In Brazil, the first decade of the twenty-first century was characterized by a successful but moderate reformist program spearheaded by its president from 2003-2010, Luiz Inácio Lula da Silva, universally known as Lula.1 His successor hoped to accelerate the project on the wings of a Rooseveltian dream: to create “in just the space of a few years” a country in which the majority could lead “recognizably similar and remarkably decent material lives.”2 What happened? What became of the horizon sketched by Dilma Rousseff  in her inaugural address that beautiful Saturday, January 1, 2011, of becoming “one of the most developed and least unequal nations of the world — a country with a solid and entrepreneurial middle class”?3 What is left of the prediction made by the economist Marcelo Neri, according to which Brazilians — “world champions of future happiness” — would have reason to be optimistic this time, since a “new middle class” would become dominant by 2014?4

There were reasons for hope. Dilma took office having at her back a GDP growth rate of 7.5%, an unemployment rate of 5.3%, and a labor share of income 14% higher than it was in 2004.5 A mass of workers took advantage of privileges formerly available to the middle class alone, such as air travel, dental care, and admission to universities. Brazil appeared to incorporate the poor into capitalist development without a single stone having crossed Brasília’s clear, blue sky. Under Lula’s direction, the Workers’ Party had squared the circle and found the path to integration without confrontation. Acclaimed urbi et orbi, Lula won applause both from the bourgeoisie, national and foreign, and from competing central unions. At the beginning of 2009, Obama declared that Lula was “the most popular politician on earth.”6 In November, under the headline “Brazil Takes Off,” the outstretched arms of Rio’s iconic Cristo Redentorgraced the cover of The Economist in the form of a rocket. In December of 2010, Lula ended his term in office with an 83% approval rating, the highest ever since Datafolha began the survey in the 1980s.7 The World Cup in 2014 and the Olympics in 2016, both in Brazil, were projected as the definitive consecration of Lulismo.

Handpicked by Lula for the post, the economist Dilma Vana Rousseff, ex-Chief of Staff, was not a professional politician and had never run for office. But in the Chamber of Deputies, the Workers’ Party (PT) formed a caucus of 88 seats, the largest in a House of 513, and enjoyed by far the greatest support among the electorate. Working with the big-tent Brazilian Democratic Movement Party (PMDB), the PT promised our first woman president a compliant congress to carry out her mandate. Protected by favorable conditions, the “mother” of the Growth Acceleration Plan (PAC) could intensify the rhythm of Lulismo’s weak reformism and, who knows, in a Rooseveltian decade, give rise to an integrated Brazil, overcoming the rift between the included and the excluded that the independent nation had inherited, nearly two centuries earlier, from the colonial era, and had continued to reproduce ever since. “The eradication of extreme poverty in the coming years will thus be one of my goals,” she affirmed after winning.8

Five years, four months and twelve days later, on a Thursday, May 12, 2016 — the day the president, accused of criminal administrative misconduct, vacated the presidential offices — the dream had become a nightmare. In 2015, the GDP had fallen 3.8%; unemployment had reached the 11% range; income had fallen by 5%; 2.7 million Brazilians had returned to extreme poverty, and nearly 3.6 million to poverty.9 Dilma, with a disapproval rating of 70%, criticized by the left and popular sectors, hated by the right and the middle class, scorned by business, abandoned by her parliamentary base, had been suspended by the legislature and had retired to the presidential residence, which she would leave three months later, removed from office. PT leaders, supposedly involved in embezzlement schemes uncovered by the massive anti-corruption investigation known as Operation Car Wash, were in prison. Lula, charged in several criminal cases, sought to arrange his own defense. The PT had lost nearly two thirds of the support it had enjoyed in March of 2013.10 Lulismo was shattered.

Vice President Michel Temer, of the PMDB, protected by a substantial congressional majority, took over the government as the head of a project that aimed not simply to revoke the integration achieved by Lulismo but to salt the earth from which it had grown: the Constitution of 1988. The new bloc in power wanted to end mandated domestic participation in the development of the massive pre-salt oil deposits, to freeze public spending for two decades, to approve the outsourcing of labor for essential services, to pass labor reforms that would gut the labor code, to approve a constitutional amendment that would limit social security benefits and, if possible, to reorient politics toward a parliamentarism that would demobilize national consituencies.Temer nominated a cabinet inclined to decrease the number of families served by the Bolsa Família guaranteed income program; toreduce funding for health, primary education, public universities, and family farming; to slow down investigations into slave labor; to stop the demarcation of indigenous lands and the recognition of property rights for descendants of maroon communities. The desire was to revoke what had been built, since the end of the dictatorship in the 1980s, on the basis of the democratization of society, the reinvention of politics, and the welfare state.

