SciELO - Scientific Electronic Library Online

vol.13 número1Estados e investidores estrangeiros: é possível alcançar cooperação?O backlog de patentes no Brasil: o direito à razoável duração do procedimento administrativo índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados


Revista Direito GV

versão On-line ISSN 2317-6172

Rev. direito GV vol.13 no.1 São Paulo jan./abr. 2017 


The ballot under the bayonet: election law in the first years of the Brazilian civil-military regime (1964-1967)


Leonardo Augusto de Andrade Barbosa1 

1 Center for Continuing Education and Professional Development (CEFOR), Chamber of Deputies Brasília – Federal District – Brazil


This paper analyzes reforms in election law introduced by the civil-military government instituted in Brazil following the 1964 coup-d’état. After a brief background on the issue, it focuses on the tenure of President Humberto de Alencar Castello Branco (April 1964 to March 1967), particularly on the modification of the ineligibilities legislation aimed at the state elections of 1965. The trajectory of Sebastião Paes de Almeida – an oppositionist representative who was prevented from running for the office of governor of Minas Gerais on the grounds of the newly enacted legislation – provides the opportunity to assess how democratic institutions worked and how political rights were interpreted under a dictatorial regime that paradoxically relied on elections as a legitimizing strategy.

Key words: Political rights; political disqualifications; state elections; Brazilian civil-military dictatorship; authoritarianism


Este artigo analisa reformas na legislação eleitoral adotadas pelo governo civil--militar instituído no Brasil após o golpe de 1964. Depois de uma rápida contextualização sobre o tema, o texto se concentra no mandato do presidente Humberto de Alencar Castello Branco (abril de 1964 a março de 1967), particularmente nas modificações na legislação de inelegibilidades que regeu as eleições estaduais de 1965. A trajetória de Sebastião Paes de Almeida – um deputado oposicionista que foi impedido de concorrer ao governo de Minas Gerais baseado na legislação recém-adotada – serve como pretexto para avaliar como instituições democráticas operavam e como direitos políticos eram interpretados sob um regime ditatorial que, paradoxalmente, recorria a eleições como estratégia de legitimação.

Palavras-Chave: Direitos políticos; inelegibilidades; eleições estaduais; ditadura civil-militar brasileira; autoritarismo

1 Democratic practices under authoritarian rule?

One of the most puzzling aspects of the Brazilian civil-military dictatorship that came to power in 1964 was the continued presence of elections and electoral debate in the political landscape (SANTOS, 1999, p. 218). Unlike Chile and Argentina (PEREIRA, 2005, p. 121), and the previous Vargas dictatorship in Brazil, the 1964 regime did not shut down the legislative branch, although Congress was eventually purged and forced into adjournment on three occasions by presidential decree.

The first official act of the civil-military regime was to call presidential elections. In the words of Milton Campos, the future Minister of Justice, “the revolution was born under the sign of elections.”1 And yet the country experienced a few days of institutional ambiguity after President João Goulart’s deposition. The presidency was declared vacant in the first hours of April 2 and Ranieri Mazzilli, the Speaker of the Chamber of Deputies, took office, as prescribed by the 1946 Constitution, making him the constitutional president. However, as Mazzilli took his seat in the presidential chair at the Planalto Palace, the victorious military leaders were deciding the fate of the country elsewhere. On April 9, the self-proclaimed “Supreme Commanders of the Revolution” [Comando Supremo da Revolução] enacted an “Institutional Act” [Ato Institucional] that provided a course of action (BRASIL, 1964a, p. 3.193).

According to the 1946 Constitution, as more than half of the presidential term had already elapsed, the new president should be elected by Congress in an indirect election held 30 days from the vacancy. Although the Constitution remained in place, the Institutional Act allowed for adjustments so that the Constitution could better meet the “revolutionary” needs. Regarding the provision about the election after a vacated presidency, the Act bypassed the Constitution and determined that elections would be held only two days after its proclamation (therefore, nine days after the vacancy). It also declared that the constitutional criteria yielding ineligibility would not apply to this election.2 This last detail is critical, since Article 138 of the Constitution prevented military officials, such as the soon-to-be-president-elect Humberto de Alencar Castello Branco, to run for elected office.

The casuistic manipulation of election law was not only woven into the fabric of the Brazilian authoritarian regime, it was its predominant political strategy.3 The government resorted to Congress whenever possible, but circumvented it (through the use of force or through an appeal to the omnipresent “constituent power of the revolution”) when necessary (BARBOSA, 2012, p. 49 ss.). Although Castello Branco sponsored the enactment of an Election Code in July 1965, hinting at a desire for lasting election regulation, he and his successors never ceased to introduce casuistic adjustments into the constitutional text or the pertinent statutes. There were several waves of election law reform, modifying every relevant aspect of the legislation. Indeed, it was a time of high-tempo constitutional politics: Between April 1964 and October 1965, nine constitutional amendments were enacted, five of which were related to the elections. These amendments were only able to pass, we should note, because the Institutional Act had lowered the required quorum for constitutional reform (BARBOSA, 2012, p. 66).

Throughout the civil-military regime there were many experiments with electoral legislation, including the adoption of the Uruguayan institution of sublema for elections for the offices of mayor and senator, as a way of circumventing strong political opposition by cunningly transforming majority elections into proportional elections (COSTA PORTO, 2002, p. 346-347); the creation of what skeptics referred to as “bionic senators” [senadores biônicos], elected not by popular vote, but by an Electoral College formed by members of the state legislature and delegates from the municipalities (BRASIL, 1977); limitations on electoral campaign advertisement, curbing any form of real political debate in the media (ALVES, 2005, p. 229); the prohibition of private corporations’ money for campaigns as a way of impairing the ability of the opposition to raise funds and gain momentum among disgruntled companies (BRASIL, 1971), among others.4

Although such manipulations were omnipresent during the 21 years of dictatorship, Castello Branco’s tenure remains a privileged moment for observing the paradoxes of election law under authoritarian rule. Indeed, the regime defined its approach towards election law during his term. As Carlos Fico notes, historians agree that although there was an absence of detailed plans in the beginning of the dictatorship, there was no doubt about the government’s motivations in this early period. Quoting Schmitter, Fico argues that the military progressively and systematically sought to reinforce a “repressive apparatus” that would eliminate any hurdles to the achievement of the so-called “permanent national objectives” (FICO, 2004, p. 75). Manipulating the rules governing elections was one of many tools within this apparatus, and one that would prove relevant until the final moments of the political transition in the mid-80s.5

This article aims to develop a better understanding of the manipulation of election law by carefully examining the beginning of this practice under the authoritarian regime origins. In the state elections of 1965, election rules were once again tailored to meet the government’s needs. These were sophisticated manipulations that defined the dictatorship’s strategy towards elections for the next twenty years. There are controversial opinions in the Brazilian historiography about the connections between the 1965 elections, the broader agenda of the federal government (involving the plans for Castello Branco’s succession, for instance), and the constitutional policy that would follow at least until 1969 (with numerous Institutional Acts, based on the doctrine of the “permanent revolution,” created in October 1965).6 The arguments advanced in this paper are meant to shed light on some of these connections.

Castello Branco considered himself a person deeply committed to democracy (or to what he believed to be democracy). In his annual message to Congress in March 1965, he urged for “democratic reform, realistically conceived.” He believed that the conditions for an authentic manifestation of popular would depended on “an electoral process freed from the vices that have been impairing it,”7 and that those against the proposed reforms were blocking democracy itself.

