Movimento Brasil Livre: analysing a political
movement through its online indicators
J.P. Guarnieri
University of Amsterdam
Keywords
Brazil, politics, right-wing, neoliberal, activism, social media
Abstract
In less than four years, Movimento Brasil Livre has been able to garner a lot of attention from the press in Brazil. The neoliberal, conservative movement comprised of young people has been involved in its fair share of polemics, but has also gained supporters throughout the country quite quickly. This study attempts to compare their online activity to their presentation of themselves. It does so by looking at it's Facebook pages and Twitter accounts. Through a combination of data extraction, map generation and content analysis, the study obtains information about the movement's localization, goals and relationships. It finds that they are concentrated in the south and south-east of Brazil, that they are a strongly reactive
movement (based on anti-petista sentiment, specifically) who grew through the radicalization of right-wing young people, that there are a few core members and that it is highly self-centred on Facebook, but interacts with both other members of the Brazilian right-wing and with their political opponents on Twitter.
Introduction
As social media becomes intimately intertwined with our daily lives, it starts to permeate, in some form, every aspect of social life. Our relationships with our friends, family and colleagues, that is, the people we know, the people who are physically and emotionally close to us, are often mediated through social networks in some form. It is not surprising, then, that social media has also become part of our interactions with society at large, and thus is now an essential tool for political organization.
What does a social movement need in order to, if not succeed, at least grow and make its voice heard in the political debate? Lopes identifies five key aspects:
“communication, organization, mobilization, validation and scope enlargement” (8), an analysis that draws upon resource mobilization theory, which, since the 70's, has been a very popular approach to understanding social movements.
There are, however, more modern approaches. Bennet and Segerberg talk about a trend of individualization in collective action (744), as paradoxical as this may sound at first. In their view, in individualized action formations, the nominal issues, that is, what is being discussed, may resemble those seen in older movements or partisan organizations, such as environmental concerns or reforms in education. However, the ideas and mechanisms used to organize action become more personalized. This process is linked to a propensity to developing flexible political identities based on personal lifestyles. They note that, while people still join in collective action, and might do so in quite large numbers, “the identity reference is more derived through inclusive and diverse large-scale personal expression rather than through common group or ideological identification”. Therefore, a form of “personalized communication” is essential for the growth of social movements. This personalized communication is characterized by political content being presented in the form of “easily personalized ideas” and the use of personal communication technologies (that is, social networks) that enable sharing these ideas. The very act of sharing them explains how the action is both communicated and organized, and the patterns defined by the technologies used often become organizational mechanisms (746). It is an interesting approach to understanding the growth of social movements in a more organic lens.
There are many aspects of the participation in social media that makes it so efficient in all of these areas. Users are frequently exposed incidentally to news or
political content while browsing, and might be exposed to mobilizing information without looking for it by sources they already somewhat trust: the users they choose to follow (Boulianne, 3), the people they have decided they want to hear from. It also introduces speed and interactivity that are lacking in more traditional forms of mobilization (Eltantay & West, 7). Individuals participate in collective action when they recognize themselves as part of this collective. The feeling of participation comes from social identity, the collective self, which involves personal connection to others through shared membership in social groups (Wright, 2001). Again, social networks provide an opportunity for such connection and the forming of social groups (that is, after all, their very function, at least from the user's point of view).
Lopes hypothesizes that, in current days, “a social movement is the effect of opportunity structures such as the economic, institutional, and social contexts of a country conditioned by its access to social media.”, and “the opportunity structures take into consideration the grievances that drive a social movement.” These
grievances, she goes on to say, derive from change (be it a transformation or a
deterioration) in social, political and economical conditions. (4). This thesis will study a current social movement that is active in Brazil, the Movimento Brasil Livre (Free Brazil Movement), which was founded in late 2014 and is, or, at least, claims to be, mostly composed by young people. It will do so by extracting data from its official communication channels on Facebook and Twitter, but will also gather data from the accounts of some of their leaders and other prominent members of the Brazilian political debate, such as politicians and media vehicles, especially those aligned with the right-wing. However, before delving into the movement itself, some
contextualizations are necessary. What are the grievances that fuel the current political crisis Brazil is facing? What are the “opportunity structures” that led to the creation and success of the movement in the first place?
The Brazilian political crisis
2013 was a very tumultuous year in Brazil. It was the second half of Dilma Rousseff's first mandate, and the social and political pressures that would, eventually, lead to her impeachment (or to the “legal coup” that toppled her, depending on which
political narrative you subscribe to) were already bubbling up. The World Cup was set for the next year, and the Dilma government already had to defend itself from
controversy over the policies enacted in order to prepare the country to host the event. These, however, became secondary after a chain of events, not directly linked to
Dilma's policies, led to a series of political demonstrations all over the country, frequently involving altercations with the police. The spark of such massive popular outrage? A “mere” 20 cents.
Protests over raises in bus fare had already been popping all over the country. In Rio de Janeiro, the protests had started in January of 2012, over a year before this discussion became a national concern. In early 2013, protests erupted in Natal and Porto Alegre. These protests contained the first few incidents of what would
characterize the debate around the protests as a whole: acts of political vandalism. The media and the public would attempt to create distinctions around “vandals” and protesters who were fighting for their rights “properly”, pointing towards a separation between what were considered “fair reivindications” and the means through which they were made, that is, “acts of vandalism”, which were seen as demonstrations of the lack of a statist in the government (Barreira, 149).
While these events started early in the year, they radically changed in June, when they gained massive appeal, which would lend them the name of “Journeys of June”. In the span of a single week, there were four separate protests in the city of São Paulo (Pires), where bus fare had been raised from R$3,00 to R$3,20. As had been the case in other cities before, police repression was swift and violent. Police brutality, coupled with feelings of indignation over the path Brazilian politics had been taking, led to “a national wave of mobilizations the likes of which the country had not seen in 20 years” (Arcary, 359). Now, as many protesters would say in their protest signs, it was no longer about the 20 cents.
While local governments, eventually, conceded, with bus fare in several cities moving back to their previous prices, the protests did not stop. They were, now, indicative of something deeper. Where, previously, they were calling for the the
reduction of bus fare and general increase in the quality of public transportation, they now had more generalised claims, such as asking for better public education and healthcare systems and criticism of corruption in the government, making it clear that the public's belief in the country's political institutions had been shaken (Barreira,
147).
