501906 2013 CRS39510.1177/0896920513501906Critical SociologySaad Filho Article Critical Sociology Mass Protests under ‘Left 39(5) 657 –669 © The Author(s) 2013 Reprints and permissions: Neoliberalism’: Brazil, June-July 2013 sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0896920513501906 crs.sagepub.com Alfredo Saad-Filho University of London, UK Abstract The mass movements in June and July 2013 were the largest and most significant protests in Brazil for a generation, and they have shaken up the country’s political system. They expressed a wide range of demands about public service provision and governance, and concerns with corruption. Their social base was broad, starting with students and left-wing activists and including, later, many middle-class protesters and specific categories of workers. The deep and contradictory frustrations expressed by those protests were symptomatic of a social malaise associated with neoliberalism, the power of the right-wing media, the limitations of the federal administrations led by the Workers’ Party (PT), the rapid growth of expectations in a dynamic country, and the atrophy of traditional forms of social representation. This article examines the political dilemmas posed by those movements, and suggests constructive alternatives for the left. Keywords Brazil, neoliberalism, protest, social movements Introduction The mass movements starting in June 2013 were the largest and most significant protests in Brazil for a generation. They can be compared to the mobilizations that triggered the collapse of the mili- tary dictatorship in early 1984, which spread across the country and culminated in demonstrations with more than one million people in São Paulo and in Rio de Janeiro. They exceeded, by a con- siderable margin, the size of the protests leading to the resignation of President Fernando Collor, in 1992. In contrast with those earlier movements, which were narrowly focused around demands for direct presidential elections (the symbol of the national campaign for democracy) and the ousting of a president that was widely regarded as being both incompetent and corrupt, the 2013 protests expressed a wide range of (sometimes conflicting) demands. They included the immediate trigger of the protests – an increase in public transport fares – to concerns over other public services, espe- cially health and education and, finally, broader issues of governance, focusing especially on Corresponding author: Alfredo Saad Filho, Department of Development Studies. SOAS, University of London, London, WC1H 0XG, UK. Email: [email protected] 658 Critical Sociology 39(5) corruption. The recent protests were also distinctive in terms of their social base. They started with students and young left-wing activists and rapidly broadened to include hundreds of thousands of (mainly) middle-class protesters, overtly with little in common. In parallel, the movement also included certain categories of workers with specific (economic) demands, and neighbourhood mobilizations seeking local service improvements. The protests were also deeply influenced by the widespread use of internet tools for their organization and the expression of demands. The June-July movements have shaken up Brazil’s political system. Their explosive growth, size and extraordinary reach caught everyone – the left, the right, and the government – by surprise. This article examines these movements in light of the achievements and limitations of the demo- cratic transition of the mid-1980s, and the experience of the federal administrations led by the Workers’ Party (Partido dos Trabalhadores, hereafter PT) since 2003. The article has three sub- stantive sections. The first reviews the events of June-July 2013 and their social and political con- text. The second examines, from a political economy perspective, six key lessons from the protests. The third summarizes the main points of the argument and draws the relevant conclusions. The Events of June-July On 6 June, the radical left Free Fare Movement (Movimento Passe Livre, MPL), an autonomist non-party organization founded in the early 2000s,1 led a demonstration of around 2000 people demanding the reversal of a recent increase from R$3 to R$3.20 in public transport fares in the city of São Paulo. A parallel increase had taken place simultaneously in Rio de Janeiro. In previous years, the MPL had led similar movements in other cities, some of which achieved partial victories, including the reversal of fare increases and the introduction of free (or half-price) public transport for students in some municipalities. The MPL demonstration in Avenida Paulista, São Paulo’s main thoroughfare, where the largest enterprises, banks and business associations are based, was criticized by the press for disrupting the flow of traffic, for vandalism, and