On a Wednesday afternoon, August 31, dressed in red, surrounded by ex-ministers, representatives, senators, and community leaders, Dilma delivered her last speech from the presidential residence, where since May she had awaited the Senate’s vote. At 1:36 pm, the Senate had proclaimed, 61 to 20, that she was guilty of infringing the sixth clause of Article 85 of the Constitution: violating the budget law. “It is the second coup d’état I’ve faced in my life. The first, the military coup, supported by the savagery of arms, of repression, and of torture, affected me when I was a young militant. The second, the parliamentary coup delivered today by means of a juridical farce, deposes me from office to which I was elected by the people.”11 Recorded by documentary filmmakers, the ex-president spoke to history. The senatorial toga of 2016 stood in for the tanks of 1964. Dilma Roussef for João Goulart. Lula for Getúlio Vargas. The PT for the old PTB, PSDB for UDN, PMDB for PSD.12 The lawyer Michel Miguel Elias Temer Lulia for the field marshal Humberto de Alencar Castelo Branco. If all went well for the coup, the PT would be out of the way for at least a decade, just as the coup of 1964 put the brakes on the growth of its precursor, the PTB, itself a product of the earlier era’s political realignment. The popular party would only reenter the lists under cover of the official opposition party in 1974.

But a parliamentary coup is not a coup d’état, which “in the vast majority of cases” involves a seizure of power by the armed forces.13 The process of impeachment, replete with dramatic reversals, had been approved by the Chamber of Deputies on April 17, after four months of disputation, public and free, between prosecution and defense. During a session that lasted nine hours and forty-seven minutes, televised in its entirety, 367 of 513 representatives voted in favor of impeachment, each making a fifteen-second declaration. The president of the House, Eduardo Cunha (PMDB-RJ), asked “that God have mercy on this nation.” The television clown Tiririca (PR-SP) voted “for my country.” The gaucho Sérgio Moraes (PTB-RS) wished a “happy birthday to Ana, my granddaughter.” The ex-army captain Jair Bolsonaro (PSC-RJ) evoked the “memory of Coronel Carlos Alberto Brilhante Ustra,” a torturer during the military dictatorship. A clumsy variety show was interrupting the lulista sequence, all within the limits of the law even as it struck at the heart of the Constitution.

The 137 lawmakers opposed to the impeachment had the right to use their fifteen seconds to defend the government, the legitimacy of the popular vote that reelected Dilma in 2014, the personal honesty of the president; to pay tribute to Leftist martyrs such as Luís Carlos Prestes, Olga Benário and Carlos Marighella; and, above all, to attack the corruption of Eduardo Cunha, the man presiding over the session with a permanent ironic smile on his lips. Eighteen days after the impeachment, he would be relieved of his duties by the Supreme Federal Court; in September, expelled by the Chamber; and in March 2017, sentenced to fifteen years in jail for corruption, money laundering, and tax evasion.

In the Senate, the session that culminated in Dilma’s impeachment lasted seven days. For three days, presided over by the minister Ricardo Lewandowski of the Supreme Court, the prosecution and defense presented arguments. The president defended herself in a 47-minute speech and during a marathon thirteen hours responded to questions from 47 senators.14 At the last moment, in special deference, the senators decided, 42 to 36, notto ban her from running for office for the next eight years. Eight PMDB senators, including two ex-ministers and the father of a third, scandalously voted against Dilma, but supported preserving her right to run for office, delicately differentiating themselves from their colleagues in the Chamber. Temer, who personally led the fight to impeach Dilma, had to promise positions in state-affiliated banks, utility companies, hydroelectric projects, the transport ministry, and so on, to get the 54 votes he needed.15

The coup took place withinthe limits of the Constitution: a postmodern coup, according to the political scientist Bernardo Ricupero, although one nourished by Brazil’s old backwardness.16 The PMDB and the PSDB had united, without any proof of misconduct on the part of the president, to remove the PT from the executive branch, after it had obtained, in 2014, its fourth consecutive victory in presidential elections. The Folha de São Paulo, a newspaper with little sympathy for the deposed government, noted that the evidence presented did not meet the standard required by the Constitution: “Although there are reasons to impeach, not least because [Article 85] establishes a wide range of options, none of these is irrefutable. Not that there is no evidence of misconduct; what has been lacking, so far, is definitive proof. Fancy footwork with the books is a questionable reason in a permissive budget culture.”17