The protests hit the popularity of politicians all across the board. Both Haddad, mayor of the city São Paulo and member of PT (Partido dos Trabalhadores, “Workers' Party”), and Geraldo Alckmin, the governor of the state of São Paulo and member of PT's main political opponent, PSDB (Partido da Social Democracia Brasileira, “Brazilian Social Democracy Party”), recorded losses in their respective popularity percentuals (Mendonça, pars. 4 - 16). In Rio de Janeiro, the same happened to the mayor of the city, Eduardo Paes, from DEM (Democratas, “Democrats”), and the governor of the sate, Sérgio Cabral Filho, from MDB (Movimento Democrático Brasileiro, “Brazilian Democratic Movement”). Dilma Rousseff herself was also strongly hit, losing 27 points of approval over the course of three weeks (Mendonça, par. 2). She was barely able to secure her re-election in 2014, winning with 51,6% of the votes over opposition candidate Aécio Neves' 48,3%, the tightest gap since the country's redemocratization (Benites, par. 1). As rumours of election fraud spread through the internet, Aécio Neves went as far as requesting an audit of the votes. While there was no official audit, his party later published a report on their
independently conducted audit, mentioning that, despite no irregularities having been found, the current electoral system lacked transparency in the voting machines. They would then call for measures that would increase accountability in the electoral process, such as printed votes (Venceslau & Chapola, pars. 1 – 3).
Dilma Rousseff would never recover from the consequences of these protests. While she did end her first term with record approval for the period (BBC, par. 1), on the first year of her second term she would hit record rejection ratings, with 71% of those polled in August of 2015 evaluating her government as bad or terrible and only 8% giving positive evaluations to her government (G1, par. 2). A few months later, a request for her impeachment was accepted by the president of the Chamber of Deputies, the lower chamber of the National Congress.
Despite her low popularity, the impeachment process what surrounded by controversy. There were no allegations of corruption against her at the time. Rather, her impeachment process was based on supposedly illegal fiscal manoeuvrers,
considered to be a form of illegal loans. These manoeuvrers, however, did not, according to the attorney general, fit the legal concept of credit operations (Fabrini, par. 3). They had also been used by other governments in the past and had never been
considered irregular by the Federal Court of Accounts (Passarinho, par. 3). Due to these proceedings, the impeachment has been regarded by the Brazilian left-wing as politically motivated, and a form of “legal coup”.
Whether or not it was actually a coup, it cannot be argued that it did not receive wide support from the Brazilian population. A poll made in March of 2016 found that 68% of those interviewed supported her impeachment (Canzian, par. 1), and the protests in support of the proceedings were the biggest political act in Brazilian
history, mobilizing between 3.6 and 6.8 million people across the country (G1, pars. 1 – 3). Despite these high numbers, society was very much divided over the issue, with support for the proceedings falling as the date of the voting approached. According to a new poll made in July of that year, the Brazilian people would prefer new elections over a Temer government (Mortari, par. 1). Nonetheless, Nonetheless, on the 31st of
August of 2016, the Senate removed Dilma from office by a 61 – 20 vote. Vice-president Michel Temer, who had been presiding the country since May, officially assumed the role of president of Brazil.
MDB, Michel Temer's party, had solid standing on the National Congress, with the second highest number of representatives elected in 2014. But the new president was widely rejected by the Brazilian population. By the time Dilma took a leave of absence due to the investigations, Temer's approval ratings were very similar to hers, and he had less than half her rejection (Datafolha, par. 1). These were the best ratings he would achieve on his entire government. As the new president governed, his ratings continually plummeted, eventually reaching 3% in late 2017, the worst approval ratings a president has had since the end of the military dictatorship, and his rejection was even higher than Dilma's, with 77% of those polled evaluating Temer's
government as bad or terrible (G1, pars. 1 – 8). The same poll indicated that 92% of the country did not trust Temer. “Primeiramente, fora Temer” (“First of all, Temer, out!”) became a tremendously popular expression, not only being uttered during protests and in his public appearances, but by celebrities on interviews, by teenagers on Tinder and by students in their political research.
In 2018, Temer's government is coming to a close, as presidential elections are scheduled to be held this year. But the electoral process itself has been shaking by the political turmoil the nation has found itself in. Since 2002, the Brazilian elections, much like the American elections, had been basically a showdown between two
parties: PT and PSDB. However, their chances of winning the election this year seem slimmer then ever before.
Operação Lava-Jato (“Operation Car Wash”), which is investigating corruption in the semi-public multinational Petrobrás, has found ex-president Luís Inácio Lula da Silva, who was leading the polls with 33% of vote intentions (Cardoso, par. 1), guilty of the crimes of corruption and money laundering, giving him a sentence of 12 years. This investigation, like the impeachment process, has also been accused of being politically motivated by the Brazilian left. The prison of the ex-president again has shown a strong division in the Brazilian population, sparking protests both against (Peduzzi, par. 1) and in favour of the sentence (IstoÉ, par. 1). Due to still having legal recourse to appeal the decision, Lula can still, in theory, run, but his situation might change by October. And Geraldo Alckmin, PSDB's candidate, is not showing good numbers on the polls, currently sitting with 7% of the vote intentions (Gazeta do Povo). The second most popular candidate, who attains the leadership in polls where Lula is taken out of the run, is Jair Bolsonaro, a far-right candidate who is constantly at the centre of polemic, due to his support for the Military Dictatorship who ruled the from the 60s to the late 80s, and his many controversial remarks, such as telling a fellow congresswoman that he “wouldn't rape her because she's not deserving of it” and calling her a “whore” during a session of the Parliament (IstoÉ, pars. 1 & 7).
Brazil is currently facing a terrible institutional crisis. It is hard to say what will happen in the next few months, but, whatever the result is, it will be instrumental in defining not only the future of the country, but of Brazilian democracy itself. Doing political research on Brazil is more necessary than ever, and studying the movements that shape political opinion, an imperative.
The Journeys of June
The Journeys of June had tremendous impact in Brazilian politics, affecting the events that would transpire in the following five years. But their relevance to this study goes beyond how they helped transform the political landscape. The 2013 protests were marked by their relationship with social media at every level.
year were grassroots political movements, such as the Movimento Passe Livre (Free Pass Movement) in São Paulo, the Fórum de Lutas Contra o Aumento das Passagens (Forum of Struggles Against the Raise of Fares) in Rio de Janeiro, the Assembleia Popular Horizontal de Belo Horizonte (Horizontal Popular Council of Belo Horizonte) in Belo Horizonte and the Revolta do Busão (Bus Revolt) in Natal. Since they were groups of organized citizens, without sponsors or ties to any established political organizations, they had limited resources. Consequently, they did not have access to traditional media in order to mobilize people for their cause. It might have been this necessity that led to the use of social media in order to organize protests. A poll that interviewed protesters during one of the demonstrations in June found that 92% of them had heard about the protests through the internet (G1, pars. 23 – 26).