for making unrealistic demands.2 The demonstration was attacked by the police, and many arrests followed. The MPL returned in larger numbers in the days following, and the police responded with increasing brutality, beating up demonstrators and pass- ers-by indiscriminately.3 Several journalists were wounded, and some were shot with rubber bul- lets.4 Savage police repression and the plight of the journalists helped to bring the protests to the attention of the country. After those events the municipal, state and federal governments were thrown into confusion and the fallout from those street clashes threatened to engulf President Dilma Rousseff, who was already on the defensive because of the country’s escalating macroeco- nomic problems (see below).5 In mid-June, the (right-wing) mainstream press and TV networks, having just covered the dem- onstrations against the Erdoğan government in Turkey, aggrieved at the attacks suffered by their reporters, and sensing an opportunity to embarrass the federal government, abruptly changed sides.6 They suddenly stopped criticizing the movement, and started supporting it.7 The protesters were now portrayed as expressing the energy of youth and popular rejection of the country’s dys- functional political system. Having switched sides, the mainstream press immediately engaged in a full-scale attempt to lead the mobilizations, offering blanket coverage, even at the expense of some beloved telenovelas (soap operas).8 Under the appearance of ‘coverage’, the TV and the main newspapers effectively called people to the streets, and – very importantly – sponsored the multiplication and de-radicalization of their demands. For example, in a popular morning show on TV Globo, one of Brazil’s best-known pre- senters taught mothers how to help their children manufacture tasteful placards for their next Saad Filho 659 demonstration.9 The press also claimed that the movement was non-partisan, but it was at risk of being captured by the radical left parties. Press coverage pushed the rapidly growing mobilizations towards a cacophony loosely centred on civic issues and, especially, state inefficiency and corrup- tion, in order both to drown out the left and to delegitimize the federal government. In rapid succession, the demonstrations exploded in size and spread across the country. By late June they had already attracted well over one million people in hundreds of cities. Movements continued to take place almost every day in July, including a large nationwide mobilization and strikes led by the left on 11 July (to be followed by another mobilization on 30 August). However, under the gaze of the press the protests lost focus and came to encompass a great variety of issues, often taking on those that had been highlighted by the media in the previous days. The demands ranged widely, The demands ranged widely, from city transport to public services more generally (for); FIFA, the 2013 Confederations Cup and the 2014 World Cup (against)10; gay rights and the legalization of drugs (mainly for, but most churches are against); compulsory voting (mostly against); abortion and religious issues (all over the place); inflation (against); public spending, privatizations and the state monopolies (unclear); PEC 37, a proposed Constitutional amendment, later rejected, limiting the right of the police to lead criminal investigations, at the expense of the office of public prosecutions (against); President Dilma Rousseff and the PT (divided, with a strong constituency against and, sometimes, asking for Dilma’s impeachment); the return of military rule (a pipe-dream of the far right); and, especially, corruption (against which everyone can happily march together).11 It was not uncommon to see meaningless slogans in the demonstrations (‘The Giant Has Awoken’), vacuous demands (‘Brazil Must Change Now!’), needless requests (‘Pray for Brazil’), empty threats (‘End Corruption or We Will Stop the Country’), and bizarre fashion statements (‘I Want a Louis Vuitton Bag’).12 Anyone could come up with their own demand, and if they were individualist, anti-political and humorous this would increase their chances of appearing on TV. It was a veritable Facebookization of protest, in which anyone could come up with a personal state- ment about the state of the nation and offer potential ‘followers’ their own (self-centred) remedy. Paradoxically, many in the middle class took the opportunity to express their bitter indignation about public services they do not currently use, and which they have no intention of using any time soon.13 In common with recent movements elsewhere, most famously in Iran, Spain, Egypt and Turkey, the Brazilian demonstrations were largely organized through social media and TV.14 They included mainly young workers, students and the middle class, they pulled