The Overthrow of Getúlio Vargas in 1945: Diplomatic Intervention, Defense of Democracy, or Political Retribution?

Late in October 1945, after 15 years in power and in the midst of what he presented as an irreversible redemocratization program, the final phase of which would be presidential elections early in December, Brazilian dictator Getúlio Vargas became the major political casualty of a coup d’état engineered by senior army commanders and embittered liberal adversaries. A year later, Vargas publicly would claim that he had been the “victim of agents of international finance” who were bent on impeding Brazil’s industrialization, and that the “inopportune” meddling of United States Ambassador Adolf Berle, Jr. was conclusive proof of that accusation, one that he would repeat during his successful bid for the presidency in 1950.¹ Berle’s tenure in Rio de Janeiro was short (January 1945–February 1946), but he became the most controversial modern United States envoy to Brazil and gave its political crisis in 1945 a diplomatic dimension that requires analysis. What precisely was Berle’s role in the events leading to Vargas’s overthrow? How did Washington react to these developments and to Berle’s management of U.S. interests in Brazil? How did Berle define those interests? The search for answers to these questions requires scrutiny of Brazil’s domestic crisis, particularly the respective intentions of Vargas and his adversaries. Such an assessment casts strong doubt on what might be called the liberal-military thesis regarding Vargas’s deposition, i.e., that in 1945 he maneuvered to perpetuate his personal rule and therefore had to be removed forcibly to ensure free elections. This was the rationale advanced by liberal politicians at the time to justify their anti-Vargas conspiracy, and it was dramatically reinforced when military leaders finally intervened to depose the dictator. The coup d’état of October 29, in other words, gave legitimacy to the liberals’ interpretation of Vargas’s conduct. Then the explanation was adopted by the U.S. Embassy and went on to become an integral part of most accounts of that period.² Reexamination of the events in 1945, however, based on new documentary evidence buttressed by logic and an unavoidable dose of conjecture, not only clarifies Berle’s relationship to them, but suggests a more convincing view of the process that culminated in military intervention against Vargas.

The Berle Mission

The credentials that Berle carried with him to Brazil were impressive. A lawyer, he had served as an army intelligence officer during World War I, and in the 1920s he taught corporate finance at the Harvard Business School, later becoming a professor of law at Columbia University. His coauthorship with Gardiner Means of The Modern Corporation and Private Property (1932) consolidated his national reputation as an authority on corporate law. That same year, Berle joined Franklin Roosevelt’s “brains trust,” and went on to become one of the architects of the New Deal. During the 1930s, he occasionally took an active part in Latin American affairs, serving on the delegations to the inter-American conferences in Buenos Aires in 1936 and in Lima two years later. Early in 1938, at the age of 43, he became an assistant secretary of state, a post he would hold until receiving his Brazilian assignment. During World War II, he handled various special tasks: as senior adviser to the delegation to the Havana Conference in July 1940, as director of the Department of State’s intelligence activities, and as supervisor of negotiations for postwar air bases, which brought him into intimate contact with Brazilian affairs and Brazilian leaders. Roosevelt judged Berle to he a man of skill and vision, and when Jefferson Caffery ended his successful seven-year mission in Rio de Janeiro in 1944, the president thought Berle an excellent replacement. “You know fully the great importance of that post at this particular time,” Roosevelt wrote to him in November, “and . . . I have come to the conclusion that I need your ability and experience in order to insure . . . the whole hemispheric policy against the insidious influences of which you are aware.” Brazilian authorities, for their part, were gratified when word of Berle's forthcoming appointment surfaced late the following month. “We ought to be pleased because . . . he is a truly eminent personality,” Foreign Minister Pedro Leão Velloso confided to Vargas.³

Aside from his wide-ranging knowledge of public issues, Berle possessed two other attributes that influenced his approach to relations with Brazil: a deep-rooted social conscience and a Calvinistic zeal to solve problems and do good works. His father, a Congregationalist minister, had been active in the turn-of-the-century social reform movement in Massachusetts, and his mother, before marriage, had lived among the Sioux Indians as a missionary. As a result of his upbringing, a “puritanical” Berle himself displayed a civic-minded, reformist bent. In 1923, he served as attorney for the Pueblo Indians in Arizona and New Mexico, defending their land titles against government encroachment. His research on the modern business enterprise made him sharply critical of the concentration of wealth and the abuse of corporate power, and his attachment to the Roosevelt camp in the 1930s was born of a shared conviction that the federal government should assume greater social responsibility.⁴ Given this mind-set, Berle subsequently proved to be acutely interested in Brazil’s social and economic situation, which he examined with the impatience of the dedicated reformer.

The ambassador arrived in Rio de Janeiro late in January 1945, enthusiastic about the challenges and convinced of the long-range significance of Brazil to the United States. “Twenty-five years from now,” he predicted in a letter to Harry Truman shortly after Roosevelt’s death, “American relations with Brazil will determine our position in the hemisphere.”⁵ Two international political factors reinforced Berle’s determination to strengthen and give what he considered to be new directions to relations with Brazil. For one thing, the military regime in Buenos Aires was repugnant to him, and he feared what he saw as the fascistic, expansionist policies advocated by Colonel Juan Perón and his associates. “Perón means business,” he recorded in mid-February. “He is building up his army, has made an armed camp of the Chilean frontier and the salient between Rio Grande do Sul and Paraguay; [and] is shooting for a total army of 400,000. . . . ” He therefore endorsed a hard line toward the Argentine government—“build a high wall around Argentina” was his recommendation⁶—at the same time that he lobbied for a special regard for Brazil’s wartime services and general cooperativeness. “Our policy towards Argentina,” he urged Truman, “has to he handled with care for Brazilian feelings.” Berle’s conception of Brazilian-American relations stopped short of a permanent bilateral alliance—he was too egalitarian and too suspicious of power politics for that—but certainly the contrast between the international policies of the two South American countries was so stark, he thought, that Brazil clearly merited preferential consideration, and he bristled at suggestions that Vargas and Perón represented the same phenomenon. “To . . . confuse the Vargas Government with the Government of Perón . . .,” he told a colleague in the Department of State, “is simply exhibiting ignorance or attempting to mislead the public: they are quite different in origin and do quite different things.” The ambassador also sent a reminder to Truman in September: “Vargas kept his obligation to the hemisphere and was our most active ally. Argentina and Perón did the opposite.”⁷

Berle was also concerned about undefined European designs on South America, which made him all the more keen to bolster the long-term U.S. position in Brazil. His uneasiness about perceived European machinations in that country was rooted in “an innate, old-fashioned distrust of perfidious Albion,” an intense hostility toward Charles de Gaulle and his Free French movement,⁸ and a strong aversion to communism. During the war, Berle had been shocked by the Machiavellianism of proposals by the British secret service for covert action in South America,⁹ and as soon as he arrived in Brazil he found apparent signs of “French intrigues” there. “As far as I can see,” he wrote in his diary, “the French . . . are already engaged in setting the stage for an imperialist plunge; and there is no question whatever that they regard South America as vulnerable.” The ambassador kept an attentive eye on real or alleged European activities in Brazil in following weeks, and saw the problem as an enduring one. “If anything is clear it is that the great powers in Europe are not going to let South America alone,” he privately commented in April. In a letter to Truman a few days later, he repeated his warning about the designs of “our European allies” on Brazil.¹⁰ Ultimately, the Russian threat loomed as the most serious, in Berle’s view. Unless the United States offered resistance to the Soviet “concept of world empire,” Moscow would attempt to dominate the Western Hemisphere, he pondered in May. “Though [the] communist organization [in Brazil] has not yet proceeded very far,” he pointed out to the Department of State in July, “its lines here follow almost exactly [the] lines of German organization from 1937 to 1939.. . .”¹¹

To meet the problems involved in protecting United States interests in Brazil in the face of such challenges, Berle insisted, first of all, that the embassy and consulates needed knowledgeable officers and analysts. In this regard, the skepticism that he had expressed in Washington became sharp disgruntlement once he reached his post. Indeed, after his first day at the embassy, he noted critically that Caffery had run it “as a personal show with half a dozen men.” Over the next few days, his mood about the embassy remained sour. “It is a collection of spare parts,” he lamented, “needing organization to replace the small, ultrasecret and centralized group which Caffery built up and which he has pulled out with him.” He complained to both the Brazil desk and the personnel division at the Department of State that he did not have “a really top-flight economic man,” and that he was “woefully weak” in the area of political reporting, since only his press attaché had any experience and contacts. In ensuing weeks, further transfers and the periodic arrival of replacements chosen, he thought, in a manner “incomplete, hasty and without reference to the needs of the post,” led Berle to break protocol and send a “painfully blunt” report to Secretary of State Edward Stettinius, Jr. and Assistant Secretary of State Nelson Rockefeller. “[The] Personnel Board should also have in mind that the language of Brazil is Portuguese, and its second language, French,” he said tartly, pointing to the fact that none of his new subordinates spoke Portuguese. “Spanish will get you meals at a hotel and the ill will of the Brazilians.” In addition, he continued, whereas the embassy in Havana had 16 officers, that in Mexico City 19, and the mission in Buenos Aires 15, he had only 9 and 2 of those were being transferred. “But we have three times the population and six times the territory of any other American post,” he argued. He further remonstrated that “you cannot go on dealing with Brazil as though it were a vast empty space populated by Hottentots.”¹²