And this allowed the protests to grow as they did. Social media allowed the protesters to be, themselves, part of the organization, by spreading the word and inviting more people to join in. In the same poll, 77% of those interviewed said that they mobilized on social media, and 75% had used some form of social media to invite more people to come. Naturally, this use of the internet as a political tool led to
comparisons between the Brazilian protests, the Arab Spring and Occupy Wall Street. Bárbara Becker Ferreira has made a very well-rounded analysis of the
similarities and differences between the three political movements. They have all made use of the occupation of squares, alternative communication methods and political articulations that refuse traditional institutional dialogue spaces (18), and counted with massive participation from their countries' youth (20). She also suggests that, in all those cases, the grievances of those young protesters were linked to their relationships with the job market (21). But maybe the most important comparison drawn within her work was supported in a citation by Slavoj Žižek:
“What unites these protests is that none of them can be reduced to a single issue.: they all deal with a specific combination of (at least) two issues, a more or less radical economic one (from corruption and inefficiency to outright anti-capitalism) to a politico–ideological one (from demands for democracy to demands for overthrowing standard multipartidary
The protests, in Brazil, were not for democracy (as, despite unrest, whether or not they were living under a democratic regime was never in question). Žižek
mentions that Dilma was, herself, “delighted” with the protests, and the protesters themselves did not know precisely who was responsible for the problems they were fighting against (107). This lack of a clear targer translated itself in the development of a strong anti-partidary sentiment. They were not against any specific party of politician, but against parties and politicians as a whole. This can, again, be seen on the poll made during one of the late manifestations, where 89% of those interviewed said that they did not feel represented by any political party, and 83% said the same about any individual politician (G1, pars. 40 – 41). Calil notes that an “apartidary” discourse had already been present from the beginning in the groups that organized the protests. They were, however, different from the feelings that would develop as the protests grew. To these groups, being apartidary meant maintaining their autonomy by not being subordinate to any party. This did not implicate, as would happen later on, refusing their support or forbidding their members from participating while
carrying flags with their parties' iconography. Calil characterises the dissemination of anti-partidary sentiment as “clearly fabricated by corporate media”, but notes that, despite the success of this narrative, it was not enough to fully control the movements. (390 – 391)
Though it was impossible to fully control the processes set in motion by the initial protests, there were clear attempts of instrumentalizing the power of these demonstrations. In one example, Revista Veja, the highest selling magazine in the country, who, since the redemocratization, has been a supporter of the neoliberal sector of the Brazilian right-wing, published an article on a young man they hailed as “the voice that emerged from the streets”, shining the spotlight on an until then unknown movement: the União Contra a Corrupção (Union Against Corruption). He would then attempt to mobilize far-right groups in a rally named “The Families' March Against Communism”, which failed spectacularly.
Despite the massive adhesion towards the last cycle of protests, the demonstrations eventually lost steam. As more people joined, the demands that initially fuelled the protests got lost in a sea of general dissatisfaction. Those were mainly demands from the Brazilian middle class (De Barros Braga, 40), and Singer
(25) noted that the presence of multiple fractions of society in the last wave of the protests multiplied their potency, but made them more vague.
In short, the Journeys of June because of specific demands from the populations of urban centres, linked to the bus fare and the quality of public transportation.
However, as they grew, they became a form of release for sentiments that had been building up in Brazilian society for a long time, which helps explain the lasting impact they would have. They can be related to other protests around the globe due to their demands, tactics and demographics, and are noteworthy for introducing the use of social media a tool for political organization and participation to Brazilian as a whole.
Movimento Brasil Livre
The object of this study, Movimento Brasil Livre, exists as a consequence of the
Journeys of June. The online organization practices and the participation of the youth in political demonstrations paved the way to the creation of the group in November 2014. Upon their foundation, they published an image on their Facebook page, back then, the only one, where they wrote their “manifesto”, a declaration of their goals and demands:
“WHAT DO WE WANT?
– A FREE AND INDEPENDENT PRESS, without government funding or regulation influencing their positioning
– TRANSPARENCY AND HONESTY in the investigations of all crimes against Petrobrás, a patrimony of all Brazilians
– EXTERNAL AUDITING of the voting machines used in the elections – INVESTIGATIONS ON THE ACTING OF THE POST OFFICE IN THE
PRESIDENTIAL ELECTIONS
– RESPECT TO THE SEPARATION OF POWERS AND THE CONSTITUITIONAL ORDER
– THE END OF DIRECT AND INDIRECT SUBSIDES TO DICTATORSHIPS”. (MBL, par. 5)
Their relationship to the existing political crisis was clear in their choice of demands. First, there were no clear political or economical leanings in their
manifesto. They seemed to be an expression of the dissatisfaction young people had been feeling, the same dissatisfaction that led to many of them to participate in the 2013 protests. The manual they wrote for the creation of new branches (which will be referred to as “manual for branches”, for short) clarifies their political positioning: according to the guide, the movement defends “liberalism as the political philosophy that should orient the actions of the Brazilian state” (MBL, 13). Within their
mandatory readings, are right-wing authors such as Ludwig von Mises, Henry Hazlitt and Fréderic Bastiat (15). While their coordinators can be affiliate to political parties in order to participate in the official political institutions, they are forbidden from being affiliated to“left-wing or statist parties” (18). In fact, these restrictions apply not only to the coordinators, but to members of the movement as a whole, who are
forbidden from membership in PT, PSOL, PCB, PCdoB, PSTU and PCO (19). Despite existing for a little over four years, the movement has already been involved in many controversies. Like other Latin-American right-wing institutions, they are linked to the Atlas Network (Bruza, par. 1), an international network of think-thanks dedicated to promoting neoliberal thought around the world. Due to these links, the movement has been accused of being an American tool for influencing Brazilian politics and of being directly supported by right-wing foundations tied to the Koch brothers (Amaral & Santos, pars. 1 – 3). These allegations are further fuelled by the lack of disclosure in the origins of their funding (Talento & Reverbel, pars. 1 – 2). The group had active participation on the impeachment of Dilma Rousseff, both by helping organize protests and by participating in reunions with politicians, watching the proceedings using irregular credentials provided by members of PMDB and DEM (Lindner et. al., par. 1). An ex-member of the movement, who used be a coordinator for their Espírito Santo branch, denounced MBL for receiving financial and structural support from PMDB in exchange for supporting the party (Sacramento, pars. 15 – 16). The movement uses the formal structure of (and therefore receives all of its revenue through) Movimento Renovação Liberal, which is registered as a private association in the name of the family of Renan Santos, one of MBL's leaders. Together, the Santos family has 125 lawsuits to their name and a debt of over 20 million reais (Betim et. al., pars. 1 – 2).
Movimento Brasil Livre makes ample use of the internet, as social media is its main tool for recruitment, dissemination of ideas and organizing and mobilizing their members. In the manual for branches, there are standards for the creation of blogs, which they consider the most important tool for communication, as they “allows for the confection of more complex texts, and the publishing of research and results” (MBL, 20). Blogs posts can then be shared on their Facebook pages, the official online presence of the movement (21). They also mention Facebook groups and What's App chats as tools to be used by the members to communicate amongst themselves. While there is an official Twitter profile for the movement, and some of the local branches and members are also present in the platform, there is no mention of it in the manual. It is possible, however, that they had not considered the possibility of using Twitter as a tool at the time of the manual's writing. Given their active participation in the platform, this is likely to be the case.