in localized movements of poor communities demanding improvements to their neighbourhoods, and they attracted categories of workers with demands that might be more or less specific to their circumstances (bus drivers, truck drivers, health sector workers, and so on). Some of the demonstrations immediately following the explosive growth of the protest movement, in early to mid-June, tended to be disproportionately white and middle-class in composition.15 More unusually, many demonstrations had no clear lead- ers, and there were no speeches. Frequently, groups of loosely connected people organized them- selves ‘spontanously’ on Facebook and Twitter, met somewhere, and then marched in unclear directions, depending on decisions made by unknown persons more or less on the spot.16 The police came under heavy scrutiny, and they failed miserably most of the time. Brutal repres- sion was sometimes accompanied by riots; at other times the police pulled back, perhaps tactically (for good or bad reasons), and sometimes out of concern with their own image. In some cases, the police attacked the demonstrators while leaving the rioters alone.17 At least once, repression led to a massacre as Rio de Janeiro police confronted a gang which had, allegedly, used an ongoing dem- onstration to facilitate a spree of robberies. Ten people lost their lives at Favela da Maré as a direct 660 Critical Sociology 39(5) result of this operation.18 Infiltration by the police and the far-right was both evident and wide- spread in most demonstrations.19 At the fringes of vast marches, unidentified gangs trashed banks, shops, offices and public buildings, sometimes under the gaze of an inert police force, while the left was blamed for the destruction. Since ‘all politicians are corrupt’ (a message endlessly, if subliminally, repeated by the media for several years), some marches were, somehow, proclaimed to be ‘party-free’, and left-wing mili- tants and trade unionists were harassed and beaten up by thugs, at times wrapped in the national flag and screaming ‘my party is my country’. There was growing speculation of a plot to create an atmosphere of chaos which might justify a military coup.20 This is far-fetched: the democratic consensus remains very strong in Brazil, the military have not overtly intervened in politics during the last decade, and any attempt to do so now would face significant difficulties. More prosaically, the military are unlikely to believe that they can easily resolve the problems of public service pro- vision which have triggered these demonstrations. Nevertheless, those rumours and events support the view that the mobilizations were in danger of being kidnapped by a media-sponsored upper middle-class political agenda that included an aggressive push to exclude the left from the streets under the cover of ‘patriotism’. The mobilizations continued to grow until late June. As they did so, they became both more radicalized and more fragmented, and the movement lost any semblance of having a unifying plat- form. It became, instead, the expression of widespread indignation with the insufficiencies of the public sector, the state and the political institutions, with corruption providing a symbol of every- thing that is wrong in the country. On 19 June the federal government eventually pushed São Paulo and Rio de Janeiro to reverse the transport fare increases, with the offer of tax breaks and the threat of leaving them alone to sort out the mess otherwise. This surrender was immediately followed by reversals of planned toll price increases, in the state of São Paulo, and of electricity price increases in the state of Paraná. They have also stalled, for the foreseeable future, further transport fare increases elsewhere in Brazil. These outcomes show that a movement with apparently ‘unreasonable’ demands can achieve sig- nificant victories if it is large and sustained enough. The success of the MPL has put public transport firmly on the political agenda, and it has opened the space for more aggressive demands concerning public service provision. However, the demonstrations were largely out of control, and they had acquired a strong middle-class and anti-left edge proving difficult to contain. In response, the left made a coordinated effort to regain leadership of the movement, including a declaration by the MPL that it would no longer call street demonstrations,21 and that it would work closely together with the radical left parties to regain the streets.22 At the same time, the President attempted to take the initiative from above with a call for political reform and proposals to increase spending on public services and to improve health provision.23 The demonstrations subsequently became increasingly scattered and dwindled in size, except for the large nationwide strike on 11 