A constitutional maneuver had been undertaken to distort the spirit of the law. In the name of the Law of Fiscal Responsibility (LRF), loans from public banks to the Treasury, issued under the direction of the president, and delays in Treasury’s payments to the Banco do Brasil, which she did not approve, were described as criminal misconduct. These were pretexts, since such loans were part of administrative routine until October 2015, when the Federal Court of Accounts (TCU) declared the practice censurable. The senators who charged Dilma with corruption sought to apply the decision retroactively, which makes no sense. The delayed payments to the Banco do Brasil depended not on her but directly on the Treasury.18

It has been observed that impeachment constitutes a new type of instability in Latin America, replacing the old military coups. The Argentine political scientist Aníbal Pérez-Liñán has listed seven Latin American impeachments from 1992 to 2015.19 According to Kathryn Hochstetler, the impeachments that followed the era of military coups were as often the result of popular protests against neoliberal policies as of pressure from elites to check progressive governments, constituting a quasi-parliamentary mechanism for ousting a government.20 This new mechanism does not necessarily entail, as it had in the 1960s and 1970s, a loss of democratic liberties. In the case of Brazil, the instability created by the parliamentary coup threatened democracy, but did not dissolve it.

A Brazilian Rag-and-Bone Shop

In 2002, when Lula won the presidential election for the first time, the Constitution was secure and democracy in full swing. How was it possible, to borrow an expression from Marx, that “Society now seems to have fallen back behind its point of departure”?21 How to explain a counterrevolution without a revolution? What hidden connections would allow for the formulation of hypotheses capable of explaining the catastrophe?22 Is class struggle the key to the enigma? I believe it is, though it is necessary to point out that, contrary to the expectations of the Communist Manifesto, according to which “the epoch of the bourgeoisie … has simplified the class antagonisms,” the contemporary scene is marked by the conspicuous fragmentation and complexity of that conflict.23 Rather than deal solely with bourgeoisie and proletariat, it is necessary to take account of divisions whose taxonomy the analyst does not find ready to hand. Transnational bourgeoisies, rent-seeking industrialists,trade unions of contract laborers, a precariat with access to the university, poor entrepreneurs, agro-ecological peasants; the social management of extreme poverty, the active role played by professionals belonging to the judicial and media apparatuses, and so on. In the rag-and-bone shop of the twenty-first century, one must make use of whatever is at hand to generate interpretations. Hence a functionalist use of the term “class,” defining them à la Weber in terms of access to goods and status (as when we speak of a middle class), will appear in a text, like this one, that seeks inspiration in the Marx of the EighteenthBrumaire. As far as I know, there is no general theory of class that gives an account of the rapid processes currently underway.

Beyond reference to the middle class, then, it will be necessary to mobilize the category of “the poor,” which refers to a lack of access to goods, since it is fundamental for understanding how Lulismo was torn apart. Following a sequence that Francisco de Oliveira defined as the “reinvention of politics,” during which class directly occupied the scene — roughly the decade that extends from 1978 to 1988 — poverty began to become a politicalcategory in Brazil.24 While Lulismo was forged in the organized fraction of working class, it has addressed itself since 2002 mainly to “the poor.” In so doing, it relinquished the advances represented by a class orientation — the proletariat having, as Marcuse would say, “decayed into the generality of the working masses” — but it touched a nerve of the peripheral formation.25 Lulismo is, therefore, profoundly contradictory, and because it is regressive and progressive at the same time, it lends itself to countless mystifications.26

The essential characteristic of the poor in Brazil is that, as Caio Prado Jr. showed, they have no place in the organized nucleus of production.27 In the colonial era, the capitalist engine was composed of masters and slaves, linked by the export economy to the most advanced world economies, and therefore to class. But, as Prado Jr. demonstrated, what predominated in the colonial era was the “inorganic” character of sectors that did not have a place in the capitalist engine. Without a defined role, they orbited around the dynamic center, offering here and there whatever service they could, in a relationship of favorthat, as Roberto Schwarz discovered later, provided the dominant class with considerable latitude.28 Celso Furtado, in turn, shows that, after independence, “the existence of a relatively amorphous labor pool, which had been accumulating in the previous centuries, was of fundamental importance for the development of the new economic system based on wage-earning labor.”29 Into this pool flowed the stratum of former slaves, who became part of the poor when the mode of production substituted free immigrants for slave labor. In the same key, the sociologist José de Souza Martins affirmed that the “groups that remained on the margins of dominant processes, abandoned and discarded by the State’s lack of an extensive, integrative and participatory political project” defined Brazilian anomie.30