Research questions
As was noted, MBL uses the internet as its primary tool for communication, mobilization and growth. It is, therefore, a prime object for digital research, as any questions about it are necessarily intertwined with their use of digital tools in some way.
Can we use their online indicators to find concrete measures of the movement itself, to make a map of it? It is a decentralized movement, where is it localized? How are the different branches linked? How strong is it in each state? What are its
relationships with the rest of the Brazilian right-wing?
This study is also interested in analysing their ideological position. Looking at their posting patterns, do we find their actions consistent with their stated goals? Are the political platforms they tend to actually defend the ones they use to define
themselves?
Lastly, Movimento Brasil Livre is an ultra-liberal movement. Is there any indication that their growth came from radicalization of citizens who were already somewhat right-wing?
Geographical presence
Movimento Brasil Livre is a national organization. But, at the same time, it is a local movement, or, rather, an agglomerate of local movements. In the introduction of the manual for branches, the movement declares: “The Movimento Brasil Livre (MBL) aims to fulfil its role by bringing the political debate based on liberal ideas and values to the municipal level, where the state can be provoked and transformed with higher efficacy” (MBL, 10). They later reinforce this idea in the second chapter of the manual: “The fundamental goal of the groups, as explained earlier, is the occupation of political spaces and the systematic destruction of statist and leftist thought throughout
counties around the country” (MBL, 24). They make clear that despite aiming to transform the country, which, consequently, denotes an aspiration to having a national reach, they aim to do so by acting locally, as they believe the municipal level to be more adequate for the exposition of their ideas and values, as well as allowing them to have enough influence to pass legislation and impact society.
In the same document, later in that chapter, there are instructions for the creation of Facebook pages, where it is stated that “The official appearance of the group happens in the form of the release of its page” (MBL, 21). We have two pieces of information: that the movement aims to act at the local level and that every branch should have its own Facebook page. The pages seem, then, to be an adequate indicator of their geographical presence, and they can be used as a starting point to try and measure the movement's strength around the country.
The movement does not have any sort of database for their Facebook pages. Therefore, in order to plot its geographical presence, it is first necessary to catalogue their Facebook presence. This was done through manual querying on Facebook's search bar for the expressions “Movimento Brasil Livre” and “MBL”. This approach, of course, is limited by Facebook's search result display. Searches on Facebook will not return a complete list of pages containing the terms, and inactive pages, especially, might not be displayed in the results. Pages connected to the pages found were also looked at, which revealed some additional MBL branches.
Once the list of MBL branches seemed finished, Zeemaps was used to create a visualization of their geographical spread. Red pinpoints were used to represent the municipal branches found. Since MBL also has pages representing the movement at a state level, additional markings were made in blue pinpoints. There were also pages that did not represent cities, such as “MBL – Movimento Brasil Livre Unicamp”, which represents the movement in a university, and “MBL – Movimento Brasil Livre –
Baixada Fluminense”, as it is a region of a state, and part of the counties in contains had their own branches of the movement. Despite Brasília, being the Federal District, having the legal status of a state, it was marked as a city in the map. This is due to its population levels being quite similar to that of other urban centres, such as São Paulo and Rio de Janeiro.
Fig. 1.0. Distribution of branches throughout the country
Through this map, it is possible to see that MBL has some form of online
Brazil. However, a significant number of states did not have municipal branches. It is likely that these states have at least one active branch, presumably, in the state's capital. There were states that did not have online representation at any level. These were Roraima, Amapá, Maranhão, Piauí and Ceará. There is not enough information to tie the lack of presence in these states to their political preferences. While Ceará has consistently voted for PT in the past, so has Bahia, where the movement is present, while Roraima has traditionally voted for PSDB, a centre-right party (Mortari, par. 3).
The distribution of MBL branches correlates with at least two other maps: a map of population density and, even more strongly, a map of voting patterns in the 2014 election. The presence of MBL branches seemed to correlate strongly with counties where Aécio Neves, from PSDB, received the majority of votes. This relation does not seem lost on MBL themselves. They mention PSDB in their manual: “At the same time, the opposition represented by PSDB has shown itself to be distant from the aspirations of a growing section of the electorate. From the absence of a truly
combative opposition emerged a wave of formal and informal groups promoting the basis of the liberal and conservative thought to a variety of public” (MBL, 9). This is further supported by the lack of support for their current presidential candidate, Geraldo Alckmin. In other words, MBL's numbers seem to have grown in areas where there was already a strong affiliation to milder right-wing politics.
Fig. 1.1. Population density in 2010, mapped by IBGE.
Fig. 1.2. Electoral results by county, superimposed over the map of MBL branches. In blue, areas where Aécio Neves, from PSDB, won. In red, areas where Dilma Rousseff, from PT, won. Due to differences in the chosen method for map creation, some stretching was necessary. Map of electoral results made by Felipe Menegaz.
The Facebook pages provide an additional dimension for analysis: their fan count. The fan count can be understood as one of the indicators of the movement's strength in a determined state. In order to provide more precise measurements, only the pages representing states were analysed. This is to avoid the duplication of likes: members of a county might have liked both their county's page and the page
representing their state, which would lead to skewed data. The main page (MBL – Movimento Brasil Livre) was excluded for the same reason.
The average fan count for the states' pages is 7384 likes, with a median of 2380 likes. Above average fan count correlated, again, with population density. This
correlation points towards the need for data normalisation. An article published on the website DigitalGov mentions similar problems: when researchers attempted to
analyse data on online traffic throughout the country, the results would consistently show more traffic in highly populated areas (Holden, 2016).
Fig. 1.3. States with an above average Fig. 1.4. Population density by state, fan count data from IBGE
While this data was useful for understanding where most traffic was coming from, it did not accurately represent which states were more engaged. In order to represent more accurately interest on the website in each state, they divided the number of accesses by the states' populations. In order to improve readability, the results were then multiplied by 100,000, as that was the order of magnitude of the states with the lowest populations.
This approach could give further insight into the relative strength of MBL in each state. The same equation was applied, and the numbers were used to generate a map. A blank map of Brazil was coloured in solid green, with each state being coloured in a separate layer. Then, varying degrees of opacity were applied based on the the number of likes per 100,000 people. The state with the highest amount of likes per person, the Federal District, was left with 100% opacity. The opacity of the colouring in other states was calculated as proportions of that number. The layers were then doubled in order to improve readability.