While he battled the department bureaucracy in an effort to build up a staff commensurate with the tasks of diplomacy in Brazil, Berle also sought to refocus the work of the embassy. Recording his thoughts about Franklin Roosevelt the day the president died, Berle revealed the social dimension of his own conception of mission. “So his foreign policy not merely included friendship with other countries, hut also friendship with the various classes, especially the large humbler classes who do not usually find expression in their governments,” he wrote with obvious approval. “So his policies had not only length and breadth, hut also depth, and brought into the consideration of current affairs problems which normally do not find the place to which they are entitled in foreign relations.” Thus, Berle the New Deal creative thinker, who had counseled the White House on hanking, transportation, agriculture, education, and a host of other socioeconomic issues, saw the role of Berle the ambassador as one requiring greater attention to, and even paternalistic involvement in, the developmental problems of Brazil. “Broadly speaking,” he commented in a private letter late in March, “the best use that can be made of this Embassy seems to he found on three lines: elementary and vocational education; solution of food and agriculture [problems]; and a means by which the benefits of industrialization will be distributed among the population rather than concentrated in the hands of cartels.”¹³

One of Berle’s initial surprises on taking charge was his discovery of “the detachment of the Embassy from the life of what must he 98% of the population.” The contacts of chief subordinates appeared to be limited to upper-class society; coordination between the embassy and other United States agencies in Brazil engaged in socioeconomic projects was minimal; and the embassy seemed likewise to be isolated from São Paulo, the country’s industrial center. The new envoy therefore quickly set about getting his staff “focused on some long-range work respectively on food, health, education, and price levels."¹⁴ At the same time, he set up lines of more intense communication not only with the various U.S. consulates and agencies, hut also with private companies, Brazilian government offices, and even representatives of various Brazilian political and social groups.

Berle conceived of the embassy as not only a center for study and observation, but for advocacy as well. In other words, he was convinced that he both understood the country’s problems and had the solutions—one sympathetic biographical essayist has acknowledged that Berle had a permanent reputation for “arrogance”¹⁵—and that, in the interests of both nations, he should tutor the Brazilian government and public, especially since, in his eyes, Brazil’s heritage was one of incompetence and disinterest, if not worse. “Every once in a while,” he confessed at one point, “I get angry at the Portuguese Empire: what were they doing during the 300 years that they had this country?”¹⁶ From the outset, he evinced little respect for the intelligentsia as a source of governing talent and developmental impulse, commenting typically that “the terrible thing is that these people are not showing very much in the way of deadweight lift for the masses,” and he was likewise suspicious of the role played by the entrepreneurial class in creating an unjust social system. With great confidence in his own judgment in such matters, Berle showed no reticence about offering advice to Brazilian audiences. Early in February, for example, he called at Itamaraty where, with Foreign Minister Velloso and other Brazilian diplomats present, the question of how to raise Brazil’s standard of living came up. “I pointed out that I had no right to intervene,” he wrote afterward, and then with obvious self-satisfaction recalled, “However, since they had asked for a discussion on the subject, I opened up . . . [and] for two mortal hours we talked Brazilian economics.” Berle also looked around for younger members of the elite that he could instruct about developmental issues—in his words, he wanted to find “some young men to talk to about handling of obvious social and economic problems”—and was delighted when one appeared saying that he represented a group called Resistência Democrática. The nature and tenor of their initial conversation was revealed by the Brazilian youth in a subsequent letter to Berle, thanking him for his decision “to see regularly a small group of us in order to help us analyse in the right light our major problems and plan a rapid and thorough solution for them. . . ."¹⁷

A well-meaning paternalism characterized Berle’s reaction to various Brazilian problems. During his first visit to São Paulo in March, he discovered, for example, that the city did not have an adequate system of public schools. “This settled me on one thing: we have got to make Brazil want education,” he confided to his diary. Two days after returning to Rio de Janeiro, he huddled with a publicity specialist from the local branch of Rockefeller’s Office of the Coordinator of Inter-American Affairs, “and we got to work on a campaign to try to work up a popular demand for elementary education,” he noted. “This is a very un-Ambassadorial line of activity but with luck we might do some good.” In ensuing weeks, Berle and Rockefeller’s agents hammered out further details of the publicity drive, and late in April he placed the prestige of the embassy behind a campaign by the country’s largest newspaper enterprise for “a country-wide chain of schools,” contributing an article on the New Deal’s educational program. In midyear, when the Brazilian navy requested the transfer to its fleet of more than 30 U.S. warships, including 2 cruisers and 2 aircraft carriers, Berle admonished the State Department that “one or two cruisers” would he sufficient. “The money and effort used in organizing a naval force . . . would be infinitely better spent on putting in an internal transport system, and building and maintaining public schools,” he argued.¹⁸

Berle’s mistrust of Brazilian industrialists, born of his study of the behavior of capital in the United States and his observations of Brazilian conditions, enhanced his interventionist tendencies. His early impression of Brazil was that it was in a position similar to that of the United States “in 1880 or perhaps better 1900,” i.e., the era of unbridled captains of industry. Within another ten days, he had concluded that the Paulista manufacturers were simply “looking for profits and plunder in a huge way,” and seized the occasion of his aforementioned talk with Brazilian diplomats in February to lambast “the high price policy of businessmen.” In his sincere desire to see a more equitable distribution of income in Brazil, Berle was concerned about the lack of economic discussion in the presidential campaign growing out of recent moves by Vargas to open up the political system. The ambassador therefore set about “cooking up a scheme” whereby he would give a speech that he apparently believed would stimulate or even force Brazilian political leaders into dealing with socioeconomic questions. He managed finally to wrangle an invitation from the Associação Brasileira de Imprensa to deliver such a speech on April 8. “In an effort to put some substance behind the current political maneuvering,” Berle wrote afterward, “I picked industrialization, and tried to show that the social standards forced on American industry in the early days by the combined agrarian and labor movements had proved the best aid to their prosperity.” Berle also endeavored to prod businessmen, U.S. and Brazilian, into raising wages before workers were compelled to go on strike. “Why should not the manager of a company become the representative of the laborers?” he asked an officer of the American Foreign Power Company two days after his speech—a pitch he also made in a talk to the American Chamber of Commerce early in May. Not surprisingly, Berle saw the antitrust law that Vargas promulgated in June as a reasonable, even necessary, measure. “It seems to me that the people who are yelling are the people who have been making two or three hundred percent a year and don’t want to let go of any of it,” he said.¹⁹

The ambassador recognized that Brazil’s long-range industrial growth was dependent in part on a solution to the fuel problem, and he also thought that the country’s dependence on external sources of petroleum was dangerous. The quickest way to develop oil resources would be for Brazil to open its doors to foreign companies, but Berle had no illusions about the U.S. oil industry. When his petroleum attaché alerted him late in March to the fact that the Vargas government was debating such a step, he worried that Brazil might repeat the examples of Mexico, Venezuela, and Bolivia, which had made politically explosive monopolistic concessions to foreign companies. Vargas, he hoped, could find a “somewhat better and safer” way to handle foreign involvement. One possible solution, he suggested to the Department of State in July, would be for the two governments to set up a bilateral watchdog committee to monitor the activities of foreign companies and ensure that “the [Brazilian] public gets the benefit of the resources.”²⁰ The plan of the Conselho Nacional do Petróleo to establish refineries in Brazil to process imported crude met with Berle’s approval, even though Brazilian law forbade foreign participation in refining. The foreign firms that controlled the Brazilian market—Standard Oil of New Jersey, Atlantic Refining Company, Shell-Mex, and the Texas Oil Company—vigorously objected, but as Berle saw it, their opposition stemmed from the fact that Gulf Oil had offered to supply crude to the projected refineries, thus breaking their stranglehold. Consequently, he was incredulous when a Standard Oil spokesman told a member of his staff that the company was concerned about the expense of the project to Brazil. “Standard Oil of Brazil made a net profit after depreciation of 114% last year . . .,” Berle pointed out. “Their sudden solicitude lest Brazil might pay too high a cost—to Brazilians—for refining oil, in Brazil, strikes me as merely funny.”²¹

Berle’s penchant for reformist activism, i.e., his desire to do something about Brazil’s fundamental social and economic problems, made him an especially keen observer of the national political scene. One thing stands out clearly from his official and private comments in that regard: a strong sympathy for Vargas. His first meeting with the Brazilian leader occurred on January 30, when he presented his credentials. “Getúlio Vargas is a quiet, solid, retiring man, who impresses me as being the common denominator of Brazil, at least partly because no leader of his caliber has emerged,” he jotted in his diary afterward. “In this respect he is in somewhat the position of President Roosevelt: [he is] not particularly anxious to seize power for its own sake, but [it is] by no means clear that anybody else can steer the country as well.” Berle seems to have been predisposed to like Vargas in any case, despite the repressive nature of the regime. The new envoy obviously was familiar with Rio de Janeiro’s war record and, too, he probably was influenced by Roosevelt’s undisguised sympathy for Vargas, whom the United States chief executive viewed as a reformer somewhat in his own mold. “He said that he knew President Vargas . . . very well,’’ James Farley, former postmaster general, recalled Roosevelt as having remarked to him in 1941, “adding [that] Vargas was not the least like other South American leaders, being more progressive in his views and a more capable administrator.” Berle himself heard that and more from Roosevelt during a stopover in Washington on his way back to Brazil from the Mexico City Conference in March 1945. Along with Stettinius and Velloso, he listened to Roosevelt declare “that he hoped General [sic] Vargas would be reelected President but did not want to take any hand in it lest it do more harm than good.” Roosevelt went on to state that “Vargas had a world point of view and had done the best he could with the situation, his dictatorship being a result of the wide illiteracy in the country."²²