July. Sobering Lessons The protest movements in June-July 2013 have transformed the Brazilian political scene. They have also highlighted both the vulnerabilities of the PT and its left-wing allies, and the fragilities of the federal administration. Dilma Rousseff’s lead in the opinion polls, which looked unassail- able only a few weeks ago, has largely disappeared.24 Examination of the recent demonstrations suggests six trigger factors and key socio-political and economic determinants, which might help to shape their long-term implications. Saad Filho 661 (1) The changing character of the state Since Lula’s first administration, starting in 2003, many hundreds of left-wing political, trade union and NGO cadres have been appointed to the federal administration. While this has, effec- tively, ‘nationalized’ most social movements through the close alignment of the material interests of their leaders with the government’s policy agenda, it has also changed the social composition of the Brazilian state. For the first time, poor citizens could recognize themselves in the bureaucracy and relate to friends and comrades who had become ‘important’ in Brasília. These personnel changes, and the progressive policy shifts in Lula’s second administration, enormously increased the legitimacy of the Brazilian state.25 They also provided a strong platform supporting the claim of the poor for better public service provision, and for a larger share of the products of their labour. This does not necessarily amount to collective action and, in some cases, it is inimical to it; but the affirmations of citizenship, and the closely related change in social atti- tudes, are centrally important for Brazilian democracy. These changes have been called a ‘demo- cratic revolution’ by some analysts (Wu, 2010): this is an exaggeration, but it illustrates the significance of the new relationship between citizens and the state. However, these developments have, simultaneously, increased Lula’s and Dilma’s distance from large swathes of the middle classes, the traditional bourgeoisie and financial market interests, which used to have almost exclu- sive access to the state administration. (2) The social and political isolation of the government The June-July movements have confirmed the unremitting rejection of former President Lula, President Dilma Rousseff and the PT by large segments of the upper and middle classes, and the mainstream media. Although these social groups are relatively small, internally divided and politi- cally unstable, they are highly influential because of their economic power, privileged access to the media and the social movements, and ideological influence over the working class. The revulsion of large segments of the upper strata of the population, which was displayed prominently in the recent marches and in the press, is not due to narrow economic concerns. Lula has plausibly insisted that the Brazilian elite never made so much money as they did during his administration,26 and this may still hold true under Dilma. Nevertheless, wealthy and influential fractions of the bourgeoisie and the middle classes resent their loss of privilege because of the expansion of citizenship due to the democratization of the country under the federal administra- tions led by the PT. To their chagrin, the Brazilian elites have realized that they alone can no longer drive Brazilian politics.27 The redistribution of income and the expansion of social programmes in the last 10 years have benefitted tens of millions of people (see Coutinho, 2013; Mattei, 2012; Souza, 2012), and the wider availability of consumer credit has allowed many relatively poor people to visit shopping centres, fly across the country, and buy a small car (see, for example, Mendonça and Sachsida, 2013; Paula, 2011). Some of these aspirations may be criticized because they are socially undesirable, macroeco- nomically destabilizing or environmentally unsustainable, or because some of these policy priorities were promoted by the government in order to support large capital (see, for example, Boito, 2010). Moreover, rapidly rising incomes at the bottom of the pyramid (see Morais and Saad-Filho, 2011) and rising auto sales28 have not been accompanied by improvements in infrastructure, leading to an overall deterioration in the quality of urban life. Despite these difficulties, those consumption aspi- rations are currently shared by tens of millions of people. The result is that Brazilian roads and air- ports are full,29 and their previous (elite) users complain bitterly about the lack of space to accommodate all those poor people moving about. The lower classes now have a sense of 662 Critical Sociology 39(5) entitlement, and possibly even worse, they also fail to demonstrate the deference to which their taller, whiter, thinner and better