Fig. 1.5. Map of likes per 100,000 people
This approach, however, did not seem to function as well in this study. It seemed to favour states with lower population instead of properly measuring how strong MBL was in that region. In that sense, absolute numbers seemed to provide more useful insights, as they were more representative of the number of people the movement can mobilize, and, therefore, the amount of pressure they can put on the political
structures in order to attain their goals and the amount of support they can harness for any given project. Counting members among the general population could be useful for studying movements more intimately linked to traditional power structures: they can be used to estimate the number of votes they might receive and, therefore, the likelihood of their members being elected into office. However, despite some members participating on electoral disputes in order to facilitate the discussion and
implementation of their projects, MBL is still a movement highly linked to direct mobilization.
was use: the calculation of standard scores (also referred to as z-scores). Standard scores represent the number of standard deviations by which a value differs from the mean. In other words, by using standard scores in order to make comparisons, we compare the movement to itself, instead of using external factors for comparison. The resulting scores were between -1.189259381 and 1.922530005 . These results were normalised again to provide numbers in a range between 0 and 1. This second normalisation was done by using the equation Normalised(ei) = ei – Emin / Emax – Emin (that is, normalised value equals original value minus minimum value divided by maximum value minus minimum value). In order to avoid having areas with either 100% or 0% opacity, 2 and -2 were chosen as the maximum and minimum values. The resulting values were then rounded to have no decimals.
Fig. 1.6. Relative strength by state
Comparing this map to Fig. 1.0 provides interesting insights. The states of São Paulo, Paraná and Santa Catarina concentrated the highest amount of MBL branches. They were not, however, the states with the highest amount of likes. This can either indicate that the manual query was indeed lacking in a significant number of results,
or that these states have a high number of small branches, while Minas Gerais, Goiás and the Federal District has fewer, but more concentrated branches. It is possible that users are liking the pages of their local branches and not liking the page for their states. This is, however, improbable, given that São Paulo, despite not being at the top, still maintained a high amount of likes.
Now, having a rough idea of the movement's strength in each state, we can put these numbers into perspective by comparing them to the strength of the right-wing in general in each state. But, of course, in order to do so, we must find numbers that provide this data. Voting data from the past elections could be an excellent point of comparison. They can, however, be quite hard to work with: it would be necessary define parties as right-wing, left-wing or centre, and this can be quite hard to do as some parties are quite heterogeneous in this aspect. The data would also be at risk of being outdated, as the political landscape has changed quite a lot in recent years. There is, instead, a more indirect way of measuring the strength of the right-wing: using data from the protests in favour of the impeachment of Dilma Rousseff. As mentioned earlier, the protests that happened across the country in the 13th of March
of 2016 were, together, the biggest political act in the country's history. The same process of using standard scores can be used in order to measure the right-wing's strength relative to itself. The numbers used were, whenever possible, those provided by official government entities (G1), with the numbers provided by the organizers being used whenever there was a lack of official data
Working with those numbers was, however, a bit different, as there was a higher degree of deviation. The highest score, for the state of São Paulo, was 4.304710259, while the lowest score, for Roraima, was -0.4481634038. Most of the results were between the range of -0.5 and 0.5. It is very likely that São Paulo and, to a lesser degree, Rio de Janeiro, have skewed the results: those are not only massive urban centres, they are cities that have always had a culture of protesting. It is
important to note, however, that there wasn't, necessarily, a pure correlation between population and attendance: despite being the 10th most populous state in the country,
Maranhão registered one of the lowest attendances. In order to correct the skewing of the data, the maximum value set for the normalisation equation was 1, and both São Paulo and Rio de Janeiro were represented on the map with 100% opacity. Data for the states that did not have MBL presence was factored into the calculations, but those
states were not painted in the map in order to facilitate comparisons. The resulting map was superimposed on Fig. 1.6, with both maps being set to 50% opacity.
Fig. 1.7. Attendance on protests Fig. 1.8. Map resulting of the across the country superimposing of Fig. 1.6 and Fig. 1.7
In most states, MBL was relatively as strong as the right-wing was in general. This appears to indicate that they comprise a relatively stable portion of the right-wing in most of the country. There are, however, states where MBL seemed to
represent a bigger section of the right-wing: Minas Gerais, Goiás, the Federal District and Pernambuco. The first three were also the states with the highest amount of likes on their pages, which seems to confirm that they are where the movement is strongest. On the other hand, they were proportionally weaker in São Paulo, Rio de Janeiro, Paraná and Santa Catarina. São Paulo is the state where Movimento Brasil Livre has the highest amount of branches, and their page for the state has a high number of likes. Rio de Janeiro, too, has a reasonable amount of branches and a high number of likes in the page for the state. Paraná and Santa Catarina, despite having a high number of branches, had a below average amount of likes on the pages representing their states. This seems to point not towards MBL being weak in these states, as they have a strong physical presence there, but towards the right-wing in general being stronger, and MBL, therefore, representing a smaller fraction.
Basically, this study has found that MBL's geographical presence is
concentrated in the south and south-east regions of Brazil, strongly correlating with areas where Aécio Neves won the last presidential elections, indicating that the movement stems from both dissatisfaction with PT and their traditional opposition, PSDB. This physical concentration, however, did not correlate with the movement's strength, as states that concentrated the most branches did not appear to be the states where they represent a higher fraction of the right-wing.
The Social Network
Social media, in general, are particular forms of social networks. Common parlance, in fact, often equates the two. If you ask Google to define “social network”, the search engine will return two definitions, which, complementing each other, explain what social media as a whole is: “1. a network of social interactions and
personal relationships. 2. a dedicated website or other application which enables users to communicate with each other by posting information, comments, messages, images, etc.” This conflation of the two separate terms is likely to the very central role social media has gained in managing (and, in fact, maintaining) our social networks. And it makes explicit the most important aspect of social media platforms: their function is to, at some level, create links. Therefore, it follows, logically, that in studying the relationship between social media platforms and social movements, we must study these links, which can be done through network analysis. Network analysis as a form of research allows us analyse a web of links among entities and identify key people (or organizations) among this mesh of ties (L. Hansen et. al, 2011). This is particularly useful in studying a decentralised movement such as MBL.
Pages, like individual users, can like other pages. The act of a page liking another one can be understood as akin to an user adding another user as a friend on their personal profile. It is a form of association, of homophily. It is also a clearer indicator of homophily than mere similarities in content. Homophily and influence have been shown to be statistically confounded and that, therefore, similarity in behaviour due to homophily and similarity of behaviour due to influence of one of the actors over the other one cannot be differentiated based on statistical data alone (Shalizi & Thomas IN Nahon & Hemsley, 1295). While similar posting behaviours or
content might be indicative of influence, as ideas seep from an actor to the other one, a page liking another page is a conscious expression of homophily by the author. Liking a page is a deliberate act, which denotes they feel some form of proximity between these pages. It has also been demonstrated that blogs of a certain ideology tend to link to content of the same political affiliations (Nahon & Hemsley, 1305).
In the manual for branches, the movement states that “The job of the municipal branches of Movimento Brasil Livre consist in acting in their complete capacity as political agents in their affiliates' respective municipalities, agglomerating liberal forces in a suprapartidary manner” (MBL, 15). Social networks can make explicit their relationships to other participants in the political debate, such politicians, media vehicles and other citizen-led political movements. They can also be used to
investigate the ties between separate local branches and the internal dynamics of the movement. The manual states that when a Facebook page for a local branch is
created, it will be “announced on the official page of Movimento Brasil Livre and on the pages of other sibling-groups, strengthening links and creating synergy” (21).