The future of the regime was unclear during those first years, but the reason the military took control of government was not. The military made a straightforward claim about its two-fold mission. The Institutional Act of April 9, 1964 spoke of “draining the communist abscess, which had a purulence that had already reached the government and the administration,” and of “economically, financially, politically and morally reconstructing the country.” Edward C. Burks, writing for the New York Times just three days after the enactment of the Institutional Act, described it as an effort to “clean up, shake up and ‘decommunize’ the Government,” “all with the idea of restoring democracy.” Burks asserted, “Brazil’s democratic Constitution, the Congress, and public opinion had all failed to halt the lunge toward leftist extremism of the Castro sort” (BURKS, 1964). The responsibility to stop this movement to the left fell to the military government, and Castello Branco was aware of this.

The “clean up” aspect of the mission was crucial. The military and their civil allies removed from office those politicians they considered unfit for the job, on the grounds either of corruption or of what they took to be “communist” or “extremist” tendencies. These purges took place in Congress (BRAGA, 2014, p. 92),8 in public administration, and even within the military forces, all based on the Institutional Act provisions that were in force from April 1964 to January 1966. Another element of the “clean up” strategy was to prevent the “wrong” candidates from ascending (or re-ascending) to office through elections, once more on the grounds of securing the “morality” in government. This task was accomplished in two ways. First, through the disfranchisement of the most prominent competing political leaders. The Institutional Act authorized such suspension of political rights, and also made such measures immune to any form of judicial review. Second, and more subtle, was the passing of customized electoral legislation. The new rules put the responsibility of enforcing biased restrictions against potential undesirable candidates into the hands of the judicial branch. Removing political rights from all relevant political opposition – instead of only its high ranked and more menacing cadres – would be inconsistent with almost any view of democracy. Castello Branco needed to present those decisions as the natural enforcement of properly enacted legal rules aimed at protecting the morality of the government and the administration.

The first step of this tactic was the enactment of Constitutional Amendment No. 14 (BRASIL, 1965a).9 Limitations on the right to run for elected office are usually entrenched in the constitution, which render them less prone to casuistic manipulation. This amendment, however, not only added a new case of ineligibility to those already written in the Constitution, but also authorized Congress to enact legislation establishing additional cases of ineligibility. Allegedly drawing on the first Brazilian republican Constitution of 1891 – which allowed the legislature to establish cases of “electoral incompatibility” [incompatibilidade eleitoral] – Castello Branco initiated the amendment in April 1965. Although it barely passed Congress, the amendment was promulgated in early June, just in time for the upcoming October elections. The reflections of Castello Branco’s right-hand-man, Minister Viana Filho, help to shed light on the president’s stance toward this issue. Viana Filho stated that with the first prospect of complexities in the 1965 elections, he felt that “the remedy would be in the ineligibilities legislation,” that is, maneuvering the rules regarding the right to run for elected office would allow Castello Branco to reconcile “democratic intent” with “revolutionary needs” (VIANA FILHO, 1975, p. 305, 313).

2 The looming State elections

Castello Branco soon understood that he would have to put aside his democratic beliefs for a while, because no official candidate would be able to win a direct presidential election as early as 1965. For that reason, in July 1964, a constitutional amendment extended his mandate to March 1967 (BRASIL, 1964b) and, as a consequence, postponed the presidential elections for one year, to 1966. That measure solved one problem, but created another.

State governors Carlos Lacerda and José de Magalhães Pinto (from Guanabara and Minas Gerais respectively), both potential candidates for the presidential run, were not happy. Neither of them was sure about the real intentions of Castello Branco, and they were then bound to eight months out of office right before the elections. At that time, according to Article 139 of the 1946 Constitution, state governors were required to resign six months before the elections. In practice, the postponement would keep them out of office for two additional months since their tenure would already have come to an end in late January 1966, eight months before the election. This sort of “political ostracism” could only harm a presidential bid. And, more importantly, if the 1965 state elections were maintained, they would face an additional electoral test before the presidential run: would they manage to get a supportive successor elected in their home states?

However, nothing about the looming state elections, including the timing, was certain. On the one hand, Castello Branco faced adverse political conditions in some key states, especially Minas and Guanabara. On the other, he was committed to the recommendation of the Superior Electoral Court (delivered at the same time as the Election Code draft) to standardize the term-lengths and the election calendar so that all state elections would coincide. Three alternative solutions were considered: First, to hold direct elections in 1965, for a one-time five-year-term. After that, the mandates would be for four years and elections would coincide. Second, to hold indirect elections for a short one-year-supplementary-mandate, using the Senate as Electoral College. Third, to declare a one-year prorogation of the current mandates, postponing elections.

The possibility of postponing elections triggered different responses from Governors Lacerda and Magalhães Pinto. Lacerda wrote to the president, making clear his plans to seek the “honor of becoming his successor in office” (VIANA FILHO, 1975, p. 294). He was worried about the postponement of elections becoming a pattern, and willing to agree with any solution, provided that it would not interfere with the 1966 presidential election schedule. Lacerda did not want to lose the chance of running without real competition, since the major opposition leaders had been disfranchised. Magalhães Pinto, on the other hand, took a more pragmatic approach to the situation. He stood for the postponement of state elections and sponsored a legislative solution to prorogate his own mandate, in the same fashion as Castello Branco. During the discussion of the proposal, the state legislature claimed that they were doing “nothing more than reproducing the recent constitutional amendment enacted by Congress, prorogating the current presidential mandate for one year” (ASSEMBLEIA LEGISLATIVA DO ESTADO DE MINAS GERAIS, 1964).

In early February 1965, Castello Branco was leaning toward the second alternative, endorsing indirect elections for the office of state governor for a one-year term. According to Viana Filho, another letter from Lacerda shook the president’s resolution. Lacerda overtly criticized indirect elections, arguing that “the better, the right, the bravest and the more democratic thing to do is to stand for direct elections.” He claimed that if the government “officially recognizes the inconvenience of direct elections,” then the most reasonable alternative would be to adopt the same solution the president embraced for himself, that is prorogation (VIANA FILHO, 1975, p. 293-294).10

Lacerda had thus put the president’s commitment to democracy into question and attempted to make him look “undemocratic” before the public. Castello Branco was incensed by this attack. He considered Lacerda’s declarations a bluff, given the unpromising prospects for the candidacy he supported in his own state of Guanabara. He decided to call the bluff and give Lacerda what he purportedly wanted: direct elections (VIANA FILHO, 1975, p. 294).

On March 23, 1965, Minister of Justice Milton Campos appeared at a congressional hearing to inform the legislative branch of the government’s official position toward state elections. He announced that the president would soon initiate a constitutional amendment granting state governors elected in 1965 a five-year term after which state elections would be synchronized. Therefore, he declared, the 1965 elections were being maintained. Congressmen emphatically applauded Campos when he got to this point, and Campos made sure to mention the “two ambitions” of the Revolution of March 31: “First, to be a revolution. Second, to be a democratic revolution.”11

In his statement, Campos emphasized Castello Branco’s Presidential Message that had been sent to Congress at the beginning of the legislative year, in which the president asserted that genuine democracy would require yet further adjustments to the electoral system. The government had been working on these details. By the end of April, it submitted to Congress the draft of the Election Code, and two months later, it initiated the bill regulating ineligibilities, based on the recently enacted constitutional amendment. It took only three weeks for Congress to deliberate on the bill. Both statutes were promulgated on the same date, July 15, 1965,12 less than three months before the elections.

Campos gave an extensive address on the motives of the ineligibilities bill, presenting the bill as a tool to fight the abuse of political and economic power in elections and as a weapon against those seeking to destroy the regime. Every civilized country in the world, Campos remarked, was taking precautions to fight unacceptable political ideas and organizations. The ineligibilities bill, Campos argued, stemmed from the same reasoning. It “perfectly distinguishes opposition to government from opposition to the regime. This latter is what the bill seeks to avoid, so we do not mistake insurrection for opposition, seditious manipulations for political freedom.”13

The two dissidents directly targeted by the ineligibilities legislation were politicians connected to Juscelino Kubitschek, the former President: Sebastião Paes de Almeida, a wealthy businessman and representative of the state of Minas Gerais, and Henrique Teixeira Lott, a retired general and former presidential candidate, defeated by Jânio Quadros in 1960. Both served in Kubitschek’s administration, the former as Minister of the Treasury and the latter as Minister of the Armed Forces. Lott was a renowned figure in mid-20thcentury Brazilian politics (CARLONI, 2010) while Paes de Almeida, on the other hand, was famous for having one of the largest fortunes of the country.