dressed social superiors had become accustomed.30 While large capital did well economically over the last decade or more, the middle class did not share in this prosperity. So-called ‘good jobs’ in the private and public sectors are relatively scarce, higher education no longer guarantees ‘good’ income, and the young find it hard to do better eco- nomically than their parents.31 Middle-class groups desperately want economic growth, but they remain ideologically attached both to a neoliberal-globalist project which slows growth,32 and to clientelistic politics, landowner interests and a neoliberal ideology. They are also frightened by the supposed ‘radicalism’ of the government, despite the PT’s extraordinary moderation, and terrified of Brazil becoming another Venezuela.33 Isolation from these key social groups has cost the PT dearly – twice. First, during the mensalão scandal, when the elites and the media attempted to replicate the demonstrations leading to the impeachment for corruption of President Fernando Collor, in 1992 (see Valença, 2002), and tried to bring the masses into the street in order to topple or, at least, imobilize Lula with another cam- paign against corruption. Second, while their attempt failed in 2005, the challenge looked easier in 2013 since the masses were already on the street; all that was needed was a shift in motivation and a suitable set of slogans generously provided and heavily promoted by the mainstream media. (3) Expectations rose faster than incomes The achievements of the PT administrations have raised expectations even faster than incomes. The emerging poor want to consume more, larger masses of people want social inclusion, and both want better public services. The middle classes oscillate between indifference and hostility to the poor, but they would also like to benefit from good public services in the future.34 They are, how- ever, firmly opposed to paying higher taxes in order to have them.35 They claim that they pay too much already, that corruption spirits away a large chunk of the government’s revenues, and that ‘their’ taxes have been supporting a parasitic mass of undeserving poor through federal transfer programmes. At the same time the media, and society more widely, completely ignore the fact that nearly half the federal budget is absorbed in the service of the domestic public debt – effectively a welfare programme for the rich – and that these transfers dwarf the cost of social spending and the federal transfer programmes.36 (4) The limited power of ‘left neoliberalism’ Under favourable economic circumstances, the ‘left neoliberal’ policies implemented by the PT, and the greater legitimacy of the state which has accompanied Lula’s election, can disarm the political right and disconnect the radical left from the masses of the population. Lula ended his second administration, in 2010, with approval rates touching on 90 percent,37 and Dilma Rousseff had nearly 80 percent approval rates until March 2013,38 which had never happened to any Brazilian president in their third year in office. Those high approval rates sucked dry the base of support which might have been available to their potential rivals. No party has ever prospered to the left of the PT and, until recently, the right- wing opposition was hopelessly disorganized. For a brief period, the PT achieved something close to political hegemony in Brazil. Yet, its prominence was not built on solid foundations. Dilma Rousseff was elected by Lula’s social base, mainly the poor, with the support of large capital (see Morais and Saad-Filho, 2011, 2012). But Dilma had a limitation: she had always been a technocrat, had never been elected to public office, and did not have her own political base. Even worse, the Saad Filho 663 economy was bound to decelerate during her administration following the boom years in the mid- 2000s, and Brazil’s extraordinarily successful bounce back after the 2008 global crisis. (5)The economic drag The space to manage the contradictions and limitations outlined above has shrunk in the recent past. Although inflation rates have remained stable, around 0.5 percent per month (6% annually) since early 2007, and despite the fact that expected inflation has also remained stable around these values,39 in early 2013 there was a media-sponsored panic in the country because of the allegedly ‘imminent’ prospect of runaway inflation.40 More significantly, the current account deficit has increased rapidly, from around 2 percent of GDP between mid-2010 and mid-2012, to almost 4 percent of GDP in the first half of 2013.41 The rapid deterioration in the country’s external accounts is due to the declining trend in global commodity prices, poor exports to (stagnant) advanced econ- omy markets, growth slowdown in China, and capital outflows because of the