Fig. 2.1. Network of MBL's pages
In order to study the relationships between the local branches, like networks for each of them were generated with Netvizz. Then, pages that did not represent MBL were excluded. Pages representing individual members of MBL were also excluded, as the goal was to study the local groups. The size of each note is determined by its in-degree, that is, how many nodes link to them. Their colour is based on a gradient from light red to bright red, determined by their out-degree, that is, how many nodes they link to. The map was organized through the ForceAtlas 2 model, in which nodes repeal each other but their edges attract them, bringing nodes with high in-degrees towards the centre and nodes with low in-degrees towards the periphery (Mathieu et. al, pars. 14 – 15). This makes identifying central figures very intuitive, as they, adequately, tend to be in the centre of the graph or, in bigger networks, at the centre of clusters.
As would be expected, the page with the highest in-degree was the main page, with 45 out of the 100 nodes linking to it. While its out-degree was quite high
main page only links to 19 other pages. This might be indicative of the flow of content. Liking a page causes its content to appear on your feed. The admins of the main page do not see content posted by most local branches, but a high amount of local branches might be seeing the content posted by the main page on their feeds. This could point towards a high amount of centralization on the movement's Facebook presence.
MBL – São Paulo received the most links, but the number is also very small compared to the amount of links received by the main page: only 14 pages are linking to MBL – São Paulo. It is, on the other hand, the page with the highest out-degree, linking to 43 other pages from MBL. It seems like the admins of MBL – São Paulo are the most preoccupied in creating synergy between the branches of the movement. This could also indicate simply that they have more familiarity with social media tools.
The graph does not seem to be indicative of actual cooperation between branches outside of the internet. If that was the case, pages would be clustered by geographical proximity, as these branches would have more links between them. his is clearly not the case: nodes from cities in the state of São Paulo are not necessarily clustered around the page for the capital (which also ends up representing the state, since they have the same name). Similarly, pages from Santa Catarina or Paraná are spread all over the graph, instead of close to their states' or even to each other in general.
Of course, MBL has relationships other than with itself. These relationships, too, can be visualised through like networks. Using Netvizz, two like networks were created for the movement's main page: the first with 1 degree of separation (that is, containing only the pages that page likes), the second with 2 degrees of separation (also containing the pages those pages like). The first network can be used to discern close proximity, who the admins of the main page regarded well enough to link to in the form of a like.
Fig. 2.2. MBL – Movimento Brasil Livre's like network, 1 degree of separation Almost all the pages present are directly linked to the movement. Kim
Kataguiri, Fernando Holiday, Marcel von Hattem, Paulo Eduardo Martins, Thomaz Henrique Barbosa and Renato Battista are all members of MBL. Luciano Ayan, owner of an independent right-wing news website that came under fire for publishing fake news about the death of Marielle Franco (one of the many controversies the movement has been involved in), is a business parter of Pedro D'Eyrot, one of the founders of the movement (Guedes & Chastinet, par. 1). JornaLivre, another independent right-wing news website, has been exposed as having articles written by members of MBL (Vice, par.1; Catraca Livre, pars. 6 – 8). Modo Espartano is also an independent right-wing news website. Its owner, Roger Roberto, who also writes for JornaLivre, is from the same party as Fernando Holiday and has been reported as having ties with the movement (Catraca Livre). La Banda Loka Liberal is a carnival block that played songs in protests against Dilma Roussef. One of it's founders, Felipe Camozzato, was a member of MBL. Lastly, O Reacionário, which is also an independent right-wing news
page, has open ties with MBL: in the side-bar that appears when you click any article, the author, in his profile, declares his membership of the group. The only pages that cannot be directly linked to the movement are Socialista de iPhone and Corrupção Brasileira Memes, two humoristic Facebook pages. These entities, to the exception of the last two, can be understood as the core of the movement, as they appear on this research in a range of areas.
The second network, with 2 degrees of separation, indicates less direct ties. It can be used to situate the movement broadly among other entities of the Brazilian right-wing and of the online political debate in general. Two things are immediately noticeable in this graph: first, that not many entities had high out-degrees, indicating that the liking function is not widely used by the actors present in this network. This could be indicative of Facebook usage for political entities as a whole, but this cannot be verified in this study. Secondly, MBL had by far the highest in-degree in the
network, receiving significantly more likes than more established figures within the right-wing, such as Reinaldo Azevedo or Rachel Sherazade, or even outsider figures with strongly established fan bases, such as Jair Bolsonaro and Olavo de Carvalho. This, however, happens because MBL's main pages receives a lot of connection from local branches of the movement.
Fig. 2.3. MBL – Movimento Brasil Livre's like network, 2 degrees of separation This network reveals a few pages that act as links between MBL and other political and non-political entities. They connect to more varied groups, both in type of group (social movements, artists, authors, humorists, etc.) and in ideology. These likes, which are forms of endorsement, are, of course, mediated by the subjectivities of these pages' admins, which are not necessarily completely in line with MBL's beliefs.
However, their presence on this network still reveals some degree of homophily between them, especially in cases where thematic or ideological proximities can be verified.
Of these pages, La Banda Loka Liberal is situated the closest to Movimento Brasil Livre's main page on the map, meaning they have the most links in common. It links to a lot of foreign right-wing personalities, both alive and dead, such as Ben Shapiro, Margaret Thatcher and Boris Johnson, and to author that are reference within MBL, such as Ludwig von Mises, Milton Friedman, Fréderic Bastiat and Friedrich Hayak. It also links MBL to other, smaller social movements, such as Movimento Apartidário dos Estudantes (Apartidary Students' Movement), Students for Liberty Brazil, Gigantes Brasileiros (Brazilian Giants) and Movimento Liberal Acorda Brasil (Liberal Movement Wake Up Brazil).
O Reacionário links MBL to other independent news vehicles and conservative entities, such as Reaçonaria, Portal Conservador, Observatório
Conservador, Luciano Ayan and others. It also links to more mainstream figures, such as Danilo Gentili (who figures in some of MBL's most engaged with videos), Rachel Sherazade, Reinaldo Azevedo, Marco Antonio Villa and magazines Veja and Época, but these links are more distant.
There are other entities linked to the Atlas Network present on the map, such as the Instituto Ludwig von Mises (Ludwig von Mises Institute), the Instituto Liberal (Liberal Institute), the Instituto Liberal de São Paulo (Liberal Institute of São Paulo), Instituto Liberdade (Liberty Institute), Instituto Millenium (Millenium Institute), Instituto de Formação de Líderes (Institute for the Formation of Leaders) and Students for Liberty Brazil. Most of them appear in a clustered, centred around Instituto Ludwig von Mises. Some of them are directly linked to MBL pages. Notably, Instituto Ludwig von Mises likes MBL's main page, but isn't liked back.