Castello Branco’s concerns about these two particular candidacies, however, had to do with more than their political affiliations: Lott and Paes de Almeida were running against the candidates supported by Governors Lacerda and Magalhães Pinto. Both Lacerda and Magalhães Pinto were directly involved in the coup (probably envisioning themselves as soon-to-be civil successors of Castello Branco), but both had been growing progressively suspicious about the real intentions of the military (FICO, 2014, p. 120).14 It is likely that, different from what is generally accepted, Castello Branco secretly thought of himself as a candidate for reelection (VIANA FILHO, 1975, p. 377).15 In this case, he could profit from the weakening of his internal competition. Who could support presidential candidates who did not manage to win elections in their own home states? However, Castello Branco would have to “set the stage” for a “tolerable defeat” in these crucial states. If they fell into the hands of politicians obviously associated with the “enemies of the Revolution,” the radical sectors within the armed forces [linha dura] would question Castello Branco’s commitment to their goals, and he would lose important civilian and military support. Thus, he had to ensure less incendiary candidates for the opposition. On the other hand, if Lacerda and Magalhães Pinto, the candidates aligned with the government, emerged victorious, Castello Branco could claim that his legislative policy promoted that outcome. And if they lost, some serious suspicion would have been cast over the main non-military names for the presidential succession, which could be beneficial for a potential Castello Branco candidacy.

“Setting the stage,” however, proved a complex task. Viana Filho’s candid description of the government’s effort to tailor the ineligibilities legislation through Congress reveals their thinking about the process:

It was easy to remove Hélio de Almeida [a possible PTB16 candidate in Guanabara]: a provision making ineligible those who served as ministers between January 23, 1963 and March 31, 1964 [the period during which João Goulart governed under the presidential system] would do. Afterwards, we realized that the criterion was excessively strict, and we made it looser by excluding those who had held legislative mandates or had served as a minister for any of the armed forces. It was the way to safeguard former minister Carvalho Pinto, whose dignity everyone respected, and General Kruel, who helped the revolution. […] In Congress, each provision represented a battle, because approval by the majority of the members of each House was required, and as the politicians targeted by the rules were somewhat recognizable, resistances were fierce. […] [Pedro] Aleixo [the leader of the majority], having reserved face-to-face persuasion for Rondon [Pacheco, a UDN17 representative from Minas Gerais], polished the language with patience and malice, to attract supporters and to dissipate suspicion, without compromising the objectives. In fact, Aleixo was not drafting the statute: he was preparing the presidential vetoes (VIANA FILHO, 1975, p. 314).

His description aligns with the original record containing the legislative debates, the committee reports, and votes in the Chamber of Deputies, where all the modifications to the original bill were made. One particular provision exemplifies Viana Filho’s point. The original bill made ineligible for the position of state governor those who, “(…) through the abuse of economic power, corruption of unduly influence while in office or mandate had or have been compromising, personally or through an intermediary, the fairness and regularity of the elections.”

This wording raised doubts about the constitutionality of the rule, and eight representatives voted against it on the Committee on Constitution and Justice, arguing that it would infringe on the constitutional prohibition of retroactive laws. At least three amendments were proposed to this provision. Martins Rodrigues, the Leader of the PSD,18 presented one of them, and Doutel de Andrade and Chagas Rodrigues, of the PTB, sponsored the other two. The aims of the amendments were the same: first, to clarify that the provision would be applied only prospectively, and, second, to establish that the abuse of economic or political power had to be declared by a final judicial decision before it could be validly claimed as grounds for ineligibility.

If the amendments succeeded, the provision would become irrelevant for the upcoming elections, frustrating the government’s plan. However, defeating the amendments was beyond the available political support for the government, since (as soon will become clear) one of the targets of the rule was a PSD representative. A different strategy was in order. The rapporteur of the bill, representative Oliveira Britto, working closely with Pedro Aleixo, presented a sub-amendment merging the three amendments and acknowledging their suggestion. But he did so (in the carefully “polished” language to which Viana Filho referred19) to make ineligible for office “those who had been convicted for compromising, personally or through an intermediary, the fairness and regularity of the elections by abuse of economic power, by acts of corruption or by unduly political influence while in office or mandate, or those who come to compromise it by the same abuses, acts or influence.”

When the enrolled bill was presented to Castello Branco for approval, he vetoed the words “convicted for,” subverting the rule actually passed by Congress, and allowing the electoral courts to decide, during the registration of the candidate, if there had been any relevant past abuses of economic or political power. Overriding a presidential veto, according to the 1946 Constitution, was difficult. A two-thirds majority on a secret ballot was required. By omitting those words, Castello Branco had substantially enlarged the scope of the ineligibility rule. And, although Constitutional Amendment No. 14 required the majority of the members of each house to approve new cases of ineligibility, with this maneuver, unless 317 congressmen (out of the 409 representatives and 66 senators) were against the president, he would prevail, as he actually did.20

It was clear that many provisions of the new legislation had been crafted to provide legal grounds for immediate elimination of high-ranked anti-revolutionary political competition. The barred candidate in Minas was Sebastião Paes de Almeida, who went by the peculiar nickname, “Tião Medonho.

3 A “dreadful candidate”

In 1962, Brazilian film director Roberto Farias released a movie entitled Assalto ao Trem Pagador [Assault on the Pay Train]. The plot, based on a real crime that drew the attention of the media in the early 1960’s, develops around “Tião Medonho” (something like “Dreadful Sebastian”), a small time criminal involved in a successful robbery. Tião, a black man living in one of Rio de Janeiro’s several favelas, now had millions, but spending it immoderately would reveal him. He thus hid part of the money and used the rest of it discretely, making sure that his partners in crime did the same. Tião was an ambiguous character, capable of acting with cruelty and violence toward his adversaries, but with tenderness and compassion toward his family and neighbors. He was pursued relentlessly by the police and ended up being killed as the result of injuries sustained during a gunfight with the cops.

Nothing could be further from the story of Sebastião Paes de Almeida. Born in the tiny city of Estrela do Sul, west of Minas Gerais, in 1912, Paes de Almeida moved to São Paulo with his family two years later, where his father built a successful career as a banker. He earned a law degree in 1937, when he was already working in the financial market and running several different enterprises, including glass and agribusiness companies. During the 1950s, Paes de Almeida served as the president of the Banco do Estado de São Paulo [Bank of the State of São Paulo] and as the secretary of the treasury for São Paulo’s governor, Lucas Garcez. After that, President Juscelino Kubitschek invited him to be the president of the Banco do Brasil [Bank of Brazil]. He later became Minister of the Treasury during the last years of Kubitschek’s government (CPDOC, 2001).

The white, rich capitalist based in São Paulo also had more than ten times the property than the black and unfortunate Tião Medonho was able to raise in his successful crime: almost 400 million cruzeiros.21 In 1962, Paes de Almeida was elected representative of the State of Minas Gerais, despite the fact that he had long been based in São Paulo. And not only was he elected without consistent political activity within his native state, but also with the largest number of votes ever cast thus far for a Minas Gerais representative in the Chamber of Deputies, more than 80,000 votes.22 That was an impressive number by any standard. Tancredo Neves, who had just served as prime minister under the short parliamentarian government of João Goulart, had received 58,000 votes, and he was a longstanding traditional politician within the state, who had already run (and lost) for governor against Magalhães Pinto in the last elections. Paes de Almeida’s electoral success had to be associated with economic power, at least according to his horrified UDN adversaries. As reported by the press:

Mr. Sebastião Paes de Almeida appeared in Minas Gerais precisely at the time a bandit named Tião Medonho, responsible for the robbery of the pay train at Central do Brasil, was spreading terror in the favelas of Rio de Janeiro. Mr. Paes de Almeida’s adversaries nicknamed the then PSD candidate for a seat as representative in Congress after the robber: “the man who is spreading fear, because he is buying everything and everyone with his money” (BARROSO, 1965a).