The intermittent pattern of the poor’s activity impedes their self-identification as workers, even though that is what they in fact are. They are a class in itself, but not for itself. It is common, nonetheless, in popular speech, to refer to “the poor,” as those who have a sense that their interests are opposed to those of “the rich.” Within the realm of politics, the opposition between rich and poor tends, simultaneously, to refract and obfuscate the subsistent conflict between capitalists and workers. In a skewed manner, the refraction allows one to see, under the labels of “poor” and “rich,” the fundamental conflict — even as it elides the effective center of the antagonism: the ownership of the means of production. It is this dual refraction and obfuscation that makes a class analysis of Lulismo, and therefore of the turbulent and entangled process that enveloped it between 2011 and 2016, difficult.

To arrive at a better understanding, it is necessary to take up again the concepts of the massesand class, the masses being the form of appearance that class takes in politics when it is not organized as a class. In Marx’s writing, “the small-holding peasants form a vast mass, the members of which live in similar conditions but without entering into manifold relations with one other. … Insofar as millions of families live under conditions of existence that separate their mode of life, their interests and their culture from those of the other classes, and put them in hostile opposition to the latter, they form a class. Insofar as…the identity of their interests forms…no political organization among them, they do not constitute a class.”31 In the Brazilian case, the masses are composed of “the poor,” which is the name given to the subproletariat, the segment of workers that do not reach the condition of the proletariat, as I suggested, following Paul Singer, in Os Sentidos do Lulismo.32

The Communist Manifesto’s prediction that the masses would be absorbed by class did not come to pass. Various authors saw, from the periphery, that the masses were reproduced in parallel with the development of productive forces. Antonio Gramsci, considering the Italian situation, discovers an interesting approach. “What does the Italian situation consist of, according to this formulation?” the Communist leader wondered. “It consists in the fact that demographic growth contrasts with the relative poverty of the country, in other words, in the existence of a surplus population.” At first, Gramsci thinks this is a matter of “a parasitic population, that is, a population that lives without in any way intervening in the productive world,” but soon thereafter, reflecting in 1934-35, wonders if it would not be more advantageous to Italian industrialists to place their bets “with low labor costs and governmental privileges rather than with technically developed production.”33

Gramsci raised questions that would reemerge in South America: could it be that surplus populations seemingly disconnected from the systemcompose, with the productive sectors, another system, invisible, but not for that reason any less connected? As Francisco de Oliveira pointed out during the economic miracle of the 1970s, “Pertinent and necessary from the point of view of a mode of accumulation that was beginning or seeking to gain strength was the conversion of enormous swathes of the population into a ‘reserve army’ adequate to the reproduction of capital.”34 As he noted, with respect to a third peripheral trajectory, that of Russia, “the very incompleteness of the system represents a new level of complexity, one that will only be understood well into the twentieth century by a line of Latin American thinkers such as Raúl Prebisch, Celso Furtado and Florestan Fernandes.”35 According to Oliveira, “The Russian transition from feudalism to capitalism results in a hybrid system that will never come to term, combining the ferocity of the new with the backwardness of the old.”36 The periphery reinvented Marx in order to understand itself. As Schwarz observed, in peripheral situations “a consequent Marxist inspiration would involve a certain displacement of the classical problematic of Marxism itself, requiring us to theorize historical experience with our own heads.”37

Oliveira continuously calls attention to the fact that Brazil’s “incomplete” peripheral capitalist system — which he christened the “platypus” — functions well from the point of view of accumulation.38 In 2014, Brazil’s GDP was larger than that of India, Russia, and Italy, despite our lost decades. Perhaps this is a specific variant of what Barrington Moore Jr. conceptualized as conservative modernization.39 Specific because, as Oliveira pointed out, there is “no pre-capitalist residue” here.40 There remain in Brazil remnants neither of feudalism, nor of Hindu castes, nor of Russian peasant communes, but rather, as I see it, of the foundations of a mercantile-slave society connected to the capitalism of the coreeconomies that dominated the 300 years of colonization. One of the peculiar traits of this society is the structural limbo from which the poor can escape (and into which they can fall again) individually, but never as a class.41 In other words, some of the poor may cease to be poor, but poverty cannot cease to exist. As a result, the masses are the majority. This means that Lula could not square the circle, and that Lulismo, sped up by Dilma in the bosom of Rooseveltian ideology, foundered on its own contradictions, which are just as much the contradictions of Brazil. Although a quarter of the population still remained in poverty in 2014, there was a path from subproletariat to proletariat — and this put pressure on Brazilian capitalism’s conditions of reproduction.