Looking at categories, pages labelled as “Politician” (excluding MBL's own page, which is labelled so, and Stalin da Depressão, a meme page) can be divided in two categories: international influences, which are Abraham Lincoln, Boris Johnson, Margaret Thatcher, Mauricio Macri, Rajnath Singh, Roberto Campos, Rick Perry, Thomas Jefferson and Winston Churchill; and Brazilian politicians: Ana Amélia Lemos, Ronaldo Caiado, Anderson Ferreira, Betinho Gomes, Bruno Araújo, Carlos Sampaio, Eduardo Cunha, Fernando Holiday, Jean Wyllis, Kaio Maniçoba, Luiz Felipe Lampreia, Nelson Marchezan Júnior, Onyx Lorenzoni and Ronaldo Caiado. Two other politicians, Jair Bolsonaro, Marcel van Hattem and Rodrigo Novaes, appear on the map as “Public Figures”.
After doing a bit of cleaning up (which here means mostly changing the labels of humoristic pages that define themselves as “Politician” or “Political Organization”) it is possible to attempt a visualization of ties between politicians and social movements and other political organizations. To do so, the politicians were marked in red, and other entities (marked as Political Organization, Non-Governmental Organization or Nonprofit Organization) in green. Everything else was greyed out.
Fig. 2.3. MBL – Movimento Brasil Livre's like network, politicians and political entities marked
There was a lack of correspondence between movements and politicians. This does not discount any real ties between them offline, it just shows that, if they exist, they do not manifest themselves through online linking. Most politicians were clustered together, likely due to the influence of Betinho Gomes and Nelson Marchezan Júnior, who had liked, or been liked, by many of the politicians present in the map. Betinho Gomes notably has a link to Jean Wyllis, the only left-wing politician in the map.
So, in summary, MBL seems to have a quite centralised network on Facebook, interacting directly only with pages that are explicitly a part of the movement. Their positioning in a broader right-wing network is given through indirect relationships, as smaller pages affiliated with the movement, such as those of individual participants, create links to other entities.
Networks on Twitter
Discourse around the use of social media for political activism has become more mainstream in the last decade. As these tools become a more intertwined with out lives and concrete examples of their usage in social action become more numerous and more prominent, declarations from both the supporters of the political potential of social media and detractors of the techno-optimism surrounding them show their usefulness in comprehending political struggle in the 21st century.
Research on political movements so far has focused on democratic or somewhat left-wing movements: the Arab Spring, Indignados, Occupy Wall Street or Black Lives Matter. There have also been some focus on how Twitter data can be used to predict elections, though most of the research done in this area seems to be retroactive. MBL fits neither of these categories: it's a neoliberal movement, calling not for a more just government, but for a smaller, less interfering one. Despite their members running for political offices, and establishing relations with politicians, elections are not the
movement's main goal, but rather aspirations of individual members. Lastly, Twitter, as a tool, doesn't seem to get as much focus as Facebook. While their manual explicitly talks about Facebook pages as a way of establishing a digital presence, nothing is said about Twitter. Many of the individual members, and many of the leaders, have
accounts, but not every branch has an official account representing the movement locally. The study of the movement through Twitter seems more promising if less focus is given to it as a mobilizing tool, but rather, as a dataset than can be used to discern relationships between individual members, find popular platforms and situate them among their allies. The manual for branches, in the section that talks about forming municipal leaders, notes that politics involve the occupation of institutional spaces and interactions with leaders who presently occupy them (MBL, 29). And Twitter, given its more personal nature, might be the most adequate network for visualising the ties that local leaderships form with their surroundings. It is also important to note that here, too, homophily plays an important role. When Twitter is used as a social
medium, it tends to present a high degree of homophily, creating an “echo chamber” like effect (Colleoni et. al., 328). Therefore, seeing who they interact with can give an idea of who they identify with on a political level.
The initial dataset was formed by looking at MBL's official profile, @MBLivre. Through manual search on the people the profile follows, 20 profiles linked to the movement were identified. Their tweets were captured using a script that works with the Rest API. This means that the data that could be captured on each user was
limited: the Rest API only allows for the capturing of the last 3200 tweets. That means that users who post more will have posts comprising a smaller amount of time. The data captured was input into TCAT for analysis.
Of the 20 profiles used in creating the initial dataset, 17 are, in some capacity, leaders within the movement (the other 3 are profiles representing the movement itself: one for the entire movement, two for local branches): they either participate or manage the movement's social media, are coordinators in their local branches or hold offices in the political structure and therefore bring the movement's platforms to the institutional level. By analysing their tweets, we see not only a picture of the
movement, but of the people surrounding it: they can be celebrities who hold similar political opinions, other politicians who might have platforms in common with the movement, press vehicles, who can either be supporting their narratives, questioning them or simply reporting them, and, of course, common people who are active in Twitter.
A second dataset was made based on the findings of the initial dataset. The initial dataset only directly collected the data of 17 figures, which is good enough for a
starting point, but can create some internal bias. In order to expand the research done, a second dataset was made, with data collected from the 150 users mentioned the most in the initial dataset. This would provide a more certain ground for
comparisons between MBL and other members of the Brazilian right-wing.
Leaders within the leadership
There are two ways of searching the dataset to find leading voices. The first one is by looking at mentions. Looking at the number of tweets, we get a notion of who posts more, but not necessarily an understanding of who engages other people and who is engaged with. Leaders are likely to have a lot of mentions, as they interact with many people and participate in a great deal of conversations.
The most mentioned user is the movement's official profile, @MBLivre. It is, obviously, a centralising profile for all the voices within the movement: its positions are the movement's official positioning, it's MBL not as the sum of all the voices, but as an impersonal organization. In second, is @FernandoHoliday, one of the movement's coordinators at the national level. Fernando Holiday has been elected to a post at the City Council of São Paulo in 2016. While, among the most mentioned users, he is the one who tweets the most, there are other accounts on the dataset with a higher
number of tweets. This could either mean that he engages other users frequently, and by getting into a lot of conversations, receives a lot of mentions, or that this is a result of his position as one of the movement's most prominent members. The third most mentioned user is @kimpkat, the profile of Kim Kataguiri, another coordinator at the national level. He is, arguably, MBL's most famous member. Fernando Holiday and him are the two faces the most associated with the movement, forming a core of its online representation. This can be easily verified through a quick search on Google Images:
Fig. 3.0. Results of Google Images search for MBL
If fourth place is, quite surprisingly, YouTube's official Twitter account, @youtube. Likely, the platform is tagged when they post their channel's videos, probably in a effort to boost their tweets' visibility. Fifth is @JornaLivre_BR, the Twitter account of an independent right-wing news website that gets linked to a lot by MBL's pages. In sixth is @salsichaMBL, one of the hosts of their show on YouTube, MBL News. Seventh most mentioned is O Antagonista, another independent right-wing news website. Eighth is @renato_battista, the other host of MBL news. Ninth is @DMLU_POA, the official account of the Municipial Department of Urban Cleaning of Porto Alegre. The Municipial Secretary of Urbanism of Porto Alegre is currently headed by a politician linked to MBL, who tweets almost daily about the services being provided by the entitiy. In tenth, is @mamaefalei, the account of a YouTube channel linked to MBL.