Contrary to what his opponents expected, the parallel drawn between Paes de Almeida and Tião Medonho served as excellent propaganda for the politician, lending a “popular flavor” (GOMES, 1965) to his tentative candidacy for the office of state governor in Minas Gerais. The press reported: “The rich will vote for Sebastião Paes de Almeida and the poor for Tião Medonho” (BARROSO, 1965b). However, different from what Viana Filho (1975, p. 317) suggested in his memoirs, Sebastião Paes de Almeida never allowed himself to be registered for the elections under that alias: in 1965, the PSD official documents required the Regional Electoral Court of Minas Gerais to present the candidate’s name as “Sebastião Paes de Almeida (TIÃO),” and not “Tião Medonho.”23

The name – and the nickname – of Paes de Almeida had been circulating since the beginning of the year, but it was only on July 23, 1965 that the PSD officially confirmed his candidacy to state governor by a vast majority, in a primary vote. The registration before the Regional Electoral Court occurred a few days later, on August 4 (MINAS GERAIS, TRE-MG, 1965a).

On August 9, the UDN contested Paes de Almeida’s registration, arguing that he was ineligible for two reasons. First, because of the four-year residency requirement (added to the Constitution by the same amendment that authorized statute-enacted ineligibilities). Paes de Almeida had transferred his electoral papers to Estrela do Sul, his native town in Minas Gerais, only two months before the registration. The constitutional provision exempted those “who had served a state elected mandate” from the four-year requirement. This might have applied to Paes de Almeida because he was an elected representative of Minas Gerais in the Chamber of Deputies, however the UDN insisted that the exception was only for those who “had served” the full term, and not for those who “have been serving” a term. Moreover, they argued, a representative at the federal legislature did not hold a “state elected mandate,” but a federal mandate. Their second argument was that Paes de Almeida’s conduct during the 1962 elections for the Chamber of Deputies made him ineligible, because of his “public and notorious” abuse of economic power to obtain votes. This argument depended entirely on the newly enacted ineligibilities legislation that declared ineligible for office “those who had compromised, personally or through an intermediary, the fairness and regularity of the elections by abuse of economic power, by acts of corruption or by unduly political influence while in office or mandate (…).” As shown above, this particular rule was modified by the president’s veto, which could still be overridden by Congress, although it was not likely. The UDN claimed that the donations made by Paes de Almeida to several municipalities and charitable organizations in 1962 interfered with the regular course of legislative elections.

The Electoral Prosecution Office admitted that there may have been an infringement of the residency requirement, but urged the Court to dismiss the allegations concerning abuse of economic power or corruption. The prosecutor was skeptical about the documents the UDN was using to base its claims, calling them “useless.” In addition, he argued that the ineligibilities legislation violated the prohibition of retroactivity, and, therefore, was unconstitutional.

Nelson Hungria, a former justice of the Brazilian Supreme Court and one of the most respected jurists of the country, presented Paes de Almeida’s defense. Hungria refuted the arguments offered by the UDN one by one, pleading the unconstitutionality of the provision that grounded the charge of abuse of economic power and even mentioning the pending congressional deliberation on the crucial word-veto. He also argued that the grammatical interpretation of the residency requirement advanced by the UDN was absurd, since the rationale of the rule was to exempt from the four-year prerequisite those who were necessarily familiar with the political circumstances of the state. And the current representatives, including Paes de Almeida, were supposed to have ultimate up-to-date knowledge about that. Moreover, there was never doubt that a representative in the Chamber of Deputies, although serving the federal legislature, received his mandate from his home state.

Hungria’s defense, however, was primarily concerned with the allegations of abuse of economic power. He claimed that those opposing Paes de Almeida’s candidacy had mistaken philanthropy for corruption. As part of his argument, he appended 78 documents (most of them notarized declarations) showing that, long before his 1962 candidacy, Paes de Almeida had been making considerable donations to a variety of institutions based outside Minas Gerais. The documents reveal that his philanthropic activities extended through various states, such as São Paulo, Rio de Janeiro, Goiás, Mato Grosso, Paraná, Brasília (the capitol), and others. The majority of the beneficiaries were catholic-oriented religious, educational or charitable associations. Sometimes he would donate money directly to churches or to the Red Cross. Father Laércio Dias de Moura, the President of the Catholic University of Rio de Janeiro – one of the most important educational institutions in the country – issued a declaration certifying that Paes de Almeida had donated one million cruzeiros in 1961 and 1962 to fund underprivileged students. The donations began at the end of the 1950s and continued until 1965, that is, even after the 1962 election. There is no evidence in the documents that these donations were motivated by anything other than the religious inclination of Paes de Almeida (MINAS GERAIS, TRE-MG, 1965a).24

The trial advanced quickly. The witnesses called to testify did not help the UDN’s case, since none were able to identify a specific situation in which Paes de Almeida actually abused economic power. One even admitted that he “had never heard about Paes de Almeida negotiating votes directly or through a middleman.” In an attempt to strengthen its claims, the UDN brought documents showing that Paes de Almeida was under investigation by federal agencies such as the Administrative Council for Economic Defense for violation of anti-trust legislation (MINAS GERAIS, TRE-MG, 1965a)25 and the Secretariat of Federal Revenue, the Brazilian tax authority (for tax irregularities allegedly commissioned by a company owned by Paes de Almeida26). The first of these investigations was intended to revive a 1955 parliamentary inquiry about violations of economic competition in the glass industry, controlled by Paes de Almeida companies. The tax inquiry, on the other hand, was aimed primarily at former President Juscelino Kubitscheck. All investigations emerged or reemerged, coincidentally, at the same time as the Paes de Almeida candidacy.

The Regional Electoral Court of Minas Gerais, however, was not impressed with any of the arguments. It ruled 4 to 2 that Paes de Almeida was exempt of the residency requirement on the grounds of being a representative of Minas Gerais in Congress, and that his philanthropy neither amounted to abuse of economic power, nor compromised the normality of the elections. However, the Court found – against Paes de Almeida’s claims – that the provision of the ineligibilities legislation under which the candidacy had been questioned was constitutional. According to the Regional Court, the rule did not violate the prohibition of retroactive laws (applicable only to criminal statutes, according to the letter of the 1946 Constitution), and the word-veto (that actually subverted the meaning of the provision enacted by Congress) was just the regular exercise of a presidential prerogative.

4 The appeal to the Superior Electoral Court and the aftermath of Paes de Almeida’s interrupted candidacy

The UDN, as expected, appealed the decision to the Superior Electoral Court. The trial took place on September 7, 1965, less than one month before the elections (TSE, 1965).27

The rapporteur for the case, Justice Oscar Saraiva, agreed with the Electoral Court on the constitutionality of the provision of the ineligibilities legislation, and also with the inapplicability of the residency requirement to Paes de Almeida. However, he disagreed with the section of the opinion regarding the charges of abuse of economic power. Saraiva asserted, “The purpose of the law was to interdict any interference of economic power within the choice of the political representation.” In his opinion, the only reasonable explanation for the electoral success of Paes de Almeida was his several contributions to municipalities and charitable institutions in the state of Minas Gerais. In addition, the justice pointed out that, regarding the state of Minas Gerais, with only one exception, the beneficences were all concentrated during the electoral period. Saraiva claimed that a candidate’s motives are determinant to assess compliance of conduct with the electoral legislation. A donation is legally permissible, but if it can be proved that the intention behind this seemingly legal act is to influence the disposition of the electorate, then it shall be regarded as inconsistent with the law. Finally, he stated that the “Electoral Justice should deem as legally prohibited abuses those acts of patronage or philanthropy performed by a candidate during the electoral period, within the district of the elections.” Since the constitutional guarantee against the retroactivity of laws was circumscribed to criminal statutes, the justice found no difficulty in considering the conduct of Paes de Almeida during the 1962 elections as sufficient grounds for declaring him ineligible for the office of state governor in the 1965 elections.