By integrating surplus populations, Lulismo was diminishing the reserve of labor power. One symptom: the shortage of domestic workers between 2011 and 2013. The employment of domestic help forms a distinctive part of the lifestyle of the “modernized” portion of this society: with 7.2 million, Brazil was the country with the highest number of domestic workers in the world.42 Apart from the shortage, the extension of labor rights to domestic workers in April 2013 — with overtime, limitations on the workday, and extra pay for working at night, among other benefits — slowly raised the value of the domestic workers that remained. The film Que horas ela volta?, released in 2015 (by which time the situation had already changed), portrays that moment. But domestic work is only an example of the larger and more central process that was underway. Lulismo did not intend to confront the ruling classes. But by reducing poverty, it did so unintentionally.

A massive surplus population, lowering the value of labor, is what allows the modern sector to function. It is intuitive, but false, to imagine that backwardness holds back or sucks down the modern sector; in fact it is the opposite. “The specificity of the productive revolution without bourgeois revolution was the ‘productive’ quality of backwardness.”43 The reality is counterintuitive: the limbo functions as the atmosphere from which the modern draws the air it breathes — or better, the labor that feeds it. With superabundant labor power at its disposal, the Brazilian “modern” sector is dynamic and large enough to speculate on the housing market in Manhattan, or to number as the second largest community of Facebook users in the world, behind only the United States.

The political consequence is that the modern sector is large enough to impose vetoes on systemic change, since the backwardness of a large part of society is the precondition of the modern part punching above its weight. However paradoxical it seems, what paralyzes advancement is not backwardness, but the size of the modernized sector. I believe that this suggestion would be the political complement to Oliveira’s suggestion that in Brazil, “as a result of the elevated extraction of absolute and relative surplus value, overextraction of surplus value fulfills in the system the function of maintaining an overaccumulation, which is necessary for real accumulation to take place.”44 The success of the platypus implies the existence of an influential middle class party. If that party makes common cause with the party that manages the country’s backwardness — the party of the interior — together they will block the steps toward integration promoted by the popular party.

The point is to establish a link between the Brazilian party system and its class dynamic, believing that without such a link it would not be possible to explain the collapse of Lulismo. I begin from the premise that the Brazilian party system is comprehensible only if we take into account the dialectic between modernization and backwardness. My hypothesis is that the three “real” major parties, from 1945, when Brazil becomes a mass democracy, until 2016, are from a certain point of view the same even if the names have changed.45 They cross the modern sector and the backward one, resulting in a bipolar opposition between a popular party and a party of the middle class, but mediated by a party of the interior, in which patronage and clientelist relations prevail. Most of the time, the popular party and the party of the middle class, while acting in a populist milieu, (re)present the actually existing class struggle. At times, the clash between capitalists and workers, that is, between right and left, becomes central, as was the case during the decade of the “reinvention of politics” (1978-1988), but the strong presence of the subproletariat and the consequently elevated role played by the middle class tends to push agents toward a polarization of society between rich and poor, a polarization that was transfigured, from 2006 on, into the opposition between Lulismo and anti-Lulismo.

Such a structure produces a problem for democracy. Once the electoral realignment has taken place, the middle-class party, whose power does not lie in numbers, will have difficulty winning presidential elections. This establishes a radicalized dynamic at the expense of the popular party, an atmosphere that encourages the tendency to stage a coup. The delegation from the interior, endowed with ample ideological mobility on a spectrum that extends from center to right, since its base is notrepresentedbut rather attended to, goes whichever way the wind blows; it is as capable of stabilizing democracy, as the PSD did in 1961, as of destabilizing democracy, as the PMDB did in 2016 (and the PSD did in 1964). Without understanding the party-electoral system it becomes impossible to arrange the pieces that allowed the shift from stability to instability in the period of 2011-2016.

In Brazil, weak reformist positions like Lulismo have a dangerous content, since they involve slowly shrinking the limbo. While they do not threaten the capitalist system, they affect the platypus’s mode of reproduction. If someone, like Dilma, resolves to quicken the pace of reform, the situation becomes even tenser, because objective conditions are created in which defenses of labor and society à la Polanyi — regulating and restricting the market’s tendency to destroy workers and social bonds — are suddenly on the table.46 In a highly contradictory manner, the events of June 2013, in their leftist aspect, pointed in an anti-business direction.47 Because it laid claim to greater social investment, pressure from the left was oriented toward containing the market. Hence it activated, in the opposite direction, energies destined to undo — via outsourcing, labor reforms, and a freezing of public spending — the advances obtained during the lulista decade.