Abertura and Opposition

In Berle’s own estimation, Vargas’s stock rose considerably as the process of redemocratization got under way. The Brazilian leader, in speeches of November 10 and December 31, 1944 had promised a restoration of political liberties, and the regime then proceeded to relax press censorship in the early weeks of 1945, terminating it altogether late in February. Vargas issued a decree on February 28, moreover, guaranteeing that within three months a date would be set for presidential and congressional elections, a step that served as the green light for the launching of the presidential candidacy of Brigadier Eduardo Gomes by the liberal opposition and the counter-candidacy of Minister of War Eurico Dutra by the government. The next major step taken by Vargas (in April) was to proclaim a political amnesty that resulted in the release from prison of scores, if not hundreds, of the regime’s adversaries. During April and May, furthermore, three major political parties—the opposition União Democrática Nacional (UDN), the government-sponsored Partido Social Democrático (PSD), which endorsed Dutra, and Partido Trabalhista Brasileiro (PTB)—were organized. Then on May 28, Vargas promulgated an electoral law stipulating that national elections would be held on December 2; that the chief executive and cabinet members would have to resign office no later than September 2 in order to be eligible to run for office; and that state elections would take place on May 6, 1946.²³

The question on the minds of all political observers, however, was what precisely were Vargas’s real intentions? Leaders of the essentially middle-class UDN had the answer even before redemocratization began, and they saw in his every move in 1945 confirmation of their conviction: Vargas wanted to continue in power and constantly maneuvered with that goal in mind. Sounding a continual alarm, the UDN chiefs offered two basic scenarios in the early months of 1945: Vargas would attempt a golpe similar to that of November 1937 or he would resign the presidency to run legally for another term. The anxieties and even hatred of UDN organizers were fueled by the emergence of the famous queremista, or queremos Getúlio, movement in midyear—promoted by elements of the newly formed PTB—and the tacit alliance established at the same time between the government and the Partido Comunista Brasileiro, which joined PTB sectors in calling for election of a constitutional assembly before the holding of presidential elections, a demand that implied postponement of the latter and Vargas’s continuation in the presidency at least in the meantime.²⁴

The liberal, and ultimately liberal-military, view of Vargas’s plans is unconvincing. In the first place, it demands at the outset acceptance of the dubious proposition that Vargas plotted to thwart the elections and somehow perpetuate his own rule in the face of what he knew to be the nearly united opposition of army commanders—when nobody understood better than he the political decisiveness of military support. After all, he had first gained power through armed revolt; had survived civil war in 1932 only because the army remained loyal; and, after 1937, had headed a regime born of military intervention and sustained by force of arms. Beginning late in 1944, moreover, there was direct pressure on him from the highest military levels to redemocratize. On October 23, for example, on returning from an inspection visit to the Força Expedicionária Brasileira (FEB) in Italy, General Dutra privately underscored to Vargas the incongruity between Brazil’s international and national situations, i.e., the fact that the country was an active partner in the Allied coalition doing battle against fascism in Europe while being governed itself by a dictatorship, and he urged the reestablishment of “democratic representative institutions." A few days later, at Dutra’s insistence, General Pedro de Góes Monteiro, along with the minister of war the major architect of the 1937 golpe, talked to the chief executive with equal candor about the need to end the dictatorship. Monteiro had served as army chief of staff (1937-44) until poor health forced him to assume less demanding duties as Brazil’s representative on the Inter-American Political Defense Committee in Montevideo, but arguably he remained the most influential military figure in Brazil and certainly he was the most political. At his meeting with Vargas, he remonstrated that discontent with the regime had become generalized and that political liberties had to be restored. Late in December, Dutra even wrote to Vargas recommending that he not wait until the end of the war to hold presidential elections.²⁵ In response to this pressure, Vargas set his minister of justice to work drawing up a liberalization program in consultation with the two generals. Monteiro, early in February 1945, before Vargas’s first decree on the elections, sent another signal by publicly calling for democratic politics in an interview that was widely printed.²⁶

The announcement later that month of the candidacy of Eduardo Gomes, one of the country’s most prominent military leaders and one of the original tenentes of the 1920s (i.e., rebellious young army officers in search of national reform), was an obvious sign that undetermined but significant segments of the officer corps were among the regime’s enemies. Dutra’s eagerness to play the role of official candidate, furthermore, created a situation in which, if the liberal thesis is accurate, Vargas was conspiring on his own behalf against the very man who had been in control of the army for almost nine years (December 1936-August 1945) and therefore had built up a massive base of personal support among senior officers. When Dutra resigned in August in order to establish his eligibility to run for office, moreover, his replacement was Monteiro, who agreed to serve only after securing from Vargas a formal promise that elections would be held as scheduled on December 2 and that the government would not enter into any understanding with the communists. In a speech on assuming office on August 10, Monteiro indicated that the army would guarantee a “peaceful and honest” presidential election, a declaration that elicited enthusiastic praise from the liberal opposition press.²⁷

The return of the FEB contingents in July and August increased military pressure against any continuismo on Vargas’s part, since various key FEB officers were imbued with a desire to end the anomaly of their having made sacrifices on the battlefields of Italy to defeat dictatorships while representing one.²⁸ In the remaining weeks before Vargas’s downfall, army spokesmen made no secret of their determination to ensure the elections. Monteiro gave an interview to the opposition Diário Carioca on September 3 in which he was emphatic; “As I have already declared several times,” he said, “the government will comply with the laws dealing with the electoral process and the holding, on December 2, of the election to choose the president of the Republic.” Four days later, the press carried a note from UDN headquarters quoting renewed assurances received from Monteiro that “the armed forces will guarantee the freedom of the elections,” and that they would be “irrevocably held on December 2 next”—a statement that led Virgilio de Melo Franco, a founder of the UDN and executive director of Gomes’s campaign, to send a jubilant letter to a fellow udenista stressing the significance of Monteiro’s unmistakable commitment. Even General Renato Paquet, commander of the Vila Militar outside Rio de Janeiro, who by all accounts was the senior general most loyal to Vargas, was unequivocal in stating on September 13 that the army must maintain “constant vigilance” in order to guarantee the elections on schedule. “By this time the dictator of Brazil must know for certain that he does not have the support of the armed forces for changing the face of our social order at the whim of his [own] interests,” the Correio de Manhã editorialized two days later. “The Army already has made it known to him that it will not permit any postponement, or modification, of the decree that established the date of December 2 for the presidential elections.” The Carioca daily, which had been waging a relentless, embittered campaign against Vargas ever since suspension of censorship, was absolutely right—and military leaders continued to drive the point home. In an interview on September 19, General Angelo Mendos de Moraes, the director general of personnel, who was very close to Dutra, vowed that the army would stand behind compliance with the election schedule. That same day, General Mario Ary Pires, commander of the Fifth Military Region, wrote Monteiro from his headquarters in Curitiba, pointedly referring to the “firm decision” of the units under his command to oppose any attempt to disrupt the country’s peaceful transition to democracy, “whether it comes from below or from above.” Both Monteiro and Vargas had made reiterated statements that “free and honest” elections would take place on December 2, Pires noted. “To us the name of this or that candidate is not important,” he said. “What we want is the exact fulfillment of that promise. . . ."²⁹

At the end of the month, there occurred an even more significant demonstration of military determination. In the face of continuing effervescence in military and opposition circles because of the queremista movement and its apparent alliance with the communists, Monteiro secured from Vargas on September 28 another flat pledge to proceed with the program for electing a new president, and then Monteiro convened a meeting of generals serving in the Rio de Janeiro area to reassure them as to that commitment and authorize them to disseminate it within their respective commands. He also wrote Pires advising him that in view of the chief executive’s renewed guarantee there was no need for any “political pronouncements” by military leaders. The following day, the army chief of staff, General Cristovão Barcelos, made a strong declaration to the press in which he emphasized the “union and unanimity” of army commanders on the issue of elections on December 2 and warned that they would impede any effort by “extremisms of creed and ambition” to interrupt an orderly process of succession. In a private note to Monteiro, Barcelos explained that with that statement he had sought to demonstrate “that never in our military and political history were the generals so united as in the delicate phase we are going through. . . .” His interview was followed by a special bulletin released by the commander of the First Military Region (Rio de Janeiro) on October 3 carefully underscoring the government’s promise.³⁰