Justice Gonçalves Oliveira dissented, arguing, first, that the Superior Electoral Court could not hear the appeal, since it would require reviewing the account of the facts, and not the interpretation of the law. The Regional Court, after assessing the circumstances of the case, had decided there was not enough evidence to prove that the contributions made by Paes de Almeida were the main reason of his substantive number of votes. Oliveira claimed that reviewing facts was not the job of the Superior Court on that sort of lawsuit. He then pointed to an even stronger reason for dismissing the appeal, citing Article 121 of the 1946 Constitution, which establishes all the situations in which the Superior Electoral Court could hear an appeal against a decision issued by a regional court. There was no mention of decisions that ruled a candidate eligible or ineligible to office, and therefore, the plea of the appellants was out of the constitutional scope of the appeal. Oliveira’s dissent was based on an accurate interpretation of the Constitution. Subsequently, during the 1965 Judicial Reform (enacted less than two months after the elections), the government hurried to alter Article 121, adding decisions of the Regional Electoral Courts regarding ineligibilities to the list of decisions that could be challenged in an appeal to the Superior Electoral Court.28

The majority of justices disagreed however. On the first point, four out of the seven justices believed that reviewing the facts was not at stake. Rather, they argued, the problem was the legal characterization: were the contributions altruistic philanthropy or a manifestation of abuse of economic power? The Superior Court did have the authority to answer that question. The second part of the dissent, however, was harder to rebut, and it provoked a very technical argument in the Court. Although the Constitution did not authorize an appeal based on the recognition of a candidate’s ineligibility (or eligibility), it did grant the right to appeal a decision of the Regional Court certifying an ineligible candidate as the winner of elections. In other words, according to the law, in the case of a victory for Paes de Almeida, duly recognized by the Regional Electoral Court of Minas Gerais, the Superior Electoral Court would then be allowed to review the ineligibility claim. The drafters of Constitutional Amendment No. 14 did not anticipate this situation, and simply failed to provide for an appeal against an inconvenient interpretation of the ineligibilities legislation by regional courts. The majority of the Superior Electoral Court, however, refused to abide by this reading of the Constitution, which made no sense to them. Consenting to an election over which such a critical doubt had been cast would be deleterious to democracy. If the Court was endowed with the authority to nullify the election after it was over, it had to be endowed with the lesser power of preventing an ineligible candidate from running and causing such an embarrassment.

On the merits, three justices concurred with the rapporteur. That was enough to put an end to the candidacy of Paes de Almeida. The decision was based exclusively on the article of the ineligibilities statute manipulated by Castello Branco’s veto. There were only two dissenting opinions, and they were grounded in the procedural quarrel about the authority of the Superior Electoral Court to hear the appeal. Following a long tradition of disregard toward legislative process, the opinion of the Court did not take the discussion about the constitutionality of the veto seriously. However, even scholars with obvious ties to the military regime criticized the abuse of the partial veto, and its conversion from a power to interdict into a power to amend (FERREIRA FILHO, 1971, p. 33-35).29

Anticipating an unfavorable outcome, Paes de Almeida attempted a last desperate move. With the consent of their party, the PSD, he and his running mate authorized another party (PSP) to register them again as candidates for state governor before the Regional Electoral Court of Minas Gerais (MINAS GERAIS, TRE-MG, 1965b). The Regional Electoral Court received the new registration on September 6, the eve of the appeal’s trial before the Superior Electoral Court. Perhaps Paes de Almeida expected the Regional Court to cancel the registration made under the PSD, and reckoned that this would force the Superior Court to drop the appeal. That could buy them some time. But maybe the idea was even simpler: if the UDN’s appeal was granted, Paes de Almeida would still keep the possibility of running for governor open (albeit under another party); if it was not, they could merely take back the new registration request. However, the strategy failed, as the Regional Court was immediately confronted with the Superior Electoral Court’s ruling on the appeal and decided to simply abide by it, acknowledging Paes de Almeida’s status as ineligible and dismissing the request for a new registration.

Hypothetically, Paes de Almeida could appeal the Superior Electoral Court’s decision to the Supreme Court. However, the election was less than one month away, and that course of action would risk leaving the PSD without any candidate for the office of state governor in Minas Gerais. To avoid that, Paes de Almeida gave up the right to appeal and was called upon by his fellow party members to appoint his substitute. Israel Pinheiro, the vice president of the party in Minas Gerais, was chosen, despite some resistance from Benedito Valadares, the president of the party, who had been on a leave of absence due to illness (SEBASTIÃO, 1965). Paes de Almeida apparently was very involved in the campaign, which in the end turned out victorious.

The defeat of the UDN in Minas Gerais and Guanabara is usually described as the trigger event for the radicalization of the “revolution,” manifested in the enactment of the second Institutional Act of October 27, 1965, which terminated all existing parties. Those willing to join forces with the government (including a majority of the PSD) were allowed to migrate to the newly created ARENA [Aliança Renovadora Nacional], the official “revolutionary” party, under the artificial bipartisan system, in which the opposition was reserved to the MDB [Movimento Democrático Brasileiro]. Probably to the disappointment of Juscelino Kubitschek and Paes de Almeida, Israel Pinheiro was among those politicians who aligned with the government.

This confusing aftermath shows that Castello Branco’s plan did not fully work. He was able to dismiss Paes de Almeida’s candidacy by manipulating the electoral rules and this led to an oppositionist candidate who, once elected, quickly collaborated with the government. However, the ability to control election law did not secure victory, as the voters in Minas Gerais and Guanabara refused to endorse the candidates officially aligned with the government. Moreover, the defeat in those crucial states indicated that a victory of Lacerda (or any other UDN candidate) in the upcoming presidential run was unlikely. This allowed the linha dura to press for deeper changes in the rules of the game, as Elio Gaspari points out (GASPARI, 2002, p. 240); changes that Congress would certainly refuse to deliver. The result was the enactment of the second Institutional Act, which relied on the idea of a “permanent revolution” and represented not only the end of direct presidential elections, but also a major blow to Castello Branco’s effort toward “democratic legality.”30

5 Conclusion: “Fatta La legge, trovato l’inganno,” or democracy, according to the authoritarian regime

As Carlos Fico points out, the sheer enumeration of the articles of the second Institutional Act make it clear that this was not just an overreaction to the election results in Minas Gerais and Guanabara. “It is a set of measures with broader scope which, as a matter of fact, provided the government with an opportunity to resume the ‘clean up’ operation, as the ‘hardliners’ desired” (FICO, 2004, p. 74). The elucidation of the events surrounding October 1965 can help to shed light on a critical turning point of the Brazilian dictatorship. A little after the first anniversary of the military coup, it became clear that, if Castello Branco ever had plans for a smooth transition of power back to the civil elites, they were indefinitely postponed.

Paes de Almeida’s story is not necessarily representative of a particular practice or socio-legal pattern. However, it can help us to understand how the political discourse and the electoral politics of the civil-military dictatorship were built. Many of its features can be found in other episodes of the paradoxical relationship between the authoritarian government and election law. Some examples include: “customized legislation” enacted by a purged Congress, through a manipulated legislative process, in a scenario of prevalence of presidential (sometimes “revolutionary”) powers; a mix of pressure on and partnership with courts to deliver an interpretation of laws consistent with the expectations of the government (later secured through purges and “court packing” politics); a quest for purifying politics through law, supported by moralizing discourses which called for detaching elections from economic and political power, but – in a concealed interplay between economic interests and governmental endeavors – only prevented influence from those powers not aligned with the “revolution.”

