts, memes, and images in these posts, determining a characteristic in Brazilians to use humour to cope with difficulties. The irony emerges especially in tweets saluting the president using the word Genocidal, considering that he has signalled nuisance with such a term. The memes take over cyberspace among the images presented. They promote behaviour, being images or phrases that go viral on the internet for they communicate directly with our common sense [73]. Moreover, there is a tendency to satire social events using laughter as a collective measure to cope with public interest news [74]. Therefore, memes are a tool for coping strategies in group distressing situations [75, 76], as in the case of the coronavirus pandemic [59, 77], to mark the political position and group affiliation, for example, following the Covid-19 safety measures [59], and are object to studies to research the SR [78] The use of irony, satire, and irreverence as political strategies begin in the 70 s as a form of resistance to a pragmatic, rigid, and rational model of political activism.Footnote 8

We understand the use of #BolsonaroGenocida as a political participation of a Brazilian social group during the pandemic. We consider the definition of political participation adopted by Howarth et al. [1], which comprehends that it involves all social relations built, developed, and resisted by individuals and groups in their perspectives, ideas, and beliefs. Political participation refers to the symbolic power to build legit social knowledge, norms, and identities, as well as to disregard, marginalize, or silence alternative forms of knowing or being. This type of power is typically granted to a group of higher socio-political status who can construct what is “real” or “true”. Nonetheless, since social representations are dynamic knowledge systems, they can be potentially open to challenges and negotiation. In this case, we cannot state that this group has a socio-political status of power. On the contrary, they probably organised this political demonstration, made feasible by social media, due to being neglected by the government. However, one cannot deny the efficacy of the popular demonstrations of discontent, given that Jair Bolsonaro lost the Brazilian 2022 presidential elections against Luiz Inácio Lula da Silva.

It is worth noting that the inauguration of President Luiz Inácio Lula da Silva, on January 1, 2023, tried to differentiate itself from the previous government that was marked by conservatism, authoritarianism, and neglect of the previous government of Jair Bolsonaro (2018–2022). Symbolically, at the inauguration, Lula sought to demonstrate an inclusive position by walking up the ramp of the Esplanade accompanied by 8 representatives of marginalized groups in Brazil: a black child from the periphery; a woman waste picker; the cacique Raoni Metuketire; a metalworker and rap DJ; a teacher; a woman cook; a young activist of the anti-ableism struggle, and an artisan man.Footnote 9 This movement reinforces, reflects and demarcates once again the expressive movement of social identity through the US-THEM polarization, which reflect the political narratives that are in dispute in Brazilian territory.

Political polarization can be greatly influenced by the moral conviction of individuals and groups [80]. When controversial issues such as same-sex marriage, abortion, among others, are discussed, they tend to create a deep moral conflict between people, leading to an increase in political participation [81]. And according to Yang et al. [82] the media is a potential source of polarization, because as most political events are not personally verifiable, the way they are presented affects people’s perception of the political system, favoring polarization.

The SRs are never stationary, though they can be hegemonic and support ideological systems; however, they enable resistance and agency through the active reappropriation of knowledge and identity positions. Consequently, TRS, based on a critical approach, is capable of explaining both the patterns set in social and political relations, and the possibility of altering unequal participation patterns, challenging identity politics, and achieving social changes [1]. Hence, we can understand that the use of this hashtag carried strong political participation by a portion of Brazilians displeased with the federal government’s management of the pandemic. It is therefore important to demarcate the nature of Twitter as an specific platform for the mobilization of SR through the sharing of images and texts, which ends up evoking emotions that contribute to accentuating the social identity of the group that is expressed.

5 Final considerations

This study did not aim to exhaust the discussion regarding the complexity involving the political polarisation of the Covid-19 pandemic in Brazil. However, it is important to study this form of political participation in the present, which in this case is reflected in the dissatisfaction of at least a portion of the Brazilian population regarding the measures employed by the Federal Government during the pandemic. We understand that this study also provides a historical document spotlighting and creating a written record of both the disregard of the Brazilian state concerning this humanitarian breakdown and the collective mobilisation that took place during the Covid-19 pandemic in Brazil.

We consider as limitations to this study the fact we analysed a sole hashtag that distinguishes an almost homogeneous position of a social group displeased with the Government, which may have limited our comprehensiveness capacity of the phenomenon. Future analyses can prioritize a more significant number of hashtags, maybe opposite to one another, to deepen the understanding of the studied issues. Moreover, we understood that the cut-off of the specified periods due to the large volume of data also restricts this work, and we suggest future analyses to cover other important moments of this period of Brazilian history.

Among the highlights of this study, we could perceive a strong influence of political positioning on the construction of senses, and also Twitter as a means of political demonstration of such positions, as well as the promotion of this information among the public. We also realised the strong influence of the presidential image, because either in support or opposition groups the population observes and uses social media to promote this information or organize collective movements opposing or supporting them. Thus, we emphasize that future analyses of the representatives of institutionalised politics are immensely relevant. In conclusion, we stress the severity and the specific details that stood out in the pandemic in Brazil, because it involved political polarisation, negationism, disregard, violence, and arbitrariness by the State in the management of such a crisis in which many deaths could have been avoided.

Recently, Twitter underwent changes to its platform, even changing its name (now called X), but with more structural changes including limitations on data access. These changes harm the scientific field, making it difficult to replicate the current research and conduct future studies on the platform.