The Dilma Moment

When Lula won in 2002, he perceived a window of opportunity for a weak reformism thanks to the commodity boom, and he took advantage of the opening in an effective manner. He raised the real value of the minimum wage; generated millions of jobs; created a payroll loan program, a national drug benefit, and a guaranteed income for poor families; extended social security, federally funded college scholarships, and home ownership subsidies for low-income workers; promoted and expanded, among other things, tuition support for low-income students, the construction of cisterns in the semi-arid northeast, the recognition of property rights for descendants of maroon communities, and credits for family farming. Although more than 90% of the jobs created were low wage, average income rose close to one third between 2003 and 2014, thanks to collective bargaining agreements favorable to workers.

Encouraged by the political capital amassed by Lula, Dilma took seriously the idea of accelerating the rhythm of the reformist venture, simultaneously initiating both a developmentalist approach to political economy and a republican action against networks of corruption embedded in the State. The orientation underlying Dilma’s economic strategy can be summarized in four points adapted from Ricardo Bielschowsky’s description of the developmentalist school: 1) Wholesale reindustrialization would provide a path to the end of poverty; 2) There is no way of achieving reindustrialization in Brazil by means of the spontaneous forces of the market; 3) The State must plan the process; 4) Such planning should coordinate the desired expansion of economic sectors, the instruments for promoting that expansion, and its execution.48

Meanwhile, the orientation underlying Dilma’s “ethical clean sweep” can be summarized in three ideas: 1) The republican State is defined by a public sphere that must be immunized against private influence; 2) This immunization reins in corruption in the state apparatus; 3) Corruption is understood not as moral decadence of public officials, but specifically as the theft of public assets.49 In Dilma’s vision, only a republicanized State would be capable of reindustrializing Brazil.

Rather than merely giving continuity to what Lula had accomplished from 2003-10 and allowing him to run again in 2014, the president had thus opted for a relatively autonomous direction, pushing Lulismo a bit closer to a strong reformism, although within the limits of a transformation from above — a reformism, that is, without a mobilization of society. I christened these two orientations the developmentalist experimentand the republican experiment, to allude to projects that would not be completed. But the reactions they produced, fomenting both an anti-developmentalist front and an anti-republican front, form part of the explanation for the impeachment.

Between 2011 and 2013, Dilma took on interest rates, rules concerning savings accounts, the interest spreads of private banks, the exchange rate, the capitalization of the national development bank, and the regulatory framework of the electrical sector.50 The aim was to draw private investment to the productive sector in such a way as to leverage economic growth with the reindustrialization of the country. These decisions responded to the worries expressed by a productivist coalition composed of an organized working class collaborating with industrial leaders. During the first half of 2011, that coalition complained about the growing “reprioritization of the export agenda,” of the “substitution of imported products and industrial inputs for domestic production” and of the “significant decline of national content in production.”51 In brief, they pointed to deindustrialization. The government decided to devalue the currency, to tax speculative capital, to reduce interest rates and to pressure private banks, by means of state banks, to reduce their rates, financing internal production and consumption. This combination of measures would explain the ferocious campaign against “state interventionism” that crashed down on Dilma beginning in 2012.

Dilma also made key decisions concerning the presence within the government of ministers accused of embezzlement; the problem of crucial posts within the public sector being filled on the basis of patronage; and the PMDB, the party most associated with clientelism. Although the President had expressed discomfort with the expression “ethical clean sweep,” utilized by the press to describe her policies for combating corruption, the actions that the Administration undertook during 2011-12 focused on areas that managed a significant volume of resources, signaling a new standard for public conduct. Very old practices were checked, producing harsh reactions from those affected. In particular the PMDB, supported by a growing bloc of rent-seeking representatives, challenged the president on every important legislative vote between 2011 and 2014.52

Despite pushback, the outcome was positive until the first half of 2013. In January, after announcing the reduction of the price of electricity, the President had the support of the Federation of Industries of the State of São Paulo (Fiesp). Its president, Paulo Skaf, declared in a communiqué: “Dilma has demonstrated sensibility, and her concrete actions demonstrate the government’s concern for the competitiveness of the country.”53 With the “ethical clean sweep” the President had also gained the approval of the modernizing sector. In an article published in piauíin September 2012, the sociologist Celso Rocha de Barros observed “that many middle-class supporters of [centrist and former PSDB presidential candidate José] Serra have become enthusiastic for Dilma.”54 In October of 2012, Dilma visited São Paulo in support of Fernando Haddad’s candidacy for mayor of a capital city that the PSDB, the party of the middle class, had governed for eight years. “You can’t run Brazil without sticking your nose into São Paulo,” the President said from the platform, in the city that would later be the epicenter of the impeachment.55 On the 28^thof that month, Haddad would beat that same José Serra in the mayoral race with 54% to 46% of valid votes.