unwinding of quanti- tative easing in the USA, the UK and the Eurozone. These factors have triggered a rapid slide of the real, from R$1.56 per US dollar, in July 2011, to R$2.27 per US dollar in mid-July 2013.42 The ensuing inflationary pressures have led the Central Bank of Brazil to raise interest rates, the state-owned banks to restrict new loans, and the federal government and the state-owned enter- prises to hold back spending and public investment.43 Economic growth has almost stalled, making it difficult to maintain the PT policy of reducing inequality gradually without directly hurting established privileges. These tribulations have been magnified by a relentless press campaign sug- gesting that the government is out of touch, that corruption is more prevalent than usual, and that the economy is spiralling out of control. In the wake of these economic problems, and the demon- strations in June-July, Dilma Rousseff’s approval rates tumbled by more than 40 percentage points in less than three months, down to 30 percent in mid-July.44 The PT and the country have found themselves locked in political confusion. The economic slowdown, coming soon after the country’s good performance in the wake of the global crisis, would inevitably create social and political tensions because of existing dissatisfac- tions, conflicting aspirations, and the shrinking ability of the state to reconcile them. These diffi- culties are compounded by the government’s political limitations in Congress, where the entire (broadly conceived) left controls less than one-third of the seats. This makes it impossible for the PT to govern without alliances with undisciplined right-wing parties and unsavoury individuals, all managed under the gaze of a hostile press and the scrutiny of a right-wing judicial system. (6) A confluence of dissatisfactions The recent protests in Brazil express a confluence of competing dissatisfactions: a) protests against the expansion of citizenship, which has helped to squeeze the middle class, versus b) demands on the part of the workers and the poor for the further widening of citizenship, better public services and improved living conditions. Yet these groups also protest together because of their perception of dysfunctionality and corrup- tion in state institutions, which the mainstream press has highlighted with gusto as if they had been created by the PT.45 664 Critical Sociology 39(5) The outcome is that while the middle class is confused, angry and disorganized, the workers are unhappy for different reasons, marginalized, and also disorganized. Finally, young people are frus- trated and confused in equal measure. These contradictory demands could probably be managed if the Brazilian economy was growing, but its slowdown makes every grievance more urgent, and every constraint tighter. This is a recipe for political volatility, and it poses difficult problems for the country’s political left. Summing up The June-July protest movements in Brazil express deep frustrations across several layers of soci- ety. Their explosive character and the amorphous set of demands put forward by the protesters are, first, symptoms of a social malaise associated with the contradictions of left-wing policy-making under neoliberalism. Second, they illuminate the limitations arising from the achievements of these administrations, including higher expectations of economic performance and public service provi- sion. Third, they reveal the atrophy of traditional forms of social representation (the media, trade unions, NGOs, political parties, and so on), which have become unable to channel discontent and resolve disputes between social groups, either because they are tightly controlled by the elite, or because they have been disempowered by neoliberal reforms (see Morais and Saad-Filho, 2011; Singer, 2009). Furthermore, the Brazilian protests show that, without organization, dissatisfaction – however legitimate, wide and deeply felt it may be – tends to be fruitless, and that spontaneous mass move- ments with a mixed class base and fuelled by unfocused anger can be destabilizing without being constructive. The Brazilian left reacted with maturity to the complex challenges posed by the protests and the attempted kidnap of the movement by the mainstream media and the far right. After a brief period of astonishment, left political parties, trade unions, social movements and NGOs have worked towards the convergence of their sectional programmes and aspirations, espe- cially through common activities and a national co-ordination of movements, organizations and parties.46 They have proposed specific goals for the movement around political reform, the limitation of working hours, state investment in health, transport and education, the decom- modification of public services, the democratization