In order to further establish the validity of these findings, another form of visualization proves itself to be useful: the creating of a network. TCAT produces a file compatible with Gephi, where the links between users can be more easily perused. This also allow us to see relationships that do not become obvious just by looking at numbers.
Fig. 4.0. Mention network based on the first dataset
At the centre of the network is the movement's official profile, with Kim
Kataguiri and Fernando Holiday close by at each side of it. This further confirms them as the centre of the movement as a whole. There is a central mesh, where other
members of the movement can be found, which seems to be the nucleus of MBL.
Within this nucleus, there seem to be two other secondary nucleus: one centred around Raquel Gerde, statual coordinator of the movement in Espírito Santo, and one around
the profile for MBL Campinas. MBL members of this nucleus seemed to be connected to almost the entire network.
Other members are more distant, and have their own smaller networks centred around them: Isadora Caprini, a local coordinator from Espírito Santo, Ramiro
Rosário, a founding member who leads the Municipal Secretary of Urban Services of Porto Alegre, and Carmelo Neto, who coordinates the movement in Ceará. This
configuration is closer to what MBL intends in their branch manual, with local leaders centralising their alliances with other local groups and politicians. MBL Santa
Catarina and Ramiro Zinder, who coordinates the movement in that state, do not have strong networks around them, with the profile for MBL Santa Catarina at the fringe of the profiles gravitating around MBL Campinas. This seem to confirm that, despite the high amount of municipal branches in that state, the movement is not quite strong in that area. It is also possible, however, that the leaders of these movements are either not very active on Twitter or are disconnected enough from conversations with the most central figures in the movement that they did not appear on the graph.
The movement's relationship to politicians can also be more clearly seen in that map. Of course, proximity doesn't necessarily imply endorsement, only engagement. João Doria Jr., elected mayor of São Paulo in 2016, appears quite close to Fernando Holiday. The movement endorsed him until they had a falling out in 2017 due to the mayor's regulations on Uber. Flávio Rocha, owner of the chain of clothing stores Riachuelo, appears quite close to MBL's profile. He is a candidate to the presidential election this year, and is being supported by MBL. Also close is João Amoedo, from Partido Novo, a recently founded political party cited in MBL's manual as the
preferential party for affiliation. There is a small cluster of politicians next to Renato Battista, containing Jair Bolsonaro, Lidbergh Farias, Luciana Genro, Geraldo Alckmin and Rogério Marinho, with Marina Silva, Aécio Neves and Roberto Requião close by but out of the cluster. These politicians represent a wide variety of views indicating that this cluster probably represents conversation in disagreement. Some of them either are or have been linked to the presidential elections: Aécio Neves and Luciana Genro ran in the 2014 elections, and Geraldo Alckmin and Jair Bolsonaro are running in 2018. Marina Silva is the only candidate there who has been present in both
elections.
noticeable clusters, showing a high degree of interactivity between the right-wing (and other participants of the political debate on Twitter, such as some left-wing candidates and media vehicles) in general, with an average degree of 20.365. This indicates a strong political debate occurring on Twitter, as the number of mentions between users is very high.
There are relationships that are maintained through the two graphs: Ramiro Rosário (@curtaramiro) and Isadora Caprini (@isadoracaprini) are still at the edges of the graph, with their own clusters linked to them. This reinforces the idea that they are strong regional leaderships, as they have networks centred on them, but, it now shows that they are disconnected from the national political debate. Mainstream media vehicles, such as Veja and Estadão, which were clustered around Raquel Gerge (@raquelgerde) are now close to MBL's main profile (but not as close to Fernando Holiday and Kim Kataguiri). Their data had not been collected in the first dataset, which indicates that, while they mention MBL with some frequency, they are not mentioned by them as much. Jair Bolsonaro (@jairbolsonaro) and João Dória
(@jdoriajr) have moved away from MBL, which shows that, despite some interactions between them, which, in the first graph, made them appear quite close, they are more participant in other networks of users. Fábio Ostermann, one of the founders of the movement, did not have a significant presence on the first graph. He is now at the center of a small cluster. He also moved further way from MBL's cluster and more towards Partido Novo, to which he is affiliated. This is a visual representation of his distancing from the movement in late 2015, which happened over disagreements over the alliances the movement had made. He appears quite on the edge of the graph in general, indicating that, despite being a pre-candidate for the Chamber of Deputies, he is not a central figure within the party he is a part of.
Fig. 4.1. Mention network based on the second dataset
This second network supports the idea that Twitter is used mostly for
interactions between users, instead of as a creator-to-user form of spreading content, as is Facebook. A significant portion of the users whose data was collected is
represented by nodes with a darker shade of pink, representing a high out-degree. They are also quite similar in size (which was used to indicate their in-degree): there are, of course, bigger nodes and smaller nodes, but most of the big nodes are about the same size. Fidgeting around with the maximum side of nodes did not create huge discrepancies. If mentions are a measurement of participation, this indicates a high degree of integration. This is supported by the number of edges: there are so many connections between the users that the edges almost blend on a solid background.
There was, indeed, a high degree of homophily. There are very few left-wing entities present in the network, and those were organized in a cluster. Most of these profiles were from figures linked to PT, such as the ex-president Lula, ex-president Dilma Rousseff, Gleisi Hoffman and Lindbergh Farias. There were also some figures from PSOL, such as Jean Wyllis and Guilherme Boulos. There was, however, a high degree of interaction between the figures from PT and MBL. Lula's Twitter profile (@lulapelobrasil) appears relatively close to MBL's main account, Kim Kataguiri and Fernando Holiday. It was right beside @mamaefalei, which indicates a lot of
interactions in common. This might be due to MBL's positioning against PT, of which Lula is the most representative figure.
In short, a cluster of figures central to the movement has been identified, and it is consistent with the perception of the movement's leadership from the outside: fernando Holiday and Kim Kataguiri are the dominant figures in MBL. The
movmeent's organizational structure has affected their online presence, with leaders of local branches having their own clusters of users centred around them. It has been demonstrated, however, that proximity on Twitter is not necessarily, like it is on Facebook, a sign of homophily, as interaction can come from a place of disagreement.
Hashtag analysis
Twitter is a massive hub of information. Every second, on average, 6000 tweets are published all over the world, which means that, daily, 500 million tweets are published (Internet Live Stats). In 2009, Dorsey, in explaining the bird theme of Twitter, compared tweets to bird calls: meaningless to us, but meaning is applied by other birds; just as well, messages can appear completely useless, but the recipient