In March 2013, crowning the success of her first two years in office, Dilma received a 65% “good” and “excellent” rating in Datafolha’s survey. In June, however, the situation suffers an unexpected reversal. Like a lightening bolt from the blue, massive street protests invert the trend. Different groups, on the left and on the right, dissatisfied with the situation, take to the streets, delivering a “substantial [blow] to institutional credibility.”56 The government’s approval rating falls to 30%. Fiesp initiates an anti-tax campaign, which will later become imbricated with the impeachment campaign, both symbolized by a yellow duck, 12 meters high, emblazoned with the slogan “We won’t be played for suckers.”57 A young opposition movement emerges on the left, which will lead to the occupation of high schools in 2015 and 2016.

The immediate causes behind the events of June remain a mystery. The “mensalão” scandal of 2005, in which the PT had used public funds to secure the support of corrupt incumbents, had reached judgments in court at the end of the previous year. Unpopular fare hikes for public transportation, some inflationary pressure — the economist Marcos Lisboa recalls that “the inflation of food prices exceeded 10% that year” — and plenty of criticism from the media, especially of “interventionism,” were all part of the environment that preceded the explosion.58 In isolation, none of these explain the size of the demonstrations. The protests originated in proposals made by leftist groups outside the PT that sought to connect with the new working class. The relatively limited mobilization for the revocation of fare increases on buses, metro, and trains in São Paulo city spilled over after an unusually brutal police repression on June 13. Suddenly, the protests included millions of people in hundreds of municipalities, for the most varied reasons.

Although the events of June had begun as a left opposition to Lulismo, the center and the right would take the issue of corruption to the streets, mobilizing crowds dressed not in red but in the national colors of green and yellow. The uprising of the middle class began at that moment, allying itself with Operation Car Wash, which had been initiated in March of 2014. The connection needs to be studied further, but it is worth mentioning a fact noted by the journalist Eugênio Bucci, according to whom “the fight against corruption resonated neither with the Presidential Palace nor with the National Congress, but rather with Operation Car Wash.”59 Despite Dilma’s republican experiment, the leadership of the federal judge Sérgio Moro picks up the yellow thread unraveled in June and takes it all the way to impeachment. The effects of Operation Car Wash, whose effectiveness cannot be understood without the alliance it established with the media, are both partisan and republican. On one hand, it catalyzes the anti-lulista mobilization that began in June of 2013 and concluded with the largest pro-impeachment demonstrations, on March 13, 2016. On the other, Operation Car Wash uncovered the center of the corrupt system that financed Brazilian politics since 1945, a system that Dilma’s own republican experiment had begun to dismantle.

June represented, nonetheless, a break in the period of five and a half years that Dilma governed Brazil. In the face of these difficulties, the president became erratic. Seeing her cornered, the jaguars provoked by Dilma’s experiments go on the offensive, and the president, without a base from which to defend herself, zigs and zags, becoming increasingly isolated. In the economy, she calls for exemptions to the payroll tax, whose results were negative. Then, in the 2014 election, she runs on a developmentalist platform that had already lost the support of the industrial sector. Contradictorily, during the race she fires Guido Mantega, the minister who had managed the developmentalist program. Reelected, she does the opposite of what she had promised and adopts the neoliberal playbook, nominating Joaquim Levy for the Ministry of the Economy. She loses the support of the left and the lulista base.