A second factor to he taken into account, in evaluating the liberal-military thesis, is that in 1945 Vargas repeatedly affirmed, both privately and publicly, that he intended to leave office after presiding over the election of his successor. He assured Dutra on March 4, for example, that “in no way whatsoever” would he he a presidential candidate, “authorizing me to transmit his intentions to the military,” Dutra recalled; in a meeting with Monteiro at that time, the chief executive specifically rejected the idea of becoming a candidate himself.³¹ A few weeks later, on May Day, Vargas told an enthusiastic crowd in Vasco da Gama Stadium, “I will maintain order, I will hold the elections and I will transfer power to whoever is legally elected by the people.” At the end of that month, of course, he decreed December 2 as the election date. At this juncture, when his political ally João Neves da Fontoura impressed on Vargas that the situation remained volatile because the opposition did not believe in his sincerity and suggested that he hand over power to Monteiro and allow him to oversee the elections, Vargas authorized Fontoura to undertake the necessary demarches. Monteiro agreed, but insisted on UDN endorsement of the move. When UDN leaders, including Gomes, were consulted, however, they rebuffed the offer, not trusting either Vargas or Monteiro. The general later recalled that Vargas had been “very sensitive” to charges that he was plotting to remain in power, and had gone so far as to empower him to propose to UDN chiefs the formation of a national union cabinet to assist him until the elections were over, but that once more they had refused.³² “Dr. Getúlio Vargas . . . has been giving repeated proof that he is no obstacle to the [political] unification of Brazilians . . . ,” an exasperated Monteiro wrote in July. “I can’t understand the reasons why the leaders of the União [Democrática Nacional] reject [his proposal]. After all, an opposition has never made such tremendous conquests in so short a time: elections, amnesty, freedom of press and radio . . . and so forth.”³³

According to the subsequent testimony of his chef de cabinet, Luiz Vergara, who had worked intimately with Vargas in that capacity for nearly a decade, the president was tired during this period, and “firmly aimed to maintain order, ensure the free exercise of the vote, and transmit the government to his legally chosen successor.” In ensuing weeks, Vargas consequently continued to proclaim his intention to proceed with the elections. Late in August, he repeated to a throng of queremistas who marched on his residence that they would be held on schedule. “Nobody can stop them. . .,” he said. “I want to preside over those elections, so that the Brazilian people may freely choose their representatives. . . ." He next allowed the September 2 deadline to pass without resigning, thus relinquishing his legal opportunity to seek another period in office—and this, as Monteiro and the U.S. Embassy recognized, when he undoubtedly would have triumphed at the polls. Five days later, in an Independence Day speech, Vargas pointed to the steps he had taken to open up the political system and hammered on the now familiar key. “As Head of the Government I promised free and honest elections,” he recalled, “and I want to preside over them with absolute impartiality and certainty. I seek nothing else. I have said so on various occasions and I reaffirm it now.” Later that month, he gave Monteiro the solemn pledge in that regard which the minister of war had requested in order to be able to reassure the high command. As Monteiro informed General Pires, Vargas “once again repeated that the elections will be held indeed on the 2nd,” and “that he is not a candidate.” For Monteiro there was no doubt. “Does the president himself need to give any further satisfactions?” he asked rhetorically. Then on October 3, the anniversary of the Revolution of 1930, when a huge queremista rally took place in front of the presidential palace, Vargas took pains to emphasize anew his intention to retire from office. “Before God, . . . before the Brazilian people, . . . I reaffirm that I am not a candidate and desire only to preside over elections worthy of our political education, delivering the government to my successor legally chosen by the Nation,” he declared.³⁴

The critical thing, however, in regard to Vargas’s avowed disinterest in continuing in power was that his adversaries did not believe him—a fact related to another important dimension of the political struggle in 1945: their profound desire, even determination, evinced from the very onset of liberalization, to remove him from office before the elections, in part, one senses, as an act of political and in some cases even personal revenge. Monteiro, on his return from Montevideo late in 1944, had been deeply impressed by the obduracy of Gomes’s supporters, who were ready to “go to the ultimate consequences” and take “drastic action” against Vargas. In the following weeks, there was a concerted effort by leading members of the future UDN, among them Colonel Juarez Távora, a former tenente and hero of 1930, to mobilize support within the Rio de Janeiro garrison. In private conversation with neutral British observers, UDN spokesmen early in 1945 were candid. Ambassador (Sir) David Gainer on March 3 heard from Gomes himself that Vargas “would not risk an honest election” and planned to continue in control of the government. “The opposition were, however, determined to do all in their power to force the President to resign before the elections . . . as it was only in this way that fair elections could be held,” Gainer reported, paraphrasing Gomes. “They would continue to push this plan [to overthrow Vargas] regardless of what might happen.” According to José Américo de Almeida, the “official” candidate in 1937 whose run for the presidency had been aborted by the golpe that year and who was a prominent member of the UDN conspiracy, he and his colleagues by this time were already discouraged by their failure to organize an “armed reaction to exterminate the Estado Novo,” and, consequently, they decided to appeal to the two generals who bore primary responsibility for establishment of the dictatorship. While associates sounded Monteiro, Almeida was assigned the task of trying to convince Dutra of the need to use force against Vargas. Dutra was noncommittal, but the plotters were undaunted. “We are still in close contact with the Minister of War and I, personally, hope that we shall finally be able to persuade him to depose Getúlio and take over control of the country whilst elections are arranged,” Almeida confessed to one of Gainer’s subordinates. The UDN should be concentrating on developing solutions to national problems, Almeida realized, “but unfortunately,” he confided, “many of our people are bent only on having Getúlio’s blood."³⁵

The persistent editorial onslaught on Vargas’s alleged continuismo by the pro-Gomes press, such as the Diário de Notícias, launched in March as soon as censorship was effectively lifted,³⁶ was a public sign of the fact that, to UDN spokesmen, it really did not matter how Vargas proceeded—they wanted him out of the presidency, in any circumstances, before redemocratization was complete. Gomes and his associates made Vargas’s unsuitability to oversee the elections a major theme of campaign speeches,³⁷ and once queremismo made its appearance in midyear, their frequent charge was that he was conspiring against the elections. “Remember’37!” Gomes exclaimed in a speech on August 12. The fact that (according to Monteiro) UDN leaders scorned Vargas’s offer to form a national union cabinet because they saw it as a sign of presidential weakness, is consistent with the view that they would be satisfied with nothing less than his ouster; indeed, they informed Monteiro that the president’s resignation was a precondition for their cooperation.³⁸ The UDN press insistently challenged the sincerity and veracity of Vargas’s acts and words, and by September was hammering on the accusation that he was scheming to remain in office. “Let’s be alert for another 1937!” the Correio da Manhâ exhorted, echoing what had become one of the party’s battlecries. That the purpose of such criticism was to generate an atmosphere propitious to a golpe was made clear by the simultaneous argument that army chiefs could best atone for their blame in erecting the dictatorship by forcing Vargas out. “What the Nation hopes for, in exchange for that mistake, is that the Army will contribute now to ber definitive liberation from the Estado Novo and from Sr. Getúlio Vargas,” one editorialist exclaimed on September 15.³⁹ The conspirators naturally had maintained their recruiting drive within military ranks, and by now had found a key ally in Chief of Staff Barcelos, who in turn won over several other senior officers in the federal capital.⁴⁰

It is in the light of military insistence on a change of government, a fact that constituted an implicit rebuke, and especially in that of the oppositions bitter hostility and demands for his removal, that Vargas’s response to queremismo should be viewed. Aside from the fact that it would have been patently discriminatory and even illogical in the context of redemocratization to suppress queremista activities, it was undoubtedly flattering to Vargas after 15 years in power to see energetic signs of working-class support, particularly in view of the liberal clamor for his political head. In this regard, Ambassador Gainer sent an interesting report to London on August 21 in which he transmitted a cogent view of Vargas’s intentions obtained from “persons close to the President,” who told him “that Vargas is quite determined not to accept office but that he is allowing the ‘queremista' campaign to go on for reasons of personal vanity and to ‘show’ that even though there is demand for him, he is still determined to retire and rest.” According to Gainer’s informants, the “show of strength” by Vargas’s followers would also “deter his enemies from taking any violent measures against him personally and let him out more gracefully than if there was only a unanimous call for his retirement.” Remarks by Vargas several days later to a boisterous queremista crowd gathered in front of his residence also rang true. “I am avenged,” he declared. “To the man who draws near the end of his public activities and who has no other desire than to retire to the tranquility of his home, this demonstration that I have just witnessed is profoundly moving and eloquent.”⁴¹

The liberal conspirators, however, had their own interpretation of Vargas’s conduct, and their cause drew sustenance from two other developments during the middle months of 1945; the disaffection of Dutra and the extreme sensitivity of Monteiro about his role in 1937. The diminutive Dutra, although a man of bedrock patriotism, proven physical courage, and a reputation for honesty, was not known for a dynamic personality, breadth of vision, or ability as a public speaker. His conversion to democratic politics, moreover, was an abrupt one stemming from his political ambition, because he had been anything but liberal until 1945—a fact that broad sectors of the electorate realized. The only reason that his candidacy was imposed was that he occupied the post of minister of war and Vargas needed to act in order to forestall what seemed to be the likelihood of a golpe by Gomes’s military supporters. The relationship between Vargas and the lackluster Dutra had always been formally correct and devoid of any personal warmth, and the general in retrospect thought he detected some “bitterness” on Vargas’s part when he informed him of the decision to make him the government’s candidate. His campaign under way, Dutra sparked no enthusiasm. The general, Ambassador Gainer reported to the Foreign Office late in April, had “neither the personality nor the following to tackle the more serious problems which must confront the future President,” a judgment that Berle shared. Dutra himself later recalled incidents that reflected his lack of political resonance, even within the PSD; worried party leaders at one point thought of endeavoring to persuade him to withdraw in favor of a more attractive candidate and sounded an unreceptive Monteiro about the idea.⁴² Dutra found no shortcomings in himself, but attributed his failure to generate support solely to Vargas’s public indifference and, ultimately, to his alleged maneuvering on behalf of his own undeclared candidacy. Vargas, in fact, was not keenly supportive of Dutra, especially after he learned of the general’s frequent contacts with spokesmen for the liberal opposition—contacts that Dutra subsequently admitted. In those circumstances, Dutra’s position in the cabinet became increasingly uncomfortable, prompting him finally to submit his resignation on July 28, well before the September 2 deadline.⁴³