of the media, and reform of the police.47 In other words, they have attempted to identify platforms which can bring together the work- ers and the poor, marginalize and fragment the middle class and the right, and put pressure on the federal government, all the while allowing a radical working class movement to work together with some state institutions in order to raise, from below, its influence over policy formulation and implementation.48 The left ought to stick to this approach instead of falling into an infantile (and, fortunately, mar- ginal) ultra-radicalism, targeting the federal government and, inevitably – because of their insuf- ficient weight – joining, in a subordinate position, the ongoing destabilization campaign led by the political right, the middle class, and the right-wing media. There is no doubt that left-wing admin- istrations tend to implement more progressive policies and be more accommodating of mass move- ments than right-wing ones. If the current government lost coherence and became paralysed because of contradictory pressures coming from the right and the left, or because of the loss of its social base, it would be unlikely to accelerate a socialist revolution in Brazil because there is no ideological, organizational, social, material or international basis for it to happen. It would, instead, directly contribute to a right-wing victory in the 2014 presidential elections, and to the ensuing demoralization and disorganization of the Brazilian left. Saad Filho 665 The response of the federal government to the movements, after considerable hesitation, was precisely to seek left support, and propose a programme of political reforms and expansion of public service provision which could bring concrete gains to the workers and the poor.49 The left should engage in a dialogue with the government, while insisting that a predominantly parliamen- tary strategy to effect those Constitutional reforms is bound to fail. The government should, instead, align itself with the workers’ organizations and the left in order to push through democratic reforms that include state funding for the political parties, the break-up of media monopolies, and improved education, health and public transport services. It is disappointing, but also sobering, to conclude that Brazil is not going through a revolution- ary crisis, and that the current political mobilizations are unlikely to trigger one. Nevertheless, the mobilizations of June-July 2013 were, unquestionably, the most important social movement in Brazil in the last thirty years. The left should continue to fight on the streets, in workplaces and in schools, to broaden and radicalize the movement, bring out the working class with its specific demands, and seek to defeat the right and disorganize and attract part of the middle classes in order to win the coming plebiscite for progressive Constitutional and policy changes. If this can be achieved, it would shift the political balance in the country, and it could lead to concrete long-term gains for workers and the left in Brazil. Acknowledgements This article was completed in late July 2013. It builds upon a note available on http://www.socialistproject.ca/ bullet/851.php. I am grateful to numerous friends and colleagues for their generous comments, especially Gilbert Achcar, Ricardo Antunes, Armando Boito, Maria de Lourdes Mollo, Lecio Morais, Anthony Pereira, Plínio de Arruda Sampaio Jr, Rubens Sawaya, Antonio Carlos Macedo e Silva and Harry West. The usual disclaimers apply. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Notes 1. For a brief history of the movement, see http://tarifazero.org/mpl/ 2. See, for example, http://acervo.folha.com.br/fsp/2013/06/07/15/. Note also the changing tone of the cov- erage in the following days. 3. The Brazilian police are organized, at state level, along two parallel tracks, the civilian police (which inves- tigates crimes) and the military police (which patrols the streets and maintains public order). The federal police investigates specific federal crimes and operates as the border force. Smaller police forces operate in the railways, the armed forces, and so on. This structure was introduced under the military dictatorship. Brazil’s democratic transition did not include significant reforms of the police or the armed forces. 4. See, for example, http://portalimprensa.uol.com.br/noticias/brasil/59405 5. The municipal government of São Paulo is led by the Workers’ Party (PT). The state government is led by its main rival, the right-wing Social Democratic Party (PSDB), and the federal government is led by a broad centre-left coalition headed by the PT. The municipal and state governments agreed the fare increase (the city runs the buses, and the state runs the metro system). Their refusal to take the protests seriously, and the heavy hand of the state police, backfired badly. 