The sudden reversal of 2015 undoes the advances of Dilma’s first mandate. The statistician José Eusáquio Diniz Alves asserts that the number of people living in extreme poverty rose from 7.9% to 9.2% in that year.60 Despite the fact that the crisis created a need for more protections for those on the bottom, funding for the Bolsa Famíliawas frozen, and the number of beneficiaries, which had grown consistently since 2004, stagnated.61 Unemployment rises 38%, expelling nearly 3 million people from the labor market just as unemployment benefits, sick pay, and life insurance benefits suffer cuts.62 Informal labor returns, with an increase of 4.6% in the number of self-employed workers without recognized rights, after the increase in formal employment from 40% to 51% of the economically active population between 2002 and 2012.63 The reversal leaves 3.7 million people outside of the lowest ranks of the consumer classes.64 Part of the new working class returns to the status of subproletariat. As if that were not enough, in 2016, Dilma even proposes a public debt ceiling and pension reform, establishing a minimum age for retirement.65

In a third erratic decision, the President, although weakened by the previous changes in direction, resolves to carry on the duel with the PMDB, confronting the corrupt right-wing evangelical deputy Eduardo Cunha at the head of an enormous bloc dissatisfied with the republican experiment. When elected to president of the Chamber on the first of February, 2015, he will open the door to the impeachment process, as stipulated in Article 51 of the Constitution. Having broken with the left on the economy, Dilma had no base with which to fight the right in politics. It is well known from the President’s biography that she was a hero of the resistance to the military dictatorship. Imprisoned and tortured, she never talked. Perhaps that has something to do with the “I will break before I bend” attitude that characterized her unwillingness to put up with the rent-seeking power brokerage of PMDB politics as usual. Lula grounded himself in the opposite motto: “I bend, but I will not break.” These are antithetical orientations, both necessary for political action, but encouraging different tendencies in each situation. Lula’s privileges effectiveness. Dilma’s sticks to principles, to the flame of faith, whose potency lies, in the words of Walter Benjamin, in the “depths of time.”66

In the strategic decisions of her second mandate, Dilma parted with Lula, recognized as one of the sharpest politicians on the scene. Why did Dilma, faced with such an adverse scenario after the turnabout of 2013/2014, decide to run for a second term? From May 2012, when he had recovered from throat cancer, Lula indicated that he was willing to run for office, but the President never offered him the spot. On the contrary, irritated with cries of “Come back, Lula!” that brought together businessmen, politicians, and labor organizers, she redoubled her efforts for reelection. When she decided, in October 2015, to start taking Lula’s advice, it was too late.

The erratic character of Dilma’s actions after June opened the door to anti-Lulismo. Gears that had waited for the opportunity since 2003 were set in motion. An ambitious power coalition, led by Michel Temer and José Serra (now a PSDB senator), had been quietly mustering since at least August 2015. Resentful of their fourth consecutive loss in presidential elections, the PSDB supplied the elements for undertaking the parliamentary coup. Although the PMDB had played a leading role and assumed the presidency, it was the PSDB who provided the legal framework, the economic program, the bridge with business community, and legitimacy in the eyes of the middle class — to say nothing of the hundred votes that the PSDB galvanized in the legislature. Meanwhile the mobilization triggered by Operation Car Wash camouflaged, via the media, the demolition of the social safety net.

The popular segments did not come to Lulismo’s defense—which was to be expected, given the depoliticization and demobilization to which they had been submitted. As the gap between the continents of rich and poor narrowed slightly, the traditional middle class gave signs of growing irritation. But Lulismo had not taken care to make its beneficiaries aware that, sooner or later, there would be a reaction against policies that aimed to reduce poverty and destitution.

Authorized by the Chamber on April 17, 2016, Dilma’s prosecution needed to be approved by the Senate, which it was on May 12, as had been expected. That afternoon, Michel Temer would assume the Presidency of the Republic, pushing the historical process backward. On August 31, the Senate would ratify the nightmare, removing a president who had arrived at the Presidential Palace on the wings of Rooseveltian ideology. Roosevelt’s New Deal emerged in the capitalist center during a period of dominant Keynesianism. Applied to Brazilian material in times of globalization and neoliberalism, it tore Lulismo apart, driving society to who knows what distant shore, far from the egalitarian aspirations that the Rooseveltian dream had crystallized.

André Singer is Professor of Political Science at the University of São Paulo. He was Press Secretary for the Palácio do Planalto from 2005-2007 and, from 2003-2007, spokesperson for President Luiz Inácio Lula da Silva. His books include Esquerda e direita no eleitorado brasileiro(EdUSP, 1999), Os Sentidos do Lulismo(Companhia das Letras, 2012), and O Lulismo em crise (Companhia das Letras, 2018).

Emilio Sauri is Associate Professor of English at the University of Massachusetts Boston; Nicholas Brown teaches in the departments of English and African American Studies at the University of Illinois at Chicago.

See Robert Castel, El ascenso de las incertidumbres: trabajo, protecciones, estatuto del individuo (Buenos Aires: Fondo de Cultura Económica, 2010).