The upshot of Dutra’s alienation—and it is tempting to speculate that his understandable embarrassment and disillusionment made it easy for him to see ulterior motives in Vargas’s attitude—was that he and a group of army officers loyal to him developed their own military option long before he relinquished his cabinet post. General Mendes de Moraes, with Dutra’s blessings, found the initial converts within the ranks of senior generals, Pires and various fellow regional commanders among them, and then recruited a network of officers in the federal capital, methodically making the transfers necessary to ensure a decisive base of support. The conspiracy, which called for the capacity and readiness to intervene forcefully should Vargas somehow renege on his pledges, was “completely articulated,” Moraes subsequently wrote, by Augnst. According to a highly sympathetic analyst of Dutra’s role in the events of 1945, during his long weeks as minister-candidate he “pretended to trust the President,” knowing full well, as the “innumerable” transfers of officers were taking place, that Vargas’s freedom of action was being progressively restricted “with every day that passed.⁴⁴

Monteiro’s role in the unfolding crisis was a difficult one. He had been a close political and military ally of Vargas since the Revolution of 1930, of which he had been the major military leader. He had commanded the main federal army during the 1932 civil war, and more importantly, as a stern anticommunist and advocate of a strong, centralized regime that would impose order on society, he had been the prime mover of the antidemocratic conspiracy in 1937. He remained on intimate terms with Vargas, and believed that the chief executive was sincere in his avowed desire to turn over the government to a legally elected successor; after all, Vargas could hardly do otherwise, given the overwhelming military opposition to alternative courses of action. But Monteiro was under relentless pressure from liberal constitutionalists who had never been his allies and who scored him now for his part in setting up the Estado Novo. “The opposition became increasingly vehement and I was the favorite target of attacks during political rallies and in the press . . .,” he later remembered. “I found myself caught between two fires.”⁴⁵ The general was also under siege by close friends and long-time associates among the “historic” revolutionaries who had fought with him in 1930 and 1932, but who now had joined the liberals under the UDN banner, were committed to Vargas’s removal, and wanted Monteiro’s cooperation. During meetings of generals that he chaired and informal gatherings of former revolutionary cronies in August and September, officers stumping for Gomes openly challenged Monteiro’s view that Vargas intended to hold elections. General Osvaldo Cordeiro de Farias, a former tenente who recently had returned from service with the FEB to become one of the principal UDN conspirators, was one of those who “stirred things up” at the meetings, hammering away at the army’s responsibility for the Estado Novo and hence its responsibility to ensure a peaceful transition to democratic rule; Távora, too, at one session in mid-September, disputed Monteiro’s argument that Vargas was sincere.⁴⁶ In addition, officers who had mobilized for possible action under Dutra’s supervision insistently sought reassurances from Monteiro, implying that they also did not completely trust his repeated guarantees that the elections would be held on December 2. On Monteiro’s list of priorities, furthermore, the unity of the army ranked above all else, and he was profoundly distressed by the cleavages that the political crisis was causing within military ranks. On at least two occasions in October he wrote to fellow generals bemoaning the sharp divisions of opinion manifested in meetings of the high command—and by division of opinion he meant disagreement with him regarding Vargas as much as he meant debates between the groups supporting the two military candidates—seeing in the lack of consensus a harbinger of chaos.⁴⁷

The picture emerging from scrutiny of Monteiro’s conduct during the last few weeks of Vargas's reign is thus that of a man endeavoring to remain loyal to the president, but beset by myriad pressures that sharpened his feelings of guilt and ultimately made him eager to demonstrate his own commitment to a democratic change of government. By late October, in other words, Monteiro seems to have been in such a frame of mind that, should Vargas make a gesture that could be interpreted as provocative and, as a result, should countermeasures be demanded by those with bitter grievances against Vargas, Monteiro might overreact, especially if in so doing he could eradicate the source of dissension within the senior officer corps.

Berle’s Intervention

Political liberalization, at least in Berle’s eyes, diluted the negative political dimensions of the regime and increased the visibility of Vargas’s interest in social welfare. The favorable image that the United States envoy held of the Brazilian leader was further strengthened when neither Dutra nor Gomes addressed fundamental socioeconomic issues. “The political pot is boiling but the resulting broth is not very nourishing . . .,” he reported to the Department of State on March 21. “There is a growing feeling that in an open election at this point President Vargas, were he a candidate, could probably win, largely because the labor elements and the lower classes would vote for him.”⁴⁸ As he studied the situation and pondered Vargas’s liberalization measures, Berle concluded that Vargas was sincere in his disavowals of any intention to endeavor to succeed himself, and so indicated in various messages to both the department and Truman.⁴⁹ As the presidential campaign wound its way toward military intervention, the ambassador remained unimpressed by both the liberals’ capacity and willingness to tackle basic national problems and their assessment of Vargas’s ambitions. While UDN spokesmen voiced justifiable complaints about the Estado Novo, their own program was limited to “generalities about liberty and orderly constitutional practice,” and included no economic measures. “This is sound as far as it goes,” Berle advised Washington late in August, “but says very little to the man on the street who sees prices rising and economic service deteriorating.” As for charges that Vargas did not intend to relinquish power. Berle saw the queremista movement at this time as small, disorganized, and lacking in mass appeal, and he found no evidence that the “Constituinte com Getúlio” slogan being promoted in part by the communists represented a serious challenge to redemocratization; indeed, as he previously had pointed out, any attempt by Vargas to perpetuate his rule would provoke armed revolt.⁵⁰

From the beginning, Berle had debated the proper role for the embassy to play in the Brazilian political crisis. Liberal intellectuals, he learned immediately on arrival in Rio de Janeiro, demanded redemocratization and wanted “the United States to do something about that,” whereas the government expected continued U.S. support because of its record of wartime cooperation. Neither side, Berle realized, desired U.S. neutrality in their struggle. “The problem is to find and hold a moral position capable of comprehending all elements,” he reflected, “while we encourage a political solution capable of representing as nearly as possible the will of the Brazilian people.” Believing democratic rule to be in the interests of both countries and given his paternalistic impulses, the ambassador decided to encourage peaceful efforts to liberalize the political system. He therefore made it a point to meet privately with representatives of the opposition in order to better monitor their activities, and probably to discourage sentiment within UDN ranks in favor of a coup d’état.⁵¹ Beide's reply to a bitter anti-Vargas letter from one of Gomes’s close advisers in August can be read as a warning not to engage in extra-legal methods. United States opinion was delighted with Brazil’s “steady progress” toward effective democracy, he said, implicitly refuting UDN charges to the contrary, and then he added that “every true friend of Brazil hopes that the movement of events in that direction will continue without interruption.” In a cover note to the copies of the letter that he forwarded to the Department of State, Berle explained that the UDN leaders were trying to enlist U.S. support at a time when they were clamoring that Vargas “should get out” before presidential elections.⁵² The fact that Berle was convinced that Vargas sincerely intended to hand over the presidency peacefully to the victorious candidate strengthens the conclusion that, as of the beginning of September, his chief concern was to dissuade the opposition from resorting to undemocratic means to remove Vargas. The very day that he replied to Gomes’s adviser, in fact, he sent a lengthy telegram to Washington reporting that it was “perfectly clear” that Vargas did not intend to resign the presidency by September 2, as required by the electoral law if he desired to be a legal candidate, and that there was “no concrete evidence” that he would seek any other means of staying in power.”⁵³

Once the September 2 deadline had passed, Berle addressed at length in an official dispatch the question of whether the embassy should publicly condemn the Vargas government as Ambassador Spruille Braden was doing in Buenos Aires with regard to the Perón-dominated regime there and as “certain quarters” in Brazil—and here he meant the UDN—were urging. The answer, Berle suggested, lay in the myriad steps that Vargas had taken in recent months to create an open political system, as a result of which “every Brazilian now has available to himself all of the recourses available to any American during a political campaign.” The ambassador reiterated that all indications were that Vargas, although “far and away the most popular individual in the country,” was acting properly in the matter of the elections. “Under these circumstances, the Embassy proposes to continue its present policy of quietly encouraging the progress of Brazil towards democratic government,” especially since there was no reason to adopt “a controversial attitude, based on the assumption that all steps to date have been taken in bad faith.” Berle raised the same issues in a separate letter to Truman, pointing to the contrasts between Vargas and Perón, and repeating that the Brazilian leader had established a democratic framework for national politics. Consequently, he concluded, there was no reason for the embassy to adopt a different attitude. Truman and his advisers accepted Berle’s assessment, and on September 13 the president signed a reply agreeing that “it would be disastrous to interfere with the internal affairs of Brazil at the present time.” Indeed, Truman said, “it seems to me that things are going along as well as anyone would want. Vargas certainly has been our friend.”⁵⁴ Nonetheless, within two weeks Berle reached the conclusion that he had to intervene in the Brazilian presidential campaign by publicly stating the American position. What happened to provoke that demarche? What was the ambassador hoping to accomplish?