Castello Branco, particularly, attempted a risk-free, “managed democracy.” He was willing to rely on elections, insofar as they worked as a process of “ratification,” in which the “sovereignty of the people” could only reveal itself under the strong tutelage of institutionalized politics. The idea of neutralizing economic and political power in elections was justified as protecting the voter’s liberty. However, the 1965 Election Code, the new legislation for the political parties, and the ineligibilities statute were all part of an effort to push elections and voters into the “right” – or expected – direction. The outcome of democracy could only be the confirmation of the Revolution, of the ideals that inspired the civic and military movement, as contradictory as it may seem.

In a speech delivered by Castello Branco before the Superior Electoral Court a couple of months before the 1965 elections, he criticized those who imagined “bad elections” as “a means of making Brazil move back to a past that will never be resurrected.” He threatened, “There is no alternative to democratic improvement than voting. However, we must consider that there is no alternative for the country other than the existence of a legal government of the Revolution. It is definitive and irreversible” (VIANA FILHO, 1975, p. 317). The hierarchy was clearly established. “Democracy” was only useful to the regime as far as it could provide it with a political scenario consistent with the status quo. That is why Castello Branco envisioned “political debate as a hurdle rather than an intrinsic part of a democratic system.” He assumed that the “problem” with elections could be fixed, or at least managed, with the right set of rules.

As Milton Campos, the first Minister of Justice of Castello Branco, helped to prepare the drafts of the new electoral legislation in 1965, he recommended caution, and quoted an Italian proverb, “Fatta la legge, trovato l’inganno” [Enacted the law, found the trick]. Castello Branco actually hoped to avoid trickery by strengthening his control over the law and over the legal system as a whole. As he and his successors struggled to conform popular sovereignty to the “revolutionary” needs, they eventually had their way, but were ultimately exposed enough times to compromise the credibility and legitimacy they sought for the political system their many reforms aspired to shape.


This paper was presented during the 2014 Annual Meeting of the American Society for Legal History, as part of the panel “Paradoxes of Authoritarian Legal Orders and the Ambiguity of Transitions to Democratic Rule in South America (1960s-1980s).” I am indebted to Prof. Mark J. Osiel for his comments during the event, as well as to the other participants of the panel, Prof. Cristiano Paixão and Marcelo Torelly. I have benefited greatly from discussions with Prof. Rebecca J. Scott, who chaired the panel and provided important suggestions on the first draft of the paper. Claudia Carvalho read earlier versions of the text and provided valuable comments. Finally, Victoria Langland, Sueann Caulfield, and Emad Ansari organized a fruitful roundtable on this paper with scholars and students connected to the Center for Latin American and Caribbean Studies, the History Department, and the Law School at the University of Michigan. The research for this paper was sponsored by a Michigan Grotius Fellowship granted by the University of Michigan Law School and by the Brazilian Chamber of Deputies, through a sabbatical leave. I am particularly grateful to Roopal Shah and to the team of the Center for International and Comparative Law of the University of Michigan Law School, for their support during the time I spent as a Michigan Grotius Research Scholar. I also thank the reviewers for Direito GV Law Review for valuable recommendations and insights. Kristin McGuire copyedited the paper.


ALVES, Maria Helena Moreira. Estado e oposição no Brasil: 1964-1985. 2ª ed. Bauru: Edusc, 2005. [ Links ]

AARÃO REIS, Daniel. Ditadura e sociedade: as reconstruções da memória. In: AARÃO REIS, D.; RIDENTI, M.; SÁ MOTTA, R. P. O Golpe e a Ditadura Militar: 40 anos depois (1964-2004). Bauru: EDUSC, 2004. p. 29-52. [ Links ]

ASSEMBLEIA LEGISLATIVA DO ESTADO DE MINAS GERAIS. Projeto de Emenda Constitucional n. 1.083, de 1964. Parecer da Comissão de Constituição e Justiça. Arquivos da Assembleia Legislativa do Estado de Minas Gerais. [ Links ]

AZEVEDO, Débora Bithiah; RABAT, Márcio Nuno. Parlamento mutilado: deputados federais cassados pela Ditadura de 1964. Brasília: Câmara dos Deputados, Edições Câmara, 2012. [ Links ]

BARBOSA, Leonardo Augusto de Andrade. História constitucional brasileira: mudança constitucional, autoritarismo e democracia no Brasil Pós-1964. Brasília: Câmara dos Deputados, Edições Câmara, 2012. [ Links ]

BARROSO, Jadir. Apelido pode eleger o novo governador de Minas Gerais. Jornal do Brasil, 4 abr. 1965a, p. 33, 1º caderno. [ Links ]

BARROSO, Jadir. Minas calcula sua abstenção em 600 Mil. Jornal do Brasil, 5 set. 1965b, p. 5, 1º Caderno. [ Links ]

BRAGA, Ricardo de João. A agenda legislativa do Governo Castelo Branco: um regime político em transição, E-Legis, Brasília, n. 14, p. 87-105, maio-ago. 2014. [ Links ]

BRASIL. Poder Executivo. Ato Institucional de 9 de abril de 1964. Diário Oficial da União. Brasília, DF, 9 abr. 1964a, Seção 1, p. 3.193. [ Links ]

BRASIL. Constituição (1946). Emenda Constitucional n. 9, de 22 de julho de 1964. Diário Oficial da União. Brasília, DF, 24 jul. 1964b, Seção 1, p. 6.593. [ Links ]

BRASIL. Emenda Constitucional n. 14, de 8 de abril de 1965, de 8 de abril de 1965. Diário Oficial da União. Brasília, DF, 9 abr. 1965a, p. 3.609. [ Links ]

BRASIL. Emenda Constitucional n. 15, de 5 de julho de 1965. Diário Oficial da União. Brasília, DF, 6 jul. 1965b, Seção 1, p. 6.291. [ Links ]

BRASIL. Emenda Constitucional n. 16, de 26 de novembro de 1965. Diário Oficial da União. Brasília, DF, 6 dez. 1965c, Seção 1, p. 12.374. [ Links ]

BRASIL. Emenda Constitucional n. 17, de 26 de novembro de 1965. Diário Oficial da União. Brasília, DF, 6 dez. 1965d, Seção 1, p. 12.375. [ Links ]

BRASIL. Câmara dos Deputados. Projeto de Emenda à Constituição n. 7, de 4 de novembro de 1965. Diário do Congresso Nacional, Seção II. Brasília, DF, 5 nov. 1965e, p. 3.860. [ Links ]

BRASIL. Lei n. 4.740, de 15 de julho de 1965. Lei Orgânica dos Partidos Políticos. Diário Oficial da União. Brasília, DF, 19 jul. 1965f, Seção 1, p. 6.764. [ Links ]

BRASIL. Lei n. 5.682, de 21 de julho de 1971. Lei Orgânica dos Partidos Políticos. Diário Oficial da União. Brasília, DF, 21 jul. 1971, Seção I, p. 5.673. [ Links ]

BRASIL. Constituição (1967). Emenda Constitucional n. 8, de 14 de abril de 1977. Altera artigos da Constituição Federal, incluindo-se em seu Título V os artigos 208, 209 e 210. Diário Oficial da União. 14 abr. 1977, Seção 1, p. 4.195. [ Links ]

BRASIL. Tribunal Superior Eleitoral. Recurso Ordinário n. 2.869, Relator: Min. Oscar Saraiva. Arquivo do Tribunal Superior Eleitoral, 1965. [ Links ]

BRASIL. Tribunal Superior Eleitoral. Resolução n. 12.017, de 27 de novembro de 1984. Diário da Justiça. Brasília, DF, 10 dez. 1984, p. 21.160. [ Links ]