6. The Brazilian press is dominated by four virtually indistinguishable national newspapers, two pub- lished in São Paulo (Folha de S. Paulo and O Estado de S. Paulo) and two in Rio de Janeiro (O Globo and Jornal do Brasil). TV ownership is also heavily concentrated, with the Globo network (the parent 666 Critical Sociology 39(5) company of the newspaper with the same name) commanding 32 percent of the audience (half of the entire audience of open TV channels). Other networks include Record (controlled by an evangelical church), SBT and Band. Although their audience has been diluted by the new cable and satellite chan- nels, the main networks retain a virtual monopoly of the national news programmes, which are substan- tively indistinguishable. 7. The most remarkable turnaround was by TV Globo. Their shift was so startling that it was the subject of satire in neighbouring Argentina; see http://www.youtube.com/watch?v=q8tcojHkPjs and http://gilberguessantos. blogspot.co.uk/2013/07/tv-argentina-ironiza-arnaldo-jabor.html; for a critical analysis of this episode, see http:// blogdaboitempo.com.br/2013/06/21/tecnicas-para-a-fabricacao-de-um-novo-engodo-quando-o-antigo-pifa/ 8. See http://noticias.terra.com.br/brasil/cidades/globo-deixa-de-exibir-novelas-para-cobertura-de-protes tos,ffc24201aea5f310VgnVCM5000009ccceb0aRCRD.html; and http://www.conversaafiada.com.br/ brasil/2013/06/20/globo-derruba-a-grade-e-o-golpe/ 9. See http://www.redebrasilatual.com.br/blogs/blog-na-rede/2013/06/a-tv-organiza-a-massa-2773.html. This is the same presenter who, in April 2013, mocked the government’s alleged loss of control of inflation by wear- ing a necklace made of tomatoes in her programme (tomatoes were particularly expensive at that point in time and, for the press, they symbolized runaway inflation). She declared that it was like she was ‘wearing gold’; see http://veja.abril.com.br/noticia/celebridades/ana-maria-braga-usa-colar-de-tomate-e-satiriza-preco 10. The cost of hosting the World Cup has been estimated at US$13b. The World Cup will follow the FIFA Confederations Cup (held in June 2013), and will be followed, in turn, by the 2016 Olympic Games. The government expected that hosting these events would help to consolidate a new image of Brazil on the world stage. This goal has been compromised, especially since Confederations Cup matches were eclipsed by riots in several cities; see, for example, http://esportes.terra.com.br/futebol/copa-das- manifestacoes-deixa-legado-misto-para-brasil,e0abe64f2b59f310VgnCLD2000000dc6eb0aRCRD.html 11. It is significant that the Commander of the Military Police of São Paulo, allegedly speaking ‘in his personal capacity’, suggested to MPL representatives that their demonstrations should target political corruption; see http://ultimosegundo.ig.com.br/brasil/sp/2013-06-17/comandante-geral-da-pm-sugere- politizacao-de-protestos-em-sao-paulo.html 12. For a sample, see http://g1.globo.com/brasil/cartazes-das-manifestacoes/platb/ 13. Most middle-class Brazilians would never consider taking the bus to work; they would also not normally use public hospitals or send their children to state schools. This social group regularly complains that, although they pay high taxes, public services remain very poor. 14. It has been estimated that up to 91 percent of demonstrators heard about the marches they eventually joined on the internet; see http://www.unicamp.br/unicamp/ju/567/o-que-vemos-nas-ruas 15. For an anecdotal account of key demonstrations, see http://www.rededemocratica.org/index. php?option=com_k2&view=item&id=4637:a-ditadura-n%20percentC3%20percentA3o-tem-vez- golpista-no-xadrez. An opinion poll in eight state capitals on 20 June (a day of large demonstrations) suggested that 63 percent of the demonstrators were aged 14–29, 92 percent had completed at least secondary school and 52 percent were students, 76 percent were in paid employment, and only 45 per- cent earned less than 5 minimum wages. In other words, they had attended school for much longer and had much higher incomes than the population average; see http://g1.globo.com/brasil/noticia/2013/06/ veja-integra-da-pesquisa-do-ibope-sobre-os-manifestantes.html and http://thesmokefilledroomblog. com/2013/06/27/who-is-protesting-in-brazil/ 16. ‘There were several demonstrations in the evening of Friday, June 21st and Saturday, June 22nd. At Paulista Avenue, on Friday evening, three demonstrations took place with different demands and, in the absence of any points of contact, they did not engage with one another: the doctors, the “gay cure” [a bill in Congress, supported by the evangelical churches, allowing psychologists to treat homosexu- als seeking a cure for their supposed ‘illness’; the bill was strongly opposed by the left and most social