The constituinte com Getúlio and queremista movements that Berle had discounted in August appeared to gain momentum daily in mid-September. He began to fear the possibility that their adherents might generate sufficient pressure on the government to lead it to attempt to postpone the December 2 presidential elections and convoke a constitutional assembly instead, a move that inevitably would precipitate military intervention against the regime. On September 18, he wired the State Department suggesting that, if Vargas seemed inclined to cede to that pressure, the United States should endeavor to discourage the notion by publicly expressing satisfaction with the progress toward democracy in Brazil. Analysts at the Brazil desk liked the idea, but apparently Berle received no instructions in that regard. Meanwhile, he attentively followed developments and became increasingly uneasy. The interesting point is that he did not associate Vargas with the conspiracy that he thought was taking shape. When Braden passed through Rio de Janeiro on September 23 on his way to Washington, where he was to assume the post of assistant secretary of state, Berle discussed the political situation with him, pointing to the “somewhat unholy combination” between the PCB and an undefined “group of officeholders” who were said to be stimulating queremista sentiment. But he did not accuse Vargas of complicity; indeed, a few days later, after the alleged antidemocratic plot had assumed, in his eyes, a clearer profile, he specifically absolved Vargas. Because of the apparent conspirators’ plans to hold a massive demonstration on October 3, the anniversary of the Vargas-led 1930 revolution, Berle decided that the time had come for him to intervene with the public statement designed to deter them and any countermeasures by liberal plotters, a move that Braden had approved during his stopover. It was “fairly clear,” Berle said in an explanatory telegram to Washington on September 28, that in recent days “certain government elements ”—he mentioned as culprits Chief of Police Joáo Alberto Lins de Barros, Minister of Justice Agamenon Magalhães, and Minister of Labor Alexandre Marcondes Filho— had prepared a decree for Vargas’s signature that would transform the presidential elections into elections to a constitutional assembly and continue the dictatorship in the interim, but that Vargas had refused to sign such a document. “Embassy’s view is that President Vargas is endeavoring to keep things on track and go through with taking [the] country into democracy despite great pressure being exercised on him from Communist elements on one hand and Queremistas . . . on the other,” Berle reported.⁵⁵

There was, to be sure, no reason for the ambassador to think that Vargas was a party to any such maneuvering, and he justifiably did not have Vargas as his target as he made preparations to deliver his warning speech at a luncheon offered him by the Sindicato dos Jornalistas on September 29. Indeed, the U. S. envoy understandably thought that he would be assisting Vargas vis-à-vis those elements who apparently wanted to impede a transfer of power. That was why, when the Brazilian leader invited him to Guanabara Palace the evening before to discuss other matters, Berle took along a copy of the projected speech. Toward the end of their meeting, he brought up the subject of internal politics and praised Vargas. “We now saw various groups talking about a coup d’état,” be remembered saying, “in exact opposition to a policy which would place Vargas on a par with the greatest statesmen of South America.”⁵⁶ The Brazilian chief executive was emphatic in telling Berle that “under no circumstances" would he attempt reelection, because he had pledged his word not to do so and because he wanted to leave office in the near future while he still enjoyed the good will of the people. The ambassador then showed him the text of the speech he planned to give the next afternoon and explained its rationale: “Our fear was lest the hotheads would defeat the policy he had so wisely and brilliantly worked out during the last year.”⁵⁷ Vargas read the text, “had no objection and on the whole seemed pleased."⁵⁸ Indeed, “he greeted me kindly as I left,” Berle noted in his diary.⁵⁹ The ambassador’s impression was that Vargas was “a tired, sincere man struggling with great forces, no longer anxious for great power.” It is important to register that the reason Berle gave Vargas for wanting to deliver the speech represented, in fact, his real motive. “It is made on the theory that everything the Brazilian Government has done to date has been done in good faith,” he wrote in his diary the very next day, “and it rejects the charges of the enemies of the Government that the Government is secretly fomenting a Fascist coup d’état.”⁶⁰

On the morning of September 29, the U.S. envoy sent a joint telegram to Truman and Braden reporting his meeting with Vargas and the latter’s reaction to his imminent speech. “Subsequently he seems to have discussed it with two of his ministers,” Berle said, “one of whom has just called me to say that the govt [sic] is very happy it is being made.” Thus confident that he had the blessing of Catete Palace and that he actually would be doing something that would bolster the hand of the Brazilian leader, Berle made his controversial speech that afternoon. There was nothing subtle about the lengthy, entirely political address in which the ambassador, point by point, praised the various measures taken to open the political system and challenged Vargas’s adversaries. “The solemn promise of free elections in Brazil, set for a definite date, by a government whose word the United States always found inviolable, was acclaimed with as much satisfaction in the United States as in Brazil itself,” Berle declared. “The Americans do not agree with those who are trying to depict those solemn promises and declarations as insincere or mere verbal deceit.” Having thus admonished the queremistas in general and the UDN, he sent a signal to the queremista-communist bloc behind the constituinte-com-Getúlio movement, saying that it would be “tragic” if the presidential elections were needlessly interrupted until a constitutional assembly had been formed and completed its work.⁶¹

Berle seems to have been beset by doubts immediately as to the wisdom of his intervention, and, perhaps as part of an understandable process of self-reassurance, he began to magnify the nature of the challenge he was seeking to overcome. Thus, in both a letter to Truman and a telegram to the State Department on October 1, he suggested that the queremistas had planned to attempt a golpe on October 3, at best an unlikely notion given the army’s unmistakable opposition to anything but completion of the redemocratization program, but one to which Berle tenaciously clung as he implied that his speech had interrupted those alleged plans. Interestingly enough, he had pleasant, chatty words about Vargas for Truman, and in his message to the department he hailed Vargas’s “good sense” in dealing with the confused situation.⁶² In other words, Berle still believed that he and Vargas were acting in concert, even though the regime and its outspoken backers were in the process of rebuking the ambassador.

Probably because the opposition press immediately hailed the speech as a criticism of the dictatorship and a warning to the queremistas, Vargas did become upset. In conversation with Monteiro, he complained about Berle’s interference in Brazil’s domestic affairs, explaining, in reply to a query from the general, that he had been tired when the ambassador had gone over the speech with him on the evening of September 28 and had not realized its true significance. The chief executive later had Itamaraty inform the Brazilian Embassy in Washington that the prevailing opinion in Brazil was that the whole incident was “deplorable” and that it revealed “the goal, without precedent in Brazil’s history, of intervening in our internal life.” The outcry by progovernment publicists against Berle's remarks, which they attacked as an infringement of national sovereignty, probably also generated pressure on Vargas, who decided that a public rejection of foreign tutelage was in order. The occasion presented itself on October 3, when the huge queremista rally took place in front of the presidential palace. Reiterating that he was not a candidate, Vargas indirectly addressed Berle. "I do not need to search for examples or lessons abroad," he pointedly declared. "We possess the principles of democracy in our own tradition of social, ethnic, and political democracy.”⁶³

Berle remained on the alert that day and, after learning of Vargas’s statements, preferred to ignore the presidential rebuke, commenting in his diary only on Vargas’s reiteration that he would not be a candidate. "I still think he means what he says and is not engaged in double talk,” he wrote. In a telegram to the State Department the following day, Berle also avoided reference to Vargas’s reprimand, indulging instead in self-congratulation. "Assessing gains and losses from my own speech on Saturday I think it did contribute materially to stabilizing [the] situation . . .,” he offered. “Certainly [the] position of [the] US [sic] and of this Embassy in case of [a] coup d’état would have been extremely difficult.” The Brazilian government was not yet satisfied, however, and that same day Velloso called the ambassador in for a polite confrontation. Berle admitted that he had acted without explicit instructions from Washington, but that as "a friend of Brazil" he had deemed it necessary to make clear the position of the American government. “In view of this, I think it is best to forget everything in the hope that he does not repeat the gaffe . . .,” Velloso then informed the Brazilian embassy in Washington.⁶⁴