BURKS, Edward C. Brazil now embarks upon a new course. N.Y. Times , 12 abr. 1964, p. E4. [ Links ]

CAMPOS, Milton. Revolução e processo revolucionário. Digesto Econômico, v. 22, n. 192, p. 53-54, nov.-dez. 1966. [ Links ]

CARLONI, Karla Guilherme Carloni. Marechal Henrique Teixeira Lott: a opção das esquerdas. 2010. 250f. Tese (Doutorado em História). Universidade Federal Fluminense, Niterói, 2010. [ Links ]

CARNEIRO, André Corrêa de Sá. O veto parcial no sistema constitucional brasileiro. E-Legis, n. 2, p. 12-21, 1o semestre 2009. [ Links ]

CASTELLO BRANCO, Carlos. Um dado para decisão: definição de Lacerda. Jornal do Brasil, 13 fev. 1965. p. 4 (1º caderno). [ Links ]

CASTELLO BRANCO, Carlos. Movimento em favor da reeleição de Castelo. Jornal do Brasil, 24 dez. 1965. p. 4 (1º caderno). [ Links ]

CENTRO DE PESQUISA E DOCUMENTAÇÃO DE HISTÓRIA CONTEMPORÂNEA DO BRASIL (CPDOC). Dicionário Histórico-Biográfico Brasileiro pós 1930. 2ª ed. Rio de Janeiro: Editora FGV, 2001. [ Links ]

CONCORDATA: trem mau pagador. Revista Veja, 17 set. 1975, p. 24. [ Links ]

CONSELHO ADMINISTRATIVO DE DEFESA ECONÔMICA (CADE). Processo Administrativo n. 3, Relator: Conselheiro Geraldo de Rezende Martins, Diário Oficial da União, 19.9.1969, p. 7.925. [ Links ]

COSTA PORTO, Walter. O voto no Brasil: da Colônia à 6ª República. 2ª ed. Rio de Janeiro: Topbooks, 2002. [ Links ]

FERREIRA FILHO, Manoel Gonçalves. O veto parcial no direito brasileiro. Revista de Direito Público, v. 4, n. 17, p. 33-37, jul.-set. 1971. [ Links ]

FICO, Carlos. Além do Golpe: versões e controvérsias sobre 1964 e a Ditadura Militar. Rio de Janeiro: Record, 2004. [ Links ]

FICO, Carlos. O Golpe de 1964: momentos decisivos. Rio de Janeiro: FGV, 2014. [ Links ]

GASPARI, Elio. A ditadura envergonhada. São Paulo: Cia. das Letras, 2002. [ Links ]

GOMES, Pedro. Informe JB – Tião avança. Jornal do Brasil, 11 mai. 1965. p. 10, 1º Caderno. [ Links ]

KINZO, Maria D’Alva Gil. Oposição e autoritarismo: gênese e trajetória do MDB, 1966-1979. São Paulo: Vértice, 1988. [ Links ]

LEMOS, Renato. Poder Judiciário e poder militar: 1964-1969. In: CASTRO, Celso; IZECKSOHN, Vitor; KRAAY, Hendrik (Org.). Nova história militar brasileira. Rio de Janeiro: FGV, 2004, p. 409-438. [ Links ]

MINAS GERAIS. Tribunal Regional Eleitoral do Estado de Minas Gerais (TRE-MG). Registro de Candidatos n. 22, de 1962. Partido Social Democrático. Arquivo do TRE-MG. [ Links ]

MINAS GERAIS. Registro de Candidatos (para Governador e Vice-Governador) n. 1, de 1965a. Relator: Des. João Martins de Oliveira. Arquivo do TRE-MG. [ Links ]

MINAS GERAIS. Registro de Candidatos (para Governador e Vice-Governador) n. 7, de 1965b. Relator: Des. Valle Ferreira. Arquivo do TRE-MG. [ Links ]

NICOLAU, Jairo. Eleições no Brasil: do Império aos dias atuais. Rio de Janeiro: Zahar, 2012. [ Links ]

O’DONNELL, Guillermo; SCHMITTER, Philippe. Transitions from authoritarian rule: tentative conclusions about uncertain democracies. Baltimore: John Hopkins University Press, 1986. [ Links ]

PEREIRA, Anthony W. Political (in)justice: authoritarianism and the Rule of Law in Brazil, Chile, and Argentina. Pittsburgh: University of Pittsburgh Press, 2005. [ Links ]

SANTOS, Fabiano. Recruitment and retention of legislators in Brazil, Legislative Studies Quarterly, v. 24, n. 2, p. 209-237, maio 1999. [ Links ]

SEBASTIÃO vai indicar novo candidato. Correio da Manhã, 10 set. 1965. p. 3, 1º caderno. [ Links ]

VIANA FILHO, Luís. O Governo Castelo Branco. Rio de Janeiro: J. Olympio, 1975. [ Links ]

1 Cf. Diário do Congresso Nacional, Seção I, 23.3.1965, p. 1.128.

2 The Institutional Act also bypassed a statute enacted two days before by Congress, which regulated the indirect elections. See Lei n. 4.321 de 7 de abril de 1964, DIÁRIO OFICIAL DA UNIÃO [D.O.U.] de 8.4.1964. According to the statute, the elections would be decided in a secret ballot. The Institutional Act, however, established that the votes should be cast openly (see Article 2 of the Institutional Act). That procedure was challenged in a point of order swiftly dismissed by the Speaker of the Senate. Questão de Ordem, Senador Aurélio Viana, DIÁRIO DO CONGRESSO NACIONAL [D.C.N.] de 12.4.1964, 95-66. Castello Branco had massive congressional support, winning with 224 votes, against 1 vote for each of his opposing candidates and 40 abstentions (probably stimulated by the open vote).

3 The idea that law should work as means to an end to government was burgeoning quickly after World War II. In Brazil, Francisco Campos, Carlos Medeiros Silva, and Manoel Gonçalves Ferreira Filho represent three generations of constitutional scholars, aligned with the military, supporting this view: governing no longer meant acting within a preexisting set of laws, but controlling the enactment of law itself (BARBOSA, 2012, p. 23). This mentality assumes that the executive branch has higher democratic credentials, and that its ability to shape legislation represents a more effective way to channel popular aspirations into law. This ability, nevertheless, might also support cunning manipulation of the law on the behalf of the executive branch’s own interests. The interesting question here – one that this article engages – is whether the law (particularly the election law) enacted during the dictatorship can be, for that reason, understood simply as a “democratic facade.” Renato Lemos (2004), discussing Kinzo’s (1988) and O’Donnell/ Schmitter’s (1986) work, criticizes this approach. He suggests that the “political hybridism” of the military dictatorship reflects deeper problems of legitimation of the regime. Therefore, the fact that the law under the civil-military rule was commonly the result of the manipulation or abuse of legislative power does not mean that it is irrelevant to develop a proper understanding of the period. On the contrary. As I have discussed elsewhere (BARBOSA, 2012, p. 353), the 1964 regime failed to control the institutions in charge of creating and applying the law more than once. Rules that were enacted with one purpose were interpreted and applied differently from the intent of the lawmakers. The dynamics of constitutionalism could hardly by contained and controlled as the architects of the 1964 legal system expected. For that reason, the motives of the Institutional Act No. 5 read: “The legal instruments that the victorious revolution granted to the Nation to its defense, development and well-being of its people are now serving as means to fight and destroy the revolution.”

4 For a schematic analysis of the rules governing the Brazilian electoral system during this period, see (NICOLAU, 2012, p. 104-118).