The Unnecessary Golpe

The political crisis now entered its decisive phase, as the UDN and pro-Dutra conspirators plugged the gaps in their positions and waited for the signal. Juracy Magalhães had gone on leave from his post in the South to organize the udenista revolt in Bahia, and by October he had a revolutionary general staff in place. In Minas Gerais the regional commander had prepared his units for action, and that month Juarez Távora helped him to remove some of the possible obstacles by persuading the commander of the Fourth Infantry Division that the army had to redeem its honor which, he argued, had been compromised in 1937. In Rio de Janeiro, meanwhile, Távora, Cordeiro de Farias, and other Gomes supporters had been promoting a spirit of rebellion against Vargas. As Cordeiro de Farias put it, in view of the universal demand within the garrison for compliance with the election schedule, "it fell to us, the most active ones, only to throw wood on the fire.” Various generals in other regions who formed part either of Dutra’s rearguard or the liberal constitutionalist group kept up the pressure on Monteiro, implicitly urging him to an even more dramatic display of fidelity to a change of government. On October 3, for example, the commander of the Third Military Region (Rio Grande do Sul), General Cesar Obino, reminded him of the “grave responsibilities” that the army had assumed in 1937—a way of pointing a guilty finger once again at Monteiro himself—which obligated it to guarantee, “by all possible means,” that Vargas respect his electoral pledges. The next day General Amaro Bittencourt, who was in charge of the Second Military Region (São Paulo), informed the minister that the officers under his command were “truly apprehensive and disconcerted, and pointedly recalled Monteiro’s repeated promises regarding the elections. Ary Pires further contributed to the exacerbation of Monteiro’s sensitivity and uneasiness by writing to him on October 8 suggesting that his attitude had been ambiguous. Pires indirectly criticized Monteiro's past role by condemning a hypothetical situation—and here he undoubtedly was making a not-so-subtle reference to the Estado Novo—in which the armed forces, in the absence of a legal, constitutional government, became "an instrument of despotism and oppression" at the service of arbitrary rulers. Pires ended his letter on an ominous note that revealed how far the pro-Dutra conspiracy had progressed: “Spontaneously or by force, the replacement of Sr. Getúlio represents the surest path for taking the Nation to the realization of its democratic longings.”⁶⁵

On October 10, Vargas generated a furor in liberal circles by issuing a decree that would move state elections, previously set for May 1946, up to December 2 to coincide with national balloting for president and Congress and giving federal interventors in the states 30 days to resign if they desired to run for office. To the opposition, the decree meant that Vargas was trying to rig the state elections by having them held under the supervision of his appointees. It also reinforced fears that he might suddenly institute further changes to ensure his own continuation in power. The UDN and two other opposition parties immediately issued a joint manifesto blasting the decree as “a political, moral and juridical monstrosity aimed at creating confusion propitious to his own ambitions, while sympathetic army officers expressed their anger to Monteiro. Távora, in the presence of Cordeiro de Farias, warned the minister that the decree could well “plunge the country into a blood bath,” and that some of it would “spatter his head.”⁶⁶ A group of 14 generals, headed by Chief of Staff Barcelos, sent a memorandum to Monteiro, for transmission to Vargas, demanding that no further electoral changes be made and that nothing occur to impede the elections on December 2.⁶⁷ General Bittencourt wrote Monteiro from São Paulo informing him that the majority of officers under his command were “convinced that the President intends to convoke the constituent assembly, even in opposition to the army and the guarantees given regarding elections on December 2.” The liberal press, meanwhile, seized on the episode to intensify its onslaught on Vargas and its simultaneous call for military intervention.⁶⁸

The decree did not provoke a coup d'état against Vargas for the simple reason that Dutra endorsed it—a fact omitted by his biographers and overlooked by scholarly analysts.⁶⁹ The measure resulted from pressure by federal interventors, typically members of the PSD, who hoped to protect their own positions at the state level in the post-Vargas era, and it represented a step that Dutra had recommended nearly a year earlier, and that was in consonance with the political interests he had developed since then. In his letter to Vargas in December 1944 on the need to restore the electoral process, the general had argued that, in order to avoid prolonged political instability, there should be simultaneous across-the-board elections rather than voting at different times for different levels of office.⁷⁰ As a presidential hopeful whose campaign was manifestly foundering, Dutra well may have reasoned that by endorsing the interventors' appeal he could win more enthusiastic support from them. At any rate, as Monteiro later testified, Dutra and PSD strategists approved the measure, and to Távora's disgruntlement, army officers in Dutra’s camp hailed it because they also thought it would enhance his chances of winning the presidency.⁷¹

The decree failed, then, to provide the catalyst that the opposition leaders sought, but they did not relent. Indeed, with an uneasy eye on Buenos Aires, where Argentine labor unions had mobilized in defense of Perón, who was under attack by some military sectors, the anti-Vargas press launched a full-scale offensive against the queremistas, their accusations reaching the proportions of near-hysteria as the month wore on. The Diário Carioca, Correio da Manhã, and the Diário de Notícias all kept the sinister trilogy of Vargas, queremistas, and communists constantly before the public, and by the third week of October were making a prediction that strained credibility: the chief executive within days would postpone the elections and convoke a constituent assembly, presenting the armed forces with a fait accompli.⁷² Conspiratorial activities within the Rio de Janeiro garrison gained momentum during the period. As Berle noted on October 26, the “main source” of political agitation was now “pro-Gomes army officers,” some of whom were “seriously” involved in preparations for an armed uprising.⁷³

In the face of redoubled attacks by his adversaries, Vargas could only hold to his course and give renewed assurances. In his speech of October 3 to the queremistas and in another several days later, he recognized their right to push for a constitutional assembly before presidential elections were held and declared the issue worthy of study—statements cited by the opposition as evidence of continuismo—but his decree of October 10 actually was a strong blow to the constituinte-com-Getúlio faction because it reaffirmed December 2 as national election day. The president also gave what Monteiro labeled “a very satisfactory reply” to the demands presented by the group of generals regarding the elections. On October 14, furthermore, he declared publicly that, despite the charges leveled by the opposition, the elections would be held on schedule and that there would be no further changes in electoral proceedings.⁷⁴ Then on October 23, after queremista organizers had announced plans for a massive demonstration in the federal capital on October 26, Vargas ordered its cancellation. Moreover, as the president of the Associação Brasileira de Imprensa informed Ambassador Gainer, the chief executive, at the same time, directed the newspapers that were backing the idea of transforming the December 2 elections into balloting for a constitutional congress to “modify their language and cut down space and importance hitherto devoted to this campaign.” On October 25, at his weekly official conference with the minister of war, Vargas once more assured Monteiro “that he intended to preside over an honest election and that he would deliver power to whoever was legally elected." Convinced that Monteiro was conspiring against the president, the latter’s chef de cabinet, Luiz Vergara, urged Vargas the following morning to have the general arrested. Vargas, however, rejected the idea, explaining that such a step would create more problems at a time when he wanted “nothing but to leave in peace."⁷⁵

For his part, Monteiro during the last three weeks of October found himself on the defensive, forced constantly to reiterate his own and the government’s commitment to presidential elections. He made the point in a message to Bittencourt on October 10 and five days later, after receiving a skeptical letter from the São Paulo commander, had to renew the pledge, assuring him that he would resign if Vargas should attempt to block the elections. But that would not happen, he asserted. “I have categorical statements from the President in that regard, all else being a war of nerves and partisan maneuvers by those from all political circles—opposition and government—interested in creating confusion and provoking a schism in the armed forces, he explained. Pires, too, remained doubtful, a fact that obviously wounded Monteiro. "It never entered my head to attempt to betray the commitments that I assumed before the army and, in its name, before the country to see that what is stipulated about the elections is carried out," he wrote in an entreating tone. A week later, he made yet another public vow that “the Government will not change its decision” regarding the elections; repeated it on October 24, calling on the contending parties to proceed more calmly; and still again on October 26. The next day, moreover, he issued a special proclamation to the army in which he was unequivocal; “In view of the present political situation . . .,” he declared, “I affirm to my comrades that the elections will be held on December 2 next and that, without a public agreement among the various political parties, no alteration of a political or electoral character will be made in existing legislation.”⁷⁶

Monteiro by now was obviously a man frustrated and made extremely sensitive by the enervating role he had been forced to play since assuming the portfolio of war. One of the movers and shakers of Brazilian politics since 1930, his influence and moral authority seemed to have waned along with his health. It was clear, at any rate, probably in large part because of his machinations in 1937—and there was the Correio da Manhã to remind the nation on October 28 that the army had abandoned the constitution eight years earlier⁷⁷—that he did not enjoy the complete confidence of his military colleagues. Indeed, his proclamation of October 27 could be read as an appeal for credibility. At the same time, moreover, that his anxiety to refurbish his image seemed to be reaching a peak, he thought he detected a certain “coldness” on Vargas’s part toward him. On the morning of October 27, Cordeiro de Farias called on Monteiro to inform him that he, the minister of war, was being “very undermined and defamed in Palace circles,”⁷⁸ all of which may have reinforced Monteiro’s desire to reassert his authority by renewing his credentials as an idealistic leader. Monteiro, in sum, seems to have been ready to show that he was a man of his word.

The opportunity came on October 29, when Vargas decided to appoint his younger brother Benjamin to the position of police chief in place of João Alberto Lins de Barros, who was shifted to the post of prefect of the Federal District. That decision has never been adequately explained, but Vargas apparently was seeking to do a last-minute favor for long-time ally Barros, who had political ambitions, and saw city hall as a trampoline to electoral power. Barros himself recommended the president’s brother as his successor.⁷⁹ Whatever Vargas’s motives, the opposition instantly interpreted his action as a sign that he was going to attempt a political golpe and call off the elections. Cordeiro de Farias learned from Barros the evening before of the impending appointments, “and I went about alerting people like mad,” he later recalled. “That same night I spread the news among the conspirators, mobilizing all the generals who, during our meetings, had declared themselves against Vargas’s continuing in office.” After he was informed, Dutra, early on the morning of October 29, summoned the manager of his military team, General Monies, and excitedly told him that Vargas had made his long-expected move—“lt’s the golpe,” he said— and Monies, who later admitted to having held his fellow conspirators in check on more than one occasion that month, then mobilized the generals and other senior officers loyal to Dutra.⁸⁰ For the lackluster PSD candidate, of course, acting forcefully to help remove Vargas would represent revenge of sorts for the president’s indifference to his candidacy, and it might also refurbish his public image.

For Monteiro, his outburst was testimony to his state of raw sensitiveness and emotional vulnerability—in short, to his readiness to be provoked into bold action. Since Cordeiro de Farias had alerted him the previous evening, he had several hours to convince himself that Vargas was actually throwing the gauntlet down. When Barros called at his residence early the next morning to inform him officially of the changes, Monteiro did not wait for an explanation of the president’s motives. While Barros tried to calm him and assure him that no irrevocable steps had been taken, Monteiro exclaimed among other things that “it was inconceivable and inadmissible that I should come to suffer, at the end of my life, a blow so brutal”; that Vargas’s “political career was liquidated”; that the chief executive apparently judged him to be “like the scoundrels he was used to dealing with”; and that “as a soldier, I preferred to die before submitting to a humiliation that would be the negation of my past and would cast me into disgrace.”⁸¹ Once at the Ministry of War, Monteiro put army units on alert, and sent a telegram to regional commanders saying that he was resigning and was “going to take a stand.” Later that day, he wrote out a letter of resignation to Vargas, citing physical exhaustion and the fact that “I am surprised by events that lead me to believe that I am not being well understood in my goals of serving well the Nation and Your Excellency’s government.” He then assumed official control of what was now a unified conspiracy and began moving the necessary military pieces around the tactical board.⁸²

Vargas’s conduct during the crisis was hardly that of someone in the middle of preparations for a show of force with the military establishment. Cordeiro de Farias was surprised to see that no effort was being made by presidential or police agents to keep him under surveillance, a fact he attributed to the probability that Vargas was “so confident” that he deemed it unnecessary. When Dutra kept a prearranged appointment with Vargas early on the evening of the 29th, the president’s demeanor was perfectly normal. His understandable initial reaction, when Dutra informed him of the high command’s opposition to the plan to place his brother in charge of the police department, was one of surprise and irritation that his authority to make appointments was being challenged. But on hearing from Dutra that troops were already in the streets, Vargas, in the general’s words, “showing a good deal of calm, proposed to me that he cancel the act of Benjamin’s appointment, saying that he was willing to designate to the position an officer agreeable to the army.” When Dutra carried that proposal to a gathering of senior officers, including Gomes and other representatives of the air force and navy, it was rejected and Cordeiro de Farias was then given the assignment of informing the president that he was being deposed. “He did not lose his poise,” Cordeiro wrote of Vargas’s response to the news.⁸³

Berle had watched the unfolding of the crisis that month with foreboding and confusion. “I have been in a good deal of mental turmoil,” he confessed to his diary on October 4, “because . . . I don’t like to get into political controversy and always torment myself with wondering whether I did the right things.”⁸⁴ It is difficult to escape the conclusion that he ultimately resolved his dilemma simply by adopting the opposition’s thesis that Vargas was plotting to perpetuate his own rule. The need to justify his interference, in other words, perhaps best explains Berle’s subsequent about-face regarding the dictator. Certainly his reports up to the fateful October 29 offer no convincing reasons either for the golpe or for the ambassador’s subsequent judgments. On October 11, for example, he informed Washington that Vargas was giving “every evidence of going forward with national elections on December 2,” and he pointed to the weakness of the queremista movement, indicating that the opposition had adduced no proof of its charge that the government secretly was subsidizing it. Four days later, he reported Vargas’s public denial that any behind-the-scenes arrangements had been made which were contrary to his stated program or that any political surprises were in store, and on October 24 and 26 he emphasized the administration’s order to cancel the queremistas’ scheduled demonstration. They now had only “faint hope” of somehow achieving an extension of Vargas’s tenure, he commented in one telegram on October 26, and in a second telegram he suggested that the police and army had the draft-Vargas movement well in hand. A report by Ambassador Gainer during this period on the concern in the United States embassy “over the situation and especially at the possibility of the Army assuming [the] Government”⁸⁵ is implicit support for the argument that Berle did not see any threats to liberalization emanating from Catete Palace.

As soon as the coup d’état of October 29 was over, however, Berle suddenly saw virtue in the opposition’s claims—an exercise that required him to reinterpret previous events, but that conveniently vindicated his speech of September 29. In his first lengthy explanation of developments to the Department of State on October 30, he was cautious and basically limited himself to describing the victors’ view of history, but notes in his diary reveal how completely he had succumbed to the temptation. If Vargas had resigned to become a legal candidate, he would have won the elections, the ambassador reflected, “but . . . with a rather disreputable Palace gang, he tried to continue his dictatorship . . ..” Over the next few days, Berle became more openly bold. In a letter to a diplomat-friend on November 6, he incredibly described Vargas’s recent moves as “an attempt to reestablish the Fascist dictatorship as it was in 1937,” a maneuver “almost immediately and spontaneously defeated by a country which really wants Democracy.” That Berle now cast his speech in a new light and was beginning to claim a critical role for himself in Vargas’s downfall is clear from a letter he wrote to his father that same day. “As you see, I have been moving out here and events seem to have justified the result,” he boasted, going on to cite his speech. The following evening, in a telegram to the department, Berle was emphatic. It was now evident that Vargas, in the “latter part of September,” had been party to a plot to renew the dictatorship on October 3, a move postponed until late October and then scheduled for early in November, the ambassador reported. “Had [the] coup d’état [on] October 3 come off, with [the] planned resumption of [the] situation as it stood when dictatorship was established in 1937 with [the] U.S. sitting by and saying nothing,” he offered, “our moral stock would have been at [an] all-time low.” But if the United States had attacked the new dictatorship, he continued, it would have provoked a crisis in relations with Rio de Janeiro and possibly pushed Vargas into an understanding with the military regime in Buenos Aires. “In actual circumstances I think it will appear that my speech [of] September 29, so far from being meddlesome, was [the] safest course to take,” he concluded.⁸⁶

Reassurance from Washington eliminated whatever remaining doubts Berle may have entertained. On November 1, Braden signed a letter to Berle congratulating him for his “splendid handling” of developments in Brazil, and a few days later, Truman sent Berle a letter in which he implicitly endorsed the ambassador’s intervention. Secretary of State Byrnes also sent Berle a telegram of support and congratulations. The Department of State had “carefully reviewed” the whole incident and was satisfied that on net balance you were correct in making your address after first clearing it with President Vargas as you did,” he informed Berle, praising the “skillful manner” in which the ambassador had conducted himself. The commendation seemed to call for drama: Berle replied that the decision to deliver the speech had been “close and difficult,” but, he insinuated, it had helped to avert a bloody civil war in Brazil.⁸⁷

It is now clear that Berle and the ex-dictator both distorted the ambassador’s role in the process by which the Estado Novo came to an end. On the one hand, nothing in Berle’s past nor in his conduct as ambassador supports Vargas’s wild charge a year later that he had been part of some conspiracy by undefined international economic forces.⁸⁸ On the other hand, the notion that Vargas was scheming to prolong his regime, propitiously borrowed by Berle from Brazilian liberal and military conspirators, seems equally groundless. Vargas was universally recognized as an archrealist and probably the cleverest politician in the country. He certainly understood political muscle, and he knew how the Estado Novo had been established and why it had endured for eight years. To suppose that he would defy two military candidates and the determined opposition of Minister of War Monteiro and base a power play on the queremistas not only was tantamount to labeling Vargas an unintelligent, power-mad liar, but challenged logic as well. Berle himself, before Vargas’s fall, correctly characterized queremismo as weak and disorganized; indeed, once the military had ousted Vargas, it disintegrated. Thus Berle, in a private postmortem dated November 14, could write that “apparently it was almost artificial.”⁸⁹ Did Berle perceive something that Vargas did not? If not, and Vargas thought that sufficient working-class strength could be rapidly mobilized to enable him to override the high command, why did he not act when Dutra informed him that the army was moving against him, especially since General Odylio Denys, the commander of the Military Police, and General Renato Paquet, commander of the strategic Vila Militar, were both loyal to Vargas and were ready to defend him against the conspirators? Instead the dictator sent orders to both generals not to take any action.⁹⁰

“. . . I don’t know what Vargas was basing himself on to believe that he had sufficient force to impose his continuation in power . . .,” General Cordeiro de Farias later commented. “The situation was clearly defined: it was Getúlio on one side and the army on the other.”⁹¹ The puzzle solves itself if Vargas was not plotting against the elections as part of some scheme to remain in office. To challenge the liberal-military thesis about the downfall of Vargas is not to gainsay the validity of the complaints of his adversaries about his past political conduct, especially the conspiracy of 1937 and the whole array of arbitrary acts—arrests, imprisonment, exile, censorship, and general suppression of civil liberties—committed by the regime during the period 1937-44, nor is it to deny the probable sincerity of their perceptions of his intentions during the redemocratization process. It is simply to point out that the activities of interested parties in late 1945 strongly indicate that Vargas was deposed because of those previous events and what they signified for his credibility rather than because of what he was doing and intended in 1945. Vargas, in other words, in October of that year, became a victim of the past and not the present.