5 See, for instance, the work of Maria José de Rezende (2011), discussing Raymundo Faoro’s analysis of the 1984 political scenario within Congress, and the negotiations that preceded the indirect election of Tancredo Neves and José Sarney in 1985. Note that the government attempted to control the Electoral College by asking the Superior Electoral Court to impose the constitutional principle of party loyalty on the representatives and senators who would vote in the Electoral College, which would, ultimately secure the victory of Paulo Maluf, the government’s candidate. The Court, however, refused this interpretation and declared that the members of the Electoral College were free to vote in any candidate regardless of their party orientation. See Resolução n. 12.017, de 27 de novembro de 1984 (BRASIL, Tribunal Superior Eleitoral, 1984).

6 Aarão Reis, for instance, affirms: “Defeated in the first post-coup elections, in 1965, [the military] turned to brute force, reenacting the institutional acts” (AARÃO REIS, 2004, p. 41). On the other hand, Carlos Fico insists that the second Institutional Act relates to the episode of the 1965 elections, “but not only.” “It was also a hard-liners’ victory”: the success of the oppositionist candidates in Minas Gerais and Guanabara can be understood as “pretext” for the enactment of the Institutional Act No. 2, “not its main cause” (FICO, 2004, p. 75).

7 Cf. Mensagem Presidencial, Diário do Congresso Nacional de 2.3.1965, p. 30.

8 During Castello Branco’s government alone, one senator (Juscelino Kubitschek) and 67 representatives (out of the 409 representatives of the 1963-1967 legislature) were purged (BRAGA, 2014).

9 For an example of the government’s effort to lend credibility to elections, see Emenda Constitucional n.15, de 5 de julho de 1965 (BRASIL, 1965b), adopting a constitutional mandatory disclosure of the candidates’ assets and income sources to the electoral courts. The same amendment established a number of prohibited conducts in public administration, such as hiring new staff, or authorizing official loans to states or municipalities within the 90 days before elections.

10 It is important to clarify that these exchanges were not private. Political journalists were aware of the content of the letters, and heightened the polemic through their writings (CASTELLO BRANCO, 1965, p. 4).

11Diário do Congresso Nacional – Seção I, 23.3.1965, p. 1.129.

12 The electoral reform package promulgated on July 15 also included a new statute regulating political parties, the Law on Political Parties (Lei n. 4.740, de 15 de julho de 1965) (BRASIL, 1965f).

13Diário do Congresso Nacional – Seção I, 23.3.1965, p. 1.128-1.129.

14 According to Fico (2004), “civilians like Magalhães Pinto and Carlos Lacerda intended just another ‘moderator-style’ [moderadora] military intervention.”

15 According to Viana Filho, Castello Branco resisted that alternative, despite the insistence of many of his political allies. However, Castello Branco had conceded the prorogation of his own term after a resolute refusal of the idea. He might have reconsidered his initial position about reelection, had the conditions become favorable to his bid (which ultimately did not happen). Even after the enactment of the second Institutional Act, which expressly prohibited reelection, the press reported about maneuvers to convince Castello Branco of running for a second term (CASTELLO BRANCO, 1965, p. 4).

16 The Brazilian Labor Party [Partido Trabalhista Brasileiro] was the party of Vargas and of the deposed President João Goulart, and was the source of the most immediate political opposition to the new regime. At the beginning of the 1963-1967 legislature, it had 116 representatives. For the number of seats of each political party during the beginning of the military regime see (AZEVEDO; RABAT, 2012).

17 The party called the National Democratic Union [União Democrática Nacional] was the foremost supporter of the newly instituted government, and, in its origins, united most of the political opposition to the Vargas regime. The UDN had 91 representatives at the beginning of the 1963-1967 legislature (the Chamber was, then, formed by 409 representatives).

18 The Social Democratic Party [Partido Social Democrático] was the biggest party in the Chamber of Deputies, with 118 representatives at the beginning of the 1963-1967 legislature. Founded by allies of Vargas, it was a dominant political force under the 1946 Constitution, having elected two presidents and several governors.

19 In Portuguese, the text reads: “[São inelegíveis] os que tenham sido condenados por haver comprometido, por si ou por outrem, a lisura e a normalidade de eleição, através de abuso do poder econômico, de ato de corrupção ou de influência no exercício de cargo ou função pública, ou que venham a comprometê-las, pela prática dos mesmos abusos, atos ou influência.” Emphasis added to the expression that would later be suppressed by the presidential veto.

20 See Diário do Congresso Nacional, 20.10.1965, p. 727. However, it seems the government itself felt that keeping that procedure in place after this episode would impair the credibility of inter-branch negotiations. In the Legislative Reform, initiated by the government in October 1965, Castello Branco sponsored the adoption of a line item partial veto. The provision was laconically justified: “On the partial veto, considering its origins and reason, the proposal declares it should encompass only the full text of a single item [artigo, parágrafo, inciso, item, número ou alínea].” See Projeto de Emenda à Constituição n. 7, de 4 de novembro de 1965 (BRASIL, 1965e) and Emenda Constitucional n. 17, de 26 de novembro de 1965 (BRASIL, 1965d).

21 According to the official statement presented to the Regional Electoral Court of Minas Gerais, as part of the documents required for Paes de Almeida’s registration as a candidate for the office of state governor. The document was signed on July 31, 1965. Pio Soares Canedo, his running mate, had something around 12 million cruzeiros (MINAS GERAIS, TRE-MG, 1965a).

22 Information on the state elections is available on the Tribunal Superior Eleitoral website: (last visited Sept. 23, 2014).

23 In Portuguese, “Tião” is a standard nickname for “Sebastião.” In the 1962 elections for the Chamber of Deputies, when the voter had to actually write down the candidate’s name, Paes de Almeida authorized some variations of his own name, including a very common and disseminated misspelling, “Sebastião Pais de Almeida” (MINAS GERAIS, TRE-MG, 1962).

24 Although there are good reasons to imagine that Paes de Almeida expected political return for all the charity he was involved in, it is possible that philanthropy was really part of his family tradition, perhaps reinforced by the connection he made with the Jesuits during his school years at São Bento College and São Luís College. Apparently, he had even attended a seminary during his youth (CONCORDATA, 1975).

25 See the appended copies of Processos n. 20.198/65 e 20.360/65, Conselho Administrativo de Defesa Econômica (CADE). In September 1969, the Council held Paes de Almeida’s companies not guilty of the charges of anti-trust violations (CADE, 1969).

26 See the appended copy of Processo n. 173.526/65, Receita Federal do Brasil.

27 For all the following citations, see Recurso Ordinário n. 2.869 (BRASIL, Tribunal Superior Eleitoral, 1965).

28 See Emenda Constitucional n. 16, 1965 (BRASIL, 1965c).

29 Actually, Castello Branco turned to the same strategy twice, as he vetoed words of the legislation regulating the political parties, enacted on the same day as the Election Code and the ineligibilities legislation. This time, the veto aimed at making it harder for political parties to continue legally exist. See also (CARNEIRO, 2009).

30 The expression “democratic legality” [legalidade democrática] was employed by Milton Campos to depict the goal of the “revolutionary process.” Campos’s tenure as Castello Branco’s first Minister of Justice came to an end after the 1965 state elections, when it became clear that the second Institutional Act was inevitable. He wrote a short article explaining his decision to run for senator in the 1966 elections. His commitment to the idea of democratic legality was still solid, despite the “many obstacles that had been opposing this ambition.” The words of Campos provide a good account of the dilemma faced by Castello Branco’s government: “We must distinguish between the Revolution and its process. The Revolution shall be permanent as idea and inspiration (…). The revolutionary process shall be a short transitional period, because its continuity tends to underpin arbitrariness, eliminating the law, fostering unrest amongst the citizenry, and paralyzing social evolution” (CAMPOS, 1966, p. 53).

Received: June 22, 2015; Accepted: March 06, 2017

Leonardo Augusto de Andrade Barbosa

Doctor of Law, University of Brasília. Professor in the Master’s Program in Legislative Affairs (Center for Continuing Education and Professional Development/ CEFOR – Chamber of Deputies)

Creative Commons License  This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.